ISSN:0003-2654    

DOI:10.1039/AN98106FX005

出版商:RSC    

年代:1981

数据来源: RSC

摘要:

ANALAO 106 (1 259) 129-256 (1 981 )ISSN 0003-2654February 19811291351461531601661727 8218819820621 3221227231243247250254THE ANALYSTTHE ANALYTICAL JOUKNAL OF THE ROYAL SOCIETY OF CHEMISTRYCONTENTSDetermination o f Dithiocarbamates by Liquid Chromatography Using Transition-metalSalts as "lon-pair" Reagents-Roger M. Smith, R. L. Morarji and W. G. SaltDetermination of Polynuclear Aromatic Hydrocarbons in Food, Water and Smoke UsingHigh-performance Liquid Chromatography-Neil T. Crosby, David C. Hunt, Lesley A.Philp and lndu PatelVoltammetric Determination o f 2-, 3- and 4-Chloroaniline in Mixtures-John P. Hart,Malcolm R. Smyth and W. Franklin SmythSeparation of Protein-bound Copper and Zinc in Human Plasma by Means o f GelFiltration - lon-exchange Chromatography-J.B. Dawson, M. H. Bahreyni-Toosi, D. J.Ellis and A. HodgkinsonGas - Liquid Chromatographic Determination of Major Constituents o f Piper meth-ysticum-R. N. DuveGas-chromatographic Determination o f Trace Amounts o f Lower Fatty Acids in AmbientAir Near and in Exhaust Gases o f Some Odour Sources-Yasuyuki HoshikaDetermination o f Trace Elements in Plant Materials b y a Dry-ashing Procedure.Part I.Determination of Trace Elements in Plant Materials by a Dry-ashing Procedure.Part II.Assessment o f Phosphorescence Spectroscopy f o r Crude Oil Identification-Mark M.Corfield, H. Lesley Hawkins, Phillip John and Ian SoutarSimultaneous Spectrophotometric Determination o f Palladium(ll) and Gold( Ill) withMethiomeprazine Hydrochloride: Analysis o f Alloys and Minerals-H.Sanke Gowda,K. A. Padmaji and K. N. ThimrnaiahApplication o f an Automatically Triggered Digital Integrator t o Flame Atomic-absorptionSpectrometry o f Copper Using a Discrete Nebulisation Technique-T. Uchida, I .Kojirna and C. lidaDetermination o f "Inorganic" Arsenic(ll1) and Arsenic(V), "Methylarsenic" and"Dimethylarsenic" Species by Selective Hydride Evolution Atomic-absorptionSpectroscopy-A. G. Howard and M. H. Arbab-ZavarStudies in Chemical Phase Analysis. Part II. Determination of t h e Solubilities o fCarbides, Nitrides, Oxides and Sulphides in Certain Organic Solvent - BromineMixtures-I. S. Busheina and J. B. HeadridgeDetermination of Microgram Amounts o f Calcium in Small Biological Saniples by EDTATitration Using Patton and Reeder's Indicator-Poul Prentp and Annette PrentplAutomatic Titration by Stepwise Addition o f Equal Volumes o f Titrant.Part VI.Further Extension o f the Gran I Method for Calculation o f the Equivalence Volumein Acid - Base Titrations-Gunnar Gran and Axel JohanssonCobalt and Molybdenum-D. L. HeanesCopper, Manganese and Zinc-D. L. HeanesSHORT PAPERSSpectrophotometric Determination o f Nitrite in Waters-S. Flamerz and W. A. BashirDetermination of Water in NN-Dimethylformamide by the Kinetic Method of TangentsUsing t h e Oxidation of Catechol as the Indicator Reaction-L. R. ShermanSpectrophotometric Determination o f Vanadium(V) Using 4-Benzoyl-3-methyl-l-phenyl-5-pyrazolone-Yoshifumi Akarna, Toshio Nakai and Fumikazu KawamuraBOOK REVIEWSSummaries of Papers in this Issue-Pages iv, vi, vii, viii, ix, xii, xiii, xiv, xviPrinted by Heffers Printers Ltd Cambridge EnglandEntered as Second Class at New York, USA, Post OfficAnalytical Sciences MonographsNo.6 lsoenzyme AnalysisBy D. W. MossThis monograph attempts to drawtogether the most important experimentaltechniques which have resulted fromthe modern recognition that enzymesfrequently exist in multiple molecularforms. This monograph also indicatesthe advantages and limitations inisoenzyme studies of these modernanalytical techniques.Brief Contents:Multiple Forms of Enzymes; Separationof Multiple Forms of Enzymes; SelectiveInactivation of Multiple Forms ofEnzymes; lmmunochemistry of MultipleForms of Enzymes; Catalytic Differencesbetween Multiple Forms of Enzymes;Met hods of 0 btai ni ng StructuralInformation; Selection of Methods ofAnalysis.Hardcover 171 pp 8%" x 59" 0 851 86 800 2 f 12.00 ($32.50)Orders to: The Royal Society of Chemistry DistributionCentre, Blackhorse Road, Letchworth, Herts.SG6 1 HNNotice to SubscribersSubscriptions for The Analyst, Analytical Abstracts and Analytical Proceedingsshould be sent to:The Royal Society of Chemistry, Distribution Centre,Blackhorse Road, Letchworth. Herts., SG6 1 HN, EnglandRates for 1981 (including indexes)U K/ Rest ofEire USA WorldThe Analyst, and Analytical Abstracts . . . . . . . . f168 $416 f177The Analyst, Analytical Abstracts, and Analytical Proceedings . . f 1 90 $41 7.50 f 200.50Analytical Proceedings alone* . . .. . . . . . . f30 $70.50 f31.50Analytical Abstracts alone . . . . . . . . . . . . f129.50 $322 f137Subscriptions are not accepted for The Analyst alone*NEW FOR 198

ISSN:0003-2654    

DOI:10.1039/AN98106BX007

出版商:RSC    

年代:1981

数据来源: RSC

ISSN:0003-2654    

DOI:10.1039/AN98106FP013

出版商:RSC    

年代:1981

数据来源: RSC

摘要:

February , 1981 SUMMARIES OF PAPERS I N THIS ISSUE xiiiStudies in Chemical Phase Analysis. Part 11. Determination ofthe Solubilities of Carbides, Nitrides, Oxides and Sulphides inCertain Organic Solvent - Bromine MixturesAs part of a study on the quantitative separation of carbide, nitride, oxideand sulphide inclusions from metals, the solubilities of five carbides, sevennitrides, sixteen oxides and eleven sulphides have been determined a t 25 “Cin methyl acetate - and acetonitrile - bromine mixtures (10 + 1 V / V ) bothafter shaking a t room temperature and after refluxing. Aluminium nitrideand the carbides and nitrides of chromium, niobium, titanium and vanadiumhave low or very low solubilities, particularly in methyl acetate - bromine a troom temperature, but iron carbide and iron and manganese nitrides areextensively decomposed, with the iron and manganese passing into solution.None of the oxides is more than sparingly soluble but most of the sulphidesare appreciably soluble.Appropriate conditions are suggested for achievinga clean separation of oxides and stable carbides and nitrides from metals.Keywords : Solubilities of carbides, nitrides, oxides and sulplaides ; methylacetate - bromine ; acetonitrile - bromineI. S. BUSHEINA and J. B. HEADRIDGEDepartment of Chemistry, University of Sheffield, Sheffield, S3 7HFAnalyst, 1981, 106, 221-226.UNSEAL@ “TWO-IN-ONE” THE HIGH PRESSURESERIES OF SIX DECOMPOSITION VESSEL SIZESIN THREECONTAINERSfor any laboratorysample size up to ,bench scaleFOR :DECOMPOSITIONDIGESTIONSOLUBlLlZATlONSELECTIVEEXTRACTIONLEACHINGHYDROLYSISDESTRUCTION OF IORGANIC MATTERIThe new generationof UnisealB High Pressure Decomposition Vessels for practically anyquantity of specimen comes in six sizes: 5, 10, 25, 50, 70 and 110 mlwhich safely operate at pressures of up to 350 atm./5,250 p s i .andtemperatures of up to 180OC. UnisealB now markets the mostcomprehensive and versatile range of decomposition vessels.HIGHLIGHTS : 0 “TWO-IN-ONE” : Two Teflon vessels f i t one steel jacketA choice of six volumesMaximum temperature : 1800 C No volatilization losses Controlled-torque closure for perfect sealing Closure tool set for use with all units0 Durability - Economy 12-month warrantyAPPLICATIONS : Inorganic - organicAgricultural Chemistry / soils, plants, fertilizers 0 Analytical ChemistryBio-Clinical Chemistry / blood, urine, tissue, protein hydrolysis, amino acids0 Cement Ceramics Corrosion 0 Environmental Pollution 0 FoodForensic Toxicology Geochemistry Glass Industry 0 Marine Research0 Materials Research 0 Metallurgy Mining Oil prospectingPharmaceuticals Plastics 0 Refractories.VISIT US AT THE PITTSBURGH CONFERENCE EXHIBITION 1981. BOOTH No.4052P.O.B. 9463. HAIFA 31094. ISRAEL. Tel. 04-244990. Telex: BXHA IL 46400 EXT 8610Maximum pressure : 350 atm./5,250 p.s.1.PATENTS PENDINGUNISEAL DECOMPOSITION VESSELS LTDxiv SUMMARIES OF PAPERS I N THIS ISSUE February, 1981Determination of Microgram Amounts of Calcium in SmallBiological Samples by EDTA Titration UsingPatton and Reeder's IndicatorAn EDTA titration method has been developed for the determination ofcalcium in small biological samples.Fresh or dried material is extractedwith 1 N nitric acid and the extract is alkalinised to pH 13.4 and titrated withEDTA in the presence of Patton and Keeder's indicator and sodium tartrate.Barium, magnesium, manganese, iron, lead, zinc, oxalate, phosphate andsulphate do not interfere owing to the high pH and the presence of tartrate.Copper interferes, but up to 5 pg of copper can be masked with cyanide if thecopper concentration does not exceed one tenth the calcium concentration.This method permits the quantification of calcium in tlte range 0.5-40 p gand with a reproducibility of f2.5y0 or better.Keywords : Calcium microdetermination ; EDTA titration ; biological samplesPOUL PRENT0 and ANNETTE PRENT0Institute of General Zoology, University of Copenhagen, Universitetsparken 15,2 100 Copenhagen 0, Denmark.Analyst, 1981, 106, 227-230.INTRODUCING 4UNISEAL@PRESSURE MEASURING SYSTEM- l U C Y l I V I6 An innovation that countsUNSEAL@ sets newstandards of accuracyand simplicity.)pens upnew horizons f )r non-invasive pressuremeasurements in closedsystems by externalcontact onlyUN1SEAl.a adc 5 a vitaldimension to * ourexperimental evaluationsby obtaining absolutevalues of pressure withUniseal's PressureMeasuring System.'3 ;elf-contained closed system : simple sealing-, leakage problems * Pressure values by:rted into reactor; hence.no volume change/itamination of specimen/ no pressure sensore data at temperatures up to 1800 CiHLIGHTS :' ?ma1 surface contact only No pressurersuu8nu=mve systems Simple to operattAPPLIC ATIONS :Analytical chemistrySample weight optimizationby pressure determinationOperational safety evaluationPressure characteristics ofunknown samplesChemical EngineeringInorganic ChemistryCeramicsFertilizersGeochemistryGlassesMetallurgyRefractoriesExplosivesOrganic Chemistry Critical pressures0 Bio applicationsClinical applications studiesCarbonyl compiexesOrgdno-metallic compounds Solubility studiesPolymerization studiesPhysical and Theoretical precipitationChemistry PbpellantsReaction kineticsGaseous reactions - rateNoble gas reactionsEnergy calculationsVapour pressure studiesChemical equilibriumOxidation - reduction - protein hydrolysisequi I ibriaSolubility products andPITTSBURGH CONFERENCEEXHIBITION 1981studies* PATENTS PENDINGUNISEAL DECOMPOSITION VESSELS LTD.P 0 Box 9463.HAIFA 31094. ISRAEL Tel 04-244990 Telex ' BXHA IL 46400 EXT 861February, 1981 THE ANALYST XVThe Royal Society of Chemistry-PublicationsAnnual Reports on Analytical Atomic SpectroscopyVOl. 9Edited by J. B. Dawson and B. L. SharpThis series provides the practising analytical chemist and spectroscopist with ahandbook of current practice and recent advances in instruments and methodsfor the determination of elements in the form of comprehensive, critical annualreports.“This is a worthwhile series, providing a survey of a tre nendous bulk of originalliterature.”-Analytica Chimica Acta reviewing Vol.5Brief Contents : Atomization and Excitation; Instrumentation; Methodology;Applications; New Books; Reviews; Meetings; References; Author Index;Subject IndexHardcover 357 pp 82’’ x 6” 0 85186 727 8 f34.00 ($94.25)RSC Members f22.00Selected Annual Reviews on the Analytical SciencesVol. 4Edited by L. S. BarkThe fourth volume continues the policy adopted in previous volumes ofpresenting critical reviews of selected topics in modern analytical science. Eachof these reviews embodies the work considered pertinent by a practisingchemist over the four or five years up to 1974.Softcover 80pp 82 x 6” 0 85990 204 8RSC Members f7.50f14.50 ($39.25)Hazards in the Chemical laboratory 2nd EditionEdited by G.D. MuirIn the five years since the first edition was published, Hazards in the ChemicalLaboratory has become established as a vital handbook in ail types of laboratoryenvironment. However over this period many developments have taken placewhich justify change in scope and emphasis.This second edition contains completely new chapters on recent developmentsin laboratory safety.“As with the first edition, Hazards in the Chemical Laboratory second edition isa must for any laboratory where chemicals are handled, and it is sure to be anunqualified success both in this country and abroad.“-S.S. Chissick, reviewing for Laboratory Practice, July 1977.Protective PVC Binding 489pp 83” x 6” 0 85186 699 9f 9.50 ($25.75) RSC Members f 7.00RSC Members should send their orders to: The Membership Officer, The RoyalSociety of Chemistry, 30 Russell Square, London WC1 B 5DT. All other ordersshould be sent to: The Royal Society of Chemistry, Distribution Centre,Blackhorse Road, Letchworth, Herts, SG6 1 HNxvi SUMMARIES OF PAPERS IN THIS ISSUEAutomatic Titration by Stepwise Addition of EqualVolumes of Titrantof the Equivalence Volume in Acid - Base TitrationsIn Part V of this series a very simple metliod for the calculation of the equiva-lence volume (V,) from titration data was presented. It was based on themethod of stepwise addition of equal volumes of titrant combined with anextended version of the Gran I method.The method presented had somelimitations, however, as it could not be used for acids with log K between 2and 4.2 or for very weak acids (log K > 7 ) .In this paper equations of a more complicated nature are given, which canbe used for monoprotic acids with log K < 10. The equations contain, besidestitrant volumes and hydrogen ion concentration, values for K and K w , which,however, need not be known but are calculated from the titration data a t thesame time as Ve. The same type of calculation can be uFzd in the calculationof Ve for most di- and triprotic acids.A discussion is given of possible causes for errors in the equivalence volume,calculated using the equations presented.When iccurate automatic pipettesare used for adding the titrant the uncertainty in the determination of thehydrogen ion concentration is the main source of error.Keywords : Gran I method; acid - base titratinn ; automatic titration; equi-February, 1981Part VI. Further Extension of the Gran I Method for Calculationvalence volume calculation ; potentiometric t'crationGUNNAR GRAN and AXEL JOHANFSONDepartment of Analytical Chemistry, The Royal Institute of Technology, S-10044Stockholm 70, Sweden.Analyst, 1981, 106, 231-242.Spectrophotometric Determination of Nitrite in WatersShort PaperKeywords : Nitrite determination ; water ; 4-aminosalicylic acid reagent ;diazotisation ; spectrophotometryS. FLAMER2 and W.A. BASHIRDepartment of Chemistry, College of Education and College of Science, University ofMosul, Mosul, Iraq.Analyst, 1981, 106, 243-247.Determination of Water in NN- Dimethy'lformamide by the KineticMethod of Tangents Using the Oxidation of Catechol as theIndicator ReactionShort PaperKeywords : NN-Dimethylfornzamide ; sodium metaperiodate ; catechol; spectro-photometry; kineticsL. R. SHERMANDepartment of Chemistry, University of Akron, Akron, Ohio 44325, USA.Analyst, 1981, 106, 247-250.Spectrophotometric Determination of Vanadium(V) Using4- Benzoyl-3-methyl- 1 -phenyl- 5-pyrazoloneShort PaperKeywords : Vanadium determination ; spectrophotometry ; 4-benzoyl-3-methyl-1 -phenyl-5-pyrazolone ; solvent extractionYOSHIFUMI AJCAMA, TOSHIO NAKAI and FUMIKAZU KAWAMURADepartment of Chemistry, Meisei University, Hino, Tokyo 191, Japan.Analyst, 1981, 106, 250-253February , 1981 THE ANALYST XViiAnalytical Sciences MonographsAnalysis of Airborne Pollutants inWorking AtmospheresThe Welding and Surface Coatings IndustriesbyJ.A.Moretonand N.A.R.Falla Part II: The Surface CoatingsIndustryThis volume is one of a series of short Origin of Airborne Pollutants in theSurface Coatings Industry monographs on topics of interest toCollection and Analysis of Gaseous Analytical Chemists.Atmospheric Pollutants in the Surface This monograph covers the following mainareas: Coatings IndustryCollection and Analysis of ParticulateAtmospheric Pollutions in the SurfaceCoatings IndustryFurther Trends Relating to Sampling andAnalysis in the Welding and SurfacePart I: The Welding IndustryAirborne Pollutants in WeldingSampling of Welding WorkshopAtmospheres Coatings IndustriesGasesAnalysis Of Welding Fumes and Pollutant Hardcover 192 pp.0 85186 860 6 f12.00($32.50)RSC Members f9.00Orders should b e s e n t to: The Royal Society of Chemistry, Distribution Centre, Blackhorse Road,RSC Members should send their orders t o :The Membership Dept., The Royal Society of Chemistry, 30 Russell Square, London, WC1 B 5DT.Letchworth, Herts, SG6 1 HN.ANALYTICAL CHEMISTSThe NATIONAL INSTITUTE FOR METALLURGY (NIM), which is one of the largest laboratories formineral processing research in the world, is located in the heart of an attractive residential area northof Johannesburg.NIM is expanding rapidly because of increased demand from the mining industry for new processesto assist them in the exploitation of South Africa’s vast low-grade mineral resources.The Analytical Chemistry Division is equipped with the most sophisticated equipment available to-day.Techniques used to analyse the wide variety of materials encountered include the following: * Emission Spectroscopy withInductively Coupled Plasma * X-ray Fluorescence * Atomic Absorption * Spark Source Mass Spectrometry* U-V Visible Spectrophotometry * Electro-Analytlcal Techniquesj , Various types of elemental andgas analyzers * Classical wet-chemistry methodsThe Division devotes a substantial effort to the preparation and analysis of international referencematerials.2 o r 3 YEAR CONTRACT OPPORTUNITIES ARE NOW AVAILABLE TO SUITABLYQUALIFIED PEOPLE FOR RESEARCH AND DEVELOPMENT I N ANALYTICALTERESTING NATURE.Fringe benefits include the following:rt Relocation assistance (return airfare, removal and settling-in allowances, etc.) * Free commuterbus service It Subsidized lunches and sports facilities * Generous leave allowance * A servicebonus equal to one month’s salary * Opportunities for advanced studv.Please write in confidence to: The Office of t h e Scientific Counsellor.South African Em-bassy, Chichester House, 278 Hi h Holborn, LONDON WCIV, 7 HE or phone him atCHEMISTRY METHODS AS WELL AS FOR SERVICE WORK OF A VARIED AND IN-01-242-1766 before 23 Februarv 1881. 6545~~~~~~~ ~ ~A246 for further information. See page xviiTUCK IN UNDER FLAP ATHE ANALYST February, 1981READER ENQUIRY SERVICEFor further information about any of the products featured in the advertise- $ments in this issue, please write the appropriate A number in one of the 2Postage paid if posted in the British Isles but overseas readers must affixa stamp.boxes below. n(Please use BLOCK CAPITALS)NAME .........................................................................................................................................................................................0 C C U PATI 0 N .................................................................................................................................................................AD D R ESS .............................................................................................................................................................................SECOND FOLDPostagewill beDo not affix Postage Stamps if posted inGt. Britain, Channel Islands or N. IrelandI Paid byLicenseeI I BUSINESS REPLY SERVICELicence No. W.D. 106Reader Enquiry ServiceThe AnalystThe Royal Society of ChemistryBurlington HousePiccadilly London W1 E 6WFENGLANDTHIRD FOLDn;D vl -I-6 r

