摘要:

THE ANALYSTTHE ANALYTICAL JOURNAL OF THE CHEMICAL SOCIETYEDITORIAL ADVISORY BOARD"Chairman: J. M. Ottaway (Glasgow)R. Belcher (Birmingham)L. J. Bellamy, C.B.E. ( Waltham Abbey)L. S. Birks (U.S.A.)E. Bishop (Exeter)L. R. P. Butler (South Africa)E. A. M. F. Dahmen (The Netherlands)A. C. Docherty (Bil/ingham)D. Dyrssen (Sweden)W. T. Elwell (Birmingham)J. Hoste (Belgium)H. M. N. H. Irving (feeds)M. T. Keiley (U.S.A.)W. Kemula (Poland)"J. H. Knox (Edinburgh)G. W. C. Milner (Harwell)*H. J. Ciuley (Wemh'ey)'P. Gray (Leeds)G. H. Morrison (U.S.A.)H. W. Nurnberg (West Germany)E. Pungor (Hungary)D. I. Rees (London)"R. Sawyer (London)P. H. Scholes (Sheffield)"W. H. C. Shaw (Greenford)S. Siggia (U.S.A.)"D. Simpson (Thorpe-/e-Soken)A. A.Smales, O.B.E. (Harwell)*A. Townshend (Birmingham)A. Walsh (Australia)T. S. West (Aberdeen)A. L. Wilson (Medmenham)P. Zuman (U.S.A.)*G. E. Penketh (Billingham)"J. Whitehead (Stockton-on- Tees)*Members of the Board serving on The Analyst Publications CommitteeREG I0 NAL ADVl SO RY ED I T 0 RSDr. J. Aggett, Department of Chemistry, University of Auckland, Private Bag, Auckland, NEW ZEALAND.Professor G. Ghersini, Laboratori CISE, Casella Postafe 3986,201 00 Mifano, ITALY.Professor L. Gierst, Universit6 Libre de Bruxelles, Facult6 des Sciences, Avenue F.-D. Roosevelt 50,Professor R. Herrmann. Abteilung fur Med. Physik., 63 Giessen, Schlangenzahl 29, W. GERMANY.Professor W. A. E. McSryde, Faculty of Science, University of Waterloo, Waterloo, Ontario, CANADA.Dr.W. Wayne Meinke, KMS Fusion Inc., 3941 Research Park Drive, P.O. Box 1567, Ann Arbor,Dr. 1. Rubegka Geological Survey of Czechoslovakia, Kostelni 26, Praha 7, CZECHOSLOVAKIA.Dr. J. R65i6ka. Chemistry Department A, Technical University of Denmark, 2800 Lyngby, DENMARK.Professor K. Saito, Department of Chemistry, Tohoku University, Sendai, JAPAN.Dr. A. Strasheim, National Physical Research Laboratory, P.O. Box 395, Pretoria. SOUTH AFRICA.Bruxelles, BELGIUM.Mich. 481 06, U.S.A.Published by The Chemical SocietyEditorial: The Director of Publications, The Chemical Society, Burlington House,London, W1 V OBN. Telephone 01 -734 9864. Telex No. 268001Advertisements: Advertisement Department, The Chemical Society, Burlington House, Piccadilly,London, W1 V OBN. Telephone 01 -734 9864Subscriptions (non-members) : The Chemical Society, Distribution Centre, Blackhorse Road,Letchworth, Herts., SG6 1 HNVolume 103 No 1232 November 1978Q The Chemical Society 197

ISSN:0003-2654    

DOI:10.1039/AN97803FX041

出版商:RSC    

年代:1978

数据来源: RSC

摘要:

ANALAO 103 (1 232) 1089-1 184 (1 978)ISSN 0003-2654November 1978THE ANALYSTTHE ANALYTICAL JOURNAL OF THE CHEMICAL SOCIETYCONTENTS1089 Beam-shaped Electrothermal Graphite Tube Furnace for Atomic-absorptionSpectrophotometry-P. Frigieri and R. Trucco11 00 Interference of Calcium on Barium as a Means o f Assessing Atomic-absorptionSpectrophotometers-R. C. Rooney and J. F. Woolley11 04 Direct Determination of Ammonium-nitrogen by Flame Emission Spectro-metry in a Hydrogen - Nitrogen Diffusion Flame-J. M. S. Butcher andG. F. Kirkbright11 16 Spectrophotometric Determination of 6,7- Dihydroxycoumarin and Its MethoxyDerivatives-E. Celon, A. Guiotto and P. Rodighiero11 21 Spectrophotometric Determination of Methylmercury i n Fish Tissue w i t hDithizone Using a Dual-wavelength Procedure-P.Jones and G. NicklessI 1 27 Determination of a Non-volatile N-Nitrosamine on a Food Matrix-C. L. Walters,M. J. Downes, M. W. Edwards and P. L. R. Smith1134 Sulphonated Alizarin Fluorine Blue (AFBS). Part IV. A Critical Comparisonof the Use of AFBS Against Alizarin Fluorine Blue (AFB) and the FluorideElectrode for the Determination of Low Fluoride Concentrations: Inter-ferences w i t h the AFBS Method and Their Removal-S. F. Deane, M. A.Leonard, Victoria McKee and G. SvehlaN-Substituted Phenothiazines as Redox Indicators in Titration w i t h N-Bromo-succinimide-H. Sanke Gowda and S. Akheel AhmedDetermination of Readily Oxidised Compounds by Flow Injection Analysis andRedox Potential Detection-Bo Karlberg and Sidsel ThelanderEnzymic Digestion of Liver Tissue t o Release Barbiturates, Salicylic Acid andOther Acidic Compounds in Cases of Human Poisoning-M.D. Osselton,I. C. Shaw and H. M. StevensDetermination o f a 24-hour Time-weighted Average Value o f EnvironmentalVinyl Chloride Monomer Concentration by Charcoal Adsorption Followedby Gas Chromatography-B. Miller, P. 0. Kane, D. B. Robinson and P. J.Whittingham114811 54116011 65SHORT PAPERS11 73 Colour Reactions in the Mixed-ligand Chelates Zirconium(1V) - Ethylene-diaminetetraacetic Acid - Pyrocatechol Violet and Thorium(lV) - Diethyl-enetriaminepentaacetic Acid - Pyrocatechol Violet-B. Maiti and R. M. SatheEffect o f Iron on the Determination of Proline and Hydroxyproline i n Formalin-fixed Coalworkers' Lungs-Roy LoxleyDetermination o f Trace Amounts of Indium i n Minerals and Rocks by EmissionSpectroscopy-Greta M.Eskenazy and Ekaterina I. Mincheva11 7611791182 Book ReviewsSummaries of Papers in this Issue-Pages iv, vi, vii, x, xi, xiiiPrinted by Heffers Printers Ltd Cambridge EnglandEntered as Second Class at New York, USA, Post OfficAnnual Reports on AnalyticalAtomic SpectroscopyVOLUME 6,1976This comprehensive and critical report of developments in analytical atomicspectroscopy has been compiled from over 1650 reports received fromworld-wide correspondents who are internationally recognised authorities inthe field and who constitute the Editorial Board. i n addition to surveyingdevelopments throughout the world published in national or internationaljournals, a particular aim has been to include less widely accessible reportsfrom local, national and international symposia and conferences concernedwith atomic spectroscopy.Paperbound 282pp 8s" x 6" f 18 (CS Members f 13.50)(Still available: Vols.3-5 covering 1973 to 1975)Obtainable from: The Chemical Society, Distribution Centre,Blackhorse Road, Letchworth, Herts., SG6 1 H NNOTICE TO SUBSCRIBERS(other than Members of the Society)Subscriptions for The Analyst, Analytical Abstracts and Proceedings shouldbe sent to:The Chemical Society, Distribution Centre,Blackhorse Road, Letchworth, Herts., SG6 1 HNRates for 1978The Analyst, Analytical Abstracts and Proceedings (including indexes):(a) The Analyst, Analytical Abstracts and Proceedings . . . . .. . . €99.00( b ) The Analyst, Analytical Abstracts printed on one side of the paper, andProceedings . . . . . . . . . . . . . . . . . . €105.00The Analyst and Analytical Abstracts without Proceedings (including indexes):( c ) The Analyst, and Analytical Abstracts . . .. .. . . . . . . f87.00(d) . . f93.00 The Analyst, and Analytical Abstracts printed on one side of the paper(Subscriptions are NOT accepted for The Analyst and/or for Proceedings alone)Analytical Abstracts only (two volumes per year. including indexes):(e) Analytical Abstracts . . .. .. .. . . .. . . . . f67.00(f) Analytical Abstracts printed on one side of the paper . . .. .. . . f73.0

ISSN:0003-2654    

DOI:10.1039/AN97803BX043

出版商:RSC    

年代:1978

数据来源: RSC

摘要:

iv SUMMARIES OF PAPERS I N THIS ISSUE November, 197sSummaries of Papers In this IssueBeam-shaped Electrothermal Graphite Tube Furnace forAtomic-absorption SpectrophotometryCommercial graphite tubes (Massmann type) generally have a cylindricalshape and only a portion of the internal volume is available for absorption,owing t o the geometry of the focused light beam. The aim of this work wasto find the best geometry of the atomisation chamber to exploit the absorbingpower of the optical system.Starting from the geometry of the light beam of an atomic-absorptionspectrophotometer (Pye Unicam SP1900), the size and geometry of the tubehave been defined. The temperature distribution along the axis of the tube,the diffusion of the atomic vapour from the centre to the ends of the tubeand the residence time of the atoms within the tube were studied to determinetheir influence on the analytical sensitivity.As theoretical calculations demonstrated that the improved tube shouldgive an appreciable enhancement of the sensitivity, a graphite tube wasconstructed according to this new geometry.The experimental resultsobtained with different elements confirmed the expected improvement in theanalytical performance.Keywords : A tomic-absovption spectvophotounetvy ; electvothevmal niomisev ;beam-shaped graph ite tubeP. FRIGIERI and R. TRUCCOCentro Informazioni Studi Esperienze, P.O. Box 3986, 201 00 Milan, Italy.Andysi, 1978, 103, 1089--1099.Interference of Calcium on Barium as a Means of AssessingAtomic-absorption SpectrophotometersThe well documented interference of calcium on the atomic-absorption deter-mination of barium has been studied by using a number of different com-mercially available instruments.The widely differing results obtained onvarying the calcium to barium ratios using these instruments are thought toindicate that the interference is essentially an instrumental artefact, andshould be largely avoidable.Keywovds: Atomic-nbsorptioiz spectvophotonzetry; instramentation; bariztmdetermination ; calcium interfevenceR. C. ROONEYRooney and Ward Ltd., Blackwater Station Estate, Cambcrlcy, Surrey, G U l i 9XF.and J. F. WOOLLEYStandard Telecommunication Laboratorics Limited, London Road, Harlow, Essex,CM17 9NA.A g d y s t , 1975, 603, 1100-1103.Direct Determination of Ammonium-nitrogen by Flame EmissionSpectrometry in a Hydrogen - Nitrogen Diffusion FlameA method for the determination of trace amounts of ammonium ion in aqueoussolution has been developed that utilises the chemiluminescent emission at336.0nm of the NH species produced when ammonia gas generated fromalkaline sample solutions is introduced into a hydrogen - nitrogen diffusionflame.The method has been applied successfully to the determination of theexchangeable ammonium-nitrogen content of soils.A practical detectionlimit of 0.2 pgml-l has been obtained for a 5-ml sample.Keywords ; Ammonium-nitrogen detevnzination ; NH emission spectvonzetyy ;hydrogen - nitrogen di,ffusion flame ; soil analysisJ.M. S. BUTCHER and G. F. KIRKBRIGHTDepartment of Chemistry, Imperial College, London, SW7 2AY.Analyst, 1978, 103, 1104-1115November, 197'8 THE ANALYST VANALYSIS 79Atomation in Industrialand Clinical ChemistryThe next Analysis conference will be held at The City University,London on 16th-18th July, 1979.This conference will focus attention on the cross-fertilization ofideas and concepts of automation between workers in clinical,industrial and academic environments.Automation implies a system approach that can be applied at alllevels of the analytical procedure and when applied in a completesense has in addition to a technical and scientific impact aninfluence on managerial, organisational and economic considera-tions. The conference will focus attention on all these aspectswithin the general framework of the following sessions:- Education- New Instrumentation- Costing and Management- Applications- StandardisationThe programme committee comprising Dr.Peter Stockwell,Dr. Fred Mitchell, Derrick Porter and Fred Fearn, have invited aninternational team of authoritative speakers to provide the keynotelectures and to chair the various sessions. Whilst the majority ofspeakers have been invited, there will be sessions for short, sub-mitted papers. Authors wishing to present papers in these areasshould submit abstracts for consideration by the programmecommittee before 30th November 1978.For further details on Analysis 79 contact Beverly Humphrey,Scientific Symposia Ltd., 33/35 Bowling Green Lane, LondonECI R ODA.Tel: 01 837 121 2vi SUMMARIES OF PAPERS I N THIS ISSUEand Its Methoxy DerivativesNovember, 1978Spectrophotometric Determination of 6,7-DihydroxycoumarinA spectrophotomctric method for the determination of the single componentsin a mixture of 6,7-dihydroxycouniarin and its methoxy derivatives isdescribed. The method is based 011 the determination of 6,7-dihydroxy-coumarin after complexation with molybdophosphoric acid in bufferedsolution and of 6-hydroxy-7-meth~oxy-, 7-hydroxy-6-methoxy- and 6,7-dimethoxycoumarin by treatment with sodium methylate solution followedby difference absorbance measurements. This procedure has also beenapplied successfully to the study of the methylation reaction of 6,7-dihydroxy-coumarin.Keywovds : 6,7-Dihz,droxycounzarin ; ~~-1~ydroxy-7-methoxycoz~~znvin ; 7-lzydvozy-6-methoxycoumavi.tz ; 6,7-dimethoxycouvnavin ; spectvophotometvyE.CELONIstituto di Chimica Organica dell’Univer:;it&, Via Marzolo, 1, 35100 Padova, Italy.A. GUIOTTO and P. RODIGHIEROCentro di Studio della Chimica del Farmaco e dei Prodotti Biologicamente attivi delCNR, Istituto di Chimica Farinaceutica dell’Universit&, Via Marzolo, 5, 35 100 Padova,Italy.Analyst, 1978, 103, 1116-1120.Spectrophotometric Determination of Methylmercury in Fish Tissuewith Dithizone Using a Dual-wavelength ProcedureA ditliizone spectrophotometric procedure is described for the measurementof trace concentrations of methylmercury salts.The application of a simpleequation using absorbance measurements taken a t two wavelengths cancels outsmall differences in excess of dithizone arising between blank and sample, thusensuring good precision in the range 0.1-4.0 p g ml-l.The developed method is used in combination with the Westoo extractionprocedure to determine methylmercury concentrations in fish tissue. A crabmeat sample contained less than 0.04 p g g-l, and values for eight tuna fishranged from 0.08 to 0.41 p g g-l.Keywovds : Methylmercury detevmination ; fish tissue analysis ; dual-wavelength spectroplzotometry ; ditltnzoneP. JONESDepartment of Chemistry, University of the West Indies, St. Augustine, Trinidad.and G. NICKLESSDepartment of Inorganic Chemistry, School of Chemistry, The University, Bristol,BS8 1TS.Analyst, 1978, 103, 1121-1126November, 1978 SUMMARIES OF PAPERS I N THIS ISSUISDetermination of a Non-volatile N-Nitrosamine on a Food Matrix1711A method devised for the determination of N-nitrososarcosine, in which theN-nitrosamine in solution is denitrosated with hydrogen bromide to formvolatile products that are rapidly removed and determined in a chemi-luminescence anal yser, has been applied successfully to the same compoundon powdered corn flakes. Ilifferentiation of N-nitrososarcosine and a numberof other N-nitrosamines and N-nitrosamides from inorganic nitrite wasachieved by decomposing the nitrite with acetic acid prior to the denitrosa-tion of the N-nitroso compounds. In the presence of a secondary aminereceptor limited nitrosation can occur during the process of differentiationbut this can be prevented through the use of ascorbyl palmitate.Indifferentiating between large amounts of nitrite and much lower levels ofN-nitrososarcosine on corn flakes, using a chemiluminescence analyser, theduration of the response from the nitrite can be shortened by freeze-dryingthe food matrix in the presence of ascorbic acid. The spectrophotometricdetermination of N-nitrososarcosine as nitrosyl bromide released into solutionby the action of hydrogen bromide was hindered by the presence of powderedcorn flakes.Keywords : N-Nitroso compound determination ; N-Nitrososarcosine deterunin-ation; food analysis ; chevniluvninescence analysevC. L.WALTERS, M. J. DOWNES, M. W. EDWARDS and P. L. R. SMITHBritish Food Manufacturing Industries Research Association, Randalls Road,Leatherhead, Surrey, KT22 7RY.Analyst, 1978, 103, 1127-1133.Sulphonated Alizarin Fluorine Blue (AFBS)Part IV. A Critical Comparison of the Use of AFBS AgainstAlizarin Fluorine Blue (AFB) and the Fluoride Electrode for theDetermination of Low Fluoride Concentrations ; Interferenceswith the AFBS Method and Their RemovalThe AFBS and AFB positive absorptiometric methods for the determinationof fluoride in water were evaluated with respect to sensitivity, range, repro-ducibility, rate of complex formation and stability of the colours formed.The spectrophotometric methods were also compared statistically with theuse of the fluoride electrode over the same working range. The AFBS methodis shown to be slightly superior t o the parent AFB method and more repro-ducible than the electrode except a t very low concentrations. The interferenceof 24 common ions in the AFBS method was examined. Most interferingcations are removed by ion exchange ; aluminium and iron require alternativetreatment. The important anionic interferences of phosphate and sulphateare readily avoided. The interference study led to the identification of a1 : 1 : 1 lanthanum - AFBS - mercury ternary mixed-cation complex.Keywords : Sulphonated alizarin fluorine blue ; l a n t h a n u m ( I I I ) complexes ;fluoride determination ; jlaoride electrode ; ternary bimetallic complexesS. F. DEANE, M. A. LEONARD, VICTORIA MGKEE and G. SVEHLADepartment of Analytical Chemistry, The Queen’s University of Belfast, Belfast,BT9 5AG.Analyst, 1978, 103, 1134-1147

ISSN:0003-2654    

DOI:10.1039/AN97803FP101

出版商:RSC    

年代:1978

数据来源: RSC

摘要:

November, 1978 THE ANALYST ixSILICONES UNDER THE MONOGRAM:A Story of Industrial Researchby H.A. Liebhafsky, Texas A & M University(with Sybil S . Liebhafsky and G. Wise)Tracing the origin and evolution of the silicone industry under the General Electric monogram, thisvolume demonstrates the industrial research process and its consequences. It stresses howresearchers worked and were managed, and includes references to original laboratory notebooks thatshow how research progressed and how discoveries were made. Scientific information is included onthe union of chemistry (inorganic, organic, physical, analytical, and polymer) and chemical engineeringwhich helped to create the silicone industry.0471 04610 8 approx. 400 pages In Press approx. $34.95/f 19.40ANALYSIS WITH ION-SELECTIVE ELECTRODESby J .Vesely and D. Weiss, Geological Survey, Pragueand K. Stulik, Charles University, PragueTranslation Editor: R.A. ChalmersThe simplicity of the apparatus and measuring technique of ion-selective electrodes has made theirapplication a growth field. This practical and up-to-date survey covers all aspects of the method, withemphasis on analytical uses. It gives detailed descriptions of experimental techniques, and discussesentire procedures from sample preparation and measurement to result-handling. {Ellis HorwoodSeries in Analytical Chemistry; Editors: R.A. Chalmers and Mary Masson, University of Aberdeen)08531 2 092 7 approx. 240 pages In Press approx. $33.90/f 16.00Published by Ellis Horwood Ltd., Chichester, and distributed by John Wiley & Sons Ltd.~ ''LABORATORY HANDBOOK OF CHROMATOGRAPHIC ANDALLIED METHODSedited by 0.Mikes, Czechoslovak Academy of Sciences, Prague.Provides a comprehensive survey of a wide range of the modern laboratory separation methods mostfrequently in use. The book does not specialise in any one chemical branch but covers separationmethods in organic, inorganic, analytical and preparative chemistry, the study of natural compoundsand biochemistry. The unifying feature throughout is the continued orientation to laboratory use (€11;~Horwood Series in Analytical Chemistry; Editors: R.A. Chalmers and Mary Masson, University ofAberdeen)08531 2 080 3 approx. 820 pages In Press approx. $80.55/f38.00Published by Ellis Horwood Ltd., Chichester, and distributed by John Wiley & Sons Ltd.SULFUR IN THE ENVIRONMENT Part 2: Ecological Impactsedited by J.O.Nriagu, Canadian Center for Inland Waters, Ontario.A comprehensive interdisciplinary volume of articles, written by leading international experts, thatfocuses on the chemical behaviour and biological effects of pollutant sulphur in the biosphere. Part 2includes chapters on the biological, ecological, and health significance of sulphur pollution.(Environmental Science and Technology Series)0471 042552 approx. 496 pages In Press approx. $32.00/f 17.10Available from all good booksellers or from Wiley. If you wish to use American Express, Diners Club,Bar& ycard or Access, please quote your card and number.~ X SUMMARIES OF PAPERS I N THIS ISSUEN- Substituted Phenothiazines as Redox Indicators inTitrations with N-BromosuccinimideNovember, 1978Butaperazine dimaleate, trifluoperazine dihydrochloride, diethazine hydro-chloride, promethazine hydrochloride, prochlorperazine maleate and chlor-promazine hydrochloride are proposed as redox indicators in the macro- andmicro-titration of hydroquinone, Met01 (N-methyl-4-aminophenol sulphate)and ascorbic acid with N-bromosuccinimide in hydrochloric, sulphuric andacetic acids. They give very sharp and reversible colour changes a t theequivalence point. A simple titrimetric method for the determination ofhydroquinone, Metol and ascorbic acid and a potentiometric method for thedetermination of Metol are described.Keywords : N-Substituted phenothiazine redox indicators ; hydvoquinone,Metol and ascorbic acid determination ; titrimetry ; potentiometry ; N-bvomo-succiniwzideH. SANKE GOWDA and S.AKHEEL AHMEDDepartment of Post-graduate Studies and Research in Chemistry, Manasa Gangotri,University of Mysore, Mysore-570006, India.Analyst, 1978, 103, 1148-1153.Determination of Readily Oxidised Compounds by Flow InjectionAnalysis and Redox Potential DetectionRedox potential detection has been applied to the determination of reducingagents injected into a non-segmented stream of cerium(1V) solution. Bydispersion the plug of injected sample was mixed with the reagent and theconsequent redox potential shift, determined by the ratio of cerium(1V) tocerium(III), was detected by a platinum or graphite electrode with suitablyplaced indicator and reference electrodes.The flow-rate of the reagentstream was about 0.7-1.0 ml min-l. With 0.3-ml samples, well definedpeaks were obtained. Millimolar or even weaker solutions of iron(I1) andascorbic acid were analysed with good reproducibility, the sample rate beingabout 45-60 per hour.Keywords : Flow injection analysis ; automated analysis ; redox fiotentialdetection ; ascorbic acid determinationBO KARLBERG and SIDSEL THELANDERAstra Pharmaceuticals AB, Analytical Control, S-151 85 Sodertalje, Sweden.Analyst, 1978, 103, 1154-1159.Enzymic Digestion of Liver Tissue to Release Barbiturates,Salicylic Acid and Other Acidic Compounds in Cases ofHuman PoisoningThe proteolytic enzyme subtilisin Carlsberg has been used to digest humanliver tissues containing barbiturates, salicylic acid and a mixture of 2.4-dichlorophenoxyacetic and 2,4-dichlorophenoxypropionic acids.The levelsof the drugs released by this method were significantly higher than thosemeasured by using conventional procedures for tissue degradation. Theenzymic digestion of tissues is easy to manipulate, yields reproducible resultsand is inexpensive.Keywords : Liver tissue digestion ; enzymic digestion ; acidic drug isolationM. D. OSSELTON, I. C. SHAW and H. M. STEVENSHome Office Central Research Establishment, Aldermaston, Reading, Berkshire,RG7 4PN.Analyst, 1978, 103, 1160-1161November, 1978 SUMMARIES OF PAPERS IN THIS ISSUEDetermination of a 24-hour Time- weighted Value of EnvironmentalVinyl Chloride Monomer Concentration by Charcoal AdsorptionFollowed by Gas ChromatographyxiThe method described was developed under the auspices of the ChemicalIndustries Association Vinyl Chloride Monomer Analysis and MonitoringCommittee a t the request of the Chief Alkali Inspector.A peristaltic pumpdraws the atmosphere through a tube packed with activated charcoal whenvinyl chloride monomer is adsorbed on the charcoal. At the end of the 24-hsampling period, the charcoal is extracted with carbon disulphide and thevinyl chloride in the extract determined by gas chromatography. Themethod has a limit of detection of 0.005 p.p.m. V/V and a precision underlaboratory conditions of about f 10% relative.Interference by other knownor expected atmospheric contaminants is negligible. Its accuracy in thefield cannot be defined, but experiments using two different flow-ratessimultaneously showed agreement to about & 10% relative.Keywords : Vinyl chloride determination ; atmospheric monitoring; charcoaladsorption ; gas chromatographyB. MILLER, P. 0. KANE and D. B. ROBINSONImperial Chemical Industries Limited, Mond Division, Research and DevelopmentDepartment, The Heath, Runcorn, Cheshire, WA7 4QD.and P. J. WHITTINGHAMImperial Chemical Industries Limited, Plastics Division, P.O. Box No. 3 (ThorntonCleveleys) , Hillhouse Works, Blackpool, FY5 4QB.Analyst, 1978, 103, 1165-1172.Colour Reactions in the Mixed-ligand Chelates Zirconium(1V) -Ethylenediaminetetraacetic Acid - Pyrocatechol Violet andThorium( IV) - Diethylenetriaminepentaacetic Acid -Pyrocatechol VioletShort PaperKeywords : Spectrophotometry ; mixed-ligand chelates ; zirconium(I V ) -ethylenediaminetetvaacetic acid - pyrocatechol violet ; thorium(I V ) -diethylenetriaminepentaacetic acid - pyvocatechol violetB.MAITI and R. M. SATHEAnalytical Chemistry Division, Bhabha Atomic Research Centre, Trombay,Bombay-400 085, India.Analyst, 1978, 103, 1173-1176.Effect of Iron on the Determination of Proline and Hydroxyprolinein Formalin-fixed Coalworkers’ LungsShort PaperKeywords : Proline determination ; hydroxyproline determination ; iron inter-ference ; clinical analysis ; coal miningROY LOXLEYHealth and Safety Executive, Safety in Mines Research Establishment, Red Hill,Sheffield, S3 7HQ.Analyst, 1978, 103, 1176-1179sii THE ANALYST November, 1978ANALYTICAL SCIENCES MONOGRAPHSHigh-Precision Titrimetryby C.Woodward and H. N. RedmanThis monograph was written in the hope that it will prove both helpful and interesting topractising analytical chemists.Brief contentsThe first section, on visual titrations, covers apparatus, preparation and assay of standardsubstances and preparation of standard solutions.The second section deals with instrumented titrations, including photometric and electro-metric techniques as well as miscellaneous instrumented Methods.There are 83 key references to the literature on high-precision titrimetry.Paperbound 71 pp 8%” x 6” 0 85990 501 2 f 2.50 ($5.50)CS Members f2.00The Chemical Analysis of WaterGeneral Principles & Techniquesby A.L. WilsonThe volume covers all stages of the complete analytical process including: deciding on theanalytical information required; sampling, including place, time and frequency, as well asdevices and techniques; the analysis proper and the reporting of results, their statisticaltreatment, and the factors involved in the choice of analytical methods (including on-lineand automatic methods) for particular purposes; and data handling.Clothbound 196pp f7.50 ( $1 6.50)CS Members f5.7582’’ x 6$” 0 85990 502 0Pyrolysis-Gas Chromatographyby R. W. May, E. F. Pearson and D. ScothernMany papers have been published, particularly over the past decade, on aspects of pyrolysis-gas chromatography.A large number of different types of apparatus have been used, on awide range of samples. This monograph attempts to present the available knowledge in aform useful to the practising analyst, helping in the choice of an appropriate method and inthe avoidance of the more common pitfalls in this, perhaps deceptively, simple technique.Clothbound 11 7pp €7.20 ( $1 5.75)CS Members f5.508$” x 6” 0 851 86 767 7Electrothermal Atomization forAtomic Absorption Spectrometryby C. W. FullerSince the introduction of atomic absorption spectrometry as an analytical technique, byWalsh, in 1953, the use of alternative atomization sources to the flame has been explored.At the present time the two most successful alternatives appear to be the electrothermalatomizer and the inductively-coupled plasma.In this book an attempt has been made toprovide the author s views on the historical development, commercial design features, theory,practical considerations, analytical parameters of the elements, and areas of application ofelectrothermal atomization.Clothbound 135pp f 6.75( $1 4.75)CS Members f5.0082” x 52’’ 0 85186 777 4Dithizoneby H. M. N. H. IrvingThe author of this monograph, who has been closely associated with the development ofanalytical techniques using this reagent for many years, and who has made extensiveinvestigations into the properties of its complexes, has gathered together a body of historicaland technical data that will be of interest to many practising analytical chemists.Clothbound 11 2pp f7.25 ($1 6.00)CS Members f5.508%” x 5%” 0 851 86 787 1THE CHEMICAL SOCIETY,Distribution Centre, Blackhorse Road, Letchworth,Herts., SG6 IHN, Englandxiv THE ANALYST November, 1978ANALYTICAL SCIENCESMONOGRAPH No.5Dithizoneby H. M. N. H. IrvingThe author of this monograph,who has been closely associatedwith the development of analy-tical techniques using this re-agent for many years, and whohas made extensive investiga-tions into the properties of itscomplexes, has gathered togethera body of historical and technicaldata that will be of interest tomany practising analytical chem-ists.Brief contentsIntroductionThe Properties of DithizoneMetal-Dithizone Complexes andThe Photochemistry of MetalThe Extraction of MetalThe Less Familiar DithizoneOrgan ometal lie DithizonatesPractical ConsiderationsSome Additional Applications ofD i t h izo neSome Unresolved ProblemsBibliographyClothbound IlOpp 8:‘’ x 59”0 85186 787 1 f7.25 ($14.50)CS Members f5.50Their FormulaeDithizonatesD it h izonatesComplexesTHE CHEMICAL SOCIETYDistribution Centre, BlackhorseRoad, Letchworth, Herts.,SG6 1 HN, EnglandiRLAG CHEMIE-qOISnrn -4- O,OlSnrn--c/Wavelength ---+rtomic AbsorptionIpect roscopyy Bern hard Welzi i s volume is an English-language edition ofe second edition of a highly successful Germaniok on the technique.fter an introductory chapter on the physical-)ectroscopical principles, a knowledge of whichrequired for an understanding of the funda-entals of atomic absorption, instrumentationi d technique are treated in further chapters.ere, the analyst will find the necessary know-3w for complete command of the method.inally,short sectionsaredevoted tothe individualements and to various specific applications.omprehensive bibliographical references arerovided.The lucid, almost tabular, presentationowhere allows the reader to become lost iniinutia, and the narrow interweaving of theorynd practice will be greatly appreciated.;rief contentsitrod uction ; Light Sources ;Atom it at ion ; 0 ptics;lectronics and Readout; Technique; Related,natytical Methods; The Individual Elements;pecific Applications.:loth bound 277pp 95” x 7” 3 527 25680 6 f 20.00brders to: THE CHEMICAL SOCIETY,bistribution Centre, Blackhorse Road,.etchworth, Herts. SG6 1 HN

ISSN:0003-2654    

DOI:10.1039/AN97803BP107

出版商:RSC    

年代:1978

数据来源: RSC

摘要:

NOVEMBER 1978 The Analyst Vol. 103 No. 1232 Beam-shaped Electrothermal Graphite Tube Furnace for Atomic-absorption Spectrophotometry" P. Frigieri and R. Trucco Centro Informazioni Studi Esperienze, P.O. Box 3986, 20100 Milan, ItaZy Commercial graphite tubes (Massmann type) generally have a cylindrical shape and only a portion of the internal volume is available for absorption, owing to the geometry of the focused light beam. The aim of this work was to find the best geometry of the atomisation chamber to exploit the absorbing power of the optical system. Starting from the geometry of the light beam of an atomic-absorption spectrophotometer (Pye Unicam SP1900), the size and geometry of the tube have been defined. The temperature distribution along the axis of the tube, the diffusion of the atomic vapour from the centre to the ends of the tube and the residence time of the atoms within the tube were studied to determine their influence on the analytical sensitivity.As theoretical calculations demonstrated that the improved tube should give an appreciable enhancement of the sensitivity, a graphite tube was constructed according to this new geometry. The experimental results obtained with different elements confirmed the expected improvement in the analytical performance. Keywords : A tomic-absorption spectrophotornetry ; electrothermal atoiniser ; beam-shaped graphite tube An improvement in the sensitivity of analyses by atomic-absorption spectrophotometry with electrothermal atomisation can be achieved by optimising some characteristic parameters of the analytical system.The study of the atomisation chamber in relation to the geometry of the incoming light beam leads to significant increases in sensitivity, without losses in energy throughput, when consideration is given to the diffusion of the atomic vapour as well as to all the other phenomena occurring during the atomisation process. Consequently, studies relating to the shape of the absorption chamber can be useful in obtaining some improvements in the performance of the atomisation system. Geometry of an Atomisation Chamber The absorbance measured at any instant during the atomic-absorption process is directly proportional to the number of absorbing atoms in the path of the light beam. If the internal geometry of the chamber reproduces the shape of the light beam, point by point, then all of the atoms in the fundamental state can be excited by the primary radiation.In the well known atomisation furnaces (Massmann type),l the internal volume of the cylindrical graphite tube is larger than the volume of the double cone of light formed by the lenses or mirrors of the optical system. The light from the source impinges on only a limited number of atoms in the central volume of the furnace, where the atomic population is the greatest, and this results in a decrease in sensitivity. There would clearly be an advantage if the space available in the tube were reduced to the dimensions of the light beam. In practice, these conditions cannot be fully satisfied because the liquid sample, when in the electrothermal atomiser, needs a surface in the chamber able to retain the drop during the drying step.The internal geometry of the atomisation chamber that best fits these two requirements is then represented by a tube with a cylindrical section in the centre, long enough to contain a suitable sample volume, and with cones at its two ends reproducing the geometry of the light beam. * Presented at the XXth Colloquium Spectroscopicurn Internationale, Prague, August 30th to Septem- ber 7th, 1977. 10891090 FRIGIERI AND TRUCCO : BEAM-SHAPED ELECTROTHERMAL GRAPHITE Analyst, Vol. 103 The performances of this chamber and of the traditional chamber can be compared by means of theoretical calculations based on geometrical and physical parameters of the systems. For this purpose it is necessary to distinguish between the two most common methods for the determination of absorption values in atomic-absorption spectrophotometry by electrothermal atomisation, namely the measurement of the peak height or the peak area of the absorbance signal.The peak height is proportional to the number of atoms on which the source photons impinge. When the volume of the tube traversed by the light beam (analysis volume) does not coincide with the volume in which expansion of atomic vapours occurs, the height of the absorption peak is proportional to the ratio of these two volumes. The value of the absorption-peak area depends on the number of atoms and also on their residence time in the analysis volume. Therefore, both parameters must be taken into account when evaluating the expected performances of chambers with different internal geometries.Neglecting the discussion on volumes, as the effect of this parameter does not need further explanation, it is convenient to consider the atomic vapour residence time, which is the other important parameter in the evaluation of analytical sensitivity. Calculations of Atomic Vapour Residence Times As the aim of the work was a comparison between the performances of two possible geo- metries of the atomisation chamber under the same operating conditions, the calculations were carried out on an electrothermal atomiser working under static conditions (gas stop) in order to simplify the problem. Under such conditions, the losses of atomic vapours from the chamber can be associated with the following processes : diffusion through tube apertures and the porous wall, convection of the gaseous cloud due to heating of the tube wall and expulsion of excess of atomic vapours if their volume is too great with respect to the cell.The effects of the last two processes can reasonably be neglected when the volume of the vapours is less than 10% of the total volume of the tube and when there are no large apertures in the tube Hence, the longitudinal diffusion process of atomic vapours through the tube ends is the main factor to be considered. In a chamber of variable geometry as previously described (Fig. l), the total mass of atoms in the chamber at instant t is given by 1 -t m M(t) = 2jC(x, t ) S(x) dx . . .. .. * ' (1) 0 where C is the atomic concentration, as a function of time t and distance x from the centre, along the longitudinal axis of the tube, S is the area of cross-section of the tube, as a function of distance x, m is the length of the terminal cones and 21 is the length of the cylindrical central section.0 Fig. 1. Variable-geornetry atomisation chamber. The radius of the chamber, depending on the distance x from the centre, is R(x) = R, when 0 < x < I and R(x) = R, + (R, - A!,) (x - l)/m when I < x < 1 + m and the area of cross-section is S(x) = nR2, when 0 < x < 1 and S(x) = T[R, + (R, - R,) (x - Z)/mI2 when 1 < x < 1 + m. The concentration can be expressed as C(x, t ) = C&) = A (maximum concentration) when x = 0 at the centre of the tube, C(x, t ) = C,(t) = 23, when x = 1 at the ends of theNovember, 1978 TUBE FURNACE FOR ATOMIC-ABSORPTION SPECTROPHOTOMETRY 1091 cylindrical section and C(x, t ) = 0 when x = I + m at the ends of the tube.The concentra- tion, highest in the centre, decreases to C,(t) at the ends of the cylindrical section, and becomes zero at the tube ends. To simplify the calculation, the mass flux is assumed to be constant through each section at each instant: This relationship, applied to the cylindrical portion, becomes x vR: = K(t) dx .. .. .. * (3) The solution of the differential equation within the above-mentioned limits is Equation (2) applied to the conical portion gives x - z .. (5) dC dx K(t) = - x 7~ [Rl + (R, - Rl) . . From equations (4) and (5) and integrating m * (7) C,(t) - C,(t) Kl - c C ( X , t ) = I (cc - 1) where R,/R, = a and Kl = mlor.obtain According to condition B, C(Z, t ) = Cl(t) and, introducing x = I into equation (7), we where y = K,/(K, + I). Substituting the value for Cl(t) in equation (7), is valid for Z < x < Z + m, whereas substituting Cl(t) in equation (4) we obtain .. . . (10)1092 FRIGIERI AND TRUCCO BEAM-SHAPED ELECTROTHERMAL GRAPHITE A ?U?dySt, vd. 203 and integrating equation (3) over the cylindrical section, From the two last equations, it follows by substitution that .. .. . . (11) C(x, t ) = C,(t) 1 - ~ .. (1 K,: 1 ) ’. . ’ is valid for 0 4 x < 1. and the latter for the cone, and setting Equations (9) and (12) can be shown to coincide if x = 1. From the resolution of integral (1) into the sum of two integrals, the former for the cylinder .... .. .. . . (13) A = - - Y - - l 1 (a - 1) .. .. .. . . (14) B = l - - - ( a - 1) 1 m and a - 1 m E = - .. .. .. .. . . (15) the equation can be derived, where S(x) = ~TR:(B + a s is demonstrated by substitution of expressions (13), (14) and (15) in the expression for section S(x) for an x value ranging from I to I + m. Equation (16) can therefore be written as :follows: The solution of the integrals through the following steps: . . (18) .. 1 12 . . (19) I J4-m J Kl(B + Ex)2dX = [B + E(Z + m)I3 - (B + El)3 1 14-m 1 - m f (B + Ex)dx =: - [m2B +- $ x E(m + 21)] . . . . (20)November, 1978 TUBE FURNACE FOR ATOMIC-ABSORPTION SPECTROPHOTOMETRY 1093 [B + E(Z + m)I3 = a3 .. .. .. . . (21) (B +El)3 = 1 .. .... .. . . (22) and .. . . (23) m2 m2 m2B + - x E(m + 21) = - (a + 1) 2 2 . . gives + A [g(ct3 - 1) - 12 2 (Kl + 4 *) = C&f(t)nR? 2 and, with further simplification, As and (a2 + a - 2) = (a + 2) (a - 1) consequently .. .. . . where 1 1 .. . . (26) . . (27) Taking the term C M ( t ) from equation (26) : and introducing it into equation (10) : - 1 1 K(t) = y- x .@ x M(4 As and 3a 6ml + 312a + m2 (a + 2) K(t) = - x M(t) . .. . . (29)1094 FRIGIERI AND TRUCCO : BEAM-SHAPED ELECTROTHERMAL GRAPHITE Analyst, vol. Using the expression of the mass flux through one section [equation (2)] and Fick’s of diff usion3 : - = 2 D [ -- a(:;, t, x S(x) ] . . .. .. .. dik! dt where D is the diffusion coefficient, gives x M(t) . . .. dM - 6aD - = 2DK(t) = ~ d t 6ml + 3Z2a + m2 (a + 2) which, if /I = m/E, becomes x M(t) .. .. .. - GD (a/Z2) - dM dt - 6 s + 3a + p2 (a + 2) and, by integrating, - 6D (a/Z2) [qc- 3cc + p 2 (a + 2) t ] M(t) = M, exp where M, is the M(t) value for t = 0. It is therefore possible to obtain an expression such as M(t) = M , exp(- At) where and .. .. which represents the residence time of the atomic vapour inside the absorption chamber, that is, the time interval between vapour fonnation and complete removal of the sample by the diffusion process. With a cylindrical geometry of the tube, where a = 1, and 18 = 0, ~ ~ ( 1 , 0) = Z2/2D .. .. .. . . (34) The last equation is in complete agreement with that derived by L’Vov.2 Beam-shaped Chamber In order to put these theoretical assumptions into practice, a tube has been designed and constructed for use in the equipment availa.ble in our laboratories, a Pye Unicam SP1900 atomic-absorption spectrophotometer equipped with a Pye Unicam SP9-01 electrothermal atomiser.This electrothermal atomiser nonnally uses cylindrical graphite tubes (Fig. 2A) with the walls thinner in the centre in order to obtain faster heating of the central zone. The beam-shaped tube is shown in Fig. 2B. The thickness of the walls has been designed in order to maintain the same total resistance and temperature profile as the standard tube. The internal shape of the tube follows, as closely as possible, the geometry of the light beam focused by the Pye Unicam spectrophotometer. In order to simplify the calculationsNovembw, 1978 TUBE FURNACE FOR ATOMIC-ABSORPTION SPECTROPHOTOMETRY 1095 when comparing the two tubes, the internal volume of the beam-shaped tube is assumed to be fully coincident with the volume of the light beam.The evaluation of the tube performance is then carried out according to both the absorption measurement met hods. As previously discussed, the peak height values in each tube depend on the ratio between the analysis volume and the expansion volume. Consequently, the sensitivity will be higher when using the beam-shaped tube and the gain should be proportional to the ratio between the volumes of those portions of the internal volumes of the two tubes that are at a high temperature and act as absorbing chambers. This volume is defined by the temperature profile, which varies according to the operating conditions. As the two tubes have the same temperature profile under the same operating conditions, the comparison of the sensitivity gain is obtained through calculation of the volume ratios as functions of the optical path length (Fig.3). This diagram shows that the gains expected with the beam-shaped tube are about 2.8 x when only cylindrical central paths are involved, 2.7-2.6 x for paths 15-20 mm long and 1.7 x if the whole furnace length is taken into account; however, this last condition is not available with the power system used. 2.88 2.64 z 2.40 ' 2.16 \ 1.92 1.68 0 12 24 36 48 I -- -- Optical path length (2x)lmm Fig. 3. Variation with the Fig. 2. Electrothermal graphite optical path length of the volume tubes for atomic absorption: A, ratio between a cylindrical tube cylindrical; and B, beam shaped.B and a beam-shaped tube. The second method of absorption measurements consists in the evaluation of peak areas. As already seen, this quantity depends on the number of atoms, N , and their residence times in the analysis volume: Q x N r The residence times can be calculated as functions of the optical path length by using the previously defined equations (33) and (34) ; the ratio can also be determined from the residence times in the beam-shaped and cylindrical tubes. Fig. 4, which illustrates this ratio, also shows variations ranging from 1 to 19.6 for the same volume as that previously considered. As the number of atoms available for absorption with respect to the atoms produced during atomisation is proportional to the ratio between the volume where the light beam is located and the volume of the cell, the ratio of the areas is found to be where V, is the internal volume of the cylindrical tube and Vv is the internal volume of the beam-shaped tube, which, in turn, is equal to the volume of the light beam.The product of the previously calculated values gives an estimate of the possible gain in the peak-area measurements carried out by means of the new graphite tube. Fig. 5 shows the sensitivity gain as a function of the optical path length. The expected gains start from 2.8 x for the cylindrical central portion, and reach 5.7 x and 9.4 x at 15 and 20 mm along the optical path, respectively.1096 FRIGIERI AND TRUCCO : BEAM-SHAPED ELECTROTHERMAL GRAPHITE Analyst, Vd.103 Some prototypes of the atomisation chamber have been constructed from spectroscopically pure graphite similar to that used for the commercially available types. Optical path length (2x)/mm Fig. 4. Variation with the optical path length of the atomic residence time ratio between a beam-shaped and a cylindrical tube. 26.00 1 / I 2.00 - 0 12 24 36 48 Optical path length (2x)/mm Sensitivity gain as a function of the optical path length in peak area measure- ments (calculated from atomic residence time and volume ratios between the tubes). Fig. 5. Experimental Measurements have been made on a series of elements, chosen for their different behaviour during atomisation, in order to test the above assumptions in practice. The experiments were carried out with the tube already described, and also with two others of slightly different design.TA:BLE I SENSITIVITY AND SENSITIVITY RATIOS OBTAINED BY MEANS OF PEAK-HEIGHT MEASUREMENTS I N BOTH CYLINDRICAL AND BEAM-SHAPED FURNACE TUBES Sensitivjtylg x P Gain in sensitivity G d - Element tube tube A1 . . .. As . . .. Sb .. .. Cd . . .. Cr . . .. co a . .. cu . . .. Au . . .. Fe . . .. Pb .. .. Mn . . .. Ni .. .. Ag .. .. Sn .. .. 31 91 31 2 23 38 37 13 36 12 6 90 4 44 174 220 80 3 90 88 44 25 55 30 30 150 10 300 5.6 2.4 2.6 1.5 3.9 2.3 1.2 1.9 1.5 2.5 5.0 1.7 2.5 6.8 TABLE I1 SENSITIVITY AND SENSITIVITY RATIOS OBTAINED BY MEANS OF PEAK-AREA Sensitivitylg x 10-12 Element tube tube MEASUREMENTS I N BOTH CYLINDRICAL AND BEAM-SHAPED FURNACE TUBES Gain in sensitivity zEz&- A1 .. .. 15 72 4.8 Cd .. .. 2 5 2.5 cu .. .. 20 48 2.4 Sn .. .. 25 120 4.8November, 1978 TUBE FURNACE FOR ATOMIC-ABSORPTION SPECTROPHOTOMETRY 1097 000 000 000 0 I I Fig. 6. Sections and temperature profiles of the beam-shaped tubes B and C. Table I shows the results of the experimental tests and the true gains in analytical sensi- tivity obtained by means of the new tube. The sensitivity is defined as the mass in grams of the element giving a peak absorbance of 0.0044. Table I1 shows the results obtained by using the peak-area method. The sensitivity is defined as the mass in grams of the element that gives an integrated absorbance of 0.0044 absorbance unit seconds. The experimentally obtained gain values appear to be of the same order of magnitude as the theoretical values, and this can be considered a satisfactory agreement as the experi- mental measurements are affected by the unavoidable differences between the geometries of the tube and the light beam, and more important, by the volatilisation process and the other peculiarities of the different elements.As previously described, in both the cylindrical and beam-shaped tubes the area of the cross- sections of the conducting material are exactly the same, point by point, so that the total resistance of the tubes and the variation along the longitudinal axis are also the same. TABLE I11 COMPARISON OF SENSITIVITIES OBTAINED BY PEAK-HEIGHT MEASUREMENTS WITH BEAM-SHAPED FURNACE TUBES OF DIFFERENT EXTERNAL GEOMETRIES Element A1 . . . . As . . .. Sb . . .. Cd .. .. Cr . . .. c o . . .. cu .. ,. Au . . .. Fe . . .. Pb . . .. Mn . . .. Ni . . .. Se . . .. Ag , . .. Sn . . * . Sensitivitylg x 10-ls - Type c Type B 26 31 85 91 28 31 1 2 20 23 35 38 28 37 11 13 31 36 9 12 4 6 88 90 44 46 3 4 38 44 Gain in sensitivity, 19 7 11 100 15 9 32 18 16 33 50 2 6 33 16 %1098 FRIGIERI AND TRUCCO : BEAM-SHAPED ELECTROTHERMAL GRAPHITE Analyst, VoZ. 103 TABLE IV COMPARISON OF SENSITIVITIES 0BTAINE.D BY PEAK-AREA MEASUREMENTS WITH BEARI-SHAPED FURNACE TUBES OF DIFFERENT EXTERNAL GEOMETRIES Sections of the tubes are shown in Fig. 6. Sensitivitylg x 10-l2 Gain in {-h---_q sensitivity, Element Type c Type B % A1 . . .. 13 15 15 Cd . . .. 1 2 100 Sn . . .. 20 25 25 The temperature profile characteristic of the heating system depends directly on the variations of conductor resistance and on the thermal dissipation of the cooled electrodes.As the temperature profile affects the size of the reaction volume, it also determines the length of the absorption cell. Moreover, as the measured value of the absorbance is directly proportional to the length of the optical path, further improvement in the sensitivity of the new tubes can be obtained by varying the temperature profile. The external shape of the tube was therefore modified by elimination of the steps between the central area and the end zones. The resistance of the tube increases continuously from the ends to the centre without affecting the total resistance value. In Fig. 6 the sections of the two tubes and the temperature profiles, experimentally obtained in the two instances, are shown.The optical path lengths are expressed as functions of the temperature. TABLE V COMPARISON OF SENSITIVITIES OBTAINED BY PEAK-HEIGHT MEASUREMENTS WITH BEAM-SHAPED FURNACE TUBES, WITH AND WITHOUT PYROLYTIC GRAPHITE COATING The shape of tube is shown as C in Fig. 6. Element Cr . . .. c u . . .. Fe . . .. Pb , . .. Mn . . ,. Mo . . * . Ni . . .. Sr . . .. Ti . . .. v .. .. Sensitivit.y/g x 10-l2 Normal graphite graphite 13 20 18 28 20 31 7 9 3 4 33 90 36 88 23 44 200 660 118 380 Gain in sensitivity 1.5 1.6 1.5 1.3 1.3 2.7 2.4 1.9 3.3 3.2 TABLE VI COMPARISON OF SENSITIVITIES OBTAINED BY PEAK-AREA MEASUREMENTS WITH BEAM-SHAPED FURNACE TUBES, WITH AND WITHOUT PYROLYTIC GRAPHITE COATING The shape of the tube is shown as C in Fig. 6. Sensitivity/g x 10-l2 Element graphite graphite sensitivity Pyrolytic ----7- Normal Gain in cu . . .. 13 18 1.4 Mo . . .. 6 8 1.3 v .. .. 25 30 1.2November, 1978 TUBE FURNACE FOR ATOMIC-ABSORPTION SPECTROPHOTOMETRY 1099 Tables I11 and IV show the experimental results obtained by use of peak-height and peak- area measurements with the two tubes and the gain in sensitivity by using the tube with the stepless outer surface. This gain is small but these tubes are easy to manufacture. Tubes have also been coated with pyrolytic graphite in order to minimise carbide formation of the elements and improve the atomisation yield. The results obtained for some elements with two beam-shaped tubes, with and without pyrolytic graphite coating, are shown in Tables V and VI. This modification has also increased the sensitivity. The authors thank F. Rossi for a critical review of the calculations, R. Anzani for the construction of the beam-shaped tubes and E. Croce for technical assistance. References 1. 2. 3. Massmann, H., in “International Symposium on High-purity Materials in Science and Technology, Dresden, Germany, September 1965,” Volume 2, Akademie-Verlag, Berlin, 1966, pp. 297-308. L’VOV, B. V., “Atomic Absorption Spectrochemical Analysis,” translated by Dixon, S . H., Adam Hilger, London, 1970, pp. 201-205. Jost, W., “Diffusion in Liquids, Solids, Gases,” Academic Press, New York, 1952. Received February 6th, 1978 Accepted May 8th, 1978

ISSN:0003-2654    

DOI:10.1039/AN9780301089

出版商:RSC    

年代:1978

数据来源: RSC

摘要:

1100 Analyst, November, 1978, VoE. 103, pp. 1100-1103 Interference of Calcium on Barium as a Means of Assessing Atom ic-a bsorption Spectro p hotometers R. C. Rooney and J. F. Woolley Rooney and Ward Ltd., Blackwater Station Estate, Camberley, Surrey, GU17 9AF Standard Telecommunication Laboratories Limited, London Road, Harlow, Essex, CM17 9NA The well documented interference of calcium on the atomic-absorption deter- mination of barium has been studied by using a number of different com- mercially available instruments. The widely differing results obtained on varying the calcium to barium ratios using these instruments are thought to indicate that the interference is essentially an instrumental artefact, and should be largely avoidable. Keywords : A tomic-absorption spectrophotometry ; instrumentation ; barium determination ; calcium interference Since the original report by Billings,l many workers have reported interference by calcium on the atomic-absorption determination of b a I i ~ m ~ - ~ ; most of the recent work has disagreed with that of earlier worker^^*^ who claimed that the use of the dinitrogen oxide - acetylene flame removed the effects. There is little agreement on the maximum permissible amount of calcium, and there have been many suggestions as to the mechanism of the interference.Explanations range from spectral interference such as molecular band absorption by CaOH, which is known to have a fairly strong emission band in this region of the spectrum, to a number of purely instru- mental causes such as overloading of the detector or some part of the measuring circuitry.One of us (R.C.R.) has previously encountered instrumental problems when working with strongly emitting flames (and an instrument with poor stray-light characteristics), and these problems were definitely attributable to failure of the phase-selective demodulator to cope with the high direct-current levels being developed at the detector when strongly illuminated. It seemed possible to us that other workers may have experienced similar problems and not recognised them as instrumental artefacts ; the calcium - barium interference seems a possible example of this type of effect, because of the wavelengths involved. The barium resonance wavelength is 553.5nm, and the strongly emitting CaOH band-head peaks almost exactly at this wavelength.Under these conditions, of course, the resolution or stray-light per- formance of the monochromator is of no consequence, and the separation of the absorption and emission signals is effected solely by th.e rejection of direct-current signals from the detector by the a.c. measuring system. The hypothesis that the “interference” may be instrumental in origin can be tested by making a series of measurements on various calcium - barium mixtures using a wide variety of instrument designs; few commercial design teams will have arrived at identical solutions to the problem, and some systems will almost certainly be more efficient than others. If the hypothesis is found to be true, the information has value far beyond the specific instance of calcium - barium interference ; the increasing use of electrothermal atomisation devices, with their black-body continuum radiation, will impose ever greater demands on precisely this part of the design of atomic-absorption spectrophotometers. Experimental In order to test the hypothesis, it was decided to make measurements on as many different commercially available instruments as possible, using identical calcium - barium mixtures.Our contention is that even if only one instrument can be found that does not show the interference, there must be a strong possibility that the effect is at least partly instrumental. Calcium carbonate and barium carbonate were dissolved in a minimum amount of concentrated hydrochloric acid to make stock solutions, from which working solutions were prepared.All of the final A set of solutions was prepared from high-purity materials.ROONEY AND WOOLLEY 1101 working solutions contained 3000 pg ml-l of potassium, added as potassium chloride, to act as an ionisation suppressor. The working solutions were as follows : A. B. C. D. E. F. 10 pg ml-l of barium. 10 pg ml-l of barium + 100 pg ml-l of calcium. 10 pg ml-l of barium + 500 pg ml-1 of calcium. 10 pg ml-l of barium + 1000 pg ml-l of calcium. 10 pg ml-l of barium + 5000 pg ml-l of calcium. 10 pg ml-l of barium + 10000 pg ml-l of calcium. These solutions were made up in sufficiently large volumes to allow portions of the same solution to be used by all of the co-operating laboratories; portions were circulated to 14 laboratories.The laboratories were asked to set up their instrument in the normal way for routine determinations, using a dinitrogen oxide - acetylene flame and the maximum per- missible lamp current. They were asked to use solution A (the pure barium solution) to optimise such parameters as burner height and flame conditions, and then to measure the remaining solutions without changing the conditions. In this way it was hoped that all of the various instruments would be working at comparable measurement heights and flame stoicheiometries ; in our laboratory on three different instruments the optimum measurement height was about 0.6 cm and the flame was optimum when slightly fuel rich. Nine labora- tories completed the work, and a total of 15 sets of results from 13 different types of instrument were received. We believe that the results indicate that the original hypothesis is valid, as considerable differences were found between the results obtained from various commercial instruments, all used under their own optimum conditions.The instruments used are listed in Table I, the conditions of use in Table I1 and the results obtained in Table 111. Some of the instruments are current and some are obsolete, but they are thought to represent a reasonable cross-section of those currently in use in labora- tories in the UK; the performance of some of the newer designs suggests that even the latest models may not yet be perfect in this respect, but it is in no way intended that this work should suggest a “best buy” instrument. Instrument A B c D E F G H I JK L M N 0 TABLE I INSTRUMENTS USED Type Remarks Perkin-Elmer 107 Perkin-Elmer 107 Perkin-Elmer 303 Perkin-Elmer 306 Perkin-Elmer 403 Varian AA6 EEL 140 EEL 240 Unicam SP 90 Series 2 Unicam SP 90 Series 2 Unicam SP 1900 Instrumentation Laboratories IL 463 Southern Analytical A3000 Shandon Southern A3300 Pye Unicam SP 2900 4 years old, mirrors in poor condition Instrument A after re-mirroring and Just serviced and overhauled by manufacturer’s service manufacturer - - Flame run rather leaner than optimum to avoid rapid carbonisation - - Requiring service Instrument I after manufacturer’s service - - Run during manufacturer’s demonstration (by potential customer) Results The results in Table I11 indicate that, of the instruments used, only five, vix., D, E, H, N and 0, appear to cope with the full range of calcium contents.The emission measurements indicate that, at the highest level of calcium used, the ratio of flame emission to hollow- cathode lamp emission varies from about 4: 1 (instrument D) to 250: 1 (instrument L). It is unlikely that this range represents the actual variation of emission from the hollow-cathode lamps used, and it is reasonable to assume that the emission from all of the flames is similar.1102 RoONEY AND WOOLLEY: INTERFERENCE OF CALCIUM ON BARIUM FOR Analyst, Vd. 103 TABLE I1 CONDITIONS OF USE OF INSTRUMENTS All instruments were adjusted for optimum flame conditions while aspirating solution A, barium and ionisation suppressor alone; subsequent solutions were aspirated without a change in the conditions.Optimum conditions include the use of the maximum permissible lamp current and minimum usable slit width. Instrument A l3 C D E F G H I JK L M N 0 Lamp type Lamp currentlmh Cathodeon . . .. .. Cathodeon . . .. .. Cathodeon . . .. .. Perkin-Elmer .. .. Varian .. .. .. Activion . . .. .. Activion . . .. .. Cathodeon . . .. .. Activion . . .. .. Activion . . .. .. Southern Spectral Sources . . Cathodeon , . .. .. Pye Unicam . . .. .. Pye Unicam . . .. .. Pye Unicam . . .. .. 13 15 15 15 25 2 0 10 18 15 15 13 20 14 12 15 Band passlnm 0.2 0.2 0.2 0.14 0.2 0.1 0.08-nm slit 0.28 0.05-nm slit 0.07-nm slit 0.1-nm slit 0.43 0.3 0.15 0.2 The variation probably represents differences in de-focusing of the flame image, and hence rejection of some of the emitted light by the external optical system.Of the instruments that coped well, N was available for more detailed study. It was found that reduction of the lamp current to 5 mA caused a marked deterioration in performance; the results obtained for the test solutions were as follows: Solution . . . . A B C D E F Absorbance . . .. 0.10 0.10 0.25 0.80 1.0 1.0 TABLE 111 RESULTS OBTAINED WITH THE VARIOUS INSTRUMENTS Absorption figures are given in absorbance units, Working solution Instrument A B C D E F G H I k L M N 0 Notes* 1 2 2, 3 2, 3 2, 3, 4 5 5 2 2 2, 4 5 - - - r A 0.060 0.13 0.18 0.128 0.120 0.078 0.02 0.064 0.03 0.12 0.225 0.308 0.10 0.12 0.350 B 0.060 0.12 0.16 0.127 0.119 0.077 0.02 0.066 0.035 0.11 0.231 0.319 0.12 0.12 0.346 C 0.056 0.10 0.16 0.125 0.121 0.080 0.03 0.058 0.035 0.11 0.224 0.312 0.12 0.12 0.351 D 0,046 0.10 0,17 0.126 0.120 0..086 0.04 0.065 0.06 0.218 0.321 0.13 0.12 0.360 0.1-0.2 :: E 0.042 0.045 0.25 0.138 0.127 0.124 0.25 0.067 0.2 0.845 2.8 0.45 0.12 0.336 0-0.8$ I? 0.015 0.050 0.70 0.130 0.129 0.173 0.50 0.076 0-0.6: N.M.§ 0.887 2.8 0.80 0.13 0.327 Emissiont -----Gz Lamp Water F 100 0 1000 140 0 1000 290 25 1000 192 29 1000 160 0 1000 - - - - - - 5 30 1000 10 110 1000 47 8 1000 4 0 1000 0 1000 lo 5 0 1000 Not measured - *Notes : 1.The energy meter fell sharply to zero on aspirating solution F; the absorbance simultaneously increased to above 2. Both readings slowly settled to their final values; readings on solution E and F were very noisy. Readings on solutions E and F were very noisy.The energy meter fluctuated wildly while aspirating solutions E and F. The absorbance reading on solution F increased to above 2 before settling back to its final value. All readings were noisy, the noise increasing with increasing calcium content. 2. 3. 4. 5. t All emission measurements have been normalised for ease of comparison. $ Range represents fluctuation of readings. 5 N.M. = not measurable owing to noise.November, 1978 ASSESSING ATOMIC-ABSORPTION SPECTROPHOTOMETERS 1103 The only difference in operating conditions was that the photomultiplier gain had to be increased; the ratio of flame emission to hollow-cathode lamp emission was, of course, greatly increased and so the a.c. to d.c. current ratio through the measuring circuits was made much more unfavourable. Finally, the same instrument was used for experiments at low barium levels; the pure 10000 pg ml-l calcium solution plus 3000 pg ml-l of potassium was aspirated, using 10-fold scale expansion and displaying the results on a strip-chart recorder.An absorbance of 0.0115 was obtained, with a noise level of about 0.001; a time constant of 0.5 s was intro- duced into the measurement circuit via the damping control. An amount of 3000 pg ml-l of potassium alone gave no discernible absorption, with a noise level of less than 0.0005, and a solution containing 0.5 pg ml-1 of barium (containing potassium) gave an absorbance of 0.0070 with a noise level of less than 0.0005. Standard addition of 0.5 pg ml-l of barium to the calcium solution gave an increase in absorbance to 0.020, which is very close to the expected value.These results suggest that the barium content of the Johnson Matthey Specpure calcium carbonate is 33 pg g-1, which is much higher than the manufacturer’s certificate of analysis and also much higher than results obtained by a method similar to that of ban^.^ The most likely explanation is light scattering by the fairly high salt content of the solution; background measurements using a deuterium arc are not possible at this wavelength, but measurements using the 535.0-nm thallium line showed that there is a background absorbance of 0.02. Conclusions It is apparent that instrument N, at least, will measure barium at the 0.5 pg ml-l level in the presence of a 2 x lo5 excess of calcium, with no interference other than that due to high salt concentrations, and it is likely that instruments D, E, H and 0 would perform in the same way.The calcium level of 10000 pg ml-l used is probably near the upper practical limit for flame work if problems due to burner and nebuliser blockage or severe scattering of the light are to be avoided. This seems to us to imply that instruments that show an inter- ference more serious than this are exhibiting some form of instrument failure; at least one form has been identified previously as overloading of a phase-selective demodulator, and cathode-ray oscilloscope measurements on instrument C showed that its failure was due to the d.c. level and shot noise from the photomultiplier increasing to such an extent that the modulated signal was completely buried. We now believe that a number of hitherto un- explained interference effects reported on a particular instrument and not found on others may be due to a similar effect, in particular those arising with electrothermal atomisers.The series of measurements used here is therefore an excellent further test to apply to a spectrophotometer. The authors are indebted to the following friends and colleagues who made the measure- ments requested in their own laboratories and on their own instruments: Mr. B. Bach, Associated Portland Cement Manufacturers Ltd. ; Mr. F. J. Bano, London and Scandinavian Metallurgical Co. Ltd.; Dr. L. E. Coles, Glamorgan County Public Health Laboratory; Mr. R. E. Coulson, Borax Consolidated Ltd.; Mr. W. H. Hill, Metal Box Ltd.; Mr. G. T. P. Humphreys, International Nickel Ltd. ; Mr. C. R. Jackson, Charter Consolidated; Mr. I. K. Kaduji, Lankro Chemicals Ltd.; Mr. R. J. W. Powell, General Electricity Co.; Mr. G. Presswell, B.A.C. Ltd.; Mr. D. Utley, Bristol Aerojet Ltd.; and Mr. R. Walker, S. gt J. Juniper and Co. Ltd. 1. 2. 3. 4. 5. 6. 7. References Billings, G. K., Atom. Absorpion Newsl., 1965, 4, 357. Cioni, R., Mazzucotelli, A., and Ottonello, G., AqzaZyst, 1976, 101, 956. Bano, F. J., Analyst, 1973, 98, 655. Frache, R., and Mazzucotelli, A., Talanta, 1976, 23, 389. Ingamells, C. O., Suhr, N. H., Tan, F. C . , and Anderson, D. H., Analytica Chim. Ada, 1971, 53, Koirtyohann, S. R., and Pickett, E. E., Analyt. Chem., 1966, 38, 585. Capacho-Delgado, L., and Sprague, S., Atom. AbsorPtzon Newsl., 1965, 4, 363. 345. Received March 30th, 1975 Accepted June 19th, 1975

ISSN:0003-2654    

DOI:10.1039/AN9780301100

出版商:RSC    

年代:1978

数据来源: RSC

摘要:

1104 Anatyst, November, 1978, VoJ. 103, PP. IIOP1115 Direct Determination of Ammonium-nitrogen by Flame Emission Spectrometry in a Hydrogen - Nitrogen Diffusion Flame J. M. S. Butcher and G. F. Kirkbright Deflartment of Chemistry, Imperial College, London, SU'7 2A Y A method for the determination of trace amounts of ammonium ion in aqueous solution has been developed that utilises the chemiluminescent emission a t 336.0nm of the NH species produced when ammonia gas generated from alkaline sample solutions is introduced into a hydrogen - nitrogen diffusion flame. The method has been applied successfully to the determination of the exchangeable ammonium-nitrogen content of soils. A practical detection limit of 0.2 pg ml-1 has been obtained for a 5-ml sample. Keywords A mmonium-nitrogen determination ; NH emission sflectrometry ; hydrogen - nitrogen diffusion flame ; soil analysis Several methods for the determination of ammonia using molecular emission and absorption spectrometry have been reported recently. Cre~serl-~ has utilised the ultraviolet absorption spectrum of ammonia in the gas phase after its liberation from strongly alkaline sample solutions. Belcher et aL4 have used a similar generation procedure followed by the observa- tion of the emission from the NH, species at 600 nm produced when ammonia was decom- posed in a heated stream of oxygen by molecular emission cavity analysis. A technique has recently been described in another paper from this laboratory6 in which ammonium-nitrogen is oxidised to nitrogen gas by using hypobromite reagent.The nitrogen released is then determined by monitoring the emission of the NH species at 336.0 nm after passage of the nitrogen generated in a stream of argon into a radiofrequency inductively coupled plasma source. To the best of our knowledge, no determination of nitrogen using direct flame emission spectrometry that has sufficient sensitivity for trace analysis has been reported. This paper reports the development of a method for the direct determination of ammonium- nitrogen by flame emission spectrometry with the use of a hydrogen-nitrogen diffusion flame. In this method, aqueous ammonium-ion sample solutions are made alkaline in order t o liberate gaseous ammonia, which is swept into a hydrogen-nitrogen flame where its concentration is determined by measurement of the emission intensity produced at 336.0 nm by the NH species produced on decomposition of the ammonia.Experimental Apparatus Schematic diagrams of the instrumentation and gas-generation system employed are shown in Figs. 1 and 2. The spectrometer was a modified atomic-absorption spectrophotometer (Model AA4, Varian Techtron Ltd., Melbourne, Australia) arranged to permit flame emission spectrometry. Details of the components employed are shown in Table I. The design and dimensions of the gas-generation cell are shown in Fig. 2. Reagents All reagents were of analytical-reagent grade. A 10 mg ml-1 ammonium-ion stock solution was prepared by dissolving analytical-reagent grade ammonium sulphate in distilled water. This stock solution was diluted as required with distilled water to prepare working s o h t ions.BUTCHER AND KIRKBRIGHT 1105 Fig.1. Schematic diagram of the instrumentation employed for the determination of ammonium- See Table I for key to the individual components. nitrogen in aqueous solutions. n -- Magnetic follower Fig. 2. Gas-generation apparatus employed. Results and Discussion Optimisation of Flame and Spectrometer Operating Conditions The emission spectrum of the oxygen - ammonia flame was first observed by Eder,6 who reported two main band systems, which he called the ammonia cc and /3 bands. These bands have since been assigned to the emission of the species NH, for the a band and NH for the /3 band.' The molecular emission from these bands has also been observed in the TABLE I KEY TO SCHEMATIC DIAGRAM OF INSTRUMENTATION (FIG.1) Letter A B C D E F G H I JK Component Gas generation cell (Quickfit, Type FP50/1/1A) Stirrer motor Electrothermal heating tape 5-cm N,O - C,H, slot burner 2.5 cm focal length silica lenses Rotating chopper (285 Hz) Grating monochromator with Ebert mounting. Reciprocal linear dispersion 3.3 nm mm-l Photomultiplier (RCA 1P28) A.c. amplifier and E.H.T. supply Electronic integrator Chart recorder1106 BUTCHER AND KIRKBRIGHT: DIRECT DETERMINATION OF Analyst, VoZ. 103 much cooler hydrogen - nitrogen diffusion flame by Dagnall et aZ.8 when a 1 + 1 mixture of concentrated ammonia solution and methanol is nebulised into the flame. It was decided to study the possibility of using one of these two bands in the hydrogen - nitrogen diffusion flame as the basis of an analytical method for the determination of ammonium ion in aqueous solution.The emission spectrum of a hydrogen - nitrogen diffusion flame when ammonia is intro- duced into the nitrogen supply is shown in Fig. 3. The ammonia was generated by adding 10 mg ml-l ammonium sulphate solution dropwise, with a syringe, to a flask containing 40% m/V sodium hydroxide solution. This procedure was found to give a reasonably steady supply of ammonia to the flame during the recording of the spectrum. As can be seen, the spectrum of the flame alone, without added ammonia, shows no emission due to either NH or NH, species, as there is presumably insufficient energy available in the flame to dissociate molecular nitrogen and thus no interaction between nitrogen and hydrogen can occur.This confirms the report of Gaydong that, unlike some much hotter flames, cool flames supported by nitrogen show no such emission. x l ‘ x5 14 - 9 13 Y I 0 - Ammonia CY band [ N H 2 j A,,,, 420-425 nrn Wavelength/nm 300 310 320 330 340 350 360 370 380 390 400 410 420 2.8 2.6 2.4 2.2 2.0 g .- C 7.8 3 > l . G 5 1.4 5 1.2 >: 1.0 0.8 5 0.6 0.4 0.2 4- .- co c-’ .- CI Fig. 3. Emission spectrum of the hydrogen - nitrogen diffusion flame obtained when ammonia gas was added to the nitrogen sweep gas (solid line), and the spectrum of the same flame in the absence of ammonia (dotted line). In the presence of ammonia in the nitrogen supply to the flame, the emission intensity of the NH band was found to be 10 times greater than that of the NH, system and a similar background intensity was observed at each wavelength with the resolution used here, although by using a broad-band filter the NH, band could be made relatively more intense.It has been noticed, however, that under the conditions used here the wavelength of maxi- mum emission for the NH, band varied between 425 and 575 nm, depending on the position of its observation in the flame and the amount of ammonia introduced; indeed, when viewed on a long-slot burner, instead of a yellow - $green emission two regions of emission were noticed. When small amounts of ammonia were introduced a persistent blue emission at the base of the flame was observed, but when higher concentrations of ammonia were intro- duced a yellow emission was observed higher in the flame, above the blue region.As yet no explanation for this effect has been found. The presence of this effect, together with the possibility that spectral interferences from easily excited ions such as Na+ and K+ may occur when using a broad band system with low spectral resolution, led to the choice of the NH 336.0-nm band system for use in all subsequent work. Attempts to study the NH species in absorption were undertaken. The use of a xenon arc continuum source was investigated as no suitable hollow-cathode lamp with a line ofNovember, 1978 AMMONIUM-NITROGEN BY FLAME EMISSION SPECTROMETRY 1107 reasonable strength coincident with the 336.0-nm band head was available. Although no absorption of the xenon arc radiation was observed when using a cool diffusion flame, absorption signals were obtained at 336.0 nm when the hydrogen - nitrogen flame was burnt in an atmosphere of oxygen.These absorption signals were, however, too small to be analytically useful, and the much stronger emission from NH was thought more likely to provide the analytical sensitivity necessary for the determination of ammonium-nitrogen. This emission is therefore probably due to chemiluminescence in which the molecules are. formed directly in the excited state and thus yield intense emission; the ground-state population is probably very small and thus absorption is very weak. A comparison was made of three different flames, namely diffusion flames of (a) hydrogen - nitrogen - oxygen, (b) hydrogen - argon - air and (c) hydrogen - nitrogen - air.The first of these flames showed intense NH emission but also high noise levels; the tail of the OH band head at 306.4nm caused considerable emission at 336.0nm. Very little difference was observed in the emission spectra and intensities observed in flames supported with argon rather than nitrogen. Possibly owing to the higher heat capacity of the nitrogen compared with argon, the background emission from the tail of the OH band head at 306.4 nm was lower when using the nitrogen-based flame, under otherwise similar conditions. The cool and inexpensive hydrogen - nitrogen diffusion flame was therefore chosen for all subsequent work because of the high signal to background intensity ratio obtained at 336.0 nm for NH.The emission from the OH 306.4-nm band head is strongly degraded to the red and causes sufficient background emission to make the detection of low emission intensities from the NH species very diffic~lt.~ It has been found, however, that the use of a long-slot burner of the type generally used with acetylene - air and dinitrogen oxide - acetylene flames is preferable to the adoption of a circular burner of the type frequently used for flame emission spectrometry, as the hydrogen - nitrogen diffusion flame burning on a long-slot burner exhibits a central non-luminous channel of low background at 336.0 nm. This region shows very little emission due to the OH species and also shows the maximum emission due to NH from ammonia added to the flame. This effect can be clearly seen in Fig.4, which shows the lateral variation of the signal and background emission intensities across the flame. -3.6 -1.2 +1.2 +3.6 -2.4 0 +2.4 Lateral displacement/m m Fig. 4. Variation of the NH signal and background emission intensities at 336.0 nm with lateral displacement of the burner. Solid line, signal + background ; and dotted line, background. The flow-rate used for nitrogen was chosen carefully so as to produce a stiff flame of sufficient dimensions that did not show excessive noise due to turbulence or lifting of the flame from the burner. The analytical signal intensity is dependent on the rate of delivery of ammonia to the flame and the optimum conditions for this occur at high nitrogen flow- rates. The highest possible flow-rate of nitrogen was therefore chosen to yield a stable, noise-free flame that was not easily extinguished.1108 BUTCHER AND KIRKBRIGHT : DIRECT DETERMINATION OF Analyst, VoZ.103 The effect of variation of the hydrogen flow-rate on both the NH signal and background emission intensities a t 336.0 nm was investigated at the optimum nitrogen flow-rate chosen. The result obtained is shown in Fig. 5, when signals from solutions containing 10 pg ml-1 of ammonium ion were recorded using the technique finally adopted, together with the integrated background intensities obtained. 160 140 .E 120 L- 2 100 2 f + .- p 80 2 60 2 w v) .- C .- - 40 r c - 20 0 1 1 1 1 1 1 1 1 1 1 i.0 1.4 1.8 2.2 2.6 3.0 1.0 1.4 1.8 2.2 2.6 3.0 Hydrogen flow-rate/l min-l Fig. 5. (a) Effect of variation in hydrogen flow-rate on signal and background emission intensities observed a t 336.0 nm (obtained when using the optimum nitrogen flow-rates) .A, Signal; and B, background. (b) Variation of signal to background ratio with hydrogen flow-rate. The effect of variation of the viewing heigh.t in the flame on the signal and background intensities at the flow-rates chosen was investigated; the results are shown in Fig. 6. The best signal to background intensity ratio was found to be above the point of maximum NH emission intensity, and this was chosen as the optimum viewing height. The effect of spectral band pass of the spectrometer on the signal to background intensity ratio was also investigated; this was found to be relatively constant for slit widths in the range 100-300 pm, which in Fig.7 corresponds to a spectral half-band pass of 0.33-0.99 nm, although an even greater spectral band pass (up to 2 nm) could possibly be used as, owing to the low light levels encountered, some improvement in the signal to noise ratio at the detection system would possibly counteract the slightly poorer signal to background ratio obtained. The optimum flame and spectrometer operating conditions established are summarised in Table 11. Optimisation of Gas-generation Technique for Ammonia All of the methods described earlier for the determination of ammonium-nitrogen rely upon passing the carrier gas through the sample and reagent to liberate the analyte in gaseous form from the sample solution. Cresser2 used various designs of bubbler with air as the carrier gas to generate ammonia.In previous work in this laboratory,5 a stream of argon has been employed to liberate nitrogen gas into an inductively coupled plasma. In this work, the nitrogen carrier gas was bubbled through the sample in a manner similar to that employed in the other techniques, prior to passing the gaseous ammonia to the flame. This procedure, however, was found to have several disadvantages in that an unacceptable level of flame noise was caused by the bubbler device and splash-over of sodium and potassium contained in the sample and reagent solutions caused intense unwanted flame emission and stray light interference. In an attempt to eliminate these effects, the generation of ammoniaNovember, 1978 AMMONIUM-NITROGEN BY FLAME EMISSION SPECTROMETRY 1109 0 2 4 6 8 1 0 1 2 Viewing height in flame/mrn Fig.6. Effect of variation in height of 0 50 100 150 200 250 300 Slit width/pm obsekation in the flame (obtained "when Fig. 7. Variation of the signal and back- using the optimum flow-rate). A, Signal; B, ground emission intensities at 336.0 nm with background; and C , signal to background spectrometer slit width. A, Signal; and B, ratio f 4. background. S/B = signal to background ratio. from stirred aqueous solutions was investigated in detail. The efficient generation of ammonia using the simple method of making the sample solution alkaline requires a cell assembly and facilities that permit rapid, efficient stirring to obtain the maximum rate of renewal of the solution surface and a high surface area of the solution in the cell in comparison with the dead volume in the cell above the liquid.The use of a magnetic follower and stirrer motor beneath the cell adds the constraint that the cell must also be easily removable for washing and the addition of the reagent. The actual procedure employed in making the sample solution alkaline is also important; the most effective procedure has been found to be to add the sample to the alkali in the generation cell through a septum cap using a syringe. The suit ability of several commercially available borosilicate-glass flasks was evaluated ; although a conical flask should in theory be the most suitable, in practice the best system utilised a 50-ml pear-shaped flask with two ground-glass B-14 necks.This met all of the above requirements and gave a large surface area of solution with a low dead volume; using a magnetic follower 25 mm in length it also allowed better and more even stirring than that obtained from conical and round-bottomed flasks. The cell could easily be detached from the cell head and re-positioned accurately by means of a short length of tubing, which was rigidly fixed to the stirrer motor, on which the cell assembly rested during use. The cell and cell head assembly employed for the nitrogen sweep gas are shown in Fig. 2. This system TABLE I1 OPERATING CONDITIONS Location Conditions Flame . . . . Nitrogen flow-rate = 6.7 1 min-1 Hydrogen flow-rate = 2.3 1 min-l Viewing height = 7.5 mm, along central axis Slit width = 300 pm (equivalent to 0.99 nm E.H.T.= 800V Amplifier = x 5 scale expansion Spectrometer . . spectral half-band pass)1110 BUTCHER AND KIRKBRIGHT: DIRECT DETERMINATION OF Analyst, VuZ. 103 was found to give satisfactory results, provided that care was taken to inject the sample reproducibly . The effect on the analytical signals obtained at 336.0nm of changing the sample volume used at a fixed reagent volume of 5 ml was in.vestigated, as was the effect of changing the sample volume at a constant sample to reagent volume ratio of 1 : 1 ; the results are shown in Fig. 8. Maximum sensitivity was obtained by using a sample volume of 5 ml with a reagent volume of 5 ml. A total volume of 10 ml was the practical upper limit for the cell used. In order to avoid problems during the transfer of the generated ammonia to the flame, the stainless-steel tubing used was wrapped with an Electrothermal tape heater, which was operated at 180 V to obtain an outlet temperature greater than 100 "C, so as to prevent any condensation of water vapour in the interconnecting tubing.0 1 2 3 4 5 Sample vol u me/m I Fig. 8. Effect on the signal in- tensities of variation in sample volume and total volume in the gas generation cell. A, Reagent volume = 5 ml; and B, reagent volume = sample volume. After optimisation of the ammonia-generat ion procedure, and using the established con- ditions for the observation of the NH emission intensity at 336.0nm described earlier, a detection limit of 2 pg ml-l was obtained for ammonium-nitrogen, employing measurement of the maximum peak height obtained during the gas-generation procedure.The gas generation is, however, relatively slow and signal durations of between 6 and 8 min were obtained. Improvement of Analytical Sensitivity of the Technique Measurement of the peak area as a function of concentration was shown in early experi- ments to provide greater reproducibility than measurement of the peak height and also to be much less sensitive to the effects of variation in temperature, injection rate, stirring speed, presence of foreign ions and reagent concentrations. Signal integration for the complete duration of the signal with the procedure developed, however, would entail excessively long integration periods. In order to allow the integration to be performed for a time period less than that of the total signal duration, a system for automatically measuring this time from the injection of the sample was developed.A schematic diagram of the electronic integrator constructed is shown in Fig. 9. The four-stage sequential timer was used to provide the necessary functions by controlling four semiconductor switches. The cycle of operation used was as follows:Noveunber, 1978 AMMONIUM-NITROGEN BY FLAME EMISSION SPECTROMETRY 1111 Reset. The capacitor holding the charge generated from the previous integration was shorted to return the output to 0 V. Zero. The input was connected to the signal channel before the injection of the sample to record the background level, which should show no signal provided that the backing-off control was correctly adjusted. The sample was injected when the start of this timing period was indicated, and the resulting signal was thus recorded.On completion of the integration time the signal was isolated from the integrator and the value obtained was held by the capacitor. Integrate. Hold. I I I Differential Backing i Electronic I Integrator input amplifier I off I switches ; I I I tput A I Fig. 9. Schematic diagram of electronic integrator. This cycle automatically repeated itself while the operator performed the washing and re-charging of the cell and injection of the sample. The system has been found to give a marked increase in sensitivity as the integral method is much less affected by random noise on the signals. The other main limit on the analytical sensitivity is the rate of generation of ammonia gas from the sample solution.This is mainly controlled by the temperature of the solution in the cell; in order to provide a simple, rapid and reproducible method of increasing the temperature in the cell during the generation procedure, the addition of sulphuric acid to the sample solution prior to its injection into the 40% m/V sodium hydroxide reagent was investigated. The heat of neutralisation of the sulphuric acid enabled the sample solution to reach temperatures as high as 80 O C , depending on the initial concentration of the acid used. This enabled signals of much shorter duration to be produced, with a marked increase in the peak height of the signals obtained. Problems were encountered, however, owing to excessive amounts of water vapour generated from the sample solutions, which caused depression of the analytical signal and background emission from the flame at 336.0 nm when high sulphuric acid concentrations were used.The effects of varying the sulphuric acid concentration on the integrated signals produced, the reproducibility obtained, the final sodium hydroxide concentration and the maximum temperature reached during the generation are shown in Fig. 10. The signals obtained using the integration method are largely unaffected by the acid concentration, even though the same integration time of 2 min was used for all signals, which were of different duration. This result was due to the depression of both the analytical signal and background emission intensities by the generated water vapour, which was more noticeable using this method than the measurement of the maximum peak heights of the signals, where the peak height obtained increased greatly with increased acid Concentration.Even though the use of the integral method did not benefit directly by the inclusion of the sulphuric acid in the sample solutions, the benefits of increased reproducibility and the reduction in the duration of the analytical signals led to an improved detection limit. The1112 BUTCHER AND KIRKBRIGHT: DIRECT DETERMINATION OF AnaZyst, VoZ. 103 140 1 (” 1 8 1 0 1 2 3 4 5 0 I I I I 0 1 2 3 4 5 80 70 60 0 \ 50 w 2 40 $ 2 30 ’ ZL + 20 10 Volume of conc. HzS04 added to 20 mi of sample/ml Fig. 10. Variation of (A) integrated signal intensity, (B) reproducibility, (C) maximum temperature obtained during the generation and (D) final concentration of sodium hydroxide obtained with the volume of concentrated sulphuric acid added to 20 in1 of sample. optimum sulphuric acid concentration was obtained when 2.5 ml of concentrated acid were added to each 20ml of the sample solution.The use of longer integration periods was investigated but led to poorer detection limits owing to the reduction in the signal to back- ground ratio obtained when the exponentially decaying signal was measured above a constant background emission intensity. Using the addition of 2.5ml of concentrated sulphuric acid to 20ml of sample and an integration time of 2 min, with a sample volume of 5 ml and 5 ml of the 40% m/V sodium hydroxide solution, the following performance characteristics were obtained for the method developed.The linear range was 0-200 pg ml-l with a detection limit of 0.2 pg ml-1 at three times the standard deviation of the blanks above the mean of the determination of the blanks. The reproducibility for the repetitive determination of 10 pg ml-l of ammonium ion was evaluated; the relative standard deviation was 0.03. ,$O 350 i (b’ P 0 2 4 6 8 ‘10 0 100 200 300 400 500 NH; concentration/pg ml-’ Fig. 11. Calibration graphs obtained in the regions (a) 0-10 pg ml-1 and (b) 0-500 pg ml-1 of ammonium ion.November, 1978 AMMONIUM-NITROGEN BY FLAME EMISSION SPECTROMETRY 1113 A typical calibration graph is shown in Fig. 11 for the concentration range 0-10pgml-1 and the signals from which the graph was obtained are shown in Fig.12. This illustrates the shape of the signals obtained by using the integration technique employed. Also shown in Fig. 11 is a calibration graph for the concentration range 0-500 pg ml-l, showing slight curvature towards the concentration axis above 200 pg ml-l. 8 2 100 E g 80 >: + .- + .- 60 .+ c .- - 2 40 en S + 20 - 50 40 30 20 10 0 Ti me/m i n Fig. 12. Typical analytical signals for solutions in the range 0-10 pgml-1 of ammonium ion (indicated above the peaks), showing the shape of the integrals obtained using the electronic integrator system. Interference Studies The effect of various ions commonly found in soil extracts on the method developed for the determination of ammonium ions was investigated. An interference was considered to have occurred if the mean of the signals obtained in the presence of the foreign ion concerned differed by more than twice the standard deviation from the mean of the signals obtained from a pure aqueous ammonium-ion solution of the same concentration.The experiments to investigate the effects of foreign ions were undertaken with 10pgml-1 ammonium-ion solutions; this concentration is appropriate as it may be expected that many soils would give exchangeable ammonium-ion concentrations of this magnitude. It was found that a 100-fold mass excess (1000 pg ml-l) of Na+, K+, Ca2+, Mg2+, SO,2-, C1-, PO,3-, NO3- and acetate caused no significant interference and that a 10-fold mass excess (100 pg ml-l) of Cu2+, Zn2+, Fe3+, Mn2+ and AP+ also caused no interferences.Although no interferences were obtained when using the integration technique employed, the presence of precipitated metal ions in the alkaline medium could have a noticeable effect on the peak heights of the signals obtained by decreasing the rate of generation of ammonia. The effect of the use of 2 M potassium chloride medium, i.e., that used for the soil extraction procedure, was investigated. Although no actual interference was observed when using this medium, the signals obtained from the presence of trace amounts of ammonium ion in the 2 M potassium chloride reagent were significant. For this reason, standards prepared in 2 M potassium chloride solution were used for the analysis of soil extracts; blank 2 M potassium chloride solution and 10 pg ml-1 ammonium ion in 2 M potassium chloride solution were the only standards required in routine analysis.Results of Soil Analysis by courtesy of the Macaulay Institute for Soil Research, Aberdeen. The soil samples examined had previously been analysed independently and were provided1114 BUTCHER AND KIRKBRIGHT : DIRECT DETERMINATION OF Analyst, liol. 103 The soils were extracted using a procedure identical with that previously used for the analysis of the soil samples examined, and the exchangeable ammonium-ion content of the extracts was determined by use of the method developed. A 10-g sample of each soil, which had previously been air-dried and sieved to 2 mm, was weighed and shaken for 1 h with 40 ml of neutral 2 M potassium chloride solution. This extract was allowed to settle and the clear supernatant liquid was removed for analysis.Aliquots of 20 ml of each standard and soil extract were pipetted into a series of 50-ml beakers and 2.5 ml of concentrated sulphuric acid were added to each. The solutions were mixed thoroughly and left to cool for 5-10 min before a 5-ml aliquot of each was taken for the analysis. Each sample was injected into 5ml of 40% m/V sodium hydroxide solution, which was being stirred at a constant rate in the generation cell, when the integrator signalled the start of the 2-min integration. The emission signal at 336.0 nm was integrated and displayed on the chart recorder. The results obtained for the analysis of the six soil samples studied are given in Table 111. It is evident that the results obtained using the proposed method are higher than those previously rep~rted.~ owing to the nature of soil samples the nitrogen level can change considerably with time by bacterial action; the sulphuric acid added to the samples affected the results obtained, possibly by breakdown of organic nitrogenous materials present in the soil extracts ; the use of integration yields greater accuracy than peak-height measurement owing to the decrease in the rate of generation of the gaseous analyte caused by precipitate formation when interfering ions are present in soil extracts.Various explanations can be suggested : 1. 2. 3. Soil sample Drumforber . . Ardconnon . . Smiddyhowe Wartle Waulkmill Strachan Logie Newton . . Average recovery Shaggart . . . . (proposed method taken as loo%), % TABLE: I11 RESULTS OF ANALYSIS OF SOIL EXTRACTS NH,+/cLg g-l - Proposed From method ref.5 . . 14.1 10.0 . . 13.5 10.1 . . 15.2 11.6 . . 18.7 13.6 . . 16.7 15.2 . . 20.4 16.2 Recorrerv of 20 p g g-1 NH,+ added t o soil NH,+/ Recovery, tG 8-l % 34.3 101 33.9 104 35.2 100 38.1 97 37.3 104 39.5 96 .. 100 78 100 Recovery in the absence of H,SO, 2 xnin integration b - 7 NH,+/ Recovery, NH,+/ Recovery, M8-l % tG g-' 0 0 I 7 L-- 3 min integration 32.0 89 34.1 100 30.9 87 32.5 95 32.8 88 35.5 102 38.0 97 37.6 94 35.2 92 37.8 106 35.6 76 39.5 96 88 99 To test these possible explanations, a number of recovery experiments were carried out under various conditions. The recovery of 20 pg g-l of ammonium as ammonium sulphate added to the soils was tested.Firstly, the recovery using the proposed method was investi- gated; as can be seen from Table I11 essentially complete recovery of ammonium ion was found, indicating that no cumulative interference effects are present, the mean value of the recoveries was found to be 100% and the spread of values was acceptable as the repetitive determination of a single soil extract gave a relative standard deviation of 0.04. The effect of the sulphuric acid was tested by omitting its addition prior to the analysis, using otherwise identical conditions. As can be seen from Table 111, the average recovery was only SSyO of that obtained previously, although this was still higher than that obtained in earlier work,5 which was about 78% of that obtained using the method proposed here.As the omission of the acid resulted in signals of greater duration being obtained, the integra- tion time was increased to 3 min and the recovery experiment repeated. As can be seen in Table 111, an average recovery of 99% was obtained; the slightly larger spread in the values obtained reflects the decrease in precision obtained when the acid was omitted. The lowNovember, 1978 AMMONIUM-NITROGEN BY FLAME EMISSION SPECTROMETRY 1115 recovery obtained using the second method suggests that there is indeed some decrease in the rate of generation of the analyte as a result of precipitate formation; slight precipitates were observed with all of the samples examined. Although it is difficult to compare the present method of generation of ammonia from alkaline solution with the method involving hypobromite oxidation to nitrogen, it was reported5 that, when using the nitrogen generation technique, interferents such as A13+ and Mn2+ did cause precipitation, which slowed down the analytical signals; integration was suggested as a means of overcoming this problem although in the actual analysis the peak height of the analytical signals was used.This could explain the difference in the results obtained with the two methods although no interferences were indicated by the recovery test on 20 p g g-l obtained using the method described earlier.5 It therefore appears that the use of sulphuric acid in the proposed technique does not affect the results of the soil analysis and can be used to obtain more rapid analyses with better precision and lower detection limits.The discrepancy between the results obtained using the two techniques is probably due to degradation of the sample in the time interval between the development of the two techniques of analysis and an interference in the hypobromite oxidation technique, which could be solved by integration of the signals obtained . Conclusion It has been shown that, by using the system described, ammonium-nitrogen in aqueous solution can be determined in trace amounts by using flame emission spectrometry a t the NH 336.0-nm band head, after displacement of gaseous ammonia from alkaline solution. Both the precision obtained and the detection limit compare favourably with those given by the currently available techniques and the modification of the gas-generation procedure using the heat of neutralisation of sulphuric acid and stirred solutions has proved a reliable method of liberating ammonia from solution when coupled to an electronic integrator to record the signals obtained. The method should also be readily adaptable to the determination of ammonium ion in other matrices as well as the detection of nitrate by using a system similar to that described by Cresser3 and the determination of the total nitrogen content of soils after the use of the Kjeldahl procedure. 1. 3. 3. 4. 5. 6. 7. 8. 9. References Cresser, M. S., Analytica Chim. A d a , 1976, 85, 253. Cresser, M. S., Lab. Pract., 1977, 26, 19. Cresser, M. S., Analyst, 1917, 102, 99. Belcher, R., Bogdanski, S. L., Calokerinos, A. C., and Townshend, A., Analyst, 1977, 102, 220. Alder, J . F., Gunn, A. M., and Kirkbright, G. F., Analytica Chim. Acta, 1977, 92, 43. Eder, J. M., Denkschr. Wein Akad.. 1893, 60, 1. Pcarse, R. W. B., and Gaydon, A. G., “The Identification of Molecular Spectra,” Third Edition, Dagnall, R. M., Smith, D. J., Thompson, K. C . , and West, T. S., Analyst, 1969, 94, 871. Gaydon, A. G., “The Spectroscopy of Flames,” Chapman and Hall, London, 1974. Chapman and Hall, London, 1965. Received May 19th, 1978 Accepted June Sth, 1978

