摘要:

1995 EUROPEAN WINTER ROYAL SOCIETY OF CHEMISTRY LON FERENCE ON PLASMA 8 -23 January 2995 CAMB RI DG E ANALYTI CAL D IVI S I0 N Great Britain SC I ENTI F I C P ROG RAMM E S P EAKE RS The 1995 European Winter Conference will for the first time be held in Plenary lecturers will include:- Professor M BIades (Univ British Columbia Canada) Professor K Niemux (ISAS Dortmund Germany) Professor H M (Skip) Kingston (Duquesne Univ USA) the UK in the historical University City of Cambridge the centre of some of the most important scientific developments of recent times. The conference will be based upon a single lecture stream featuring five plenary and five invited speakers plus contributed papers. Symposia will include the following topics:- 1) Sample introduction phenomena 2) Glow discharge techniques 3) Fundamental plasma processes 4) Sample preparation and handling 5) Elemental speciation Professor L Ebdon (Uniu Plymouth UK) Professor J Caruso (Univ Cineinnat4 USA) Invited lecturers will include:- Professor J Oksik (Ohio State Univ USA) Professor R K Marcus (Clemson Univ USA) Dr S J HasweU (Univ HuU UK) 6) Plasma source mass spectrometry Professor R Nesbitt (Univ Southampton UK) 7) Laser sampling techniques Dr K Kawabata (Yokogawa Analytical Systems Japan) 1995 EUROPEAN WINTER CONFERENCE ON PLASMA SPECTROCHEMISTRY If you would like to receive the final circular and registration form please complete and return to:- Janice M Gordon European Winter Conference Royal Society of Chemistry Thomas Graham House Science Park Milton Road Cambridge CB4 4WF United Kingdom NAME ADDRESS I intend to submit a paper/poster on:1995 European Winter Conference on Plasma Spectrochemistry 8-13 January 1995 CAMBRIDGE UK Short Courses A series of short courses of one half day duration will take place on Sunday 8th January.Notes and tuition material will be distributed with each course. Courses 1 and 2 Short Courses on ICP-MS Professor R.S. Houk Ames Laboratory Iowa State University USA Course 1 (AM) Instrumentation and Theory The course will cover fundamental aspects of ICP-MS including:- a) Molecular beam sampling b) Quadrupole and high resolution c) Vacuum technology d) Ion sources e) Detection systems and data hand1 ing f) Sample introduction technologies analys ers Course 2 (PM) Advanced Topics The course will cover more advanced topics on ICP-MS particularly relevant to problem solving.Each topic will be illustrated with relevant applications examples. a) Interferences (spectroscopic and non-spectroscopic and methods of alleviation b) Isotopic analysis c) Chromatographic methods d) Overview of commercial instrumentation Course 3 (PM) Sample Preparation for ICPs Dr S.J. Haswell Hull University UK The course will focus on important aspects of sampling and sample preparation with particular emphasis on ICP measurements. a) Batch methods f o r wet oxidation b) Recent trends in microwave preparation for ICP-MS atomic spectrometry general analytical techniques c) On-line sample preparation d) Extraction methods e) On-line chemical processing f) Miniaturization Course 4 (PM) Speciation Professor O.X.Donard University of Bordeaux France The course will focus on practical aspects of speciation analysis with particular emphasis on ICP and other plasma sampling systems. Sample collection and handling preservation and preparation prior to injection into hyphenated systems using atomic spectrometry and ICP-AES or ICP-MS as detectors will be illustrated with applications from current topical fields . a) Sampling and sample pretreatment b) Separative techniques Differential chemistry Gas liquid ion and SCF c ) Interfacing chromatography techniques to ICPs and other plasma sources and detectors chromatographies Course 5 (AM) Quality Systems in the Laboratory Professor L. Ebdon Dr E.H. Evans University of Plymouth UK The course will discuss how high quality analytical data can be produced in the laboratory that are accurate reliable and adequate f o r the intended purpose. a) Quality assurance principles b) Sampling and sample preparation c) Personnel aspects d) Statistics for quality control e Use of reference materials and f) Equipment and records maintenance g) Audits and accreditation. traceability Course 6 (AM) Sample Presentation for ICPS Dr C McLeod Sheffield Hallam University UK The course is intended as a problem solving workshop and will attempt to rationalise the choice of sampling system for ICP spectrometries by use of practical examples. a) Nebulisation techniques Traditional and high efficiency The role of desolvation Hydride Other vapour techniques e . g . b) Vapour generation Hg oso c) Microsampling systems d) Flow injection e) Laser ablation

ISSN:0267-9477    

DOI:10.1039/JA99409BP015

出版商:RSC    

年代:1994

数据来源: RSC

摘要:

JOURNAL OF ANALYTICAL ATOMIC SPECTROMETRY OCTOBER 1994 VOL. 9 55N Symposium (7th 1994 Seventh Biennial National Atomic Spectroscopy BNASS) Hull UK 20-22 July Jane Emma Pallanca Most delegates to the 7th BNASS at Hull arrived in good time despite the best efforts of British Rail staff who (with a sense of timing usually lacking) had called a strike for the first day of the conference. After lunch Professor A. Townshend gave us a gracious welcome to the University of Hull and promised that we could look forward to a confer- ence featuring an array of renowned plenary speakers plus invited and contributed lectures covering a wide range of topics associated with atomic spectroscopy. Proceedings got off to a rousing start with a plenary lecture from Professor Rick Browner (Georgia USA) who gave us a detailed and fascinating overview of the recent progress of sample introduc- tion methods in ICP-AES and ICP-MS.Sample introduction has traditionally been considered the ‘Achilles heel’ of ICP-based methods as typically 98-99% of analyte solution is lost on nebuliz- ation. This is not only wasteful but can also lead to peak broadening in ‘hyphen- ated’ techniques such as LC-ICP-MS. Professor Browner explained how new nebulizers and sample interfaces com- bining low sample volumes and efficient desolvation methods largely do away with these problems. With the exception of the plenary lec- tures the subsequent presentations were divided into two streams. Dr Helen Crews (MAFF) gave an invited lecture on Multi-element Analysis of Biological Materials by ICP-MS- The Reality.We were asked to take a long cool look at the problems that the analyst faces with such a remit. Quality assurance in par- ticular requires a degree of compromise and a more flexible approach than for single element analysis and presentation of results should be tailored to the cus- tomer’s needs. Of the contributed lectures I would like to single out those of Dr Andreas Stroh on Analysis of Dificult Sample Matrices by ETV-ICP-MS and Martin Bloxham on On-line Preconcentration and Matrix Removal for Sea-water Analysis by Flow Injection with ICP-MS Detection. Both lectures stressed the problems associated with trace element analysis in complex samples and showed us how to avoid some of the pitfalls. The first of two poster sessions followed.There were approximately two dozen posters of a generally very high stan- dard and a good deal of lively discussion and exchange of ideas ensued. Faced with the prospect of giving my first conference lecture the following day I had planned a quiet evening. My good intentions were scotched when I was approached by Dr Steve Hill and Janice Gordon and asked to write this report- from a ‘BNASS virgin’s eye view’. My participation in the evening’s entertain- ment was therefore essential. A treasure hunt through Beverley had been arranged passing many sites of historic interest and an even greater number of attractive public houses-a detail much appreciated by the participating teams of delegates. A good time was had by all despite accusations of ‘insider deal- ing’ of vital information! Thursday began with a plenary lecture on Clinical and Environmental Appli- cations of Atomic Spectroscopy by Dr Trevor Delves (Southampton) with emphasis on analysis of trace levels of analytes by methods such as ETAAS and ICP-MS.Complex organic samples such as blood can be ashed with oxygen or merely diluted. High levels of aluminium are associated with renal dialysis and drinking soya milk whereas manganese can reach toxic levels in the blood of patients on parenteral nutrition. In ICP-MS analysis of mercury use of an ammonia-EDTA diluent was shown to reduce ‘memory’ effects. Mention was also made of the novel use of ICP-MS boron isotope ratio measurements in the monitoring of brain cancer patients treated with radioactive boron- containing drugs.Following through the clinical theme Dr David Halls (Glasgow) gave an invited lecture on Analytical Minimalism Applied to the Determination of Trace Elements in Clinical and Biological Samples indicating how we can reduce time cost amount of sample energy requirements and waste by judicious sample preparation and analytical prac- tice. Sample preparation time can be reduced by analysing sample slurries and digestion procedures can be adapted for microsamples by careful optimiz- ation. He also showed how fast furnace analysis can be made even faster using hot injection and high temperature drying. After giving my own contributed lec- ture I stayed in the same stream to hear lectures by Mr I. Harrison on The Determination of Selenium in Human Hair and Nail by Microwave Digestion and ETAAS and Mr T.J. Burden on Optimal Accuracy Precision and Sensitivity Using ICP-OES Examples with the Bioanalysis of Aluminium-which emphasized the importance of line choice and spectral interpretation. Dr Reinhold Klockenkamper ( I S Germany) faced with the unenviable task of keeping the delegates awake in the stifling heat of the post-lunch session rose to the occasion and spoke with great aplomb on Total Reflection XRF-Physical Principles and Analy- tical Applications. Dr Klockenkamper described how TXRF requires only microgram or microlitre samples on a totally reflecting glass carrier. The normal problems of low detection (due to continuous background) and low accuracy (due to matrix effects) associ- ated with other methods when directly analysing complex samples such as dusts pigments and sections of organic tissue are largely overcome by this technique with absolute detection limits for 70 elements in the range of 1-1Opg.Professor Stan Greenfield followed with The ICP in Fluorescence Spec- trometry -Source and Atomllon Reser- voir. He described how a low power ICP can be used as an atomizer in which the source is either another low power56N JOURNAL OF ANALYTICAL ATOMIC SPECTROMETRY OCTOBER 1994 VOL. 9 Invited speakers and guests enjoying the conference dinner plasma a high power plasma or a hollow cathode lamp. Low power ICP-AFS gives lower limits of detection for refrac- tory elements such as Al B and Ba whereas ASIA (atomizer source induc- tively coupled plasma in atomic fluor- escence spectrometry) is superior for non-refractory elements such as Ca Ce Cu and Fe.HCL-ICP-AFS rivals ETAAS for limits of detection particu- larly for elements with lines below 200 nm. The contributed lectures in the remain- der of the afternoon sessions focused on aspects such as assessment of spectral interference and background correction (Professor David Littlejohn) multivari- ate analysis (I. F. Savage) and wave- length calibration using charge coupled device detectors (D. A. Sadler). The sub- sequent poster session was every bit as impressive as the previous one. After a long hot and very interesting day we were whisked back to Halls of Residence for a quick wash and brush up then returned to the Myton Room at the University for the Conference Dinner.After a convivial meal we were then treated (?) to a performance of ‘Ode to a British Rail Steak and Kidney Pie’ by Conference Secretary Dr Steve Haswell. Entertainment was continued by a local magician who rounded off his act by guillotining Dr Haswell-to the delight of his audience. The evening con- cluded with the results of the treasure hunt-a dead heat! In the light of dubi- ous practices attributed to the Plymouth team there was a ‘sudden death’ play- off in which the team leaders had to say certain words and phrases with a north- ern accent. Dr Hill’s accent was roundly derided by the local delegates and the prize was awarded to the AT1 Unicam team led by Dr Kathryn Thomas. The final day got off to an excellent start with the plenary lecture by Dr Dave Styris (USA) on The Elucidation of Electrothermal Atomizer Mechanisms- Surface Science at Very High Tem- peratures.He gave an overview of surface science in atomization processes and surface/sub-surface reaction mech- anisms. By vaporizing under high vacuum as well as atmospheric pressure it is possible to separate homogeneous gas phase reactions from the hetero- geneous reactions involving the con- densed phases which can be charac- terized by MS or extended X-ray absorp- tion fine structure spectroscopy. Dr Gillian Greenway (Hull) followed with an invited lecture on On-line Sample Pre-treatment for Atomic Spectrometric Detection. In doing so she returned to a recurring theme of this conference-how to achieve trace element analysis in a complex sample matrix.In this case trace elements were preconcentrated from sea-water on mini columns using chelates such as L-cysteine and 8-hydroxyquinoline analogues or more exotically with algae on controlled- pore glass. Continuing the theme of sample prep- aration Angelo Grillo spoke on Adaptive Control of Microwave Digestions show- ing how simultaneous optimized sample digestion can be achieved on several samples using a closed vessel system. This was followed by Graham Coe who talked about the use of FI microwave digestion showing how this approach reduces both the amount of reagents used and time taken for sample prep- aration. Philip Rock concluded the morning session with his lecture on the use of hydroxylamine hydrochloride as a reducing agent for Pd used as a matrix modified in ETAAS.Following lunch I)r Clive Thompson (Yorkshire Water) kicked off the final session with his Invited lecture on The Utilization of Vapour Generation Tech- niques in Environmental Analysis. The need to perform trace level analysis on large numbers of samples at Yorkshire Water requires simple robust method- ologies with large linear calibration ranges. Vapour generation techniques fulfil these requirements and also remove the analyte from the sample matrix (thus reducing potential interferences) and allow preconcentration of the analyte prior to quantitation. Despite same problems (e.g. the limited number of commercially available intense stable sources background from atomization systems and light scatter interference) 11 elements can now be usefully analysed using vapour generation techniques with linearity from 1 ng 1-l to 10 mg 1-l.Dr Hywel Evans then gave a lecture on Atomic and Molecular Speciation by GC-LP-ICP-MS showing how adjust- ment of forward power and gas flow allows the production of atomic and molecular ions allowing analysis of organometallic and organohalide species. The conference ended with a bang (rather than a whimper) with Professor Sam Houk’s plenary lecture on Innovations in Elemental Analysis by ICP- MS. Likening his campaign to further improve the performance of ICP-MS to the WW2 landings on the Normandy beaches he described making multiple ‘pincer’ movements on polyatomic inter- ferences and precision problems in iso- tope ratio measurements. Polyatomic interferences were attacked by destroying oxides removal of solvents and moving Members of JAAS editorial board discussing policy!JOURNAL OF ANALYTICAL ATOMIC SPECTROMETRY OCTOBER 1994 VOL.9 57N Delegates f rom Spain enjoying strawberries on a fine summer evening the MO’ mountain (by reshaping the plasma). Cryogenic desolvation and the use of graphite injectors were dis- cussed. Alternatively a micropump can be used to disperse the sample which is then introduced as a dry jet into the plasma using a direct injection nebulizer. Precision problems of isotope ratio measurements were largely put down to noise associated with plasma sources but can be circumvented by the use of double beam (magnetic sector type) instruments or twin quadrupole ICP-MS. No doubt Professor Houk has plans to take ICP-MS ‘where no man has gone before... !’ Dr Steve Hill had the pleasant duty of awarding some prizes prior to closing the meeting. The Royal Society of Chemistry 1994 Hilger Spectroscopy Prize for a young scientist making an outstanding contribution to atomic spec- troscopy was awarded to Dr Hywel Evans (Plymouth). The prizes for the best student oral presentation and the best student poster were won by Martin Bloxham and Robert McNeill respect- ively. Finally Dr Barry Sharp (ASG) thanked the Organizing Committee for their excellent handling of the conference after which the delegates dispersed. For those who stayed on the evening’s entertainment began with an oppor- tunity to take part in a space-age laser gun battle at Laserquest. Evidently some people just couldn’t resist the chance to shoot their colleagues in the back! The hungry survivors returned for a barbecue at Cleminson Hall. My final impressions of the 7th BNASS? It’s a great way of being part of the cutting edge of atomic spec- troscopy exchanging ideas and making friends. Don’t forget the 8th BNASS at the University of East Anglia in 1996; I wouldn’t want to miss it and neither should you! Jane Emma Pallanca Department of Environmental Sciences University of Plymouth Drake Circus Plymouth UK

ISSN:0267-9477    

DOI:10.1039/JA994090055N

出版商:RSC    

年代:1994

数据来源: RSC

摘要:

1995 European Winter Conference on Plasma Spectrochemistry 8-13 January 1995 CAMBRIDGE UK Short Courses A series of short courses of one half day duration will take place on Sunday 8th January. Notes and tuition material will be distributed with each course. Courses 1 and 2 Short Courses on ICP-MS Professor R.S. Houk Ames Laboratory Iowa State University USA Course 1 (AM) Instrumentation and Theory The course will cover fundamental aspects of ICP-MS including:- a) Molecular beam sampling b) Quadrupole and high resolution c) Vacuum technology d) Ion sources e) Detection systems and data hand1 ing f) Sample introduction technologies analys ers Course 2 (PM) Advanced Topics The course will cover more advanced topics on ICP-MS particularly relevant to problem solving. Each topic will be illustrated with relevant applications examples.a) Interferences (spectroscopic and non-spectroscopic and methods of alleviation b) Isotopic analysis c) Chromatographic methods d) Overview of commercial instrumentation Course 3 (PM) Sample Preparation for ICPs Dr S.J. Haswell Hull University UK The course will focus on important aspects of sampling and sample preparation with particular emphasis on ICP measurements. a) Batch methods f o r wet oxidation b) Recent trends in microwave preparation for ICP-MS atomic spectrometry general analytical techniques c) On-line sample preparation d) Extraction methods e) On-line chemical processing f) Miniaturization Course 4 (PM) Speciation Professor O.X. Donard University of Bordeaux France The course will focus on practical aspects of speciation analysis with particular emphasis on ICP and other plasma sampling systems.Sample collection and handling preservation and preparation prior to injection into hyphenated systems using atomic spectrometry and ICP-AES or ICP-MS as detectors will be illustrated with applications from current topical fields . a) Sampling and sample pretreatment b) Separative techniques Differential chemistry Gas liquid ion and SCF c ) Interfacing chromatography techniques to ICPs and other plasma sources and detectors chromatographies Course 5 (AM) Quality Systems in the Laboratory Professor L. Ebdon Dr E.H. Evans University of Plymouth UK The course will discuss how high quality analytical data can be produced in the laboratory that are accurate reliable and adequate f o r the intended purpose.a) Quality assurance principles b) Sampling and sample preparation c) Personnel aspects d) Statistics for quality control e Use of reference materials and f) Equipment and records maintenance g) Audits and accreditation. traceability Course 6 (AM) Sample Presentation for ICPS Dr C McLeod Sheffield Hallam University UK The course is intended as a problem solving workshop and will attempt to rationalise the choice of sampling system for ICP spectrometries by use of practical examples. a) Nebulisation techniques Traditional and high efficiency The role of desolvation Hydride Other vapour techniques e . g . b) Vapour generation Hg oso c) Microsampling systems d) Flow injection e) Laser ablation1995 European Winter Conference on Plasma Spectrochemistry 8-13 January 1995 CAMBRIDGE UK Short Courses A series of short courses of one half day duration will take place on Sunday 8th January. Notes and tuition material will be distributed with each course.Courses 1 and 2 Short Courses on ICP-MS Professor R.S. Houk Ames Laboratory Iowa State University USA Course 1 (AM) Instrumentation and Theory The course will cover fundamental aspects of ICP-MS including:- a) Molecular beam sampling b) Quadrupole and high resolution c) Vacuum technology d) Ion sources e) Detection systems and data hand1 ing f) Sample introduction technologies analys ers Course 2 (PM) Advanced Topics The course will cover more advanced topics on ICP-MS particularly relevant to problem solving. Each topic will be illustrated with relevant applications examples.a) Interferences (spectroscopic and non-spectroscopic and methods of alleviation b) Isotopic analysis c) Chromatographic methods d) Overview of commercial instrumentation Course 3 (PM) Sample Preparation for ICPs Dr S.J. Haswell Hull University UK The course will focus on important aspects of sampling and sample preparation with particular emphasis on ICP measurements. a) Batch methods f o r wet oxidation b) Recent trends in microwave preparation for ICP-MS atomic spectrometry general analytical techniques c) On-line sample preparation d) Extraction methods e) On-line chemical processing f) Miniaturization Course 4 (PM) Speciation Professor O.X. Donard University of Bordeaux France The course will focus on practical aspects of speciation analysis with particular emphasis on ICP and other plasma sampling systems.Sample collection and handling preservation and preparation prior to injection into hyphenated systems using atomic spectrometry and ICP-AES or ICP-MS as detectors will be illustrated with applications from current topical fields . a) Sampling and sample pretreatment b) Separative techniques Differential chemistry Gas liquid ion and SCF c ) Interfacing chromatography techniques to ICPs and other plasma sources and detectors chromatographies Course 5 (AM) Quality Systems in the Laboratory Professor L. Ebdon Dr E.H. Evans University of Plymouth UK The course will discuss how high quality analytical data can be produced in the laboratory that are accurate reliable and adequate f o r the intended purpose. a) Quality assurance principles b) Sampling and sample preparation c) Personnel aspects d) Statistics for quality control e Use of reference materials and f) Equipment and records maintenance g) Audits and accreditation. traceability Course 6 (AM) Sample Presentation for ICPS Dr C McLeod Sheffield Hallam University UK The course is intended as a problem solving workshop and will attempt to rationalise the choice of sampling system for ICP spectrometries by use of practical examples. a) Nebulisation techniques Traditional and high efficiency The role of desolvation Hydride Other vapour techniques e . g . b) Vapour generation Hg oso c) Microsampling systems d) Flow injection e) Laser ablation

ISSN:0267-9477    

DOI:10.1039/JA99409FX057

出版商:RSC    

年代:1994

数据来源: RSC

摘要:

58N JOURNAL OF ANALYTICAL ATOMIC SPECTROMETRY OCTOBER 1994 VOL. 9 Diary of Conferences and Courses 1994 6th International Colloquium on Solid Sampling With Atomic Spectroscopy October 11-13 Amsterdam The Netherlands Details can be found in J. Anal. At. Spectrom. 1993 8 59N. For further information contact Dr. R. F. M. Herber Coronel Laboratory University of Amsterdam Meibergdreef 15 NL-1105 AZ Amsterdam The Netherlands. Third Rio Symposium on Atomic Spectrometry November 6-12 Venezuela Details can be found in J. Anal. At. Spectrom. 1993 8 64N. For further information contact Pro- fessor Jose Alvarado Universidad Simon Bolivar Departamento de Quimica Laboratorio de Absorcion Atbmica Apartado postal No. 89000 Caracas 1080-A Venezuela. Fax (0058-2-) 938322/5719134/5763355/9621695. Anal ytica '94-Second National Sym- posium on Analytical Science December 1994 Western Cape South Africa Details can be found in J.Anal. At. Spectrom. 1993 8 60N. For further information contact Dr. I. M. Moodie c/o PO Box 1970 Tygerberg 7505 South Africa. Fax 021-932-4575. 1995 1995 Winter Conference on Plasma Spectrochemistry January 8-13 Cambridge UK For further information contact Janice M. Gordon The Royal Society of Chemistry Thomas Graham House Science Park Milton Road Cambridge CB4 4WF UK. Telephone +44 (0) 223 420066; fax $44 (0) 223 420247. Pittcon '95 The Pittsburgh Conference on Analytical Chemistry and Applied Spectroscopy March 5-10 New Orleans Louisiana USA Details can be found in J. Anal. At. Spectrom. 1994 9 49N. For further information contact The Pittsburgh Conference 300 Penn Center Boulevard Suite 332 Pittsburgh PA 15235-5503 USA.Telephone (412) 825-3220; toll free (800) 825-3221; fax (412) 825-3224. Fourth International Conference on Pro- gress in Analytical Chemistry in the Steel and Metals Industry May 16-18 Jean Monnet Building Luxembourg Details can be found in J. Anal. At. Spectrom. 1994 9 50N For details of providing a contribution to the programme or other information contact CEC/CETAS Conference R. Jowitt British Steel plc Technical Teesside Laboratories PO Box 11 Grangetown Middlesbrough Cleveland TS6 6UB. Telephone $44 642 467144; fax +44 642 460321 43rd ASMS Conference on Mass Spec- trometry and Allied Topics May 21-25 Atlanta GA USA For further details contact ASMS 1201 Don Diego Avenue Santa Fe NM 87501 USA.Telephone 505 989 4517. Fax 505 989 1073. Vth COMTOX Symposium on Toxi- cology and Clinical Chemistry of Metals University of British Columbia Van- couver British Columbia Canada Details can be found in J. Anal. At. Spectrom. 1994 9 26N. July 10-13 Colloquium Spectroscopicum Inter- nationale (CSI) XXIX August 27-September 1 Leipzig Germany Details can be found in J. Anal. At. Spectrom. 1993 8 50N. Colloquium Spectroscopicum Internation- ale (CSI) XXIX Post Symposium ICP- MS September 1-4 WernigerodelHartz Germany Details can be found in J. Anal. At. Spectrom. 1994 9 46N.JOURNAL OF ANALYTICAL ATOMIC SPECTROMETRY OCTOBER 1994 VOL. 9 59N International Symposium on Environmen- tal Biomonitoring and Specimen Banking December 17-22 Honolulu Hawaii USA This symposium is being held as part of the International Chemical Congress of Pacific Basin Societies (PACIFICHEM 95) sponsored by the American Chemi- cal Society Canadian Society for Chem- istry Chemical Society of Japan New Zealand Institute of Chemistry and the Royal Australian Chemical Institute.Papers for oral and poster presentations are solicited on topics that will focus on monitoring of organic pollutants; moni- toring of trace metal pollutants; exposure assessment; and biomarkers and risk assessment/management. The deadline for receipt of abstracts on the official Pacifichem 95 abstract form is MARCH 31 1995. For further details contact K. S. Subramanian Environmental Health Directorate Health Canada Tunney's Pasture Ottawa Ontario K1A OL2 Canada (phone 613-957-1874; fax 613-941-4545) or G.V. Iyengar Center for Analytical Chemistry Room 235 B 125 National Institute of Standards and Technology Gaithersburg MD 20899 USA (phone 301-975-6284; fax 301-921-9847) or M. Morita Division of Chemistry and Physics National Institute for Environmental Studies Japan Environmental Agency Yatabe- Machi Tsukuba Ibaraki 305 Japan (phone 81-298-51-6111 ext. 260; fax 81- 29 8 -5 6-4678). 1996 1996 Winter Conference on Plasma Spectrochemistry Fort Lauderdale Florida USA January 8-13 Details can be found in J. Anal. At. Spectrum. 1994 9 53N. For further information contact Dr R. Barnes ICP Information Newsletter Department of Chemistry Lederle GRC Towers University of Massachusetts Box 34510 Amherst MA 01003-4510 USA. Telephone (413) 545 2294; telefax (413) 545 4490. International Schools and Conferences on X-Ray Analytical Methods January 18-25 Sydney Australia Details can be found in J. Anal. At. Spectrom. 1994,9 47N. For further information contact AXAA '96 Secretariat GPO Box 128 Sydney NSW 2001 Australia. Telephone 61 2 262 2277. Fax 61 2 262 2323. Telex AA 176511 TRHOST.

ISSN:0267-9477    

DOI:10.1039/JA994090058N

出版商:RSC    

年代:1994

数据来源: RSC

摘要:

1995 European Winter Conference on Plasma Spectrochemistry 8-13 January 1995 CAMBRIDGE UK Short Courses A series of short courses of one half day duration will take place on Sunday 8th January. Notes and tuition material will be distributed with each course. Courses 1 and 2 Short Courses on ICP-MS Professor R.S. Houk Ames Laboratory Iowa State University USA Course 1 (AM) Instrumentation and Theory The course will cover fundamental aspects of ICP-MS including:- a) Molecular beam sampling b) Quadrupole and high resolution c) Vacuum technology d) Ion sources e) Detection systems and data hand1 ing f) Sample introduction technologies analys ers Course 2 (PM) Advanced Topics The course will cover more advanced topics on ICP-MS particularly relevant to problem solving. Each topic will be illustrated with relevant applications examples.a) Interferences (spectroscopic and non-spectroscopic and methods of alleviation b) Isotopic analysis c) Chromatographic methods d) Overview of commercial instrumentation Course 3 (PM) Sample Preparation for ICPs Dr S.J. Haswell Hull University UK The course will focus on important aspects of sampling and sample preparation with particular emphasis on ICP measurements. a) Batch methods f o r wet oxidation b) Recent trends in microwave preparation for ICP-MS atomic spectrometry general analytical techniques c) On-line sample preparation d) Extraction methods e) On-line chemical processing f) Miniaturization Course 4 (PM) Speciation Professor O.X. Donard University of Bordeaux France The course will focus on practical aspects of speciation analysis with particular emphasis on ICP and other plasma sampling systems.Sample collection and handling preservation and preparation prior to injection into hyphenated systems using atomic spectrometry and ICP-AES or ICP-MS as detectors will be illustrated with applications from current topical fields . a) Sampling and sample pretreatment b) Separative techniques Differential chemistry Gas liquid ion and SCF c ) Interfacing chromatography techniques to ICPs and other plasma sources and detectors chromatographies Course 5 (AM) Quality Systems in the Laboratory Professor L. Ebdon Dr E.H. Evans University of Plymouth UK The course will discuss how high quality analytical data can be produced in the laboratory that are accurate reliable and adequate f o r the intended purpose.a) Quality assurance principles b) Sampling and sample preparation c) Personnel aspects d) Statistics for quality control e Use of reference materials and f) Equipment and records maintenance g) Audits and accreditation. traceability Course 6 (AM) Sample Presentation for ICPS Dr C McLeod Sheffield Hallam University UK The course is intended as a problem solving workshop and will attempt to rationalise the choice of sampling system for ICP spectrometries by use of practical examples. a) Nebulisation techniques Traditional and high efficiency The role of desolvation Hydride Other vapour techniques e . g . b) Vapour generation Hg oso c) Microsampling systems d) Flow injection e) Laser ablation1995 European Winter Conference on Plasma Spectrochemistry 8-13 January 1995 CAMBRIDGE UK Short Courses A series of short courses of one half day duration will take place on Sunday 8th January. Notes and tuition material will be distributed with each course.Courses 1 and 2 Short Courses on ICP-MS Professor R.S. Houk Ames Laboratory Iowa State University USA Course 1 (AM) Instrumentation and Theory The course will cover fundamental aspects of ICP-MS including:- a) Molecular beam sampling b) Quadrupole and high resolution c) Vacuum technology d) Ion sources e) Detection systems and data hand1 ing f) Sample introduction technologies analys ers Course 2 (PM) Advanced Topics The course will cover more advanced topics on ICP-MS particularly relevant to problem solving. Each topic will be illustrated with relevant applications examples.a) Interferences (spectroscopic and non-spectroscopic and methods of alleviation b) Isotopic analysis c) Chromatographic methods d) Overview of commercial instrumentation Course 3 (PM) Sample Preparation for ICPs Dr S.J. Haswell Hull University UK The course will focus on important aspects of sampling and sample preparation with particular emphasis on ICP measurements. a) Batch methods f o r wet oxidation b) Recent trends in microwave preparation for ICP-MS atomic spectrometry general analytical techniques c) On-line sample preparation d) Extraction methods e) On-line chemical processing f) Miniaturization Course 4 (PM) Speciation Professor O.X. Donard University of Bordeaux France The course will focus on practical aspects of speciation analysis with particular emphasis on ICP and other plasma sampling systems.Sample collection and handling preservation and preparation prior to injection into hyphenated systems using atomic spectrometry and ICP-AES or ICP-MS as detectors will be illustrated with applications from current topical fields . a) Sampling and sample pretreatment b) Separative techniques Differential chemistry Gas liquid ion and SCF c ) Interfacing chromatography techniques to ICPs and other plasma sources and detectors chromatographies Course 5 (AM) Quality Systems in the Laboratory Professor L. Ebdon Dr E.H. Evans University of Plymouth UK The course will discuss how high quality analytical data can be produced in the laboratory that are accurate reliable and adequate f o r the intended purpose. a) Quality assurance principles b) Sampling and sample preparation c) Personnel aspects d) Statistics for quality control e Use of reference materials and f) Equipment and records maintenance g) Audits and accreditation. traceability Course 6 (AM) Sample Presentation for ICPS Dr C McLeod Sheffield Hallam University UK The course is intended as a problem solving workshop and will attempt to rationalise the choice of sampling system for ICP spectrometries by use of practical examples. a) Nebulisation techniques Traditional and high efficiency The role of desolvation Hydride Other vapour techniques e . g . b) Vapour generation Hg oso c) Microsampling systems d) Flow injection e) Laser ablation

ISSN:0267-9477    

DOI:10.1039/JA99409BX059

出版商:RSC    

年代:1994

数据来源: RSC

摘要:

JOURNAL OF ANALYTICAL ATOMIC SPECTROMETRY OCTOBER 1994 VOL. 9 267 R ATOMIC SPECTROMETRY UPDATE-ATOMIC MASS SPECTROMETRY AND X=RAY FLUORESCENCE SPECTROMETRY Jeffrey R. Bacon The Macaulay Land Use Research Institute Craigiebuckler Aberdeen UK A69 2QJ Andrew T. Ellis* Oxford Instruments Industrial Analysis Group 19-20 Nuffield Way Abingdon Oxfordshire UK OX1 4 1 TX Adam W. McMahon Department of Chemistry Manchesfer Metropolitan University John Dalton Building Chester Street Manchester UK M I 5GD Philip J. Potts Department of Earth Science Walton Hall The Open University Milton Keynes Buckinghamshire UK MK76AA John G. Williams NERC ICP -MS Facility Royal Holloway and Bedford New College Department of Geology Egham Hill Egham Surrey UK TW20 OEX Summary of Contents 1. Atomic Mass Spectrometry 1.1 Introduction 1.2.Accelerator Mass Spectrometry (AMS) 1.3. 1.4. Glow Discharge Mass Spectrometry (GDMS) Inductively Coupled Plasma Mass Spectrometry (ICP-MS) 1.4.1. Reviews 1.4.2. Fundamental studies 1.4.2.1 Operating parameters and instrument development 1.4.2.2. Sample introduction 1.4.2.3. Interference effects 1.4.3.1. Geochemical 1.4.3.2. Environmental 1.4.3.3. Biological 1.4.3.4. Industrial 1.4.3. Applications 1.5. Laser Ionization Mass Spectrometry (LIMS) 1.6. Resonance Ionization Mass Spectrometry (RIMS) 1.7. Secondary Ion Mass Spectrometry (SIMS) 1.8. Spark Source Mass Spectrometry (SSMS) 1.9. Sputtered Neutral Mass Spectrometry (SNMS) 1.1 Q. Stable Isotope Ratio Mass Spectrometry (SIRMS) 1.1 1. Thermal Ionization Mass Spectrometry (TIMS) 1.1 2.Other Methods 2. X-ray Fluorescence Spectrometry 2.1. 2.2. 2.3. 2.4. 2.5. 2.6. 2.7. Reviews Instrumentation 2.2.1. X-ray optics and microfluorescence instrumentation 2.2.2. Excitation 2.2.3. Detectors Total Reflection X-ray Fluorescence (TXRF) 2.3.1. Chemical analysis 2.3.2. Surface analysis and depth profiling X-ray Microprobe Field Analysis Calibration and Data Processing 2.6.1 Fundamental parameter (FP) calculations 2.6.2. Matrix correction using scattered radiation Applications 2.7.1. Specimen preparation 2.7.2. Chemical analysis 2.7.3. Thin films 2.7.4. Geological * Review Co-ordinator to whom correspondence should be addressed.268 R JOURNAL OF ANALYTICAL ATOMIC SPECTROMETRY OCTOBER 1994 VOL. 9 2.7.5. Environmental 2.7.6. Archaeology and forensic 2.7.7. Industrial 2.7.8.Clinical and biological 2.7.9. Chemical effects Strong developments continue in most areas of atomic spectrometry and this review year has seen the introduction of new instrumentation and methodologies refinements and improvements of established procedures and applications to new fields of study. The large number of papers published on ICP-MS and TIMS in particular demonstrates the high level of activity in these areas and the continuing need to lower the detection limits improve sensitivities and precisions and to explore new areas of application. Many of the techniques are very versatile and atomic MS in general can be used to undertake virtually any type of elemental or isotopic analysis on any type of matrix. Although developments in the coupling of chromatography with MS have been slow there is still seen to be an interest in this promising development.1. ATOMIC MASS SPECTROMETRY 1.1. Introduction This year's Update follows the same format used last year. It does not set out to be comprehensive in its coverage but is based on critical selection of developments in instrumentation and methodology notable for their innovation originality or achievement of significant advances. It has been necessary to include a small number of conference papers which were only considered for inclusion if they were judged to include enough detail and information to give evidence that they meet the above criteria. Routine applications of atomic MS are not generally included in the Update and readers are referred to the Updates on industrial analysis metals chemicals and advanced materials (93/3400) environmental analysis (93/3417) and clinical and biological materials foods and beverages (94/739).Foreign language papers that are unlikely to reach a wide readership have not been included but those that appear to be making a significant contribution to develop- ments in atomic MS have received consideration. The large number of papers in the current review year gives further evidence of the very healthy activity in most of the techniques covered. Most atomic MS techniques continue to receive close attention and advances have been considerable. There is an ever-growing requirement for all types of analysis (bulk microprobe isotope) in a wider range of matrices. The need to press limits (detection resolution spatial or depth resolution precision and accuracy) grows ever further and advances continue to be made in all these areas.It is notable that every year more MS techniques associated mainly with organic analysis are evaluated for application to atomic analy- sis. The versatility of MS in general is apparent and it is clear that many analyses can be performed through the use of one of a number of different MS techniques. 1.2. Accelerator Mass Spectrometry (AMS) A review with many references (93/2977) discussed the history of AMS and its advantages over decay counting methods. Another review with 40 references (94/1440) covered the basic concepts of AMS and some of the new areas of application such as biomedical studies measurement of 59Ni and selective excitation of ions with lasers.The principal sources that are currently being or could potentially be used for applications of AMS to materials science have been outlined in a review with over 300 references (93/3654). Emphasis was placed on recent improvements to existing sources and new source devel- opments which show promise. Developments in new instrumentation have been limited this review year. A new ion source and sample manipulation system has been designed at the Toronto facility which incorporated remote Csf primary ion beam generation and rapid inter- change between samples. The design included dual Cs primary beams improved primary and secondary beam optics and the facility for high capacity pumping sample viewing optics and a capacity of 72 targets.A tandem instrument designed for highly sensitive analysis (93/2976) combined a laser source time-of-flight MS a charge-exchange chamber and a MS for measurement of both positive and negative singly-charged ions. Of particular significance for the development of cheaper instrumentation was the use (93/3681) of a cyclotrino for the detection of negative ions produced at relatively low energy (40 keV). Chinese papers have described the design and set-up of an electrostatic deflector for AMS (93/3792) and the use of a gas ionization chamber to identify and eliminate isobars for example 36S in the determination of 36Cl. The most widely reported application of' AMS continues to be to radiocarbon dating. The principles instrumentation methodology and applications of radio-isotope dating by AMS have been described (94/1378) with particular regard to archae- ological and paleoanthropological studies.An example of such an application was the dating of iron artefacts (93/3571) which required only the equivalent of < 5 mg of carbon (10 g wrought iron or 250 mg high-carbon steel or cast iron) for analysis. Improvements in instrumentation for radiocarbon dating have included a dedicated tandem-based instrument at the Peking facility (94/1448) which was designed to achieve highly sensitive analysis by paying attention to the background and fraction- ation effects. Preliminary applications to the carbon dating of archaeological samples were performed at a background level of with an accuracy of better than 1.7%.A 40keV negative ion injector has been installed (93/3505) on an instru- ment fitted with a Cs sputter ion source and an achromatic arrangement of four dipole magnets with two electrical slot lenses. This arrangement was designed to achieve minimal aberrations for the beam injected into a tandem. Analytical improvements in radiocarbon dating continue to be made to try to push back the limits on reliable dating. One approach (94/1447) was to understand better the sources of background and contamination associated with the sample preparation and the ion source. By understanding the individ- ual contributions it was possible to date samples reliably up to 60000 BP. A procedure has been developed (94/1481) for the combustion of several hundred microgram amounts of a chemically-treated organic sample and conversion of the result- ant CO to graphitic carbon.The latter yielded relatively high I3C- ion currents and low blanks to give I4C count rates at <0.20% modern consistently giving a 2s age limit of > 50 000 BP. Sources of error in radioazrbon dating were shown (94/1473) to be dominated by the background problems in the MS arising from contamination of very small samples. Although the atmospheric concentrations of 14C were found to vary with time and latitude the short variations in the production rate of atmospheric I4C were considered to be strongly attenuated in the relatively large atmospheric I4C pool. The variations doJOURNAL OF ANALYTICAL ATOMIC SPECTROMETRY OCTOBER 1994 VOL. 9 269 R provide however essential data for the calibration of dynamic models of the global carbon cycle.Dates for coral samples (94/1474) displayed discrepancies from those obtained by U-Th dating. This was attributed to contamination by a small (0.5-2.5%) amount of modern carbon which can produce considerable offsets in 14C dates of samples older than about 10000 years. The AMS measurements of cosmogenic 14C in terrestrial rocks at high altitudes were found (94/1475) to give lower numbers of 14C atoms per gram of rock than were expected from other estimates of the rock surface age. This was interpreted as evidence for the importance of both erosion and complicated exposure histories for the samples. It was considered that measurement of more than one cosmogenic isotope is probably necessary to define a real exposure age.The combination of SIMS with AMS continues to receive attention through the use of high-intensity Cs sputter-ion sources. Such a source has been equipped (94/1439) with a spherical Mo ionizer a 20-position target wheel and a vacuum lock for loading and unloading batches of samples. Reliable insulation of the cathode and control of the Cs vapour were achieved through the use of a metal-ceramic bonded section protected by a specially designed labyrinth shielding system. Three versions of the source have been constructed-a single target version a multitarget version with manual sample change and a multitarget version with automated sample change. The combination of ion or laser excitation with superconducting minicyclotron-based AMS was considered (93/3645) to hold considerable advantages over conventional SIMS with better analytical sensitivity and isotope resolution.Sensitive detection and submicron resolution are also advan- tages that can be exploited in the quantitative imaging of radiocarbon labels widely employed in biomedical research (94/1449). The potential of AMS in the study of the geochemistry of the platinum group elements (PGEs) and the metallogenesis of Ni-Cu-PGE sulfide ore deposits was demonstrated (94/1359) by the in-situ determination of all the PGEs and Au in individual minerals (pyrrhotite pentlandite and chalcopyrite). Enrichment of Os Pd Rh and Ru and depletion of Au Ir and Pt in the sulfide minerals suggested that the first four existed as metal sulfides during crystallization of the sulfide melt whereas the other three remained as stable metals. Although there were problems of calibration and quantitation in the determination of PGEs in sulfide minerals (94/1356) the sub- ng g-' limits of detection and short data collection times made it the most sensitive method available. An interesting development has been the determination qf' 59Ni (t+= 76 000 years) produced by the effects of cosmic rays in extraterrestrial matter (94/1443). With a beam energy of 641 meV and using the full-stripping technique it was possible to separate cleanly the 59Ni28+ ions from the 59C027+ back- ground ions.A 59Ni Ni ratio of (2.3 k0.4) x 10- l1 was meas- ured in a sample of stony-iron meteorite. Attention has also been paid to other long-lived radionuclides. Existence of negative ions for all the lanthanide elements except Er Ho and Pm has been demonstrated (94/1450).The relative ion yields varied for each element with that for Dy being dependent on the terminal voltage. It was suggested that use of negative ions would require a reduction of the electric field gradients used to accelerate the ions. A simple ion- exchange method for the extraction of 1291 from the heavy water system of a nuclear reactor allowed (93/2914) concen- trations of 1011-1013 atoms 1-1 to be determined on samples of 0.5 ml. Procedures based on leaching with an HF-HN03 mixture in a heated ultrasonic tank have been developed (93/2942) for the separation of the quartz phases from terres- trial rocks and soils (10-200 g) for the purpose of measuring in situ "Be and 26Al in the quartz phase.The absolute measurement of "Be 9Be ratios in certified reference materials identified (94/1454) discrepancies of up to 14% from the certified values. 1.3. Glow Discharge Mass Spectrometry (GDMS) Developments in the instrumentation and applications of GDMS continue to be very active and it is notable that they follow closely the path taken by spark source MS three decades ago. In this review year reports have appeared on fundamental aspects of ion formation development of instrumentation methodologies using relative sensitivity factors (RSFs) and an understanding of interference effects. Although the majority of applications are for the analysis of high-purity metals the analysis of non-conducting materials and more complex matrices (for example soils) are being explored.Isotopic analy- ses have also been reported. Some of these developments have been covered in reviews by Marcus and co-workers on new developments (94/1501) GD devices (including a table of commercially-available instrumentation) (94/1138) and oper- ation principles and design considerations for r.f. powered GD devices (94/2176). Developments in instrumentation have concentrated in the main on the ion source. In the co-axial magnetron GD device a co-axial magnet pair was placed (93/3421) behind the cathode to achieve stable plasma operation for all pressures from 0.05 Pa to over 330 Pa. Measurements of mass loss atomic emission intensities and ionization characteristics suggested that the plasma was dominated by ionization processes rather than recombination processes.This has implications for the success- ful use of lower pressures than are commonly used in GD devices. Analytical detection limits (0.9-2.0 p.g g-' for a GD operated at 4 mA and 1000 V) and relative ion yields (0.57-3.5 relative to the copper matrix) were measured (93/2908) for a jet-type ion source. Substantial advances have been made in the design of r.f. powered GD sources. The design of the r.f. discharge source coupled to a double-focusing MS and electrical system has allowed (94/1282) resolution of problems of inadequate r.f. shielding maintaining the necessary d.c. bias potential on the sample surface preventing r.f. modulation of ion energies and coupling of the accelerating potential to the discharge.Preliminary results showed relatively small differences in ion yields across the Periodic Table. An r.f. plasma sputter type source has been developed (94/1503) for the production of heavy negative ions. The use of a 13.56 MHz r.f. discharge with a three-turn r.f. coil ( 5 mm diam.) produced a dense plasma of 1 0 " ~ m - ~ . Total negative ion currents of 6.5 mA (Cu) and 4.2 mA (C) were achieved for sputtering targets (42 mm diam.) with a negative bias of 600 V against the plasma. Considerable effort is being placed on a better understanding of plasma processes in the discharge. This has ranged from numerical simulation of the discharge to model the behaviour of electrons and ions (94/1090) to more experimental approaches based on time-resolved studies of ion production and the use of cooling to change the reactivity of plasma species.Time-resolved studies (93/3433 94/1290) showed that the signals for analyte atoms and ions and metastable argon atoms reached a sharp maximum within 2 ms of termination of discharge power. The rapid recombination of argon ions with electrons brought about by the termination of discharge power resulted in the production of the metastable argon atoms. Sputtered atom ionization by the Penning mechanism domi- nated during this period. The signal profiles of ionized sputter atoms were influenced by the discharge parameters (discharge power duty cycle support gas compositions and sampling distance). The time-dependent nature of ion signals confirmed the possible use of time-gated detection to discriminate against spectral background species.Interactions between trace gaseous impurities in the discharge have been studied (94/604 94/1284) using a cryogenic coil to remove the impurities by freezing. The presence of water vapour and other gaseous species in the discharge resulted in reactions with matrix element species and the consequent reduction of analyte species in the plasma. Removal of the270 R JOURNAL OF ANALYTICAL ATOMIC SPECTROMETRY OCTOBER 1994 VOL. 9 con taminan ts cryogenically resulted in significant increases in both the sputtering gas and the cathode matrix ion signals. This was demonstrated for the m/z range from 10 to 40 in which interferences on A1 and Mg signals could be removed effectively. Other aspects of instrument operation have been considered.A study of the influence of operating parameters on ion intensity showed (93/3445) that whereas sample shape did not influence ion intensity signals the sample position sample size discharge current and discharge voltage did have a significant influence. The maximum ion intensities for elements were obtained with the sample 8-9mm from the ion exit slit. In recognition of the potential problem of interferences a spectral interference database has been compiled (93/3365). Inter- ferences were divided into groups (isobarics oxides doubly charged background dependent matrix dependent argides and dimers) and could be calculated for all masses of all elements. The number of studies on the optimization of discharge parameters (voltage current and pressure) for depth projiling indicates the growing importance of this application.In the analysis of multi-layer samples it was possible (94/605) to produce flat-bottomed sputter cratFrs with a depth profile resolution as good as 0.03 pm (300 A). Sputter rates could be controlled to within *lo% and could be chosen to profile various coating thicknesses from 0.1 to 200 pm. The most significant parameter that affected the formation of flat- bottomed craters was found (93/3361) to be the inner diameter of the anode. A sputter rate of 0.2 pm min-' was achieved (93/3488) in the analysis of coated copper wires using a flat- type ion source operated under optimum conditions of 6 mm anode-cathode separation source parameters of 4 mA - 1000 V and 1 mbar and sampler and skimmer voltages of 40 and -280 V respectively. Detection limits of several pg 8-l were obtained for most elements but some interferences from residual gases were observed.The major application of GDMS is the analysis of high- purity metals in particular of steels. Although this must now be considered relatively routine standardization and cali- bration of analysis continue to receive attention. A method for the calibration of steel analysis was (93/3384) based on the preparation of high-purity graphite used for the supporting electrode doped with a synthesized standard solution contain- ing 18 elements. Such a method does not seem to take into account differences in matrix and interference effects between the two types of sample.Methodologies are being developed for the analysis of other high-purity metals. Comparison of Ar and Ne plasma gases found (93/2904) that the spread of relative ion-yield values was considerably smaller for Ne (0.2-2.5) than for Ar (0.1-5). Particular attention to sample preparation is required in the analysis of gallium (93/2893) due to problems of impurity segregation during melting and resolidification. Such problems were best avoided by preparation of samples by rapid freezing of molten metal and it was possible to determine up to 70 impurity elements with detection limits at the ng g-' level and accuracies better than 40%. Analysis of high-purity molyb- denum was achieved (93/2921) using a high-sensitivity dis- charge cell (the Mega cell) and a liquid nitrogen sample cooling system.Samples shaped mechanically into rods were acid washed thoroughly before analysis and were sputtered for up to 2 h before determination at the ng g-' level. Precisions were about 2% at the pg 8-l level. Problems of interference on the Ti lines were noted in this work also. The analysis of non-conducting materials by GDMS is more demanding than the analysis of metals and is usually achieved through the use of a secondary electrode or by mixing with a conducting powder. Recent studies have shown that in both methods tantalum is the best matrix to use. Tantalum was the only suitable secondary electrode material for the sputtering of oxide glass samples (93/3368). The sputtering yield of glass atoms was highly dependent on the discharge pressure because of the redeposition of cathode atoms.Of several conducting matrices used to mix with samples tantalum produced the simplest spectrum with reduced formation of oxide inter- ferences (94/974). In a detailed study of the analysis of alu- minium oxide powder by mixing with high-purity copper powder it was shown (94/607) that sample pre-treatment and sample structure had an influence on the elemental sensitivities. The operational conditions for optimum analytical perform- ance could not be assumed from experience of metals analysis and metal SRMs could not be used to calibrate analysis of non-conducting powders. Two preparation methods have been investigated (93/3596) for the analysis of solutions. In the first the analyte was deposited from solution (0.2 ml) by solvent evaporation onto a pure silver-pin electrode made from compressed silver powder.In the second the sample solution was slurried with the silver powder dried at 100°C and compressed to form the electrode. The analyte signal was stable to better than 5% for 30-45 min internal and external reproducibilities were within 5% and the RSD was <15% for 17 measured elements in two SRMs. GDMS is being increasingly used for the analysis of soils and other complex matrices. In an initial application to the analysis of soils (94/606) 51 elements were determined in a range of SRMs using ion-beam ratios and user-generated RSFs. In the absence of isobaric interferences accuracies in the range 0.6-73% were observed. Problems were noted in achieving reproducibility and homogeneity in sample prep- aration.At present the method is suitable as a screening method for priority metals. A full survey analysis from Li to U in coal fly ash was possible (93/2933) with accuracies of 20-30% and sub-pg 8-l detection limits. The analysis of bituminous coals was only marginally successful however and the analysis of sub-bituminous coals failed owing to the high levels of volatile components. Iron meteorites are good conduc- tors of electricity so direct analysis was possible (94/567) of major and trace constituents at the sub-ng g-l level. Reports of isotopic analysis by GDMS included the direct measurement of uranium isotope ratios in soils (93/3427). Precisions of about 5.7% were achieved in an analysis time of 10 min. Lead isotope ratios in lead-rich iron meteorites were measured (94/267) with precisions of about 4% and were found to be close to primordial. The authors also observed an excess of 53Cr in one meteorite which would be consistent with production by cosmic rays.1.4. Inductively Coupled Plasma Mass Spectrometry 1.4.1. Reviews The last year saw an interesting range of review articles. In addition to extensive reviews of the capabilities and appli- cations of ICP-MS (93/3639) and plasma source MS (PSMS) (93/2950) useful accounts of interferences spectroscopic and non-spectroscopic (93/3401) and of introduction methodology (93/3713) have been published. Ross (94/1472) reviewed the use of ICP-MS for the determination of long-lived radio- nuclides and outlined possible future uses for the technique in this area.Dulski (93/3453) included ICP-MS as a com- parative technique in a review of analytical techniques used for residual-element determination of elements in high- performance alloys. (ICP-MS) 1.4.2. Fundamental studies Instrument development and operating parameters. Takaku et al. (94/576) reported on a lateral application of a single collector high resolution ICP-MS instrument. In order to overcome interference problems from GdO GdH and GdOH on Tb Yb Tm and Lu in the determination of trace impurity 1.4.2.1.JOURNAL OF ANALYTICAL ATOMIC SPECTROMETRY OCTOBER 1994 VOL. 9 271 R REEs in high purity Y,03 and Gd203 doubly charged ions were deliberately propagated as it was believed that although M2+ ions formed for monatomic species they did not form in detectable levels for interfering species. Detection limits for the analytes in question were improved to < 3 ng I-' and the concentrations of trace REEs were found to be 'variable between sample'. Although an interesting ICP-MS application it highlights the limitations of a 'high resolution' system for some real world analysis.Walder et al. have continued to extol the virtues of their double focusing multiple collector instrument. In one appli- cation (93/3402) isotope ratios of Hf and Hf-Lu mixtures Nd and Nd-Sm mixtures and Pb were determined in a series of RMs. For Pb ratios the precision at 2s was about 0.05% about 2-4 times better than can be achieved on a quadrupole system. For Nd and Hf ratios precisions of around 0.007% were achieved equivalent to the TIMS benchmark but with the additional benefit of only 100 s analysis times.Between 200 and 400ng of analyte was required for analysis. In a second application a high efficiency nebulization system was used for sample introduction allowing high accuracy and high precision measurements to be made on 50 ng ml-' samples of Hf Pb and U. Transmission calculations indicated an ion atom ratio of 1 5 500 for each of the three elements. This was considered superior to other plasma source mass spec- trometers and in the case of Hf exceeded that achieved with TIMS. Garcia Alonso et al. (94/574) reported on the performance characteristics of a glove box ICP-MS for the analysis of nuclear material. A flange had to be included between the skimmer cone and the ion lens assembly which increased the separation by 15 mm and slightly decreased analyte sensitivity.Optimum conditions in the modified instrument were evaluated for plasma operation signal stability and isotopic ratios together with levels of interfering species for selected fission products and actinides. Non-spectroscopic matrix effects arising from up to 500ppm of uranium or 2000ppm of Pu on fission products at masses 80-160 m/z were studied. Montaser et al. (94/1285) investigated an He ICP-MS system for the detection of metals and non-metals in particular those that suffer from spectral interference in Ar ICP-MS. The following four approaches to reducing the secondary discharge exhibited by the system were examined ( i ) the use of a centre- tapped coil (ii) modified load coils (iii) electrostatic shields and (iv) application of sampling cones made from non- conducting materials and aluminium.Plasma torch parameters were also optimized. The LODs obtained with an analogue detector and a prototype spectrometer for the metals studied were in the range 0.2-9 ng ml-' whilst those for As Br I and Se were 0.2 1 0.04 and 0.2 ng ml-' respectively. Vandecasteele and co-workers (94/563) used a 'zone model' as an explanation for signal behaviour and non-spectral inter- ferences in ICP-MS. According to the model for every nuclide there is a zone in the central channel of the ICP where a maximum density of singly charged ions occurs. The position of such a zone of maximum Mf density is a function of mass number of the nuclide and the zone can undergo a spatial displacement under the influence of an alteration of an instru- mental parameter or the introduction of a different matrix. Although the zone model did not completely reflect the genuine physical reality in all its facets it did provide a phenomenologi- cal model for the variation of ion signals with mass operating conditions and matrix composition.This group also reported on the use of internal standards in ICP-MS (93/3087). Careful study of the matrix effect showed that the magnitude of the signal suppression or enhancement depended in a regular way on the mass number Hence accurate correction for non- spectral interferences and improvement of precision is only possible using an internal standard with a mass number close to that of the analyte element(s).In both cases the ionization energy of the internal standard appeared to be of little signifi- cance. The authors advised that for multi-element analysis several internal standards should be used. Williams et al. (94/609) used colour-pho tographic evidence and computerized signal processing to evaluate fundamental limiting noise sources in ICP-MS. These were estimated to be 50 Hz from the electrical mains peristaltic pump rotation and plasma audio-frequency peaks (and their harmonics) and a l/f noise factor arising from the sample introduction system. A bonnet sheathing device could be used to reduce the 1Fnoise and peristaltic-pump noise and eliminate the plasma audio frequency peak. Houk and co-workers (93/3594 93/3595) described an ICP-MS instrument with an enlarged sampling or$ce and ofset ion lens system.The instrument had four pumping chambers and the 1.31 mm sampling orifice both improved ion signals and resisted plugging from deposited solids. The lens system without photon stop deflected ions off centre and then back on centre into the differential pumping orifice. The performance of a channel electron multiplier (CEM) was compared to that of a Daly detector. Both yielded similar sensitivities of about lo6 counts s-' per ppm with detection limits of around 1 ppt. Background counts were 0.4 counts s-' for the CEM and 4 counts s-l for the Daly a significant improvement compared with earlier use with ICP-MS. The instrument was reported to yield very low levels of many troublesome polyatomic ions but metal oxide species remained at about 1% typical of most ICP-MS systems.Grounding of the first lens electrode reduced the severity of analyte signal suppression in the presence of matrix elements to about 20%. It was also found that these matrix effects could be mitigated by adjusting the potential of the first lens in the presence of the matrix. 1.4.2.2. Sample introduction. Reports on studies of sample introduction methods for ICP-MS continue to be the most prolific. Amongst these laser ablation chromatographic tech- niques flow injection and electrothermal vaporization are still the greater proportion. In a recent review of the significance to analytical spectrometry of laser radiation interaction with solid materials Darke and Tyson (94/280) made the observation that much of the reported work on LA-ICP-MS is of little significance either contained in conference proceedings or in instrument manufacturers' literature a view long held by this author.This review year has seen a gratifying increase in the number of refereed reports of LA-ICP-MS however many are concerned merely with reporting analytical figures of merit and others accuracies of >20%. Moenke (93/3713) has com- piled a useful 122 reference review of LA-ICP spectrometry. Some highly original fundamental studies have been reported. Walder et al. (93/3451) determined the isotopic composition of Pb within NIST SRM 610 Glass using LA in conjunction with an ICP coupled to a double focusing magnetic sector muss analyser equipped with seven Faraday detectors.The laser was operated in Q-switched mode with an energy output of about 2 mJ. Craters were approximately 40 pm in diameter and 60-80 pm deep. The total amount of ablated Pb per crater was approximately 9 x lo-" g. Precision as good as 0.088% (2s) was achieved for the 207Pb 'O'Pb isotope ratio of 0.9096 which was in good agreement with TIMS results. This technique shows enormous potential for the direct determination of high precision isotope ratios in solid samples and will have many applications in geological and nuclear sciences. Mermet et al. (94/2212) compared IR and UVlasers for solid sampling ICP Spectrometry. It was found that direct LA is the major process responsible for the removal of material in the case of a UV laser as opposed to with an IR laser where shielding of the laser radiation by the absorbing plasma limits direct LA and increases the temperature of the (LA) plasma.It was concluded that the UV laser is superior in all analytical respects including quantification reproducibility and matrix effects.272 R JOURNAL OF ANALYTICAL ATOMIC SPECTROMETRY OCTOBER 1994 VOL. 9 Prabhu et al. (93/3225) reported on the development of a technique to analyse small volumes of solution using laser vaporization. The ablation cell consisted of a PTFE boat containing 200pl of sample solution and a rotating graphite wheel partially immersed in the solution. A film of solution adhered to the wheel passed through the laser beam and the resulting vapour was swept into the ICP by a stream of Ar.This technique was felt to be particularly useful for overcoming the poor efficiency of nebulizer systems and be attractive in the analysis of solutions containing radioactive elements. Accuracy was assessed using NIST SRM 1606 Stainless Steel and was considered good for the elements studied. Precision was <8%. It is not clear from this work what advantages this technique has over more traditional solution vaporization techniques such as ETV. Anderson et al. (93/3388) described a quick and simple preparation method for tantalum and niobium oxide powders prior to analysis by LA-ICP-MS. Samples were ground mixed with an aqueous solution of polyvinyl alcohol dried at 50°C and pressed into a pellet. Limits of detection for trace elements were < 1 pg 8-l. Problems such as memory effects and poor reproducibility were discussed. A good proportion of reports describe the application of LA-ICP-MS to geochemistry. Jarvis and Williams (93/3403 94/1364) described quantitative multi-element determination in silicate rock reference materials prepared as pressed powder pellets.Accuracy was better than 5% for major minor and trace elements (including REEs) and precision was <lo% RSD. They concluded that the quantitative analysis in this matrix was possible providing that standards and samples were closely matched both in terms of bulk chemistry and mineralogical composition. Jackson et al. (94/1357) reported similar work for the analysis of minerals but using a Q-switched Nd YAG laser with power attenuation and steer- ing optics to guide the laser light through the phototube of a petrographic microscope. This was focused onto a petrographic section contained in a sample cell.Minerals with a high absorptivity could be ablated controllably in thin section. In mineral with low absorptivity catastrophic ablation could occur caused by absorption at interfaces beneath the mineral surface particularly at the mineral-glass slide interface in standard thin sections. Good accuracy and precision was obtained for a diverse suite of elements. An SRM Silicate Glass was used for calibration but major elements were used as internal standards to correct for differences in ablation yield drift and matrix effects. Routine LODs were about 0.5 ppm. Perkins et al. (93/3 102) also analysed some silicate rocks prepared both as glasses and as pressed powder pellets.Limits of detection at or close to chondritic levels were obtained calibration lines were linear over four orders of magnitude and accuracy was ‘evaluated using reference materials’. Fedorowich et al. (94/1363) determined major minor and trace elements in six silicate rock RMs. These were prepared as glasses using a tungsten strip heater cell under an argon atmosphere. Standard reference material NIST SRM 612 Glass was used for calibration and A1 selected as internal standard. Of the 43 elements determined 25 were accurate to within 20% of the recommended values with precision of between 5 and 15% RSD. Two groups have investigated the use of LA-ICP-MS for Pb geochronology ofzircons. Ludden et al. (94/1387) determined 207Pb 206Pb ages in single zircon grains using a Nd YAG laser for ablating leaving a cylindrical pit of 30-60 pm.The precision of the Pb ratio was generally 0.5-6%. This was considered adequate to obtain reliable Pb ages from high quality limpid zircons with a simple Pb-loss history. In comparison with the sensitive high resolution ion microprobe (SHRIMP) technique LA-ICP-MS had the disadvantage of partially destroying the sample and inability to obtain accurate and precise U:Pb ratios and poorer spatial resolution. However sensitivity was better population precision was the same and purchasing and operating costs were much lower. Fryer et al. (94/2227) illustrated the application of LA-ICP-MS to the direct dating of single pitchblende and zircon grains.The major advantages of the technique were highlighted. Analysis of small sample areas (20-30 pm) was rapid and precise and facilitated studies on the timing of primary and secondary U mineralization. Chenery and Cook (94/292) described a preliminary study of the determination of REEs in single mineral grains by high spatial resolution L A microprobe ICP-MS. The REEs were determined to sub-ppm levels using a 40 pm crater and quanti- fication at the 100 ppm level using a 4 pm crater was demon- strated. A dual gas flow sample introduction system was used to permit calibration of the ICP-MS using aqueous standards. Repeatability of about 10% was achieved with the use of an internal standard when ablating a grain of clinopyroxene. REE distributions were determined in a monazite crystal using a 5 pm crater size.Limits of detection were considered superior to those obtained using an electron probe microanalysis technique. It is ironic that although in general good accuracy is required in the analysis osf environmental samples where LA-ICP-MS is used accuracy of ~ 2 0 % is considered accept- able. Denoyer (94/738) used the semi-quantitative procedure with LA-ICP-MS to determine trace elements in environmental particulates to an accuracy of 20-40% using internal stan- dardization and ‘within’ 2-3 orders without internal stan- dardization. Fuge et al. (94/1336) studied the minor and trace metal contents of the hard parts of modern shelled organisms. Spatial resolutions of <25 pm were achieved with sub-ppm sensitivities.This allowed seasonal variation of Mg to be observed together with irregular variations in Cu Pb and Zn which were believed may reflect major pollution incidents. Levels of metals do not appear to have been quantified. The requirement for both qualitative and quantitative infor- mation on the forms of metals in samples has sustained the level of interest of previous years in chromatographic techniques coupled to ICP-MS although the majority of development lies with the chromatography. A number of useful reviews are available this year. Hill et al. (93/3219) in a 127 reference review considered the use of GC and LC coupled to both ICP-MS and ICP-AES for environmental. industrial and gen- eral analytical samples. Caruso and co-workers (94/592 94/1276) reviewed interfacing criteria for LC GC and SFC including instrumental modifications and overall performance together with a summary of current LODs and illustrative method comparison.Indeed this group has made several new contributions to the field. They investigated the applicability of using time resolved acquisition (TRA) software with SFC-ICP-MS (94/273) for quantitative multi-element determination of organ- omercury and organolead compounds and the results were compared with those obtained using single-ion monitoring. Improved or equivalent LODs were obtained for organolead (0.7-100 pg) and organornercury (5-10 pg) using TRA. Calibration lines were rectilinear for up to three orders. Using an MIP with SFC (93/2070) halogenated hydrocarbons could be determined at trace levels.Calibration graphs were rectilin- ear for three orders and the LOD for C1 in l-chloronaphthalene was reported as 15 pg and 0.75 pg for Br in l-bromo-2- methylnaphthalene. This group also reported on interfacing a reduced-pressure water-cooled He-MIP (93/3226) and a water-cooled low pressure Ar ICP (94/562) to GC for plasma mass spectrometry. The former was used for the determination P S and halogens in both prepared mixtures and a pesticide providing sub-ng LODs for all the elements studied. The Ar ICP was used in conjunction with GC to measure an analytical signal for a sample of l-bromononane. The background spectrum con- tained many of the polyatomic species seen for atmosphericJOURNAL OF ANALYTICAL ATOMIC SPECTROMETRY OCTOBER 1994 VOL. 9 273 R plasmas and the authors concluded that there were small leaks in the vacuum seals.Larsen et al. (93/3224) separated eight As compounds in human urine by anion and cation HPLC-ICP-MS using anti- monite as a chromatographic internal standard. The ArCl interference was eliminated by chromatographic separation of the chloride present in the samples. The recovery of all As species in urine was close to 100%. The chromatographic peaks were evaluated by their peak heights and calibration was carried out by the method of standard additions. The calibration graphs were linear for all species and the LODs were 3-6 ng cm- for the cations and 7-10 ng cm-3 for the anions. Peters and Beauchemin (93/2898) reported on the charac- terization of an interface which allowed the coupling of a GC with an ICP-MS as well as conventional solution analysis with a nebulizer spray chamber system.The application of heat to the interface between spray chamber and plasma torch resulted in improved performance and LODs compared well with those obtained with a conventional nebuliser spray chamber. The use of the interface with GC was characterized for the determi- nation of two C1 and two Sn containing compounds. Elemental LODs for C1 were < 15 ng and < 12 pg for Sn. Houk and co-workers investigated the use of the direct injection nebulization system in conjunction with anion exchange and size exclusion chromatography (94/1287) and microbore column (MC)-LC (93/1943). The former techniques were used to determine Se compounds in human serum with LODs of 0.5-3 pg of Se.These absolute LODs were considered to be 1-2 orders of magnitude better than those obtained with conventional nebulizers. The MC-LC technique was used to separate Hg and Pb compounds with LODs of 7 and 0.2 pg respectively. Four Hg compounds and three Pb compounds were separated and determined in urine following a nine-fold dilution of the samples. There have been some interesting developments in the use of ETVwith ICP-MS during the last review year. Several studies have reported on the use of the technique as a diagnostics tool rather than just an analytical tool. Byrne et al. (93/323 1 ) used ETV-ICP-MS to investigate the interference by NaCl on the atomization of Mn in ETAAS. Use of JCP-MS allowed direct observation of the signals of Mn along with matrix components during the pyrolysis and atomization steps of ETAAS.The loss of Mn was reported to be due to vapour- phase interference caused by the formation of manganese chloride during atomization. The addition of ascorbic acid as a chemical modifier was found to remove the NaCl interference. Al-Maawali and Chakrabarti (93,4642) reported a five-fold enhancement in peak area sensitivity for Ag Sn and Pb in ETV-ICP-MS when samples were combined with a mixed chemical modifier Pd(N03)2 + Mg( NO,),. A smaller enhance- ment effect was observed when Pd modifier was used in the chloride form. Analytical precision was significantly improved with the use of the mixed chemical modifier. As the addition of the modifier did not enhance atom formation in ETAAS the authors concluded that the phenomena were largely attributable to a change in the mass transport efficiency of the sample from the ETV to the ICP. Richner and Wunderli (94/278) reported on the use of ETV sample introduction to distinguish between organic and inorganic chlorine.Using an optimized temperature programme includ- ing slow heating from 200 to 300"C some resolution of dichloro- tetrachloro- and nonachloro-biphenyls was achieved. Organic and inorganic C1 could be separated by using ramped heating from 400 to 2650°C in 1 s. Richner (94/614) assessed the trade off between LODs and matrix effects when using FI-ICP-MS for the analysis of high purity nickel samples. Using FI solutions of 3% m/m nickel could be analysed resulting in LODs in theng g-' range of Ni.However analyte signal suppression occurred and varied between 5% for Rb and 80% for B. This was overcome by use of an automated standard additions procedure because it was found impossible to compensate for the different suppres- sion effects using internal standards. Stroh et al. (93/3389) also illustrated the advantage of FI in improving LODs as well as the short and long term stability. In a 3% m/v sodium chloride matrix LODs were found to be 2-5 times better than those for continuous flow sample introduction of the same matrix. Accuracy was demonstrated with AGV-1 geological RM pre- pared by lithium tetraborate fusion and analysed at > 1.2% m/v dissolved salts. Recoveries of between 89 and 119% with 1 and 3% NaCl solutions were considered acceptable.Beauchemin (94/1296) reported on a preliminary characteriz- ation of ICP-MS with FI into a gaseous carrier. Air was used in place of the normal aqueous carrier for several analytes. For most analytes the background increased but the corre- sponding enhancement in sensitivity was such that the corre- sponding LODs were in general significantly improved particularly for elements with high ionization potentials. The only exception was Ni with a factor of 30 degradation in LOD and this was attributed to the increased removal of Ni from the interface components by the no longer water cooled ICP. For all elements a significant memory effect was observed which could only be eliminated by flushing with water between injections. A method of matrix elimination was developed that facili- tated the interference-free determination of Cd Cu Mn Pb I/ and Z n in biological matrices (94/577).Analytes were chelated onto an iminodiacetate-based resin in a microcolumn with simultaneous matrix removal. The analytes were then eluted into the ICP-MS with 3 moll-' nitric acid using FI. Close to 100% recoveries were achieved for Cd Cu Mn and Pb 90% for V and 85% for Zn. Chromium determination proved unsuccessful. The authors state however that the analysis of RMs yielded results in 'reasonable agreement' with certified values. For a 175 pl sample loop LODs were in the range 0.6-9.9 ng m1-l. A similar technique was applied to the deter- mination of brines (94/2182). Analysis of RM NASS-3 yielded results in 'excellent agreement' with certified values.McLaren et al. (94/290) modified a commercial ion chroma- tography system to allow the simultaneous determination of Cd Co Cu Fe Mn Ni Pb and Z n in 5 ml of sea-water samples in <15 min by TCP-MS. The results obtained on standard MetPac CC-1 columns were compared with those using silica- immobilized 8-hydroxyquinoline (I-8-HOQ) columns. The LODs in sea-water ranged from 1.6 ng 1-1 for Pb to 55 ng 1-' for Ni on the MetPac columns and from 0.3 ng 1-I for Cd to 47 ng 1-1 for Fe on the I-8-HOQ columns. Accurate analysis of Coastal and Open Ocean Sea-water RMs were achieved with at least one of the column types for all elements except Co and Fe. Stroh and Vollkopf (93/3404) combined FI with HG gener- ation for the determination of As Hg and Sb in four water RMs.Flow injection parameters such as sample volume purge gas flow rate and concentration of reductant were optimized together with sample pre-reduction techniques. Results obtained agreed with reference values and LODs were in the range 0.5-7 ng 1-'. Hydride generation-ICP-MS has been used for the determi- nation of Se stable isotope tracer enrichment in standard solutions and biological materials (93/2917). Three enriched isotopes 76Se 77Se and s2Se were determined while '*Se served as the reference isotope. Tracer enrichment was linearly meas- ured with or without simultaneous additions of two other Se stable isotopes. Limits of detection for Se were sufficient to detect up to a 200-fold dilution of tracer following an initial tracer enrichment equivalent to 10% of the mass of the natural Se pool.McLaren et al. (94/2196) used HG for the determi- nation of As and Se in sea-water and marine biological tissues with LODs of 0.3 pg ml-' for As and 2.5 pg ml-' for Se. The authors highlighted the importance of careful cleaning of the HG apparatus the central tube of the ICP torch and the274 R JOURNAL OF ANALYTICAL ATOMIC SPECTROMETRY OCTOBER 1994 VOL. 9 skimmer cone in order to avoid degradation of LODs by memory effects. Caruso et al. (93/2957) used a polypropylene gas-liquid separator for HG to completely eliminate the chlor- ide interference in the separation and determination of As species by HPLC-ICP-MS. An LOD for As"' of 0.46 ppb was achieved with the system. The effectiveness of the method was demonstrated by the determination of As in NIST SRM Urine. Haraldsson et al.(93/3387) used the HG batch technique to determine As Sb and Se simultaneously in natural waters. The volatilized hydrides were purged out of the sample and trans- ported directly to the plasma by the nebulizer gas flow. In this method APDC replaced I- as the reducing agent for Sb because I- interfered in the HG of Se. The LODs for a 10 ml sample were 1.5 ngml-' for As 2.4ngml-' for Sb and 6.3 ng ml-' for Se. Although the analytical precision was reported as >6% no indication of accuracy was given. Houk et al. (94/1297) used a Doppler particle analyser to measure aerosol droplets and velocities from a direct injection nebulizer (DIN) with radial and axial spatial resolution. The droplets on the central axis of the aerosol became finer and their size more uniform when 20% methanol was added to the usual aqueous solvent which was coincident with signal maxima when this composition was used for ICP-MS. Mean droplet velocities were 12-22m s-'.The outer fringes of the aerosol plume tended to be enriched in large droplets. The velocity of the droplets varied substantially with axial position in the plume. Powell et al. (93/2163) used a DIN to introduce drinking waters into an ICP-MS for the determination of Hg. The LOD was comparable to that obtained with a conven- tional spray chamber while the inherent problems of memory effects were drastically reduced. Accuracy was assessed with RMs and agreed well with certified values in two out of three instances.Houk et al. (94/128) also measured As Ni and V in Sea- water and Urine RMs by ICP-MS with cryogenic desoluation. Addition of 2% H2 to the aerosol gas flow enhanced analyte signals by a factor of 2-3 which compensated for the loss of analyte signal that accompanied the use of cryogenic desolv- ation. The interfering polyatomic ions ClO+ CaO' and ArC1' were attenuated to 'manageable' levels by the desolvation. The samples were diluted with 2% nitric acid so that the chloride could be removed as HCl. Analytical results were in good agreement with certified values and the detection limits were at best 4 10 and 30 ng 1-l for As V and Ni respectively. The samples were introduced by FI to minimize clogging of the orifice. Rattray et al. (94/2192) reported a new approach to sample preconcentration.The method involved deposition of the sample solution as an aerosol into an inductively heated graphite direct sample insertion probe. For a deposition time of 2 min and a sample volume of 0.5 ml LODs of 0.10 0.05 0.07 and 0.12 ppb were obtained for Cd Cu Pb and Zn respectively which represented an improvement of almost 3 orders of magnitude over those obtained on the same instru- men t by conventional nebulization. Longerich (94/554) described an easily assembled electronic controller that allowed automatic selection of the 'high-speed purge' mode commonly used on some commercial peristaltic pumps when used in conjunction with an autosampler. The controller placed the peristaltic pump in high-speed purge mode when the 'sipper' was at the autosampler wash station and could maintain this condition for a pre-set period after the sipper had moved to a sample position.Sample and wash solution uptake times could be reduced by more than 1 min using the system. Slurry nebulization methods are still in use for certain applications. Totland et a!. (93/3097) determined PGEs and Au in geological materials. Samples were ground to < 5 pm using zirconia beads in polypropylene bottles. An acidified dispersing agent (tetrasodium pyrophosphate and aqua regia) ensured that grains were maintained in suspension during grinding and led to a partial digestion of the material. Samples were diluted to < 2000 pg ml-' prior to analysis. Instrument operating conditions were optimized for slurry introduction by increasing forward power and carrier gas flow rate com- bined with a 3 mm torch injector tube.Calibration standards and blanks were made up in acidified dispersing agent. Limits of quantification were 0.04-0.2 pg g-' and results from the analysis of PGE in RMs were in good agreement with reference values for all elements in all samples. 1.4.2.3. Interference studies. Tanner considered both space charge and the implications of ion kinetic energy for the source of diatomic oxide ions. The first paper described an iterative procedure for the calculation of space charge in ion optical systems with cylindrical symmetry (93/3364). Ion beam param- eters which are specific to ICP-MS were discussed with relevance to their impact on ion-trajectory calculations. Most important among these was the mass-dependent kinetic ener- gies of the ions resulting from the supersonic expansion of the beam through the interface.The calculations qualitatively account for the mass dependence of concomitant matrix element effects. In a second paper ion kinetic energies were measured (94/608) for a suite of atomic ions and diatomic oxide ions as a function of nebulizer flow for an ICP-MS instrument. The results were discussed in terms of the source of the oxide ions. The author argued that the dominant source of the observed oxide ions was not within the interface and that the results of the study were consistent with a model that I envisaged the plasma as a turbulent reactor containing pockets of cooler gas in which oxide ion formation and persistence were promoted. The results were also consistent with a model that associated the oxide ions with the cooler region in the vicinity of vaporizing droplets and particles. Van Heuzen and Nibbering (94/619) described the use of stable isotopic labelling for the ussignment of the elemental composition interference ions by comparison of spectra of labelled compounds with those of non-labelled ones.To this end mass spectra of H,160 D2160 H2l80 and also H14N0 ' and H"NO diluted in the labelled water were compared. A number of unreported polyatomic ions were identified. The technique allowed origin of nitrogen entrainment to be estab- lished. Additional observations led to the conclusion that the water injected into the ICP was not completely dissociated and that ionization of the undissociated water molecules may occur uia charge transfer processes.There still exists some interest in the application of methane addition to the nebulizer gas for the removal of polyatomic integerences (93/3385). The effect of nebulizer gas flow rate power and percentage methane on a range of species in the plasma was studied. Analyte response was not 'seriously' affected by the addition of methane but polyatomic ions oxide ions and background levels were reduced. Recovery tests for As Se and V demonstrated the utility of removing the interfer- 1 ing species in comparison with typical operating conditions. Several RMs high in chloride were 'successfully analysed for As Se and V'. Goossens et al. (94/1294) used a similar approach to the elimination of interferences on As and Se in biological sample analysis.Instead of methane addition to the nebulizer gas flow a 4% ethanol addition was made to all ' analytical solutions and nebulizer gas flow was adjusted. The method was applied to the determination of As and Se in human serum and urine and results were considered to be in excellent agreement with certified values. Longerich (94/564) reported on digerences in mass spectra on 0.2 rnol I-' HC1 and 0.2 mol I-' HC104. An enhancement was observed of the ion signal in the mass spectra of several oxychlorine species in a solution of HC10 compared with the spectra of an equimolar solution of HCl. Spectra of Ca in HC104 showed enhanced formation of CaClO' CaClO,' CaClO ' and CaClO + compared with equimolar solutions of Ca in HCl.Similar effects were seen with Mg in HClO 'JOURNAL OF ANALYTICAL ATOMIC SPECTROMETRY OCTOBER 1994 VOL. 9 275 R compared with Mg in HCI. Other geological major elements (Al Fe K Mn Na Sc and Ti) did not produce the same effects. Little attempt was made at an explanation of these phenomena. Kershisnik et al. (93/2931) described an improved method for identifying the ArCl interference in the determination of As in urine. The authors observed that the signal intensities for 35C1160 and 40Ar35C1 were proportional in water and urine over a broad range of chlorine concentrations and that the proportionality was stable within an analysis sequence. The increased sensitivity (50 fold) for detecting the 40Ar35C1 inter- ference provided improved accuracy for urine arsenic quantifi- cation which was demonstrated by a split sample comparison with ETAAS.1.4.3. Applications Most reported applications of ICP-MS have little novel con- tent which is a reflection of how established the technique has become. The reader is directed to the Atomic Spectrometry Updates concerned with analysis for a broad review of ICP-MS applications. This review will deal with those pieces of work with notable TCP-MS developments. 