ISSN:0003-2654    

DOI:10.1039/AN98106BP021

出版商:RSC    

年代:1981

数据来源: RSC

摘要:

FEBRUARY 1981 Vol. 106 No. 1259 The Analyst Determination of Dit hiocarbama tes by Liquid Chromatography Using Transition-metal Salts as "Ion-pair" Reagents Roger M. Smith, R. L. Morarji and W. G. Salt Department of Chemistry, Loughborough University of Technology, Loughborough, Leicestershire, LE 11 3T U A technique has been developed for the liquid-chromatographic determina- tion on an octadecylsilyl silica reversed-phase column of N-alkyl- and NN- dialkyldithiocarbamates by adding transition-metal salts of cobalt(I1) or nickel(I1) to the mobile phase. The unsuccessful use of mercury(II), copper(I1) and lead(I1) is discussed and the mixed complexes formed when two dithiocarbamates are injected are described. The retentions of the dithiocarbamates are compared with those of related thiuram disulphides.Keywords: Dithiocarbamate determination; thiuram disulphide determina- tion ; liquid chromatography ; complex formation ; transition-metal salts A number of methods have been developed for the determination of N-alkyl- and NN- dialkyldithiocarbamates and their salts and related thiuram disulphides,l because of their widespread use as fungicides, rubber vulcanisers and as pharmaceuticals.2 However, many of these methods, including ultraviolet spectroscopy, colorimetry with copper salts and the widely used acid degradation to give carbon disulphide, are specific only for the dithiocarba- mate group and cannot distinguish between compounds with different N-alkyl substituents. As part of a study of the degradation of the cyclic fungicide dazomet (3,Ei-dimethyltetra- hydro-l,3,5-thiadiazine-2-thione), we developed and briefly reported a liquid-chromatographic technique in which the individual dithiocarbamates were separated using a solvent containing nickel or cobalt salts as "ion-pair" reagent^.^ Although the sodium salts were unretained on chromatography, the non-aqueous soluble metal-ion complexes formed in the eluent were readily separated on a reversed-phase column.In this paper we describe our qualitative and quantitative studies in detail and discuss the examination of further metal ions as potential complexation reagents. The interaction of dithiocarbamates, in particular NN-diethyl- and NN-tetramethylene- di t hiocarbamat e, with transit ion-met a1 ions to form complexes extract able into organic solvents has been used in analytical chemistry for the separation4 and spectrophotometric determination of a number of metal^.^ The complexes, after extraction, can be separated by thin-layer chromatography (e.g., reference 6), gas chromatography (q., reference 7) or liquid chromatography on silica gel,*-14 bonded nitrile15-16 or hydrocarbon (reversed-phase) columns.17-19 In most instances the separation has been aimed at the determination of the metal ion rather than of the ligand.Experimental Apparatus Liquid chromatography was carried out using a Water Associates 6000 pump connected to a Shandon Southern column (10 cm x 5 mm i.d.) packed with Hypersil-ODS and fitted with a Rheodyne 7010 valve injector with a 10-pl loop. The eluates were detected using an ALC 202 detector at 254 nm.The solvent flow-rate was 1.5 cm3 min-l. Reagents and Standards Metal salts. Analytical-reagent grade nickel sulphate, cobalt nitrate, copper nitrate, 129 mercury(I1) nitrate and lead nitrate were used.130 SMITH et al. : DITHIOCARBAMATES BY LIQUID Analyst, Vol. 106 The sodium salts of N-methyl-, NN-diethyl- and NN-dimethyldithio- carbamate were supplied by J. D. Cambell and Sons Ltd., Warrington, Sigma London Chemical Co. Ltd., Poole, and Aldrich Chemical Co. Ltd., Gillingham, respectively. Ammonium NN- tetramethylenedithiocarbamate was supplied by Sigma London Chemical Co., Ltd., Poole. Sodium N-ethyldithiocarbamate and ammonium dithiocarbamate were synthesised from the corresponding amine and carbon disulphide.2 Thiram (NNN’N‘-tetramethylthiuram disulphide) was supplied by Robinson Bros.Ltd., West Bromwich, and disulfiram (NNN’N’-tetraethylthiuram disulphide) was supplied by Sigma London Chemical Co. Ltd., Poole. Dithiocarbamates. Thiuram disulphides. Methanol. HPLC grade from Fisons Scientific Apparatus, Loughborough. Procedure Solutions of the dithiocarbamates or thiuram disulphides, approximately 500 mg dm3 in methanol - water, were injected on to the column. Different ratios of methanol and water, containing 0.1% wz/V of the metal salts, nickel sulphate, cobalt nitrate, copper nitrate or lead nitrate, were used as indicated for the eluent. In some runs 0.05 M phosphate buffer, pH 5, containing the metal ions was used. As prolonged elution of methanol - water mixtures did not remove all the trace amounts of cobalt(I1) from the column, a different column was used for each metal ion. Results and Discussion On examination of solutions of the sodium salts of N-methyl- or NN-diethyldithiocarbamic acid by liquid chromatography using an octadecylsilyl silica reversed-phase column, the solutes were eluted at the solvent front (k’ = 0) even if water alone was used as the solvent.In some runs the sample appeared to interact with the column and no peaks were observed. Attempts to control the ionisation of the dithiocarbamates by using a phosphate buffer at pH 5 as the eluent had no effect. A lower pH was not used because the dithiocarbamates are sensi- tive to acid and are rapidly degraded to carbon disulphide and an amine.5 The addition of the organic ion-pair reagents tetrabutylammonium hydroxide and cetrimide to the eluent was also studied but their use appeared to result in complete degradation of the sample.The thiuram disulphides thiram and disulfiram could be readily chromatographed without decom- position. As a number of metal-ion complexes of NN-diethyl- and NN-tetramethylenedithio- carbamates have been successfully chromatographed on RP-8 and RP-18 reversed-phase columns,1g it was decided to investigate the use of transition-metal ions as potential ion-pair reagents. It was hoped that on injection of the dithiocarbamate salts, the metal ions would form stable neutral complexes, suitable for chromatography. Despite the common use of organic ion-pair reagents to improve the separation or efficiency of polar compounds on liquid chromatography, there are few reports of the use of metal ions.Examples include the inter- action of zinc ions with aminobenzoic acids,20 nickel ions with aniline21 and its metabolites and a zinc or cadmium C,,-dien complex with the dansyl amino acids and s ~ l p h o n a m i d e s . ~ ~ ~ ~ The more common addition of silver ions to solvents to alter the retention of olefins results not in the formation of a neutral ion-pair complex but of a polar n-complex with a shorter retention time on a reversed-phase column than the original olefin (e.g., reference 24). After initial trials using separately prepared complexes a range of solvents containing different proportions of methanol and 0.1% solutions of nickel(I1) sulphate or cobalt(I1) nitrate were examined.A series of sodium or ammonium salts of dithiocarbamates and related thiuram disulphides were individually injected and their capacity factors were determ- ined (Table I). The complexes were readily detected at 254 nm, with no background inter- ference from the metal salts in the solvent. The peak efficiencies of the complexes were com- parable to those of uncomplexed eluates and suggested that mixing and complex formation occurred rapidly on injection. A number of important fungicides, including zineb, maneb and mancozeb, are based on ethylenebisdithiocarbamic acid, (CH,NHCS,), 2-; however, its metal complexes are polymeric and attempts to examine the sodium salt using cobalt as a reagent were unsuccessful. Many dithiocarbamates form stable silver(1) salts but a trial using silver nitrate appeared to result in decomposition of the ethylenebisdithiocarbamate.February, 1951 CHROMATOGRAPHY USING TRANSITION-METAL SALTS TABLE I 131 CAPACITY FACTORS OF DITHIOCARBAMATES AND THIURAM DISULPHIDES ON LIQUID CHROMATOGRAPHY WITH NICKEL AND COBALT REAGENTS Solvent r A \ Methanol - water containing 0.1 yo Ni(I1) as sulphate r A \& 30+70* 50+50 60+40 70+30 60+40 70+30 Methanol - water containing 0.1% Co(I1) as nitrate Dithiocarbamate- N-H,- .. . . . . . . N-Methyl-? . . . . . . N-Ethyl- . . . . . . NN-Dimethyl- . . . . . . NN-Diethyl- . . . . . . NN-Tetramethylene- . . Thiuram disulphide- NNN’N’-Tetramethyl (thiram) NNN’N’-Tetraethyl (disulfiram) . . . . . . . . . . . . . . - - - - 0.25 - 11.8 1.25 0.71 0.50 1.25 0.57 - 7.25 2.71 1.37 3.37 1.28 - 10.6 3.85 2.25 4.12 1.57 - - >30 24.7 32.1 7.14 - - 26.5 11.8 - - 2.75 1.75 0.88 1.25 0.86 - 12.7 6.37 11.8 3.71 - * Methanol - 0.05 M phosphate buffer adjusted to pH 5.These results were obtained using methanol - water containing 0.1% Ni(I1) as sulphate: 50f50, K’ = 2.86; 40+60, k‘ = 10.4; 30+70, k’ = 25.4. In addition to the cobalt and nickel complexes, Schwedt also successfully chromatographed the complexes of copper, mercury and lead, although the last was unstable.19 As each of the complexes has a different capacity factor, it appeared that it would be possible by selection of the appropriate metal ion to adjust the retention times of dithiocarbamates, when they are present in a mixture, to give the optimum resolution.The complexes of most other metal ions including iron(III), manganese, silver, chromium(III), molybdenum and vanadium were found either to decompose or to give several bands.19 Although the reaction with diethyldithiocarbamate followed by extraction and spectro- photometry of the complex is a widely used assay for c ~ p p e r ( I I ) , ~ , ~ when copper nitrate was used as a reagent the results were erratic and reproducible peaks could not be obtained. In addition, a strong interaction between the copper ions and thiuram disulphides appeared to take place, presumably similar to the reactions that cause colour changes on mixing aqueous solutions of disulphides and copper(I1) ions,l whereas with nickel or cobalt ions any effect was minimal. In order to determine the product that was being formed on injection, a solution of copper( 11) ions was mixed with diethyldithiocarbamate and the precipitate was dissolved by the addition of methanol.The resulting solution has a Amax. at 386 nm, whereas if the copper - dithiocarbamate complex was extracted from the aqueous solution into chloroform and then diluted with methanol - water the absorbance of the solution had a A,,,,. at 434 nm. The latter band corresponds to the neutral ML, complex, whereas the former is very similar to the ML+ water-soluble