摘要:

/- one-stop immediate access to all atomic spectrometry literature published since 7 985 including conference papers. jAASbase is a unique database that provides a fully comprehensive up-to-date source of over 23,. 000 analytical atomic spectrometry references. It is designed to meet every atomic spectroscopists information needs - a convenient desktop tool. As a subscriber you will enjoy the following benefits of JAASbase @ Simplicity of use even for the non-specialist @ Economy of effort and expense @ A vast store of references @ Flexibility that fosters thorough searches @ Adaptability - you can add your own data @ Helpdesk and user literature gives added assurance that you can quickly master JAASbase Idealist Software f 21 0.00 $368.00 1994 Subscription Details JAAS Backfile (1 986-93) jAASbase Updates EC €99.00 EC €280.00 EC USA $174.00 USA $490.00 USA (VAT chargeable in the UK) To order JAASbase and for further information please contact Sales and Promotion Department Royal Society of Chemistry Thomas Graham House Science Park Milton Road Cambridge CB4 4WF United Kingdom. TeI +44 (0)223 420066.Fax +44 (0)223 423623. ROYAL SOCIETY OF Information Services/- one-stop immediate access to all atomic spectrometry literature published since 7 985 including conference papers. jAASbase is a unique database that provides a fully comprehensive up-to-date source of over 23,. 000 analytical atomic spectrometry references. It is designed to meet every atomic spectroscopists information needs - a convenient desktop tool. As a subscriber you will enjoy the following benefits of JAASbase @ Simplicity of use even for the non-specialist @ Economy of effort and expense @ A vast store of references @ Flexibility that fosters thorough searches @ Adaptability - you can add your own data @ Helpdesk and user literature gives added assurance that you can quickly master JAASbase Idealist Software f 21 0.00 $368.00 1994 Subscription Details JAAS Backfile (1 986-93) jAASbase Updates EC €99.00 EC €280.00 EC USA $174.00 USA $490.00 USA (VAT chargeable in the UK) To order JAASbase and for further information please contact Sales and Promotion Department Royal Society of Chemistry Thomas Graham House Science Park Milton Road Cambridge CB4 4WF United Kingdom. TeI +44 (0)223 420066. Fax +44 (0)223 423623. ROYAL SOCIETY OF Information Services

ISSN:0267-9477    

DOI:10.1039/JA99409BP003

出版商:RSC    

年代:1994

数据来源: RSC

摘要:

JOURNAL OF ANALYTICAL ATOMIC SPECTROMETRY MARCH 1994 VOL. 9 11N Foreword XXVlll Colloquium Spectrsscopicum Internationale June 29-July 7 1993 York UK The XXVIII Colloquium Spectro- scopicum Internationale was the fourth CSI to be held in Great Britain and like the last CSI in Cambridge in 1979 was organized by the Association of British Spectroscopists. This is a confederation of all the specialized spectroscopy groups in the UK including groups of The Royal Society of Chemistry and the Institute of Physics. We acknowledge gratefully the support which we received from the RSC staff particularly those associated with JAAS and the Analyst and from the Institute of Physics. However we did not use a permanent conference organization and relied heav- ily on volunteers. For various reasons we adopted a slightly different timing to previous conferences. Delegates arrived on Tuesday evening June 29 and the working sessions began on Wednesday morning and ran until late Saturday afternoon; delegates then dispersed or went on to the post-CSI Symposia on the Sunday morning.Although there were some doubts expressed before the CSI the arrangement worked well in practice and was particularly appreci- ated by those whose flight tickets required a Saturday night stay in the United Kingdom. The recession and the plethora of conferences involving spectroscopy in 1993 took their toll and there were rather fewer delegates than we had originally hoped for though more than we at one time feared. Some 600 del- egates were present from about 45 countries; this included about 150 from the United Kingdom and 55 from Germany the next largest group.We offered reduced registration fees to del- egates from Eastern Europe and other less favoured areas and this increased their number; we also gratefully acknowl- edge funding from the Royal Society and from the International Science Foundation (‘The Soros Fund‘) for 15 delegates from the former Soviet Union and the support given to individual delegates from various firms. Those from Eastern Europe greatly appreciated the chance to attend and made a significant contribution to the conference but the arrangements for them did involve a great deal of time and effort. The Organizing Committee had adopted the policy endorsed by the National Delegates’ meeting some years ago and tried to include all branches of analytical spectroscopy.However the majority of the papers submitted were on atomic spectroscopy; at least two of the four parallel sessions were devoted to some aspect of atomic spectroscopy or ICP-MS and some of the lectures on molecular spectroscopy were poorly attended. As there are many conferences on molecular spectroscopy on NMR and on organic mass spectrometry it is open to debate whether this overall coverage should be maintained or whether the CSI should be a specialized atomic spectroscopy conference However about 60 papers have been submitted for the CSI issue of the Analyst and about 105 papers for the CSI issue of JAAS (40 and 60 accepted respect- ively) so there is clearly a case for maintaining the wide coverage.We requested all the plenary and invited lecturers to submit manuscripts of their papers; unfortunately a significant number felt unable to do so but those papers which have been published will be collected and sent as a single volume to all delegates. Apart from the 5 plenary and 28 invited lectures 92 contributed lectures were presented in the four parallel ses- sions and there were about 300 posters. We had laid down a format for the abstracts but the instructions were inter- preted in so many ways that it was decided to use a scanner so that they could all be reproduced in a common style. We achieved the common style thanks to a great deal of effort at Loughborough University of Technology by Barry Sharp and his team; the amount of work involved for the abstract book was very large but indexing then became relatively easy.The exhibition involved some 40 firms both large and small. It had to be spread over three sites in the University but we have helped to make future exhibitions better! We expressed our strong support for a new exhibition facility being planned at the University; it was decided to go ahead with the plan but only as soon as the XXVIII CSI was over! Preliminary work in the Physics Department started during the CSI and went into full gear as soon as we left. The new enlarged exhibition area has now been opened! We had full cooperation from the weather4ry and sunny throughout (typical English summer weather!) so we enjoyed the campus to the full and the excursions for the accompanying persons had excellent conditions.One of the main social events was an organ and choir concert in York Minster on the Wednesday evening. Thursday afternoon was free with a choice of visits to various stately homes and gardens to the North Yorkshire Moors steam railway to Bempton Cliffs and Flamborough Head and to the Drax power station. The final social event on the Saturday evening was a visit to the National Railway Museum and everyone was impressed with the museum display. On the Thursday evening a meeting was held under the aegis of the Association of British Spectroscopists to discuss closer collaboration between spectroscopists in Europe. The dis- cussion with about 50 participants centred on three topics. (i) Formal collaboration as arranged through IUPAC national Chemical Societies etc.,-information should be published more widely.(ii) Informal collabor- ation-names and addresses would be circulated to assist the exchange of information about meetings and methods. (iii) Technical collaboration including financial resources and multi- laboratory projects-outside the scope of the meeting but would be encouraged by contacts made through (i) and (ii). The National Delegates’ meeting took place on the Friday afternoon. The principal business was the choice of location for the XXX CSI in 1997; there were two bids-Australia and the United States and it was decided by two votes to hold the 1997 CSI in Melbourne Australia. There were a number of pre- and post- symposia associated with the XXVIII CSI. One day tutorial meetings on vapour generation and chemometrics were held on Tuesday June 29.The 3rd Kingston Conference on Analytical Spectroscopy in the Earth Sciences took place immediately prior to CSI (papers will appear in a special issue of Chemical Geology.) After the CSI the post- symposium on Graphite Atomizer Techniques in Analytical Spectroscopy took place at the University of Durham followed by a one day meeting on Trace Elements in Clinical Chemistry; the 5th Surrey Conference on Plasma Source Mass Spectrometry took place at Lumley Castle Hotel near Durham whilst the post-symposium on the Analytical Applications of Glow Discharges was at the University of York. (Forewords from12N JOURNAL OF ANALYTICAL ATOMIC SPECTROMETRY MARCH 1994 VOL. 9 the organizers and papers from these meetings can be found in this issue Personally I managed to get to very few of the lectures but I felt that it was a very friendly and happy conference and I think that most of the delegates gained the same impression.I am con- scious of things that we could and should have done better but it is easy to see the problems afterwards. I would like to take this opportunity of thanking all my colleagues on the Organizing Committee particularly the Scientific and Exhibition Committees the Secretary Barry Sharp with his team at of JAAS). Loughborough University of Technology who dealt with registrations and the Treasurer Terry Threlfall at the University of York. Our thanks are also due to the York University Conference Office and to Neil James there who always responded promptly to our requests reasonable and unreasonable.Some may recollect that the chairman- designate of the 1981 CSI died during the 1979 CSI in Cambridge; again a tragic event has befallen the chairman of the next CSI. His many friends and colleagues were shocked in December by the sudden death of Klaus Dittrich chairman of the XXIX CSI Organizing Committee. Our deepest sympathy goes to Frau Dittrich and to the German spectroscopic community. The XXIX CSI will go ahead in Leipzig on the planned dates (August 27-September 1 1995). The position of chairman has been taken over by Professor Dr Hubertus Nickel of the Forschungs- zentrum Julich. We wish him and his col- leagues every success and look forward eagerly to the next CSI. Edward Steers Conference Chairman University of North London London UK XXVlII CSI Post-Symposium on Graphite Atomizer Techniques in Analytical Spectroscopy July 4-7 1993 Durham UK The papers in this issue from the XXVIII CSI Post-Symposium on Graphite Atomizer Techniques in Analytical Spectroscopy held from the 4th to 7th July 1993 at the University of Durham UK represent a selection of the material presented at the meeting.A pre- or post- CSI symposium on electrothermal atom- ization has come to be a regular feature of the CSI meetings and the material presented and discussed covers a wide range of electrothermal atomization topics. This meeting was no exception in that the 98 participants enjoyed hearing and discussing topics ranging from fun- damental considerations of atomization and interferences modifications to atom- izer design calibration considerations and interesting applications presented in 26 and 23 oral and poster presen- tations respectively.I am grateful to both the participants of the meeting and those who have managed to submit a manuscript within the time frame required for publication in this special issue as I know from personal experience just how difficult it can be to find the time to convert ones lecture or poster presentation into a finished manuscript. I would also like to acknowledge the honesty of a number of presenters who discussed their work at the post- symposium but declined to submit a paper on the grounds that they felt that they still had further experiments to complete before they would consider the investigations to be in a form suitable for publication.A number of last-minute papers from this symposium will follow in subsequent issues of JAAS as a few did not manage to make the final deadline. In conclusion I would like to take this opportunity to publicly thank my fellow co-organizers Dr David Halls (Trace Element Unit Glasgow UK) and Dr John Dean (University of Northumbria at Newcastle upon Tyne UK) for all their hard work and efforts in organizing the meeting. At times I felt a little embarrassed sitting in Germany making suggestions that I knew other people would have to implement. Dr Ian L. Shuttler Bodenseewerk Perkin-Elmer GmbH Postfach 10 17 61 0-88647 Uberlingen Germany XXVlll CSI Post-Symposium 5th Surrey Conference on Plasma Source Mass Spectrometry July 4-6 1993 Durham UK The 5th Surrey Conference on Plasma Source Mass Spectrometry held in July was attended by over 90 delegates from more than 20 countries.Due to the coincidence of timing and location with the XXVIII CSI the conference was on this occasion organized as a post CSI symposium. The subject areas covered during the three days included several contributions on fundamental instru- ment studies two papers concerned with sector ICP-MS instruments and a multi- tude of presentations using alternative sample introduction techniques. By far the most comprehensively covered topic was sample introduction using laser ablation both as a probe tool and for bulk analysis. However the not insig- nificant calibration problems which exist with this technique are still a major factor in limiting the widespread appli- cation of laser ablation.The conference paper published in this issue by Grkgoire and Lee applied elec- trothermal vaporization (ETV) for intro- duction of blood and animal tissue before isotope ratio measurement. Cadmium and Zn from digested samples were separated from the matrix and determined after introduction by ETV. The analysis of the small sample volumes is an obvious advantage with this method. The paper by Alves et al. discusses the removal of organic solvents by use of cryogenic desolvation and introduction of the resulting aerosol into an ICP mass spectrometer. Kym Jarvis N ERC I CP-MS Facility Imperial College at Silwood Park Ascot Berkshire U K SL5 7TEJOURNAL OF ANALYTICAL ATOMIC SPECTROMETRY MARCH 1994 VOL.9 13N XXVlll CSI Post-Symposium on the Applications of Glow Discharge Sources in Optical Spectroscopy and Mass Spectrometry July 4-7 1993 York UK Glow discharge sources (GDS) have been used for many years in optical spectroscopy particularly since the pion- eering work of Schuler on hollow cath- ode discharges. The development by Grimm of a convenient source using plane metallic samples as the cathode led to the increasing use of GDS in combination with optical emission spec- troscopy (OES) for analytical purposes. Much of the early investigations on these sources were carried out in Germany particularly at the Institute for Spectrochemistry and Applied Spectroscopy (ISAS) in Dortmund and in many cases practitioners in other countries had their interest first aroused by working for a period at ISAS or attending short courses and user groups held there. A number of specialist meet- ings have been held in recent years-at Baden-Baden in 1986 Braunschweig in 1988 and in Julich in 1990 but these were mainly German speaking meetings with relatively few participants from outside Germany. More recently there have been a number of developments extension of the technique to non-conducting samples by the use of r.f.discharges instead of d.c. discharges increasing use of GDS for depth-profiling of corroded and coated surfaces (including non- conductive coatings) renewed appli- cation of GDS as ion sources in mass spectrometry (MS) and development of boosted sources using auxiliary dis- charges. These have all encouraged the renewed interest in GDS shown by additional papers at conferences particu- larly those on plasma spectrochemistry. In March 1992 the First European Workshop on Surface Analysis by GD-OES was organized by GAMS and LECO in Paris.At this meeting the European Working Group on Depth Profiling by GD-OES was set up. Dr Arne Bengtson of the Swedish Institute for Metals Research was elected as the first chairman and Dr John Murphy of Leco UK became the secretary of the steering committee. When it was decided that a XXVIII CSI Post-symposium would be held at the University of York on the applications of GDS in OES and MS it was agreed by the steering committee that the meeting would include the Second European GDS Workshop. There were 65 participants from 13 countries including Professor K. Laqua formerly head of the Atomic Spectroscopy Group at ISAS Professor R. K. Marcus who has carried out pioneering work on r.f.-GDS for OES and MS and Dr Bengtson; some 41 papers and were presented at the meet- ing 21 as lectures and the rest as posters. All aspects of GDS were covered funda- mentals and applications surface and bulk analysis plane and hollow cathodes d.c. and r.f. excitation. On the first afternoon (Monday) there were demon- strations particularly of commercial software and the main poster session. In the evening a very enjoyable confer- ence dinner was held at the Merchant Adventurers Hall in the city of York. The final session on Wednesday morn- ing was devoted to a report on the progress made by the European Working Group and a discussion on its future activities. It was agreed that the Group be expanded to cover all aspects of glow discharge spectrometry that the next workshop would be held in conjunction with the XXIX CSI in Leipzig (August 1995) if possible and that a sub-group on theoretical aspects of GDS should be established which I was asked to chair- so far I have not found time to act on this! Edward Steers University of North London London U K

ISSN:0267-9477    

DOI:10.1039/JA994090011N

出版商:RSC    

年代:1994

数据来源: RSC

摘要:

14N JOURNAL OF ANALYTICAL ATOMIC SPECTROMETRY MARCH 1994 VOL. 9 Conference Reports XXVIII Colloquium Spectroscopicum Internationale (CSI) June 29- July 4 1993 York UK A Scientific Journey to Old England- CSI XXVlII + Famous English History Four years ago in July 1989 we met Dr Edward Steers in Sofia at the XXVI CSI. It had just been announced that the UK would host the XXVIII CSL. We were in a cool conference room looking out into the heat of Sofia and yearning for Sunshine and good weather for York. Since that time Dr Steers and the rest of the committee worked extremely hard towards the conference and the dream of good weather came true! If we are to believe the movies and stories or indeed the information in the conference circular the weather for the meeting was very untypical for the UK.It tempted everyone to visit the sights of York such as York Minster the Jorvik Viking Centre the ARC Clifford’s Tower Treasurer’s House or simply to walk around the attractive campus of the University of York instead of sitting in lecture room V045 pushing through the crowds of people in the confined space available for the posters and instrument exhibition. (New instruments were not being demonstrated we have to wait for PittCon for that.) However we do think it was a good idea to organize the CSI at the University of York. This enabled both scientific dis- cussions and meeting of old and new friends to take place throughout the whole day beginning with the morning jog and coming across Boris L‘vov during the meals we all took together and finishing in the late evening in the college bars over a few pints of beer.Unfortunately even though the scien- tific programme was interesting and the conference fees which included all meals and accommodation were moderate it seemed to be a long way for colleagues from North America to come. It could be however that the late announcements and rumours that the meeting would actually start on June 29th (a Tuesday which is unusual) caused problems with booking tickets to Europe. Despite the problems more than 500 delegates from around the globe enjoyed the ‘conference mixer’ at Heslington Hall which was a good welcoming party on the first evening listened to a fanfare of trumpets at the opening ceremony and stayed until the end of the conference to receive the invitation from Professor Klaus Dittrich to go to the XXIX CSI in Leipzig in 1995.In between delegates had the chance to listen to the plenary lectures (given by K. Niemax C . L. Wilkins R. E. Hester D. A. King and M. L. Gross) and to choose from 28 invited and 88 contributed lectures given in four parallel streams. We counted ourselves lucky that we were attending as a ‘team’ so that we could resolve the ‘coincidences’ in the field of spectroscopy which were demonstrated by the lecture schedule. Some of the ‘all-round’ spec- troscopists will have to wait for publi- cation of the results presented and so it is to be hoped that many lecturers have submitted their manuscripts. The area allocated for viewing the approximately 400 posters (about 130 each day) and the space in which they were to be viewed was too limited to do justice to the interesting results pre- sented even though some of those listed in the programme were missing.For most of us ‘younger’ scientists working in different areas of atomic and inorganic mass spectrometry this was the first major conference we had attended hence it is difficult to say what the highlights of this CSI were. Therefore we have noted what was of particular interest to one or other of us. Electrothermal atomizers/vaporizers were discussed with several techniques in combination with hydride AAS (e.g. G. Schlemmer and Y. Z. Zinbi) or with chromatographic separation for speci- ation analysis (e.g. D. C. Baxter and G. Knapp) and with both ICP-AES (e.g. H. Nickel) and ICP-MS (e.g. E.Hoffmann and A. Stroh). It would seem that speciation which is very important for environmental and biological research is becoming a ‘boom’ area in analysis. The use of GC-MIP- AES (e.g. N. W. Barnett J. Mierzwa and R. Lobinski) and GC/HPLC/ SFC-ICP-MS (e.g. J. A. Caruso R. S. Houk E. H. Evans and A. V. Hirner) to study the binding of elements in gaseous samples and in solution was discussed. Another approach to investigating such problems is the combination of HPLC-hydride generation (AAS/AES/ ICP) or FI with AAS/AES or ICP-MS as the method of detection (e.g. J. F. Tyson M. Gomez S. F. Durrant D. Stuewer H. Emteborg and A. A. Brown). Of particular interest to us also were papers dealing with discrete sampling in ICP-AES and ICP-MS. Laser ablation as a sampling technique with the poten- tial for distribution analysis in different types of solid samples (including isotope ratio analysis in for example environ- mental and geological material) was the subject of several presentations (e.g.M. Thompson J. G. Williams V. Majidi H. P. Longerich P. J. Rommers D. J. Bate and F. E. Lichte). It would be a serious omission not to say a word about ICP-MS one of the most expensive but also most efficient techniques used in ‘atomic’ spectrometry. However as with the situation for graph- ite furnace techniques some of the interesting developments were presented at the graphite atomizer Post-CSI Symposium held in Durham. However in addition to the lectures which gave reports of combinations of ICP-MS with both separation techniques and laser ablation as mentioned above we had the opportunity to listen to an interesting approach to dealing with the accuracy of the results from ICP-MS (B.L. Sharp) the efficiency of ion formation and ion processes in a complex system a high temperature-normal pressure plasma source and low pressure ion transport system (e.g. L. S. Dale S. D. Tanner R. F. Browner J. W. Olesik and M. J. Ford) and the application of ICP-MS in environmental research (e.g. L. Moens). It is unfortunate that it is not possible to mention here all the authors of the interesting posters pre- sented. We saw posters with excellent layout not just those prepared using the resources of large companies and organ- izations but a lot of authors had worked diligently to produce their results and to present them clearly.There were also posters which looked as if they had been scribbled down during the journey to York! Despite the full scientific programme the conference organizers had arranged trips so that the foreign delegates in particular could see something of the area other than just the University. There was a journey into the history and music of York Minster through an organ and choir concert. CSI tradition- ally includes an excursion for delegates during the meeting. Although some of the destinations were a bit far for half a day they certainly whetted everyone’s appetite to come back to Yorkshire even for a holiday.JOURNAL OF ANALYTICAL ATOMIC SPECTROMETRY MARCH 1994 VOL. 9 15N The final conference gathering took place at the National Railway Museum a good place for train enthusiasts.However it did leave us longing for the refectory at the University where we have fond memories of the ample meals. After the six days in York our team then had an interesting journey to Durham (via the open-air museum at Beamish and with a fish and chip lunch) to the Post-Symposium on Graphite Atomizer Techniques. We stayed in Durham Castle where the meeting was held the next place on our journey through English history. Many interes- ting lectures were presented at the meet- ing which was well organized by I. L. Shuttler. We also had the opportunity to spend a few hours at the ICP-MS Post-Symposium which was being held nearby. (The differences in comfort between the two symposia were related to the prices of each one!) It is clear that the organizers of the XXVIII CSI and the associated Post- Symposia spared no efforts in ensuring everything went well and we thank them all.We will endeavour to organize the XXIX CSI to include interesting speakers and under good conditions. We wish all spectroscopists luck with their work so that they can report lots of good news at the end of August 1995 The Leipzig Team University of Leipzig Germany See you in Leipzig. XXVlll Colloquium Spectroscopicurn lnternationale (CSI) Post-Symposium Graphite Atomizer Techniques in Analytical Spectroscopy July 4-7 1993 Durham UK The XXVIII CSI Post-Symposium on Graphite Atomizer Techniques in Analytical Spectroscopy was held in Durham from the 4th to 7th July 1993. For me this was the first time I had participated in a really great scientific meeting quite an extraordinary and exciting experience.For those who had already attended the CSI meeting in York the Symposium began on American Independence Day (marked by the Stars and Stripes outfits of some of the participants from the USA) with a bus trip to the mediaeval city of Durham. This gave everyone a beautiful tour through the North Yorkshire Moors scenery with a break for lunch at the legendary Harry Ramsden’s Fish and Chips restaurant. So lunch was celebrated in typical English manner a meal that really matched all expectations of the 90 par- ticipants from 20 different countries. After lunch the tour continued with a visit to the Beamish ‘Open Air Museum’ just outside Newcastle where we felt as if we had stepped back in time to life in Northern England in the early part of the 20th century.This is a ‘living’ museum with fully restored villages a coal mine farms and actors dressed accordingly telling everyone what life was like during this period. The dentist’s surgery was particularly gruesome! The next stop was Durham a city in Northern England with a wonderful historic centre providing an excellent location for the Symposium. The meeting took place at the univer- sity but some other arrangements had also been made for example all meals were eaten together in the Norman Castle built in 1072. Some participants were even accommodated within its old walls. So on arriving in town the main question was whether you were lodged in the Castle (the lucky ones) or in one of the university’s student houses (the not so lucky ones!).Everyone gathered before dinner on the Sunday evening for a social mixer in the Tunstall gallery in the Castle an appropriately grand location for such a gathering of atomic spectroscopists. On Monday morning the official part of the meeting began with the opening ceremony by Ian Shuttler. After some ‘housekeeping’ announcements (which we later became quite used to) he quickly passed over to the first scientific lecturer. This was Jim Holcombe who presented an interesting lecture on ETV-MS with an electron impact ionizer instead of an ICP atomizer. Other highlights of the day were Ralph Sturgeon speaking about FAPES (furnace atomization plasma emission spectrometry) and Dave Styris about bulk diffusion in graphite layers.After lunch the first of two poster and discussion sessions were held in a small room in the cellar of the Castle. The first poster session concentrated on funda- mental and theoretical areas and ranged from investigations of the use of a CID detector for continuum source ETAAS presented by Jim Harnly to the use of metallic platforms for standardless analyses from Professor Ma. From the general level of noise coming from the room where the posters were displayed it was clear that the participants were finding a lot to discuss. The late afternoon lecture session continued with an excellent overview by Dimiter Tsalev on the development of the analysis of biological matrices with the GFAAS over the last 10 years and was complemented by the following lecture from Yngvar Thomassen show- ing the advantages of coupling hydride Delegates enjoying their fish and chip lunch Discussions during one of the poster sessions16N JOURNAL OF ANALYTICAL ATOMIC SPECTROMETRY MARCH 1994 VOL.9 Ian Shuttler (L) Frank Portala ( C ) and Werner Schrader entering into the spirit of the pub tour generation with the graphite furnace for the determination of antimony in body fluids. In the evening a guided tour through Durham Castle had been organized. We all listened intently to the eventful and exciting history of the Castle and of course its ghosts. After dinner we were offered the opportunity of participating in a pub tour through the old town with the aim of having at least one pint of good English ale in each of the town’s pubs.The organizers who had warned that only a half pint should be drunk by those not used to English beer had anticipated that only few participants would join in but hardly anyone wanted to miss such an extraordinary experience of English culture and virtually all attendees waited at the alloted time to come along. Even splitting the group into two did not help. Such masses of ‘CSI-tourists’ simply had to overwhelm the traditional pub atmosphere in every pub we went into. Nevertheless we all enjoyed the night although because of the early closing hours we did not quite have the chance to achieve the aim of drinking in all the pubs on the schedule. Early the next morning it was Nancy Miller-Ihli who had the responsible job of presenting the first lecture on high-accuracy ultrasonic slurry GFAAS determinations fortunately not in beer and with confidence considering how late to bed we all were! The next session was chaired by Vahid Majidi who is always entertaining.He introduced the lecture given by Mike Hinds (for whom is gold not like gold) on a comparison of different determinations of silicon in gold. Uwe Heitmann proceeded to dis- cuss the use of laser excited atomic fluoresence with ETAAS for the determi- nation of selenium arsenic and anti- mony. Sadly the next lecturer Phil Riby could not be with us due to business reasons and so Clare Smith’s lecture on using continuum source ETAAS with a linear photodiode array detector was moved forward. The Tuesday afternoon poster session concentrated on applications of ETAAS and these ranged from the novel such as the determination of iodine as mercury iodide by ETAAS presented by Professor Bermejo-Barrera and her students to improvements in methods for the deter- mination of molybdenum and vanadium in sediments by Rhodri Thomas. There was much to expect of the late after- noon’s programme as it was dominated by the most well known names in atomic spectrometry.Boris L‘vov spoke about precision and detection limits in Zeeman GFAAS followed by Greet de Loos- Vollebregt with a study of the dynamic range in Zeeman GFAAS. She paid special attention to the precision meas- ured in concentration units. Walter Slavin continued with a lecture about the working range and the stability of characteristic masses in GFAAS. He emphasized the influence of the lamp current of a hollow cathode lamp.Sadly as David Littlejohn could not be present due to illness the organizers had twisted Jim Harnly’s arm to step in for David The ‘Prince Bishop’ with all the ‘CSI-crew on the River Wear below Durham Castle Professor Bermejo-Barrera (R) and some of her group from Spain and chair this session a job he performed well in controlling the questions and developing discussion on what appears to be one of the current ‘hot’ topics of discussion in ETAAS work. Unfortunately the discussion session was cut short by the university security guard who insisted on locking the lecture theatre for the night! There was a social programme also organized for the Tuesday evening a boat trip on the ‘Prince Bishop’ cruising on the River Wear below the ancient Durham Cathedral and Castle.All day long we had doubted whether the boat would stand the weight of the whole CSI-crew. As is usual in science the problem was finally solved by ‘trial and error’. We were lucky. The boat did not founder and we could enjoy the beautiful evening weather on board with a barbe- cue and yet more good English ale. A jazz band had been arranged to provide the musical entertainment. A special challenge was given for the presentation of short serenades from different countries. If there were enough voices from one particular country people had to present a typical song. So we had the chance to listen to songs from Russia North America England (which seemed to go on and on!) Scotland (the subject of which I was The barbecue!JOURNAL OF ANALYTICAL ATOMIC SPECTROMETRY MARCH 1994 VOL.9 17N informed concerned fighting the English!) Norway and Germany. For Italy G. Torsi formed his own group and gave an excellent solo performance of an aria from ‘Aida’. For everyone present the evening was very enjoyable. On Wednesday morning the sym- posium continued with Vahid Majidi presenting a lecture about gas-phase and surface reactions in ETV-AAS (despite the audience participation with a number of people assisting with their laser pointers directed at his slides on the screen!) and Albert Gilmutdinov who then explained the basics of atomic absorption with a few very very simple equations. At short notice Dmitri Katskov filled David Littlejohn’s lecture session and talked about a new design of graphite tube containing a graphite filter. Trevor McAllister provided the final lecture of the Symposium with a presentation about the detection of gase- ous oxides from nitrate decomposition in graphite furnaces by mass spec- trometry. There has been much dis- cussion in the literature about this topic in the last few years and it was interesting to observe the discussion that developed between Trevor McAllister Boris L‘vov Jim Holcombe and Dave Styris who have all been involved in these investigations. In conclusion the meeting was a great success and for me it was a new exciting experience. I was surprised about the good relationships between all the par- ticipants. Many of them have known each other for many years and meet regularly at such meetings. During the day they were involved in long often heated discussions and in the evening they sat together to have a drink. I got the impression that the Post-CSI partici- pants were like a huge family and even I felt included in this. I would like to thank the organizers Ian Shuttler David Halls and John Dean and all the companies that helped to ensure the success of the meeting with their financial support. Frank Portala Bodenseewerk Perkin-Elmer GmbH Uberlingen Germany