1.4.3.1. Biological. Vanhoe et al. (94/2213) described a method for the determination of 11 ultra-trace elements (B Ba Bi Cd Cs Hg Li Mo Pb Sb and Sn) in human serum. Sample preparation was a minimal 5-fold dilution with 0.14 mol 1-' HNO,. Matrix effects and signal instability were corrected using Be In and T1 internal standards.With optimiz- ation of the ion lenses and nebulizer gas flow LODs of between 0.007 ng ml-' (Bi) and 0.5 ng ml-' (B) were obtained taking into account the original 5-fold dilution. Memory effects were negligible with blank levels <0.1 ng ml-' for all elements except Pb and B. Results were provided for a 'second gener- ation' Freeze Dried Human Serum RM and NIST SRM Human Serum. Serum samples from healthy individuals were analysed to determine typical element concentrations for normal human serum. Templeton et a!. (94/565) described a method for the measurement of Ni in human serum and urine. Interferences from Ca- Na- and K- based species were cor- rected using principal components analysis with Rh internal standard. The method was accurate to a level of 5 ng ml-' as determined with available RMs.Within run precision was 6% RSD at the 1.5 ng ml-' level and the pattern of interference was stable across an analytical run of at least 15 samples. Nickel was directly determined in diluted urine following quantitative precipitation of Ca2+ with oxalic acid. The method was accurate to 40 ng ml-' with a within run precision of 7% RSD at 4.5 ng ml-' of Ni. The method indicated the feasibility of dual isotope tracer studies for human biokinetic investigations. Friel et al. (94/1352) determined isotope ratios in human tissues enriched with Zn stable isotope tracers. Enriched faecal and urine samples were prepared using ion-exchange column chromatography. Both ICP-MS and FAB-MS were used for analysis in order to compare the precision of the two tech- niques.A least squares fit of the ICP-MS results against the FAB-MS for 70Zn 64Zn gave a slope of 0.98 with a 0.7% RSD. The results for 68Zn:64Zn gave a slope of 0.82 and a 14% RSD. It was concluded that Poisson counting statistics were a major contribution to the total analytical error indicating the usefulness of this procedure for enrichment studies. Buckley and Ihnat (93/3544) described the determination of Cu Mo and Se by ID-ICP-MS in a contribution to the elemental characterization of 10 new biological RMs. Copper and Mo were quantified by measurement of 65Cu 63Cu and 97Mo looMo ratios following spiking with 65Cu and 97Mo and digestion with nitric acid. Selenium was separated as hydrogen selenide from the matrix using sodium borohydride after spiking with 82Se and acid digestion-dry ashing and quantified by measurement of the 82Se 78Se ratio.Results from this work were compared with those obtained by other techniques and shown to be without analytical bias. 1.4.3.2. Environmental. Sturgeon et al. (94/2194) described the application of ETV-ICP-MS to the direct determination of trace heavy metals in surface samples of Arctic snow. For the elements Cd Co Cr Cs Cu Mn Ni Pb T1 and V the LODs were in the range 29-3200 pg. Direct quantification against simple aqueous standards was feasible using NaCl as a carrier. The latter was obtained in a high purity form by dilution of a Sea-water RM NASS-3. Accuracy was assessed from the analysis of River Water RM SLRS-2 and by comparative analysis of the samples by ETAAS.Hall and Pelchat (94/2195) developed a method to analyse fresh waters for Pd and Pt by ICP-MS following preconcen- tration onto activated charcoal. Conventional nebulization and ETV sample introduction were compared. Using sample vol- umes of 1 1 and a 200-fold preconcentration factor LODs were in the range 0.3-0.8 ng ml-' for both elements with little difference between the two introduction methods. Precision was typically 10% RSD at the 5 ng 1-' level and 4-5% at 10-fold higher concentrations of Pd and Pt. Addition of HC1 (1% v/v) to samples was found to be necessary in order to maintain the analytes in solution at their natural (< 50 ng 1-') levels although stability after 20 d was questionable. Analyte levels in samples collected from Canadian mineralized sites were in general < 5 ng 1 - '.The preconcentration procedure could be extended to Au and Rh. Miyazaki and Reimer (94/566) described the determination of Pb isotope ratios and concentrations in open and coastal sea- water by a combination of chelating resin (Chelex-100) precon- centration and ICP-MS. Isotope ratio precision (RSD) was in the range 0.1% (207Pb 208Pb)-1.2% (206Pb 204Pb) and the LOD was 0.8 ng 1-'. Accuracy was established by comparison with sea-water RMs and the precision was considered adequate to be able to discuss the origin of Pb in sea-water. Vanhaecke et al. (94/1322) used ICP-MS to determine Mo in a sea-water candidate RM. The determination was hampered by both spectral and non-spectral (analyte suppression) inter- ferences.Ten-fold sample dilution and the use of carefully selected internal standards allowed accurate correction for suppression. Correction for spectral interference on Mo iso- topes from BrO+ and BrOH+ species was achieved with sufficient accuracy by matrix matching of the blank for Br or by application of a mathematical correction procedure using simultaneous equations. Results obtained after anion exchange to separate Br from Mo confirmed the results obtained using the correction methods. There was good agreement between the results obtained with these methods and those obtained by other techniques and laboratories. Colodner et al. (93/3442) developed methods using ID- ICP-MS to measure Ir Pt and Re in sediments Pt and Re in sea-water and Re in sediment pore waters and river waters.In each case a stable isotope spike was added to the sample before processing. Sediments were dissolved in Teflon vessels using a microwave digestion method. Anion exchange of the chloro-complexes of Ir and Pt and of the perrhenate ion was used to preconcentrate the elements and to separate them from concomitants that produce polyatomic-ion interferences. Samples of 300-600 p1 were introduced using FI. Recoveries of 90+ 10% were achieved for all elements. The method had LODs for Ir Pt and Re of 6 14 and 5 pg respectively. Smith (94/1283) described the application of ID-ICP-MS to the determination of Hg in sediment and water samples. Sediment samples were analysed after spiking with enriched 201Hg and a 1 h acid digestion.The procedural LOD was 2 ng g-' using a 0.5 g sample. Quantitative recovery was confirmed from the preparation of RM BEST-1. Because of the levels in water samples preconcentration of Hg onto gold traps was required with subsequent electrothermal heating and purging276 R JOURNAL OF ANALYTICAL ATOMIC SPECTROMETRY OCTOBER 1994 VOL. 9 of the traps with Ar directly into the ICP. The procedural LOD was 0.2 ng 1- for a 200 ml sample. The accuracy of the method was evaluated using a river water RM (ORMS-l) certified at 6.8 & 1.3 ng m1-l. Results from the method yielded a value of 6.3 f 0.4 ng ml - '. Brown et al. (93/3509) used enriched IIB and "B to trace B uptake in intact plants (soil and nutrient solution application) isolated protoplasts and leaf and stem tissues (foliar appli- cation). Results suggested that stable isotope labelling was an effective and inexpensive method of monitoring B movement in plants and soil.Enriched B isotopes were used effectively to monitor B uptake into isolated protoplasts and whole plants. With the application of loB at about 5-fold the natural abundance ng amounts of translocated labelled B could be measured as well as soil B movement. Since results obtained were concentration independent and only two adjacent iso- topes were under investigation variation of instrument and sample parameters showed little effect on results. In general the use of ICP-MS for B isotope ratio determination was considered sensitive reproducible and straightforward. 1.4.3.3. Geochemical. Isotope dilution ICP-MS was used to determine the carbonate-bound Ba in foraminifera and corals (93/3095).The trace element content of these biogenic carbon- ates only reflects sea-water concentrations if the portion of element bound into the Ca lattice site can be selectively determined. The authors developed a procedure to purify benthic foraminifera shells of extraneous sedimentary organic and diagenic Ba. The effectiveness of the technique was con- firmed using paired-sample studies and partial dissolution of benthic foraminifera. Purified shells were analysed for Ba by ID-ICP-MS and Ca by FAAS. The long term precision of the element ratio was between _+ 2 and _+ 3%. The procedure was also used to determine Ba in aragonite corals with routine precisions of better than *2%.Bailey et al. (93/3223) determined T h and U in basalts by E TV-ICP-MS after a multi-stepped HF-HNO digestion pro- cedure. To improve vaporization efficiency trifluoromethane (Freon 23) was added to the carrier gas flow. Results from the analysis of RMs agreed well with certified values with LODs < 1 ng ml-'. The complexation of U with EDTA prior to analysis improved detection limits to 0.5 fg ml-'. Reiner (93/3375) investigated the eficiency with which the elements Nb Mo Ta and W could be extracted from silicate rock solutions (in HCl) as a function of fluoride activity (from HF). At pF 3.5 N-benzoyl-N-phenylhydroxylamine (BPHA) dissolved in CHC1 proved to be an ideal system for the separation of the four elements from the rock matrix after HF dissolution as all major rock constituents except Ti were left in the aqueous phase.Results from the analysis of 20 RMs 'documented' the practical value of the described procedure although accuracy figures were not quoted in the abstract. Using the above method LODs were obtained that were sufficiently low to allow results to be obtained for Th and U in some RMs for which there were no previous values. 1.4.3.4. Industrial. Taddia and Poluzzi (94/587) compared ETAAS with ICP-MS for the determination of Sn in doped indium phosphide slices. For ICP-MS samples were dissolved in HC1 and calibration was carried out with matrix matched standards after some investigation into matrix effects. Results with means that did not differ significantly (P=0.05) were achieved. The limit of quantification (6s) for Sn by ETAAS was 5ppm and 0.5ppm by ICP-MS and the RSDs for 35-149 ppm Sn were 3.4-2.0% (ETAAS) and 12.9-4.3% Fucsko et al.(94/1408) developed a highly sensitive analyt- ical method termed vapour-phase decomposition flow injection ICP-MS to measure the concentration of trace metals on silicon wafer surfaces. The method used HF to decompose and release metal contaminants from a surface oxide. A single drop (ICP-MS). of HF was placed on the wafer and carefully rolled over the surface to extract the trace metals. The drop was then trans- ferred to a small tube and injected into the ICP. Recoveries of 90-110% were achieved for Al Cr Fe and Na at surface concentrations in the order of lo1' atom cm-2. Recovery for Cu was <8.3%. The technique had the potential to measure 60 elements with LODs of 10' to 10" atom cm-2.1.5. Laser Ionization Mass Spectrometry (LIMS) There are a number of areas of overlap between the techniques which employ lasers in atomic mass spectrometry. The label LIMS will be used when lasers are used both to ablate material from a solid sample and generate atomic ions in the same step. If the material is laser ablated thermally evaporated or sputtered with an ion beam and subsequently resonantly excited and ionized using tuneable lasers then these various techniques are labelled as branches of resonance ionization spectroscopy (RIS). If material is sputtered with an ion beam and subsequently ionized non-resonantly with a laser then the technique is considered to be a type of sputtered neutral mass spectrometry (SNMS) sometimes known rather confusingly as surface analysis by laser ionization (SALI).This year a technique has been described which involves laser desorption followed by non-resonant laser ionization and the technique finds itself categorized under the heading LIMS (94/1315). A comprehensive review of LIMS has been provided in a recent book Laser Ionization Mass Analysis (eds. Vertes A. Gijbels R. and Adams F. Wiley 1993). Fundamentals of the ablation and ionization processes were discussed in the different laser irradiance regimes. The development of LIMS instrumen- tation was reviewed as well as applications of the technique to both chemical structure determination and elemental analy- sis. A much briefer review was provided in Application of Laser Spectroscopy in the chapter by Ledingham and Singhal (93/3472). Becker and Dietze (94/13@7) have described the use of LIMS for the analysis of silicon carbide zirconium dioxide and superconducting materials.They reported reproducibilities of between 10 and 20% and detection limits as low as 5 ng g-' using a pulsed Nd:YAG laser for ablation and ionization combined with a double focusing mass spectrometer. The paper provided a comparison of the detection limits spatial resolution and matrix compatibility of a variety of techniques used for surface analysis. The use of picosecond laser desorption pulses for solid sampling specifically applied to GaAs samples has been investi- gated by Trappe et al. (94/1315). This non-thermal desorption process avoids bulk melting and the concomitant problem of mixing and surface rippling effects. The authors reported a desorption yield of 5 x lo5 atoms per pulse from an area on the surface of 150mm2 which corresponded to lo- of a monolayer.The desorbed atomic plume was non-resonantly ionized by a delayed portion of the ablation laser pulse. A useful yield of 3 x was achieved. This technique is closely related to the SNMS methods particularly SALI. 1.6. Resonance Ionization Mass Spectrometry The proceedings of the biannual RIS conference provide a major outlet for publications in the field. No RIS conference proceedings were published during this review period and there are therefore correspondingly fewer publications cited in this update. A number of general reviews of the subject area have been published.Ledingham and Singhal (93/3472) reviewed RIMS in a chapter of Applied Laser Spectroscopy (eds. Vertes A. Gijbels R. and Adams F. Wiley 1993). The review concen- trated on semiconductor applications and the authors' own 'Resonant Laser Ablation' technique. Grey-Morgan and Telle (93/2973) have presented a brief overview of RIMS including some preliminary results from their new SIRIS instrument.JOURNAL OF ANALYTICAL ATOMIC SPECTROMETRY OCTOBER 1994 VOL. 9 277 R The RIMS technique has also been compared with resonance- enhanced ionization spectroscopy (REIS) and LEAFS using a variety of atom sources (94/687). Geo-chronologicical appli- cations have been reviewed particularly the measurement of 230Th 232Th isotope ratios (94/1489).Arlinghaus et al. (94/1418) reviewed applications of sputter-initiated RIS (SIRIS) to depth profiling in semiconductors in the concen- tration range i013-i020 atoms crnp3. Published RIS excitation schemes are being compiled and presented in a uniform format suitable for general use by Saloman (94/633) who has now published his fourth set of RIS/RIMS excitation schemes which covers Ag Be In K Li Rb Ti and V with an updated scheme for Ni. Knowledge of ionization schemes is critical if RIMS is to see widespread use. Relevant data are also provided by allied techniques such as resonance enhanced ionization (REI) in flames optogalvanic spectroscopy (OGS) in hollow cathode lamps and laser-induced fluorescence. Axner et al. (93/2878) have for example provided a comprehensive survey of atomization systems and resonant excitation schemes for Cr.The paper contained valuable data but did not strictly fall into the category of RIMS. Documentation of auto-ionizing states and high Rydberg states is also important to the development of RIS as these states are often critical to the ionization step. Krustev et al. (94/2191) report the measurement of Lu in geological samples at the lo-'% level using a two-step ionization route involving an auto-ionizing state. New ionization routes for B have been reported (94/1415) which involve states in an np-Rydberg series. The theory and practicalities of Hg ionization by both (4+ 1) and (2+ 1) ionization schemes and the use of an auto- ionizing state have been discussed (94/919) and auto-ionizing states of Ru have been found which allow its determination at the 0.05 pg g-' level (93/3788).Instrument developments included a continuous-wave res- onant excitation and field-ionization system (94/1002) which overcame the duty-cycle problem associated with instruments which use pulsed lasers. If used with semiconductor lasers this system could dramatically reduce the cost of RIMS measure- ments bringing the technique into more widespread use. However this goal awaits the development of commercially available diode lasers with higher powers and greater wave- length coverage. This continuous RIS system has been applied to only a few elements and in this publication its application to Li isotope ratio measurement was discussed. The authors reported a detection limit of approximately 20 fg for 7Li and an isotope selectivity better than 4 x loi3 for 6Li 7Li.A different approach to overcome the duty cycle problem was the use of a thermally-pulsed filament atom source (93/3686). This instru- ment used a 200 pg Mo filament which was evaporated at a rate of 80 pg in lo4 pulses. The system as described was not suited to a trace analytical r6le although the filament could presumably be loaded with a low-concentration sample. Pulsing filaments in an attempt to synchronize gas-phase atom production with laser pulses is not a new idea and has not been particularly successful owing to the high thermal mass of such filaments. A new quadrupole-SIMS instrument with res- onant post-ionization has been constructed (93/3476) which was capable of determining Cr in stainless steel with a detection limits of 0.18 pg g-'.A singly resonant (1+1) ionization scheme was used for these measurements but the spatial resolution corresponding to this detection limit was not reported. The RIMS method promises to be able to determine isotope ratios without isobaric interferences. However there are a number of practical spectroscopic limitations which can lead to significant isotopic fractionation. Wunderlich (93/3443) has discussed the wavelength dependence of even-mass isotope ratios of Ti indicating how a careful choice of resonance transitions and optimization of laser parameters can lead to accurate measurement of isotope ratios. Variations in measured Pb isotope ratios as a function of laser wavelength have been studied (94/1479) the authors described the relationship between the measurement error and the ratio of isotope shift to laser bandwidth.Developments in the continuous-wave resonance ionization of Th have been reported (94/626). An overall detection efficiency of 0.41% and the capability to measure 230Th 232Th isotope ratios on 25 ng volcanic samples was reported. This represents an order of magnitude reduction in sample requirement compared with previous TIMS measurements. The development of SIRIS-type RIMS instruments has been largely driven by the need to analyse semiconductor materials with high sensitivity and a high degree of spatial resolution. Depth profiles for Co and Ni implants in silica and silicon have been measured (93/3717).The authors reported inter- ferences due to molecule formation when low energy (< 4 keV) Xe primary ions were used. The interferences were suppressed by the use of Ar primary ions. Measurement of impurity Na in n-type silicon has been demonstrated at ng g-' levels using a laser microprobe and a three orders of magnitude enhance- ment was achieved with resonant post-ionization (93/2981). As the RIMS technique develops more diverse applications to complex matrices are reported including environmental geological and biological samples. Estler and Nogar (94/1006 94/1321) reported the measurement of Cr in urine at the 0.093 pg ml-' level showing good agreement with certified values. Using a 50Cr calibration spike the urine was wet-ashed and loaded onto tantalum ribbon filaments for measurement by RIMS.Anderson et al. (94/1393) have also reported the use of RIMS for the determination of Cr in urine and have applied the technique to the measurement of Cu in single cells. The determination of 237Np at trace levels has been demon- strated and a series of both two- and three-resonant step ionization schemes were compared (94/1338). The preconcen- tration and separation chemistry for tap-water samples was outlined leading to electrodeposits containing 237Np on rhe- nium filaments for RIMS analysis. The group have measured the filament loading of Np by a-spectrometry and from these data calculated a detection efficiency of 3 x lop8 which corre- sponded to 160fg of 237Np. The group did not address the problem of internal standards for overall yield calculations.This is a problem for 237Np as all other Np isotopes are short-lived. An interesting development was the use of RIMS in conjunc- tion with electrophoretic ion focusing to allow speciation of Np and Pu (93/3781). Different oxidation states of Np and Pu were separated electrophoretically and RIMS allows more sensitive detection of the separated species than do radiometric methods. This hyphenated RIMS technique is similar in concept to the approach used for DNA sequencing using Sn isotopic labels (J. Vac. Sci. Techno!. 1991 A9 1312). 1.7. Secondary Ion Mass Spectrometry The SIMS literature is supplemented each year by the proceed- ings of the annual conference. This year's review includes papers presented at SIMS VIII (Second. Ion Mass Spectrom. SIMS 8 Proc.Int. Con$ 8th 1991 (Pub. 1992)) as well contributions from a variety of other journals. It is always valuable to compare the advantages and disad- vantages of a technique with those of other methods. In this vein Grasserbauer et al. (94/1318) have reviewed the state of the art in a number of methods for surface analysis. Developments in these techniques have been driven mainly by the miniaturization of electronic devices and the purity require- ments of advanced materials. The authors discussed which techniques were best suited to the solution of the various problems and discussed artefxts such as ripple formation in SIMS which they ascribed to thermal recrystallization rather than the direct effect of surface bombardment. Gijbels (93/3662) has reviewed sputtering techniques in general with an emphasis on the instrumentation used.A comprehensive review of278 R JOURNAL OF ANALYTICAL ATOMIC SPECTROMETRY OCTOBER 1994 VOL. 9 sputtering theory by Greenlief and White (93/3663) discussed depth profiling and applications to organic and biological samples. Legget and Vickermann (94/1335) have reviewed static SIMS for the measurement of organic species and adsorbates. Despite its emphasis on molecular measurements the review is of general interest because of its discussion of secondary-ion formation mechanisms. The SIMS technique has been reviewed in the context of competitive and comp- lementary techniques for surface characterization including AES AFS elastic recoil detection analysis (ERDA) PIXE Rutherford back scattering (RBS) RIMS and X-ray methods.Williams et al. (94/1412) have discussed instrumentation for high lateral resolution imaging as well as a number of appli- cations. The mechanism of formation of multiply charged ions in SIMS by a SNMS-type process has been investigated (93/3720). Ionization was found to occur in the gas phase due to collisions of sputtered neutrals with primary ions or more importantly with secondary electrons. The authors examined the formation of M2+ and M3’ ions on 0- bombardment. Herion (93/3376) in a short communication challenged the logical processes of induction based on experimental ‘facts’. In this philosophical paper the validity of experimentally-derived ‘facts’ was challenged using SIMS as a vehicle for the argument.A number of new instruments have been described although none appear to depart very significantly from existing designs. A Poschenreider TOF-SIMS instrument suitable for the analy- sis of highly resistive samples ( z 1013 R cm) has been described (93/3782). Each primary ion pulse produced lo4 ions and there was found to be no need for charge compensation under these conditions. A SIMS instrument incorporating a photodiode array detector has been constructed which was able to detect between 10 and lo5 ions per pixel with good linearity. Relative ion sensitivities were found to vary by less than a factor of two for elements from H to Pb and imaging applications of the device were discussed (93/3579). Magnetic sector instru- ments generally have higher ion throughputs than quadrupole instruments. This was demonstrated in an instrument where the quadrupole mass analyser was replaced by a magnetic sector leading to a 100-fold improvement in sensitivity (93/3732).The SIMS ion signals are not representative of local concen- trations unless sputter equilibrium has been reached. Thus bulk standards cannot be used with confidence for the charac- terisation of ultrathin layers. This problem has been addressed by the use of spin-coated thin film standards of Al Cr Na Ti and Zn on silica surfaces (93/3755). Ion-implanted materials are frequently used as matrix matched standards. However the matrix in an ion-implanted standard may differ from that in a uniformly doped matrix due to the effects of ion-beam damage during the implantation process.For As B and P in a silicon substrate it has been found that within experimental error ion implants can act as valid standards for quantita- tive work with other implants crystalline or recrystallized materials as long as peak concentrations do not exceed 6 x lo1’ cmP2 (93/3696). In another study useful ion yields from bulk-doped mercury cadmium telluride were compared with ion implanted standards (93/2554). Quantitative measurements by SIMS ideally require matrix- matched standards. However a knowledge of relative sensitivity factors ( RSFs) can allow semi-quantitative measurements to be made. The RSFs are a relative measure of the ‘useful yield’ of the analyte ions from the matrix of interest when bombarded by a specific ion beam. Useful yields for a number of elements in semiconductor metallic and biological matrices using Ar’ Csf O,’ Ga+ and Xe+ primary ion beams have been reported (94/1517).The authors concluded that the ‘reactive’ primary ions Csf and 0,’ give the highest ion yields for electronegative and electropositive species respectively. Lower ion yields were reported for Ga and the inert gases although these differences in yield were less pronounced for oxidized matrices or biological matrices. Measurements of RSFs have been reported (94/1419) for a large number of elements in TiSi TIN and TiW which are important matrices in semicon- ductor fabrication. It was noted that depth resolution was adversely affected by surface roughening of the matrix during ion bombardment. Measurements of RSFs for a range of elements in GaAs InP and Si (93/3704) were made over the course of five years and found to be stable to within +50% suggesting that the RSFs should be ‘portable’ within these limits.The RSFs for an extensive set of analytes ranging from H to U have been determined for a number of metal matrices (93/3749). Values were reported for both Cs+ and 0,’ primary ions and limits of detection reported for positive negative atomic and molecular secondary ions. The RSFs for Al B Ba Ca Fe K Li Mg 0 Rb Si Sr U and UO in glasses using a 25 keV 69Ga+ primary ion beam have been reported (93/3695). Data for Ga primary ion beams are scarce but important because of the high spatial resolution possible with such instruments. The authors noted that electrical charging of insulating samples can alter RSFs for elements which differ in their secondary ion energy distribution.The results from a round-robin study involving 20 partici- pants allowed the determination of the variability of measure- ments of peak depth and concentration at that depth for a number of ion-implanted samples (93/3744). The value of a good vacuum for background reduction was emphasized by the results. McRae et al. (93/3094) have compared the sample isolation and conventional energy filtering modes of discriminating against molecular interferences. Ion yields were measured for REEs in fused amphiboles and fused clinopyroxenes. They concluded that the two techniques were equally good for quantitative measurements giving precisions and accuracies of the order of 10Y0 for the silicate minerals studied.Energy filtering has also been used to reduce matrix effects in the determination of B Be and Li in silicon (94/1279). The authors reported precisions and accuracies of z +20% at concen- trations in the pg 8-l range. One approach to the removal of molecular interferences in REE determinations involved the monitoring of the doubly charged ions of odd-mass isotopes thus yielding half-in tegral mass-to-charge ratios (93/3441). The group reported measure- ments of Dy Er Eu Gd La Nd Pr Sm Tb Tm and Yb at concentrations below 15ng g-’ with <20mm depth reso- lution. The method clearly worked for this application but was limited by the low yield of doubly charged ions and the need to have an odd-mass isotope of the analyte with no isobaric interference.Canteri et al. (93/3697) have implanted F and 0 into boron- doped silicon to examine the enhancement efSects induced by these ‘reactive’ elements. They reported that the enhancement was greater for B than for Si signals and that F caused a greater enhancement than 0. A SIMS study of Pt/Si structures revealed profile distortions due to the presence of oxygen (94/1406). The effect was shown to vary with the oxygen pressure. The effect of surface H on secondary ionization yield has also been investigated (94/1370) with an examination of the nature of the H-surface bonds and the influence of hydrogen on surface work-function. Depth resolution in all techniques based on ion sputtering is limited both by substrate mixing and the development of surface topography caused by the bombardment of the primary ion beam itself. The problem of depth resolution in the measurement of ultra-thin layers has been reviewed by Barlow (93/3651).Friedbacher et al. (94/1301) have examined AlGaAs/GaAs structures by SIMS and during depth profiling measurements have examined the surface topography by atomic force microscopy. The authors described the evolution of both cone and ripple structures. Ion-beam induced swelling of GaSb has been shown to reduce depth resolution and to affect elemental ion yields (94/1452). A comparison of ion- beam and plasma sputtering processes. (94/1492) showed thatJOURNAL OF ANALYTICAL ATOMIC SPECTROMETRY OCTOBER 1994 VOL. 9 279 R needle formation at the surface was associated with the ion- beam technique only.Studies of ion-beam mixing in thermally deposited lo7Ag-lo9Ag layers have indicated that mechanisms other than ballistic mixing were at play (93/3741). Ion-beam mixing examined as a function of impact energy angle of incidence and primary ion type (93/3742) caused an exponential decay in signal at boundaries that could hide real structures. Measurements of Al Ca and Na profiles in thin oxide layers have shown that sample charging should be avoided by the high mobility of Na in silica which also can lead to profile distortions (94/1313). The use of sample rotation during analysis has been successful in the reduction of ion-beam induced topography. The method has been reviewed by Cirlin (93/3765) in the context of both AES and SIMS depth profiling. It was noted that the technique reduces the detrimental effect of inhomogeneity of low energy ion beams.The effects of sample rotation on depth resolution in the analysis of Ni/Cr multilayers has been investigated in a round-robin study (94/1495). Surface topography efSects can also be reduced by careful choice of the primary ion and its energy. Likonen et al. (93/3599,94/45 1) have investigated ion-beam mixing at various energies for Ar and Xe primary ions in Au/Pt multilayers. Depth resolution was found to be inversely proportional to the square root of primary ion energy and the experimental results agreed with the theoretical model used. Oxygen can enhance sensitivity and depth resolution in some circumstances but its influence varies depending upon the analyte the matrix and whether it is used as the primary ion or as a residual gas.Houlton and Blackmore (93/3718) reported the surface roughening effects of oxygen-ion bombard- ment on AlGaAs which reduced the attainable depth reso- lution. Depth resolution was also found to worsen at higher A1 concentrations. Prudon (93/3475) has attempted to optimize depth resolution in multilayer systems by oxygen doping. The work has examined a Si1602 and Si1802 bilayer and Fe-Ti multilayers. Oxygen doping was found to limit ion-beam induced roughness and help to yield the correct composition profile. A depth resolution of 5.5 nm was reported using a 2.5 keV primary beam. A more complex method for achieving high depth resolution involves ion-beam etching in SIMS instruments to produce 'mesa' structures prior to depth profi- ling analysis (93/2992).The authors reported that the main disadvantage of the technique was a memory effect caused by the redeposition of material sputtered from the surrounding region. This memory effect is a more general problem which has been modelled theoretically and tested against experimental measurements on B in silicon samples (94/1493). Numerical simulations have been used in an attempt to 'reclaim' lost depth resolution. For example the broadening of Al Be and Si &doped layers in GaAs under ion bombardment has been simulated (94/1505). Allen et al. (94/1497) have used a maximum entropy deconvolution approach which allowed accurate depth profiling of ion-implanted standards. However a critical study of deconvolution procedures which attempts to tackle the problem of ion-beam mixing during the depth profile analysis of &doped structures (94/1346) concluded that improvements in depth resolution were only likely to accrue from instrumental developments or techniques such as the use of reactive gases and not from deconvolution procedures.One approach to the reduction of matrix effects in SIMS is to monitor molecular secondary ions such as MCs'. This technique has received a great deal of attention this year and is discussed both in this section of the Update and in section 1.9 where MCs' SIMS has been compared with SNMS. The method is used both to reduce matrix effects and for the measurement of elements which give low secondary ion yields.Thomas and Torabi (93/3653) have compared X f SIMS with CsX' SIMS for a Ge/Pd/GaAs system and concluded that the CsX' was 'free' from matrix effects. Although this may be something of an overstatement the group were able to suppress spurious enhancement effects at interfaces. An extreme signal- enhancement effect was observed in which analyte signals at interfaces were higher than in the pure analyte phase. The authors proposed a mechanism for CsX' ion formation in which emerging X+ ions reacted with surface Cs whilst the surface was still accessible to act as a collisional third body. Gnaser and Oeschner (93/3724) have reviewed the field of CsM+ SIMS and cite a number of applications. Wittmack (94/1519) has discussed the use of MCs' SIMS for the measurement of low-yield elements such as Zn and the inert gases.Deposition of Cs on the surface was found to alter the surface work function and influence ion yield so it was rec- ommended that surface Cs concentration should be kept at a constant concentration to avoid a further 'matrix effect'. A yield depression effect related to surface Cs concentration was also reported which could be avoided by the use of primary impact angles greater than 60". The CsHe' ion has been used to measure He concentrations in an A1-Mg alloy (94/1354). Improved quantitative measurement was possible at the expense of sensitivity. Owing to its high ionization potential a very poor secondary ion yield would be expected for He yet a detection limit of 1OOpg 8-l was reported. The same approach has been used to measure 0 and N concentrations in a Nb-l%Zr matrix (93/3725).Depth profiling of C in tantalum films has been performed (94/1390) using the matrix- derived TaC- ion (the primary ion was Cs'). The authors reported a loss in depth resolution with increasing C concen- tration and ascribed this to matrix amorphization with increas- ing C-content. Non-linearities in response in the analysis of InP/InGasAs layers were overcome by a combination of oxygen flooding and the use of 02+ primary ions (94/1506). Oxygen pressures used were higher than that required to saturate the surface owing to the inertness and low sticking probability for O2 on these surfaces. Monitoring the CsM' ion did not give a signal which was representative of the true analyte distribution but no explanation was given.Chew and Sykes (94/1504) have used sample-derived molecu- lar secondary ions to distinguish between discrete B atoms (true solid solution) and B precipitates in silicon. Dis- crimination was based on cluster formation where cluster size was related to concentration by the equation [B,] = k[B]" for discrete B atoms implying that significant matrix effects were likely where local concentrations vary dramatically. There is a proliferation of examples of the applications of SIMS in the literature which will not be discussed in any detail in this review. The reader is referred to the proceedings of SIMS VIII for a taste of these applications. Only a few of the more novel applications are considered here. An important field is the study of corrosion for which a few papers are cited (93/2967 93/2968 93/2970 93/3474). SIMS spectra have been used for 'fingerprinting' minerals within geological samples (93/35 15).Oxygen isotope ratios have been measured using Cu- primary ions which were themselves generated by sputtering( 94/1484). Microvolume SIMS allows the determi- nation of elemental concentrations in solution. Using this method Chia et al. (93/3700) reported the determination of S with a detection limit of 4ng g-' by depositing 1-10 nl of solution on to a high purity silicon substrate. The solution contained 1 pg g-' Y as an internal standard and 100 pg g-' U as a matrix stabilizer. The same group (94/1541) reported the use of the technique for the determination of Ca Mg and Zn in urine bovine liver serum and faecal samples and obtained good agreement with certified values. The analysis of micro-particles by surface methods is not straightforward because of the surface topography and electrical charging effects.The subject has been reviewed by Xhoffer (93/3519) both for SIMS and LIMA. In the analysis of clay microspheres (93/3757) containing Pu and its decay products it proved possible to determine less than lo6 Pu atoms but problems were encountered measuring isotope ratios owing to isobaric interferences from 238U and 241Am. Measurements of the280 R JOURNAL OF ANALYTICAL ATOMIC SPECTROMETRY OCTOBER 1994 VOL. 9 composition of salt aerosol particles (93/2964) were found to require the use of a low energy Xe primary beam to reduce damage to the sample. 1.8.Spark Source Mass Spectrometry (SSMS) The very limited use of this technique was emphasized in this review year by the lack of reported developments. Only one paper in the English language has been considered worthy of inclusion and the remainder were in the Russian or Chinese languages and unlikely to achieve wide circulation. Only two papers concerned fundamental aspects or method development. Ramendik has discussed (94/1323) a method based on his quasi-equilibrium model for quantitative multi-element analy- sis. Aspects of this work have been reported in previous year's reviews. The same group of workers have optimized (93/2997) an improved method for isotope dilution analysis. Full con- sideration of this work was not possible because a translation was not available at the time of review.Applications of SSMS have emphasized its suitability for survey analysis of a wide range of materials and for almost all elements of the periodical table. Up to 70 impurity elements could be determined (94/1509) in high-purity arsenic with detection limits in the range 1-100 ng g-'. Other novel applications included the analysis of atmospheric dust using the ID technique for 24 trace elements with precisions in the range 10-25% (93/2939) and the multi-element analysis of paper products (94/1374). 1.9. Sputtered Neutral Mass Spectrometry (SNMS) The SNMS technique encompasses a variety of techniques in which ionization of sputtered neutrals is performed by electron beams a low-pressure gas discharge (also known as plasma or electron-gas SNMS) or by laser photo-ionization.The latter method encompasses both resonant ionization which is dis- cussed in section 1.6 and non-resonant ionization or SALI which is discussed in this section. Wucher (94/1302) has reviewed electron-gas- electron-beam- and laser-SNMS using quadrupole magnetic sector or TOF mass analysers. It was noted in the review that high useful yields (> 1% atomic) were required to achieve high lateral and depth resolutions and that the efficient laser post-ionization was the most promising approach for three-dimensional imaging applications. Laser SNMS was found to be best-suited to static surface analysis whilst electron gas postionization was more suited to ultratrace bulk analysis at concentrations down to long g-'.Hopfe et al. (94/1303) have analysed multilayer structures for X-ray diffraction by X-ray diffraction AES SNMS and TEM provid- ing a useful intercomparison of techniques. Bartella et al. (94/1306) reported the results of a round-robin study of aluminium and steel analyses by a group of 9 INA-3 electron-gas SNMS instruments. Results highlighted the impor- tance of careful mass calibration and the use of a good and constant mass resolution across the mass range. Variations in reported concentrations were within a factor of three and this was considered satisfactory. Instrumental developments included the use of a focused high current electron beam for post-ionization (93/3621). The authors reported a significant enhancement in ion yield and a small SIMS interference.Albers et nl. (94/1344) examined metal oxide multilayer optical coatings by SIMS and SNMS paying particular atten- tion to the interfacial signal enhancement effects. By examining surface chemistry by XPS the group reported that the enhance- ment was due to the chemical state of the analyte and to non- equilibrium sputtering at the interface rather than due to electrical charging effects. The SALI technique is similar in many ways to SIRIS and it is useful to compare the methods and the relative benefits of resonant and non-resonant photoionization. The advantage of SALI is that it allows simultaneous multi-element measure- ments but ion yields can vary and molecular ions can be created. Terhorst et al. (93/3716) used intense 500 fs UV laser pulses for photoionization of sputtered atoms and molecules and reported useful yields of up to 1%.The authors discussed the difficulties of trace analysis at high spatial resolution. An extensive study of the variation in ionization efficiency with laser power in a non-resonant photoionization SNMS (SALI) system using Cu and Si substrates has been performed (93/2946). Variables included laser intensity beam profile pulse duration and the delay of the laser pulse after sputtering. Two mechanisms were described for atomic ion formation the direct non-resonant ionization of sputtered neutral atoms and the quasi-resonant dissociative ionization of sputtered neutral clusters. The authors noted that the latter process was more efficient and discussed the conditions required to optimize detection by this mechanism.Hayashi et al. (93/3752) have reported a practical application of the method to measurement of metallic elements in silicon wafers. The authors used non- resonant ArF-exciplex-laser postionization. Irradiances of 5 x 10" W cm-' allowed the determination of analytes at the lo1' cm-' level with a useful yield of 1.3 x Sensitivity was limited by the available ion beam pulse widths and laser beam dimensions. There have been a number of comparisons of SIMS MCs+ SIMS and SNMS. With the exception of SALI SIMS generally provides the highest sensitivity but is prone to large matrix effects. Matrix effects reduced in MCs' SIMS using the same SIMS instrumentation but this approach can lead to mass spectral interferences. SNMS also leads to a reduction in matrix effects but requires significantly different instrumen- tation.SIMS has been compared with laser-postionization SNMS for the measurement of Ar concentrations in AlAs/GaAs multilayers (93/3623) and the authors reported that SNMS gives more reliable quantitative results. Gnaser (94/1343) has compared the use of MCs+ SIMS with SNMS for the analysis of amorphous semiconductor thin films. The MCs+-based measurements were shown to be more sensitive but there were problems with mass interferences. However both methods showed general agreement. The authors have contrasted the use of CsM + molecular-ion SIMS with electron-gas SNMS for depth profiling in Si,Ge -, H layers and multilayered photovoltaic devices. The two techniques show good agreement with the major differences being due to mass interferences in the molecular SIMS spectra.Pfeifer et al. (94/1311) analysed oxide films and ion-implanted sapphire standards. Measurements by the two techniques were found to agree within a factor of three. The authors determined RSFs for Al Cr and Fe for the oxide systems under investigation. Matrix effects were reported for both CsM+ SIMS and SNMS due to the presence of different solid phases within the sample. The dynamic range sensitivity and depth resolution of SNMS and MCs' SIMS have been compared in a review by More et al. (93/299 1 ). Some matrix effects are common to both SIMS and SNMS and result from the sputtering process itself. Grunenberg et ul. (94/1308) have examined the quantitative capabilities of elec- tron gas SNMS for surface analysis.For Zn surface coatings they reported cone formation which was ascribed to pre- existing or sputter-induced poorly-conducting areas on the surface. These areas cause surface charging under ion-beam irradiation which in turn causes a reduction in sputter rate. These effects have been overcome by the use of the high- frequency sputter mode. The difficult problem of microparticle analysis has also been tackled by SNMS. Goschnick et al. (94/1345) have examined airborne microparticles (0.3-0.8 mm diameter) by both SIMS and electron-gas SNMS. The group have measured elemental depth profiles through particles which show a surface enrichment in organic compounds. The authors reported the presence of a series of molecular ions derived from inorganic salt particles.Both SIMS and SNMS were used in a comp-JOURNAL OF ANALYTICAL ATOMIC SPECTROMETRY OCTOBER 1994 VOL. 9 281 R lementary fashion and allowed the authors to postulate the source and evolution of these particles. Schweiker (93/3624) has compared sputter yields for carbonate nitrate and sulfate salt particulate matrices and reported on chemical effects on the sensitivity of both SIMS and SNMS. 1.10. Stable Isotope Ratio Mass Spectrometry (SIRMS) This review year has been highlighted by the publication of some substantial reviews. In a review of the stable isotope analysis of small samples Pillinger discussed (93/2949) the reduction of sample requirements for gas source SIRMS to allow analysis in the picomol range the use of ion microprobes for isotope analysis the extraction of gases for measurement using focused laser beams and the potential of the GC- combustion-SIRMS procedure.Some interesting applications were included in the review which had 167 references. The review of Mulvaney (94/1438) although principally on nitro- gen analysis gave a comprehensive consideration of sample preparation instrumental methodology and recent advances in automation. Preston described (93/3667) the basic principles of the technique giving particular attention to sample size gas production and transfer to the MS. In addition the conven- tional dual inlet technique which gives highly precise but labour intensive analysis was compared with the relatively new continuous flow GC-combustion technique. A new design of instrument for stable isotope measurement was optimized (94/1397) for use with the static vacuum tech- nique and in particular with the continuous-fl ow combustion technique.Differential pumping was employed to cope with the high flow of He used in the latter technique and to achieve a good abundance sensitivity of +0.05%0 for 13C in 25 mol of CO and +0.2%0 for 15N in 3.7 mol of N,. Other instrumental developments have involved evaluations of relatively well-established techniques for new areas of study. The use of GC-combustion-SIRMS provided (93/2918) a simple and practical alternative to the conventional preparative GC-ninhydrin-SIRMS method for the determination of (13C)leucine in the measurement of muscle protein fractional synthetic rate. The new procedure avoided problems of labori- ous sample preparation and contamination errors and pro- vided more reproducible data.A quadrupole-based system has been evaluated (93/3593) for the measurement of 13C0, 12C02 in breath samples. An inlet system of a control valve connected to a microbore stainless-steel tube allowed manual adjustment of source pressure to 4 x Torr and analysis of a standard before each sample. Precision of ratio measurement in both standard and samples was &0.2%. The use (93/2960) of 3He MS for the determination of low concentrations of 3H (0.5-5 Bq 1 - l ) in environmental samples was novel. Further method development is required but the procedure could become preferred to the conventional decay-counting methods. A new procedure for the determination of C N 0 and S was based (93/3501) on the ID principle.Changes in 6 values produced by the mixing of samples with reference materials of known 6 values were used to calculate the value in the original sample which might be too small in volume for direct analysis. For meaningful analysis this approach requires accurate dis- pensing of sample and reference and complete mixing of the two. Control of blanks becomes critical with very small sample sizes. Pillinger and co-workers considered (93/2924) this in a review of static MS for carbon isotope determination. A procedure was described which limited the total blanks to 1 nmol in a stepped combustion from room temperature to 1200" and to < 50 pmol in the range 600-1200". It was also possible to measure the isotopic composition of the blanks using the static MS procedure and thereby to identify the source of the blanks and to make corrections for them.The same group have constructed (94/1366) an inlet system with the purpose of identifying the cause and magnitude of memory effects observed in the measurement of D H ratios in natural substances. They found that the major source of memory effects was the adsorption of water on to surfaces in the glass components of the inlet system. Identification of memory effects allowed corrections to be applied but it was also recognized that it would be preferable to eliminate the memory effects. Problems have also been identified (94/1365) in the outgassing of air from PTFE valves used in the determination of hydrogen isotopic composition.This led to apparent deu- terium depletion in samples. Attention has been given to sample preparation for nitrogen isotope analysis. In very similar applications Liu and Mulvaney described the use of diffusion discs used to collect ammonia produced in Kjeldahl digestion (93/2985) or ammonia diffused from soil extracts (93/2986) in automated nitrogen analysis using the Rittenberg technique. The discs containing 50-150 pg of nitrogen were transferred to disposable sample trays treated with concentrated HF (10~1) and dried in a desiccator containing anhydrous CaSO,. If corrections were made for ID by natural abundance nitrogen results using this procedure were in close agreement with results obtained with conventional steam distillation and concentration.Nitrate at ultratrace levels in ice cores has been prepared (93/2941) for isotope analysis using HPLC to preconcentrate the samples and the Dumas method to prepare molecular nitrogen for MS analysis. Overall analytical precision was If 0.4%0 for a sample size of 35 bar yl N,. Results were presented for deep ice-core samples from Greenland. Developments in sample preparation for other isotope analy- ses have included a method (94/1296) for the determination of D in brines and hypersaline solutions without the removal of alkaline earth metal cations. Only 8 pl of sample were needed for reduction with zinc metal (1.0-1.5 g) at 460 "C in a special glass container. Deuterium concentrations in brines were meas- ured with an accuracy of + 1%. A procedure for the determi- nation of 34S in petroleum and rock samples involved (93/3647) hydrogenation of the samples previously dried over CaC1 in a stream of hydrogen without the presence of a catalyst and passage of the gaseous products over a silica gel-NaX zeolite (1 2) adsorbent followed by a platinum catalyst for hydrogen- ation-hydrocracking The H2S produced was captured in aqueous cadmium or zinc acetate.The narrow range of the natural variations in chlorine stable isotopic composition required (94/2237) highly precise measure- ment. Any CO in chloride solutions produced if necessary by digestion of samples was removed by boiling and the chloride precipitated as AgC1. This step was repeated until all traces of S were removed. The AgCl was converted to methyl chloride by reaction with methyl iodide for isotope analysis of CH3Cl+ at masses 50 and 52.A peak hopping procedure was used on a modified instrument because the collector design did not allow exclusion of masses 49 and 51. Determination of Standard Mean Ocean Chloride over an extended period gave a precision of 0.009% (Is). The range of 637Cl was - 1.5 to 0.8%0 in groundwaters and -1.2 to 0.8%0 in brines. It was suggested that the method could be used to identify different sources of potable waters and to indicate leakage of brine into groundwaters. The observed range of 637Cl in nature was ten times narrower than results reported from TIMS analysis which places doubt on the reliability of the latter. Applications of SIRMS indicate its value in the study of biological geological and industrial processes.Collagen type I11 was found (93/3482) to have a higher turnover rate in growing rats than collagen type I after determination of '*02 in isolated collagen-derived hydroxyproline. Measurement of 15N in ammonium released by crustal rocks allowed (94/1386) pro- cesses to be proposed for the release of ammoniacal nitrogen from different rock types. The gasification of carbon composite materials (94/1360) and coal-oil coprocessing under thermal and severe hydrotreatment conditions (94/1469) have been studied by measurement of the I3C -. 12C values.282R JOURNAL OF ANALYTICAL ATOMIC SPECTROMETRY OCTOBER 1994 VOL. 9 The SlRMS method appears to be used increasingly for the identijication of adulterants in foodstuffs and related materials.Such investigations rely on different sources of materials having different stable isotope ratios. The quality control and verifi- cation of authenticity of honey (93/2909 93/2996 94/1487) is clearly important around the world and one such study in Germany (93/2909) of 13C:12C values in honey proteins showed that 79 of 160 samples had been adulterated with sugar. Natural differences in 13C "C values allowed (93/2954) detection of as little as 10% corn-derived acetic acid in apple cider vinegar. Adulteration of pure frozen concentrated orange juice with beet invert syrups can be identified (93/3591) by measurement of 6l8O values. The last two procedures have been adopted for action by AOAC International. 1.1 1. Thermal Ionization Mass Spectrometry (TIMS) The very large number of reports in this review year on developments and new applications of TIMS indicates the very healthy and active status of this technique. In both the determination of isotopic composition and the application of ID there have been considerable advances in analytical meth- odology and the range of elements studied.Heumann's review on ID methods (93/2951) covered not only TIMS but also other MS methods for example ICP-MS and GC-MS. It gives an excellent coverage of the potential of the technique in particular for the determination of ultra trace levels of elements and of the use of negative ion TIMS which is finding increased application in a number of areas. Instrumental developments tend to be for specific applications but the description of the multi-purpose (TIMS SIMS and RIMS) Isolab 54 instrument illustrated (93/3582) the perform- ance of a modern and highly-sophisticated design.A second electrostatic analyser placed after the multicollector (conven- tional Faraday cups plus an ion counting system) was equipped with an ion-counting Daly detector for high abundance sensi- tivity measurements over large dynamic ranges. Sources collec- tors and a- and energy slits were all selectable. New methods of sample preparation for strontium analysis have been investigated. The more promising appears to be that (93/2900) based on adsorption by a solution of crown ether in octanol sorbed on an inert substrate packed in small columns (5 cm x 5 mm). Samples were loaded on to the col- umns in 3ml of 3 mol lP3 HNO and washed with volumes of 5 and 3.5ml of 3 mol I-' HN03.The strontium was subsequently eluted with 4ml of 0.05 niol 1-l HNO,. Clean separation from interfering elements in particular calcium and rubidium was achieved with small reagent volumes. The method was seen to have advantages of high selectivities and almost total recoveries but memory effects were observed as a result of incomplete cleaning of the columns and as a conse- quence re-use of the packing materials was not advised. Publication in Chinese prevented evaluation of a so-called double ion-exchange procedure (94/1390) for the separation of strontium from brines. Conventional U-Pb age determination requires measurement of the U Pb ratio in samples using ID. This has been avoided by a procedure which measured (94/2225) only lead isotope ratios.Detection of the 210Pb isotope which is generally in radioactive equilibrium with 238U on a TIMS fitted with a high abundance sensitivity lens system allowed precise measurement of the 210Pb 206Pb ratio in zircons. Ratios in the range 1.3 x to 9 x lo-' corresponded to ages of 330 to 3365 Ma respectively and were in good agreement with conventional age determinations. It was suggested that the technique could be applied to single grain measurements but only for U concentrations > 1000 pg g-l. Highly sensitive and selective analysis is also required in the measurement of thorium isotope ratios. This has been achieved on both conventional single-focusing and double focusing instruments. The group at the Open University has developed (93/3496) techniques for the measurement of 230Th 232Th ratios (5 x in young volcanic rocks using a conventional and unmodified instrument.The better counting statistics of MS in comparison to a-spectrometry gave enhanced precision (&1.5% 2s over a two year period) reduced analysis time (1.5 h) and a lower sample requirement (300 ng). Improved performance was achieved (93/2925) on an instrument fitted with a 30 cm energy filter. Abundance sensitivities of 1 x and 5 x lo-' for the 237Th:238Th and 230Th:232Th ratios respectively were obtained with external precisions of Although the negative ionization TIMS method for Re-0s measurements has been developed for a few years its appli- cation has been restricted by problems of inhomogeneity in samples incomplete equilibration between sample and spike and high Re blanks in the platinum filaments.An ingenious method has been developed (94/2235) to avoid these problems in which both the Re:Os ratio and the osmium isotopic composition were achieved with a single measurement of osmium isotopes. Samples were irradiated with thermal neu- trons in a nuclear reactor to produce activated rhenium atoms which decayed to the stable isotopes 1860s and 1880s. The production of lS60s and 1S80s was proportional to the Re concentration which could therefore be determined from the measured osmium ratios. Good agreements were obtained in the dating of molybdenite samples but uncertainty in the half- life of 187Re limits accuracy of determination. Although an innovative technique nuclear reactors are not widely accessible and a waiting period of two months for decay of the activated Re is a limitation.The determination of very low levels (108 atoms per sample) of plutonium in uranium ores required (93/2979) particular care in the use of ultra-clean equipment and reagents and of clean room protocols. Use of this procedure to determine naturally-produced 239Pu in uranium-bearing rocks should make it possible to assess the geochemical behaviour of plutonium over long time periods. The 239Pu 242Pu ratio in a French isotopic standard mixture has been measured (93/2980) using both direct deposition on to a triple filament and the resin bead technique in order to study isotopic fractionation which was similar in both cases. A new procedure for the isotopic analysis of lithium required (94/1467) simpler sample separation and smaller sample size than other procedures.Lithium was measured as Li' ions emitted from lithium phosphate deposited on rhenium double filaments. A stable ion beam of 8 x 10-l' A for 7Li was obtained at a low filament temperature of 850°C at which isotopic fractionation was under much better control than at the higher temperatures necessary in other methods. An analyt- ical repeatability of 0.026% (Is) was achieved for the 7Li:6Li ratio. The group in Xining China have reported a method for the determination of bromine and chlorine isotopic compositions. Significant variance in bromine isotopic composition was observed (93/3586) in bromides of different origin. Bromine was prepared as CsBr and loaded on to filaments with graphite added as an ionization enhancer.Measurement of the Cs,Br+ ion gave a value of 1.02654+0.00012 (precision of 0.011% at 95% confidence limit) for the 79Br 81Br ratio in KBr chemical reagent. The same method of addition of graphite and measure- ment of the dicaesium halide ion was used (93/2945) for the determination of chloride isotopic composition in high-purity HC1 with a precision of 0.034%. The doubts raised in sec- tion 1.10. about TIMS determination of chloride isotopic composition should be noted. Interest in the determination of boron isotopic composition remains considerable but there appears to be little consensus about the best procedure other than the use of a molecular beam technique the accuracy of which has been put into doubt by the study of Datta et al.