complex formed with NN-diethanolarninodithiocarbamate and copper ions (MLf, Amax. at 380 nm, and ML,, Amax. 435 nm).25 Thus, when used as a reagent the copper(I1) ions, which are in excess, appear to form the MLf complex rather than the neutral ML, com- plex, which can be chromatographed.Samples of the ML, copper complex prepared by extraction could therefore be readily chromatographed. Mercury(I1) ions are reported to form the strongest complexes with diethyldithiocarbamate and will hsplace all other metal ions from complexes.26 However, attempts to chromatograph mercury( 11) diethyldithiocarbamate were unsuccessful and no peaks were obtained. Although the extracted complex is very stable, the equilibrium constant for the reaction 2HgL+ + HgL, + Hg2+ is 0.27 Therefore, the complex forms stepwise so that in the presence of an excess of mercury(I1) ions only HgL+ will be present. In contrast to mercury and copper, lead ions form much weaker complexes26 and it was felt that they would be a useful comparison.Schwedt found the complexes to be unstable,lg and in two papers reported very different retentions, relative to the other c o r n p l e x e ~ . ~ ~ ~ ~ ~ In this study an attempt to use lead nitrate as a reagent was unsuccessful, and the complex when pre- pared by extraction was not eluted on chromatography. A recent report has suggested that1 32 SMITH et al. : DITHIOCARBAXATES BY LIQUID Analyst, Vol. 106 th? lead complex decomposes in solution in isobutyl methyl ketone to give the free metal after 2-3 h2* although other papersB considered the complex to be stable under these conditions. Extraction of lead ions by ammonium tetramethylenedithiocarbamate was also found to fail with both high and low proportions of dithiocarbamate.28 The disulphides chromatographed essentially unaltered in the presence of lead ions.Two previous studies have examined in detail the separation of nickel(I1) complexes on silica gel columns8y12 and as in our degradation studies the cobalt complex gave a more suitable retention time relative to the other components in the mixture, a detailed examination of the use of cobalt(I1) was carried out. The pH of the aqueous solution was not critical and was usually not controlled except for the degradation studies.3 In the degradation studies, as the samples were in phosphate buffers at a range of pH values it was necessary to use a pH 5 phosphate buffer in the solvent to prevent precipitation of cobalt phosphate. In extraction studies both cobalt and nickel complexes are completely extracted between pH 2 and 12.4 Different concentrations of cobalt nitrate, 0.01,0.5 and O.l%, gave the same capacity factors and it is assumed that complex formation is essentially complete.As part of the degradation study a calibration graph was prepared for N-methyldithio- carbamate using caffeine as an internal standard (Table 11). The graph is linear from 100 to 1000 pg ml-l, but a small negative intercept, also found on other calibration runs, suggested a small but constant amount of decomposition. TABLE I1 CALIBRATION OF N-METHYLDITHIOCARBAMATE USING METHANOL - pH 5 PHOSPHATE BUFFER [0.1% CO(I1) AS NITRATE] (30+70) AS SOLVENT Peak height, scale units A I 7 Sample concentration*/ N-Methyl- Caffeine pg ml-l dithiocarbamatet (internal standard) t 100 6 183 200 14.5 183 300 23 183 400 31 177.5 500 40.5 177 600 49 178 700 58 179 800 65 178.5 1000 79 177 Slope .. .. .. . . 0.4682 Intercept . . . . . , -11.5 Correlation coefficient. . . . 0.9991 Standard deviationlpg ml-l . . 5.894 Ratio, sample x lo3 to internal standard 32 79 126 175 229 275 324 364 446 * 10-p1 injection of a sample solution containing caffeine (800 pg ml-l) in pH 7.0 phosphate buffer. t For caffeine k’ = 2.0 and for N-methyldithiocarbamate k’ = 18.8. In all the studies so far, only one dithiocarbamate was injected at a time. In view of the exchange of ligands found with nickel complexes the effect of injecting mixtures of two dithio- carbamates was determined (Table 111). Four peaks were obtained in each instance corre- sponding to a random formation of mixed complexes of cobalt(II1).The observation con- firmed that, as in the test-tube reaction between cobalt and dithiocarbamates, the cobalt(I1) reagent was being oxidised to yield the very stable cobalt(II1) complex. This reaction is reported to occur spontaneously by oxidation with atmospheric 0xygenl59~~ and explains the marked colour change from pink to green on complex formation. The mass spectrum of the extracted complex is reported to agree with the formation of a cobalt(II1) - tridithiocarbamate complex.15 The four peaks found in this work with cobalt ions can therefore be assigned to MX,, MX,Y, MXY, and MY, complexes. Although a mixture of tetramethyl- and tetraethylthiuram disulphides gave two peaks on chromatography immediately after mixing, on re-examination after standing for 90 min an intermediate peak, presumably the mixed NN-diethyl-N’N’-dimethylthiuram disulphide, was also present, an effect that had also been noted earlier31 (Table 111).February, 1981 CHROMATOGRAPHY USING TRANSITION-METAL SALTS 133 TABLE I11 CAPACITY FACTORS OF MIXED DITHIOCARBAMATE METAL COMPLEXES FORMED BY INJECTION OF TWO DITHIOCARBAMATES Dithiocarbamates Capacity factors r f A Solvent Sample (X) Sample (Y) MX, MX2Y MXY, 30+ 70* -NH2 -NHMe 0.25 1.37 4.90 50+ 507 -NHEt -NMe, 7.25 7.89 9.00 -NHMe' -NMe2 1.25 2.63 5.25 -NHMe -NHEt 1.25 2.50 4.12 70+ 307 -NHMe -N(CH,), 0.50 1.25 4.00 -NHEt -N(CH,), 1.37 2.87 5.87 -N(CH2) 4 -NEt, 11.8 15.0 19.2 -NHMe -NHEt 0.50 0.75 1.00 -NHEt -NEt, 1.37 2.75 10.0 -NMe2 -N(CH2), 2.00 3.62 6.75 I MY3 1.47 10.6 10.6 11.8 11.8 11.8 24.7 24.7 7.25 1.37 * 30+70 methanol - 0.05 M phosphate buffer (pH 5 ) containing 0.1% Co(I1) as nitrate.7 Methanol - water containing 0.1% Co(I1) as nitrate. If a disulphide and dithiocarbamate were mixed and injected (Table IV), interchange apparently also occurred, leading to all possible combinations of disulphides, mixed disulphides and mixed complexes, although some of the assignments are tentative as not all the possible standards were available. Whether this exchange occurs via oxidation, reduction and cleav- age of the S-S bond or by amine ex~hange3~ is not known. TABLE IV CAPACITY FACTORS FOR MIXTURES OF THIURAM DISULPHIDES AND DITHIOCARBAMATES Disulphide Dithio- Capacity factors of complexes A A r , carbamate I > Solvent* A-A B-B X A-A A-B B-B MX, MX,A MXA, MA, Other 50+50 -NMe - -NMe 2.75 10.6 70 + 30 -NMe -NEt - 0.75 6.25 (after 90 &in) (after 90 ',in) 0.75 2.25 6.25 -NMe: - -NHde 2.75 10.5 1.5t 3.0t 1.75 3.5 6.5 11.8 - 2.12 2.62 4.75 - 1.12 -NMe, - N W A - -NEt, -NMe, 6.25 - -NEt, -N(CH& 6.37 4.37t 3.0t 11.8 14.4 19.1 24.7 8.4 * Methanol - water containing 0.1% Co(I1) as nitrate.t A tentative assignment assuming that tetramethylenedithiocarbamate is yielding the corresponding disulphide (B-B) and a mixed disulphide (A-B). Hence, the use of nickel or cobalt ions as complexation reagents is a useful technique for the determination of individual dithiocarbamates but the interpretation of the results can be complicated if dithiocarbamates or thiuram disulphides with different amino functions are present in the sample.The method has been successfully applied to the determinations of N-methyldithiocarba- mate formed during the degradation of dazomet in vitro, in cell cultures and in bacterial cell cultures under the same chromatographic conditions used to monitor dazomet and methyl isothi0cyanate.~3 We thank the SRC for a studentship to R.L.M. and J. D. Campbell and Sons Ltd. and Robinson Bros. Ltd. for dithiocarbamates.SMITH, MORARJI AND SALT References 2. 3. 4. 5. 6. 7 . 8. 9. 10. 11. 12. 13. 14. 15. 16. 17. 18. 19. 20. 21. 22. 23 24. 25. 26. 27. 28. 29. 30. 31. 32. 33. Raizman, P., and Thompson, Q. E., in Karchmer, J. H., Edztor, “The Analytical Chemistry of Thorn, G. D., and Ludwig, R. A., “The Dithiocarbamates and Related Compounds,” Elsevier, Smith, R.M., Morarji, R. L., Salt, W. G., and Stretton, R. J . , Analyst, 1980, 105, 184. De, A, K., Khopkar, S. M., and Chalmers, R. A., “Solvent Extraction of Metals,” Van Nostrand Hulanicki, A., Talanta, 1967, 14, 131. Liska, O., Lehotay, J., Brandstetorova, E., and Guiochon, G., J . Chromatogr., 1979, 171, 153. Krupcik, J.. Leclcreq, P. A., Garaj, J . , and Masaryk, J., J . Chvomatogr., 1979, 171, 285. Liska, O., Guiochon, G., and Colin, H., J , Chromatogr., 1979, 171, 146. Lehotay, J . , Liska, O., Brandsteterova, E., and Guiochon, G., J . Chromatogv., 1979, 172, 379. Liska, O., Lehotay, J., Brandsteterova, E., Guiochon, G., and Colin, H., J . Chromatogr., 1979, 172, Moriyasu, M., and Hashimoto, Y., Anal.Lett., 1978, A l l , 593. Moriyasu, M., and Hashimoto, Y., Chem. Lett., 1980, 117. O’Laughlin, J . W., and O’Brien, T. P., Anal. Lett., 1978, A l l , 829. Edward-Inatimi, 1:. B., and Dalziel, J. A. W., Anal. Proc., 1980, 17, 40. Gaetani, E., Laurcri, C. F., and Mangia, A., A n n . Chim. (Rome), 1979, 69, 181. Bannister, S. J . , Sternson, L. A,, and Repta, A. J . , J . Chromatogr., 1979, 173, 333. Schwedt, G., Fresenius 2. Anal. Chem., 1977, 288, 50. Schwedt, G., Chromatographia, 1978, 11, 145. Schwedt, G., Chromatographia, 1979, 12, 289. Walters, V., and Raghavan, N. V., J , Chromatogr., 1979, 176, 470. Sternson, L. A,, and Dewitte, W. J . , J . Chromatogr., 1977, 137, 305. Cooke, N. H. C., Viavattene, R. L., Eksteen, R., Wong, W. S., Davies, G., and Karger, B. L., J . Karger, B. L., Wong, W. S., Viavettene, 13. L., Lepage, J. N., and Davies, G., J . Chromatogr., 1978, Vonach, B., and Schomburg, G., J . Chromatogr., 1978, 149, 417. Cullen, T. E., Anal. Chew., 1964, 36, 221. Eckert, G., Fresenius 2. Anal. Chew., 1957, 155, 23. Usatenko, Y. I., and Tulyupa, F. M., Russ. J . Inovg. Chem., 1959, 4, 1148. Subramanian, I<. S., and Meranger, J. C., I n t . J . Environ. Anal. Chem., 1979, 7, 25. Tessier, A., Campbell, P. G. C., and Bisson, M., I n t . J . Environ. Anal. Chem., 1979, 7, 41. Regenass, W., Fallab, S., and Erlenmeyer, H., Helv. Chim. Acta, 1955, 38, 1448. Johnson, E., “Liquid Chromatography a t Work,” Varian Associates, Palo Alto, Calif., 1977, No. 39. Facklcr, J . P., Coucouvanis, D., Seidcl, W. C., Masek, R. C., and Holloway, W., Chewz. Comwu.tz., 1967, Morarji, K. L., Salt, W. G., Smith, R. M. and Stretton, R. J., in preparation. Sulfur and Its Compounds,” Wiley-Interscience, New York, 1972, Part 11, p. 491. Amsterdam, 1962. Reinhold, New York, 1970. 384. Chromatogr., 1978, 149, 391. 167, 253. 924. Received J u l y 14th, 1980 Accepted September 17th, 1980

ISSN:0003-2654    

DOI:10.1039/AN9810600129

出版商:RSC    

年代:1981

数据来源: RSC

摘要:

Agzalyst, February, 1981, Vol. 106, p p . 135-145 135 Determination of Polynuclear Aromatic Hydrocarbons in Food, Water and Smoke Using High-performance Liquid Chromatography Neil T. Crosby, David C. Hunt, Lesley A. Philp and lndu Patel Department of Industry, Laboratory of the Government Chemist, Cornwall House, Stamford Street, London, SE1 9NQ A method for the separation of 17 different polynuclear aromatic hydro- carbons (PAHs) has been developed. After separation by thin-layer chro- matography, the PAHs are examined by high-performance liquid chromato- graphy using ODS and PPS stationary phases. The elution times of all 17 compounds relative to benzo[a]pyrene have been computed. Six representa- tive compounds have been added to bierwurst and to water to check recovery values.Work on stubble smoke has been confined to benzo[a]pyrene only. The method of extraction and clean-up varies with the substrate but the method of determination of the PAH compounds in the final extract is the same in all three cases. Confirmation of identity is obtained by measuring the peak-height ratios for each PAH a t two different excitation and emission wavelength combinations. The limit of detection of benzo[a]pyrene is 0.02 pg kg-1 in food, 0.3 ng 1-1 in water and 150pg per filter in smoke. Recoveries of the six representative compounds are generally in the range 75-100% a t levels approaching the limit of detection. Keywords : Higlz-performance liquid chronzatograplzy ; polynuclear aromatic hydrocarbons ; food ; water ; stubble smoke Polynuclear aromatic hydrocarbons (PAHs) occur in crude coal tar products and they are also formed during the pyrolysis of coal, oil and other forms of organic matter.Contamination of the environment with PAHs may occur through the discharge of industrial wastes, accidental spillages, combustion processes of industry, motor vehicle emissions and in food during preparation and cooking. Lo and Sandil have reviewed the occurrence of PAH residues in food, and Harrison et aL2 have paid particular attention to their distribution in water supplies. The identification and determination of these compounds in various parts of the environment, air, food, water, etc., is of interest as some of them are known to be strongly carcinogenic,3 whilst others are non- or only weakly carcinogenic.A number of analytical techniques have been employed for the extraction, separation and identification of PAHs in various different substrates. Chromatographic techniques have been widely used, especially thin-layer chromatography (TLC)4 and gas - liquid chr~matography.~ In recent years, however, high-performance liquid chromatography (HPLC) has been developed and this technique possesses certain advantages with its fast analysis times and ability to utilise directly certain molecular properties of the compounds of interest, such as ultraviolet absorption and fluorescence. Most workers using high-performance liquid chromatography in the field of PAH analysis have preferred to use reversed-phase columns, particularly octa- decylsilane (ODS) .6-g There is also an increasing tendency to use fluorescence detectors.*-ll Generally a number of PAH compounds, not all of which are carcinogenic, will be present concomitantly.Hence the analytical method must be capable of separating and identifying minute amounts of individual compounds in a complex mixture. Methods are proposed for the separation and determination of PAHs in foodstuffs, water and smoke. For water samples attention has been confined to the separation and measurement of the six compounds listed by the World Health Organization (WHO) as indicators of pollution.12 Additionally, some samples of smoke produced during controlled straw stubble burning were analysed for benzo [ajpyrene only. Although the sample preparation and clean-up necessarily require The food method is capable of separating 17 different PAH compounds.Crown Copyright.136 CROSBY et al. : PAH IN FOOD, WATER Analyst, Vol. 106 different approaches for the three different types of sample, the eventual separation and measurement of the PAHs in the final extract utilises the same HPLC system under identical conditions. Fig. 1 shows a flow diagram of the different procedures used for the three types of sample. The methods as presented in this paper will detect 0.02 pg kg-l in food, 0.3 ng 1-1 in water and 150 pg of benzo[a]pyrene when deposited on a filter. final extract 4 in methanol , Experimental Foods The extraction procedure is essentially that described by Hanus et ~ l . , ~ followed by clean-up using a silica gel Sep Pak cartridge. The PAHs are then separated into two groups by TLC on acetylated cellulose.After recovery from the TLC plate, each group is subjected to HPLC. Detection is by fluorescence at two different wavelength combinations. Reversed-phase I HPLC with Concentrate, fluorescence final extract detection in methanol Reagents Unless stated otherwise, all A cet onitrile. Acetylated cellulose for TLC Celite 545. Diethyl ether. Ethanol. solvents are re-distilled over charcoal. (21 yo acetate). 3 Lead acetate reagent. ml of acetic acid and dilute to 1 1 with water. Methanol. Analytical-reagent grade. .Potassium hydroxide. Analytical-reagent grade. Sodium sulphate (granztlar, anhydrous). Silica gel. Toluene. Analytical-reagent grade. 2,2,4-Trimethylpentane. Dissolve 200 g of lead(I1) acetate trihydrate in warm water, then add Heat for 5 h at 300 "C.Sep Pak from Waters Associates Ltd. Procedure blend for 3 min with a top-drive macerator. for a further 3 min at high speed. Weigh 100 g of homogeneous sample into a 500-ml beaker, add 275 ml of acetonitrile and Add 25 ml of the lead acetate reagent and blend Saponify with alcoholic KOH Clean-up on Sep-Pa k Preparative TLC v I Concentrate. I on polythene Soxhlet extraction Precipitate polythene with methanol Clean-up by preparative HPLCFebruary, 1981 AND SMOKE USING HPLC 137 Filter the mixture with suction through a Whatman No. 541 filter-paper into a Biichner flask. Collect 150 ml of filtrate plus the equivalent volume of the calculated water content of 50 g (based on a previously dried sample) in a measuring cylinder.The total volume of extract will now be equivalent to 50 g of the original sample. Transfer the collected filtrate to a 500-ml separating funnel and shake vigorously with 50 ml of 2,2,4-trimethylpentane for 30 s. Add 200 ml of distilled water and 30 ml of saturated sodium sulphate solution and shake for 2 min. Allow the layers to separate then transfer the lower aqueous phase into a second 500-ml separating funnel. Pass the organic extract through a 3-cm layer of anhydrous granular sodium sulphate, contained in a filter-paper and funnel, into a 250-ml round-bottomed flask. Partition the aqueous phase in the second separating funnel with 50ml of 2,2,4- trimethylpentane. Allow to separate then discard the aqueous layer, and pass the organic phase into the first separating funnel to rinse it, then pass through the same sodium sulphate layer into the 250-ml flask.Reduce the combined solvent in the flask nearly to dryness using a rotary evaporator at 60 "C, and remove the last traces of solvent with a stream of nitrogen. Add 25 ml of ethanol, 1.5 g of potassium hydroxide and a few boiling chips to the residue. Fit a reflux condenser and boil under reflux for 40 min. Transfer the solution whilst still hot into a 250-ml separating funnel and rinse the flask successively with 30ml of hot ethanol followed by two rinses with 50 ml of distilled water, adding each in turn to the separating funnel. Add 25 ml of saturated sodium sulphate solution, mix thoroughly and allow the contents to cool. Rinse the flask finally with 50 ml of 2,2,4-trimethylpentane, add this to the separating funnel and shake the mixture for 2 min.Allow the layers to separate and transfer the lower aqueous layer to a second 250-ml separating funnel. Wash the organic phase in the first separating funnel by swirling with two 100-ml portions of distilled water. Discard the aqueous portions in turn and pass the organic portion through a 3-cm layer of anhydrous granular sodium sulphate contained in a filter-paper and funnel, into a 250-ml flask. Partition the reserved aqueous phase contained in the second separating funnel with 50 ml of 2,2,4- trimethylpentane. Discard the aqueous phase, transfer the organic phase to the first separat- ing funnel, rinse and then pass the solvent through the same sodium sulphate layer into the 250-ml flask.Rinse the two separating funnels successively with 30 ml of 2,2,4-trimethyl- pentane, finally passing the solvent through the sodium sulphate layer into the 250-ml flask. Concentrate the combined extracts to 2 ml on the rotary evaporator. Connect a silica gel Sep Pak to a suitable syringe, then wash it with 10 ml of 2,2,4-trimethylpentane. Add the sample extract to the Sep Pak, rinse the flask with two 2-ml portions of 2,2,4-trimethylpentane and add each of the washings to the Sep Pak. Discard all the eluate from the Sep Pak up to this point. Elute the PAHs with 20 ml of 2,2,4-trimethylpentane followed by 20 ml of acetonitrile - 2,2,4-trimethylpentane (2 + 98). Collect the eluent in a 100-ml pear-shaped flask, con- centrate to about 200 pl and apply the extract as a 10-cm streak to a TLC plate of acetylated cellulose prepared as described below.Rinse the flask with 1 ml of 2,2,4-trimethylpentane, concentrate and transfer to the TLC plate. Thin-layer chromatography Thin-layer chromatography plates (20 x 20 cm) are prepared by thoroughly mixing 50 g of 21% acetylated cellulose for TLC (Schleicher and Schiill) with 100 ml of 95% ethanol. Spread the plates to a thickness of 0.5 mm. Allow the plates to dry in air for 1 h and then store in a desiccator cabinet until required. Place the development solvent of ethanol - toluene - water (17 + 4 + 4) in a lined tank to equilibrate for 30 min. At the start line on the TLC plate at a distance of 2.5 and 3 cm from both ends of the sample streak, spot standards of fluoranthene and anthanthrene.When the TLC plate has fully developed examine it under ultraviolet light to establish the position of the PAH standards. Using the edge of a spatula, scribe straight lines on the plate as follows: (i) Horizontal line at 0.5 cm above the start line.138 CROSBY et al.: PAH IN FOOD, WATER (iv) Horizontal line at R, 0.6. r I I I I I I I I I I I I I I I I I I I I I T - - - - - 4 ------- r---------- ---- I I I I I C I I I I 3 -----------___I- _ _ _ _ _ - I I I I I B I I I I I I A I I I I I v 2-.- - - - - - - - - - - - - - - - - - - - - - - - - - l m I Analyst, Vol. 106 Solvent front RF = 0.6 Fluoranthene Antha nth rene (v) Vertical lines between the sample and standards. Fig. 2 illustrates the scribing of the developed plate.Scrape off the side portions of the plate containing the standards and discard. Similarly, remove and discard the lower and upper portions of the plate. Designate the three remaining portions of the plate as A, B and C in ascending order then remove and bulk portions A and C into a 10-ml stoppered test-tube. Similarly transfer portion B into a second stoppered test-tube. To the combined A and C portions add 5 ml of diethyl ether and shake for 1 min, or use an ultrasonic bath. Set the tube aside for the powder to settle. Prepare a small filter column by plugging the first 1 cm of the shoulder of a long-form Pasteur pipette with glass-wool, followed by a 1-cm layer of Celite. Wash the column with 10 ml of diethyl ether before use. Transfer the supernatant solvent in the test-tube to the small column and collect the filtered extract in a 25-ml pear-shaped flask.Repeat the extraction and filtering procedure a further three times to use a total volume of 20 ml. Extract the TLC powder from portion B separ- ately, in a similar manner. Concentrate each portion to about 1 ml and transfer quantitatively to two small vials and evaporate off the diethyl ether using a stream of nitrogen. Re-dissolve each residue in 500 pl of methanol ready for the injection of 20 pl on to the HPLC system (see below). Carry out a blank determination omitting the sample, and subtract the result from the sample results . I I 1 5 Fig. 2. Scribing of the acetylated cellulose TLC plate. Results Table I shows the R, values of 17 PAHs separated on acetylated cellulose under the condi- tions of the method.Table I1 gives the recovery of six PAHs from a sample of spiked bier- wurst, whilst Table I11 gives three results from a survey of 46 samples of a wide range of smoked foods purchased in the UK. For spiking the bierwurst the PAH compounds were dissolvedFebruary, 1981 AND SMOKE USING HPLC TABLE I THIN-LAYER CHROMATOGRAPHY OF PAH COMPOUNDS ON 21 yo ACETYLATED CELLULOSE 139 PAH Renzo[b]fluoranthene . . Indeno[l,2,3-~d]pyrene . . Anthanthrene . . . . Fluoranthene . . . . Benzo[ghi]fluoranthene . . 9,l O-Dimethylanthracene Benzo[a]pyrene . . . . Pyrene . . . . . . RF Position 0.16 A 0.16 A 0.20 A 0.21 A 0.41 C 0.41 C 0.42 C 0.44 C PAH RF 7-Methylbenz[a]anthracene 0.28 12-Methylbenz[a]anthracene 0.29 Benzo[k]fluoranthene .. 0.29 Dibenz[a,h]anthracene . . 0.30 3-Methylcholanthrene . . 0.30 Perylene . . . . . . 0.33 Benzo [ghilperylene . . 0.33 Benzo[e]pyrene . . . . 0.36 Benz [a] an thracen e . . 0.36 Position n €3 13 €3 B B €3 B I3 in methanol and the required volume was added from a syringe in droplets to the minced bierwurst. I t was then left for at least 1 h in a stream of air to remove methanol. Levels of fluoranthene found were significantly higher in most foods than the levels of the other PAH compounds detected. However, smoking of foods by modern controlled processes appears to add only very low levels of carcinogenic PAH compounds to the diet. Water of sampling to the analytical laboratory. sample with 2,2,4-trimethylpentane. HPLC system consisting of a reversed-phase column and fluorescence detector.This method has been designed for water samples which need to be transported from the site The extraction procedure consists of shaking the After concentration, the extract is injected on to the Reagents 2,2,4-Trinzethylpentane. Re-distilled over charcoal. Sodium sulphate (anhydrous, granular). Methanol. Analytical-reagent grade. Heat at 300 "C for 5 h. TABLE I1 RECOVERY OF PAH COMPOUNDS FROM SPIKED BIERWURST PAH Spiking level/pg kg-1 Recovery, % Fluoranthene . . . . .. Benzo [klfluoranthene . . .. Benzo [blfluoranthene . . .. Benzo[a]pyrene . . . . .. Indeno[l,2,3-~d]pyrene . . .. Benzo [ghilperylene . . .. 