ISSN:0267-9477    

DOI:10.1039/JA994090014N

出版商:RSC    

年代:1994

数据来源: RSC

摘要:

JOURNAL OF ANALYTICAL ATOMIC SPECTROMETRY MARCH 1994 VOL. 9 17N Diary of Conferences and Courses 1994 ANATECH 9 4 4 t h International Symposium on Analytical Techniques for Industrial Process Control April 10-13 Royal Hotel Casino Mandelieu La Napoule France Details can be found in J. Anal. At. Spectrom. 1993 8 48N. For further information contact ANATECH 94 Secretariat Elsevier Advanced Technology Mayfield House 256 Banbury Road Oxford UK OX2 7DH. Telephone +44 (0) 865 512242; fax +44 (0) 865 310981. MARC-111 Conference April 10-16 Kona Hawaii The Third International Conference on Methods and Applications of Radio- Analytical Chemistry (MARC-111) will be held in conjunction with BERM-6 April 10-16 1994 at the King Kamehameha Hotel Kona Hawaii. For further information contact Dr Ned Wogman Battelle Pacific North-west Laboratories PO Box 999 P7-35 Richland WA 99352.Fax 509 376 2373. Sixth International Symposium on Biological and Environmental Reference Materials (BERMd) Topical Conference of AOAC International April 17-21 Details can be found in J. Anal. At. Spectrorn. 1994 9 5N. For further information contact Dr Wayne R. Wolf BERM-6 Standard Reference Materials Program Bldg. 202 Room 211 National Institute of Standards and Technology Gaithersburg MD 20899 USA. Short Courses at Loughborough University of Technology Loughborough Leicestershire UK Gas/Liquid Chromatography April 18-22 Fee E750 including residence and all meals. Non-residents E600. High-performance Liquid Chromatog- raphy July 4-8 Fee E750 including residence and all meals.Non-residents E600. AA/ICP-AES/ICP-MS September 5-9 Capillary Electrophoresis December 15- 17 For further information contact Mrs S. J. Maddison Department of Chemistry Loughborough University of Technology Loughborough Leicester- shire UK LE11 3TU. Telephone (0509) 222575. 24th Annual Symposium on Environmental Analytical Chemistry May 16-19 Ottawa Canada For further information contact M. Malaiyandi CAEC Chemistry Department Carleton University 1255 Colonel By Drive Ottawa Ontario Canada. International Symposium on Microchemical Techniques (ISM '94) May 16-20 Montreux Switzerland Details can be found in J Anal. At. Spectrom. 1993 8 63N. For further information contact Nicko & C.R.I. Associes 7 Rue d'Argout F-75002 Paris France. Telephone +33-1-42 334766; fax +33-1-40 419241.ASMS Short Courses on Mass Spectrometry May 22-29 Hyatt Regency Hotel Chicago IL USA Introduction to Interpretation of Mass Spectra Advanced Interpretation of Mass Spectra LC-MS The Art and the Practice GC-MS for Environmental Analysis Practical MS-MS Analysis 4th Annual Flow Injection Atomic Spectrometry Short Course May 24-26 University of Massachusetts Amherst MA USA This three day intensive short course18N JOURNAL OF ANALYTICAL ATOMIC SPECTROMETRY MARCH 1994 VOL. 9 will cover all aspects of the theory and practice of flow injection techniques in combination with atomic spectrometry. In addition to lectures and discussion sessions the course will feature hands-on experiments with a variety of equipment including atomic absorption and plasma mass spectrometry as well as commer- cially available FIAS systems.Some emphasis will be given to environmental and clinical applications. The course will be tutored by members of the UMass Analytical Chemistry Group and a number of external experts from both academic and industrial organizations. Fees The fee for the course is $980 including 3 nights accommodation all meals full documentation lecture notes and practi- cal schedules. For further information contact; Dr Julian Tyson Department of Chemistry University of Massachusetts Amherst MA 01003 USA Telephone (413) 545 0195; fax (413) 545 4846/4490. 42nd ASMS Conference on Mass Spectrometry and Allied Topics May 29-June 3 Hyatt Regency Hotel Chicago IL USA For further information contact Judith A.Sjoberg ASMS 815 Don Gaspar Sante Fe NM 87501 USA. Telephone (505) 989-4517; fax (505) 989-1073. Scandinavian Symposium on Infrared and Raman Spectroscopy SSIR-94 May 30-June 1 Department of Chemistry University of Bergen N-5007 Bergen Norway Details can be found in J. Anal. At. Spectrorn. 1993 8 48N. For further information contact Dr Alfred A. Christy Department of Chemistry University of Bergen. Telephone + 47-55-213363; fax + 47-55-329058. Laila Kyrkjebo Department of Chemistry University of Bergen telephone + 47-55-213342. Biosensors '94 The Third World Congress on Biosensors June 1-3 New Orleans LA USA Details can be found in J. Anal. At. Spectrom. 1993,8,42N. For further information contact Kay Russell Conference Department Elsevier Advanced Technology Mayfield House 256 Banbury Road Oxford UK OX2 7DH.Telephone f 4 4 (0) 865 512242; fax +44 (0) 865 310981. 6th International Conference on Flow Analysis June 8-11 Toledo Spain Details can be found in J. Anal. At. Spectrom. 1993 8 42N. For further information contact M. Valcarcel/M. D. Luque de Castro Flow Analysis VI Departmento de Quimica Analitica Facultad de Ciencias E-14004 Cordoba Spain. Telephone 34 57 218616; fax 34 57 218606. Short Course in Radioisotope Techniques June 19-24 Loughborough University of Technology Loughborough U K Details can be found in J . Anal. At. Spectrom. 1994 9 6N. For further information contact Dr Peter Warwick Department of Chemistry Loughborough University of Technology Loughborough Leicester- shire UK LEll 3TU. Telephone (0509) 222585; fax (0509) 233163.7th International Symposium on Resonance Ionization Spectroscopy and Its Applications (RIS-94) July 3-8 Bernkastel-K ues Germany Details can be found in J. Anal. At. Spectrom. 1993 8 59N. For further information contact RIS-94 R. Chitty Institut fur Physik Universitat Mainz Postfach 39 80 D-55092 Mainz Germany telephone 0049-61 3 1-393628; fax 0049-6131-393428; telex 418-7155 phmz d; email RIS94@VIPMZA. PHY SIK.UN1-M AINZ.DE. Spectroscopy Across the Spectrum IV Techniques and Applications of Analytical Spectroscopy July 11-14 University of East Anglia Norwich UK Details can be found in J. Anal. At. Spectrom. 1993 8 59N. For further information contact Dr D. L. Andrews School of Chemical Sciences University of East Anglia Norwich UK NR4 7TJ.Fax 0603 25936. Seventh Biennial National Atomic Spectroscopy Symposium July 20-22 University of Hull Hull UK Details can be found in J. Anal. At. Spectrom. 1994 9 7N. For further information contact Dr Steve Haswell School of Chemistry University of Hull Hull UK HU6 7RX. 40th Canadian Spectroscopy Conference August 8-1 0 Halifax Nova Scotia Canada The 40th Canadian Spectroscopy Conference organized by the Spectroscopy Society of Canada will be held jointly with the 1994 Annual Conference of the Canadian Society for Mass Spectrometry at Dalhousie University of Halifax Nova Scotia Canada. Contributions from all areas of spectroscopy (atomic magnetic mass molecular vibrational etc.) are invited. The programme will include the follow- ing sessions. (i ) (ii) (iii) For Analysis of groundwater and potable water; analysis of marine toxins; applications of mag- netic resonance; applications of spectroscopy in geoscience; applications of vibrational spec- troscopy; chemical speciation; gamma ray spectroscopy; graphite furnace techniques; and marine applications of spectroscopy.Mass spectrometry (joint with C S M S ) biochemical applications; environmental and industrial appli- cations; fundamentals; and ioniz- ation methods. Plasma spectroscopy; and recent advances in ultratrace analytical techniques. further information contact Dr W. D. Jamieson Fenwick Laboratories Ltd. 5595 Fenwick St. Suite 200 Halifax NS Canada B3H 4M2. Telephone (902) 420-0203; fax (902) 420-86 12. 13th International Mass Spectrometry Conference August 29-September 2 Budapest Hungary For further information contact Hungarian Chemical Society; FO u.68 H-1027 Budapest. Hungary. Telephone 361 201 6883; fax 316 15 61215.JOURNAL OF ANALYTICAL ATOMIC TSPECTROMETRY MARCH 1994 VOL. 9 19N 7th Conference on Atomic Spectroscopy in Chemical Analysis August 30-September 2 Pardu bice Czech Repu bl ic The Conference is organized by the Spectroscopic Society of J. Marcus Marci in cooperation with the University of Chemical Technology Pardubice at the University Conference Centre and is held every 3-4 years. Scope and Topics The Conference will cover all types of optical and mass atomic spectroscopy as applied to chemical analysis. New instru- mental and/or methodical developments analytical applications including data treatments protocols for QC/QA use of SRM and laboratory accreditation will be discussed.Separate sessions each introduced by a selected paper will be held on these topics. Oral presentations should preferably be in English however lectures in Czech or Slovak will also be accepted. Posters must be exclusively in English. An instrument exhibition will take place concurrently. Location The conference will be held at Pardubice a town about 100 km east of Prague and easily accessible by bus or by train (main railway line from Prague to the East). All meetings sessions and the instrument exhibition will take place at the University Conference Centre. Accommodation Participants will be accommodated in the University campus or at Hotel Garni Synthesia located within walking dis- tance of the Conference Centre.Deadlines Submission of abstracts 30 June 1994 Registration 30 June 1994 For further information contact Spectroscopic Society JMM Thakurova 7 166 29 Praha 6 Czech Republic. Tele- phone 042-2-31 12343; fax 042-2-3 112343. East European Furnace Symposium Sep tem ber 4-7 Warsaw Poland A symposium on electrothermal atomiz- ation sponsored by the Military University of Technology Warsaw the Polish Academy of Sciences and Perkin- Elmer will be held at the campus of the Military University in Warsaw Poland. The symposium continues the series of conferences organized by Professor Boris L'vov in St. Petersburg over many years but with more international partici- pation. One of the main intentions is to stimulate scientific discussions between spectroscopists from Eastern and Western Europe America and the Far East.Topics Furnace techniques and materials; atom formation and distribution; calibration techniques; coupling techniques (matrix separation analyte preconcentration speciation); multi-element furnace deter- minations; solid sampling techniques; and applications. Programme The symposium will consist of Plenary Lectures Oral Presentations and Posters. There will be no parallel ses- sions. The conference language will be English. Short Course on Graphite Furnace AAS A 1 day course on advanced furnace techniques will be held on Sunday September 4th. An extra fee of $20 will be charged for this course. Accommodation and Conference Fees The conference fee will be US $250 including accommodation and full board at the University Hall of Residence and a book of abstracts. Participation of Eastern European Scientists is to be encouraged through a limited number of reduced conference fees.For further information contact Dr Gerhard Schlemmer Bodenseewerk Perkin-Elmer GmbH P.O. Box 10 17 61 D-88647 Uberlingen Germany. Fax +49(7551)813511. EUCMOS XXII XXII European Congress on Molecular Spectroscopy September 11 - 16 Essen Germany Details can be found in J. Anal. At. Spectrom. 1993 8 49N. For further details contact Gesellschaft Deutscher Chemiker Abt. Tagungen PO Box 90 04 40 W-6000 Frankfurt 90 Germany. Telephone + 49 697917-366; fax +49 69 7917-475; telex 4 170 497 gdch d. Geoanalysis 9 4 An International Symposium on the Analysis of Geological and Environmental Materials September 18- 22 Charlotte Mason Conference Centre Ambleside UK Details can be found in J.Anal. At. Spectrom. 1993 8 49N. For further information contact Mr D. L. Miles Analytical Geochemistry Group British Geological Survey Kingsley Dunham Centre Keyworth UK NG12 5GG. Telephone 0602 363100; fax 0602 363200. 7th International Symposium on Environmental Radiochemical Analysis September 21-23 Bournemouth U K Dates to Note Synopses of papers January 28 1994. Final data for registration July 15 1994. For further details contact Dr. P. Warwick Department of Chemistry Loughborough University of Tech- nology Loughborough Leicestershire UK LEll 3TU. Telephone 0509 222585 or 0509 222545; fax 0509 233163. 1994 European Workshop in C herno me tr ics September 25-30 Bristol University Bristol UK Details can be found in J.Anal. At. Spectrom. 1994 9 8N. For further information contact Janice Green School of Chemistry University of Bristol Cantock's Close Bristol BS8 lTS UK. Telephone (0272) 303030 extn. 4421 or (0272) 303672; fax (0272) 251295. 21st Annual Conference of the Federation of Analytical Chemistry and Spectroscopy Societies (FACSS) October 2-7 St. Louis MO USA Details can be found in J. Anal. At. Spectrom. 1994 9 (1). For further information contact FACSS 198 Thomas Johnson Drive Suite S-2 Frederick MD 21702-4317 USA. Telephone (301) 846 4797. 6th International Colloquium on Solid Sampling With Atomic Spectroscopy October 1 1 - 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.20N JOURNAL OF ANALYTICAL ATOMIC SPECTROMETRY MARCH 1994 VOL. 9 Third Rio Symposium on Atomic Spectrometry November 6- 12 Venezue La Details can be found in J. Anal. At. Spectrom. 1993 8 64N. For further information contact Professor Jose Alvarado Universidad Simon Bolivar Departamento de Quimica Laboratorio de Absorcion Atomica Apartado postal No. 89000 Caracas 1080-A Venezuela. Fax (0058-2) 938322/57 19 134/5763355/ 962 1695. Analytica '944econd National Symposium 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 Colloquium Spectroscopicum Inter- nationale (CSI) XXIX August 27-September 1 1995 Leipzig Germany Details can be found in J. Anal. At. Spectrom. 1993 8 50N. For further details contact Prof. Dr. H. Nickel Universitat Leipzig FB Chemie FG Atomspektroskopie Linnestr. 3 D-04103 Leipzig Germany. Telephone and fax (49)-341-6858377. UFZ-Centre for Environmental Research Department of Analytical Chemistry Permoserstr. 15 D-043 18 Leipzig Germany. Telephone (49)-341-235-2370; fax -2625.