(94/1483). They monitored different pairs of Li2B02+ ions and found that although precisions +0.85% (IS).JOURNAL OF ANALYTICAL ATOMIC SPECTROMETRY OCTOBER 1994 VOL. 9 283 R were comparable the absolute calculated values of boron isotope ratios varied. This was attributed to measurement errors being enlarged to differing degrees in the computation stages. Guidelines were proposed for minimizing these errors. The use of Na2B02+ ions as in the determination of boron isotopes in boron carbide (93/2994) or Cs2B02+ ions reduces the isotopic complexity of the measured molecular ions. The group at Xining used (93/3099) the latter method to detect isotopic fractionation of boron between brines which are enriched in 1 lB and lake sediments.Single tantalum filaments were loaded with a slurry of graphite (about 100 pg) in ethanol CsOH to give a mole ratio of B Cs in the range 1 2 to 1 4 and sample solution (1-2 pg B). Ratios in the NIST SRM 951 were measured with a precision of 0.026% (2s). The significant differences in boron isotope ratios in fly ash leachates and in groundwaters have been used (93/2937) to identify contami- nation of groundwaters. The method is more sensitive than using boron concentrations alone but may be limited to simpler situations with a well-defined source of contamination. A fascinating application of TIMS was (94/1388) the measurement of fallout 135Cs in coastal sediments and the proposal to use 137Cs 135Cs ratios to estimate recent sedimen- tation and erosion rates.An instrument fitted with a retarding potential quadrupole lens filter was used to measure 135Cs 133Cs ratios of 1 x lo-' and 137Cs 135Cs ratios of 0.5. The 137Cs 135Cs isotope ratios are more powerful chron- ometers and tracers than 137Cs concentrations alone. The isotopic compositions of Ag Ru and Te in several types of uranite have been determined (94/2223) by TIMS using the silica gel-phosphoric acid procedure on V shaped rhenium filaments. The chemical separation of the three elements was based mainly on ion exchange both anion and cation pro- cedures. In a few samples clear isotopic anomalies resulting from spontaneous fission of uranium were observed for 99Ru IolRu lo2Ru lo4Ru 12*Te and 130Te but not for the silver isotopes. It was possible to show that one uraninite had reached a critical condition for neutron-induced fission of 235U.Measurement of the Ru isotope ratios is not straightforward and the precautions necessary for successful analysis have been described (93/3431). A mixture of silica gel-phosphoric acid- boric acid-alumina (1 1 1 0.1) was used as ionization enhancer the sample was kept moist at all times before loading into the instrument and the filament temperature was only raised very slowly. Results were presented for the natural isotopic composition of Ru. Problems of bias encountered in the isotope dilution deterrni- nation qf U in Rb,U(SO4)3 were identified (94/1328) as arising from the incomplete isotopic exchange between sample and spike solutions.Accuracies of better than 0.1 % were achieved using a perchloric acid procedure in the equilibration. Whereas there was (93/2892) no systematic difference in accuracy between ID-SIMS with a monocrystalline silicon wafer as substrate and ID-TIMS for the determination of U in soils the precision of TIMS was six-fold better than that of SIMS. In the same study the traditional AG1X8 and new U/TEVA.SPEC resins were compared for sample preparation. The latter generated cleaner fractions and was prefered for TIMS analysis but use of AGlX8 removed the need for hydride corrections in SIMS analysis. Papers in Chinese have presented methods for the determination of U in tea (93/3569) soil (93/3791) and irradiated fuels (93/3768). Attention is being paid to the ID determination of the rare earth elements.The determination of rare earth elements in high-purity rare earth oxides was achieved (94/1319) using isotopic spiking of the sample a modified HPLC procedure to separate the determinands from the matrix and TIMS analysis. Although the detection limits of 0.5 pg 8-l were higher than by MS analysis without HPLC separation (0.05 pg g-') the latter could only be used if isotope measure- ment was not affected by the matrix. The total evaporation technique originally developed for the determination of Pu and U has been applied (93/2953) to the determination of Gd Lu Nd and Sm using a specific software program written for real time data acquisition and processing. Sample requirement was 20-50 times less than with the conventional measurement method of stabilized currents.A new value of 0.72333 was proposed for the 146Nd 144Nd normalizing ratio. A new pro- cedure for the determination of Gd involved (94/1414) the loading of Gd as nitrate on to single rhenium filaments use of colloidal carbon as modifier and measurement of the Gd + ion. Methods have been presented for the ID determination of other elements. Plant samples were digested (93/2940) in a microwave oven and Mo separated by liguid-liquid extraction for TIMS analysis as the MOO,- ion. New procedures for the determination of Pu in a range of different matrixes involved (93/3780) the use of 239Pu and 240Pu as spikes in place of the commonly used but difficult to obtain 242Pu and 244Pu. Meticulous procedures involving clean chemical separation by anion exchange chromatography and precise isotope measure- ment by negative TIMS allowed (93/3101) Re concentrations to be measured in 200ml of sea-water with a precision of 0.5 YO.This was a 100-fold improvement over previous analyses of sea-water and it was possible to demonstrate that Re was relatively well-mixed throughout the water column. Although ID-TIMS is essentially limited to single element determination Heumann and co-workers have developed pro- cedures using a thermal ionization quadrupole MS for the accurate determination of trace contaminants in high-purity materials. In the determination of metals (Cd Cr Cu Fe Ni Pb Th and U) in high-purity titanium used in the microelec- tronics industry various separation methods and ionization procedures were optimized (94/1292) for each element to give detection limits in the ng 8-l range.Chlorine was determined (94/1317) in high-purity molybenum and tungsten by dissolu- tion in H202 precipitation as AgCl and measurement of negative ions produced on double filaments. The limits of detection influenced by variable blank values were in the range 0.006-0.1 pg 8-l. Trace impurities (Ca Cd Cr Cu Fe Ni Th and U) were determined (93/2891) in refractory metals niobium metal and silica by positive ion TIMS following selective chromatographic extraction and electrolytic pro- cedures for the separation of trace elements from matrix. These procedures allowed Th and U to be determined with precision at the pg 8-l level necessary for the characterization of microelectronic devices.The certification of reference materials is an activity of considerable importance. The use of ID-TIMS often con- sidered as a definitive method in establishing trace element concentrations has been reported for the development of CRMs (93/3650) the determination of Li in lyophilized serum (93/3448) and the determination of Cd in polyethylene (93/3447 93/3583). In addition the 143Nd 144Nd ratios have been determined (94,4381) for 12 Geological Survey of Japan rock CRMs. Such data are important so that quality control data do not rely on one or two SRMs. Developments continue to be made in the use of enriched stable isotope spikes in human nutrition studies in particular those requiring high precision and sensitivity. Turnlund et al. developed (94/2183) a high-precision procedure for the measurement of enriched 46Ca administered to pre-term infants in a study of calcium absorption and excretion in very low birth weight infants fed different formulas.The Ca was separ- ated from biological samples by precipitation as the oxalate which was dissolved in nitric acid for simultaneous measure- ment of the 44Ca 46Ca and 48Ca isotopes in 10 pg samples. Within-run and between-run precisions were d 0.14 and <0.09% respectively for urine and faecal samples. A two tracer study was proposed which would use enriched 42Ca and 46Ca and simultaneous measurement of the 42Ca 44Ca 46Ca and 48Ca isotopes. Molybenum isotopes were measured (94/1277) by the same group in 1 pg of Mo isolated from284 R JOURNAL OF ANALYTICAL ATOMIC SPECTROMETRY OCTOBER 1994 VOL.9 biological samples. Samples were loaded with boric acid and silica gel on to double filaments and five isotopes (94 96 97 98 100) were measured simultaneously. It has been estimated (94/1351) that the TIMS measurement of Ca and Mg tracers administered orally and simultaneously to adults is of sufficient precision and accuracy to be able to detect the tracers in the blood plasma for up to 600 h after ingestion. An improved MS procedure for the determination of iron isotopes in sub-g samples with precisions of 0.1% involved (94/1280) a low temperature (1200 "C) silica gel-boric acid technique on an improved source filament assembly composed of platinum and ceramic. The source region was re-designed to minimize iso- baric interferences and the use of clean working environment allowed analysis of 100-1000 ng filament loads.A review with many references presented (93/3587) compiled data on the isotopic compositions and atomic masses for all stable elements. Atomic masses reported in this review year have included those for Fe (93/3581) Ir (94/1396) for which Ir02- and Ir03- ions were produced from (NH4)JrC16 on double platinum filaments Li (93/3502) and Ti (93/3584). 1.12. Other Methods Although the ions produced in laser ionization are normally measured using a time-of-flight (TOF) MS the suitability of ion traps in such analysis has also been investigated (94/288). The use of an ion trap allows the collection and storage of ions generated from several laser pulses and makes them available for various alternative methods of detection.Analysis of ceramic metallic and polymeric materials identified some disadvantages but these were all associated with the method of ionization rather than the ion trap itself Optimum con- ditions of laser fluence and number of laser shots were sample dependent and sample surface preparation influenced the spec- tra produced. A limited dynamic range resulting from space charge effects and sample matrix interferences were also noted. It was suggested that a decoupling of ablation and ionization would be preferable but the potential advantages of the ion trap were seen to be real. A perfect Penning trap MS for the measurement of unstable isotopes should combine capture of external ions with high efficiency high resolving power and accuracy and general applicability to all elements and isotopes available from an on-line isotope separator.Such an instrument has been set up (93/3665) at the ISOLDE/CERN facility and used (93/3664) to resolve the ground and isomeric states of a nucleus (78Rbm,g and *'Rbm7g) for the first time. The effects of ion-ion interaction were investigated for two ion species differing in mass and stored simultaneously in the trap. A Penning trap MS with a resolution > 1 in lo9 was used (94/1463) to measure the atomic masses of 3H and 3He. The results provided support for a setting of new limits on the neutrino's rest mass. The potential of Fourier transform ion cyclotron resonance MS (FTICR-MS) has attracted a number of interesting investi- gations.The general principles of the technique have been covered (93/3635) in a wide-ranging review while the use of laser ablation FT-MS was covered (93/2938) in a review of laser ablation. The combination with laser ablation has very high mass and spatial resolutions and has been exploited in the analysis of integrated circuits with a spatial resolution of 5ym (94/1477) and in the measurement of long-lived radio- nuclides produced by fission or activation and for which high mass resolution is required to separate isobaric interferences (94/1446). This high mass resolution (> 290 000) has also been exploited (94/1286) in the coupling of a GD source and FTICR-M S. Resolution of isobaric atomic interferences resulted in detection limits in the high ng g-' to low pg g-' range in the analysis of stainless steel.The potential of FTICR-MS for the measurement of isotope ratios using either electron ionization (EI) or laser ionization has been investi- gated (94/1471) for Kr Pb Ti and Zr isotopes. Precisions of better than 0.4% were achievable for the EI analysis of Kr whereas only 9-12% could be achieved for trace elements using laser ionization. The number of ions in the analytical cell needed to be optimized to maintain a high SIN ratio but to avoid space charge and magnetron effects. Initial results have been reported (93/2965) for the isotopic analysis of Xe but errors were high and further development was required. A new method of isotope separation has been proposed (94/1420) using ion cyclotron resonance in a magnetic field with a radial component. In the new concept isotopes would be ionized in a magnetic field in a solenoid and confined axially in an electrostatic potential well.Computer simulations were carried out to demonstrate the feasibility of separating the isotopes through exploitation of oscillations in the axial component of the force on the isotopes. The use of gas-phase electron ionization MS is rare but has been demonstrated (94/1289) for the measurement of iron isotope ratios in the volatile iron-containing compound penta- kis(trifluorophosphane)iron Fe( PF,) . Precisions of 6 x lop4 and 5 x were obtained for abundance ratios of major and minor isotopes respectively. The GC-MS method for the measurement of enriched stable isotopes in clinical studies has been expanded for the determi- nation of additional elements.Aggarwal and co-workers are particularly active in this area and have reported methods for the determination of Cd (93/3597) and Pt (94/769) using the lithium bis( trifluoroethy1)dithiocarbamate chelate and of Ni (94/1325). The cadmium isotopes could be measured with a precision of 1-3% on 10 ng samples but memory effects were experienced between successive samples. Lead in sea-water was converted (94/1486) to tetraethyllead by boiling with sodium tetraethylborate for determination of lead isotopic composition in samples with concentrations as low as 0.1 ng ml-'. Analysis of NIST SRMs gave results within 1% of the certified values. Molybdenum isotope ratios were measured (93/3716) with a detection limit of 50 ng by analysis of the chelate with sodium bis( trifluoroethy1)dithiocarbonate.A method has also been reported (93/3605) for the determination of Se in clinical studies using an enriched stable isotope as internal standard. A new development holding promise was (94/1739) the use of capillary electrophoresis (CE)-MS with an ion spray-sheath flow interface coupling for the determination of inorganic cations and anions in synthetic solutions. Solutions containing 30-300 pg of up to 12 individual metals were separated by CE using an aqueous buffer (pH 4.8) containing 30 mmol I-' creatinine and 8 mmol 1- ' a-hydroxyisobutyric acid. Adducts formed between creatinine and cations required a relatively high (120 eV) declustering energy for dissociation and resulted in the production of singly charged positive ions in the mass spectra.The method was applied to the analysis of used engine oils and proved satisfactory for the detection of wear metals. The procedure did not prove satisfactory for the determination of anions because of the formation of complex interfering cluster ions.JOURNAL OF ANALYTICAL ATOMIC SPECTROMETRY OCTOBER 1994 VOL. 9 2. X-RAY FLUORESCENCE SPECTROMETRY 285 R Considerable advances in TXRF spectrometry and X-ray microfluorescence (XRMF) are again evident in this year's Update. There has been a reduction in the total number of XRF papers available for the Update which has combined with our consciously reduced coverage of Applications where we have limited citations to those in which there is some identifiable analytical innovation.Where appropriate additional Applications papers are mentioned only in passing and identify the sister Update in which coverage should be sought e.g. Industrial materials. All accessible Application papers will at least be included in the Tables in the relevant sister Update. By doing this it was possible to concentrate more on instrumentation and data processing aspects of XRF spectrometry without detracting from the ability of the Update to identify important trends throughout the XRF literature of the review period. 2.1. Reviews It is a rare pleasure to include a new book on XRF even more so when it is as good as the recently published Handbook of X-ray Spectrometry Methods and Technique (93/3 184). The editors (Van Grieken and Markowicz) have brought together high quality contributions from experts throughout the XRF community to produce an excellent book covering both modern instrumentation and practical methology.This is truly a handbook and if you are involved in XRF spectrometry or need a complete practical update in the subject you can do no better that to consult this volume. Torok and Van Grieken (94/1560) reviewed comprehensively the literature of 1991- 1992 and concentrated on instrumen- tation and data processing. Advances in beam techniques (SRXRF and PIXE specifically) and TXRF were identified as major areas of growth and the authors also considered in some detail developments in microanalysis using electron X-ray and charged particle beams. Wobrauschek (94/677) presented the current state of development in WDXRF EDXRF and TXRF techniques and emphasized the increases in sensitivity which had been obtained in recent years.An extensive Russian review of EDXRF was presented by Revenko (93/3885) who made particular reference to the benefits to be gained from monoch- romatization and polarization of the excitation radiation. In an authoritative and wide-ranging general review of XRF analysis Jenkins (93/3911) included a useful historical perspec- tive of the development of the technique. The principles of WD and EDXRF were presented by Kundra (94/2322) who also reviewed some of the more practical aspects of XRF such as problems arising from spurious diffraction peaks and the presentation of small amounts of sample as thin film specimens. Bruna (94/937) described the characterization and advantages of XRF in comparison to alternative spectrometric techniques.Recent applications of radioisotope-excited XRF at the IAEA in Austria were presented by Valkovic et al. (94/1566) who described the wide range of isotopes available (to the IAEA) and their useful characteristics for XRF. In particular the benefits of each source for portable and on-line instrumentation were stressed and examples given. In their review of XRF in Latin America Vazques et al. (94/2368) showed that more than two thirds of the XRF instruments were situated in Brazil and that the IAEA had been a substantial contributor to the availability of EDXRF instrumentation and knowledge in the region. Unfortunately the extensive reviews of the use of XRF spectrometry for geological analysis by Ma and Liang (94/201) Wang (93/3909) and Yi (94/204) were only available in Chinese.The unusual and often difficult analysis of transuranic elements particularly in the complex mixtures found in nuclear fuel re-processing plants was the subject of a review by Myasoedov and Lebedev (94/109). X-ray techniques including XRF spec- trometry were included in the comprehensive review of surface analysis and characterization by McGuire et al. (94,4275) and in the review of residual element (< 100 ppm) analysis of high performance alloys by Dulski (93/3453). Finally the prior review in the ASU Updates series on XRF can be consulted for publications in the year up to March 1993 (93/3250). 2.2. Instrumentation A historical perspective of how advances in analytical technol- ogy including WD and EDXRF have influenced the develop- ment of modern inorganic geochemistry was presented by Potts et al.(94/2374). Although recent developments in scientific instruments tend to greater sophistication sometimes there is elegance in sim- plicity. In this category Lavrinenko and Faingol'd (93/3873) presented calculations for predicting the performance of a simple two-layer filter designed for the non-diffractive XRF determination of S in hydrocarbons containing C1. The optimal S/Cl ratio on the filter surface was about 3. Turning to WD instrumentation developments Hoover (93/3888) Steinmeyer (93/3868) and Nikiforov (3/3907) claimed patent applications for multi-crystal X-ray monchroma- tors.In the case of Steinmeyer the monochromator was designed to diffract two separate wavelengths. Hoover's instru- ment comprised multiple pairs of multilayer dispersion elements designed so that the most appropriate may be selected for the diffraction of a particular wavelength. In Nikiforov's design the monochromator comprised dual dispersive crystals installed so that their edges were angled at 180"-2a (where a is the Bragg angle). A double crystal high resolution X-ray spectrometer was described by He et a/. (93/3878) using a design that offered high dispersion and unit-magnification one-dimensional spat- ial resolution in the direction perpendicular to the specimen direction. The instrument produced two spectra simultaneously disposed about a central wavelength and for this reason was especially useful for measuring spectrum line shifts.The instru- ment was used for measuring the X-ray spectra observed from laser-produced plasmas but may also be used for characteriz- ing chemical shift effects. Longerich (93/3804) described the application of a new commercially available treated LiF200 crystal to the XRF analysis of geological samples. This crystal was recommended for the determination of Ni to Nb (Ka lines) and elements having L lines in the same energy range offering up to a 2-fold increase in sensitivity compared with a standard crystal. For elements of lower atomic number than Ni the standard LiF( 200) crystal offered better performance. The performance of a commercial sequential XRF instrument was described by Kuiperes and Smit (94/2292) incorporating close-coupled optics and claiming to offer a significant increase in the analytical performance for all elements and with a capability of determining elements down to Be in the Periodic Table.The XRF instrumental applications of layered synthetic microstructure (LSM) devices was reviewed by Biltoft (94/1580) with particular interest in their use in modifying the spectral emission from a copper anode X-ray tube the fabrication of these devices and characterization techniques as they apply to EDXRF. A compact monochromator for hard X-rays and gamma rays (8-140 keV) based on LSM devices in a novel orientation was described by Hoover (93/2877). To operate in this energy range an LSM with a 2d spacing of 1.5 nm was required.However the multilayer was configured so that the incoming polychromatic beam subtended an angle of between 3" down to 10 arc minutes. This consideration placed constraints on the design of the multilayer in terms of length (12 cm) rigidity,286 R JOURNAL OF ANALYTICAL ATOMIC SPECTROMETRY OCTOBER 1994 VOL. 9 minimal coefficient of t hetmal expansion and smoothness for which a value of 0.2-0.5 A rms was achieved using advanced flow polishing methods. Sapphire or silicon carbide slabs were selected as substrate material and a metal foil filter was required at the input aperture to absorb radiation below 8 keV which could otherwise satisfy criteria for total external reflec- tion (rather than diffraction). A monochromatic excitation source was applied by Manninen and Karhumaa (94/2366) to the determination of the rare earth elements.Samples were excited with the Ka radiation from a tungsten anode tube operated at 150 kV 3 mA. The tube output was transmitted through a monochromator com- prising a symmetrically cut Ge(400) crystal bent into a cylindri- cal shape to focus the 59.318 keV W Ka line at the sample position. Excitation characteristics were claimed to be superior to those of an 241Am source because in the latter gamma emissions suffer inelastic scatter which leads to a low energy tail to the excitation peak. The instrument was evaluated in the analysis of standard solutions and yttrium wire containing trace REE abundances. Turning now to energy dispersive technology Yokhin and Tisdale (94/1588) considered advances in instrumentation in Part I1 of a review of high sensitivity EDXRF.The requirement for REE determinations referred to in the previous paragraph was tackled in a different way by La1 et al. (94/2362). These workers used a large 3.7 GBq (100 mCi) 241Am source closely coupled to the sample with a high purity germanium (HPGe) detector which showed superior detection efficiency to a previously employed Si( Li) detector. Detection limits of 20-50 ng were reported in the analysis of REE solutions aliquots of which had been evaporated to dryness on thin Mylar sheets. Counting times of 1 h were employed. These developments are likely to be of more general interest providing clear advantages can be demonstrated in comparison with REE analysis by existing solution techniques such as ICP-AES or ICP-MS.An instrumental solution to the problem of determining adjacent elements of grossly diflerent concentration exemplified by Mn (at ~ 2 % ) in steel was investigated by Yacout and Dunn (94/2369). Two possibilities were considered. The first was using excitation source energies just less than the K absorption edge of the heavier element (Fe) and the second was preferentially filtering the higher energy fluorescence X-rays before detection. To enhance detection of Mn in the above example these investigators undertook a simulation study based on a Monte Carlo code supplemented by a few experiments using various combinations of '09Cd or 244Cm excitation sources iron or cobalt as secondary targets (fluor- escers) and chromium foil as a detector filter.It is perhaps relevant to note that in a similar application one of the present reviewers achieved some success in the direct excitation of samples using a cobalt anode tube using an iron primary beam filter to enhance the detection of Cr Ka and suppress the excitation of Fe (Potts et al. J. Anal. At. Spectrom. 1986 1 467). 2.2.1. X-ray optics and microJuorescence instrumentation Once again this topic represents an area of significant activity with many contributions describing developments in beam lines available on SR X-ray sources or optical elements designed to be used on such beam lines. In this area of endeavour analytical chemists need to confront the rigours of theoretical physics head-on to achieve success in practical applications and it is encouraging to see that a significant number of such applications are now being reported in the literature.Indeed the X-ray microprobe currently represents one of the exciting areas of development in XRF with the capability of making a unique contribution in several areas of scientific endeavour. These studies continue to be comple- mented by the use of more conventional X-ray tube sources used for selected area analysis. For the purposes of this review instrumental considerations are covered in the present section whereas applications are considered in section 2.4. Almost all 'second generation' SR X-ray microprobes used ED detectors as X-ray spectrometers but there is currently much interest in the provision of WD spectrometers for this application. Thus Ohashi et al.(94/1) described a crystal spectrometer consisting of a flat analysing crystal and a position sensitive detector for use in SRXRF. This instrument was claimed to offer the advantages of high resolution multi- element detection and high efficiency offering detection limits in the pg g-' (or pg) range. ' Some of the recent developments on synchrotron storage rings have seen the introduction of undulators and wigglers both devices being capable of generating more intense X-ray beams than conventional bending magnets with some degree of energy tuneability. Thus Sasaki et al. (93/3906) described a new undulator for generating variably-polarized radiation and Shoji and Kato (93/3917) described an XRF analysis system for light element analysis using undulator radiation.This development at the Photon Factory in Japan was evaluated by measuring the B Ka X-ray emission and fluorescence XANES spectra of B BN and B20,. Treusch et al. (94/1595) ;used ray-tracing and finite element analysis programs to calcu- late the performance of an X-ray wiggler beam line at DORIS 111 a storage ring at HASYLAB Hamburg the main emphasis being the thermal distortion of optical elements caused by the bigh intensity of the wiggler-generated X-ray beam. Experimental measurements were made to check calculations. The exciting (!) opportunities presented by the European Synchrotron Radiation Facility (ESRF) were described by Van Langevelde et al. (93/2869) who made predictions of the minimum achievable spot sizes and photon flux densities using a Monte Carlo-based ray-tracing program.Using a combi- nation of planar and ellipsoidal crystals to focus the beam typical spot sizes were predicted to be 2-7 1-5 and 0.8-3 pm for bending magnet wiggler and undulator sources respect- ively with microanalysis detection limits in the 1-10 pg g-' range. During the current review period much has been published lon developments in mirrors and rejectors for use in microfluo- rescence (XRMF) instrumentation. These devices were included in a review of X-ray optics for SR by Malgrange (93/3892). The fabrication and testing of grazing incidence hirrors designed for X-rays in the 'hard' (0.1 nm region) was described by Uchida and Suzuki (93/3866). The particular interest here was the use of diamond cutting machines to minimize the surface roughness of the resultant mirrors as a means of obtaining a sharp reflection profile suitable for microprobe applications.Optical elements were also fabricated and tested by Zhou (93/3861) in particular spherical quartz laminar gratings scanning cylindrical mirrors for X-ray lith- iography and X-ray multilayer monochromator silicon sub- strate the emphasis again being on minimum surface roughness. Shaw (93/3877) described a method for calculating the effects of surface roughness on the grazing incidence reflectivity and transmissivity of a synthetic multilayer X-ray reflector. The effect of interfacial roughness on scattering from multilayer mirrors was examined by Spiller et al. (94/2371). Although this work was based on the image quality in a soft X-ray telescope the general conclusions appear to be relevant to other X-ray focusing applications. Chapman and co-workers (94/2242 and 94/2243) described the theory and practice of X-ray focusing using circular-pore microchannel plates or arrays of cylindrical capillaries where the focusing effect resulted from total reflection at grazing incidence on the interior surfaces of the hollow channels of the microchannel plate.The interest in these devices was in improv- ing the flux and spatial resolution in X-ray scattering and spectroscopic instruments or as a condenser in soft X-ray microscopy. Performance was found to depend on geometric parameters of the capillary arrays surface properties of theJOURNAL OF ANALYTICAL ATOMIC SPECTROMETRY OCTOBER 1994 VOL.9 287 R channels and X-ray wavelength. In a complementary paper Fraser et al. (93/3905) presented soft X-ray images produced by the focusing action of a planar square-pore microchannel plate and reported the three-component ‘cruxiform’ structure expected from previous studies indicating that there was a high degree of geometric regularity within and between about 46 000 active channels of the microchannel plate structure. Several patent claims were presented in the area of X-ray optics. Two patent applications from Tamura (93/3869 93/3884) described a multilayer mirror for use in X-ray mono- chromators and microscopes comprising either Be0 or BN as the first layer and a metal (Co Cr Fe Ni) as the second layer. Kondo’s patent claim (93/3883) also related to multilayer film X-ray mirrors comprising alternating non-metal element-based layers and alloy-based layers in which the alloy contained the element.Yamada and Adachi (93/3882) described X-ray reflec- tor plates comprising a Ge (111) or (100) single crystal substrate with a Au ( 11 l} or Bi-( 100) single crystal epilayer formed thereon or a Si{ 100) wafer with a Ge{ 111} epilayer. These devices were claimed to offer mirrors of long life and high reflectivity. A patent claim was also made by Taylor (93/3889) who described a heat exchanger designed to mini- mize heat-induced distortion of an X-ray monochromator presumably for third generation SR beam lines where X-ray intensities are so high that the heat transferred when the X-ray beam strikes an optical element is a significant problem.Although the use of glass capillaries as waveguides is dis- cussed in the following section Dubchik et al. (93/3881) described the transmission properties of multilayer thin films as X-ray waveguides. X-ray radiation was excited in a silicon Bremsstrahlung target by 100 keV electrons and directed at the end face of waveguide comprising 250 alternate layers of Ta and Al. The angular distribution of photons passing through this waveguide was measured. Sprangle et al. (94/2313) described a compact laser synchro- tron source designed to give a tuneable short pulse hard X-ray output. This device was based on Thompson backscattering of the intense laser radiation from a counter-streaming electron beam. Using current technology this device could generate picosecond pulses of X-rays of > lo9 photons per pulse with a peak brightness of 3 lo2’ photons s-’ mm-2 and mrad-2 with photon energies ranging from 50 to 1200 keV.2.2.2. Excitation Much of the recent interest in excitation sources has been centred on SR beam lines and the novel characteristics of glancing angle excitation TXRF. Both of these topics are treated elsewhere in this Update. The developments in polarized X-ray tube radiation over recent years have culminated in the availability of the first commercial EDXRF spectrometer now being available. As a contribution to these developments Swoboda and colleagues at Bremen (94/1584) reported on the suitability of an alumina Barkla scatterer as a polarizing element for the above type of spectrometer.The authors presented in detail a model for calculating the efficiency of Barkla scatterers and used it to calculate the scatter efficiency of boron carbide graphite and alumina. The excitation efficiency over the range 5-60 keV was calculated and measured for each scatterer and it was confirmed that the scatterer density controlled the absolute intensity of scattered radiation while the mean atomic number controlled the energy range over which maximum scatter intensity occurred. Alumina was recommended on the grounds of the wide energy range over which it could be used and freedom from contamination by other X-ray excitable elements. Using a high power tube (2 kW) an alumina scatterer and an 80 mm2 Si( Li) detector the authors claimed detection limits for the elements Ba Fe Sn Y and Zn in the range 0.25-2.5 ppm (1000 s livetime at 50% dead).In the area of X-ray tubes Benz et al. (93/3836) claimed a patent for a range of X-ray tube targets which have a reactive barrier layer made from molybdenum or niobium alloy and which are suitable for use in high power especially rotating anode X-ray tubes. A high voltage (50-250 kV) high power (3.75 kW) tungsten anode X-ray tube was characterized for its use in a secondary target geometry (93/1747). The beam uniformity intensity and purity were measured using both a spherical ionization chamber and an HPGe semiconductor detector. An interesting movable multi anode X-ray tube was described by Tirsell et al. (93/3855) incorporating four individual copper- based anodes. The filament and active anode were in a circular symmetry and the anode position was computer controlled by means of a three-axis translator which allowed the authors to produce a very symmetrical spot size from 1 to 5mm in diameter. Targets made from cobalt germanium iron and magnesium were used to generate four sets of intense moder- ately pure X-ray lines below 1.5 keV which were used for the energy and intensity calibration of X-ray imaging detectors.The X-ray tube output spectrum of a PW2188/00 chromium anode tube operated at 45 kV was carefully measured by Marchut (94/2354) over the wavelength range 0.035-0.55 nm using a modified sequential WDXRF spectrometer incorporat- ing a lead diaphragm in front of the detector. The tabulated intensities were normalized to the Cr Ka line.Using similar measurements by direct WD Jha and Agrawal (94/2361) reported the relative intensities of the principal lines for a PW2181/00 gold anode 100 kV 3 kW X-ray tube operated at very low Au L edge overpotentials. Calculations based on a Bethe equation of the expected L intra-series line ratios for tube potentials in the range 14-20 kV showed substantial deviation from their measured data. The long-standing use of selective excitation for EDXRF spectrometry was revitalized by Ladisch and Wobrauschek (94/2269) for the determination of minor concentrations of A1 in iron and Fe in copper using a secondary target EDXRF instrument. In the first case a titanium secondary target was excited by a 1.6 kW chromium anode X-ray tube operated at 40 kV to yield a detection limit (100s) of 400 ppm Al.In this example the A1 K lines were preferentially excited compared to the Fe K lines resulting in a more ‘balanced’ spectrum from which the A1 K and Fe K peak intensities could be extracted and used as a ratio for quantitative analysis. In the case of Fe in copper a nickel secondary target and a 2 kW molybdenum anode X-ray tube were used. The characteristic lines of the nickel target excited the Fe K line but not the Cu K lines which were excited by the residual Bremsstrahlung. Under these conditions of selective excitation an impressive detection limit for Fe in the ppm range was claimed. Selective excitation was also noted as a potential benefit in the study by Karydas and Paradellis (94/1573) who used proton bombardment of thick secondary targets of copper iron and titanium to produce monochromatic radiation for XRF analysis.One example quoted was the use of a copper target to yield a detection limit of 11 ppm Fe in a copper matrix. The authors also investigated the generation of polarized excitation beams from a proton- excited target but despite the production of ‘satisfactory’ amounts of polarization only the scattered target line intensity and not the spectrum background was reduced. The lack of reduction in background was attributed to the Compton scatter events from the interaction in the Si(Li) detector of gamma rays emitted by the bombarded nickel target. Efforts to produce X-ray lasers continue and the mechanism of their generation in plasma experiments was reviewed by Luk et al. (93/3901).Experimental results from the bombard- ment of a magnesium target with Kr’ ions (94/882); from steel and gold micro capillary targets (94/884) and from multilayer targets (94/883) were reported although none of these methods seems set to offer much scope for XRF spec- trometry for some time to come!288 R JOURNAL OF ANALYTICAL ATOMIC SPECTROMETRY OCTOBER 1994 VOL. 9 2.2.3. Detectors In their review of the status of solid state energy dispersive X-ray detection Kalinka and Taniguchi (94/2378) considered Si(Li) High Purity Silicon (HPSi) HPGe HgI silicon CCDs and superconducting tunnel junction (STJ) detection systems. The state-of-the-art for resolution in each of these device types was reported and the potential improvements in the newer devices were considered. The authors were careful to emphasize the importance of the first amplification stage and showed the progressive reduction of electronic noise for the detector FET (field effect transistor).The authors considered the performance of the five-terminal FET (PENTAFET) ‘probably to represent the ultimate limit (in electronic noise and therefore resolution) which can ever be achieved with the combination of an individual Si(Li) and FET’. Other parts of the amplification chain and of detector entrance window characteristics such as metallization and dead layers were considered in valuable detail. An outline of the status of X-ray detector systems at Kebangsaan University Malaysia was presented (94/2323) and included results from Na( TI) scintillators and four HPGe detectors as well as a discussion of aspects of resolution and efficiency.Bertuccio et al. (94/2333) reviewed the performance and use of silicon drift detectors in the 1-10 keV range. These authors also stressed the importance of the front end electronics and considered it necessary to integrate fully the electronics on the detector chip. Electronic devices and circuits suited for such integrated systems were summarized and the best results obtained to date were presented. The subject of low noise signal processing electronics was reviewed by Iwanczyk (94/2331) who concentrated on room temperature detector systems. Reflecting the author’s long involvement in the development of mercuric iodide detectors recent developments in this important area were presented and the design difficulties associated with the production of highly dense two-dimensional detector arrays with integrated low- power miniaturized processing electronics were highlighted.Details of the electronic characteristics of the PENTAFET were presented by Nashashibi (93/3860)-one of the inventors of this novel detector FET. The low attainable electronic noise and rapid (<1 ys) reset time with no associated after effects make the device suitable for both high resolution and high count rate Si(Li) detector applications. The best measured detector resolution at the time (1992) was shown to be 128 eV FWHM for the MnKa line. The author also indicated the potential suitability of the device for integration into a planar silicon room temperature X-ray detector.Four types of low noise FET suitable for use with semiconductor detectors of various capacitance were presented by Gostilo (94/2334). The best detector resolution was obtained when the detector and FET characteristics were closely matched. At the present time most room temperature detector systems use a mercuric iodide crystal as their detector element. Van den Berg (94/2328) reviewed recent developments at EG&G Energy Measurement Inc. in this technology. In particular the all- important subjects of crystal purification growth and detector fabrication were addressed and it was shown that increases in detector active area continue to be made. A mercuric iodide detector system will once more be incorporated in the X-ray analytical payload of a space mission.This time an ambient temperature mercuric iodide EDXRF spectrometer is to be included in the Russian Mars ’94 mission (94/2340) where it will be used to obtain a comprehensive elemental analysis of Martian surface material at several sites. The energy resolution is expected to be 200-250 eV FWHM at 5.9 keV. The develop- ment and performance of an impressive 20-element mercuric iodide detector array has been described in detail by Iwanczyk et al. (94/2316). The detector array which achieved an energy resolution of 250 eV FWHM at 5.9 keV was mounted on an SR beam line at SSRL (Stanford USA) and was fully charac- terized prior to its use in EXAFS measurements. Preliminary tests were reported (94/2336) for a mercuric iodide microstrip detector which was prepared for use as a position sensitive detector in a Bragg X-ray spectrometer.Energy and spatial resolutions of a device with six differently spaced microstrips (0.05-1 mm inter-strip distances) were measured. Energy reso- lution for a single strip was reported to be worse than that obtained for conventional mercuric iodide detectors and the target spatial resolution was about 50 ym. Conti et al. (94/2319) presented the results of a detailed Monte Carlo simulation of the response of a mercuric iodide detector in terms of efficiency energy and spatial resolution versus photon energy over the extended energy range from 20-100 keV. Despite the expected better charge carrier mobility expected for gallium arsenide room temperature X-ray detectors the achievable energy resolution remains unfavourable in compari- son with silicon or mercuric iodide X-ray detectors.Benz et al. (94/2338) reported fabrication of 2 x 2 mm and 3 x 3 mm room temperature gallium arsenide detectors which gave detection of X-rays down to 20 keV but an energy resolution of only 18% at 122 keV. Resolution obtained from a 2 x 5 mm detector was reported by Sumner et ul. (94/2339) to be 6% at 60 keV. Carrier trapping and device instability were identified as the dominant limitations to better performance. As with most detectors the careful preparation of the crystal starting material has been identified as a key step in fabricating lead iodide X-ray detectors. An improved technique for prepar- ing large oriented PbI single crystals for detectors has been described in detail by Zhang et al.(93/3908). The same group briefly reviewed (94/2337) progress in lead iodide detector fabrication and performance indicating that resolutions better than 1 keV (FWHM at 5.9 keV) can now be achieved. Similarly the availability of high purity thallium bromide crystals was claimed (94/2330) no longer to represent a limi- tation to its use in the production of X-ray detectors. Despite this there seems little evidence to support the use of such devices in the energy range where XRF spectrometry is commonly practised. Tsutsui et al. (93/3854) characterized a 32 element cadmium telluride detector which was fabricated on a single wafer. The spectrum obtained from one of the 32 elements was investigated and found to show spectral distortion attributed to the small size of the detector element.Escape peaks and the energy dependence of the photoelectric absorption process caused appreciable spectrum artifacts but the authors claimed that these could be corrected and that they only affected narrow and specific energy ranges. Despite the above interest in new detector materials silicon and germanium remain the most widely used detector materials in EDXRF spectrometers. E;quer (94/2329) reviewed the use of thin film semiconductors as X-ray detectors and highlighted the current rapid growth of interest in high purity amorph- ous silicon. One particularly active area of development with the poten- tial to form the next generation of X-ray detectors is that of superconducting tunnel junctions (STJs).Kraus briefly reviewed the subject (94/2321) then described in more detail the fabrication and characterization of detector systems intended for use in a multi-detector imaging system (93/3846). Using a device 470pm long the author demonstrated a positional resolution better than 5 pm and an energy resolution of 60 eV (FWHM at 5.9 keV). The application of modern micro engineering techniques to improve the performance of STJ detectors was described by Rippert et al. (93/3830). Gabutti and Gray (93/3865) described a superconducting strip detector which demonstrated a very rapid recovery time making it suitable for very high count rate applications. A study of the so called ‘proximity effect’ with Nb/AI/ Alumina/Al/Nb junctions (93/3831) and of a new design for a trans-impedence amplifier for junction biasing and signal amplification at low magnetic fields (94/23 14) was reported by a group in Utrecht.Kurakado et al. (93/3826) continued their development of Nb/Al STJs with the study of series-connectedJOURNAL OF ANALYTICAL ATOMIC SPECTROMETRY OCTOBER 1994 VOL. 9 289 R devices which yielded increased detection efficiency. Using a device with a large (for STJs!) active area of 178 x 178 pm2 an energy resolution (FWHM at 5.9 keV) of 88 eV was obtained. A series-connected array of aluminium STJs was fabricated and mounted in a specially designed adiabatic demagnetization refrigerator capable of holding the detector at the required temperature (110 mK) for use as an X-ray detector (94/2315). Tests on this new device indicated an achievable energy resolution approaching 100 eV FWHM at 5.9 keV.The energy resolution of the devices reported by Labor et al. (93/3852) and Saulnier et al. (93/3829) were rather less inspiring with FWHM at 5.9 keV of 190 eV and about 700 eV respectively. Clearly the energy resolutions obtained thus far for these devices are at least one order of magnitude greater than theoretically predicted. Recognising this Van Vechter et al. (94/2339) discussed several ways by which the applied magnetic field which is normally used to suppress the Josephson pair current may itself be contributing to the degradation in resolution generally experienced in STJ detectors to date. Thermal detectors (e.g. silicon microcalorimeters) operating at 100 mK have mostly been of interest to date in X-ray astronomy.This is perhaps surprising in view of the impressive energy resolution of 7 eV at 5.9 keV available from the devices described by Juda et al. (93/3835). Although also initially targeted for X-ray astronomy the devices produced by McCammon et al. (94/2320) have a similarly impressive energy resolution (12 eV at 5.9 keV) and were identified by the authors as being potentially of use for laboratory X-ray spectrometry. The high flux of SR beams which could offset the low collection efficiency arising from the small detector size coupled with the physics expertise associated with SR facilities could perhaps see these high resolution X-ray detectors being incorporated in SRXRF equipment before any of the STJ detectors.Although there is considerable research effort directed at the very high resolution ultra-low temperature STJ and other exotic detector types the gas proportional detector which could perhaps be regarded as more mundane and certainly at the opposite end of the attainable resolution scale still attracts some attention. This is particularly the case for the Coimbra group who have continued their work in this field with a theoretical (93/38 16) and experimental (93/38 14) investigation of energy linearity and distortion effects in xenon-filled gaseous detectors. The theoretical study used a Monte Carlo simulation approach to investigate energy non-linearity in the 0.1-5 keV range. Distortions were shown to correspond to the loss of electrons in the detector window and to the influence of the electric field strength on the X-ray absorption zone.The experimental study performed over the wide energy range of 2-60 keV demonstrated the presence of a noticeable discon- tinuous decrease of 92 f 10 eV near the 4.8 keV Xe L shell threshold. Throughout the rest of the energy range the response was linear including the region around the Xe K edge (34.6 keV) although the experimental error may well mask a slight discontinuity at the K edge. A modified xenon gas proportional detector was described by Dietz et al. (93/3828) in which a pre-amplification region was added to improve energy resolution by a factor of two. The gas pressure vessel was built using molybdenum to reduce background and high energy sensitivity in the 50-70 keV range was claimed to be improved by a factor of three by increasing the active gas depth from 9 to 14 cm.Dos Santos and co-workers at Coimbra (93/3813 93/3817 94/1586) described in detail their design for a simple gas proportional scintillation counter for use in the 3-30 keV range. Using a well-collimated 55Fe radioisotope source the resolution of the detector (5.9 keV) rose from 8 to 9.3% for window diameters in the range 10-25 mm. Using the same range of window sizes but making measurements on a manganese disc placed 10mm away from the window and excited by a 244Cm source the detector resolution ranged from 8.2 to 10.6%. Providing greater collimation and smaller window sizes are used and depending upon the authors’ definition of inexpensive (a claim which is made for these detectors) these room temperature detectors could be an attractive proposition for portable or bench-top low-resolution EDXRF instruments.An interesting multilayer detector window for use with a space-borne X-ray CCD was described by Stephan et al. (93/3851). The window was made from layers of Parylene- N (25 pg crnp2) aluminium (30 pg cm-2) and carbon (25 pg cmP2) and had an unsupported area of 30 x 10 mm2. Several types of ultra-thin detector window were investigated by Viitanen and colleagues (93/3833). Active areas up to 140 mmZ were achieved and the best transmissions were > 20% at B Ka and >40% at N Ka. A new gas impermeable layer utilizing aluminium nitride was also investigated. The accurate calibration of energy scale and efJiciency of X-ray detectors typically requires the availability of a number of isotope sources.A compact calibration source based upon the secondary fluorescence of pure element targets by an encapsulated 55Fe source was described by Lepy et al. (93/2777). In cases where pure element targets were unavail- able ion-exchange membranes were soaked in solutions such as calcium chloride dried and embedded in a PTFE matrix. The moveable four-anode X-ray tube developed by Tirsell et al. ( 93/3855) and described in the Excitation section above is a particularly useful high-energy source for the calibration of detector energy scales. Wernisch et al. (94/2318) described a novel method for the in-situ efficiency calibration of an Si(Li). The method involved the varying of detector incident radiation angles and energies and used a knowledge of the metallization layer material and thickness.Using these data the thickness of the silicon dead layer and of the beryllium window were determined from the measured data. The overall efficiency was determined by the measurement of the Bremsstrahlung spectrum collected by the detector. Although providing a useful and fairly straightforward method for meas- uring detector efficiency it should be borne in mind that the initial assumption about contact layer material could well be confounded by what it actually is. A fast X-ray counting system in which each individual component in the measuring chain was carefully selected to minimize overall system dead time was described by Short et al. (93/3812). The authors detailed the components used and how the speed of each contributed to the attainment of an experimentally determined dead time of only 42 ns.The system could operate at input count rates of up to 36 Mcounts s-l and was shown to exhibit negligible dead-time losses at 10 kcounts sC1 and only 4.2% counting losses at 1 Mcount s - ’. The limiting component in this interesting measurement chain was the 60 MHz counter-timer PC interface board. Mikhailov (94/1693 and 94/1597) described a new pulse pro- cessor circuit for correction of ballistic deficit in semiconductor detectors. Results were presented for a germanium gamma-ray detector and the circuit was claimed to be much simpler (though not as effective) than pulse processors incorporating a series switch.Cousins et al. (93/3876) discussed the special problems of pile-up rejection and sum peaks in SR experiments using solid- state detectors. Pile-up (sum) peaks were particularly prevalent when the storage ring was operated in single bunch mode with an inter-bunch interval of up to an order of magnitude shorter than the pulse processor time constant. Such effects are not limited to SRXRF systems and are seen in conventional EDXRF systems especially when pulsed X-ray tube operation is possible. The authors indicated a need for a full evaluation of pulse processing parameters to permit correction of spectra for sum peaks but they were unable to provide a scheme for that correction. 