摘要:

THE ANALYSTTHE ANALYTICAL JOURNAL OF THE CHEMICAL SOCIETYE 0 I T 0 R IAL ADVlSO RY BOARD'Chairman: H. J. Cluley (Wembley)'L. S. Bark (Salford)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 (Billingham)D. Oyrssen (Sweden)'W. T. Elwell (Birmingham)J. Hoste (Belgium)'J. A. Hunter (Edinburgh)H. M. N. H. Irving (Leeds)M. T. Kelley (U.S.A.)W. Kemula (Poland)'G. F. Kirkbright (London)G. W. C. Milner (Harwell)G. H. Morrison (U.S.A.)H. W. Nurnberg (W. Germany)'J. M. Ottaway (Glasgow)'G. E. Penketh (Wilton)'T. B. Pierce (Harwell)E. Pungor (Hungary)D. I. Rees (London)'R. Sawyer (London)P. H. Scholes (Sheffield)'W.H. C. Shaw (Greenford)S. Siggia (U.S.A.)A. A. Smales, O.B.E. (Harwell)A. Walsh (Australia)T. S. West (Aberdeen)A. L. Wilson (Medmenham)P. Zuman (U.S.A.)'A. Towns h end (Birmingham)'Members of the Board serving on The Analyst Publications CommitteeREGIONAL ADVISORY EDITORSDr. J. Aggett. Department of Chemistry, University of Auckland, Private Bag, Auckland, NEWDr. G. Ghersini, Laboratori CISE, Casella Postale 3986, 20100 Milano, ITALY.Professor L. Gierst, Universit6 Libre de Bruxelles, Facult6 des Sciences, Avenue F.-D. Roosevelt 50,Professor R. Herrmann, Abteiiung fur Med. Physik., 63 Giessen, Schlangenzahl 29, W GERMANY,Professor W. A. E. McBryde. Dean of Faculty of Science, University of Waterloo, Waterloo, Ontario,Dr. W.Wayne Meinke, KMS Fusion Inc., 3941 Research Park Drive, P.O. Box 1567, Ann Arbor,Dr. I. Rubeika, Geological Survey of Czechoslovakia, Kostelni 26, Praha 7, CZECHOSLOVAKIA.Dr. J. R&i&ka. Chemistry Department A, Technical University of Denmark, 2800 Lyngby, DENMARKProfessor K. Saito, Department of Chemistry, Tohoku University, Sendai, JAPAN.Dr. A. Strasheim, National Physical Research Laboratory, P.O. Box 395, Pretoria, SOUTH AFRICA.ZEALAND.Bruxelles, BELGIUM.CANADA.Mich. 48106, U.S.A.Published by The Chemical SocietyEditorial: The Director of Publications, The Chemical Society, Burlington House,London, WlV OBN. Telephone 01 -734 9864. Telex No. 268001Advertisements: Advertisement Department, The Chemical Society, Burlington House, Piccadilly,London, WIV OBN. Telephone 01-734 9864Subscriptions (non-members): The Chemical Society, Distribution Centre, Blackhorse Road,Letchworth, Herts., SG6 1 HNVolume 103 No 1225@ Thz Chemical Society 1978April 197

ISSN:0003-2654    

DOI:10.1039/AN97803FX013

出版商:RSC    

年代:1978

数据来源: RSC

摘要:

ANALAO 103 (1225) 305-416 (1978)ISSN 0003-2654April 1978THE ANALYSTTHE ANALYTICAL JOURNAL OF THE CHEMICAL SOCIETY30531 733234134635435936837538239139740040340640941 141 3CONTENTSAutomatic Titration by Stepwise Addition of Equal Volumes of Titrant. Part 111. Use o f LinearEquations t o Calculate Equivalence Volumes i n Acid - Base Titrations-Axel Johansson and StenJohanssonAutomated Digestion and Extraction Apparatus f o r Use i n the Determination o f Trace Metals i nPolarographic Studies on Some Organic Compounds o f Arsenic. Part IV. Diphenylarsinic Acid-A. WatsonDetermination of Lead i n Atmospheric Particulates Using an Automated Atomic-absorption Spectro-photometric System w i t h Electrothermal Atomisation-C. J.Pickford and G. RossiAnalysis o f Steroids. Part XXXI. Mechanism o f the Tetrazolium Reaction o f Corticosteroids-Sdndor Gorog and Peter HorvdthSpectrophotometric Determination of Hydrochlorothiazide and Reserpine i n Combination-H. Abdine, M. Abdel-Hady Elsayed and Yousry M. ElsayedFluorescence Properties o f Metoclopramide and Its Determination i n Pharmaceutical DosageForms-W. Baeyens and P. De MoerlooseDetermination of Substituted Ureas and Some Related Herbicide Residues i n Soils by Gas Chroma-tography-David J. Caverly and Ronald C. DenneyREPORTS BY THE ANALYTICAL METHODS COMMITTEEFoodstuffs-C. J. Jackson, D. G. Porter, A. L. Dennis and P. 6. StockwellApplication of Gas - Liquid Chromatography t o the Analysis o f Essential Oils.Part VI.Microbiological Determination o f Zinc Bacitracin i n Animal FeedingstuffsStandardised General Method f o r the Determination o f Iron with 1.10-PhenanthrolineSHORT PAPERSGas-chromatographic Determination o f Some Sulphur Gases at the Volumes per Million Level i nDevelopment o f Acidity in Non-ionic Surfactants: Formic and Acetic Acid-M. Donbrow, R. Ham-Comparison o f Diffesent Forms o f Cadmium as Reducing Agents f o r the Batch Determination o fDetermination o f Nitrogen w i t h Copper as Catalyst f o r High-temperature Digestion-Per TingvallImprovement of the Nebuliser Pattern o f a Flame Photometer-R. van EckCOMMUNICATIONEffect of Cyanoethylation on End-point Sharpness i n Catalytic Thermometric Titrations w i t hBook ReviewsSummaries o f Papers in this Issue-Pages iv, vi.X. xi, xiiDetermina-t i o n of Limonene and 1.8-Cineole in Oils of Peppermint (Varieties Mentha)Air Using Tenax-GC-D. S. Walkerburger, E. Azaz and A. PillersdorfNitrate-W. Davison and C. WoofAcrylonitriie as the Indicator Reagent-E. J. Greenhow, A. Nadjafi and L. Dajer de TorrijosPrinted by Heffers Printers Ltd Cambridge EnglandEntered as Second Class at New York USA, Post Offic“ A N A L O I D ”COMPRESSED CHEMICALREAGENTSoffer a saving in the use of lab-oratory chemicals. A range of over50 chemicals includes Oxidizingand Reducing Agents, Reagents forColorimetric Analysis and Indicatorsfor Complexometric Titrations.For full particulars send for ListNO. 458 to:-RIDSDALE & CO.LTD.Newham Hall, Newby,M i dd les broug h,Cleveland TS8 9EAortelephone Middlesbrough 31 721 6A simple message. . . . . . . .SODIUM BOROHYDRIDEPELLETS(specially produced for atomic absorp-tron work) are a lot cheaper fromCAMBRIAN CHEMICALS.1 oog f 9.005 x 1OOg f40.00CAMBRIAN CHEMICALSBeddington Farm Road,Croydon CRO 4XBTelephone 01 -686 0544Telex 262781 (KKG RF)NOTICE 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):The Analyst, and Analytical Abstracts printed on one side of the paper(c) The Analyst, and Analytical Abstracts . . .. .. . . .. , . €87.00(d) . . €93.00(Subscriptions are NOT accepted for The Analyst and/or for Proceedings alone)Analytical Abstracts only (two volumes per year, including indexes) :(e) Analytical Abstracts . . . . .. .. . . .. .. . . €67.00(f) Analytical Abstracts printed on one side of the paper . , * . .. . . €73.0

ISSN:0003-2654    

DOI:10.1039/AN97803BX015

出版商:RSC    

年代:1978

数据来源: RSC

摘要:

Aeril, 1978 THE ANALYST iiiRS solvents forUV and IRAcetone UVand IRAcetonitrile UV and IRBenzene UV and IRCarbonium sulfideUV and IRCarbonium tetrachlorideUV and IRChloroform UV and IRCyclohexane UV and IRN-N-DimethylformamideUV and IRDichloroethane IRDi met h y Isu If oxide UVDioxane UVEthyl acetate IREthyl alcohol UV95" and abs.Ethyl ether UVn-Heptane UVn-Hexane UVlsoctane UV and IRlsopropyl alcohol UVMethylene chlorideUV and IRMethyl Alcohol UVn-Pentane UVPotassium bromide IRTetrachloroethylene IRTetrahydrofuranUV and IRToluene IRTrichloroetilene IRCHEMICALS DIVISIONP 0 Box 3996120159 MilanolVia lmbonatl 24 (IIaly)Telex Erba MI 36314tTel 6995a MONTED!SON S p A PEG TRADEMARiv SUhl>IARIES O F PAPERS IS THIS ISSYE April, 1978Summaries of Papers in this IssueAutomatic Titration by Stepwise Addition of Equal Volumes of TitrantPart 111.Use of Linear Eqnations to Calculate Equivalccce Volurricsin Acid - Base TitrationsPotentiometriz titrations in which the titration curve is plotted or the titrationis interrupted a t a pre-set end-point have important disadvantages. Thefollowing conditions must be met to obtain satisfactory results: (1) reactionbetween titrant and titrand must be rapid; (2) the response time of theindicator electrode must be short; and (3) electrode potentials near theequivalence point must bz stable and reproducible.These disadvantages can be avoided by using the method of stepwiseaddition of titrant. In this method, equivalence volcmes cannot be readdirectly and must be calculated.A computational procedure allon.ing thecalculation of equivalence volumes for the titration of monoprotic acids,bases and ampholytcs with log K values up t o 1 1 is described. Thc proccdureis based on the solution of a set of linear cqcations and requires only a fewsets of measurements of e.m.f. z'evsus volume of titrant plus initial vclL-meand titrant concentration. Log K values are not required art1 e.m.f.measurements may has-e a constant systematic error L-P t o $ 6 m\-, corre-sponding to 1 0 . 1 pH unit. Only concentrations and concentration constactsare used. No estimated values of equivalence XTolumes or stability constantsare necessary.Keywovds : Acid - base titvinzetvy ; equivalence volunzes ; co2~1$utev cnlczalationsAXEL JOHANSSON and STEN JOHANSSONDepartment of Analytical Chemistry, Thc Royal Institute of Technology, S-10014Stockholm 70, Swedexl.AYtnZJJst, 1978, 103, 305--316.Automated Digestion and Extraction Apparatus for Use in theDetermination of Trace Metals in Foodstuffs.An automated system has been constructed that uses a displacement techniquet o transfer foodstuffs, previously converted into a fluid state, into a modifiedTechnicon cligestor where a wet oxidation is carried out at 400 "C with the aidof sulphuric acid, nitric acid and hydrogen peroxide.The digest is pumpedinto a neutralisation unit where its pH is adjusted to a pre-set value and thenflows into an extraction unit where the metals of interest are chelated andseparated and collected for subsequent analysis b y plasma-emission spectro-metry.The operation of the system is fully automatic and it processes 5-8-gsamples a t the rate of three per hour. Results obtained with the automaticsystem for the determination of iron, copper and zinc in a wide range of food-stuffs compare favourably with those obtained by manual techniques. Somepreliminary results for the determination of cadmium, cobalt, nickel and leadare presented.Keywords: Tvace metal determination ; foodstufls analysis ; automated digestionand extractionC. J. JACKSON, D. G. PORTER, A. L. DENNIS and P. B. STOCKWELLDepartment of Industry, Laboratory of the Government Chemist, Cornwall House,Stamford Street, London, SEl 9SQ.A ~ u Z J J S ~ , 1978, 103, 317-331vi SUMMARIES OF PAPERS I K THIS ISSUEPolarographic Studies on Some Organic Compounds of ArsenicPart IV.Diphenylarsinic AcidX study has been made of the polarographic behaviour of diphenylarsinicacid. I t gives rise to a single cathodic wave below pH 6, displaying anadsorption pre-wave, and some limited inhibition effects. The latter areremoved by addition of a surface-active agent, thus yielding a well formed,diffusion-controlled wave, the height of which is proportional to concentration(up to 1 x RI) and independent of pH. The current - potential relation-ships and the reduction mechanism are discussed. An irreversible reductionto tetraphenyldiarsine has been found. The use of polarography is proposedfor the quantitative determination of diphenylarsinic acid.In mixtures withphenylarsonic acid andjor triphenylarsine oxide, the wave a t pH 1 gives thetotal concentration of all three, while that a t pH 5.3 gives the concentrationof diphenyl and triphenyl species and that a t pH > 7 triphenylarsine oxidealone.Keywovds : Dipheutylavsinic acid; polavogvapliyApril, 1978A. WATSONDepartment of Chemistry, The New Univerjity of Ulster, Coleraine, Co. Londonderry,BT52 ISA, Northern Ireland.A n a l y s t , 1978, 103, 332-340.Determination of Lead in Atmospheric Particulates Using anAutomated Atomic-absorption Spectrophotometric Systemwith Electrothermal AtomisationA laboratory constructed automatic sampling system is used for the deter-mination by atomic-absorption spectrophotometry with electrothermalatomisation of up to 50 air particulate samples, after acid dissolution.Whenused on a routine basis, this method results in a high precision and permitsa large number of samples to be analysed in a minimum time without operatorattendance.Keywords : Lead detevmination ; atmosphevic particulates ; atomic-absorptionspectrophotoiiaetvy with electvotheviulal atontisation ; automated samplingC. J. PICKFORD and G. ROSS1Commission of the European Communities, Euratom C.C.R., Ispra Establishment,Chemistry Division, 21020 Ispra (Varese), Italy.Analyst, 1978, 103, 341-345.Analysis of SteroidsCorticosteroidsThe stoicheiometry and mechanism of the reaction between tetrazoliumreagents and the cr-ketol side-chain of corticosteroids was studied.On thebasis of literature data plus some experimental results it is concluded that theside-chain is oxidised to the 20-keto-2 1-aldehyde, which then partly under-goes an intramolecular Cannizzaro reaction, leading to the 20-hydroxy-21-carboxylic acid. The doubled sensitivity of 16-hydroxycorticosteroidstowards blue tetrazolium can be explained by a mechanism involving D-homorearrangement of the D ring.Equilibrium studies showed that under the usual experimental conditionsthe reduced blue tetrazolium reagent exists almost quantitatively in themonoformazan state.Keywovds : Covticostevoid detevmination ; tetvazoliztna veaction mechanism ;Part XXXI. Mechanism of the Tetrazolium Reaction oftviamcinolone oxidation; blue tetvazoliulvz veductionSANDOR GOROG and PfiTER HORVATHChemical Works Gedeon Richter Ltd., GyomrG u t 21, H-1475 Budapest, Hungary.Analj~st, 1978, 103, 346-353viii THE AXALYST April, 1978 I Qfi ANN ARBOR SCIENCETRACE METALS IN THE ENVIRONMENTSeries Editors: I.C.Smith and Bonnie L. Carson,Midwest Research Institute, Kansas City, Missouri.These comprehensive in-depth works give a wide scope of information and data on processesby which trace metals reach the environment, their impact on plant and animal life, and onhumans. Each volume of the series analyzes sources of pollution, routes to the food chain,potential health hazards, and also provides useful information on industrial uses.There are references for each chapter which will prove valuable to pollution control andindustrial engineers and scientists, biologists, geologists, chemists, toxicologists, and govern-ment officials.Vol.1: THALLIUMby I.C. Smith and Bonnie L. Carson.Pointing out that chronic intakes of thallium by humans from 10 to 25mg. per day are possible a t presentcontamination levels, this detailed report emphasizes the need for immediate study into this pollutionproblem. It examines current uses for thallium compounds, industrial pollution sources, and physiologicaleffects and health hazards for life forms.0250402149 approx. 395 pages due Apri1'78 approx. 826.40/€16.10Vol. 2: SILVERby I.C. Smith and Bonnie L. Carson.This is a comprehensive report examining silver contamination of our environment with in-depthdiscussions on distribution levels, sources of losses to the environment and physiological effects.It pointsout the danger of silver concentration in sewage sludge and water basin sediments t o microorganisms, fish,shellfish, birds and mammals.0250402157 approx. 469 pages due Apri1'78 approx. %26.40/€16.10Vol. 3: ZlRCOiVlUMby I.C. Smith ant Bonnie L. Carson.This is a comprehsnsive report on zirconium, i t s alloys and compounds - with detailed coverage ofenvironmental occurrence, uses, processing, disposal and physiological effects. I t examines sources of lossto the environmertt and their magnitude, while giving an especially complete review of zirconium'sbiological effects on plants, animals and humans.0250 4p216 5 approx. 405 pages due Apri1'78 approx. $26.40/€16.10Vol. 4: PALLADIUM/OSMIUMby I.C. Smith, Bonnie L. Carson, and T.L. Ferguson.This describes uses of osmium by the chemical, electrical and medical industries, and examines the toxiceffects of osmium tetroxide. The section on palladium gives detailed coverage to potential problems thatmay develop from the use of palladium in catalytic converters. Results of studies are presented showingthat palladium compounds cause cancer in rats and mice, and are toxic to mammals, microflora, fish andsome plants.0250402173 approx. 188 pages due April'78 approx. $26.40/f16.10Vol. 5: INDIUMby I.C. Smith and Bonnie L. Carson.As toxic to mammals as thallium, indium i s especially dangerous because the effects of long-term, low-levelexposure are unknown. Since toxicity studies show that indium is absorbed into the human system, thereis an immediate need for further studies on its potential effects.025040232 7 due December 1978 amrox. %26.40/f15.1

ISSN:0003-2654    

DOI:10.1039/AN97803FP025

出版商:RSC    

年代:1978

数据来源: RSC

摘要:

April, 1978 THE ANALYST ix1 PARTICLE;SIZE ANALYSISedited by J. D . Stockham, Manager, Fine Particles Research, and E . G . Fochtman, Manager,Chemical Engineering Research, IIT Research Institute, Chicago.This book provides both a theoretical and practical guide to techniques for particle size analysis. A l l thepapers were prepared by the IlTRl Fine Particles Research Section and presented at IITRl’s FiltrationDay 1975.0250 401 89 4 152 pages November 1977 €1 4.75js26.40iMODERN METHODS OF TRACE ELEMENTS ANALYSISby M . Pinta, Office de la Recherche, Scientifique et Technique, Outre-Mer, Bondy, France.This reference book examines the principle procedures and techniques in use today in industrial,analytical and research laboratories for the determination of trace elements in various media.0250 401 52 5 approx.21 6 pages In Press approx. f 19.75/$32.45ANN ARBORSCIENCX SUMMARIES OF PAPERS I N THIS ISSUESpectrophotometric Determination of Hydrochlorothiazide andReserpine in CombinationAn accurate and simple method is proposed for the determination of hydro-chlorothiazide and reserpine. The former is determined by the applicationof the differential spectrophotometric method and the latter by formation ofa charge-transfer complex with iodine. The procedure has been appliedsuccessfully to the determination of hydrochlorothiazide and reserpine whenpresent in combination with a concentration of hydrochlorothiazide 100 xthat of reserpine.Keywords : Hydrochlorothiazide determination ; reserpine determination ;A p r i l , 1978spectropkotometryH.ABDINE, M. ABDEL-HADY ELSAYED and YOUSRY M. ELSAYEDDepartment of Pharmaceutical Analytical Chemistry, Faculty of Pharmacy,University of Alexandria, Alexandria, Egypt.Analyst, 1978, 103, 354-358.Fluorescence Properties of Metoclopramide and Its Determination inPharmaceutical. Dosage FormsThe native fluorescence characteristics of rnetoclopramide in various solventsand a t different pH values were determined qualitatively and quantitatively.Fluorimetric determinations can be performed directly and are based on anintense emission a t 360 nm t h a t occurs when the sample is excited at 310 nmfollowing dilution in p H 2.0 buffer solution to a final concentration up to 6p g ml-1, The detection limit is 3 x pg rnl-l.A comparison is made between the fluorescence behaviour of metoclo-pramide and the structurally analogous procainamide molecule.Finally,metoclopramide has been converted into its monoacetyl derivative a t theprimary aromatic amine function. Spectral data confirming the structureare reported. Apart from the second-order excitation shoulder at 276 nm,no significant difference concerning fluorescence behaviour, qualitative orquantitative, between the acetyl derivative and the original molecule isestablished.Keywords : Fluorimetry ; metoclopramide fluorescence properties ; meto-clopramide determination ; pharmaceutical analysisW. BAEYENS and P. De MOERLOOSEState University of Ghent, Faculty of Pharmaceutical Sciences, Department ofPharmaceutical Chemistry and Drug Quality Control, Akademisch Ziekenhuis,De Pintelaan 135, B-9000 Ghent, Belgium.Analyst, 1978, 103, 359-367.Determination of Substituted Ureas and Some Related HerbicideResidues in Soils by Gas ChromatographyA method for the determination of 11 substituted ureas and related herbi-cides in soils is described.Acetone extraction is followed by alkalinehydrolysis, steam distillation and concentration of anilines in toluene, thelast three steps being carried out in a single operation using a liquid - liquidextractor. The anilines, after partition into hydrobromic acid, are bromi-nated and determined by gas chromatography with an electron-capturedetector. The procedure is sufficiently sensitive for investigations intoproblems of crop damage and can be applied t o a wide variety of soil types.An additional step t o remove interference due t o the presence of anilinemetabolites is also described.The limit of detection is 0.01 mg kg-1 andrecoveries at residual levels are generally better than 80%.Keywords : Substituted urea determination ; herbicide residue determination ;gas chromatographyDAVID J. CAVERLYMinistry of Agriculture, Fisheries and Food, Agricultural Development and AdvisoryService, Olantigh Road, Wye, Ashford, Kent, TN25 5EL.and RONALD C. DENNEYSchool of Chemistry, Thames Polytechnic, lVoolwich, London, SE18 6PF.Analyst, 1978, 103, 368-374A p r i l , 1978 SUMMARIES O F PAPERS I N THIS ISSUE xiApplication of Gas - Liquid Chromatography to the Analysis ofEssential OilsPart VI.Determination of Limonene and 1,s-Cineole in Oils ofPeppermint (Varieties Mentha)Report prepared b y the Essential Oils Sub-committee.Keywords : Limonene detevnzination ; 1,8-cineole determination ; oils of pepper-m i n t ; gas - liquid chromatographyANALYTICAL METHODS COMMITTEEThe Chemical Society, Burlington House, London, W1V OBN.Analyst, 1978, 103, 375-381.Microbiological Determination of Zinc Bacitracin inAnimal FeedingstuffsReport prepared by the Antibiotics in Animal Feedingstuffs Sub-committee.Keywords : Zinc bacitracin determination ; microbiological assay; animalfeedingstuffs; antibioticsANALYTICAL METHODS COMMITTEEThe Chemical Society, Burlington House, London, W1V OBN.Analyst, 1978, 103, 382-390.Standardised General Method for the Determination of Ironwith 1,lO-PhenanthrolineReport prepared by the Iron Sub-committee.Keywovds : Ivon detevinination ; 1,lO-phenanthvoline ; spectvophotometryANALYTICAL METHODS COMMITTEEThe Chemical Society, Burlington House, London, WIV OBN.Analyst, 1978, 103, 391-396.Gas-chromatographic Determination of Some Sulphur Gases a t theVolumes per Million Level in Air Using Tenax-GCShort PaperKeywords : Sulphuv gas determination ; gas chromatography; Tenax-GC; airanalysisD.S . WALKERDepartment of Industry, Warren Spring Laboratory, P.O. Box 20, Gunnels WoodRoad, Stevenage, Hertfordshire, SG1 2BX.Analyst, 1978, 103, 397-400.Development of Acidity in Non-ionic Surfactants : Formic andAcetic AcidsShort PaperKeywords : Non-ionic surfactant decomposition; pharmaceutical adjuvants ;degradation ; storageM.DONBROW, R. HAMBURGER, E. AZAZ and A. PILLERSDORFPharmacy Department, School of Pharmacy, Hebrew University of Jerusalem,P.O. Box 12065, Jerusalem, Israel.Analyst, 1978, 103, 400402xii SUMMARIES OF PAPERS Ih’ THIS ISSUEComparison of Different Forms of Cadmium as Reducing Agentsfor the Batch Determination of NitrateApril, 1978Shovt PapevKeywords : Nitvate batch detevmination ; cadmium yeduction ; spectvophoto-vnetvy ; water analysisW. DAVISON and C. WOOFFreshwater Biological Association, Windermere Laboratory, Ambleside, Cumbria,LA22 OLP.Analyst, 1978, 103, 403-406.Determination of Nitrogen with Copper as Catalyst forHigh-temperature DigestionShort PaperKeywovds ; Nitvogen determination ; clinical analysis ; automatic analysis ;digestion; spectvophotometvyPER TINGVALLVitrum Institute for Human Sutrition, Box 121i0, 10224 Stockholm 12, Sweden.Analyst, 1978, 103, 406-409.Improvement of the Nebuliser Pattern of a Flame PhotometerShort PapevKeywovds ; Flame photometev; nebzilisev pattern ; glycevol; plant digestsR.VAN ECKDepartment of Soils and Fertilisers, Agricultural University, De Dreijen 3,Wageningen, The Netherlands.Aqzalyst, 1978, 103, 409-410.Effect of Cyanoethylation on End-point Sharpness in CatalyticThermometric Titrations with Acrylonitrile as the Indicator ReagentComnzunicationKeywords : Cyanoethylation ; anionic polymerisation ; catalytic thermometrictitrimetryE. J.GREENHOW, A. NADJAFI and L. DAJER DE TORRIJOSDepartment of Chemistry, Chelsea College, University of London, Manresa Road,London, SW3 6LX.Analyst, 1978, 103, 411-412April, 1978 THE AN4L'iSTI should from now on be addressed t o our ownAdvertisement Department,The Chemical Society, Burlington House,Piccadilly, London W1V OBNTel: 01-734 9864 ' Please send all space orders, copy, enquiries~ etc. to this address.xiiiCLASSIFIED ADVERTISEMENTSThe Rate /or Classified A clz,eTtisements i s L2.30 per sin&!coluiitn centimetre lminiiniim f4.60).Box IVuniheTs are charEed an extra 500.Deadline foT classifid copy is 20th o j the month precrdingnionth of issue.All space orders, copy instructions aiid enquiTies shozrld beaddressed to The Advertisement Department.The Chentirai Society, Burlington House, Pzccadiilv,London WrV o B STelepkorid 01.774 9864 Telex 268001THE UNIVERSITY OF THE SOUTHPACIFIC-FI JIXpplicaLions are invited for the position ofXSALPTICIL CHEMIST (Post 78/9) with theInstitute of Xatural Resources.Candidates shouldhave a PhD or equix-alent professional qualificationsand have experience in a range of analytical methodsincluding G.L.C., 1 . 4 . S . and spectrophotometrictechniques. Preference will be given t o applicantswho have familiarity with agricultural analyses suchas soils, water, pesticides and crop nutrients.Experience in areas such as air and water quality,food and industrial analyses, is also relevant to theposition.The ability to adapt methods and impro-vise together with a sound practical knowledge ofinstrumentation xi11 be an advantage. Preparationof reports and liaison with Regional and Govern-mental organizations will be required. The successfulapplicant will be expected to unleitalte analyses as aconsultancy service for outside bodies and to initiateand participate in projects carried out as part ofL'nirersity activities. The work will be generallynon-routine. Opportunity may be available to partici-pate in Institute sponsored workshops and instructionalprogrammes and to a limited extent in areas of theundergraduate teaching, in particular in specialisedfinal-!rear courses. Salary in accordance withexperience and qualifications in one of the followingscales: (a) F$7,640-10,808 p , a .; (b) F$11,178-13,209 p.a. ( L l sterling = F S 1 . 6 i ) . In addition theUniversity provides 15 oi gratuity, superannuationcontribution, appointment allowance, partly furn-ished accommodation a t a maximum rental at presentof 150; of salary. Appointments will be for a con-tract period of three years a.nd will be renewable bymutual agreement. Candidates should send THREECOPIES of their curriculum vitae, quoting the rele-vant post reference indicated above with full personalparticulars and names and addresses of three refereesto the Registrar, the University of the South Pacific,PO Box 1168, Suva, Fiji to reach him no later than22 April 1978.Applicants resident in UK should alsosend 1 copy to Mrs. C. Xacroman, Inter-UniversityCouncil, 90/91 Tottenham Court Road, LondonTVlP ODT. Further particulars can be obtained fromeither address.n United Arab Emirates1 ' DubaiPublic AnalystOur client, Dubai Municipality, intends toappoint a Public Analyst who, in addition tocarrying out normal chemical and micro-biological analysis of food and drugs, will becharged with the responsibility of improvingand expanding existing facilities.Applicants should hold the approved quali-fication of Mastership in Chemical Analysis(M.Chem.A.) and have 10 years' experiencein the analysis of food, drugs, and relatedsubstances. Experience should include notonly analytical chemistry but also microscopyand microbiology.Salary will be not less than f10,000 p.a. tarfree. The initial contract is for two years withprobability of renewal and terms includeannual gratuity, free furnished, air con-ditioned accommodation, car, leave passagest o UK, and four weeks paid leave per annum.Dubai has a large expatriate British com-munity and is an exciting and pleasant place inwhich to live. Europeans enjoy a high standardof living and there are no restrictions onpersonal freedom. All salaries and benefitsare tax free and money is freely transferableto the UK.Applicants shou!d telephone or write to theManagement Selection Division as soon aspossible. Only applicants with the relevantprofessional qualifications and appropriateexperience will be considered.Scotconsult LtdManagement and Resource DevelopmentConsultanis17 Regent Terrace,Edinburgh z ScotlandTelephone 031 556 002

ISSN:0003-2654    

DOI:10.1039/AN97803BP031

出版商:RSC    

年代:1978

数据来源: RSC

摘要:

APRIL 1978 Vol. 103 No. 1225 Automatic Titration by Stepwise Volumes of Titrant Addition of Part 111." Volumes in Acid - Base Titrations Axel Johansson and Sten Johansson Use of Linear Equations to Calculate Equivalence Department of Analytical Chemistry, The Royal Institute of Technology. S-10044 Stockholm 70, Sweden Potentiometric titrations in which the titration curve is plotted or the titration is interrupted at a pre-set end-point have important disadvantages. The following conditions must be met to obtain satisfactory results : (1) reaction between titrant and titrand must be rapid; (2) the response time of the indicator electrode must be short; and (3) electrode potentials near the equivalence point must be stable and reproducible. These disadvantages can be avoided by using the method of stepwise addition of titrant.In this method, equivalence volumes cannot be read directly and must be calculated. A computational procedure allowing the calculation of equivalence volumes for the titration of monoprotic acids, bases and ampholytes with log K values up to 11 is described. The procedure is based on the solution of a set of linear equations and requires only a few sets of measurements of e.m.f. veiysus volume of titrant plus initial volume and titrant concentration. Log K values are not required and e.m.f. measurements may have a constant systematic error up to f 6 mV, corre- sponding to &O. 1 pH unit. Only concentrations and concentration constants are used. No estimated values of equivalence volumes or stability constants are necessary, Keywords Acid - base titrimetry ; equivalence volumes ; computer calculations The first two papers1s2 in this series were published in this journal in 1970.A third paper dealing with the evaluation of titrations in more detail than in Part I was intended, but for various reasons was not published. Since then, several computer programs have been developed, one of which has been applied to the titrations of mixtures of acids,3 and a similar program was recently published by Nowogrocki et aL4 Other papers from this laboratory have described other non-approximate computational methods based upon the same principle~.~s~ These programs have been shown to be applicable to titration by stepwise additions. So far, however, these procedures have not been combined into a general-purpose program that is capable of handling experimental data from all types of acid - base titrations.In this paper, we present a new computational procedure that is applicable to the titration of monoprotic acids or bases of almost any vahe of the stability constant. The procedure can also handle data from the titration of ampholytes. With the titration procedure described by Johansson,' a variant of the program can also be used with some polyprotic acids and to determine the total acid content of mixtures. These computational procedures are intended primarily for automatic titrations, but they give equally satisfactory results with manual data. The program is based upon the solution of a set of linear equations. The approach presented has a number of important advantages : (a) it requires only a few pairs of measurements (volume of titrant v e y s m e.m.f.or veysws pH) plus initial volume and titrant concentration ; (b) no preliminary estimate of equivalence volume is necessary; (c) 305 * For Part I1 of this series, see reference list, p. 316.306 JOHANSSON AND JOHANSSON AUTOMATED TITRATION BY STEPWISE Analyst, VOL. 103 the stability constant of the acid need not be known; (a) exact pH values are not required (this is very important, particularly in routine analysis, as it is inconvenient to re-calibrate a pH meter before each titration); and (e) the program can automatically select appropriate measurements (for example, it is often desirable to use only data obtained before the equivalence point).In some instances, the calculations require iterations in order to obtain acceptable data. In future papers, we shall discuss problems in handling data from titrations of polyprotic acids and of mixtures of several acids. The mathematical procedures used in these instances are similar to those described here. Principles The primary aim of a quantitative titrimetric analysis is the determination of an equivalence volume. With visual end-point determination and with some types of automatic titration, the titration is stopped at the equivalence point and the equivalence volume is measured directly, With other titration methods, titrant is added continuously, with simultaneous measurement of, for example, potential. The equivalence volume is read directly from the resulting graph of titrant volume oeysus potential.With the latter mode of titration the titrant must be added slowly, particularly near the equivalence point, to allow the solution to reach equilibrium. When titrant is added by stepwise addition of equal volumes of titrant, equivalence volumes cannot be read directly, but must be calculated. After each addition of titrant, the e.m.f. is measured and the pairs of titrant volume - e.m.f. data are used to calculate the equivalence volume. On the other hand, it is not necessary to determine the equivalence point, as this is defined in terms of volume and (in this instance) potential. This mode of titration has several advantages: (1) titrations can be carried out more rapidly, as it is necessary to reach equilibrium at only a few points; in particular, it eliminates waiting for slow equilibration near the equivalence point; (2) titrations can be carried out more accurately, as (a) the risk that equilibrium has not been reached is minimised, (b) titration points can be chosen such that the effects of interfering equilibria are minimised and (c) several titration points and not only the equivalence point are used for the evaluations; and (3) random errors can be estimated by statistical analysis of data taken at several points during the titration.Concentration, Activity and E.m.f. E.m.f. values obtained from the electrochemical cells used in potentiometric titrations are a function of activity rather than concentration. Hence equilibrium constants determined by potentiometry are often “mixed” constants, based upon both activities and concentra- tions.However, the use of activities is impractical for the analytical chemist, as these values must eventually be converted into concentrations. It is simpler to calibrate the cell directly with solutions of known concentration. For example, the e.m.f. of a cell used to measure hydrogen-ion concentration can be described by These constants were used for the calculations presented in Part 1.l .. - * (1) E = E’O + Q log [HI + E’j . . . . (for simplicity, ionic charges are omitted throughout this paper). ElO is the standard potential including the reference potential and the components of the activity coefficients and the liquid junction potential which are independent of acidity.Q is the factor in the Nernst equation (Q = RT lnlO/F = 59.158 mV at 25 “C). The term E’, accounts €or the fractions of the activity coefficients and the liquid junction potential which are a function of acidity. Equation (1) has the same form as the expanded Nernst equation except that it uses hydrogen-ion concentration rather than hydrogen-ion activity. However, the primed terms, EfO is constant a t constant ionic strength while E’, varies with acidity according to and E f i , are conditional, i.e., they are a function of solution composition. E’j = ia CHI + joH [OH] . . .. .. ..April, 1978 ADDITION OF EQUAL VOLUMES OF TITRANT. PART I11 307 The constants jH and joH are a function of the composition of the ionic medium and of the condition of the glass electrode.For a good glass electrode, jH is about -25 mV in a 0.5 M potassium chloride solution. Elj varies from 0.25 mV to negligible levels during titration of 0.01 M hydrochloric acid. It has even less effect with weaker acids as the hydrogen-ion concentration is low during the entire titration. In order to calculate hydrogen-ion concentration from measured e.m.f. (E) values, both are required. These can be obtained by carrying out an “E’* titration.”5,8 E’, for a given pair of electrodes is constant, but E’, values show day-to-day variations. Also, transfer of electrodes from one solution to another of the same concentration will alter by about h0.l mV. A procedure for calculating equiv- alence volumes must therefore be insensitive to small variations in E‘*.The program described in the next section is formulated so that variations in E’, of &6 mV, corresponding to hO.1 pH unit, have no effect on the calculated equivalence volume. The electrode system can be calibrated by measuring the e.m.f. of a 0.01 M hydrochloric acid solution containing sufficient neutral salt to give the desired ionic strength. Q values vary with temperature, and temperatures must be both measured (in order to calculate Q) and controlled during the titration. E’j values may appear to have a negligible effect on the titration of solutions with an initial pH value of, say, 2. E’* can have an error of &6 mV, which is much larger than a typical However, if even these low E’j values are neglected there will be an error of a few tenths of a per cent.in the final result. With the Gran method, non-linear curves are obtained if the liquid-junction potential is not considered. and an approximate value of value of 0.25 mV. Calculation of Equivalence Volumes Consider the case where V , ml of a monoprotic acid is titrated with V ml of a strong base, The following relationship applies e.g., sodium hydroxide solution of concentration C,. [equation ( 5 ) in Part Ill: V + K (V[H] + ‘F [HI2 - ~ I“ I‘ Kw) + GLLY ( [HI - 3) - V e = 0 (3) CB C B where Vo is the initial volume of sample, V is the volume of titrant of concentration CB added, K is the stability constant of the acid, Kw is the ionic product of water and V e is the equivalence volume. K and Kw are concentration constants, i.e., the electrode couple is calibrated by using concentrations instead of activity in a certain medium.Their values are unknown. V,,, V and CB are known, and [HI is calculated from E values measured after each addition of base. The calculation can first be simplified by ignoring the term KK,(V, + V)/C,, as it is negligible except at very high K values, i.e., for very weak acids. Equation (3) then becomes linear with respect to Kw, K and Ve. Ve is to be calculated. The function F(K,, K , V,) = 0 can be written as A(1,i)Kw + A(2,i)K + A(3,i)V, + A(4,i) = 0 .. * - (4) where A(1,i) = A(2,i) = A(3,i) = A(4,i) = F(K,, K , V,) is a linear function and the set of equation (4) is first order with three unknown variables. Thus, at least three values of A(l,i), A ( 2 , i ) , A(3,i) and A(4,i) (i.e., at least three pairs of titrant volume versus e.m.f.measurements) are needed. With more than three measurements, the set of equations is over-determined and should be solved using standard mathematical techniques. In this study, a least-squares procedure was first used to calculate coefficients in the normal equations, then these equations were solved using the Gauss elimination method with pivoting.308 JOHANSSON AND JOHANSSON : AUTOMATED TITRATION BY STEPWISE Analyst, VoZ. 103 The resulting values of K,, K and V e include both random and systematic errors derived from errors in Vo, V and [HI. In practice, accurate pipetting will make volume errors negligible, leaving the error in [HI as the limiting error in the calculation. Random errors can be minimised by increasing the number of measurements, but systematic errors due to errors in values cannot be easily eliminated.Values of [HI should therefore be assumed to include systematic errors. If the measured value of ElO differs from its true value by AE’,,, then AE‘o Aph = - Q (the value of -log[H] has been denoted by ph to distinguish it from -log(H) = pH, where [HI is hydrogen-ion concentration and (H) is hydrogen-ion activity). The hydrogen-ion concentration calculated from the erroneous value of ElO is written as [H’], hence [H‘] = lO-(ph + Aph) = [H]lO-*ph = [H]/f. Thus [HI = f[H’] where f = 1OAph. As is unknown, f is therefore unknown (this is not important with sigmoidal titration curves as the only effect is to displace the curve along the ph axis). Equivalence volume calculations should therefore allow for the possibility of errors in ElO values.Equation (3) can be written as5 V + K ( V [ H ’ ] f -I- v*v [H’12f2 - c, vo + Kw) + or where K’ = fK and K’, = Kw/f. f typically has a value of 1.0 but can vary between 0.8 and 1.25 as A F o varies by up to &6 mV. These equations are no longer first order, and in principle cannot be solved using a linear regression. However, this problem can be avoided by assuming a value off = 1.0 to solve equation (6) and then, if necessary, re-calculating f. Methods of calculation vary, depending on the stability constant of the acid. The dimensions of all terms in equation (6) are volume, and their contribution to the equivalence volume varies according to the type of titration.5 For example, for a strong acid, K = 0, and the K‘ term in equation (6) is zero.Calculation Procedure T i t r a t i o ~ of monoprotic acids The function F(K’,, K’, V,) = 0 can be re-written as A(1,i)K’w + A(2,i)K’ + A(3,i)Ve + A(4,i) = 0 . - .. * . (7) where A(1,i) = - (Vo + V)/[H’ICB A(291) = v[H’] A(3,Z) = - 1 A(4,i) = V + ( V O (VO + V ) LH’]”flCB - ( V O $- v)K‘W/cB V)[H‘]f/CB Equation (7) can be solved by givingf an initial value of 1.0 with K’, in coefficient A(2,i) set at 2.5 x 10-14 [the ionic product of water in a barium chloride solution at 25 “C with p = 0.5 (ref. 9)]. This yields approximate values of Klw, K’ and Ve that should not be seriously in error and that can be improved with further calculation.April, 1978 ADDITION OF EQUAL VOLUMES OF TITRANT.PART I11 309 The calculation proceeds along one of four paths, depending on the value of K’: (a) strong acids . . .. .. . . .. .. .. .. log K’ < 1 (b) relatively strong acids . . . . . . .. .. . . 1 < l o g K ‘ < 4 (c) weak acids . . .. . . .. .. .. .. .. 4 <logK’<9 (d) very weak acids . . .. .. .. .. .. . . logK’ > 9 Strong acids. For very strong acids, equation (6) has the form This equation is linear with respect to the three unknowns, Ve, f and K’,. All three quantities can be calculated, together with a correct value of K,, as Kw = fK‘,. This equation has the form where 4 L i ) = - (Vo + V)/C,[H’] A(3,i) = -1 A(4,i) = V = (vo + V ) [H’]/CB At this stage, it is desirable to reject data for which V >Ve. The titrant can contain carbonate and carbon dioxide can be absorbed from the atmosphere into alkaline solution.Further, with many glass electrodes, El0 values change at high pH values. The estimated equivalence volume obtained from equation (7) is used to reject data for which V >Ve. The term A(l,i)Kfw, representing [OH], in equation (8) is eliminated, as [OHIwO. Equation (8) reduces to A(1,i)f + A(2,i)Ve + A(3,i) = 0 .. .. . . (8a) where A(l,i) = ( V O + V ) [H’]/CB A(2,i) = - 1 A(3,i) = V This is basically the same as Gran’s equation for strong acids.1° Equation (8a) yields the concentration of strong acid whereas equation (8) is affected by the presence of weak acids. Relatively strofig acids. The term K‘K‘,(Vo + V)/C, is negligible (for log K = 4, Vo = 100 ml and C, = 0.1, this term is about lO-’rnl).As with strong acids, data for which V >Ve can be rejected, resulting in the equation A(1,i)K‘ + A(2,i)f + A(3,i)Ve + A(4,i) = 0 . .. ’ - (9) where A(l,i) = v[H] + ( Y O + V ) [H’]2flcB A(2,i) = (Vo + V ) [H’]/C, A(3,i) = - 1 A(4,i) = V Calculations are iterated three times with f set at 1 in coefficient A (1,i) in the first calcu- lation (f is assumed to be within the range 0.8-1.25). This procedure may give an imprecise estimate off, but a correct value for equivalence volume. Weak acids. When log K’ is large, the term K’K’,(Vo + V ) / P is no longer negligible. However, the estimated value of K’, from equation (7) is sufficiently precise for stability constants up to log K’ = 9. The appropriate equation is therefore310 JOHANSSON AND JOHANSSON : AUTOMATED TITRATION BY STEPWISE Analyst, Vol. 103 A(1,i)f + A(2,i)K‘ + A(3,i)Ve + A(4,i) = 0 .. .. . . (10) where A(1,i) = ( V O A ( 2 , i ) = v[H’] $- K‘W(V0 + V)/cB v) [H’]/CB + ( V O + V ) [H’]2K‘/cB A(3,i) = - 1 A(4,i) = V - ( V o + v)K’,/C,[H’] For the weakest acids in this category, the coefficient A (1,i) is very small, resulting in an imprecise estimate of f . Very weak acids. In this instance, all terms that include f are negligible, except when V = 0. The estimated value from equation (7) must be further refined, requiring use of measurement data after the equivalence point. On the other hand, an accurate value of Klw is required. The appropriate equation is A(1,i)K’w + A(2,i)K’ + A(3,i)Ve + A(4,i) = 0 - . .. . . (11) where A(1,i) = - ( V o + V)/C,[H’l 4 3 , ; ) = -1 A(4,i) = V A(2,i) = V[H’] - ( V o + V)K’w/CB The value of Ktw obtained from equation (7) is used to compute A ( 2 , i ) .Titration of ampholytes acid completely neutralised at the start of the titration. initially affected by the presence of titrated acid. result in an error of several per cent. in the equivalence volume if it is ignored. Correction for this effect is simple. stability constants k and K (K > k). thus An ampholyte can be regarded as a mixture of two acids of equal concentration, with one Titration of the second acid is Normally this effect is not large, but it may Assume that the ampholyte is a diprotic acid with An additional term must be added to equation ( 3 ) , In equation (7), this term is not considered, and only approximate values of Ktw, K‘ and V e are obtained.As with monoprotic acids, ampholytes are divided into groups with similar values of log K’ [obtained by solving equation (7)]. There are three groups (apart from ampholytes, where both stages are fully dissociated; these can be treated as strong monoprotic acids). Initial approximate values are obtained from equation (9), (10) or ( l l ) , respectively, and then term (12) is added to produce three new equations. For a relatively strong ampholyte (e.g., sodium hydrogen sulphate) : A(1,i)K’ + A(2,i)k’ + A(3,i)Ve + A(4,i) = 0 . . .. . . (13) where A(1,t) = V[H’] + ( V O + V ) rH‘I2flcB A(2,z) = Ve[H’l (1 + KEH’1)/(1 + R’CH’I) 4 3 , ; ) = -1 A(4,i) = v + (Vo + V ) [H’If/CB The computation is iterated four times. In the first loop, previously determined values of V , and K’ are used, with k‘ set at 0 in coefficient A ( 2 4 (k’ is defined as fk).Similarly, previously determined values of f are used in coefficients A(1,i) and A(4,i). The resulting determined values are then used in the next iteration.April, 1978 ADDITION OF EQUAL VOLUMES O F TITRANT. PART I11 311 For a weak ampholyte (e.g., potassium hydrogen phthalate, log k = 2.8, log K = 4.3) the following equations are used: A(1,i)K’ + A(2,i)k’ + A(3,i)Ve + A(4,i) = 0 .. . . (14) where A(1,i) = V H ‘ I + (‘vo + V ) [H’l2flCB A(3,i) = -1 A(4,i) = ‘v - (Vo + V)K‘WK’/CB + (Vo + V ) [H’lflCB A(2,i) = V,[H’] (1 + K’[H’])/(l + k’[H‘]) This computation is also iterated four times following a procedure similar to that for relatively strong ampholytes.Amino acids such as glycine are very weak ampholytes with log k = 2.5 and log K = 9.7. In this instance the equations are . . . . (15) A(1,i)K‘w + A(2,i)K’ + A(3,i)Ve + A(4,i)k‘ + A(5,i) = 0 where A(1,i) = - (Vo + V)/C,[H‘l A(3,i) = -1 A(5,i) = v A(2,i) = V[H’] - (Vo + V)K’w/CB A(4,i) = Ve[H’] (1 + K’[H’])/(l + k ’ [ H ’ ] ) Calculations are carried out using the same procedures as with other types of ampholytes. Titration of dipyotic acids With diprotic acids, there are three cases: (a) the step around the first equivalence point is much greater than that around the second; (b) both steps are of about equal size; and (c) the second step is much greater than the first. Neutralisation of the first proton is readily detected and titration can be stopped shortly after the first equivalence point.Computation assuming a monoprotic acid gives a correct value of Ve, but an incorrect value of Krw. For acids in category (b) and particularly ( c ) , it is desirable to titrate until the second equivalence point has been passed. The first equivalence point is then treated as the starting point, resulting in the same relationship as in the titration of ampholytes. As Ve is unknown, it is first estimated as half the total volume of base added, and the calculations are iterated in order to obtain more precise estimates. This procedure is adequate, but in future papers we shall describe better procedures. Acids in category ( a ) are best treated as monoprotic acids. Hence V - ‘ve is substituted for V in the appropriate coefficients. Titration of a n acid in the pyesence of its conjugate base A mixture of an acid and its conjugate base (e.g., acetic acid and sodium acetate) can be treated as a titration that has been interrupted and then continued without obtaining data from the first part.The titration curves are similar to those obtained with ampholytes, except that K = k and the salt concentration, CA, is unknown. The term VoCAIH’]K‘/C, is added, resulting in the equations A(1,i)K + fft(2,i)CA + A(3,i)Ve + A(4,i) = 0 . . * . . (16) where = ‘v[H’] + ( V O + V)[H‘]”flCB A(2,i) = -Vo[H’]K’ 4 3 , ; ) = -1 4 4 4 = I‘ + (I‘o + V)[H‘If/C, - (Vo + v)K’w/C,[H’l Calculations are iterated four times, starting with values of K’ and Klw estimated from Equation (16) is valid for relatively strong acids and weak acids for which equation (7).log K < lo9 with a negligible term - (V, + V)K’,K‘/C,.312 JOHANSSON AND JOHANSSON : AUTOMATED TITRATION BY STEPWISE Analyst, VOZ. 103 Titration of fiolyprotic acids and mixtures of acids The computational procedure described in this paper is not directly applicable to poly- protic acids or to mixtures of acids. The method described by Johansson' can be used in some instances ; this method implies that a relatively concentrated solution of , for example, sodium acetate is added to the sample. If the acids in the sample are stronger than acetic acid they will release an equivalent amount of this acid, which can then be titrated. From the equilibrium equation ITxAo = [HAc]/ [HI [Ac] .. .. .. (17) and it follows that Kw = [H][OH] . . .. .. . . (18) [HAc][OH] = D . . .. .. . . (19) where the constant D is given by D = K,K,,[Ac] . . .. .. . . (20) Equation (19) has the same form as that used for the titration of a strong acid with a strong base, with the exception that [HI is replaced with [HAc]. The concentration of acetic acid during titration can be monitored with a glass electrode as [HAc]/[H] is constant at high (hence constant) acetate concentrations. To calculate [HAc], a special value of Elo, E',(HAc), is required. A 25-ml volume of 0.02 M hydrochloric acid is added to 25 ml of 1.0 M sodium acetate solution, giving a mixture with an acetic acid concentration of 0.01 M. If the measured e.m.f. of this solution is designated as Eo.ol,m then E',(HAc) = E,,, + 2Q.With this E',(HAc) value, [HI can be replaced with [HAc] in the appropriate equations. Acetate ion is formed during titration and the acetate concentration cannot be considered to be absolutely constant. This can be corrected for during computation by using the equation [equation (13) in ref. 71 : This value can be determined as follows. (V - Ve) + D (1 + VC,/VoC;,) [HAc] = 0 . . . . (21) where Ci, is the initial acetate concentration. mated value of Ve but an incorrect value of Kw. to solve Equation (2) is valid only before the equivalence point. Equation (7) yields an esti- Further calculation is carried out in order A(1,i) D + A(2,i)Ve + A(3,i) = 0 .. .. . . (22) where A(1,i) = (1 + VCB/V,C~,)[HA~] A(2,i) = -1 A(3,i) = v All measurements after the equivalence point are rejected.A constant error in the deter- mination of ElO (HAc) results only in a different value of D, and does not change the value of Ve- Using this procedure, the acid content of a sample can be determined, provided that their stability constants are less than lo5. For weaker acids, sodium sulphite (log K = 7.0) is used instead of sodium acetate.' Computer Program Initially, it uses El,,, jrr and temperature data to convert e.m.f. measurements, E , into ph, then initial and final ph measurements are compared. If the latter value is lower, a base has been titrated, and the program will calculate poh values from poh = pKw - ph. This requires an approximate value for pKw. Bases are treated in a similar manner to acids. Information is also required about initial sample volume ( V,) , titrant concentration (C,) and added A computer program (EKVOL) has been developed to carry out these calculations.April, 1978 ADDITION OF EQUAL VOLUMES OF TITRANT.PART 111 313 volume (V). The last value is calculated from the number of titrant additions and the volume of the pipette used for the titrant. Equation (7) is used to calculate Krw, K’ and Ve. The program utilises two routines. One computes coefficients in normal equations using the least-squares method. The other, a standard IBM routine, SIMQ, solves these equations. Subsequent calculations depend upon the value of the stability constant, as described above. Titrations of ampholytes and of acids in the presence of conjugate base must be identified.Computations for conditional titrations are carried out as for strong acids, except that the coefficients using [H’] are altered. The program outputs the values of Ve, Ktw, f and K’ (and additionally for ampholytes k’ and for conjugate bases C,). For each pair of measurements, total number i, these con- stants are used to calculate an equivalence volume Ve(i) using the complete equations. This gives the possibility of establishing whether or not the approximations made during the calculations were valid. The differences between V e and Ve(i) are used to calculate the standard deviation. Fig. 1 shows a flow diagram of the computer program. Evaluation of Computer Program Synthetic Titration Data The program was first evaluated using synthetic data, which were calculated from equation (3) for the titration of 100.0 ml of 0.01 M acid with 1.0-ml additions of a 0.1 M solution of strong base.Log K values of -10 and 0, 1, 2, . . ., 12 were chosen for the acid, with additions of 1, 2, . . ., 15 ml of base. These values correspond to a titration producing perfect data, without systematic and random errors. Equation (7) gave correct values for Ye, K and Kw. Data of this quality are unobtainable in practice, with errors in EIO measurements being the major problem. This problem has not been adequately handled in previously published programs except those described by Pehrsson et ~ 1 . ~ 9 ~ To evaluate the effect of systematic errors in Calculations based on equation (7) were incorrect, with the largest error for an acid with log K = 3.Calcula- tions using the methods described in this paper greatly improved the results; the relative error in equivalence volume was less than 0.1% for all acids with log K <11. The program was also tested using synthetic data for titrating ampholytes. ph values were calculated for an ampholyte with log k = 2.58 and log K = 4.32. A sample of volume 100.0 ml was titrated with 0.1 M sodium hydroxide solution with 1.0-ml additions. Calcula- tion using equation (7) resulted in V e = 10.35 (instead of 10.00) and log K = 4.44. Further calculation resulted in V e = 9.999, log K = 4.32 and log k = 2.57. all ph values were altered by + O . l and then -0.1 unit. Experimental Data of different acids and bases. The procedure was further tested with real data obtained from the titration of a number Apparatus A Metrohm No.EA 875-200 flask was used as the titration vessel, the titrant being added in 0.5-ml portions using a pneumatically operated pipette (Auto Chem Instruments AB, Bromma, Sweden). The volume of the pipette was determined by weighing ten aliquots of distilled water. The tip of the sample pipette was made of PTFE and had an external diameter of 1.8 mm. It was not allowed to come into contact with the titration solution. Pneumatic operation of the pipette results in a continuous stream of reagent, with no drops remaining on the tip of the pipette. The titration vessel was not thermostated, but was protected from heat from the magnetic stirrer with an insulated plate. The temperature was measured before and after every titration, and the mean value used to calculate Q.The e.m.f. measurement system consisted of a combination glass - silver - silver chloride micro-electrode (HA-401 M5, Dr. ing. chem. W. Ingold, Zurich, Switzerland) with a digital millivoltmeter (Model S 101 H, AB Systemteknik, Lidingo, Sweden).314 JOHANSSON AND JOHANSSON AUTOMATED TITRATION BY STEPWISE Analyst, vd. 103 Final output Start Q or e.m.f. data (IE = 1) are used; whether an ampholyte (IW = 1) or a mixture of an acid and its conjugate base (IW = 3) is Data in Q 1-1 e.m.f. data 1-1 --+ poh I Classification 4 Krw, K‘, V, I titration IW = 3? conjugate base Moderately strong Ampholyte -+ K’, f, Ve Weak +f, K: Ve Very weak + KL,K:Ve Amp h ol yte KtW, K’, k’, V, I 1 I Calculation of standard deviation Fig.1. Flow diagram. The input data consist of: the number of titrant portionsApril, 1978 ADDITION O F EQUAL VOLUMES OF TITRANT. PART I11 315 Reagents and materials The titrant was 0.1 M sodium hydroxide solution that was 0.133 M in barium chloride, giving an ionic strength of 0.5. Samples were dissolved in 25.00ml of water and then 25.00 ml of 0.33 M barium chloride solution were added. ElO was determined using a cali- bration solution that was 0.01 M in hydrochloric acid and 0.163 M in barium chloride (reagent grade). Procedure With each new electrode pair, jH and EtO values were determined by carrying out an EfO t i t r a t i ~ n . ~ The electrode pair used in this work had a jH value of -23 mV, and this was constant over a period of several months.EfO varied from day to day, but measurement of the e.m.f. of the calibration solution, Ecal, gave a sufficiently accurate value of Efo, calculated from EfO = Ecal + 2Q - O . O l j H . Titrations were carried out in a conventional manner except that the titrant was added in discrete aliquots, with an e.m.f. measurement after each addition. The electrodes required about 5 min after initial immersion to reach equilibrium. After each addition of titrant, 10-20 s were usually required for a stable potential to be attained, with a longer period after the large changes in potential near the equivalence point. This time delay should not be longer than necessary so as to minimise errors due to a slow drift in Efo. The first titration point at V = 0 ml is imprecise and should be rejected.Titrations are terminated when the equivalence point has been passed (except for very weak acids, which require an accurate value for Ktw and need three or four additions of titrant after the equivalence point). Results The procedure was evaluated by carrying out a number of different titrations, the results of which are given in Table I. Acids, bases and ampholytes with a wide range of stability constants were studied. The results are median values for five titrations (in some instances only three titrations) . They show excellent accuracy and precision, with relative standard deviations of less than 0.2% (0.3% for alanine). The stability constants listed in Table I were computed from titration data and may be biased because no attempt was made to determine accurate ElO values. The values will be valid only in the ionic medium used in TABLE I RESULTS OF TITRATION OF ACIDS, BASES AND AMPHOLYTES Type Acids Bases Ampholytes Conditional titrations Amount/mmol f A Recovery, Compound Added Found % Hydrochloric acid 0.500 0 0.500 5 100.1 Sulphamic acid 0.562 8 0.562 3 99.9 Sulphanilic acid 0.434 3 0.434 0 99.9 Picric acid 0.526 0 0.526 2 100.0 Formic acid 0.664 0 0.663 0 99.8 Acetic acid 0.528 3 0.528 2 100.0 Barium hydroxide 0.500 0 Ammonia 0.508 0 Sodium acetate 0.528 0 Potassium hydrogen tartrate 0.707 4 Potassium hydrogen phthalate 0.311 4 Glycine 0.647 4 /3- Alanine 0.607 4 0.499 5 99.9 0.508 1 100.0 0.527 8 100.0 0.707 4 100.0 0.311 7 100.1 0.647 8 100.1 0.605 0 99.6 Hydrochloric acid 0.500 0 0.500 0 100.0 Oxalic acid 0.435 4 0.434 0 99.7 Sulphanilic sulphamic and acids ::;:: }0.557 3 0.558 4 100.2 h g K <O 0.38 1.21 3.02 3.42 4.59 4.60 9.20 < O 2.45 2.57 <O <O 3.24 4.29 9.62 10.15316 JOHANSSON AND JOHANSSON these titrations, and they may be influenced by the formation of complexes between barium ions and the various acids.Values for ampholytes are “titration constants” (Part IV of this series). The authors express their gratitude to Mrs. Karin Lindgren for skilful experimental assistance. They are also indebted to Professor Folke Ingman and Dr. Lennart Pehrsson for stimulating discussions and to Dr. Roland Ekelund for valuable assistance in preparing the computer program. They also thank Dr. Douglas Mitchell for the translation of the manuscript. Financial support from the Swedish Board for Natural Sciences is gratefully acknowledged. References 1. 2. 3. 4. 5. 6. 7. 8. 9. 10. Johansson, A., Analyst, 1970, 95, 535. Johansson, A., and Pehrsson, L., Analyst, 1970, 95, 652. Ingman, F., Johansson, A., Johansson, S., and Karlsson, R., Analytica Chim. Acta, 1973, 64, 113. Nowogrocki, G., Cannone, J., and Wozniak, M., Bull. SOC. Chim. FY., 1976, 5. Pehrsson, L., Ingman, F., and Johansson, A., Talanta, 1976, 23, 769. Pehrsson, L., Ingman, F., and Johansson, S., Talanta, 1976, 23, 781, Johansson, A., Talanta, 1975, 22, 945. Biedermann, G., and Sillh, L. G., Ark. Kemi, 1953, 5, 425. Harned, H. S., and Geary, C . G., J . Am. Chem. SOC., 1937, 59, 2032. Gran, G., Analyst, 1952, 77, 661. NOTE-References 1 and 2 are to Parts I and I1 of this series, respectively. Received September 20th, 1977 Accepted October loth, 1977

ISSN:0003-2654    

DOI:10.1039/AN9780300305

出版商:RSC    

年代:1978

数据来源: RSC

摘要:

Analyst, A@d, 1978, Vol. 103, pp. 317-331 317 Automated Digestion and Extraction Apparatus for Use in the Determination of Trace Metals in Foodstuffs C. J. Jackson,* D. G. Porter, A. L. Dennis and P. B. Stockwell Department of Industry, Laboratory of the Government Chemist, Cornwall House, Stamford Street, London, SE1 9NQ An automated system has been constructed that uses a displacement technique to transfer foodstuffs, previously converted into a fluid state, into a modified Technicon digestor where a wet oxidation is carried out a t 400 "C with the aid of sulphuric acid, nitric acid and hydrogen peroxide. The digest is pumped into a neutralisation unit where its pH is adjusted to a pre-set value and then flows into an extraction unit where the metals of interest are chelated and extracted into an immiscible organic solvent.Finally, the organic phase is separated and collected for subsequent analysis by plasma-emission spectro- metry. The operation of the system is fully automatic and it processes 5-8-g samples at the rate of three per hour. Results obtained with the automatic system for the determination of iron, copper and zinc in a wide range of food- stuffs compare favourably with those obtained by manual techniques. Some preliminary results for the determination of cadmium, cobalt, nickel and lead are presented. Keywords : trace metal determination; foodstufls analysis ; automated digestion and extraction Widening interest in the quality of the environment has led to an increased demand for information on a wide range of trace-metal contents of foodstuffs.Trace metals in food- stuff s are normally determined by spectroscopic techniques after complete destruction of the organic matrix. Destruction is achieved either by wet oxidation or by dry ashing and, whichever procedure is used, additional treatment is normally required in order to obtain the metals of interest in a form suitable for analysis. Both methods of destruction are time consuming and tedious ; this is particularly true of the wet-oxidation procedure, which has the additional disadvantage that it is potentially hazardous. In addition, these methods require considerable analytical skill and experience. Both methods are prone to produce erroneous results either by the loss of an element of interest or by adventitious contamination from the component parts of the apparatus used.In this laboratory, increased analytical demands and the expense of suitable staff had resulted in a requirement for the complete or partial mechanisation or automation of this work. A review of the literature together with an examination of the procedures adopted at that time within the laboratory suggested that a wet-oxidation system would be the most suitable for automation. The specification of the automatic system was that it should be modular in construction in order to facilitate modification and maintenance and that it should have a multi-element capability. After careful consideration of the range and types of foodstuffs to be analysed, it was decided not to attempt to automate the sample preparation stages as only manual procedures provided sufficient flexibility.A system based on the above design criteria has been constructed and evaluated. This paper discusses the salient features of the design, the method of operation and the performance of the digestor system. In addition, the relative merits of the automated and manual procedures are compared. Apparatus It consists of five component modules: a sample introduction unit, a digestion unit, a neutralisation unit, a A schematic diagram of the automatic system is shown in Fig. 1. * Present address : Occupational Hygiene Laboratory, 403 Edgware Road, London, NW2 6LN. Crown Copyright.Nitric acid m-1 Hydrogen peroxide 1 II I Sulphuric , /I 1 acid (displacement) + I I 1 r- I I I I Digestor helix L--.---A Sampler Key I>] Pump I>] Dispenser Valve (non-return) @ Valve (2-or 3-way) Waste l / i Water (wash) Potassium sodium tartrate solution ~- -----I II ii I I Neutralisation vessel I I Extraction -Air Ammonia (anhydrous) _I DDDC solution Ethanol -water Phase boundary sensor Extract (collected in turntable) b 2 + Waste Fig.1. Schematic diagram of complete system.April, 1978 APPARATUS FOR DETERMINATION OF TRACE METALS IN FOODSTUFFS 319 chelation and extraction unit and an extract collection unit. The system is controlled by a series of interacting cam and electronic process timers. The method of sample preparation, the design of the sample introduction module and the general problem of transferring samples and acids from one module to another have been the subject of considerable investigations. Two particular problems place serious constraints on the types of materials used in construc- tion of the modules : the risk of physical damage to the system due to the extremely corrosive chemicals used (concentrated sulphuric and nitric acids and ammonia), and the possibility of trace elements from the components being introduced into the system and producing false results. These problems were overcome to a large extent by the choice of inert non- corrodible materials, for example, Kel-F, or by plating components with a noble metal such as gold.Further practical considerations that have influenced the design and construction of the modules are discussed in the ensuing sections. Sample Introduction Unit The method of sample preparation and the design of the sample introduction module are the result of a considerable amount of preliminary investigation to overcome a number of problems.The system must be capable of dealing with a wide range of samples having very different chemical and physical properties. Contamination of the sample by metals during the preparation stage must be avoided. The digestor must be fed with sample and acids a t constant rates and for a specified period. The whole of the sample or sub-sample must be transferred to the digestor without leaving a residue. As it is both difficult and dangerous to make accurate volumetric measurements of foodstuffs suspended in 33% sulphuric acid, these measurements should be avoided. These problems were solved in the following way.A manual method of sample prepara- tion was chosen so as to allow maximum flexibility in treatment of a wide range of samples. Whenever possible, the samples were freeze-dried and powdered. To minimise the possibility of contamination, samples were prepared for analysis in polyethylene bags. Quantitative measurements were performed by weighing the sample and bag at various stages in the preparative procedure set out below using an automatic balance and data recording system comprising a Mettler PT 1200 top-loading balance connected to an ASR-33 teletype through a Mettler CT-10 interface module. This balance has a capacity of 1200 g and weighs with a precision of &lo mg. The teletype produces a paper-tape record of the indicated mass on receipt of a command initiated by a push-button on the interface unit.The sample introduction unit is constructed from inert materials, which minimise the introduction of metal contamination into the system. The samples or digestion acids make contact only with PTFE, Kel-F, glass, acid-resistant rubber and platinum - iridium (9 + 1) alloy. In addition, the construction materials are limited to acid-grade Arborite (urea - formaldehyde laminate), Perspex and stainless steel. The unit is constructed in three contiguous sections : a heated sample compartment, a turntable mechanism and heat exchanger compartment and a pump compartment. Heated sample compartment The heated sample compartment is maintained at 60 “C and contains an Arborite turn- table that holds 20 glass sample displacement vessels with side-arms fitted at the 25-cm3 level. A pneumatic actuator is connected to a pair of platinum - iridium probes that can be lowered to the bottom of the sample tubes or raised to clear the lid of the compartment.A funnel-shaped mixing chamber, made from Kel-F and PTFE, is located below the sample tube side-arm, the tail of the funnel passing through the wall of the sample compartment and into the digestor helix. Two ports above the funnel allow the introduction of streams of concentrated sulphuric and nitric acids. A reed switch fitted to the wall of the sample compartment is actuated by a small permanent magnet that can be fixed to the turntable near the last tube in the batch, thus shutting down the system and leaving it in a “stand-by” condition.Transparent Perspex top and front panels are fitted to the sample compartment to allow inspection during operation; these panels can be removed to allow insertion and removal of the turntable. The turntable base is sealed and fitted with a rim large enough to contain 150 cm3 of acid in the event of a spillage.320 JACKSON et aZ. : AUTOMATED DIGESTION AND EXTRACTION Analyst, VoZ. I03 Turntable mechanism and heat exchanger The turntable mechanism and heat exchanger is situated below the sample compartment. An Arborite disc is fixed securely to a central. stainless-steel spindle on which the turntable is also mounted. A horizontal pneumatic actuator is fitted beneath this disc in such a way that when it is extended it pushes against one of a set of 20 pegs mounted on the under- surface of the disc and thus rotates the turntable through 18".A second pneumatic piston mounted vertically below the disc can be extended to force a conical pin into one of a set of 20 holes drilled in the underside of the disc. This locking device has two functions: it positively locates the side-arm of the sample vessel over the funnel and it prevents the turntable from being moved accidentally. Electrically interlocking switches provide an additional safeguard by preventing the locking and rotating actions being carried out simultaneously and by inhibiting rotation when the probes are lowered. The heat exchanger consists of a small centrifugal fan that takes air from outside the unit and forces it, via ducts, to the four corners of the sample chamber.A heater made from Nichrome wire wound around a ceramic former is situated within the ducting immediately upstream of the fan. The temperature within the sample chamber is monitored by a thermistor set in the floor of the chamber and a proportional temperature controller (RFL Electronics Ltd., MOD 70) maintains the temperature at 60 "C. Pump compartment In the pump compartment situated at the base of the unit separate pumps are provided for the displacement sulphuric acid, the by-pass sulphuric acid, nitric acid, hydrogen peroxide and the removal of waste displacement acid from the sample tubes. The displacement acid is dispensed by a purpose-built syringe pump ; single-channel peristaltic pumps are used for the other liquids. The syringe pump, apart from the stepping motor and control circuitry, is constructed from corrosion-resistant materials; the acid coimes into contact only with glass, PTFE and Kel-F.The syringe (a modified 50-cm3 Hamilton Gas-Tight Model 1050) is driven by a stepping motor, which permits discharge at a constant and reproducible rate of 2.0 cm3 min-1 and refilling at 6.0 cm3 min-l, coupled with fast manual operation (25 cm3 min-1) in either direction. The limits of syringe travel, and therefore the volume dispensed per stroke, are determined by a rod of suitable length that is attached to the piston mounting and which moves between two microswitches. Operation of either of these microswitches stops the syringe drive and also switches a pneumaticably actuated three-way valve (Chromatronix CAV-3060-K), fitted at the end of the syringe, from the fill to the discharge position, or vice veysa.The by-pass sulphuric acid, the nitric acid and the hydrogen peroxide streams are pumped by Hiloflo peristaltic pumps (Metering Pumps Ltd.) that are fitted with Acidflex tubing (Technicon Instrument Co.) on the acid streams and Tygon tubing on the hydrogen peroxide stream. These pump tubes are connected by modified Chromatronix couplings (Type G2-C) to 0.125-in 0.d. PTFE tubing, which is used for most of the plumbing in the sampler unit. An Ismatek Mini-micro-2 pump is used to remove waste displacement acid. The flow-rates are monitored with all-glass flow meters (S.G.E. Ltd.). Digestion Unit The digestion unit consists of a modified Technicon continuous digestor with ancillary equipment for introduction of samples, removal of the digest and disposal of fumes.The continuous digestor consists of a borosilicate-glass helix, which is rotated over three banks of heaters; the samples and digestion acids are fed in at one end, transported over the heaters by the rotation of the helix and the resulting digest is pumped out of the other end of the helix for subsequent analysis. A long glass tube, inserted at the output end of the helix, allows the addition of 100-volume hydrogen peroxide to complete the digestion process, The control of the temperature of the standard digestor was found to be inadequate for our application, the desired operating temperature being attained by adjusting the current to the heaters so that it balanced heat losses. Fitting proportional heating controllers to each of the three banks of heaters has resulted in greater flexibility in the operation of the digestor, a more rapid attainment of the working temperature and the virtual eliminationApril, 1978 APPARATUS FOR DETERMINATION OF TRACE METALS IN FOODSTUFFS 321 of temperature drift.A description of these modifications and the resulting improved performance of the digestor has been published previous1y.l The output of the digestor is pumped from the helix through a pneumatically actuated valve (Chromatronix CAV-3060-K) either into the neutralisation vessel or to waste. A pump was designed and built in the laboratory for this purpose. Its action is based on an oscillating piston, it has no moving valves and the hot acid comes into contact only with Kel-F and PTFE.This pump has been described in more detail elsewhere.2 The maximum flow-rate is about 12 cm3 min-l. Neutralisation Unit This unit consists of a jacketed glass vessel with an Arborite lid and a motor-driven three- way glass stopcock with PTFE key fitted to the outlet. Fitted into the lid are a demountable, re-fillable combination glass pH electrode (Activion Ltd.), and inlet tubes for digest, a dilution - buffer solution, the gaseous ammonia - air neutralisation mixture and the wash water. The jacket is maintained at 35 "C by a bath of circulating water; this temperature was chosen to give adequate cooling without risk of formation of ammonium sulphate crystals, which would tend to block the outlet valve.The pH signal is fed via a pH meter to a potentiometric recorder and to a neutralisation controller (designed and built within this laboratory), which compares the electrode e.m.f. with pre-selected values (determined by two calibrated 10-turn potentiometers) and opens or closes gas control valves accordingly. Depending on the pH, it is possible to obtain a flow of air, ammonia or a mixture of both gases. By careful adjustment of the potentio- meters and the flow-rates of ammonia and air, it is possible to control the final pH and also to keep the digest within a fairly closely defined pH range during the neutralisation process. Efficient and continuous mixing is essential for proper control of the neutralisation process. This is achieved by introducing a stream of air into the stem of the vessel immediately above the outlet valve.It is not sufficient to rely on mixing by the incoming stream of neutralising gas because when pure ammonia is being used the gas dissolves so readily that no turbulence is induced and a dense layer of acid tends to collect at the bottom of the neutralisation vessel. An automatic dispenser (Fisons LFA 100) is used to add 150 cm3 of a 2% m/V solution of potassium sodium tartrate to the neutralisation vessel before the digest is introduced. This protects the electrode and helps to reduce the concentration of dissolved salts. Two jets mounted in the lid provide a spray of water to wash the walls of the vessel, the electrode and the gas inlet tube. Initially, washing is carried out with the outlet valve closed and a vigorous stream of air passing through the inlet tube; the valve is then opened to drain the vessel and washing is continued with the outlet valve opened.Chelation and Extraction Unit This unit is positioned directly below the neutralisation unit and is constructed from a modified glass reaction vessel (Quickfit FR 700F) fitted with a standard reaction head (Quickfit MAF 2/2) with inlets for sample, water wash, ethanol - water wash, chelating reagent solution, compressed air, stirrer and extract removal probe. An outlet in the bottom of the vessel is connected to a phase boundary detector, which is connected in turn to a valve (Chromatronix CAV-2060-K) that can be switched to a waste line or to a collection vessel. Automatic dispensers are used to add both the chelating reagent [a 2% m/V solution of diethylammonium diethyldithiocarbamate (DDDC) in heptan-2-oneI and the ethanol - water wash liquid.The latter is used to remove trace amounts of residual organic solvent and it is therefore stirred vigorously, allowing it to splash off the top of the extraction vessel and run down the walls. This liquid is run off before the vessel is rinsed finally with water admitted through a perforated circular glass tube fitted within the reaction vessel head. The extraction vessel outlet valve can be controlled manually from the control unit or automatically by the phase boundary detector unit. This detector, the operation of which is based on the difference in refractive index between the organic and aqueous phases, has been described elsewhere.3 The organic phase is removed by applying a positive air pressure to the extraction vessel, thus forcing the extract up a PTFE tube and through a probe into a glass or polypropylene322 JACKSON et a,!.: AUTOMATED DIGESTION AND EXTRACTION Analyst, ‘Vd. 103 extract collection vessel situated in the extract collection module. During the washing cycle, the tube and probe are washed out with ethanol - water and then blown dry with air. Extract Collection Unit This unit consists of a turntable with provision for holding 20 glass or polypropylene tubes (28-mm diameter, 25-cm3 capacity). Two pneumatic pistons are used to raise and lower the probe and to rotate it through 90” to position it either over a tube for collection of extract or over a waste outlet for rinsing during the wash cycle. Both the rotation of the turntable and the movement of the probe are controlled. by the central control unit.Control System Most of the control system is situated in a series of 19-in rack-mounted units. As far as possible, each of these units is self-contained with one controlling the sample introduction unit, one the neutralisation unit, one the chelation - extraction and extract collection units and one being used as the source of the various power supplies. The neutralisation controller and phase boundary sensor are situated in small separate units and the digestion unit controller is incorporated into the digestor module. The system control is based on a series of six cam timers, which control the following basic operations : sample introduction unit, sampling phase; sample introduction unit, re-filling phase; neutralisation unit, neutralisation phase.; neutralisation unit, wash phase ; chelation - extraction unit, extraction phase; and chelation - extraction unit, collection and wash phase. Linked to these cam timers are individual electronic process timers that are used to control the over-all sampling rate, to control the valve that diverts the sample digest to the neutralisation module, to fix the beginning and end of the neutralisation process in relation to the initial sampling period, and in a number of processes such as dispenser operation and valve rotation where short and accurately fixed time intervals are required. Fig. 2 shows the control system in diagrammatic form, Fig.3 shows the timing sequence and Table I lists the various components and their functions. In normal operation, the control cycle is initiated by pressing a push-button switch (“Manual Start” in Fig. 2) ; the system will continue to operate automatically until a sequence of operations is initiated by a last-tube sensor in the sampler module to stop the instrument in a safe manner. IT1 CT 1 Pump CT2 IT3 CT4 CT5 Valve CT6 I T2 CT3 / A ‘B I I I I L 0 10 20 30 4( Timdmin Fig. 3. Control system: timing diagram. A, Start of first cycle; B, end of first cycle; and C, start of second cycle.I 413 I I I I I I I I I I I I I I I I I I I I L + I I I I324 JACKSON ef d. AUTOMATED DIGESTION AND EXTRACTION Analyst, Vd.103 TABLE I CONTROL SYSTEM TIMING UNITS : FUNCTIONS AND DESCRIPTIONS Unit* Description Cam timer 1 Sample introduction unit- Cycle time 3 rnin : 6 switches Sample introduction unit-sample Cycle time 3.75 min : 8 switches sampling process. Cam timer 2 change-over process. Cam timer 3 Neutralisation unit-neutralisation process. Cycle time 9 rnin : 6 switches Neutralisation unit-wash process. Cycle time 6 rnin : 15 switches Chelation - extraction unit- Cycle time 7.5 min : 8 switches Chelation - extraction unit- collection and wash process. Cycle time 9 min : 12 switches Cam timer 4 Cam timer 5 extraction process. Cam timer 6 Unit* Interval timer 1 Interval timer 2 Interval timer 3 Interval timer 4 Interval timer 5 Interval timer 6 Interval timer 7 Interval timer 8 Description Over-all cycle time.Type D 100H2 :time 20 rnin Helix transit time delay. Type D lOOHl : time 5 min Digest collection period. Type D 200H1 :time 7.6 rnin Tartrate reagent dispenser Type SlOOC4 : time 4 rnin 55 s Neutralising unit outlet valve- Type D lOOCl :time 1.75 s Neutralising unit outlet valve- Type D lOOCl :time 1.7.5 s Chelating reagent dispenser Type S 100C3 : time 36 s Extract collecting unit turn- Type SlOOCl :time 3.8 s control. collect. waste. control. table motor. * Cam timers were Edgcumbe-Peebles Type Atcotrol 324C. Interval timers were obtained from Deltic Automation Ltd. The “Manual Start” signal sets cam timer 1 (CT1) into action. CT1 controls the sample (i) It starts interval timer 1 (ITl), which determines the over-all cycle time of 20 min.IT1 and CT1 will continue to operate sequentially, three times per hour, until the loop is broken by a “stop” signal from the last-tube sensor. (ii) It starts the displacement sulphuric acid pump, initiating the flow of sample into the digestor helix. When the pump is fully discharged, a microswitch is closed thus completing a circuit which sets CT.2 into operation. (iii) It sends a pulse to start interval timer 2 (ITZ), which, after an interval of 5 min, the time required for the sample front to pass through the helix, sets CT3 into operation and thereby initiates the digest collection and neutralisation process. CT2 sends a pulse to IT3, which, after an interval of 7.5 rnin (by which time the whole of the sample has been passed through the helix), starts CT4 and initiates the neutralisation unit washing cycle.CT4, in turn, starts CT5, which controls the chelation and extraction process. When the extraction is complete, CT5 opens the valve on the extraction vessel, the phase boundary detector closes this valve when it senses the interface between the organic and aqueous phases and simultaneously transmits a pulse to start CT6, which controls the processes of collecting the extract and washing the extraction vessel. The analytical cycle is terminated when CT6 reaches the end of its operating cycle, the complete process taking about 37 min. introduction process and has three separate functions in the control cycle : Use of a cascading series of cam timers has several operational advantages: (i) It allows independent control of the various phases of the cycle. (ii) It simplifies adjustments to the timing cycle; in many instances changes can be made without appreciably affecting subsequent operations.