ISSN:0267-9477    

DOI:10.1039/JA994090017N

出版商:RSC    

年代:1994

数据来源: RSC

摘要:

20N JOURNAL OF ANALYTICAL ATOMIC SPECTROMETRY MARCH 1994 VOL. 9 Future Issues will include- Atomic Absorption Another Way for Phase Transition Characterization- Walter Serge and Andre Hatterer Theoretical Calculation of the Standard Deviation and Detection Limit in Induc- tively Coupled Plasma Emission Spec- trometry-Evgeniy D. Prudnikov Jaap W. Elgersma and Henri C. Smit Determination of Total Mercury in Scalp Hair of Humans by Gold Amalgamation Cold Vapour Atomic Absorption Spec- trometry-Carlos G. Bruhn Calibration in Flame Atomic Absorption Spectrometry Using a Single Standard and a Gradient Technique-Ignacio Lopez Garcia Pilar Vinas and Manuel Hernandez Cordoba Determination of Inorganic Arsenic in Seafood Products by Microwave-assisted Distillation and Atomic Absorption Spectrometry-Jose C.Lopez Carmen Reija R. Montoro Maria Luisa Cervera and Miguel De La Guardia Spectral Interference on the Lead 283.3 nm line in Zeeman-effect Atomic Absorption Spectrometry-U. Kurfurst and J. Pauwels Determination of Arsenic Chromium Selenium and Vanadium in Biological Samples by Inductively Coupled Plasma Mass Spectrometry Using On-line Elimination Interference and Pre- concentration by Flow Injection-Les Ebdon Andrew S. Fisher and Paul J. Worsfold Inductively Coupled Plasma Mass Spec- trometric Determination of Low-level Rare Earth Elements in Rocks Using Potassium-based Fluxes for Sample Decomposition-Alessandro Rivoldini and Sandro Fadda Direct Determination of Metals in Oils by Inductively Coupled Plasma Atomic Emission Spectrometry Using High Temperature Nebulization-Johann L.Fischer and C. J. Rademeyer Reduction of Matrix Effects and Mass Discrimination in Inductively Complex Plasma Mass Spectrometry with Optim- ized Argon-Nitrogen Plasmas-Diane Beauchemin and Grace Xiao Gas-phase Re-distribution of Analyte Species in the Integrated Contact Cuvette Furnace Atomization Plasma Emission Spectrometry Source-Shoji Imai and Ralph E. Sturgeon Speciation and Preconcentration of Trace Elements With Immobilized Algae for Atomic Absorption Spectrophoto- metric Detection-Hayat A. M. Elmahadi and Gillian M. Greenway Choice of Fluorescence Wavelengths for the Determination of Trace Amounts of Chlorine by Graphite Furnace Laser- excited Molecular Fluorescence Spec- trometry of Indium Monochloride- Evelyn G. Su and Robert G. Michel Determination of Silver by Electrother- mal Atomic Absorption Spectrometry After Complexation and Sorption on Carbon-A. K. Avila and A. J. Curtius Capacitively Coupled Plasma with Tip- ring Electrode Geometry for Atomic Emission Spectrometry. Analytical Performance and Matrix Effect of Sodium Chloride and Potassium Chlor- ide-Emil A. Cordos Sorin D. Anghel Tiberiu Frentiu and Andrian Popescu

ISSN:0267-9477    

DOI:10.1039/JA994090020N

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/JA99409FX025

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 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 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/JA99409BX027

出版商:RSC    

年代:1994

数据来源: RSC

摘要:

JOURNAL OF ANALYTICAL ATOMIC SPECTROMETRY MARCH 1994 VOL. 9 131 Theoretical and Practical Limits in Atomic Spectroscopy* Plenary Lecture J. D. Winefordner G. A. Petrucci C. L. Stevenson7 and B. W. Smith Department of Chemistry University of Florida Gainesville FL 3261 1 USA Theoretical expressions are given for the efficiency of detection and the efficiency of measurement for several atomic methods including atomic absorption atomic emission atomic ionization atomic fluor- escence and mass spectrometry where flames plasmas and furnaces are used to produce atoms or ions and in some cases excited atoms and ions. These unique expressions are then used with noise expressions to develop detection limit expressions. Assuming reasonable values of instrumental and spectroscopic parameters efficiencies of detection and measurement and detection limit of atoms in the sample are estimated to within an order of magnitude.Several methods have the capability of being single atom measurement approaches and therefore potentially useful for atom counting. Considerable discussion of noise sources and a comparison of atomic methods with respect to a variety of analytical figures of merit are given. The present fundamental approach is used to predict the future potential of various atomic methods. Keywords Atomic spectroscopy; efficiency of detection; efficiency of measurement; limit of detection; limit of guarantee; shot noise flicker noise The ability to detect atoms1 of a given element in analytical chemistry is largely governed by the signal-to-noise (S/N) ratio of the measurement.Kaiser’72 was the most influential scientist in elevating the limit of detection (LOD) and the limit of guarantee of purity (LOG) to a firm foundation based upon statistical concepts. The LOD corresponds to an S/N=3 and the LOG to an S/N=6. The limit of quantitation (LOQ) corresponds to an S/N= 10. The reciprocal of S/N times 100 is the percentage relative standard deviation (%RSD) of the measurement. The LOD has a low probability of false positives (type I error) 0.14% but a high probability of false negatives (type I1 error) 50% whereas the LOG a more conservative analytical figure of merit has very low probability of a false positive below 0.14% and a low probability of a false negative about 0.14%. In this paper a review or discussion of the statistical concepts of LOD LOG and LOQ will not be given since these have been thoroughly discussed by Kaiser3 and by the present group of but rather the major analytical figures of merit that limit the LOD and LOG will be discussed.These are the efficiency of detection (Ed) and the efficiency of measurement (E,) of the most prominent and/or potentially most analytically useful atomic methods. In addition there will be a discussion of critical background noise sources in each atomic method and how these background and noise levels affect the LOD and LOG and estimations of Ed E the critical noises and the limiting detectable numbers and concen- trations of species. A synopsis will also be given of the atomic methods in terms of their overall detection powers their use in counting atoms their use in analytical chemistry and a biased view of the future of these techniques in analytical chemistry.A glossary of all parameters their definitions and units is given in Appendix I and a definition of all acronyms in Appendix 11. Analytical Figures of Merit Spectral Selectivity Spectral selectivity is not generally an analytical figure of merit although and Fujiwara et aL7 have developed quanti- * Prepared for presentation at the XXVIII Colloquim Spectroscopicurn Internationale (CSI) York UK June 29-July 4,1993. t Present address Advanced Monitoring Development Group Health and Safety Research Div. Oak Ridge National Laboratory Oak Ridge TN 37831-6101 USA. 1 The term atom will be used throughout even though the detected species could be atoms or ions.tative approaches for it. In this paper spectral selectivity will be discussed on a qualitative basis. All atomic methods have inherently high spectral selectivities atomic emission being the poorest and atomic absorption slightly better. Atomic ioniz- ation methods with single-wavelength excitation are somewhat better but are surpassed still by atomic ionization with two- wavelength (two colour) excitation atomic fluorescence with single wavelength excitation and atomic mass spectrometric methods the last three being nearly equivalent. Atomic fluor- escence with two colour excitation has the highest spectral selectivity. The assignment of this order is based on the complexity of the spectrum of each element and the number of independent excitation and/or measurement steps.Efficiency of Detection &d The efficiency of detection first coined by Alkemade8*9 in landmark papers was defined as the probability that a given atom appearing in the probed volume produced an event during the probing time. Since Alkemade was only referring to laser induced fluorescence and laser induced ionization techniques the probed volume was the effective volume irradiated by the laser and ‘observed’ by the detector and the probing time was the duration of a single laser pulse. In addition Alkemade was concerned only with detecting atoms in the absence of extrinsic noise sources. In the present paper the efficiency of detection’@’’ will be defined more generally to include atomic emission atomic absorption and mass spectrometric methods as well as laser induced fluorescence/ionization and will also take into account the presence of extrinsic noise.Therefore the efficiency of detection Ed will be defined here as the probability that a given atom appearing in the probed (measured) volume produces a signal that is detected above the background noise (if any) during the residence time of the atom within the observation region. Note that this definition differs from Alkemade’s in two important aspects (i) &d is defined with respect to a residence time of the atom in the detection volume not simply over a single ‘probing’ of the laser; and (ii) Ed takes into account the need to detect counts due to analyte atoms over the back- ground noise. Thus &d is the probability that a single atom in the detection volume produces a sufficient number of counts to give a signal above the detection limit X,.Efficiency of Measurement E The overall efficiency of meas~rement,~*~’~ E is defined as the probability that a given atom in the sample is detected above132 JOURNAL OF ANALYTICAL ATOMIC SPECTROMETRY MARCH 1994 VOL. 9 the background noise in the 'observation' region and so is the counts per atom in the sample per residence time and is also dimensionless. Unlike Ed the value of E is related to analyte concentration in the sample since it accounts for analyte atom losses between the sample and the detection region and for insufficient spatial and temporal probing of the analyte atoms as they appear within the detection volume.The relationship between E and Ed is given"." by Em = EIEVEa,i&TEsEtEd (1) where is the sample introduction efficiency which is less than unity for continuous sample introduction systems such as nebulization of a sample solution into a spray chamber; cV is the sample matrix vaporization efficiency which accounts for vaporization of desolvated particles in nebulization systems and vaporization of solids in systems using discrete samples such as furnaces; &a,i is the efficiency (also called free atom or free ion fraction) of atomization or ionization depending upon the measurement method; E~ is the transport efficiency of the analyte atoms to the detection system; E is the spatial probing efficiency accounting for the fraction of analyte signal being spatially measured; and E is the temporal probing efficiency accounting for the fraction of analyte species within the detection region during interaction with the laser.All efficiencies are dimensionless. The first three efficiencies E E ~ E ~ can be sample and analyte matrix dependent but in the estimations of order of magnitude of values of E no attempt will be made to include sample specific conditions. The fourth and fifth terms E ~ E ~ depend primarily upon how well the observation system views the process spatially and how well the analyte is temporally observed; for example in a pulsed laser fluorescence or ionization experiment it is possible only one in every 100 analyte species is detected with a low repetition rate pulsed laser yielding an E of 1 x Background Noise and Detection Limits Noise is certainly not an analytical figure of merit but does directly affect the LOD LOG and precision and so will be considered in this section.Noise in a measurement can be c l a ~ s i f i e d ~ ~ ~ ~ ~ ~ ' ' as either extrinsic or intrinsic. Extrinsic noise8.9~1s17 is the noise which arises owing to a non-specific background signal that is present even in the absence of analyte. Extrinsic noise sources include for example dark current source light scatter background emission and non- selective detection of atoms/ions in the blank and are generally classified as shot and flicker noises. In a counting experiment it could be possible to reduce all extrinsic noise sources to the extent that the probability of registering one or more counts from the blank is negligible during the measurement time. For such measurement the only noise on the signal is due to the statistical nature of the analyte detection process itself; this is intrinsic noise.Intrinsic noise arises from such sources as the varying number of atoms within the detection region during the measurement and shot noise in analyte signal production. As intrinsic noise arises from the detection of analyte atoms it cannot be removed. Signal noise in extrinsic-limited measurements is due to contri- butions from both extrinsic and intrinsic noise. Since there are no background counts at the intrinsic limit it can be assumed that every count is due to the presence of analyte atoms in the detection region. Thus at the intrinsic noise limit the value of Ed is simply the probability that an atom appearing within the detection volume will produce at least one count.The detection efficiency at the intrinsic limit is a special case and will be denoted by the symbol &do. Likewise E,' is defined as the probability that any analyte atom in the sample will produce one or more counts and can be calculated by using cdo in eqn. (1). The values of Ed' and 8,' are characteristic figures of merit of an analytical method and will be called the intrinsic detection (&do) and measurement (E,,.,') efficiencies. They represent the capability of analyte atoms to produce signal counts regardless of the noise level. Thus while the values of Ed and E provide a means of determining the capability of a method to detect single atoms above the current background noise level (see Appendix 111) the values of &do and E,' allow for a comparison of the signal production probabilities of various analytical methods (independent of the noise level). At the intrinsic limit any detected event is due to the presence of analyte atoms.At a given analyte sensitivity as the noise level increases (owing to extrinsic noise sources) &d will fall below &do and E will correspondingly fall below E,'. This occurs for extrinsic-limited measurements because it is not possible to determine whether a given count is due to analyte signal or extrinsic noise. However with detection techniques which use non-destructive detection (e.g. resonance fluorescence) it is possible that a single atom can give rise to more than one count.For such techniques it is still possible to detect the presence of single atoms in the detection region in the extrinsic noise limit case if the sensitivity (in counts per atoms) is high enough. Destructive techniques like those involving ionization or those involving traps in fluorescence processes can give rise to only one count per atom. The sensitivity Y required to detect single atoms using non- destructive detection at various mean blank signal levels is given in Table 1. Note that for destructive techniques such as mass spectrometry r/( 1 count per atom)=&d' and single atom detection (SAD) is only possible at the intrinsic limit. The second column of Table 1 gives the sensitivity necessary to give a signal probability distribution centered on the signal detection limit Xd (assuming a Poisson distribution).' At this sensitivity it could be claimed that a limit of detection of a single atom is achieved.However this claim might be mislead- ing although single atoms can be detected there is a high probability (approximately 0.5) that atoms passing through the detection region are not detected. A technique which is capable of true SAD as defined by Alkemadesi9 and by the present group of will be capable of detecting each and every atom that passes through the detection region; this corresponds to &d z 1. The necessary sensitivities for non- destructive detection are given in the third column. Destructive detection can only achieve SAD at the intrinsic (LOD) limit with Y z 1 count per atom. More details of SAD theory can be found in refs.10-12. Table 1 Two limits for an SAD experiment. Distribution of back- ground and signal counts assumed to follow a Poisson distributionlG12 Mean blank level/ counts 0.00 0.05 0.25 1 .oo 5.00 10.00 100.00 Y (mean sensitivity)/counts per atom LOD= 1 atom* a G0.0014 1 2 4 5 9 12 32(30)f LOG= 1 atom? p~0.0014 6.6 8.9 12.4 15 23 29 (69)$ * This column gives the mean sensitivity necessary to produce a signal distribution centered at the signal detection limit Xd.10-12 At LOD the probability of false positives a is low but there is a relatively high probability (approximately 0.5) of false negatives p. t This column gives the mean sensitivity required to reduce the false negative probability p to an acceptable level.Note that LOG = 1 atom is necessary to achieve single atom detection (SAD) which is defined as the capability of detecting each and every atom which passes through the detection volume.'0-'2 f The values in parentheses were found by assuming a Gaussian distribution with the appropriate LY and p values. At these higher signal levels the Poisson distribution can be approximated by a Gaussian distribution with p = c2.133 JOURNAL O F ANALYTICAL ATOMIC SPECTROMETRY MARCH 1994 VOL. 9 Atomic Methods The atomic methods to be compared in this paper are the following optical emission spectrometry13J4 (OES) in flames (F) and inductively coupled plasmas (ICP); atomic absorption spe~trometryl~.'~ (AAS) in flames and electrothermal atomiz- ers (ETA); laser induced fluorescence spectr~rnetry'~-'~ (LIFS) in flames inductively coupled plasmas glow discharges (GD) electrothermal atomizers and atomic beams (AB); laser induced ionization s p e c t r ~ m e t r y l ~ - ~ ~ J ~ ~ ~ ~ (LIIS) in flames and in flames with introduction by electrothermal vaporization (ETV); res- onance ionization spectromet ry20-24 (RIS) in atomic beams and atomic mass s p e c t r ~ m e t r y ' ~ ~ ~ - ~ ~ (MS) with sample intro- duction via an ICP a GD or an AB.The symbolism for electrothermal vaporization of a sample into a flame with two colour (TC) (two different excitation wavelengths) excitation of laser induced ionization spectrometry would be ETV-F-TC- LIIS. Other combinations result by similarly combining other systems and will be used in this paper.No review of the theoretical basics instrumentation and applications of the individual atomic methods will be given here. It will be assumed that the reader is familiar with the general aspects of the individual methods or if not familiar the reader is referred to the appropriate references cited previously. Estimation of Detection Limits The requirement for SAD (also see Appendix 111) is that &d % 1. It is also possible to calculate detection limits using &do and E,' values even in cases when &do< 1. For both destructive and non-destructive detection the following can be written N,= YNA (2) where N = number of detected count events during t (measurement time) or z (counts) Y = sensitivity (counts per atoms) and NA=number of analyte atoms detected in the sample during the atom residence time in the region of observation.The probability of an analyte atom producing one (or more) counts is given by &do (3) where Ati is the interaction time (s). The meaning of $s (counts s-l) depends on whether destructive or non-destructive detec- tion is For non-destructive detection & is the mean flux of detected events; for destructive detection $s is the reciprocal of the mean detection time once the atom has entered the detection region. For OES AAS and CW AFS Ati=z the residence time for the atom in the observation region. For pulsed (laser) source methods LIFS LIIS RIS RIMS etc the following can be written 0 - 1 -e-@sAti &d - Ed0(1)= 1 -e-@-qAtl (4) where &do( 1) is the &d value for one single laser pulse and Atl is the laser pulse width.If zfi < 1 and z,>> At then &do = &do( 1 );A is the laser repetition rate (s-l). However if there are p probings (where p = z f i and zfi> 1) per atom residence time in the detection region then &do@) is given by where &do(p)=&dO for p pulses. It should be noted that &do( 1) is identical with the Alkemades.9 definition of Ed. It is clear that the sensitivity of measurement Y is given by (6) &do(p)=f-[ 1 -&do(1)lp ( 5 ) Y = &I&V&,,i&T&,&t&dO counts per atom for destructive methods (LEI RIMS RIS etc.) and for non- destructive CW methods such as AAS OES and CW and pulsed laser fluorescence methods where &do<< 1 and by for non-destructive pulsed methods (LIF) where &do > 0.05. Y = &I&Vea,i&T&s&t$sAtlp counts per atom (7) To be detected the analyte atoms must produce a large enough signal N so that Note that this value is given for single atoms in the second column in Table 1 for various mean blank z values.The LOD in atoms in the sample N is given by (9) Xd - xb NL=- Y This equation provides a link between the LOD (in terms of numbers of atoms) and E,' as long as the &d and values used to determine N are calculated at the intrinsic noise limit i.e. &d=&dO and E,=E,O. Eqn. (9) is valid for the extrinsic as well as the intrinsic noise limits with only the numerator changing depending upon the noise type. For example for the case of the background shot noise limit the numerator Xd - Xb becomes three times the square root of the number of back- ground counts during the species residence time or measure- ment time i.e.3 X where 3 is the confidence factor as - defined by Kaiser. G Evaluation of cdo and 8,' for Atomic Methods In Table 2 expressions for &do for various atomic methods are given. In Tables 3-6 estimations of &do and E,' for the various atomic methods are given. Such estimates have never been given in a consistent manner for the major analytical atomic methods. All values used for the various parameters for each atomic method are given at the end of each table. The values for the parameters represent in all cases possible magnitudes for existing experimental systems. The intent here is to estimate &do and 8,' values that are possible with each atomic method under good operating conditions.It would of course be impos- sible to give an exhaustive listing of &do and &,O values for all possible conditions. It should also be pointed out that certain liberties were taken with the choice of parameters; for example it was assumed for the ICP-OES case that all of the atomic species were in the correct form (atoms or ions) when in fact this fraction ( E ~ ) could be much less than unity for certain elements. In addition for consistency and simplicity an atom residence time of 1 ms was chosen for all cases except for ETA where 1 s was chosen. Also only two hypothetical resonance lines (200 and 500 nm) were chosen for ICP-OES F-OES and AAS and only two flame temperatures (2000 and 3000 K) and one plasma temperature (6000 K) were assumed.For the laser based and the MS methods only a single set of reasonable (except for the optimistic selection of a 3000 Hz tunable laser) conditions were chosen to minimize the number of cases possible. In the case of the AB (in LIFS RIS and RIMS) it was assumed that the atoms produced in the furnace (ETA) were either excited (and fluoresced) or ionized immediately adjacent to the furnace orifice so that E ~ E = 1. If a true atomic beam is produced and excited/ionized down-field from the orifice E~E,<<I resulting in much smaller &do and E,O values and much larger N and cL values. The reader is encouraged to estimate &do and em0 values for their own specific atomic systems. Estimation of Magnitudes of Extrinsic Noises in Atomic Methods The intrinsic noise limit at the detection limit exceeds the extrinsic noise if the following equation is valid (Poisson distribution of analyte signal is assumed) next < (NL&rn0)li2 (10) It should be pointed out that in reality the intrinsic noise limit is achieved when there is no extrinsic noise not simply when the intrinsic noise on the signal exceeds the extrinsic134 JOURNAL OF ANALYTICAL ATOMIC SPECTROMETRY MARCH 1994 VOL.9 Table 2 Specific expressions for &do for atomic methods*? I( I Optical emission spectrometry (0ES)- Ed" = [ 1 -exp(-4Ezrrem~el?d)l 4 ~ = Vex& I1 Atomic absorption spectrometry (AAS)$- A Narrow line source if EPL<EpSat &do = [ -exp(-$ATr)I B Broad line or pseudo-continuum source if EplC<Eppt -exp(-#Azr)I 111 Pulsed laser induced resonance Juorescence spectrometry (LIFS)*+ A Single-colour excitation )= -exp(-gA,lAtirflrdrel)l B Two-colour excitation &do( = [ -exp(-g'Au,l,Atiyfl?d'l,l)l IV Pulsed laser induced ionization spectrometry ('LIIS)*k A Single-colour excitation B Two-colour excitation1 -exp(-g'ku,iAtiqdr,l)l C Two-photon (single wavelength) excitation1 [ -exp(-~Z,hEpz((nu,)A~i~drel)l V Pulsed resonance ionization spectrometry (two or more colour excitation) *F VI Mass spectrometry (MS)- * The parameter p accounts for the probing of the same atom in a pulsed laser experiment by more than one laser pulse.p = 1 if residence time of the atom times the laser repetition rate is < 1.However if that product exceeds unity then eqn. ( 5 ) is used; e.g. p would be 3 if the residence time of an atom is 1 ms and a 3000 Hz repetition rate laser is used ( 1 x lop3 x 3 x 103=3 pulses per atom). ? See Appendix I for definition of all parameters and their units. $ Line source = hollow cathode lamp; pseudo-continuum source = xenon arc lamp with spectrometer. ij Assumptions for &d expressions for LIFS LIIS and RIS. The laser is assumed to be broad band i.e. its spectral profile is larger than the absorption profile of the atoms in all the atomizers considered. In addition the rate equations approach was used15 to derive the expressions and the laser is assumed to be characterized by a rectangular temporal profile. For the ion yield expressions recombina- tion between ions and electrons was neglected.In all expressions both the first and second (where applicable) transitions are considered to be optically saturated during the entire interaction time. For LIFS no metastable states are considered. 1 Assumes no collisional or radiational losses from level u'. 