2.4 0.8 2.0 2.0 40 10 70 110 110 100 95 95 TABLE I11 LEVELS OF PAH COMPOUNDS FOUND IN THREE DIFFERENT SMOKED FOODS PAHI pg kg-l A r \ Benzo[k]- Benzo[b]- Benzo[a]- Indeno[l,2,3- Benzo[ghi]- cdlpyrene perylene Food Fluoranthene fluoranthene fluoranthene pyrene Bacon 7.8 0.05 0.30 0.05 2.5 3.75 Kippers 2.4 0.10 0.35 0.10 2.7 4.3 Cheese 4.2 0.15 0.30 0.20 ND* 0.60 * ND = less than 0.02 p g kg-l.140 Analyst, Vol.106 Procedure Prepare the bottle for collecting the water sample as described below. To a 2.5-1 brown- glass Winchester bottle (such as an empty methanol reagent bottle) fitted with a screw-cap and a PTFE insert, add 2 1 of tap water and mark the bottle at the meniscus. Discard the water, rinse the bottle with distilled water, then methanol, then finally with 2,2,4-trimethylpentane and invert the bottle to dry. Add exactly 100 ml of 2,2,4-trimethylpentane to the bottle and screw the cap on firmly.The bottle is now ready for collecting the water sample. The water sample is poured into the bottle until the lower meniscus reaches the 2-1 mark, then the bottle is tightly sealed. The sample can now be transported or left for a number of days before the analysis takes place with little loss of PAH compounds. CROSBY et al. : PAH IN FOOD, WATER Extraction If the sample appears cloudy add 10 g of sodium sulphate; if the water is clear this is not necessary. Shake the bottle on a laboratory shaker for 10 min, then transfer the contents into a 2-1 separating funnel. Allow the two layers to separate and run off the lower, water layer to waste. Rinse the sample bottle with 20 ml of 2,2,4-trimethylpentane, add the rinsings to the solvent in the separating funnel and mix.Filter the bulk of the solvent extract through a 10-g layer of anhydrous sodium sulphate contained in a Whatman No. 1 paper in a filter-funnel. Collect 60 ml of the filtered extract (= 1 1 of water sample), concentrate it nearly to dryness on a rotary evaporator at 60 "C, then transfer it quantitatively into a small vial and evaporate to dryness using a stream of nitrogen. Re-dissolve the residue in 200 pl of methanol ready for the injection of 20 p1 on to the HPLC system (see below). Development of the method The method was tested by the addition of a solution in methanol, of the six PAH compounds listed in Table 11, to water taken from the River Thames at Waterloo, where the river is tidal and well below the last point where it is abstracted for use as drinking water.Immediate extraction of filtered and spiked samples gave satisfactory recovery figures. After storage for 3 d in glass bottles, recoveries were only 59431%; if 2,2,4-trimethylpentane was present at the time of sampling, however, the recoveries after 3 d were 74-96%. These results are shown in Table IV and indicate that PAHs can be adsorbed on to a glass surface. Further confirmation of this was obtained when spiked tap water was examined using a membrane filter concentra- tion step when recoveries of 60-S5% were obtained, dropping to 45-60% after 24 h in a glass bottle. The decision whether to examine a filtered or unfiltered sample of water should be made before the bottle is filled, as the removal of particulate matter can considerably reduce the PAH level.Table V shows the effect on PAH recoveries of both filtering and adding 2,2,4- trimethylpentane at the sampling stage. Work carried out on sampling procedures indicates TABLE IV RECOVERIES OF PAHS FROM FILTERED WATER AFTER 3 DAYS IN GLASS PAH Solvent treat men t Fluoranthene . . . . . . Without solvent Benzo[k]fluoranthene . . .. Benzo[b]fluoranthene . . .. Benzo[a]pyrene . . . . Indeno[1,2,3-cd]pyrene . . Benzo [ghilperylene . . .. Fluoranthene . . .. . . With solvent Benzo [klfluoranthene . . . I Benzo [blfluoranthene . . .. Benzo[a]pyrene . . .. Indeno[l,2,3-cd]pyrene . . Benzo [ghilperylene . . .. Original PAH Final determination/ added/ determination/ Recovery, ng 1-1 ng 1-l % ng 1-' 10.2 0.7 0.2 1.8 2.9 Not detected 15.3 1.3 0.3 4.2 4.4 Not detected 50 4 1 10 20 15 50 4 1 10 20 15 41.5 3.1 0.9 9.6 12.0 10.6 48.3 4.4 1.1 13.6 17.6 15.9 69 66 75 81 60 59 74 83 81 96 88 82February, 1981 EFFECTS AND SMOKE USING HPLC TABLE V OF FILTRATION AND SAMPLING PROCEDURE ON RIVER THAMES WATER PAH concentration/ng 1-1 141 I 3 Filtered Unfiltered f A > r A PAH No solvent With solvent No solvent With solvent Fluoranthene .. .. .. . . 10.2 15.3 540 667 Benzo[k]fluoranthene . . . . . . 0.7 1.3 85 99 Benzo[b]fluoranthene . . .. . . 0.2 0.3 17 20 Benzo[a]pyrene . . . . .. 1.8 4.2 294 430 Indeno[1,2,3-cd]pyrene . . . . Not detected Not detected Not detected Not detected Benzo [ghilperylene . . .. . . 2.9 4.4 500 54 1 that those which are based on concentration steps involving a degree of filtration, such as the on-column concentration methods and the membrane methods, may not be appropriate for unfiltered samples or for those samples requiring transportation from the sampling site, as the PAHs will be partially retained by particulate matter and by the glass container.With polluted water emulsions may form when the sample is shaken with the solvent. If sodium sulphate is added after the emulsion has formed the emulsion can still be difficult to break, whereas the addition of the salt before the extraction starts will prevent the emulsion from forming. Peaks were obtained corresponding to indenopyrene but the ratio was 1.0 instead of 0.1 as found for standards. Other PAHs detected possessed ratios very close to standard values. Hence, the presence of indenopyrene was not confirmed.This observation was supported by further measurements at 355 nm excitation and 500 nm emission. Benzo[a]pyrene in Smoke Interest has been expressed in the possible formation of benzo [alpyrene during the burning of straw and stubble. Experiments were designed by AERE, Harwell, in which straw was burned under controlled conditions in a wind tunnel. Samples were taken in a high-volume sampler with a glass-fibre filter backed by two bubblers filled with hexane. Approximately 0.5-1 m3 of air was sampled at each burn. In later experiments, the concentration of benzo[a]- pyrene was determined as a function of particle size using an Anderson cascade impactor by which samples are collected at several stages on thin polythene sheets.The results of the stubble burning study have now been p~b1ished.l~ Analysis of samples The hexane extracts presented no problems; they were evaporated to dryness and the residue was re-dissolved in methanol before examination by HPLC (see below). As charcoal is a very powerful adsorbent for PAHs, difficulties were envisaged for the recovery of benzo[a]pyrene from filters ingrained with soot from the smoke. Recovery experiments were carried out to test the efficiency of various solvents for extracting benzo[a]- pyrene from activated charcoal, as this material should represent the most difficult case en- countered with the samples. Glass-microfibre filter-papers were used for a series of experi- ments in which activated charcoal was spread over the surface, then ground in by rolling with a glass rod.Excess of charcoal was removed by shaking the papers. A 2 pg p1-1 standard solution of benzo[a]pyrene (50 pl) was then spotted evenly over the surface using a syringe and the paper was then allowed to dry in air. Soxhlet extraction was carried out on the impreg- nated filters by placing them in pre-extracted thimbles and using 150-200 ml of solvent for 7 h. The solvents selected for trial were benzene, cyclohexane, acetone and dichloromethane, these solvents having been used in the past for the extraction of PAHs from air samples.14-17 After 7 h, the extracts were evaporated and re-dissolved in methanol for quantitation by HPLC. Table VI shows the recoveries obtained for each solvent and the superiority of benzene (with a mean recovery of 88%) over the other solvents for this particular work.142 CROSBY et al.: PAH IN FOOD, WATER Analyst, Vol. 106 TABLE VI RECOVERY OF BEN20 [a]PYRENE FROM CHARCOAL-IMPREGNATED FILTERS Solvent Recovery, % Mean recovery, % Benzene . . .. .. 100,82,80,88 88 Cyclohexane . . . . 5, 5, 6, 6 6 Acetone . . .. . . 17,17, 8,15 14 Dichloromethane . . .. 4, 5, 6,20 9 Reagents All solvents should be re-distilled over charcoal. Benzene. Caution-Benzene is highly toxic and appropriate precautions should be taken. A cetonitrile. Methanol. 2,2,4-Trimethylpentane. Procedure Extract the sample of smoke-impregnated glass-fibre filter-paper in a Soxhlet apparatus for 7 h with benzene. Concentrate the extract to a low volume using a rotary evaporator at 60 O C , quantitatively transfer it into a small vial and take just to dryness with a stream of nitrogen.Re-dissolve the residue in 200 p1 of acetonitrile and inject 100 pl into a semi-preparative HPLC column composed of a 300 x 7 mm i.d. stainless-steel tube packed with Spherisorb ODs, 5 pm, using methanol - water (9 + 1) as the mobile phase at a flow-rate of 2 ml min-1. Then collect the portion of the eluate containing benzo [alpyrene, previously determined by the injection of a standard, in a stoppered test-tube. Add an equal volume of distilled water and mix the contents. Extract the benzo[a]pyrene by shaking with half the original volume of 2,2,4-trimethylpentane. Repeat the extraction and concentrate the combined extracts just to dryness, re-dissolve in 100 p1 of methanol and inject 20 pl on to the HPLC system (see below).If samples of smoke are collected on an Anderson impactor using polythene discs, then the same extraction procedure is used with a minor modification. A t the end of the Soxhlet extraction period the polythene will have partially dissolved in the benzene and is removed by precipitation by the addition of 100 ml of methanol, followed by filtration through a Whatman No. 40 paper prior to the concentration stage. Fig. 3 shows an extract of smoke subjected to semi-preparative HPLC on an ODS column together with the final determination of benzo [alpyrene on a PPS analytical column. ' 2 100 pI extract 1 I - 20 10 0 10 0 Timelmin \ extract I/ I 1 I , 1 1 1 1 , , , 1 0 8 6 4 2 0 8 6 4 2 0 Time/min Fig. 3. High-performance liquid chromatogram of smoke extract on (a) preparative column of ODS, and (b) benzo[a]pyrene portion from (a) on an analytical column of PPS.February, 1981 AND SMOKE USING HPLC 143 High-performance Liquid Chromatography In recent years the separation and detection of PAHs by HPLC has been effected primarily with bonded phases using a fluorescence detector, which imparts both selectivity and high sensitivity to the system. In this work two different bonded phases, octadecylsilane (ODS) and phthalimidopropylsilane (PPS) , have been examined.HPLC columns containing these two packings give different orders of elution for many PAHs and their preparation and per- formance have been compared in a previous paperls using methanol - water (9 + 1) as the mobile phase.Additionally, PPS has been used with a non-polar solvent in a normal-phase system to give a separation of the six PAHs designated by the WHO, found in water,lg which is superior to that obtained using ODS or other similar systems. A wide-range filter instrument has proved useful for general surveillance work, whilst a more expensive double monochromator scanning instrument, which can be programmed to give a number of pre-set wavelength combinations, has been used as a further aid to PAH identification. Two types of fluorescence detector have been examined during this work. A pparalus Services Ltd., Model 750/03, reciprocating pump. HPLC pumps. Waters Associates M6000 reciprocating pump and Applied Chromatography Detectors. The HPLC system now in routine use consists of a dual-piston reciprocating pump supplying the mobile phase of methanol - tetrahydrofuran - water (20 + 2 + 3) at 1.5 ml min-l through a Rheodyne valve fitted with a 20-pl loop, to a 15 cm x 4.6 mm i.d. stainless-steel column of either ODS or PPS bonded to LiChrosorb and Partisil5, respectively.The eluate is monitored at two wavelength combinations and the ratio of the peak heights of each PAH at the two wavelength combinations lends a further aid to identification. The settings used for the six PAH compounds listed by the WHO as indicators for water pollution are excitation a t 290 nm and emission at 430 nm (Al), and excitation at 282 nm and emission at 457 nm (A2). Compounds are identified both by their elution times and by comparison of the values of the ratios of peak heights h,/h2 with standards.If necessary further confirmation can be obtained by using the alternative column packing and by scanning the peak. Fig. 4 shows the chroma- tograms obtained for the six WHO polynuclear compounds at the two different wavelength combinations stated in the above method and the ratio of peak heights obtained. Table VII lists the elution times of 17 PAHs relative to benzo[a]pyrene on the two columns, and their Aminco FluoroMonitor and Perkin-Elmer 3000 Fluoresence Spectrometer. TABLE VII ELUTION TIMES RELATIVE TO BEN20 [a]PYRENE OF PAHS ON TWO REVERSED-PHASE COLUMNS AND THEIR PEAK-HEIGHT RATIOS AT TWO WAVELENGTH COMBINATIONS PAH Benzo[b]fluoranthene . . .. . . Benzo[a]pyrene . . . . .. .. Indeno[1,2,3-~d]pyrene .. .. .. Anthanthrene . . . . .. .. Fluoranthene .. . . .. . . Benzo[ghi]fluoranthene . . .. . . 9,lO-Dimethylanthracene . . .. . . Pyrene . . .. .. .. . . 7-Methylbenz[a]anthracene . . . . 12-Methylbenz [alanthracene . . .. Benzo[R]fluoranthene . . .. .. Dibenz[a,h]anthracene . . .. .. 3-Methylcholanthrene . . . . .. Perylene . . . . .. . . .. Benzo [ghi]perylene . . .. . . .. Benzo[e]pyrene . . .. . . .. Renz[a]anthracene . . . . . . .. Relative elution times* (benzo[a]pyrene = 1.00) f A \ Peak-height A, : A, ODS on LiChrosorb PPS on Partisil ratio? 0.86 1.00 1.42 1.76 0.42 0.60 0.52 0.45 0.81 0.80 0.90 1.24 1.85 0.84 1.34 0.81 0.60 0.84 1.00 1.32 1.63 0.52 0.83 0.75 0.61 0.82 0.72 0.78 1.00 1.11 1.21 1.68 1.07 0.64 1.3 3.8 0.1 3.4 0.5 0.9 1.5 6.4 3.6 2.3 3.4 4.8 2.7 0.8 3.3 3.7 1.8 * Mobile phase = methanol - tetrahydrofuran - water (20 + 2 + 3) t For values of A, and A, see text.144 CROSBY et al.: PAH IN FOOD, WATER Analyst, Vol. 106 Time/min Fig. 5. Comparison of high-performance liquid chromatograms of mussels with different detection systems: (a) ultraviolet a t 280 nm and 0.01 a.u.f.s. and (b) fluorescence a t 10% of full sensitivity. A, Benzo[a]- pyrene peak.February, 1981 AND SMOKE USING HPLC 145 peak-height ratios at the two wavelength combinations. Fig. 5 illustrates the advantages of sensitivity and selectivity of fluorescence detection over ultraviolet detection for the determina- tion of PAHs in mussels. We thank the Government Chemist for permission to publish this paper and AERE, Harwell, for permission to publish some details of the stubble-burning study. 1 . 2. 3. 4. 5. 6. 7. 8. 9. 10. 11. 12. 13. 14. 15. 16. 17. 18. 19. References Lo, M.-T., and Sandi, E., Residue Rev., 1978, 69, 35. Harrison, R. AT., Perry, R., and Wellings, R. A., Water Res., 1975, 9, 331. Hartwell, J ., “Survey of Compounds Which Have Been Tested for Carcinogenic Activity,” Publica- Howard, J . W., Teague, R. T., Jr., White, R. H., and Fry, B. E., Jr., J . Assoc. 08. Anal. Chem., Grimmer, G., and Bohnke, H., J . Assoc. 08. Anal. Chem., 1975, 58, 725. Schmit, J. A., Henry, R. A., Williams, R. C., and Dieckman, J . F., J . Chromatogr. Sci., 1971, 9, 645. O’Hara, J . R., Chin, M. S., and Kilbuck, J. H., J . Food Sci., 1974, 39, 38. Wheals, B. B., Vaughan, C. G., and Whitehouse, M. J., J . Chromatogv., 1975, 106, 109. Hanus, J. P., Guerrero, H., Riehl, E. R., and Kenner, C. T., J . Assoc. OH. Anal. Chem., 1979, 62, 29. Pellizzari, E. D., and Sparacino, C. M., Aqtal. Chern., 1973, 45, 378. Vaughan, C. G., Wheals, B. B., and Whitehouse, M. J.. J . Chromatogr., 1973, 78, 203. ‘‘U’orld Health Organization International Standards for Drinking Water,” World Health Organiza- Atkins, D. H. F., Wiffen, R. D., Healy, C., and Tarrant, J . B., Rep. U.K. Atom. Energy Aztth., Abdoh, Y., Aghdaie, N., Darvich, M. R., and Iihorgami, M. H., Atmos. Environ., 1972, 6, 949. Liberti, A., Cartoni, G. P., and Conthuti, V., J . .Chromatogr., 1964, 15, 141. Cooper, R. L., Analyst, 1954, 79, 573. Sawacki, E., Stanley, T. W., Elbert, W. C., Meeker, J., and McPherson, S., Atnzos. Environ., IMi, Hunt, D. C., Wild, P. J., and Crosby, N. T., J . Chvomatogv., 1977, 130, 320. Hunt, D. C., Wild, P. J., and Crosby, N. T., Water Res., 1978, 12, 643. tion No. 149, Public Health Service, Washington, D.C., 1951. 1966, 49, 596. tion, Geneva, 1971. AERE-9426, 1979. 1, 131. Received May 5th, 1980 Accepted September 8th, 1980