(iii) It allows parts of the system to adjust automatically to fluctuations in some of the processes, e.g., the volume of neutralised digest varies depending on the amount of acid consumed in the digestion process; this will influence the time taken to separate the phases after chelation and extraction and the system is therefore arranged so that CT6 is started only when the phase boundary detector senses the interface. Indicator lights on the control panel show the current state of the various components of the control system. A number of manually operated switches are provided to over-ride the automatic controls when required. The control system incorporates a number of safety interlocks and cut-outs, e.g., the sampler turntable cannot be rotated while the probe is inApril, 1978 APPARATUS FOR DETERMINATION O F TRACE METALS I N FOODSTUFFS 325 the “down” position and the displacement sulphuric acid pump will not operate while the probe is in the “up” position.In the event of an electric power or pneumatic line failure, valves are arranged so that they close or switch to waste. Leak detectors activate circuits to sound alarms and switch the system to a safe condition when an acid leak has occurred. Method of Operation Sample Preparation Weigh a 5-10-g amount of the freeze-dried and ground sample into a weighed polyethylene bag and add 30 cm3 of 33% sulphuric acid with the aid of a dispenser. Suspend the bag in an oven at 60 “C for 20 min, transfer the bag to a Colworth Stomacher and homogenise for 30 s, then replace it in the oven for 90 min.Weigh the bag and its contents, pour a sub-sample of the slurry into a sample tube, filling it to a point just below the level of the side-arm, re-weigh the bag with its residual contents and place the sample tube in the turntable. When the turntable is fully loaded with samples, standards and blanks, place the last-tube marker in position and rotate the turntable manually to position the first tube under the sampling probe with the side-arm extending over the mixing chamber that leads to the digestor helix. At some convenient stage during the sample preparation process, the digestion system must be prepared for operation.Switch on the services (240 V a.c., compressed air, demineralised water, ammonia, vacuum pump), switch on the digestor helix rotation motor and the supply to the heaters, check and if necessary adjust the sulphuric acid, nitric acid and hydrogen peroxide flow-rates and use the manual over-ride controls to drain the neutralisation and extraction vessels. All other functions of the system are activated automatically when the 240 V ax. supply is switched on. Press the “Manual Start” switch when the system is loaded with samples and has been prepared for operation. Prepare a blank in a similar manner. All processes that follow are completely automatic. Digestion The platinum-iridium probes are lowered into the sample vessel and into the previous sample tube (if present).Concentrated sulphuric acid is pumped through the first probe into the bottom of the sample tube, thereby displacing the sample slurry into the mixing chamber at a constant rate (2 cm3 min-l). The ratio of sulphuric acid to nitric acid is preserved by stopping the direct input of sulphuric acid into the mixing chamber while the sample is being introduced. The nitric acid flows continually at a rate of 8 cm3 min-l. Sufficient displacement acid is delivered to transfer all of the sample into the digestor. The second probe is connected to a continuously operating peristaltic pump, which drains the acid from the previously used sample tube and discharges it to waste. The probes are then raised and allowed to drain.The displacement acid pump is re-filled. The sample and nitric acid mixture flows from the mixing chamber into the digestor helix by gravity and is transported over the heating zones by the rotation of the digestor helix. The three heating zones are maintained at 380, 390 and 400 “C. Hydrogen peroxide (100 volume) is pumped at 2 cm3 min-l into the digestor helix at a point where the sample has traversed approximately 90% of the heated area. The residual digest is partially cooled in the end zone of the helix and then pumped through a valve to the neutralisation vessel. The valve is switched to waste when the wash-acid between samples is flowing. Neutralisation The digest flows into the neutralisation vessel that has previously been filled with 150 cm3 of a 2% rn/V solution of potassium sodium tartrate.The pH controller is activated and valves switch automatically to control the admission of a neutralising stream of anhydrous ammonia. The valves switch to admit pure ammonia when the pH is less than 2.5, a mixture of air and ammonia when the pH is between 2.5 and 3.5 and air only when the pH exceeds 4.0. Neutralisa- tion takes place continuously while the sample digest enters the vessel. The controller remains activated for about 1.5 min after the end of the sample collection period to allow the system to attain the pre-set pH value. The gas control valves are then closed and the With these settings the pH never exceeds 7.0 and the final pH is 4.5 & 0.2.326 JACKSON et al. : AUTOMATED DIGESTION AND EXTRACTION Analyst, VoZ.103 outlet valve is opened, allowing the neutralised digest to drain into the extraction vessel beneath it. The outlet valve is switched to waste and the neutralisation vessel is washed with a stream of water admitted through spray-jets. Chelation and Extraction A 2O-cm3 volume of a 2% m/V solution of diethylammonium diethyldithiocarbamate in heptan-2-one is added to the neutralised digest in the extraction vessel and the mixture is stirred vigorously for 2 min. The two phases are allowed to separate over a period of 3.5 min, the phase boundary detector is switched on and the outlet valve is opened, allowing the lower aqueous phase either to run to waste or to be collected for further treatment. When the phase boundary detector senses the phase boundary, the outlet valve is closed, retaining the organic phase in the extraction vessel.A pressure of 4 l b k r 2 is applied to the extraction vessel, forcing the extract through a PTFE tube into a collection vessel held in a turntable. The extraction vessel is then washed with 160 cm3 of a 50% aqueous ethanol solution, which is sufficient to provide efficient washing of the walls and cover of the vessel when the stirrer is operated. The extract collection flow line is rinsed by applying a pressure of 4 lb in-2 to the vessel, which is then drained via the outlet valve. Any residual ethanol is removed by repeating the washing process with water. The outlet valve is shut just before complete drainage is achieved, to ensure that water rather than air is present in the phase boundary sensor tube when the next sample is introduced.The sequence of operations described above takes about 35 min to complete; the cycle time of 20 min is obtained by processing two samples simultaneously. When sampling begins on the sample in the turntable, the preceding sample is just completing the digestion stage, the sample previous to that will have been processed completely and the extraction vessel washing cycle will be almost complete. When the last-tube marker reaches the sampling position, it activates a microswitch that inhibits further operation of the master cycle timer. The subsidiary timers complete their cycles, leaving the neutralisation and extractions vessels clean. The operator completes the shut-down process by switching off the three digestor heaters, operating manual over-ride controls to fill the neutralisation and extraction vessels with water, switching off the sulphuric acid, nitric acid and hydrogen peroxide flows, stopping the helix rotation and finally shutting down all services to the instrument.Performance The performance of the system is shown in Tables II-IV. The analysis of the NBS bovine liver (Table 11) shows that the automatic system is capable of giving accurate results. The figures in Table I11 were obtained on samples that were freeze-dried on receipt. Measure- ments on manually digested samples were made by atomic-absorption spectrophotometry and on automatically digested samples by plasma emission spectrometry. The agreement between automatic and manual determinations is very satisfactory.There is some indication that in the iron determinations the automatic method gives slightly lower results than the manual method, but there are insufficient data to test the significance of this difference. A possible explanation for this observation is discussed under Neutralisation. The recovery figures in Table IV confirm the performance demonstrated in Tables I1 and 111. TABLE I1 ANALYSIS OF NBS STANDARD REFERENCE MATERIAL 1577 (BOVINE LIVER) Element NBS value*/pg g-l This worktlpg 8-l Fe 270 f 20 262 -& 7 cu 193 & 10 189 f 7 Zn 130 -& 10 133 f 6 * The NBS values are based on the results of 6-12 determinations by a t least two methods. t The values from this work are the means of 9 determinations.April, 29'18 APPARATUS FOR DETERMINATION OF TRACE METALS IN FOODSTUFFS 327 Discussion Sample Preparation and Introduction During the earlier stages of the development of this automatic system, during which solid sampling techniques were investigated, a number of severe problems arose.This laboratory is required to analyse a wide range of foodstuffs, and to build a fully automatic system that will cope with all types of material would be not only technically difficult but also very expensive. The sampling approach that has been devised is simple, flexible and relatively inexpensive but does involve a certain amount of manual work. TABLE I11 COMPARISON OF AUTOMATIC AND MANUAL DETERMINATIONS OE IRON, COPPER AND ZINC I N VARIOUS FOODSTUFFS Most values given are the mean of two or three determinations carried out on separate portions of the freeze-dried sample.I Description Canned chicken soup . . .. .. Canned vegetable soup . . .. Dried chicken noodle soup . . . . Dried tomato soup . . .. .. Curried beef in gravy . . .. .. Green pea soup .. . . .. Drinking chocolate . . .. .. Minced beef in gravy . . .. .. Processed peas .. .. .. Chocolate pudding . . .. .. Plain biscuits . . .. .. .. Fruit biscuits . . . . .. . . Baked beans . . .. .. .. Ironlpg g-l Auto Manual 3.0 3.6 4.6 5.8 V f 21 27 15.2 18 74 76 67 68 13.0 12 20 25 14.4 19 34 34 11.6 13 17.3 20 41 42 Copper/@ g-l - r Auto Manual 0.2 0.2 0.6 0.6 2.2 2.6 3.2 3.3 3.2 3 6.4 6 1.6 (1 3.5 3 4.0 4 0.6 <1 2.4 1.5 2.0 1.5 3.9 5 Zinclpg g-l - Auto Manual 3.2 3.3 2.9 2.9 11.2 11.5 7.2 7.5 69 68 22 21 6.0 5.0 4.6 3.4 21 21 29 28 10.0 8.8 5.8 6.0 8.2 8.5 At one stage, the use of a Technicon Mark I SOLIDprep sampler was investigated.It was recognised that, because the homogeniser chamber was made from stainless steel, the SOLID- prep would not be suitable for trace-metal work; however, it was thought to be worth examining this approach in order to determine what other problems might arise. It was found that the macerator would not cope with all foods (it was not designed for this purpose), but, if the suspension medium was warm 50% sulphuric acid rather than water, almost all foods could be reduced to an adequately homogeneous state. To transfer a sub-sample of the homogenised sample from the SOLIDprep to the digestor requires some form of pump, the obvious choice being a peristaltic pump.Other pumps introduce metallic contamination, are not inert to acids, require priming or have large dead- volumes. However, even when using Acidflex pump tubes it was found that the effective TABLE IV RECOVERY OF KNOWN AMOUNTS OF ADDED METALS Recovery, yo Element Spike*/pg Fe 250 c u 250 Zn 100 Cd 50 Cd 25 Pb 60 c o 50 Ni 50 Ni 25 Range 96-99 96-100 97-107 98-105 - 90-1 15 97-107 96-103 - 1 Mean 97 99 104 102 102 104 103 100 106 No. of determinations 3 3 3 3 1 3 3 3 1 *The smallest spike size of 25 p g corresponds to a level of approximately 7 p g g-1 in a freeze-dried food sample (approximately 1-4 pg g-l in most food samples as received).328 JACKSON et d. : AUTOMATED DIGESTION AND EXTRACTION Analyst, Vd.103 pump delivery rate decreased with time because the pump tube was becoming flattened by use. This made it difficult to determine the amount of the sub-sample as the only simple way of doing this was to pump at a known rate for a fixed period of time. In this application, the performance of the pump tube deteriorated almost immediately and even changing the pump tubes twice daily was not adequate. An attempt to devise an automatic method of monitoring flow-rates by making measurements during the short period between the intro- duction of successive samples was unsuccessful If a method could be devised to measure precisely low flow-rates of a stream of a corrosive fluid containing some particulate matter and whose composition varies, then it should be possible to feed-back a signal to a servo- motor mounted on a peristaltic pump, such as the Hiloflo, where the flow-rate is adjustable by means of a 10-turn micrometer, and thus maintain a constant output flow-rate, To overcome the problem of the varying pump delivery rate, the use of a calibrated reservoir into which the sub-sample could be pumped and held until required for digestion was investigated, the whole of the contents of the reservoir being pumped into the digestor.It was found that there were some problems in storing partially digested semi-solid materials; valves tended to block with particulate matter but more serious problems were caused by the evolution of gases from the stored digest. This line of investigation was not pursued any further as it was at this time that the current sampling method was devised.However, it is believed that it might be worth investigating further the use of a Technicon Mark I1 SOLIDprep sampler (the SPS-11, in which the sample cannot come into contact with any metals), together with a storage coil held in a refrigerated bath, possibly combined with an ultrasonic de-gassing treatment of the digest. The use of media other than 33% sulphuric acid for sample preparation has been investi- gated, with particular reference to mixtures of sulphuric and nitric acids. Such mixtures are more effective than sulphuric acid alone, but were rejected because of serious problems of frothing and gas evolution. The proposed sample preparation and introduction technique has several advantages : The manual method of sample preparation gives greater flexibility than could be obtained with any fully automatic procedure.Investigation has shown that the sulphuric acid concentration is fairly critical, the optimum strength being about 33% for most types of sample. If the strength is reduced appreciably, dissolution of the sample is not very efficient, while if it is increased there is a tendency for charring to occur. The manual procedure makes it easy to adjust the amount and concentration of the acid used, thereby compensating for changes in the water content of the sample. It is simple, also, to give special manipulation or treatment to any sample that is still not homogeneous after the initial treatment with warm acid. No volume measurements are necessary.If gases are being evolved from the sample preparation, volume measurements can be difficult and inaccurate. Transfer of the sub-sample to the digestor is quantitative. Initially a slender glass probe was used to introduce the displacement acid into the bottom of the sample tube. The glass probe was very fragile and tended to retain a thin coating of sample when withdrawn from the tube. This problem was overcome by using a platinum - iridium alloy probe. The rate of sample addition is very easily controlled and remains constant as it is determined by the movement of a motor-driven syringe rather than a peristaltic pump. A constant rate of sample introduction is very important as the flow-rates of sulphuric and nitric acids are designed to be sufficient for normal use.Large excesses would not only give problems in the neutralisation stage but would also give rise to an unnecessarily large blank value. A surge in the sample introduction rate could give a local deficiency of digestion acid within the helix. Use of an automatic balance linked to a paper-tape punch or, as is now intended, directly to a computer, eliminates a potential source of transcription errors. Very little washing up is necessary. There are some minor drawbacks associated with the proposed technique, the main one occurring with fatty samples where there is a risk of fat separating and rising to the top of the sample tube on storage before analysis. This can cause problems in the digestion stage if all of the fat enters the helix at the same time, and in the worst instance can actuallyApril, 1978 APPARATUS FOR DETERMINATION OF TRACE METALS IN FOODSTUFFS 329 prevent the sample from flowing into the helix if the sample tube side-arm should become blocked with congealed fat.I t has been found that by maintaining the sample at 60 "C it is possible to prevent the fat from separating from the homogenised sample preparation. Samples that contain bone fragments can be a problem as the bags will become perforated during homogenisation in the Stomacher. Hard materials, such as spaghetti, are best analysed after soaking overnight in water. Digestion Early in these investigations it became obvious that limitations in the design of the standard Technicon AutoAnalyzer digestor module were severely restricting its utility for our purposes.Its performance was measured by pumping an inert, high-boiling liquid through the helix and monitoring the temperature at several points within the helix with the aid of a thermocouple. It was observed that the temperature fluctuated with time in an unpredictable manner, often by as much as &5 "C over a period of 30 min, and over longer periods there was a significant drift in the mean temperature, probably in response to changes in ambient conditions and in the characteristics of the heaters. Fluctuations of this magnitude seriously affect the performance of the digestor if it is working near the limits of its capabilities, when a small reduction in temperature may result in incomplete digestion, and a small increase may char the sample.Replacement of the two variable transformers in the digestor with three digital proportional temperature controllers, one for each bank of heating elements, considerably improved the performance.1 Close temperature control makes it possible to minimise the excess of digestion acid used, which helps to keep blank levels low. Other operational advantages ensue, the chief of which are the ability to control the temperature of each of the three heating zones separately and the ability to attain the working temperature much more rapidly. Most samples are digested very smoothly and real difficulties have been experienced with only two types of substrate. Sucrose tends to froth vigorously, causing the contents of a helix to spill over into the adjacent helices; this problem is easily avoided, either by decreasing the temperature in the initial stage of the digestion or by digesting less sample. With samples that are rich in fat (e.g., margarine) volatile fatty acids distil within the helix and condense, undigested, at the cooler outlet end of the helix. It has not yet been established whether this causes a reduction in the recovery of trace metals, but the presence of solid fatty materials in the subsequent neutralisation and extraction stages can cause problems.It may be possible to prevent this happening by reducing the temperature of the initial stage of the digestion process and simultaneously reducing the speed of rotation of the helix, thus increasing the residence time within the digestor. Neutralisation The use of gaseous ammonia, rather than an aqueous solution of an alkali, to neutralise the acid digest has several advantages: it will not add to the reagent blank provided that the gas is filtered, it is much easier to control the addition and mixing of a gas stream than a liquid stream, less heat is evolved and there is no increase in liquid volume.The neutralisa- tion reaction is very smooth and is always easily controlled. The performance of the neutralisation system is illustrated in Table V, which lists the final pH values attained with batches of samples analysed on three different occasions. The pH controller was set at pH 6.50. The location of the gas inlet is critical; if placed lower than the glass electrode bulb, the electrode rapidly becomes contaminated and gives false readings.The outlet of the gas supply tube should be slightly above the electrode. Even with this precaution the electrolyte gradually absorbs ammonia, but this is not a serious problem provided that the electrode is calibrated daily with a standard buffer solution. A demountable combined glass - reference electrode is used so that it is simple to renew the electrolyte when contamina- tion becomes excessive. The addition of a 2 7 , solution of sodium potassium tartrate to the neutralisation vessel before the digest is introduced has three purposes: (i) it protects the glass electrode from damage by the hot concentrated acid in the initial stage of the neutralisation; (ii) it provides a certain amount of buffering action; and (iii) it minimises the loss of iron and copper as The digestion stage has given much less trouble than was anticipated initially.330 JACKSON et al.: AUTOMATED DIGESTION AND EXTRACTION Analyst, vol. 103 Sample No. 1 2 3 4 5 6 7 8 9 10 11 12 13 14 Mean .. Standard deviation Range .. TABLE V PRECISION OF pH ADJUSTMENT pH attained r I Day 1 Day 2 Day 3 6.60 6.65 6.40 6.50 6.50 6.40 6.60 6.30 6.35 6.30 6.35 6.65 6.65 6.75 6.40 6.60 6.60 6.45 6.55 6.60 6.75 6.60 6.60 6.45 6.60 6.45 6.80 6.30 6.30 6.50 6.50 6.35 6.40 6.20 6.70 6.45 6.60 6.65 - 6.60 6.70 - . . 6.51 6.54 6.50 . . 0.142 0.159 0.149 A . . 6.20-6.65 6.30-6.75 6.35-6.80 insoluble hydroxides should the pH exceed 6-7 during the neutralisation process. We have observed that this loss can be serious if the pH reaches 8 or 9, even for only a few seconds.Extraction The extraction stage has been designed with a considerable amount of in-built flexibility. It is a simple matter to change chelating reagent, pH or extraction solvent. However, the equipment would require some modification if a solvent more dense than water were to be used. When this work was started ammonium tetramethylenedithiocarbamate in 4-met hyl- pentan-2-one was used as the chelating reagent solely because it was very widely used in manual analyses. Considerable trouble was experienced with this reagent ; frequently the metal-chelate extracts were unstable and had very short storage life (about 2 h), and some- times the reagent decomposed visibly when added to the neutralised digest. Much of the evidence pointed to the presence of something in the neutralised digest, perhaps free nitrogen dioxide, that was causing the decomposition.Greater success was obtained by changing to a solution of quinolin-8-01 in heptan-2-one. Both the reagent and the metal-chelate extracts appeared to be stable indefinitely. Recently, however, with the change to using hydrogen peroxide in the later stage of the digestion, problems have arisen with using quinolin-8-01 in the determination of zinc and copper, but by substituting a 2y0 solution of diethylammonium diethyldithiocarbamate in heptan-2-one it is now possible to make satisfactory determina- tions of iron, copper, zinc, nickel, cobalt and cadmium. Although heptan-2-one is more expensive than 4-methylpentan-2-one, it has two advantages that justify the increased cost. Firstly, it improves the reliability of the phase boundary dete~tor,~ and secondly, it is much less soluble in water.Because of this reduced solubility, changes in the volume of the neutralised digest (as a result of greater or lesser amounts of acid being consumed in the digestion process) ha.ve a much smaller effect on the final volume of the organic solvent. This makes it possible to use smaller volumes of solvent and thus achieve larger concentration factors in the extraction from the aqueous to the organic phase. At present, extractions are made with a nominal 20 cm3 of heptan-2-one; it should be feasible to extract with only 5 cm3 of organic solvent and still have sufficient extract for a single measurement. With such small volumes, however, it might be desirable to incorporate an internal standard .Measurement Initially, all measurements were made by atomic-absorption spectrophotometry, which can be time consuming when four to six elements have to be determined on each extract. A more serious problem was that occasional periods were experienced when it was found to be impossible to obtain reproducible measurements. However, satisfactory measurementsA p d , 1978 APPARATUS FOR DETERMINATION OF TRACE METALS I N FOODSTUFFS 33 1 are now being made with the aid of an ARL plasma emission spectrometer. No problems of sample nebulisation have been experienced with this instrument, which is capable of making simultaneous measurements of up to 24 elements. Conclusion The automatic system described here has several advantages over normal manual pro- cedures.It eliminates the tedious, unpleasant and potentially hazardous work of wet digestion and solvent extraction. It provides better control of experimental conditions, which gives increased analytical precision and increased safety. The automatic system makes it possible to control the problem of contamination from the laboratory environment (which can be a serious problem in the absence of a properly constructed clean-room) and also from the apparatus used. Blank levels tend to be more consistent. The automatic system saves staff time and there is scope for considerable further improvement should an automated sample preparation procedure be developed. Automatic sample preparation would be a valuable facility for organisations, such as quality control laboratories, that examine reasonably large numbers of samples of similar composition.Use of a continuous flow digestion module considerably simplifies the task of automating the digestion procedure. In order to analyse the maximum amount of sample, the system has been designed to operate on a discrete sample processing basis. Even the digestion stage is not a true continuous flow operation, as each sample is processed in turn until the whole sample has been treated. Automation of the digestion step on a discrete basis would be a formidable task, whereas the use of the continuous digestor is a relatively simple, safe and easily controlled operation. One of the undoubted advantages of the continuous digestor is its speed of operation, which is brought about by the creation of a thin film of hot digest moving over the surface of the helix as it rotates.The benefit of this phenomenon was recognised by Technicon at an early stage in the development of the digestor module.* Equipment manufacturers appear to have lost sight of this advantage in recent years. A number of companies, including Technicon, now market block digestors, but unfortunately, none is amenable to use as part of a continuous flow system. One manufacturer of block digestors has recognised the benefit of heating a thin film and incorporates provision for mechanically agitating the digestion tube. The system in its present form has several limitations, all of which are the subject of current or planned further development work. Because the system has been optimised to cope with a wide spectrum of sample types, it is fed with an excess of acid in the digestion process. This slightly increases the running costs of the system, but perhaps more importantly it increases the levels of the blanks. A compensating advantage is that blanks are very consistent and can be determined precisely. If the system were to be used with large numbers of samples of similar composition, then the acid consumption could be reduced to a level sufficient for the type of substrate under analysis. The system will not cope with all types of sample. Two examples have already been discussed: samples rich in sucrose froth vigorously, but this is not an insuperable problem, and samples that contain volatile lipids are not completely digested; the latter problem may be solved by modification of the digestion conditions. In the present system the sample preparation stage is not automated, which is largely a reflection of the wide range of sample types that this laboratory is required to examine. Considerable further saving of staff time could be achieved if sample preparation were automated and it is hoped to examine this possibility soon. An appropriately modified Technicon AutoAnalyzer SPS-I1 SOLIDprep sampler would be suitable for this purpose. Current development work with the system is aimed at increasing the range of metals that can be determined and improving the detection limits. References 1. 2. 3. 4. Jackson, C . J., Morley, F., and Porter, D. G., Lab. Pract., 1975, 24, 23. Lidzey, R. G., Jackson, C. J., and Porter, D. G., Lab. Pract., 1977, 26, 400. Porter, D. G., Jackson, C. J., and Bunting, W., Lab. Pract., 1974, 23, 111. Marten, J. F., and Ferrari, A., “Proceedings of the 1963 Technicon Symposium on Automated Received September 16th, 1977 Accepted October 31st, 1977 Analytical Chemistry,” Technicon Instruments Co., Chertsey, 1963, pp. 20-28.