11 The relationship between the absorption cross-section (iA and the Einstein coefficient of spontaneous emission Aul is given by where A& is the absorption FWHM which is assumed to be Lorentzian. noise level. Otherwise the atomic system is limited by extrinsic noise. For all atomic methods near the limit of detection "background' shot or flicker extrinsic noise will be limiting. 'The 'background' or 'blank' refers to all non-analyte noises such as those due to background emission background related to the source of excitation detector etc.Background shot noise It& only on the magnitude of the rate of background detected events Ri (count s-') and the time of measurement. For a CW detection system and one residence time and for a pulsed dete'ction ~ y s t e r n ~ ~ ~ ~ where Atg is the gate width of the detection system andfi and z are as defined above. For the case of ETA cells z is 1 s. Typical extrinsic noise estimates are given in Table 7. For the case of extrinsic noise it should be stressed that 'background' shot noise predominates at low background fluxes (photon s-l) but 'background' flicker noise (tb flicker factor for background emission or t flicker factor for source emission) predominal es at high background fluxes.Comparison of Absolute and Relative Detection Limits In Table 8 a theorei:ical comparison of detection limits NL (absolute number of atoms at detection limit) and cL (concen- trational detection limit in mass of analyte divided by mass of sample; for example 1 x lo-'' is a part per trillion) is given for the various atomic methods with several cell and/or source types and the limiting nine sources. The theoretical possibility of atom counting with each method is also given. It should be stressed that the masses of sample assumed for the various sample introduction devices have been liberally chosen; for example a mass of 1 x g for all nebulization and ETV experiments (an asp:iration rate of 6 ml min-' for zr= 1 ms) was chosen.For all furnaces (ETA) a sample mass of 0.1 g was assumed during the atom residence time of 1 s and for the GD and AB a sample mass of 1 x lop9 g was assumed to be sputtered vaporized atomized and/or ionized and/or atom- ized during each millisecond (7 = 1 ms). In Table 9 an experimental comparison of atomic methods for solution samples is given. The reader should also refer to the excellent review by Sjostrom and M a ~ c h i e n . ~ ~ Conclusions Based Upon 8 2 E,' NL cL and Noise Estimates The estimations given in Tables 3-9 for &do E,' noise NL and cL should be used only in comparing one method with another. If one wishes to estimate any of these parameters for a given system then specific values of the parameters must be used. However several conclusions can be made.and E,' for OES AAS and a variety of other atomic systems are given. It is clear that even if &do were unity and extrinsic noise were absent NL would slill exceed unity since cm0<<1 because of atomic losses probing inefficiencies detection inefficiencies etc. (2) The values of ,:do and E,' for OES for atoms in the ICP and for visible emission lines can approach values of 1 x lop5 and 1 x lo-' respectively and for AAS for both hollow cathode lamp and xenon arc spectrometer excitation can approach values of 1 x lop7 for flames and 1 x lop3 for furnaces (ETAS). Despite the rather impressive detection and measure- ment efficiencies in OES and in AAS the 'background' noise levels in both cases limit the calculated values of NL and cL to values consistent with experimental results.(1) For the first time expressions to estimateJOURNAL OF ANALYTICAL ATOMIC SPECTROMETRY MARCH 1994 VOL. 9 135 Table 3 conditions at end of table) for a measurement time (z,) of 1 ms Order of magnitude values of efficiency of detection and efficiency of measurement for optical emission spectrometry (see experimental 2000 K (flame) 3000 K (flame) 6000 K (plasma) 6000 K (plasma) (NEB) (NEB) (NEB) (ETV) &do Emo Ed0 Emo Ed Emo Ed Em0 i,Jnm A&' 200 1 105 2 x 10-15 1 x 10-19 3 x 10-13 2 x 10-14 5 x 10-5 5 x 10-l0 1 x 1 10-5 1 106 2 10-20 1 10-21 3 10-15 2 1 0 - l ~ 5 x 1 0 - l ~ 5 x 1 0 - l ~ 1 x 1 x 10-l0 1 106 2 10-22 1 3 x 10-1~ 2 x 1 0 - l ~ 5 x 1 0 - l ~ 5 x 1 0 - l ~ 1 x 1 0 - l ~ 1 x 1 0 - l ~ 1 ~ 1 0 9 4x10-9 2 x 10-10 5 x 10-7 3 10-5 8 x lop6 8 x lop8 2 10-4 2 x 10-6 1 106 4 x 10-11 2 x 10-12 5 x 10-9 3 x lo-'' 8 x lop8 8 x lo-'' 2 x 2 x 10-5 500 1 x 104 4 x 2 x 5 x lo-" 3 x lo-'' 8 x lo-'' 8 x 2 x lo-' 2 x lo-'' Notes Conditions for 0ES13,17,29-32 3 x (500 nm 2000 K) 4 x lo-" (200 nm 3000 K); 6 x lo-' (500 nm 3000 K) 6 x 1 x lo8 s-' (highly probable transition) 1 x lo6 s-' (less probable transition) 1 x lo4 s-l (improbable transition) 0.001 cm x 1 cm x 0.01 sr 4nsr 0.1 z = 1 ms 1 0.05 (F); 0.01 (ICP); 1 (ETV) 1 (F or ICP) 1 (F or ICP) 1 (F or ICP with nebulization) (200 nm 2000 K); 5 x (200 nm 6000 K); 1 x lop3 (500 nm 6000 K) cm2=8 x 1 (ETA-ICP-AES) Order of magnitude values of efficiency of detection and efficiency of measurement for atomic absorption spectrometry (see experimental conditions at end of table) for a measurement time (2,) of 1 ms HCL XeS A1,Jnm A,JsC1 (F) 200 1 x 108 5 x 10-7 5 x 10-9 5 x 10-11 500 1 x 105 5 x 10-7 5 x 10-9 5 x 10-11 1 x 106 1 x 104 1 x 106 1 x 104 &do (ETA) E,' (F) E,' (ETA) 5 x 3 x lo-'' 5 x 5 x 3 x 10-l0 5 x 5 x 1 0 - 4 3 x 10-5 5 x 10-4 5 x 10-5 3 10-12 5 x 10-5 5 10-4 3 10-5 5 x 10-4 5 x 10-5 3 x 10-12 5 x 10-5 &do (F) &do (ETA) E,' (F) E,' (ETA) 5 x 10-9 5 x 3 x 10-1° 5 x 5 x lo-" 5 x 3 10-12 5 x 10-5 5 10-13 5 10-10 3 x 10-14 5 x i o - l o 5 x 5 x 10-3 3 x 10-7 5 x 10-3 5 x 10-5 5 x 10-5 3 10-9 5 x 10-5 5 x lo-'' 5 x 3 10-11 5 x 10-7 Notes Conditions for AAS133'7,31*32 A = 1 x lo's-' (highly probable transition) 1 x lo6 s - l (less probable transition); 1 x lo4 s-l (improbable transition). oA = 1 x t = T,= 1 ms (F); t,= 1 s (ETA) A = 200 and 500 nm E = 1 = 0.05 (F); E,= 1 (ETA) cV = E ~ = E = E ~ = 1 (HCL or XeS) cV = cT=l and ~ = ~ ~ = 0 . 0 1 Sb = 1 cm2; WH=0.001 cm2 T,=0.5 Ijdqel=0.1 xtL = 5x10-'cm2 EpLc= 1 x 10l6 photons sK1 cm-2 nm-l at 500 nm EpLc= 1 x photons s-l cm-' nm-l at 200 nm Sb = 1 cm2 WH = 0.001 cm2 T = 0.5 AA = 0.01 nm qdqel = 0.1 xtc = 5 ~ 1 0 - ~ c m ~ n m cm2 (Aul = 1 x lo8 s- l) 1 x 10- l4 (Aul = 1 x lo6 s -') and 1 x cm2 (Au1 = 1 x lo4 s-l).HCL EpL = 1 x lo1' photons s-l cmP2 (z 100 pW cm-2) at 200 or 500 nm XeS (3) The wide range of cdo and emo (and N and cL) and values for OES result because of the Boltzmann distribution of excited states and therefore the significant variation in the fraction of species excited vex with temperature. (4) The narrow range of cdo and emo (and N and cL) values for AAS is based on the rather constant transition probability or absorption cross-section for resonance absorption lines.( 5 ) Of the LIFS and LIIS techniques considered here SAD in the sample is theoretically possible by several LIFS (ETA GD AB) methods. These are methods theoretically capable of single atom counting as shown in Table 8. In fact the intrinsic noise limit is theoretically achieved in those techniques. Certainly SAD in the sample has not yet been achieved but the predictions are encouraging and intriguing. The great variations in c values for single atoms (see Table 8 e.g. LIFS methods have N z 1 atom and cL z 1 x lop2' whereas RIMS and MS (AB) have N L z 1 atom and c L z 1 x is owing to the amount of sample introduced within one atom residence time (0.1 g for ETA and 1 x g for GD and AB).Certainly ETA-LIFS appears to be the simplest method with the greatest potential for achieving near SAD. The new method ETV-F- LIIS should achieve detection limits ( N ) of z 1 x 103-1 x lo4 atoms with c values of 1 x 10-15-1 x like ETA-LIFS ETV-F-LIIS is a relatively simple approach for both solid and solution samples. Of the remaining LIFS techniques GD-LIFS has the greatest potential since it should achieve detection136 JOURNAL OF ANALYTICAL ATOMIC SPECTROMETRY MARCH 1994 VOL. 9 Table5 experimental conditions at end of table) for a measurement time of 1 ms for all methods except those involving the ETA ( 1 s) Order of magnitude values of efficiency of detection and efficiency of measurement lor several pulsed laser atomic methods (see Method LIFS (SC)$ LIFS (TC)$ LIIS (SC)\i LIIS (TC)/I LIIS (TP)JJ RIS (TC)** Bp=l x 106s-l Bp= 1 x 1O'O s-l Edo 30 Hz laser*? 5 x 10-4 5 10-4 5 x 10-4 5 10-4 2 x 10-4 2 x 10-4 2 x 10-4 2 x 10-4 0.02 0.01 5 x 10-3-5 x lo-' 1 .o 1 .o 0.01 4 x 10-3 1.0 3000 Hz laser*? 2 x 10-3 2 x 1 0 - 3 2 x 10-3 2 x 10-3 0.8 (1.5) 6 x lop3 6 x lop3 6 x lop3 0.6 (1.0) 6 x lop3 1 .o 1 .o 0.03 2 x 10-2-2 x 10-4 0.0 1 1 .o 30 Hz laser*? 5 x lo-' 5 x lo+ 3 x 10-7 2 x 10-4 5 x 10-4 2 x 10-7 2 x 2 x 8 x 2 x 3 x 10-'-3 x lo-' 5 x 10-4 5 x 10-4 5 x 4 x 0.01 3000 Hz laser"? 1 x 10-4 2 x 10-3 2 x 10-3 3 x 10-5 2 x 0.8 (1.5) 6 x lo-' 6 x lob3 0.6 6 x ( 1.0) 5 x 1 x 10-3-1 x 10-5 2 x 10-3 1 .o 0.01 1 .o Notes * For fi = 30 Hz and z = 1 ms p = 1 since AT < 1 for all nebulization cases (F ICP) and the GD and AB and p = 30 for the ETA.t For J; = 3000 Hz and zr = 1 ms p =AT = 3 for (F ICP) nebulization cases and for the GD and AB and 3000 for the ETA. $ Conditions for g = 0.5; g'=0.3 Ad Ati zr vfl = 1 x 10' s-1; A,*,,= 1 x lo8 s-' = At,=1 x lo-' s = 1 ms (F ICP GD AB); z = 1 s (ETA) = wfi i22,T0/47dfl= 1 X lo-' VdVel = 1x10-' &a,i = 1 El EV = 1 E r = 1 Ei Et = 0.05 (F); 0.01 (ICP); 1 (GD ETA AB) = 1 x = 1 (30 Hz or 3000 Hz laser) for ETA (30 Hz pulsed laser) and Et= 1 (3000 Hz pulsed laser) for F ICP GD and AB Es = 1 (GD F ICP AB ETA) 9 Measurement time is 1 ms (q). 7 Measurement time is 1 s (tm). 11 Conditions for L I I S ' ~ * ~ O g = 0.5; g'=0.3 kui g2ph Ati V d v e l = 1 &a = 1 EV = 1 Es = 1 Zum = 1x108s-' g = 0.5 A tl v d v e l = 1 Ea = 1 El = 1 EV = 1 ET = 1 Es = 1 = 1 x lo6-1 x 10' s-'; ZUm= 1 x lo8 s-'; ku,i= 1 x 10" s-' = At,= 1 x lo-' s = 1 x cm4 s; E (A,,)= 1 x lo2' photons s - l cm-2 (100 W cmP2) El = 0.05 (F); 1 (ETV-F) ET = 1 ( F or ETV-F) ** Conditions for R I S ' ' S ~ ~ Buipui(Aui) = 1 x lo6 s-' and 1 x 101os-l = At,= 1 x lo-' s Ei = 1 x lop2 (30 Hz pulsed laser); 1 (3000 Hz pulsed laser) limits of near single atoms with concentrations of 0.1-1 pptr in solid samples is simple and could possibly be improved dramatically if the GD cell design minimized dilution and loss of the atomic vapour.( 6 ) It should be stressed that of the LIFS approaches the sensitivity Y in counts per atom exceeds unity (parenthetical values in Table 5) only twice and is only 1.5 in those cases.In other words the sensitivity requirements for SAD (LOG) as shown in Table 1 are not achieved in those two cases or in any method consid,ered in this paper. However it should be possible to achieve a Y>> 1 in LIFS as long as the transition probability is z 1 x lo8 s-' if the residence time of the atom within the observation region is long as in ETAS and if the laser interaction time (Atl) is increased. In the latter instance if it is assumed that At is only increased to 100 ns instead of 10 ns then Y would be 15 instead of 1.5 for the two cases in Table 5 where Y is listed as 1.5. Of course if more efficient collection of the fluorescence and detection of fluorescenceJOURNAL OF ANALYTICAL ATOMIC SPECTROMETRY MARCH 1994 VOL.9 137 Table6 Order of magnitude values of efficiency of detection and efficiency of measurement for several atomic spectrometric methods (see experimental conditions at end of table) for measurement time of 1 ms Ed &mO Method Cell 1 x QPMS NEB-ICP 1.0 TOFMS RIMS Continuous wave Pulsed ETV-ICP GD AB AB 1.0 1 .o 0.01 4 x lo-’ (30 Hz)* 0.01 (3000 Hz)* 1.0 (30 Hz)t 1.0 (3000 Hz)t ~~ 1 x 10-4 1 x 10-4 1 x 10-4 4 x 10-5 (30 HZ)* 0.01 (3000 Hz)* 0.01 (30 Hz)t 1.0 (3000 HzH Notes ~ $ 2 5 2 7 ci = 1 (ICP for many elements) ci qdqel = = 0.01 (GD for many elements) = 0.1 (AB for many elements electron impact) El Ev = 1 E = 1 (ICP); 1 (GD); 1 (AB) E ET = 1 x (ICP-QPMS); 1 x (GD-QPMS); 1 x lo-’ (TOFMS) E = 1 EV = 1 E = 0.01 E = 0.01 (ICP); 1 (GD AB); 1 (ETV) = 1 (QPMS TOFMS); E,< 1 for ITMS and ICRMS qdqel = E I = 1 ET = 1 = 1 (CW RIMS); 1 x lo-’ (30 Hz pulsed laser); 1 (3000 Hz pulsed laser) * Bp= 1 x lo6 s-l.t B p = 1 x 1o’O s-l. photons were achieved this would also increase Y. Therefore non-destructive LIFS methods deserve careful attention for achieving the ultimate goal of SAD. (7) The AB-RIS technique is also theoretically capable of N L % 1 atom and atom counting as shown in Table 8. However the concentration detection limit cL is much poorer (x 0.1 pptr) than the values for LIFS because of the low sample introduc- tion rate in RIS. Of the MS methods only AB-RIMS (with a TOFMS)” and possibly AB-MS are capable of SAD (N,= 1 atom) but both have corresponding cL values of ~ 0 .1 pptr because of the low sample introduction rate. The other MS methods theoretically achieve better detection limits (0.1 pptr for NEB-ICP-MS and 0.1 ppq for ETV-ICP-MS) but cannot achieve SAD because of the losses occurring in the transfer process (cT<<l). The major difficulty with the AB-MS method involving an ideal atomic beam AB produced some distance from the ion source/ion trap is the significant loss in transport ( E ~ ) and spatial probing ( E J of the atoms/ions produced. Falk17 proposed the use of a TOFMS in an AB-MS system and estimated an overall measurement efficiency of the order of = l x 10-5-1 x The use of a TOFMS rather than a QPMS enables this approach to be a truely multi-element method. However no experimental results are yet available.and E,’ values represent maximum values and the NL and cL values represent minimum values in many cases since the efficiency of producing the species of concern (atom or ions) was chosen to be unity for all methods the best case scenario. This assumption is particularly severe in the case of a flame cell where compound formation and ionization can greatly reduce the atomic species e.g. Zr W Hf Os Mo and V in the former case and Na K Rb and Cs in the latter case. If the values of E or ci were known then the E ~ ’ E,’ N and cL values could be readily corrected. The efficiency of vaporization E ~ was also assumed to be unity for all methods which again will be too high for some samples introduced into flames and to a lesser extent for some samples introduced into ETVs and ETAS.The other efficiencies were (8) The less sample and analyte dependent and so their values will be more accurate except possibly for E ~ . Finally in those cases where analyte is transferred from and ETV to a cell e.g. ETV- F-LEIS the E ~ which was assumed to be unity may also be considerably less for some samples. (9) The NL values in this manuscript differ to some extent from those given by Stevenson and Winefordner.” The reasons for this are as follows (i) in this paper all N values were obtained from estimates of the noise and the sensitivity Y and efficiency of measurement E,’ for each method by the general approach given in eqns. (2)-( 9) whereas in ref. 11 the approach differed greatly depending upon the noise source; (ii) in this paper all N values are given for a single atom residence time whereas in ref.11 more arbitrary times were chosen; therefore the N and cL values and their ranges are more consistent than those given in ref. 11; (iii) in this paper values of E ~ E ~ E ~ E E ~ E ~ are given whereas in ref. 11 this product was assumed to be unity in most cases; (iu) in this paper the intrinsic efficiencies of detection and measurement for each atomic method were estimated whereas in ref. 11 this was not done; and (v) in this paper more efforts were taken to be consistent and fair with regard to choice parameters for all atomic methods whereas in ref. 11 the NL values were taken from a number of references. Nevertheless ref. 11 as well as this paper are valuable resources to use in estimating the detection power of atomic methods.(10) It should be stressed that the solution techniques of flame and ICP OES AAS LIFS and LIIS suffer by an additional factor of about 100 when solid samples are to be analysed. Solids commonly require grinding weighing 1 part solid to 100 parts solvent and dissolving. A direct solids approach neglecting the difficulties of solid standards is faster and does not suffer in loss in detection power owing to dilution but does posses difficulties with standardization. The direct solids approaches involve ETA (solutions can also be used) GD (dried solutions can also be used) AB and AT (furnaces with solids or dried solutions can be used),138 Table 7 JOURNAL OF ANALYTICAL ATOMIC SPECTROMETRY MARCH 1994 VOL.9 Estimates of extrinsic noise magnitudes near the LOD in atomic method^^^-^^ ‘Background’ noise counts Method OES AAS LIFS LIF LIIS RIS RIMS MS Cell F ETA F ICP ETA (2000 K) ETA (3000 K) GD AB F AB AB ICP GD Source F ICP ICP HCL HCL XeS XeS HCL HCL PL (SC or TC) PL (SC or TC) PL (SC or TC) PL (SC or TC) PL (SC or TC) PL (SC or TC) PL (SC or TC) PL (TC) TC CL TC PL (TC) TC Noise type BES BES BES SBS SBF SBS 200 nm SBS 500nm SBF 200 nm SBF 500 nm ETA BEF 2000 K 200 nm 2000 K 500 nm ETA BEF 3000 K 200 nm 3000 K 500 nm BES 30 Hz 3000 Hz BES 30 Hz 3000 Hz BES 200 nm 30 Hz 200 nm 3000 Hz 500 nm 30 Hz 500 nm 3000 Hz 200 nm 30 Hz 200 nm 3000 Hz 500 nm 30 Hz 500 nm 3000 Hz DN 30 Hz 3000 Hz DN 30 Hz 3000 Hz BCS 3000 Hz 1 x A 3000 Hz 1 x lO-’A DN 30 Hz 3000 Hz DN 30 Hz 1 x A 30 HZ 1 x 10-9 A DN 30 Hz 3000 Hz DN DN z,=1 s 30 800 5000 1 x 104 5 x 105 5 x 103 2 x 105 1 x 105 1 x lo8 0.25 8 x lo2 80 1 104 0.02 0.2 0.5 5 1 x 10-4 i x 10-3 0.5 5.5 0.009 0.09 8 80 1 x 10-4 1 x 10-3 8 x lop4 8 x lop3 2.5 x 103 a x 104 77 8 x f03 1 1 1 1 1 1 1 z,=1 ms 1 .o 25 5 300 50 2 x lo2 6 x lo3 I 103 1 x 105 8 x 25 2 3 x 10’ 0.0006 0.006 0.005 0.15 3 x 3 x 10-5 3 x 10-4 3 x 10-3 0.01 5 0.16 0.25 2.5 3 x 3 x 10-5 4 x 10-5 4 x 10-4 3 x 103 80 2 3 x f02 8 x 8 x lop3 0.03 8 x 8 x 0.03 0.03 Notes Typical ‘Background’ levels in cells sources detectors Flame2’ (1 not in OH bands or in CN CH C2 bands; photon radiance would be higher by =:.LO-fold in spectral regions of these bands) BpAz 1 x loll photons s-’ cm-2 sr-l nm-I (conventional C2H,-air flame) Typical spectrometer conditions WH = 0.001 cm2 sZE = 0.01 sr; A1 = 0.01 nm To = 0.5; qdqe = 0.1; 4; = 0.01 Background count rate x 5 x 10’ s - ’ ICP30 BpAz 1 x 1014 photons s-’ cm-’ sr-l nm-‘ Typical spectrometer conditions same as for flame Background count rate z 5 x lo5 s-l E TA3’ BpAz 1 x 1014 photons s-l cmP2 sr-l nm-‘ at 500 nm and 2000 K BpAz2 x 10l6 photons s-’ cmP2 sr-l nm-‘ at 500 nm and 3000 K BpAz 7 x lo1’ photons s-’ cm-’ sr-l nm-’ at 200 nm and 3000 K BpAx5 x lo6 photons s-l cm-2 sr-’ nm-’ at 200 nm and 2000 K Typical spectrometer conditions same as for flame Background count rate = 5 x lo5 s-’ at 500 nm and 2000 K Background count rate= 1 x 10’ s - l at 500 nm and 3000 K Background count rate = 0.03 s-’ at 200 nm and 2000 K Background count rate=4 x lo3 s-l at 200 nm and 3000 KJOURNAL OF ANALYTICAL ATOMIC SPECTROMETRY MARCH 1994 VOL.9 139 Table 7 (continued) GD No absolute B values could be found. Based on relative measurements of signals to background levels in HCLs the values of BPn should be < 1 x lo7 photons s-l cm-2 sr-l nm-' at all wavelengths between atomic lines of the fill gas and the elemental species in the gas phase. Using spectrometer conditions for the flame case above the background count rate should therefore be ~ 0 . 0 5 s-l PMTdetector Electron multiplier count rate33-35 1-3 s-l Flame background current20'21 Optogalvanic detector x 1 x 10lO-1 x 1013 s-l for H2 based to C2H2 based flames Source count rate (AAS) HCL? B P z l x 1013-1 x 1014 photons s-l cm-2 sr-l nm-I Typical spectrometer conditions WH = 0.001 cm2 R = 0.1 sr T = 0.5 qdqel =O.1 tS = 0.001 HCL EPL= 1 x 1013 photons s-' cm-2 Xenon arc lamp3' BPA (200 nm)x2 x photons s-' cm-2 sr-l nm-' BPn (500 nm)x2 x 10" photons s-' cm-2 sr-' nm-' Typical spectrometer conditions WH = 0.001 cm' R = 0.001 sr To = 0.5 A& = 0.01 nm q&.l= 0.1 5 = 0.01 Xenon arc E = 2 x 10"-2 x 1OI2 photons s-l cm-' nm-' Source scatter count rates (LIFS)15,16 Laser scatter and environmental fluorescence count rates are assumed to be less than cell background emission count rates and are neglected. This is an excellent assumption if the experimental system is optimized for negligible laser scatter for low fluorescence optics assuming proper baffling is used as well as measurement of atomic fluorescence in the UV Typical pulsed laser conditions Ati = Atl = 1 x f; = 30 or 3000 Hz Atg= 1 x lo-' s Dark count rate x 1 x 10'-1 x lo2 s-l Table 8 Theoretical comparison of several atomic methods Method OES I1 AAS** LIFStT LIIS$$ RIS$§ RIMS$§ MS% Cell NEB-F NEB-ICP ETV-ICP NEB-F ETA NEB-F NEB-ICP ETA ETA GD AB NEB-F ETV-F NEB-F AB AB NEB-ICP ETV-ICP GD AB Source HCL XeS HCL XeS XeS SC TC SC TC SC TC SC TC SC TC SC TC TC TP TC TC Noise limit BES BEF BEF SBS SBS (200 nm) SBS (500 nm) SBF SBF (200 nm) SBF (500 nm) BES BES BES BES DN DN BCS BCS BCS BCS DN DN DN DN DN DN Atom counting* No No No No No No No No No No No Yes Yes Yes Yes No No No No No Yes No No No Yes Theoretical N (atoms)tS 1 x lo8-1 x 10l8 1 x 108-1 x 1014 1 x 107-1 x 1013 1 x 1010-1 x 10'4 1 x 1011-1 x 1015 1 x 109-1 x 1013 1 x 1011-1 x 1015 1 x 1011-1 1015 1 x 102-1 x 104 1 x 104-1 x 106 1 x 10l2-1 x 10l6 x 1 x loo-1 x 10' % 1 x loo-1 x lo1 % 1 x loo-1 x lo1 %l x loo 1 x 107-1 x 1011 1 x 105-1 x 107 1 x 103-1 x 106 1 100-1 103 1 x 104 1 x 103 M 1 x 100-1 x 103 1 x lo6-1 x lo8 z 1 x loo-1 x lo2 1 x lo6 Theoretical cL (fraction)$§l 1 x 10-lo-1 1 x 10-10-1 x 10-4 1 x 10-11-1 x 10-5 1 x 10-8-1 x 10-4 1 x 10-7-1 x 10-3 1 x 10-6-1 x 1 x 10-"-1 x 1 x 10-'O-1 x lo+ 1 x 10-lo-1 x 1 x 10-16-1 x 10-14 1 x 10-14-1 x 10-12 1 x 10-13-1 x 10-12 1 10-13 1 10-11-1 x 10-7 1 x 10-13-1 x 10-11 1 x 10-15-1 x 10-12 1 x 10-l2-1 x 10-1° x 1 x 10-13-1 x 10-10 x 1 x 10-13-1 x 10-10 1 x 10-l2 1 x 10-14 1 x 10-'O 1 x 10-13-1 x 10-10 % 1 x 10-21-1 x % 1 x 10-21-1 x _ _ _ _ _ _ _ _ _ _ _ _ ~ ~ ~ ~ ~ ~~ ~ * Yes means atom counting is theoretically possible if NL= 1 x 10' (unity). -f Values of N and cL (rounded to decades) calculated using eqn.( 5 ) with values of E from Tables 3-6 and noises from Table 7 assuming a single atom/ion residence time measurement. $ For those cases where the calculated N for the extrinsic noise limit was < 1 atom NL is reported to be x 1 atom since the intrinsic noise takes over as N approaches 1 atom. Q cL= NL x 60/6 x A where A is the relative atomic mass of the analyte. The detection limit cL is a fraction 1 x is 1 pptr; N is the number of detectable atoms within the observation time 60 is the assumed relative atomic mass of the atom and m is the mass of sample. For the flame and ICP 6 ml of sample are assumed to be nebulized in 1 min giving a sample mass of 0.1 g sC1 or 1 x g in 1 ms.For the ETA the sample mass is assumed to be 0.1 g (100 pl of an aqueous sample). For the ETV a sample mass of g is assumed to be present during one residence time. For the GD and AB the sample mass is assumed to be 1 x g introduced in one residence time. 7 If the value of NL is so large as to make c,> 1 NL was limited to the value making c unity. 