2.3. Total Reflection X-ray Fluorescence (TXRF) The literature of TXRF spectrometry continues to grow yet its widespread acceptance in the wider atomic spectrometry290 R JOURNAL OF ANALYTICAL ATOMIC SPECTROMETRY OCTOBER 1994 VOL.9 community remains low and heavily overshadowed in the simultaneous multi-element chemical analysis arena by ICP-MS and ICP-AES. This situation may yet change when such an influential member of the ICP community as Paul Boumans recognizes as he did in his 40 year historical personal view of Plasma Spectrochemistry (94/590) TXRF spectrometry as ‘a perhaps curious intruder and possibly serious competitor’. This personal but perhaps defensive view by Boumans of an intruding competitor partly explains why TXRF spectrometry is not being developed or used as extensively as perhaps it deserves. No technique is an analytical panacea and when TXRF spectrometry is widely considered to be a valuable complement in the modern analytical atomic spectrometry armoury as it is I believe by those requiring non-destructive surface analysis and depth-profiling in the semiconductor industry will the prejudice lack of understanding and reticence to accept TXRF have been dispelled.In May 1992 the ‘Fourth Workshop on Total Reflection X-ray Fluorescence (TXRF)’ returned to GKSS at Geesthacht in Germany the site of the first workshop in 1986. The proceedings of this important conference were published in Spectrochim. Acta Part B 1993 48 which constituted as for previous proceedings of the workshop a unique status report of all aspects of TXRF spectrometry. The original idea for TXRF came from a paper by Yoneda and the Workshop was fortunate to host a rare presentation (94/651) by his colleague Professor Horiuchi (sadly Professor Yoneda died in December 1992).The author described how they discovered the high sensitivity of TXRF analysis of optical flats as a result of initial work on anomolous surface reflection of X-rays. The develop- ment of the equipment used in the original series of experiments was described which led to the production of their analytical results on the direct microanalysis of U in sea-water and Fe in human blood. The author concluded by describing the continuing research of their group into Compton and Rayleigh scattering in TXRF measurements and the interesting combi- nation of TXRF and total reflection X-ray diffraction (TXRD) which has been used for studies of molecular structure in ultra- thin organic films and was the subject of a second paper at the workshop (94/2218).Tolg and Klockenkamper (93/3373) provided an authoritat- ive and comprehensive review of the role of TXRF in atomic spectrometry showing its favourable comparison with ETAAS and ICP-MS for trace chemical analysis and with RBS and SIMS for surface analysis. The authors began with a well- considered philosophical discussion of the balance between power of detection accuracy and economy which raised important topical questions of training of analytical staff the validity and utility of quoted detection limits and the ever- important topic of blanks in ultra-trace analysis. The bulk of this excellent review was given to a comparison of the perform- ance of a wide range of trace analytical techniques for trace micro and surface analysis.A brief German review of TXRF for ultra-trace analysis was given by Hegedus (93/3858). Usui (93/3893) briefly reviewed SEM-TRAXS which combined SEM and TXRF geometry. Possible areas forfurther development of TXRF spectrometry were considered in some detail by members of the GKSS group (94/662) who continue to push forward development on many aspects of TXRF. The authors suggested that surface analysis has most to gain from advances in instrumentation noting that the extension of chemical analysis to the light elements by means of vacuum spectrometers and thin window or windowless X-ray sources and detectors was unlikely to yield quantitative analysis unless the problem of calibration (sensitivity and X-ray correction) was also addressed.However the calibration problems were claimed to be less in surface analysis and it was suggested that the development of WD spectrometers in place of ED detectors and of rotating anode X-ray tubes to offset the substantial reduction in sensitivity with the WD detector would be of benefit for light element (C,N,O) determination. Perhaps even more interestingly the authors suggested using a beta spectrometer and extending the technique into the realm of total reflection XPS in much the same way that Kawai has described (93/1081). In the case of chemical analysis the best opportunities for improvement in TXRF were suggested to be in the area of sample preparation where methods such as the direct use of sample carriers in cascade impactors and in electrochemical cells showed much promise for maximizing the main benefits of TXRF measurements.Streli and co-workers (94/654 94/1596) at the Atominstitut in Vienna described in detail the design and performance of a new TXRF spectrometer for the measurement of light elements (C F N Na 0 specifically). The complete system was optimized for this low energy region of the spectrum and all components were incorporated into a sample chamber which could operate at high vacuum yet still provide a convenient means of changing samples. The authors used for excitation either a fine focus chromium anode X-ray tube or a specially- developed windowless copper anode X-ray tube with high incidence and takeoff angles.Both tubes were operated at 25 kV and 20 mA and the additional low energy Bremsstrahlung and contribution from the Cu L lines yielded lower detection limits for C 0 and F of 10 7 and 1.4ng respectively than for the chromium anode tube (13 10 and 3.4 ng respectively). Once the effects of the Cu L line excitation were lost detection limits for elements with higher energy lines were substantially better using the fine focus chromium anode tube owing to its higher brilliance. Samples were placed on pure silicon reflectors and a commercially available HPGe with a 0.4 pm DIAMONDTM (diamond-like carbon) entrance window and a resolution of 123 eV at Mn Ka. Calibration curves which were truly curves were produced for C and 0 and the reported absorption effects observed with more than 10mg sample masses tends to support the view expressed by Schwenke and colleagues (94/662) that the major limitations in light element TXRF analysis lies mainly in calibration not in the availability of instrumentation.The design detail of a versatile multifunction vacuum chamber for TXRF measurements using various excitation and detection regimes was given by Rieder et al. from Atominstitut Vienna (94/1248). The unit was used with many types of X-ray tube system and also at the SSRL (Stanford USA) SR facility for measurements with SR tuned to 8.88 keV using a double crystal monochromator. In their own laboratory the authors used a 12 kW rotating anode (either copper or molybdenum) X-ray tube either for direct excitation or in conjunction with a tungsten-carbon LSM monochromator (94/1594).Benefits either in shorter measurement times or lower detection limits were as expected from the increased current available although there was no real benefit in detection limits as a result of the improved peak background ratio obtained from the monoch- romatized incident radiation. Detection limits (1000 s) as low as 610 fg Sr and 170 fg Mn were obtained. When operating at higher power limitations in the authors’ pulse processing electronics and detector backgrounds became evident. Further instrumentation developments by the highly-active Atominstitut group were concerned with excitation (94/652). Using the new multifunction vacuum chamber described in the previous paragraph the authors investigated the use of conventional fine focus and rotating anode tubes and a synchrotron source.Aspects of this work were also covered in papers referenced above (94/1248 and 94/1594) but perhaps of most interest was the use of a 100 kV gold anode X-ray tube in conjunction with an HPGe detector for the measure- ment of the REE K series emission lines. Single element detection limits (1000 s) for elements in the 2 range 57-72 were better than 1 ng with this new fine focus 100 kV tube. With their previous 90 kV broad focus tube the detection limits for elements with atomic number greater than 65 had quickly deteriorated to reach 15 ng for 2 = 70. Finally usingJOURNAL OF ANALYTICAL ATOMIC SPECTROMETRY OCTOBER 1994 VOL. 9 291 R perhaps the ultimate state-of-the-art excitation for Ni K lines of an SR beam monochromatized at 8.88 keV the authors managed to obtain a detection limit (1000 s) of 50 or 150 ppb Ni in oil (depending on use of beam polarization). It is perhaps important to bear in mind that state-of-the-art commercially available WDXRF (and some EDXRF) spectrometers will achieve these detection limits for Ni in oil an application for which TXRF spectrometry is less well suited.Van den Hoogenhof and De Boer (93/3850) described the characteristics of glancing angle X-ray analysis (GIXA) which includes conventional TXRF. The information available by scanning incidence and detector angles was presented and the authors showed how the reflectivity data were influenced by interfacial roughness in multilayer samples. Overall the com- bined data which were generated from a single instrument provided conventional TXRF surface analysis; TXRF elemen- tal composition depth profiling of layered structures and interfacial roughness from reflectivity and rocking curve measurements.Bormann and Schwenke (93/3887) have been granted a German patent for a method and a device using what appears to be GIXA. 2.3.1. Chemical analysis As is the case in conventional XRF and for that matter all analytical atomic spectrometry sample preparation and speci- men presentation must be attended to with considerable care. Prange and colleagues (94/653) at GKSS examined the use of pyrolytic boron nitride as a sample carrier for TXRF spec- trometry. Its high purity chemical and thermal stability mechanical strength and absence of any elements where emis- sion lines occur in the energy region typically used for TXRF analysis are particular benefits.It shares many of the beneficial properties of quartz but has the added benefit for TXRF of being naturally hydrophobic. The sole disadvantages were reported to be its higher surface roughness (20-30 nm) com- pared to quartz (lonm) which tended to increase spectral background and its high cost. Improved polishing techniques were expected by the authors to reduce the former problem but could be expected to further exacerbate the latter. A new and highly efficient electrostatic precipitator has been devel- oped for the collection of sub-pm airborne and vehicle exhaust gas particulates directly on to TXRF sample carriers (94/659). The precipitator was shown to give 100% collection of particles smaller than 1 ym and SEM measurements showed the depos- ition to be extremely uniform which is a particular benefit for TXRF analysis. With such uniformity the authors suggested that internal standardization would be unnecessary and could be replaced by external calibration and use of constant tube current.Under such conditions it was hoped that a measure of total deposited sample mass could still be made. Future work was also outlined to investigate the use of glassy carbon sample carriers which would provide stronger attractive elec- trostatic field forces as a consequence of its higher electrical conductivity. Sinner et al. (93/3374) used TXRF analysis for the determination of Fe" and Fe"' in eluate from microbore ion chromatography columns. The use of a TXRF finish increased measurement time by a factor three over the use of ETAAS with no matching benefit in detection limit.The only time that the TXRF finish would seem to be of benefit is when much smaller sample volumes (10 pl instead of 100 pl) are used but even then the limitation of the ETAAS system may lie more in the autosampler than the technique's attainable detection limit. Hegedus (93/3825) described a sampling method in which irradiated high alloy steel samples were scraped and the scrapings dissolved before a chemical separ- ation of Nb and Ta from the chromium-iron-nickel matrix. The resulting Nb-rich sample was analysed by TXRF spectrometry. The difficulties in the determination of trace elements in open sea-water is an area of environmental analysis where TXRF has been extensively applied during the past decade.The continuing acceptance of TXRF in this application is evidenced by the results from two major North Sea research projects in which TXRF analysis has replaced ETAAS for the routine analysis of sea-water (94/655). The method used a sample preparation method involving complexation with sodium dibenzodithiocarbamate followed by reversed-phase chroma- tography which provided both preconcentration and matrix separation. The TXRF results were highly reliable and showed excellent agreement in an interlaboratory comparison. Schmidt et al. (94/44) used an identical sample preparation scheme in an on-board clean laboratory during a survey cruise to analyse open Atlantic Ocean water.Once more the performance of TXRF compared very favourably with ETAAS and ASV for these very challenging analyses. Perhaps underlining these continuing successes Freimann et al. (94/45) reported on the performance of TXRF analysis in the certification of Cu Ni Pb and Zn in BCR CRM 403 a sea-water RM. For the elements which could not be certified the TXRF method showed good agreement with other techniques such as ETAAS and ASV. A related intercomparison exercise on a tentative estuarine water RM showed similarly high quality TXRF results. Prange and colleagues (94/657) used TXRF for a study of trace elements in river water from the Elbe. Direct analysis of filtered fresh water samples was carried out with detection limits in the low pg 1-l region.Samples from the lower brackish region of the river were subjected to the same matrix separation and preconcentration procedure used above for sea-water analysis which yielded detection limits in the range 0.05-0.5 pg I-'. Battiston et al. (94/2219) used TXRF spectrometry for heavy element speciation in coastal sediments. Five fractions were evaluated exchangeable metals; metals bound to carbonates; metals bound to iron and manganese oxides and hydroxides; and metals bound to organic matter and residual metals. Full details of the sequential chemical separation of each of the fractions was given and data were presented for 12 environmen- tally important elements in the 19 < 2 < 82 range. TXRF spectrometry is equally suited to the simultaneous multi-element analysis of airborne particulate matter as described in the report of work carried out in collaboration with the Dortmund group by Salva (94/1601).The first of the simple straightforward sample preparation schemes adopted was to collect dust on a cellulose nitrate membrane filter then punch out a 1 mm diameter disc and collapse the filter with a drop of tetrahydrofuran. The second technique was to collect dust directly on Plexiglas sample carriers serving as impaction plates in a cascade impactor. In both cases the internal standard was applied and dried prior to TXRF measurements and detection limits were reported to be in the 1-10 ng range for elements in the 2 range 20-82. The use of 1 mm filters allowed the distribution of elements on a collection filter to be measured and in the case of the cascade impactor it was found that memory effects and erosion of the impactor itself gave high and unstable blank values.TXRF spectrometry was used to analyse both rain and air particulates in a study at GKSS (94/658) into trace element deposition into a forest ecosystem. The elements in filtered rainwater were concentrated 20 fold by freeze drying and particulate samples on filter paper were digested and then analysed as solutions in the normal way. Up to 25 elements with 2 ranging from 16 to 82 were determined with detection limits in the range 20-180 pg g-' for freeze-dried rain water and 10-100 pg m-3 for dust samples using a rather long measurement time in the range 3000-7000 s. Moving away from the stock-in-trade environmental appli- cation of TXRF Dogan et al.(94/1728) used TXRF for the simultaneous determination of 12 trace elements in blood serum. Four sample preparation procedures were investigated direct analysis; digestion with a nitric and perchloric acid292 R JOURNAL OF ANALYTICAL ATOMIC SPECTROMETRY OCTOBER 1994 VOL. 9 mixture; high pressure nitric acid digestion; and digestion with nitric acid at normal pressure. Only these last two procedures gave reliable results and the normal pressure digestion could be completed in almost half the time of the high pressure variant. Results for the element Cu Fe Rb Se and Zn seemed to be of greatest utility in the authors' investigation of Behcet disease. Schirmacher et al. (94/656) used TXRF spectrometry for the determination of 10 elements in marine fuel oil and oil sludge.Five elements (Ca Fe Ni V and Zn) could be used to distinguish the two types of oil in support of maritime legal shipping cases. Detection limits (1000 s) were in the range 2-17 mg kg-' which compares unfavourably with those attain- able with conventional bulk WDXRF and EDXRF methods for elements in oil. The high recoveries (107-113% and 137% for Ca) of the TXRF sample preparation method and the small sample size (3-10 mg) combined with the uninspiring detection limits suggest this is unlikely to become a widespread appli- cation for TXRF spectrometry. Ojeda et al. (94/2220) on the other hand produced a simple direct TXRF method for the determination of Fe Ni S and V in crude oil. The oils were heated and homogenized and a few mg placed directly on the quartz sample carrier.Unlike most other TXRF quantitation methods but in common with conventional XRF analysis these authors used external stan- dardization and plotted the ratio of the analyte line to the W La Compton backscatter signal against concentration. Some non-linearity was experienced due to limitations of the spec- trum processing package but a wide variety of crude oil types was successfully analysed and detection limits (200 s) for Fe Ni and V were better than 1 ppm for the ideal excitation by W L lines. The authors did note higher than expected values in light crudes and these were attributed to evaporation of the sample matrix during irradiation. The ability of TXRF spectrometry to measure very small samples was put to great effect by Klockenkamper et al.(94/660) in the analysis of fragments of only about 1 pg of artist's pigment from valuable works of art. These minute samples were taken either by means of a toothpick or a dry Q-tip which was gently touched on the painting and then tapped onto a quartz sample carrier. The resulting micro fragments were analysed directly by conventional TXRF. A simple quantification process using the 'normal' sensitivity factors and then normalizing concentration to 100% was adopted and shown to be perfectly adequate for the require- ments of pigment characterization and providing advice for restoration. As a consequence of the minute amount of sample taken and the convenience of the method it should be accept- able for cases where extremely valuable works of art require analysis of the pigments used.A large multinational team used the Atominstitut's flexible TXRF chamber described earlier in this Update for the TXRF determination of trace transmutational elements in iron and copper alloys used in the structures of nuclear fusion reactors (94/1577). The SR beam used for excitation was monochromat- ized at an energy just below the absorption edge of the matrix element and detection limits (1000 s) of 8 pg g- ' were achieved for Co and Fe in copper. Although TXRF is of benefit when only small samples are available and can be digested if sufficient material is available for a conventional XRF sample then either modern WDXRF or EDXRF with selective exci- tation as advocated in a paper from workers at the Atominstitut (94/2269) may provide a more accessible analytical alternative.Certainly the use of selected anodes or rotating anode tubes and a monochromator should be capable of approaching these detection limits for conventional samples. Through adjustments to the flexible sample chamber the authors were able to investigate the benefits to be gained from using the natural polarization of the SR beam. As a result the detection limit (1000 s) for Ni in oil was improved from 150 to 50 ng g-'. The difficult determination of Ce Eu and Tb in gadolinium- and lanthanum-containing electroluminescent materials was reported by Freiburg et id. (94/2221). The extreme peak overlaps in the analysis caused some problems for the spectrum processing software and resulted in detection limits (1000 s) of around 0.1% rising to almost 1% when an atomic number neighbour was present in much higher concentration.As a consequence the authors recommended the use of WDXRF and fundamental parameter (FP) quantitative analysis but provided no data to support their decision. Sensitivity factors were interpolated using the useful technique of plotting log (sensitivity) uersus reciprocal energy. 2.3.2. Surface analysis and depth proJiling TXRF spectrometry continues to grow in stature in this important area particularly so for contamination control in semiconductor wafer fabrication plants. Bernieke (94/2222) presented a review of the basic features of the TXRF technique and instrumentation used for silicon wafer surface contami- nation studies.Instrumental aspects of the subject were pre- sented by Schwenke et al. (93/3894) who showed detection limits of the order of 10" atomscm-2 corresponding to the detection of 0.2 pg in a 2mm diameter spot. Both papers showed how measurements with varying incidence angles could be used to show true surface particulate contamination and also sub-surface bulk contamination. Kroninger et al. (93/3837) used TXRF spectrometry to investigate the release of particu- late and metallic contaminants onto the rear surface of wafers. The amount of contamination released was found to depend upon the type of chuck and nature of the wafer surface coating. The deposition characteristics of metal contaminants from HF process solutions used in wafer cleaning operations was inves- tigated by Hsu et al.(93/3870) while Frost et al. (93/3755) compared SIMS TXRF and vapour-phase decomposition (VPD)-AAS for the determination of near-surface contami- nants on silicon wafers. The challenge of depth profiling of layered structures by TXRF was taken on by Knoth et al. at GKSS (94/661). The authors presented a discussion of their work including exper- imental measurements and theoretical modelling. In the diffi- cult application of measuring a very thin cobalt layer between two copper layers on float glass they were unable to determine non-destructively from TXRF alone whether the cobalt layer was thin and of high concentration or broad and of lower concentration. In general the authors considered that the depth profiling capability of TXRF depended very much on the type of multilayer system under study and the technique tended to fail with increasing depth.In some cases they suggested the combination of sputtering and TXRF could be used to good effect in some multilayer systems. In the specific case of a SiGe layer on a silicon substrate Kunii (94/1591) was able to obtain valuable non-destructive but rather simple depth profile infor- mation by means of TXRF measurements. The Ge Si ratio was determined by Auger electron spectroscopy and could be used to calibrate the TXRF measurement of Ge La Si Ka peak ratio. The glancing angle was changed between two settings to allow the Ge Si atomic ratio to be determined in the first 3 nm and then in the depth up to 200nm. The thickness of the SiGe layer was also determined from the TXRF measurements at the same time as the composition data.2.4. X-ray Microprobe Although many applications that are considered in this section are based on the use of SR sources interest continues in the use of lower intensity X-ray tube sources some of which use glass capillary coupling optics. There is also an extensive interest which this reviewer finds surprising in the use of X-ray microprobes in imaging applications involving elemental mapping. General aspects of analysis using SR sources were reviewed by Rindby (94/1564) Sutton et al. (94/2380) Brown (94/688) and for Chinese language readers Chai (93/3126).JOURNAL OF ANALYTICAL ATOMIC SPECTROMETRY OCTOBER 1994 VOL. 9 293 R One aspect to which reference was made was the complemen- tary nature of SRXRF EXAFS and XANES microanalysis as well as the opportunities that will shortly be offered by the very high intensity beams available from third generation synchrotron X-ray sources.One important area of application of the X-ray microprobe is in the analysis of geological samples. Indeed the synchrotron has particular advantages in the non-destructive analysis of fluid inclusions as described by Vanko et al. (94/2228) who developed protocols for the analysis of individual synthetic fluid inclusions in Brazillian quartz containing fluids of known NaCl-CaCl,-H,O and KCI-H,O composition. The effect on element ratios of variable inclusion depth size geometry and salinity was investigated and results were usually to within 30% of the nominal ratio.In another geological application undertaken at the Brookhaven National Laboratory Treiman and Sutton (94/2370) studied the composition and zoning in the Zagami meteorite to evaluate its magmatic history. Determinations were made by electron and SR X-ray micro- probe; the elements Cu Ga Ni and Zn being determined by the latter technique. Figueiredo et al. (94/1572) used the LURE synchrotron facility to analyse natural lithium aluminosilicates (lepidolites) analytical results being reported for Au Nb Ta T1 and W. Daryin et al. (92/3007) described results by scanning SRXRF microanalysis of phosphate-ferrous-manganese nod- ules from the NW Pacific prepared as thin sections. Samples were excited at 23-25 keV and using an exposure time of 1 to 3 s per point with a raster size of 75 x 100 pm2 and detection limits of 1Opg g-’ were claimed for the Sr to Mo group of elements.Another important area of application where the non- destructive nature of X-ray microprobe analysis can be exploited is in the analysis of biological samples. Iida and Noma (94/1599) analysed human hair by synchrotron X-ray microprobe using a 5 pm focused X-ray beam obtained using either double crystal or multilayer monochromators offering X-ray fluxes of 107-10s and 109-1010 photons s-’ respectively. Their results demonstrated that mapping the distribution of trace elements in round slices of human hair in two dimensions was important in understanding the inhomogeneous distri- bution of these elements. Wu et al. (94/1069) in a Chinese language publication used a synchrotron X-ray microprobe technique to analyse womens’ hair during pregnancy reporting the Ca C1 Cu Fe Zn contents and claiming detection limits of 0.46 ng (Ca) 47 pg (Cu) 5.3 pg (Fe) and 0.65 ng (Zn).Other biological applications included the analysis by SRXRF of growth plate cartilage from calf scapula by Vittur et al. (94/2307) the elements of interest being Ca Ni S Sr Zn measured with an analytical resolution of 10 x 10pn2. Yao et al. (94/2381) analysed single cells of liver by both micro- PIXE (a technique favoured for its high resolution) and SRXRF favoured because of its low damage to the specimen. The feasibility of experimental methods and specimen prep- arations were discussed. Dzfusion coejicients in aqueous gels were studied by Anderson et al.(94/1575) using the Daresbury SR X-ray microprobe. Concentration-time and concentration-distance profiles for the elements Co Cu and Zn with and without hydrobromic acid being present were measured and results indicated that the diffusive flow of these elements was coupled with the diffusive flow of the acid. Aside from X-ray microprobes based on SR there has also been interest in the use of Conventional X-ray tubes as excitation sources using apertures glass capillaries and/or microfocus X-ray tubes. Some of these devices have been based on the use of parallel-bore glass capillaries but the more complex task of optimizing designs based on tapered-bore capillaries was tackled by Lindgren and Selin of Chalmers University (94/1568) with a view to deriving conditions (including entrance and exit diameters and length of capillary) for opti- mum intensity enhancement.Working at the same laboratory Selin et al. (94/1578) described results from an instrument based on a conventional X-ray tube source and secondary target (normally molybdenum) in which the exciting beam was allowed to pass through a pure silver slit collimator 0.1 x 1 mm. This instrumentation was applied to the analysis of tree rings in an investigation related to the effects of acid rain. In a contribution from Rindby’s research group also based at Chalmers University Voglis et al. (94/1570) used an XRF instrument fitted with a conical glass capillary wave- guide for the elemental mapping of bone tissue. It seems possible that X-ray tube systems that incorporate capillary-coupling optics have not yet achieved their full poten- tial and one way forward has been investigated by Yan and Ding (94/1598) who described instrumentation incorporating an ’X-ray lens’ comprising bundles of capillaries claimed to increase power densities by lo4 and to offer a probe diameter of 500pm.This present source cannot be described as a ‘microprobe’ in the normat sense and the relationship of this instrumentation to the recently announced ‘Kumakhov’ lens system is not clear to this reviewer but instrumentation of this nature clearly has potential. Resolution of a similar magnitude (this time 300 pm) was reported by Valdre (93/3809) for XRMF instrumentation based on a scanning electron microscope using different targets to excite different regions in the spectrum and a windowless ED detector.Although this design of instrumentation has been described previously this contribution further demonstrates the variety of configurations possible for XRMF. In the analysis of particulate material. Lankosz et al. (93/3811) tested experimentally a Monte Carlo method for simulating matrix and particle size effects in the XRMF analysis of individual particles having diameters of less than 200 pm. This technique was evaluated in the analysis of spherical glass particles for Ca and Fe and results of Monte Carlo simulation agreed with experimental data to within 10%. Experimental measurements were undertaken with a molybdenum anode X-ray tube and a single aperture X-ray source collimator having a nominal hole size of 70 pm.Of the contributions considering the imaging capabilities of X-ray Juorescence Ebel et al. (94/1581) described a 3-D imaging system applied to the analysis of the lateral thickness distributions of Zn layers on steel presenting in addition the principle of the technique and experimental background. The instrumentation was based on a Cu anode X-ray tube with appropriate collimator and an ED detector which could be adjusted to view the sample at a variable take-off angle of between 30 and 80”. A fine focus X-ray tube was used by Fukumoto et al. (93/3 168) for the non-destructive elemental mapping of biological samples. The beam size was constrained by a 20 pm i d . glass capillary and fluorescence spectra were detected using two large area Si( Li) detectors.Instrumentation was applied to elemental distributions in leguminous plants and brain tissue. Using a synchrotron-based system at the HASYLAB facility Bavdaz and Gurker (93/3797) compared image data acquisition by line scanning (termed ‘coded image’) and conventional point scanning (‘direct’ imaging) and con- cluded that the line scanning technique gives a better S/N ratio thereby offering superior image quality. Multi-element imaging in computerized X-ray tomography was described by Fryar et al. (94/2375) and in a second contribution Markham and Fryar (94/2377) presented the merits of undertaking computerized tomography using an X-ray tube as the source and a low energy-resolution detector system. Iida et al. (94/1600) described high spatial resolution XAFS and its imaging applications using a SRXRF microprobe. 2.5.Field Analysis One important instrumental development of recent years is that of portable X-ray fluorescence analysers. Such instrumen- tation has already made an important mark in environmental294 R JOURNAL OF ANALYTICAL ATOMIC SPECTROMETRY OCTOBER 1994 VOL. 9 applications particularly in the analysis of soils and hazardous waste sites and has a clear role to play in geological archaeol- ogical and some industrial applications where immediate decisions can be made on sampling strategies for example. One of the more valuable recent developments has been the introduction of the room temperature high resolution mercuric iodide detector and the use of instrumentation incorporating this device was described by Bernick et (11.(94/2) in the field screening of hazardous metal wastes. Results were presented for soil analyses from different sites to demonstrate the potential of the fundamental parameter approach to obtain acceptable data quality with minimum calibration effort and avoiding the need to prepare site-specific standards. Piorek and Pasmore (94/2326) also applied field portable instrumentation to haz- ardous waste screening and identification using samples of natural soil spiked with the elements of interest for calibration and 30-200 s count time for each soil analysis. A portable XRF analyser was used by Zhou et al. (93/3845) for the analysis of ore samples in particular copper prospecting in NE China either by the direct analysis of samples for Cu or for other pathfinder elements.Also for the analysis of ore samples Bai et al. (94/2311) made a patent claim for a portable XRF analyser incorporating 55Fe 238Pu and 241Am excitation sources and a ‘high resolution’ proportional counter although from the limited details available for review it is not clear how this instrumentation differs substantialy from existing commer- cial instrumentation. Recent applications of radioisotope excited X R F especially portable analysers but also other applications including on-line and in vivo were reviewed by Valkovic et al. (94/1566). 2.6. Calibration and Data Processing In Part 3 of a series of papers addressing problems of automated qualitative analysis in WDXRF Jordanov et al. (93/3802) presented the details of the algorithms in a computer program for the automatic selection of elements present in an unknown sample.The authors reported that an approach using fuzzy set data did not yield reliable results but their use of ‘coefficients of presence’ did. The rules embodied in their algorithm were presented and showed that they chose to ignore with good reason the perturbation of intra-series line ratios caused by matrix effects as only the spectrum data were used and no attempt was made to use quantitative information. The ‘coefficients of presence’ allowed an easy ranking of likely-to- least-likely elements but the cut-off point for determining presence remained subjective. The problems of spectrum processing in EDXRF become ever more challenging as increasingly complex materials are analysed and lower detection limits are required.Members of the Antwerp group have applied a partial least squares (PLS) method to the problem and reported results of their investi- gation on sulfur-graphite synthetic samples (94/2252). A straightforward least squares fitting procedure proved unre- liable owing to the presence of backscattered Rh L lines from the X-ray tube anode emission and diffraction peaks. The PLS method fared better and when a data set was used to calibrate the system the mean difference in S content was 2.5%. An additional benefit of the PLS method was its ability to isolate some of the physical effects (diffraction etc.) causing spectrum artifacts which helped in their interpretation. Kis-Varga and Vegh (94/2271) reported a detailed study of the influence of in-sample scattering of fluorescence radiation on Si( Li) detec- tor line shapes.The so-called Compton shelf which is included in the Campbell model was investigated for thick pellets of low atomic number matrix (starch) spiked with elements in the 2 range 39-53. The contribution of the Compton shelf to the spectrum was determined by least squares fitting using a model based upon the Hypermet function adopted by Campbell. The Compton shelf increased with increasing line energy and was found to be smaller than that predicted by the SHAPE code for higher energy lines. Poor amplifier electronics resulted in high energy peak tailing due to pile-up. The intensity of the Compton shelf increased with detector aperture which was attributed to increased likelihood of large angle scattering but may have had a component arising from polarization effects as identified by the Chalmers group (93/1972).A fast stable recursive filtering algorithm was described by Gerasimov (93/3900) for the subtraction of background in gamma-ray spectra and Kalinin and Plotnikov (94/2385) presented a generalized scheme for the estimation of errors in peak area measurements. Wegrzynek and Holynska (94/1590) concluded unsurprisingly that ‘serious systematic errors can be introduced in the analysis of Pb if one does not take into account the presence of As’. This caveat is true for WD and EDXRF alike and the authors showed how using a simple overlap factor method the As Ka intensity could be determined by reference to the Pb L/? peak providing the latter is not overlapped.In addition to the requirement for no overlap on the Pb Lp peak it should be noted that the authors’ use of thin samples minimized the otherwise substantial error arising from changes in the Pb Lp La intra-series line ratio. The effect on KP Ka intra-series line ratios for Cr Ni and Ti intensities in ternary alloys of these materials was studied by Bhuinya and Padhi (93/3879) who attributed the observed increase over the pure metal values for Cr and Ti to transfer of 3d electrons to the Ni atoms rather than to inadequately compensated absorption effects. In a similar study of Cr Fe Mn and V compounds Raghavaiah et al. (94/2322) observed deviations of up to 9% in the KP:Ka intra-series line ratio for Mn and V.Energy shifts in K satellite lines of A1 and Si were studied by Murthy et al. (93/3849) using a high resolution WDXRF spectrometer. An empirical expression was presented which allowed the authors to predict the shift to within l e v for elements in the 2 range 11-19. Sanchez et al. (94/2382) described how multi- electron transition effects could affect XRF spectrometry and presented evidence from SRXRF measurements for the pres- ence of double K vacancies. The problem of determining back- ground in a multi-channel WDXRF spectrometer was investi- gated by Gunicheva et al. (94/2344). These authors calibrated the background using a specially-designed blank sample then compensated for differences between it and the unknowns to obtain accurate determinations of major elements in rocks.The large effect that particle sizes in the range 25-300 pm had on the accuracy of analysis of iron (111) oxide and copper sulfide samples was investigated by Abdunnabi and Wasilewska-Radwanska (94/1202). Theoretical predictions of accuracy of the emission-transmission calculation method for inhomogeneous and intermediate thickness samples were veri- fied using two different radioisotope sources in annular geometry. Finkelshtein et al. (93/3913) derived equations for the calculation of X-ray fluorescence and scattered primary radiation intensities from slurry samples. The space distribution of the particles was considered and the proposed calculation showed reasonable agreement with measured data. The theoretical details of a method for the quantitative analysis of inhomogeneousfy loaded thin film samples such as those obtained from air particulate samplers was described by Markowicz et a2.(94/2270). The method required knowledge of the sample distribution and the excitationdetection charac- teristics of the spectrometer before reliable data were produced. The simple process of spinning the sample would remove these requirements and probably result in more accurate data in any case. Couture (93/3801) studied the effect of sample transparency on WDXRF measurements of the high energy Sn Ka for trace Sn determinations in geological materials. The presented semi- empirical calibration procedure was based on a Compton correction procedure for mass absorption correction which was extended to account for transparency effects.The overall technique was numerically simple and allowed the use of synthetic calibration samples.JOURNAL OF ANALYTICAL ATOMIC SPECTROMETRY OCTOBER 1994 VOL. 9 295 R La Chance a name long associated with influence and alpha coeficients comprehensively reviewed (94/938) the underlying principles of influence coefficients and described the correction procedures which use this approach. Staats and Strieder (94/2357) anticipated the demise of classical analytical chemistry in their report citing the use of a primary calibration XRF method for the certification of sulfide- bearing mineral reference samples. Computer-designed syn- thetic calibration samples derived from pure oxides were used in conjunction with a WDXRF spectrometer. All samples were prepared as borate fusions and sulfur recoveries were found not to deviate significantly from 100%.Calibration samples typically four closely bracketed the unknown and removed the need for mathematical matrix correction. The overall method was capable of high accuracy yet retained traceability simplicity repeatability and a high degree of reliability for a variety of sulfide ores. A similar but less rigorous approach was taken by Lebedev (94/2342) in the analysis of low alloy steels. As with the fusion method the highly homogeneous nature of the synthetic calibration samples was a substantial benefit . The calibration maintenance and routine operation of an XRF spectrometer for the analysis of silicone paper coatings described by Price and Morrison (94/2324) allowed for the use of statistical process control techniques for the verification of instrument performance.The use of Monte Carlo (MC) simulations in XRF studies continues to grow. Members of the Antwerp group have used MC calculations to simulate the complete EDXRF spectrum from samples excited using an X-ray tube in conventional geometry (94/1001). The simulation took account of all exci- tation and scatter processes and showed good agreement with measured spectra taken from NIST Orchard Leaves and alloy samples. In an extension to this study (94/1571) the MC code was used to investigate the analytical qualities of currently operating and proposed mono-and poly-chromatic SR sources. The value of the code in simulating spectra for alloy calibration samples in which extreme absorption and enhancement effects occur was also demonstrated. An MC model was developed by Lankosz (94/2268) for the simulation of the particle size and edge effects encountered in the XRMF analysis of geologi- cal samples.The model assumed cylindrical grain cross sections and accounted for primary and secondary fluorescence events. Calculations were performed on a system of zinc sulfide grains in a calcium carbonate matrix and showed the strong depen- dence of detected intensities on the grain diameter and thick- ness and enabled optimization of XRMF measurement conditions. Lankosz et al. (93/3811) verified experimentally their MC computations on a system containing Ca and Fe in spherical glass particles of known composition and known diameters in the range 50-200pm.Sartori and Fernandez (93/3864) used an MC procedure to extend their work (based on the transport theory) on the contribution of multiple scattering events to the observed XRF spectrum to higher order terms. The work showed good agreement with exper- imental data and that higher (third and fourth) order inter- actions were only important for low 2 elements. The extensive knowledge and diverse problems available in the XRF field make it a likely candidate for the application of chernometvic techniques. Van den Bogaert et a!. (93/39 14) proposed an expert system for the determination of analytical strategy in a well-equipped laboratory for elemental analysis. The system was based upon a commercial expert kernel and was designed to guide a user through the decisions required to solve analytical problems using all the procedures and instrumentation available in the particular laboratory.In order to test the system an XRF problem domain was selected for the second stage of the project (93/3915). A single XRF expert in the knowledge gathering process was consulted which resulted in a site-specific system although the process was expected to be generally applicable. The XRF process focused on sample preparation and steered clear of instrumental factors of the analysis. Bos et al. (94/2255) used a back-projection neural network technique to model the difficult Cr-Fe-Ni ternary problem in which there are very large absorption- enhancement effects. The neural network results for Fe and Ni were found to be better than those given by the Rasberry- Heinrich empirical regression method (which was originally designed for the solution of this problem!) but depended on the availability of calibration samples in much the same way as a regression model.Following on their work on the limit dilution method from Part 1 (93/2754) Bosch-Reig et al. (94/27) investigated the correction of the diluent effect and the all-important question of selection of calibration samples. The authors presented the theoretical basis for their proposed empirical correction scheme in which the calibration lines were linearized and the accuracy of the procedure was confirmed by the successful analysis of mixtures of zirconia and iron (111) oxide in lithium metaborate fusion disc samples.Andersson and Olin (94/1011) in their report of the determination of Zn in fly ash and Fe in cement used an attenuation modifier (mercury (11) acetate) in their modified standard additions method which has been described previously (92/134) for the determination of Pb in domestic waste incinerator fly ash. The method of double dilution was described in useful detail by Leenanupan et al. (94/2372) who then chose the rather inappropriate (for XRF which is gener- ally rather good at analysing the metal as the solid) analysis of a cupro-nickel coin to demonstrate the utility of the method. Once digested the coin was subjected to the double dilution method yielding results of 25.65 &0.21% Ni and 72.75 & 0.50% Cu which were slightly worse than those from a conventional AAS determination.One can only presume that the matrix correction procedures available to the authors were inadequate for this analysis or that the coin was so inhomogeneous as to render analysis of the solid impracticable. Markowicz et al. (93/3912) have consolidated and extended their earlier publications (93/1738 and 93/1710) on the emis- sion-transmission technique for the correction of matrix effects. 2.6.1. Fundamental parameter (FP ) calculations De Boer et al. (93/3808) have assessed the accuracy of FP-MULTI a new FP calculation program suitable for both bulk and multilayer samples. In the analysis of NIST Cr-Fe-Ni Alloys the Fe and Ni results by FP were within YO relative of the certified values but the Cr data showed some positive bias because the FP model did not account for tertiary fluorescence effects.The accuracy deteriorated slightly if all the trace and minor elements were treated as a single element (Si) which was calculated by difference. Such a method is all but essential when assuming as was the case here that all analysed components sum to 100%. Accuracy for multilayer analysis was of the order of 2% relative. The authors used a chi-squared (x2) term in a particularly valuable way to deter- mine the intrinsic validity of the FP calculations for a particular sample and especially in the case of thin multilayer films where the composition of a layer may not be well known. The authors concluded with a very useful summary of the current position and future trends of FP analysis. An FP method for the analysis of samples of intermediate thickness was described by Wegrzynek et al.(93/3799,93/3790). A monochromatic radioisotope source was used and the mass absorption coefficients of the sample were determined using the emission-transmission method in order to simplify and increase the accuracy of the FP calculations and remove the need for an iterative calculation procedure. The enhancement correction was studied in detail and when data were corrected using the authors’ method errors of up to 50% relative were reduced to less than 10% relative in synthetic cellulose-based samples and ore reference samples. The convergence has been improved of the direct determi-296 R JOURNAL OF ANALYTICAL ATOMIC SPECTROMETRY OCTOBER 1994 VOL. 9 nation method which is a variant of the FP method for standardless analysis in which FP calculations are combined with measured instrument sensitivity factors (94/2363).An internal reference line not used as a conventional internal standard is used to simplify calculations and the speed of iteration was improved through the use of an improved algorithm for the determination of the starting concentration value and the reduction of numerical integrations required. Excellent accuracy was reported for the analysis of highly alloyed steels and the method was claimed to be suitable for rough samples. The difficult simultaneous determination by EDXRF spec- trometry of REE using L series emission lines excited using a Io9Cd radioisotope source was reported by Marco Parra et al. (94/2365).The AXIL program was optimized for the difficult task of spectrum processing and the light matrix option in QXAS was used for matrix correction using FP calculations. Ground mineral samples were diluted in 70% corn starch for analysis in order to comply with the requirements of the QXAS calculations. Sample mass absorption coefficients were carefully determined using the emission-transmission method and accu- racy of analysis was in the range 1.4-24% relative. The difficulty in correcting the severe overlaps in the spectrum was a substantial contributor to the achievable analytical precision of the method. Verkhovodov (93/3803) presented a complete theoretical derivation of an expression used for the calculation of X-ray line intensity in real samples and in real spectrometers with divergent excitation beams from X-ray tubes.Both side- and end-window X-ray tubes were considered and calculated inten- sities for first and second transition element K lines in lantha- num oxide were within 6-15% relative of their measured values. The same author (94/2367) gave details of the calcu- lations which he had implemented in his TEREN program for the theoretical calculation of L line intensities from the L1 L2 and L absorption edges. A theoretical approach for the description of the primary X-ray beam by means of the K fluorescence which it excites from a set of thin pure element targets was described by Martinez and Hidalgo (94/1583). The authors proposed that such a method removed the need for explicit knowledge of the actual spectral distribution of the X-ray tube which is difficult and inconvenient to measure.A simplified mathematical approach to the calculation of secondary fluorescence (enhancement) effects was proposed by Bui and colleagues (93/3806). The complete derivation was presented and the algorithm could be applied to samples of finite and infinite thickness. Results were presented for coinage metal samples where very large (up to a factor 2.5) enhancement effects are encountered and these showed accuracy of better than 10% relative for wide composition ranges. Rossiger (93/3871) investigated the validity for the analysis of thin film and bulk samples of a simplified calculation procedure for secondary fluorescence correction in which the isotropic distri- bution of the enhancing radiation was represented by a few selected discrete angles.Marchut (94/2355) applied an FP calculation method to the WDXRF determination of Ba Ca Fe S and Zn in hydro- carbons. When a solid calibration standard was used the accuracy was about 15% relative in the range 20-10000 pg g-' but this was improved substantially when an aqueous cali- bration sample was used. The poor results when using the solid calibration sample were attributed to particle size effects increased susceptibility to inaccuracy in the mass absorption coefficients and geometry differences between the solid and liquid sample cups. The rapid standardless analysis of combustible liquid waste samples using the UNIQUANT program was described by Wall (93/3181). Samples were analysed in their received state in a helium atmosphere and inaccuracy was typically 20% relative.Using the same program inaccuracy of 18-38 Yo relative was achieved for the semi-quantitative analysis of metal samples. Finally the use of FP calculations would be the worse were it not for the diligence of various groups in their determination of the parameters needed. Gerward (94/2374) reviewed the current state of knowledge and availability of mass absorption coefficients Balicinska-Church and McCammon (93/3863) par- ameterized mass absorption coefficients for 17 elements over the energy range 0.03-10 keV Orlic and Tang (93/3902) par- ameterized mass absorption coefficients for elements in the 2 range 1-40 over the energy range 0.1-1000 keV and Bhandal and Singh (94/1589) determined mass absorption coefficients for several multi-element materials over a wide energy range. Jahagirdar et al.(94/2360) measured total photoelectric cross sections for 52.4 keV photons in Ag Ba and Sn Kerur and colleagues (93/3805 94/2359) measured X-ray attenuation coefficients around the K edge of elements in the Z range 16-30 using Fe Ka excitation Garg et al. (93/3815 94/2312) measured XRF cross sections and yields for elements in the 2 range 41-52 using Mn Ka excitation and Rao et al. (93/3848) measured XRF L series cross sections and L series line intensity ratios for Au Ho Pr Pb and Yb using 23.62 and 24.68 keV excitation. Sanchez and Rubio (93/3800) used a method based on the tilt of the propagation plane to determine fluorescence yield and emission probability.The effectiveness of the pro- cedure was demonstrated in binary alloys of Cr Fe and Ni. Puri and colleagues (94/2364) calculated and parameterized L shell fluorescence yields and Coster-Kronig transition prob- abilities for elements in the 2 range 25-96. Garg et al. (93/3847) measured coherent and incoherent photon scattering cross- sections for the elements Au C Cu Fe Mo Pb Si V and Zn using 34.6 keV photons. The measured values were found to be higher than those calculated from theory. 2.6.2. Matrix correction using scattered radiation The dearth of papers specifically in this subject area may be a consequence of the fact that many FP calculation procedures now include the use of scatter corrections within them in much the same way that many regression programs have for a great many years.Certainly the practical applications of the methods are spread throughout XRF spectrometry. Karydas and Paradellis (94/1567) developed a model for the calculation of the enhancement of characteristic X-ray fluor- escence due to scattering in radioisotope EDXRF spectrometry. The authors' model was based upon the independent atomic model approximation and the scatter corrections were calcu- lated using exact differential scatter cross sections In a carbon matrix the scatter enhancement correction was as much as 40%. The ability of the proposed model accurately to correct for these large effects was demonstrated by the successful analysis of doped cellulose pellets. The effect of polarization on the Compton and Rayleigh scatter peaks was described in depth by Fernandez (94/1579).Intensity contributions from multiple scatter terms of the Compton and Rayleigh effects were computed using two models and the effect of polarization was shown to be up to 10% in some specific cases. The author considered it essential to include the polarization effect in models treating X-ray interactions with matter when using highly polarized sources such as those in SRXRF experiments. 2.7. Applications 2.7.1. Sample preparation Sample preparation is of course critical to successful XRF analysis and represents both a fundamental advantage of the technique (in terms of the simplicity with which many solid materials can be prepared for analysis) as well as an Achilles heel (in terms of sensitivity to particle size and surface irregu- larity effects). There are no important break-throughs to reportJOURNAL OF ANALYTICAL ATOMIC SPECTROMETRY OCTOBER 1994 VOL.9 297 R this year but rather some interesting innovations in extending the range and applicability of the XRF technique. Holynska (94/1565) offered a timely reminder of the principles of Sam- pling sample storage and sample preparation in the analysis of environmental biological and geological materials including procedures for the preconcentration and separation of elements after sample digestion all points aimed at improving the quality of results obtained by EDXRF. Zagorodnii (93/3171) confirmed the well-known fact that the fluorescence X-ray signal observed from powdered samples increases with decreas- ing particle size and increasing pressure of compacting pelleted samples (all to do with the packing density) investigated in this case in the XRF analysis of powdered ferro-alloys.In the ferrous metallurgical industry Ivanov et al. (94/2384) reported the virtues of using a re-melting sample preparation procedure. Baechmann et al. (93/3872) described an interesting adaptation of a thin film sample preparation technique tested in the analysis of Co304 and Y203 and a mixture of these with Ag20 and MnO,. Powdered samples (of particle size 1 pm) were mixed with a cold-setting polymer and a few drops of the mixture made into a thin film by squeezing between glass slides. Advantages claimed for this procedure were that it was particularly suitable for samples that are difficult to decompose (for analysis by other techniques) or could easily be contami- nated.Detection limits by EDXRF were between 6 and 20 ng of these elements and variations in film thickness and matrix effects were compensated by normalizing intensities to the Compton scatter peak. 2.7.2. Chemical analysis As in previous years a number of schemes of analysis have been described in which XRF is used to analyse samples that Table 1 CHEMICAL ANALYSIS have been chemically treated to preconcentrate and separate the analyte(s). There is a general (though not exclusive) trend that such procedures have been applied to water samples which in many laboratories would be analysed by alternative solution-orientated atomic spectroscopy techniques. In other cases techniques have been developed to extend the analytical range of XRF to lower concentrations in the analysis of low- level samples that could not otherwise be analysed directly by XRF.One separation procedure solvent extraction was covered in a wide ranging review by Kertes et al. (93/3670). More specific details of other procedures are listed in Table 1. 