ISSN:0003-2654    

DOI:10.1039/AN9810600135

出版商:RSC    

年代:1981

数据来源: RSC

摘要:

146 Analyst, February, 1981, Vol. 106, pp. 146-152 Voltammetric Determination of 2-, 3- and 4-Chloroaniline in Mixtures John P. Hart Department of Orthopaedics, Charing Cross Hospital, Fulham Palace Road, London, W6 8RF Malcolm R. Smyth Institute of Chemistry, Institute 4: Applied Physical Chemistry, Nuclear Research Centre (KFA ), D-5170 Julich, Federal Republic of Germany and W. Franklin Smyth Department of Chemistry, University College, Cork, Republic of Ireland The oxidative voltammetric behaviour of 2-, 3- and 4-chloroaniline has been investigated a t the glassy carbon electrode over the pH range 1-12. The optimum pH values for the linear-sweep and differential-pulse voltammetric determination of these compounds were found to be 1.95 for 2- and 3-chloro- aniline and 8.1 for 4-chloroaniline.Differentiation of the three compounds in a mixture was achieved by using a combination of high-performance liquid chromatography with voltammetric detection. This method could be used to determine concentrations of 2- and 4-chloroaniline down to 2 ng and 3- chloroaniline down to 1 ng for a 20-4 injection on to the chromatographic column. Keywords : 2-, 3- and 4-chloroaniline determination ; Oxidation ; Jaigh- per foriiiance liquid clzroiizatography ; voltammetric detection Chlorine-substituted phenylamides are widely used as herbicides and have been found to metabolise in vivo to monochloroanilines.1-6 For example, isopropyl N-( 3-chloropheny1)- carbamate is converted into 3-chloroaniline through microbial action in the soil .2-6 N - Demethylation of 4-chloro-N-methylaniline has also been reported.’ For ex- ample, gas chromatography has been used to separate mixtures of mono- and dichloroanilines8-lO and thin-layer chromatography with fluorimetric detection has been used to determine the hydrolysis products of carbamate- and urea-containing herbicidesll However, both methods are time consuming.Colorimetric methods have also been used in this regard, but these usually involve derivatisation (by diazotisation) and are generally non-specific. Monochloroanilines can be oxidised at a variety of solid electrode surfaces and the potential of oxidation has been shown to be dependent on the nature of the substitution in the ring system.12 It was the purpose of this investigation to study the inherent voltammetric be- haviour of 2-, 3- and 4-chloroaniline at the glassy carbon electrode with a view to developing a method for their determination using differential-pulse voltammetry.However, this tech- nique was found not to have the required selectivity, and a method was therefore developed for their determination using a combination of high-performance liquid chromatography (HPLC) with voltammetric detection. This method has recently been applied to the deter- mination of many amino- and phenolic-containing compounds of biological importance.13914 Since acceptance of this paper another paper dealing with the determination of 2-chloraniline by HPLC with voltammetric detection has been p~b1ished.l~ This paper demonstrates the advantageous possibilities of using this technique in environmental analysis.A variety of methods exist for the determination of these monochloroanilines. Experimental Apparatus Linear-sweep, differential-pulse and cyclic voltammograms were recorded using a PAR, Model 174A, Polarographic Analyser in conjunction with a Servoscribe, Model 1s 541.20, potentiometric recorder. A three-electrode cell system was employed that incorporated a glassy carbon indicator electrode (Tokai GC-305 ; calculated area 0.39 cm2), a saturated calomelHART, SMYTH AND SMYTH 147 reference electrode and a platinum counter electrode. A Spectra-Physics, Model 3500, liquid chromatograph was operated in conjunction with a Metrohm, Model EA 1096, electrochemical detector, the potential of which was controlled by the PAR 174A.The detector was operated in the amperometric mode with glassy carbon as the material of both the indicator and counter electrodes and silver - silver chloride as the reference electrode. Procedures Solutions of the monochloroanilines were prepared by dissolving the compounds in either 0.1 N sulphuric acid or in Britton - Robinson buffers of pH 1.95-12.0 containing 50% V/V of ethanol, so that the final concentration was approximately 5 x M. Linear-sweep voltam- metry was carried out on these solutions using the following operating conditions: initial potential, 0.0 V; scan rate, 10 mV s-l; and current range, 0.02 mA. Between successive runs, the indicator electrode was cleaned by washing it with distilled water and drying it with a tissue. In the cyclic voltammetric experiments a scan rate of 50 mV s-l and a chart speed of 60 cm min-l were employed.Following selection of the optimum pH for the determination of the individual mono- chloroaniline, calibration graphs were constructed using the linear-sweep and differential- pulse modes. A scan rate of 50 mV s-l was employed in the former instance and 5 mV s-1 in the latter. The pulse height was 100 mV, the “drop time” 1 s and the time constant 0.3 s for different ial-pulse voltammetry. For the construction of chromatovoltammetric curves, the current arising from the oxidation of 2-, 3- or 4-chloroaniline following HPLC separation was plotted as a function of the applied potential (in the range 0.6-1.2 V), which was increased stepwise (50 mV) over a series of injections.The separation was carried out on a 5-pm LiChrosorb RP-8 column (250 x 4.6 mm i.d.) with a water - methanol - acetonitrile mixture (75 + 20 + 5) containing sodium dihydrogen orthophosphate (6 g 1-l) and 85% orthophosphoric acid (2.5 ml l-l) as eluting agent. For analytical determinations, the column was operated at a flow-rate of 2 ml min-l and the compounds were detected arnperometrically a t an applied potential of +1.2 V us. silver - silver chloride. Results and Discussion Linear-sweep Voltammetry range 1-12. these chloroanilines using linear-sweep voltammetry is shown in Fig. 1. Linear-sweep voltammograms of 2-, 3- and 4-chloroaniline were recorded over the pH The variation of E, and i, with pH for the single peak (peak A) exhibited by The first break in Fig.1. Variation of E , (solid lines) and i, (broken lines) with pH for 2-, 3- and 4-chloroaniline.148 HART et al. : VOLTAMMETRIC DETERMINATION OF 2-, Analyst, VoZ. 106 each of the respective E, veysus pH curves corresponds to the pKa value relating to deprotona- tion of the anilinium species, and although the values obtained using this method are higher than those obtained using spectrophotometry (Table I), the trend of increasing pKa values in the series 2-chloroaniline < 3-chloroaniline < 4-chloroaniline is maintained. It is interesting that at pH < pKa, 4-chloroaniline behaves differently from the other two isomers in that the i, value is small in strongly acidic media and increases as the pH is increased, whereas the reverse occurs for the other two compounds.In addition, 4-chloroaniline does not exhibit a pKa value in the pH region 8-9. Values of arta, obtained from the equation16 0.048 ana = EP - EPi2 where EP,2 is the potential corresponding to iPl2, were calculated at various pH values (Table 11), and were shown to decrease with increasing pH, indicating an increasing irreversibility of the electrode process with increasing deprotonation of the compounds in the bulk of solution. TABLE I pK VALUES FOR 2-, 3- AND 4-CHLOROANILINE Compound 2-Chloroaniline . . .. 3-Chloroaniline . . .. 4-Chloroaniline . . .. PKa pKa’ PKC3 (voltammetry) (voltammetry) (spectrophotometry) 4.8 8.0 2.65 5.4 8.6 3.46 5.5 - 4.15 Cyclic Voltammetry The cyclic voltammetric behaviour of 2- and 3-chloroaniline was found to be similar over the pH range 1-12.At pH < pKa, the cyclic voltammetric pattern was as shown in Fig. 2 (for 2- chloroaniline). From this it can be seen that the process giving rise to peak A is irreversible, but that some reduction of the product of oxidation occurs on the reverse scan. In 0.1 N sulphuric acid - 50% ethanol, two peaks appear for 2-chloroaniline on the first reverse scan at +0.47 V and +0.33 V, the more negative of which appears to be quasi-reversible. On increas- ing the pH these two cathodic peaks decrease in size until at pH > 8.0 they disappear, leaving only peak A. The cyclic voltammetric behaviour of 4-chloroaniline was found to be similar to that of the other two isomers except that this compound exhibited only one cathodic process on the reverse scan at +0.37 V.This peak, which was reversible, was also found to disappear in solutions of pH > 8.0. Chromatovoltammetric Behaviour The hydrodynamic chromatovoltammograms obtained for 2-, 3- and 4-chloroaniline are shown in Fig. 3. From this it can be seen that 4-chloroaniline (El,2 = +0.91 V) is more easily oxidised than either 3-chloroaniline = + 1.00 V) or 2-chloroaniline = +0.96 V, +1.13 V) in water - methanol - acetonitrile (75 + 20 + 5) containing phosphate ions as supporting electrolyte (pH It is interesting that for 2-chloroaniline (and to a 3). TABLE I1 ana VALUES FOR 2-, 3- AND 4-CHLOROANILINE PH 2-Chloroaniline 3-Chloroaniline 4-Chloroaniline 1.95 0.9 0.6 0.9 4.15 0.7 0.5 0.6 6.10 0.7 0.6 0.7 8.10 0.6 0.5 0.7 10.40 0.4 0.5 0.6 11.80 0.4 0.4 0.5Fcbruary, 1951 3- AND 4-CHLOROANTLINE IN MIXTURES 149 lesser extent 3-chloroaniline) there are two processes involved in the over-all oxidation reaction at the glassy carbon electrode in the hydrodynamic mode, a feature that was not observed in either the linear-sweep voltammetric or the cyclic voltammetric experiments in quiescent solution.Mechanism of Oxidation Based on these experimental findings, and bearing in mind the work carried out by previous workers,l7 the following mechanisms can be postulated to explain the oxidation of 2-, 3- and 4- chloroaniline with pH < pKa : 4-C h I oroani I i ne Head-to-tail - 2.5-1 coupling loss of CI- and 1 No further oxidation 2- and 3-chloroaniline y z Head-to-tail coupling + Tail-to-tail coupling of + - 2e- + NHZ NH2 hHzw - Similar mechanisms are likely to be in operation at pKa < pH < pKrt', but at pH > pKa' there is likely to be a change in the nature of the coupling process with more head-to-head coupling.This has been shown to be the case for aniline in the pH range 7-14, where increas- ing amounts of azobenzene were formed following oxidation a t a variety of solid indicator electrodes.lB Analytical Applications Optimum pH values for the determination of these compounds by linear-sweep voltammetry and differential-pulse voltammetry were found to be 1.95 for 2- and 3-chloroaniline and 8.1 for 4-chloroaniline. Calibration graphs of i, (nA) versus concentration (ng ml-1) were constructed and the response factors calculated. These data, together with the concentration range stud- ied and coefficients of variation obtained using both voltammetric techniques, are given in150 HART et al.: VOLTAMMETRIC DETERMINATION OF 2-, Analyst, VoZ. 106 1.2 1 .o 0.8 0.6 0.4 0.2 0 E N Fig. 2. Cyclic voltammogram of 2-chloroaniline in 0.1 N H,SO, - 50% EtOH (f = forward; r = reverse). Scan rate, 50 mV s-l; starting potential, 0 V. Table 111. As expected, greater sensitivity was achieved using the differential-pulse mode. The limit of detection of this technique was found to be 0.2 pg ml-l for all three compounds. Diff erential-pulse voltammetry was also investigated as a means of differentiating between 3- and 4-chloroaniline in mixtures. A supporting electrolyte of Britton - Robinson buffer (pH 3.0) - 50% ethanol gave rise to the best resolution (Fig.a), but it is obvious that this selectivity is not sufficient for trace analytical studies. It was decided, therefore, to employ the combination of HPLC with voltammetric detection for this purpose. Using the conditions 50 40 30 P .> 20 10 0.8 0.9 1 .o 1.1 1.2 EIVVS. Ag - AgCl Fig. 3. Hydrodynamic chromatovoltammograms of 2-, 3- and 4-chloroaniline. Concentration, 100 ng per 20 pl; flow-rate, 1.2 ml min-1.February, 2981 3- AND 4-CHLOROANILINE I N MIXTURES 0.6 0.7 0.8 0.9 1.0 EN Fig. 4. Differential pulse voltammogram of a mixture of 3-chloroaniline (1.17 x M) and 4-chloroaniline (1.04 X M) in Britton - Robinson buffer, pH 3 - 50% EtOH. Scan rate, 5 mV s-l; pulse height, 100 mV. T 0 2 4 6 8 1 0 Timelmin Fig. 5. SeDaration of a 10-ni mixture 'bf 2-, 3- and 4- chloroaniline by HPLC with voltammetric detection.Flow-rate, 2 ml min-'; ap- plied potential, + 1.2 V. 151 described previously under Experimental, a good separation of the three isomers was effected (Fig. 5 ) and the method could determine concentrations of 2- and 4-chloroaniline down to 2 ng and 3-chloroaniline down to 1 ng for a 20-4 injection on to the column. This method should prove to be useful for the trace analysis of these compounds in environmental samples, whereas the diff erential-pulse voltammetric method might be of application in formulation studies. TABLE I11 LINEAR-SWEEP (LSV) AND DIFFERENTIAL-PULSE VOLTAMMETRIC (DPV) DATA FOR 2-, 3- AND 4-CHLOROANILINE Polarographic mode Compound DPV . . . . 2-Chloroaniline DPV .. .. 3-Chloroaniline DPV . . . . 4-Chloroaniline LSV . . . . 2-Chloroaniline LSV . . . . 3-Chloroaniline LSV . . . . 4-Chloroaniline Coefficient of variation, Response factor/ (-A-, Supporting electrolyte Ep/V nA ng-l ml (a)* (b)? 0.1 N H,SO, - 50% EtOH 0.940 *0.010 0.93 6.8 2.2 0.1 N H,SO, - 50% EtOH 0.990 iO.010 0.88 7.1 2.4 Britton - Robinson buffer (pH 7) 0.665 *0.005 0.82 6.2 2.0 - 5OoL EtOH 0.1 N HiSO, - 50% EtOH 1.015 &0.010 0.63 8.2 3.3 0 1 N H SO, - 50% EtOH 1.060 *0.010 0.60 8.4 3.4 Britton8- Robinson buffer (pH 7) 0.730 *0.005 0.55 8.0 2.9 - 50% EtOH Concentration range studiedlpg ml-1 0.64-12.70 0.64-12.70 0.64-12.70 0.64-12.70 0.64-12.70 0.64-12.70 * (a) Based on 10 determinations at 0.64 pg d-l. j. (b) Based on 10 determinations at 12.70 pg ml-l.The authors thank Dr. B. J. Birch and Prof. Dr. H. W. Nurnberg for their interest in this work and for valuable discussions. 1. 2. 3. 4. 5. 6. 7. 8. References Barabanova, G. S., and Motenkov, Yu. M., Nauch. Osn. Okhr. Prir., 1973, 2, 202. Burge, W. D., and Gross, L. E., Soil Sci., 1972, 114, 440. Kaufman, D. D., and Kearney, P. C., Appl. Microbiol., 1965, 13, 443. Kaufman, D. D., J . Agric. Food Chem., 1967, 15, 582. Kearney, P. C., J . Agric. Food Chem., 1965, 13, 561. Kearney, P. C., and Helling, C . S., Residue Rev., 1969, 25, 25. Kreiger, R. A., J . Econ. Entomol., 1976, 69, 1. Bombaugh, K. J., Anal. Chem., 1965, 37, 72.152 9. 10. 11. 12. 13. 14. 15. 16. 17. 18. HART, SMYTH AND SMYTH Harris, R. J.. and Whiteoak, R. J., Analyst, 1972, 97, 294. Kaufman, VCT. M., Bills, D. D., and Hannan, E. J., J . Agric. Food Chem.. 1972, 20, 628. Frei, K. W., Lawrence, J. F.. and LeGay, D. S., Analyst, 1973, 98, 9. Suatoni, J . C., Snyder, R. E., and Clark, R. O., Anal. Chem., 1961, 33, 1894. Kissinger, P. T., Anal. Chem., 1977, 49, 447A. Smyth, M. R., Frischkorn, C. G. B., and Nurnberg, H. W., Anal. Proc., in the press. Purnell, C. J., and Warwick, C . J., Analyst, 1980, 105, 861. Adams, R. N., “Electrochemistry a t Solid Electrodes,” Marcel Dekker, New York, 1969, p. 126. Adams, R. N., “Electrochemistry a t Solid Electrodes,” Marcel Dekker, New York, 1969, pp. 327- Desideri, P. G., Lepri, L., and Heimler, D., J . Electroanal. Chem., 1971, 32, 225. 336. Received July 16th, 1980 Accepted September llth, 1980

ISSN:0003-2654    

DOI:10.1039/AN9810600146

出版商:RSC    

年代:1981

数据来源: RSC

摘要:

Analyst, February, 1981, Vol. 106, p p . 153-159 Separation of Protein-bound Copper and Zinc in Human Plasma by Means of Gel Filtration - Ion-exchange Chromatography 153 J. B. Dawson, M. H. Bahreyni-Toosi, D. J. Ellis and A. Hodgkinson Department of Medical Physics, University of Leeds, and Medical Research Cozcncil Mineral Metabolism Unit, The General Infirmary, Leeds, LS1 3EX A procedure for the separation of protein-bound copper and zinc fractions and of potassium and magnesium in plasma, involving chromatography on columns of DEAE-Sepharose CL-GB, is described. Copper, magnesium and zinc were measured by atomic-absorption spectroscopy and potassium by flame photometry. Most plasma samples yielded one copper and two zinc fractions and these were identified by the use of protein “markers.” Most of the copper appeared to be bound to caeruloplasmin and a lesser amount to albumin, whereas zinc appeared to be bound mainly to globulins with a smaller, variable, amount bound to albumin.Magnesium and potassiuni appeared as single peaks eluting in approximately the same fractions, ahead of the albumin peak. Keywords : Copper ; zinc ; plasma proteins ; chromatograplay ; atornic-absorp- tion spectroscopy Over 90% of the copper in human plasma is firmly bound to an a,-globulin, caeruloplasmin. The remainder, constituting the labile pool, is less firmly bound, in large part to albumin and in smaller part to amino acids, especially to histidine, threonine and glutamine.lY2 Zinc in plasma is likewise mainly bound to albumin and to an a,-globulin, although there are conflict- ing reports as to the relative distribution between these two protein^.^ Some zinc may also be bound to amino acids4 Various techniques have been used to study the binding of copper and zinc to specific proteins in plasma, including salt fra~tionation,~ electroph~resis,~-* gel filtration9 and sucrose density gradient ~entrifugation.~ Some of these techniques are time consuming, subject to contamination or require specialised apparatus that is not always available in a clinical chem- istry laboratory.A suitable method for clinical investigations should provide a rapid and reproducible separation of large and small molecules, with minimum dilution, contamination or loss of trace metals, and, moreover, should be suitable for automated analysis. In this study we examined several types of gel filtration and ion-exchange chromatography, in combination with atomic-absorption and -emission spectroscopy ; a modified ion-exchange procedure is described.Experimental Apparatus Acrylic plastic columns with an internal diameter of 0.9 cm and 30 or 60 cm long were used (Pharmacia Fine Chemicals, Uppsala, Sweden, Type K9/30 or K9/60). Solutions were fed to the tops of the columns by a variable-speed peristaltic pump (Stalprodukter, Uppsala, Sweden) and a constant-pressure head was maintained by means of an over-flow pipe. Ionic strength gradient elution was used in some experiments, the gradients being produced by using a second channel of the peristaltic pump to inject, with stirring, a concentrated solution of the eluent into a large reservoir of starting buffer.A schematic diagram of the apparatus is shown in Fig. 1. This system produced a nearly linear gradient according to the equation154 DAWSON et al. : SEPARATION OF PROTEIN-BOUND COPPER AND Analyst, VuZ. 106 Fig. 1. Schematic diagram of apparatus used for separation of plasma proteins and the determination of Cu, Zn, Mg, Na and K. Theabsorbance of the effluent at 280 nm was monitored by a Uvicord I1 with chart recorder (LKB Produkter, Bromma, Sweden). A laboratory-made electronic unit, based on a logarith- mic circuit (Module 757P, Analog Devices Ltd.), was inserted between the detector and chart recorder to produce a signal that was proportional to absorbance instead of absorption, as was the case with the original system. Fractions of effluent were collected in a Radi Rac collector fitted with a drop counter (LKB Produkter).The concentrations of copper and zinc in the fractions were de :rmined by flame atomic- absorption spectroscopy (Pye Unicam, Model SP2900). A progr mmable laboratory-made turntable with continuously variable sampling and wash times and Nith facilities for replicate determinations was used to supply samples to the instrument. Tk sampling and wash times in the present application were 3 and 7 s, respectively, and this cy e was performed twice on each fraction, the total consumption of sample being less than C 2 ml. The output of the instrument was presented on a chart recorder (Philips), an extra ci )acitor being incorporated to give a response time of about 1 s. Sodium, potassium and i,,agnesium were measured simultaneously, using a modified atomic-absorption spectrophotometer (Pye Unicam, Model SP90).Magnesium was determined by atomic-absorption spectroscopy and sodium and potassium by flame-emission spectroscopy, the light emitted from the flame being collected and transmitted to external filter photometers by means of a fibre-optic system. Reagents Reagents of the highest available purity, usually AnalaR or Aristar grade, were used in order to minimise the trace-element background signal. Chelex 100 (Bio-Rad Laboratories, Watford) was used in some experiments to reduce contamination still further. All gels were manufactured by Pharmacia Fine Chemicals. Solutions were prepared with singly glass- distilled water.Eluting solutions contained tris(hydroxymethy1)methylamine (Tris), hydro- chloric acid and sodium chloride. Commercial standard solutions for atomic-absorption spectroscopy (BDH Chemicals) were diluted with eluting solutions to minimise errors arising from contamination or interference. Albumin, caeruloplasmin and oc-, /3- and y-globulins were purchased from Sigma London Chemical Co. Procedure 0.05 mollA1. the eluting fluid (initial eluting fluid in the case of gradient elution). Powdered gels were prepared in Tris - hydrochloric acid buffer, pH 7.4 or 8.6, ionic strength After loading on to the column, all gels were washed with three bed volumes of Heparinised plasma fromFebruary, 1981 ZINC IN HUMAN PLASMA BY ION-EXCHANGE CHROMATOGRAPHY 155 normal adults (0.5 ml) was added to the column and the flow-rate of the eluting fluid was adjusted to about 15.0 ml h-l.Between 30 and 40 fractions (1.5 ml) were collected in plastic sample cups, which were then transferred manually to the turntables of the SP2900 and SP90 instruments and analysed without further treatment, using the instrument settings recom- mended in the manufacturer's handbooks. Total concentrations of copper and zinc in plasma were determined after 10-fold dilution with 0.1 mol 1-1 hydrochloric acid.l0,l1 The wave- lengths used were sodium 589.0, potassium 766.5, magnesium 285.2, copper 324.8 and zinc 213.9 nm. Results Gel Filtration Chromatography on Sephadex G-150 gave only partial separation of protein-bound copper or zinc under the experimental conditions used [Fig.2(a)]. The use of Sepharose CL-6B and a longer column gave improved separation of both proteins and metals [Fig. 2 ( b ) ] . There appeared to be at least two zinc fractions but only one copper fraction. Some improvement in the fractionation of both copper and zinc was achieved with Sephacryl S-200 [Fig. 3 ( a ) ] , with both copper and zinc exhibiting at least two fractions. Similar results were obtained with Sephacryl S-300, for both copper and zinc [Fig. 3 ( b ) ] , but the separation of plasma proteins was still poor. Ion-exchange Chromatography proteins than could be achieved with gel filtration. peaks [Fig. 4 ( a ) ] . The use of DEAE-Sephadex A-50 with gradient elution gave a better separation of plasma There were also several copper and zinc However, the reproducibility of results on repeated runs with the same Eluent volumeh I Fig.2. Fractionation of (A) plasma proteins, (B) copper and (C) zinc. (a) Sephadex G-150; bed height, 30 em; flow-rate, 12 ml h-l; eluent, 0.1 mol 1-l Tris - HCl buffer, pH 8.0, and 1.0 moll-' NaC1. (b) Sepharose CL-6B; bed height, 57 cm; flow-rate, 14 ml h-l; eluent, 0.1 mol 1-1 Tris - HC1 buffer, pH 7.4, and 0.2 mol 1-1 NaCl.156 DAWSON et al. : SEPARATION OF PROTEIN-BOUND COPPER AND Analyst, VoZ. 106 0 10 20 30 40 0 10 20 30 40 Eluent volumelml Fig. 3. Separation of (A) plasma proteins, (B) copper and (C) zinc. (a) Sephacryl S-200; bed height, 57 cm; flow-rate, 7 ml h-l; eluent, 0.1 mol 1-1 Tris - HCl buffer, pH 7.4, and 0.2 mol 1-1 NaCl.(b) Sephacryl S-300; bed height, 57 cm; flow-rate, 14 ml h-l; eluent, 0.1 mol 1-1 Tris - HC1 buffer, pH 8.0, and 0.5 mol 1-1 NaC1. sample was poor, owing to changes in bed volume and flow-rates as the gel settled. A com- parable degree of separation of proteins and associated metals was achieved by using DEAE-Sepharose CL-6B in place of DEAE-Sephadex [Fig. 4(b)]. The use of the more rigid agarose gel gave better reproducibility than was the case with DEAE-Sephadex but both ion- exchange procedures suffered from the disadvantage that a heterogeneous mixture of proteins and metals was eluted in the first fractions owing to the salts already present in the plasma samples. Preliminary de-salting on a small column of Sephadex G-25 was unsatisfactory because it resulted in unwanted dilution of the sample and loss of copper and zinc.The possibility of using an eluting buffer with a fixed ionic strength, comparable to that of plasma, was therefore investigated. Chromatography on DEAE-Sepharose CL-6B with an Eluent of Constant Ionic Strength The protein and metal elution patterns for a system using DEAE-Sepharose CL-6B and an eluent of constant ionic strength are shown in Fig. 5. With this procedure the early “self- elution” of protein, copper and zinc was eliminated and an ultraviolet absorbance pattern with four peaks was obtained. Copper and zinc were each partially separated into two fractions. In Figs. 2 4 , the curves for copper and zinc have an elevated base line attributable to con- tamination of the eluting solution by about 0.01 p.p.m.of those elements. The contamination arose from the presence of copper and zinc in the Tris and sodium chloride used to prepare the eluting solution. In the final procedure this contamination was reduced by passing the eluting fluid through a column of Chelex 100 (30 x 0.9 cm). This treatment removed at least 98% of the copper and SOYo of the zinc, as shown by the near zero base lines in Fig. 5. Identification of Ultraviolet Absorption Peaks An attempt was made to identify the ultraviolet absorption peaks in Fig. 5 by eluting u-, 18- and y-globulins, albumin, caeruloplasmin and lithium urate under the same experiment a1 conditions and using copper as a marker for caeruloplasmin. This procedure was followedF~bYZcUy, 1981 ZINC IN HUMAN PLASMA BY ION-EXCHANGE CHROMATOGRAPHY 157 0 10 20 30 40 50 60 Eluent volume/ml Fig.4. Separation of (A) plasma proteins, (B) copper and (C) zinc. (a) DEAE-Sephadex A-50; Tris - HC1 buffer, pH 8.3, with a linear salt gradient of 0.0-0.4 mol 1-1 NaCl; bed height, 57 cm; flow-rate, 18 ml h-l. (b) DEAE- Sepharose CL-GB; Tris - HC1 buffer, pH 8.6, with a linear salt gradient of 0.0- 0.25 moll-' NaCl; bed height, 57 cm; flow-rate 11 ml h-l. because the ultraviolet absorbance peak for caeruloplasmin would be masked by the higher concentrations of other proteins present in the sample. Fig. 6(a) shows the pattern obtained with a mixture of albumin, caeruloplasrnin and lithium urate. Albumin and caeruloplasmin both gave peaks at 24 ml whereas lithium urate emerged much later at 56 ml.A mixture of a- and /3-globulins, albumin, caeruloplasmin and lithium urate, on the other hand, gave peaks at 10, 19, 27 and 31 ml and a late peak at 50 ml [Fig. 6 ( b ) ] , while a mixture of a-, /3- and y- c (0 Caeruloplasmin Urate 8 (a) -P a 0 a I C 0 .- c C 0 0 0 10 20 30 40 50 60 E luen t vo I u meim I Fig. 5. Separation patterns of (a) plasma proteins, (b) copper and zinc and (c) magnesium and potassium on DEAE-Sepharose CL-GB, using an eluent of constant ionic strength (Tris - HC1 buffer, pH 8.6, containing NaCl, 0.2 mol 1-l; total ionic strength = 0.25 mol 1-l). Bed height, 54 cm; flow-rate, 14 ml h-l.158 DAWSON et al. SEPARATION OF PROTEIN-BOUND COPPER AND Analyst, VOZ. 106 globulins, caeruloplasmin and lithium urate gave major peaks at 11 and 20 ml, small peaks a t 30 and 40 ml and a late peak at 57 ml [Fig.6(c)]. Finally, a mixture of a-, /3- and y-globulin with additional albumin, caeruloplasmin and lithium urate gave four peaks at 10, 18, 25 and 52 ml [Fig. 6 ( d ) ] . Lithium urate eluted at a relatively constant volume, whatever the composition of the mixture, but the proteins showed some variability, particularly caeruloplasmin, which tended to elute with albumin at about 25 ml if albumin was present and at a higher value if albumin was absent. The identity of the absorbance peaks in Fig. 5 therefore cannot be deduced with absolute certainty from Fig. 6(a)-(4. It may be noted that the albumin peak in Fig. 5 is abnormally low; this is because the sample was obtained from a patient suffering from mal- absorption and mild renal failure.I (a' F I I 0 10 20 30 40 50 60 Q) I C a. m 10 20 30 40 50 0 10 20 30 40 50 60 0 10 20 30 40 50 Eluent volume/ml Fig. 6. Separation of mixtures of proteins (-), copper (o--o) and lithium urate on DEAE-Sepharose CL-6B. The sample volume was 0.5 ml in all instances ; other conditions were as in Fig. 5 . The test solutions had the following composition: (a) albumin 20 g l-l, caeruloplasmin 20 g 1-l, lithium urate 0.3 g 1-l; (b) a-globulins 27.2 g 1-l, /3-globulins 9.2 g l-l, albumin 5.6 g l-l, caeruloplasmin 0.8 g l-l, lithium urate 0.05 g 1-I; (c) a-globulins 27.2 g l-l, /3-globulins 9.2 g 1-l, y- globulins 20.0 g l-l, albumin 5.6 g 1-l, caeruloplasmin 6.0 g l-l, lithium urate 0.05 g 1-l; (d) a-globulins 27.2 g 1-l, /3-globulins 9.2 g l-l, albumin 25.6 g 1-I, caeruloplasmin 0.8 g l-l, lithium urate 0.05 g 1-l.Peaks: A, a-globulins; B, 6- globulins ; C , caeruloplasmin; D, albumin; E, y-globulins; and F, lithium urate. Discussion and Conclusions Recent reports indicate that both copper and zinc concentrations are higher in serum than in plasma, possibly owing to liberation of these metals from platelets or other cells during the process of c10tting.l~~~~ Plasma is therefore to be preferred to serum and was used throughout this study. Moreover, the protein elution pattern of plasma showed fewer signs of change with ageing than was the case with serum. The ion-exchange procedure described above departs from usual practice in that an eluting buffer of fixed ionic strength is used instead of one with a steadily increasing concentration.The use of a single-strength buffer is less effective in separating fractions but it avoids the needFebruary, 1981 ZINC IN HUMAN PLASMA BY ION-EXCHANGE CHROMATOGRAPHY 150 to de-salt the specimen, with the attendant risk of loss of metals. The procedure finally adopted gave good reproducibility of both protein and metal patterns on repeated runs on the same sample, a feature of some importance in comparing one patient with another and in deriving quantitative data. The identity of the protein peaks has been partly established but more detailed examination of the degree of homogeneity of each peak is needed, using more refined techniques such as electrophoresis or ultracentrifugation. The distribution of copper and zinc between albumin and the various globulins is broadly in agreement with published data,1,3.* most of the copper being bound to caeruloplasmin and to a lesser extent to albumin, whereas zinc appeared to be bound mainly to globulins, although the relative distribution of zinc between albumin and globulins appeared to differ from one individual to another; this aspect is being examined in greater detail.So far, it has not been possible to detect copper or zinc bound to small molecules such as amino acids because of the low signal to noise ratio in this area of the chromatogram, but this limitation can probably be overcome by the use of atomic-absorprion spectroscopy with electrothermal atomisation and work on this problem is in progress. 1. 2. 3. 4. 5. 6. 7. 8. 9. 10. 11. 12. 13. References Mason, K. E., J. Nutr., 1979, 109, 1979. Harrison, P. M., and Treffry, A., in “Inorganic Biochemistry,” Volume 1, Specialist Periodical Song, Air. K., and Adham. N. F., Clin. Chim. Acta, 1979, 99, 13. Prasad, A. S., “Trace Elements in Human Health and Disease,” Volume 1, Academic Press, New Dennes, E., Tupper, R., and Wormall, A., Biochem. J . , 1962, 82, 466. Bearn, A. G., and Kunkel, H. G.. Proc. SOC. Exp. Biol. Med., 1954, 85, 44. Prasad, A. S., and Oberleas, D., J . Lab. Clin. Med., 1970, 76, 416. Teape, J., Kame], H., Brown, D. H., Ottaway, J. M., and Smith, W. E., Clin. Chim. Acta, 1979, Falchuk, K. H., New Engl. J . Med., 1977, 296, 1129. Dawson, J . B., Ellis, D. J., and Newton-John, H., Clin. Chim. Acta, 1968, 21, 33. Dawson, J . R., and Walker, €3. E., Clin. Chim. Acta, 1969, 26, 465. Foley, B., Johnson, S. A., Hackley, B., Smith, J . C., and Halstead, J. A., Proc. SOC. Exp. Bid. Med., Rosenthal, R. W., and Blackburn, A., Clin. Chem., 1974, 20, 1233. Report, Chemical Society, London, 1979, p. 151. York, 1976, p. 13. 94, 1. 1968, 128, 265. Received August 12th, 1980 Accepted September 15th, 1980

