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/JA99409FX045

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/JA99409BX047

出版商:RSC    

年代:1994

数据来源: RSC

摘要:

203 R JOURNAL OF ANALYTICAL ATOMIC SPECTROMETRY JULY 1994 VOL. 9 ATOMIC SPECTROMETRY UPDATED REFERENCES The address given in a reference is that of the first named author and is not necessarily the same for any co-author. 9412 176. 9412177. 9412 178. 9412 179. 9412 1 80. 941218 1. 9412182. 9412 1 8 3. 9412 184. 9412 18 5. 94/21 86. Marcus R. K. Operation principles and design con- siderations for radiofrequency powered glow discharge devices J. Anal. At. Spectrom. 1993 8 935. (Dept. Chem. Howard L. Hunter Chem. Lab. Clemson Univ. Clemson SC 29634-1905 USA). Riby P. G. Harnly J. M. Characterization of a helium discharge for hollow anode furnace atomization non- thermal excitation spectrometry J. Anal. At. Spectrom. 1993 8 945. (United States Dept. Agric. Beltsville Human Nutr.Res. Center Nutrient Composition Lab. Building 161 BARC-East Beltsville MD 20705 USA). Hettipathirana T. D. Blades M. W. Furnace atomiz- ation plasma excitation spectrometry effects of sodium chloride and sodium nitrate on lead and silver emission J. Anal. At. Spectrom. 1993 8 955. (Dept. Chem. 2036 Main Mall Univ. British Columbia Vancouver British Columbia Canada V6T lZl). Matusiewicz H. Use of the Hildebrand grid nebulizer as a sample introduction system for microwave-induced plasma spectrometry J. Anal. At. Spectrom. 1993 8 61. (Politech. Poznanska Dept. Anal. Chem. 60-965 Poznan Poland). Beinrohr E. Bulska E. Tschopel P. Tolg G. Determination of lead by electrothermal vaporization microwave-induced plasma atomic emission spec- trometry after flow-through electrolytic deposition in a graphite tube packed with reticulated vitreous carbon J.Anal. At. Spectrom. 1993,8 965. (Dept. Anal. Chem. Slovak Tech. Univ. CS-812 37 Bratislava Slovakia). Jakubowski N. Feldmann I. Stuewer D. Diagnostic investigations of aerosols with varying water content in inductively coupled plasma mass spectrometry J. Anal. At. Spectrom. 1993 8 969. (Inst. Spektrochem. ange- wandte Spektrosk. Postfach 10 13 52 D-44013 Dortmund Germany). Ebdon L. Fisher A. Handley H. Jones P. Determination of trace metals in concentrated brines using inductively coupled plasma mass spectrometry on-line preconcentration and matrix elimination with flow injection J. Anal. At. Spectrom. 1993 8 979. (Plymouth Anal. Chem. Res. Unit Dept. Environ. Sci. Univ. Plymouth Drake Circus Plymouth UK PL4 8AA).Turnlund J. R. Keyes W. R. Scott K. C. Ehrenkranz R. A. Isotope ratios of calcium determined in calcium-46 enriched samples from infants by automated multiple-collector thermal ionization mass spec- trometry J. Anal. At. Spectrom. 1993 8 981. (Western Human Nut. Res. Center United States Dept. Agric. Agric. Res. Service PO Bos 29997 Presidio of San Francisco CA 94129 USA). Smith C. M. M. Nichol R. Littlejohn D. Evaluation of linear photodiode array detection for continuum source atomic absorption spectrometry with electrother- mal atomization J. Anal. At. Spectrom. 1993 8 989. (Dept. Pure Appl. Chem. Univ. Strathclyde 295 Cathedral St. Glasgow UK G1 1XL). Ni Z.-m. He B. Han H.-b. In situ concentration of selenium and tellurium hydrides in a silver-coated graphite atomizer J.Anal. At. Spectrom. 1993 8 995. (Research Center for Eco-Environ. Sci. Acad. Sin. P.O. Box 2871 Beijing China). Johannessen J. K. Gammelgaard B. Jons O. Hansen S. H. Comparison of chemical modifiers for simul- taneous determination of different selenium compounds in serum and urine by Zeeman-effect electrothermal 9412 18 7. 9412188. 9412 189. 9412 190. 9412 9412 91. 92. 9412 193. 9412 194. 9412 1 9 5. 9412 196. atomic absorption spectrometry J . Anal. At. Spectrom. 1993,8,999. (Royal Danish Sch. Pharmacy Dept. Gen. Chem. 2 Universitetsparken DK-2 100 Copenhagen Denmark). Tahan J. E. Granadillo V. A. Sainchez J. M. Cubillan H. S. Romero R. A. Mineralization of biological materials prior to determination of total mercury by cold vapour atomic absorption spectrometry J.Anal. At. Spectrom. 1993 8 1005. (Lab. Inst. Anal. Fac. Exp. Ciencias Univ. Zulia Maracaibo Zulia Venezuela). Elmahadi H. A. M. Greenway G. M. Immobilized cysteine as a reagent for preconcentration of trace metals prior to determination by atomic absorption spectrometry J. Anal. At. Spectrom. 1993 8 101 1. (Sch. Chem. Univ. Hull Hull UK HU67RX). Robles L. C. Garcia-Olalla C. Aller A. J. Determination of gold by slurry electrothermal atomic absorption spectrometry after preconcentration by Escherichia Coli and Pseudomonas Putida J . Anal. At. Spectrom. 1993 8 1015. (Dept. Biochem. Mol. Biol. Univ. Leon E-24071 Leon Spain). Coedo A. G. Dorado T. Rivero C. J. Cobo I. G. Study of X-ray fluorescence spectrometry and spark ablation inductively coupled plasma atomic emission spectrometry for chromium determination in ferrochro- mium from bulk metal samples J.Anal. At. Spectrom. 1993 8 103. (Censejo Superior de Investigaciones CientifiCas Gregorio del Am0 8 28040 Madrid Spain). Krustev T. B. Mincheva S. T. Angelov D. A. Vidolova-Angelova E. P. Determination of traces of lutetium in geological samples by resonance ionization spectroscopy J. Anal. At. Spectrom. 1993 8 1029. (Optics and Spectrosc. Lab. Inst. Solid State Phys. Bulgarian Acad. Sci. Boulevard Tzarigradsko chaussee 72 Sofia 1784 Bulgaria). Rattray R. Miiioso J. Salin E. D. Rapid sample preconcentration by aerosol deposition for the determi- nation of trace elements by inductively coupled plasma spectrometry J .Anal. At. Spectrom. 1993 8 1033. (Dept. Chem. McGill Univ. 801 Sherbrooke St. West Montreal QuCbec Canada H3A 2K6). Walder A. J. Koller D. Reed N. M. Hutton R. C. Freedman P. A Isotope ratio measurement by induc- tively coupled plasma multiple collector mass spec- trometry incorporating a high efficiency nebulization system J. Anal. At. Spectrom. 1993 8 1037. (Fisons Instruments Elemental Analysis Ion Path Road Three Winsford Cheshire UK CW7 3BX). Sturgeon R. E. Willie S. N. Zheng J. Kudo A. GrCgoire D. C. Determination of ultratrace levels of heavy metals in arctic snow by electrothermal vaporiz- ation inductively coupled plasma mass spectrometry J. Anal. At. Spectrom. 1993 8 1053. (Inst. Environ. Chemistry National Res. Council Canada Ottawa Ontario Canada K1A OR9).Hall G. E. M. Pelchat J. C. Determination of palladium and platinum in fresh waters by inductively coupled plasma mass spectrometry and activated char- coal preconcentration J. Anal At. Spectrom. 1993 8 1059. (Geol. Survey Canada 601 Booth St. Ottawa Ontario Canada K1A OE8). Tao H. Lam J. W. H. McLaren J. W. Determination of selenium in marine certified reference materials by hydride generation inductively coupled plasma mass spectrometry J. Anal. At. Spectrom. 1993 8 1067.204 R JOURNAL OF ANALYTICAL ATOMIC SPECTROMETRY JULY 1994 VOL. 9 9412197. 9412 198. 9412 199. 9412200. 94/2201. 9412202. 9412203. 9412204. 9412205. 9 412206. 94/2207. 9412208. (Inst. Environ. Chem. Natl. Res. Council Canada Ottawa Ontario Canada K1A OR6). Larsen E. H. Pritzl G. Hansen S.H. Arsenic speciation in seafood samples with emphasis on minor constituents an investigation using high-performance liquid chromatography with detection by inductively coupled plasma mass spectrometry J. Anal. At. Spectrom. 1993 8 1075. (Natl. Food Agency Inst. Food Chem. and Nutr. 19 Marrkhsj Bygade DK-2860 Saborg Denmark). Han H.-b. Liu Y.-b. Mou S.-f. Ni Z.-m. Speciation of arsenic by ion chromatography and off-line hydride generation electrothermal atomic absorption spec- trometry J. Anal. At. Spectrom. 1993 8 1085. (Res. Center Eco-Environ. Sci. Acad. Sin. P.O. Box 2871 Beijing 100085 China). Duan Y.-x. Li X.-y. Jin Q.-h. Electrothermal vaporization for sample introduction in microwave- induced plasma atomic absorption spectrometry J. Anal. At. Spectrorn.1993,8 1091. (Dept. Chem. Jilin Univ. Changchun 130023 China). Fernandez B. A. Fernandez de la Campa M. R. Sanz- Medel A. Improvement in mercury cold vapour atomic techniques by resorting to organized assemblies and on-line membrane drying of vapour J . Anal. At. Spectrom. 1993,8 1097. (Dept. Phys. and Anal. Chem. Fac. Chem. Univ. Oviedo c/Julihn Claveria 8 Oviedo Spain). Schnurer-Patschan C. Zybin A. Groll H. Niemax K. Improvement in detection limits in graphite furnace diode laser atomic absorption spectrometry by wave- length modulation technique J. Anal. At. Spectrom. 1993 8 1103. (Inst. Spektrochem. Angewandte Spektrosk Univ. Dortmund Bunsen-Kirchhoff-Str. 11 D-44139 Dortmund Germany). Zhuang Z. Yang P.-y. Wang X.-r. Deng Z.-w. Huang B.-l. Preliminary study on the use of palladium as a chemical modifier for the determination of silicon by electrothermal atomic absorption spectrometry J.Anal. At. Spectrom. 1993 8 1109. (Res. Center Environ. Sci. Xiamen Univ. Xiamen FJ 361005 China). Todorovic M. Vidovic S. Ilic Z. Effect of aqueous organic solvents on the determination of trace elements by flame atomic absorption spectrometry and induc- tively coupled plasma atomic emission spectrometry J. Anal. At. Spectrorn. 1993,8 1113. (Fac. Chem. Univ. Belgrade P.O.B. 550 1 1000 Belgrade Yugoslavia). Eisman M. Gallego M. Valcircel M. Indirect flame atomic absorption spectrometric determination of papaverine strychnine and cocaine by continuous precipitation with Dragendorff's Reagent J . Anal. At. Spectrorn. 1993 8 1117. (Dept.Anal. Chem. Fac. Sci. Univ. Cbrdoba 14004 Cbrdoba Spain). Sharp B. L. Chenery S. Jowitt R. Sparkes S. T. Fisher A. Atomic spectrometry update-atomic emis- sion spectrometry J . Anal. At. Spectrom. 1993,8 151R. (Chem. Dept. Loughborough Univ. Technol. Loughborough Leicestershire UK LE113TU). Hill S. J. Dawson J. B. Price W. J. Shuttler I. L. Tyson J. F. Atomic spectrometry update-advances in atomic absorption and fluorescence spectrometry and related techniques J. Anal. At. Spectrom. 1993 8 197R. (Dept. Environ. Sci. Univ. Plymouth Plymouth Devon UK PL48AA). Bacon J. R. Ellis A. T. McMahon A. W. Potts P. J. Williams J. G. Atomic spectrometry update-atomic mass spectrometry and X-ray fluorescence spectrometry J . Anal. At. Spectrom. 1993 8 261R (The Macaulay Land Use Res.Inst. Craigiebuckler Aberdeen UK AB9 245). Marshall J. Carroll J. Crighton J. S. Barnard C. L. R. Atomic spectrometry update-industrial 9412209. 94/22 10. 94/22 1 1. 9412212. 94/22 13. 9412214. 9412215. 94/22 16. 94/22 17. 9412218. 94/22 19. analysis metals chemicals and advanced materials J. Anal. At. Spectrom. 1993 8 337R. (ICI plc Wilton Res. Centre P.O. Box 90 Middlesbrough Cleveland UK TS6 8JE). Dahl K. Thomassen Y. Martinsen I. Radziuk B. Salbu B. Thermal stabilization of antimony in electro- thermal atomic absorption spectrometry J. Anal. At. Spectrom. 1994 9 1. (Natl. Inst. Occup. Health P.O. Box 8149 DEP N-0033 Oslo Norway). Dadfarnia S. Thompson K. C. Hoult G. Development of a simple method for the determination of toluene extractable organotin by electrothermal atomic absorp- tion spectrometry and its application to effluent analysis J .Anal. At. Spectrom. 1994 9 7. (Yorkshire Water LabServices Sheffield UK S2 4EQ). Zhuoer H. Silicon measurement in bone and other tissues by electrothermal atomic absorption spec- trometry J. Anal. At. Spectrom. 1994,9,11. (Guangzhou Environ. Monitoring Centre Guangzhou 5 10030 China). Geertsen C. Briand A Chartier F. Lacour J.-L. Mauchien P. Sjostrom S. Comparison between infra- red and ultraviolet laser ablation at atmospheric pressure-implications for solid sampling inductively coupled plasma spectrometry J . Anal. At. Spectrom. 1994 9 17. (Lab. Spectrosc. Laser Commissariat A L'Energie Atom. Service de Phys. d'Expkriment. et d'Analyse Centre d'Etudes de Saclay B2t.391 91191 Gif-sur-Yvette Cedex France). Vanhoe H. Dams R. Versieck J. Use of inductively coupled plasma mass spectrometry for the determi- nation of ultra-trace elements in human serum J. Anal. At. Spectrorn. 1994 9 23. (Lab. Anal. Chem. Univ. Ghent Inst. Nucl. Sci. Proeftuinstr. 86 B-9000 Ghent Belgium). Branch S. Ebdon L. O'Neill P. Determination of arsenic species in fish by directly coupled high- performance liquid chromatography-inductively coupled plasma mass spectrometry J. Anal. At. Spectrom. 1994 9 33. (Plymouth Anal. Chem. Res. Unit Dept. Environ. Sci. Univ. Plymouth Drake Circus Plymouth UK PL4 8AA). Berndt H. Schaldach G. High-performance flow electrothermal atomic absorption spectrometry for on-line trace element preconcentration-matrix separ- ation and trace element determination J.Anal. At. Spectrom. 1994 9 39. (Inst. Spectrochem. and Appl. Spectrosc. Bunsen-Kirchhoff-Str. 11 D-44139 Dortmund Germany). Absalan G. Chakrabarti C. L. Hutton J. C. Back M. H. Lazik C. Marcus R. K. Effect of discharge conditions on the sputtering and spatial distribution of atoms in a radiofrequency glow discharge atomizer for atomic absorption spectrometry J. Anal. At. Spectrom. 1994 9 45. (Dept. Chem. Carleton Univ. Ottawa Ontario Canada K1S 5B6). BudiE B. Hudnik V. Matrix effects of potassium chloride and phosphoric acid in argon inductively coupled plasma atomic emission spectrometry J. Anal. At. Spectrom. 1994,9,53. (Natl. Inst. Chem. Hajdrihova 19 POB 30 61115 Ljubljana Slovenia). Horiuchi T. Matsushige K. Total-reflection X-ray diffractometry and its applications to evaporated organic thin films Spectrochim.Acta Part B 1993 48 137. (Dept. Appl. Sci. Fac. Eng. Kyushu Univ. Hakozaki Higashi-ku Fukuoka 8 12 Japan). Battiston G. A. Gerbasi R. Degetto S. Sbrignadello G. Heavy metal speciation in coastal sediments using total-reflection X-ray fluorescence spec- trometry Spectrochirn. Acta Part B 1993 48 217. (1st. Chim. Tecnol. dei Radioelementi del C.N.R. Corso Stati Uniti 4 1-35020 Padova Italy).JOURNAL OF ANALYTICAL ATOMIC SPECTROMETRY JULY 1994 VOL. 9 205 R 9412220. 941222 1. 9412222. 9412223. 9412224. 9412225. 9412226. 9412227. 9412228. 94/2229. 9412230. 9412231. 9412232. Ojeda N. Greaves E. D. Alvarado J. Sajo-Bohus L. Determination of V Fe Ni and S in petroleum crude oil by total-reflection X-ray fluorescence Spectrochim.Acta Part B 1993 48 247. (Univ. Simon Bolivar Apartado 89000 Caracas 1080A Venezuela). Freiburg C. Krumpen W. Troppenz U. Determinations of Ce Eu and Tb in the electroluminesc- ent materials Gd202S and La,O,S by total-reflection X-ray spectrometry Spectrochim. Acta Part B 1993 48 263. (Forschungszentrum Julich GmbH Postfach 1913 W-5170 Julich Germany). Berneike W. Basic features of total-reflection X-ray fluorescence analysis on silicon wafers Spectrochim. Acta Part B 1993,48,269. (ATOMIKA Analysetechnik GmbH Bruckmannring 6 W-8042 OberschleiDheim Germany). Hidaka H. Masuda A. Isotopic search for spon- taneous fission-produced ruthenium silver and tel- lurium in uraninite Chem. Geol. 1993 106 187. (Dept.Chem. Tokyo Metropolitan Univ. Hachioji Tokyo 192-03 Japan). Stone W. E. Crocket J. H. Determination of noble and allied trace metals using radiochemical neutron activation analysis with tellurium coprecipitation Chem. Geol. 1993 106 219. (Dept. Geol. McMaster Univ. Hamilton Ontario Canada L8S 4M1). Wendt J. I. Wendt I. Tuttas D. Determination of U-Pb a es of zircons by direct measurement of the Planck-Inst. Chemie Postfach 3060 W-6500 Mainz Germany). Staud W. J. Oswald E. J. Schoonen M. A. A. Determination of sodium chloride and sulfate in dolomites a new technique to constrain the composition of dolomitizing fluids Chem. Geol. 1993,107,97. (Dept. Earth and Space Sci. State Univ. New York at Stony Brook Stony Brook NY 11794-2100 USA). Fryer B. J. Jackson S. E. Longerich H.P. Application of laser ablation microprobe-inductively coupled plasma mass spectrometry (LAM-ICP-MS) to in situ (U)-Pb geochronology Chem. Geol. 1993 109 1. (Dept. Earth Sci. and Centre for Earth Resources Res. Memorial Univ. Newfoundland St. John's Newfoundland. Canada A1B 3x5). Vanko D. A Sutton S. R. Rivers M. L. Bodnar R. J. Major element ratios in synthetic fluid inclusions by synchrotron X-ray fluorescence microprobe Chem. Geol. 1993 109 125. (Dept. Geol. Georgia State Univ. Atlanta GA 30303 USA). Poitrasson F. Pin C. Telouk P. Imbert J. L. Assessment of a simple method for the determination of Nb and Ta at the sub-pgg-' level in silicate rocks by ICP-MS Geostand. Newsl. 1993 17 209. (Dkpt. Sci. Terre Univ. Blake Pascal U.R.A. 10 C.N.R.S. 5 rue Kessler 63038 Clermont-Ferrand France).Hallett R. B. Kyle P. R. XRF and INAA determi- nations of major and trace elements in geological survey of Japanese igneous and sedimentary rock standards Geostand. Newsl. 1993 17 127. (Dept. Geosci. New Mexico Inst. Mining and Technol. Socorro New Mexico 87801 USA). Mannan A. Waheed S. Rahrnan A Ahrnad S. Qureshi I. H. Multi-element analysis of coastal marine sediment an IAEA proposed reference material (IAEA-356) Geostand. Newsl. 1993 17 223. (Nucl. Chem. Div. Pakistan Inst. Nucl. Sci. and Technol. P.O. Nilore Islamabad Pakistan). Croudace I. W. Randle K. Fluorine abundances of twenty nine geological and other reference samples using fast-neutron activation analysis Geostand. Newsl. 1993 17 217. (Geol. Dept. Univ. Southampton Southampton UK SO9 5NH).210Pb/20 fF Pb ratio Chem. Geol. 1993 106 467. (Max- 9412233. 9412234. 9412235. 9412236. 9412237. 9412238. 9412239. 9412240. 9412241. 9412242. 9412243. 9412244. 9412245. 9412246. 9412 24 7. Ghazi A. M. Vanko D. A. Roedder E. Seeley R. C. Determination of rare earth elements in fluid inclusions by inductively coupled plasma mass spectrometry (ICP-MS) Geochim. Cosmochim. Acta 1993 57 4513. (Dept. Geol. Georgia State Univ. Atlanta GA 30303 USA). Palmer M. R. Edmond J. M. Uranium in river water Geochim. Cosmochim. Acta 1993,57,4947. (Dept. Geol. Univ. Bristol Wills Memorial Building Queen's Rd. Bristol UK BS8 1RJ). Yin Q. Z. Jagoutz E. Verkhovskiy A. B. Wanke H. 1870s-'860s and 1870s-1880s method of dating an introduction Geochim. Cosmochim.Acta 1993,57,4119. ( Max-Planck-Inst. Chem. Abteilung Kosmochem. Postfach 3060 D-55020 Mainz Germany). Suzuki K. Lu Q. Shimizu H. Masuda A. Reliable Re-0s age for molybdenite Geochim. Cosmochim. Acta 1993 57 1625. (Dept. Chem. Fac. Sci. Univ. Tokyo Hongo Bunkyo Tokyo 113 Japan). Long A. Eastoe C. J. Kaufmann R. S. Martin J. G. Wirt L. Finley J. B. High-precision measurement of chlorine stable isotope ratios Geochim. Cosmochim. Acta 1993 57 2907. (Lab. Isotope Geochem. Dept. Geosci. Univ. Arizona Tucson AZ 85721 USA). McKelvey B. A. Orians K. J. Dissolved zirconium in the North Pacific Ocean Geochim. Cosmochim. Acta 1993 57 3801. (Dept. Chem. and Oceanogr. Univ. British Columbia Vancouver BC Canada V6T 124). Galoisy L. Calas G. Structural environment of nickel in silicate glasslmelt systems Part 1.Spectroscopic determination of coordination states Geochim. Cosmochim. Acta 1993 57 3613. (Lab. Minkral. Cristallogr. URA CNRS 09 Univ. Paris 6 et 7 and IPGP 75252 Paris 05 France). Miller J. H. Elreedy S. Ahvazi B. Woldu F. Hassanzadeh P. Tunable diode-laser measurement of carbon monoxide concentration and temperature in a laminar methane-air diffusion flame Appl. Optics 1993 32 6082. (Dept. Chem. George Washington Univ. Washington DC 20052 USA). Slangen P. Martinez C. Weber G. Lion Y. Measurement of chromium content in dichromated gelatin by X-ray fluorescence Appl. Opt. 1993,32,6132. (Hololab Service Phys. Gknerale Phys. Inst. B5 4000 Sart-Tilman Likge Belgium). Chapman H. N. Nugent K. A. Wilkins S. W. X-ray focusing using cylindrical-channel capillary arrays.I. Theory Appl. Opt. 1993 32 6316. (Sch. Phys. Univ. Melbourne Parkville Victoria 3052 Australia). Chapman H. N. RodC A. Nugent K. A. Wilkins S. W. X-ray focusing using cylindrical-channel capillary arrays. 11. Experiments Appl. Optics 1993 32 6333. (Sch. Phys. Univ. Melbourne Parkville Victoria 3052 Australia). Savory J. Wills M. R. Trace metals essential nutrients or toxins Clin. Chem. 1992 38 1565. (Dept. Pathol. Univ. Virginia Charlottesville VA 22908 USA). Willis J. B. Birth of the atomic absorption spec- trometer and its early applications in clinical chemistry Clin. Chem. 1993 39 155. (211 Orrong Rd. East St. Kilda Victoria Australia 3183). Milne D. B. Johnson P. E. Assessment of copper status effect of age and gender on reference ranges in healthy adults Clin.Chem. 1993 39 883. (US Dept. Agric. Agric. Res. Ser. Grand Forks Human Nutr. Res. Center Grand Forks ND 58202-9034 USA). Miller-Ihli N. J. Greene F. E. Direct determination of lead in sugars using graphite furnace atomic absorption spectrometry At. Spectrosc. 1993 14 85. (US Dept. Agric. Beltsville Human Nutr. Res. Center,206 R 9412248. 9412249. 9412250. 9412251. 94/22 5 2. 9412253. 9412254. 94/2255. 9412256 9412257. 9412258. 94/2259. 94/2260. JOURNAL OF ANALYTICAL Nutr. Composition Lab. Bldg. 161 BARC-East Beltsville MD 20705 USA). Burguera J. L. Burguera M. Brunetto M. R. In Vivo sample uptake and on-line measurements of zinc and copper in whole blood by microwave-assisted mineraliz- ation and flow injection AAS At.Spectrosc. 1993 14 90. (Dept. Chem. Fac. Sci. Univ. Los Andes P.O. Box 542 MCrida 5101-A Venezuela). Rabadan J. M. Galban J. Aznarez J. Direct determination of tin in rigid poly(vinylch1oride) by hydride generation atomic absorption spectrometry in organic media At. Spectrosc. 1993 14 95. (Dept. Anal. Chem. Fac. Sci. Univ. Zaragoza Zaragoza- 50009 Spain). Memon M. A. Wang X.-r. Huang B.-l. Flow injection extraction with mixed solvents its application to Cu determination with flame AAS At. Spectrosc. 1993 14 99. (Dept. Chem. Xiamen Univ. Xiamen 361005 China). Oktavec D. Lehotay J. Determination of Ti and A1 in polypropylene by AAS and ICP-AES in the presence of Na At. Spectrosc. 1993 14 103. (Dept. Anal. Chem. Slovak Tech. Univ. Radlinskeho 9 812 37 Bratislava Slovakia).Swerts J. Van Espen P. Geladi P. Partial-least- squares techniques in the energy-dispersive X-ray fluorescence determination of sulfur-graphite mixtures Anal. Chem. 1993 65 1181. (Dept. Chem. Micro and Trace Anal. Centre Univ. Antwerp 2610 Wilrijk Belgium). Mixon P. D. Griffin S. T. Bray C. W. Williams J. C. Temporal shape of the photocurrent and the discharge current in a pulsed hollow cathode discharge Anal. Chem. 1993 65 1406. (Dept. Chem. Memphis State Univ. Memphis TN 38152 USA). Fang Z. Welz B. Sperling M. Expansion of dynamic range of flame atomic absorption spectrometry by an efficient flow-injection dilution system based on disper- sion of microlitre-volume samples Anal. Chem. 1993 65 1682. (Dept. Appl. Res. Bodenseewerk Perkin- Elmer GmbH W-7770 Ueberlingen Germany).Bos A. Bos M. Van der Linden W. E. Artificial neural networks as a multivariate calibration tool modelling the iron-chromium-nickel system in X-ray fluorescence spectroscopy Anal. Chim. Acta 1993 277 289. (Dept. Chem. Anal. Univ. Technol. Twente 7500 AE Enschede Netherlands). Emteborg H. Baxter D. C. Frech W. Speciation of mercury in natural waters by capillary gas chromatogra- phy with a microwave-induced plasma emission detector following preconcentration using a dithiocarbamate resin micro-column installed in a closed flow-injection system Analyst (London) 1993 118,1007. (Dept. Anal. Chem. Univ. Umea 901 87 Umea Sweden). Remy I. Brossier P. Atomic absorption spectrometric detection of biotin-cymantrene as a metallo-tracer for the avidin-biotin system Analyst (London) 1993 118 1021.(Unite Immunoanal. Fac. Pharm. 21033 Dijon France). Ohta K. Yamanaka N. Inui S. Winefordner J. D. Mizuno T. Separation of atomic vapours by sequential metal vapour elution analysis Analyst (London) 1993 118 1031. (Dept. Chem. Materials Mie Univ. Tsu 514 Japan). Shlifshteyn A. Lang F. D. Quantitative spectrographic analysis for in situ continuous emission monitoring Appl. Spectrosc. 1993 47 479. (Exergetic Systems Inc. Livermore CA 94550 USA). Lust A. Atomic spectroscopy bibliography for January-June 1993 At. Spectrosc. 1993 14 106. (Perkin-Elmer Corp. Norwalk CT 06859-0012 USA). 941226 1. 9412262. 9412263. 9412264. 94/2265. 9412266. 9412267. 9412268. 9412269. 9412270. 9412271. 9412272. 9412273. ATOMIC SPECTROMETRY JULY 1994 VOL.9 Yu D. K. Determination of chemical elements in montmorillonite by X-ray fluorescence spectrometry using the cellophane glue-paper sample preparation method Fenxi Huaxue 1993 21 704. (Lanzhou Inst. Geol. Chinese Acad. Sci. Lanzhou 730000 China). Yuan Q. H. Zhang H. Q. Determination of trace elements in Humulus Eupulus L. by oxygen-rich flame atomic absorption spectrometry Fenxi Huaxue 1993 21 741. (Ningxia Anal. and Test Centre Yinchuan 750021 China). Angeli J. Haselgruebler K. Achammer E. M. Burger H. Surface analysis of steel sheets with GDOS [glow-discharge AES] and EPMA [electron micro- probe analysis] Fresenius’ J. Anal. Chem. 1993 346 138. (Res. Dev. and Testing Techniques Voest-Alpine Stahl Linz GmbH 4031 Linz Austria). Hoffmann V. Application of glow-discharge optical- emission spectroscopy (GDOES) to the analysis of PVD [plasma vapour deposition] and CVD [chemical vapour deposition] layers Fresenius’ J.Anal. Chem. 1993 346 165. (Inst. Festkoerper-und Werkstofforschung Dresden eV (IFW) D-8080 Dresden Germany). Andersson J T. Schmid B. Atomic emission detector in gas chromatographic trace analysis. Some studies on the performance and applications Fresenius’ J. Anal. Chem. 1993 346 403. (Anorg. Chem. Inst. D-4400 Munich Germany). Burba P. Rocha J. C. Schulte A. Cellulose TETPA a chelating collector designed for multi-element pre- concentration in flow systems Fresenius’ J. Anal. Chem. 1993 346 414. (Inst. Specktrochem. u Angewandte Spectroscopie D-4600 Dortmund 1 Germany). Lam C. K. Y. Zhang Y.K. Busch M. A. Busch K. W. Design and performance of a new continuous- flow sample-introduction system for flame infra-red emission spectrometry applications in process analysis flow-injection analysis and ion-exchange high- performance liquid chromatography Talanta 1993 40 867. (Dept. Chem. Baylor Univ. Waco TX Lankosz M. Monte Carlo simulation of the grain size and edge effects in X-ray fluorescence micro-analysis of geological samples X-Ray Spectrom. 1993 22 125. (Inst. Phys. and Nuclear Techniques Acad. Mining and Metallurgy Krakow Poland). Ladisich W. Wobrauschek P. Quantitative X-ray fluorescence analysis of iron-aluminium and copper- iron binary alloys using selective excitation X-Ray Spectrom. 1993 22 15 1. (Atominst. Oesterreichischen Univ. 1020 Vienna Austria).Markowicz A. Haselberger N. Whai Zin 00 U. Method for quantitative XRF analysis of inhomo- geneous thin specimens X-Ray Spectrom. 1993 22 160. (Instrumentation Unit IAEA 2444 Seibersdorf Austria). Kis-Varga M. Vegh J. Influence of in-sample scat- tering of fluorescent radiation on-line shapes of lithium- drifted silicon detectors in XRF studies X-Ray Spectrom. 1993 22 166. (Inst. Nuclear Res. Hungarian Acad. Sci. 4001 Debrecen Hungary). Al-Ammar A. S. Determination of impurities in nuclear-grade uranium compounds by X-ray fluor- escence spectrometry. [Reply] X-Ray Spectrom. 1993 22 178. (Baghdad Univ. Baghdad Iraq). Shelpakova I. R. Kosyakov V. I. Chanysheva T. A. Rossin A. E. Borisova E. S. Modelling of impurity preconcentration by distilling off the matrix in the analysis of high-purity substances Zh.Anal. Khim. 1993 48 610. (Inst. Inorg. Chem. Siberian Branch Russian Acad. Sci. Novosibirsk Russia). 76798-7348 USA).JOURNAL OF ANALYTICAL ATOMIC SPECTROMETRY JULY 1994 VOL. 9 207 R 9412274. 9412275. 9412276. 9412277. 9412277a. 9412278. 9412279. 94/2280. 9412281. 9412 2 8 2. 94/228 3. 9412284. 941228 5. 9412286. Kolosova L. P. Ushinskaya L. A. Kopylova T. N. Atomic-absorption determination of osmium in HGA-500 graphite furnace Zh. Anal. Khim. 1993 48 644. (Sci. Res. and Design Inst. Mech. Treatment Mineral Resour. St. Petersburg Russia). Chirkova G. D. Bondareva N. V. Zolotovitskaya E. S. Potapova V. G. Granova N. S. Pleskach L. I. Atomic absorption determination of fluorides using selective dissolution of lead zirconate-titanate Zh.Anal. Khim. 1993 48 648. (Inst. Single Crystals Kharkov Ukraine). Retonerova E. Korsakova M. Kashkan G. V. Shaibovich M. V. Atomic absorption and atomic emission determination of trace elements in human tooth enamel Zh. Anal. Khim. 1993 48 754. (Polytech. Univ. Tomsk Russia). Pupyshev A. A. Muzgin V. N. Methodological problems of thermodynamic simulation of element atomization in electrothermal atomic absorption spec- trophotometry Zh. Anal. Khim. 1993 48 774. (Ural Polytech. Inst. Sverdlovsk Russia). Bashin G. M. Savichev A. T. Khots M. S. Optimal detection of trace elements from the X-ray spectra of characteristic radiation of unknown intensity Zh. Anal. Khirn. 1993 48 827. (Inst. Lithosphere Russian Acad. Sci. Moscow Russia). Tarnopol’skii M. G.Duimakaeva T. G. Shpolyanskii A. Ya Duimakaev Sh. I. Vershinin A. S. Skomorovskii 0. S. Procedures and software for automated X-ray fluorescence analysis Zh. Anal. Khirn. 1993 48 870. (Rostov State Univ. Rostov Russia). Kuz’min N. M. Pukhovskaya V. M. Varshal G. M. Spivakov B. Ya. Maryutina N. A. Volynets M. P. Ryabukhin V. A. Chkhetiya N. N. Grebneva 0. N. Pavlutskaya V. M. Inductively coupled plasma atomic emission determination of rare earth elements in geological samples Zh. Anal. Khim. 1993 48 898. (V. I. Vernadskii Inst. Geochem. and Anal. Chem. Russian Acad. Sci. Moscow Russia). Teissedre P. L. Cabanis M. T. Cabanis J. C. Comparison of two mineralization methods for determi- nation of lead by electrothermal atomic absorption spectrometry. Application to soils vine leaves grapes musts grape residues and lees Analusis 1993 21 249.(Centre Formation et Rech. Oenol. Fac. Pharm. Univ. Montpellier 34060 Montpellier France). Jin C. Xu B. Xu T. Fang Y. Determination of tin(1v) in drinking water and mineral water by chemi- cally-modified probe atomic absorption spectrometry Fenxi Ceshi Xuebao 1993 12 64. (Zhejiang Fish Coll. Zhejiang China). Hang W. Yang P. Wang X. Huang B. Recent advances in inductively coupled plasma mass spec- trometry and glow-discharge mass spectrometry Fenxi Yiqi 1993 2 1. (Dept. Chem. Xiamen Univ. Xiamen 361005 China). Liu J. Duan Y. Hou M. Li Y. Jin Q. Improved construction of a microwave-plasma torch Fenxi Yiqi 1993 2 22. (Dept. Chem. Jilin Univ. Changchun 130023 China). Zhuang Y. Determination of neodymium in lanthanum tungstate laser crystals by inductively coupled plasma atomic emission spectrometry Guangpu Shiyanshi 1993 10 (3) 1.(Fujian Inst. Struct. Matter Acad. Sin. Fuzhou 350002 China). Mi R. Determination of niobium in steel by inductively coupled plasma atomic emission spectrometry Guangpu Shiyanshi 1993 10 (3) 12. (Chem. Anal. Centre Northeast Univ. Tech. Shenyang 110006 China). Shan H. He Y. Determination of potassium in starch by atomic absorption spectrometry Guangpu Shiyanshi 9412287. 9412288. 9412289. 9412290. 941229 1. 9412292. 9412293. 9412294. 941229 5. 9412296. 9412297. 941229 8. 94/2299. 9412300. 1993 10 (3) 16. (Yunnan Anal. Test. Inst. Kunming 650051 China). Wu C. Lin Y. Analysis of rare-earth elements erbium and ytterbium in dual-doped laser crystals of aluminium yttrium oxide by inductively coupled plasma atomic emission spectrometry Guangpu Shiyanshi 1993,lO (3) 20 15.