2.7.3. ThinJilms Tsuji and Hirokawa (94/2356) described in detail the technique of glancing incidence variable takeof angle XRF spectrometry. The technique is really an interesting variant of TXRF (or GIXA) in which the incident radiation strikes the quartz or Plexiglas sample support at fixed glancing incidence (just above the critical angle) and the fluorescence radiation is viewed both at 90 degrees and also at a very low take-off angle.The take-off angle was varied from 0 to 20 mrad to characterize thin layers of gold on the substrate. To compensate for the heavy collimation and low intensities a 7.5 kW molyb- denum rotating anode X-ray tube was used. The thickness of gold films up to 30nm could be determined by the method. The similar technique of refracted beam XRF was used by Shoji and Hirokawa (94/2327) for the characterization of chromium and gold thin layers on silicon wafers. The depth profiling of aluminium gallium arsenide thin films on a gallium arsenide substrate was reported by Ebel et al. (94/1582). These workers used an in-house secondary target Specific details Reference Anal yte Application Method Procedure V Environmental Extraction samples Cd Cu Pb Waters Zn Ex traction Trace Rocks and soils Solvent extraction elements Metals Waters Ag Ores Trace metals Water Co Cu Ni Ores U Th Water Ion exchange Precipitation Coprecipitation Sorption Selective Extract with 8-hydroxyquinoline at 150 "C from the solid.Palmitic acid was used to dissolve the V oxinate into a form which could be analysed directly by XRF Analytes extracted with higher aliphatic monocarboxylic acids at between 60-90°C for 3-5 min HF-HN0,-HCIO acid attack followed by solvent extraction into N7402 isobutyl methyl ketone from a KI-ascorbic acid aqueous phase DETATA filter containing aminocarboxylic groups. Dry filter then analyse directly Analyte with added Au carrier was precipitated from dissolved sample solution using hydrazine hydrate Analytes coprecipitated with aluminium hydroxide (A1 added as carrier) after adding ammonia solution to a filtered and acidified water sample Method involves sorption onto silica gel treated with an acetone solution of quinoxalin-2,3-dithiol Sample acidified with nitric acid and the analytes adsorbed onto a precipitate formed by adding 2,2,4-trimethylpentyl hydrogen methylphosphonate and ferric nitrate solution Sample pumped through a Quantitative extraction of 0.001-5% of V Analytes determined as a thin film on Mylar after back extracting into nitric acid.Detection limits reported as 0.04-0.6 pg g-' Elements determined Cd Co Cu Fe Mn Ni Pb V Zn to detection limits in the 1-3 pg 1-' range Filter membrane pasted on aluminium ring dried sprayed with ethanolic PVA and analysed by XRF Residue after filtering through membrane filter was dried and analysed directly.Not suited to the analysis of sea-water Mo tube excitation U detection limit was 0.06 (U) and 0.05 (Th) pg 1-' on precipitate prepared as pellet 941690 9411 602 9413841 941 1 199 9411 152 94J231 9313839 94/2 1 7298 R JOURNAL OF ANALYTICAL ATOMIC SPECTROMETRY OCTOBER 1994 VOL. 9 EDXRF system in which both the incidence and take-off angles could be adjusted. A complete theoretical description of the X-ray interactions in this interesting thin film system were presented. Using the equations which the authors derived the composition and thickness of the film could be determined reliably without any need for reference samples. Gries (93/3183) described a variable take-off angle XRF method which was used for depth profiling of zinc sulfide optical coatings on silica glass and phosphorus-doped layers on silicon wafers.A simple straightforward direct WDXRF method was described by Lehto et al. (94/1562) for the determination of the Pb concentration and film thickness in electroluminescent calcium sulfide thin films. A standard commercially available sequential WDXRF spectrometer was used for the measure- ment of Pb L S K and Ca K lines. A calibration procedure based on synthetic PVA standards covering the range 0-40% Pb yielded reliable results only up to about 15% Pb as a result of extreme matrix effects. The use of an FP program (Uniquant) was able to offer similar accuracy and could be applied reliably up to 40% Pb. A simple empirical calculation technique was used to determine the film thickness with high accuracy and excellent agreement with results from profilometry.Finally Vazquez and De Leyt (93/3856) reviewed (in Spanish) methods for the preparation and XRF analysis of both the composition and thickness of thin film samples. 2.7.4. Geological Despite the advent of more recently introduced atomic spec- trometry techniques XRF remains one of the corner stones of many geochemical laboratories and a technique into which innovations are still being introduced. Jackson et al. (94/771) reviewed the analysis of geological and inorganic materials by a comprehensive range of techniques including XRF. One useful contribution by Eastell and Willis (93/3798) examined the accuracy of three correction procedures based on influence/alpha coefficient procedures in the XRF analysis of samples prepared as glass discs using the low dilution fusion technique (2 1 flux sample).The work was instigated because the technique is capable of determining an extended range of trace as well as major elements and a complete set of alpha coefficients was not available in the literature. It was found that the multiple regression analysis and Norrish and Hutton methods gave poor accuracy when applied to rock types of widely varying composition whereas the computer program NBSGSC (based on fundamental parameters and the COLA algorithm) required variations in coefficients with varying mass fraction of the analyte elements to be taken into account. XRF is the standard technique for the determination ofRb and Sr in silicate rocks in support of mass spectrometric Rb Sr isotope studies.Measurements are conventionally made by WDXRF but the Open University group evaluated EDXRF in this application (94/299). The precision of EDXRF measure- ments was found to be comparable with that of WDXRF at concentrations below 20 pg 8-l with detection limits at the 1.5-2 pg 8-l level. This surprising result was attributed to the suppression in background caused by the use of a primary beam filter (127 pm silver foil) in EDXRF analysis using a silver anode X-ray tube. Above 20 pg g-l WDXRF became increasingly more precise owing to the superior sensitivity of that technique. Although XRF is generally considered to offer a best detec- tion limit in the 1-5 pg g-’ range two Chinese language contributions showed that this performance could be improved significantly using a chemical separation procedure.Li et al. (93/3841) digested rocks and soils using a conventional acid attack ( HF-HN03-HC104) followed by a solvent extraction into ‘N,40,’-isobutyl ketone (sic) from an aqueous phase which had been taken up in HCl and treated with a KI-ascorbic acid solution. The organic layer was back-extracted into nitric acid solution 1 ml of which was evaporated onto a Mylar film for XRF analysis. Detection limits for the elements Bi Cd Cu Mo Sn were reported to be 0.04-0.6 pg g-‘ between one and two orders of magnitude better than could be achieved by the direct analysis of solid samples. An acid attack (HF-HN03) was also used by Zeng and Chen (94/1224).An HCl solution of the sample was extracted with isobutyl methyl ketone to remove interferences and an HCl-NH4SCN solution of the residue in the aqueous phase passed through an ion-exchange column. After further treatment the eluate was evaporated onto a Mylar film for the XRF analysis of Co Ni Pb Zn Zr (using Sm as an internal standard) to detection limits of 0.11-0.29 pg 8-l. Cojocaru and Spiridon (94/1092) used an aqua regia digestion coupled to an anion-exchange separation for the XRF analysis of Au using a 238Pu excitation source. Measurements were made directly on the dried ion-exchange resin and recoveries of 97% were reported although it should be noted that the addition of HF is recommended in aqua regia procedures for some geological matrices to ensure the quantitative release of Au.The interfering elements Hg and Zn were removed by the ion-exchange procedure. The REE were determined after anion-exchange separation by Bauer-Wolf et al. (93/3377) after either acid attack of samples using a microwave oven or lithium metaborate fusion. Determinations were made both by ICP-AES (from HCl solution) and by XRF (after co-precipitation with rhodizonate and tannin and presentation as a thin film on a membrane filter). The preconcentration procedures were optimized and tested with radiotracers and accuracy evaluated by the analysis of reference materials. XRF was one of the techniques con- sidered in a Chinese language review by Yi (94/204) of tech- niques for the determination of the REE in geological samples.Nonferrous metallic ores were analysed by Cai et al. (94/1157) in another Chinese language paper using a fairly conventional lithium tetraborate-lithium metaborate-lithium nitrate fusion (3 1 flux sample) but in which the resultant glass bead was powdered and pressed to form a pellet for XRF analysis. Precision in the analysis of Cu Mo Pb W and Zn ores were reported to be better than 5% (relative). The XRF technique was used to analyse the industrial mineral mullite by Maruta et al. (93/3 169) using a low dilution lithium tetraborate-lithium metaborate fusion (3 1 flux sample) with lithium bromide added as a releasing agent. Con- versely montmorillonite was analysed by Yu (94/2261) samples (10 mg) being placed on ‘cellophane glue’ paper for direct XRF analysis.Hallet and Kyle (94/2230) used XRF together with INAA in a detailed characterization of major and trace elements in igneous and sedimentary refmnce materials from the Geological Survey of Japan. Finally in a link to the next section Bao and Shen (93/3822) in a Chinese language publication described the use of XRF in the determination of 17 elements in calcareous mollusc shell fossils. Samples collected as part of paleo-environmental and paleo-ecological research were crushed and pressed directly into pellets using a 3 g lining of boric acid in the mould. This sample (0.1 g) and an equal mass of cellulose binder were mixed and pressed into a disc 22 mm in diameter. This disc was supported on a filter paper or polyester film for XRF analysis.Detection limits except for the elements of low atomic number were in the range 1-10 pg g - l . 2.7.5. Environmental Although XRF is widely used in the analysis of environmental samples one of the problems with some sample types is the relatively low mass of material available for analysis. To overcome this difficulty in the characterization of environmental dusts Moore of the Pretoria Cement Company (93/3807) described a micro-bead method of sample preparation in which 0.07 g of dust was mixed with 0.35 g of a lithium fluoride- lithium tetraboride (20-80%) flux. After fusion in a muffleJOURNAL OF ANALYTICAL ATOMIC SPECTROMETRY OCTOBER 1994 VOL. 9 299 R furnace (7 min at 1050-1100°C) the melt was poured into a specially designed mold and subsequently analysed by XRF using a polyethylene X-ray resistant spacer in the normal sample holder.Precisions of 1-10% were reported for various analytes. An alternative approach in the analysis of atmospheric pollutants was followed by Matherny and Balgava (94/1561) who evaluated the optimization of thin-layer methods in the EDXRF analysis of inorganic atmospheric pollutants. Samples prepared as thin layers (as collected on a cellulose filter) and thin tablets (i.e. thin pressed pellets) were considered. The same group (94/1091) applied these techniques to a study of inorganic pollutants in gravitation urban dust sediments and residues formed after the evaporation of rain water. A further contribution to the thorny problem of preparing representative aerosol filter reference materials was made by Waetjen et al.(94/649) specifically ‘high volume’ uniform filter deposits for XRF analysis in the 10 ng cmP2 to pg cm-2 range. Particular attention was paid to checking the uniformity and measurements gave RSDs in the 515% range. An interesting study was undertaken by Kitsa et al. (94/2309) into the environmental hazard represented by Cr in respirable particles in soilsfrom contaminated sites in New Jersey (USA) where the soil is mixed with chromate slag. Re-suspended dust was collected on filters and analysed for 38 elements by XRF. These data were complemented by wet chemistry results for hexavalent Cr. Contaminated soil was also studied by Hans et al. (94/838) who compared the performance of XRF and AAS with German and Dutch regulatory standards.XRF was considered to offer benefits in terms of cost speed and quality of analyses with the advantage that the localization of environ- mental anomalies could be undertaken more effectively. Maruyoshi and Tamaki (94/144) considered the benefit of XRF in the primary screening of Cr Cu Pb and Zn in industrial waste water in comparison with TCP-AES and the ‘official’ method based on AAS. Other XRF applications were reported by Sharma and Patil (93/3521) (source identification of industrial aerosols in Bombay) Injuk et al. (94/1569) (variation in the composition of aerosols in air masses crossing the North Sea) Alsenoy et al. (94/928) (heavy metal pollution in sediments and sus- pended matter from the Belgium North Sea and Scheldt Estuary) Raoof and Al-Sahhaf (93/3823) (particulant pol- lutants in the air of Riyad Saudia Arabia) Harangozo et aE.(93/3824) (metals in soils near the Slovakian highways) and Marques et al. (94/1563) (pollution monitoring in the Lisbon region using the leaves of the evergreen shrub Nerium Oleander). In addition details of environmental applications are covered in the latest ASU review on environmental analysis by Cresser et al. (93/3417). 2.7.6. Archaeological and forensic The particular advantages of XRF in the analysis of archaeol- ogical artefacts which cannot be destroyed or altered because of their importance in historical studies are illustrated in an important contribution by Giauque et al. (93/3810) who used the technique in a provenancing study of obsidian artefacts. The method was applicable to both thick and thin samples with wide variations in size and shape.Thirteen trace and three major elements were determined from one spectrum using a 1000 s count time and average precisions in the determination of the most sensitive elements Ba Fe Rb Sr Y and Zr was 2.3%. Mean concentrations for these elements (except Y) deviated by only 2% from published values determined by NAA and XRF. An XRF study of marble by Yap (94/991) showed that radioisotope excitation (55Fe Io9Cd and 241Am sources) could be used reliably to determine As Cu Fe Mn Sr Y and Zn and that the use of Fe Y and Sr in a triangular plot could be used to distinguish between pieces of marble from Portugal and Thailand. In a review of the use of XRF in the analysis of works of art in Rome Cesareo et al.(93/3875) emphasized the usefulness of EDXRF instruments incorporating mercuric iodide room temperature semiconductor detectors applications of which are almost certain to feature strongly in future reviews. In a similar area of application Schreiner et al. (94/3) described a new EDXRF instrument designed for the pixel-by-pixel non- destructive analysis of objects of art and archaeology. The elemental distributions of Ag Au Ca Cu Fe Hg and Pb were mapped to identify the distribution of specified pigments. Punch-marked silver coins (5th to 2nd Century BCE) were analysed by WDXRF by Rao et al. (94/2308). Results by SEM X-ray analysis agreed well with WDXRF data which demon- strated the method of manufacture of the coins. Ancient gold coins were analysed by EDXRF and PIXE by Araujo et al.(94/2379) results between the two techniques agreeing for samples with Au >94% differences at lower gold contents being attributed to surface enrichment effects. In the only forensic application featured here Misra et al. (93/3843) used source excited EDXRF to analyse chips/films of nail varnish to evaluate the potential of the method in forensic applications. 2.7.7. Industrial Industrial applications of XRF continue to feature strongly in the research literature an indication of the important contri- bution made by XRF in this area of activity and the fact that there is a continuing interest in diverse and innovative developments. An interesting adaptation used in the XRF analysis of oil was described in a Chinese-language publication by Liu et al.(94/1060 94/1068) in which the oil sample (1.3 g) was mixed with MgO (1.1 g) and heated at 270°C with stirring for 20 min. After pulverising the residue was compressed into a powder pellet for the EDXRF analysis of Cr Cu Fe Mn Ni V and Zn to detection limits in the range 0.6-11.5 pg 8 - l . An alternative approach was followed by Yu (94/992) in which a mixture of crude oil (0.5-1 g) and IBMK was melted and then three 50 p1 portions spotted into a wax ring on a filter paper. The elements Cu Mn Ni and V were determined with detection limits in the 0.14-0.37 pg range. Gimeno Adelantado et al. (94/698) analysed the toxic metals AS’” Ba Cd CrII’ Hg Pb Sb”’ in plastic components used in making toys using an XRF method that involved a detailed chemical separation.The sample was fused with sodium hydroxide using sodium nitrate as an additional oxidant. The elements of interest were precipitated with 1 % sodium diethyl- dithiocarbamate-0.1 ?Lo sodium rhodizonate-Fe3 solution and after filtering and drying the precipitate was analysed directly by XRF. Tetrafluoroethylene/perfuoro( propylvinyl ether) plas- tics were analysed for the co-monomer content by Yonemori et al. (93/3170) using an XRF method in which a multilayer (2d = 8.06 nm) dispersing element and a gas flow proportional counter (0.6 pm polyester film window) were used to determine the 0 content. Calibration lines were linear from 0.1-2% 0 corresponding to 1-10% of the co-monomer the method being rapid and non-destructive. Controversy has raged over alleged interference effects which Hurley (94/2272 two comments) claimed affected the results of Al-Ammar (94/2272) in the analysis of impurities in nuclear- grade uranium compounds.In an independent investigation by Qadeer et al. (94/1592) solutions of uranium were analysed by XRF for the elements Cd Gd Hf and Hg. Sample solutions contained in polyethylene bottles were excited directly and calibration undertaken using matched standard solutions although it should be noted that polyethylene bottles are not considered to be suitable for the storage of mercury solutions. An interesting patent application was made by Brune (93/3842) to detect corrosion on metal surfaces by comparing the EDXRF analysis of corroded areas with adjacent non-300 R JOURNAL OF ANALYTICAL ATOMIC SPECTROMETRY OCTOBER 1994 VOL.9 corroded areas. In the case of iron-based alloys corrosion caused changes in the concentration of elements such as Cr Fe Mo Ni. For example lower concentrations of Fe in rusty areas were caused by the dilution effect of oxide and selective release mechanisms from the alloy surface which resulted in other elements such as Cr Mo and Ni becoming enriched in the metal surface. A rapid XRF method for the analysis offerroalloys was described in a Korean language contribution by Moon et al. (93/3891). Of interest was a comparison reported by the authors on sample preparation methods. Briquetting methods were found to be most satisfactory; fusion to form a glass bead caused problems with alloy damage to the platinum crucible and remelting the sample caused compositional changes during the remelting process. Ladisich and Wobrauschek (94/2269) investigated the optimum excitation conditions for the determi- nation of A1 in Fe-A1 binary alloys and used a secondary target configuration to achieve selective excitation of A1 to a precision of kO.2 at.% (for 40-80 at.%Al) and an extrapolated detection limit of 0.04 at.%Al (0.08 at.%).These procedures were also applied in the analysis of Cu-Fe binary alloys. Rare-earth element oxides were analysed by Chen (94/1226) using an XRF procedure designed to minimize matrix effects as follows the sample (50 mg) was fused with lithium tetrabo- rate-lithium fluoride then pulverized and compressed into a thin disc (27 mm diameter x 0.3 mm thick).The Lucas-Tooth and Pyne method was used to correct matrix effects (which were reduced by the dilution during sample preparation) and the method offered detection limits of 14-83 pg g-'. A complete method was described for the analysis of clinker by Bonvin et al. (94/1196) using a 'total cement analyser'. Instrumentation comprised an X-ray diffraction channel inte- grated with an XRF spectrometer. The X-ray diffraction chan- nel was used to measure the free lime content on compressed powder pellets under the same excitation conditions as were used for XRF analysis. As a contribution in the mining and mineral extraction industries Bazykina et a!. (94/1015) described a method for the determination of the elements Fe Pb and Zn in lead concentrates by XRF using a chemical separation procedure involving an acid attack followed by the precipitation of analytes from a solution to which calcium nitrate and sodium hydrogen phosphate had been added.The analysis of drilling fluids was the subject of a patent application by Jones et al. (93/3844). The method was designed for the analysis of well fluids containing solid or liquid components dissolved or dispersed in a liquid carrier using either infra red or XRF as the determination stage. There has been continuing interest in the analysis of ceramics and semiconductor materials during the current review period. In particular Kotrly et al. (93/2725) developed an XRF method for analysing small areas on the natural cleavage faces of layered single crystals. The method was devised as a screen- ing technique prior to the selection of samples for physical measurements and was evaluated by the XRF determination of In in Bi(,-x,In,Se single crystals in comparison with wet chemical determinations of In.Piezo-ceramics based on lead zirconate were analysed using an XRF technique by Kvaratskheli et al. (94/1166). Powdered samples were fused with silica and lithium metaborate in the ratio 3:3:22. After fusion at 950 "C (15 min) the melt was poured into a vitreous carbon crucible and cast with centrifugation. A wide range of elements was determined. For a more comprehensive review of other contemporary applications in the analysis of industrial materials the reader is advised to consult the latest ASU review in this area ( 9 3/ 3400). 2.7.8.Clinical and biological The non-destructive capabilities of XRF give the technique a unique advantage for in viuo measurements and the technique is now widely used for the in viuo analysis of Pb in bone to evaluate the body burden of this element. Thus Erkkila et a!. (94/2306) used this technique to analyse bone lead concen- trations in the tibia wrist sternum and calcaneus of current and retired workers from two lead-acid battery factories. Production workers office staff and a control group were studied. The most precise estimates of the cortical bone content were obtained from the tibia rather than the ulna. The most precise measurement of the lead content of trabecular bone were obtained from the calcaneus rather than the sternum. Furthermore groups of individuals occupationally exposed to lead were better distinguished on the basis of the calculated skeletal lead burdens rather than lead blood concentrations suggesting that the the latter measurement might seriously underestimate lead body burden.This study also indicated that tibial lead increased consistently as a function of both intensity and duration of exposure. Calcaneal lead was strongly dependent on the intensity rather than duration of exposure. The general advantages of using the XRF technique to deter- mine the Pb contents of bone in viuo were considered by Todd et al. (93/3896). Slatkin et al. (93/3874) considered the require- ments of a report by the International Commission on Radiation Protection (ICRP 60) one of the recommendations of which was that the dose from any radiation exposure must be minimized. Using published absorption data the dose during the non-invasive assessment of tibial bone lead by L line EDXRF was 1.0 pSv for a 5 year old child and about 0.3 pSv for an adult as compared with figures of 1.7 and 0.7 pSv calculated from older ICRP data for tissue weighting functions.The importance of in uivo XRF measurements in contributing data to epidemiological studies especially chronic effects were considered by Skerfving and Nilsson (93/3897) in assessing the accumulated body burden of Cd in kidney and Pb in finger bone tibia or calcaneus. Mori et al. (93/3818) described a procedure using XRF for measuring regional blood flow in the heart and other organs. These workers demonstrated the stability of eight kinds of heavy-element loaded microspheres used in conjunction with a high sensitivity WDXRF instrument.Results were validated by comparison with duplicate flow measurements made with radioactive microspheres in measurements on acute and chronic dogs. An SR EDXRF system was also used and was found to offer higher S/B ratios for Zr-loaded microspheres with the potential of analysing smaller regions (<20 mg) than the alternative radioactive method. Tartari et al. (94/1585) investigated the 'SHAPE' code for simulating ED-XRF spectra with particular reference to the determination of trace Cd in soft tissue using 241Am (59.54 keV) excitation and various irradiation geometries. Results of this code which can be run on a PC agreed well with those obtained by Monte Carlo simulation.The analysis ofhair samples has been proposed as a means of assessing the internal body burdens of environmental pol- lutants but the interpretation of these measurements has been questioned by some workers. The results of an IAEA research programme were described by Cortes et al. (94/676) in correlat- ing the As Cd Cu Hg Pb Se and Zn contents of five tissue samples taken from corpses with the analysis of hair using XRF as well as NAA and PIXE. Some correlation was found for As Hg and Se but not for Cu or Zn. Other applications in the analysis of hair are considered in the present review in the X-ray microprobe section. XRF was used by Allwood et al. (94/2310) in nutritional studies to assess the trace element compatibility in total parenteral nutrition infusions.The elements determined in derived precipitates were Ca Cu Fe P and Se. The analysis of plant tissue for non-mobile elements can supplement or indeed replace soil analysis when assessing the soil's ability to supply nutrients to next year's crop. Frank et al. (93/3820) used an XRF method to compare results forJOURNAL OF ANALYTICAL ATOMIC SPECTROMETRY OCTOBER 1994 VOL. 9 301 R elements including S Ca Cu and Mn obtained from both pelleted leaf samples and leaf discs prepared from dried and ground leaves collected from corn plants. Leaf punch discs appeared to provide adequate plant nutrient values for diag- nostic purposes. Plant tissue in this case the leaves and roots of barley seedlings were anaIysed by Calliari et al. (94/1587) to investigate the effect of Cr (and other bio-essential elements including Cu Fe Mn and Zn) in compost contaminated with sewage sludge from tanneries.A secondary target EDXRF technique was used in conjunction with a microwave acid digestion procedure in which the sample was finally presented for analysis as a thin film. Sample preparation in the analysis of marine algal communi- ties was addressed by Boman et al. (94/1574) who submerged algal samples (microbial periphyton communities) mounted on glass substrates in marine waters off the West coast of Sweden. Phytoplankton from the same marine waters were collected on polycarbonate filters. The preferred sample presentation scheme involved a rapid wash of algal communities with distilled water to remove interstitial sea-water with the sub- sequent transfer of the sample onto a foil prior to EDXRF analysis using a secondary target configuration.As in previous reviews there is continued interest in the XRF analysis of S in rapeseed. Wagstaffe et al. (93/2733) described the certification of three rapeseed reference materials by ICP-AES ID-MS and ion chromatography and established their suitability for calibrating an XRF spectrometer. In a second contribution from the same group Schnug et al. (93/3890) described the calibration of an XRF spectrometer for the determination of total glucosinolates using three rap- eseed reference materials. Other contributions in which XRF was applied in biological applications include the determination of Cu Fe Mn and Zn in soybean (Custo et al.93/3489) and the determination of Se in animal tissue (Tariq and Preiss 93/3819) using a sample preparation procedure that involved a solvent extraction after digestion of the sample. 2.7.9. Chemical efects This year only a relatively small number of chemical state studies are available for review. Of these Kucukonder et al. (93/3827) investigated changes in the KP Ka ratio measured from a series of octahedral and tetrahedral coordination com- pounds of Cr Cu and Mn. A Ge(Li) detector was used and the KP:Ka ratio was found in general to be larger for tetrahedral complexes. Kawai et al. (94/978) measured the Cu Lp La ratio in various copper compounds and alloys and found that the intensity ratio changed with the copper concen- tration as well as the coordination state in the order Cd' > Cu' > Cu metal.Differences between metals and alloys were attributed to self absorption effects and differences between alloys and Cu coordination states to differences in the L2-L,V Coster-Kronig transition probabilities. Chromium Kcr and Ka emission lines were studied by Sarode (94/2358) in various chromium compounds and chemical shifts spin-orbit splitting effects asymmetry indices and half- widths were correlated with Cr oxidation state. In a Russian language contribution Borovoi et al. (93/3904) measured changes in the intensity of the Ka, group of emission lines in Sc and Sc203. Finally Leiro et al. (94/1576) studied the linearly polarized Fe K absorption and Fe L emission spectra of Fe,O and compared the measurements with the results of multiplet theory and cluster calculations.LOCATION OF REFERENCES The full list of references cited in this Update have been published as follows 93/2711-9313353 J. Anal. At. Spectrom. 1993 8(7) 313R-336R. 9313354-9314131 J. Anal. At. Spectrom. 1993 8(8) 377R-404R. 9411-941614 J. Anal. At. Spectrom. 1994,9( l) 1R-23R. 94/615-941960 J. Anal. At. Spectrom 1994 9(2) 73R-85R. 941961-9411264 J. Anal. At. Spectrom. 1994,9(4) 135R-146R. 9411265-9411830 J. Anal. At. Spectrom. 1994 9( 5) 149R-169R. 9411831-9412175 J . Anal. At. Spectrom. 1994 9(6) 189R-200R. 9412176-9412412 J. Anal. At. Spectrom. 1994 9(7) 203R-212R. 941241 3-94/2867 J. Anal. At. Spectrom. 1994 9( 8 ) 249R-265R. Abbreviated forms of the literature references quoted (excluding those to Conference Proceedings) are given on the following pages for the convenience of the readers.The full references names and addresses of the authors and details of the Conference presentations can be found in the appropriate issues of JAAS cited above. Abbreviated List of References Cited in Update 921134 Talanta 1991 38 385. 9213007 Nucl. Instrum. Methods Phys. Res. Sect. A 1991 308 318. 93/1081 Spectrochim Acta Part €3 1992,47,983.93/1642 Spectrochim. Acta Part B 1992 47 1123. 9311710 J. Radioanal. Nucl. Chem. 1992 158 409. 9311738 Appl. Radiat. Isot. 1992,43,777.93/1747 Appl. Radiat. Isot. 1992 43 767. 93/1943 Anal. Chem. 1992 64 2444. 9311972 X. Ray Spectrom. 1991 20 325. 9312070 J. Anal. At. Spectrom. 1992 7 993. 9312163 Anal. Chem. 1992 64 2253. 9312554 Proc.Aust. Conf. Nucl. Tech. Anal. 1991,7,9.93/2725 AnaIyst 1993,118 79.9312733 Fresenius' J. Anal. Chem. 1992 344 1.9312754 Fresenius' J. Anal. Chem. 1992,344,22.93/2777 Appl. Radiat. Isot. 1992 43 847. 9312869 Nucl. Instrum. Methods Phys. Res. Sect. A 1992 A317 383. 93/2869 Nucl. Instrum. Methods Phys. Res. Sect. A 1992 A317 383. 9312877 A p p l . Opt. 1993 32 785. 9312878 Appl. Opt. 1993 32 861. 9312891 Anal. Chem. 1992 64 2942. 9312892 Anal. Chem. 1992 64 2945. 9312893 Anal. Chem. 1992 64 2958. 9312898 Anal. Chem. 1993,65,97.93/2900 Anal. Chim. Acta 1992,269 249. 9312904 Anal. Sci. Technol. 1990 3 215. 9312908 Anal. Sci. Technol. 1992 5 169. 93/2909 Anal. Sci. Technol. 1992 5 229. 9312914 Appl. Radiat. Isot. 1992 43 1420. 9312917 Biol. Mass Spectrom. 1992 21 473.9312918 Biol. Mass Spectrom. 1992 21 486. 93f2921 Bunseki Kagaku 1992 41 425. 9312924 Chem. Geol. 1992 101 81. 9312925 Chem. Geol.,302 R JOURNAL OF ANALYTICAL ATOMIC SPECTROMETRY OCTOBER 1994 VOL. 9 1992 101 157.9312931 Clin. Chem. (Washington D. C.) 1992 38 2197. 9312933 Elem. Anal. Coal Its By-Prod. Int. Conf. Proc. 2nd 1991 (Pub. 1992) 17.9312937 Environ. Sci. Technol. 1993 27 172. 9312938 Eur. Mater. Res. SOC. Monogr. 1992 4 167. 9312939 Fenxi Shiyanshi 1992 11(3) 51. 9312940 Fresenius’ J . Anal. Chem. 1992 344 109. 9312941 Fresenius’ J. Anal. Chem. 1992 344 279. 9312942 Geochim. Cosmochim. Acta 1992 56 3583. 9312945 Int. J. Mass Spectrom. Ion Processes 1992 116 183. 9312946 Int. J. Mass Spectrom. Ion Processes 1992 116 219. 9312949 Int. J. Mass Spectrom.Ion Processes 1992,118,477.9312950 Int. J. Mass Spectrom. Ion Processes 1992 118 519. 9312951 Int. J. Mass Spectrom. Ion Processes 1992 118 575. 9312953 Int. J. Mass Spectrom. Ion Processes 1992 120 163. 9312954 J. AOAC Int. 1992 75 725. 9312957 J. Chromatogr. Sci. 1992 30 427. 9312960 J. Radio-anal. Nucl. Chem. 1992 161 443. 9312964 Kernforschungszent. Karlsruhe [Ber.] KfK 1992 KfK 5017 9 1 pp. 9312964 Kernforschungszent. Karlsruhe [ Ber.] KfK 1992 KfK 5017 91 pp. 9312965 Mantle Meteorites 1990 71. 9312967 Microsc. Oxid. Proc. Int. Conf. lst 1990 (Pub. 1991) 10. 9312968 Microsc. Oxid. Proc. Int. Conf. lst 1990 (Pub. 1991) 36. 9312970 Microsc. Oxid. Proc. Int. Conf. lst 1990 (Pub. 1991) 311.9312973 Phys. World 1992,5( 12) 28.9312976 Radiocarbon 1992 34 271. 9312977 Radiocarbon Four Decades [Pap.Conf.] 1992 214. 9312979 Radiochim. Acta 1992 57 15. 9312980 Radiochim. Acta 1992 57 21. 9312981 Rev. Sci. Instrum. 1992 63 5299. 9312985 Soil Sci. SOC. Am. J. 1992 56 1151. 9312986 Soil Sci. SOC. Am. J. 1992 56 1185. 9312991 Surf. Interface Anal. 1992 18 765. 9312992 Surf. Interface Anal. 1992 18 777. 9312994 Tongweisu 1991 4 181. 9312996 Z. Lebensm.- Unters. Forsch. 1992 195 307. 9312997 Zh. Anal. Khim. 1992 47 820. 9313087 Talanta 1992 39 737. 9313094 Chem. Geol. 1993,103,45.93/3095 Chem. Geol. 1993 103,73.93/3097 Chem. Geol. 1993,104,175.93/3099 Geochim. Cosmochim. Acta 1992,56,1561.93/3101 Geochim. Cosmochim. Acta 1992,56,4099.93/3102 Geochim. Cosmochim. Acta 1992 57 475. 9313126 Fenxi Shiyanshi 1992 11(6) 39. 9313168 Bunseki Kagaku 1992,41,545.93/3169 Bunseki Kagaku 1992 41 589.9313170 Bunseki Kagaku 1992 41 593. 9313171 Zh. Anal. Khim. 1992 47 1428. 9313181 Lab. Equip. Dig. 1992 30(3) 15. 9313183 Mikrochim. Acta 1992 107 117. 9313184 Marcel Dekker New York NY USA. 9313219 J. Anal. At. Spectrom. 1993 8 499. 9313223 J. Anal. At. Spectrom. 1993 8 551. 9313224 J. Anal. At. Spectrom. 1993 8 557. 93/3225 J . Anal. At. Spectrom. 1993 8 565. 9313226 J. Anal. At. Spectrom. 1993 8 571. 9313231 J. Anal. At. Spectrom. 1993 8 599. 9313250 J. Anal. At. Spectrom. 1992 7 279R. 9313361 Anal. Sci. 1993 9 137. 9313364 Spectrochim. Acta Part B 1992 47 809. 9313365 Spectrochim. Acta Part B 1992 47 E1621.9313368 Spectrochim. Acta Part B 1993,48,39.93/3373 Spectrochim. Actu Part B 1993 48 11 1. 9313374 Spectrochim. Acta Part B 1993 48 255.9313375 Fresenius’ J. Anal. Chem. 1992 344 326. 9313376 Fresenius’ J. Anal. Chem. 1992 344 353. 9313377 Talanta 1993 40 9. 9313384 Bunseki Kagaku 1993 42 21. 9313385 J. Anal. At. Spectrom. 1992 7 1157. 9313387 J. Anal. At. Spectrom. 1992 7 1183. 9313388 J. Anal. At. Spectrom. 1992 7 1195. 9313389 J. Anal. At. Spectrom. 1992 7 1201. 9313400 J. Anal. At. Spectrom. 1992 7 349R. 9313401 J . Anal. At. Spectrom. 1993,8 1.9313402 J. Anal. At. Spectrom. 1993 8 19. 9313403 J. Anal. At. Spectrom. 1993 8 25. 9313404 J. Anal. At. Spectrom. 1993 8 35.9313417 J. Anal. At. Spectrom. 1993 8 1R. 9313421 Appl. Spectrosc. 1992 46 1762. 9313427 Appl. Spectrosc. 1993 47 243. 9313431 Anal. Chem. 1993 65 660. 9313433 Anal. Chem. 1993 65 735. 9313441 Anal. Chem.1993 65 1186. 9313442 Anal. Chem. 1993 65 1419. 9313443 Anal. Chem. 1993 65 1411. 9313445 Anal. Chim. Acta 1993 274 327. 9313447 Fresenius’ J. Anal. Chem. 1993 345 310.9313448 Fresenius’ J. Anal. Chem. 1993 345 330. 9313451 Spectrochim. Acta Part B 1993 48 397. 9313453 Adv. Mater. Processes 1992 141 20. 9313472 Appl. Laser Spectrosc. 1992 365. 9313474 Appl. Surf. Sci. 1992 59 135. 9313475 Appl. Surf. Sci. 1993 63 62. 9313476 Appl. Surf. Sci. 1993 63 75. 9313482 Biochim. Biophys. Acta 1992 1156 57. 9313488 Bull. Korean Chem. SOC. 1992 13 620. 9313489 Can. J. Appl. Spectrosc. 1992 37 176. 9313496 Chem. Geol. 1993 103( 1-4) 283. 93/3501 Chin. J. Geochem. 1991 10 372. 9313502 Chin. Sci. Bull. 1992 37 915. 9313505 Conf. Rec. IEEE Part. Accel. Conf. 14th 1991 2 878.9313509 Commun. Soil Sci. Plant Anal. 1992 23 2781. 9313515 Crit. Rev. Anal. Chem. 1992 23 355. 9313519 Environ. Part. 1992 1 107. 9313521 Environ. Technol. 1992 13 1043. 9313544 Fresenius’ J. Anal. Chem. 1993,345,217.9313569 He Huaxue Yu Fangshe Huaxue 1992 14 126. 9313571 Hist. Metall. 1991 25 78. 9313579 IEEE Trans. Electron Devices 1993 40 82. 9313581 Int. J. Mass Spectrom. Ion Processes 1992 121 111. 9313582 Int. J. Mass Spectrom. Ion Processes 1992 121 201. 9313583 Int. J. Mass Spectrom. Ion Processes 1993 123 1. 9313584 Int. J. Mass Spectrom. Ion Processes 1993 123 29. 9313586 Int. J. Mass Spectrom. Ion Processes 1993 123 117. 9313587 Int. J. Mass Spectrom. Ion Processes 1993 123 149. 9313591 J. AOAC Int. 1992 75 1107. 9313593 J. Am. SOC. Mass Spectrom.1992 3 742. 9313594 J. Am. SOC. Mass Spectrom. 1993 4 16. 9313595 J. Am. SOC. Muss Spectrom. 1993 4 28. 9313596 J. Am. SOC. Mass Spectrom. 1993 4 47. 9313597 J. Anal. Toxicol. 1993 17 5. 9313599 J. Appl. Phys. 1992 72 5898. 93/3605 J. Chromatogr. Biomed. Appl. 1992 583 35. 9313619 J. Trace Microprobe Tech. 1992 10 191. 9313621 J. Vuc. Sci. Technol. A 1993 11 125. 9313623 Jpn. J. Appl. Phys. Part 1 1993 32 982. 9313624 Kernforschungszent. Karlsruhe [Ber.] KfK 1992 KfK 4981 86. 9313635 Mass Spectrom. Rev. 1992 11,495. 9313639 Microchem. J. 1992 46 302. 9313645 Mikrochim. Acta Suppl. 1992 12 187. 9313647 Neftepererab. Neftekhim. (Kiev) 1991 41 62. 9313650 Nippon Genshiryoku Kenkyusho [Rep.] JAERI-M 1992 JAERI-M 92-036,12. 9313651 Nucl. Instrum. Methods Phys.Res. Sect. B 1992,72,442. 9313653 Nucl. Instrum. Methods Phys. Res. Sect. B 1993 73 214. 9313654 Nucl. Instrum. Methods Phys. Res. Sect. B 1993 73 221. 9313662 Phys. Mag. 1992 14 49. 9313663 Phys. Methods Chem. (2nd Ed.) 1993,9A 321.9313664 Phys. Rev. C. Nucl. Phys. 1992,46 R2140. 9313665 Phys. Scr. 1992 46 581. 9313667 Plant Cell Enuiron. 1992 15 1091. 9313670 Princ. Pract. Solvent Extr. 1992 51 1. 9313681 Rare Nucl. Processes Proc. Europhys. Conf. Nucl. Phys. 14th 1990 (Pub. 1992) 93. 9313686 Rev. Sci. Instrum. 1993 64 265. 9313695 Second. Ion Mass Spectrom. SIMS 8 Proc. Int. Conf. 8th 1991 123. 9313696 Second. Ion Mass Spectrom. SIMS 8 Proc. Int. Conf. 8th 1991 131. 9313697 Second. Ion Mass Spectrom. SIMS 8 Proc. Int. Conf. 8th 1991 135. 9313700 Second.Ion Mass Spectrom. SIMS 8 Proc. Int. Conf. 8th 1991 163. 9313704 Shitsuryo Bunseki 1992 40 217. 9313713 Spectrochim. Acta Rev. 1993 15 1. 9313716 Surf. Interface Anal. 1992 18 824. 9313717 Surf. Interface Anal. 1993,20 53. 9313718 Surf. Interface Anal. 1993 20 69. 9313720 Surf. Sci. 1993 280 208. 9313724 Second. Ion Mass Spectrom. SIMS 8 Proc. Int. Conf. 8th 1991 95. 9313725 Second. Ion Mass Spectrom. SIMS 8 Proc. Int. Conf. 8th 1991 103. 9313732 Second. Ion Mass Spectrom. SIMS 8 Proc. Int. Conf. 8th 1991 179. 9313741 Second. Ion Mass Spectrom. SIMS 8 Proc. Int. Conf. 8th 1991 363. 9313742 Second. Ion Mass Spectrom. SIMS 8 Proc. Int. Conf. 8th 1991 367. 9313744 Second. Ion Mass Spectrom. SIMS 8 Proc. Int. Conf. 8th 1991 403. 9313749 Second. Ion Mass Spectrom. SIMS 8 Proc.Int. Conf. 8th 1991 487. 9313752 Second. Ion Mass Spectrom. SIMS 8 Proc. Int. Conf. 8th 1991 575. 9313755 Second. Ion Mass Spectrom. SIMS 8 Proc. Int. Conf. 8th 1991 637. 9313757 Second. Ion Mass Spectrom. SIMS 8 Proc. Int. Conf. 8th 1991 715. 9313765 Thin Solid Films 1992 220 197. 93/3768 Tongweisu 1992 5( l ) 8. 9313780 Transuranium Elem. Symp. 1990 (Pub. 1992) 126. 9313781 Transuranium Elem. Symp. 1990 (Pub. 1992) 159. 9313782 Vacuum 1992 43(5-7) 481. 9313788 Yuanzi Yu Fenzi Wuli Xuebao 1992,9 2346. 93/3790 Yuanzineng Kexue Jishu 1991 25(3) 45 61. 93/3791JOURNAL OF ANALYTICAL ATOMIC SPECTROMETRY OCTOBER 1994 VOL. 9 303 R Yuanzineng Kexue Jishu 1991 25( 3) 68. 9313792 Yuanzineng Kexue Jishu 1991 25(4) 29. 9313797 X-ray Spectrom. 1993 22 65. 9313798 X-ray Spectrom.1993 22 71. 9313799 X-ray Spectrom. 1993 22 80. 9313800 X-ray Spectrom. 1993,22 89. 9313801 X-ray Spectrom. 1993 22 92. 9313802 X-ray Spectrom. 1993 22 97. 9313803 X-ray Spectrom. 1993 22 103. 9313804 X-ray Spectrom. 1993 22 114. 9313805 X-ray Spectrom. 1993 22 13. 9313806 X-ray Spectrom. 1993 22 17. 9313807 X-ray Spectrom. 1993 22 28. 9313808 X-ray Spectrom. 1993 22 33. 9313809 X-ray Spectrom. 1993 22 39. 9313810 X-ray Spectrom. 1993 22 44. 9313811 X-ray Spectrom. 1993 22 54. 9313812 X-ray Spectrom. 1993 22 58. 9313813 X-ray Spectrom. 1993,22,000.93/3814 Nucl. Instrum. Methods Phys. Res. 1993 A324 61 1. 9313815 Nucl. Instrum. Methods Phys. Res. 1992 B72 147. 9313816 Phys. Rev. A 1992 46 237. 931C3817 Presented at the IEEE-1992 Nucl. Science Symposium Orlando (USA) Oct.27th-31th 1992. 9313818 Am. J. Physiol. 1992 263(6 Pt. 2) H1946. 9313819 J. Radioanal. Nucl. Chem. 1992 163 355. 9313820 Commun. Soil Sci. Plant Anal. 1992 23 2415. 9313822 Fenxi Huaxue 1992 20 688. 9313823 Atmos. Environ. Part B 1992 26 421. 9313824 J. Radioanal. Nucl. Chem. 1993 175 33. 9313825 Proc. ASTM-EURATOM Symp. React. Dosim. 7th 1990 (Pub. 1992) 861.9313826 Proc. SPIE-Int. SOC. Opt. Eng. 1992 1743 351. 9313827 J. Phys. B At. Mol. Opt. Phys. 1993 26 101. 9313828 Proc. SPIE-Int. SOC. Opt. Eng. 1992 1743 236. 9313829 Proc. SPIE-Int. SOC. Opt. Eng. 1992,1743 2.9313830 Proc. SPIE-Int. SOC. Opt. Eng. 1992 1743 12. 9313831 Proc. SPIE-Int. SOC. Opt. Eng. 1992 1743 24. 9313833 Proc. SPIE- Int. SOC. Opt. Eng. 1992 1743 245. 9313835 Proc SPIE-Int.SOC. Opt. Eng. 1992 1743 398. 9313836 U.S. US 5,159,619 (Cl. 378-143; H01J35/08) 27 Oct 1992 Appl. 760,262,16 Sep 1991; 6 pp. 9313837 Appl. Surf. Sci. 1993,63,93. 9313839 Ukr. Khim. Zh. (Russ. Ed.) 1992 58 746. 9313841 Guangpuxue Yu Guangpu Fenxi 1992 12 111. 9313842 PCT Int. Appl. WO 92 07,246 (Cl. GOlN17/00) 30 Apr 1992 NO Appl. 90/4 504 18 Oct 1990; 10 pp. 9313843 Indian J. Forensic Sci. 1992 6 79. 9313844 Eur. Pat. Appl. EP 507,405 (CI. GOlN21/55) 07 Oct 1992 GB Appl. 91/7 041 04 Apr 1991; 53 pp. 9313845 Nucl. Sci. Tech. 1992 3 191. 9313846 Proc. SPIE-Int. SOC. Opt. Eng. 1992 1743 36. 9313847 Nucl. Instrum. Methods Phys. Res. Sect. B 1993 73 109. 9313848 Phys. Rev. A 1993 47 1087. 9313849 Indian J. Pure Appl. Phys. 1993 31 137. 9313850 Spectrochim. Acta Part B 1993 48 277.9313851 Proc. SPIE- Int. SOC. Opt. Eng. 1992 1743 192. 9313852 Proc. SPIE-Int. SOC. Opt. Eng. 1992 1743 328. 9313854 IEEE Trans. Nucl. Sci. 1992 39 2282. 9313855 Proc. SPIE-Int. SOC. Opt. Eng. 1993 1736 106. 93/3856 An. Asoc. Quim. Argent. 1992 80 305. 9313858 Chimia 1992 46 477. 9313860 Nucl. Instrum. Methods Phys. Res. Sect. A 1992 322 551. 9313861 Proc. SPIE-Int. SOC. Opt. Eng. 1992 1720 272. 9313863 Astrophys. J. 1992 400 699. 9313864 X-Ray Spectrom. 1992 21 299. 9313865 Low Temp. Detect. Neutrinos Dark Matter I V Proc. Int. Workshop 4th 1991 (Pub. 1992) 333. 9313866 Proc. SPIE-Int. SOC. Opt. Eng. 1992 1720 264. 9313868 U.S. US 13 Jun 1991; 9pp. 9313869 Jpn. Kokai Tokkyo Koho JP 04,164,297 [92,164,297] (Cl. G21K1/06) 09 Jun 1992 Appl. 90/291,053 29 Oct 1990; 4 pp.9313870 J. Electrochem. SOC. 1992 139 3659. 9313871 X-Ray Spectrom. 1992 21 245. 9313872 Mikrochim. Acta 1992 108 299. 9313873 Zh. Prikl. Spektrosk. 1992 57 208. 9313874 Radiat. Prot. Dosim. 1992 42 319. 9313875 Nucl. Instrum. Methods Phys. Res. Sect. A 1992 322 583. 9313876 Nucl. Instrum. Methods Phys. Res. Sect. A 1993,324 598.9313877 J. Opt. SOC. Am. A Opt. Image Sci. 1992 9 2231. 9313878 Rev. Sci. Instrum. 1993 64 26. 9313879 J. Phys. B At. Mol. Opt. Phys. 1992,25,5283.93/3881 Nucl. Instrum. Methods Phys. Res. Sect. A 1993 325 343. 9313882 Jpn. Kokai Tokkyo Koho JP 04,133,000 [92,133,000] (Cl. G21K1/06) 07 May 1992 Appl. 90/256,035 25 Sep 1990; 4pp. 9313883 Jpn. Kokai Tokkyo Koho JP 04,250,400 5,164,975 (Cl. 378-85; G21K1/06) 17 NOV 1992 Appl.714,805 [92,250,400] (Cl. G21K1/06) 07 Sep 1992 Appl. 91/921 09 Jan 1991; 4pp. 9313884 Jpn. Kokai Tokkyo Koho J P 04,169,899 [92,169,899] (Cl. G21K1/06) 17 Jun 1992 Appl. 90/298,202 02 Nov 1990; 3 pp. 9313885 Zavod. Lab. 1992 58(6) 12.9313887 Ger. Offen. DE 4,028,044 (Cl. GOlN23/203) 12 Mar 1992 Appl. 05 Sep 1990; 5pp. 9313888 U.S. US 5,173,930 (Cl. 378-85; G21K1/06) 22 Dec 1992 Appl. 803,268 22 Nov 1991; 5pp. 9313889 Eur. Pat. Appl. EP 520,671 (Cl. G02B7/18) 30 Dec 1992 US App. 723,081,28 Jun 1991; 13 pp. 9313890 Fett Wiss. Technol. 1992 94 297. 9313891 Yongu Kipo-Sanop Kwahak Kisul Yonguso 1992 6 70. 9313892 Acta Phys. Pol. A 1992 82 13. 9313893 ISTEC J. 1992,5(2) 13. 9313894 Part. Surf. 1991 3 131. 9313896 Environ. Res. 1992 59 326. 9313897 Toxicol.Lett. 1992,64-65 17. 9313900 Nucl. Instrum. Methods Phys. Res. Sect. B 1992 72 467. 9313901 Adv. At. Mol. Opt. Phys. 1992 (Suppl. 1 At. Intense Laser Fields) 207. 9313904 Ukr. Fiz. Zh. (Russ. Ed.) 1992 37 1563. 9313905 Nucl. Instrum. Methods Phys. Res. Sect. A 1993 324 404. 9313906 Jpn. J. Appl. Phys. Part 2 1992 31 ( 12B) L1794. 9313908 Nucl. Instrum. Methods Phys. Res. Sect. A 1992 322 499. 9313909 Fenxi Huaxue 1992 20 850. 9313911 Muter. Sci. Technol. 1992 2A 619. 9313912 X-Ray Spectrom. 1992 21 271. 9313913 X-Ray Spectrom. 1992 21 287. 9313914 Anal. Chim. Acta 1992 270 107. 9313915 Anal. Chim. Acta 1992 270 115. 9313917 Rev. Sci. Instrum. 1992 63 5597. 9411 Adv. X-Ray Anal. 1992 35B 1027. 9412 Adv. X-Ray Anal. 1992 35B 1047. 9413 Adv. X-Ray Anal. 1992 35B 1157.94127 Fresenius’ J.Anal. Chem. 1992,344 16.94144 Spectrochim. Acta Part B 1993 48 171. 94/45 Spectrochim. Acta Part B 1993 48 193. 941109 GITFachz. Lab. 1993 37 111. 941128 J. AOAC Int. 1992 75 949. 941144 Kogyo Yosui 1992 409 88. 941201 Yankuang Ceshi 1992 11 37. 941204 Yankuang Ceshi 1992 11 173. 941217 Zavod. Lab. 1992 58 30. 941231 Anal. Chim. Acta 1992 268 177. 941267 Fresenius’ J. Anal. Chem. 1992 344 535. 941273 J. Anal. At. Spectrom. 1992 7 1173. 941278 J. Anal. At. Spectrom. 1993 8 45. 941280 J. Anal. At. Spectrom. 1993 8 145. 941288 J. Anal. At. Spectrom. 1993 8 261. 941290 J. Anal. At. Spectrom. 1993 8 279. 941292 J. Anal. At. Spectrom. 1993 8 299.941299 J. Anal. At. Spectrom. 1993,8,293.94/451 J. Trace Elem. Electrolytes Health Dis. 1992 6 75.941554 J. Anal. At. Spectrom. 1993 8 371. 941562 J. Anal. At. Spectrom. 1993 8 427.941563 J. Anal. At. Spectrom. 1993,8,433.94/564 J. Anal. At. Spectrom. 1993,8 439. 941565 J. Anal. At. Spectrom. 1993 8 445. 941566 J. Anal. At. Spectrom. 1993 8 449. 941567 J. Anal. At. Spectrom. 1993 8 453. 941574 J. Anal. At. Spectrom. 1993 8 673. 941576 J. Anal. At. Spectrom. 1993 8 687.941577 J. Anal. At. Spectrom. 1993,8,691.94/587 J. Anal. At. Spectrom. 1993,8 755. 941590 J. Anal. At. Spectrom. 1993 8 767. 941592 J. Anal. At. Spectrom. 1993 8 787. 941604 J. Anal. At. Spectrom. 1993 8 859. 941605 J. Anal. At. Spectrom. 1993 8 867. 941606 J. Anal. At. Spectrom. 1993 8 875. 941607 J. Anal. At. Spectrom. 1993 8 881. 941608 J. Anal. At. Spectrom. 1993 8 891. 941609 J. Anal. At.Spectrom. 1993 8 899. 941614 J. Anal. At. Spectrom. 1993 8 927. 941619 Spectrochim. Acta Part B 1993 48 1013. 941626 Spectrochim. Acta Part B 1993 48 1065. 941633 Spectrochim. Acta Part B 1993 48 1139. 941649 Fresenius’ J. Anal. Chem. 1993 345 261. 941651 Spectrochim. Acta Part B 1993 48 129. 941652 Spectrochim. Acta Part B 1993 48 143. 941653 Spectrochim. Acta Part B 1993 48 153. 941654 Spectrochim. Acta Part B 1993 48 163. 941655 Spectrochim. Acta Part B 1993 48 183. 941656 Spectrochim. Acta Part B 1993 48 199. 941657 Spectrochim. Actu Part B 1993 48 207. 941658 Spectrochim. Acta Part B 1993 48 223 941659 Spectrochim. Acta Part B 1993 48 231. 941660 Spectrochim. Acta Part B 1993 48 239. 941661 Spectrochim. Acta Part B 1993 48 285. 941662 Spectrochim. Acta Part B 1993 48 293.941676 J. Radioanal. Nucl. Chem. 1993 167 413. 941677 J. Radioanal. Nucl. Chem. 1993 167 433. 941687 Spectrochim. Acta Rev. 1993 15 153. 941688 Spectrosc. Eur. 1992 4 10 12 14 16 18 20. 941690304 R JOURNAL OF ANALYTICAL ATOMIC SPECTROMETRY OCTOBER 1994 VOL. 9 Zauod. Lab. 1992 58 1. 941698 Anal. Chim. Acta 1993 276 39. 941738 J. Anal. At. Spectrom. 1992 7 1187. 941769 Anal. Biochem. 1993 210 113. 941771 Anal. Chem. 1993 65 12. 941838 Fresenius’ Environ. Bull. 1992 1 741. 941882 Laser Part. Beams 1992 10 737. 941883 Laser Part. Beams 1992 10 753. 941884 Laser Part. Beams 1992 10 861. 941919 Report 1990 Order No. N92-11349 145 pp. 941928 Sci. Total Environ. 1993 133 153. 941937 Tec. Lab. 1993 15(178) 32. 941938 Text. Prax. Int. 1992 47 826 828 833.941974 Spectrochim. Acta Part B 1993 48 1237. 941978 Spectrochim. Acta Part B 1993 48 1281 941991 Appl. Spectrosc. 1993 47 330.941992 Fenxi Huaxue 1993,21 190.94/1001 Spectrochim. Acta Part B 1993 48 553. 9411002 Spectrochim. Acta Part B 1993 48 589. 9411006 Spectrochim. Acta Part B 1993 48 663. 9411011 Talanta 1993 40 669. 9411015 Zh. Anal Khim. 1992 47 1893. 9411060 Guangpuxue Yu Guangpu Fenxi 1992 12 83. 9411068 Guangpuxue Yu Guangpu Fenxi 1992 12 91. 9411069 Guangpuxue Yu Guangpu Fenxi 1992,12 99.9411090 J. Nucl. Muter. 1993 200 282. 9411091 J. Radioanal. Nucl. Chem. 1993,170 171.9411092 J. Radioanal. Nucl. Chem. 1993 170 259. 9411138 Spectroscopy (Eugene Oreg.) 1992 7 12. 9411152 Yankuang Ceshi 1992 11 217. 9411157 Yankuang Ceshi 1992,11,260.94/1166 Zauod.Lab. 1992,58,17.94/1196 X-Ray Spectrom. 1992 21 i. 9411199 Zh. Anal. Khim. 1993 48 166. 9411202 Appl. Radiat. Isot. 1993 44 659. 9411224 Guangpuxue Yu Guangpu Fenxi 1993 13 131. 9411226 Guangpuxue Yu Guangpu Fenxi 1993 13 149. 9411248 Nucl. Instrum. Methods Phys. Res. Sect. A 1993 A327 594. 9411275 Anal. Chem. 1993 65 311r. 9411276 Anal. Chem. 1993 65 585A. 9411277 Anal. Chem. 1993 65 1717. 9411279 Anal. Chem. 1993 65 1960. 9411280 Anal. Chem. 1993 65 2125. 9411282 Anal. Chem. 1993 65 2478. 9411283 Anal. Chem. 1993 65 2485. 9411284 Anal. Chem. 1993 65 2542. 9411285 Anal. Chem. 1993 65 2784. 9411286 Anal. Chem. 1993 65 2801. 9411287 Anal. Chem. 1993 65 2972. 9411289 Anal. Chem. 1993 65 3166. 9411290 Anal. Chem. 1993 65 3187. 9411292 Anal. Chem. 1993,65 3199.9411294 Anal.Chim. Acta 1993 280 137. 9411296 Analyst 1993 118 835. 9411297 Appl. Spectrosc. 1993 47 575. 9411301 Fresenius’ J. Anal. Chem. 1993 345 615. 9411302 Fresenius’ J. Anal. Chem. 1993 346 3. 9411303 Fresenius’ J. Anal. Chem. 1993 346 14. 9411306 Fresenius’ J. Anal. Chem. 1993 346 131. 9411307 Fresenius’ J. Anal. Chem. 1993 346 134. 9411308 Fresenius’ J. Anal. Chem. 1993 346 147.9411311 Fresenius’ J. Anal. Chem. 1993 346 186. 9411313 Fresenius’ J. Anal. Chem. 1993 346 310. 9411315 Fresenius’ J. Anal. Chem. 1993 346 368. 9411317 Fresenius’ J. Anal. Chem. 1993 346 426. 9411318 Fresenius’ J. Anal. Chem. 1993,346 594.9411319 Spectrochim. Acta Part B 1993 48 507. 9411321 Spectrochim. Acta Part B 1993 48 663. 9411322 Talanta 1993 40 975. 9411323 Zh. Anal. Khim.1992 47 1804. 9411325 Adv. Environ. Sci. Technol. 1992 25 77. 9411328 Anal. Lett. 1993 26 981. 9411335 Annu. Rep. Prog. Chem. Sect. C 1991 (Pub. 1993) 88 77. 9411336 Appl. Geochem. Suppl. 1993 2 111. 9411338 Appl. Phys. B 1993 B56 275. 9411343 Appl. Surf. Sci. 1993 70 44. 9411344 Appl. Surf. Sci. 1993 70 49. 9411345 Appl. Surf. Sci. 1993 70 63. 9411346 Appl. Surf. Sci. 1993 70 73. 9411351 Biol. Mass Spectrom. 1993 22 441. 9411352 Biol. Trace Elem. Res. 1993 37 123. 9411354 Bull. Muter. Sci. 1993 16 45. 9411356 Can. Mineral. 1992,30 1023. 9411357 Can. Mineral. 1992 30 1049. 9411359 Can. Mineral. 1993,31 19. 9411360 Carbon 1993,31 273.9411363 Chem. Geol. 1993,106,229.9411364 Chem. Geol. 1993 106 251. 9411365 Chem. Geol. 1993 106 485. 9411366 Chem. Geol. 1993,107,147.94/1370 CLB Chem.Labor Biotech. 1993 44 284. 9411374 Dokl. Bulg. Akad. Nauk. 1992 45 21. 9411378 Euro Courses Adv. Sci. Tech. 1991 1 37. 9411381 Geochem. J. 1992,26,105.94/ 1386 Geochim. Cosmochim. Acta 1993 57 1339. 9411387 Geochim. Cosmochim. Acta 1993 57 3479. 9411388 Geochim. Cosmochim. Acta 1993 57 3493. 9411390 Haiyang Yu Huzhao 1992,23,550.94/1393 Inst. Phys. Con$ Ser. 1992 128 293. 9411396 Int. J. Mass Spectrom. Ion Processes 1993 123 139. 9411397 Int. J. Mass Spectrom. Ion Processes 1993 125 241. 9411406 J. Appl. Phys. 1993 73 4649. 9411408 J. Electrochem. SOC. 1993 140 1105. 9411412 J. Microsc. (Oxford) 1993 169 163. 9411414 J. Radioanal. Nucl. Chem. 1993 175 427. 9411415 J. Opt. SOC. Am. B Opt. Phys. 1993 10 763. 9411418 J. Vac. Sci. Technol. A 1993 11 2317. 9411419 J.Vac. Sci. Technol. A 1993 11 2373. 9411420 Jpn. J. Appl. Phys. Part I 1993 32 2167. 9411438 Nitrogen hot. Tech. 1993 11. 9411439 Nucl. Instrum. Methods Phys. Res. Sect. A 1993 A328 207. 9411440 Nucl. Instrum. Methods Phys. Res. Sect. A 1993 A328 330. 9411443 Nucl. Instrum. Methods Phys. Res. Sect. B 1993 B74 156. 9411446 Nucl. Instrum. Methods Phys. Res. Sect. B 1993 B79 617. 9411447 Nucl. Instrum. Methods Phys. Res. Sect. B 1993 B79 620. 9411448 Nucl. Instrum. Methods Phys. Res. Sect. B 1993 B79 624. 9411449 Nucl. Instrum. Methods Phys. Res. Sect. B 1993 B79 627. 9411450 Nucl. Instrum. Methods Phys. Res. Sect. B 1993 B79 63 1. 9411452 Nucl. Instrum. Methods Phys. Res. Sect. B 1993 BSO 543. 9411454 Nucl. Instrum. Methods Phys. Res. Sect. B 1993 82 399.9411463 Phys. Rev. Lett. 1993 70 2888. 9411467 Proc. Jpn. Acad. Ser. B 1993 69 123. 9411469 Prepr. Am. Chem. SOC. Div. Pet. Chem. 1993,38 363.9411471 J. Mass Spectrom. Ion Processes 1993 124 223. 9411472 Radioact. Radiochem. 1993 4 24. 9411473 Radiocarbon 1992 34 219. 9411474 Radiocarbon 1992 34 611. 9411475 Radiocarbon 1992 34 737. 9411477 Rapid Commun. Mass Spectrom. 1993 7 323. 941 1479 Rapid Commun. Mass Spectrom. 1993 7 524. 9411481 Report 1991 UCRL-CR- 108120; Order No. DE92003966,5 pp. 9411483 Rapid Commun. Mass Spectrom. 1993 7 581. 9411484 Rev. Sci. Instrum. 1993 64 1146. 9411486 Shigen to Kankyo 1993 2 331. 9411487 Shipin Kexue (Beijing) 1993 159 70. 9411489 Spectroscopy (Eugene Oreg.) 1993 8 32. 9411492 Surf. Interface Anal. 1993 20 121. 9411493 Surf. Interface Anal. 1993 20 206. 9411495 Surf. Interface Anal. 1993 20 621. 9411497 Surf. Interface Anal. 1993 20 696. 9411501 Trends Anal. Chem. 1993 12 86. 9411503 Vacuum 1993 44 203. 9411504 Vacuum 1993 44 223. 9411505 Vacuum 1993 44 331. 9411506 Vacuum 1993 44 389. 9411509 Vysokochist. Veshchestva 1993 2 91. 9411517 Second. Ion Mass Spectrom. SIMS 8 Proc. Int. Con$ 8th 1991,49.94/1519 Second. Ion Mass Spectrom. SIMS 8 Proc. Int. Conf. 8th 1991 91. 9411541 Second. Ion Mass Spectrom. SIMS 8 Proc. Int. Conf. 8th 1991 661. 9411560 Anal. Chem. 1992,64 180R. 9411561 Fresenius’ J. Anal. Chem. 1993 346 162. 9411562 Fresenius’ J. Anal. Chem. 1993 346 608. 9411563 X-ray Spectrom. 1993 22 244. 9411564 X-ray Spectrom. 1993 22 187. 9411565 X-ray Spectrom. 1993 22 192. 9411566 X-ray Spectrom. 1993 22 199. 9411567 X-ray Spectrom. 1993 22 208. 9411568 X-ray Spectrom. 1993 22 216. 9411569 X-ray Spectrom. 1993 22 220. 9411570 X-ray Spectrom. 1993 22 229. 9411571 X-ray Spectrom. 1993 22 234. 9411572 X-ray Spectrom. 1993 22 248. 9411573 X-ray Spectrom. 1993 22 252. 9411574 X-ray Spectrom. 1993 22 260. 9411575 X-ray Spectrom. 1993 22 265. 9411576 X-ray Spectrom. 1993 22 272. 9411577 X-ray Spectrom. 1993 22 277. 9411578 X-ray Spectrom. 1993 22 281. 9411579 X-ray Spectrom. 1993 22 286. 9411580 X-ray Spectrom. 1993 22 293. 9411581 X-ray Spectrom. 1993 22 300. 9411582 X-ray Spectrom. 1993 22 305. 9411583 X-ray Spectrom. 1993 22 312. 9411584 X-ray Spectrom. 1993 22 317. 9411585 X-ray Spectrom. 1993 22 323. 9411586 X-ray Spectrom. 1993 22 328. 9411587 X-ray Spectrom. 1993 22 332. 9411588 Am. Lab. 1993 25(14) 36H. (Jordan Valley Appl. Radiation Migdal Haemek Israel 9411589 Appl. Radiat. Isot. 1993 44 929. 9411590 Appl. Radiat. Isot. 1993 44 1101. 9411591 Jpn. J. Appl. Phys. 1993 32 1279. 9411592 J. Radioanal. Nucl. Chem. Art. 1993 172 137. 9411594 Nucl. Instrum. Methods Phys. Res. Sect. A 1993 330 501. 9411595 Nucl. Instrum. Methods Phys. Res. Sect. A 1993 332 314. 9411597 Nucl. Instrum. Methods Phys. Res. Sect. A 1993 334 447. 9411598 Nucl. Instrum. Methods Phys. Res. Sect. B 1993,JOURNAL OF ANALYTICAL ATOMIC SPECTROMETRY OCTOBER 1994 VOL. 9 305 R 82 121. 9411599 Nucl. Instrum. Methods Phys. Res. Sect. B 1993 82 129. 9411600 Jpn. J. Appl. Phys. 1993 32 160. 9411601 Quim. Anal. 1993 12 57. 9411602 Zh. Anal. Khim. 1992 47 629. 9411693 Asian J. Chem. 1993 5 743. 9411728 Clin. Chem. (Washington D.C.) 1993 39 1037. 9411739 Electrophoresis (Weinheim Fed. Repub. Ger.) 1993 14 531. 9412176 J. Anal. At. Spectrom. 1993 8 935. 9412182 J. Anal. At. Spectrom. 1993 8 979. 9412183 J. Anal. At. Spectrom. 1993 8 981. 9412191 J. Anal. At. Spectrom. 1993 8 1029. 9412192 J. Anal. At. Spectrom. 1993 8 1033. 9412194 J. Anal. At. Spectrom. 1993 8 1053. 9412195 J. Anal. At. Spectrom. 1993 8 1059. 9412196 J. Anal. At. Spectrom. 1993 8 1067. 9412212 J. Anal. At. Spectrom. 1994 9 17. 9412213 J. Anal. At. Spectrom. 1994 9 23. 9412218 Spectrochim. Acta Part B 1993 48 137. 9412219 Spectrochim. Acta Part B 1993 48 217.9412220 Spectrochim. Acta Part B 1993,48,247. 9412221 Spectrochim. Acta Part B 1993 48 263. 9412222 Spectrochim. Acta Part B 1993 48 269. 9412223 Chem. Geol. 1993 106 187.9412225 Chem. Geol. 1993,106,467.9412227 Chem. Geol. 1993 109 1. 9412228 Chem. Geol. 1993 109 125. 9412230 Geostand. Newsl. 1993 17 127. 9412235 Geochim. Cosmochim. Acta 1993,57,4119.94/2237 Geochim. Cosmochim. Acta 1993 57 2907. 9412242 Appl. Opt. 1993 32 6316. 9412243 Appl. Optics 1993 32 6333. 9412252 Anal. Chem. 1993 65 1181. 9412255 Anal. Chim. Acta 1993 277 289. 9412261 Fenxi Huaxue 1993 21 704. 9412268 X-Ray Spectrom. 1993 22 125. 9412269 X-Ray Spectrom. 1993 22 151. 9412270 X-Ray Spectrom. 1993 22 160. 9412271 X-Ray Spectrom. 1993 22 166. 9412272 X-Ray Spectrom. 1993 22 178. 9412292 Int. Lab. 1993 23 14 16. 9412306 Br. J. Ind. Med. 1992 49 631. 9412307 Biochem. Biophys. Res. Commun. 1992 188 1010. 9412308 Indian J. Chem. Sci. 1991 5 91. 9412309 Aerosol Sci. Technol. 1992 17 213. 9412310 Pharm. Weekbl. Sci. Ed. 1992 14 321. 941231 1 Fuming Zhuanli Shenqing Gongkai Shuomingshu CN 1,058,470 (CI. GOlN21/64) 1992 Appl. 91,104,263,02 Jul 1991; 22 pp. 9412312 Nucl. Instrum. Methods Phys. Res. Sect. B 1992 72 147. 9412313 J. Appl. Phys. 1992 72 5032. 9412314 Low Temp. Detect. Neutrinos Dark Matter IV Proc. Int. Workshop 4th 1991 (Pub. 1992) 265. 9412315 Low Temp. Detect. Neutrinos Dark Matter I V Proc. Int. Workshop 4th 1991 (Pub. 1992) 285. 9412316 IEEE Trans. Nucl. Sci. 1992 39 1275. 9412318 Mikrochim. Acta Suppl. 1992 12 147. 9412319 Nucl. Instrum. Methods Phys. Res. Sect. A 1992 322 59 1. 9412320 Low Temp. Detect. Neutrinos Dark Matter III [Proc. Int. Workshop] 3rd 1989 (Pub. 19901 213. 9412321 Low Temp. Detect. Neutrinos Dark Matter I V Proc. Int. Workshop 4th 1991 (Pub. 1992) 275. 9412322 Adv. Crystallogr. Cryst. Growth Proc. Indo-Sov. Symp. Cryst. Growth 1989 (Pub. 1991) 143.9412323 J. Sains Nukl. Malays. 1991,9,69. 9412324 Spectroscopy (Eugene Oreg.) 1992,7 32 36. 9412326 Proc. Natl. Conf. Hazard. Wastes Hazard. Muter. 7th 1990 58. 9412327 Surf. Interface Anal. 1992 18 773. 9412328 Nucl. Instrum. Methods Phys. Res. Sect. A 1992 322 453. 9412329 Nucl. Instrum. Methods Phys. Res. Sect. A 1992 322 457. 9412330 Nucl. Instrum. Methods Phys. Res. Sect. A 1992 322 504. 9412331 Nucl. Instrum. Methods Phys. Res. Sect. A 1992 322 543. 9412333 Nucl. Instrum. Methods Phys. Res. Sect. A 1992 322 538. 9412334 Nucl. Instrum. Methods Phys. Res. Sect. A 1992 322 566. 9412336 Nucl. Instrum. Methods Phys. Res. Sect. A 1992 322 449. 9412337 Nucl. Instrum. Methods Phys. Res. Sect. A 1992 322 464. 9412338 Nucl. Instrum. Methods Phys. Res. Sect. A 1992 322 493. 9412339 Nucl. Instrum. Methods Phys. Res. Sect. A 1992 322 514. 9412340 Nucl. Instrum. Methods Phys. Res. Sect. A 1992 322 633. 9412342 Zavod. Lab. 1992 58 61. 9412344 X-Ray Spectrom. 1992 21 253. 9412354 Fresenius’ J. Anal. Chem. 1993 345 737. 9412355 Fresenius’ J. Anal. Chem. 1993 347 11 1.9412356 Spectrochim. Acta Part B 1993,48,1471.94/2357 X-ray Spectrom. 1993,22,132.94/2358 X-ray Spectrom. 1993 22 138. 9412359 X-ray Spectrom. 1993 22 156. 9412360 X - ray Spectrom. 1993 22 163. 9412361 X-ray Spectrom. 1993 22 345. 9412362 X-ray Spectrom. 1993 22 349. 9412363 X - ray Spectrom. 1993 22 352. 9412364 X-ray Spectrom. 1993 22 358. 9412365 X-ray Spectrom. 1993 22 362. 9412366 X - ray Spectrom. 1993 22 368. 9412367 X-ray Spectrom. 1993 22 377. 9412368 Adv. X-Ray Anal. 1992 35 1097. 9412369 Appl. Radiat. hot. 1993 44 1373. 9412370 Geochim. et Cosmochim. Acta 1992 56 4059. 9412371 J. Appl. Phys. 1993 74 107. 9412372 J. Radioanal. and Nucl. Chem. Articles 1993 173,9. 9412374 Magmatic Processes and Plate Tectonics Geol. SOC. Spec. Publ. (London) 76 1993 501.9412375 Nucl. Instum. Methods Phys. Res. Sect. A 1988 271 671. 9412377 Nucl. Instum. Methods Phys. Res. Sect. A 1993 324 383. 9412378 Nucl. Instum. Methods Phys. Res. Sect. B 1993,75,91. 9412379 Nucl. Instum. Methods Phys. Res. Sect. B 1993 75 450. 9412380 Nucl. Instum. Methods Phys. Res. Sect. B 1993 75 553. 9412381 Nucl. Instum. Methods Phys. Res. Sect. B 1993 75 563. 9412382 Phys. Rev. A 1993 47 4078. 9412384 Ind. Lab. 1993 58 828. 9412385 Ind. Lab. 1993 58 816.