11 The range of NL and cL for OES values is dependent mainly on the temperature of the flame or plasma the wavelength of the transition and ** The range of N and cL values for AAS is dependent mainly on the range of A,] (or oA) values for the absorption transition. tt The range of NL and cL values for LIFS depends mainly on the values of the laser repetition frequency the noise source and the residence $$ The range of N and c values for LIIS depends mainly on the laser repetition frequency values and the background current shot noise.the value of A,]. time for an atom. @ The single values for RIS RMS and MS are a result of choosing one noise level and optimized ionization systems.140 JOURNAL OF ANALYTICAL ATOMIC SPECTROMETRY MARCH 1994 VOL. 9 Table 9 Experimental comparison of several atomic methods for solution samples Selectivity Matrix Range of Multi- - LOD*/ Sample RSD Method ng ml-l LODt/pg volumeipl ( O h ) Elemental$ Isotope§ isotopq elements I/ element8 F-AAS 100- 104 104-108 5 x 103-104 1 G N S M N 1 o -2- 102 10 - 4- 100 5-100 3 G N S M N 10 - lo2 103-106 5 x 103-104 G 1 F N M A Y ETA-LIFS 10-~-10 10-~-10-l 1-50 5 E N S M N ICP-LIFS 10- '-lo2 103-107 5 x 103-104 5 G N S M N F-LIIS 10-4-102 100-106 5 103-104 5 G N H M N ICP-MS 10 - 3- 100 10-104 5 x 103-104 G I E Y M A Y ETV-ICP-MS 10-4-100 10- 6-10-2 5-100 3 E Y M A Y GD-MS 1oo-1o3 10-~-10O 10-~-10 5 G Y S A Y ETA-RIMS 10-~-10 10-~-10-~ 10-~-10 5 E Y VS A Y ETA-RIS 10-~-10 1 0 - ~ - i o - ~ 10-~-10 5 E Y S A Y ETA-A AS ICP-OES * LOD =Limit of detection in concentration units.All values rounded to nearest decade. LOD = Limit of detection in absolute amount; 10 ml sample volumes assume for all flame and ICP nebulization cases and 10 pl assumed for all ETA and ETV cases and 1 pl for GD-MS. All values rounded to nearest decade. 1 E = Excellent G =good and F =fair. §N=No and Y=yes. 7 VS =Very small S = small M =moderate and H = high. 11 M =Mostly metals and A = almost all elements.( 11) The only present technique which is truly multi-element is ICP-OES. However ICP-MS and GD-MS are rapid sequen- tial multi-element approaches. The quadrupole (QP) mass spectrometer has been by far the most popular system for ICP-MS. The ion trap (IT) and the ion cyclotron resonance (ICR) mass spectrometers have dynamic range difficulties as well as difficulties with handling high flow rate of ions (1 x 10l6 argon ions s-l; a 1 ppm solution of some analyte will contrib- ute about 1 x lo1' analyte ions s - ~ ) . ~ ~ Therefore it is difficult to envision the routine analytical use of ITMS or ICRMS for the QPMS in ICP-MS. The time-of-flight mass spectrometer (TOFMS) has the capability for rapid multi-element analysis but so far has a duty factor problem; Hieftje27 believes this problem can be overcome. Interfacing the GD to the IT or ICR could lead to a substantial improvement in the through- put i.e.&,' could approach &do. (12) The reader might note that laser ablation (vaporization) was not included with the ICP-MS. The calculated detection limits NL and cL for those cases should be similar to those using the GD-MS since the sample introduction rate of ~1 ngms-l of the laser ablation is similar to the use of GD. Also no calculated detection limits NL and cL are given for GD-OES laser ablation of sample with transfer to ICP-OES or laser ablation-excitation-emission where the laser vaporizes the sample and the vapour enters into a plasma formed by the laser interaction with the surface.In the case of GD-OES FANES;' HA-FANES,40 FAPES,41 and the laser produced plasma excitation sources the essential characteristics namely the electronic excitation temperature of the plasma and the background spectral radiance are not well known negating calculations of &do E,' NL and cL. In the case of laser ablation of the sample with transfer to the ICP-OES the values of E,' NL and cL should be similar to those for ETV-ICP-OES. (13) Several other atomic methods including ETA coherent forward scatter spectrometry flame or ETA degenerate four wave mixing ETA intracavity laser atomic absorption spec- trometry flame concentration modulation atomic absorption spectrometry and flame-ring down laser as well as electrother- mal OES have also not been considered in this paper because the authors felt they were not viable analytical atomic methods.However all of the above flame and furnace-based methods except for electrothermal OES will have similar &do E,' NL and cL values to FAAS and ETAAS depending upon the atom cell. Electrothermal OES will have poorer NL and cL values than ETAAS for all elements except possibly for those with resonance line having wavelengths longer than z 500 nm because of the high black-body emission. (14) A novel technique which could have future analytical utility and which has been omitted from the estimations of E ~ ' E,O NL and cL is atom trap (AT) LIFS.14 No experimental results for analytical AT-LIFS have been published and so the analytical future of this method is still questionable.In addition the technique would appear to suffer considerable losses owing to transport (E:) and spatial (E,) efficiencies much like the AB-MS and the ion trap based methods [see items (7) and (ll)]. (15) The atomic techniques listed in Tables 5 and 6 are all fairly spectrally selective. Even so AB-RIMS certainly has the highest selectivity and ETA-LIFS (TC) most likely has the second highest spectral selectivity although ETV-F-LIIS ICP-MS F-LIIS (TC) ETA-LIFS (SC) and AB-RIS (TC) are a close third. Electrothermal AAS and ICP-OES are certainly less spectrally selective but the vast literature available for virtually all analytes in all sample types minimizes such difficulties. (16) The atomic technique with the greatest freedom from matrix interferences is certainly ICP-OES although ETAAS is a close second and the LIFS techniques using ETA fall in this same category.Glow discharge LIFS and GD-MS have been found25 to exhibit minimal matrix interferences and would fall third in this comparison. Some matrix interferences problems are encountered with ICRMS; but many have been overcome or accounted for with proper choice of experimental conditions Certainly all flamc-based methods including AAS LIIS and LIFS have matriK interferences which are also well docu- mented in the literature. Insufficient data are available on techniques using PLB or AT to know the extent of such matrix interferences. (17) It would seem based upon the above conclusions that ICP-OES and ETAAS will continue to be used for years to come as a back-ulp for ICP-MS or as stand-alone techniques known for their simplicity and reliability and low cost of equipment and operation.Certainly ICP-MS will continue to grow as the premier multi-element method; the next major improvement will probably involve improved operation for the rapid simultaneous analysis of 20 or more elements based upon a true 'multi-element' measurement approach (e.g. ICP-MS with a TOFMS).27 Up till now the outstanding detection limits ( ~ 0 . 1 - 1 fg) in ETV-ICP-MS have been obtained with single ion monitoring. Considerably poorer detection limits result if for example 20 elements are to be measured in the same sample. This degradation in detection limits does not occur in ICP-OES when using the direct reader approach with photomultipliers or a CCD array detector.The laser-based techniques all single element approaches willJOURNAL OF ANALYTICAL ATOMIC SPECTROMETRY MARCH 1994 VOL. 9 141 continue to find use primarily for the determination of selected elements at concentrations below the LOD of ICP-OES ETAAS and ICP-MS or in small sample amounts (sub- microgram samples) where the number of analyte atoms is below NL or cL for the three conventional approaches. Such samples can be found,'* for example in the environmental biological forensic high-purity materials geological and nuclear areas. (18) The introduction of sample as particles molecular vapour atoms or ions into the atom/ion 'cell' and the type of atom/ion 'cell' are and will undoubtedly remain the research areas of greatest activity.Certainly the use of hybrid techniques will continue to flourish because of the possibility of optimizing atom/ion production as well as excitation. This research was supported by DOE-DEOFGOS-88ER- 13881. Professor Winefordner was unable to present the paper at the XXVIII CSI owing to illness. Appendix I = total surface area of emission (cm') = total surface area of fluorescence (cm2) = Einstein coefficient of spontaneous emission from level u to level 1 (s-') = Einstein coefficient of spontaneous emission from level u' to level 1' (the primes indicate uppermost laser populated level resulting from two colour excitation) (s-') = Einstein coefficient of spontaneous emission from level u to level m (a metastable level) (s-l) = photon radiance of a source or a cell back- ground (photons s-' cm-' sr-' nm-') = radiative ionization rate (s-l) for the transition between level u and the ionization continuum induced by the laser radiation tuned at Aui of spectral energy density pui (J cm-3 Hz-l).BUi is the Einstein coefficient for stimulated absorp- tion (J-' cm3 Hz s-') Definition of all Symbols with Units Aem Afl All1 All,*! Aum BPk Buipui(Aui) Bui where a(A) is the cross section (cm2) c is the velocity of light (cm s-I) Aui is the wavelength of transition (cm) and h is the Planck constant (J s) C = speed of light (3 x 10" cm s-') CL = limiting detectable analyte concentration Ei Ell EpSat Ep(Auv) (fraction) = ionization energy of atom (eV) = energy of upper energy level (J) = saturation irradiance (photons s-' crnd2) = photon irradiance of photoionization laser in two-photon ionization (photons s-' cm-') = photon irradiance of line source (photons s-' cm-2) = photon spectral irradiance of continuum source at All (photons s-' cm-') = frequency of laser (source) pulsing (s-') = the statistical weight of level i (dimensionless) E,L EPIC f; gi g = (A) (dimensionless) g' = (gl + ::+ gu,) (dimensionless) H = slit (or aperture) height (cm) h = Planck constant (6.7 x J s) k = Boltzmann constant (1.38 x J K-l) kui ku'i = 'effective' collisional ionization rate coefficients from levels u into the ionization continuum (s- I); 'effective' here means that in collision- dominated systems this coefficient cannot be referred to a single level u since collisions might be effective in redistributing the excited atoms between neighboring levels; u' represents the level reached by the second excitation step (s-') ku,m = total collisional deactivation rate from level u to level m (s-l) K = 4.83 x 10'' T3'2[Z(T)A+/Z(T)A] x 10-5040Ei,/T next = extrinsic ('background') noise in an atomic ne = electron number density ( ~ m - ~ ) = background shot noise for CW (source) system %hPU = background shot noise for pulsed (source) N = noise level (counts) Ne = number of detected (signal) count events (counts) NA = number of analyte atoms measured during one residence time (atoms) NL = limiting detectable number of atomic species in the sample (dimensionless) P = number of laser pulses during atom residence time (minimum value of p = 1 even if temporal efficiency is less than unity) (dimensionless) = probability that signal is at least equal to or greater than as assigned value (dimensionless) = rate of detected events (detected electrons per second) due to phenomenon i (s-') = cross sectional area of source beam (cm2) = transmittance of optical (spectrometer) system = temperature of emission source (K) = slit-width (or aperture) (cm) = mean of one blank measurement (counts) = signal limit of detection (counts) = signal limit of guarantee (counts) = signal counts (counts) = total signal level (counts) method (counts) cw nsh system P Ri Sb TO Ts W Xb Xd xg XS xt Y = sensitivity of method (counts per atom in = electronic partition function for temperature T Z(T) Z(T)A Zum (dimensionless) - sample) (dimensionless) Z(T),+ = Z(T) for atom A and ion A' respectively (dimensionless) = (=Aum + kum) total deactivation rate coefficient for atoms from level u to metastable level m taking into account radiative (Aum) and quench- ing (/cum) transitions (s-') a P 4 Ati = probability of a type I error (false positives) = probability of a type I1 error (false negative) = gate or aperture width of boxcar detector (s) = interaction time.This time is defined as the probing time of the process (fluorescence and ionization) and is given by the laser pulse duration since a pulsed laser is assumed here whose duration does not exceed 1 ps (s) (dimensionless) (dimensionless) = laser pulse width (s) = absorption line full width at half maximum = 'source' line full width at half maximum or = efficiency of atomization/ionization of analyte = efficiency of detection of event X (photons or = same as &d defined at intrinsic noise limit (nm) spectral bandpass of spectrometer (nm) in the cell or source (dimensionless) ions) (dimensionless) (dimensionless)JOURNAL OF ANALYTICAL ATOMIC SPIECTROMETRY MARCH 1994 VOL.9 = efficiency of sample introduction (dimen- sionless) = efficiency of overall measurement process accounting for losses of atoms or ions and for spatial and temporal probing of the atoms (dimensionless) = same as E defined at intrinsic noise limit (dimensionless) = efficiency of nebulization of nebulizer systems (dimensionless) = efficiency of vaporization of analyte containing material (dimensionless) = efficiency of probing analyte in observation region = E,E (dimensionless) = spatial probing efficiency accounting for inefficient spatial excitation (dimensionless) = temporal probing efficiency accounting for inefficient temporal excitation (dimensionless) = transport efficiency of analyte to observation (detection) region (dimensionless) = detection efficiency of the photons reaching the PMT in emission or fluorescence or of the ion produced in ionization experiments (dimensionless) = efficiency of electronics counts per photoelec- tron (dimensionless) = efficiency of thermal excitation of atoms (or ions) in cell gJZ(T)e-EukT (dimensionless) = collection efficiency in emission spectrometry = WHQE T0/47rAe (dimensionless) = collection efficiency in fluorescence spec- trometry = WHQF T0/4nA (dimensionless) = wavelength of l+u or u-rl transition (cm) = true mean = flicker factor for source (dimensionless) = flicker factor for background (dimensionless) = 3.1418 ...= spectral energy density for transition u+i = true standard deviation = absorption cross-section (cm2) = absorption cross-section for the two photon excitation process (cm4 s) = residence time of atom (ion) in observation region (s) = absorption photoelectron count rate (counts due to absorption/s atom) = emission photoelectron count rate (counts due to emission/s atom) = photoelectron count rate due to laser process (counts per atom) = throughput for spectrometric system with line source = WHToVdVel assuming entrance optics matched to spectrometric optics (cm2) = throughput for spectrometric system with con- tinuum source = WHToVdVel assuming entrance optics matched to spectrometric optics (cm2 nm) (J cm-3 Hz-l) = solid angle of collection of emission (sr) = solid angle of collection of fluorescence (sr) Appendix I1 HA-FANES = Hollow anode FANES LIFS = Laser induced (enhanced) fluorescence spec- tronietry (also LEAFS) LIIS = Laser induced (enhanced) ionization spec- trornetry (also LEIS) MS = Mass spectrometry OES RIMS RIS = Resonance ionization spectrometry Cells/sources terminology AB = Atomic beam AT = Atom trap cw = Continuous wave operation CWL = CMT tunable laser ETA = electrothermal atomization ETV = electrothermal vaporizer F = Flame GD = Glow discharge HCL = Hollow cathode lamp ICP = Indiuctively coupled plasma NEB = Nebulizer PL = Pulsed laser PU XeS Laser terminology sc = Single colour excitation TC TP Mass spectrometry terminology IT = Ion trap ICR = Ion cyclotron resonance = Quadrupole = Time-of-flight QP TOF Other terms AFOM = Analytical figure of merit BCS = Background current shot noise BEF = Background emission flicker noise BES = Background emission shot noise CCD = Charge coupled device DN = Detector noise INT = Intrinsic noise LOD = Limit of detection LOG = Limit of guarantee LOQ = Limit of quantitation PMT = Photomultiplier tube RSD = Rclative standard deviation SAD = Single atom detection SBF SBS uv = Ultraviolet = Optical (atmic or ionic) emission spectrometry = Resonance ionization mass spectrometry = Pulsed operation = Xenon continuum source Spectrometer = Two-colour excitation (through real levels) = Two-photon excitation (through virtual level) = Source background flicker noise = Source background shot noise Appendix 111 Requirements for SAD The general requirement for an SAD method is that the method detects ea’ch and every atom with near The SAD methods are only applicable to the laser based methods.An SAD method is therefore a method in which where P is the probability of a false negative and Ed is defined as Definitions of Acronyms AAS = Atomic absorption spectrometry where P ( X 3 x d ) is the probability that the measured signal FANES exceeds or equals the detection limit. At the intrinsic limit x d = l count therefore &d becomes &do and E becomes E ~ ’ FAPES = Furnace atomization plasma emission which gives the probability that an atom will produce a signal count.The probabilities &do and gmo do not change even if the Methods terminology Ed P(xt 3 x d ) (A2) = (Hollow cathode) furnace atomization using non-thermal excitation spectrometry spectrometryJOURNAL OF ANALYTICAL ATOMIC SPECTROMETRY MARCH 1994 VOL. 9 143 extrinsic noise increases. The application of the above general requirement for SAD differs for the cases of destructive detec- tion (e.g. LIIS RIS and RIMS) and non-destructive detection (e.g. LIFS). To achieve SAD with a destructive detection method a single atom can only give rise to one count at most. The detection efficiency is thus the binomial probability of success namely that an atom will be ionized during the laser interaction time Ati.This probability is given by a Poisson process. However the only way the SAD requirement can be reached is for the intrinsic noise limit i.e. xd to be equal to 1. Thus for true SAD using a laser based ionization method the following two conditions must hold P(X > 0) <a (A31 (A4) y = 1 -,-@ti> 1 - ' P where is the number of detected events per atom per unit time and Ati is the laser-atom interaction time. To achieve SAD with a non-destructive detection method the guaranteed detection limit X is used; X is defined as the probability of a variable in a distribution with mean X being less than xd being negligible i.e. less than a desired probability fl (false negative). If both background and signal are described by Poisson probability distributions it is easy to assign values of Xd and X for any value of Xb by using tables of Poisson values.The procedure in determining these signal limits is as follows. From the values of xb X is chosen so that P(Xb>Xd)%M. With this value of Xd a Poisson distribution is found such that P(X 2 X,) % P. The mean of this distribution is X,. If it is assumed that $s and Ati are both constants then if X is found in the Poisson tables the requirement for SAD is X,>,Xg-Xb (A5 From Table 1 it can be seen that when xb= 1 count xd= 6 counts and X,= 16 counts. Thus SAD is possible by LIFS if X,> 15 counts per atom even in the presence of extrinsic (background) noise. Note however that for the intrinsic noise case (zb=o) a value of x',=6.6 counts per atom is needed for SAD by LIFS (of course this number of counts per atom is not possible by destructive methods).References Kaiser H. Z . Anal. Chem. 1965 209 1. Kaiser H. Two Papers on the Limit of Detection of a Complete Analytical Procedure Jafner Publishing New York 1969. Kaiser H. Foundations for the Critical Discussion of Analytical Methods in Methodicum Chimica ed. Korte F. Academic Press New York vol. lA 1974. Stevenson C. L. and Winefordner J. D. Appl. Spectrosc. 1991 45 1217. Long G. L. and Winefordner J. D. Anal. Chem. 1983,55 712A. Epstein M. S. and Winefordner J. D. Prog. Anal. At. Spectrosc. 1984 7 67. Fujiwara K. McHard J. A. Foulk S. J. Bayer S. and Winefordner J. D. Can. J. Spectrosc. 1980 25 18. Alkemade C. Th. J. Appl. Spectrosc. 1981 35 1. Alkemade C. Th. J. in Analytical Applications of Lasers ed.Piepmeier E. H. Wiley-Interscience New York 1986 ch. 4. 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 Stevenson C. L. and Winefordner J. D. Chemtracts-Anal. Phys. lnorg. Chem. 1990 2 217. Stevenson C. L. and Winefordner J. D. Spectrochim. Acta Part B 1993 48 757. Stevenson C. L. and Winefordner J. D. Appl. Spectrosc. 1992 46 407. Omenetto N. and Winefordner J. D. Prog. Anal. At. Spectrosc. 1980 3 181. Falk H. Prog. Anal. At. Spectrosc. 1980 3 181. Winefordner J. D. Smith B. W. and Omenetto N. Spectrochim. Acta Part B 1989 44 1397. Omenetto N. Smith B. W. and Winefordner J. D. Spectrochim. Acta Part B 1988 43 1111. Falk H. J. Anal. At. Spectrom. 1992 7 255. Sjostrom S. Spectrochim. Acta Rev. 1990 13 407. Travis J. C. Turk G. C. De Voe J. K. Schenk P. K. and Van Dijk C. A. Prog. Anal. At. Spectrosc 1984 7 199. Travis J. C. and De Voe J. R. in Lasers in Chemical Analysis ed. Hieftje G. M. Travis J. C. and Lytle F. E. The Humana Press Clifton NJ 1981. Hurst G. S. and Payne M. G. Principles and Applications of Resonance Ionization Spectroscopy Adam Hilger Bristol 1988. Young J. P. Hurst G. S. Kramer S. D. and Payne M. G. Anal. Chem. 1979 51 1050A. Hurst G. S. Payne M. G. Kramer S. D. and Young J. P. Rev. Med. Phys. 1979 51 767. Letokhov V. S. Laser Photoionization Spectroscopy Academic Press New York 1987. Date A. R. and Gray A. L. Applications of Inductively Coupled Plasma Mass Spectrometry Blackie Glasgow 1988. Jarvis K. E. Gray A. L. and Houk R. S. Handbook of Inductively Coupled Plasma Mass Spectrometry Blackie Glasgow 1992. Hieftje G. M. J. Anal. At. Spectrom. 1992 7 783. Harrison W. W. and Bentz B. L. Prog. Anal. Spectrosc. 1988 11 53. Gilbert P. T. in Analytical Flame Spectroscopy ed. Mavrodineau R. MacMillan New York 1970. Boumans P. W. J. M. Fresenius' 2. Anal. Chem. 1986 324 397. Priigger H. Spectrochim. Acta Part B 1969 24 197. Pivonsky M. and Nagel M. R. Tables of Blackbody Radiation Functions MacMillan New York 1961. Alkemade C . Th. J. Hollander Tj. Snelleman W. and Zeegers P. J. Th. Metal Vapours in Flames Pergamon Press Oxford 1982. Alkemade C. Th. J. Snelleman W. Boutilier G. P. Pollard B. D. Chester T. L. Omenetto N. and Winefordner J. D. Spectrochim. Acta Part B 1980 35 261. Boutilier G. D. Pollard B. D. Chester T. L. Omenetto N. and Winefordner J. D. Spectrochim. Acta Part B 1978 33 401. Voigtman E. and Winefordner J. D. Prog. Anal. Spectrosc. 1986 9 7. Voigtman E. Appl. Spectrosc. 1991 45 237. Sjostrom S. and Mauchien P. Spectrochim. Acta Rev. 1993 13 153. Falk H. Hoffman E. and Ludke Ch. Prog. Anal. Spectrosc. 1988 11 417. Ballou N. E. Styris D. L. and Harnly J. M. J. Anal. At. Spectrom. 1988 3 1141. Liang D. C. and Blades M. W. Spectrochim. Acta Part B 1989 44 1059. Paper 3/04012G Received July 9 1993 Accepted August 24 1993