ISSN:0003-2654    

DOI:10.1039/AN9780301104

出版商:RSC    

年代:1978

数据来源: RSC

摘要:

1116 Anal-yst, November, 1978, VoE. 103, pp. 1116-1120 Spectrophotometric Determination of 6,7=Dihydroxy- coumarin and Its Methoxy Derivatives E. Celon A. Guiotto and P. Rodighiero Istituto di Chimica Organica dell' Universit&, Via Marzolo, 1 , 35100 Padova, Italy Centro di Studio della Chimica del Farmaco e dei ProdotCi Biologicamente attivi del CNR, Istituto d i Chimica Farmaceutica dell' Universith, Via Marzolo, 5, 35100 Padova, Italy A spectrophotometric method for the determination of the single components in a mixture of 6,7-dihydroxycoumarin and its methoxy derivatives is described. The method is based on the determination of 6,7-dihydroxy- coumarin after complexation with molybdophosphoric acid in buffered solution and of 6-hydroxy-7-methoxy-, 7-hydroxy-6-methoxy- and 6,7- dimethoxycoumarin by treatment with sodium methylate solution followed by difference absorbance measurements.This procedure has also been applied successfully to the study of the methylation reaction of 6,7-dihydroxy- coumarin. Keywords : 6,7-Dihydroxycoumari?z ; 6-hydroxy-7-~tethoxycoumarin ; 7-hydroxy- 6-methoxycouunaran ; 6,7-dimet?toxycoumarin ; sfiectrophotometry 6,7-Dihydroxycoumarin and its monomethyl ethers are natural products that frequently occur in plant extractives.l In many instances two or more of these coumarins2-6 can be found in the same plant and therefore a method for their identification and determination is useful in establishing the composition of several natural products. Many studies have been carried out on the identification and characterisation of individual coumarins by spectroph~tometric,~ flu~rimetric,~,~ mass spectrometriclOJ1 and gas - liquid chromato- graphic12 procedures. However, no methods have been reported for their simultaneous determination. In this paper a spectrophotometric method for the determination of single components in a mixture of 6,7-dihydroxycoumarin with its methoxy derivatives is described.The availability of this method has also permitted a quantitative investigation of the methylation react ion of 6,7-dihydroxycoumarin. Experimental Reagents All reagents for microanalysis were used without further purification unless otherwise specified. Acetone and methanol (reagent grade) were dried by the usual methods; the water content was less than 0.1% by the Karl Fischer method.Potassium carbonate was dried at 270 "C for 24 h and kept under anhydrous conditions. 6,7-Dihydroxycoumarin and its 6-methoxy, 7-methoxy and 6,7-dimethoxy derivatives were prepared as described in the literature.13J4 Standard sample solutions. Dissolve accurately weighed amounts of 6,7-dihydroxy- coumarin, 6-hydroxy-7-methoxycoumarin, 7-hydroxy-6-methoxycoumarin and 6,7- dimethoxycoumarin in methanol. Dilute each solution to a concentration of 0.5 mmol l-1 with methanol. Mix 50 ml of 0.1 M potassium hydrogen phthalate solution with 2.6 ml of 0.1 M hydrochloric acid. Dilute to 100 ml with distilled water and adjust the pH of the solution to 3.80. Aqueous molybdophosphoric acid solution, 1 yo. Sodium methylate solution, 1 N. This solution was prepared according to the AOAC method.15 Clark - Lubs bz@er solution, p H 3.80.CELON, GUIOTTO AND RODIGHIERO 1117 Apparatus was used.Spectrophotometer. A Hitachi, Model 156 , digital double-wavelength spectrophotometer Procedure Determination of 6,7-dihydroxycoumarirt Place a suitable aliquot (0.2-2 ml) of the standard solution of 6,7-dihydroxycoumarin in a 10-ml calibrated flask, add 5 ml of buffer solution followed by 0.4 ml of molybdophosphoric acid solution and dilute with water to the mark. Transfer some of the solution into a 1-cm cell and after 3 min record at 410 nm the absorbance of the solution against a blank solution. The molar absorptivity at 410nm, as determined by the least-squares method, is 8105 1 mol-l cm-l. The intensity of the colour appears to be stable for 1 - 45 min.Determination of 6,7-dihydroxycoumarin and its methoxy derivatives Place a suitable aliquot (0.2-2 ml) of the standard solution of each compound in a 10-ml calibrated flask, add 1 ml of sodium methylate (sodium methoxide) solution and dilute with methanol to the mark. Transfer some of the solution into a 1-cm cell and after 3 min record the absorbance of the solution against a blank solution at 401 and at 341 nm, and the difference between the absorbances at 325 and 358.8 nm. The molar absorptivities at 401 and 341 nm and the difference between the molar absorptivities at 325 and 358.8 nm are, respectively: 15364, 4365 and 5 151 1 mol-l cm-l for 6,7-dihydroxycoumarin; 20511, 4947 and 8750 1 mol-l cm-l for 6-methoxy-7-hydroxycoumarin; 7328, 2401 and 0 1 mol-l cm-l for 6-hydroxy-7-methoxycoumarin; and 0, 12 030 and 0 1 mol-l cm-1 for 6,7-dimethoxy- coumarin.The absorbances were measured after 3 min because the intensity of the colour appears to be stable for 1 - 6 min. For unknown mixtures the concentrations of the individual coumarins can be calculated by using the following equations: A 410 = ~~,41oCa - A358.8 = (%325 - %,%8.tdca -k (Eb,325 - Eb,%8.s)Cb A401 = ~a,4oiCa + ~b,4oiCb + ~c,goiCc A341 = ~a,34iCa + ~b,34iCb + ~c,34iCc + ~,84iCd where and Ca are the molar absorptivity and concentration, respectively, of 6,7-dihydroxy- coumarin ; E b and cb are the molar absorptivity and concentration, respectively, of 7-hydroxy- 6-methoxycoumarin ; E~ and C, are the molar absorptivity and concentration, respectively, of 6-hydroxy-7-methoxycoumarin ; and Ed and cd are the molar absorptivity and concentration, respectively, of 6,7-dimethoxycoumarin.Methylation of 6,7-Dihydroxycoumarin with Dimethyl Sulphate To a stirred solution of 0.2 mmol of 6,7-dihydroxycoumarin in 20 ml of dry acetone were added 0.2 mmol of anhydrous potassium carbonate and then 2 mmol of dimethyl sulphate in 20 ml of acetone, under an inert atmosphere. The course of the reaction was checked at regular intervals as follows. A 1.5-ml sample was taken from the reaction mixture and passed through a Millipore filter (3.0 pm) and 1 ml of the filtrate was transferred into a 10-ml calibrated flask. To stop the reaction 1 drop of concentrated ammonia solution was added and the solution was evaporated to dryness under vacuum.The residue was dissolved in methanol and diluted to the mark. The spectrophotometric determination was carried out on an aliquot of the solution obtained. Results and Discussion 6,7-Dihydroxycoumarin and its methoxy derivatives show similar ultraviolet - visible spectra with an absorption maximum at about 345 nm in solution in methanol (Fig. 1). However, on treating the reaction mixture with a pH 3.80 buffer solution and 1% molyb- dophosphoric acid, as reported for catechol determination,16 the spectrum exhibits a1118 CELON et al. : SPECTROPHOTOMETRIC DETERMINATION OF Analyst, VoZ. 103 500 250 300 350 400 450 Wavel engt h/n m Fig. 1. Spectra of 5 x moll-]. solutions of coumarins in metha- nol : A (-), 6,7-dihydroxycoumarin; B (- .-), 7-hydroxy-6-methoxy- coumarin; C (--), 6-hydroxy-7-methoxycoumarin; and D ( a .), 6,7- dimethoxycoumarin. strong absorption band centred at 377nm, due to the complex formed between 6,7- dihydroxycoumarin and molybdophosphoric acid (Fig. 2). The absorption bands of the methoxycoumarins remain unchanged. The methoxycoumarins can then be determined by treating the mixture with 0.1 N sodium methylate solution (Fig. 3).' Further, in order to determine the individual coumarins in the mixture, a double-wavelength spectrophotometer was employed. In the double-wavelength spectrophotometer the light emitted by the source enters the two photometers to form a dual beam of wavelengths A, and A2. The two beams are re- combined and guided to the sample cell in order to measure the difference in absorbance, AA.Beer's law is generally valid at any wavelength at which there is absorption. Thus the difference in absorbance, AA, at two wavelengths A, and A, is proportional to concentra- tion. Direct determination of a single component is possible if the interfering compounds show the same absorbance at two different wavelengths. At these wavelengths variation of the concentration of the interfering compounds does not influence the AA value of the 0.800 1 a, 6 0.600 $ 0.400 2 0.200 0 - 300 350 400 450 500 . Wavelengthhm Fig. 2. Spectra of 5 x rnol 1-1 solutions of coumarins in buffered solution (pH 3.8) containing 1% molybdophos- phoric acid: A (-), 6,7-dihydroxycoumarin; B (-. -), 7-hydroxy-6-methoxycoumarin ; C (- -) , 6-hydroxy-7- methoxycoumarin; and D (., .), 6,7-dirnethoxycoumarin.November, 1978 6,7-DIHYDROXYCOUMARIN AND ITS METHOXY DERIVATIVES 1119 1.200 o) 1.000 u 2 0.800 L 0 2 0.600 a 0.400 0.200 200 250 303 350 400 450 500 Wave I en gt h/n rn Fig. 3. Spectra of 5 x 10-5 mol 1-1 solutions of coumarins in methanol that is 0.1 N in sodium methylate: A (-), 6,7-dihydroxycoumarin ; B (-.-), 7-hydroxy-6-methoxy- coumarin C ; (- * -), 6-hydroxy-7-methoxycoumarin ; and D ( a -), 6,7-dimethoxycoumarin. mixture and the concentration of one component can be read directly. As both the 6-hydroxy-7-methoxycoumarin and 6,7-dimethoxycoumarin show the same absorbance a t 325 and 358.8 nm, the determination of the 7-hydroxy-6-methoxycoumarin is then possible as the amount of 6,7-dihydroxycoumarin is known through the molybdophosphoric acid method.After the determination of 6,7-dihydroxycoumarin and 7-hydroxy-6-methoxycoumarin has been carried out, the 6-hydroxy-7-methoxycoumarin is determined by measuring the absorb- ance at 401 nm at which wavelength the 6,7-dimethoxycoumarin does not interfere. Having determined the concentrations of these three coumarins, it is possible to detect quantitatively the concentration of the last component of the mixture, 6,7-dimethoxycoumarin, by measuring the absorbance at 341 nm. Beer’s law is obeyed for all the compounds examined in the concentration range from 1 X lod4 to 1 x loa6 M. This method appears to be reliable, as is shown by the mean recoveries from a mixture of the single components. The mean percentage recoveries (& standard deviations), calculated from the results of ten determinations, were 101 ( *0.5) , 98 (50.8), 98 (& 1 .l) and 99 (* 1 .O) for 6,7-dihydroxycoumarin, 7-hydroxy-6-methoxy- coumarin, 6-hydroxy-7-methoxycoumarin and 6,7-dimethoxycoumarin, respectively.By using the proposed analytical method it was possible to study the course of the methylation reaction of 6,7-&hydroxycoumarin reported by Zagorevskii and Sovzenk0,~~1~~ who identified the products only qualitatively. The methylation reaction was carried out with a large excess of methylating reagent in order to ensure complete methylation and a reaction rate suitable for sampling. The amount of potassium carbonate was kept low in order to prevent loss of 6,7-dihydroxycoumarin by absorption on the salt.Moreover, the experiments were performed under dry conditions as small amounts of water affect the reaction rate and the product distribution. The results obtained show that methylation of 6,7-&hydroxycoumarin gives 6-hydroxy- 7-methoxycoumarin and not 7-hydroxy-6-methoxycoumarin. Moreover, 50% of the 6,7- dihydroxycoumarin appears to be converted into 6-hydroxy-7-methoxycoumarin after 42 min. These results, obtained by determination of all the products involved in the methylation reaction, suggest that methylation does not occur at the 6-position under the experimental conditions described. Attempts to determine the kinetic parameters of this methylation reaction are now in progress. Our thanks are due to A. Berton and G. Biasioli for their skilful technical assistance in performing this work.1120 CELON, GUIOTTO AND RODIGHIERO References 1.2. 3. 4. 5. 6. 7. 8. 9. 10. 11. 12. 13. 14. 15. 16. 17. 18. Soine, T. O., J . Pharm. Sci., 1964, 53, 231. Jain, M. P., Thakur, R. S., and Rao, P. R., Curr. Sci., 1976, 45, 640. Dargaeva, T. D., and Brutko, L. I., Khim. Prir. Soedin., 1976, 536; Chew. Abstr., 1976,85, 174 265 k. Dzhumyrko, S . F., Khim. Prir. Soedin., 1976, 63’7; Chem. Abstr., 1976, 85, 174 266 m. Stohr, H., and Herrmann, K., 2. Lebensmitteluntcrs. u. -Forsch., 1975, 159, 219. Brown, D,, Asplund, R. O., and McMahon, V. A., Phytochemistry, 1975, 14, 1083. Boheme, H., and Severin, T., Arch. Pharm., B e d , 1957, 290, 486. Crosby, A. G., and Berthold, R. V., Analyt. Biochem., 1962, 4, 349. Ichimura, Y., J. Pharm. Soc. Japan, 1960, 80, 771. Barnes, C. S., and Occolowitz, J. L., Aust. J . Chem., 1964, 17, 975. Shapiro, R, H., and Djerassi, C., J . Org. Chem., 1965, 30, 955. Furuya, T., and Kojima, H., J . Chromat., 1967, 29, 382. Braymer, H. D., Shetlar, M. R., and Wender, S. H., Biochim. Biophys. Acta, 1960, 44, 163. Bargellini, G., and Monti, L., Gazz. Chim. ItaZ., 1915, 45, 90. Horwitz, W., Editor, “Official Methods of Analysis of the Association of Official Agricultural Chemists,” Strachota, J., and Kotasek, Z., Chemickd Listy, 1958, 52, 1093; Analyt. Abstr., 1959, 6, 1804. Zagorevskii, V. A,, and Sovzenko, 2. D., J . Gen. Chem. U.S.S.R., 1964, 34, 4046. Zagorevskii, V. A., and Sovzenko, Z. D., J . Gen. Chem. U.S.S.R., 1963, 33, 1655. Ninth Edition, Association of Official Agricultnral Chemists, Washington, D.C., 1960, p. 343. Received December 29th, 1977 Accepted May Sth, 1978

ISSN:0003-2654    

DOI:10.1039/AN9780301116

出版商:RSC    

年代:1978

数据来源: RSC

摘要:

Analyst, November, 1978, Vol. 103, pp. 1121-1126 Spectrophotometric Determination of Methylmercury in Fish Tissue with Dithizone 1121 Using a Dual-wavelength Procedure P. Jones and G. Nickless Department of Chemistry, University of the West Indies, St. A ugustine, Trinidad De9artment of Inorganic Chemistry, School of Chemistry, The University, Bristol, BS8 1TS A dithizone spectrophotometric procedure is described for the measurement of trace concentrations of methylmercury salts. The application of a simple equation using absorbance measurements taken a t two wavelengths cancels out small differences in excess of dithizone arising between blank and sample, thus ensuring good precision in the range 0.1-4.0 pg ml-l. The developed method is used in combination with the Westoo extraction procedure to determine methylmercury concentrations in fish tissue. A crab meat sample contained less than 0.04 pg g-l, and values for eight tuna fish ranged from 0.08 to 0.41 pg g-l.Keywords : Methylmercury determination ; fish tissue analysis ; dual- wavelength spectrophotometry ; dithizone Most legislation concerning the level of mercury in foodstuffs generally requires the deter- mination of total mercury only, although maximum allowable concentrations were arrived at by assuming the presence of highly toxic methylmercury compounds.l For this reason, it is important to be aware of the proportion of methylmercury compounds present in any type of sample. The most commonly used method for the determination of methylmercury in biological tissue, based on Westoo’s original work,2 is a three-stage solvent extraction procedure followed by gas - liquid chromatography with electron-capture detection.The results of round-robin surveys carried out by Finger and Bennett3 indicated that not many routine testing laboratories (10 out of 64) were equipped or confident enough to carry out the determination of methylmercury by gas - liquid chromatography. A simpler method was developed, still based on the Westoo extraction process, but finishing with a diphenylthio- carbazone (dithizone) spectrophotometric procedure. The absorbance of a methylmercury solution in the presence of excess of dithizone is measured against a blank containing dithizone at approximately the same concentration. The method relies on the application of a simple equation, involving absorbance measurements at two wavelengths to cancel out the effect of the small, unavoidable variations in dithizone concentrations between blank and sample, thus giving the absorbance due only to the methylmercury - dithizone complex at 628 nm.This ensures a good precision, even close to the detection limit, with sufficient sensitivity for the determination of methylmercury in fish flesh. Experimental Reagents All reagents were of analytical-reagent grade, except methylmercury( 11) chloride. Benzene, distilled. Dithixone solution. Prepare a solution of approximately 0.02% m/ V dithizone in benzene, Store in a such that on 10-fold dilution the absorbance at 628 nm will be close to 1.8. refrigerator when not in use and renew monthly.Hydrochloric acid, 6 N. Sulphuric acid, approximately 1 N. Sodium chloride. Cysteine solution. Dissolve 1 .OO g of cysteine hydrochloride monohydrate, 0.775 g of sodium acetate trihydrate and 12.5 g of anhydrous sodium sulphate in distilled water and dilute to 100 cm3. Prepare freshly every 3 d.1122 JONES AND NICKLESS : SPECTROPHOTOMETRIC DETERMINATION Analyst, VoZ. 103 Prepare a stock solution of concentration 1000 p.p.m. m/V (with respect to mercury) in benzene. Dilute the stock solution as necessary for partition and recovery experiments and spectrophotometric standards. Apparatus MethyZmercuyy(II) chloride solution. A Hilger and Watts Spectrochem non-scanning visible spectrophotometer was used. A centrifuge capable of a speed of 2000 rev min-l and a homogeniser are required. Procedure Extraction The extraction of methylmercury from fish tissue is based on that used by Westoo2 (see also Kamps and McMahon*).The figures given in parentheses below are for 40-g samples expected to contain very low concentrations of methylmercury. Homogenise 20 (40) g of the sample with water and rinse the homogeniser quantitatively. Use a total of 55 (110) cm3 of water for these procedures. Add 14 (28) cm3 of concentrated hydrochloric acid and 10 (20) g of sodium chloride and mix. Add 70 (140) cm3 of benzene and shake the mixture for 5 min by hand or 15 min on a shaking machine. Allow the mixture to settle and transfer the top layer, including any emulsion, into a centrifuge tube or flask. Cover the container with aluminium foil and centrifuge it for 5 min at 2000 revmin-1.Transfer as much of the benzene layer as possible into a separating funnel, noting the volume. Add 6.0 (80) cm3 of cysteine solution and shake the funnel vigorously for 2 min. Allow the mixture to settle and run off the emulsified bottom layer into a small tube and centrifuge it for 5-min periods, stirring the emulsion between runs with a glass rod, until the solution is clear. Transfer as much of the clear solution as possible, again noting the volume, into a small separating funnel and acidify it with 0.6 cm3 of 6 N hydro- chloric acid for every 1.0 cm3 transferred. Extract the solution with 5.00 (6.00) cm3 of benzene by shaking for 2 min and allow the mixture to settle. Measurement Pipette 4.00 cm3 of the above benzene extract into a 10-cm3 test-tube (with a ground- glass neck) containing 4.0 cm3 of 1 N sulphuric acid.Add 0.40 cm3 of 0.02% dithizone solu- tion, stopper the tube and shake it for 1 min. Allow the layers to separate and transfer the benzene solution into a 1-cm glass cell, removing aqueous droplets with a small strip of filter-paper between the pipette and the cell, and close it with a lid. Prepare a blank with 4.00 cm3 of benzene in the same way. At 628nm. Zero the instrument with the sample and measure the absorbance of the blank. If the absorbance is less than zero, interchange the cells and record the absorbance as negative. At 475nm. Zero the instrument with the blank and measure the absorbance of the sample. If the absorbance is less than zero, interchange the cells and record the absorbance as negative.Then, the absorbance due to the methylmercury - dithizone complex a t 628 nm is 0.5s + T. Compare the absorbance against a calibration graph constructed by carrying aliquots of methylmercury( 11) standards through the same procedure. Measure the absorbances as follows. Let the absorbance be S. Let the absorbance be T . Results and Discussion Development of Dual -waveleng th Procedure Dithizone, a well known reagent for the determination of trace amounts of inorganic mercury, can also be used for the spectrophotometric determination of organomercury salts,5,6 but has rarely been applied to environmental samples.' The dithizone reagent can be made highly specific for organomercury compounds when used in conjunction with the solvent-extraction process, because interfering ions such as mercury( 11) , copper( 11) and silver(1) are not The sensitivity is lower than that for inorganic mercury, but this can be compensated for by using a fairly high sample mass and a low final extraction volume, What is more important is that the sensitivity can be seriously limited by poorNovember, 1978 OF METHYLMERCURY IN FISH TISSUE WITH DITHIZONE 1123 precision at low concentrations of organomercury salts.This became clear when it was found that excess of dithizone had to be present in order to stabilise the lowest concentrations of methylmercury - dithizone complex, and subsequently there was difficulty in exact matching of blanks for single-wavelength determinations. The fact that dithizone has two peaks in the visible spectrum can be exploited in a dual-wavelength procedure to maintain high precision for organomercury - dithizone measurements.Irving and Coxg used absor- bances at 628 and 475nm measured against a non-absorbing blank to calculate the absorbance due only to the organomercury complex at 477 nm. Even higher precision is possible if this is adapted to the method of Allen and Hammakerlo for the analysis of a two- component system, where the blank contains one of the components at approximately the same concentration as in the sample. When measured against a blank containing dithizone at the same initial concentration as in the sample, the methylmercury concentration can be related to the decrease in absorbance at 628 nm or the increase at 475 nm.The different Amax. value for methylmercury - dithizone (477 nm) accounts for the bathochromic peak shift from 450 nm with increasing amounts of methylmercury. The decrease in absorbance at 628 nm is the most sensitive measure of the methylmercury complex, but for small amounts of methylmercury [down to 0.5pg (0.1 p.p.m.)] it can be seen that small variations in dithizone concentration between the blank and sample will seriously distort or even mask the absorbance due to the methylmercury. For example, when using calibrated pipettes and a volatile solvent it is not unreasonable to assume a relative precision of 0.25% for each pipetting operation. There are four such opera- tions involved in preparing a blank and sample, giving a relative precision of 1 yo.Therefore, the variation in concentration of dithizone between blanks and samples in absorbance terms is 1.80 5 0.018. This is small as far as the dithizone is concerned, but would lead to a precision of 50% and 25% for 0.2 and 0.4 p.p.m. of methylmercury, respectively. When other errors inherent in a spectrophotometric procedure, such as photometric error, are also taken into account there is a very poor and unacceptable precision in the concentration range of interest. The dual-wavelength procedure, which overcomes this problem, is derived as follows. Fig. 1 shows the effect of adding methylmercury(I1) chloride to excess of dithizone. 2.0 1.8 1.6 8 1.4 m 1.2 $ 0.8 0.6 0.4 0.2 f $ 1.0 I- A 1 ~ _ _ 400 450 500 550 600 650 700 Wavelengthhm Fig.1. Change in dithizone absorption curve (A) with the addition of (B) 10 p g and (C) 20 pg of mercury as methylmercury(I1) chloride. The absorbance (S) of a sample containing a small amount of methylmercury salt in the presence of excess of dithizone, measured against a dithizone blank at 628 nm, will be the sum of the decrease in absorbance due to the formation of the complex ( y ) and the increase or decrease in absorbance due to inexact matching of blank and sample dithizone concentra- tions (x). Similarly, the absorbance (T) of a sample measured at 475 nm will be the sum of the increase in absorbance due to the complex (4) and the increase or decrease in absorbance due to inexact matching of sample and blank ( p ) . Therefore, s = x + y T = p + q Assuming a linear relationship between x and j5 and between y and q, we have a pair of1124 JONES AND NICKLESS : SPECTROPHOTOMETRIC DETERMINATION Analyst, VOz.103 simultaneous equations, S = x + y and T = mx + ny, where m and n are proportionality constants. Solving for y , which is the absorbance change a t 628 nm due only to the methyl- mercury complex, we have mS - T y = -- m -- n m is simply the ratio between the dithizone absorbances at 475 and 628 nm. The value of m found was 0.52. n is calculated by constructing calibration gra.phs of the change in absorbance with con- centration of methylmercury(I1) chloride (in the presence of excess of dithizone) at each of the two wavelengths. The calibration range of methylmercury( 11) chloride standards should be 1-8 p.p.m.so as not to be significantly affected by small variations between the blank and sample dithizone concentrations. The ratio of the two slopes, n, was found to be -0.43. Hence, 0.52s - T = 0.95 which can be simplified, without loss of accuracy, to JJ = 0.5s - T The calculated values of y (the absorbance change due only to the methylmercury complex a t 628 nm) will be negative because of the negative correlation between the absorbance change at 628nm and the concentration of methylmercury (see Fig. 1). For convenience only, it was decided to reverse the sign of the measured absorbance a t 628nm ( S ) , and to change the relationship between S and T from negative to positive such that calculated values for y would then be positive, i.e., y = 0.5s + T Hence, as stated above, a decrease in absorbance between the blank and sample at 628 nm ( S ) will be given a positive sign and an increase a negative sign, whereas the positive and negative values at 475 nm ( T ) will remain as measured.The dual-wavelength procedure was used to calculate the absorbances of methylmercury standards a t 628 nm for two calibration ranges. The results are given in Table I. TABLE I CALIBRATION OF METHYLMERCURY(II) CHLORIDE STANDARDS Amount of Hg added/pg 1.0 2.0 3.0 4.0 5.0 Check blank Total volume*/ml 4.50 4.62 4.73 4.84 4.95 4.40 Concentration of Hg, p.p.m. 0.22 0.43 0.63 0.83 1.01 0 Absorbance Corrected J-, absorbance 628 nm (S) 475 nm ( T ) at 628 nm ( y ) +0.230 -0.075 0.040 + 0.220 - 0.035 0.075 +0.242 -0.018 0.103 +0.255 +0.025 0.153 + 0.290 +0.032 0.177 +0.175 - 0.090 - 0.002 2.5 4.43 0.56 1-0.130 + 0.040 0.105 5.0 4.45 1.12 +0.190 +0.110 0.205 10.0 4.50 2.22 + 0.425 +0.175 0.388 15.0 4.55 3.30 +0.590 + 0.265 0.560 20.0 4.60 4.35 + 0.880 +0.345 0.785 Check blank 4.40 0 + 0.040 -0.023 -0.003 * This includes the addition of methylmercury standards with a micropipette.The corrected absorbances at 628 nm using the equationy = 0.5s + T gave good straight- line plots, and Table I1 shows a comparison of single-wavelength (uncorrected) and dual- wavelength (corrected) methods after regression analysis of the calibration values. It is clear from the regression analyses that poor precision and a large uncertainty in the slope will give meaningless results for single-wavelength determinations, particularly for theNovember, 1978 OF METHYLMERCURY I N FISH TISSUE WITH DITHIZONE TABLE I1 COMPARISON OF SINGLE- AND DUAL-WAVELENGTH CALIBRATION GRAPHS AFTER REGRESSION ANALYSIS 1125 Uncorrected measurements a t 628 nm Regression analysis A - I f A Corrected measurements a t 628 nm parameter* 0.2-1 p.p.m.0.5-5 p.p.m. 0.2-1 p.p.m. 0.5-5 p.p.m. A 0.186 0.012 -0.001 0.00007 B 0.095 0.189 0.178 0.177 S.D.,,, . . .. . . 0.0148 0.037 8 0.005 3 0.0152 S.D.B . . . . . . 0.0175 0.010 1 0.006 2 0.004 1 Confidence interval of the slope (90%) . . 0.095 & 0.041 0.189 f 0.023 0.178 f 0.013 0.177 f 0.008 * Regression equation: y = A + Bx. y = absorbance measurement for a concentration of methylmercury x ; A = intercept on y axis; B = slope (absorbance corresponding to 1.00 p.p.m.of Hg) ; S.D.,,. = standard deviation of y values about the regression line; S.D.B = standard deviation of the slope. Confidence interval of the slope (90%) = B f S.D.B X t,-,, where is the value of Student’s t forn-2 observations and a 90% probability level. lower range. On the other hand, the dual-wavelength method shows very good agreement between the slopes and little loss in relative precision for the lower range. A more quantita- tive test of precision is shown in Table 111, where it can be seen that the coefficients of variation for eight standards at the 0.4 p.p.m. level are 5.6% and 38.7% for corrected and uncorrected absorbances, respectively. TABLE I11 PRECISION TEST FOR METHYL MERCURY(I1) CHLORIDE STANDARDS (0.43 p.p.m. ??i?/V Hg) Absorbance r- 628 nm 475 nm + 0.083 + 0.027 + 0.092 + 0.085 + 0.107 +O.lSO +0.095 +0.095 +0.018 Average .. .. . . 0.093 Standard deviation . . 0.036 Coefficient of variation . . 38.7% + 0.034 +0.057 +0.032 +0.027 +0.014 -0.001 + 0.025 + 0.025 -0.015 Corrected absorbance a t 628 nm 0.078 0.071 0.078 0.070 0.068 0.079 0.073 0.073 0.074 0.004 1 5.6% -0.006 (check blank) Application to Analysis of Fish Flesh Eight tinned tuna fish samples and one crab meat sample were extracted by the Westoo procedure2 and analysed for organomercury content by the dual-wavelength dithizone method. For accurate calculation of methylmercury levels in the fish flesh, over-all extrac- tion efficiencies need to be known. This is the product of the efficiency due to partition and the efficiency due to volume losses.The efficiency due to partition, using standard solutions of methylmercury(I1) chloride, was found to be 70% (83% in the first stage, 100% in the second stage and 84% in the third stage). The efficiency due to volume losses varies with the sample, depending on the intractability of the emulsion formed. For example, with sample 2, 50 ml of benzene were recovered from 70 ml in the first stage and 4.0 ml of cysteine solution were recovered from 6.0 ml in the second stage. This gives an extraction efficiency of 48%, and therefore an over-all extraction efficiency of 32%. The results for eight samples of tuna fish and one of crab meat are given in Table IV.1126 JONES AND NICKLESS TABLE IV DETERMINATION OF METHYLMERCURY IN FISH FLESH Sample Tuna fish .... Crab meat. . .. Sample No. 1 1, repeat 1, repeat? 2 3 4 5 5, repeat? 6 7 8 .. .. Mass taken/g 20 20 20 20 40 20 40 20 40 40 40 20 Over-all extraction efficiency,* % 32 27 27 32 33 32 25 32 24 42 25 32 Methylmercury found in final extract/ p g 2.64 1.98 3.80 1.38 0.99 1.21 1.65 4.40 1.74 2.92 1.54 t 0 . 2 5 Concentration of methylmercury in samplelpg g-l 0.41 0.37 0.70 0.22 0.08 0.19 0.17 0.69 0.18 0.17 0.15 (0.04 * Over-all extraction efficiency = [efficiency due to partition (70%) X efficiency due to volume loss]/lOO. t 10 pg of Hg as methylmercury(I1) chloride added to sample. Although dithizone reacts with most organoniercury salts of the type RHgX,g where X is any anion, all results are reported as methylmercury. This assumption is based on Westoo’s work,2s11 carried out over a number of years with many types of fish samples. Methylmercury was the only organomercury compound found and constituted a high percentage of the total mercury present.Thin-layer chromatography can be used as a check if other organomercury salts are thought to be present.12,13 The dual-wavelength procedure gives sufficient precision to permit the determination of methylmercury salts at concentrations down to 0.1 p.p.m. This level is equivalent to 0.07 pg g-l for a 20-g fish sample, assuming an over-all extraction efficiency of 30%, which is more than adequate for a reliable estimate of the maximum allowable concentration (0.5 pgg-1). Any problems involving accuracy are likely to arise from the extraction procedure itself. Westoo reported2 an average recovery of 91%, with a range of S2-102~0, for a wide variety of sample types.Double extractions should give better over-all extrac- tion efficiencies as in the modification by SCOPE.l* A number of semi-micro modifications have been put forward, such as that by Uthe et a1.,l5 but may not give sufficient sensitivity for a dithizone finish. For a recent review of the solvent extraction process, see the work of Sumino,16 with a discussion by Westoo.ll 1. 2. 3. 4. 5. 6. 7. 8. 9. 10. 11. 12. 13. 14. 15. 16. References Report from an Expert Group, Nord. Hyg. Tidskr., 1971, Suppl. 4. Westoo, G., Acta Chem. Scand., 1967, 21, 1790. Finger, J . H., and Bennett, T. B., in Krenkel, P. A., Editor, “Heavy Metals in the Aquatic Environ- ment,” Supplement to “Progress in Water Technology,” Pergamon Press, Oxford, 1975, p. 63. Kamps, L. R., and McMahon, B., J . Ass. Off. Analyt. Chem., 1972, 55, 590. Miller, V. L., Polley, D., and Gould, C. J., Analyt. Chem., 1951, 23, 1286. Kiwan, A. M., and Fouda, M. F., Analytica Chim. Acta, 1968, 40, 517. Murakami, T., and Yoshinaga, T., Jafian Analyst, 1971, 20, 878; Analyt. Abstr., 1972, 23, 2872. Sandell, E. B., in Clarke, B. L., Elving, P. J., and Kolthoff, I. M., Editors, “Chemical Analysis,” Volume 111, “Colorimetric Determination of Traces of Metals,” Third Edition, Interscience, New York, 1959, p. 633. Irving, H., and Cox, J. J., J . Chem. Soc., 1963, 466. Allen, E., and Hammaker, E. M., Analyt. Chem., 1952, 24, 1295. Westoo, G., in Krenkel, P. A., Editor, “Heavy Metals in the Aquatic Environment,” Supplement to “Progress in Water Technology,” Pergamon Press, Oxford, 1975, pp. 47-50. Westoo, G., Acta Chem. Scand., 1966, 20, 2131. Takeshita, R., Akagi, H., Fujita, M., and Sakagami, Y., J . Chromat., 1970, 51, 283. Scientific Committee on Problems of the Environment (SCOPE), “Environmental PoIlutants- Selected Analytical Methods,” SCOPE 6, Butterworths, London, 1975, p. 177. Uthe, J . F., Solomon, J., and Grift, B., J . Ass. Off. Analyt. Chem., 1972, 55, 583. Sumino, K., in Krenkel, P. A., Editor, “Heavy Metals in the Aquatic Environment,” Supplement to “Progress in Water Technology,” Pergamon Press, Oxford, 1975, pp. 35-45. Received February 13th, 1978 Accepted May 16th, 1978

ISSN:0003-2654    

DOI:10.1039/AN9780301121

出版商:RSC    

年代:1978

数据来源: RSC

摘要:

Analyst, November, 1978, Vol. 103, pp. 1127-1133 Determination of a Non-volatile N-Nitrosamine on a Food Matrix 1127 C. L. Walters, M. J. Downes, M. W. Edwards* and P. L. R. Smith British Food Manufacturing Industries Research Association, Randalls Road, Leatherhead, Surrey, KT22 TRY A method devised for the determination of N-nitrososarcosine, in which the N-nitrosamine in solution is denitrosated with hydrogen bromide to form volatile products that are rapidly removed and determined in a chemi- luminescence analyser, has been applied successfully to the same compound on powdered corn flakes. Differentiation of N-nitrososarcosine and a number of other N-nitrosamines and N-nitrosamides from inorganic nitrite was achieved by decomposing the nitrite with acetic acid prior to the denitrosa- tion of the N-nitroso compounds.In the presence of a secondary-amine receptor limited nitrosation can occur during the process of differentiation but this can be prevented through the use of ascorbyl palmitate. In differentiating between large amounts of nitrite and much lower levels of N-nitrososarcosine on corn flakes, using a chemiluminescence analyser, the duration of the response from the nitrite can be shortened by freeze-drying the food matrix in the presence of ascorbic acid. The spectrophotometric determination of N-nitrososarcosine as nitrosyl bromide released into solution by the action of hydrogen bromide was hindered by the presence of powdered corn flakes. Keywords : N- N itroso compound determination ; N-nitrososarcosine determination ; food analysis ; chemilwminescence analyser While volatile N-nitrosamines can be separated readily from a biological matrix by distilla- tion in steam, no similar separation procedure applicable to all non-volatile N-nitrosamines is available, some compounds of this type may be associated with the components of the matrix itself and therefore extraction procedures would be inefficient. As most N-nitroso compounds are carcinogenic to experimental animals, it would be advantageous to have available a direct method for the determination of N-nitrosamines and N-nitrosamides on a biological or other matrix without the necessity for prior extraction.In some situations, moreover, the identities of the precursors of the N-nitroso compounds involved are unknown and a method of screening for total non-volatile N-nitrosamines and N-nitrosamides would be valuable.A method has been devisedl in which the volatile products of denitrosation of a non- volatile N-nitroso compound such as N-nitrososarcosine can be determined by using a chemiluminescence analyser designed to respond selectively to nitrogen oxide. The most suitable denitrosating agent for this purpose was hydrogen bromide, the use of which was first proposed by Eisenbrand and PreussmannS2 In solution in acetic acid the release by this reagent of nitrosyl bromide, detected as inorganic nitrite, was restricted to N-nitroso compound^.^ Nitrite itself, or an alkyl nitrite, also gave a response, which could be differentiated from that due to N-nitroso compounds by means of a control determination in the absence of the denitrosating agent ; some N-nitrosamines derived from weakly basic secondary amines are, however, hydrolysed in aqueous solution without the use of hydrogen bromide.Further, nitrosyl bromide is a very reactive compound, which would be expected to react in solution with unsaturated and other compounds extracted concurrently from a biological or other matrix. Its reactivity is probably responsible for the low recoveries obtained for N-nitroso compounds added to foods or extracts of foods when assayed by the method of Eisenbrand and Preussmann,2 which is excellent for the determination of such compounds in simple solution. For these reasons, conditions have been chosen such that nitrosyl bromide arising from the denitrosation of N-nitroso compounds is liberated and dispersed rapidly as nitrogen oxide, * Present address: IPC Industrial Press, Dorset House, Stamford Street, London, S.E.l.1128 WALTERS et al.: DETERMINATION OF A Analyst, Vol. 103 for determination in a chemiluminescence analyser. Further, the addition of acetic acid prior to that of hydrogen bromide results in the evolution of nitrogen oxide from any nitrite present, thereby differentiating this nitrogen oxide from the gas produced from the denitrosa- tion of N-nitroso compounds, and providing a measure of the availability of metal or alkyl nitrites in the system. This methodl has been extended to the determination of N-nitrososarcosine on a food matrix in order to determine whether the release of nitrogen oxide by use of hydrogen bromide is impaired in this situation.In addition, the proposed method has been found to be equally applicable to N-nitroso derivatives of other amines representing a wide range of basicities. Finally, it has proved possible to differentiate between nitrite and N-nitrososarcosine impregnated on a food matrix when the nitrite is present in large excess. N-Nitrososarcosine was chosen in these studies as a typical non-volatile N-nitrosamine likely to be formed in a biological matrix. Alternative procedures for its determination after extraction have been devised by Iwaoka and Tannenbaum,* who employed photo- hydrolytic detection of the compound as inorganic nitrite, and by Eisenbrand et al.,5 who made use of single-ion mass fragmentography after trimethylsilylation.Experimental Caution-N-Nitroso compounds are highly carcinogenic and strictly controlled procedures should be adopted for all aspects of experimental work involving their use. Method of Assay of N-Nitrososarcosine by Using a Chemiluminescence Analyser This procedure was carried out in the manner described by Downes et aZ.,1 but with modifica- tions. In particular, the traps immersed in ice and acetone saturated with carbon dioxide were replaced with two traps containing 6 N sodium hydroxide solution. The converter was omitted. A 0.5-ml volume of a 15% m/V solution of hydrogen bromide in glacial acetic acid was used for denitrosation and, as before, the responses of individual amounts of N-nitrososarcosine were calculated from the integrated areas under the recorder peaks in combination with the relevant range factor.Method of Assay of N-Nitrososarcosine by Determination of Nitrosyl Bromide Released in Solution Samples of powdered corn flakes (10 g) on which various amounts of N-nitrososarcosine had been distributed as uniformly as possible were each suspended for 15min at room temperature in 25 ml of glacial acetic acid containing 3.0% m/V of hydrogen bromide. After sedimentation of the solid by centrifugation, 0.5 ml of 5.0% m/V sulphanilamide in hydrochloric acid (1 + 3) was added to aliquots of the supernatants followed after 3 min by 0.5 ml of 0.1% N-1-naphthylethylenediamine dihydrochloride solution. The resulting pink colours were allowed to develop for 15 min at room temperature prior to the determination of the absorbances at 545 nm, using as controls preparations obtained in the same way but in the absence of the denitrosating agent hydrogen bromide.Control experiments were also carried out without the addition of N-1-naphthylethylenediamine dihydrochloride in order to cover any changes in the spectra due to treatment of extracted pigments with hydrogen bromide. Freeze-drying of Powdered Corn Flakes Impregnated with Nitrite and Ascorbic Acid Slurries of powdered corn flakes in aqueous solutions containing the requisite amount of sodium nitrite were distributed as evenly as possible around the surface of a round-bottomed flask and dried by lyophilisation in the frozen condition. From additions of sodium nitrite of 200 mg kg-l dry mass, recoveries of 15-20 mg kg-l were obtained after freeze-drying; similar losses of nitrite in contact with food under refrigerated storage were noted by Shank and Newberne.6 Deter mination of Nitrite The method of the Draft International Standard ISO/DIS 29.18.2 was employed.November, 1978 NON-VOLATILE N-NITROSAMINE ON A FOOD MATRIX 1129 Results Determination of N-Nitrososarcosine on a Food Matrix by Using a Chemi- luminescence Analyser Fig.1 illustrates the response of the chemiluminescence analyser to the volatile products of the denitrosation with hydrogen bromide of 2.0 pg of N-nitrososarcosine applied to 10 g of powdered corn flakes, using the procedure of Downes et aZ.1 As with N-nitrososarcosine at the microgram level in simple solution, the detection of volatile products giving a response in the analyser was rapid, with little background noise although the pen did not return exactly to its original position.2 min H I 4-- Time Fig. 1. Response of chemiluminescence analy- ser to 2.0 pg of N-nitroso- sarcosine on 10 g of pow- dered corn flakes. An essentially linear relationship was obtained between the response of the chemi- luminescence analyser and the amount of N-nitrosamine over at least the range of 2-80 pg of N-nitrososarcosine. When the relationship was examined in greater detail at the lower end of the concentration range, a small intercept was noted on the response axis that was due to small responses obtained from unspiked corn flakes treated with hydrogen bromide in glacial acetic acid in refluxing 1,2-dichloroethane.The mean value obtained for l o g of corn flakes without the addition of N-nitrososarcosine was equivalent to 0.49 pg of N-nitro- sosarcosine, with a standard deviation of 0.0707 pg. Thus, the mean value plus three times the standard deviation gives a value of 0.7 pg of N-nitrososarcosine, which can be taken as the limit of detection under these conditions. Determination of N-Nitrososarcosine on a Food Matrix by Denitrosation to form Nitrite in Solution Table I presents the recoveries of various amounts of N-nitrososarcosine on powdered corn flakes suspended in glacial acetic acid when determined as the inorganic nitrite liberated following denitrosation with hydrogen bromide. No response at all was detected for added amounts of N-nitrososarcosine of less than 50 pg, which is equivalent to a concentration in the food of 5.0 rng k g l .1130 WALTERS et al.: DETERMINATION OF A TABLE I Analyst, VoE. 103 RECOVERY OF N-NITROSOSARCOSINE ADDED TO POWDERED CORN FLAKES WHEN DETERMINED AS INORGANIC NITRITE FOLLOWING DENITROSATION I N SOLUTION Amount of N-nitrososarcosine added/pg 10 23 34 41 50 100 200 500 1.0 Concentration of N-nitrososarcosine added to food/ mg kg-l 0.1 1.0 2.3 3.4 4.1 5.0 10 20 50 Recovery of added nitrosamine, yo 0 0 0 0 0 13 21 38 73 Repeatability and Reproducibility of Determinations Involving the Use of a Chemiluminescence Analyser Nine 10-g samples of powdered corn flakes each with 22 pg of added N-nitrososarcosine were analysed during 1 d by one operator with foreknowledge of the amount added, using the method of Downes et al.1 The results obtained were in the range 15.4-19.3 pg of N-nitrososar- cosine, the mean value being 17.52 pg, the standard deviation 1.53 pg and the coefficient of variation 8.7%.In comparison with the values reported for N-nitrososarcosine in solution, the repeatability for this compound on the food matrix was less good and the mean recovery was slightly reduced. Repeated determinations, using equal aliquo ts of a stock solution, of N-nitrososarcosine on 10-g samples of powdered corn flakes, were carried out on 1 d by two operators, neither of whom was aware of the concentration chosen, and the results are presented in Table 11; the amount of N-nitrosamine used for each determination was 2.0pg. The difference observed between the mean values obtained by the two operators was not statistically significant.TABLE I1 REPRODUCIBILITY OF DETERMINATION OF EQUAL ALIQUOTS OF N-NITROSOSARCOSINE ON POWDERED CORN FLAKES BETWEEN TWO OPERATORS USING THE CHEMILUMINESCENCE ANALYSER ON THE SAME DAY N-Nitrososarcosine detected/ p g r > Operator 1 Operator 2 1.73, 1.87 2.42, 2.22 A 1.58, 1.58, 1.98, 1.66, 2.08, 1.96, Mean .. .. * . 1.75 2.07 Standard deviation . . 0.170 0.285 Coefficient of variation . . 9.7% 13.8% t = 0.775 for 8 degrees of freedom. Application of the Chemiluminescence Analyser Procedure to Other N-Nitroso Compounds In addition to N-nitrososarcosine, other N-nitrosamines, such as N-nitrosodimet hylamine and N-nitrosodiphenylamine, gave volatile products following denitrosation with hydrogen bromide that were detectable by using a chemiluminescence analyser, as also did the N- nitrosamides N-nitroso-N-methylurea, -N-et hylurea and -N-met hylurethane. Further, the procedure developed for the differentiation of N-nitrososarcosine and nitrite1 has been extended successfully to these further N-nitrosamines and N-nitrosamides.Thus, the N-nitroso derivative of the weakly basic amine diphenylamine was sufficiently stable in the presence of acetic acid alone to permit the dispersal of nitrogen oxide from nitrite priorNovember, 1978 NON-VOLATILE N-NITROSAMINE ON A FOOD MATRIX 1131 to that from the denitrosation of N-nitrosodiphenylamine using hydrogen bromide in the same solvent. During the differentiation process, however, a small extent of nitrosation of secondary amines present was noted.This is illustrated by Fig. 2, which shows the chemiluminescence analyser response due to nitrogen oxide liberated from nitrite following the addition of acetic acid and the additional peak after the inclusion of hydrogen bromide, which occurred only when morpholine or another receptor amine was included. The extent of conversion of this secondary amine of intermediate basicity into its N-nitroso derivative ranged from 0.70% at a molar ratio of nitrite t o morpholine of 20:l to a mean value of 1.9% with a corresponding ratio of 1.2:l. However, the inclusion of 60 pmol of ascorbyl palmitate to a concentration of 2.0 mM prevented completely the nitrosation of 1.8 pmol of diphenylamine by 0.14 mmol of nitrite following the addition of acetic acid under the test conditions.As in practice it may be necessary to distinguish between N-nitroso compounds and nitrite in a food or other matrix with the latter present at much higher concentrations than those of N-nitrosamines or N-nitrosamides, the procedure developed for the differentiation in solution1 has been extended to N-nitrososarcosine on powdered corn flakes impregnated with sodium nitrite. However, corn flakes freeze-dried following the addition of 200 mg of sodium nitrite per kilogram dry mass resulted in the evolution of large amounts of nitrogen oxide following treatment with acetic acid, which took about 60 min to disperse sufficiently to permit the determination of N-nitroso compounds also added but a t the microgram level.Nevertheless, the treatment of a food matrix containing 200 mg k g l of sodium nitrite with a 0.1% aqueous solution of ascorbic acid not only reduced the nitrite content further during the freeze-drying process but also shortened the time required to complete the breakdown of nitrite to at most 10 min prior to the determination of the N-nitrosamine at high sensi- tivity. Fig. 3 illustrates the ability to differentiate in this manner between 2.7 pg of N-nitrososarcosine and nitrite added originally to a concentration of 200 mg kg-l to powdered corn flakes. r 2min H -Time Fig. 2. Responses of che- miluminescence analyser to 19.6 pg of sodium nitrite and 500 pg of morpholine. I -Time Fig. 3. Responses of chemiluminescence analy- ser to l o g of powdered corn flakes freeze-dried with 2.0mg of sodium nitrite and spiked with 2.7 p g of N-nitrososarco- sine.Discussion The choice of powdered corn flakes as a matrix in which to determine a non-volatile N-nitrosamine was prompted by the relative freedom from water of this commodity and its composition, which includes several of the major types of food components, and in no way implies the possible occurrence of such compounds in practice. The method of Eisenbrand and Preussmann2 is entirely suitable for the determination of N-nitrososarcosine in simple solution in many non-polar solvents such as lJ2-dichloroethane. In view of the reactivity of nitrosyl bromide liberated during the denitrosation process,1132 WALTERS et al. : DETERMINATION OF A Analyst, Vol.103 however, it is not surprising that its recovery was far from complete in the presence of powdered corn flakes. Further, the sensitivity of this procedure is diminished by the comparatively large volumes of acetic acid containing hydrogen bromide required to ensure contact between the denitrosating agent and the food matrix. In many instances, the concentrations of inorganic nitrite, alkyl nitrites, etc., would be greatly in excess of that of any N-nitroso compounds present. In such circumstances, it would be extremely difficult using this procedure to provide an exactly similar aliquot of a food or other matrix to take account of the relatively high background levels of inorganic nitrite itself and other com- pounds, such as the S-nitrosothiols, which are likely to break down to nitrite.On the other hand, it is apparent that the method devised for the determination of N-nitrososarcosine in solution by using a chemiluminescence analyser is directly applicable to this compound when supported on the food matrix chosen without prior extraction, in which circumstances it can also be differentiated from inorganic nitrite. This improvement over the method in which nitrosyl bromide is determined as nitrite in solution probably arises from the rapid dispersal of this denitrosation product as nitrogen oxide from the food matrix in a stream of nitrogen with a flow-rate of about 400 ml min-l. The dehalogenation of the nitrosyl bromide formed is indicated by the survival of the volatile denitrosation products following passage through the alkali traps currently interposed between the reaction vessel and the chemiluminescence analyser .Not surprisingly, the coefficients of variation obtained under these circumstances were greater than those resulting from the determination of N-nitrososarcosine in simple solution. Nevertheless, they remained reasonable for the direct determination of this nitrosamine a t the microgram level on a matrix without prior extraction and clean-up, and the possibility remains of using larger samples in effecting even better replication at lower levels of spiking. Further, this procedure has permitted the differentiation from inorganic nitrite of N-nitroso- diphenylamine and therefore appears to be applicable to N-nitroso derivatives of weakly basic secondary amines in its presence, unlike the method involving denitrosation in solution.Ascorbic acid is considered to be a nitrite scavenger and thereby to inhibit the nitrosation of secondary and tertiary amines in simple solution,’ in foods8 and in vivog when animals are fed concurrently nitrite and a nitrosatable amine. It has not, however, been found to decompose pre-formed N-nitrosamines but in a limited number of circumstances its use has stimulated nitrosation.l0*l1 In cured meat systems, for instance, the use of ascorbate has resulted in the incorporation of more nitrogen-15 from isotopically labelled nitrite into water-soluble components.12 In these studies, it has proved useful in reducing the extended duration of the response of the chemiluminescence analyser due to residual nitrite and probably other compounds derived from it within the food matrix. However, ascorbic acid is not very soluble in 1,2-dichloroethane and therefore it has not been particularly useful in restricting the small extent of irrelevant nitrosation observed during the differentiation of nitrite and an N-nitroso compound in the presence of a secondary amine.Ascorbyl palmitate, which has also been found to inhibit strongly the rate of nitrosation of sarcosine in solution,l3 is more soluble in the solvent system employed and has proved to be effective in suppressing irrelevant nitrosation in this instance. The financial support of the US National Cancer Institute, Contract No. N01-CP-43337, is gratefully acknowledged. 1. 2. 3. 4. 5. 6. 7. 8. 9. References Downes, M. J., Edwards, M. W., Elsey, T. S., and Walters, C. L., Analyst, 1976, 101, 742. Eisenbrand, G., and Preussmann, R., Arzneimittel-Forsch., 1970, 20, 1513, Johnson, E. M., and Walters, C. L., Analyt. Lett., 1971, 4, 383. Iwaoka, W., and Tannenbaum, S. R., in “Environmental N-Nitroso Compounds, Analysis and Formation,” IABC ScientiBc Publications, No. 14, International Agency for Research on Cancer, Lyon, 1976, p. 51. Eisenbrand, G., Janzowski, C., and Preussmann, R., J . Chromat., 1976, 115, 602. Shank, R. C., and Newberne, P. M., Fd Cosmet. Toxic., 1976, 14, 1. Mirvish, S. S., Wallcave, L., Eagen, M., and Shubik, P., Science, N.Y., 1972, 177, 65. Fiddler, W., Pensabene, T. W., Piotrowski, E. G.. Doerr, R. C., and Wasserman, A. E., J . Fd Sci., Kamm, J. J., Dashman, T., Conney, A. G., and Burns, J. J., Proc. Nutn. Acad. Sci. U.S.A., 1973, 1973, 38, 1084. 70, 747.November, 1978 NON-VOLATILE N-NITROSAMINE ON A FOOD MATRIX 1133 Sen, N. P., and Donaldson, B., “N-Nitroso Compounds in the Environment,” IARC Scientific Walters, C . L., Edwards, M. W., Elsey, T. S., and Martin, M., 2. Lebensmittelunters. u. -Forsch., Woolford, G., and Cassens, R. G., J . Fd Sci., 1977, 42, 586. Walters, C. L., and Manning, K., 2. Lebensmittelunters. u. -Forsch., 1977, 165, 21. 10. 11. 12. 13. PubZications, No. 9, International Agency for Research on Cancer, Lyon, 1974, p. 103. 1976, 162, 377. Received April 13th, 1978 Accepted June 5th, 1978

ISSN:0003-2654    

DOI:10.1039/AN9780301127

出版商:RSC    

年代:1978

数据来源: RSC