ISSN:0267-9477    

DOI:10.1039/JA994090267R

出版商:RSC    

年代:1994

数据来源: RSC

摘要:

306 R JOURNAL OF ANALYTICAL ATOMIC SPECTROMETRY OCTOBER 1994 VOL. 9 Glossary of Abbreviations Whenever suitable elements may be referred to by their chemical symbols and compounds by their formulae. The following abbreviations are used extensively in the Atomic Spectrometry Updates. a.c. AA AAS AE AES AF AFS AOAC APDC ASV BCR CCP CMP CRM cv cw d.c. DCP DDC DMF DNA ECD EDL EDTA EDXRF EIE EPMA ETA ETAAS ETV EXAFS FAAS FAB FAES FAFS FANES FAPES FI FPD FT FTMS GC GD GDL GDMS Ge ( Li ) HCL h.f. HG HPGe HPLC IAEA IBMK ICP ICP-MS alternating current atomic absorption atomic absorption spectrometry atomic emission atomic emission spectrometry atomic fluorescence atomic fluorescence spectrometry Association of Official Analytical Chemists ammonium pyrrolidinedithiocarbamate anodic-stripping voltammetry Community Bureau of Reference capacitively coupled plasma capacitively coupled microwave plasma certified reference material cold vapour continuous wave direct current d.c.plasma diethyldithiocarbamate N N-dimeth ylformamide deoxyribonucleic acid electron capture detection electrodeless discharge lamp ethylenediaminetetraacetic acid energy dispersive X-ray fluorescence easily ionizable element electron probe microanalysis electrothermal atomization electrothermal atomic absorption spectrometry electrothermal vaporization extended X-ray absorption fine structure flame AAS fast atom bombardment flame AES flame AFS furnace atomic non-thermal excitation furnace atomization plasma excitation flow injection flame photometric detector Fourier transform Fourier transform mass spectrometry gas chromatography glow discharge glow discharge lamp glow discharge mass spectrometry lithium-drifted germanium hollow cathode lamp high frequency hydride generation high-purity germanium high-performance liquid chromatography International Atomic Energy Agency isobutyl methyl ketone (4-methylpentan-2-one) inductively coupled plasma inductively coupled plasma mass spectrometry (ammonium pyrrolidin-1-yl dithioformate) spectroscopy spectrometry spectrometry ID IR IUPAC LA LC LEAFS LEI LMMS LOD LTE MECA MIP MS NAA NaDDC NIES NIST NTA OES PIGE PIXE PMT PPb PPm PTFE QC r.f.REE(s) RIMS RM RSD SEC SEM SFC Si (Li) SIMAAC SIMS SR SRM SSMS STPF TCA TIMS TLC TMAH TOP0 TXRF u.h.f. uv VDU vuv WDXRF XRF SIB SIN isotope dilution infrared International Union of Pure and Applied Chemistry laser ablation liquid chromatography laser-excited atomic fluorescence spectrometry laser-enhanced ionization laser-microprobe mass spectrometry limit of detection local thermal equilibrium molecular emission cavity analysis microwave-induced plasma mass spectrometry neutron activation analysis sodium diethyldithiocarbamate National Institute for Environmental Studies National Institute of Standards and ni trilo triace tic acid optical emission spectrometry particle-induced gamma-ray emission particle-induced X-ray emission photomultiplier tube parts per billion parts per million polytetrafluoroethylene quality control radio frequency rare earth element(s) resonance ionization mass spectrometry reference material relative standard deviation signal to background ratio size-exclusion chromatography scanning electron microscopy supercritical fluid chromatography lithium-drifted silicon simultaneous multi-element analysis with a continuum source secondary ion mass spectrometry signal to noise ratio synchrotron radiation Standard Reference Material spark source mass spectrometry stabilized temperature platform furnace trichloroacetic acid thermal ionization mass spectrometry thin-layer chromatography tetramethylammonium hydroxide trioctylphosphine oxide total reflection X-ray fluorescence ultra-high frequency ultraviolet visual display unit vacuum ultraviolet wavelength dispersive X-ray fluorescence X-ray fluorescence Technology

ISSN:0267-9477    

DOI:10.1039/JA994090306R

出版商:RSC    

年代:1994

数据来源: RSC

摘要:

307 R JOURNAL OF ANALYTICAL ATOMIC SPECTROMETRY OCTOBER 1994 VOL. 9 ATOMIC SPECTROMETRY UPDATE REFERENCES The address given in a reference is that of the first named author and is not necessarily the same for any co-author. 9412868. 9412869. 9412870. 9412 87 1. 9412872. 9412873. 9412874. 94/28 7 5. 94/2876. 94/28 77. 941287 8. 9412 8 7 9. 9412880. 94/2881. Yago H. Furuta K. Ishikawa K. Komura H. Time- resolved emission spectra from laser-induced plasma in sodium chloride Phys. Status Solidi B 1993 179 223. (Grad. Sch. Electron. Sci. Techno]. Shizuoka Univ. Mahamatsu Japan 432). Bykovskii Yu. A. Zapov V. V. Uzienko D. A. Laser analyser for the elemental composition of aerosols in gases Prib. Tekh. Eksp. 1993 (2) 244. Berg P. A. Sloan J. J. Compact standalone data acquisition system for submicrosecond time-resolved Fourier transform spectroscopy Rev. Sci.Instrum. 1993 64 2508. (Dept. Chem. Univ. Waterloo Waterloo Ontario Canada N2L 3G1). Weber R. Atomic emission spectroscopy in gas chromatography Spurenanal. Hum.- Umweltbereich Beitr. Symp. Ges. Toxikol Forensische Chem. 199 1 (Pub. 1992) 87. (Hewlett-Packard W-7517 Waldbronn Germany). Ricard A. Spectroscopy of flowing discharges and post-discharges in reactive gases (Lab. Phys. Gaz Plasmas Univ. Paris-Sud 9 1405 Orsay France). Wichmann N. Trends in analytical instrumentation Umwelt 1993 23 508. (Hamburg Germany). Hayakawa S. Kobayashi K. Gohshi Y. A trace element quantification method using energy tuneable X-ray fluorescence analysis X-sen Bunseki no Shinpo 1992 24 35. (Fac. Eng.Univ. Tokyo Tokyo Japan 113). Fang G.-z. Huang R. Application of organic reagents in atomic emission spectroscopy Yejin Fenxi 1992 12( 6 ) 3 1. (Dept. Chem. Sichuan Univ. Chengdu 610041 China). Yang J.-f. Zheng Z.-j. Huang B.4 Argon inductively coupled plasma with introduction of ethanol solution. I. Effect of ethanol on the spectral line intensity of rare earth elements Zhongguo Xitu Xuebao 1992,10( l) 31. (Changchun Inst. Appl. Chem. Acad. Sin. Changchun 130022 China). Via J. Taylor L. T. Packed-column supercritical fluid chromatography/chemical ionization mass spec- trometry of energetic material extracts using a thermo- spray interface Anal. Chem. 1994 66 1385. (Dept. Chem. Virginia Polytech. Univ. and State Univ. Blacksburg VA 24061 USA). Maquieira A Elmahadi H.A. M. Puchades R. Use of saccharomyces cerevisiae in flow injection atomic absorption spectrometry for trace metal preconcen- tration Anal. Chem. 1994 66 1462. (Dept. Chem. Polytech. Univ. Valencia Camino de Vera s/n 46071 Valencia Spain). Pretty J. R. Blubaugh E. A. Caruso J. A. Davidson T. M. Determination of chromium(v1) and vanadium(v) using an on-line anodic stripping voltammetry flow cell with detection by inductively coupled plasma mass spectrometry Anal. Chem. 1994 66 1540. (Dept. Chem. Univ. Cincinnati Cincinnati OH 45221-0172 USA). Morita M. Ito H. Linscheid M. Otsuka K. Resolution of interelement spectral overlaps by high- resolution inductively coupled plasma mass spec- trometry Anal. Chem. 1994 66 1588. (Natl. Inst. Environ. Studies 16-2 Onogawa Tsukuba Ibaraki 305 Japan).Richner P. Evans D. Automated laser ablation sampling (ALAS) for ICP-MS At. Spectrosc. 1993 14 9412882. 94/28 8 3. 9412884. 9412 8 8 5. 94/28 86. 9412887. 9412888. 9412889. 9412890. 9412891. 941289 2. 9412893. 9412 8 94. 9412895. 9412896. 157. (Swiss Federal Labs. Mater. Testing Res. CH-8600 Duf bendorf Switzerland). Denoyer E. R. Lu Q.-h. Characterization of operating parameters in flow injection ICP-MS At. Spectrosc. 1993 14 162. (Perkin-Elmer Corp. 761 Main Ave. Norwalk CT 06859-0215 USA). Al-Swaidan H. M. Trace determination of vanadium and nickel in Saudi Arabian petroleum and petroleum products by microemulsion ICP-MS At. Spectrosc. 1993 14 170. (Dept. Chem. Coll. Sci. King Saud Univ. P.O. Box 2455 Riyadh 11451 Saudi Arabia).Balaram V. Characterization of trace elements in environmental samples by ICP-MS At. Spectrosc. 1993 14 174. (Natl. Geophys. Res. Inst. Hyderabad 500 007 India). Satyanarayana K. Nayeem M. A. Determination of Nb Ta Ti and Sn in niobate tantalates and Nb/Ta- bearing tin slag samples by ICP-AES and FAAS At. Spectrosc. 1993 14 180. (Chem. Lab. At. Miner. Div. Dept. At. Energy Begumpet Hyderabad 5000 16 India). Bendicho C. Sancho A. Determination of selenium in Wheat Flour by GFAAS using automated ultrasonic slurry sampling At. Spectrosc. 1993 14 187. (Pure App. Chem. Dept. Univ. Vigo Campus at Ourense As Lagoas s/n. 32004 Ourense Spain). McLaren J. W. ICP-MS applications bibliography update At. Spectrosc. 1993 14 191. (Inst. Environ. Chem. Natl. Res. Council of Canada Ottawa Canada K1A OR6).Zu L.4 Li A.-m. Interference mechanism of calcium chloride in the determination of Mn by GFAAS Guangpuxue yu Guangpufenxi 1993 13( 6) 100. (Dept. Chem. Peking Univ. Beijing China). Dai L.-j. Xiao Y.-g. Huang J. GFAAS determination of Ru in water by using ascorbic acid as matrix modifier Lihua Jianyan Huaxue Fence 1994 30( l) 15. (Central Lab. Fujian MGMR Fuzhou 350002 China). Zhang Y.-l. FIA-FAAS determination of Fe in caustic soda Lihua Jianyan Huaxue Fence 1994,30( l) 9. (Res. Inst. Qilu Petrochim. Corp. Zibo China). Xu S.-k. FAng Z.4 Determination of trace amount of As by FI-HG-AAS with on-line reduction Fenxi Shiyanshi 1994 13(2) 20. (Inst. Appl. Ecol. Acad. Sin. Shenyang 110015 China). Li G.-k. Zhang Z.-x. Yang X.-h. Standardless analysis by graphite furnace AAS Fenxi Ceshi Xuebao 1994 13(1) 49.(Dept. Chem. Zhongshan Univ. Guangzhou 510275 China). Bai W.-m. Wang Y. Study on GC-AAS with silica furnace and on the analysis of chemical species of lead alkyl Guangpuxue yu Guangpufenxi 1994 14( 1) 103. (Dept. Chem. Tsinghua Univ. 100084 Beijing China). Qi W.-q. Chen Y . 4 Zhang M.-s. Cao J.-s. Speciation of As"' and AsV by Pd-coated graphite furnace and their determination in environmental aqueous samples Guangpuxue yu Guangpufenxi 1994 14( l) 81. (China Natl. Environ. Monitoring Center 100012 Beijing China). Chi X.-z. Zhou J. Determination of trace Mn with on-line FI-solvent extraction preconcentration-FAAS Guangpuxue yu Guangpufenxi 1994 14( l) 94. (Dept. Chem. Beijing Normal Univ. 1000875 Beijing China).Cai Q.-x. Dai J. Zhou Y.-c. Determination of Ca in rat's cardiac muscle by AAS Guangpuxue yu308 R JOURNAL OF ANALYTICAL ATOMIC SPECTROMETRY OCTOBER 1994 VOL. 9 9412897. 94/2898. 9412 8 99. 9412900. 941290 1. 94/2902. 9412903. 9412904. 9412905. 9412906. 9412907. 9412908. Guangpufenxi 1994 14( 2) 103. (Shanghai Inst. Org. Chem. Acad. Sin. 200032 Shanghai China). Zheng J.-g. Zhang Z.-x. Monte Carlo technique for studying chemical and physical processes in ICP-AES. 1. Simulation of nebulization process Guangpuxue yu Guangpufenxi 1994 14(2) 103. (Dept. Chem. Zhongshan Univ. 5 10275 Guangzhou China). Huang H.-j. Fang Y. Zhu M.-h. Shao J.-c. Huang H. Kalman filtering technique for correcting spectral interferences in ICP-AES Guangpuxue yu Guangpufenxi 1994 14(2) 72.(Anal. Res. Center East China Univ. Chem. Tech. 2000237 Shanghai China). Winefordner J. D. Petrucci G. A. Stevenson C. L. Smith B. W. Theoretical and practical limits in atomic spectroscopy J. Anal. At. Spectrom. 1994,9 131. (Dept. Chem. Univ. Florida Gainesville FL 32611 USA). Sheppard B. S. Caruso J. A. Plasma mass spec- trometry consider the source J. Anal. At. Spectrom. 1994 9 145. (US Food and Drug Admin. Natl. Forensic Chem. Center 1141 Central Pkwy Cincinnati OH 45202 USA). Janssens K. Vincze L. Rubio J. Adams F. Bernasconi G. Microscopic X-ray fluorescence analysis J. Anal. At. Spectrom. 1994,9 151. (Dept. Chem. Univ. Antwerp Universiteitsplein 1 B-2610 Antwerp Belgium). Nakahara T. Mori T. Analyte volatilization pro- cedure for the determination of low concentrations of iodine by inductively coupled plasma atomic emission spectrometry J.Anal. At. Spectrom. 1994,9 159. (Dept. Appl. Chem. Univ. Osaka Prefec. Sakai Osaka 593 Japan). Jackson J. G. Fonseca R. W. Holcombe J. A Detection of surface aggregates of trace amounts of copper and silver on graphite using secondary ion mass spectrometry at elevated temperatures J . Anal. At. Spectrom. 1994 9 167. (Dept. Chem. and Biochem. Univ. Texas at Austin Austin TX 78712 USA). Begley I. S. Sharp B. L. Occurrence and reduction of noise in inductively coupled plasma mass spectrometry for enhanced precision in isotope ratio measurement J. Anal. At. Spectrom. 1994 9 171. (Dept. Chem. Loughborough Univ. Technol. Loughborough Leicestershire UK LE113TU). Vanhoe H.Goossens J. Moens L. Dams R. Spectral interferences encountered in the analysis of biological materials by inductively coupled plasma mass spec- trometry J. Anal. At. Spectrorn. 1994 9 177. (Lab. Anal. Chem. Univ. Ghent Inst. Nucl. Sci. Proeftuinstr. 86 B-9000 Ghent Belgium). Moens L. Vanhoe H. Vanhaecke F. Goossens J. Campbell M. Dams R. Application of inductively coupled plasma mass spectrometry to the certification of reference materials from the Community Bureau of Reference J. Anal. At. Spectrom. 1994 9 187. (Ghent Univ. Lab. Anal. Chem. Proeftuinstr. 86 B-9000 Ghent Belgium). Jakubowski N. Jepkens B. Stuewer D. Berndt H. Speciation analysis of chromium by inductively coupled plasma mass spectrometry with hydraulic high pressure nebulization J. Anal. At.Spectrom. 1994 9 193. (Inst. Spektrochem. angew. Spektrosk. Postfach 10 13 52 D-44013 Dortmund Germany). Durrant S. F. Krushevska A. Amarasiriwardena D. Argentine M. D. Romon-Guesnier S. Barnes R. M. Matrix separation by chelation to prepare biological materials for isotopic zinc analysis by inductively coupled plasma mass spectrometry J. Anal. At. Spectrom. 1994 9 199. (Univ. Massachusetts Dept. Chem. Lederle Grad. Res. Center Towers Amherst MA 01003-0035 USA). 9412909. 94/29 10. 94/29 1 1. 94/29 12. 94/29 94/29 94/29 15. 94/29 16. 94/29 17. 94/29 18. 94/29 19. 9412920. Carribn N. Itriago A. Murillo M. Eljuri E. Fernandez A. Determination of calcium potassium magnesium iron copper and zinc in maternal milk by inductively coupled plasma atomic emission spec- trometry J.Anal. At. Spectrom. 1994 9 205. (Centro de Quim. Anal. Escuela de Quim. Facult. Ciencias Univ. Central de Venezuela P.O. Box 47102 Caracas 1041 -A Venezuela). Manninen P. K. G. Determination of extractable organic chlorine by electrothermal vaporization induc- tively coupled plasma mass spectrometry J. Anal. At. Spectrom. 1994 9 209. (VTT Reactor Lab. P.O. Box 200 Fin-021 5 1 Espoo Finland). Hu Y.-p. Zhang Z.-x. Zheng J.-g. Simulation of the vaporization process in inductively coupled plasma atomic emission spectrometry with a modified model using the Monte Carlo technique J. Anal. At. Spectrom. 1994 9 213. (Dept. Chem. Zhongshan Univ. Guangzhou 510275 China). Fernandez A. Murillo M. Carribn N. Mermet J.-M. Influence of operating conditions on the effects of acids in inductively coupled plasma atomic emission spec- trometry J.Anal. At. Spectrom. 1994 9 217. (Centro Quim. Anal. Escuela de Quim. Facult. Ciencias Univ. Central de Venezuela P.O. 47102 Caracas 1041-A Venezuela). Coedo A. G. Dorado M. T. Cobo I. G. Analysis of electrical arc furnace flue dusts by spark ablation inductively coupled plasma atomic emission spec- trometry J. Anal. At. Spectrom. 1994 9 223. (Centro Nacl. Investigaci6nes Metal. Consejo Superior de Investigaciones Cientificas Gregorio del Amo. 8 28040 Madrid Spain). Manickum C. K. Verbeek A. A. Determination of aluminium barium magnesium and manganese in Tea Leaf by slurry nebulization inductively coupled plasma atomic emission spectrometry J. Anal. At. Spectrom. 1994 9 227. (Dept. Chem. Univ.Natal P.O. Box 375 Pietermaritzburg 3200 South Africa). ValdCs-Hevia y Temprano M. C. Fernandez de la Campa M. R. Sanz-Medel A. Sensitive inductively coupled plasma atomic emission spectrometric determi- nation of cadmium by continuous alkylation with sodium tetraethylborate J. Anal. At. Spectrom. 1994 9 231. (Dept. Phys. Anal. Chem. Fac. Chem. Univ. Oviedo clJulian Claveria 8 33006-Oviedo Spain). Murillo M. Chirinos J. Use of emulsion systems for the determination of sulfur nickel and vanadium in heavy crude oil samples by inductively coupled plasma atomic emission spectrometry J. Anal. At. Spectrom. 1994 9 237. (Centro de Quim. Anal. Fac. Ciencias Univ. Central de Venezuela P.O. Box 47102 Caracas 1041 -A Venezuela). Kubova J. Nevoral V. Stregko V. Determination of rare earth elements in mineral waters by inductively coupled plasma atomic emission spectrometry J.Anal. At. Spectrom. 1994 9 241. (Fac. Nat. Sci. Comenius Univ. 842 15 Bratislava Slovakia). HlavaEek I. Hlavatkova I. Analysis of some low silicon content alloys by inductively coupled plasma atomic emission spectrometry J . Anal. At. Spectrom. 1994 9 245. (Analytica Co. U Elektry 650 194 05 Prague Czech Republic). HlavaEkovB I. Hlavatek I. Multi-element analysis of some high silicon content ferroalloys by inductively coupled plasma atomic emission spectrometry J. Anal. At. Spectrom. 1994 9 251. (Analytica Co. U Elektry 650 194 05 Prague Czech Republic). Flbrian K. Fischer W. Nickel H. Direct solid sample analysis of silicon carbide powders by direct current glow discharge and direct current arc emission spec- trometry J.Anal. At. Spectrom. 1994 9 257. (TUJOURNAL OF ANALYTICAL ATOMIC SPECTROMETRY OCTOBER 1994 VOL. 9 309 R 941292 1. 9412922. 9412923. 9412924. 941292 5. 9412926. 9412927. 9412928. 9412929. 9412930. 9412931. Kosice Dept. Chem. Letna 9 SK-04001 Kosice Slovakia). Webb C. Cooper C. B. 111 Zander A. T. Arnold J. T. Lile E. S. Anderson S. E. Direct solid sampling for analysis with inductively coupled plasma using a novel electronic spark source J. Anal. At. Spectrom. 1994 9 263. (Varian Assoc. Ginzton Res. Center Palo Alto CA 94303 USA). Krieger B. L. Kimber G. M. Selby M. Smith F. O. Turak E. E. Petty J. D. Peachey R. M. Automatic sample handling and calibration methods for analysis by flame atomic absorption spectrometry using discon- tinuous flow analysis J.Anal. At. Spectrom. 1994 9 267. (Anal. Spectrosc. Group Centre Instrum. Devl. Chem. Queensland Univ. Technol. G.P.O. Box 2434 Brisbane Queensland 4001 Australia). Foster R. D. Howe A. M. Development of a proposed international standard for determining arsenic in work- place air using hydride generation atomic absorption spectrometry J. Anal. At. Spectrom. 1994 9 273. (Occup. Med. Hyg. Lab. Health and Safety Exec. Broad La Sheffield UK S37HQ). Wrobel K. Gonzalez E. B. Sanz-Medel A. Progress with the speciation of aluminium and silicon in serum of chronic renal patients using atomic spectroscopic techniques J. Anal. At. Spectrom. 1994 9 281. (Dept. Phys. Anal. Chem. Univ. Oviedo c/Julian Claveria 8 33006 Oviedo Spain).Veber M. tujes K. GomiSEek S. Determination of selenium and arsenic in mineral waters with hydride generation atomic absorption spectrometry J. Anal. At. Spectrom. 1994 9 285. (Dept. Chem. Chem Technol. Univ. Ljubljana P.O.B. 537 Ljubljana SLO-61001 Slovenia). Lopez-Gonzalvez M. A. Gomez M. M. Camara C. Palacios M. A. On-line microwave oxidation for the determination of organoarsenic compounds by high- performance liquid chromatography-hydride gener- ation atomic absorption spectrometry J. Anal. At. Spectrom. 1994 9 291. (Dept. Quim. Anal. Fac. Ciencias Quim. Univ. Complutense de Madrid 28040 Madrid Spain). Emteborg H. Hadgu N. Baxter D. C. Quality control of a recently developed analytical method for the simultaneous determination of methylmercury and inorganic mercury in environmental and biological samples J.Anal. At. Spectrom. 1994 9 297. (Dept. Anal. Chem. Univ. Umei S-901 87 UmeA Sweden). Kratzer K. BeneS P. Sp6vaEkov5 V. Kolihovh U. Zilkova J. Determination of chemical forms of mercury in human hair by acid leaching and atomic absorption spectrometry J. Anal. At. Spectrom. 1994 9 303. (Dept. Nucl. Chem. Fac. Nucl. Sci. Phys. Eng. Czech Tech. Univ. in Prague 115 19 Prague 1 Brehova 7 Czech Republic). Burakov V. S. Xsaevich A. V. Misakov P. Ya. Naumenkov P. A. Raikov S. N. Intracavity laser spectroscopic method for determining trace amounts of iodine and barium in water and biological samples J. Anal. At. Spectrorn. 1994 9 307. (Inst. Mol. At. Phys. Acad. Sci. of Belarus Minsk 220072 Belarus). PCrez-Arantegui J.QuerrC G. Castillo J. R. Particle- induced X-ray emission thick-target analysis of inor- ganic materials in the determination of light elements J. Anal. At. Spectrom. 1994 9 311. (Dept. Quim. Anal. Fac. Ciencias Univ. Zaragoza 50009 Zaragoza Spain). Cheam V. Lechner J. Sekerka I. Desrosiers R. Direct determination of lead in sea-waters by laser- excited atomic fluorescence spectrometry J. Anal. At. Spectrom. 1994 9 315. (Environ. Canada Natl. Water Res. Inst. Res. and Appl. Branch Burlington Ontario Canada L7R 4A6). 9412932. 941293 3. 9412934. 9412935. 941293 6. 941293 7. 9412938. 9412939. 9412940. 9412941. 9412942. 9412943. 9412944. Katskov D. A. Shtepan A. M. McCrindle R. I. Marais P. J. J. G. Applications of a two-step atomizer and related techniques for investigating the processes of sample evaporation and atomization in electrother- mal atomic absorption spectrometry J.Anal. At. Spectrom. 1994 9 321. (State Inst. Appl. Chem. Dobrolyobova pr. 14 St. Petersburg 197198 Russia). Doner G. Akman S. Effect of cobalt chloride on the atomization of zinc in electrothermal atomic absorption spectrometry J. Anal. At. Spectrom. 1994 9 333. (Istanbul Tek. Univ. Fen-Edebiyat Fak. Kimya Bolumu 80626 Maslak/Istanbul Turkey). Manzoori J. L. Saleemi A. Determination of chro- mium in serum and lake water by electrothermal atomic absorption spectrometry using vanadium and molyb- denum modifier J. Anal. At. Spectrom. 1994 9 337. (Dept. Anal. Chem. Fac. Chem. Univ. Tabriz Tabriz Iran). Bruno S. N. F. Campos R. C. Curtius A.J. Determination of lead and arsenic in wines by electro- thermal atomic absorption spectrometry J. Anal. At. Spectrom. 1994 9 341. (Minist. Agric. Lab. Bebidas e Vinagres Rio de Janeiro Brazil). Cserfalvi T. Mezei P. Direct solution analysis by glow discharge electrolyte-cathode discharge spec- trometry J. Anal. At. Spectrom. 1994 9 345. (Aqua- Concorde R&D Lab. 1145 Budapest Bosnyak u. ll. Hungary). Weiss Z. New method of calibration for glow discharge optical emission spectrometry J. Anal. At. Spectrom. 1994 9 351. (Lab. GDOS-LECO Appl. Lab. Dept. Metall. Mater. Sci. Univ. West Bohemia Americka 42 306 14 Plzeii Czech Republic). Quentmeier A. Role of depth resolution in quantitative glow discharge optical emission spectrometry depth analysis J. Anal. At. Spectrom. 1994 9 355.(Inst. Spektrochem. Angew. Spektrosk. Postfach 10 13 52 D-44013 Dortmund Germany). Payling R. Brown N. V. Gower S. A. Correcting for background and inter-element effects in glow discharge optical emission spectrometry J. Anal. At. Spectrom. 1994 9 363. (BHP Steel-Sheet & Coil Prod. Div. Res. Technol. Centre PO Box 77 Port Kembla NSW 2505 Australia). Jones D. G. Payling R. Gower S. A. Boge E. M. Analysis of pigmented polymer coatings with radio- frequency glow discharge optical emission spectrometry J. Anal. At. Spectrom. 1994 9 369. (BHP Steel Sheet and Coil Prod. Div. Res. Technol. Centre PO Box 77 Port Kembla New South Wales 2505 Australia). Fischer W. Naoumidis A Nickel H. Effects of a controlled addition of nitrogen and oxygen to argon on the analytical parameters of glow discharge optical emission spectrometry J.Anal. At. Spectrom. 1994 9 375. (Res. Centre Julich GmbH Inst. Mater. for Energy Systems D-52425 Julich Germany). Outred M. Rummeli M. H. Steers E. B. M. Microwave boosted glow discharge source using a slab- line cavity J. Anal. At. Spectrom. 1994 9 381. (SECEAP Univ. North London Holloway Rd. London UK N7 8DB). Betti M. Rasmussen G. Hiernaut T. Koch L. Milton D. M. P. Hutton R. C. Adaptation of a glow discharge mass spectrometer in a glove-box for the analysis of nuclear materials J. Anal. At. Spectrom. 1994,9 385. (Comm. Eur. Commun. Joint Res. Centre Inst. Transuranium Elements Postfach 2340 76125 Karlsruhe Germany). GrCgoire D. C. Lee J. Determination of cadmium and zinc isotope ratios in sheep’s blood and organ tissue by electrothermal vaporization inductively310R 9412945.9412946. 9412947. 9412948. 9412949. 9412950. 941295 1. 9412952. 9412953. 9412954. 941295 5. 9412956. JOURNAL OF ANALYTICAL ATOMIC SPECTROMETRY OCTOBER 1994 VOL. 9 coupled plasma mass spectrometry J . Anal. At. Spectrorn. 1994,9 393. (Geol. Surv. Canada 601 Booth St. Ottawa Canada K1A OE8). Alves L. C. Minnich M. G. Wiederin D. R. Houk R. S. Removal of organic solvents by cryogenic desolvation in inductively coupled plasma mass spec- trometry J. Anal. At. Spectrorn. 1994 9 399. (Ames Lab. US Dept. Energy Dept. Chem. Iowa State Univ. Ames IA 50011 USA). Tsalev D. L. Electrothermal atomic absorption spec- trometry in occupational and environmental health practice-a decade of progress and establishment a review J.Anal. At. Spectrorn. 1994,9,405. (Fac. Chem. Univ. Sofia Sofia 1126 Bulgaria). Scheie A. J. Holcombe J. A. Multi-element detection using second surface trapping with electrothermal vaporization mass spectrometry J. Anal. At. Spectrorn. 1994 9 415. (Dept. Chem. Biochem. Univ. Texas at Austin Austin TX 78712 USA). Smith C. M. M. Harnly J. M. Moulton G. P. O’Haver T. C. High current pulsing of a xenon arc lamp for electrothermal atomic absorption spectrometry using a linear photodiode array J. Anal. At. Spectrorn. 1994 9 419. (USDA ARS BHNRC NCL Bldg. 161 BARC-East Beltsville MD 20705 USA). McAllister T. Equilibrium and mass spectrometry of nitrate decomposition in electrothermal atomic absorp- tion spectrometry J. Anal.At. Spectrom. 1994 9 427. (CSIRO Mater. Sci. Technol. Locked Bag 33 Clayton Victoria 3168 Australia). Katskov D. A. Schwarzer R. Marais P. J. J. G. McCrindle R. I. Use of a furnace with a graphite filter for electrothermal atomic absorption spectrometry J. Anal. At. Spectrom. 1994 9 431. (Centre Appl. Chem. Tech. Pretoria Private Bag X680 Pretoria South Africa). Heitmann U. Sy T. Hese A. Schoknecht G. High- sensitivity detection of selenium and arsenic by laser- excited atomic fluorescence spectrometry using electro- thermal atomization J. Anal. At. Spectrorn. 1994 9 437. (Inst. Strahlungs- und Kernphysik Tech. Univ. Berlin HardenbergstraDe 36 D- 10623 Berlin Germany). Tittarelli P. Biffi C. Kmetov V. Vaporization of silicon and germanium as molecular species in electro- thermal atomizers J.Anal. At. Spectrom. 1994 9 443. (Stazione Sperimentale Combust. Viale A. De Gasperi 3 20097 San Donato Milanese MI Italy). Hinds M. W. Kogan V. V. Determination of silicon in fine gold by solution and solid sample graphite furnace atomic absorption spectrometry and inductively coupled plasma atomic emission spectrometry J. Anal. At. Spectrorn. 1994 9 451. (Royal Canadian Mint 320 Sussex Dr. Ottawa Ontario Canada K1A OG8). Patriarca M. Fell G. S. Determination of nickel in serum of haemodialysed patients by means of electro- thermal atomic absorption spectrometry with deuterium background correction J. Anal. At. Spectrom. 1994 9 457. (Dept. Pathol. Biochem. Univ. Glasgow Royal Infirmary Glasgow G4 OSF UK). Hauptkorn S. Schneider G.Krivan V. Determination of silicon in titanium dioxide and zirconium dioxide by electrothermal atomic absorption spectrometry using the slurry sampling technique J. Anal. At. Spectrom. 1994 9 463. (Sektion Anal. Hochstreinigung der Univ. Ulm Albert-Einstein-Allee 11 D-89069 Ulm Germany). Bermejo-Barrera P. Barciela-Alonso C. Aboal- Somoza M. Bermejo-Barrera A. Slurry sampling for the determination of lead in marine sediments by electrothermal atomic absorption spectrometry using palladium-magnesium nitrate as a chemical modifier 941295 7. 94/29 58. 9412959. 9412960. 941296 1. 9412962. 9412963. 9412964. 9412965. 94/2966. 9412967. 9412968. J . Anal. At. Spectrom. 1994,9,469. (Dept. Anal. Chem. Nutr. Bromatol. Fac. Chem. Univ. Santiago de Compostela Spain). Cabon J.Y. Le Bihan A. Direct determination of zinc in sea-water using electrothermal atomic absorp- tion spectrometry with zeeman-effect background cor- rection effects of chemical and spectral interferences J. Anal. At. Spectrom. 1994 9 477. (Unite Rech. Assoc. au CNRS No. 322 Univ. Bretagne Occidentale 6 Ave. Le Gorgeu 29275 Bfest-Ckdex France). Bermejo-Barrera P. Moreda-Piiieiro A. Aboal- Somoza M. Moreda-Piiieiro J. Bermejo-Barrera A. Indirect determination of iodide as an Hg,I complex by electrothermal atomic absorption spectrometry J. Anal. At. Spectrorn. 1994 9,483. (Dept. Anal. Chem. Nutr. Bromatol. Fac. Chem. Univ. Santiago de Compostela 15706-Santiago de Compostela Spain). Imai S. Sturgeon R. E. Gas-phase re-distribution of analyte species in the integrated contact cuvette furnace atomization plasma emission spectrometry source J.Anal. At. Spectrorn. 1994 9 493. (Dept. Chem. Joetsu Univ. Educ. Joetsu Niigata 943 Japan). Su E. G. Michel R. G. Choice of fluorescence wavelengths for the determination of trace amounts of chlorine by graphite furnace laser-excited molecular fluorescence spectrometry of indium monochloride J. Anal. At. Spectrorn. 1994 9 501. (Dept. Chem. Univ. Connecticut 215 Glenbrook Road Storrs CT Xiao G. Beauchemin D. Reduction of matrix effects and mass discrimination in inductively coupled plasma mass spectrometry with optimized argon-nitrogen plasmas J. Anal. At. Spectrom. 1994 9 509. (Queen’s Univ. Dept. Chem. Kingston Ontario Canada K7L 3N6). Rivoldini A. Fadda S. Inductively coupled plasma mass spectrometric determination of low-level rare earth elements in rocks using potassium-based fluxes for sample decomposition J.Anal. At. Spectrom. 1994 9 519. (1st. Giacimenti Minerari Fac. Ingegneria Univ. Cagliari Piazza D’Armi 09 123 Cagliari Italy). Walter S. Hatterer A Atomic absorption another way for phase transition characterization J. Anal. At. Spectrom. 1994 9 525. (Ecole Natl. Superieure de Chim. de Mulhouse Lab. C.M.A. 3 Rue Alfred Werner F68093 Mulhouse Cedex France). Kurfiirst U. Pauwels J. Spectral interference on the lead 283.3 nm line in Zeeman-effect atomic absorption spectrometry J. Anal. At. Spectrorn. 1994,9 531. (Univ. Fulda (Fachhochschule) Marquardstr. 35 D-36039 Fulda Germany). Bruhn C. G. Rodriguez A. A. Barrios C. Jaramillo V. H. Becerra J.Gonzalez U. Gras N. T. Reyes O. Salud S. Determination of total mercury in scalp hair of humans by gold amalgamation cold vapour atomic absorption spectrometry J . Anal. At. Spectrom. 1994 9 535. (Dept. Anal. Instrum. and Dept. Farmacia Fac. Farmacia Univ. Concepcion Casilla 237 Concepcion Chile). Avila A. K. Curtius A. J. Determination of silver in waters and soil by electrothermal atomic absorption spectrometry after complexation and sorption on carbon J. Anal. At. Spectrorn. 1994 9 543. (Dept. Quim. Anal. Univ. Federal do Rio de Janeiro Ilha do Fundiio Rio de Janeiro Brazil). Elmahadi H. A. M. Greenway G. M. Speciation and preconcentration of trace elements with immobilized algae for atomic absorption spectrophotometric detec- tion J. Anal. At. Spectrorn. 1994 9 547.(Sch. Chem. Univ. Hull Hull UK HU67RX). Garcia I. L. Viiias P. Cbrdoba M. H. Calibration in flame atomic absorption spectrometry using a single 06269-3060 USA).JOURNAL OF ANALYTICAL ATOMIC SPECTROMETRY OCTOBER 1994 VOL. 9 311R 9412969. 9412970. 9412971. 9412972. 941297 3. 94/2974. 9412975. 9412976. 9412977. 9412978. 9412979. 9412980. standard and a gradient technique J . Anal. At. Spectrom. 1994 9 553. (Dept. Anal. Chem. Fac. Chemistry Univ. Murcia E-30071-Murcia Spain). Greenfield S. Inductively coupled plasmas in atomic fluorescence spectrometry J. Anal. At. Spectrom. 1994 9 565. (Dept. Chem. Univ. Technol. Loughborough Leicestershire UK LE113TU). Horlick G. Inductively coupled plasma mass spec- trometry has it matured already? J. Anal. At. Spectrom. 1994 9 593.(Dept. Chem. Univ. Alberta Edmonton Alberta Canada T6G 2G2). Lord C. J. 111 Determination of lead and lead isotope ratios in gasoline by inductively coupled plasma mass spectrometry J. Anal. At. Spectrom. 1994 9 599. (Phillips Petroleum Co. Res. and Develop. Bartlesville OK 74004 USA). GrCgoire D. C. Miller-Ihli N. J. Sturgeon R. E. Direct analysis of solids by ultrasonic slurry electrother- mal vaporization inductively coupled plasma mass spectrometry J. Anal. At. Spectrom. 1994 9 605. (Geol. Surv. Canada 601 Booth St. Ottawa Ontario Canada K1A OE8). Ebdon L. Fisher A. S. Worsfold P. J. Determination of arsenic chromium selenium and vanadium in biological samples by inductively coupled plasma mass spectrometry using on-line elimination of interference and preconcentration by flow injection J.Anal. At. Spectrom. 1994 9 611. (Plymouth Anal. Chem. Res. Unit Dept. Environ. Sci. Univ. Plymouth Drake Circus Plymouth Devon UK PL4 8AA). Dean J. R. Ebdon L. Foulkes M. E. Crews H. M. Massey R. C. Determination of the growth promoter 4-hydroxy-3-nitrophenyl arsonic acid in chicken tissue by coupled high-performance liquid chromatography- inductively coupled plasma mass spectrometry J. Anal. At. Spectrom. 1994 9 615. (Plymouth Anal. Chem. Res. Unit Univ. Plymouth Drake Circus Plymouth Devon UK PL48AA). Prudnikov E. D. Elgersma J. W. Smit H. C. Theoretical calculation of the standard deviation and detection limit in inductively coupled plasma atomic emission spectrometry J . Anal. At. Spectrom. 1994 9 619. (Earth's Crust Inst.State Univ. St. Petersburg 199034 Russia). Fischer J. L. Rademeyer C. J. Direct determination of metals in oils by inductively coupled plasma atomic emission spectrometry using high temperature nebuliz- ation J . Anal. At. Spectrom. 1994,9 623. (Dept. Chem. Univ. Pretoria Pretoria 0002 South Africa) Duan Y.-x. Du X.-g. Jin Q.-h. Comparative studies of surfatron and microwave plasma torch sources for determination of mercury by atomic emission spec- trometry J. Anal. At. Spectrom. 1994 9 629. (Dept. Chem. Jilin Univ. Changchun 130023 China). Cordos E. A. Anghel S. D. Frentiu T. Popescu A. Capacitively coupled plasma with tip-ring electrode geometry for atomic emission spectrometry. Analytical performance and matrix effect of sodium chloride and potassium chloride J.Anal. At. Spectrom. 1994 9 635. (Dept. Chem. Univ. Cluj 3400 Cluj-Napoca Romania). Gilmutdinov A. Kh. Nagulin K. Yu Zakharov Yu. A. Analytical measurement in electrothermal atomic absorption spectrometry-how correct is it? J. Anal. At. Spectrom. 1994 9 643. (Dept. Phys. Univ. Kazan 18 Lenin Str. Kazan 420008 Russia). Lbpez J. C. Reija C. Montoro R. Cervera M. L. Guardia M. de la Determination of inorganic arsenic in seafood products by microwave-assisted distillation 941298 1. 9412982. 9412983. 9412984. 9412985. 9412986. 9412987. 9412988. 9412989. 9412990. 9412991. 9412992. and atomic absorption spectrometry J. Anal. At. Spectrom. 1994 9 651. (Inst. Agroquimi. Tecnol. Alimentos (CSIC) Jaime Roig 11 46010 Valencia Spain). Arruda M. A. Z. Gallego M.Valcarcel M. Determination of selenium in fruit juices by flow injection electrothermal atomization atomic absorption spectrometry J. Anal. At. Spectrom. 1994 9,657. (Dept. Anal. Chem. Fac. Sci. Univ. Cordoba E-14004 Cordoba Spain). Mendes P. C. S. Santelli R. E. Gallego M. Valcarcel M. Determination of nickel in rocks by use of a continuous precipitation-preconcentration system coupled on-line to a flame atomic absorption spec- trometer J. Anal. At. Spectrom. 1994 9 663. (Dept. Geochem. Univ. Fed. Fluminense Niteroi-24020-007 Brazil). Su K.-d. Lin KA. Ion enhancement effect of laser- enhanced ionization induced by one-step excitation and two-step excitation Appl. Spectrosc. 1994 48 241. (Dept. Chem. Natl. Taiwan Univ. Taipei Taiwan 10764). Morgan C. A. Davis C.L. Smith B. W. Winefordner J. D. Evaluation of a microcavity hollow cathode discharge emission source Appl. Spectrosc. 1994 48 261. (Dept. Chem. Univ. Florida Gainesville Florida 32608 USA). Wichems D. N. Jones B. T. Simple direct coupling for a flame detector and an HPLC column Appl. Spectrosc. 1994 48 273. (Dept. Chem. Wake Forest Univ. Winston-Salem NC 27109 USA). Seely J. F. Brown C. M. Multilayer-coated grating spectrometer operating in the extreme-ultraviolet region and based on the Seya-Namioka mount Appl. Opt. 1993 32 1288. (E. 0. Hulburt Center for Space Res. Naval Res. Lab. Washington D.C. 20375 USA). Kauffman R. L. Phillion D. W. Spitzer R. C. X-ray production z 13 nm from laser-produced plasmas for projection X-ray lithography applications Appl.Opt. 1993 32 6897. (Lawrence Livermore Natl. Lab. P.O. Box 5508 Livermore CA 94550 USA). Bowyer oS. Continuous emission source covering the 50-300A band Appl. Opt. 1993 32 6930. (Center for EUV Astrophy. Univ. California at Berkeley Berkeley CA 94720 USA). Richardson M. Silfvast W. T. Bender H. A. Hanzo A. Yanovsky V. P. Jin F. Thorpe J. Characterization and control of laser plasma flux parameters for soft X-ray projection lithography Appl. Opt. 1993 32 6901. (Center for Res. in Electro-Opt. and Lasers Univ. Florida Orlando FL 32826-357 USA). Cerjan C. X-ray plasma source design simulations Appl. Opt. 1993 32 6911. (Lawrence Livermore Natl. Lab. P.O. Box 808 L438 Livermore CA 94550 USA). Mende B. Claflin E. S. Rairden R. L. Swenson G. R. Hadamard spectroscopy with two-dimensional detecting array Appl.Opt. 1993 32 7095. (Lockheed Palo Alto Res. Lab. 3251 Hanover St. Palo Alto CA 94304 USA). McMillin B. K. Palmer J. L. Hanson R. K. Temporally resolved two-line fluorescence imaging of NO temperature in a transverse jet in a supersonic cross flow Appl. Opt. 1993,32,7532. (High Temperature GasDynamics Lab. Dept. Mech. Eng. Stanford Univ. Stanford CA 94305 USA).312R JOURNAL OF ANALYTICAL ATOMIC SPECTROMETRY OCTOBER 1994 VOL. 9 94/2993. Robichaud J. L. Wong W. K. Van Tassel R. A. 94/2994. Grinstead J. H. Laufer G. Krauss R. H. McDaniel Evaluation of a Hadamard-coded photodiode-array Jr. J. C. Calibration source for OH laser-induced spectrometer under low illumination Appl. Opt. 1994 fluorescence density measurements with thermally dis- 33 75.(Optical Systems Solutions Inc. 50 Dalton St. sociated H,O in atmospheric air Appl. Opt. 1994 33 Tewksbury MA 01876 USA). 1115. (Aerospace Res. Lab. Univ. Virginia 570 Edgemont Road Charlottesville VA 22903 USA). WHY PAY A FORTUNE GBC Scientific Equipment U.K Ltd. 13 Frederick Sanger Road The Surrey Research Park Guildford Surrey GU2 5YD Tel 0483 304988 Fax 0483 303071 Circle 001 for further information

ISSN:0267-9477    

DOI:10.1039/JA994090307R

出版商:RSC    

年代:1994

数据来源: RSC

摘要:

JOURNAL OF ANALYTICAL ATOMIC SPECTROMETRY OCTOBER 1994 VOL. 9 1087 Electron Density and Hydrogen Distribution in an Ethanol-loaded Inductively Coupled Plasma Robert 1. McCrindle Department of Chemistry and Physics Technikon Pretoria Private Bag X680 Pretoria South Africa Corn el i us J . Rade me ye r Department of Chemistry University of Pretoria Lynnwood Pretoria South Africa The effect of an ethanol-loaded sample solution when aspirated into an inductively coupled plasma was investigated. It was found that increasing concentrations of ethanol resulted in a proportional increase in the intensity of the Ha line. The electron density as calculated both with and without Abel inversion also increased with increasing ethanol concentration. Mass flow rates of the aerosol increased with an increase in the ethanol concentration.This resulted in a relative mean drop-size distribution which increased from 2 ym for water to 24.5 pm for 25% ethanol. A mechanism is suggested to explain the increases encountered. Keywords Inductively coupled plasma; electron number density; aerosol mass flow rate; aerosol drop size distribution; organic solvent Ethanol remains a useful organic solvent for extraction pro- cesses. It has also been found to improve detection limits in inductively coupled plasma atomic emission spectroscopy (ICP-AES).'y2 This enhancement has been attributed to the changes the ethanol (or other organic solvent) makes to the sample aspiration rate and median aerosol drop diameter.3 Changes in surface tension and viscosity due to the solvent alter the uptake rate.The dependence of the median drop diameter on the surface tension viscosity density and solvent flow rate for a particular nebulizer and gas flow is given by the equation of Nukiyama and Tana~awa:~ d = 585/V(/(a/p)'.'+ 597(~/[/(ap]0~5)0~45( 103Q1/Q,)'.' (1) where V is the difference in velocities of the gas and liquids entering the nebulizer c p and q the surface tension density and viscosity of the liquid respectively and Q and Q the volume flow rates of the liquid and gas respectively. The equation determines the Sauter median diameter of the aerosol d,. The mass median diameter d for unit density and spherical droplets is identical to the Sauter diameter. The density of ethanol is 0.7893 g ml-' but due to dilution will approach unity and eqn.(1) can be used to explain nebulizer performance. Ethanol remains one of the cheapest organic solvents and because of its polarity its importance in the laboratory cannot be underestimated. Unfortunately if the concentration of etha- nol in the water is increased above a certain limit (in the case of the instrument used above 25% v/v ethanol) the plasma is extinguished. The tolerance of the ICP to the percentage of ethanol in the sample before the plasma is extinguished depends on the type of r.f. generator and the power a ~ p l i e d . ~ This limitation results in other organic solvents like 4-methyl pentan-2-one (IBMK) and xylene being used more extensively than ethanol. Inductively coupled plasmas exhibit deviations from local thermodynamic equilibrium.6 Hence a single value for tem- perature does not exist as it is dependent on the method of calc~lation.~ The free electron concentration or electron den- sity is an important parameter because the free electrons are the ones most likely to cause analyte excitation and ionization.8 Blades and Caughlin' determined the excitation temperatures and electron densities in an argon plasma using xylene as organic solvent.They found that the excitation conditions for the xylene plasma at 1.75 kW were similar to those of water when the power was 0.5 kW less. The characteristics of the aerosol as it reaches the plasma are governed by the tertiary drop-size distribution. Impaction gravitational settling turbulence and evaporation influence the changes that the primary drops undergo before reaching the plasma.Major changes are also caused by the nebulizer gas flow and the design of the spray chamber. The pre-atomization drop size or tertiary drop size is determined primarily by the evaporation of the solvent from the surface according to Boorn et a1." In this work the diameter d at time t(s) after formation of the tertiary drops was calculated from (2) where do is the initial drop diameter (pm) and E the evaporation factor given by d = (d,3 - E p 3 E=48 D o P M ~ ( ~ R T ) - ~ (3) where D is the diffusion constant for the solvent vapour P the saturated pressure M the molecular mass p the density c the surface tension R the gas constant and T the absolute temperature. From the calculated evaporation curves it was shown" that the amount of solvent evaporation even in a short period of time (tenths of a second) caused a decrease at the micrometre level in the diameter.The actual particle size distribution can be determined using microscope slide^.^ A more recent method is that of using a laser particle analyser." To understand the mechanism of the reactions which are taking place within the plasma while ethanol enriched solutions are being aspirated it is necessary to determine the electron density n,. Although this has been done with another organic solvent,' xylene ethanol has different chemical characteristics to xylene. It was therefore appropriate to compare the n distribution in ethanol plasmas with that found in other solvents. Since the ICP is extinguished possibly by either large particles or by a high mass transport rate these parameters were also investigated.Experimental Reagents Analytical-reagent grade ethanol was used. Apparatus Spectrometer A Hilger Analytical microprocessor-controlled 1.5 m spec- trometer was used. It was equipped with a Czerny-Turner grating (1800 lines mm-l). The monochrometer was place on1088 JOURNAL OF ANALYTICAL ATOMIC SPECTROMETRY OCTOBER 1994 VOL. 9 an xz-table and focused through a grid onto the torch by means of a helium laser. Particle size measurement A Micro Laser Particle Spectrometer pLPS (Particle Measuring Systems) was used for drop-size determination. ICP spectrometer The ICP used was a Spectro Analytical Instruments Spectroflame operating under the conditions given in Table 1. It was fitted with a standard cross-flow nebulizer and a glass Scott spray chamber.Read-out system A Hamamatsu photomultiplier tube (PMT) R955 was used as detector. It was powered by a Brandenburg Power Supply Model 4728 and the analogue signal was sent through a Philips PM 3350 oscilloscope to an analogue-to-digital con- verter (PCL-711s) in an IBM-compatible PC computer. The voltage applied to the PMT was 850V in the case of HP readings and 600 V when Ha was measured. Procedure Electron density determination The n has been determined from the Stark broadening of the HP line (486.1 nm) of hydrogen by several ~ o r k e r s . ~ ~ ' ~ - ~ ~ The formula of Grieg et a l l 5 is as follows n = [C + C1 (1nAA) + C,( lnAll)2 + C3( lnAll)3]Al13/21013 (4) where the coefficients are Co = 36.84 C1 = - 1.430 C2 = - 0.133 and C3=0.0089. The All being the full width at half maximum (FWHM) of the HP line.Eqn. (4) was applied both with and without Abel inversion. Although the contribution from Doppler and instrumental broadening are relatively small in comparison with the Stark broadening correction was made for these effects. The temperature was assumed to be 5500K while both the entrance and exit slits were 15 pm. Correction was then calculated according to the method used by Xiao et aL14 Measurement of the emission intensities The spectrometer was positioned so that the emission from just above the load coil (z=O) and at the extreme edge of the torch (x= -8 mm) could be measured. Both the HP and Ha (656.2 nm) lines were scanned while water was being nebulized and their intensities were recorded. This recording was made for both the forward and reverse direction of scan.The lateral position of the spectrometer was changed (moved through 1 mm) and the scans repeated. This process was continued until the emission from the other edge of the torch was measured (x=8 mm). The height of the spectrometer was Table 1 ICP operating parameters Forward power/kW Intermediate plasma gas flow Outer gas flow rate/l min-' Nebulizer Aerosol carrier gas flow rate/ 1 min-' Sample uptake rate/ml min-' Spray chamber Torch rate/l min-' 1.3 1 14 Crossflow 0.4 2.5 Scott glass (Demountable quartz Fassel type); 16mm i.d. changed so that the recording could be made from a position 5 mm above the load coil (z= 5). The lateral measurements were then repeated.This procedure was carried out for z = 10 and z= 15. The entire procedure was repeated for solutions containing 5 10 15 20 and 25% v/v ethanol. The digital signal of the emission intensities received was recorded in ASC I1 on floppy disks using a program written for this purpose. The data were transferred to a software package Asystant (Asyst Software Technologies). The data were transformed and the array thus obtained smoothed by applying a Blackman window low pass filter to each row. The cut-off frequency of the filter was set and entered in units of cycles per point. The wave generated by the program was either processed further in that the FWHM was calculated in order to determine the electron density or the wave data was applied to a program for calculating the Abel inversion.In the case of the HE lines only the magnitude of the intensity was determined. Abel inversion The Abel integral inversion was calculated according to the technique of Freeman and Katz.16 The intensity data I ( x i ) were obtained for equidistant lateral xi-coordinates from the cylindrical symmetric source. The wavelength index for the peak of the spectral line was determined for each lateral position and the wavelength distributions synchronized. The maximum and minimum wavelength indices that were used in the integral inversion calculation were determined from this skewness. The radial origin of the plasma was determined from symmetry considerations of the lateral intensities at the peak wavelength of the spectral line.The Abel integral inversion was then calculated for each wavelength by means of fitting a sixth order polynomial over the lateral intensities at that particular wavelength and a summation process to calculate the radial intensities at the different radial indices for that wavelength channel. Drop-size distribution The aerosol emitted from the sample injection tube of the ICP torch was transferred to the measuring cell of the laser particle counter by means of a plastic tube of internal diameter 5 mm. It was found that the aerosol flow was so dense that the laser beam was totally absorbed by the cloud. In order to obtain comparative measurements the aerosol cloud had to be diluted. This dilution was achieved by inserting a T-piece into the tube from the sample injector immediately above the tip of the injector and passing argon at a fixed flow rate through one end of the T.This diluted the aerosol which now could be measured. The drop-size distribution was measured for water and the samples containing increasing amounts of ethanol. MassJow rate The mass flow rate was determined by passing the aerosol from the sample injection tube into a U-tube containing anhydrous calcium chloride. The aerosol was collected in the U-tube for 2 min and the mass determined gravimetrically. This gravimetric determination was carried out for the water and ethanol samples. Results and Discussion Hydrogen Intensity During the determination of electron density (taking measure- ments of the Ha and HP lines) it became apparent that the magnitude of the signal increased with increasing concen- trations of ethanol.The magnitude of the Ha line was used to determine relative intensities for each ethanol concentration because of its sensitivity. The magnitude of relative intensitiesJOURNAL OF ANALYTICAL ATOMIC SPECTROMETRY OCTOBER 1994 VOL. 9 1089 of the various concentrations of ethanol at the positions measured is given in Fig. 1. As each measurement was made under identical conditions they can be compared with each other. The intensity normalized to 100 increased as the ethanol concentration was increased since by moving from water to 25% ethanol (the maximum ethanol concentration) a 4-fold increase was obtained. As could be expected the intensity decreased away from the load coil up to approxi- mately 4-fold (Fig.1). Electron Density The method used to determine the FWHM involved recording a number of data points as the line was scanned. The program Asystant allows 256 data points to be evaluated in one determination. The limited number of data points together with the problems associated with smoothing may have led to an inherent error in the system. In the smoothing process the peak may be reduced in size and this will influence the FWHM. The magnitude of this error is dependent upon the amount of smoothing required [i.e. the signal-to-noise ratio (S/Nj] and consequently was larger for peaks with small S/N values. In the extreme case an error of 15% was possible which decreased to 7.5% when a large S/N was obtained. The uncertainties due to the application of the Abel inversion have been discussed previously by Cremers and Birkebak.l7 Even taking these possible errors into consideration the values calculated for n and given in Fig.2,(a)-(d) are in agreement with those found in the l i t e r a t ~ r e . ' ~ . ~ ~ The values that were obtained from Abel inversion were checked by determining the n for z = 0 without inversion. The results can be found in Fig. 3. From the results presented in Fig. 2(a)-(d) and Fig. 3 it can be seen that n increased as the concentration of ethanol in the sample solution was increased. A further observation can be made from studying Fig. 2. The n as shown in Fig. 2(a) reached a maximum 5 mm from the centre of the torch when water was aspirated. The position of this maximum changed as the ethanol concentration was increased and was situated in the centre of the torch when the highest concentration ie.25% ethanol was aspirated. Similar observations can be made from Fig. 2(bj (c) and (d) when z=5 10 and 15 mm. These changes in n can also be noted in Fig. 3 but since the results have not been inverted they are not as easily evaluated. The increase in n for a 25% ethanol solution as compared with water was 1.44. 120 I I I I I 0 5 10 15 He i g ht/m m Fig. 1 Relative intensity of the Ha line at the centre of the torch of various concentrations of ethanol for various positions above the load coil Mass Transfer Rate The values obtained for the mass of aerosol aspirated for 2 min for each of the water and ethanol solutions are represented graphically in Fig.4. An increase of 5% in the concentration of ethanol caused very little increase in the mass flow rate (43.6-43.8 mg min-l). The mass flow rate then increased as the ethanol concentration increased as can be seen in Fig. 4. Drop-Size Distribution The drop-size distribution for the various concentrations of ethanol investigated are given in Fig. 5. The laser particle size analyser was limited in that it could only measure up to a magnitude of 4.5 pm. Clearly the maximum count for the solution containing 25% ethanol is beyond or at this limit. It is not suggested by the authors that these be considered actual distributions because of the dilution of the aerosol required in making these measurements. Furthermore the additional tubing and the T-piece would have altered the drop size of the aerosol. For the purpose of comparing the results obtained for the different concentrations of ethanol the measurements will suffice since they were made under identical conditions.The tendency of the mean drop size to increase as the concentration of the ethanol increased is clearly demonstrated (Fig. 5). This finding that the mean drop size increased with increas- ing ethanol concentration is in agreement with the calculations of Todorovic and Vidovic from eqn. ( l ) . 1 9 The gas flow rate in an ICP is lower than for an atomic absorption spectrometer which results in the first term in eqn. (1) being less significant. With an increase in the aspiration rate the ratio QI/Q increased resulting in larger drops. This increase in droplet diameter with aspiration rate has been calculated and plotted by Boorn and Browner.20 This is contrary to the assumption that organic solvents increase transport efficiency by reducing the primary mean drop diameter.3 The transport efficiency can be influenced by the evaporation factor given in eqn.(3) but the drop-size distribution was found to increase with an increase in ethanol. Hence the evaporation factor can be discounted as an important contributor in determining the particle-size distribution. The increase in the concentration of hydrogen in the plasma as ethanol was added can be readily deduced from the increase in intensity of the Ha line (and the HP line not given). The increased hydrogen therefore can only be a breakdown prod- uct of the ethanol in the plasma.The increase in n associated with the increase in hydrogen concentration can result from one of two sources; either the breakdown of the ethanol to produce hydrogen resulting in an increase in free electrons or the hydrogen produced by the breakdown is more easily ionizable by the r.f. source and an increase in n results. From the ionization potentials of hydrogen and argon it would seem likely that if the hydrogen was present in an atomic form it would undergo ionization more readily than argon. The ther- mal breakdown of ethanol is unlikely to be such that n will increase. The following mechanism is proposed. The initial thermal decomposition of the ethanol will result in the release of CO and OH 2C2H20H + 702+4C0 + 120H The CO and OH react to form C 0 2 and atomic hydrogen CO +OH+CO,+ H and it is the atomic hydrogen produced in this manner that causes both the increase in intensity of the Ha and HP lines and n,.The proposed mechanism is substantiated by the findings of Goldfarb and Goldfarb21 who reported that injec- tion of an alcohol aerosol produced quantitatively the same effect as the introduction of gaseous hydrocarbons or carbon oxides.21 Also the consumption in this way of the OH radical would explain why when organic solvents are used the OH band is very weak even when alcohols are aspirated.,'1090 JOURNAL OF ANALYTICAL ATOMIC SPECTROMETRY OCTOBER 1994 VOL. 9 12 (a) 25% 10 8 6 4 ’” Pb’ 25% d I 0 5 I I I 1 I I I 1 0 1 2 3 4 5 6 7 0 1 2 3 4 5 6 7 Radial position/rnrn Fig. 2 Radial n of the plasma across the torch for increasing ethanol concentration where (a) z=Omm (b) z=5 mm (c) z=10mm and (d)z=15mm r 1 0% 2t I I I 1 I I I 0 -8 -6 -4 - 2 0 2 4 6 8 Lateral position/mrn Fig.3 concentrations of ethanol; z = 0 mm Lateral n of the plasma across the torch for increasing Conclusions The effect of ethanol loading in the solution being aspirated into an ICP has clearly been demonstrated.The ethanol caused the following to increase (i) n,; (ii) the intensity of the hydrogen lines; (iii) the mass flow rate of the aerosol; and (iv) the mean drop size. 0.04 0.06 5 0 2 4 6 8 10 12 14 16 18 20 22 24 Ethanol (%) Fig.4 Mass of aerosol released from the sample injection tube in 2 min for increasing concentrations of ethanol As the electron density increased its maximum value moved from near the edge of the torch closer to the centre of the plume.This was also the region where the greatest concen- tration of sample solution was to be found after aspiration. It is suggested that the increase in n resulted from the production of atomic hydrogen from the thermal decompo- sition products of ethanol. The atomic hydrogen was respon- sible for the increase in the intensity of the hydrogen lines and because of the lower ionization potential for hydrogen than for argon the increase in a,. The increase in the drop-size distribution resulted from anJOURNAL OF ANALYTICAL ATOMIC SPECTROMETRY OCTOBER 1994 VOL. 9 1091 6 1 0 0.3 0.8 1.3 1.8 2.3 2.8 3.3 3.8 4.3 Drop size/pm Fig. 5 concentration Drop-size distribution of the ICP aerosol for increasing ethanol increase in the mass flow rate and this is the fundamental cause of the plasma being extinguished.References 1 Benli H. Yang J. Pei A. Zeng X. and Boumans P. W. J. M. Spectrochim. Acta Part B 1991 46 407. 2 Benli H. Spectrochim. Acta Part B 1983 38 81. 3 Greenfield S. McGeachim H. McD. and Smith P. B. Anal. Chim. Acta 1976 84 67. 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 Nukiyama S. and Tanasawa Y. Trans. SOC. Mech. Eng. Tokyo 1938-1940 4 5 and 6 (transl. Hope E. Experiments on the Atomization of Liquids in an Air Stream Defence Research Board Dept. of National Defence Ottawa Canada 1950). Boorn A. W. and Browner R. F. Inductively Coupled Plasma Spectroscopy Part 11 Applications and Fundamentals ed. Boumans P. W. J. M. Wiley New York 1987 ch. 6 p. 151. Blades M. W. Spectrochim. Acta Part B 1982 37 869. Huang M. and Hieftje G. M. Spectrochim. Acta Part B 1989,44 Jarosz L. Mermet J. M. and Robin J. P. Spectrochim. Acta Part B 1979 33 365. Blades M. W. and Caughlin B. L. Spectrochim. Acta Part B 1985 40 579. Boorn A. W. Cresser M. S. and Browner R. F. Spectrochim. Acta Part B 1980 35 823. Rademeyer C. J. Collins C. S. and Bulter L. R. P. J. Anal. At. Spectrom. 1991 6 329. Caughlin B. L. and Blades M. W. Spectrochim. Acta Part B 1985 40 987. Montaser A. Fassel V. A. and Larson G. Appl. Spectrosc. 1981 35 385. Xiao J. Li Q.-Y. Li W.-C. Qian H.-W. Tan J.-Y. and Zhang Z.-X. J . Anal. At. Spectrom. 1992 7 131. Grieg J. R. Lim C. P. Moo-Yourig G. A. Palumpo G. and Griem H. R. Phys. Rev. 1968 172 148. Freeman M. P. and Katz S. J. Opt. SOC. Am. 1963 53 1172. Cremers C. J. and Birkebak R. C. Appl. Opt. 1966 5 1057. Hill R. A. J. Quant. Spectrosc. Radiat. Transfer 1967 7 401. Todorovic M. and Vidovic S. J . Anal. At. Spectrom. 1993 8 1113. Boorn A. W. and Browner R. F. Anal. Chem. 1982 54 1402. Goldfarb V. M. and Goldfarb H. V. Spectrochim. Acta Part B 1985 40 177. Paper 41013246 Received March 7 1994 Accepted May 23 1994

ISSN:0267-9477    

DOI:10.1039/JA9940901087

出版商:RSC    

年代:1994

数据来源: RSC

摘要:

JOURNAL OF ANALYTICAL ATOMIC SPECTROMETRY OCTOBER VOL. 9 1093 Minimization of Spectral Interferences in Inductively Coupled Plasma Mass Spectrometry by Simplex Optimization and Nitrogen Addition to the Aerosol Carrier Gas for Multi-element Environmental Analysis* Trijntje van der Velde-Koerts and Jan L.M. de Boert RIVM P.O. Box 1 3720 BA Bilthoven The Netherlands Low-level trace element determinations by inductively coupled plasma mass spectrometry in environmental samples e.g. groundwater can be seriously hampered by spectral interferences originating from matrix- components and/or the argon plasma. Two approaches to reduce the level of interference were investigated variable step size simplex optimization with a number of response factors (analyte signals precision oxide and other polyatomic interferences) was applied to reduce simultaneously a number of molecular ion interferences; and nitrogen was added to the aerosol carrier gas and the nitrogen flow rate was used as an additional variable in the simplex optimization.Simplex optimization alone resulted in a more consistent choice for the settings of the instrumental variables and in reduction of the levels of polyatomic interferences. By the addition of nitrogen spectral interferences were further reduced by a factor of 1.5 to 3. For each set of optimum instrumental settings the spectral interferences were quantified as apparent analyte concen- trations. Final levels of interference were < 3 pg I-' for Cr Cu Nil Se and Zn and <10 pg I-' for As and V in 0.02 mol I-' Na 0.02 mol I-' CI 0.01 mol I-' Ca 0.01 mol I-' Mg or 0.004 mol I-' SO solutions.Matrix- induced sensitivity changes were also reduced by the addition of nitrogen. Keywords Inductively coupled plasma mass spectrometry; nitrogen addition; polyatomics; simplex optimization In principle inductively coupled plasma mass spectrometry (ICP-MS) is a very suitable technique for the determination of trace elements in solutions. Favourable aspects of the technique are the low detection limits multi-element analysis capability extended linear dynamic range and limited sensi- tivity for matrix-induced sensitivity changes of the analyte signal when an internal standard is used. However low-level determination in real samples for instance groundwater can be seriously hampered by interferences caused by polyatomic ions having the same M / Z ratio as the analyte ions to be determined.'-'' Tan and Horlick' showed that the spectral background in ICP-MS may be quite complex particularly in HC1 and H2S04.The minor isotopic species of the background spectra cannot be ignored when elemental determinations are being made at the ultra-trace level. These isotopic species also involve compounds related to the argon plasma and oxides and hydroxides. Tables of possible polyatomic ion interferences are given in references. '-,J ',12 Various approaches can be followed to reduce the effects of polyatomic ion interferences. These approaches are (i) optimiz- ation of instrumental parameters like gas flow rates power etc. and/or instrumental configurati~n~>~*'~ (ii) mathematical correction procedure^^'^*'' (iii) the introduction of organic solvents like propanol into the p l a ~ m a ~ ? ~ (iu) the application of matrix matching for certain well-defined samples like human serum,' (v) addition of gases other than argon to the nebulizer gas 5*6,13-15 coolant gas,16+17 or all gases," and (vi) reduction of the waterload to the plasma to reduce the level of oxides and hydroxides by spray chamber coolinglg or desolvation of the aeros01.~'-~~ A combination of the various approaches is also possible.Hill et applied simplex optimization of the operating parameters of the spectrometer in combination with the addition of nitrogen to the outer or aerosol carrier gas to reduce the interference of ArCl + on 75As. Mathematical correc- tion of the analyte signal [approach (ii)] can be carried out * Presented at the XXVIII Colloquium Spectroscopicurn Inter- t To whom correspondence should be addressed.nationale (CSI) York UK June 29-July 4 1993. straightforwardly for isobaric overlap and for polyatomic ions having another isotope combination at another m/z value that is free of interference. In principle correction can also be carried out for polyatomic interferences4,'' based on an assumed fixed ratio of a parent ion signal and the polyatomic ion signal. However this ratio is possibly not constant in all circumstances and also the method is inherently limited. Ketterer et al.7 applied multiple linear regression (MLR) and principle components regression (PCR) to determine low levels of cadmium in the presence of Zr Mo Ru In and Sn.However this approach is limited to multi-isotope element systems. Lam and Horlick16 observed modest signal enhance- ments (up to a factor of 4) and an order of magnitude reduction for certain background species by the addition of nitrogen to the outer flow of an argon plasma [approach ( u ) ] . In contrast Beauchemin and Craig17 found a reduction of the sensitivity by a factor of up to 5 upon the addition of nitrogen to the argon outer gas. Yet the detection limit for iron was improved by up to a factor of 4. Addition of xenon to the aerosol carrier argon gas also reduced the sensitivity considerably while the polyatomics were even more reduced.14 Multiple desolvation at -80°C in combination with an ultrasonic nebulizer [approach (ui)] considerably reduced the amount of water and HC1 introduced into an argon inductively coupled plasma.22 Polyatomic ion interferences from ArO' ClO' and ArCl+ were reduced by several orders of magnitude.Oxide levels for a number of metals were also considerably reduced e.g. UO' :U+ ratios were reduced to 0.06% when cryogenic desolv- ation was used in combination with the addition of acetylene to the central channel.22 The approaches mentioned above all suffer from limited (practical) applicability such as applicability to only one or a few elements or insufficient reduction of the level of polyatomic ion interferences to be able to determine all trace elements of interest at the desired low concentration levels in 'real world' samples with sometimes high levels of dissolved solids.None of these approaches have led to a procedure that takes full advantage of the low-level multi-element analysis capabilities of ICP-MS. Moreover quantitative information is still lacking in the literature about the actual level of the interferences in a particular case.1094 JOURNAL OF ANALYTICAL ATOMIC SPECTROMETRY OCTOBER VOL. 9 This study gives the quantification of spectral interferences by polyatomic ions for a number of environmentally important trace elements in relation to the analysis of groundwater. Elements considered were As Cr Cu Ni Se V and Zn while the matrix components were Na Ca Mg C1 and SO4,-. Quantification was done after variable step size simplex optim- ization of the instrumental settings with a series of criteria to reduce simultaneously a number of spectral interferences.Optimization was carried out both in the presence and absence of Cl (HC1). In the second approach to reduce the effects of polyatomic ion interferences nitrogen was added to the aerosol carrier gas and the nitrogen flow rate was used as an additional variable in the simplex optimization procedure. The choice was made for addition to the carrier gas because this was the most effective one compared with addition to the outer gas and/or auxiliary gas.' After optimization the levels of poly- atomic ion interferences were established again. Experiment a1 Chemicals Working solutions were prepared from 65% m/m HNO (Suprapur Merck Darmstadt Germany); 1.000 g In (Fixanal Riedel-de Haen Seelze Germany) 1.000 g Cr Ni Co Cu Zn As" Se Ba (Titrisol Merck); 1.000 g I-' V (Fluka Buchs Switzerland); NaCl CaC12.4H,0 Na2S0 Mg( NO3),*6H2O Ca( N03),.4H20 (Suprapur Merck).Instrumentation An ICP mass spectrometer (PQ2plus FI Elemental Winsford Cheshire UK) upgraded with a 'high-performance' interface was used equipped with a Pt sample cone a Ni microskimmer cone a Meinhard nebulizer and a water-cooled spray chamber. Nitrogen was added to the aerosol carrier gas via a T-piece before the inlet using a mass flow controller (Hi-Tec F-201C range 0-50 ml min - ' N Bronkhorst Ruurlo The Netherlands). Simplex Optimization of the Instrumental Variables Optimization of the instrumental variables was carried out under four different conditions resulting in four sets of measure- ment conditions optimization condition I (OC I) in the presence of HNO,; optimization condition I1 (OC 11) in the presence of HC1; optimization condition I11 (OC 111) in the presence of HN03 while adding nitrogen; and optimization condition IV (OC IV) in the presence of HCl while adding nitrogen. R.f.power outer gas flow intermediate gas flow Ar carrier gas flow N2 carrier gas flow and sampling depth (i.e. distance between coil and sample cone) were optimized using a variable step size simplex procedure. The initial simplex was constructed according to the scheme of Yarbro and D e m i ~ ~ g . ~ ~ The simplex was moved to the region of optimum response by reflection contraction and expansion as described by Deming and Morgan,,' Deming and Parker26 and Shavers et The "'In signal the relative standard deviation (RSD) of the "'In signal the BaO+:Ba+ ratio and the background signals at rn/z=52 (ArO' Arc' ArN' ClOH') m/z=60 (Ni contribution of the cones) and m/z=78 (Ar,+) were used as optimization criteria for all four optimization conditions.The background signals at rn/z=51 (C10') and m/z=75 (ArCl+ ) were used as extra criteria when chloride was present (OC I1 and OC IV). The simplex optimization procedure was continued until each response reached the limit shown in Table 1. During the simplex optimization a solution of 10 pg 1-' of In + 25 pg 1 -' of Ba in either HNO or HCI (Table 1 ) was continuously nebulized. The spray chamber temperature and the sample uptake rate were kept constant at 2°C and 0.7 ml min-'.At each combination of instrumental settings the lenses were retuned for maximum signal at m/z = 115. The final simplex was repeated on a different day This resulted in the same choice of instrumental settings for each of the four optimization conditions. The four sets of measurement conditions chosen are shown in Table 2. Quantification of Spectral Interferences Spectral interferences were quantified as apparent analyte concentration i.e. the analyte concentration that was found when a matrix solution without analyte was aspirated. The apparent analyte concentration was established for seven matrix solutions at the four sets of optimum measurement conditions (Table 2). Each matrix study was carried out using a set of three solutions in the following order a standard a blank and a matrix solution.The pre-measurement stabiliz- ation time was 3 min for the standard solution (10 pg 1-' of V Cr Ni Cu Zn As Se In in 1% v/v HNO,) and 10 min for the blank (10 pg 1-l of In in 1 % v/v HNO,) and matrix solution (matrix+ 10 pg 1-' of In in 1% v/v HNO,). The 10min period for the blank served as wash-out and stabiliz- ation period. The matrix solutions were aspirated for 10 min prior to the measurement because earlier experiments showed that some of the interferences increased with time and this equilibration period was taken to represent the worst case situation. This increase was attributed to a build-up of deposits on skimmer and/or sample cone during the aspiration of a matrix solution. For at least one type of interference (increase of signal at m/z= 52 and 56 in Mg-containing solutions) it was shown that a deposit on the skimmer cone was responsible for the interference.The interference remained at the same level when a blank solution was aspirated following the aspiration of the Mg-containing solution and only disappeared after cleaning of the skimmer cone. The following matrix solutions were investigated 0.02 moll-' NaCl 0.01 mol I-' CaCl 0.01 ml1-' Ca(NO,) mixed 0.02 moll-' NaCl+0.01 mol I-' Ca(NO,) (final con- centrations) 0.004 moll-' Na,SO 0.01 mol I-' Mg(N03)2 mixed 0.004 mol I-' Na,SO,+0.01 moll-' Mg(N0,)2 (final concentrations). The concentrations reflect the somewhat higher concentrations of the matrix components in Dutch groundwater. Cones were cleaned before each matrix study.Every matrix study was carried out in duplicate (on two different days). Data acquisition was performed in multi-element peak jump mode with 3 runs per sample 50 sweeps dwell time 10240 ps 5 points per peak 6 digital-to-analogue (DAC) steps and pulse counter detector. Isotopes chosen for analysis were 51V ',Cr ',Cr 58Ni 60Ni 63Cu 65Cu 66Zn 68Zn 75A~ 82Se and '''In as internal standard. Results and Discussion Simplex Optimization For routine analysis it is important that as many elements as possible can be determined in one run under the same set of operating conditions. Simplex optimization is a quick and efficient method of optimizing various interrelated variables at the same time. Optimization is usually performed with one optimization criterion as is described by Hill et aLi3 who used the Sb" :ArCl+ ratio as criterion for the reduction of ArC1'.Where different isotopes suffer from different spectral inter- ferences as in routine analysis several optimization criteria are important. Because optimization towards one response can result in a detrimental effect on another response the final simplex is reached when all of the chosen responses have reached a 'compromise' value. When doing so the optimum response area for the instrumental variables could be obtained in one or two days by testing 18-25 different instrument settings.JOURNAL OF ANALYTICAL ATOMIC SPECTROMETRY OCTOBER VOL. 9 1095 Table 1 Limits for response values in the simplex optimization procedure for the four optimization conditions (OC I-IV) Limits for response values Response "% signal/counts s-' 154( BaO) 138Ba m/z = 5211 15 m/z = 60/115 m/z = 7811 15 m/z = 51/115 m/z = 75/115 RSD 1 1 5 1 ~ (yo) oc I (1% v/v HNO no nitrogen) > 200 000 < 2 < 0.001 < 0.05 <0.1 <0.1 - - oc I1 (0.2 mol 1 - HCI no nitrogen) > 200 000 < 2 < 0.00 1 < 0.5 < 0.1 <0.1 < 20 < 1.0 oc I11 (1% v/v HNO nitrogen added) > 200 000 < 2 < 0.001 < 0.05 < 0.01 <0.1 - - ~ ~ ~~ oc IV (0.2 mol I-' HC1 nitrogen added) > 200 OOO < 2 (0.001 < 0.5 < 0.05 < 0.05 < 10 < 0.5 Table 2 Optimum measurement conditions (MC) after simplex optimization at the four optimization conditions mentioned in Table 1; MC I is the result of simplex optimization at optimization condition I etc.Simplex optimized values for instrument variables Instrument variable Intermediate gas/l min - ' Outer gas/l min-' Ar aerosol carrier gas/l min - N aerosol carrier gas/ml min-' Power/kW Sampling depth/mm Sample uptake*/ml min-' Spray-chamber temperature*/"C MC I 1.7 13.50 0.70 0.0 1.225 12.5 0.70 2 MC I1 1.7 13.75 0.76 0.0 1.250 12.5 0.70 2 MC I11 1.5 14.00 0.7 1 4.0 1.275 12.5 0.70 2 MC IV 0.8 14.00 0.73 6.0 1.325 12.5 0.70 2 * Variable kept constant during simplex optimization. During the simplex optimization it was found that the optimum measurement condition differed for different sets of glassware (torch nebulizer spray chamber elbow). The spec- tral interferences were therefore quantified using one set of glassware. The optimum N carrier gas flow is quite low (4-6 ml min-').At higher N flows (20-40 ml min-') the indium signal was significantly reduced ( 50-90% reduction). The BaO+:Baf ratio and the background signal at rn/z=52 were increased by a factor of 3-4 at a flow of 20 ml min-l. Also the doubly charged ion levels (Ba2+) were increased at higher N flows. In contrast relatively low interference levels were found for 75As at higher N flows at the beginning of the simplex optimization in the presence of HC1. Because of the unfavourable responses mentioned above the corresponding vertices were discarded and the simplex figure developed in another direction i.e. towards lower N flows. However the final interference level for 75As was similar to the levels found in the beginning of the procedure. So the choice of optimization for compromise conditions is the reason for the low optimum N flows compared with the 30-50 ml min-' found by other investigator^.^,^.'^.'^ The background signals of the blank and the net analyte signals in a 1% HN03 solution without matrix component remained similar for all four sets of optimum measurement conditions.Louie and Soo" obtained a small increase in both the background signal of the blank and the analyte sensitivity when nitrogen was added to the main inlet of the argon gas whereas Beauchemin and Craig28 observed an increase in the background signal of the blank and a decrease in analyte sensitivity when nitrogen was added as a sheathing gas to the sample aerosol. Matrix Induced Sensitivity Changes of the Indium Signal Table 3 shows the matrix induced sensitivity changes of the indium signal for the four sets of optimum measurement conditions.For most matrix solutions the sensitivity changes for MC 111 and MC IV where nitrogen was added (see Table 2) were less than for MC I and MC 11 where nitrogen was absent. The reduction of matrix-induced sensitivity changes by the addition of nitrogen was also shown by Hill et a[.,13 Branch et a[." and Beauchemin and Craig17 who found an elimination or reduction of matrix-induced sensitivity changes in a 0.01-0.3 mol 1-' sodium solution by addition of 3-5% nitrogen to the outer or aerosol carrier gas. Spectral Interferences Table4 shows the apparent V Cr Ni Cu Zn As and Se concentrations in seven different matrix solutions for the four Table 3 Sensitivity changes of the internal standard indium if matrix was added to 1% v/v HNO for the four sets of optimum measurement conditions (MC) in Table 2 Relative signal change of "'In (YO) Matrix/mol I-' 0.02 NaCl 0.004 Na,S04 0.01 CaC1 0.01 Mg(N0,)2 0.01 Ca(NO,) mixed 0.02 NaCl + 0.01 Ca( NO,) mixed 0.01 Mg(N03) + 0.004 Na2S04 MC I -51 - 13 - 34 - 18 - 43 -44 - 25 MC I1 - 37 - 15 - 26 -8 -31 - 38 - 26 MC I11 - 18 -21 -11 0 1 - 32 - 14 MC IV - 27 -8 -9 7 5 - 25 - 181096 JOURNAL OF ANALYTICAL ATOMIC SPECTROMETRY OCTOBER VOL.9 Table 4 Apparent analyte concentrations (pg 1-') in different matrix solutions for the four sets of optimum measurement conditions (MC) in Table 2 0.02 moll-' NaCl in 1% HNO 0.004 moll- Na,SO in 1 YO HNO MC I 2.0 0.3 6.3 0.6 0.5 6.3 0.4 <0.1 <0.1 2.5 1.1 MC I1 MC I11 2.5 <0.1 8.0 <0.1 <0.1 5.0 0.1 <0.1 <0.1 1.8 0.4 1.5 < 0.1 4.3 <0.1 <0.1 2.1 0.1 0.9 1.2 1.1 0.2 MC IV 2.2 <0.1 5.8 t 0 .1 0.2 2.3 0.2 0.6 1 .0 1.9 0.4 MC I <0.1 < 0.1 <0.1 <0.1 t 0 . 1 1.4 1 .o 2.2 <0.1 <0.1 2.9 <0.1 0.2 <0.1 <0.1 <0.1 1.3 0.7 1.1 <0.1 <0.1 2.8 <0.1 <0.1 <0.1 <0.1 <0.1 0.7 1 .o 2.5 0.5 <0.1 2.6 0.01 moll-' Ca(NO,) in 1% HNO 0.01 moll-' Mg(NO,) in 1% HNO ~~ MC I MC I1 MC IT1 MC IV <0.1 0.3 0.2 2.1 13.3 0.1 1.2 <0.1 t 0 . 1 <0.1 0.5 <0.1 0.2 < 0.1 3.1 20.9 0.1 1.3 <0.1 <0.1 <0.1 0.4 <0.1 <0.1 <0.1 1.8 12.4 0.1 1 .o 0.7 0.7 <0.1 0.1 0.02 mol I-'NaCl+0.01 mol I-' Ca(NO,) < 0.1 0.3 <0.1 1.7 10.4 <0.1 0.9 0.5 0.4 <0.1 0.3 MC I 3.5 0.3 11.3 <0.1 5.3 5.9 1 .o <0.1 0.2 6.1 1.1 MC TI MC 111 2.4 0.2 8.4 1.5 10.1 10.5 1.2 <0.1 <0.1 3.2 0.5 1.5 <0.1 4.1 1.5 8.6 11.3 1.2 0.5 0.5 1.7 0.2 MC IV <0.1 <0.1 <0.1 <0.1 <0.1 0.9 0.5 1.1 0.2 <0.1 1.2 MC I MC I1 MC I11 MC IV < 0.1 1.5 0.4 1.5 0.2 0.1 0.4 0.9 0.2 0.1 0.7 <0.1 0.7 0.1 0.5 t 0 .1 <0.1 0.3 0.5 <0.1 <0.1 0.6 <0.1 0.6 <0.1 0.5 <0.1 <0.1 0.2 0.5 0.1 0.2 0.3 <0.1 0.8 <0.1 0.7 <0.1 <0.1 0.2 0.2 <0.1 0.2 0.4 0.004 moll-' Na,SO,+0.01 moll-' Mg(NO,) MC IV 3.7 0.2 8.7 1 .o 6.5 2.0 1 .o 0.3 0.4 3.4 0.3 MC I <0.1 <0.1 0.1 <0.1 <0.1 2.1 0.8 2.0 <0.1 < O . l 2.7 MC I1 <0.1 0.2 <0.1 <0.1 <0.1 3.0 0.8 1.8 <0.1 <0.1 4.6 MC IT1 <0.1 <0.1 <0.1 <0.1 <0.1 0.9 0.4 1 .o 0.3 <0.1 1.6 MC IV < 0.1 <0.1 0.1 <0.1 <0.1 0.8 0.4 0.8 <0.1 < 0.1 1 .o MC I 24.5 1.3 75.4 2.2 12.2 <0.1 1.1 <0.1 <0.1 22.7 0.6 MC I1 15.7 0.9 48.9 3.2 19.9 <0.1 1.2 < 0.1 <0.1 14.0 0.4 MC I11 9.7 0.6 27.0 2.2 14.0 <0.1 1.1 0.8 1 .o 8.6 0.3 MC IV 9.8 0.8 24.4 1.9 10.6 < 0.1 0.9 0.6 0.7 6.6 0.3 sets of optimum measurement conditions.The variations of the measurements (difference between the duplicate measure- ments) are not shown for reasons of readability. In many cases the agreement between the duplicate measurements of the interferences was remarkably good with a difference between the duplicates of <30% of their average. In the case of the CaC1 measurements the difference was < 15% for the signifi- cant/relevant interferences.This makes a fair comparison of the different measurement conditions feasible. The apparent analyte concentrations at MC 111 and MC IV (nitrogen added) were lower than at MC I and MC I1 (no nitrogen added) as is shown in Table 4. Measurement condition I11 gave a reduction of 1.5-3 times (relative to MC I) for chloride interferences [NaCI and mixed NaCl + Ca(NO,) solutions] and sodium interferences (NaCl solution). Measurement condition IV gave a reduction of 1.5-3 times (relative to MC I) for chloride interferences (CaCl solution) sodium interferences [Na2S04 and mixed Na,SO + Mg( NO3) solutions] sulfate interferences (Na,SO solution) calcium interferences [CaCl and Ca( NO3) solutions] and magnesium interferences [ Mg( NO,) and mixed Mg( NO,) + Na,SO solutions].JOURNAL OF ANALYTICAL ATOMIC SPECTROMETRY OCTOBER VOL.9 1097 The level of the ClO' and ArCl' interferences at m/z=51 53 75 was influenced by the other matrix components present in the solutions. The apparent V Cr and As concentrations in the NaCl and the mixed NaCl+Ca(NO,) solutions were lower than in the CaCl solution although the chloride concen- tration is the same in all three solutions (Table 4). This effect is not caused by a matrix-induced sensitivity change of the indium signal because a calculation of the apparent analyte concentration without internal standard also gave lower inter- ference levels for the NaCl and mixed NaCl + Ca( NO3) solu- tions. A contamination of the CaCl solution is also not likely because the ratio of the signals at m/z = 5153 and m/z = 75:77 is 3.0-3.1 which is equal to the expected ratio of 3.07 for 35C1:37C1.Possibly sodium acts as an electron donor for the reduction of positively charged polyatomic ion interferences. The level of ArO' interferences at m/z=52 and MgO,+ interferences at m/z= 58 was also matrix dependent. The apparent Cr and Ni concentrations in the Mg(NO,) solution were higher than in the mixed Na,S04 + Mg(N03)2 solution. This effect could again be caused by sodium. The NaC1' interferences at m/z = 58 and 60 can effectively be reduced (below 0.2 pg 1-') if the proper measurement conditions are chosen (Table 4). Interferences at m/z = 66 and 68 were enhanced in the presence of nitrogen (MC I11 and MC IV); this could be caused by enhanced 38Ar14N,+ and 40Ar14N + interferences.The apparent Se concentration at m/z = 82 was more than 0.2 pg 1-' in the matrix solutions investigated. When the apparent Se concentration at m/z = 82 was calculated without an internal standard the spectral interference was less than 0.2 pg 1-' in all matrix solutions except the sulfate- and mag- nesium-containing solutions. For sulfate-containing solutions the interference is ascribed to 34S1603+ and in the magnesium- containing solutions the interference could be caused by an enhancement of 40Ar,'H2+ In the other matrix solutions the apparent Se concentration is an effect of a matrix induced sensitivity change of the indium signal. The apparent Cu concentration at m/z=65 was in most cases more than 0.2 pg I-'.In sulfate- or calcium-containing solutions the signal at m/z=65 increased when the sulfate or calcium concentration was increased (data not shown). The spectral interference in these solutions is therefore assigned to 33S1602+ 32S33S' and 48Ca'601H+. In the NaCl solution the apparent Cu concentration at m/z=65 is an effect of a matrix induced sensitivity change of the indium signal because the apparent Cu concentration was less than 0.2 pg 1-' if indium was omitted from the calculation. The interference in the Mg(NO,) solution could not be attributed to a matrix induced sensitivity change of the indium signal because calculation without internal standard gave the same results. Also spectral interference by ArMg' had to be excluded because of the absence of a concentration-interference relation.Copper con- tamination of the Mg(NO,) solution is also very unlikely because there is no Cu present at m/z=63. So the origin of this interference is not clear up to now. A comparison with measurements without simplex optimiz- ation is hard to make. Pilot studies without optimization using the old standard interface (2 x lo5 counts s-' for 100 pg 1-' of In) showed very variable and sometimes high levels of inter- ferences even when the instrumentation met the BaO' and Ba2+ specifications (0.2 and 2%) of the manufacturer. Reductions of the levels of interference of 5-100 times were then observed when simplex optimization was applied. In principle any factor of improvement can be obtained (within certain boundaries) depending on the starting situation.Simplex optimization after installation of the new high- performance interface with a combined set of optimization criteria (Table 1) resulted in more consistent sets of instrumen- tal settings and more reproducable levels of interference as was shown above. When comparing the levels of interference found in this study to those mentioned in the literature comparable levels were found for "V by Lam and Hor1ick.l6 These authors found an apparent concentration for this element of 78 pg 1-' in 2% HC1. This is equivalent to about 7 pgl-' for the 0.02moll-' C1 concentration used in this study. Lam and Horlick16 investigated the addition of nitrogen to the outer gas of their SCIEX Elan 250 instrument.Hill et ~ 1 . ' ~ found an apparent concentration for arsenic of 12.5 pg 1-' for NASS-2 certified reference material (CRM) Seawater (National Research Council of Canada) (& 0.5 mol - ' NaCl ) when adding nitrogen to the aerosol carrier gas. This is equivalent to about 0.5 pg 1-' for 0.02 mol 1-' NaCl and thus about 2-2.5 times lower than the level of interference found in this study. The instrument they used was a VG PQ2 with a standard interface. However Hill et only optimized for the reduction of the 40Ar35C1 ' interference on 75As while in this paper optimization for a set of responses including the background signal at m/z = 52 was carried out which resulted in a compromise set of measurement conditions. Increase of the N2 addition' would most certainly result in an increase of this background signal and deterioration of the detection limit for chromium at mlz = 52.Conclusion Variable step size simplex optimization proved to be a powerful tool to select optimum measurement conditions in ICP-MS analysis not only for one response factor but also for a combined set of response factors (analyte signals precision oxide and other polyatomic interferences). This is especially important for routine analysis when a suite of elements has to be measured in the same run. Optimization in this way resulted in a more consistent choice for the settings of the instrumental variables in more reproducable levels of inter- ferences and in minimization of the levels of polyatomic interferences. The addition of N to the aerosol carrier gas further reduced the levels of polyatomic interferences.1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 References Tan S. H. and Horlick G. Appl. Spectrosc. 1986 40 445. Vaughan M. A. and Horlick G. Appl. Spectrosc. 1986 40 434. Munro S. Ebdon L. and McWeeny D. J. J. Anal. At. Spectrom. 1986 1 211. Date A. R. Cheung Y. Y. and Stuart M. E. Spectrochim. Acta Part B 1987 42 3. Evans E. H. and Ebdon L. J. Anal. At. Spectrom. 1989 4 299. Evans E. H. and Ebdon L. J. Anal. At. Spectrom. 1990 5 425. Ketterer M. E. Reschl J. J. and Peters M. J. Anal. Chem. 1989 61 2031. Vanhoe H. Vandecasteele C. Versieck J. and Dams R. Anal. Chem. 1989 61 1851. Lam J. W. H. and Horlick G. Spectrochim. Acta. Part B 1990 45 1327. Ashley D. At. Spectrosc 1992 13 169. Vaughan M.-A. and Templeton D. M. Appl. Spectrosc. 1990 44 1685. Shao Y. and Horlick G. Appl. Spectrosc. 1991 45 143. Hill S. J. Ford M. J. and Ebdon L. J. Anal. At. Spectrom. 1992 7 719. Smith F. G. Wiederin D. R. and Houk R. S. Anal. Chem. 1991 63 1458. Branch S. Ebdon L. Ford M. Foulkes M. and O'Neill P. J. Anal. At. Spectrom 1991 6 151. Lam J. W. H. and Horlick G. Spectrochim. Acta. Part B 1990 45 1313. Beauchemin D. and Craig J. M. Spectrochim. Acta Part B 1991 46 603. Louie H. and Soo S. Y.-P. J. Anal. At. Spectrom. 1992 7 557. Hutton R. C. and Eaton A. N. J. Anal. At. Spectrom. 1987,2,595. McLaren J. W. Lam J. W. and Gustavsson A. Spectrochim. Acta Part B 1990 45 1091. Tsukahara R. and Kubota M. Spectrochim. Acta Part B 1990 45 581.1098 JOURNAL OF ANALYTICAL ATOMIC SPECTROMETRY OCTOBER VOL. 9 22 Alves L. C. Wiederin D. R. and Houk R. S. Anal. Chem. 1992 28 Beauchemin D. and Craig J. M. in Plasma Source Mass 64 1164. Spectrometry ed. Jarvis K. E. Gray A. L. Jarvis J. and 23 Jakubowski N. Feldmann I. and Stuewer D. Spectrochim. Acta. Williams J. Special Publication No. 85 Royal Society of Part B 1992 47 107. 24 Yarbro L. A. and Deming S. N. Anal. Chim. Acta 1974,73 391. 25 Demina S. N. and Morgan S . L. Anal. Chem. 1973 45 278A Chemistry Cambridge 1990 p. 25. 26 Deming; S. N. and Parker Jr. L. R. CRC Crit. Rev. Anal. Chem. 1978 7 187. 27 Shavers C. L. Parsons M. L. and Deming S. N. J. Chem. Ed. 1979 56 307. Paper 3/06899D Received November 28 1993 Accepted April 18 1994

ISSN:0267-9477    

DOI:10.1039/JA9940901093

出版商:RSC    

年代:1994

数据来源: RSC