ISSN:0267-9477    

DOI:10.1039/JA9940900131

出版商:RSC    

年代:1994

数据来源: RSC

摘要:

JOURNAL OF ANALYTICAL ATOMIC SPECTROMETRY MARCH 1994 VOL. 9 145 Plasma Mass Spectrometry Consider the Source" Invited Lecture Brenda S. Sheppard US Food and Drug Administration National Forensic Chemistry Center 7 147 Central Pkwy Cincinnati OH 45202 USA Joseph A. Caruso Department of Chemistry University of Cincinnati Mail Location 7 72 Cincinnati OH 45221 USA The use of mixed-gas helium and nitrogen plasmas as alternative sources for plasma mass spectrometry is discussed in this paper. These plasmas are used to alleviate some of the problems inherent in argon inductively coupled plasma mass spectrometry (ICP-MS). Spectroscopic and non-spectroscopic interferences as well as sensitivities for high ionization potential elements are addressed. Reduced-pressure plasmas are used for the determination of P and S.In addition applications of chromatographic techniques such as gas supercritical-fluid and high-performance liquid chromatography with alternative plasma sources are included. Some problems found in the application of argon ICP-MS can be reduced or eliminated with alternative sources and sub-ng to sub-pg levels of detection for halogens can be achieved. Keywords Inductively coupled plasma; microwave-induced plasma; mass spectrometry The need for ultra-trace level elemental analysis has been the stimulus for the development and improvement of analytical techniques. Plasma source mass spectrometry is one technique that is currently of interest for multi-element analysis at the part per billion and part per trillion levels. Methods used to increase the applicability and sensitivity of this technique are of interest. The inductively coupled plasma was developed as a source for use in atomic emission but has also been widely used as an ion source for elemental mass spectrometry.Plasma mass spectrometry was developed in the 1970s and early 1980s'-'1 and has been reviewed el~ewhere.'~-'~ In the last decade or so inductively coupled plasma mass spectrometry (ICP-MS) has become a powerful elemental analysis tool and is in some cases more valuable than atomic emission spectrometry (AES). Some advantages of ICP-MS over ICP-AES are the excellent sensi- tivity selectivity and ability for isotope dilution. The argon ICP has proven to be the most useful and widely applied source for plasma MS. It has been utilized extensively with solid liquid and gaseous sample introduction techniques. However there are ionization difficulties and problems with spectral and matrix interferences that warrant the investigation of alternative plasma source^.'^-^^ Several alternative plasma sources have been reported including helium nitrogen mixed gas reduced pressure ICPs and microwave-induced plasmas (MIP). Certain applications require investigations of glow discharge sources although these will not be discussed here.In this paper the use of mixed-gas plasma^,^^-^' helium ~ l a s m a s ~ l - ~ ~ and nitrogen plasmas5s57 as alternative sources is discussed. Studies have been carried out with both ICPs and MIPS. Reduced pressure helium plasmas formed either in a m i c r ~ w a v e ~ ~ - ~ ~ or r.f.field63*64 offer attractive possibilities with gaseous sample introduction. Also interesting are the potential applications of these plasmas for the determination of non-metals that are difficult to carry out with the argon ICP source; determination of phosphorus sulfur and halogen containing compounds are examples of these. Gas and supercritical fluid chromatographic sample introduction pro- vides an excellent opportunity to introduce gaseous samples for these compounds with sub-nanogram to sub-picogram levels of detection. * Presented in part at the XXVIII Colloquium Spectroscopicurn Internationale (CSI) York UK June 29-July 4 1993. Inductively Coupled Plasmas Argon ICPs are the most widely used ion sources for plasma MS because of their stability excellent detection limits and wide linear ranges.Trace level detection capabilities make the Ar ICP ideal for analysis of environmental and biological samples where the elements of interest are sometimes present at parts per billion to parts per trillion levels. The analysis of some of these samples are complicated by high concomitant element concentrations which could lead to polyatomic and other matrix interferences. The argon ICP is not an efficient ionization source for difficult to ionize halogens and other elements with higher ionization potentials. Additionally the formation of analyte oxides diminishes the sensitivity for some elements. Alternatives to the 100% argon ICP have been investigated to alleviate some of these problems and have included mixed gas as well as helium ICPs.Spectroscopic interferences on analytes of interest can be a problem in ICP-MS work. For example the use of hydrochloric acid for sample dissolution or the presence of a high chloride matrix will interfere with the determination of 51V+ or 75As' because of interferences by 35Cl'60 + and 40Ar35C1 + respect- ively. Mixed-gas plasmas have been investigated for reduction of these and other matrix effects. The addition of hydrogen,33 air,34 xenon4' and helium2* to various portions of the plasma gas have been studied. These gases primarily have been added to the aerosol carrier and outer gas flows to reduce the formation of polyatomic interferences and the formation of analyte oxides. Improvements in detection limits as well as reduction of oxides and interfering species have resulted.Interfering polyatomic ions such as N2+ HN2+ NO' ArH' ClO' Arc' ClOH+ and ArO' were reduced significantly with the addition of xenon to the carrier gas. However the use of xenon can be cost prohibitive. The addition of nitrogen to the carrier and outer gas flows has been successfully used to reduce the ArCl+ and C10+ interferences on arsenic and vanadium as well as to reduce the formation of MO' and ArO' species. The main attribute of the mixed gas plasma is its ability to decompose refractory elements and reduce the formation of some plasma gas polyatomic species. However the use of some gases such as nitrogen increase the amount of mass spectral interferences at other masses and therefore must be used only in appropriate cases.Alternative gas plasmas are not only used to reduce oxides and interfering oxide species but they have also been used to146 JOURNAL OF ANALYTICAL ATOMIC SPE;CTROMETRY MARCH 1994 VOL. 9 improve the sensitivity of high ionization potential elements. For example a helium-argon ICP has been used to improve sensitivity for high ionization potential (IP) element^.^^-^' The partial replacement of argon with helium produces a plasma with better analyte ionization capabilities for some elements than a 100% argon ICP. Such a plasma is capable of ionizing elements with high first IPS such as the halogens in halide salts. Detections limits were improved for the higher IP elements such as arsenic and bromine and for some metals. For example detection limits with argon ICP-MS have been reported as 0.40 1.7 and 0.02 ng ml-' for arsenic bromine and iodine respectively.These detection limits are improved using the helium-argon ICP (20% helium) to 0.006 0.07 and 0.006 ng ml- for arsenic bromine and iodine re~pectively.~~ The helium-argon ICP can be used for the mass spectrometric detection of high IP elements without degrading sensitivity for other elements of interest. This source can be used with only a few modifications to existing commercial instrumentation. Helium ICPs have also been studied as possible sources for plasma MS.4143 Several modifications to existing instrumen- tation were necessary and include modifying the load coil plasma impedance network and torch. The helium ICP pro- duced a mass spectrum above m/z 40 that was free from background interferences. Increased sensitivity for gaseous samples containing bromine chlorine sulfur and fluorine was also found in comparison with argon ICP-MS.In further work this group has reported on the introduction of aqueous samples to helium ICP-MS. Significant improvements in sensi- tivity in comparison with argon ICP-MS have yet to be achieved. Helium ICPs show promise as ion sources for elements with high IPS. This plasma source warrants further investigation. Helium Microwave-induced Plasmas The determination of halogenated compounds at increasingly low levels is of great environmental importance. The halogens have higher ionization energies in comparison with most of the elements of interest The ionization and excitation capabili- ties of the argon ICP are not sufficiently great to achieve the sensitivity needed.In addition major argon background inter- ferences exist that hinder the determination of some elements. A helium plasma is better suited for the determination of halogens. The helium MIP is an attractive alternative to an ICP because it is compact and relatively inexpensive. The formation of helium plasmas can be accomplished with relative ease with MIPs in comparison with ICPs. Several modifications to the ICP mass spectrometer are required in order to use an MIP. The ICP torch box has to be removed and replaced with an MIP cavity a microwave generator is used instead of the r.f. generator and additional pumping capacity is added to the expansion stage of the spectrometer.The sampling orifice is also usually less than 1 mm. A fundamental study of the sampling process in a helium MIP mass spectrometer has been undertaken by Chambers et ~ 1 . ~ ' The ion transportation process was studied in order to determine what modifications to the ICP-MS interface were needed to operate an MIP-MS instrument effectively. These workers concluded that with this type of plasma it is important to minimize air entrainment and maintain a high ion flux through the interface. Smaller sampling cone and skimmer cone orifice diameters can be used to reduce air entrainment; however this will limit the ion flux. They also discussed the proper placement of the skimmer cone in relation to the Mach disk. Helium MIPs have been used as ion sources for detection of gas-phase species species in aqueous solutions as well as coupled to gas chromatography (GC) high-performance liquid chromatography (HPLC) supercritical-fluid chromatography (SFC) and electrothermal vaporization (ETV). Most of the work with the helium MIP has been conducted in the area of detection of gas-phase species.Many of the background inter- ferences associated with an argon plasma can be effectively reduced with helium plasmas as has been shown by Brown et and Satzger ct who were able to detect halogens in the low pg s-l range by modifying the MIP-MS interface. A nitrogen sheath gas was used in conjunction with a quartz bonnet to reduce the amount of entrained air in the plasma. Sensitivity for bromine chlorine and iodine introduced as gaseous mixtures of CH,Br CH,Cl and CH,I was improved by reducing the amount of air entrained in the plasma.47 Detection limits for 79Brf 35Cl+ and 1271+ were 1.2 21 and 1.8 pg s-l respectively.Additionally Douglas and co- w o r k e r ~ ~ ~ ~ found that the helium MIP source is free from ioni- zation interferences for sodium concentrations up to 100 ppm and found good correlation with the certified values of several Standard Reference Materials (Orchard Leaves Water Low- Alloy Steel) and reported on the determination of lead in blood by isotope dilution. Reduction in background interferences and improved sensi- tivity for halogens make the helium MIP ideal for use with GC. Microwave-ind uced plasmas can be easily interfaced with gas chromatographs. A gas chromatograph was coupled to an MIP-MS system for the detection of chlorinated brominated and iodinated compounds.48 A tangential flow torch was used in this study and the sampling orifice was reduced to 0.4mm with an aluminium sampler.Absolute detection limits of approximately 1 pg for bromine and iodine were found while the absolute detection limit for chlorine was 10-20 pg. Higher detection limits were found for chlorine because of raised backgrounds at m/z values of 35 and 37. Organotin compclunds are also of considerable environmen- tal importance. The most common separation method for the speciation of organotin compounds is GC. A gas chromato- graph was interfaced to a helium MIP-MS system for the detection of six orgatnotin species.49 The compounds of interest were tetravinyltin i.etraethyltin tetrabutyltin triethyltin bro- mide tripropyltin chloride and tributyltin chloride.Initially a standard tangential flow MIP torch was used in this study. It was found that tin forms refractory oxides in an MIP which deposit on the walls of the torch thereby decreasing sensitivity. Additionally the analyte can diffuse throughout the plasma prior to reaching the sampling orifice. A tantalum tube was used to introduce the effluent from the gas chromatograph directly into the centre of the plasma and reduce the distance between the plasma and the sampling orifice. The sensitivity for tin compounds was increased by a factor of ten with use of the tantalum injector. Detection limits at sub-picogram levels were achieved and linear dynamic ranges of three orders of magnitude were obtained.One disadvantage associated with helium plasmas has been their inability to tolerate aerosol introduction. The solvent load imparted by direct solution nebulization can de-stabilize the plasma. Most helium MIPs are operated at powers of less than 400 W whereas ICPs are operated at approximately 1 kW. A helium MIP has been successfully interfaced with a mass spectrometer for the analysis of aqueous aerosols. A tangential flow torch with a glass aerosol injector was used and is described in detail in the original m a n ~ s c r i p t . ~ ~ The aerosol injector prclduces an analyte rich region similar to that in an ICP. The sample introduction system consisted of a MAK nebulizer and a cooled double-pass chamber.Detection limits for chloride bromide and iodide were 39 0.18 and 0.04 ppb respectivdy. These elements were detected as positive ions at m/z 35 37 79 and 127. Detection of metals with this experimental set-up was also investigated. Detection limits were comparable or slightly improved when compared with an argon ICP. A reversed-phase HPLC system was directly coupled to a helium MIP-MS instrument for the element-selective detection of halogenated organic compounds. The tangential flow torchJOURNAL OF ANALYTICAL ATOMIC SPECTROMETRY MARCH 1994 VOL. 9 147 with the glass aerosol injector used by Creed et aLS1 was used in this experiment. However a helium concentric nebulizer was used instead of a MAK nebulizer. The plasma was able to tolerate up to 70% methanol.Absolute detection limits were 50 pg of bromine for brominated compounds 1 pg of iodine for iodinated compounds and l o n g of chlorine for chlorinated compounds. Linear dynamic ranges were 3-4 orders of magnitude for bromine and iodine. This work shows that helium MIP-MS is a viable detector for some applications of HPLC. Helium MIP-MS can be successfully used as a mass selective detector for both HPLC and GC separations. However there are some compounds that would be better separated by SFC. Non-volatile thermally labile compounds of fairly high relative molecular mass can be separated. Also unlike HPLC SFC provides gaseous sample introduction to the plasma with virtually 100% transport efficiency. The coupling of SFC with helium MIP-MS was investigated for 1-chloronaphthalene and l-bromo-2-methylnaphthalene.s3 In this study a demountable tangential flow torch was used.The SFC and MIP systems were interfaced by connecting a transfer line from the SFC to the torch. The transfer line was run through a stainless-steel tube and wrapped in heating tape to maintain a constant temperature. Detection limits at the low to sub-picogram levels were achieved for chlorine and bromine thus indicating the potential of this technique. Electrothermal vaporization has also been used as a sample introduction device for helium MIP-MS.s4 The electrothermal vaporizer has several advantages over solution nebulization including greater sample transport efficiency and the possibility of separating analytes from interfering matrix species by use of temperature programming.Sample sizes are generally in the pl range not in them1 range. In this work a tantalum-tipped electrothermal vaporizer was used to ensure that no analyte condensed during transport to the plasma and also to facilitate the formation of an annular helium plasma. Detection limits for silver cadmium lead and bromine were 0.03 0.09 0.75 and 1.5 pg respectively. Detection limits were blank limited by contamination from furnace components. Microwave-induced plasmas are under-utilized sources for plasma mass spectrometric detection. The MIP is an excellent source for gas-phase sample introduction providing figures of merit comparable to or better than ICP. Liquid samples can be analysed with this source at sufficient powers however this source is more susceptible to non-spectroscopic matrix inter- ferences than is the argon ICP.Reduced Pressure Microwave-induced Plasmas Background interferences from entrained gases can be mini- mized with the use of a low-pressure plasma. The plasma is isolated from contamination owing to atmospheric entrainment and uses lower gas flow rates. Gas flow rates for reduced-pressure use are 250 ml min-l compared with 1 1 min- ’ for atmospheric pressure MIPs. These features mini- mize the ionization of plasma gas impurities and/or the forma- tion of polyatomic ions that interfere with the determination of certain low-mass elements. The low-pressure plasma could be the solution to certain low-mass interferences such as P (m/z 31) S (m/z 32) and C1 (m/z 35 and 37). Interferences associated with these m/z values could include l4NI6OH+ l6O 2 9 + 160180H+ I6O2H3+ and 36ArH+.The signal-to- background ratios associated with m/z values 35 37 31 and 56 were reduced using the low-pressure MIP in comparison with the atmospheric pressure MIP.” This background reduction should now allow for more accurate determinations of chlorine phosphorus and sulfur. The low-pressure MIP-MS system has been used as a detector for GC for the detection of phosphorus and sulfur in pesticide^.^^ Malathion and diazinon were chosen as two widely available pesticides containing these elements. It was found that phosphorus reacted with the hot quartz plasma discharge tube to form phosphorus oxides on the walls of the torch and thus resulted in decreased sensitivity. The torch was modified by the addition of an air cooling jacket.A 100-fold improvement in the detection limit for phosphorus was realized with the air-cooled torch. However oxide formation was still a problem even with this improvement. Phosphorus in triethyl phosphite was detected in the 1-90ng range depending on the torch cooling. Detection of sulfur at m/z 32 was still not possible owing to a large background signal presumably from 02+. Nitrogen has a lower IP than helium and thus produces a less energetic plasma. A significant reduction in the formation of 0,’ was noted with the nitrogen plasma and made the detection of sulfur possible. Less reaction of phosphorus with the torch walls was also observed. Detection limits for phosphorus and sulfur in diazinon were 0.79 and 0.51 ng respectively .Further modifications to the low-pressure helium MIP torch were made in order to improve detection of phosphorus and some of the halogens. A water-cooled torch was developed to reduce the phosphorus interaction with the torch walls.62 A small percentage of hydrogen gas was added to the plasma to act as a reagent gas to scavenge the phosphorus before it reacted with the hot quartz walls and to reduce formation of other polyatomic species. A tantalum injection tube was used to transfer the GC effluent directly into the plasma. A seven component pesticide mixture with chlorine phosphorus sulfur and bromine containing compounds was analysed. Sub- nanogram detection limits were achieved for all elements. Several groups have investigated the use of low-pressure MIPs to obtain fragmentation of organic and organometallic corn pound^.^^^' Fragmentation in combination with total compound decomposition would allow both structural and quantitative elemental analysis with the same instrumental set-up.Poussel et aL6’ used a low-pressure surfatron MIP interfaced to a mass spectrometer and obtained soft ionization and fragmentation comparable to the conventional electron impact source for a variety of compounds including propanoic acid chloroform limonene and dodecane. The sample was injected into the expansion part of the plasma otherwise total decomposition occurred. Heppner” combined GC with low- pressure MIP-MS and obtained fragmentation of organic compounds to a greater degree. A preliminary study was conducted by Olson et using a low-pressure MIP for fragmentation of organic compounds where the sample was introduced into the expansion stage of the mass spectrometer and thus into the tail flame of the plasma.The sampling cone was modified to accommodate this method of sample introduc- tion by adding a 1 mm channel to the side of the sampler cone. In this configuration the sample was introduced midway between the sampler and skimmer orifices. This paper describes the interface in full detail.61 Low powers and flow rates produced parent ion peaks and major fragments similar to those arrived at using conventional electron impact sources for compounds such as hexane toluene o-xylene 1-chlorohexane 1-chloroheptane chlorobenzene and bromo benzene. Nitrogen Microwave-induced Plasmas Moderate-power nitrogen MIPs have been investigated as alternative sources for plasma MS.55-57 Shen et a1.55756 investi- gated the use of nitrogen MIPs for plasma mass spectrometric detection of potassium calcium chromium arsenic and sel- enium.The background mass spectrum of the argon ICP contains many argon-containing polyatomic ions (36ArH+ 40Ar2+ etc.) which interfere with the determination of potass- ium calcium chromium arsenic selenium and iron. The background mass associated with a nitrogen MIP is much less 3 8 ~ ~ ~ + 4 0 ~ ~ + 4 0 ~ ~ ~ + 4 0 ~ ~ 1 2 ~ + 4 0 ~ ~ 1 4 ~ + 4 0 ~ ~ 1 6 0 + Y 9 7148 JOURNAL OF ANALYTICAL ATOMIC SPECTROMETRY MARCH 1994 VOL. 9 complicated in comparison with that of an argon ICP.ss,s7 Background species interfering with potassium calcium and selenium are greatly reduced or eliminated.In addition ArCl' and Arc + polyatomic species are not present in significant amounts to hinder the detection of chromium arsenic and selenium.ss However an interfering polyatomic species was present at m/z 56 probably N4+ which complicated the determination of iron at this isotope. The reduction or elimin- ation of argon-containing polyatomic species allows for the determination of calcium potassium chromium arsenic and selenium at their major isotopes. Significant improvement over the argon ICP for some detection limits was achieved using the nitrogen MIP. Detection limits for 39K and 40Ca were 0.48 and 0.22 ppb with the nitrogen MIP and 1000 ppb for 39K and 5 ppb for 44Ca with the argon ICP.s6 Arsenic and selenium were free from ArCl' interferences with the nitrogen plasma.The nitrogen MIP-MS system allowed determination of iso- tope ratios for 10 ppm of calcium (40Ca:44Ca) and 100 ppb of potassium (39K:41K) chromium (s2Cr:s3Cr) and selenium (soSe:78Se) with less than 5% error.s6 Nitrogen MIPS have been shown to be an effective source for mass spectrometric detection. High-purity nitrogen can be obtained almost any- where at reasonable cost. Low-pressure Inductively Coupled Plasma Atmospheric pressure ICPs have numerous polyatomic inter- ference species associated with them. The entrainment of air coupled with the plasma gas causes the formation of these species. Low-pressure ICPs can be generated to reduce the impact of these interferents on analytical determinations. The low-pressure ICP is interfaced to the mass spectrometer in a similar manner as the low-pressure MIP.A modification to the sampling cone has been made incorporating an ultra-Torr fitting.63 The torch was connected to the fitting. The pumping capacity of the expansion stage was also increased. No modifi- cations were made to the matching network. Two types of low-pressure torch were investigated. The first was a regular Fassel style torch with a water-cooled jacket.63 However it was found that the jacket was not necessary and the torch was re-designed. The second torch was a quartz discharge tube similar to an MIP torch connected to gas lines with ultra- Torr fittings.64 Low-pressure ICPs have been generated using argon air carbon dioxide nitrogen and helium.The first torch was coupled to a gas chromatograph for the detection of 1 -bromon~nane.~~ The second low-pressure torch was also interfaced with a gas chromatograph for the detection of halides in organic compounds.64 Bromobenzene benzylbro- mide chlorobenzene and chloroheptane were investigated. Detection limits were in the 3-8 pg range for all compounds. These detection limits are comparable to those found with helium MIP.48 The advantage was the relative ease of setting up the low-pressure ICP system in comparison with an MIP. Further studies to improve the matching network are necessary to reduce the reflected power. This should assist in improving the analytical performance. Conclusions Inductively coupled plasma mass spectrometry has become an accepted method for trace metal analysis.Alternative plasma sources can be used successfully to improve determinations of various elements such as the halides and to eliminate interfering polyatomic species. The use of an argon-nitrogen ICP is an effective means of controlling some interferences and is not cost prohibitive as are other methods. This plasma has been successfully used by a number of workers to improve the detection of elements such as arsenic. In specific cases mixed- gas plasmas can be easily used with present instrumentation and minimal cost to improve detectibility of elements of interest in particular high ionization potential elements. Modest modifications to existing instrumentation can be made in order to use low-pressure plasmas when air entrainment creates interfering po1;yatomic species.The MIP shows high potential for use as an alternative source for analysis of gas-phase samples. At present this technique is under-utilized because commercial instrumen- tation is not readily available. In addition the nitrogen MIP shows great promise as an alternative source since it provides comparable detection levels to arsenic ICP-MS but with a less expensive more readily available and highly purified plasma gas. The development of commercial instrumentation should take the MIP from use as a research tool to everyday use in service laboratories. There is still much work to be done in the area of alternative plasma sources to improve trace element determinations. 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 References Houk R.S. Fassel. V. A. Flesch G. D. Svec H. J. Gray A. L. and Taylor C. E. Anal. Chem. 1980 52 2283. Gray A. L. Proc. Xoc. Anal. Chem. 1974 11 182. Gray A. L. Analyst 1975 100 289. Gray A. L. Anal. Chem. 1975 47 600. Houk R. S. Svec H. J. and Fassel V. A. Appl. Spectrosc. 1981 37 380. Houk R. S. Montaser A. and Fassel V. A. Appl. Spectrosc. 1983 37 425. Douglas D. F. Quan E. S. K. and Smith R. G. Spectrochim. Acta Part B 1983 38 39. Date A. R. and Gray A. L. Analyst 1981 106 1255. Date A. R. and Gray A. L. Analyst 1983 108 159. Date A. R. and Gray A. L. Spectrochim. Acta Part B 1983 38 29. Gray A. L. and Date A. R. Analyst 1983 108 1033. Douglas D. J. and Houk R.S. Prog. Anal. At. Spectrosc. 1985 8 1. Houk R. S. and Thompson J. J. Mass Spectrom. Reu. 1988 7 425. Hieftje G. M. and Vickers G. H. Anal. Chim. Acta 1989 216 1. Beauchemin D. McLaren J. W. and Berman S. S. Spectrochim. Acta Part B 1987 42 467. Gregoire D. C. Spectrochim. Acta Part B 1987 42 895. Tan S. H. and Horlick G. J. Anal. At. Spectrom. 1987 2 745. Gillson G. R. Douglas D. J. Fulford J. E. Halligan K. W. and Tanner S. D. Anal. Chem. 1988,60 1472. Vandecasteele C. Nagels M. Vanhoe H. and Dams R. Anal. Chim. Acta 1988 211 91. Hutton R. C. and Eaton A. N. J. Anal. At. Spectrom. 1988,3,547. Brotherton T. J. Shen W. L. and Caruso J. A. J. Anal. At. Spectrom. 1989 4 39. Thompson J. J. and Houk R. S. Appl. Spectrosc. 1987 41 801. Crain J. S. Hod; R. S. and Smith F. G.Spectrochim. Acta Part B 1988 43 1355. Olivares J. A. and Houk R. S. Anal. Chem. 1986 58 20. Gregoire D. C. Appl. Spectrosc. 1987 41 897. Satzger R. D. Anal. Chem. 1988 60 2500. Sheppard B. S. Shen W. L. Davidson T. M. and Caruso J. A. J. Anal. At. Spectrom. 1990 5 697. Sheppard B. S. Shen W. L. and Caruso J. A. J. Am. SOC. Mass Spectrom. 1991 2 355. Sheppard B. S. Caruso J. A. Heitkernper D. T. and Wolnik K. A. Analyst 19'32 117 971. Wang J. Evans 13. H. and Caruso J. A. J. Anal. At. Spectrom. 1992 7 929. Craig J. M. and Beauchemin D. J. Anal. At. Spectrom. 1992 7 937. Hill S. J. Ford M. J. and Ebdon L. J. Anal. At. Spectrom. 1992 7 719. Louie H. and Soo S. Y.-P. J. Anal. At. Spectrom. 1992 7 557. Lam J. W. H. and Horlick G. Spectrochim. Acta Part B 1990 45 1313.Lam J. W. H. arid Horlick G. Spectrochim. Acta Part B 1990 45 1327. Lam J. W. and McLaren J. W. J. Anal. At. Spectrom. 1990 5 419. 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.JOURNAL OF ANALYTICAL ATOMIC SPECTROMETRY MARCH 1994 VOL. 9 149 39 40 41 42 43 44 45 46 47 48 49 50 51 52 Beauchemin D. and Craig J. M. Spectrochim. Acta Part B 1991 46 603. Smith F. G. Wiederin D. R. and Houk R. S. Anal. Chem. 1991 63 1458. Montaser A. Chan S. K.. and Koppenaal D. W. Anal. Chem. 1987 59 1240. Koppenaal D. W. and Quinton L. F. J . Anal. At. Spectrom. 1988 3 667. Nam S. Masamba W. R. L. and Montaser A. Anal. Chem. 1993 65 2784. Douglas D. J. and French J. B. Anal. Chem. 1981 53 37.Chambers D. M. Carnahan J. W. Jin Q. and Hieftje G. M. Spectrochim. Acta Part B 1991 46 1745. Brown P. G. Davidson T. M. and Caruso J. A. J. Anal. At. Spectrom. 1988 3 763. Satzger R. D. Fricke F. L. Brown P. G. and Caruso J. A Spectrochim. Acta Part B 1987 42 705. Mohamad A. H. Creed J. T. Davidson T. M. and Caruso J. A. Appl. Spectrosc. 1989 43 1127. Suyani H. Creed J. Caruso J. and Satzger R. D. J. Anal. At. Spectrorn. 1989 4 777. Creed J. T. Mohamad A. H. Davidson T. M. Ataman G. and Caruso J. A. J. Anal. At. Spectrom. 1988 3 923. Creed J. T. Davidson T. M. Shen W. L. Brown P. G. and Caruso J. A. Spectrochim. Acta Part B 1989 44 909. Heitkernper D. Creed J. and Caruso J. A. J . Chrom. Sci. 1990 28 175. 53 54 55 56 57 58 59 60 61 62 63 64 Olson L. K. and Caruso J. A. J . Anal. At. Spectrom. 1992,7,993. Evans E. H. Caruso J. A. and Satzger R. D. Appl. Spectrosc. 1991 45 1478. Shen W. L. Davidson T. M. Creed J. T. and Caruso J. A. Appl. Spectrosc. 1990 44 1011. Shen W. L. Davidson T. M. Creed J. T. and Caruso J. A. Appl. Spectrosc. 1990 44 1003. Wilson D. A. Vickers G. H. and Hieftje G. M. Anal. Chem. 1987 59 1664. Story W. C. Olson L. K. Shen W. L. Creed J. T. and Caruso J. A. J. Anal. At. Spectrom. 1990 5 467. Heppner R. A. Anal. Chem. 1983 55 2170. Poussel E. Mermet J. M. Deruaz D. Beaugrand C. Anal. Chem. 1988 60 923. Olson L. K. Story W. C. Shen W. L. and Caruso J. A. J. Anal. At. Spectrom. 1990 5 471. Story W. C. and Caruso J. A. J. Anal. At. Spectrom. 1993,8,571. Evans E. H. and Caruso J. A. J . Anal. At. Spectrom. 1993,8,427. Castillano T. M. Giglio J. J. Evans E. H. and Caruso J. A. presented at FACSS XIX Philadelphia PA USA September 20-25 1992 paper no. 532. Paper 3/04991 D Received August 17 1993 Accepted November 4 1993