ISSN:0003-2654    

DOI:10.1039/AN9780300317

出版商:RSC    

年代:1978

数据来源: RSC

摘要:

332 Analyst, A@& 1978, Vol. 103, 99. 332-340 Polarographic Studies on Some Organic Compounds of Arsenic Part IV." Diphenylarsinic Acid A. Watson Department of Chemistry, The New University of Ulster, Coleraine, Co. Londonderry, BT52 1SA , Northern Ireland A study has been made of the polarographic behaviour of diphenylarsinic acid. It gives rise to a single cathodic wave below pH 6, displaying an adsorption pre-wave, and some limited inhibition effects. The latter are removed by addition of a surface-active agent, thus yielding a well formed, diff usion-controlled wave, the height of which is proportional to concentration (up to 1 x The current - potential relation- ships and the reduction mechanism are discussed. An irreversible reduction to tetraphenyldiarsine has been found.The use of polarography is proposed for the quantitative determination of diphenylarsinic acid. In mixtures with phenylarsonic acid and/or triphenylarsine oxide, the wave at pH 1 gives the total concentration of all three, while that at pH 5.3 gives the concentration of diphenyl and triphenyl species and that at pH a7 triphenylarsine oxide alone. M) and independent of pH. Keywords : Diphenylarsinic acid ; polarography The organic compounds of arsenic have many important applications in agriculture and industry, and as laboratory reagents. They display an interesting, varied polarographic behaviour, thus permitting specific determination according to oxidation state and number of organic substituents. This property is valuable as the compounds vary considerably in toxicity.Previous papers in this series have been studies of phenylarsonic acid,l phenyl arsenoxide2 and triphenylarsine oxide.3 A detailed study of diphenylarsinic acid is necessary to complete the investigation of this range of compounds. Diphenylarsinic acids are formed as by-products4 in the preparation of phenylarsonic acids, some of which are used as additives in animal fee ding stuff^.^ Previously, determina- tion of the arsinic acid impurity had proved very difficult. Diphenylarsinic acid is a stable, simple product easily obtained by hydrolysis and oxidation of unstable irritants, such as diphenylcyanoarsine and diphenylchloroarsine, and is, therefore, useful in their de termina- tion. The latter compounds, known as Clarke I and 11, have military applications.6 For these reasons, the following study of the polarographic behaviour of diphenylarsinic acid was undertaken.Experimental Apparatus with the Radiometer DLTl drop-life timer (Copenhagen). was used in combination with the saturated calomel electrode. apparatus were described in Part 1.l Current - potential curves were recorded by using the Polariter PO4 polarograph equipped The dropping-mercury electrode Details of the Instantaneous current veYsZts time curves were recorded oscillographically. Reagents Diphenylarsinic acid was obtained from Kodak Ltd. and was checked for purity by thin- layer chromatography and elemental analysis. Further purification of the reagent was not necessary. M were prepared in distilled water and were found to be stable for several weeks. Stock solutions of 5 x * For details of Part I11 of this series, see reference list, p.340.WATSON 333 The supporting electrolytes, 0.1 M hydrochloric acid and buffers of phosphate, acetate and borate were prepared from AnalaR-grade reagents; details were given in Part 1.l Experimental Techniques These were described in detail in Part 1.l Results and Discussion The polarographic behaviour of solutions of diphenylarsinic acid in 0.1 M hydrochloric acid was examined over the concentration range 2.5 x to 1 x 1 0 M 3 ~ . One main cathodic wave was found at -0.8 V, which displayed a maximum and was preceded by an adsorption pre-wave (Fig. 1). The potential of the main wave lay at a similar potential to the waves given by phenylarsonic acid1 and triphenylarsine oxide.3 6.0 5.0 4.0 21 --.9 c 3.0 L 3 0 2.0 1 .o I I I I I I I I I 0.3 0.4 0.5 0.6 0.7 0.8 0.9 1.0 1.1 1.2 Potential/-V Fig. 1. Current - potential curves for A, 0.0003 M and B, 0.0005 M solutions of diphenylarsinic acid in 0.1 M hydrochloric acid. The adsorption pre-wave has been shown to correspond to the classical adsorption pre- wave as described by BrdiEka.7 The height of the pre-wave becomes limited at about 1 x 1 0 - 4 ~ , and above this concentration the remaining current is carried by the main diff usion-controlled wave. Instantaneous current - time curves were recorded a t -0.71 V on the upper plateau of the pre-wave for a series of concentrations from 1 x to 1 x 1 0 - 3 ~ and were found to possess the same typical shape predicted by BrdiCka,7 that is, a sharp increase to a maxi- mum near the birth of the drop followed by slower decrease in the current [Fig.2 ( a ) ] . The graphs of the logarithm of the current veYsZts the logarithm of the time were found to be strongly curved near to the beginning of the drop life time. The slope measured over the last 2 s of the drop life time, corresponding to the exponent x in the instantaneous current - time relationship i = k t x , was found to lie close to the theoretical value of -0.33 that was calculated by BrdiEka' for an adsorption process. The relationship between the height of the pre-wave and the height of the mercury column was determined by linear regression analysis of the wave height veysZts the height and the square root of the height of the column (these being the independent variables) for several concentrations between 1 x and 1 x M of diphenylarsinic acid in 0.1 M hydro- chloric acid.The high positive values of the correlation coefficients indicated that the wave was not a kinetic or a catalytic hydrogen wave, while the higher coefficients against the height of the column suggested adsorption control. However, correlation coefficients alone do not offer sufficient proof and so the intercepts of the regression line on both the axis for the height and for the square root of the height were compared with the intercepts334 WATSON : POLAROGRAPHIC STUD~ES ON SOME Analyst, Vol. 103 predicted1 for adsorption and diffusion control and the set of eight heights of the mercury column used. The predicted intercepts for adsorption control fell within the tolerance range of the experimentally obtained intercepts while those for diffusion control did not.al 3 L 0 0.5 0 Time (2 s per division) Fig. 2 . Instantaneous current - time curves for a 0.0002 M solution of diphenylarsinic acid in 0.1 M hydrochloric acid: ( a ) , a t -0.710 V; ( b ) , a t -0.99OV; and in the presence of 0.01% of Triton X-100 at (c), -0.99OV and (d), -0.750 V. Finally, the pre-wave showed the characteristic decrease in height with increasing tempera- ture or on the addition of surface-active agents. The pre-wave can thus be seen to display all the characteristics of the classical adsorption-controlled wave. Linear regression analysis showed that a proportional relationship exists between the combined heights of the main wave and pre-wave and the concentration of diphenylarsinic acid.In the absence of a surface-active agent the standard error of estimate of the line ( 5 4 % of the average wave height) is higher than would be expected. Higher correlation coefficients were found om linear regression analysis of the combined wave height versus the square root of the height of the mercury column than were found versus the height of the column, indicating diffusion control. The predicted intercepts on the axis of the height and the square root of the height of the column for diffusion control fell within the tolerance interval of the experimentally obtained intercepts, while those for adsorption control did not. Instantaneous current - time curves on the rising part and upper plateau of the main wave showed the rising curve characteristic of diffusion control, with a distinct maximum superimposed near the beginning of the drop life time indicating the presence of inhibition effects [Fig.2 ( b ) ] . The small, sharp discontinuity in current a t much more negative potentials (Fig. 1) is also characteristic of a limited degree of inhibition. The presence of this inhibition on the reduction explains the poor reproducibility in the wave height. The behaviour of diphenylarsinic acid in this respect is intermediate between monophenylarsonic acid, which shows no inhibition effects, and triphenylarsine oxide,3 which shows strong, complex inhibition effects as a result of adsorbed species. The effect of a surface-active agent was investigated by recording the current - potential curve for 2 x 10-4 M diphenylarsinic acid in 0.1 M hydrochloric acid in the presence of an increasing concentration of Triton X-100.The small discontinuity moved rapidly to more positive potentials to merge with the main wave while the pre-wave and maximum were suppressed. By a concentration of 0.005~0 of Triton X-100 a single, well formed wave resulted (Fig. 3). Above this concentration no further change was detectable. The half- wave potential and the over-all height underwent no major change, indicating that there had been no change in the basic reduction process.April, 1978 ORGANIC COMPOUNDS OF ARSENIC. PART IV 335 In the presence of O.Olyo of Triton X-100 good proportionality has been found between the wave height and the concentration of diphenylarsinic acid, with much improved standard errors of estimate of the graph (2-3y’ of the average wave height).Calibration graphs 1 .a 4 t 1.2 z 3 3 0 0.6 I I I 0.6 0.8 1.0 1.2 o i ’ Potential/-V Fig. 3. Current - potential curve for a 0,0002 M solution of diphenyl- arsinic acid in 0.1 M hydrochloric acid containing 0.01% of Triton X-100. were prepared from several solid samples of diphenylarsinic acid. Within the limits of experimental error the slopes for all of the samples are equal, confirming that solid diphenyl- arsinic acid yields a reproducible concentration of the electroactive form (Table I). TABLE I CALIBRATION GRAPHS (TEN POINTS, 0.1-1.0 mmoll-1) IN HYDROCHLORIC ACID CONTAINING O.Olyo TRITON X-100 FROM SIX SOLID SAMPLES OF DIPHENYLARSINIC ACID Slope/pA per mmol 1-1 - Sample Slope Tolerance 1 8.01 0.27 2 8.15 0.17 3 8.31 0.20 4 8.24 0.24 5 8.08 0.29 6 8.29 0.18 Intercept on the wave height axislpA & Intercept Tolerance 0.04 0.18 0.01 0.10 0.06 0.12 -0.02 0.16 0.02 0.18 -0.03 0.11 Intercept on the concentration axis/ mmoll-1 & Intercept Tolerance -0.005 0.022 -0.001 0.012 -0.004 0.014 0.002 0.018 -0.006 0.002 0.003 0.013 Standard error of estimate/ 0.11 0.07 0.08 0.10 0.12 0.07 PA Mean value of the wave height/pA 4.41 4.48 4.55 4.52 4.47 4.57 Re-investigation of the relationship between the height of the wave and the height of the mercury column in the presence of O .O l ~ o of Triton X-100 showed an increase in all of the correlation coefficients, while the predicted intervals for diffusion control agreed more closely with those determined experimentally.Instantaneous current - time curves on the upper plateau and on the rising part of the wave show, in the presence of 0.01% of Triton X-100, the rising curve typical of diffusion control, without the maxima which previously indicated inhibition effects [Fig. 2 (c)]. Graphs of the logarithm of the current veysus the logarithm of the time were found to be curved near the beginning of the drop lifetime. The slope, measured over the last 2 s of the drop life- time, corresponding to the exponent x in the instantaneous current - time relationship i = At”, was found to decrease with increasing potential until it reached the theoretical value 0.19 for a diffusion-controlled process,* at the upper plateau of the wave (Table 11).The increase in the exponent towards the base of the wave is characteristic of irreversible processes.8 At potentials on the very lowest part of the wave maxima in the instantaneous current - time curves showed that inhibition effects still persisted, which increasing the concentration of Triton X-100 did not eliminate [Fig. 2 (41.336 WATSON : POLAROGRAPHIC STUDIES ON SOME Analyst, Vol. 103 Because of the greater simplicity of behaviour, increased general precision and the near elimination of inhibition effects in the presence of a surface-active agent, the remaining work described in this paper was carried out in the presence of O.Olyo of Triton X-100. TABLE I1 EXPONENT X I N THE INSTANTANEOUS CURRENT - TIME RELATIONSHIP i = kt' FOR FIVE SOLUTIONS OF DIPHENYLARSINIC ACID I N 0.1 M HYDROCHLORIC ACID AND O.Olyo TRITON X-100 ConcentrationlM Potential/V 0.000 1 0.0002 0.0005 0.0007 -0.825 0.49 0.53 0.51 0.46 - 0.850 0.45 0.46 0.47 0.47 -0.875 0.38 0.41 0.43 0.43 - 0.900 0.35 0.36 0.37 0.35 -0.925 0.33 0.34 0.35 0.34 - 0.950 0.32 0.29 0.31 0.31 - 0.975 0.29 0.24 0.29 0.25 - 1.000* 0.24 0.19 0.22 0.16 - 1.050* 0.23 0.22 0.21 0.22 - 1.100* 0.18 0.22 0.19 0.18 - 1.150* 0.21 0.20 0.26 0.23 1 0.001 0.52 0.44 0.39 0.38 0.33 0.30 0.26 0.24 0.22 0.21 0.20 * Upper plateau.Current - Potential Relationship and the Effect of pH Current - potential curves were recorded in a variety of buffers over the pH range 1-13 at intervals of 0.5 unit. From pH 1 the height of the wave remains constant until about pH 6 in the vicinity of the pK, value of diphenylarsinic acid, when the height falls away rapidly until, at higher pH values, diphenylarsinic acid as an anion is no longer electroactive.This behaviour contrasts with triphenylarsine oxide, which without any acidic protons is electroactive throughout the pH range, and with the more strongly acidic phenylarsonic acid, which is electroactive only below pH 3. At some values between pH 3 and 5 the wave due to diphenylarsinic acid may be partially obscured by the dissolution of the hydrogen ion. The wave moves to more negative potentials with increasing pH and below pH 6 the half-wave potential displays an almost linear relationship with pH without any breaks and with a slope suggesting an irreversible reduction. The reversibility of the electrode process was investigated by logarithmic analysis of the shape of the wave.Five functions of the current were plotted against the electrode potential on the rising part of the wave, corresponding to six possible current - potential relationships and six different types of process, for a series of eleven concentrations of diphenylarsinic acid in 0.1 M hydrochloric acid containing 0.01% of Triton X-100. The functions of current used were log [ (&--i)/i], corresponding to most irreversible and simple reversible reduc- t i o n ~ , ~ log(id-i) for reversible reduction to an insoluble product,lO log [ (id-i)'/i], corre- sponding to reversible reduction followed by reversible dimeri~ation,~ log [ (id-i)/i2] for the reversible reduction of a dimerg and, finally, log [ i d ( & ~ - i ) ~ / i ~ ] , corresponding to a reversible reduction followed by an irreversible dirneri~ation.~ In these functions, i d is the limiting diffusion current and i the current at each potential. The potential was chosen as the independent variable in linear regression analysis.As inhibition and adsorption processes were shown, by the instantaneous current - time curves, to remain unsuppressed at the base of the wave, and as these processes would alter the current - potential relationship, the logarithmic analysis was confined to the larger, upper part of the wave where the instantaneous current - time curves indicated simple diffusion control in the absence of inhibition effects.The current was recorded during the last 15% of the drop life time without damping. The reciprocal value of the slope of these plots, corresponding to the value 2.303RT/a%F in the current - potential relationship, is not in any instance a simple fraction of 0.059 V, indicating that the process is irreversible (Table 111).April, 1978 ORGANIC COMPOUNDS OF ARSENIC. PART IV TABLE I11 PREDICTED VALUE OF 2.303RTlanF (V) FROM REGRESSION COEFFICIENTS OF SEVERAL FUNCTIONS OF CURRENT verszIs POTENTIAL AT ELEVEN CONCENTRATIONS (0.05-1.0 mmol l-l) OF DIPHENYLARSINIC ACID IN 0.1 M HYDROCHLORIC ACID 337 Function of current r A > i d -2 Log(id -2) Log- i d -2 i d ( i d - 2) 22 Log- 2 . 303RT {Msean . . . . 0.089 0.118 0.074 0.052 0.039 anF . . . . 0.006 0.006 0.003 0.003 0.002 0.79 1.26 1.79 2.38 Mean correlation coefficient versus potential .. . . 0.99974 0.991 81 0.997 24 0.997 84 0.999 11 The correlation coefficients are consistently higher for the graphs of log [ (id-i)/i] versus potential, suggesting that this function of the current has the most linear relationship with potential. For each of these functions, and the corresponding current - potential relationships, there exist equivalent relationships between half-wave potential and the logarithm of the concentration of the depolari~er.~?~~ For processes in which log [ ( i d - i ) / i ] is linearly dependent on potential, the half-wave potential is independent of the concentration. For those in which log[(id-i)/i2] is linearly dependent on potential, the half-wave potential should shift to more negative potentials at a rate of 2.303RTlanF (V) for an increase in concentration of one decade, while for the remaining processes the shift is to more positive potentials at an equal rate. A correlation coefficient of 0.1905 for ten concentrations from 1 x 104 to 1 x M, well below the critical value, shows the half-wave potential to be independent of concentration, again indicating a linear dependence of log [ (id-i)/i)] on potential.Finally, the predicted value of 2.303RT/anF, calculated from the reciprocal of the slope of the graph for each function, can be compared with the value obtained by use of an independent method. To this end, the relationship between half-wave potential and pH was examined in greater detail in steps of 0.2 pH unit in the range pH 1-2.4 for several concentra- tions of diphenylarsinic acid in a mixture of hydrochloric acid and potassium chloride. Good linear relationships were found, of which the slope, dE+/d(pH) = -0.093 & 0.005 V, should equal the product of the number of protons involved in the reduction prior to the potential determining step and the value of 2.303RTlanF. This value was found to be an integral multiple only of the reciprocal of the slope, 0.089 O.O06V, of the graphs of bg[(id-i)/i] versus potential, confirming the h e a r dependence of this function on potential, and as the integral was unity (Table 111) one protonation in the reduction prior to the potential determining step was indicated.The value of 2.303RTlanF = 0.09 V corresponds to a value of an = 0.66, a value greater than the value (0.49) for phenylarsonic acid,l while the value for triphenylarsine oxide3 tends to a value of 2. The value of 0.65 and the linear dependence of log[(;d-i)/i] on potential indicate a typical irreversible process obeying the relationship However, the correlation coefficients alone do not offer sufficient proof.2.303RT ( i d - i ) 2.303RT ccnF log- -~ x PH 2 anF E = Ef,pH=O +------- As diphenylarsinic acid (I) is electroactive only a t pH values below the pKl value, and as no breaks occur in the half-wave potential versus pH graphs, only one electroactive form is indicated, that being the undissociated acid. One protonation occurs prior to the potential determining step and if, as is generally true with most irreversible processes, this step is the addition of the first electron, then it can be suggested that the potential is deter- mined by the following process :338 WATSON : POLAROGRAPHIC STUDIES ON SOME Anaiyst, Vol.103 I Reduction Mechanism A study of relationships involving potential yields information only about the initial potential-determining step. Elucidation of the total reduction mechanism requires direct determination of the total number of electrons per molecule consumed in the reduction. For diphenylarsinic acid this was carried out by use of microcoulometry, that is, by means of prolonged electrolysis on a small volume (1 m1) at the potential of the upper plateau of the wave. From the decrease in the concentration and the current passed the number of electrons per molecule was calculated for four concentrations of diphenylarsinic acid in 0.1 M hydrochloric acid and was found to be 2.95 & 0.16.The potential determining step proposed above yields the diphenyldihydroxyarsine radical, 11. Rapid addition of a second electron - proton pair will form the hydroxide monomeric form, 111, of diphenylarsine oxide. The literature2J1 suggests that this hydroxide form can be reduced at more positive potentials (about -0.1 V) while the microcoulometric result shows that it is not the final product. Under these conditions formation of the less easily reduced dimer is unlikely. Phenylarsonic acid1 and triphenylarsine oxide3 have both been shown to be reduced initially to the arsine. The highly reactive pftenylarsine was then found to combine with the intermediate product phenyl arsenoxide, by interaction and loss of the hydride and 1 Ill IV As--H +-April, 19'18 ORGANIC COMPOUNDS OF ARSENIC.PART IV 339 hydroxy groups, to yield the insoluble oligomer arsenobenzene, containing arsenic-arsenic bonds.1 With no hydride groups triphenylarsine cannot react in this fashion and remains the final product of the reduction of triphenylarsine oxide.3 Diphenylarsine, IV, with one hydride group, is more similar to phenylarsine and is very reactive. The equivalent reaction of diarylarsines with diarylarsine oxides has been reported12 for the preparation of tetra- aryldiarsines such as that illustrated (V). A reduction path analogous to that determined €or phenylarsonic acid could therefore be expected for diphenylarsinic acid.Such a reduction mechanism requires six electrons in order to reduce two molecules of diphenylarsinic acid, that is, three electrons per molecule. The experimentally obtained value of 2.95 & 0.16 electrons per molecule confirms that this is the reduction mechanism which occurs. Further support for this belief comes from the early work of Fichter and Elkind,13 who in 1916 prepared tetramethyldiarsine by the electroreduction of dimethylarsinic acid at a large mercury pool. Note that the final product depends on the solvent. Dessy et d.ll have shown that dimeric diphenylarsine oxide in anhydrous ethylene glycol dimethyl ether (glyme) is reduced to [(C,H,),As]- and [(C,H,),AsO]-. These species could not remain stable in aqueous acid and would protonate to yield the species combining in the final reaction shown above.Dessy et aZ.ll also showed that other diphenylarsenic(II1) species in anhydrous glyme are reduced to tetra- phenyldiarsine, V. This undergoes further reduction to [ (C,H,),As]- only at the most negative potentials (-2.8 V) by virtue of the solubility of the tetraphenyldiarsine and the stabilisation of the anion product in the anhydrous solvent. In aqueous conditions the insoluble tetra- phenyldiarsine must remain the final product. Analytical Applications Diphenylarsinic acid gives rise to a single, well defined, diffusion-controlled wave in 0.1 M hydrochloric acid containing O.Olyo of Triton X-100. The wave height is reproducible, proportional to concentration in the range 2 x 10-5-10-3 M and is independent of pH.These are suitable conditions for analytical use. In analyses of known solutions of diphenylarsinic acid by use of the standard additions technique with five additions of standard solution, errors of about 2y0 were obtained. Chloride, sulphate, nitrate, phosphate and acetate were found not to interfere. The polarographic analysis of diphenylarsinic acid in hydrochloric acid lends itself to the determination of irritants, such as diphenylcyanoarsine or diphenylchloroarsine. Hydro- lysis and oxidation to diphenylarsinic acid yields a single electroactive species with simple reproducible polarographic behaviour, while hydrolysis alone yields diphenylarsine oxide, the behaviour of which is complicated by its existing both as a monomeric hydroxide and as an oxide dimer.Direct polarographic determination of the cyanoarsine is also much less convenient as it requires strictly anhydrous solvents. In 0.1 M hydrochloric acid, the waves of diphenylarsinic acid and phenylarsonic acid are not resolved but give a combined wave height. By increasing the pH above 3 the phenyl- arsonic acid becomes electroinactive. The practical working range for diphenylarsinic acid above pH 3 is considerably narrowed by partial overlap with decomposition of the hydrogen ion and the need not to work too close to the pK value. By working at pH 5.3 in phosphate buffer containing 0.01 yo of Triton X-100 specific determination of diphenylarsinic acid is possible in up to a 100-fold excess of phenylarsonic acid.Accuracies of about 3% have been obtained by using the standard additions technique with five additions. The phenylarsonic acid content is obtained from the combined wave height in solution in 0.1 M hydrochloric acid. Similarly, in 0.1 M hydrochloric acid or at pH 5.3 resolution from the wave of triphenyl- arsine oxide is not possible. Thus, by choice of a suitable pH, specific determination of phenylarsonic acid, diphenylarsinic acid and triphenylarsine oxide in mixtures is possible. Inorganic arsenic( V) is electroinactive while the tetraphenylarsonium ion is reduced at much more negative potentials.14 Arsenic(II1) species are reduced at more positive potentials. Polarography is thus of value for the quantitative speciation of organoarsenic samples. At pH 2 7 only the latter species is active.340 WATSON 1. 2. 3. 4. 5. 6. 7. 8. 9. 10. 11. 12. 13. 14. References Watson, A., and Svehla, G., Analyst, 1975, 100, 489. Watson, A., and Svehla, G., Analyst, 1975, 100, 573. Watson, A., and Svehla, G., Analyst, 1975, 100, 584. Freedman, L. D., and Doak, G. O., J . Am. Chem. Soc., 1951, 73, 5656. Malawyandi, M., MacDonald, S. A., and Barette, J. P., J . Agric. Fd Chem., 1969, 17, 51. “McGraw-Hill Encyclopedia of Science and Technology,” Volume 3, McGraw-Hill, New York and London, 1966, entry “Chemical Warfare,” p. 44. BrdiCka, R., Colln Czech. Chem. Commun., 1947, 12, 522. Zuman, P., “Elucidation of Organic Electrode Reactions,” Academic Press, New York and London, 1969, p. 16. Perrin, C. L., in Cohen, S. G., Streitwieser, A., and Taft, R. W., Editors, “Progress in Physical Organic Chemistry,” Volume 3, John Wiley, New York and London, 1965, pp. 177-184. Delahay, P., “New Instrumental Methods in Electrochemistry,” Interscience Publishers, New York, 1954, p. 57. Dessy, R. E., Chivers, T., and Kitching, W., J . Am. Chem. SOC., 1966, 88, 467. Blicke, F. F., and Webster, G. L., J . Am. Chem. SOL, 1937, 59, 537. Fichter, F., and Elkind, E., Berichte, 1916, 49, 239. Horner, L., Rottger, F., and Fuchs, H., Chem. Ber., 1963, 96, 3141. NOTE-References 1-3 are to Parts 1-111 of this series, respectively. Received Augwst 25th, 1977 Accepted October 12th, 1977

ISSN:0003-2654    

DOI:10.1039/AN9780300332

出版商:RSC    

年代:1978

数据来源: RSC

摘要:

Analyst, April, 1978, Vol. 103, pp. 341-345 341 Determination of Lead in Atmospheric Particulates Using an Automated Atomic-absorption Spectrop hotometric System with Electrothermal Atomisation C. J. Pickford and G. Rossi Commission of the European Communities, Euratom C.C.R., Ispra Establishment, Chemistry Division, 21020 Ispra (Varese), Italy A laboratory constructed automatic sampling system is used for the deter- mination by atomic absorption spectrophotometry with electrothermal atomisation of up to 50 air particulate samples, after acid dissolution. When used on a routine basis, this method results in a high precision and permits a large number of samples to be analysed in a minimum time without operator attendance. Keywords : Lead determination; atmospheric particulates ; atomic-absovption spectrophotometry with electrothermal atomisation ; automated sampling Trace amounts of heavy metals in air are often monitored by using atomic-absorption spectrophotometry with electrothermal atomisation, as this method possesses the extreme sensitivity that is required without the need for time-consuming pre-concentration methods, is reasonably rapid and does not require excessively costly instrumentation.However, the technique is prone to matrix interferences1*2 and requires a higher degree of operator skill and attention than does, for example, conventional flame atomic-absorption spectrophoto- metry. In general, the methods of collecting particulate material from air can be divided into those in which it is collected directly in a carbon cup, which forms part of the atomisation ~ystem,~p* and those in which a conventional filter is used that is subsequently dissolved in, or treated with, acids in order to liberate the trace elements from the matrix of silica, etc., that may be p r e ~ e n t .~ ? ~ In this laboratory, many samples of this nature are routinely analysed for a number of trace elements and, because of the attention and long periods of concentration required, the analysis may become extremely tedious. In order to cope with large numbers of samples in flame atomic-absorption spectrophotometry, manufacturers have often coupled their instruments to automatic sample changers and read-out devices. Clearly, this is only justified when the time saved in this way is greater than the time required to wash and fill the sample cups and load them into the sampler.Such systems are suitable for the analysis of a large number of samples of a similar nature and within a limited concentration range. However, few systems have been described for automatic sampling in atomic-absorption spectrophotometry with electrothermal atomisation.7-ll This may stem partly from the fact that, unlike flame atomic-absorption spectrophotometry, this technique is essentially a discontinuous one (in its usual commercial furnace or tube form) and it is difficult to manipulate the typically 10-pl volumes of sample required, automatically and without cross-contamination. In order to sample flowing coolant or effluent streams, we developed7y8 an automated sampling system using a linear sliding sampling valve to remove a fixed pre-selected volume of sample from the coolant loop, or any other stream of liquid, and an injection mechanism to transfer the liquid into the heated graphite atomiser.In addition to the improved convenience, there was also a marked increase in precision compared with manual injection, resulting very often in a repeatability between injections of better than 1%. This original system has been modified by the addition of a sample changer so as to be able to run a large number of samples successively and without operator attendance being required. The system has the advantages, compared with existing commercial systems, that it can readily be constructed in the average laboratory, and that it can be used with almost any atomic-absorption spectrophotometric system with electrothermal atomisation.342 PICKFORD AND ROSSI: DETERMINATION OF LEAD I N ATMOSPHERIC Analyst, Vd.103 The determination of lead in particulate matter in air is described as a typica1:example of an application where the use of an automated system is particularly advantageous. The samples represent a chosen fraction from an analytical survey carried out by the European Commission in Turin and Sardinia, and were also analysed by X-ray fluorescence spectro- metry and flame atomic-absorption spectrophotometry, when possible. Experimental Apparatus The apparatus is shown schematically in Fig. 1, except for the hollow-cathode and photo- multiplier power supplies, which are omitted. Fig. 2 is a photograph of the sampling system and the table-mounted dual-channel optical system; full details of the latter can be found elsewhere .12 L2 +-t- Fig.1. Schematic diagram of apparatus. Prog. = electromech- anical programmer (Crouzet, France) ; Pump = peristaltic pump (Ismatec, Switzerland) ; SV = sampling valve (laboratory built, see Ref. 7 ) ; ASC = automatic sample changer (Model 51, Varian, Aus- tralia) ; Dig. = digital printer (Systron Donner) + digital voltmeter (Keithley) ; Rec. = potentiometric chart recorder (Hewlett-Packard, USA) ; HGA = HGA-74 heated graphite atomiser (Perkin-Elmer) ; Inj. = pneumatic sample injector (see text); Amp. = two-channel logarithmic amplifier - integrator; BS = beam splitter (Perkin-Elmer) ; L,, L, = lenses; MI, M, = monochromators (Jarrell Ash); MC = mirror chopper (Perkin-Elmer) ; and HC,, HC, = hollow-cathode lamps (or one hollow-cathode and one deuterium lamp).Operation The operating sequence of the instrument is controlled by an electromechanical pro- grammer consisting of a rotating drum, driven by a synchronous motor. The drum is fitted with movable open - close contacts, which operate a bank of 11 microswitches. At the end of a cycle, the drum rapidly rotates back to its initial position, and re-starts. After a new tray of samples has been inserted, and the Perspex cover of the sample changer lowered into position, the programme is initiated. The sampler tube (PTFE) dips into the first sample (usually distilled water) and the peristaltic pump is started. The first 100 pl of sample are not used, in order to wash completely the PTFE tubing between the sampler and the sampling valve.Then, the electrovalve controlling the pneumatic operation of both the valve and the sampling injector is opened. Tkis transfers a volume of sampleFig. 2. Photograph of sampling system and dual-channel optical systems.April, 1978 PARTICULATES BY AUTOMATED ATOMIC-ABSORPTION SPECTROPHOTOMETRY 343 (usually lop1 was used) to the injection loop, and lowers the injection tip (a disposable Eppendorf pipette tip) to within 1 mm of the far wall of the graphite tube. The sample is blown into the tube by a second channel of the peristaltic pump. The tip is then raised, and the furnace programme starts. The recorder writes only during the final part of the HGA-74 cycle, i.e., during atomisation.The amplifier, which has been described more fully elsewhere,12 is a dual-channel a.c. ratioing instrument, designed to provide background correction up to, and beyond, 1.5 absorbance units, even when the intensities of the two beams may be very different. This is essential if a deuterium lamp is used, as otherwise (as in most commercial instruments where the background-correction and measurement-channel intensities must be similar) the intensity of the hollow-cathode lamp must be greatly attenuated when spectral regions are used in which the output of the deuterium lamp is low. This results in a considerable deterioration of the signal to noise ratio. The amplifier output is in absorbance units or integrated absorbance units. The print and re-set commands for the integrator - printer are issued at the end of the atomisation cycle.To avoid possible memory effects, each sample is injected four times; the first measure- ment is not recorded, and serves to flush the sampling valve and the 25-cm length of PTFE tubing between the valve and the injection tip. At the end of the fourth measurement, the zero base line is re-set automatically, the sampler advances one step and the cycle is repeated. After the last sample, the programmer is stopped and an alarm sounds. If more elements are required to be determined, the two monochromator wavelengths are changed and then the programmer is re-started, thus running through the samples in this way. The time required to analyse 50 samples for one element (three recorded measurements) is 3 h.Lead absorption was measured at the 283.3-nm line; the line used for background correction (simultaneous) was the lead 280.3-nm line. Reagents The lead solution used was a concentrated standard (Merck) diluted to 1000pgml-l with distilled water. Neither the water used nor the concentrated acids showed any significant blank. Preparation of Solutions The samples as supplied consisted of quartered 3-cm Millipore discs collected at various locations, bearing an atmospheric particulate deposit, the lead content of which had been requested. The filters were treated according to the following standard method. A 5-ml volume of 50% nitric acid was added to the filter in a 50-ml PTFE beaker and the beaker was covered and heated at 100 "C for 30 min.The contents were evaporated almost to dryness to destroy organic matter and the residue was re-dissolved in several drops of a mixture of concentrated nitric and perchloric acids. After subsequent evaporation to dryness, the white residue was treated with 1 ml of hydrofluoric acid and heated to fumes of perchloric acid. When reduced to a very small volume (about 20 pl), the residue was re-dissolved in 2 ml of 1 N nitric acid and the solution diluted to 100 ml. Blanks and standards containing the equivalent of 1, 2, 5, 10, 25, 100 and 200 pg of lead per whole filter were also prepared using a 1 000 pg ml-1 stock solution of lead. A known amount of lead was added to an aliquot of one of the dissolved filters, in order to test for any possible matrix effects. The sampler was then loaded with the solutions, starting with distilled water and the five standards, followed by the samples and finally by distilled water.All reagents were of Suprapur grade (Merck, Darmstadt, Germany). Results At the completion of the measurement cycle, the measured peak heights for each standard and sample were averaged and a calibration graph was drawn. This graph showed some curvature at high concentration values. Fig. 3 shows typical calibration peaks obtained in this way, The signal to signal variation of lead standards carried through the acid-treatment procedure was greater than that of simple aqueous standards (usually 0.5-1 .tiyo). Similarly, the variation in signal for the samples was also greater, but was generally in the range 0.5- 2% for high and 2-6% for low concentrations.344 PICKFORD AND ROSSI: DETERMINATION OF LEAD I N ATMOSPHERIC Analyst, Vd.103 Fig. 3. Part of an individual run. Fig. 4 shows typical results for samples together with derived concentration values. The absence of memory effects and the good precision can be clearly seen. The use of a recorder for measurement purposes was found to be the preferred method, as unacceptable peaks (due to, e.g., stray contamination or electrical noise from nearby heavy plant machinery) could be more rapidly identified and then discarded than when a digital display was used. Although signal integration offers advantages from the point of view of increased linearity and a certain immunity to some of the matrix effects that occur in atomic-absorption spectro- photometry with electrothermal atomisation,l at low signal levels the inherent base-line drift of the integrator used makes it necessary for amplitude measurement to be used instead.Direction of sampling ___+ s9 90 pg r-----7 1 S6 17.0 pug I /I /I s10 110 pg - S11 188,ug S12 136pg Fig. 4. the filter. lamp being used. Part of an individual run. The lead values are those for the total lead concentration on The difference in base-line noise compared with Fig. 3 is due to a newer hollow-cathode The signals obtained for one sample with and without the addition of an aqueous standard showed that there was no discernable matrix effect. However, in some preliminary tests it was established that this was not so if the concentrated acids used to dissolve the filters were not evaporated to a very small volume before re-dissolution in dilute acids.Leaving even 0.5 ml of concentrated acid was sufficient to depress the amplitude of the signal to 70-50% of its correct value.April, 1978 PARTICULATES BY AUTOMATED ATOMIC-ABSORPTION SPECTROPHOTOMETRY 345 Conclusion The use of an automated sampling procedure in atomic-absorption spectrophotometry with electrothermal atomisation clearly offers advantages when a large number of similar samples are to be analysed, the concentrations of which fall within a limited range that can be adequately covered by a calibration graph. If there are few samples, or many different types of sample with different concentrations of the element concerned, then clearly there will be little reason to use an automated system.The precision gained with an automated system is clearly worthwhile only if the accuracy of the determination is high; this means that matrix interferences should be avoided or, if this is not possible, recognised and the standard additions technique used for analysis. The system described can, of course, be used for the analysis of many environmental samples, e.g., waters and effluents, and in medical applications, e.g., blood, serum and urine analyses. In addition, the possibility of using a loop sampling technique renders it suitable for certain kinetic application~l~ or in situ medical applications. Also, it can be assembled fairly readily from commercially available components and in its present form has the advantage of being applicable to most atomic-absorption spectrophotometric systems with electrothermal atomisation with a minimum of modification. The use of a commercial pipette tip means that it can readily be changed if it becomes contaminated or worn. 1. 2. 3. 4. 5. 6. 7. 8. 9. 10. 11. 12. 13. References Schramel, P., Analytica Chim. A d a , 1974, 72, 414. Cruz, R. B., and Van Loon, J. C., Analytica Chim. Acta, 1974, 72, 231. Matousek, J. P., and Brodie, K. G., Analyt. Chem., 1973, 45, 1606. Noller, B. N., and Bloom, H., Atmosph. Envir., 1975, 9, 505. Omang, S. H., Analytica Chim. Acta, 1971, 55, 439. Begnoche, B. C., and Kisby, T. H., Analyt. Claem., 1975, 47, 1041. Pickford, C. J., and Rossi, G., Analyst, 1972, 97, 647. Pickford, C. J , , and Rossi, G., Analyst, 1973, 98, 329. Mislan, J. P., and Elchuk, S., “Advances in Automated Analysis,” Volume 11, Technicon Inter- Dahl, R., and Stoeppler, M., Berichte der Kernforschungsanlage, Julich, No. 1254, 1975. Welz, B., Paper presented a t the Vth International Conference on Atomic Spectroscopy, Melbourne, Pickford, C. J., and Rossi, G., Atom. Absorption Newsl., 1975, 14, 78. Cavalli, P., Pickford, C. J., and Rossi, G., t o be published. national Congress 1969, Mediad Inc., New York, 1970, p. 315. Australia, August 25th-29th, 1975. Received October 19th, 1977 Accepted November 14th, 1977

ISSN:0003-2654    

DOI:10.1039/AN9780300341

出版商:RSC    

年代:1978

数据来源: RSC

摘要:

346 Analyst, April, 1978, Vol. 103, $9. 346-353 An a I ys is of Steroids Part XXXI." Corticosteroidst Mechanism of the Yetrazolium Reaction of SAndor Gorog and Peter Horviith Chemical Works Gedeon Richter Ltd., Gy6rnro"i ut 21, H-1475 Budapest, Hungary The stoicheiometry and mechanism of the reaction between tetrazolium reagents and the cc-ketol side-chain of corticosteroids was studied. On the basis of literature data plus some experimental results it is concluded that the side-chain is oxidised to the 20-keto-2 l-aldehyde, which then partly under- goes an intramolecular Cannizzaro reaction, leading to the 20-hydroxy-2 1- carboxylic acid. The doubled sensitivity of 16-hydroxycorticosteroids towards blue tetrazolium can be explained by a mechanism involving D-homo rearrangement of the D ring.Equilibrium studies showed that under the usual experimental conditions the reduced blue tetrazolium reagent exists almost quantitatively in the monoformazan state. Keywords : Corticosteroid determination ; tetrazolium reaction mechanism ; trianzcinolone oxidation ; blue tetrazoliunz reduction The reduction of tetrazolium compounds by the C-17 side-chain of corticosteroids is one of the classical reactions in steroid ana1ysis.l Following the first applications of this reaction for the determination of corticosteroids described about 25 years more than 100 papers have been published dealing with the determination of this group of steroid hormones in biological fluids and pharmaceutical formulations based on the photometric determination of the coloured formazans produced during the reaction.These methods have been adopted by almost all of the major pharmacopoeias. Some of the above papers described the optimum conditions for the reaction, but most dealt with practical applications only. In the period 1951-66, virtually no attempts were made to describe the exact stoicheiometry and mechanism of the reaction and even the scattered data available were rather contradictory. In the last 10 years, however, two excellent series of papers have been published on this problem. The series by Mohrle and co-~orkers~--~ dealt with the mechanism of the reaction, while that of Graham and co-workerss-l1 described the kinetics of the reaction together with the evaluation of structural and physico-chemical factors that influence the colour develop- ment.As some problems still seem to remain, in this paper we try to answer the following questions : (a) what is the fate of the C-17 side-chain in the reaction, with particular respect to the anomalous behaviour of the 16-hydroxy derivatives? and ( b ) what is the fate of the blue tetrazolium in the reaction? Experimental Reagents All reagents and solvents were of analyt ical-reagent grade. Triphenyltetrazolium chloride (TTC) was obtained from Reanal (Budapest) and blue tetrazolium (BT) [3,3'-(3,3'- dimethoxy-4,4'-biphenylylene) bis(2,5-diphenyl-2H-tetrazolium chloride) ] from Dajac Labora- tories (Philadelphia, Pa.). Procedures The reactions of steroids with TTC were carried out according to the method of Johnson et aZ.12 as adopted by the British Pharmacopoeia 1973,13 while the reactions with BT were run as described in the United States Pharmacopeia XIX.14 * For Part XXX of this series, see J .Chromat., in the press. t Presented at the Fourth SAC Conference, Birmingham, July 17th t o 22nd, 1977.GOROG AND HORVATH 347 In the thin-layer chromatographic investigation of the reaction mixture, the latter was first neutralised with dilute hydrochloric acid, then its volume was reduced by evaporation. The solution was diluted with water, the precipitated formazans were filtered off and the solution was extracted several times with chloroform. The volume of the chloroform extract was reduced by evaporation and the resulting solution was applied to Kieselgel F254 plates. The thin-layer chromatograms were developed with chloroform - propan-2-01 (93 + 7) and the spots rendered visible under short-wavelength ultraviolet light.After elution of the spots with ethanol, the materials obtained were investigated by means of standard spectroscopic techniques. The 20-keto-21-aldehyde group was detected by the method of Gorog and Szepesi.15-17 The attempted determination of formaldehyde by Conway diffusion and chromotropic acid colorimetry was carried out as described by Edwards and Kellie.18 The equilibrium constants of the reaction between prednisolone and BT were determined spectrophotometrically. The reaction was carried out according to USP XIX,14 the only difference being that instead of ethanol a 1 + 1 mixture of ethanol and chloroform was used as the solvent.After allowing the reaction mixture to stand at 25 "C for 90 min in the dark, the absorbance was measured at 525 and 590 nm. Results and Discussion Oxidation of the C-17 Side-chain The stoicheiometry of the reaction between corticosteroids and one tetrazolium unit of the reagent can easily be established when using the monomeric TTC reagent, the reductive opening of which to the corresponding formazan is known to involve two electrons : N-NH // C The molar absorptivity of the triphenyl formazan is 14 50019 or 15 000,20 while the apparent molar absorptivities of various corticosteroids obtained using the method of Johnson and co-workers12v21 and calculated from the data of Garratt22 range between 15 700 and 15 900. The good agreement of the data* means that one molecule of the tetrazolium compound reacts with one molecule of corticosteroid or in other words the oxidation of the side-chain also involves two electrons.As can be seen in equation (2), in principle the two-electron oxidation of the side-chain might take place in two different ways, viz., cleavage of the C-20-C-21 bond to form etianic acid derivatives and formaldehyde, or oxidation at the C-21 position to form a 20-keto-21- aldehyde derivative. An early publication of Z a f f a r ~ n i ~ ~ seems to have supported the first reaction route; he claimed, although without giving any experimental details, to have isolated etianic acids as the reaction products from the reaction mixtures. This possibility, however, can easily be excluded. As can be seen in the reaction equation, an equivalent amount of formaldehyde also ought to be formed in this instance.However, no formaldehyde could be detected from the reaction mixture by Conway diffusion and chromotropic acid colorimetry.18 (Recoveries of over 90% were found when the expected amount of formaldehyde was added to the reaction mixture. * The active methylene group in the vicinity of the a,p-unsaturated ketone group in ring A contributes slightly to the reaction.23 Formaldehyde itself does not react with tetrazolium reagents.23)348 GOROG AND HORVATH : ANALYSIS OF Analyst, "01. 103 COOH "* + CH20 + formazan N COOH D A,. = 351 nm The above simple considerations suggest that the second reaction route, leading to the 20-keto-21-aldehyde. is the more likely.However, the authors of most of the publications dealing with this subject either do not give an opinion on this question or, if the 20-keto-21- aldehyde is described as the reaction product, no experimental evidence is given. The series of papers by Mohrle and co-workers4-7 does not seem to be well known to analytical chemists. Mohrle and co-workers investigated the tetrazolium reaction of corticosteroids and several non-steroidal a-ketol derivatives on the preparative scale, and presented a large amount of experimental evidence to support the formation of the 20-keto-21-aldehyde derivative as the primary reaction product. Our results, derived from the thin-layer chromatographic investigation of the reaction mixture of prednisolone and TTC as described under Experimental (chromatography of the steroidal oxidation product together with authentic prednisolone-21-aldehyde hydrate, and spectroscopic investigation of the material eluted from the chromatogram, including the detection of the 20-keto-21-aldehyde group as the quinoxaline derivative15-17) were in good agreement with Mohrle and co-workers' results.Under the experimental conditions adopted by Mohrle and co-workers, a high proportion of the initially formed 20-keto-21-aldehyde is converted by a base-catalysed intramolecular Cannizzaro reaction into 20-hydroxy-2 1-carboxylic acid derivatives [see equation (2)]. This polar reaction product remains at the start of the chromatogram of the reaction mixture. The rate of this reaction was studied in order to determine the proportion of this secondary reaction product in the reaction mixture. According to an early observation in this l a b ~ r a t o r y , ~ ~ the reaction is almost instantaneous in boiling 0.1 N sodium hydroxide solution and may serve as the basis for the determination of 20-keto-21-aldehydes based on the back-titration of the excess of alkali.At room tempera- ture and much lower alkali concentrations, however, the reaction is much slower and only about l0-20% of the initially formed 20-keto-21-aldehyde was converted into the 20-hydroxy- 21 -carboxylic acid under the recommended pharmacopoeia1 conditions. Even minor changes in the experimental conditions, however, may cause the primary reaction product to be converted to a greater extent into the hydroxy acid. It is important to keep strictlyApril, 1978 STEROIDS.PART XXXI 349 to the experimental conditions of the pharmacopoeial methods if the aim of the investigation is the elucidation of the mechanism of those methods. The oxidation of the a-ketol side- chain of corticosteroids is known to occur by several reaction routes,26 and major deviations from the experimental conditions of the pharmacopoeial methods may lead to entirely different reaction mechanisms. For example, Smoczkiewiczowa and J a s i c ~ a k , ~ ~ who carried out the reaction between prednisolone and TTC at 50 "C and used a concentration of potassium hydroxide solution as high as 0.06 N instead of about 0.01 N tetramethylammonium hydroxide, claimed that etianic acid and the 17-keto derivative were the reaction products.The explanation of the above-mentioned results of Zaff a r ~ n i ~ ~ may be similar. The reaction between BT and triamcinolone (9-fluoro-16a-hydroxypredniso1one) merits special attention. Smith and HalwerB were the first to report that the colour intensity in this instance was 70% higher than that with the 16-deoxy derivatives. From the calibration graphs in the paper of Ascione and F ~ g e l i n , ~ ~ we calculated double values for the molar absorptivities of triamcinolone and its diacetate in comparison with other derivatives, including triamcinolone acetonide, in which the 16- and 17-hydroxyl groups are blocked. Graham et aL1O also claimed that triamcinolone contained two reduction units, although by varying the experimental conditions an even higher reactivity towards BT could also be obtained.ll It is interesting that Johnson et aZ.12 reported that after an initial rapid period the reaction between triamcinolone and TTC became so sluggish that it could not be exploited for analytical purposes.With BT, however, the over-all four-electron reaction takes place rapidly, the reason for which may be the higher oxidation potential of BT (according to Balogh and Lad&nyi,lg the polarographic half-wave potential of the tetrazolium - formazan reaction is +0.27 V for BT and 0.37 V for TTC).* Virtually nothing has been published on the mechanism of this very high reducing capacity. Graham et aZ.10 assumed that the first step of the reaction was the ordinary oxidation of the 20-keto-21-aldehyde, which was then followed by the alkali-catalysed cleavage of glyoxal, leading to the 16a-hydroxy-17-keto derivative, which is an a-ketol and is therefore capable of a two-electron oxidation.This assumption seems to be unlikely to be correct, as we failed to detect glyoxal in the reaction mixture after distillation and attempted reaction with 4,5-dimethyl-0-phenylenediarnine.~~-~~ It is likely that the formation of the second reduction unit is associated with the base- catalysed D-homo rearrangement of the initially formed triamcinolone-21 -aldehyde. The D-homo rearrangement of triamcinolone has been extensively s t ~ d i e d . ~ l - ~ ~ In principle, three reaction routes are possible : CH20H -2e- __3 +BT HC' / I'OH OH OH 1 ,OH pJH (3) The third route does not really come into question and the ratio of the first and second routes It seems likely that in * It is worth mentioning that Smoczkiewiczowa and Jasiczak30 found an increased sensitivity even with is a function of the experimental conditions and the substituents.TTC when the reaction was carried out a t elevated temperature.350 GOROG AND HORVA'TH: ANALYSIS O F Analyst, Vol. 103 this instance the main reaction product is that formed by the second reaction route. This contains a further cc-ketol group and is oxidised by BT to form a diketone, which exists in the enolic form. In addition to this reaction, several others may also occur, such as the other route of rearrangement, D-homo rearrangement of triamcinolone itself prior to the oxidation and, in addition, splitting off formaldehyde from any of the mentioned compounds.31J6 This great variety of possible reaction products may be the reason for the fact that with triamcinolone the stoicheiometry of the reaction is far from being as unequivocal as it is with 16-deoxy derivatives.l0S1l For the same reason in this instance we failed to obtain pure material after thin-layer chromatographic separation.In contrast to the oxidation of prednisolone, in which the oxidation product was less polar than the parent compound (its R, value was higher), in this instance the main reaction products were much more polar than triamcinolone and it was extremely difficult to elute them from the layer. This finding supports the presence of an enolised cc-diketone group in the reaction product.Further evidence for this conclusion is the ultraviolet spectrum of the isolated material, which contains a peak at 260 nm shifted to 300 nm in alkaline media owing to the formation of an enolate. Reduction of Blue Tetrazolium The elucidation of the fate of the monomeric TTC during the reaction with corticosteroids does not cause any problem, but with BT, which contains two reducing tetrazolium rings, in principle the reduction can lead not only to the monoformazan derivative but also to the completely reduced diformazan : BT BT monoformazan BT diformazan Chen et observed that, depending on the experimental conditions, the reduced BT can be red or blue. They attributed this phenomenon to the formation of the above- mentioned two reduced forms of BT.* In most of the analytical methods based on this reaction the red colour is measured, but it must be noted that heating the reaction mixture favours the formation of the blue colour3* and the same effect is observed with BT as a spray reagent in the thin-layer chromatographic investigation of corticosteroids.It is extremely difficult to judge the reliability of the data published in the first 10 years after the discovery of the reaction, as the quality of the BT reagent a t that time might have been poor. Bush and Gale40 claimed that the blue product was the only real reduction product of BT; * Despite the problems described subsequently here, nowadays this theory seems to be generally ac~epted.~'April, 1978 STEROIDS. PART XXXI 351 the red pigments were considered to have been the reduction products of contaminants of BT.* To make the picture even more complicated, the presence of geometrical isomers cannot be excluded.Gosztonyi et al.42 explained the anomalous behaviour of the blue reduction product on thin-layer chromatographic plates on the basis of this theory, and Graham et aZ.11 claimed, on the basis of kinetic experiments carried out in various solvents, that the blue and red colours could be attributed to the two geometrical isomers of the same formazan. In our experiments, we tried to approach this problem from the direction of the proportions of the reactants. Using the conditions of USP XIX1* we varied the concentration of prednisolone between 10-5 and 2 x and 2 x mol l-l, i.e., the molar ratio of BT to prednisolone was varied between lo2 and 10-4.Chloroform was added to the reaction mixture to increase the solubility of prednisolone. A red solution was obtained (Amax. 525 nm) when the molar ratio was greater than 1, whereas the solution was blue (Amax. 590 nm) when the ratio was less than At intermediate molar ratios, a mixture of the two spectra was obtained. On the basis of Fig. 1, where the ratio of the absorbances measured at 590 and 525nm is plotted against the molar ratio of the reactants, the red and blue forms seem to be in equilibrium and therefore we have assumed them to be the monoformazan and diformazan of BT, respectively. In order to obtain the equilibrium constants, the concentrations of the two compounds had to be determined. Their molar absorptivities were calculated from the spectra of the solutions at extreme molar ratios, and the following values were obtained: monoformazan (red), E~~~ = 24 200 and eSg0 = 13 330; diformazan (blue), €525 = 29 070 and €590 = 42 300.Of these values, 24 200 corresponds to the generally accepted apparent molar absorptivity of cortico- steroids in the BT reaction, while 42 300 is in very good agreement with the data of Balogh and Ladiinyilg for BT diformazan (41 200). mol 1-1 and the concentration of BT between I I I I I 1 -4 -3 -2 -1 0 1 2 CB T Cprednisolone Fig. 1. Ratio of absorbances a t 590 and 525 nm veYsus the logarithm of the molar ratio of BT and prednisolone (experimental points and calculated curve). Using these data, the concentrations of the mono- and diformazans were calculated in the usual way with the aid of two equations with two unknowns. The equilibrium constants of the following two equilibria were calculated : Prednisolone (P) + BT + Prednisolone-21-aldehyde (PA) + BT monoformazan (BTMF) ..[PA] [BTMF] PI [BTI K , = P + BTMF + PA + BT diformazan (BTDF) (5) * The introduction of the high-quality Dajac BT seems to have solved this problem.*l352 GOROG AND HORVkTH: ANALYSIS OF Analyst, Vol. 103 [BTMF] and [BTDF] were calculated as described above, and [PI, [PA] and [BT] were calculated from the above values and the initial concentrations of prednisolone and BT using the following equations : .. * (7) * (8) [BT] = [BT], - [BTMF] - [BTDF] . . .. [PI = [PI0 - [PA] == [PI0 - [BTMF] - 2[BTDF] . . The equilibrium constants obtained were K , = 3.8 & 0.8 and K , = 0.08 3 0.02.The calculated curve obtained from these constants with the aid of a computer program is shown in Fig. 1. These data clearly indicate the existence of an equilibrium that can only be explained by the monoformazan - diformazan theory. It can also be seen that, by using the molar ratio of USP XIXf4 (about 15), the monoformazan is formed almost quanti- tatively. Dividing K , by K,, the disproportionation constant, K , is obtained : = 0.021 .. K , [BT] [BTDF] K = - = - IS^ [BTMFF ‘ ‘ (9) The low value of this constant characterises well the equilibrium, which is strongly shifted towards the formation of monoformazan under the given experimental conditions. Pre- liminary experiments showed that this constant increased with increasing temperature ; this is the reason for the appearance of the blue colour at high temperatures.The solvent effect observed by Graham et aZ.ll may have a similar explanation. The monoformazan-diformazan theory does not seem to conflict with the blue colour observable when corticosteroids are detected on thin-layer chromatographic plates with BT spray reagent. In addition t o the high reactivity of molecules in the adsorption layer, under the usual chromatographic conditions BT is not in excess and, moreover, a considerable excess of the corticosteroid can be determined in most instances. The authors are indebted to Mrs. M. GA, Miss 2 s . Hegedus and Miss 2 s . Vincze for techni- cal assistance. 1. 2. 3. 4. 5. 6. 7. 8. 9. 10. 11. 12. 13. 14. 15.16. 17. 18. 19. 20. 21. 22. 23. 24. 25. References Gorog, S., and SzAsz, Gy., “Analysis of Steroid Hormone Drugs,” Elsevier, Amsterdam, Oxford Mader, W. J., and Buck, R. R., Analyt. Chem., 1952, 24, 666. Chen, C., and Tewell. H. E., Fedn Proc. Fedn Am. Socs. Exp. Biol., 1951, 10, 377. Mohrle, H., and Schittenhelm, D., Pharm. Ztg, 1967, 112, 1400. Mohrle, H., and Schittenhelm, D., Arch. Pharm., B e d , 1970, 303, 771. Mohrle, H., and Schittenhelm, D., Arch. Pharm., Berl., 1972, 305, 309. Mohrle, H., Schittenhelm, D., and Federolf, E., Arch. Pharm., B e d , 1972, 305, 587. Graham, R. E., Williams, P. A., and Kenner, C. T., J . Pharm. Sci., 1970, 59, 1152. Graham, R. E., and Kenner. C. T., J . Pharm. Sci.., 1973, 62, 103. Graham, R. E.. Biehl, E. R., Kenner, C. T., Luttrell, G.H., and Middleton, D. L., J . Pharm. Sci., Graham, R. E., Biehl, E. R., and Kenner, C. T., J . Pharm. Sci., 1976, 65, 1048. Johnson, C. A,, King, R., and Vickers, C., Analyst, 1960, 85, 714. “British Pharmacopoeia 1973,” HM Stationery Office. London, 1973, p. 44. “United States Pharmacopeia XIX,” USP Convention Inc., Rockville, Md,, 1975, p. 622. Gorog, S., and Szepesi, G., Analyt. Chem., 1972, 44, 1079. Gorog, S., and Szepesi, G., Analyst, 1972, 97, 519. Szepesi, G., and Gorog, S., Boll. Chim.-Farm., 1975, 114, 98. Edwards, R. W. H., and Kellie, A. E., Biochem. J., 1954, 56, 207. Balogh, S., and LadAnyi, L., Acta Pharm. Hung., 1975, 45, 271. Fairbridge, R. A., Willis, K. J., and Booth, R. G., Biochem. J., 1951, 49, 423. Johnson, C. A., in West, P.W.. Macdonald, A. M. G,, and West, T. S., Editors, “Analytical Chemistry Garratt, D. C., “The Quantitative Analysis of Drugs,” Chapman and Hall, London, 1964, p. 696. Meyer, A. S., and Lindberg, M. C., Analyt. Chem., 1955, 27, 813. Zaffaroni, A., Recent Prog. Horm. Res., 1953, 8, 51. Gorog, S., Tuba, Z., and Egyed, I., unpublished results. and New York, 1978, pp. 331-343. 1975, 64, 226. 1962.” Elsevier, Amsterdam, 1963, p. 183.April, 1978 STEROIDS. PART XXXI 353 26. 27. 28. 29. 30. 31. 32. 33. 34. 35. 36. 37. 38. 39. 40. 41. 42. Gorog, S., and SzAsz, Gy., “Analysis of Steroid Hormone Drugs,” Elsevier, Amsterdam, Oxford Smoczkiewiczowa, A., and Jasiczak, J., P r . Zakresu Towarozn. Chem., Wyzsza Szk. Ekon. Poznaniu, Smith, L. L., and Halwer, M., J . Am. Pharm. Ass., Scient. E d n , 1959, 48, 348. Ascione, P., and Fogelin, C., J . Pharm. Sci., 1963, 52, 709. Smoczkiewiczowa, A., and Jasiczak, J., Chemia Analit., 1971, 16, 1091. Wendler, N. L., and Taub, D., J , Am. Chem. SOC., 1958, 80, 3402. Wendler, N. L., and Taub, D., J . Am. Chem. SOC., 1960, 82, 2836. Wendler, N. L., Taub, D., and Kho, H., J . Am. Chem. SOC., 1960, 82, 5701. Heller, M., Stolar, S. M., and Bernstein, S., J . Org. Chem., 1961, 26, 5036. Smith, L. L., Marx, M., Garbarini, J. J., Foell, T., Origoni, V. E., and Goodman, J. J., J . Am. Chem. Pesez, M., and Bartos, J., Annls Pharm. FY., 1962, 20, 60. Chen, C., Wheeler, J., and Tewell, H. E., J . Lab. Clin. Med., 1953, 42, 749. Bauwens, J. P., and Logghe, G. N., Farm. Weekbl., 1976, 111, 633. Nowaczynski, W., Goldner, M., and Genest, J., J . Lab. Clin. Med., 1955, 45, 818. Bush, I. E., and Gale, M. M., Analyst, 1958, 83, 532. Kunze, F. M., and Davis, S. J., J . Pharm. Sci.. 1964, 53, 1170 and 1259. Gosztonyi, T., MBrton, J., KemCny, V., and Vecsei, P., J . Chromat., 1963, 10, 29. and New York, 1978, p. 40. Zesz. Nauk., Ser., 1, 1972, No. 46, 181; Chem. Abstr., 1974, 80, 60071k. SOC., 1960, 82, 4616. NOTE-References 15, 16 and 17 are to Parts XX, XXI and XXV of this series, respectively. Received Sepember 26th, 1977 Accepted October loth, 1977