(Fujian Inst. Struct. Matter Acad. Sin. Fuzhou 350002 China). Liu L. Determination of copper lead zinc and iron in ingots of cadmium of flame atomic absorption spectrometry Guangpu Shiyanshi 1993 10 (3) 23. (Central Lab. Huidong Lead-Zinc Mine Sichuan 615205 China). Hu B. Jiang Z. Zeng Y. Electrothermal vaporization for sample introduction in inductively coupled plasma atomic emission spectrometry and mass spectrometry Henliang Fenxi 1993 9 9. (Dept. Chem. Wuhan Univ. Wuhan 430072 China). Liang S. Rapid determination of total rare-earth metals in rock soil and river sediment by X-ray fluorescence analysis Henliang Fenxi 1993 9 48.(Anhui Geol. Experimental Inst. Hefei 230001 China). Benda F. Filistein V. Hezina F. Musil J. Determination of trace elements in rainwater by ICP AES with Ionex preconcentration Int. J. Enuiron. Anal. Chem. 1993,50,9. (Inst. Landscape Ecol. Czechoslovak Acad. Sci. 37005 Ceske Budejovice Czech Republic). Kuiperes G. Smit H. Sequential spectrometer with an integrated system design Znt. Lab. 1993 23 14 16. (Philips Analytical X-ray Almelo Netherlands). Frias M. Sanchez de Rojas M. I. Garcia N. Luxan M. P. Determination of minority elements in white Portland cement using ICP Znt. Lab. 1993 23 18 20. (Inst. Eduardo Torroja (CSIC) 28080 Madrid Spain). Vela N. P. Caruso J. A. Comparison of flame ionization and inductively coupled plasma mass spec- trometry for the detection of organometallics separ- ated by capillary supercritical-fluid chromatography J.Chromatogr. 1993 641 337. (Dept. Chem. Univ. Cincinnati Cincinnati OH 45221-0172 USA). Anon. Determination of lead in blood LaborPraxis 1993 17 98. Wei J.-z. Sun H.-w. Flame atomic absorption spectrometric determination of palladium after its electrolytic preconcentration on tungsten filament Lihua Jianyan Huaxue Fence 1993 29 140 141 170. (Dept. Chem. Nankai Univ. Tainjin 30001 1 China). Li Z.q. Flame atomic absorption spectrometric deter- mination of potassium and sodium in ground water Lihua Jianyan Huaxue Fence 1993 29 159. (Xingtai Hydrol. and Water Resource Exploration Brigade Xingtai 054000 Hebei China). Song J.q. Wang X.-h. Xu F.-f. Flame AAS determi- nation of cadmium lead and copper in water after the pre-concentration by co-precipitation Lihua Jianyan Huaxue Fence 1993 29 179.(Fenghua Municipal Sanitation and Anti-Epidemic Station Fenghua 3 15500 Zhejiang China). Ohta K. Mizutani K. Itoh S.-I. Mizuno T. Electrothermal atomic absorption spectrometric deter- mination of potassium in biological materials with a molybdenum tube atomizer Microchem. J. 1993 48 184. (Dept. Chem. Materials Fac. Eng. Mie Univ. Tsu Mie 514 Japan). Buoso M. C. Fazinic S. Haque A. M. I. Moschini G. Volpe A. Caravello G. U. Heavy-element distribution profiles in archaeological samples of human tooth enamel and dentine using the proton-induced X-ray emission technique Nucl. Instrum. Methods Phys. Res. Sect. B 1992 68 269. (1st. Nazl.Fisica Nuclear Lab. Nazl. Legnaro Legnaro Padua Italy).208 R 9412301. 9412302. 9412303. 9412304. 9412305. 9412306. 9412307. 9412308. 9412 3 09. 94/23 10. 94/23 11. 94/23 12. 94/23 1 3. JOURNAL OF ANALYTICAL ATOMIC SPECTROMETRY JULY 1994 VOL. 9 Bishopp J. A. Thompson G. E. Role of electron microscopy in the study of adhesion to aluminium substrates SIA Surf. Interface Anal. 1993 20 485. (Ciba Composites Duxford Cambridge UK). Greenwood J. C. Lamb B. Prutton M. Study of the chemical composition of some cobalt silicide layer structures SIA Surf. Interface Anal. 1993 20 524. (Dept. Phys. Univ. York Heslington York UK YO1 5DD). Cao H. Li W. Check method for analytical data of certified reference materials involving precision param- eters Yejin Fenxi 1993 13 (2) 33.(Res. Inst. Wuhan Iron and Steel Co. Wuhan 430080 China). Wang L. Ga S. Uhlis S. Pater M. State of modern X-ray fluorescence analysis Yejin Fenxi 1993 13 45. (Ningbo Anal. Instrument Service Station Simens Co. Ningbo Zhejiang 315010 China). Zhao Q. Sun T. Determination of micro-amounts of manganese and chromium in the serum of patients with glycosuria by graphite furnace atomic absorption spectrophotometry Zhongguo Yiyuan Yaoxue Zazhi 1993 13 169. (Shenyang Milit. Command General Hospital Shenyang 110015 China). Erkkila J. Armstrong R. Riihimaki V. Chettle D. R. Paakkari A. Scott M. Somervaille L. Starck J. Kock B. Aitio A. In vivo measurements of lead in bone at four anatomical sites long term occupational and consequent endogenous exposure Br. J. Ind. Med.1992 49 631. (Inst. Occup. Health SF-00290 Helsinki Finland). Vittur F. Tuniz C. Paoletti S. Rizzo R. Jones K. W. Elemental analysis of growth plate cartilage by synchrotron radiation-induced X-ray emission (SRIXE) Biochem. Biophys. Res. Commun. 1992 188 1010. (Dip. Biochim Univ. Trieste Italy). Rao B. S. Lal U. S. Bhatnagar I. K. Nair M. V. Correlation study of scanning electron microscopy- energy dispersive analysis of X-rays (SEM-EDAX) with wavelength dispersive X-ray fluorescence spectrometry method ( WDXRF) on punch-marked silver coins Indian J. Chem. Sci. 1991 5 91. (Natl. Res. Lab. Conserv. Cult. Prop. Lucknow 226 020 India). Kitsa V. Lioy P. J. Chow J. C. Watson J. G. Shupack S. Howell T. Sanders P. Particle-size distribution of chromium total and hexavalent chro- mium in inspirable thoracic and respirable soil particles from contaminated sites in New Jersey Aerosol Sci.Technol. 1992 17 213. (Robert Wood Johnson Med. Sch. Univ. Med. and Dent. New Jersey Piscataway Allwood M. C. Greenwood M. Assessment of trace element compatibility in total parenteral nutrition infusions Pharm. Weekbl. Sci. Ed. 1992 14 321. (Dept. Life Sci. Derbyshire Coll. Higher Educ. Mickleover Derby UK DE3 5GX). Bai Y.-z. Wang Y. Wang Y.-m. et al. X-ray fluorescence analyser Faming Zhuanli Shenqing Gongkai Shuomingshu CN 1,058,470 (Cl. G01N21/64) 1992 Appl. 91,104,263 02 Jul 1991; 22 pp. (Chinese Academy of Building Material Sciences Institute of Cement Science). Garg R. R. Puri S. Singh S. Mehta D. Shahi J. S. Garg M. L. Singh N. Mangal P.C. Trehan P. N. Measurements of L X-ray fluorescence cross sections and yields for elements in the atomic range 41 d Z d 52 at 5.96 keV Nucl. Instrum. Methods Phys. Res. Sect. B 1992 72 147. (Dept. Phys. Panjab Univ. Chandigarh 160014 India). SprangIe P. Ting A. Esarey E. Fisher A. Tunable short pulse hard X-rays from a compact laser synchro- tron source J . Appl. Phys. 1992 72 5032. (Plasma NJ 08854-5635 USA). 94/23 14. 94/23 15. 94/23 16. 94/23 17. 9412318. 9412319. 9412320. 9412321. 9412322. 9412323. 9412324. 94/23 25. 9412 3 26. Phys. Div. Nav. Res. Lab. Washington DC 20375-5320 USA). Le Grand J. B. Bruijn M. P. Frericks M. De Korte P. A. J. Houwman E. P. Flokstra J. Superconductive junctions for X-ray spectroscopy Low Temp. Detect. Neutrinos Dark Matter IV Proc.Int. Workshop 4th 1991 (Pub. 1992) 265. (Lab. Space Res. 3584 CA Utrecht The Netherlands). Labov S. Silver E. Le Gros M. Bland R. W. Dickson S. C. Dignan T. G. Laws K. Johnson R. T. Simon M. W. et al. Aluminium tunnel junction detector operation in an adiabatic demagnetization refrigerator Low Temp. Detect. Neutrinos Dark Matter I V Proc. Int. Workshop 4th 1991 (Pub. 1992) 285. (Lab. Exp. Astrophys. Lawrence Livermore Natl. Lab. Livermore CA USA). Iwanczyk J. S. Dorri N. Wang M. Szczebiot R. W. Dabrowski A. J. Hedman B. Hodgson K. O. Patt B. E. 20-Element mercury iodide energy dispersive X-ray array detector system IEEE Trans. Nucl. Sci. 1992 39 1275. (Xsirius Inc. Marina del Rey CA 90292 USA). Lumb D. H. Nousek J. A. Energy and time response of CCD X-ray detectors IEEE Trans. Nucl.Sci. 1992 39 1379. (Dept. Astron. Astrophys. Pennsylvania State Univ. University Park PA 16802 USA). Wernisch J. Schoenthaler A. August H. J. Method for in-situ calibration of semiconductor detectors Mikrochim. Acta Suppl. 1992 12 147. (Inst. Angew. Tech. Phys. Tech. Univ. Wien A-1040 Vienna Austria). Conti M. Del Guerra A. Mazzei D. RUSSO P. Bencivelli W. Bertolucci E. Messineo A. ROSSO V. Stefaniai A. et al. Use of the EGS4 Monte Carlo code to evaluate the response of mercuric iodide and cadmium telluride detectors for photons in the diagnos- tic energy range Nucl. Instrum. Methods Phys. Res. Sect. A 1992 322 591. (Dip. Sci. Fis. Univ. Napoli 1-80125 Naples Italy). McCammon D. Edwards B. Juda M. Plucinsky P. Zhang J. Kelley R. Holt S.Madejski G. Moseley S. Szymkowiaki A. Thermal detection of X-rays Low Temp. Detect. Neutrinos Dark Matter III [Proc. Int. Workshop] 3rd 1989 (Pub. 1990) 213. (Phys. Dept. Univ. Wisconsin Madison WI 53706 USA). Kraus H. Jochum J. Kemmather B. Gutsche M. Von Feilitzsch F. Moessbauer R. L. Superconducting tunnel junctions Low Temp. Detect. Neutrinos Dark Matter IV Proc. Int. Workshop 4th 1991 (Pub. 1992) 275. (Phys.-Dept. E 15 Tech. Univ. Miinchen 8046 Garching Germany). Kundra K. D. Composition analysis of materials by X-ray fluorescence spectrometry Adv. Crystallop. Cryst. Growth Proc. Indo-Sov. Symp. Cryst. Growth 1989 (Pub. 1991) 143. (Cent. Charact. Mater. Electron. Natl. Phys. Lab. New Delhi 110 012 India). Samat S. Sarmani S. X-ray and gamma-ray spec- troscopy systems at Universiti Kebangsaan Malaysia J.Sains Nukl. Malays. 1991 9 69. (Fac. Phys. Appl. Sci. Univ. Kebangsaan Malaysia Ukm Bangi 43600 Malay.). Price L. Morrison L. Calibration maintenance and operation of XRF spectrometers for silicone coatweight measurement Spectroscopy (Eugene Oreg.) 1992 7 32 36. (Avery Dennison Res. Cent. Pasadena CA 91107 USA). Koestler R. J. Indictor N. Fiske B. Characterization of Japanese papers using energy-dispersive X-ray spec- trometry Restaurator 1992 13 58. (Objects Conserv. Dept. Metrop. Mus. Art New York NY 10028 USA). Piorek S. Pasmore J. R. Hazardous waste screening and identification with portable X-ray fluorescenceJOURNAL OF ANALYTICAL ATOMIC SPECTROMETRY JULY 1994 VOL. 9 209 R 9412327. 9412328. 9412329. 9412330.94/233 1. 9412332. 9412333. 94/2334. 94/23 3 5. 9412336. 94/23 37. 94/23 3 8. 94/23 3 9. analyser Proc. Natl. Conf. Hazard. Wastes Hazard. Muter. 7th 1990 58. (Outokumpu Electron. Inc. Round Rock TX USA). Shoji T. Hirokawa K. Reactions of gold or chromium evaporated on silicon wafers measured by the refracted X-ray fluorescence method Surf Interface Anal. 1992 18 773. (Ube Kousan Ube Japan 755). Van den Berg L. Recent developments in mercuric iodide technology at EG&G Energy Measurements Inc. Santa Barbara Operations Nucl. Instrum. Methods Phys. Res. Sect. A 1992 322 453. (Santa Barbara Oper. EG and G Energy Meas. Inc. Goleta CA 93117 USA). Equer B. Thin-film semiconductors new materials for future detectors Nucl. Instrum. Methods Phys. Res. Sect. A 1992 322 457.(Lab. Phys. Interfaces Couches Minces Ec. Polytech. 91 128 Palaiseau France). Olschner F. Shah K. S. Lund J. C. Zhang J. Daley K. Medrick S. Squillante M. R. Thallium bromide semiconductor X-ray and y-ray detectors Nucl. Instrum. Methods Phys. Res. Sect. A 1992 322 504. (Radiat. Monit. Devices Inc. Watertown MA 02171 USA). Iwanczyk J. S. Low noise low power miniaturized processing electronics for X-ray detectors Nucl. Instrum. Methods Phys. Rex Sect. A 1992 322 543. (Xsirius Inc. Marina del Rey CA 90292 USA). Raghavaiah C. V. Rao N. V. Murty G. S. K. Rao M. V. S. C. Reddy S. B. Sastry D. L. KBIKcr ratios and chemical effects in partially filled 3d-shell elements X-Ray Spectrom. 1992 21 239. (Swami Jnanananda Lab. Nucl. Res. Andhra Univ. Visakhapatnam India 530 003).Bertuccio G. Sampietro M. Fazzi A. High resolution X-ray spectroscopy with silicon drift detectors and integrated electronics Nucl. Instrum. Methods Phys. Res. Sect. A 1992 322 538. (Dip. Elettron. Politec. Milano 20133 Milan Italy). Gostilo V. V. Low-noise electronics for semiconductor detectors Nucl. Instrum. Methods Phys. Res. Sect. A 1992 322 566. (Riga Sci. Res. Inst. Radioisot. Appar. Riga Latvia 226005). Van Vechten D. Wood K. S. Lovellette M. N. Arnold G. Expected consequences of magnetic field bias on quasiparticle loss rates in X-ray tunnel junction detectors and other resolution degrading effects Low Temp. Detect. Neutrinos Dark Matter IV Proc. Int. Workshop 4th 1991 (Pub. 1992) 347. (US Nav. Res. Lab. Washington DC 20375-5000 USA). Grassi D.Hu Z. Murolo F. Perillo E. Rosato E. Spadaccini G. Vigilante M. Preliminary tests on a mercuric iodide microstrip detector with analogue readout proposed as PSD in a Bragg X-ray spec- trometer Nucl. Instrum. Methods Phys. Res. Sect. A 1992 322 449. (Dip. Sci. Fis. Univ. ‘Federico II’ 80125 Naples Italy). Lund J. C. Shah K. S. Olschner F. Zhang J. Moy L. Medrick S. Squillante M. R. Recent progress in lead iodide X-ray spectrometer development Nucl. Instrum. Methods Phys. Res. Sect. A 1992 322 464. (Radiat. Monit. Devices Watertown MA 02172 USA). Benz K. W. Irsigler R. Ludwig J. Rosenzweig J. Runge K. Schaefer F. Schneider J. Webel M. X-ray detectors based on semi-insulating gallium arsenide substrate Nucl. Instrum. Methods Phys. Res. Sect. A 1992 322 493. (Fak.Phys. Univ. Freiburg D-7800 Freiburg Germany). Sumner T. J. Grant S. M. Bewick A. Li J. P. Smith K. Beaumont S. P. Gallium arsenide revisited as a room temperature X-ray detector Nucl. Instrum. Methods Phys. Res. Sect. A 1992 322 514. (Blackett 9412340. 9412341. 9412342. 9412343. 9412344. 9412345. 9412346. 9412347. 9412348. 9412349. 9412350. 94/23 5 1. 9412352. 94/23 5 3. Lab. Imp. Coll. Sci. Technol. Med. London UK SW7 2BZ). Economou T. E. Iwanczyk J. S. Rieder R. A mercuric iodide X-ray instrument for the Soviet Mars ’94 mission Nucl. Instrum. Methods Phys. Res. Sect. A 1992 322 633. (Enrico Fermi Inst. Univ. Chicago Chicago IL 60637 USA). Hegedus F. Szaloki I. On the track of trace elements. Supersensitive X-ray spectrometer Termeszet Vilaga 1992 123 318. (Hungary). Lebedev V.V. Calibration with pure elements and heterogeneity of standard reference materials in X-ray spectrometric analysis of steels Zaood. Lab. 1992 58 61. (PO ‘Izhorskii Zaved’ St. Petersburg Russia). Karpukova 0. M. Smagunova A. N. Skribko N. N. Zhargalsaikhan D. Development of methods for XRF analysis of multicomponent materials using the internal standard technique Zh. Anal. Khim. 1992 47 678. (Irkutsk Univ. Irkutsk Russia). Gunicheva T. N. Afonin V. P. Finkel’shtein Background correction procedure for multichannel X-ray fluorescence spectrometers X-Ray Spectrom. 1992 21 253. (Inst. Geochem. Irkutsk Russia). Stroh A. Determination of Pb and Cd in whole blood using isotope dilution ICP-MS At. Spectrosc. 1993,14 141. (Bodenseewerk Perkin-Elmer GmbH P.O. Box 101761 D-88647 Uberlingen Germany).Lbpez Garcia I. Arroyo CortCz J. Hernsindez Cbrdoba M. Generation of vapours from slurried samples for the fast determination of As and Hg in coal fly ash and diatomaceous earth At. Spectrosc. 1993 14 144. (Dept. Anal. Chem. Fac. Chem. Univ. Murcia 30071 Murcia Spain). Hernandez Lopez M. Arbaizar A. Determination of trace amounts of nickel in welding fumes by solvent extraction flame atomic absorption spectrometry At. Spectrosc 1993 14 148. (Dept. Anal. Chem. Fac. Sci. Univ. Malaga E-29071 Malaga Spain). Jian F. Lou C.-q. Wang Q-w. Determination of the bath ratio in acid aluminium electrolyte by flame AAS At. Spectrosc. 1993 14 154. (Dept. Chem. Central South Univ. Tech. Changsha Hunnan 410083 China). Tewari R. K. Gupta J.R. Rapid and economical method for the determination of gold in geological materials by FAAS using bleaching powder for sample dissolution At. Spectrosc. 1993 14 152. (Northern Region Geol. Survey India Lucknow-226020 India). Maruta T. Yokoyama S. Kobayashi H. Yamane T. Simultaneous determination of the major and minor elements in high-purity mullite by XRF analysis Bunseki Kagaku 1992 41 589. (Central Res. Lab. Chichibu Cement Co. Ltd. 2-1-1 Tsukimi-cho Kumagaya-shi Saitama 360 Japan). Yonemori S. Ichikura E. Sugizaki M. Determination of the co-monomer content in tetrafluoroethylene/ perfluoro( propylvinyl ether) copolymers by XRF analy- sis Bunseki Kagaku 1992 41 593. (Res. Center Asahi Glass Co. Ltd. 1150 Hazawa-cho Kanagawa-ku Yokohama-shi Kanagawa 221 Japan).LaBrecque J. J. Schorin H. Certification of VL-2 a Venezuelan lateritic material as a standard reference material Fresenius’ J. Anal. Chem. 1992,342 299. (Inst. Venezolano de Investigaciones Cientificas-IVIC Apartado 21 827 Caracas 1020A Venezuela). LaBrecque J. J. Schorin H. Certification of SLB-1 a Suriname lateritic bauxite material as a standard reference material Fresenius’ J. Anal. Chem. 1992 342 306. (Inst. Venezolano de Investigaciones Cientificas- IVIC Apartado 21827 Caracas 1020A Venezuela).210R 9412354. 9412355. 9412356. 9412357. 941235 8. 94/23 59. 9412360. 9412361. 9412362. 94/23 63. 9412364. 9412365. 94/23 66. 9412367. 9412368. JOURNAL OF ANALYTICAL ATOMIC SPECTROMETRY JULY 1994 VOL. 9 Marchut A. Spectral distribution of the chromium X-ray tube Fresenius’ J.Anal. Chem. 1993 345 737. (Inst. Petrol. Processing ul. Lukasiewicza 1 PL-3 1-429 Cracow Poland). Marchut A X-ray fundamental parameter methods for the determination of some elements in hydrocarbons by WDXRF Fresenius’ J. Anal. Chem. 1993 347 111. (Inst. Petrol. Processing ul. Lukasiewicza 1 31-429 Cracow Poland). Tsuji K. Hirokawa K. Take-off angle-dependent X-ray fluorescence of thin films at glancing incidence Spectrochim. Acta Part B 1993 48 1471. (Inst. Materials Res. Tohoku Univ. Katahira 2-1-1 Sendai 980 Japan). Staats G. Strieder S. Validation of sulfide-bearing certified reference samples by primary calibration as applied to X-ray fluorescence X-ray Spectrom. 1993 22 132. (AG der Dillinger Huttenwerke D-6638 Dillingen/Saar Germany).Sarode P. R. Effects of chemical combination on X-ray Ka emission spectra of chromium X-ray Spectrom. 1993 22 138. (Phys. Dept. Goa Univ. Taleigao Plateau Goa 403 202 India). Kerur B. R. Thontadarya S. R. Hanumaiah B. Measurement of X-Ray attenuation coefficients around K-absorption edges using Fe Kcr X-Rays X-ray Spectrom. 1993 22 156. (Dept. Phys. Karnatak Univ. Dharwad-580 003 India). Jahagirdar H. A. Hanumaiah B. Thontadarya S. R. Measurement of total photoelectric cross-sections for 52.4 keV photons X-ray Spectrom. 1993 22 163. (Dept. Phys. Karnatak Univ. Dharwad-580 003 India). Jha S. N. Agrawal R. M. Relative intensities of La Lp and Lf12,3,15 lines of gold electron beam excitation in the near-threshold region of the L shell X-ray Spectrom. 1993 22 345.(Spectrosc. Div. Bhabha Atom. Res. Centre Trombay Bombay 400 085 India). Bajpal M. L. H. N. Jospeh D. Patra P. K. Sensitive method for the determination of rare earth elements by radioisotope-excited XRF employing a high-purity germanium detector in optimized geometry X-ray Spectrom. 1993 22 349. (Nucl. Phys. Div. Bhabha Atom. Res. Centre Bombay-85 India). Wang Y.-z. Fast convergence procedure for a direct determination method in XRF X-ray Spectrom. 1993 22 352. (Wuhan Res. Inst. Mater. Protection Pao Feng Er Lu 182 Wuhan 430030 China). Puri S. Mehta D. Chand B. Singh N. Trehan P. N. L shell fluorescence yields and Coster-Kronig transition probabilities for the elements with 25 < Z < 96 X-ray Spectrom. 1993 22 358. (Dept. Phys. Panjab Univ. Chandigarh-160014 India).Marco Parra L. M. Greaves E. D. Paz J. L. Sajo- Bohus L. Simultaneous determination of rare earths by X-Ray fluorescence spectrometry using a funda- mental parameters method X-ray Spectrom. 1993 22 362. (Dept. Fisica. Univ. Simon Bolivar Apartado 89000 Caracas 1080A Venezuela). Manninen S. Karhumaa R. Energy-dispersive W Ka spectrometer for trace element analysis of rare earth elements X-ray Spectrom. 1993 22 368. (Dept. Phys. Univ. Helsinki Siltavuorenpenger 20 D SF-001 70 Helsinki Finland). Verkhovodov P. A. Intensities of individual L X-ray spectra calculated using the TEREN program X-ray Spectrom. 1993 22 377. (Inst. Mater. Sci. Problems Ukrainian Acad. Sci. 3 Krzhizhanovsky St. Kiev 252680 Ukraine). Vazquez C. de Leyt D. V. LaBrecque J. J. Analytical X-ray analysis techniques a panorama of some of the 9412369.94/23 70. 9412371. 9412372. 9412373. 9412 3 74. 9412375. 9412376. 94/23 77. 9412 3 78. 9412379. 94/23 80. 94/23 8 1. applications in Latin America Adu. X-Ray Anal. 1992 35 1097. (Dept. Quim. Anal. Comisi6n Energ. Atom. Buenos Aires Argentina). Yacout A. M. Dunn W. L. EDXRF analysis for elements of adjacent atomic number but grossly different concentrations Appl. Radiat. Isot. 1993 44 1373. (Quantum Research Services Inc. 100 Capitola Dr. Suite 306 Durham NC 27713-4411 USA). Treiman A. H. Sutton S. R. Petrogenesis of the Zagami meteorite inferences from synchrotron X-ray (SXRF) microprobe and electron microprobe analyses of pyroxenes Geochim. et Cosmochim. Acta 1992 56 4059. (SN21 NASAIJSC Houston TX 77058 USA).Spiller E. Steams D. Krumrey M. Multilayer X-ray mirrors interfacial roughness scattering and image quality J. Appl. Phys. 1993 74 107. (IBM Res. Div. T. J. Watson Res. Center P.O. Box 218 Yorktown Heights NY 10598 USA). Srichom V. L. K. X-Ray fluorescence analysis of liquid specimens using the double dilution method J. Radioanal. and Nucl. Chem. Articles 1993 173 9. (Chem. Div. Office Atom. Energ. Peace Bangkok-10900 Thailand). Gerward L. X-Ray attenuation coefficients current state of knowledge and availability Radiat. Phys. Chem. 1993 41 783. (Lab. Appl. Phys. Tech. Univ. Denmark Bldg. 307 DK-2800 Lyngby Denmark). Potts P. J. Hawkesworth C. J. van Calsteren P. Wright I. P. Advances in analytical technology and its influence on the development of modern inorganic geochemistry a historial perspective Magmatic Processes and Plate Tectonics Geol.SOC. Spec. Publ. (London) 76 1993 501. (Dept. Earth Sci. Open Univ. Walton Hall Milton Keynes UK MK76AA). Fryar J. McCarthy K. J. Fenelon A. Multi-element imaging in computerized X-ray tomography Nucl. Instum. Methods Phys. Res. Sect. A 1988 271 671. (Sch. Phys. Sci. Natl. Inst. Higher Educ. Dublin Republic of Ireland). Pansky A. Breskin A. Chechik R. Malamud G. Detection of X-ray fluorescence of light elements by electron counting in a low-pressure gaseous electron multiplier Nucl. Instum. Methods Phys. Res. Sect A 1993 330 150. (Nucl. Phys. Dept. Weizmann Inst. Sci. 76100 Rehovot Israel). Markham C. Fryar J. Element specific imaging in computerised tomography using a tube source of X-rays and low energy-resolution detector system Nucl.Instum. Methods Phys. Res. Sect. A 1993 324 383. (Sch. Phys. Sci. Dublin City Univ. Dublin 9 Ireland). Kalinka G. Taniguchi K. Status of solid state energy dispersive X-ray detection Nucl. Instum. Methods Phys. Res. Sect. B 1993 75 91. (Inst. Nucl. Res. Hungarian Acad. Sci. P.O. Box 51 H-4001 Debrecen Hungary). Araujo M. F. Alves L. C. Cabral J. M. P. Comparison of EDXRF and PIXE in the analysis of ancient gold coins Nucl. Instum. Methods Phys. Res. Sect. B 1993 75 450. (Dept. Quim. ICENILNETI Est. Nac. 10,2686 SacavCm Portugal). Sutton S. R. Rivers M. L. Bajt S. Jones K. W. Synchrotron X-ray fluorescence microprobe analysis with bending magnets and insertion devices Nucl. Instum. Methods Phys.Res. Sect. B 1993 75 553. (Dept. Geophys. Sci. and Consortium Adv. Radiation Sources Univ. Chicago Chicago IL 60637 USA). Yao H.-y. Zhang J.-x. Wu X.-k. Zhu J.-q. Zeng X.-z. Element distributions and quantitative analysis of a single cell by micro-PIXE and synchrotron radiation XRF Nucl. Instum. Methods Phys. Res. Sect. B 1993 75 563. (Dept. Nucl. Sci. Fudan Univ. Shanghai 200433 China).JOURNAL OF ANALYTICAL ATOMIC SPECTROMETRY JULY 1994 VOL. 9 211R 9412382. 941238 3. 9412384. 9412385. 9412386. 94/238 7. 9412388. 9412 3 89. 9412390. 9412391 9412392. 9412393. 9412394. 941- 795. Sanchez H. J. Burattini E. Rubio M. Multielectron transition effects on X-ray-fluorescence spectra Phys. Rev. A 1993 47 4078. (Fac. Matematica Astron. y Fisicu Univ. Nac. Cbrdoba Laprida 854 CP 5000 Cbrdoba Argentina).Smagunova A. N. Medolazov L. Yu Molchanova E. I. Skribko N. N. Bespalova L. L. Comparison of the metrological characteristics of X-ray spectrometers Ind. Lab. 1993 58 824. (Inst. Oil and Coal Chem. Synthesis Irkutsk State Univ. Irkutsk). Ivanov A. L. Mosiyanov G. I. Grohman A. E. Metal sample preparing device using the method of remelt- ing Ind. Lab. 1993 58 828. (Sci. Ind. ASSOC. Chermetavtomatika Moscow Russia). Kalinin B. D. Plotnikov R. I. Estimation of the error of X-ray spectral analysis Ind. Lab. 1993 58 816. (Burevestnik Sci. Ind. Assoc. St. Petersburg Russia). Perry B. J. Speller D. V. Barefoot R. R. Van Loon J. C. Large sample dry chlorination ICP-MS analyt- ical method for the determination of platinum group elements and gold in rocks Can.J. Appl. Spectrosc. 1993 38 131. (Dept. Geol. Univ. Toronto Canada M5S 3B1). Sen Gupta J. G. Twenty-five years of research and applications of atomic absorption spectroscopy to geoanalysis in the geological survey of Canada Can. J. Appl. Spectrosc. 1993,38,145. (Geol. Survey Canada Ottawa Ontario Canada K1A OE8). Hlavilikova I. HlavaEek I. Analysis of ferrophos- phorus by inductively coupled plasma atomic emission spectrometry Chem. Listy 1993 87 583. (Nucl. Res. Inst. Re2 near Prague Czech Republic). Kubovi J. Polakoviliova J. StreSko V. Medved J. Determination of niobium and tantalum in geological materials by atomic emission spectroscopy with induc- tively coupled plasma Chem. Papers 1993 47 225. (Geol. Inst. Fac. Natural Sci. Comenius Univ.SK-842 15 Bratislava Slovakia). Kanickf V. Study of drift in inductively coupled plasma optical emission spectrometry. Analysis of silicate and carbonate raw materials Collect. Czech. Chem. Commun. 1993 58 2905. (Dept. Chem. Fac. Educ. Masaryk Univ. Poi'iEi 7 603 00 Brno Czech. Republic). Poussel E. Mermet J. M. Deruaz D. Dissociation of analyte oxide ions in inductively coupled plasma mass spectrometry J. Anal. At. Spectrom. 1994 9 61. (Lab. Sci. Anal. Unite Rech. AssociCe au Centre National de la Recherche Sci. (URA 435) Batiment 308 Univ. Claude Bernard-Lyon 1 69622 Villeurbanne France). Anderson D. R. McLeod C. W. Smith T. A. Rapid survey analysis of polymeric materials by laser-induced plasma emission spectrometry J. Anal. At. Spectrom. 1994 9 167. (Div.Chem. Sch. Sci. Sheffield Hallam Univ. Sheffield UK S1 1WB). Goode S. R. Thomas C. L. Determination of oxygen- containing additives in gasoline by gas chromato- graphy-microwave-induced plasma atomic emission spectrometry J. Anal. At. Spectrom. 1994 9 73. (Dept. Chem. and Biochem. Univ. South Carolina Columbia SC 29208 USA). Epler K. S. O'Haver T. C. Turk G. C. Liquid chromatography-laser-enhanced ionization spec- trometry for the speciation of organolead compounds J. Anal. At. Spectrom. 1994 9 79. (Dept. Chem. and Biochem. Univ. Maryland College Park MD 20742 USA). Barshick C. M. Smith D. H. Wade J. W. Bayne C. K. Isotope dilution glow discharge mass spec- trometry as applied to the analysis of lead in waste oil 9412396. 9412397. 9412 398. 9412399. 9412400.941240 1. 9412402. 941240 3. 9412404. 9412405 9412406. 9412407. 9412408. samples J. Anal. At. Spectrom. 1994 9 83. (Anal. Chem. Div. Oak Ridge Natl. Lab. Oak Ridge TN Okamoto Y. Sugawa K. Kumamaru T. Chemical modification for inductively coupled plasma atomic emission spectrometric determination of boron with tungsten boat furnace vaporizer J. Anal. At. Spectrom. 1994 9 89. (Dept. Chem. Fac. Sci. Hiroshima Univ. Higashi-Hiroshima 724 Japan). Marin S. R. Cornejo S. G. Arriagada L. Spectral line selection for determination of zirconium cerium thorium and titanium by inductively coupled plasma atomic emission spectrometry in zirconia-based ceramic materials J. Anal. At. Spectrom. 1994 9 93. (Chilean Commission Nucl. Energy Lo Aguirre Nuclear Centre Amunategui 95 Santiago Chile).Ronlievik S. Siroki M. Effects of low acetic acid concentrations in inductively coupled plasma atomic emission spectrometry J. Anal. At. Spectrom. 1994 9 99. (Dept. Chem. Lab. Anal. Chem. Fac. Sci. Univ. Zagreb Strossmayerov trg 14 41000 Zagreb Croatia). Luecke W. Systematic investigation of aluminium interferences on the alkaline earth elements in flame atomic absorption spectrometry. Part 111. The behav- iour of strontium and barium J. Anal. At. Spectrom. 1994 9 105. (Inst. Petrogr. and Geochem. Univ. Karlsruhe 76128 Karlsruhe Kaiserstrasse 12 Germany). PCrez Parajbn J. M. Sanz-Medel A. Determination of silicon in biological fluids uisng metal carbide-coated graphite tubes J. Anal. At. Spectrom. 1994 9 111. (Dept. Phys. and Anal. Chem. Univ. Oviedo C1 Julian Claveria 8 33006-Oviedo Spain).Cimadevilla E. A. C. Wrbbel K. Gaybn J. M. M. Sanz-Medel A. Determination of chromium in biologi- cal fluids by electrothermal atomic absorption spec- trometry using wall platform and probe atomization from different graphite surfaces J. Anal. At. Spectrom. 1994 9 117. (Dept. Phys. and Anal. Chem. Univ. Ovideo C1 Julian Claveria 8 33006 Oviedo Spain). Anzano J. M. Martinez-Garbayo M. P. Belarra M. A. Castillo J. R. Direct determination of copper at trace levels in solid samples of animal feed using electrothermal atomic absorption spectrometry J. Anal. At. Spectrom. 1994 9 125. (Dept. Anal. Chem. Univ. Zaragoza 50069-Zaragoza Spain). Cresser M. S. Armstrong J. Cook J. Dean J. R. Watkins P. Cave M. Atomic spectrometry update- environmental analysis J.Anal. At. Spectrom. 1994 9 25R. (Dept. Plant and Soil Sci. Aberdeen Univ. Meston Building Old Aberdeen UK AB9 2UE). Feng Z.-y. Li Y.-s. Xie F.-z. Direct determination of trace silver in geological samples by graphite furnace atomic absorption spectrometry Fenxi Ceshi Xuebao 1993 12 59. (Changsha Res. Inst. Mining and Metall. Changsha China). Zhang Z.-x. Li G.-k. Absolute analysis by graphite furnace atomic absorption spectrometry Fenxi Ceshi Xuebao 1993 12 7. (Dept. Chem. Zhongshan Univ. Guangzhou China). Jin C.-h. Xu B.-x. Xu .T.-g. Fang Y.-z. Determination of Sn (IV) in drinking water and mineral water by chemically modified probe atomic absorption spectrometry Fenxi Ceshi Xuebao 1993 12 67. (Zhejiang Fisheries Coll. China). He H.-k. Xie Y.-s.Chen J.-h. Study on the spectral line profiles in Zeeman atomic absorption spectroscopy Fenxi Ceshi Xuebao 1993 12 5. (Chinese Natl. Anal. Centre Guangzhou China). Lin Z.-x. Song J.-y. Mo S.+ Studies of the background absorption of Eu matrix in GFAAS Fenxi 37831-6375 USA).212R JOURNAL OF ANALYTICAL ATOMIC SPECTROMETRY JULY 1994 VOL. 9 Ceshi Xuebao 1993 12 34. (Dept. Chem. South China Normal Univ. Guangzhou China). 94/2411. Ma Y.-p. Gulikezi Fang X.-h. Determination of trace mercury in environmental and biological samples by 94/2409. Zhu X.-p. Cai S.-q. Wang K. Studies on the determination of serum selenium by Zeeman GFAAS Fenxi Shivanshi 1993 12 (4) 32. (Beiiing Gen. Res. flow-injection hydride generation GFAAS Guangpuxue Yu Guangpu Fenxi 1993 13 110. (Anal. and Testing Centre Xijiang Wulumuqi China).Inst. NonIFerrous Metals Beijing Chin;). - 94/2410. Lin K. Lin W. He Y.-z. Luo J.-h. Huang L.-j. Determination of trace tin in blood by GFAAS Fenxi Shiyanshi 1993 12 ( 5 ) 51. (Guangxi Instrum. Anal. and Res. Centre Nanning China). 94/2412. Zu LA. Li A.-m. Interference mechanism of calcium chloride in the determination of Mn by GFAAS Guangpuxue Yu Guangpu Fenxi 1993 13 85. (Dept. Chem. Peking Univ. Beijing China).