ISSN:0267-9477    

DOI:10.1039/JA9940900145

出版商:RSC    

年代:1994

数据来源: RSC

摘要:

JOURNAL OF ANALYTICAL ATOMIC SPECTROMETRY MARCH 1994 VOL. 9 151 Microscopic X=ray Fluorescence Analysis* Invited Lecture K. Janssens L. Vincze J. Rubio and F. Adamst Department of Chemistry University of Antwerp Universiteitsplein I B-2610 Antwerp Belgium G. Bernasconi International Atomic Energy Agency Laboratories A- I I40 Seibersdorf Austria The status of microscopic X-ray fluorescence analysis with tube excitation and synchrotron radiation is reviewed in terms of the lateral resolution minimum detection limits and elemental sensitivity that can be achieved. As illustrations the utilization of two typical state-of-the-art instruments for the analysis of geological material is described; one of the instruments is based on tube excitation the other is installed at a synchrotron X-ray source.The analytical implications of the use of X-ray microprobes installed at a third generation storage ring and in particular at the European Synchrotron Radiation Facility (ESRF) are discussed. Keywords X-ray fluorescence; microscopic X-ray fluorescence; synchrotron radiation and imaging; micro- analysis; trace element mapping For more than two decades X-ray fluorescence analysis (XRF) has been a well-established and mature multi-element tech- nique capable of yielding accurate quantitative information on the elemental composition of a variety of materials in a non- destructive manner.' In the last 10 years two important variants of the bulk technique have come into existence.2 Both variants are based on the confinement of the interaction volume of the primary X-ray beam with the material being analysed.In total reflection XRF (TXRF),3 by irradiating an (optically flat) sample with a parallel X-ray beam below the angle of total reflection the in-depth penetration of the primary X-rays can be confined to a few tens of nanometers below the surface. As a result the intensity of the scatter background in XRF spectra collected in this way is significantly reduced and the (surface) sensitivity of the technique enhanced. In view of the relatively simple sample preparation required TXRF has proven itself to be an extremely useful technique for the analysis of solutions and natural waters with typical detection limits below 20 ng for 40 elements for counting times of 1000 s ~ and for the ultrasensitive surface analysis of semiconductors3 (coarse resolution 2-D mapping of impurity centres and high- resolution depth profiling).The second major variant of the XRF technique which in the last 2-3 years has received a lot of attention in the literature both with respect to the methodological develop- ments and to its applications in diverse fields is micro-XRF (p-XRF). This microanalytical variant of bulk XRF is based on the localized excitation and analysis of a microscopically small area on the surface of a larger sample providing infor- mation on the lateral distribution of major minor and trace elements in the material under study. The recent increase in the popularity of p-XRF can be attributed on the one hand to the availability of commercial instrumentation and on the other to the development of (relatively) simple devices for the focusing of X-rays.In addition the potential of all the forms of XRF mentioned above has in recent years been greatly enhanced by the increasing use of synchrotron rings as sources of highly intense X-radiation. In this paper an overview of the analytical capabilities and limitations of currently existing p-XRF spectrometers using conventional and synchrotron X-ray sources is presented. In the last part of this work the implications the use of third- * Presented at the XXVIII Colloqium Spectroscopicurn Internationale (CSI) York UK June 29-July 4 1993. t To whom correspondence should be addressed. generation synchrotron rings will have on the analytical pos- sibilities of p-XRF are briefly discussed.Experimental The laboratory-scale pXRF used for collecting some of the data presented in this paper consisted of a conventional Siemens diffraction tube with an Mo anode operated at 30 mA and 50 kV. The resulting X-ray cone was transformed into a microbeam by means of a conical glass capillary of 70 ym final inner diameter. Samples were mounted on an XYZB table with a positioning accuracy of 1 pn. The set-up operated in air the Si(Li) detector being positioned at 90" to the incoming beam and at a distance of approximately 2cm from the sample. Scanning of the sample through the beam and data collection were performed by means of a personal computer and software developed by one of the authors (G. B.) using a Canberra SlOO MCA plug-in board and associated driver software.For the micro-synchrotron radiation induced XRF (p-SRXRF) measurements the X-ray microprobe station at the bending magnet beam line X26A of the NSLS (National Synchrotron Light Source Brookhaven National Laboratories NY USA) was employed. In this instrument the white synchrotron light has a maximum flux density of about lo4 photon s-' pm-2 mA-' at 8 keV. After emerging from the storage ring ultra-high vacuum (UHV) the beam is defined by four Ta slits to a size of approximately 8 x 8 pm. Soft X-rays ( < 5 keV) are heavily absorbed in the Be end-window of the beam pipe and the air path between collimator and sample. The sample is positioned at 45" to the incoming beam mounted on an XYZB stage with 0.1 ym accuracy and can be viewed by a horizontally mounted microscope equipped with a colour TV camera.X-ray spectra are detected at 90" to the incoming beam using a well collimated Si( Li) detector. The electron probe X-ray microanalysis (EPXMA) measurements were per- formed on a Jeol JSM 6300 scanning electron microscope (SEM) system equipped with a Princeton Gamma Technology (PGT) X-ray spectrum and image collection system and using a 25 kV 1 nA electron beam. Both electron and photon induced X-ray spectra were evaluated using the AXIL p a ~ k a g e . ~ Discussion p-XRF Using Conventional X-ray Sources The simplest way of producing an X-ray microbeam is to collimate the broad cone of radiation originating from an152 JOURNAL OF ANALYTICAL ATOMIC SPECTROMETRY MARCH 1994 VOL. 9 X-ray tube by means of suitable apertures or crossed-slit systems.This concept was employed in the 1 9 6 0 ~ ~ ~ ' but did not receive a lot of attention at the time the major reason being the low count rates observed for small area samples. Also although the use of photons as an excitation source intrinsically features a number of advantages over charged particle excitation the rapid development of electron microscopy (and of electron induced X-ray micro-analysis) with its much higher lateral resolution overshadowed further developments. In the 1980s p-XRF was 're-discovered' with the first use of energy dispersive (ED) detectors for X-ray imaging applications. As the solid angle of acceptance of Si( Li) detectors can be much larger than that of crystal spectrometers fluorescent signals of much lower intensity could be used to advantage.In 1986 Nichols and Ryong equipped a rotating anode X-ray generator with a modified micro-diffractometer system to generate a microbeam with a nominal diameter in the range 10-1OOpm. The use of the diffractometer allowed manual translation of the sample through the beam and included a microscope for sample viewing. Their objective was to arrive at a relatively inexpensive system built from off-the- shelf components which would (i) provide analysis to a greater depth within the sample than the near surface analysis obtain- able using electron optical instrumen tation (ii) allow for localized analysis of large objects and (iii) be capable of performing analysis under ambient atmospheric or helium conditions rather than in vacuum.In 1987 Boehme" described the use of a p-XRF system for elemental mapping of large area (400 mm2) geological materials and pointed out that p-XRF provided information that enhances and corroborates data obtained by means of optical microscopy SEM electron probe microanalysis (EPMA) and conventional XRF analysis." Nichols et a/.12 investigated the effect of instrumental param- eters such as the position and size of the aperture and the beam profile and divergence on achievable lateral resolution and sensitivity. In 1988 Ryon et ~ 1 . ' ~ summarized the advantages of X-ray imaging as follows. (a) The penetrative character of X-rays and the complex but well understood interaction of X-rays with matter allows for the determination of layer thicknesses the analysis of sub- surface structures and homogeneity testing throughout a material.(b) High-energy photons can penetrate below the surface of opaque materials and cause the emission of characteristic radiation. (c) Unlike electron microscopies which (in most instruments) must be performed in high vacuum X-ray imaging can be done in air and on large samples requiring little or no sample preparation; also non-conducting materials can be analysed without problems. Compared with charged particle micro- scopies X-ray imaging causes low thermal loading allowing e.g. volatile components or very sensitive materials to be analysed. (d) X-ray equipment is simple in comparison with that required for scanning particle beam microscopies. Also in 1988 Wherry et presented a description of 'an automated X-ray microfluorescence materials analysis system' the first commercially available p-XRF system developed by Kevex (Valencia CA USA).This set-up consists of a low- power (50 W) X-ray tube providing a small (<250 x 250 pm) spot fitted with 10 30 and 100 pm diameter apertures placed sz 60 mm from the anode and 3 mm before the sample. Samples (in an evacuable chamber) are mounted on a motorized XYZ stage illuminated from above and below and can be viewed by means of a colour charge coupled device (CCD) video camera. Fluorescent radiation is detected using a 50 mm2 Si(Li) detector at 90" to the incident beam. As a result of the close coupled detection geometry for pure element samples and a 100pm aperture net count rates in the range 2000-7000 counts s- ' could be obtained; lateral resolutions were reported to be of the same order of magnitude as the collimator opening.Cross et all5 also have reported on the multivariate processing of multiple X-ray images for automatic phase discrimination. The use of this instrument in various disciplines has been reported including the analysis of buried layers in multi-metal multi-layer materials used in computer mainframe r n a n ~ f a c t u r e ~ ~ ~ ~ ~ the screening of toxic contami- nants and precious metals in heterogeneous wastes'' and the use of the instrumeni for miscellaneous problem solving in the fuel industry.19 Pella et a1.20.21 have reported on the analysis of coarse particles (diameters between 50 and 200 pm) using a laboratory-built instrument similar to the Kevex device and on the problems associated with the quantification of the results derived from this and other heterogeneous sample types.22 By using a micro-focus X-ray tube the photon flux loss as a result of collimation can be minimized; nevertheless in view of the large distance between anode and aperture significant losses occur and only a small fraction of the total photon flux leaving the X-ray tube arrives at the sample.An alternative approach to obtaining a more intense X-ray microbeam is to employ glass wave guides instead of collimators. As shown in Fig. l(a) through repeated total reflection off the inner walls of glass capillaries photons can be 'transported' from the tube anode to the immediate vicinity of the sample surface.In early work straight glass capillaries were used as fine collimators generating fine beams for mi~ro-diffraction~~ while in the 1970s the X-ray wave guide properties of capillaries were investigated by several g r o ~ p s . ~ ~ - ~ ~ Rindby28 described the use of straight capillaries together with conventional X-ray tubes for generat- ing X-ray beams of about 100pm in size. Carpenter and ~ o - w o r k e r s ~ ~ ~ ~ replaced the Be side-window of an HOMX 160A micro-focus X-ray tube with a capillary assembly bring- ing the end of the capillary as close as 2 mm from the 15 pm spot on the anode (see Fig. 2). In view of the small cross- section of the inner channel in the glass capillary tubing the vacuum inside the tube can be maintained without serious problem.In the Kwex instrument mentioned above the use of collimators smaller than 30 pm requires long scanning times because of the low beam flux. By the maximization of the acceptance angle of the capillary described above microbeams with cross-sections in the range 4-100 pm2 still provided acceptable flux. For these beam sizes minimum detection limits (MDLs) in the range 20-100ppm for Cu W and Mo were reported in a National Institute of Standards and Technology (NIST) Standard R.eference Material (SRM) 610 Glass Trace Elements 610 glass. Using SRMs 1832 and 1833 minimum detectable amounts of 0.01- 10 pg absolute were a~hieved.~' Applications of the microprobe described by the same workers include the identification of inclusions in a carbon structure the examination of cracking in alumina cylinders by ZrO particles and the irivestigation of the P Ca ratio in bone and teeth. In the last application the non-destructive character of p-XRF made reproducible analyses possible while in the 1 -.- 100 pm -v - 20 cm---+ 1 100 pm (b) 100 pm 10 pm Fig. 1 Principle of X-ray propagation in (a) straight and (b) tapered capillaries. Typical capillary dimensions (length entrance and exit diameters) are indicatedJOURNAL OF ANALYTICAL ATOMIC SPECTROMETRY MARCH 1994 VOL. 9 153 Fig. 2 Close coupling of micro-focus anode and straight capillary; adapted from Carpenter and Taylor.31 T tube; H capillary housing; C capillary; E electron beam; and A anode. The system can be aligned by moving the anode up and down and by steering of the electron beam case of electron microprobe analysis changes of the P Ca ratio with time were observed as a result of beam damage.32,33 In order to increase the photon flux further it is also possible to use glass capillaries as X-ray concentrators.Conical or tapered capillaries were developed in the later 1980s by several g r o ~ p s . ~ ~ - ~ ~ As shown in Fig. l(b) the photon beam that enters the wide end of the capillary after a number of reflections off the walls of the conical tube are ‘squeezed‘ to the size of the inner diameter at the narrow end. Typical dimensions for conical capillaries used in p-XRF spectrometers are a length of 10-20cm and final inner diameters of 50-10pm. The advantage of this type of X-ray concentrators is the indepen- dence of the source.Owing to the numerous reflections inside the capillary the source size does not directly influence the cross-section of the generated microbeam; the latter is deter- mined only by the inner diameter of the capillary. The diver- gence of the beam is approximately equal to OJE) the critical angle for total reflection at the capillary walls for photons of energy E (for glass in the 10-20 keV energy range OC takes values from ~5 to 1 mrad). As a result of this low divergence which is far less than that of the original (tube generated) X-ray beam entering the capillary for very small beam diam- eters (< 10 pm) it is necessary that the sample is placed as close as possible (ie. in practice 0.5-1 mm) to the capillary tip. Instability and fluctuations in the position of the source do not affect the position and size of the focal spot.Because the capillary is a total-reflection device photons within a broad energy band are concentrated although the gain factor (k the ratio of the beam flux density entering and leaving the capillary) is energy dependent. As a result of the 1/E2 depen- dence of the gain factor Rindby et aL3’ have shown that the combined effect of the anode self-absorption and the high- energy filtering of the conical capillary can influence the original tube spectrum in such a way that a nearly monochro- matic microbeam is obtained. At present several manufacturers are developing com- mercial p-XRF instruments that are based on the use of conical capillaries. Rindby and co-workers have reported on a number of applications of this type of p-XRF instrument.Larsson et used a 19 pm capillary and a Cr X-ray tube to investigate the distribution of elements such as Ti and S in birch leaves. Engstrom et ~ 1 . ~ ~ reported on the analysis of soft biological tissues such as 8 pm thick pig heart muscle fibres and quoted detection limits in the 60-1 ppm range for the elements in the range Al-Ti. Using both Cr and Mo tubes they also analysed single hair strands along and across their longitudinal direction and could determine the elements in the Z range S-Br at trace levels varying from 1 to 1000ppm. Shakir et studied the leaching of elements from the leaves of the Kardadeh plant (Hibiscus Shadriif) by interaction with hot water (tea brewing) showing a depletion of Mn Ca and K levels.Stocklassa et aL4’ reported on the application of non- destructive trace and microanalysis by means of p-XRF in forensic science involving the analysis of small glass particles paint fragments ball point ink and single hair strands. An interesting application has been described by Rindby et a!?’ concerning the authentication of historic documents such as a Swedish Letter of Possession dated April 1499 which was suspected of having been forged in the 16th century. By scanning selected areas of the parchment old text wiped out or covered by more recent writing could be revealed by means of its trace element content. Voglis et aL4’ described investi- gations concerning the incorporation of heavy metals into bone and recorded the radial distribution of elements such as Al S Cl K Cr and Fe around Haversian Canals.Prior to the use of glass capillaries for pXRF purposes Gurker et ~ 1 . ~ ~ 9 ~ ~ suggested a method for partially eliminating the count rate limitations associated with collimator-based p- XRF s p e ~ t r o r n e t e r s . ~ ~ . ~ ~ Instead of using the very inefficient two-dimensional collimation (where less than 0.1 % of the total output power of the X-ray tube is employed) only one pair of slits was used thereby irradiating a line of points on the sample surface instead of a single point. Instead of collecting an image through point-by-point irradiation the sample is translated and rotated through the narrow band of radiation. The observed fluorescence intensity I(r,O) as a function of the translation co-ordinate Y and the rotation angle 8 is called a ‘sinogram’.By means of tomographic back-projection tech- niques the resulting series of sinograms can be converted into elemental maps. The price to be payed for the more efficient use of the total available photon flux is that artefacts and noise could be introduced in the reconstructed images as a result of the data-collection procedure and/or the back- projection algorithms employed. In general one can state that p-XRF provides new capabili- ties for the analytical chemist in that it fills the gap between bulk X-ray fluorescence and (high-resolution) electron probe X-ray microanalysis. As such a wide range of samples that have been excluded from microscopic examination because of their incompatibility with the vacuum and conductivity requirements of electron microscopy can be analysed.On the other hand in situations involving the determination of trace levels of high-2 elements (e.g. Fe-Mo) with high (1-10 pm) lateral resolution the applicability of the method is still seriously constrained by the limitations in available micro- beam flux and size. As an illustration of the advantages and limitations of p-XRF in comparison with elemental mapping by means of EPXMA Figs. 3 4 and 5 show elemental maps the electron backscattered (EB) image and X-ray spectra obtained using a p-XRF spectrometer equipped with an Mo tube and a 70 pm conical capillary and by means of scanning electron microscopy with an energy-dispersive X-ray assembly (SEM-EDX) from a heterogeneous geological sample.The sample a 60 pm thick section of igneous rock shows biotite crystals embedded in a feldspar-quartz matrix; it was scanned with a step size of 100 pm in both directions in the p-XRF spectrometer. A collection time of 40s per location was employed causing the total acquisition time to become about 14 h for the 30x 30 pixel images. During such a relatively short time only statistically significant data on the major constituents of the geological samples can be obtained. As such only meaningful maps of elements such as K Ca and Fe could be collected usingp-XRF providing more or less the same information as can be found in the EB image shown in Fig. 5. However as can be seen from the spectrum in Fig. 4(b) when XRF spectra are collected during a longer time from a particular spot on the sample information can be obtained on trace elements such as Rb and Zr which are not visible in the corresponding EPXMA data [Fig.4(a)].154 1 X 1 o 5 1 X l O d ' 1 x103 1 x 102 JOURNAL OF ANALYTICAL ATOMIC SPECTROMETRY MARCH 1994 VOL. 9 Fig. 3 Distributions of various elements obtained via laboratory scale p-XRF from a 60 pm thick section of igneous rock. See also Fig. 5 p-XRF Using Synchrotron X-ray Sources In Fig. 4(c) a spectrum collected from the sample shown in Fig. 3 and 5 when irradiated with white synchrotron light in the NSLS X-ray microprobe is shown. In comparing Fig. 4(b) and (c) some of the advantages of using synchrotron radiation (SR) are immediately clear. Synchrotron radiation is generated when light elementary particles (electrons positrons) at rela- tivistic speeds (k close to the speed of light) are forced to change their direction of motion (ie.are accelerated). This type of radiation is very intense (a factor of 1O6-10l2 more than that which can be produced in conventional X-ray tubes) extends from the infrared range into the hard X-ray region and when sampled in the orbital plane of the electron storage ring is linearly polarized. The radiation is also naturally collimated along a direction tangential to the quasi-circular movement of the electrons in the ring. The high intensity and natural collimation make SR ideally suited for the generation of X-ray microbeams; by means of simple collimation very intense X-ray beams with cross-sections in the range 5 x 5 to 10 x 20 pm2 can be generated by means of which white beam excitation p-XRF experiments can be performed with detection limits in theppm andfg As a result of the linear polarization of the radiation the intensity of the scatter background in SRXRF spectra can be significantly reduced allowing even with polychromatic forms of excitation the determination of ppm and sub-ppm levels of trace elements.This is illustrated in Fig. 6 which shows an SRXRF spectrum collected from a 100mgcm-2 sample of NIST SRM 1571 Orchard Leaves when irradiated in the XRF spectrometer of beamline L of Hasylab (DORIS I11 ring Hamburg Germany). Fig. 7 illustrates the ability of the NSLS X-ray microprobe (beamline X26A) t o provide information on the distribution of various trace elements with a lateral resolution of about 5-10 pm.In contrast to the p-XRF data shown in Fig. 3 in Fe 1- ' 1 I I 1- ' I I ' 0 6 pi- n Fe K I1 lo tl 0 5 10 15 20 E nerg ykeV Fig.4 X-ray spectra obtained from the sample shown in Figs. 3 5 and 6 using (a) EPXMA (b) laboratory scale p-XRF and (c) p-SRXRF this case also the distribution of trace constituents such as Rb Sr and Zr can be visualized. In addition to polychromatic forms of radiation the con- struction and use of several monochromatic X-ray microprobes has been r e p ~ r t e c / . ~ ~ . ~ * In contrast to X-ray tubes synchrotron sources have a low emittance (i.e. a low source size and a small divergence) making them well suited to use with de-magnifying optics such as Bragg-reflecting or totally reflecting curved mirrors.49 Conical capillaries have also been employed by various groups to concentrate SR yielding in some cases X-ray beams of sub-pm d i m e n s i o r ~ s .~ ~ . ~ ~ ~ However because the focused beam leaving the capillary is much more divergent than the original synchrotron light entering it the applicability of these devices for high- resolution p-SRXRF may be limited. As well as providing information on the concentration ofJOURNAL OF ANALYTICAL ATOMIC SPECTROMETRY MARCH 1994 VOL. 9 155 H 100 pm Fig. 5 Electron backscattered image of the sample shown in Fig. 3. Scale bar is 100 pm Ca K n Fe J I I I I J 0 5 10 15 20 25 EnergyJkeV Fig. 6 Synchrotron radiation induced XRF spectrum of NIST SRM 1571 Orchard Leaves ( z 100 mg cm-2) collected using a 50 x 50 pm2 white beam at Hasylab beamline L (from ref.70) certain elemental species at some of the p-SRXRF stations chemical information on sample materials can also be obtained.5k56 This is done by scanning the energy of a monochromatized X-ray beam over the absorption edge of an element of interest. The shift in the half-height of the edge reveals information on the oxidation state of the element,54.57 while pre-edge peaks and the X-ray absorption fine-structure (XAFS) above the edge contain information on its chemical surrounding^.^^^^^^^^^^ A number of overview articles (e.g. refs 45 and 61) and chapters in books (refs. 57 62 and 63) have been published which provide a comprehensive overview of the technical aspects and applications of white beam and monochromatic p-SRXRF.In the earth and environmental sciences in particular p-SRXRF has proven itself a unique and very valuable technique for trace analysis of materials which are heterogeneous at the micrometer level. Comparison of the images in Figs. 5 and 7 nevertheless shows that p-SRXRF as it can be performed at currently operating stations is still seriously limited with respect to achievable lateral resolution. The latter parameter is deter- mined by two factors on the one hand the penetrative character Fig.7 X-ray maps of various major minor and trace elements obtained using p-SRXRF (NSLS microprobe) from the central (dark) crystal shown in Fig. 5 of hard X-rays (e.g. Fe Ka X-rays can emerge from a depth of several tens of micrometers out of geological material without appreciable attenuation) and on the other hand by the relatively large photon beam sizes used (5-10 pm).Even at such second- generation sources as the NSLS smaller beam sizes can only be obtained at the expense of a considerable reduction of total beam flux and thus of analytical sensitivity. A number of the limitations of current-day p-SRXRF men- tioned above will with high probability soon be eliminated by the combined use of third-generation synchrotron X-ray sources and of X-ray optics of increasing sophistication. Around the world several large storage rings are being built which will provide synchrotron light of unprecedented intensity and brilliance. The European Synchrotron Radiation Facility (ESRF Grenoble France) is already operating and synchro- tron light will probably be available to external users from the end of 1994 onwards. Other facilities such as the APS (Advanced Photon Source Argonne IL USA) and Spring-8 (Harima Japan) will become operational in 1996 and 1998 respectively. The ALS (Advanced Light Source Berkely CA USA) began operation in October 1993.At all the third- generation rings mentioned above X-ray fluorescence micro- probes are planned or under development.6k67 For the APS and ESRF it is highly likely that undulator beamlines will be dedicated for p-SRXRF and related applications. In contrast to bending magnet sources which provide a continuous energy spectrum the output spectrum of an undulator source features sharp maxima called harmonics. By means of broad-band monochromators one of these harmonics can be isolated and used after appropriate focusing for monochromatic excitation in a p-SRXRF spectrometer.Various optical configurations can be employed for generating monochromatic microbeams from undulator sources. An overview of the various approaches investigated by the scientific groups which are active in this field can be found in ref. 68. As an example Fig. 8 shows a combination of a channel cut monochromator and two ellip- soidal mirrors mounted in a Kirkpatrick-Baez geometry. In Fig. 9(a) the attainable spot size as predicted by means of the ray-tracing code SHADOW is shown for the case of 20 keV156 JOURNAL OF ANALYTICAL ATOMIC SPECTROMETRY MARCH 1994 VOL. 9 w .- ‘ E .- - 1.000 0 al .- 4- w $ 0.100 f .- 0.010 .- 2 M2 - - - S F L MI Fig.8 Ellipsoidal mirrors in Kirkpatrick-Baez geometry focus the synchrotron beam in horizontal and vertical directions. S X-ray source; C crystal monochromator; and M1 and M2 ellipsoidal mirrors 0 10 Horizontal distance/pm 100.000 I (b’\ 1 10.000 1 ,\ layer. When using this type of advanced optics the advantages of obtaining a sub-pm beam must be weighed against the disadvantages caused by the loss of flux which results from these drastic beam. de-magnifications. At the ESRF p-XRF beamline with high probability both combinations (ie. either high flux plus a 5-10 pm spot or a lower flux plus a 1 pm spot) will be realized in two separate set-ups installed at the same beam port. Both are expected to be available for analyt- ical purposes in the near future. When in operation these spectrometers will combine the quantitative reliability and accuracy of XRF with the sensitivity and to some extent the lateral resolution of destructive microanalytical techniques such as secondary ion mass spectrometry (SIMS).Conclusions In this work an overview of recent developments in the field of microscopical X-ray fluorescence analysis is presented. Using state-of-the-art apparatus p-XRF can be performed using either conventional X-ray tubes or synchrotron storage rings as X-ray sources. Iin the former case the development of glass capillaries of different shapes as a simple and inexpensive means of focusing X-rays is opening interesting prospects for laboratory-scale p-XRF as is attested to by published appli- cations of this method in art and archeology industrial research the geosciences and clinical chemistry and biology.The use of SR provided by second-generation sources allows for trace element mapping at theppm level of detection with a lateral resolution better than 10 pm. In the near future by means of radiation originating from undulators of third- generation storage rings the sensitivity and lateral resolution of p-SRXRF are expected to reach the 10-100 ppb and 1 pm level respectively. 0.001 b 15 20 25 30 35 40 Atomic number 1 We express our gratitude to a number of people for the use of their equipment and beam time and for assistance with per- forming some of the measurements K. Jones S. Sutton M. Rivers and S. Bajt at the NSLS F. Lechtenberg S. Garbe G. Gaul and A. Knochel at Hasylab and N.Hasselberger A. Markowics and V. Valcovic of the IAEA Labs Seibersdorf. K. J. is a fellow of the Belgian National Science Fund NFWO (Brussels); this research was sponsored by FKFO (Brussels) Grant No. 2009201N. Fig. 9 (a) Dimensions of the microspot obtained by ray-tracing results using the SHADOW program for the configuration shown in Fig. 8.” (b) Predicted MDLs achievable at an ESRF low-p undulator beamline using the optical configuration shown in Fig. 8 in comparison with MDLs currently achievable at the NSLS X26A SRXRF station operated in broad beam monochromatic mode A NSLS 2 x 0.5 mm2 t = 300 s (FNsLs=4 x lo9 ph s-’ at 180 mA); and B ESRF 3 x 9 pm2 t=300s(F,,,,=5~lO~~phs-’at 100mA) photons originating from a standard ESRF low-/? undulator. Using this optical configuration a monochromatic flux density of the order of 1010photons-1pm-2 per 100mA will be attainable. As shown in Fig.9(b) the expected detection limits for a p-SRXRF spectrometer equipped with this optical con-. figuration will be situated in the 10-100 ppb region. By means of other optical elements such as Bragg-Frensel lenses or capillaries with ellipsoidal shape smaller spot sizes can be obtained but probably at the expense of a certain loss in flux. Using a Bragg-Frensel lens and an ESRF low-/? undulatoir source Kuznetsov et ~ 1 . ~ ~ obtained a lateral resolution of 0.8 pm when scanning a narrow strip of Cr; at the same beam line a tapered glass capillary was used for a micro-diffraction experiment on a Zr alIoy at 8 keV. An approximately 2 pni beam was obtained to sample a 10 pm thick ZrOz corrosion 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 References Bertin E.P. Principles and Practice of X-ray Spectrometric Analysis 2nd edn. Plenum New York 1975. Janssens K. EL and Adams F. C. J. Anal. At. Spectrorn. 1989 4 123. Spectrochim. Acta Part B 1991 46 1313 (Special Issue). Schwenke H. and Knoth J. Handbook of X-ray Spectrometry eds. Van Grieken R. E. and Markowicz A. A. Marcel Dekker New York 1993. Van Espen P. Janssens K. and Nobels J. Chemom. Intell. Lab. Syst. 1986 1 109. Adler I. Axelrod J. and Branco J. J. R. Adv. X-ray Anal. 1992 2 OOO. Heinrich K. F . J. Adv. X-ray Anal. 1992 35 15. Rose H. J. Christian R. P. Lindsay J. R. and Larson R. R. Geol. Surv. Pmf. Pap. (US) 1969 650-B B128. Nichols M. C.and Ryon R. W. Adv. X-ray Anal. 1986 29 423. Boehme D. R. Ado. X-ray Anal. 1987 30 39. Boehme D. R. and Yang N. Y. C. Microbeam Anal. 1993,2 S86. Nichols M. C. Boehme D. R. Ryon R. W. Wherry D. Cross D. and Aden G. Adv. X-ray Anal. 1987 30 45. Ryon R. Martz H. E. Hermandez J. M. Cross B. and Wherry D. A.dv. X-ray Anal. 1988 31 35. Wherry D. C. Cross B. J. and Briggs T. H. Adv. X-ray Anal. 1988 31 93. Cross B. J. Lamb R. D. Ma S. and Paque J. M. Adw. X-ray Anal. 1992 35 1255. Zaits M. A. Proc. 27th Microbeam Analysis Society Meeting,JOURNAL OF ANALYTICAL ATOMIC SPECTROMETRY MARCH 1994 VOL. 9 157 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 Boston August 16-21 1992 ed. Bailey G. W. Bentley J. and Small J. A. San Francisco Press San Francisco 1992 p.1756. Zaits M. A. paper presented at the 42nd Denver Conference on Applications of X-ray Analysis Denver CO USA August 2-6 1993 book of abstracts p. 150. Gissot R. G. and Boehme D. R. paper presented at the 42nd Denver Conference on Applications of X-ray Analysis Denver CO USA August 2-6 1993 book of abstracts p. 152. Havrilla G. J. paper presented at the 42nd Denver Conference on Applications of X-ray Analysis Denver CO USA August 2-6 1993 book of abstracts p. 153. Pella P. A. Lankosz M. and Holynska B. Proc. 27th Microbeam Analysis Society Meeting Boston August 16-21 1992 ed. Bailey G. W. Bentley J. and Small J. A. San Francisco Press San Francisco CA USA 1992 p. 1754. Pella P. A. Lankosz M. and Holynska B. paper presented at the 42nd Denver Conference on Applications of X-ray Analysis Denver CO USA August 2-6 1993 book of abstracts p.151. Lankosz M. Pella P. paper presented at the 42nd Denver Conference on Applications of X-ray Analysis Denver CO USA August 2-6 1993 book of abstracts p. 154. Hirch P. B. in X-ray Diffraction and Polycrystalline Materials eds. Peiser H. S. Roocksby H. P. and Wilson A. J. C. Plenum Press London 1955. Mosher D. and Stephanskis S. J. Appl. Phys. Lett. 1976 29 105. Spiller E. and Segmuller A. Appl. Phys. Lett. 1974 27 101. Chung P. S. and Pantell R. H. Electron. Lett. 1977 13 527. Pantell R. H. and Chung P. S. IEEE J . Quantum Electron. 1978 14 694. Rindby A. Nucl. Instrum. Methods 1986 A249 536. Carpenter D. A. Taylor M. A. and Holcombe C. E. Adu. X-ray Anal.1989 32 115. Carpenter D. A. and Taylor M. A. Adu. X-ray Anal. 1991 34 217. Carpenter D. A. and Taylor M. A. Proc. 27th Microbeam Analysis Society Meeting Boston August 16-21 1992 ed. Bailey G. W. Bentley J. and Small J. A. San Francisco Press San Francisco CA USA 1992 pp. 1758-1759. Carpenter D. A. and Taylor M. A. Microbeam Anal. 1993,2 S84. Carpenter D. A. and Taylor M. A. paper presented at the 42nd Denver Conference on Applications of X-ray Analysis Denver CO USA August 2-6 1993 book of abstracts p. 148. Stern E. A. Kalman Z. Lewis A. and Lieberman K. Appl. Opt. 1988 27 5135. Rindby A. Engstrom P. Larsson S. and Stocklassa B. X-ray Spectrom. 1989 28 109. Thiel D. J. Hoffman S. A. and Bildenback D. Physica B+C 1989 158 314. Larsson S. Engstrom P. Rindby A. and Stocklassa B.Adu. X-ray Anal. 1990 33 623. Engstrom P. Larsson S. Rindby A. and Stocklassa B. Nucl. Instrum. Methods 1989 B36 222. Shakir N. Larsson S. Engstrom P. and Rindby A. Nucl. Instrum. Methods 1990 B52 194. Stocklassa B. Nillson G. and Paulsson N. Proceedings of the European Conference on EDXRF Myconos Greece May 30- June 6 1992 p. 61 (book of abstracts). Rindby A. Voglis P. Nilsson G. and Stocklassa B. Adu. X-ruy Anal. 1992 35 1247. Voglis P. Attaelmanan A. Engstrom P. Larsson S. Rindby A. Bostrom K. and Helander C. G. X-ray Spectrom. 1993 22 229. Gurker N. X-ray Spectrom. 1979 8 149. Bavdaz M. and Gurker N. X-ray Spectrom. 1993 22 65. Jones K. W. and Gordon B. M. Anal. Chem. 1989 61 341A. Bavdas M. Knochel A. Ketelsen P. Petersen W. Gurker N. Salehi M. H.and Dietrich T. Nucl. Instrum. Methods 1988 A266 308. 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65 66 67 68 69 70 71 Van Langevelde F. Bowen D. K. Tros G. H. J. Vis R. D. Huizing A. and de Boer D. K. G. Nucl. Instrum. Methods 1990 A292 719. Goshi Y. Aoki S. Hayakawa S. Yamaij H. and Sakurai K. Jpn. J. Appl. Phys. 1987 26 L1260. Van Langevelde F. Janssens K. Adams F. C. and Vis R. D. Nucl. Instrum. Methods 1992 A317 383. Engstrom P. Larsson S. Rindby A. and Stocklassa B. Proc. 2nd European Conference on Progress in X-ray Synchrotron Research ed. Balerna A. Bernieri E. and Mobilio S. SIF Bologna 1990 pp. 283-286. Engstrom P. Larsson S. Rindby A. Buttkewitz A. Garbe S. Gaul G. Knochel A. and Lechtenberg F. Nucl. Instrum. Methods 1991 A302 547.Thiel D. J. Bilderback D. H. and Lewis A. Nucl. Instrum. Methods 1992 A317 597. Bilderback D. H. Thiel D. H. and Hoffman S. A. Microbeam Anal. 1993 2 S73. Sutton S. R. Jones K. W. Gordon B. M. Rivers M. L. Bajt S. and Smith J. V. Geochim. Cosmochim. Acta 1993 57,461. Janssens K. Vincze L. Adams F. and Jones K. W. Anal. Chim. Acta 1993 283 98 Jones K. W. and Gordon B. M. in Microscopic and Spectroscopic Imaging of the Chemical State ed. Morris D. M. Marcel Dekker New York 1993 ch. 9. Iida A. and Gohshi Y. Trace Element Analysis by X-ray Fluorescence Handbook of Synchrotron Radiation eds. Ebashi S. Koch M. and Rubenstein E. North-Holland Amsterdam 1991 Waychunas G. A. Am. Mineral. 1987 12 89. Behrens P. Felshe J. Vetter S. and Schulz-Ekloff E. K. J. Chem. Soc. Chem.Commun. 1991,61 678. Bertch P. M. Hunter D. B. Sutton S. R. Bajt S. and Rivers M. L. NSLS Annual Report 1992 eds. Hulbert S. L. and Lazarz N. M. Upton NY 1993 p. 385. Janssens K. Adams F. Rivers M. L. and Jones K. W. Proc. 11th Pfefferkorn Conf. Amherst MA August 7-14 1992 Scan. Microsc. Suppl. 1994 in the press. Vis R. D. in Applications of Synchrotron Radiation eds. Catlow C. R. A. and Greaves G. N. Blackie Glasgow 1990 ch. 13 Jones K. W. in Handbook of X-ray Spectrometry eds. Van Grieken R. E. and Markowicz A. A. Marcel Dekker New York 1993. Smith J. V. Conf. Ser. Inst. Phys. 1992 160 605. Vincze L. Janssens K. and Adams F. C. Con$ Ser. Inst. Phys. 1992 160 613. Gohshi Y. personal communication. Chapman K. L. Thompson A. L. and Underwood J. H. Proc. 42nd Denver Conference on Applications of X-ray Analysis Denver CO USA August 2-6 1993 book of abstracts p. 119. Proceedings of a Workshop on X-ray Focusing Techniques and Applications NSLS Upton NY USA May 18 1993 ed. Rothman E. Lazarz N. and McWhan D. NSLS Newsl. July 1993 p. 11. Kuznetsov S. Snigirev A. Snigireva I. Engstrom P. and Riekel C. Appl. Phys. Lett. submitted for publication. Janssens K. Vincze L. Lechtenberg F. Garbe S. Gaul G. Knochel A. and Adams F. X-ray Spectrom. 1994 submitted for publication. Lai B. Chapman K. and Cerrina F. Nucl. Instrum. Methods 1988 A266 544. VO~. 4 ch. 9 pp. 307-348. pp. 514-550. Paper 3/05717H Received September 22 I993 Accepted November 18 1993

ISSN:0267-9477    

DOI:10.1039/JA9940900151

出版商:RSC    

年代:1994

数据来源: RSC