ISSN:0003-2654    

DOI:10.1039/AN9780300346

出版商:RSC    

年代:1978

数据来源: RSC

摘要:

354 Analyst, April, 1978, Vol. 103, pp. 354-358 Spectrophotometric Determination of Hydrochlorothiazide and Reserpine in Combination H. Abdine, M. Abdel-Hady Elsayed" and Yousry M. Elsayed Department of Pharmaceutical Analytical Chemistry, Faculty of Pharmacy, University of A lexandria, A lexandria, Egypt An accurate and simple method is proposed for the determination of hydro- chlorothiazide and reserpine. The former is determined by the application of the differential spectrophotometric method and the latter by the forinationof a charge-transfer complex with iodine. The procedure has been applied successfully to the determination of hydrochlorothiazide and reserpine when present in combination with a concentration of hydrochlorothiazide 100 x that of reserpine. Keywovds : Hydrochlorothiazide determination ; reserpine determination ; spectrophotometry The spectrophotometric procedures for the determination of hydrochlorothiazide by means of the diazotisation rea~tion,l-~ reaction with phenol in sulphuric acid medium4 or with anthrone5 are not specific.Those metliod~~,~ which are based on chromatographic separation with subsequent spectrophotometric determination are time consuming. The strong fluorescence of reserpine developed by adding toluene-4-sulphonic acid8 or under the influence of ultraviolet lightgJO is sensitive to many interfering substances present in pharmaceutical formulations. The spectrophotometric methods employed for the deter- mination of reserpine using 4-dirnethylamin0benzaldehyde~~~ vanadate reagent12 or the acid-dye technique13J4 are either not specific or a.re insufficiently accurate. The presence of hydrochlorothiazide in 100-fold excess of reserpine gives rise to problems during the analysis of such two-component mixtures.Methods based on ion-exchange separation15s16 prior to the determination of these components have been described. Recently, high-performance liquid chromatography1' has been applied for the analysis of reserpine - hydrochlorothiazide mixtures in two steps. Each determination was achieved in 60 min with an accuracy of 34%. In this work, the components of this mixture were determined by independent methods within 20-30 mine Hydrochlorothiazide is determined by using the absorbance difference method1*J9 and reserpine by the formation of a charge-transfer complex.20 Experimental Materials and Reagents Hydrochlorothia~ide standard solution. A 1 mg ml-l solution of hydrochlorothiazide (sup- plied by El-Nile Company for Pharmaceutical and Chemical Industries, Cairo, Egypt) in ethanol.Reserpine standard solution. A 1 mg ml-l solution of reserpine base, USP XVI (supplied by Boehringer, Germany), in chloroform. Ravoline - Hydrex tablets. These tablets, supplied by the Memphis Chemical Company, Cairo, Egypt, each contained 0.15 mg of reserpine and 15 mg of hydrochlorothiazide. Iodine solution, 5 x M . Dissolve re-sublimed iodine in chloroform. The solution is stable for 4 weeks when kept at 4 "C away from direct light. All reagents were of analytical-reagent grade and the solvents were of spectroscopic grade.Instrument A Prolabo photoelectric spectrophotometer with a 1-cm silica cell was used. * Present address : Department of Pharmacy, Faculty of Biological and Pharmaceutical Sciences, University of Nigeria, Nsukka, Nigeria.ABDINE, ELSAYED AND ELSAYED 355 Analysis of Ravoline - Hydrex Tablets Determination of hydrochlor0thia.de content Ten tablets were powdered and mixed and an accurately weighed amount equivalent to about one tablet was taken and dissolved in ethanol in a 50-ml calibrated flask. A volume of 10ml was further diluted to 100ml with ethanol, and two equal volumes (5ml) were suitably diluted for spectrophotometric measurement. One portion was diluted with 0.1 N sulphuric acid and the other with 0.1 N sodium hydroxide solution. The absorbance of the alkaline solution was measured at 238 nm using the acidic solution as a blank.Determination of reserpine content Five powdered tablets, or an equivalent amount from the ten powdered tablets, were placed in a 30-ml beaker. The powder was extracted with lo-, 5-, 5- and 5-ml portions of chloroform, each portion decanted into a 25-ml calibrated flask and the solution made up to volume. A volume of 5.0 ml of this solution was transferred into a 25-ml calibrated flask containing 2.0ml of 0 . 0 0 5 ~ iodine solution and diluted to volume with chloroform. The absorbance of this solution was measured a t 294 nm in a 1-cm cell using a solution of 2 ml of 0.005 M iodine solution diluted to 25 ml with chloroform in the reference beam. Results and Discussion Hydrochlorothiazide exhibits overlapping Amax, and Amin.in alkaline and acidic solutions The Amin. value at 238nm was (Fig. 1). chosen for the application of the absorbance-difference method.18Jg However, the absorbance values are different. 0.6 0.5 0.4 0.3 0.2 0.1 0.0 -0.1 T 230 250 270 290 310 330 Wavelengthhm Fig. 1. A, Absorption spectra of hydrochlorothiazide (1 mg per 100 ml) . Solvents : solid line, 0.1 N sulphuric acid ; and broken line, 0.1 N sodium hydr- oxide solution. B (dotted line), AA curve. Within the concentration range 0.3-1.7 mg per 100 ml, AA versus C exhibits a linear relationship, where AA is the difference between the absorbances in alkaline and acidic solutions. The concentration of an unknown solution can be calculated from the equation .. * . (1) AA2% = 0.0178 + 0.2291 C ..which was obtained from the results given by the application of the procedure described above to known concentrations of the standard solution.356 ABDINE E t Ui?. : SPECTROPHOTOMETRIC DETERMINATION Analyst, VOZ. 103 The conventional spectrophotometric method was carried out and the results were compared with those given by the absorbance-difference method. Beer’s law is obeyed within the concentration range 0.3-1.5 mg per 100 ml, using 0.1 N sulphuric acid as a solvent. The following equation describes the calibration graph : A2,0 = -0.0053 + 0.6329 C . . .. ’ - (2) Reserpine was determined by applying the method of charge-transfer complexation. The spectrum of the reserpine - iodine complex is given in Fig. 2. This shows that the A:%cm value for iodine-complexed reserpine (Amax. 294 nm) is about 2.4 times higher than that for the uncomplexed reserpine in chloroform (Amax.268 nm), and the complexed reserpine exhibits another maximum (at 365 nm) of high intensity (A:& - 500) compared with the uncomplexed reserpine, which merely shows an inflection a t 296 nm (A:& w 150). This result is attributed to an intermolecular charge-transfer transition involving a one-electron jump from donor to acceptor.21 Beer’s law is obeyed over the concentration range 0.24- 1.68 mg per 100 ml so that the procedure could be used for the determination of reserpine. 1.2 1 .o 0.8 0.6 0.4 0.2 0 Wavelength/nm Fig. 2. Absorption spectra of (A) un- complexed reserpine (1.92 mg per 100 ml) and (B) iodine-complexed reserpine (1.92 mg per 100 ml).A calibration graph was constructed by following the procedure described above. corresponds to the following regression equation : It .. * * (3) Ag4 = 0.044 + 0.642 0 C .. where AE4 is the absorbance at 294 nm due to the charge-transfer complex. Reserpine was also determined by measuring the absorbance of the chloroform extract of tablets at 268 nm and calculating the concentration from the following regression equation : A268 7 -0.0270 + 0.3100 C . . .. - - (4) The validity of equations (1) and (2) was tested by determining hydrochlorothiazide in a laboratory-made mixture. The concentration of reserpine was calculated using equations (3) and (4) (Table I). The applicability of the procedure to Ravoline - Hydrex tablets was checked (Table 11).Conclusions The good recovery of hydrochlorothiazide shows the successful application of the differ- ential spectrophotometric method, although the only pH-induced spectral change is the hyperchromic effect. As indicated by the (Tables I and 11), the accuracy of the results is increased by using the absorbance-difference method for the major component and charge-transferA$?'iE, 1978 OF HYDROCHLOROTHIAZIDE AND RESERPINE IN COMBINATION TABLE I 357 DETERMINATION OF HYDROCHLOROTHIAZIDE AND RESERPINE IN KNOWN MIXTURES Hydrochloro thiazide Reserpine A I A T I \ Amount recovered, yo Amount recovered, yo r h 7 7 I h Amount Spectrophoto- Amount Spectrophoto- added/mg metric method AA method added/mg metric method CT method BP method* 0.72 102.19 98.70 0.48 102.08 102.75 99.47 0.80 102.30 97.87 0.56 104.38 101.25 104.52 0.88 102.21 99.78 0.64 104.25 101.72 94.16 1.04 102.01 100.34 0.72 104.17 100.14 102.98 1.12 103.00 97.74 0.80 104.64 99.88 1.20 102.92 98.33 0.96 107.78 98.40 1.28 102.56 98.41 Mean .. . . 102.46 98.74 Standard deviation . . f0.38 &0.97 Calculated t . . 9.43 104.55 100.69 100.28 f 1.83 & 1.54 k4.60 8.84 0.207 (2.179) t (2.228)t (2.306)t 8.91; * In this method 20-mg samples were taken. t The figures in parentheses are theoretical t values at the 95% probability level, 01 = 0.05. $ Calculated F value, for which theoretical value a t the 5% level is 5.41. complexation for the minor component, because the interfering absorbance is cancelled whereas it gives a positive error in the conventional spectrophotometric method.If the method involving charge-transfer complexation is compared with the pharma- copoeial method,l which is based on the reaction of reserpine with nitrite,23 the results of both methods are seen to be of equal accuracy, as the calculated t does not exceed the theoretical t. However, there is a significant difference between the precision of the two methods (F-test). As the official method is influenced by many variables (e.g., normality of acid, concentration of sodium nitrite, time of contact of reagents and temperat~re),~~ it is not surprising that utilisation of charge-transfer complexation gives more reproducible results. TABLE I1 DETERMINATION OF HYDROCHLOROTHIAZIDE AND RESERPINE IN COMMERCIAL TABLETS H ydrochlorothiazide I A I f Amount found, yo A \ Amount Spectrophoto- takenlmg metric method AA method 0.72 95.83 94.44 0.84 94.29 91.67 0.90 93.60 92.22 0.08 93.75 92.71 1.02 93.14 92.16 1.08 93.52 91.67 1.20 93.67 92.60 1.32 93.18 91.67 Reserpine A 1 Amount found, % Amount takenlmg 0.36 0.42 0.48 0.54 0.60 0.66 Spectrophoto- metric method 105.72 109.19 108.42 107.41 106.50 106.04 Mean ... . 93.87 92.38 Standard deviation . . &0.75 f0.92 Calculated f . . 3.66 (2.145) * * The figures in parentheses are theoretical t values at a = 0.05. 1 CT method 100.62 97.62 100.42 101.30 103.03 99.23 107.21 100.37 f 1.38 & 1.84 7.29 (2.228) *358 ABDINE, ELSAYED AND ELSAYED Analyst, Vol. 103 Moreover, the need for powdering about 130 tablets before carrying out the official method renders its application to the determination of reserpine in tablets containing 0.15 mg each less suitable for practical work.1. 2. 3. 4. 6. G. 7. 8. 9. 10. 11. 12. 13. 14. 15. 16. 17. 18. 19. 20. 21. 22. 23. References “British Pharmacopoeia 1973.” HM Stationery Office, London, 1973, pp. 229 and 411. Rehm, C. R., and Smith, J. B., J . Am. Pharm. .4ss.. Scient Edn, 1960, 49, 386. Magalhaes, J. F., and Piros, M. G., Rev. Farm. Bioquim. Univ. S. Paulo, 1970, 8, 273; Analyt. Abstr., Vachek, J., Cslkd. Farm., 1961, 10, 515; Analyt. Abstr., 1962, 9, 3428. Perez, B. J.. Danhier, C. A., and Brieva, A. J., A n . R. Acad. Farm., Madr., 1967, 33, 233. “United States Pharmacopeia XIX,” USP Convention Inc., Rockville, Md., 1975, p. 236. Fazzari, F. R., J . Ass, Off. Analyt.Chem., 1970, 53, 582. Jakovijevic, I. M., Fose, J. M., and Kuzel, N. R., Analyt. Chem., 1962, 34, 410. Haycock, R. P., Sheth, P. B., and Mader, W. J., J . Am. Pharm. Ass., Scient Edn, 1959, 48, 479. Kahane, E., and Kahane, M., Annls Pharm. Fr., 1958, 16, 726. Hakkesteegt, T. J., Pharm. Weekbl. Ned., 1968, 103, 1237. Stoinier, R., Farmaco Ed. Prat., 1969, 24, 167. Langel, P., and Hasselman, M., Chim. Analyt., 1962, 44, 433. Hsi-Shui, C., Chih-Ying, K., Chih-Ling. C., Chin-Yea, C., Chia-Lu, H., Shu Ling, L., and Hsi-Kung, Urbanyi, T., and O’Connell, A., Analyt. Chem., 1972, 44, 565. Chu, R., J. Ass. Off. Analyt. Chem., 1971, 54, 603. Honigberg, I. L., Stewart, J. T., Smith, A. P., Plunkett, R. D., and Hester, D. W., J . Pharm. Sci., Aulin-Erdtman, G., Chemy Ind., 1955, 74, 581. Junejo, G. M., and Glenn, A. L.. Chemy Ind., 1956, 75, 813. Taha, A. M., Ahmad, A. K. S., Gomaa, C. S., and El-Fatatry, H. M.. J . Pharm. Sci., 1974, 63, 1853. Foster, R., Editor, “Organic Charge Transfer Complexes,” Academic Press, London, 1969, pp. 23 Bauer, E. L., “Statistical Manual for Chemists,” Academic Press, London, 1971, p. 61. Szalkowski, C. R., and Mader, W. J., J . Am. Pharm. Ass., 1956, 46, 613. 1971, 21, 3723. C., Acta Pharmac. Sin., 1966, 13, 280; Analyt. Abstr., 1967, 14, 4967. 1974, 63, 1762. and 33. Received May 25th, 1977 Accepted October 6th, 1977

ISSN:0003-2654    

DOI:10.1039/AN9780300354

出版商:RSC    

年代:1978

数据来源: RSC