ISSN:0267-9477    

DOI:10.1039/JA994090203R

出版商:RSC    

年代:1994

数据来源: RSC

摘要:

745 JOURNAL OF ANALYTICAL ATOMIC SPECTROMETRY JULY 1994 VOL. 9 High-sensitivity Microwave-induced Plasma Mass Spectrometry for Trace Element Analysis Yukio Okamoto Department of Electrical and Electranic Engineering Faculty of Engineering Toyo University Ka wagoe Saitama 350 Japan A high power (< 1.5 kW 2.45 GHz) atmospheric pressure rlitrogen microwave-induced plasma mass spec- trometer is described for trace element analysis. The plasma which has an annular shape was produced by an 'Okamoto cavitj' operated in a surface-wave mode. The back round mass spectrum was dominated by 30NO+ 14N+ and ' O' and argon related ions such as 39Ar+ 40A? 52ArCf 56ArO+ and ''Ar2+ were not observed. Preliminary detection limits for 39K+ 40Caf 52Cr+ and 56Fe+ obtained directly were less than 5 ppt which are lower than those for argon inductively coupled plasma mass spectrometry.Analytical curves for the elements of interest were linear over five orders of magnitude of concentration. Keywords Microwave-induced atmospheric pressure nitrogen plasma; micro wave-induced plasma mass spectrometry; Okamoto cavity; trace element analysis The technique of nitrogen microwave-induced plasma mass spectrometry (N MIP-MS) has received a great deal of attention as a method for sensitive analytical element analy- sis.lP5 The use of a nitrogen plasma provides several advantages compared with an argon plasma the elimination of some spectral interferences induced by the argon isotopes the formation of polyatomic argon species associated with mass spectrometry i.e.the background species found in a nitrogen plasma during the nebulization of aqueous solutions are not as complex as the background found with argon inductively coupled plasma mass spectrometry (Ar ICP-MS).1-6 However existing N2 MIPS produced using a modified Beenakker cavity have a low tolerance to liquid aerosol samples because the input microwave power is limited to up to 500 W.lP4 The low-power MIP does not provide sufficient plasma energy to both desolvate and ionize the elemental mass from the directly nebulized sample solution. Therefore the detection limits for MIP-MS are higher than those of Ar ICP-MS with the exception of 39Kf 40Ca+ and 75A~f.3 The purpose of this study was to explore the analytical features of a high-power (< 1.5 kW) N2 MIP-MS ~ y s t e m . ~ * ~ The plasma was produced by an Okamoto cavity7 in a surface- wave mode which can supply up to 3 kW of microwave power.The preliminary analytical characteristics i.e. the background mass spectrum analytical curves and detection limits for several elements are presented. Problems encountered in using the high-power N MIP source analytically are illustrated. Experimental A schematic diagram of the experimental apparatus which was designed and constructed in this laboratory is shown in Fig. 1.' Briefly sample solutions are nebulized and carried into the N MIP where they are vaporized excited and ionized. Ions are extracted through a differentially pumped orifice into a vacuum interface into the Einzel lens system with a photon stop and a quadrupole mass spectrometer (filter). The appar- atus is principally the same as that for Ar ICP-MS.* Microwave Plasma Source The microwave cavity used here was an Okamoto cavity with a choke as shown in Fig.2.7 The cross-section of the cavity (a) and the radial distribution of the electric field (b) are shown in Fig. 2. The choke was used to reduce leakage of the microwave power. An annular-shaped nitrogen plasma was produced in a quartz discharge tube with a 10 mm i.d. (1 mm thickness) (see Fig. 3). The tube consists of two concentric tubes an inner and an outer. The inner tube is tulip-shaped with a large outer diameter of 9 mm and a small inner diameter of 1 mm. There are two gas flows a central carrier gas flow with the sample aerosol and a plasma support gas flow with a spiral trajectory into the inner and outer tubes respectively.These flow rates referred to as G and G are typically 0.6-1.4 and 10-141min-1 respectively. The plasma was run with 0.5-1.5 kW of forward microwave power and zero reflected power after tuning; it was ignited in argon and then changed over to nitrogen (purity 4N5 and 5N5) as soon as possible. The plasma conditions used were optimized for maximum signals of the analyte ions. Using an Okamoto cavity annular- shaped oxygen and air plasmas can be produced at atmospheric pre~sure.~ Sample Introduction The sample introduction systems used were the pneumatic and ultrasonic nebulizer of an ICP i.e. utilizing direct introduction of aqueous solutions. The spray chamber temperature was maintained at 6-25 "C by refrigerating circulation.Mixed aqueous solutions of 1-100 ppb of Ca Fe Mg etc. were used. The sample aerosol was introduced into the centre of the annular-shaped plasma through the inner tube along with the carrier gas. All sample handling was at atmospheric pressure. Plasma Vacuum Interface The sample ions extracted from the tail flame of the MIP through an orifice in a differentially pumped vacuum system were transmitted and focused before mass analysis. The system consists of a three-stage vacuum chamber which includes the plasma sampling region a transition flow region and a high vacuum region in which the quadrupole mass filter was located. The plasma sampling region consists of a dual-cone interface with a pressure of about 100Pa between the cones during plasma sampling.The sampling cone (sampler) made from Cu or Ni has a 0.7 mm diameter orifice. The majority of the gas that enters the first region through the sampler was pumped away by a rotary pump (1000 1 s-'). The plasma passing through the sampling orifice expands as a supersonic jet. The ions in the expanded plasma were extracted by the axis- symmetric cone (ion extractor) which is located approximately 7mm behind the sampler tip and has a 1.0mm diameter orifice. The second region contains the ion lens for forming and focusing the ions extracted from the expanded plasma into the mass filter. This was pumped by a 9801s-' oil diffusion pump. The pressure in this region is held at 2.7 x lo-' Pa during plasma sampling. An ion accelerator with an 8 mm diameter orifice (biased to negative) is located 8 mm746 JOURNAL OF ANALYTICAL ATOMIC SPECTROMETRY JULY 1994 VOL.9 If lecto r Gas supply Fig. 1 Schematic diagram of an atmospheric pressure MIP mass spectrometer RP = rotary pump; DP = diffusion pump; and TMP = turbo molecular pump ( a ) Microwave power (2.45 GHz 1.4 kW) Plasma support g Sample (aerosol) Side view L u 2 1 0 1 2 R a d i u s/cm Fig. 2 (a) Cross-section of Okamoto cavity and (b) radial distribution of electric field behind the ion extractor (earth potential). The lens (Einzel lens) for focusing the ion beam consists of the ion accelerator and two cylindrical electrodes. The Einzel lens stack incorpor- ates a central metal stop (4 mm diameter) which reduces the number of photons and/or neutral species leaving the plasma from reaching the detector.The lens potentials were set with the ion extractor at -17OV the stop at OV and the third focusing cylindrical electrode at 20 V. Ions enter the third high vacuum region through the lens. The third region contains the mass filter and is held at 5.3 x Pa during plasma sampling. The mass filter has a mass range of 1-420 (m/z) and is capable of unit mass resolution which is adequate for elemental analysis (Balzers QM-420). A deflector was located after the mass filter for "I 5 I I I I I I I I I I I I I \ $ $ \ 0 R ad i u s/m m 5 Fig. 3 (a) Photograph of annular-shaped atmospheric pressure N MIP (discharge tube 10 mm id.). (b) Radial distribution of emission intensity solid line N2+ (391.4 nm); and broken line Ca I1 (393.4 nm).Microwave power 1 kW G 10 1 min-'; and G 0.6 1 min-' directing the ions to an off-axis detector located perpendicular to the axis of the mass filter. Off-axis detection further reduces photon noise because photons are not deflected. Reduction of this background noise could lower the detection limits by up to a factor of 10. Signal Detection Ions were detected with the off-axis discrete and continuous dynode secondary electron multipliers which were used in analogue- and pulse- counting (digital) modes respectively. All analytical signals were determined by subtraction of the back- ground using distilled de-ionized water as the blank. For determination of the detection limits and the analytical curves the mass filter was tuned to the peak of interest and counts were totalled for 1-10 s.JOURNAL OF ANALYTICAL ATOMIC SPECTROMETRY JULY 1994 VOL.9 747 Results and Discussion Nitrogen Annular-shaped Plasma Fig. 3 shows (a) a photograph of the N2 MIP and (b) the radial distribution of the emission intensity for N2+ (391.4 nm) and analytical signals of Ca I1 (393.4 nm).' Here the microwave power is 1 kW and the gas flow rates G and G are 10 and 0.6 1 min-l respectively. As shown in this figure the plasma has an annular shape and the sample is efficiently ionized in the centre of the plasma. The temperature and the electron density of the plasma were 5500-6500 K and 1 x 1014 cmP3 respectively. Background Mass Spectrum The background mass spectra for the blank obtained from m/z 3 to 85 when a sample of distilled water was introduced into the plasma at three different detection sensitivities are shown in Fig.4.5 The main ions present in the background are due to nitrogen and water. The largest peak in the spectrum at m/z 30 is NO+ (ionization potential 9.3eV) which is about 1.5 times the magnitude of the next most concentrated ion of (14.5 eV). Oxygen ions such as l 6 0 + and "OH,+ from the water and/or air entrained into the plasma are another major source of background. Some other molecular ions formed from these elements can be seen in Fig. 4 including ,'N2+ 3202+ and 42N3+. Similar spectra have been reported by Wilson et aL2 and Shen et a1.,3 but the details of the background peaks differ in each case. From this figure it is evident that the N2 MIP-MS back- ground is free from spectral interfaces above m/z 45 so a larger background-free region is available than in Ar ICP-MS.It is interesting to note that in the present work unlike the work of Shen et aL3 who used a Beenakker cavity 56N4+ (m/z= 56) seems to play a small role. However when the sampling orifice was less than 0.6 mm a peak at m/z 56 thought to arise from 56N4+ was observed. One of the aims of the investigation of the N2 MIP-MS was to determine Fe at m/z 56 and its major isotope but the system is also suitable for the detection of 39K+ 40Caf 52Cr+ 75A~+ and "Se'. The effect of microwave power on the background ions of Fig. 5. On increasing the microwave power the concentration of these ions increases and for the low ionization potential ions such as 30NO+ (9.3 eV) and 3202+ (13.2 eV) the intensity saturates while for the high ionization potential ions such as "N2+ (15.5 eV) and 14N+ (14.5 eV) the intensity increases with increasing power.These results show that the effect of microwave power on intensity is correlated with the ionization potential of the elements i.e. higher microwave power is necessary for the elements with high ionization potential^.^ A background spectrum of a dry nitrogen plasma was obtained. The major background species were 14N+ (ionization potential 14.5 eV) "N2+ (15.5 eV) 30NO+ 17NH+ and 42N3+. The largest peak in the spectrum was 14N+ which is about 300 times the magnitude of the next ion ,'N2+. 1 4 ~ + 28N2+ 160f 3202+ and 42N3+ are shown in 3 0 ~ 0 + 1 4 ~ + Solution Analysis A typical mass spectrum obtained in the ion counting detec- tion mode for a Cr (5 ppb) + Fe (3 ppb) + Co (4 ppb) + Ni ( 10 ppb) + Cu (4 ppb) + Ga ( 15 ppb) mixed sample solution is shown in Fig.6. The microwave power was 1.3 kW with plasma and carrier gas flow rates of 13.5 and 1.4lmin-' respectively.The N MIP-MS system is capable of determining Cr Fe Ni Cu and Ga at their major isotopes and shows good experimental isotope ratios at less than 5 y 0 . ~ t z (I) C 4- .- 4- - II 10 I 30 50 70 90 10 30 50 70 10 30 m/z Fig.4 Background mass spectra of wet N MIP (analogue mode) (a) gain x lo6; (b) x 10,; and (c) x 1. Microwave power 850 W; G 11.5 1 min-' G 0.55 1 min-'; and sampler material Cu Effects of Power The effects of power on the analyte signals were evaluated for a number of elements with various ionization potentials.The analyte signals versus power for Mg Co and Y are shown in Fig. 7. This figure shows a fairly linear increase in signal with microwave power indicating that it is desirable to employ as high a power as is possible. Analytical Curves The dynamic range of the proposed plasma can be judged from Fig. 8 which shows the analytical curve (signal uersusJOURNAL OF ANALYTICAL ATOMIC SPECTROMETRY JULY 1994 VOL. 9 Microwave power/\/\/ Fig.5 Effects of microwave power on background ions A 30NO+ (x 9.4 eV); B 14N+ ( x lo-' 14.5 eV); C "N2+ ( x 0.2 15.5 eV); D l60+ (xO.1 13.6eV); E 3202+ (13.2eV); and F 42N3+ G 11.5 1 min-'; G 0.55 1 min-'; and sampler material Cu 60 65 70 75 m/z Fig.6 Typical N MIP mass spectrum for Cr (5 ppb)+Fe (3ppb)+Co (4ppb)+Ni (lOppb)+Cu (4ppb)+Ga (15ppb) mixed sample solution (pulse counting mode).Values given are natural abundance % ratio. Microwave power 1.3 kW; Gp= 13.5 1 min-'; G,= 1.5 1 min-'; and sampler Cu solution concentration) for Fe. The dynamic ranges for most elements of interest covered five orders of magnitude from the detection limit up to 100 ppb. These values are comparable to those for Ar ICP-MS. Detection Limits The detection limits were based on three standard deviations above the background and were obtained using 1 s integrations in a single-ion monitoring mode and with the experimental conditions optimized for each individual element. The detection limits for the elements of interest as a function of ionization potential are shown in Fig.9. These results indicate that detection limit is correlated with the ionization potential of the element. The microwave power used here was 1.3 kW and the flow rate of the plasma and carrier gases were 13 and 1.3 1 min-' respectively These values are about 1-3 orders of magnitude lower than those reported for other N MIP-MS system^.^ For K Ca Fe As and Se the values were improved by about 1-2 orders of magnitude and other A 600 800 1000 Microwave power/\/\/ Fig. 7 Effects of microwave power on intensity for Mg+Co +Y mixed solution sample A 24 Mg+ (7.65 eV); B s9C0+ (7.86 eV); and C "Y+ (6.38 eV); G 10 1 min-'; G 0.6 1 min-'; and sampler Cu Concentration (ppb) Fig. 8 Analytical curve for 56Fe+. Microwave Power 1.3 kW; G 13 1 min-'; G 1.3 1 min-'; and sampler Cu 1; 5 0.0 Pb co Ti Ca 0 0 4 5 6 7 8 5 10 Ionization potential/eV Fig. 9 Effects of ionization potential on detection limits. Microwave power 1.3 kW; G 13 1 min-' G 1.3 1 min-'; and sampler Cu elements with high ionization potentials are the same or one order of magnitude higher than those for Ar ICP-MS.3 The higher detection limit is because of the high concentration of 30NO+ in the nitrogen plasma as shown in Figs. 4 and 5. Further work will be directed at lowering the detection limits for elements with high ionization potentials by increasing the microwave power i.e. lowering of the 30NO+ content or by using helium as a plasma support gas.JOURNAL OF ANALYTICAL ATOMIC SPECTROMETRY JULY 1994 VOL. 9 749 Conclusions The high-power (< 1.5 kW) N MIP shows promise as an alternative ion source for elemental mass spectrometry.Using the N MIP it is possible to determine K Cay Cr Fe As and Se at their major isotopes without background interferences. These elements are difficult to determine using Ar ICP-MS as argon related polyatomic background species or matrix-related argon polyatomic species arise. Detection limits for these elements were obtained directly and were lower than those by Ar ICP-MS. The dynamic ranges for most of the elements were linear over five orders of magnitude of concentration and low-ppt to sub-ppt detection limits were obtainable for most of the elements. The new N2 MIP-MS could be used for the analysis of organic solvents,’ photoresistors and serum. Details of this work will be published in the near future.Part of this work was supported by Hitachi who provided the instruments and by the Grant-in-Aid for Scientific Research No. 05650817 (1993) from the Ministry of Education Science and Culture and also by the Special Research Fund for 1992 and 1993 of Toyo University. The authors thank Dr. Kounosuke Ooishi Masataka Koga and Satoshi Shimura of Hitachi for their valuable discussions and support. References Douglas D. J. and French J. B. Anal. Chem. 1981 53 37. Wilson D. A. Vickers G. H. and Hieftje G. M. Anal. Chem. 1987 59 1664. Shen W.-L. Davidson T. M. Creed J. T. and Caruso J. A. Appl. Spectrosc. 1990 44 1003. Shen W.-L. Davidson T. M. Creed J. T. and Caruso J. A. Appl. Spectrosc. 1990 44 1011. Okamoto Y. Simura S. Ooishi K. and Koga M. paper presented at the 1991 European Winter Conference on Plasma Spectrochemistry January 14-18 Dortmund Germany. Tan S. H. and Horlick G. H. Appl. Spectrosc. 1986 40 445. Okamoto Y. Anal. Sci. 1991 7 273. Wilson D. A. Vickers G. H. Hieftje G. H. and Zander A. T. Spectrochim Acta Part B 1987 42 29. Shirasaki T. and Yashuda K. Anal. Sci. 1992 8 375. Paper 4/00586D Received January 31 1994 Accepted March 18 1994

ISSN:0267-9477    

DOI:10.1039/JA9940900745

出版商:RSC    

年代:1994

数据来源: RSC

摘要:

JOURNAL OF ANALYTICAL ATOMIC SPECTROMETRY JULY 1994 VOL. 9 75 1 Evaluation of an Axially and Radially Viewed Inductively Coupled Plasma Using an Echelle Spectrometer With Wavelength Modulation and Second-derivative Detection* Yoshisuke Nakamura Katsuyuki Takahashi Osami Kujirai and Haruno Okochi National Research Institute for Metals 2-3- 72 Nakameguro Meguro-ku Tokyo 753 Japan Cameron W. McLeod Department of Chemistty Sheffield Hallam University Sheffield UK S I 7 WB A high-dispersion echelle spectrometer which incorporates wavelength modulation and second-derivative signal detection is used to view analyte emission from an inductively coupled plasma in both the axial (end- on) and radial (side-on) configurations. Movement of the plasma torch assembly between the viewing positions is computer controlled.Performance characteristics such as background intensity repeatability background equivalent concentration limit of detection dynamic range and interference effects are reported for 24 elements and in general measurement in the end-on position produces an improvement in sensitivity. Using end-on measurement the system is applied to the multi-element analysis of a certified reference water. Keywords End-on measurement; inductively coupled plasma atomic emission spectrometry; echelle spec- trometer; wavelength modulation; second-derivative detection One of the major areas of interest in inductively coupled plasma atomic emission spectrometry (ICP-AES) is optimiz- ation of the plasma operating parameters and spectrometer measuring parameters. Almost all development studies and applications have utilized side-on measurement.As the aerosol is introduced into the central part of the toroidal ICP end-on measurement could be expected to yield increased emission intensity for the elements owing to the extended source path relative to side-on viewing. Analytical performance of end-on measurement of a hori- zontal ICP has been investigated by several In the end-on measurements by Demers,' the optimum observation region was the same for both single and simultaneous quantifi- cation and only r.f. power was compromised in the simul- taneous quantification. An air cut-off stream was blown upwards towards the tip of the plasma. Kawaguchi et aL2 used a water-cooled low-flow ICP without a cut-off stream.However de Loos-Vollebregt et aL3 showed that analytical performance of the water-cooled low-flow ICP during end-on measurement without a cut-off stream was similar to that of a conventional high-flow ICP using side-on measurement. An echelle spectrometer has been used exclusively for side- on mea~urernents.~~ Less spectral interferences are expected with the high-dispersion kchelle spectrometer. In the present study the echelle spectrometer incorporating wavelength modulation and second-derivative detection is used to view analyte emission in both end-on and side-on configurations. Optimum operating parameters and basic performance data for end-on measurement are discussed and a critical compari- son is made with conventional side-on measurement. A novel aspect of the instrument is the rapid and automated changeover between viewing positions.Experimental Echelle Spectrometer A Kyoto Koken (Uji Japan) Model UOP-2 MARK I1 ICP atomic emission spectrometer was used. Details of this echelle spectrometer have been described el~ewhere.~ The Cchelle grat- ing was a fiat type with 79 grooves mm-'. Reciprocal linear dispersion was 0.031 0.078 and 0.12 nm mm-' at 200 500 and * Presented at the International Congress on Analytical Sciences Makuhari-Messe Japan August 25-31 1991. 800 nm respectively. Sequential measurement was used. The width and height of both the entrance and exit slits were 100 and 500 pm respectively. Signal integration time was 5 s. The spectral lines of 24 elements investigated are summarized in Table 1 in order of wavelength.The spectral lines that showed the highest emission intensity and a Gaussian profile were utilized. The spectral lines thus selected for end-on measure- ment agreed with those for side-on measurement. Real-time background correction was carried out when wavelength modulation and second-derivative detection were used. The wavelength was modulated sinusoidally by a quartz refractor plate which was positioned behind the entrance slit. The second harmonic of the modulated signal was detected by a built-in lock-in amplifier. The wavelength modulation width for each element was selected according to the wavelength of emission and plasma background interferences. Minimum background intensity at one side of a spectral line was sub- tracted from the emission intensity for background c~rrection.~ An auto-tuning system was used for impedance matching.Torch System Movement of the ICP torch assembly between end-on and side-on measurements was computer controlled. Schematic diagrams and dimensions for end-on and side-on measurements are shown in Fig. 1. A Fassel-type torch with a bonnet and a cross-flow nebulizer were used. Flow rates of the Ar outer gas intermediate gas and cut-off stream (end-on measurement) were 15 1.1 and 10 dm3 min-' respectively. Sample uptake rate was 3.0 cm3 min-'. The shape and position of the drain tubing from a Scott-type spray chamber were modified to allow easy drainage in end-on and side-on measurements. The distance between the torch and entrance slit was 550mm in side-on measurement and this was adjusted to between 550 and 610 mm for end-on measurement.With end-on measure- ment the central axial region of the ICP was projected onto the entrance slit and fine lateral adjustment was made in order to locate the best ratio of signal-to-standard deviation of background intensity (S/o) for the analyte. The S/a was used instead of signal-to-background ratio (SIB) because the back- ground was corrected with real-time background correction and became almost zero and the noise contribution became large.g The o-data acquisition had been programmed into the software of the computer system. A Cu cone was positioned between the ICP and the focal lens in end-on measurement.752 Table 1 JOURNAL OF ANALYTICAL ATOMIC SPECTROMETRY JULY 1994 VOL. 9 Operating parameters for end-on and side-on measurements for each element Element Zn Sb Cd c o As Ni B Mn Fe Cr Mg V Ga Be Ca c u Ag Mo A1 Sr U Ba Na K Position/mm Modulation width/ R.f.power/kW Carrier gas/dm3 min- nm Spectral End-on Side-on ine/nm I 213.856*,7,$ 1217.581* I1 226.502*,7 I1 228.616 1228.812*,? I1 231.6047 1249.7732 I1 257.610*,7,$ I1 259.940* I1 267.716* I1 279.553* I1 292.403 1294.3643 I1 313.042$ I1 317.933 I 324.754*,7,$ 1328.0681 I 379.825 I 396.153*,7 I1 407.771 I1 409.014 I1 455.404*,$ 1588.995* I 766.491 End-on 1.4 1.5 1.5 1.5 1.1 1.3 1.4 1.5 1.4 1.3 1.3 1.3 1.3 1.5 1.5 1.3 1.4 0.8 1.3 0.9 0.8 1.3 1 .o 1.4 Side-on 1.4 1.3 1.2 1.4 1.4 1.5 1.5 1.3 1.5 1.5 1.2 1.4 1.3 1 .o 1.4 1 .o 1.3 0.6 1.1 1 .o 0.9 1 .o 1 .o 1.2 distance 590 590 590 590 610 580 600 580 580 600 600 590 600 590 600 590 580 560 610 600 590 590 5 80 610 height 6 11 10 10 8 11 5 11 11 8 9 9 11 7 11 10 9 20 12 17 17 10 11 6 End-on 0.35 0.42 0.35 0.35 0.42 0.40 0.45 0.40 0.39 0.35 0.38 0.36 0.35 0.38 0.35 0.35 0.36 0.40 0.40 0.40 0.40 0.35 0.36 0.40 Side-on 0.40 0.37 0.40 0.36 0.40 0.39 0.40 0.35 0.40 0.40 0.40 0.35 0.33 0.40 0.40 0.35 0.40 0.39 0.40 0.38 0.40 0.40 0.40 0.40 End-on 1.70 1.36 1.26 1.26 1.44 1.66 1.36 1.20 1.40 1.84 1.89 1.76 1.80 1.92 2.16 1.96 1.47 1.96 1.74 2.40 2.10 2.34 2.97 4.00 Side-on 1.02 1.36 1.26 1.08 1.62 1.85 1.95 1.20 1.40 1.84 1.89 1.98 1.80 1.68 1.44 2.20 1.96 2.24 2.03 2.40 3.00 2.68 3.40 3.50 * Spectral line used by Demers.’ 7 Spectral line used by Kawaguchi et a1.’ $ Spectral line used by de Loos-Vollebregt et aL3 320 mm- 260 m - 150 mm I ( b ) 8- Fig.1 Schematic diagrams of (a) end-on and (b) side-on measure- ments 1 cross-flow nebulizer; 2 carrier-gas tubing; 3 sample-uptake tubing; 4 drain tubing; 5 ICP torch; 6 focal lens; 7 entrance slit; 8 echelle spectrometer; 9 Ar cut-off stream; 10 cooling water; and 1 1 Cu cone Cooling water was circulated through the cone to protect it from over-heating. An Ar cut-off stream was introduced into the cone and passed through the nozzle (6 mm diameter) of the cone against the tip of the horizontal ICP to protect the spectrometer optics from the high temperature and deposits from the sample aerosol. Reagents and Solutions Hydrochloric acid nitric acid phosphoric acid sodium chlor- ide and potassium chloride were of analytical-reagent grade.Distilled water was used throughout. Stock solutions (1 g dm-3) of diverse elements were prepared from high-purity metals or compounds (> 99.99% Johnson Matthey Materials Technology Royston Hertfordshire UK) by dissolving them in minimum amounts of hydrochloric acid and diluting with hydrochloric acid (1 + 100). Standard solutions were prepared from the stock solutions by serial dilutions. Antimony was dissolved in hydrochloric acid (1 + 4) and diluted with hydro- chloric acid of the same concentration. Silver was dissolved in nitric acid ( 1 + 1 ) and diluted with nitric acid (1 + 100). Analyte concentrations of 0.10 and 1.0 mg dm-3 respectively were used for the comparative study of end-on and side-on measure- ments according to the sensitivity unless otherwise stated.A certified reference material (CRM) from the National Research Council of Canada NRCC CRM SLRS-2 Riverine Water was analysed. Measurement For the comparative study of end-on and side-on measure- ments the following operating parameters were used unless otherwise indicated r.f. power 1.2 kW; aerosol carrier gas flow rate 0.40 dm3min-’; distance between the torch and the entrance slit (end-on measurement) 600 mm; and wavelength modulation width 80% of the maximum modulation width. Dilute standard solutions of each element were used for instrument calibration when analysing the CRM.JOURNAL OF ANALYTICAL ATOMIC SPECTROMETRY JULY 1994 VOL. 9 753 Results and Discussion Establishment of Optimum Operating Parameters The optimum operating parameters for each element were investigated as follows.Distilled water was nebulized (n = 5) and the standard deviation (a) of the background intensity was calculated. Analyte solution was then nebulized (n = 5) and the average net emission intensity (S) was calculated. The maximum S/a was used as a criterion for selection of the optimum operating parameters. A univariate study was carried out to select the r.f. power flow rates of the Ar gas spectral lines integration time slit-width and height background cor- rection method modulation width distance between the torch and the entrance slit (end-on measurement) and the observation height above load coil (side-on measurement). The optimum operating parameters for end-on and side-on measurements thus selected are summarized in Table 1.Some of the operating parameters will be discussed briefly. It can be seen in Table 1 that the optimum operating param- eters differ little from element to element and between end-on and side-on measurements. The distance between the torch and entrance slit in end-on measurement exhibited a maximum S/a at between 580 and 600 mm for almost all elements. The distance of 600mm was chosen for the comparative study between end-on and side-on measurements to minimize ther- mal damage to the Cu cone. The system allowed for fine adjustment of the plasma torch in the lateral and horizontal directions in order to locate precisely the central axial position. A 10 pg dm-3 solution of Sr and distilled water were nebulized (n = 5 each) and the S/a was calculated. The S/a in the centre of the plasma was typically 5-fold better compared with the response at the extremities.Cut-off Stream Demers' used 50-70 dm3 min-' of air as the cut-off stream. Argon was used as the cut-off stream in the present study to minimize air entrainment and molecular emission. The effect of the flow rate (4-10 dm3 min-') of the Ar cut-off stream was studied on the emission intensities. Although the emission intensities were the same above 6 dm3 min-l 10 dm3 min-' of Ar were used in subsequent studies because 8 drn3min-' or more were necessary to blow the tip of the plasma out through the Cu cone nozzle. Basic Analytical Performance Background intensity Background measurements using distilled water were carried out without real-time background correction. The background intensities during end-on measurement were higher than those in side-on measurement by 1.5- to 17-fold except for Be.The background intensity of Be was lower in end-on measurement. Next the background intensities were measured using real- time background correction. The ratios of background intensit- ies in end-on measurement to those in side-on measurement could be divided into four groups. The ratio of the background intensities was smaller than unity in group I (Be). The back- ground profile showed two small peaks at the shorter and longer wavelength sides of the Be spectral line. The ratios of background intensities in group I1 (Zn Sb Co Ni Mn Fe Cr Mg Ga Ca Mo Al U and Ba) were around unity. The background profiles showed an irregular plateau which could not be identified as a peak.The ratios of background intensities in group I11 (V Ag Sr and K) were slightly higher than unity and those in group IV (Cd As B Cu and Na) were much higher than unity. The background profiles in group I11 showed an obscure peak and those in group IV showed a clear peak. In an attempt to clarify these results profiles around the spectral lines were investigated. Spectral profiles of the elements that showed relatively high background intensities in end-on measurement are shown in Fig. 2(a)-(f). The background peaks (broken lines) were reproducible. For example a strong background peak which could be partially derived from the Ar 324.755 nm line," was found at the longer wavelength side of the Cu I 324.754 nm line in Fig.2(d). There was a back- ground peak which could be derived in part from the relatively strong Ar I 588.859 nm line at the shorter wavelength side of Na I 588.995 nm in Fig. 2(e). Fig. 3(a)-(f) shows the effect of the aerosol carrier gas flow rate in end-on measurement on the background intensities shown in Fig. 2. Two types of experiments were carried out as shown in Fig. 3 to investigate the effects of the Ar line and OH emission. In one experiment 3 cm3 min-l of distilled water (0) were passed through the sample uptake tubing. In another experiment 3 cm3 min-' of Ar (e) which is comparable to the sample uptake rate of water were introduced through the uptake tubing in order not to extinguish the ICP. The background profiles shown in Fig.3 were the same as those in Fig. 2 although the relation- ship between the background intensities and the aerosol carrier gas flow rate differed from line to line. It should be mentioned that interference of the OH emission was greater than that of the Ar line at Cu I 324.754 nm in Fig. 3(d) and Na I 588.995 nm in Fig. 3(e) and vice versa at the As I 228.812 nm line in Fig. 3(b). Contributions of the Ar line and OH emission differed according to the aerosol carrier gas flow rate at the Cd I1 226.502 B I 249.773 and K I 766.491 nm lines in Fig. 3(a) (c) and (f) respectively. Similar but less remarkable behaviour was observed for the background intensities at Cu I 324.754 Na I 588.995 and K I 766.491 nm with side-on measurement. Stability of the background intensity in end-on measurement is an important factor to evaluate.Distilled water was nebulized and the relative standard deviations (RSDs) of the background intensities (n = 10) were compared between end-on and side- on measurements using real-time background correction. Although the RSDs ranged from 9.8% for K to 107% for Be in side-on measurement they ranged from 1.6% for Mg to 36% for Sb in end-on measurement. The RSDs of the back- ground intensities were similarly improved using end-on measurement even if real-time background correction was not used. Net emission intensity ratio Ratios of net emission intensities of the 24 elements were compared for end-on and side-on measurements. Standard solutions (1.0 mg dm-3 for Sb As U and K and 0.10 mg dm-3 for the other elements) and distilled water were nebulized and the net emission intensity was obtained by subtracting the background for water.The ratio values of the net emission intensities for end-on to side-on measurement are shown in Table 2. The ratio values for most elements were from 5- to 35-fold. This improvement is mainly due to the increased light flux in end-on measurement. The ratio value for As was low. This small improvement could be due to the fact that the spectral line of As was relatively weak and that the background intensity was relatively high with end-on measurement. Severe spectral interference due to OH emission hindered the side-on measurement of K. However the increased light flux made the detection of K sensitive in end-on measurement and a high ratio value was obtained.Repeatability Repeatability is defined as the RSD of the net emission intensity after background correction (n = 10). The repeatability for 0.10mgdm-3 solutions of the elements in both end-on and side-on configurations is shown in Table 3. The RSDs for more than half of the elements studied in end-on measurement were less than 3%. However the RSDs for most elements in side- on measurement were greater than 3%. The RSDs €or Sb As,754 JOURNAL OF ANALYTICAL ATOMIC SPECTROMETRY JULY 1994 VOL. 9 226.488 226.502 226.516 r 1 1 5986 0 - Na I 588.961 588.995 589.030 I 416 - A s ' 0 228.826 228.812 228.798 I 1 324.734 324.754 324.774 766.451 766.491 766.531 Wavelengthlnm Fig.2 Spectral profile of background emission in end-on measurement (a) A 0.10mgdm-3 Cd B distilled water; (b) A 1.0mgdmP3 As B distilled water; (c) A 0.10 mg dm-3 B distilled water; (d) A 0.050 mg dm-3 Cu B distilled water; (e) A 0.10 mg dm-3 Na B distilled water; and (f) A 1.0 mg dm-3 K B distilled water U and K in side-on measurement are not shown in Table 3 because the limits of detection (LODs) which will be shown later were higher than 0.10 mg dm-3. Background equivalent concentration The background equivalent concentration (BEC) is defined as the concentration of an element which is equivalent to the background intensity.The BEC values shown in Table 4 were obtained from linear calibration curves without real-time back- ground correction. The BEC values were improved by from 1.4-fold for Mo to 39-fold for Mg by using end-on measure- ment.The slopes of the calibration curves were steeper in end- on measurement than in side-on measurement. Similar studies were performed using real-time background correction and the slopes of the calibration curves were parallel with and without background correction showing that the decreases in BEC values were equivalent to the background correction. Limit of detection The LOD is defined as the concentration equivalent to 3a of the background intensity using real-time background correc- tion. The LODs obtained are shown in Table 5 for the 24 elements using end-on and side-on measurements. The ratio values of the LODs ranged from 1.4-fold for B to 25-fold for U. Demers' obtained 4- to 27-fold improvement in the LODs with end-on measurement The LODs with end-on measure- ment were generally lower than those obtained by Demers' except for Mn Fe Cu Ba and Na.In the study of Kawaguchi et a1.,2 the LODs were 0.7- to 4-fold lower with end-on measurement than with side-on measurement and improve- ment of the LODs was greater at shorter wavelengths.2 However such a tendency is not seen in the results in Table 5. The LODs obtained in the present study were lower than those reported by de Loos-Vollebregt et This was due to high RSDs of the noise in the background intensity in the end- on viewed low-flow torch. The LOD for B was not improved as much with end-on measurement in spite of the higher emission intensity ratio and better repeatability as shown in Tables 2 and 3. This could be due to the fact that a weak background peak [Fig. 2(c)] is located at almost the same wavelength of the B I 249.773 nm line and because of the high background intensity ratio of end- on measurement to side-on measurement.The LOD for K poor in side-on measurement was improved in end-on measurement for the same reason as described under net emission intensity ratio. Comparison of dynamic range Standard solutions of 0-1000 mg dme3 were used to establish the dynamic range for the 24 elements using real-time back- ground correction. The dynamic range was about four orders of magnitude for Zn Co Ni and A1 and about five orders of magnitude for Cd Mn Mg V Cu Ag and Ba in both end-on and side-on measurements but lower concentrations could be quantified with end-on measurement. Boron (five orders of magnitude with end-on measurement uersus seven orders of magnitude with side-on measurement).Fe (four uersus five) Cr (four versus five) and Be (five versus six) showed narrower dynamic ranges with end-on measurement than with side-on measurement. Gallium exhibited the same dynamic range (four orders of magnitude) in both measurement modes. Narrower dynamic ranges were obtained in end-on measurements for Ca (five versus six) and Na (four versus five) and concentrations of about one order of magnitude lower could be quantified with end-on measurement. Antimony (five versus four) As (five versus four) Mo (five uersus four) Sr (six versus five) and U (five versus four) showed wider dynamic ranges with end-on75. JOURNAL OF ANALYTICAL ATOMIC SPECTROMETRY JULY 1994 VOL.9 200 150 100 50 0 - 250 c .- C ' 200 L- 2 c 150 a > C - .% 100 CI 5 50 0 I 1 I I 250 200 150 100 50 0 1000 7 50 500 I u' 10000 7500 5000 2500 2000 f 000 0 (el 0.30 0.35 0.40 0.45 2000 1500 1000 500 0 0.30 0.35 0.40 0.45 Carrier gas flow rate/dm3 min-' Fig. 3 Effect of aerosol carrier gas flow rate on background intensity in end-on measurement (n=3) for (a) Cd I1 226.502k0.0144nm; (b) As I 228.812+0.0144nm; (c) B I 249.773+0.0156nm; ( d ) Cu I 324.754 f0.0196 nm; (e) Na I 588.995 k0.0345 nm; and (f) K I 766.491 f0.0400 nm. Sample-uptake rate is 3 cm3 min-' of 0 distilled water and a. Ar Table 2 Ratio of net emission intensities in end-on and side-on measurements; n = 3 Element Zn Sb Cd c o As Ni B Mn Fe Cr Mg v Ratio* 4.9 29 17 6.2 2.4 7.6 9.1 10 26 33 16 26 Element Ga Be Ca cu Mo A1 Sr U Ba Na K Ag Ratio* 5.4 1.2 20 11 35 21 29 14 31 11 20 103 measurement.This could be due to the increased light flu and decreased RSD in end-on measurement. The observatior that a similar dynamic range was obtained for half of thc elements studied in end-on and side-on measurements coulc be owing to the effect of the Ar cut-off stream. Demers' founc that the linear dynamic range in end-on measurement was generally equal or slightly wider than that in side-on measure- ment. The dynamic range of the calibration curves obtained by Kawaguchi et aL2 was narrower by about one order ol magnitude in end-on measurement than in side-on measure- ment without the cut-off stream. In the study by de Loos- Vollebregt et ~ l . ~ the linear dynamic range in end-on measure- ment was similar to that in side-on measurement.Ionization and chemical interferences Ionization interferences were studied using real-time back- ground correction. Sodium was added at concentrations of up to 10000mgdm-3 as an example of an easily ionized inter- ferent to 1 mg dm-3 of a solution of Ca. The effect of Na on the Ca I 422.673 nm line (lower excitation level) was similar to that seen by Demers' and larger than that by Kawaguchi et uL2 and the effect on the Ca I1 393.367 nm line (higher excitation level) was smaller than that in both of these previous studies'V2 in end-on measurement. The effects of Na on two Cr spectral lines (Fig. 4) and on two Cd spectral lines and the effect of K on an Na spectral line (Fig. 5) were studied with end-on and side-on measure- ments.In Fig. 4 the effect of Na on the relative emission intensity at Cr I 425.435 nm (lower excitation level) was large especially when the r.f. power was high and was small on the emission at Cr I1 284.325 nm (higher excitation level) in end- on measurement. These interferences were larger than those seen by Demers.' The effect of Na was constant on Cr I 425.435 nm up to 10000 mg dm-3 and on Cr I1 284.325 nm up to 100 mg dm-3 in side-on measurement even if the obser- vation height was varied. The effect of Na was not observed on Cd I 228.802 nm (lower excitation level) and Cd I1 226.502 nm (higher excitation level) up to lo00 mg dm-3 of Na against 1 mg dm-3 of Cd in both measurement modes. A slightly positive effect of Na was found on Cd I 228.802nm and Cd I1 226.502nm at 10000mgdm-3 of Na in end-on measurement although Demers' found negative interferences of Na on Cd.In Fig. 5 the large positive effect of K can be seen on Na I 588.995 nm for concentrations of from 100mgdm-3 and greater in end-on measurement and the extent of this effect became larger when the r.f. power increased. The effect of K on Na I 588.995 nm was not observed for up to 1000 mg dm-3 in side-on measurement however some effects were seen at 10000 mg dm-3 depending on the obser- vation height. These effects of K in side-on measurement were different from the observations made by Demers.' The different ionization effects between the present results and those of Demers' and Kawaguchi et aL2 are probably attributable to the Ar cut-off stream real-time background correction and plasma conditions.Chemical interference of phosphoric acid at up to 3% v/v was investigated on the Ca I 422.673 and Ca I1 393.367 nm lines (1 mg dm-3 of Ca) in end-on and side-on measurements. The effect of phosphoric acid on the two Ca spectral lines was similar to the results reported by Kawaguchi et aL2 * End-on to side-on measurement. measurement and the concentrations quantified were one order of magnitude lower with end-on measurement. Concentrations of about three orders of magnitude were obtained as the dynamic range for K with end-on measurement. In general a lower concentration could be quantified for most elements using end-on measurement than side-on Analysis of a Water Reference Material In order to demonstrate the overall reliability of the arrange- ment used for end-on measurements with this system the direct analysis of a certified reference water was performed.As shown in Table 6 good accuracy and precision were obtained for four major elements (Mg Ca Na and K) and six trace elements (Mn Fe Cu Al Sr and Ba). Although it had been difficult to quantify trace amounts of A1 owing to the recombi-756 JOURNAL OF ANALYTICAL ATOMIC SPECTROMETRY JULY 1994 VOL. 9 Table 3 Comparison of repeatability between end-on and side-on measurements for 0.10 mg dm-3 solutions; n= 10 RSD(%) Measurement mode < 1.00 1.00 to 2.99 3.00 to x4.99 5.00 c End-on Side-on Be Sr Mg Be Cu Ag Sr Co B Mn Fe Cr V Ca Zn Cd Na K Cd B Mn Fe Cr V Ca Ag Sb As Ni Ga U Zn Co Ni Ga Mo Al Na Mo Al Ba Mg Cu Ba Table 4 on measurements; n = 5 Background equivalent concentrations for end-on and side- BEC/mg dm-3 0' I I I Element Zn Sb Cd c o As Ni B Mn Fe Cr Mg V Ga Be Ca c u Ag Mo A1 Sr U Ba Na K End-on 0.0053 0.25 0.024 0.044 0.70 0.057 0.035 0.01 1 0.040 0.014 0.0010 0.022 0.16 0.002 0.053 0.0 13 0.027 0.023 0.045 0.0006 0.23 0.0044 0.075 3.2 Side-on 0.027 0.42 0.10 0.11 2.14 0.15 0.42 0.10 0.12 0.10 0.039 0.13 1.9 0.038 0.49 0.18 0.14 0.032 0.56 0.002 3.5 0.044 2.01 12.7 Ratio* 5.1 1.7 4.2 2.5 3.1 2.6 9.1 3.0 7.1 5.9 12 39 12 19 14 9.2 5.2 1.4 3.3 12 15 10 27 4.0 ~~ ~ ~ * Side-on to end-on measurement.Table 5 Comparison of limits of detection between end-on and side- on measurements; n = 10 30 LOD/pg dm-3 Element Zn Sb Cd c o As Ni B Mn Fe Cr Mg V Ga Be Ca c u Mo A1 Sr U Ba Na K Ag End-on 3.0 30 0.81 2.2 4.6 1.6 0.81 4.2 0.87 0.063 0.99 0.075 0.48 1.3 0.81 0.72 2.7 0.018 0.21 3.3 29 16 12 35 Side-on 16 540 23 315 34 6.0 2.2 4.8 5.9 0.30 3.1 0.30 9.9 4.7 5.7 11 95 11 26 300 34 345 0.17 1.4 Ratio* 5.3 18 7.4 10 11 7.4 1.4 5.9 2.6 6.8 4.8 3.1 5.9 4.0 3.6 7.0 9.6 9.4 6.7 9.9 21 15 25 10 * Side-on to end-on measurement.200 1 (b' loo(/--- I 01 I I I I 200 1 ( d ) 0' I I I I 1000 10000 1 10 100 INal/mg dm-3 Fig. 4 Effect of Na on Cr emission intensity (1 mg dme3 of Cr); 0 Cr I 425.435 nm 0 Cr I1 284.325 nm (a) and (b) end-on measure- ment for r.f. powers of 1.5 and 1.3 kW respectively; and (c) and ( d ) side-on measurement for r.f. powers of 1.5 W and observation heights of 20 and 9 mm respectively nation continuum interference of Ca with side-on measurement trace amounts of A1 in the river water CRM were quantified with end-on measurement.This was due to the high dispersion of the echelle spectrometer real-time background correction by wavelength modulation and second-derivative detection and the decreased LOD of A1 in end-on measurement. Conclusions End-on measurement of the ICP was combined with an echelle spectrometer which incorporated wavelength modulation 'and second-derivative detection. This system proved to be more sensitive than side-on measurement although ionization inter- ferences were rather severe. The system can be switched automatically from end-on to side-on measurement and vice versa However the optimum operating parameters for end- on and side-on measurements do not normally coincide.The repeatability of net emission intensity was improved in end-on measurement and the LODs were lower in end-on measure- ment than in side-on measurement. The dynamic range for many elements was almost the same for end-on and side-onJOURNAL OF ANALYTICAL ATOMIC SPECTROMETRY JULY 1994 VOL. 9 .- ri; 200 - 757 * ( b ) 4 Ishii H. and Satoh K. Talanta 1982 29 243. 5 Nakamura Y. and Noto Y. Bunseki Kagaku 1982,31,413. - 1 r n C a Y 4- .- 100 0 2 0 measurements although quantification at lower concentrations was possible with end-on measurements. Multi-element analy- sis of a certified reference water was performed successfully. Further studies are needed for other aspects of this system and applications. The authors thank Kihachiro Murakami of Kyoto Koken for his cooperation.References 1 Demers D. R. Appl. Spectrosc. 1979 33 584. 2 Kawaguchi H. Tanaka T. and Mizuike A. Bunseki Kagaku 1984 33 129. I I I I 200 Ic’ I B l A C 0 I I I I 7 10 100 1000 10000 I KI/mg dm-3 Fig. 5 Effect of K on Na emission intensity (1 mg dm-3 of Na) at Na I 588.995 nm. (a) End-on measurement at r.f. powers A 1.3; and B l.OkW (b) and (c) side-on measurement at r.f. powers of 1.0 and 1.3 kW respectively at observation heights of A 4; B 12; and C 20 mm Table6 Results of river water (NRCC CRM SLRS-2) with end-on measurement; X & (r Element This work* Certified value Concentration/mg dm-3 Mg Ca Na K Zn Sb Cd co As Ni Mn Fe Cr v c u Mo A1 Sr U Ba 1.52 & 0.009 5.67 & 0.09 1.87 & 0.02 0.70 & 0.08 1.51 k0.13 5.70 & 0.13 1.86 k 0.1 1 0.69 k 0.09 Concentration/pg dm - 3 < 6.07 < 1.627 < 58t < 9.27 9.0 & 0.3 127 2 < 1.741- < 1.98-f 2.72 4 0.3 1 83.0 k4.0 27.1 & 0.9 < 247 14.1 f0.2 < < 4.47 < 1.44t 3.33 k0.15 0.26 f 0.05 0.028 f 0.004 0.063 k0.012 0.77 k 0.09 1.03 kO.10 10.1 k0.3 129f7 0.45 & 0.07 0.25 & 0.06 2.76 -t 0.17 0.1 6 f 0.02 84.4 4 3.4 27.3 -t 0.4 0.049 & 0.002 13.8 f 0.3 6 Ishii H. and Satoh K. Talanta 1983 30 111. 7 Nakamura Y. Takahashi K. Kujirai O. and Okochi H. J. Anal. At. Spectrom. 1990 5 501. 8 Xu J. Kawaguchi H. and Mizuike A. Appl. Spectrosc. 1983 37 123. 9 Berman S. S. and McLaren J. W. Appl. Spectrosc. 1978,32 372. 10 Phelps F. M. 111 MIT Wavelength Tables Volume 2 Wavelengths by Element MIT Press Cambridge MA 1982. Paper 3/05 75 1 H Received September 23 1993 Accepted February 24 1994 * n= 10. j- Twice the LOD.

ISSN:0267-9477    

DOI:10.1039/JA9940900751

出版商:RSC    

年代:1994

数据来源: RSC

摘要:

JOURNAL O F ANALYTICAL ATOMIC SPECTROMETRY JULY 1994 VOL. 9 759 Furnace Atomization Plasma Emission Spectrometry at Controlled Pressures Shoji lmai Department of Chemistry Joetsu University of Education Joetsu Niigata 943 Japan Ralph E. Sturgeon* and S. N. Willie National Research Council of Canada Institute for Environmental Chemistry Ottawa Ontario Canada KIA OR9 A furnace atomization plasma emission source was operated in a controlled pressure He environment (200-2000 Torr) (1 Torr=133.322 Pa). The response from several elements (Cd Pb Ag Mn Cu Fe and Co) as well as emission from He I at 388.8 nm and the temperature of the centre electrode (CE) were measured. Below 800Torr the temperature of the CE increases with increasing pressure as does emission from the analytes and He I.This can be attributed to increases in the gas density and collision frequency. Response reached an optimum slightly above atmospheric pressure and declined thereafter. At high pressures (> 1200 Torr) the emission from Cd Pb and Ag again increased probably as a result of an increase in the efficiency of deposition of these analytes onto a cooler CE (secondary site). The effect of pressure on detection limit and linear range of calibration for Cd and Mn was also examined. Keywords Furnace atomization plasma emission spectrometry; centre electrode; deposition efficiency; pressure effect Furnace atomization plasma emission spectrometry (FAPES) is a relatively new atomic emission technique for ultra-trace analysis based on a combined source that features an atmos- pheric pressure r.f.plasma as an excitation medium sustained within a graphite furnace acting as a vaporizer/atomizer.1'2 Furnace atomization non-thermal excitation spectrometry (FANES) operates at reduced pressures [ < 200 Torr ( 1 Torr = 133.322 Pa)]. The effect of this parameter on the excitation characteristics of the plasma have been in~estigated.~ With increasing pressure the mean free path of collision partners decreases resulting in more efficient heating and a loss of highly energetic particles. Several studies of the effect of pressure on response in electrothermal atomic absorption spectrometry (ETAAS) have also been Reduced pressure atomization decreases the mean residence time of the analyte thereby reducing sensitivity. Elevated pressure induces line broadening and lowers the diffusive loss rate.In FAPES atomic emission probably arises via collisions of analyte species with energetic particles such as electrons and metastable He atoms. The effect of pressure on response in FAPES would thus reflect its net effect on both plasma characteristics and on the transport of analyte species in the gas phase. Increasing pressure depletes a fraction of the highly energetic plasma collision partners and increases the frequency of energy exchange. However transport of gaseous analyte in the FAPES workhead is more complex than in ETAAS atomizers due to the presence of the centre electrode (CE). Double analyte emission peaks are frequently observed in FAPES.12-'5 Analyte initially deposited on the furnace wall (primary site) condenses on the CE (secondary site) following thermal desorption from the wall and subsequently re-atomizes from this radiationally- heated s~rface.'~*'~ The largest response is due to the latter process.With volatile elements such as Cd and Pb deposition on this secondary site is limited by its temperature which rises in the presence of the plasma during the pyrolysis stage.I5 This work was undertaken in an effort to study the effect of pressure on response in FAPES. Experiment a1 Apparatus All studies were conducted with a water-cooled integrated contact cuvette (ICC) pyrolytic graphite coated graphite * To whom correspondence should be addressed. NRCC No. 37558 furnace housed within a 10cm vacuum 6-way cross fitted with a feed-through for r.f.power as described previously.16 A coaxial pyrolytic graphite coated 1 rmn diameter graphite CE supported in a Ta holder was used to deliver power from a crystal controlled 13.56 MHz 1500 W r.f. Dionex generator (Model PM 112-1 500). Impedance matching was achieved with a manually adjusted Heathkit antenna tuner (Model SA-2060A Benton Harbor MI USA). The furnace was pow- ered by a Perkin-Elmer Model 2200 supply and fitted for maximum power heating via an optical feedback circuit. The chamber could be evacuated to 25 Torr pressure with a rotary pump and backfilled with high-purity He gas. The FAPES workhead was interfaced to a Spectrometrics Model SMI I11 echelle grating 0.75 m polychromator (Spectrometrics Andover MA USA). The transfer optics for this spectrometer have been described previ~usly.'~ A rectangular vertical slice of the plasma was always viewed with one vertical edge bounded by the edge of the CE the other by the tube wall and both horizontal edges bounded by the upper and lower tube walls.Owing to the radial symmetry of the source viewing of this image permitted a representative fraction of the source to be studied. This could be quickly verified by imaging the entire cross-section of the source onto the large photodiode detector element and observing identical effects of system parameters on measured intensity as when PMT detection of the sub-region was used. All optical components and the FAPES source were aligned with the aid of an 8 mW He-Ne laser. Analyte resonance lines were isolated with the use of appropriate hollow cathode lamps operated in d.c.mode. Atomic absorption transients were measured in the absence of the usual r.f. CE and support and also with the CE coaxially suspended within the ICC via a short piece of 1 mm i.d. stainless-steel tubing in an arrangement that permitted the hollow cathode light beam to pass through and illuminate the interior of the ICC. Photocurrents were fed to a current-to- voltage amplifier having a gain of lo9 digitized with 12-bit resolution and stored to disc using an IBM AT processor. All data manipulations were performed using in-house software written in Turbo Pascal version 4 (Borland International). Temperature measurements of the graphite surfaces were made with both an Ircon Series 1100 (Niles IL USA) auto- matic optical pyrometer and a Thermodot Model TD-6BH (Infrared Industries Santa Barbara CA USA) optical pyrom-760 JOURNAL OF ANALYTICAL ATOMIC SPECTROMETRY JULY 1994 VOL.9 eter. The latter permits focusing to a viewed region < 1 mm in diameter and was calibrated to 1300°C by focusing onto a hole drilled into a graphite block heated in a muffle furnace and the recorded temperature compared with the output from a thermocouple. The lower temperature limit was 120°C. Temperature measurements of the CE were obtained by sight- ing the Thermodot pyrometer through the sample dosing hole of the ICC furnace during the time when the furnace was not being heated. Those of the furnace wall during the atomization cycle were taken with the Ircon pyrometer (blackbody assumed) in the absence of the CE.The temperature of the furnace wall during the pyrolysis stage was measured using a chromel-alumel thermocouple. Pressure measurements in the chamber were made with one or other of two pressure transducers an Ashcroft Transducer Model ASHKlGlOOD7M0242Jl (Cole Parmer Chicago IL USA) and a Setra Systems Model 204 pressure transducer (Acton MA USA). The former was used at high pressure and the latter for low pressure conditions. Reagents High-purity He (Matheson Whitby Ontario Canada) was used as the plasma gas and for purging of the source. Stock solutions of all of the elements were prepared by dissolution of the high-purity metals (Cd Pb Ag Mn Cu Fe and Co) in sub-boiling distilled HN03. Working standards were prepared by dilution of the stocks with deionized distilled water acidified to 1% v/v with HN03.Procedures Both FAPES and AAS measurements were made according to the following procedures. Volumes of sample (5 pl) were pip- etted by hand onto the interior wall of the ICC furnace using an Eppendorf pipette fitted with polypropylene tips. The sample was then dried for 30 s at 80 "C under reduced pressure. Helium was admitted to the chamber to adjust the pressure to the desired level and all gas inlet and outlet valves were closed. The r.f. power was then applied and the plasma ignited spontaneously. Following a further 5 s plasma stabilization period the atomization stage was activated the signal recorded and the r.f. power turned off. Data acquisition commenced with a trigger pulse to the computer commensurate with the beginning of the atomization stage. The pyrolysis and atomiz- ation conditions for each element are summarized in Table 1 along with the wavelengths and heating rates used.Maximum power heating mode was used during atomization. All tempera- tures refer to the pre-set values as read from the front meter panel of the HGA-2200 power supply. The actual measured temperatures of the pyrolysis stage were 86 240 and 330°C for pre-set values of 80 250 and 400"C respectively. Measurements of He I emission at 388.8 nm and the tempera- ture of the CE were made using the same procedure. Table 1 Experimental conditions for analyte atomization Results and Discussion When pressure increases collision frequency as well as plasma gas density increase.This results in a depletion of the fraction of high-energy species due to their more efficient thermal equilibration. The intensity of the triplet He I line at 388.8 nm corresponding to the 3p3P-2s3Sl transition'* is presented in Fig. 1 as a function of pressure at various plasma forward powers. The ICC was maintained at a constant 330°C. As the power increases 4-fold from 20 to 80 W the He I emission intensity increases 60-fold at atmospheric pressure. When the total source intensity is integrated by focusing the image onto a large (5.8 x 5.8 mm) wide-band response (320-1 100 nm) pho- todiode detector (Model S1336-8BKY Hamamatsu Photonics Japan) the output was found to change 138-fold. In a low pressure r.f. plasma electron temperatures are generally inde- pendent of power whereas electron densities increase with power.lg Sturgeon et reported that the He excitation temperature increased marginally from 3000 to 3260 K as the forward power increased from 20 to 80 W in an atmospheric pressure FAPES source.Within the precision of the meas- ured excitation temperatures (i.e. k 180 K) the increase in output intensity could be accounted for by assuming the usual exponential relationship between the excitation temperature and energy i.e. the expected change in the relative emission intensity at 80 and 20 W follows from the estimated change in excitation temperature (260 K) coupled to the energy level involved (23 eV) in an exponential relationship. This is accompanied by a volume expansion of the plasma both radially and longitudinally within the ICC.At low power (< 40 W) emission intensity decreases continuously with increased pressure. This occurs as a result of decreasing excitation energy of collision electrons due to a decrease in mean free path coupled with a shrinkage in plasma volume which eventually leads to extinction of the plasma at pressures 0 400 800 1200 1600 2000 Pressu reflorr Fig. 1 Emission intensity of He I 388.8 nm as a function of pressure at a tube wall temperature of 330°C and plasma forward powers of A 20; B 30; C 40; D 50; E 60; F 70; and G 80 W Element Cd Pb Ag Mn cu Fe c o A/nm 228.8 283.3 328.1 279.5 324.8 248-3 242.5 Pyrolysis temperature*/oC 80 250 400 400 400 400 400 Atomization temperaturet/'C 1100 1 700 2000 2400 2400 2400 2400 Heating rate/"C s-l 1610 1600 1530 1530 1530 1530 1530 * Time.80 s for all elements studied; plasma ignition and stabilization occurs during final 5 s of pyrolysis stage. i Time 4 s for all elements studied.JOURNAL OF ANALYTICAL ATOMIC SPECTROMETRY JULY 1994 VOL. 9 761 of 1000 and 1200 Torr for powers of 20 and 30 W respectively. At higher powers (> 50 W) intensity initially increases with pressure reaches a maximum and then declines. At 50 W a maximum occurs at 800 Torr and the emission declines rapidly above 1000 Torr. Several competitive factors contribute to these observations. Increased pressure increases He plasma gas (emitter) density but decreases the mean free path for electron excitation thereby reducing the high-energy tail of the electron energy distribution function.Additionally as noted above the plasma volume shrinks at high pressure. Even at 80 W power the plasma is eventually extinguished at pressures > 2000 Torr. Under such conditions electron energies become too small to sustain ionization. Alternatively this may be due to a decrease in the efficiency of coupling of r.f. power into the source. Fig. 2 displays the He I line intensity at 388.8 nm as a function of pressure for a 50 W input power (plasma continuously re-tuned as the pressure is changed). Imaging the source with a 10 ym widex25 pm high entrance slit or a 200x300 pm slit shows the same dramatic decrease in intensity with pressure as that observed when the entire source output intensity is integrated by focusing its image onto the large photodiode detector.Unfortunately this measurement cannot account for the losses of power in the source in the form of simple heating of the He gas and the ICC assembly. Although the CE can be seen to cool as the pressure is increased the net IR radiation emitted from the source may be increasing and since He is such an efficient conductor of heat especially at higher pressure the 50 W delivered to the FAPES source may simply be radiated away under conditions that are not energetic enough to support ionization. The temperature of the CE is an important factor in influencing the efficiency of condensation of gaseous molecular and atomic analyte species (such as those of Cd and Pb) onto this ~urface.'~ The CE is heated by both radiation from the wall and collision of excited species from the plasma.The negative bias potential which develops on the electrode in a free running system2' induces further heating by He ion bombardment. The temperature of the CE is presented in Fig. 3 as a function of the pressure at various ICC surface temperatures and a plasma power of 50 W. For any given pyrolysis temperature the electrode temperature initially increases as the pressure rises. Increased frequency of collision of He ions with the electrode surface at high gas density elevates its temperature. After the electrode temperature reaches a maximum in the 800-1000Torr region for a tube temperature of 86 "C 800 Torr for 240 "C and 1000-1200 Torr for 330°C it then declines. At higher tube temperatures the negative bias potential on the CE increases.21 This induces increased He ion bombardment and further heating which is / +4+ B \ 0 400 800 1200 1600 2000 Pressu reflo r r Fig.2 He I intensity at 388.8 nm measured using A 1 0 x 2 5 pm entrance slit; B 200 x 300 pm entrance slit; and C 5.8 x 5.8 mm photo- diode (320-1 100 nm spectral bandwidth) 700 1 1 600 i C I 300 ' I I 1 J 200 600 1000 1400 Pressureflorr Fig.3 Temperature characteristics of the centre electrode in response to changes in pressure for a 50 W plasma. Pyrolysis temperature A 86; B 240; and C 330 "C otherwise masked by the collapse of the plasma at the higher pressures. Fig. 4 illustrates some typical emission transients at various pressures. Signals shown for Fe are representative of those obtained for Mn Cu and Co. Consequently only Cd Pb Ag and Fe were selected for display purposes along with those from the He I 388.8 nm line.When the primary site for deposition of analyte is the tube wall double peaks frequently occur as a result of re-distribution of analyte from the wall to the cooler CE.I5 Because the temperature of the CE is low enough for condensation of molecular oxides of Cd and Pb to occur (cJ Fig. 3) double peaks (resolved and unresolved) are obtained for these elements. In the case of Ag and Fe transfer of atomic vapour from the wall to the CE with their subsequent desorption as this surface heats is responsible for the signal shapes observedi5 for elements such as these. Fig. 5 illustrates the effect of pressure on the peak height and area intensities for these elements.As pressure increases there is an initial increase in both intensities. Whereas one maximum occurs for Fe at 1000 Torr Mn at 1060 Torr Cu at 900Torr and Co at 1000Torr two maxima are obtained for Cd (at 880 and 1340 Torr) Pb (at 930 and 1350 Torr) and Ag (at 760 and 1550 Torr). Under the operating conditions used the temperature of the CE begins to decrease at 1200 Torr for Cd and Pb and at 1100 Torr for Ag (cJ Fig. 3). Thus the observed intensity increase may arise as a consequence of the more efficient deposition of analyte onto the CE as the pressure rises. It is likely that excitation mechanisms involving electron collisions and re-combination dominate in the plasma. The influence of metastable He atoms is probably negligible and their concentration would decrease with pressure as a conse- quence of collisional deactivation. Moreover energy transfer by metastable species exhibits a sharp resonance behaviour restricting the process to a few energy levels and hence elements.Predicting the effect of ambient pressure on response from the analyte is difficult. Measured intensity is proportional to the analyte number density and the frequency of collision with the excitation partner. As pressure increases analyte residence time increases owing to reduced diffusional loss. Concurrently the plasma volume can be seen to shrink thereby increasing r.f. power density (for a constant forward power of 50 W) and hence collision partner (electron) density. To a first approximation these factors lead to a quadratic depen- dence of the analyte emission intensity on applied pressure.However as pressure increases the mean free path decreases and there is a loss in the high-energy tail of the electron energy distribution. This may lead to a decrease in the excitation capabilities of the plasma but in the absence of a defined distribution and detailed rate information one can speculate that this heating process has no impact on the excitation of the relatively low energy levels associated with the transitions762 JOURNAL OF ANAL,YTICAL ATOMIC SPECTROMETRY JULY 1994 VOL. 9 (d) 0 0.5 1 .o 1.5 2.0 D . * 0 0.5 1 .o 1.5 2.0 0 0.5 'I .o 1.5 2.0 Ti rn e/s Fig. 4 Typical signals at various pressures with a 50 W plasma. (a) 1 ng Cd A 300; B 510; C 760; and D 1400 Torr. (b) 1 ng Pb A 300; B 540; C 930; and D 1400Torr.(c) 1 ng Ag A 300; B 570; C 760; and D 1340Torr. ( d ) 5 ng Fe A 300; B 500 C 760; and D 1400Torr. (e) He I A 300; B 760; and C 1400 Torr. Atomization conditions as specified in Table 1; He transient obtained using conditions for Fe t A 200 600 1000 1400 1 1 I I I 200 600 1000 1400 Pressu renorr Fig. 5 Effect of pressure on emission response with a 50 W plasma for (a) Cd 1 ng; (b) Pb 1 ng; (c) Ag 1 ng; and (d) Fe 5 ng. A peak height and B areaJOURNAL OF ANALYTICAL ATOMIC SPECTROMETRY JULY 1994 VOL. 9 763 monitored for these elements. It is clear from Fig. 5 that no linear dependence of the intensity on pressure exists. A quad- ratic dependence on applied pressure is implied by the data presented in Fig. 6 wherein the response for each element has been normalized to the maximum obtained in the pressure range investigated.This may arise as a fortuitous trade-off of numerous factors which could lead to the proposed relation- ship. It is evident from the data in Fig. 4(e) that the intensity of the He I line at 388.8 nm generally decreases as the pressure increases. Since the ‘residence’ t h e of this specie3 is not altered by pressure changes (only emitter density) excitation con- ditions must be changing presumably via the change in the electron energy distribution function. Helium i s more sensitive to changes in this function than the other analytes. Above 800-1000 Torr the data presented in Fig. 6 begin to deviate from linearity. The slopes of the lines for Cd Pb and Ag are higher than that f6r Fe (as well as for Mn Co and Cu which are not displayed).The lower slope for Fe Mn Cu and Co may arise because there is a greater density of excited electronic states available for these elements over which the collisional excitation energy can be distributed. An extensive study of the parameters affecting the rate of loss of atomic vapour from the ICC operating in both the AAS and FAPES modes was undertaken. Rates were calculated from a plot of log(response) uersus time based on data from the decay side of the transient. Linear relationships were obtained and the loss rate data were reproducible within +_ 15% (relative standard deviation from several determi- nations). When the CE is present during atomic absorption measurements no plasma is present. Double peaks which arise due to analyte desorption from both the tube wall and the CE were also observed in AAS when the electrode was present.This has earlier been reported by Hettipathirana and Blades.12 In such case the late peak was used to estimate the loss rate. The following general observations could be drawn. At atmospheric pressure the presence of the CE in the AAS mode decreases the loss rate from the observation volume by 3-6-fold for the volatile elements Cd Pb and Ag and 1.5-2-fold for Mn Cu Fe and Co. The CE serves as a condensation site for atomic vapour and acts as a second surface source for re-desorption similar in operation to that of a L‘vov platform. As the ambient pressure increases the influence of the CE decreases because it is more efficiently heated by conduction and the platform effect diminishes.The rate of loss from the ICC is 2-3-fold greater in the FAPES mode with the plasma present than in the AAS mode with only the CE present. This is probably a consequence of the greater temperature of the diffusion medium (He) in the presence of the plasma as well as possible ionization of the analytes. As the pressure is increased this disparity is reduced because the plasma volume 1.20 W a 0.80 2 t; 0.60 + .- W 0.40 c g 0.20 I I . I 1 0 300 600 900 1200 1500 Pz/lo9 Torr2 Fig. 6 Normalized integrated emission intensity uersus the square of the pressure for A Cd 1 ng; B Pb 1 ng; C Ag 1 ng; and D Fe 5 ng shrinks and the gas cools. When no CE is present for the AAS measurements vapour loss rates for Cd Pb and Ag are less in the FAPES system because of the platform effect offered by the cool CE.For the less volatile elements loss rates in the FAPES system are larger than i s GAS without the CE because the temperature of the dihsion medium (He) is higher when a plasma is present. As the applied p u r e is increased the loss rate for volatile elcmtnts in $’APES tends to increase relative to AAS because t k tm~mfure of the CE increases and condensation efficiency dccmwes. Table2 summarizes data far the dative limit of detection (LOD) and linear range of calkation for Cd and Mn at various pressures. At reduced pressure a decrease in sensitivity and an increase in the LOD are reported for ETAAS systems and the upper limit of the calibration curve is extended.’T6 This occurs as a result of a decrease in the analyte number density coupled with inamwed diffusive f a s ~ mtt.At akvated pressure decreased sensitivity is also noted resulting from absorption line broadening due to the Lorentz effect. However the linear range of the calibration curve is extended 2-6-f0ld.**~ In reduced pressure FAPES there is a decrease in sensitivity and an increase in LOD but no extension to the upper limit of the calibration curve. Decreased sensitivity is likely to be due to decreased collision frequency (excitation rate) decreased analyte number density and redwed residence time. Although both of the latter parameters increase with pressure neither sensitivity and LOD nor linear range of calibration is im- proved. This arises as a result of a decrease in the excitation efficiency as a consequence of shorter mean free paths as well as enhanced self-absorption.Emission sources are typically characterized as having linear dynamic ranges of 4-6 decades. The FAPES source spans no more than 2-4 probably due to self-ab~orpti0n.l~ At high pressure the reduced efficiency of excitation leaves a greater population of ground state analyte atoms both in the ICC source as well as in a cloud between the source and the detector enhancing the self-absorption effect. At low pressure reduced excitation efficiency may occur at high analyte masses when the volume fraction of the analyte vapour approaches 1-2% of that of the He. At 940Torr it is evident that optimum analytical features are achieved reflecting the balance between the competitive factors of collision frequency analyte number density and excitation efficiency.Conclusion The complexity of effects due to pressure in FAPES occurs as a result of the influence of pressure on both the excitation characteristics and on the distribution of gaseous analyte Table 2 Normalized (to 760Torr) LOD and sensitivity and linear range for Cd and Mn Pressure/ Element Torr Cd 300 760 940 1160 1400 2000 Mn 300 760 940 1240 1400 2000 LOD*/pg Relative Log sensitivity? (linear range) 9-1 0 12 3.1 1-0 1 .o 3.7 1.6 1.1 3.9 1.1 0.83 3.9 2-5 082 3-8 1.0 0.99 3.3 1.4 1.1 1.9 1.0 1.0 2.4 0.84 2.2 2.2 0.9 1 1 *4 2.1 0.95 1.4 2.0 0.9 1 1.6 1.8 * Based on signal/s,=3. t Calculated from slope of calibration curve normalized to response at 760 Torr.764 JOURNAL OF ANALYTICAL ATOMIC SPECTROMETRY JULY 1994 VOL.9 species. An optimum pressure for analytical work may be unique for each element and is likely to be slightly higher than atmospheric pressure this being determined by the competition between increasing density longer residence time secondary site adsorption-desorption processes and decreasing excitation energy for collision. A more complete understanding of these observations will require Langmuir probe diagnostics of the plasma and two dimensional imaging of the source. The efficiency of deposition of analyte onto secondary sites is increased with increased pressure and the rate of re-desorption from such surfaces is limited by their decreased temperature at higher pressures. The effect of pressure on the interference by easily ionized elements and the influence of controlled bias potential on performance at various pressures warrant investigation.The authors thank G. Jolly Bell Northern Research Ottawa for the loan of the r.f. generator. S.I. thanks the NRCC for partial financial support while in Ottawa. References Liang D. C. and Blades M. W. Spectrochirn. Acta Part B 1989 44 1059. . Sturgeon R. E. Willie S. N. Luong V. T. Berman S. S. and Dunn J. G. J. Anal. At. Spectrom. 1989 4 669. Falk H. Hoffmann E. and Ludke Ch. Prog. Anal. Spectrosc. 1988 11 417. Donega H. M. and Burgess T. E. Anal. Chem. 1970,42 1521. Hassel D. C. Rettberg T. M. Fort F. A. and Holcombe J. A. Anal. Chem. 1988 60,2680. 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 Wang P. and Holcombe J. A. Spectrochim. Acta Part B 1992 47 1277. Sturgeon R. E. Chakrabarti C. L. and Bertels P. C. Spectrochim. Acta Part B 1977 49 1100. Sturgeon R. E. and Chakrabarti C. L. Anal. Chem. 1977 49 1100. Sturgeon R. E. and Chakrabarti C. L. Prog. Anal. At. Spectrosc. 1978 1 5. Fazakas J. Spectrochim. Acta Part B 1982 37 921 Fazakas J. and Zugravescu P. Gh. Appl. Spectrosc. 1988,42,521. Hettipathirana T. D. and Blades M. W. J. Anal. At. Spectrom. 1992 7 1039. Smith D. L. Liang D. C. and Blades M. W. Spectrochim. Acta Part B 1990 45,493. Sturgeon R. E. Willie S. N. Luong V. T. and Berman S. S. Anal. Chem. 1990 63 2370. Imai S. and Sturgeon R. E. J. Anal. At. Spectrom. 1994 9 493. Sturgeon R. E. Willie S. N. Luong V. T. and Dunn J. G. Appl. Spectrosc. 1991 45 1413. Berman S. S. and McLaren J. W. Appl. Spectrosc. 1978,32 372. Weise W. L. and Martin G. A. Wavelength and Transition Probabilities for Atoms and Atomic Ions US Department of Commerce NSRDS-NBS (US) No. 68 Washington D.C. 1980. Lin I. J. Appl. Phys. 1985 58 2981. Sturgeon R. E. Willie S. N. and Luong V. T. Spectrochim. Acta Part B 1991 46 1021. Sturgeon R. E. Luong V. T. Willie S. N. and Marcus R. K. Spectrochim. Acta Part B 1993 48 893. Paper 3/06496D Received November 1 1993 Accepted March 21 1994

ISSN:0267-9477    

DOI:10.1039/JA9940900759

出版商:RSC    

年代:1994

数据来源: RSC

摘要:

JOURNAL O F ANALYTICAL ATOMIC SPECTROMETRY JULY 1994 VOL. 9 765 Easily Ionized Element Interference Effects in Furnace Atomization Plasma Emission Spectrometry Shoji lmai Department of Chemistry Joetsu University of Education Joetsu Niigata Japan 943 Ralph E. Sturgeon* Institute for Environmental Chemistry National Research Council of Canada Ottawa Ontario Canada KIA OR9 The effects of easily ionized element (EIE) matrices on the response in furnace atomization plasma emission spectrometry is reported. Using NaCl and CsCl as model matrices interference on integrated emission intensity from Pb and Co was studied as a function of EIE mass and source pressure [200-1500 Torr (1 Torr = 133.322 Pa)] in a 50 W He r.f. plasma. Comparative response was measured in the same source using the AAS mode of operation.Radiative power losses from the plasma due to excitation of the EIE matrix species and probable alteration of the electron energy distribution function (due to EIE ionization and collisional dissociation of molecules) appear to be the major sources of interference. Classical analyte molecule formation gas-phase expulsion and de-tuning of the plasma in the presence of large masses of EIE are not significant factors in determining the extent of interference. Keywords Furnace atomization plasma emission spectrometry; interference; easily ionized element; radiat- ive power loss The technique of furnace atomization plasma emission spec- trometry (FAPES) is based on the measurement of emission from analyte species vaporized into and excited by an atmos- pheric pressure r.f.plasma contained within a heated graphite furnace (GF).ls2 Non-spectral interferences are currently prob- lematic with the He-based FAPES source. As with the furnace atomic non-thermal excitation spectroscopy (FANES) tech- nique real sample analyses limited to date rely on the method of additions for quantification due to the interference effects arising from the presence of concomitant element^.^-^ In addition to inducing potential chemical interference effects well-known and characterized in GF atomic absorption spec- trometry (AAS) these species also perturb the discharge characteristics of the plasma thereby altering the excitation of the analyte. In this respect their action can be likened to the effects of easily ionized elements (EIEs) on response in other plasma s o ~ r c e s .~ - ~ ~ Thus Falk14 noted that the presence of 2 pg of NaCl caused a 20% suppression in emission response from Cu in the FANES source. Similarly Smith et aL4 reported an initial slight enhancement in Ag emission in the FAPES source in the presence of NaCl followed by signal suppression beyond 1.2 pg of salt (in a 10 pl sample aliquot). It was suggested that the initial enhancement might arise as a result of suppression of ionization of Ag but that at higher concen- trations the excitation characteristics of the plasma may be altered. Subsequently Hettipathirana and Blades” noted sig- nificant interference from NaCl and NaNO on the response from Pb and Ag in their FAPES source. At the low powers and heating rates employed (14-40 W 360 K s- ’ respectively) problems related to molecular dissociation and changes in plasma excitation characteristics were encountered when these EIEs were present even in trace amounts (i-e.162ng). The presence of 1-3 pg of NaCl resulted in significant signal suppression of Pb although the extent of interference could be decreased as the forward power to the plasma was increased. Interference by NaCl was also reported by Gilchrist and Liang7 who observed suppression of TI emission intensity in the presence of only 0.08 pg of salt vaporized in an integrated contact cuvette (1CC)-based FAPES source. Platform technology and chemical modification can be used with the FAPES source to alleviate these problems but not * To whom correspondence should be addressed. without impunity.6 The masses of any modifiers must be minimized otherwise there is a tendency for arcing to occur between the centre electrode (CE) and the deposition site of the sample on the furnace wall.This situation can be circum- vented when using a platform but in the Massmann furnace the limitations of this device for use with involatile analytes is well-known. The nearly isothermal system based on the (ICC) should be more suitable for operation with a platform. Additionally being an emission technique the probability of line overlap from the modifier or an impurity element in the modifier is of concern and may prohibit the use of specific modifier-element combinations. Nevertheless use of Pd and atomization from a platform was found to significantly decrease the degree of interference suffered by Pb in the presence of NaCl such that no effect on the recovery of the integrated signal occurred for up to 24 pg NaCL6 The presence of EIEs alters the discharge characteristics of the plasma as evidenced by the fluctuations noted in the self- bias voltage during vaporization of Na.I6 The magnitude of the reflected power transient however is not altered even during the atomization of up to 60pg of NaC1.16 Thus interferences do not arise as a consequence of decreased power delivered to the source and hence other factors must be sought.As is common for tandem sources the extent of interference by EIEs is proportional to the degree of temporal overlap of the gas-phase populations of analyte and interferent in the observation volume which is related to the relative volatiliz- ation temperatures and residence times of the species con~idered.~.~~ The influence of EIEs on emission response in the FAPES source is of significant practical relevance since such matrices are encountered in many environmental samples.This work was undertaken in an effort to gain further insight into this problem by examining the effect of NaCl CsCl NaNO and SeOz on the emission response from Pb Co and He over a range of ambient pressure forward r.f. power and salt matrix concentration. Experimental Apparatus The FAPES source is comprised of a water cooled ICC pyrolytic graphite-coated graphite furnace housed within a 1Ocm vacuum stainless steel 6-way cross fitted with a feed-766 JOURNAL OF ANALYTICAL ATOMIC SPECTROMETRY JULY 1994 VOL.9 through for r.f. power.17 A coaxial pyrolytic graphite coated 1 mm diameter graphite CE supported in a Ta holder was used to deliver power from a crystal controlled 13.56MHz 1500 W r.f. Dionex generator (Model PM 112-1500). Impedance matching was achieved with a manually adjusted Heathkit antenna tuner (Model SA-2060A Benton Harbor MI USA). The furnace was powered by a Perkin-Elmer Model 2200 supply and fitted for maximum power heating via an optical feedback circuit. The chamber could be evacuated to 25 Torr pressure (1 Torr = 133.322 Pa) with a rotary pump and backfilled with high-purity He gas. The FAPES workhead was interfaced to a Spectrometrics Model SMI I11 Cchelle grating 0.75 m polychromator (Spectrometrics Andover MA USA).All optical components and the FAPES source were aligned with the aid of an 8 mW He-Ne laser. Analyte resonance lines were isolated with the use of appropriate hollow cathode lamps operated in d.c. mode. Atomic absorption transients were measured in the absence of the usual r.f. CE and support and also with the CE coaxially suspended within the ICC via a short piece of 1 mm i.d. stainless-steel tubing in an arrangement that permitted the hollow cathode light beam to illuminate the interior of and to pass through the ICC. Photocurrents were fed to a current-to-voltage amplifier having a gain of lo9 digitized with 12-bit resolution and stored to disc using an IBM AT processor. All data manipulations were performed using in-house software written in Turbo Pascal version 4 (Borland International Scotts Valley CA USA).Pressure measurements in the chamber were made with one or other of two pressure transducers an Ashcroft Transducer Model ASHKlGlOOD7M0242Jl (Cole Parmer Chicago IL USA) and a Model 204 pressure transducer (Setra Systems Acton MA USA). The former was used at high pressure and the latter for low pressure conditions. Reagents High-purity He (Matheson Gas Products Whitby Ontario Canada) was used as the plasma gas and for purging of the source. Stock solutions of Pb and Co were prepared by dissolution of the high-purity metals in sub-boiling distilled HNO,. Working standards were prepared by dilution of the stock solutions with de-ionized distilled water acidified to 1 % v/v with HNO,.The National Research Council Canada (NRCC) Nearshore Seawater Reference Material for trace metals CASS-2 was used as the source of the sodium chloride matrix. Matrix solutions of CsCl and SeO were prepared from the commercially available salts. Procedures Both FAPES and AAS measurements were carried out accord- ing to the following procedures. Volumes of sample (5 pl) were pipetted by hand onto the furnace wall using an Eppendorf pipette fitted with polyethylene tips. Matrix solutions were mixed in the furnace. Samples were dried for 30 s at 80 "C (measured) under reduced pressure. Following a 30 s 'char' stage at reduced pressure He was admitted to the chamber and the pressure was adjusted to the desired value. Radio- frequency power was then applied (50 W) and the plasma spontaneously ignited.Following a further 5 s plasma stabiliz- ation period the atomization stage was activated the signal was recorded and the r.f. power was turned off. A reagent (matrix) blank run was subtracted from all signals to compen- sate for the effect of any added impurities in the EIE matrix. The atomization conditions used for each experiment are summarized in Table 1. Data acquisition commenced with a trigger start pulse commensurate with the beginning of the atomization stage. Maximum power heating mode was used in all cases. All temperature values refer to the pre-set tempera- tures as read from the front panel of the HGA-2200 power supply. Results and Discussion When there is a significant temporal overlap of the vapour populations of the EIE matrix (molecular or elemental) and the analyte released into the GF several factors may be responsible for the observed interferences (both enhancements and depressions of analyte intensity); those factors which affect the residence time atomization efficiency degree of ionization and excitation efficiency of the analyte atoms have an impact on the net response.These factors include expulsion loss of the analyte from the excitation/observation volume due to co-vaporization of the EIE matrix; (classical GF) chemical interference due to analyte molecule formation; decrease of plasma power available for analyte excitation due to photon emission uia excited matrix species; loss of plasma power uia dissociation of matrix molecules; alteration of the power coupling efficiency to the plasma (ie.de-tuning) as a conse- quence of changes in the load impedance in the presence of the EIE; alteration of the self-bias voltage of the system in response to changes in electron density in the presence of the EIE; increased gas density of the diffusion medium (a significant amount of He is replaced by the matrix vapour); shifts in the ionization equilibrium of the analyte species due to ionization of matrix element(s); decreased gas kinetic temperature as a result of the EIE-induced power dissipation losses mentioned above; and alteration of the electron energy distribution func- tion (EEDF) as a consequence of the release of electrons from the EIE through ionization or attenuation of plasma electrons through molecular dissociation and matrix species excitation.The first six factors lead to a decrease in analyte emission the subsequent three to an increase and the last factor may either decrease response or have little influence depending on the excitation energy of the analyte transitions relative to the original and altered EEDF. It is clear from earlier studies that for a non-autotuned system the vaporization of large amounts of EIE (i.e. 60 pg of NaCl) into the plasma does not alter the reflected power characteristics of the system16 and lead to additional de-coupling of the source from the generator. This factor may thus be discounted as a possible mechanism responsible for interference. As analyte emission signals are integrated the effect on response of any change in analyte release kinetics arising from the presence of the EIE matrix will be minimized.Classical GF chemical interference effects caused by the forma- tion of analyte molecules with the co-vaporizing matrix species can be assessed through comparative measurement of the system response in the AAS mode in the absence of the plasma. Unfortunately measurement of the d.c. bias voltage character- istics of the CE could not be made in the present system and the effects of variations in this parameter due to the presence of EIEs were not assessed in this study. Lead and Co were selected as examples of analytes covering a range of volatilities and likely to produce atomic vapour clouds which are temporally overlapped with those of NaCl and/or CsC1. Both NaCl and CsCl serve as pertinent EIE matrices as they permit the study of the influence of the ionization potential of the matrix cation on analyte response and Na is so pervasive in the environment.Additionally SeO was chosen as an alternative concomitant in order to examine the effect of a non-EIE matrix on analyte response. The effect of EIEs on emission lines arising from the metastable states of the He plasma support gas was also investigated. It is assumed throughout that the density and energy distribution of free electrons determine the excitation conditions in the plasma all other excitation processes being ultimately influenced by e1ectr0ns.l~ Plasma electrons lose their energy by excitation or by ionizing collisions with support gas analyte and matrix species. In addition to fast electrons excited He atoms may play a role in the excitation of sample atoms.The metastable states of He (at 19.77 and 20.55 eV) dissipate their surplus energy primarily uia collisions and not by radiative relaxation.JOURNAL OF ANALYTICAL ATOMIC SPECTROMETRYy JULY 1994 VOL. 9 767 Table 1 Experimental conditions Pyrol yse Atomize Element Ilnm TemperaturePC Time*/s TemperaturePC Time/s Heating ratePC s-' Pb 283.3 250 80 2400 1 1530 Pbt 283.3 250 80 1700 4 1600 c o 242.5 400 80 2400 4 1530 * Plasma stabilization period during final 5 s of the pyrolysis stage. CE-AAS. Szilvassy-Vamos et al." have summarized a number of poten- tial energy exchange reactions. Excitation of Pb and Co must be due to electron collisions since the lowest excited levels of He are too far above even the ionization potentials of these elements for any direct interactions to be likely.'* Fig.1 illustrates the effect of vaporization of NaCl and SeO into the plasma on the emission from He I at 388.8 nm during a typical transient atomization event (atomization conditions as for Pb cfi Table 1). Temporal response from He I in the presence of these matrices has been 'normalized' to that from He I in a 'clean' plasma (taken as zero emission and represented by the horizontal line). It is clear that emission intensity from the excited state decreases in direct response to the presence of the matrix species in the plasma. This observation clearly indicates a decrease in energy in the plasma. Two distinct desorption events are also noted for both SeO and NaCl the latter being characterized by measurement of non-specific absorption (attributable to absorption by molecular NaCl) at the Co I 242.5 nm line.They arise as a result of early desorption of matrix species from the graphite tube wall followed by a second subsequent release from the CE.19 With the masses used in this study significant amounts of matrix vapour species reside within the ICC during the entire atomization transient (as evidenced by simple indirect observation of Na emission with the unaided eye) and are not temporally limited to the two major introductory events illustrated in Fig. 1. As a consequence perturbation to the He I intensity occurs through- out the entire atomization cycle but is particularly severe during both periods of rapid introduction of matrix into the plasma.Decreased response for He I may be due to several factors assuming that excitation of He occurs as a result of collision with high-energy electrons (> 23.0 eV) molecular vapour arising from the desorption of large amounts of these I I I I 1 0 0.5 1 .o 1.5 2.0 Time/s Fig.1 (a) A continuum background absorption at Co I line (242.5 nm) from atomization of 30 pg of NaCl. Atomization conditions as for Pb (Table 1). By Atomic emission by Se I at 196.0 nm from atomization of 19 pg of SeO (displayed as a 'negative' intensity). (b) Effect of NaCl and SeO on emission transients for He I line at 388.8 nm in a 50 W plasma. Data normalized to response obtained in a clean plasma (horizontal line) C 0.5 pg of NaCl atomized; D 5.0 pg of NaCl atomized; and E 19 pg of SeO atomized matrices (ie.NaCl and SeO and SeO),' will undergo plasma induced dissociation via impact with fast electrons severely attenuating this population and reducing the rate of excitation of He. The Na-C1 and Se-0 bond strengths are not signifi- cantly different (4.25 and 4.38 eV respectively) and an equal number of moles of each material was vaporized. If dissociation cross-sections are similar with respect to electron-impact energy their effect should be similar. The result however is more dramatic for NaC1. The high ionization potential of Se (9.75 eV) relative to Na (5.14 eV) is a likely factor contributing to this difference. Ionization of Na floods the plasma with low energy (secondary) electrons as well as further depleting the plasma of high energy primary electrons via a collisional ionization process.Additionally excitation of resonance states of Na is considerably more efficient than for Se as a result of the significantly higher excitation energies of the latter.21 The consequences include a further shift in the EEDF to lower energies and an increase in the power radiated from the plasma in the form of photon loss as a result of excitation of the atomic sodium v a p o ~ r . ' ~ FalkI4 has estimated the magni- tude of this effect by considering only excitation of the Na 589 nm doublet. A 10% radiative power loss occurs when Na constitutes as little as 0.01% of the plasma gas in a 50 W source. Assuming that the NaCl atomizes as a plug with no losses introduction of 30 1-18 of matrix would produce a relative gas-phase composition of 30% Na thereby causing substantial radiative power loss and a decrease in the He I intensity. Such power losses and changes in the EEDF will have an impact on the excitation of analytes in the plasma.In addition to the above large amounts of EIEs can alter the conductivity of the gas phase thereby increasing the current and collapsing voltage gradients to such low values that the extent of ionization does not produce fast electrons which further excite species with a resultant shift of the bulk electron populations to lower energy values. Fig. 2 illustrates the effects of added EIE matrix components on the transient response from Pb and Co analytes. Evidence for desorption of both analytes from two distinct surface sites is clear.Initial desorption of molecular and atomic Pb species from the wall of the furnace tube (primary site) is accompanied by their condensation on the cooler CE which with subsequent radiative and plasma induced heating re-desorbs Pb vapour (secondary site). A similar situation arises for Co wherein atomic vapour released from the tube wall condenses on and is subsequently released from the CE.19 An early shift in the signal for Pb is obtained in the presence of the chloride matrices. A slight enhancement in response occurs when 0.5 pg (8.5 nmol) of NaCl is added. If 8.5 nmol of NaNO is substituted for the NaCl no shift in temporal response nor alteration of emission intensity from Pb is observed. The effect of the anion is thus clear early vaporiz- ation in the presence of small amounts of NaCl may be due to volatilization of PbCl from both the primary and secondary deposition sites.The Pb-Cl bond energy (3.1 eV) is lower than that of Pb-0 (3.9 eV) and enhanced response may occur as a consequence of more extensive plasma induced dissociation of the chloride molecule. Molecular dissociation of analyte mol- ecules in the FANES plasma has been established3 but the extent of such processes occurring in the higher pressure768 JOURNAL OF ANALYTICAL ATOMIC SPECTROMETRY JULY 1994 VOL. 9 1 r I- \ CO 0.5 1 .o 1.5 2.0 0 0.5 1 .o 1.5 2.0 0 1 2 3 0 1 2 3 Fig. 2 Effect of added EIEs on temporal response from Pb and Co. (a) Atomic emission from 2 ng of Pb at 283.3 nm A 0 pg NaC1; B 0.5 pg NaC1; C 10 pg NaC1; D 0.73 pg NaNO (equivalent to 0.5 pg NaC1); and E continuum background absorption by 30 pg of NaCl at 242.5 nm.(b) Atomic emission from 2 ng of Pb at 283.3 nm A 0 pg CsCl; B 2.0 pg CsC1; C 20 pg CsC1; D continuum background absorption by 50 pg CsCl; and E atomic emission at 196.0 nm from 19 pg SeO (displayed as a ‘negative’ intensity). (c) Atomic emission from 10 ng of Co at 242.5 nm A 0 pg NaCl; B 0.5 pg NaC1; C 5.0 pg NaC1; D 50 pg NaCl; and E continuum background absorption by 50 pg NaC1. (d) Atomic emission from 10 ng of Co at 242.5 nm A 0 pg CsCl; B 0.5 pg CsC1; C 5.0 pg CsC1; D 50 pg CsCl; and E continuum background absorption by 50 pg CsCl FAPES source is unknown. With increasing mass of EIE matrix there is a general shift to later times for the desorption of both Pb and Co from the secondary site (CE).This might be the consequence of increased radiative power losses from the plasma which results in a cooling of the CE (whose temperature is proportional to the r.f. forward power).lg No such shifts are observed with the addition of SeO (whose vapour population overlaps that of the analytes and of CsCl) lending support to the interpretation given above. Although vaporization of the EIE matrix produces a decreased response for He I it cannot be concluded that the EEDF if altered is changed to such an extent as to have a severe impact on the excitation of these analytes which have significantly lower excitation energies than He. Further insight into the factors responsible for the inter- ference effect by EIEs can be gained by examining their concentration dependent effects.Fig. 3 shows integrated signal intensities for Pb and Co as a function of added mass of NaCl CsCl and SeO as well as the response from the analytes in the AAS mode with the CE present. The cooler CE plays an important role in the AAS mode. Fig. 4 shows that the AAS response from Pb is free of inter- ference over a much greater range of added NaCl (20-fold) when the CE is present which functions as and possesses the attributes and benefits of a platform. The range of interference free AAS operation was also enhanced for Co in the presence of the CE. More significant however are the data in Fig. 3 which show that interference from the EIE matrix occurs at substantially lower concentrations in the FAPES mode of operation for both analytes (50-fold for Pb; 10-fold for Co).It can be concluded from the above comparison that the effect of the EIE in FAPES is neither due to classical analyte molecule formation of PbCl and CoCl nor to their loss by expulsion with the rapidly expanding gas cloud produced by the desorption of the matrix and must therefore be associated with processes affecting the plasma. This is also particularly clear in the case of Pb when SeO is used as the matrix; the effects of the EIEs become evident at much lower concen- trations. Both NaCl and CsCl flood the plasma with low energy electrons via ionization of the metals. The vapour populations of the two matrices substantially overlap that of Pb [c$ Fig. 2(4 curve E and Fig. 2(b) curve D].Both matrices appear to affect response equally suggesting that for these analytes (and for these analyte resonance lines) ionization of the EIE is not the primary cause of the problem otherwise the effect of CsCl should be evident at much lower concen- trations than for NaCl. The observed interference is probably due to the loss of plasma power radiated from the observation volume as photons from the excitation of Cs and Na,14 as discussed earlier with respect to He I emission. At sufficiently high concentrations of all matrices additional loss of power arises as a result of electron impact dissociation of molecular matrix species. For constant forward power to the plasma a fixed amount of EIE radiates a fixed amount of power. As forward power increases plasma electron density increases enhancing exci- tation rates (assuming no changes in the EEDF) and hence radiative power loss as well. In such circumstances it is to be expected that the relative interference by an EIE is independant of the forward power.However there is a decrease in the degree of interference as the forward power is increased as is illustrated by the data presented in Table 2. A 3-fold increase in forward power reduces the degree of interference from 10 pg of NaCl on Pb 2-fold. For Co increased power raises theJOURNAL OF ANALYTICAL ATOMIC SPECTROMETRY JULY 1994 VOL. 9 769 t ( a ) . z I I I (b) 0.5 1 10 1 x 102 1x103 dXio3 Matrixhmol Fig. 3 Effect of added matrix on (linear) integrated atomic emission and absorption for (a) 2ng of Pb and (b) long of Co for CE-AAS mode V NaCl; and FAPES mode + SeO,; A NaCl; and 0 CsCl ( a ) I ._ t > i- .- g o 0 ‘ ‘ I I I I 0.05 0.1 1 10 100 400 NaCI/pg Fig. 4 Effect of added NaCl on (linear) integrated response for (a) Pb and (b) Co. (a) A 2 ng NaCl AAS mode; + 4 ng NaC1 CE-AAS mode; and 0 2 ng NaCl FAPES mode. (b) A 1 ng NaCl AAS mode; + 5 ng NaC1 CE-AAS mode; and e 10 ng NaCl FAPES mode relative response beyond unity probably due to the suppression of ionization of Co in the presence of the ionized Na. These data suggest that the changes in the EEDF which occur as the forward power increases have greater impact on the inter- ference system. The concentration of electrons released from Table 2 Effect of r.f. power on recovery of integrated intensity Recovery* (YO) Power/W 50 75 100 150 Pb 0.36 0.38 0.4 1 0.71 c o 0.51 0.8 1 1.26 1.49 * Net relative signals withiwithout 10 pg of NaCl present.the EIEs via ionization is reduced as a result of the higher intrinsic electron density in the higher power plasma. In turn the consequences of this added ‘buffering’ capacity are such as to reduce the impact of the presence of the EIE. The results of a more detailed study of the influence of EIEs on the loss rate of analyte vapour from the absorption and excitation volumes of the FAPES source operated in both AAS and AES modes are presented in Fig. 5. Addition of CsCl produced effects similar to those for NaCl and are thus not reproduced here. Loss rates were calculated from plots of log(response) versus time based on data from the decay side of the signal transients.Linear relationships were obtained the slopes of which were reproducible within 5% (relative standard deviation for several determinations). In all cases loss of analyte vapour is faster from the FAPES source when the r.f. plasma is on than when it is used in the AAS mode with the CE present (3-fold for Pb; 5-fold for Co). This is a consequence of the higher gas kinetic temperature of the diffusion medium with the plasma present the difference being more notable for Pb which desorbs at a lower ICC temperature than Co. Addition of SeO to the system imparts little change to the rate of loss of analyte suggesting that there is neither significant cooling of the gas phase change in the composition of the diffusion medium increased physical expulsion nor chemical loss occurring for either Pb or Co despite vapour cloud overlaps.The presence of NaCl and CsCl has an insignificant effect on the rate of loss of Pb but these matrices decrease the rate of loss of Co by nearly 8-fold (at 50 pg mass). The boiling-point of CsCl is 1563 K and that of Cs is 951 K. If it is assumed that all of the CsCl in a 50 pg mass is transferred into the ICC as a plug of vapour the equilibrium vapour pressure of CsCl would be achieved at 2000K (as would that for Cs assuming complete dissociation). In any case the volume of the ICC would be filled and Cs and/or CsCl would displace the He to produce a Cs ‘plasma’. This would dramatically decrease the excitation properties of the source and in the most acute case alter the diffusional loss rate of Co from the observation volume by a factor of 5.7.Such a scenario is however highly unlikely based on the relative population overlaps shown in Fig. 2(4. Similarly the temperature of the diffusion medium would have to fall to 700 K (assuming a T3 dependence) for this factor alone to account for the %fold decrease in the rate of loss of Co from the observation volume in the presence of this amount of matrix. In combination these factors could be responsible for the decreased loss rates. Additionally consideration should be given to suppression of the rate of loss of analyte through ionization in the presence of the EIE. This would also be consistent with the benign effects noted for Se02. Figs. 6 and 7 illustrate the effects of EIE mass and source pressure on the integrated response from Pb and Co in both the FAPES emission and ICC-AAS modes of operation.Varying the source pressure permits changes to be made not only in the extent of overlap of analyte and EIE matrix vapour clouds but also in the excitation characteristics of the plasma.23 Several features of these figures merit discussion. As the source pressure is increased there is a corresponding increase in the intensity of integrated emission and absorption primarily due to an increase in analyte residence time. As ambient pressure770 JOURNAL OF ANALYTICAL ATOMIC SPECTROMETRY JULY 1934 VOL. 9 0 -0.20 -0.40 2 3 5 1 0 rn - 0 -0.60 L -0 -2- - 0 -0.38 -0.75 -1.13 - 1.50 0.10 0.20 0.30 0.40 0.50 0.60 I I 0 0.10 0.20 0.30 0 0.10 0.20 0.30 Time/s Fig.5 Log(response)-time plots illustrating effect of added SeO and NaCl on rate of loss of atomic Pb and Co from the observation volume. (a) 2 ng P b + 0 pg SeO (FAPES); 0,2.5 pg SeO (FAPES); A 25 pg SeO (FAPES); + ,250 pg SeO (FAPES); and 0 O pg SeO (CE-AAS mode). (b) 2 ng Pb; + 0 pg NaCl (FAPES); 0,0.5 pg NaCl (FAPES); A 5.0 pg NaCl (FAPES); + 50 pg NaCl (FAPES); and 0 0 pg NaCl (CE-AAS mode). (c) 10 ng Co + 0 pg SeO (FAPES); 0 2.5 pg SeO (FAPES); A 25 pg SeO (FAPES); + 250 pg SeO (FAPES); and 0 0 pg SeO (8 ng Co CE-AAS mode). ( d ) 10 ng Co + 0 pg NaCl (FAPES); a 0.5 pg NaCl (FAPES); A 5.0 pg NaCl (FAPES); + 50 pg NaCl (FAPES); and 0 O pg NaCl(8 ng Co CE-AAS mode) C m 0.05 0.1 1 10 100 400 NaCI/pg Fig. 6 Effect of added NaCl on integrated response from Pb and Co at various pressures.(a) 2ng Pb. FAPES mode 0 200 Torr; A 760 Torr; and M 1500 Torr. CE-AAS mode + 200 Torr; and V 760Torr. (b) long Co. FAPES mode 0 200Torr; A 760Torr; and H 1500Torr. AAS mode 0 200Torr; A 760Torr; and 0 1500 Torr I I 1 1 1 1 x 1 0 ~ 1x103 4x103 10 0.5 1 Mat rix/n mo I Fig. 7 Effect of EIEs on integrated emission intensity (FAPES mode) from Pb and Co at various pressures. (a) 2 ng Pb-NaC1 0 200 Torr; A 760Torr; and H 1500Torr. 2ng Pb-CsC1 0 200Torr; A 760Torr; and 0 1500Torr. (b) long Co-NaC1 0 200Torr; A 760 Torr; and H 1500 Torr. 10 ng Co-CsCl 0 200 Torr; A 760 Torr; and 0 1500 TorrJOURNAL OF ANALYTICAL ATOMIC SPECTROMETRY JULY 1994 VOL. 9 77 1 increases the tolerance of the FAPES Pb signals to added NaCl decreases.Whereas 100 pg of NaCl does not perturb the Pb response at 200 Torr as little as 0.1 pg gives a positive interference at 1500 Torr. At 200 Torr the He plasma expands beyond the geometric confines of the ICC. The mean free path for electrons is increased at low pressures and it follows that the EEDF for such a condition should exhibit higher mean energy and possess a larger population in the high energy tail of the distribution. As a consequence molecular dissociation uia electron impact should be more efficient than at higher pressure and vapour densities will be lower as will the degree of temporal overlap of analyte and matrix vapour populations. These factors tend to decrease the extent of chemical inter- ference arising from molecule formation as well as reduce radiational power losses from the plasma due to excitation of the EIE matrix species.At 1500 Torr the plasma shrinks and is confined to a narrow annulus about the CE because electron mean free paths are so short that electron energies are rapidly attenuated to the point where collisional ionization of the He support gas becomes very inefficient. In such circumstances small amounts of added NaCl thermally ionize to produce additional electrons which although not energetic enough to ionize the He plasma gas (cf. Fig. l) may acquire sufficient energy in the r.f. field to excite the Pb analyte (4.4 eV) giving rise to an elevated response over that obtained in the absence of NaCl. Higher masses of NaCl suppress response at high pressure by draining power from the plasma as radiational losses.At 200 Torr the range of interference-free concentration of NaCl on Pb response is 2-fold greater in the FAPES mode than in the AAS mode. Under such conditions Pb and NaCl vaporize into a low (thermal) temperature environment. Assuming a classical interference the effect of NaCl on Pb in AAS is due to both expulsion and molecule formation it is clear that the presence of the plasma serves to minimize the latter. As pressure increases a larger mass of NaCl can be tolerated in the AAS mode than in FAPES; at 760Torr AAS is superior to FAPES by 50-fold. Interference from NaCl on Co is more severe in the FAPES mode than in AAS at all pressures studied. Cobalt desorbs from the graphite surface at a higher temperature than Pb and thus is less prone to classical interference due to molecule formation. Additionally overlap of the NaCl vapour popu- lation with atomic Co is more severe than in the case of Pb [ c t Fig.2(c)] and the Co-Cl bond strength is greater than that of Pb-Cl(95 and 72 kcal mol-l respectively).The conse- quences are that the plasma suffers greater power losses and attenuation of the EEDF than for Pb. The impact of these factors is all the more significant as the excitation energy for the Co line is higher than for that of the Pb line (5.11 eV). Fig. 7 enables comparison of the effects of equimolar amounts of CsCl and NaCl on the response from both Pb and Co at various pressures. The general effects of CsCl are similar to those discussed above for NaCl. However interference from CsCl in some cases manifests itself at significantly lower amounts viz.more than 20-fold for Pb at 200Torr and despite much more favorable vapour overlap of NaCl with Co attenuates response from this element to a similar degree. Greater radiative power losses from the plasma may ensue in the presence of Cs as a consequence of the lower excitation energies of the electronic states of this atom compared with Na. Additionally the ease of ionization of Cs may perturb the EEDF to a greater extent than in the case of Na. These factors will lead to an enhanced interference by this salt. Conclusions The results of these studies suggest that the major causes of interference from EIEs in the FAPES source are loss of power radiated from the plasma by the excited matrix vapour compo- nents and alteration of the EEDF by the injection of low energy electrons from ionization of the EIE in addition to attenuation of plasma electron energy through collisional dissociation of molecular matrix vapour.Investigations were admittedly limited in scope and these conclusions may not be strictly valid for other analytes and other electronic transitions. Ionic lines of analytes are undoubtedly affected more severely than atom lines by the presence of EIEs. It is clear that more information is required to characterize the problem including the need for diagnostic information on electron densities and energy distributions in the presence and absence of EIEs using electrostatic probe techniques thorough optical imaging of the source in a 2-dimensional format to detect any changes in the spatial distributions of analyte arising from the presence of the EIE and a study of the effect of voltage bias control on the response in the presence of the EIEs.It is clear that the control of EIE interference in FAPES can be achieved by clever use of ‘front-end’ chemistry on the sample to remove or minimize the amount of matrix injected into the source. It also appears that operation at reduced pressures may lead to an extension of the interference-free working range. Both of these options increase the complexity of use of the technique however this should not cause discour- agement and other means should be implemented to minimize EIE effects. These may include operation at higher forward powers to ‘buffer’ the effect of concomitants use of Ar plasmas instead of He to achieve an intrinsically higher plasma electron density,24 the application of platform and chemical modifi- cation techniques in combination with optimized atomization programs to eliminate as much EIE from the source as possible prior to analyte atomization and higher r.f.frequencies which may dissipate more power in the ‘negative glow’ With respect to use of higher powers it should be noted that as the ICC is rapidly heated source impedance changes and in the absence of autotuning circuitry reflected power rises such that for a 50 W forward power plasma only 15-20 W power is delivered to the source when the temperature of the ICC is 2500 K16 (the impact would be greater for Co than for Pb as a consequence of the higher atomization temperature required for this element).Thus EIE interferences should be reduced overall if the system could be dynamically tuned to minimize reflected power losses. In such case a free-running r.f. generator might prove more useful as a power source.26 The authors thank G. Jolly Bell Northern Research Ottawa Canada for the loan of the r.f. generator. S.I. thanks the NRCC for partial financial support while in Ottawa. 1 2 3 4 5 6 7 8 9 10 11 12 13 14 References Liang D. C. and Blades M. W. Spectrochim. Acta Part B 1989 44 1059. Sturgeon R. E. Willie S. N. Luong V. T. Berman S. S. and Dunn J. G. J. Anal. At. Spectrom. 1989 4 669. Falk H. Hoffmann E. and Ludke Ch. Prog. Anal. Spectrosc. 1988 11 417. Smith D. L. Liang D. C. Steel D. and Blades M.W. Spectrochim. Acta Part B 1990 45 493. Sturgeon R. E. Willie S. N. Luong V. T. and Berman S. S. J. Anal. At. Spectrom. 1990 5 635. Sturgeon R. E. Willie S. N. Luong V. T. and Berman S. S. J. Anal. At. Spectrom. 1991 6 19. Gilchrist G. F. R. and Liang D. C. Am. Lab. 1993 25 34U. Riby P. G. and Harnly J. M. J. Anal. At. Spectrom. 1993,8,945. Matousek J. P. Orr B. J. and Selby M. Spectrochim. Acta Part B 1986,41,415. Szilvassy-Vamos Zs. Gyorfi-Buzasi A. and Pasztor Zs. Talanta 1991,11 1265. Kitagawa K. and Horlick G. J. Anal. At. Spectrom. 1992,7 1221. Holclaj tner-Antunovic I. D. and TripkoviC M. R. J. Anal. At. Spectrom. 1993 8 359. Galley P. J. Glick M. and Hieftje G. M. Spectrochim. Acta Part B 1993 48 769. Falk H. J. Anal. At. Spectrom. 1991 6 631.772 JOURNAL OF ANALYTICAL ATOMIC SPECTROMETRY JULY 1994 VOL. 9 15 16 17 18 19 20 21 22 Hettipathirana T. D. and Blades M. W. J. Anal. At. Spectrom. 1993 8 955. Sturgeon R. E. Luong V. T. Willie S. N. and Marcus R. K. Spectrochim. Acta Part B 1993 48 893. Sturgeon R. E. Willie S. N. Luong V. T. and Dunn J. G. Appl. Spectrosc. 1991 5 1413. Pillow M. E. Spectrochim. Acta Part B 1981 36 821. Imai S. and Sturgeon R. E. J. Anal. At. Spectrom. 1994 9 493. Styris D. L. and Redfield D. A. Spectrochim. Acta Rev. 1993 15 71. Saidel A. N. Prokofjew W. K. and Raiski S. M. Tables of Spectrum Lines VEB Verlag Berlin 1961. Paper 3/07292E Fang D. and Marcus R. K. Spectrochim. Acta Part B 1991 Received September 12 1993 46,983. Accepted March 8 1994 23 Imai S. Sturgeon R. E. and Willie S. N. J. Anal. At. Spectrom. 1994 9 759. 24 Tanabe K. Haraguchi H. and Fuwa K. Spectrochim. Acta Part B 1983,38,49. 25 Marcus R. K. J. Anal. At. Spectrom. 1993 8 935. 26 The GF-CCP 100 specifications brochure Aurora Instruments Ltd. Vancouver B.C. Canada.

ISSN:0267-9477    

DOI:10.1039/JA9940900765

出版商:RSC    

年代:1994

数据来源: RSC

摘要:

JOURNAL O F ANALYTICAL ATOMIC SPECTROMETRY JULY 1994 VOL. 9 773 Electrothermal Vaporization Inductively Coupled Plasma Atomic Emission Spectrometric Technique Using a Tungsten Coil Furnace and Slurry Sampling Peter Barth and Viliam Krivan* Sektion Analytik und Hochstreinigung Universitat Ulm 0-89069 Ulm Germany A simple and inexpensive electrothermal vaporization device consisting of a double-layer tungsten coil of the type normally manufactured for halogen lamps was used for the simultaneous determination of AI Ca Cr Cu Fe Mg Mn Ni and Ti in aqueous suspensions of silicon carbide powder by inductively coupled plasma atomic emission spectrometry (ICP-AES). Possible interferences were investigated andl backgroun! correction are discussed. Excluding Al the limits of detection achievable were at the sub-pg g- and pg g- level.The accuracy was checked by comparison of the results with those obtained by instrumental neutron activation analysis (INAA) slurry sampling electrothermal atomic absorption spectrometry and ICP-AES involving decomposition of the sample. The precision expressed as relative standard deviation was between 3.3% (for 210 pg g-' of Ti) and 13.5% (for 8.9 pg g-' of Ni). Keywords lnduc tively coupled plasma atomic emission spectrometry ; elec fro t h erm a I vaporization ; tungsten coil furnace; slurry sampling; silicon carbide Because of serious limitations of conventional solution tech- niques in the analysis of solids solid sampling methods have become of great interest particularly for materials that are difficult to decompose.The solid sampling techniques that have been developed for inductively coupled plasma atomic emission spectrometry (ICP-AES) are basically of two types. The first is based on the vaporization of the original sample in the plasma. This principle is used by the direct insertion technique,'-6 the dry nebulization of powders7-'' and the slurry nebulization technique."-16 In the second type the sample is vaporized outside the plasma in a device that is connected directly to the injector tube of the ICP. Sample insertion can be achieved by laser ablati~n,'~-~' arc and spark or electrothermal vaporization (ETV). The ETV technique offers a number of advantages such as high sample introduction efficiency a low sample volume high absolute detection power [comparable to electrothermal atomic absorption spectrometry (ETAAS)] the possibility of removing certain sample components before the vaporization step and the use of chemical modifiers.However condensation during transport to the plasma memory effects soaking of the sample into the furnace material blanks chemical reactions and spectral interferences caused by the furnace material can limit the applicability of this technique. In addition for com- mercially available ICP-AES instruments the processing of the short transient signals that occur in ETV-ICP-AES and appropriate background correction or recording of the emis- sion line profiles is difficult. Graphite (tubes and rods)24-33 and metals (tungsten or tantalum tubes and have been used as furnace materials for ETV devices.Owing to the low electrical resist- ance of graphite furnaces and metal tubes currents of up to several hundred amperes obtainable only from strong and expensive power supplies are necessary to achieve the tempera- tures desired for the vaporization step. However metal fila- ments which consume less power allow sample volumes of only a few microlitres (typically 3-5 yl) to be processed. Furthermore graphite and metal tubes are relatively expensive and the laboratory-made metal filaments often suffer from low reproducibility. Berndt and Schaldach used a double-layer tungsten coil normally manufactured for halogen lamps as an ETV de~ice.~' This technique was applied to the analysis of aqueous soh- ti on^.^'-^^ The attraction of the utilization of these coils as * To whom correspondence. should be addressed.ETV furnaces is to be seen in their extremely low price and high reproducibility with respect to the geometric shape resulting in high reproducibility of the physical properties. In addition very high heating rates and temperatures of about 3000°C can be achieved by an inexpensive power supply sample volumes of up to 50 yl can be applied and only a small quartz apparatus is necessary to mount the tungsten coil which can easily be connected to the injector tube of an ICP torch. However the advantage of a sample introduction method based on ETV can most effectively be exploited in connection with solid sampling. In the present paper the first results obtained by combination of ETV using a tungsten coil furnace with the slurry sampling technique applied to the analysis of silicon carbide powder are reported.Experimental Standards and Samples For standardization of Al Ca Cr Cu Fe Mg Mn Ni and Ti stock standard solutions ( 1 mg ml-') supplied by Merck Darmstadt Germany were used. In all cases the solvent was OSmoll-' HN03 except for Ti for which the solvent was 5.0 moll-' HCl. For W and Si solutions of 1 mg ml-' in 2.5 moll-' HF and 5.0 moll-' NaOH solution respectively were used. The silicon carbide powder Type S 933 was obtained from ESK Kempten Germany. The typical average particle diameter was at the sub-pm level and the particle size did not exceed For dilution of the stock standard solutions and preparation of the sample slurries doubly dis- tilled water was used.Carrier Gas In order to lower the oxidation of the tungsten coil during heating an Ar-H2 mixture (Linde Munich Germany; 6.5% v/v H2) was used as the carrier gas.44 The gas flow was controlled by a flow meter and a needle valve (Rota Wehr FRG; Rotameter Type). Instrumentation Spectrometer A sequential Jobin-Yvon JY-24 spectrometer extended with a Jobin-Yvon JY-SOP polychromator was used. The JY-SOP allows wavelength scanning by moving the entrance slit of the774 JOURNAL OF ANALYTICAL ATOMIC SPECTROMETRY JULY 1994 VOL. 9 polychromator. Control of the spectrometer and data collection were performed using an IBM PS/2 computer system and the ISA/Jobin-Yvon software package. ETVdevice The quartz apparatus power supply and software (written in GW-BASIC) were obtained from ISAS Dortmund Germany.The power supply was controlled by an AT-286 computer equipped with a Flytech Type FPC-011 14-bit two-channel analogue-to-digital-digital-to-analogue (ADDA) card. The 24 V 250 W tungsten coils Type 64655 HLX were supplied by Osram Munich Germany. The diameter of the wire was about 0.25 mm the area of the coil was 7.0 x 3.5 mm2 the distance between the double layers was 1.0mm and the mass 211 mg. The space formed between the double layers allows the intro- duction of up to 50 pI of sample solution or slurry. In order to mount the tungsten coil in the quartz apparatus a ceramic stopper with two copper electrodes for clamping and contacting the coil was used. The ETV device could be connected to the injector tube of the ICP torch without any modification of the ICP atomic emission spectrometer.The whole instrumental set-up is shown in Fig. 1. The computer program originally written for controlling a high-pressure asher was appropriately modified for the present purpose. The software allowed a voltage-time programme to be run (similar to the tempera- ture-time programmes in ETAAS). For the voltage measure- ments a Keithley Type 130A digital multimeter was used. Integrator In measurements using the sequential spectrometer transient signals were registered by a Spectra Physics Type SP4270 integrator connected to the analogous output of the photomul- tiplier-amplifier. The integrator was cont olled by the second f\ 0 8 14 Fig. 1 Schematic set-up for the tungsten coil technique 1 tungsten coil (24 V 250 W); 2 quartz apparatus (100 x 6 mm i.d.); 3 carrier gas Ar-H (6.5% v/v H,) at a flow rate of 800mlmin-l; 4 quartz stopper; 5 micropipette; 6 silicon hose; 7 glass tube with ball joint; 8 d.c.-controlled power supply; 9 computer (AT-286); 10 14-bit D/A card; 1 1 laboratory-made interface for controlling the integrator; 12 integrator; 13 analogous output of the photomultiplier-amplifier of the monochromator (11-13 were only used for sequential measure- ments); and 14 plasma torch digital-to-analogue (DA) channel of the ADDA card via a 1 aborat ory-made interface.All of the operating conditions for the instrumentation are summarized in Table 1. .Procedures The tungsten coil was pre-treated five times in the Ar-H mixture using programme I given in Table 1.The sample :slurries (three replicates for each analysis set) were prepared by mixing 20mg of silicon carbide powder with 10ml of doubly distilled water in a 30ml plastic beaker. The mixture was treated in a Sonorex RK 255H ultrasonic bath (Bandelin Electronic Berlin Germany) for 15 min in order to disintegrate the particle agglomerates. The suspension was maintained by stirring with a magnetic stirrer. For standardization by the standard additions method the slurries were spiked with an aqueous standard solution containing the following concen- trations of the elements to be determined Al 40.0; Ca 2.2; Cr 1.6; Cu 0.8; Fe 65.0; Mg 0.8; Mn 0.2; Ni 1.0; and Ti 35.0 pg ml-I. Aliquots (20 pl) of the suspensions were pipetted manually on to the tungsten coil. One analysis set included the measurement of a water blank slurry slurry spiked with 100 pl of the standard mixture and slurry spiked with 200 pl of the standard mixture (five replicate measure- ments for each sample).The data acquisition was manually triggered off 1 s before the vaporization step of the voltage-time programme I1 (see Table 1). After execution of the temperature programme the furnace was allowed to cool down for 45 s. Table 1 Operating parameters Spectrometer- Plasma gas (Ar) Intermediate plasma Aerosol carrier gas (Ar) Nebulizer (Ar) ETV (Ar-H,; 6.5% v/v H,) R.f. power Nebulizer Integration time (simultaneous measurements) Emission lines used Element Simultaneous measurements A1 Ca Cr c u Fe Mg Mn Ni Ti Sequential measurement W 0 ETVdevice - Programme I Voltage/mV (pre-treatment) 5000 14000 Programme I1 (slurry) 250 150 250 14000 1000 18000 Sample volume Time/s 6 5 14 1 min-' 0.2 1 min-' 0.4 1 min-' 0.8 1 min-' 900 W Meinhard type 7.0 s Wavelength/nm 396.152 39 3.3 66 267.716 324.754 259.940 279.553 257.610 23 1.604 334.941 208.819 777.193 20 10 5 (vaporization) 2 4 (cleaning) 120 (drying) 20 plJOURNAL OF ANALYTICAL ATOMIC SPECTROMETRY JULY 1994 VOL.9 775 Results and Discussion Voltage-Time Programme The drying step was optimized by monitoring the evaporated water using the 0 line at 777.193 nm. When voltages of up to 250mV were applied no detectable losses of the analyte elements occured. The intermediate reduction of the voltage during the drying step of from 250 to 150 mV lead to more even evaporation and an improvement in the reproducibility.Complete decomposition and vaporization of the dried slurries was achieved at 14 000 mV and further increase of the vaporiz- ation voltage did not lead to any significant improvement of the analyte signal but it did cause a reduction in the lifetime of the tungsten coil. Without the cleaning step at 18 000 mV a memory effect for Ti was observed. Interferences Possible spectral interferences of the elements to be determined from the matrix (Si) and the furnace material (W) were investigated by recording the emission spectra over a range f0.3 nm around the emission lines used in the determinations (see Table 1). This was achieved by processing aqueous solu- tions of Si (lo00 pg ml-') W (100 pg ml-') and a mixture of the elements to be determined (1 pg ml-I for each element).Sample introduction was by pneumatic nebulization. As the superimposed spectra showed no spectral interferences from Si were observed whereas spectral interferences from W occured in the case of Cr Mn and Fe. On processing 20 pl of water as the blank the W interference caused an increase in the background at the emission lines used for Cr Mn and Fe of about 5 10 and 20% respectively. For the silicon carbide powder under investigation this increase in background amounts to about 2% for Fe 20% for Cr and 60% for Mn of the analyte signal obtained during processing of 40 pg of sample (2 mg of Sic in 1 ml of water; 20 pl aliquot). Background Correction The applicability of the usual background correction pro- cedure based on the subtraction of the blank signal from the sample signal was investigated.When programme I1 was used for processing 20 pl of doubly distilled water (in all other experiments 20 pl portions were also applied) the integrated intensity of the W line at 208.819nm was about 50% higher compared with that obtained without water owing to slight corrosion of the tungsten filament. For each element the background was measured at the left and right side of the emission line (see Table 1) in doubly distilled water and in the slurry using the same conditions as during the analysis (simultaneous determination programme 11). For all elements the values obtained for water and the slurry were very similar. For the interference-free elements (Al Ca Cu Mg Ni and Ti) the same behaviour could also be expected at the maximum of the emission lines.Consequently background correction was performed by measuring the inten- sities at the maximum of the emission lines using doubly distilled water as the blank sample and subtracting this value from the signal for the sample. When applying this background correction method to Cr Fe and Mn which are subject to interference from W it is necessary to know if the amount of W released from the coil is the same when processing the water blank and the slurry. By processing 20 pl portions of water and the slurry under the same conditions as used for analysis the amount of W released was measured using the 208.819 nm line and for the slurry and water intensities of 3760f 190 and 4100+ 130 (in both cases n=5) respectively were obtained.For the slurry the amount of vaporized W is lower by 8.3% than that for water provided that the contribution of the W contained in the slurry sample is negligible. The sample investigated contains 2 pg g-' of W,45 i.e. 80 pg of W in a 20 pl aliquot of slurry (concentration of 2 mg of Sic in 1 ml of water). By processing 20 pl of a solution containing the same amount of W no increase in the W signal compared with the water blank could be observed. Thus performing background correction by subtracting the water blank from the sample signal must lead to lower results for these three elements. Owing to the relatively high concen- trations of Cr and Fe in the material being analysed this error is negligible. However in the determination of the sub-pg g-' concentrations of Mn the background correction method applied is a source of considerable error.Analysis of Silicon Carbide The method developed was applied to the analysis of a silicon carbide material which has been well characterized by several independent method^.^^,^^ The matrix effects occuring in the determination of Cr Cu and Ti made calibration by the standard additions method necessary. As the simultaneous mode was being used anyway for the measurements this calibration method was then applied for all elements. In Table 2 the results for the proposed slurry-ETV method are compared with all other results available and these show for most elements a high degree of consistency. For Al Ca Cr Cu and Fe the present results are in good accordance with those of the other methods. For Mg considering the relatively low concentration level the agreement is satisfactory and for Ti the result is clearly outside the range of the results from the independent methods.This could be caused by the different vaporization behavior of Ti in the slurry and in the standard solution used for standard additions. As discussed under Background Correction the deviation of the results obtained for Mn by the proposed technique compared with the results from the independent methods can be explained by the error introduced by the background correction which owing to the low Mn content represents a relatively large fraction of the analyte signal measured. Only in the case of Ni does the poor consistency of the whole set of results not allow an evaluation of the accuracy of the present results by comparison with the others.In spite of the very low effective sample mass of 40 pg involved in the analyses the precision achievable is remarkably high. This is obviously due to a high degree of sample homogeneity. In fact owing to the minute sample portions applied this also seems to be a suitable method for checking the sample homogeneity. After processing three consecutive analysis sets (k. 60 runs) the tungsten coil was partially molten and very brittle and had to be replaced by a new one. Compared with the lifetime of 400-500 runs of a graphite tube used for ETAAS (slurry sampling with a Sic matrix),43 the durability of the tungsten coil is lower. However the price of the graphite tube used (pyrolytic graphite coated fork-shaped platform) is about loo0 times higher than that of the tungsten coil.In Table 3 limits of detection determined experimentally on the basis of three standard deviations of the blank fluctuation (20 pl water n = 10) divided by the sensitivity (slope of the standard additions graph) are given and compared with those obtained by graphite furnace ETV-ICP-AES,29 slurry nebuliz- ation ICP-AES46 and ~lurry-ETAAs.~~ Although the tungsten coil ETV-ICP-AES method has a high absolute power of detection the limits of detection achievable in the analysis of slurries are limited by the relatively low amount of sample that can be applied to the coil. If an amount of sample greater than 60 pg is used the ability of the tungsten coil to carry the sample is exhausted and then applying more sample to the coil leads to irregular losses during the execution of the temperature programme.This results in reducing the reproduc- ibility of the analyte signals. Thus the amount of sample that can be applied is limited by the surface of the coil. For larger amounts of sample a new coil design would be necessary.776 JOURNAL OF ANALYTICAL ATOMIC SPECTROMETRY JULY 1994 VOL. 9 Table 2 Element concentration (pg g-') in silicon carbide powder (ESK S 933) determined simultaneously by ICP-AES using slurry sampling and tungsten coil ETV (n=4) and a comparison with results from other techniques Element This work ICP-AES (ref. 45) Slurry- ETAAS A* Bt (ref. 43) INAA (ref. 45) A1 Ca Cr c u Fe Mg Mn Ni Ti 157 i- 7 10.6k0.8 9.5 & 0.6 2.61f0.1 308 & 14 3.2 If 0.2 0.46 k 0.04 8.9 f 1.2 210f7 178+4 9.3 k 0.4 6.9 f 0.5 2.9 f 0.3 320 7 3.7 & 0.2 0.73 f 0.14 158k 1 - 184k6 12k3 13f3 2.0 k 0.2 325 & 20 4.2 & 0.5 0.8,O.l 8 f 2 148f2 ~ _ _ _ _ _ 177f 19 - 7.1 f 0.5 7.1 k0.2 3.2 f 0.6 - 340 f 20 322f 14 5.2 f 0.8 0.72 & 0.08 4.8 k 0.7 130f13 - - - - 0.70 f 0.01 3.8 k 0.2 * A Decomposition with HF-HN0,-H,SO at 240 "C for 12 h.t B Decomposition with HF-HNO at 260°C and fuming with HClO,. Table 3 Achievable limits of detection in the analysis of silicon carbide by tungsten coil ETV and slurry sampling and comparison with other techniques Detection limit/pg g- Element A1 Ca Cr c u Fe Mg Mn Ni Ti ICP-AES (this work) 17 0.11 2.4 0.54 3.6 0.12 0.11 3.0 5.4 ICP-AES (graphite furnace ETV ref.29) 17 15 0.97 3.8 1.2 0.09 0.22 0.95 - ~~ ICP-AES Slurry ETAAS (slurry nebulization ref. 46) (ref. 43) - 0.2 2 2 0.04 0.1 11 2 - - - 0.2 0.05 2 0.02 0.8 5 - For ICP techniques limits of detection strongly depend on the sample concentration achievable in the carrier gas As the carrier gas flow applied to the tungsten coil and in graphite furnace ETV2' is similar (800 and 1000 ml min-' respectively) the great differences in the duration time of the signal (tungsten coil about 1 s; graphite furnace 10-20s) are the obvious reason why for both methods limits of detection achieved are at about the same level although the sample portions evapor- ated in the graphite furnace were 250 times higher; only for Ti does the graphite furnace technique lead to a significantly better limit of detection.For Ca Cr and Cu the limits of detection of this method are even lower by factors of 9 6 and 7 respectively than those of graphite furnace ETV (see Table 3). In slurry nebulization ICP-AES,46 the carrier gas flow was 730ml min-' and the amount of sample evaporated in the plasma was 370 pg s-' of Sic. Compared with the tungsten coil ETV technique the higher concentration of sample in the carrier gas in the slurry nebulization technique should lead to better limits of detection. As can be seen from Table 3 this is only achieved for Mg and Ti (improvement by a factor of about 3) whereas for Ca Cr and Mn comparable limits of detection are reached by both techniques. In the case of Cu and Ni the limit of detection of this method is lower by a factor of about 4 than that of the slurry nebulization technique. Possible reasons for the increased detection limits of the slurry nebulization technique can be seen in the additional consump- tion of plasma energy required for removal of the solvent sample decomposition and vaporization altering the atomiz- ation and excitation of the analyte elements.In addition the slurry nebulization technique is much more limited by the particle size than this slurry sampling ETV technique. In slurry-ETAAS the excellent power of detection combined with the size of the sample portions applied (500 pg)43 leads to limits of detection that are better by about one order of magnitude than those obtainable by the tungsten coil ETV technique for Cr Cu Mn and Ni whereas for Fe and Ti the limits of detection for both techniques are comparable. The increased detection limits for these two elements in slurry- ETAAS are due to strong spectral interferences and memory effects.43 Conclusions Through the analysis of silicon carbide for Al Ca Cr Cu Fe Mg Mn Ni and Ti it has been demonstrated that the combination of a tungsten coil as a simple and inexpensive ETV furnace with the slurry sampling technique for ICP-AES provides a promising method for the analysis of powdered refractory materials.Detection limits at almost the same level as those for graphite furnace ETV-ICP-AES and the slurry nebulization technique are achievable. References 1 Zaray G. Broekaert J. A. C. and Leis F. Spectrochim. Acta Part By 1988 43 241.2 Reisch M. Nickel H. and Mazurkiewicz M. Spectrochim. Acta Part B 1989 44 307. 3 Sing R. L. A. and Salin E. D. Anal. Chem. 1989 61 163. 4 Umemoto M. and Kubota M. Spectrochim. Acta Part B 1991 46 1275. 5 Blain L. and Salin E. D. Spectrochim. Acta Part B 1992,47,205. 6 Umemoto M. Hayashi K. and Haraguchi H. Anal. Chem. 1992 64 257. 7 Ng K. C. Zerezghi M. and Caruso J. A. Anal. Chem. 1984 56 417. 8 Pfannerstill P. E. Caruso J. A. and Willeke K. Appl. Spectrosc. 1989 43 626. 9 De Silva K. N. and Guevremont R. Spectrochirn. Acta Part B 1990 45 997.JOURNAL OF ANALYTICAL ATOMIC SPECTROMETRY JULY 1994 VOL. 9 777 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 Guevremont R. and De Silva K. N. Spectrochim. Acta Part B 1991 46 67. Fuller C.W. Hutton R. C. and Preston B. Analyst 1981 106 913. Fernandez Sanchez M. L. Fairman B. and Sanz-Medel A. J. Anal. At. Spectrom. 1991 6 397. Graule T. Von Bohlen A. Broekaert J. A. C. Grallath E. Klockenkbper R. Tschopel P. and Tolg G. Fresenius 2. Anal. Chem. 1989 335 637. Lathen C. Broekaert J. A. C. Tolg G. Docekal B. and Tschopel P. ICP In$ NewsL 1991 16 714. Ebdon L. and Goodall P. J. Anal At. Spectrom. 1992 7 1111. Halicz L. Brenner J. B. and Yoffe O. J. Anal At. Spectrom. 1993 8 475. Ishizuka T. and Uwamiro Y. Anal. Chem. 1980 52 125. Thomson M. Goulter J. E. and Sieper F. Analyst 1981,106,32. Carr J. W. and Horlick G. Spectrochim. Acta Part B 1982,37 1. Arrowsmith P. Anal. Chem. 1987 59 1437. Farnsworth P. B. and Hieftje G. M. Anal. Chem. 1983,55 1414.Aziz A. Broekaert J. A. C. Laqua K. and Leis F. Spectrochim. Acta Part B 1984 39 1091. Ono A. Suchi M. and Chiba K. Appl. Spectrosc. 1987,41,970. Gunn A. M. Millard D. L. and Kirkbright G. F. Analyst 1978 103 1066. Ng K. C. and Caruso J. A. Anal. Chim. Acta 1983 143 209. Swaidan H. M. and Christian G. D. Anal. Chem. 1984 56 120. Matusiewicz H. Fricke F. L. and Barnes R. M. J. Anal. At. Spectrom. 1986 1 203. Shen W.-L. Caruso J. A. Fricke F. L. and Satzger R. D. J. Anal. At. Spectrom. 1990 5 451. Zaray G. Kantor T. Wolff G. Zadgorska Z. and Nickel H. Microchim. Acta 1992 107 345. 30 Ren J. M. and Salin E. D. J. Anal. At. Spectrom. 1993 8 59. 31 Nickel H. Zadgorska Z. and WolfT G. Spectrochim. Acta Part B 1993 48 25. 32 Verrept P. Galbacs G. Moens L. Dams R. and Kurfurst U.Spectrochim. Acta Part B 1993 48 671. 33 Docekal B. and Krivan V. Spectrochim. Acta 1994 in the press. 34 Nixon D. E. Fassel V. A. and Kniseley R. N. Anal. Chem. 1974 46 210. 35 Kitazume E. Anal. Chem. 1983 55 802. 36 Tikkanen M. W. and Niemczyk T. M. Anal. Chem. 1984 56 1997. 37 Stahl R. G. Brett L. and Timmins K. J. J. Anal. At. Spectrom. 1989 4 337. 38 Tsukahara R. and Kubota M. Spectrochim. Acta Part B 1990 45 779. 39 Okamoto Y. Murata H. Yamamoto M. and Kumamuru T. Anal. Chim. Acta 1990 239 139. 40 Berndt H. and Schaldach G. J. Anal. At. Spectrom. 1988,3 709. 41 Dittrich K. Berndt H. Broekaert J. A. C. Schaldach G. and Tolg G. J. Anal. At. Spectrom. 1988 3 1105. 42 Gine M. F. Krug F. J. Sass V. A. Reis B. F. Nbbrega J. A. and Berndt H. J. Anal. At. Spectrom. 1993 8 243. 43 Docekal B. and Krivan V. J. Anal. At. Spectrom. 1992 7 521. 44 Suzuki M. Ohta K. and Yamakita T. Anal. Chem. 1981 53 9. 45 Franek M. and Krivan V. Fresenius’ J. Anal. Chem. 1992 342 118. 46 Docekal B. Broekaert J. A. C. Graule T. Tschopel P. and Tolg G. Fresenius’ J. Anal. Chem. 1992 342 113. Paper 3/06415H Received October 26 1993 Accepted March 22 1994

ISSN:0267-9477    

DOI:10.1039/JA9940900773

出版商:RSC    

年代:1994

数据来源: RSC

摘要:

JOURNAL OF ANALYTICAL ATOMIC SPECTROMETRY JULY 1994 VOL. 9 779 On-line Flow Injection Cobalt-Ammonium Pyrrolidin-I = yldithioformate Coprecipitation for Preconcentration of Trace Amounts of Metals in Waters with Simultaneous Determination by Inductively Coupled Plasma Atomic Emission Spectrometry Zhixia Zhuang Xiaoru Wang* Pengyuan Yang Chenlong Yang and Benli Huang Department of Chemistry Xiamen University Xiamen China The technique of on-line flow injection (FI) cobalt-ammonium pyrrolidin-1 -yldithioformate (Co-APDC) copre- cipitation for the preconcentration of trace amounts of the heavy metals Cd Cu Fe Ni Pb and Zn in rain water samples with simultaneous determination by inductively coupled plasma atomic emission spectrometry (KP-AES) has been developed. A precipitate collector system consisting of a poly(tetrafluoroethy1ene) (PTFE) membrane on a polypropylene support filtering device combined with a 1.5 m reaction coil was selected.An inorganic solution of concentrated nitric acid and hydrogen peroxide was applied as the dissolution reagent. The technique is characterized by high retention efficiency (which ranged from 77 to 99% for the six elements of interest) good enrichment factors (ranging from 10 to 50 for 100 s preconcentration depending on the elements studied) and satisfactory accuracy and precision (recoveries from two standard additions to a rain sample ranged from 92 to 104O/0 with relative standard deviations ranging from 1.9 to 5%). The sample throughput is 20 per hour. Keywords On-line coprecipitation; cobalt-ammonium pyrrolidin- 1 -yldithioformate coprecipitation ; heavy metals in rainwater; flow injection inductively coupled plasma atomic emission spectrometry The flow injection (FI) technique has been widely applied to the on-line separation and preconcentration of trace and ultra- trace amounts of heavy metals in various water samples to improve the sensitivity of the determination and to eliminate interferences from the sample matrix and some co-existing elements.Solvent extraction including liquid-liquid' and sor- bent extraction,2 chelating ion-exchange columns3 and hydride generation4 are common techniques used for FI on-line separ- ation and preconcentration. On-line precipitation-dissolution in non-segmented continu- ous-flow systems in combination with flame atomic absorption spectrometry (FAAS) has been studied extensively by Valcarcel and gal leg^.^ The technique has been used for the preconcen- tration of lead in water samples by on-line preconcentration of the hydroxide with ammonium solution followed by dissolu- tion with nitric acid and detection by FAAS.6 A preconcen- tration factor of 700 was achieved under their operating conditions.The preconcentration of Cu Ca and Co in silicate samples with this technique has also been Although coprecipitation is a very traditional chemical separation and preconcentration method it is not widely used in combination with the on-line FI technique. Recently on-line FI with hexamethylene ammonium hexamethylene dithio- carbamate coprecipitation combined with FAAS was estab- lished by Fang et al.," in which the coprecipitation of lead in the presence of high concentrations of iron was performed with the advantages of low sample consumption high operating efficiency and high tolerance of the iron matrix.The technique developed by Fang used a knotted reactor as precipi- tate collector. The knotted reactor promoted radial mixing of sample and reagent providing reproducible conditions for the precipitation. In the present work on-line FI Co-ammonium pyrrolidin- l-yldithioformate (APDC) coprecipitation for trace amounts of elements in water samples and simultaneous multi-element determination with inductively coupled plasma atomic emis- sion spectrometry (ICP-AES) is developed. Although the con- ventional batch method of Co-APDC coprecipitation has the advantages of high retention efficiency large enrichment fac- tors suitable pH range and the ability for multi-element * To whom correspondence should be addressed. preconcentration it has not been applied to on-line FI because of the technical difficulties involved in the on-line system.In this study the FI manifold for the Co-APDC coprecipi- tation-dissolution has been established. The effects of the concentration of coprecipitation reagents the pore size of the filter membrane and the selection and flow rate of the dissolu- tion reagent were optimized. Several on-line FI coprecipi- tation-dissolution systems including a poly(tetrafluor0- ethylene) (PTFE) membrane on a polypropylene support filtering device a PTFE membrane on a polypropylene sup- port filtering device combined with a reaction coil and a reaction coil combined with a filter device filled with tiny PTFE chips have been studied and compared.The technique developed was applied to the analysis of rain water samples. Experimental Apparatus A Baird (Bedford MA USA) PS-4 multi-channel ICP atomic emission spectrometer was used for simultaneous multi-element determinations. The ICP-AES operating conditions are sum- marized in Table 1. A transient signal collection and data processing software developed in this laboratory was used to collect and handle the FI signals. One hundred data points were collected for each injection with 0.8s integration per Table 1 ICP-AES operating conditions R.f. power/kW Frequency of r.f. generator/MHz Coolant gas flow rate/l min-' Plasma gas flow rate/l min-' Carrier gas flow rate/l min-' Observation height/mm Integration time/s Data collection points Torch Nebulizer Wavelength/nm 1.2 21 10 1 .o 1 .o 15 (above load coil) 0.8 100 Regular Fassel type High salt (Babington) Cd 226.5 Cu 324.1 Fe 259.9 Ni 231.6 Pb 220.3 Zn 213.8 Co 238.8780 JOURNAL OF ANALYTICAL ATOMIC SPECTROMETRY JULY 1994 VOL.9 point. The peak height measurement mode was used while the background was corrected with a fixed slit. A ZL 2000 FI processor (Zhaofa Institute of Automated Analysis Shengyang China) with two peristaltic pumps and a 16-port valve was applied to the on-line preconcentration and separation. Coprecipitation-dissolution Manifold and Operation Procedure The manifold of the on-line FI Co-APDC coprecipitation- dissolution system is illustrated in Fig.1. The operation sequence is summarized in Table 2 and can be interpreted as follows. In the first step coprecipitation was carried out in which the sample containing 20 pg ml-' of Co" was pumped at a flow rate of 3.3ml min-' and merged downstream with 2% APDC solution at a flow rate of 0.4 ml min-'. The precipitate was collected on the filter which was cleaned with isobutyl methyl ketone (IBMK) solution stored in the loop from the previous cycle prior to the filtration. In the second step Pump 1 was stopped whereupon the coprecipitation process was terminated. Pump 2 carried a 2% APDC solution to clean off the residues remaining on the surfaces of the filter and the reaction coil. Meanwhile the dissolution loop was filled with HN03-H202 (1 + 1) solution.In the next step Pump 2 was actuated and the valve was returned from position A to the position B (See Table 2). The doubly de-ionized water was sucked up with Pump 3 attached to the ICP atomic emission P I n 2% APDC Sample Air H2O Eluent spectrometer which carried the dissolution reagent required to dissolve the precipitate collected on the filter into the nebulizer of the ICP. During the collection of signals the IBMK solution was pumped into the loop by Pump 2 for 10 s. The procedure was then repeated. A commercial PTFE filtering device was used as the precipi- tate collector. The device has a dead volume of around 200 1-11 and 40 pl after filling with PTFE chips (3 mm in length 1.5 mm in width and 0.1 mm in thickness).Poly(tetrafluoroethy1ene) membranes with pore sizes of 0.2 0.45 and 1.0pm on a polypropylene support (Gore and Associates Membrane Products Elkton MD USA) were tested for the filtration. The membrane area of the filtering device was 0.078 cm2 in the present work. Reagent and Sample Preparation Analytical reagent grade pure chemicals were used throughout the experiments. Stock solutions of 1000pgml-' for each element of interest were used for the preparation of standard solutions. All standard solutions for calibration and sample analysis were adjusted to a pH of about 3 with dilute nitric acid and ammonium solution and contained 20 pg ml-' of Co as the coprecipitate. Isobutyl methyl ketone was selected as a cleaning reagent for the residues remaining on the surface of the coil and the filtering device from the previous sample.A saturated solution of APDC (2% m/v) was prepared with Milli-Q water (Millipore Bedford MA USA). It was then P I n 2% APDC Sample Air Eluent H2O H2O Step 1 Step 2 P I Step 3 2% APDC Sample Air H2O Eluent H20 m Step 4 I-& 2%APDC Sample Off P2 Air Eluent H2O H2O Fig. 1 and W = waste (see text for details) Manifold of the FI on-line Co-APDC coprecipitation-dissolution system; A and B refer to position of valves P1 P2 and P3 are pumps,JOURNAL OF ANALYTICAL ATOMIC SPECTROMETRY JULY 1994 VOL. 9 78 1 Table 2 Operation sequence for FI on-line coprecipitation ICP-AES Step Function Time/s Pump Flowrate/ml min-' Chemicals Valve position 1 Filter clean up 120 coprecipitation and dissolution reagent filled solution filled up signal collection 2 Rinse of precipitate 30 3 Dissolution IBMK 10 4 Dissolution and 100 *The flow rate of pump 3 is 2.5 ml min- 2 3 3 3.3 0.4 4.0 1.2 0.4 - Sample A APDC APDC A HN03-Hz02 IBMK B B - purified by extraction with an equal volume of IBMK.Doubly de-ionized water was further purified with a Milli-Q water purification device. Rain water samples were collected from Xiamen Island and were also adjusted to a pH of about 3. Results and Discussion Design of the On-line Co-APDC Coprecipitation-Dissolution Preconcentration System Precipitate collector The on-line FI precipitate collector is an important part of the system. Fang" proposed a knotted reactor as a precipitate collector for the determination of lead in biological samples with FAAS.An organic solvent was used as the dissolution reagent and an enrichment factor of 20 was obtained. Several other precipitate collector systems have been studied in the present work to accommodate the technique of on-line copre- cipitate-dissolution with ICP-AES with a low power r.f. gener- ator which cannot maintain a stable plasma discharge with the introduction of an organic solvent. The experimental results obtained are compared in Table 3 and interpreted as follows. (i)The PTFE membrane on a polypropylene support filtering device. The retention efficiencies of most elements obtained with this system are satisfactory except for Fe and Zn. (ii)The combination of a knotted coil (1.5 m long x 0.5 mm i.d.) and a filter filled with PTFE chips. In order to further improve the retention efficiencies of some elements via an increase of precipitate reaction time and reaction surface area a 1.5 m reaction coil was combined with the filter device used in (i) but filled with PTFE chips instead of a PTFE membrane.The experimental results indicate that although the retention efficiencies for iron and zinc were greatly improved that of lead was very low. This might be attributed to the low dissolution efficiency of lead under these operating conditions. (iii)A PTFE membrane on a polypropylene support filtering device combined with a knotted 1.5 m reaction coil. With this precipitate collector system all elements examined showed satisfactory retention efficiencies. Therefore this system was used in the present work. The contribution of the filtering device to dispersion was measured by the method described by Ruzicka and Hansen,13 and was 2.98.The retention efficiency (RE) was examined using the follow- Table 3 Comparison of retention efficiency (%) Element Cd co c u Fe Ni Pb Zn Filter only 98 16 99 13 97 94 35 Coil + PTFE chips 79 92 100 97 96 2 94 Coil + Filter 95 93 99 77 95 93 88 ing procedure 25 ml of 0.08 pg ml-' of a multi-element stan- dard solution was preconcentrated with three precipitate collectors under the optimized conditions; the filtrates were collected; and both standard solution and filtrate were analysed directly with ICP-AES. The retention efficiency was calcu- lated as ) x 100% [Element] in filtrate [Element] in standard solution The summary of standard solution REs using three types of the precipitate collectors is given in the Table 3.Dissolution system Isobutyl methyl ketone has been successfully used to dissolve Co,Ni-APDC coprecipitates in a batch method.12 However when IBMK is introduced into the ICP-AES system the plasma is susceptible to extinguishment with a low power r.f. generator. Owing to the high dissolution efficiency of organic solvents for the Co-APDC coprecipitate several other organic solvents (ketone methanol dimethylformamide) were tested. However none of the methods are successful under the current instrumentation conditions. Therefore an inorganic dissolution system with HN0,-H202 has been considered in this work. With the preliminary test it was found that the solution was a very strong dissolution reagent for on-line FI Co-APDC coprecipitation and did not cause any problems with the ICP discharge.Optimization of the mixing ratio of HNO3-HzO2 was performed and the results are illustrated in Fig. 2 which indicates that the best dissolution would be expected with a mixing ratio of l t l corresponding to 7 moll-' HN03 and 19% m/v H202. The dissolution efficiency with the system selected was examined. A 15 min preconcentration of a 0.4 pg ml-' multi- 1400 - 1200 4- .- C 3 >. F .= 1000 e - > v) C 4- ._ 800 - 600 0 1 1.2 1 l 2 1 1 :o HNO, H,O Fig. 2 Effect of HNO H202 ratio on intensity782 - c JOURNAL OF ANALYTICAL ATOMIC SPECTROMETRY JULY 1994 VOL. 9 1600 - In c C 3 .- ; 1200 c .- e - .g 800 In C m C c. - 400 0 2 4 6 8 Dissolution reagent flow rate/ml min-' Fig.3 Effect of dissolution reagent flow rate on intensity - 1 I .! c I / - 400 ' I I I I I 0 30 60 90 Preconcentration time/s Fig. 4 Effect of preconcentration time on intensity for zinc element standard solution was carried out and the coprecipitate formed was dissolved on-line with 300 pl of HN03-H202. The dissolution process was repeated three times. The solutions dissolved were nebulized and measured by ICP-AES. The intensities of all the elements studied in the third dissolution were identical to that of a blank solution. Therefore the dissolution efficiency was calculated as the amount of the element found from the first dissolution over the sum of the amounts from the first and the second dissolution without considering the third dissolution.The experimental results indicate that the dissolution efficiencies are over 97% for all elements studied. The effect of the flow rate of the dissolving solution ranging from 1 to 8 mlmin-' has been examined. The results are shown in Fig. 3. The maximum peak heights were obtained with a flow rate of 2.5 ml min-' which was subsequently used throughout the experiment. Optimization of On-line Co-APDC Coprecipitation Capacity of precipitate collector The capacity of the precipitate collector with the membrane filtering device was examined using 0.4 pg ml-' of a multi- element standard solution at flow rates of 3.3 ml min-' for the sample and 2.5 ml min-I for the dissolution reagent. Using these conditions no block from the flowing system and filtering device was observed.The capacity is calculated as concen- tration of standard solution (pg ml-') x preconcentration time (min) x flow rate of sample (ml min-') x the number of elements =0.4 x 15 x 1.5 x 6= 54 pg The concentration of Co used is 20 pg ml-' which is 50 times that of the element being studied. Hence the contribution of Co to the capacity of the system is as much as 450pg. Considering that the system used in the present work includes a 1.5 m reaction coil the actual capacity of the system is even larger than the value calculated (54 pg + 450 pg = 504 pg). Effect of Membrane Pore Size and the Life Time of the Membrane The effect of the membrane pore size on preconcentration was tested qualitatively with pore sizes of 0.2 0.45 and 1.0 pm.Initially high retention efficiency with a small pore size mem- brane and low efficiency with the large pore size membrane were expected. However no significant differences of the retention efficiencies among the three types of membrane studied were observed. This might be interpreted as indicating that the fine particles of the precipitate dissolved are mainly adsorbed on the surface of the knotted reaction coil. Only particles larger than 1.0 pm are deposited on the surface of the filtering membrane. Therefore the membrane with a pore size of less than 1.0 pm did not show a significant effect. The life-time of the membrane with 0.2 pm pore size was evaluated for almost a hundred working hours utilization. No surface distortion or unusual behaviour was observed during the period of operation.Concentration of Coprecipitation Reagent The Co-APDC complex acted as a carrier for the coprecipi- tation of the heavy metals of interest. The concentration of APDC in this work was fixed at 2% (saturated sol~tion)'~ with a flow rate of 0.4 ml min-' for reaction with the sample solutions containing Co. The effect of Co concentration on the preconcentration was examined with 0.05 pg ml- ' multi- element standard solutions containing different concentrations of Co ranging from 10 to 200 pg ml-'. Although high retention efficiency would be expected with a high concentration of Co there was difficulty associated with the dissolution in the inorganic solvent of test solutions containing 100 and 200 pg ml-I of Co. The experimental results indicate that the best signals with high and sharp peaks were obtained with the lowest concentration of Co (10 pg ml-') for Cd Cu Fe and Zn using the peak height measurement mode and there was no significant effect on Ni and Pb; 20 pg ml-' of Co was used in the present work.Sample Preconcentration Time and the ICP-AES Signal The linearity of sample preconcentration time ranging from 30 to 120s versus ICP-AES signals of the elements studied was evaluated with a multi-element standard solution of 0.08 pg ml-l. As an example the preconcentration time versus zinc signal is demonstrated in Fig. 4. For all elements studied the peak heights of the signals collected were proportional to the sample preconcentration time with correlation coefficients of 0.999 or above. Performance of On-line FI Co-APDC Coprecipita tion- Dissolution System The performance of the system was evaluated using the enrichment factor (EF) retention efficiency (RE) relative standard deviation (RSD) of the multi-element injection corre- lation coefficients of the calibration curves (r) and the detection limits (DL) which are presented in Table 4.Some typical signal profiles with and without preconcentration are com- pared in Fig. 5. Rain Water Analysis Four rain water samples collected from Xiamen Island were analysed using the proposed method. The recovery test from standard additions to one rain water sample (No. 6) was performed to verify the method. The results are summarized in Tables 5 and 6. Conclusions The batch method of Co-APDC coprecipitation for trace elements has been successfully adapted to FI on-line coprecipi- tation-dissolution with ICP-AES multi-element determination.Three different types of precipitate collector systems wereJOURNAL OF ANALYTICAL ATOMIC SPECTROMETRY JULY 1994 VOL. 9 783 250 200 150 100 - v) +- .- 5 50 L. h .- G o 450 c .- v) c 0 c - - 350 250 150 1 I 1 I 1 0 50 100 550 450 350 250 150 1600 1200 800 400 0 50 100 Timeh Fig. 5 Comparison of signal profiles A with and B without preconcentration for (a) Cd; (b) Ni; (c) Cu; and (d) Zn Table 4 Characteristics of F1 on-line Co-APDC coprecipitation system Element RE*(%) EFt RSD$'(%)(n=6) r DL/pg I-' Cd 95 28 4.6 0.9980 0.7 c u 99 15 5.0 0.9984 0.2 Fe 77 7 5.0 0.998 1 0.9 Ni 95 30 1.9 0.9942 0.4 Pb 93 27 2.0 0.9898 1 Zn 88 10 4.8 0.9996 0.3 *RE = Retention efficiency.tEF = Enrichment factor. $'r = Correlation coefficient of calibration curve. Table 5 Rain water analysis; element concentrations given are the mean of triplicate analyses -~ Element concentration/pg 1- ' ~~ ~ Element No. 3 No. 6 No. 7 No. 10 Cd 9.1 8.1 7.1 8.4 cu 0.6 5.4 1.8 1.2 Fe 22.4 21.3 23.4 25.3 Ni 2.1 0.6 1.1 1.1 Pb 10 30 65 1 Zn 16.5 14.3 34.2 21.4 studied with particular attention being paid to retention efficiency. The system selected for the present work is charac- terized by high retention efficiency fairly good EF and satisfac- tory accuracy and precision for the determination of trace elements; and is suitable for simultaneous multi-element determination with ICP-AES. Future work will be focused on extension of the method to other metals and application to other sample types.This work was supported by the Chinese Natural Scientific Foundation under grant No. 292351 10-11. References Memon M. A. Zhuang Z. X. and Fang Z. L. At. Spectrosc. 1993 14 50. Ruzicka J. and Amdal A. Anal. Chim. Acta 1989 216 243. Wang X.-R. and Barnes R. M. J. Anal. At. Spectrom. 1989,4,509. Wang X.-R. and Barnes R. M. J. Anal. At. Spectrom. 1988 3 1091. Valcarcel M. and Gallego M. TrAC Trends And. Chem. 1989 8 34. Martinez-JimCnez P. Gallego M. and Valcarcel M. Analyst 1987 112 1233. Santelli R. E. Gallego M. and Valcarcel M. Anal. Chem. 1989 61 1427. Adeeyinwo C. E. and Tyson J. F. Anal. Proc. 1989 26 375. Santelli R. E. Gallego M. and Valcarcel M. J. Anal. At. Spectrom. 1989 4 547. Table 6 Recoveries from standard additions to rain sample No. 6; n = 3 Recovery Sample concentration/ Found/pg 1-' Recovery Found/pg 1 - Element PLg 1-' (added 40 pg 1-') (added 80 pg 1-l) (%I 47.2 98 86.4 98 8.1 5.4 45.8 101 88.1 103 92 93 100 104 Cd c u Fe Ni Pb Zn 14.3 54.6 101 97.8 21.3 62.6 103 95.2 0.6 40.6 100 74.9 30 67 93 110784 JOURNAL OF ANALYTICAL ATOMIC SPECTROMETRY JULY 1994 VOL. 9 10 Fang Z.-l. Sperling M. and Welz B. J. Anal. At. Spectrom. 11 Ruzicka J. and Hansen G. H. Flow Injection Analysis 2nd edn. T. R. P. American Chemical Society Washington D.C. 1975 1991 6 301. p. 44. Wiley New York 1988 p. 23. 12 Boyle E. A. and Edmond J. M. Anal. Chim. Acta 1977 91 189. 13 Boyle E. A. and Edmond J. M. in Analytical Methods in Oceanography Advances in Chemistry Series No. 147 ed. Gibb Paper 3 /03 95 0 A Received July 7 1993 Accepted March 3 1994

ISSN:0267-9477    

DOI:10.1039/JA9940900779

出版商:RSC    

年代:1994

数据来源: RSC

摘要:

JOURNAL OF ANALYTICAL ATOMIC SPECTROMETRY JULY 1994 VOL. 9 785 Analysis of Zirconium Alloys by Inductively Coupled Plasma Atomic Emission Spectrometry I. Steffan Institute for Analytical Chemistry University of Vienna Wahringerstr. 38 A- 7 090 Vienna Austria G. Vujicic I WM Industriestr. 59 CH-8 7 52 Glattbrugg Switzerland In the inductively coupled plasma atomic emission spectrometric analysis of real samples influences of the matrix on the analyte spectral lines could be expected especially when the ratio of the concentration of the matrix to that of the analyte is extremely high. In addition to the elimination of stray-light effects by background correction there are also spectral interferences which must be investigated. It is of great importance to obtain information concerning the spectral line overlaps before setting up an analytical programme. The experiments performed dealt with the interferences to be expected in a Zr matrix.In general because of the nature of Zr the chemistry and technology associated with this element are very complicated therefore the methods applied in its analysis have specific requirements. Where possible for all the analytes of interest (Al 6 Cd Co Cr Cu Fe Hf Mg Mn Mo Nb Ni Sn Ta V U and W) spectral lines were selected that were free from interferences caused by the matrix element. Where no interference- free analyte line could be found it was necessary to perform line interference corrections. The analytical programme developed in this way was tested with Standard Reference Materials (National Institute of Standards and Technology).The data compare well with the certified values. The confidence intervals for 95% probability were in the range of 2-12% depending on the element and the concentration. Keywords Zirconium; zirconium alloys; line selection; spectral interference The most important Zr ore is zircon a zirconium silicate plus hafnium silicate. More than 95% of the production of Zr is used in the form of zircon or zirconia ZrO for foundry moulds and refractory ceramic and abrasive materials. The metal is also used in the construction of chemical plants. It would appear that Zr is non-toxic and is compatible with body tissue and has thus become a competitor with Ti as a component of artificial joints and limbs. The good mechanical properties combined with resistance to corrosion and a low neutron absorption cross-section are the main reasons for the important role Zr plays in nuclear reactor technology.In the ore form Zr is associated with Hf (with an Hf content of up to 7%) which offers a very high neutron cross-section. This results in difficult separation chemistry from the two adjacent elements in the Periodic Table and isolation of Zr containing less than 100 mg kg-I of Hf as an impurity. In addition Zr has a very high chemical reactivity with respect to different metals and the environment in particular 0 N H and C and it is also necessary to eliminate numerous unwanted from a metallurgical point of view impurities. All of these require- ments lead to the complicated chemistry and metallurgy for this element and its compounds and the necessity of separating Zr from Hf and other elements.’ Inductively coupled plasma atomic emission spectrometry (ICP-AES) is usually reported to be almost free from inter- ferences nevertheless some interferences have to be considered.2 It is convenient to distinguish between spectral and non- spectral interferences.Spectral interferences arise from the incorrect isolation of the net analysis signal from the composite radiation that passes the ‘spectral window’ tuned to the analysis line causing a lack of selectivity for the method. Occasionally this problem can be partially overcome by appropriate choice of excitation condition^.^ Spectral interferences can be caused by continua stray light line wings and lines or bands.The interference signals produced by a smooth and unstructured background are directly measurable in the sample spectrum at positions adjacent to the analysis lines. This enables a reliable estimate of the background below the analysis line. Interferences from lines or band components produce greater problems because their magnitude can only be determined indirectly. Background correction is one of the most difficult problems to deal with in emission spectrometry since the background under a spectral line cannot be determined directly and in addition varies with the composition of the sample. The principal method of reducing spectral interferences is selection of the appropriate analytical lines the choice of the spectrometric apparatus and the detector. For a given spec- trometer and detector line selection is based on two criteria detection limit and selectivity.A sequence of lines with decreas- ing detection limits for a certain element are usually checked for the occurrence of particular interferences. The main problem in the spectrometric analysis of Zr and Zr alloys is the line-rich emission spectrum of Zr. The object of this investigation was the selection of lines for the ICP-AES determination of trace elements present as impurities in Zr alloys in a quality control procedure for these materials. Experimental Apparatus For the studies presented an ARL 3520 ICP spectrometer was used. The technical data for this instrument are listed in Table 1. Samples In order to test the analytical procedure National Institute of Standards and Technology (NIST) zirconium Standard Reference Materials (SRM) 360a 1238 and 1239 were anafysed.Table 1 Technical data for the ARL 3520 ICP spectrometer Spectrometer Grating Torch Nebulizer Gas flows Outer Intermediate Aerosol carrier Incident power Observation height 1 m sequential Paschen-Runge 1200 grooves mm - ’ FasseI type Meinhard type 1211 rnin-l 0.8/1 min-’ i/1 min-’ 1200/w 15 mm above coil786 JOURNAL OF ANALYTICAL ATOMIC SPECTROMETRY JULY 1994 VOL. 9 Sample Dissolution Sample dissolution was performed according to the following procedure. A portion (2 g) of sample was treated with 10 ml of a mixture of concentrated hydrofluoric acid and 7 mol I-' nitric acid (1 + 2; added in portions) in a platinum dish. The dissolution was accompanied by a violent reaction.Then 2-3 ml of water were added. As soon as the reaction had calmed down 3 ml of hydrofluoric acid were added and the dissolution was completed by the digestion of the covered sample for 1 h on a steam-bath. The digest was subsequently evaporated to dryness and the residue was dissolved in 100 ml of 1.2 mol I-' hydrochloric acid. Reagents All the reagents used were of analytical-reagent grade (Merck Suprapur). Standard solutions were prepared from stock solu- tions containing 1 mg ml-' of each element (Merck Titrisol). Line Selection and Experiments For line selection in most cases a list of the 'most sensitive' or 'prominent' lines of the elements to be detem~ined~-~ was checked for applicability to ICP analysis. The 28 most interes- ting lines tested are listed in Table 2.For all the measurements performed solutions containing zirconium oxychloride with and without the addition of the Table 2 Spectral lines investigated Element A1 B Cd Cd c o c o Cr Cr c u Fe Fe Hf Mn Mg Wavelength/ nm 308.215 182.589 214.438 226.502 230.786 228.6 16 205.552 357.869 224.700 239.562 238.204 2 7 3.8 7 6 279.553 293.930 Element Mn Mo Mo Nb Ni Sn Ta Ta Ti V V U W W Wavelength/ nm 294.920 204.598 28 1.615 309.417 221.647 181.110 267.590 226.230 334.900 311.838 309.3 1 1 290.828 216.632 208.8 19 Table 3 DLs and BEC values for the lines selected analytes of interest were used. The concentration of the stock solution of Zr was 141.4 g I-' of ZrOC12.8H20. According to the expected concentrations of Zr in the real samples dilutions were made for the investigations.Detection limits (DL) and background equivalent concentration (BEC) values were calcu- lated for pure aqueous solutions of the analytes and for solutions of the same analyte concentrations in the presence of a Zr matrix for 11 integration^.^ With such solutions spectral scans were performed over a range of kO.06 nm to both sides of the lines investigated. The integration time was 2 s per step and the step size 0.004 nm. These scans were used both for selection of the analyte lines and for evaluation of the appropriate background correction position(s). The correction coefficients for the spectral inter- ferences observed and identified in the scans were determined by concentration measurements using synthetic Zr solutions of increasing concentrations.For lines suffering from inter- ferences caused by the Zr matrix correction coefficients A and B were calculated according to the equation cEl= Bczr + A where cZr is the concentration of Zr (%) and cEl the concen- tration (pg ml-l) measured at the analyte line in the presence of the Zr matrix. Calibration was performed using the Zr matrix stock solution with addition of increasing amounts of the analytes. The concentrations of Zr and the analytes were adjusted to the concentrations expected in the samples. The analytical programme established was tested by its application to the SRMs. Each standard sample was dissolved 12 times and analysed. The mean values of the concentration and the confidence intervals for 95% probability were calculated.Results and Discussion The concentrations of the unwanted elements in Zr materials are very low and therefore the influence of the Zr spectrum on the analyte lines was investigated systematically. For this purpose spectra were measured using synthetic standard samples. For line selection the 28 spectral lines listed in Table 2 were checked and eventually 19 lines were selected for analytical purposes (Table 3). Also shown in Table 3 are the detection limits and BEC values obtained for pure aqueous solutions and for comparison purposes in the Zr matrix used for the investigations. For better comparison the ratios of the DLs in water and in the Zr solution are also listed. It is evident that the DLs in the Zr matrix are worse than those obtained for pure aqueous solutions of the single-element standards.Element A1 B Cd c o Cr cu Fe Hf Mg Mn Mo Nb Ni Sn Ta Ti V U W Wavelength/ nm 308.21 5 182.589 214.438 230.786 205.552 224.700 239.562 273.876 279.553 293.930 204.59 8 309.41 7 221.647 181.110 267.590 334.900 311.838 290.828 216.632 DL*/ pg ml-l 0.028 0.016 0.006 0.042 0.019 0.014 0.027 0.055 0.001 0.009 0.174 0.034 0.030 0.071 0.160 0.044 0.004 0.084 0.293 BEC*/ pg ml - 2.73 0.59 0.3 1 3.51 0.9 1 0.81 1.38 3.60 0.06 1.24 4.78 1.14 0.93 3.54 11.56 0.67 0.70 7.62 21.76 DL-F/ pg ml-' 0.061 0.021 0.01 5 0.065 0.032 0.04 1 0.028 0.110 0.001 0.029 0.268 0.068 0.032 0.123 0.323 0.083 0.020 0.112 0.743 BECtIl m1- 3.97 1.12 0.68 3.91 1.70 2.34 1.72 4.01 0.06 1.77 8.80 1.93 1.61 7.54 12.17 1.02 1.17 8.81 22.90 DL* DLt 0.46 0.76 0.40 0.65 0.59 0.34 0.96 0.50 1 .00 0.31 0.65 0.50 0.94 0.58 0.50 0.53 0.20 0.75 0.39 * Water.t Matrix.JOURNAL OF ANALYTICAL ATOMIC SPECTROMETRY JULY 1994 VOL. 9 787 Some of the scans obtained are shown in Figs. 1-6. In each figure the measurements performed with a pure Zr matrix are compared with scans using the same matrix solution but with the addition of 10 pg ml-I of the elements of interest. The A1 line at 308.215 nm (Fig. 1) shows a simple background shift and a complex line overlap. This simple 'flat' background can be corrected by appropriate selection of the background correc- tion position (see Table 4). Because of the complex line overlap calibration should be performed in the Zr matrix. The B line at 182.587 nm (Fig. 2) was selected because the more sensitive B line (182.641 nm) suffers from additional interference caused by the S line at 182.62 nm.Therefore the background correc- tion position was also selected at the lower wavelength side of the peak. For this line matrix interference was observed. The Cd line at 214.438 nm represents an interference-free line for the Zr matrix. The Co line at 230.786nm was chosen for the determination of Co in the Zr matrix. The Cr line at 205.552 nm (Fig. 3) was chosen for the determination of this trace element in the Zr matrix after application of matrix interference correc- tion. The Cu line at 224.700 nm is an interference-free spectral line for the Zr matrix. For the determination of Fe the line at 239.562nm was selected. The selection of a suitable Hf line 308.14 308.18 308.22 308.26 308.30 308.16 308.20 308.24 308.28 Wavele ngt hln rn Fig. 1 Scan around the 308.215 nm A1 line A HCl; and C Zr matrix+ 10 mg 1-l of A1 50 I -3j 40 i i I I I B ' I '7 ' I I / I I I \ I I \ I I \ I I 3 Zr matrix; 182.52 182.56 182.60 182.64 182.54 182.58 182.62 182.66 Wavelengt hln m Fig.2 Scan around the 182.589 nm B line A Zr matrix; and B Zr matrix + 10 mg 1-l of B 25 1 .- 3 201 3 P B :, * I 0 I I I t 1 I I 205.48 205.52 205.56 205.60 205.50 205.54 205.58 205.62 Wavelengthlnm Fig. 3 Scan around the 205.552 nm Cr line A Zr matrix; and B Zr matrix+ 10 mg I-' of Cr 600 ; 500 Y .- 5 3 L. 400 E e .- < 300 w .- m al .- E 200 - m C 0 i7j 100 0 309.34 309.38 309.42 309.46 309.50 309.36 309.40 309.44 309.48 Wavelengthlnm Fig.4 Scan around the 309.417 nm Nb line A Zr matrix; and B Zr matrix+ 10 mg 1-1 of Nb 25 I ; I \ ' I i l l 0 221.58 221.62 221.66 221.70 221.60 221.64 221.68 221.72 Wavelengt hln m Fig. 5 Scan around the 221.647 nm Ni line A Zr matrix; and B Zr matrix+ 10 mg 1-' of Ni788 JOURNAL OF ANALYTICAL ATOMIC SPECTROMETRY JULY 1994 VOL. 9 was of special interest because of the importance of the efficient separation of Zr and Hf (as discussed previously). The line at 273.876 nm was found to be suitable for the determination of Hf in Zr matrices. The Mg line at 279.553 nm is a very sensitive 0 181.04 181.08 181.12 181.16 181.06 181.10 181.14 181.18 Wavelengthlnm Fig.6 Scan around the 181.110nm Sn line A Zr matrix; B Zr matrix+ 10 mg I-' of Sn; and C 250 mg 1-' of Sn Table 4 Background correction positions for the lines selected Element A1 B Cd c o Cr c u Fe Hf Mn Mo Nb Ni Sn Ta Ti v U W Mg Wavelength/ nm 308.215 182.589 214.438 230.786 205.552 224.700 239.562 273.876 279.553 293.930 204.598 309.417 221.647 181.110 267.590 334.900 311.838 290.828 216.632 Background correction position/ nm + 0.060 - 0.025 - 0.030 + 0.048 + 0.036 + 0.048 + 0.036 - 0.036 + 0.036 + 0.036 + 0.036 - 0.030 + 0.040 - 0.048 + 0.036 - 0.048 - 0.030 + 0.034 - 0.0182 Table 5 Spectral lines tested in the Zr matrix but not selected for the analysis line without spectral interferences in the matrix investigated.The Mn line at 293.930nm was selected as it offers the possibility of background correction and compared with other Mn lines investigated the surroundings are relatively inter- ference free.Similar to the Mn 293.930nm line the intensity of the surroundings of the Mo line at 204.598 nm are almost constant. This offers a good possibility for background correc- tion on both sides of the peak Of all the Nb lines investigated the line at 309.417 nm (Fig. 4) shows the best spectral charac- teristics independent of the possible influences of the peak- wing interference produced by a Zr line at 309.507 nm. This interference had to be corrected for. Other Nb lines exhibit lower detection power or suffer from even stronger inter- ferences. The scan around the Ni line at 221.647 nm is shown in Fig. 5. The small peak on the right side of the Ni peak was identified as a spectral line of Si at 221.669 nm. This interference was corrected for in the analytical programme.The Sn line selected (181.110 nm Fig. 6) shows only weak interferences and relatively poor detection power. Of all the Ta lines investigated the line at 267.590 nm showed the best detection limit and only weak interferences. The V line at 311.838 nm is practically free from interferences. Of all the U lines investi- gated only the spectral line at 290.828 nm shows relatively good properties suitable for analytical purposes. The results of the investigations of W lines were similar to those of U. The W line at 216.632 nm was finally selected for determination in the Zr matrix. The scans were also used to select the background correction positions of the spectral lines chosen (Table 4). The behaviour of the lines tested for their applicability for measurements in the Zr matrix but not selected is shown in Table 5.For the A1 (308.215 nm) B (182.589 nm) Cr (205.552 nm) Nb (309.417nm) Ni 221.647nm and Sn 181.110nm lines interferences caused by the Zr matrix were observed The necessity for on-peak corrections appears to be due to both partial spectral overlap and changes in the sensitivity of the analytical calibration with respect to changes in concentration of the matrix. Correction coefficients were calculated by a linear approach. The correlation coefficients vary in the range The detection limits calculated for solids using the DLT values from Table 3 and the mean values of the concentrations obtained for the NIST SRMs after 12 determinations are shown in Table 6. Unfortunately not all of the 19 elements investigated were certified in the SRMs.The DLs calculated vary between 1.4 pg g-' for Fe and 13 pg g-' for Mo. For W the DL is 37 pg g-' which is extremely high compared with the other elements nevertheless a value of 44 pg 8-l with a confidence interval of 6.8 % was obtained for the determination of W in SRM 1239. The detection limit for U was not sufficient for a direct determination in SRM 360a. Similar to the determination of U in many complex matrices a preconcen- tration step could be advantageous. For the determination of 0.985-0.999. Element Cd c o Cr Fe Mn Mo Ta v W Wavelength/nm 226.502 230.786 357.869 238.204 294.920 28 1.61 5 226.230 309.3 11 208.819 Type of interference Not serious Two interfering peaks Intense line wing Line overlap Line wing and line Matrix Line Complex line overlap Line Comment Usable Not usable Usable (limited) Not usable Not usable overlap Usable (limited) SBR* poor Not usable Difficult to correct * SBR signal-to-background ratio.JOURNAL OF ANALYTICAL ATOMIC SPECTROMETRY JULY 1994 VOL.9 789 Table 6 Analytes determined in the Zr SRMs from NIST Element c u Cr Fe Hf Mn Mo Ni Sn (YO) Ti U W DL*/ 2.1 3.3 1.4 5.5 1.5 1.6 6.2 4.2 5.6 Pg 8-l 13 37 SRM 360a SRM 1238 Certified 140.00 1060.00 144 1 .OO 3.00 554.00 1.42 27.00 0.15 - - - ~ Found 138.0+ 6 1048.0 & 44 1450.0 f 30 2.8 k 0.1 538.0 f 18 1.4 & 0.1 5 24.0 + 2 t D L - - - Certified 60 580 2500 178 60 120 100 100 90 - - Found 62k4 592 f 30 2420 & 100 180f6 58$2 124$-5 110i-5 95f5 95+4 - - Certified 130 1055 2300 77 50 45 45 40 45 - - Found 129&8 1050 & 42 2180&80 74&4 48k2 40+4 40+5 38+3 44f3 - - * Detection limits calculated in the solid form using DLf values from Table 3.all other elements certified in the NIST SRMs the confidence intervals for 95% probability vary from 2 to 12% depending on the elements and the concentration. The data obtained were in good agreement with the certified values. Conclusion Spectral interferences can seriously limit the performance of emission spectrometric analyses. The idealized DLs should not be over-emphasized. Detection limits conventionally deter- mined on blank matrices are not necessarily applicable to the samples themselves. The most important decision in AES is the selection of appropriate analysis lines and wavelength@) position(s) for the background measurements. Even with the use of modern computer technology these decisions must still be made by the analyst. In some cases the choice of the interference-free analyte lines is not possible and these diffi- culties must be overcome by applying interference corrections in the course of the analytical programme or even by appli- cation of separation techniques for the analytes. This work was sponsored by the Fonds zur Forderung der Wissenschaftlichen Forschung Vienna Austria project no. L j795. References Schemel J. H. Metal Handbook American Society for Metals OH USA 9th edn 1980 vol. 3. Inductively Coupled Plasma Emission Spectroscopy. Part I ed. Boumans P. W. J. M. John Wiley New York 1987. Boumans P. W. J. M. in Inductively Coupled Plasma Emission Spectroscopy. Part I ed. Boumans P. W. J. M. John Wiley New York 1987 p. 22. Boumans P. W. J. M. Line Coincidence Tables for Inductively Coupled Plasma Atomic Emission Spectrometry Pergamon Press Oxford 1984. Winge R. K. Fassel V. A. Peterson V. J. and Floyd M. A. Inductively Coupled Plasma Atomic Emission Spectrometry. An Atlas of Spectral Information Elsevier Amsterdam 1984. Parsons M. L. Forster A. and Anderson D. An Atlas of Spectral Interferences in ICP Spectroscopy Plenum New York 1980. Paper 4/00171 K Received January 11 1994 Accepted March 31 1994

ISSN:0267-9477    

DOI:10.1039/JA9940900785

出版商:RSC    

年代:1994

数据来源: RSC