ISSN:0003-2654    

DOI:10.1039/AN9810600153

出版商:RSC    

年代:1981

数据来源: RSC

摘要:

160 Analyst, February, 1981, Vol. 106, p p . 160-165 Gas - Liquid Chromatographic Determination of Major Constituents of Piper methysticum R. N. Duve Research Division, Ministry of Agriculture and Fisheries, Koronivia Research Station, P.O. Box 77, Nausori, Fiji A procedure is described for the quantitative determination of seven known major constituents in sun-dried roots, rhizomes and commercially powdered samples of Piper methysticum. A 3.0-8.0-g amount of powdered sample is extracted with chloroform in a Soxhlet apparatus for 6 h. After evaporation of solvent the extract (about 0.35 g) is dried a t 100 “C for 2 h and then dissolved in chloroform to about 0.7% m/V concentration. The resulting solution is analysed by gas - liquid chromatography using dual 1.5 m x 4 mm i.d.glass columns, containing 3% m/m of OV-1 on Chromosorb W HP, and dual differential flame-ionisation detectors with nitrogen as carrier gas, the column temperature being 210 “C. There is no interference from the eight other trace constituents, non-polar low-boiling compounds or polar “tarry” material. Keywords : Piper methysticum analysis ; gas - liquid chromatography Chemical and pharmacological investigations of constituents of the sun-dried roots and rhizomes (known locally as “waka” and “lawena,” respectively) of Piper methysticum, which began in 1860,1-3 led to the publication of numerous papers and reviews.*-6 To date at least seven major and eight trace constituents have been isolated from the roots of the shrub4-’ (see Table I). A number of these constituents have been shown to possess interesting physiological properties such as sleep-producing,*s9 antimycotic,lO local anaesthetic,ll anti- convulsivel2-14 and smooth muscle contractionl5 effects.The beverage prepared by straining the powdered roots or rhizomes of Piper methysticum with water is a national drink, and is held in high esteem in all traditional Fijian ceremonies. I t is also a common traditional drink in other Pacific islands such as Samoa, Tonga and Hawaii. During an investigation by the author of the quality of commercially powdered samples of Piper methysticum (known locally as “yaqona” or “kava”), it became necessary to determine the amounts of major constituents. Numerous reports5-8J6 exist on the column and thin-layer chromatographic separation of the active constituents, but these methods are unsatisfactory and there does not appear to be a simple and rapid method for the quantita- tion of these constituents. This paper, which deals with the gas - liquid chromatographic analysis of the seven major constituents of Piper methysticum, aims to bridge this gap. Experimental Structures of the compounds concerned are shown in Fig.1. An electrothermal melting-point apparatus, a Hewlett Packard, Model 5730A, gas - liquid All glassware was cleaned with chromic acid, washed with distilled water and oven dried Apparatus chromatograph, a Model 3380A integrator and a Model 7123A recorder were used. at 100 “C. Reagents solvents were chloroform, methanol and benzene. I1 (Merck) was used. Standard Solutions Standard solution A of m j o ~ constituents Prepare the following standard solution in chloroform (m/‘V> : 0.12% 7,8-dihydrokawain (l), 0.12% kawain (2), 0.03% 5,6-dehydrokawain (6), 0.02y0 5,6,7,8-tetrahydroyangonin (7), 0.075% 7,8-dihydromethysticin (5), 0.075% yangonin (4) and 0.12% methysticin (3).All reagents and solvents were of analytical-reagent grade unless otherwise stated. The Aluminium oxide of Brockmann activityDUVE 161 13 Numbering system 2. Kawain 3. Methysticin 4. Yangonin Fig. 1. Structures of the three parent compounds of Piper methysticum. The structural formulae for all other kava pyrones, 1 and 5-15, can be derived by using the numbering system above and diagrams 2, 3 and 4. Standard solution B of trace constituents The trace constituents were provided by Professor R.Hansel of Frieie Universitat, Ger- many. Prepare a 0.003% m/V solution of each of cis-5-hydroxykawain (8), 7,s-dihydro- yangonin (9), 5,6-dihydroyangonin (lo), 5,6-dehydromethysticin (1 1), 11-methoxyyangonin (12), 11-hydroxyyangonin (13), 11-methoxy-12-hydroxy-5,6-dehydrokawain (14) and 10- methoxyyangonin (15). Standard solution C of all active constituents ents as in the original solutions. in the samples and sample solutions. least 6 months if stored in a refrigerator. Isolation of Major Constituents Yangonin (4), methysticin (3), 7,s-dihydromethysticin (5), kawain (2), 5,6-dehydrokawain (6) and 7,8-dihydrokawain (1) were separated by column chromatography on aluminium oxide and purified by repetitive crystallisation as described elsewhere.8 5,6,7,8-Tetra- hydroyangonin (7) was isolated by a combination of column chromatographya and the proposed gas - liquid chromatography and purified by crystallisation from methanol. 7,8-Dihydromethysticin (5), 7,s-dihydrokawain (1) and 5,6,7,8-tetrahydroyangonin (7) were obtained by catalytic hydrogenation of the parent compounds methysticin (3), kawain (2) and yangonin (4), respectively.Hydrogenation was carried out at room temperature on a 2-g sample using 0.1 g of platinum dioxide and 75 ml of ethyl acetate and was complete in 4 h. After filtration and distillation of the solvent in vacuo, the hydrogenated products were recrystallised.8 The hydrogenated products had similar melting-points and mixed melting-points to the naturally occurring compounds.Preparation of Samples ground to pass a No. 25 sieve. sieve. Prepare this solution by combining solutions A and B. It has concentrations of constitu- Standard solutions A, B and C have similar constituent ratios and concentrations to those The standard and sample solutions are stable for at The dried roots and rhizomes were obtained from local markets, cut into thin slices and Commercial powders were further ground to pass a No. 25162 DUVE: MAJOR CONSTITUENTS OF PIPER Moisture Determination Dry 5 g of prepared sample at 100 "C for 5 h. Analyst, Vol. 106 Extraction of Constituents Extract 3.0 g of prepared roots or 5.5 g of rhizomes or 8.0 g of commercial powders with chloroform in a Soxhlet apparatus for 6 h. Evaporate the solvent on a water-bath and dry the extract at 100 "C for 2 h.Dissolve the dried extract (about 0.35 g) in chloroform and make up to 50 ml with chloroform in a calibrated flask to give a concentration of about 0.7% m/V (sample solution D). Gas - Liquid Chromatography Inject 3 pl each of standard solution A and sample solution D into the gas chromatograph using the following conditions: column, dual 1.5 m x 4 mm i.d. glass containing 3% m/m of OV-1 on Chromosorb W HP; detector, dual differential flame-ionisation ; carrier gas, nitrogen at 300 kPa and 60 ml min-1; fuel gases, hydrogen at 150 kPa and 60 ml min-l and air at 180 kPa and 240 ml min-l; column temperature, 210 "C; detector temperature, 300 "C; injection port temperature, 250 "C; and integrator, sensitivity 0.3-1.0 mV min-l, attenuation 64, chart speed 5 mm min-l.Determination of Polar "Tarry" Material in Extract by Clean-up on an Aluminium Oxide Column Wash 4-5 g of aluminium oxide contained in a glass column (12 mm diameter) with 20 ml of chloroform. Apply a suitable amount (generally about 0.35 g) of extract in 5ml of chloroform to the washed column, elute with 50 ml of chloroform and collect the eluate in an already dried and weighed 100-150-ml conical flask. Evaporate the solvent on a water- bath and dry the extract at 100 "C for 2 h to give a cleaned-up extract. The difference in mass represents polar ' 'tarry ' ' material. Thin-layer Chromatography Thin-layer chromatography was carried out on 0.5-mm silica gel G plates with benzene - methanol (98.5 + 1.5) as developing solvents and the spots were revealed with iodine vapour.Results and Discussion The seven major and eight trace constituents of the roots and rhizomes of Piper methysticum are as presented in Table I and Fig. 1. The constituents are all derivatives of the three variant compounds, namely kawain (2), methysticin (3) and yangonin (4). The major constituents have traditionally been isolated by column chromatography8*l6 and the trace constituents by preparative thin-layer chr~matography.~ We have found that neither column nor thin-layer chromatography could be used for the quantitative determination of the constituents, as in the former the separation is incomplete and in the latter all of the major constituents have similar R, values (Table I). The proposed gas - liquid chromato- graphic analysis of the major constituents overcomes this problem and offers a simple and rapid technique for their determination in Piper methysticum. Extraction of Constituents Chloroform,8 diethyl etherls and ethanoP have been used for the extraction of constituents of Piper methysticum.Some of the constituents, such as methysticin (3) and yangonin (4), are insoluble in diethyl ether but all constituents are soluble in chloroform. We therefore used chloroform for the extraction work. It was found that 95% and 99% of the constitu- ents are extracted in the first 2 h and 4 h, respectively, and we therefore used 6 h as the extraction time. The difference in the amounts of roots, rhizomes and commercial powders of Piper methysticum taken for extraction is to ensure that uniform amounts of extract (ca.0.35 g) are obtained. Quantification and Linear Range times of the various constituents are reported in Table I. A typical chromatogram is shown in Fig. 2, and the tested linear range limits and retentionFebruary, 1981 METHYSTICUM BY GLC TABLE I CHROMATOGRAPHIC CHARACTERISTICS AND CONCENTRATIONS OF CONSTITUENTS OF Piper Methysticum 163 Compound No. Constituents 1 7,8-Dihydrokawain 2 Kawain 3 Methysticin 4 Yangonin 5 7,8-Dihydromethysticin 6 5 ,g-Dehydrokawain 7 5,6,7,8-Tetrahydroyangonin Total of 1-7 7a Tarry material Total of 1-7a Chloroform extract Major constituents- Other constituents- I Non-polar compounds I1 T ~ a c e constituents- 8 cis-5-Hydroxykawain 9 7,8-Dihydroyangonin 10 5,6-Dihydroyangonin 11 5,6-Dehydromethysticin 12 11-Methoxyyangonin 13 11 -H ydrox yyangonin 14 11 -Me thoxy-12-h ydrox y- 15 10-Methoxyyangonin dehydrokawain Average content of constituents on Retention time Tested Retention I L > (GLC] a t limit of time a t Commercial Melting- 210 C/ linear 240°C/ RF Roots Rhizomes powder point/”C min rangelwg min (TLC) (6 samples) (6 samples) (12 samples) dry mass basis, yo m/m 55-5717 5.88 4.8 2.4 0.41 2.37 1.20 0.70 .~~ 10717 8.01 4.8 2.8 0.43 1.90 i.ii 0.84 139-140.5l’ 27.0 - 8.0 0.39 2.12 1.00 0.69 155-15719 25.63 3.0 8.0 0.48 1.73 0.70 0.47 117-11818 18.33 3.0 6.4 0.39 1.12 0.69 0.61 138-139* 10.27 1.2 3.6 0.52 0.81 0.32 0.18 99-100’ 13.12 0.75 4.4 0.43 0.39 0.20 0.14 - - - - - 10.44 5.28 3.63 - - - - 0.00 0.71 0.55 0.41 - - - - - 11.15 5.83 4.04 - - - - - 12.21 6.61 4.69 - - - - - 1.06 0.78 0.65 <5.0 <2.0 >0.65 120-1227 9.58 - 4.0 0.10 104-106’ 11.80 - 3.2 0.48 122-1247 Decomposes - Decomposes 0.37 230-231’ 37.54 - 10.8 0.50 155-1577 50.18 - 13.6 0.40 196-2007 50.18 13.6 0.18 119-220’ 50.18 - 13.6 0.19 191-1927 56.12 - 14.8 0.42 For 7,s-dihydrokawain (l), kawain (2), 5,6-dehydrokawain (6), 5,6,7,8-tetrahydroyangonin (7) and 7,s-dihydromethysticin (5), quantification was effected by comparisons of either standard and sample peak heights or integrated peak areas.However, the peak of yangonin (4) could not be completely separated from that of methysticin (3), which appears as a shoulder on the former. However, the peak height of yangonin (4) is linearly related to amount, irrespective of the amount of methysticin (3) present, and therefore can be deter- mined.However, as the peak height of methysticin (3) is not linearly related to amount, the amount of methysticin (3) is determined by matching its peak heights in a standard and a sample. This is made simple by the fact that the ratio of yangonin (4) to methysticin (3) in roots, rhizomes and commercial powders of Piper methysticum is reasonably constant (1.7:Z.l, 1.54:Z.l and 1.5:2.1, respectively). Clean-up of Extract The total amount of the major constituents 1-7 (see Table I) was found to be lower than that in the total chloroform extract. Thin-layer chromatography of standard C and a chloroform extract of Piper methysticum root sample showed that in addition to the major and trace constituents, the root extract contained some non-polar compounds and polar “tarry” material.The polar “tarry” material had an R, value of 0. The amount of “tarry” material could be determined as described under Experimental. Between 10 and 60 ml of standard solution A, representing 0.056-0.336 g of major constituents, and between 10 and 40 ml of standard solution B, representing 0.002 1-0.0084 g of trace constituents were subjected to the clean-up procedure, and subsequent analysis showed no significant adsorption of active constituents on the column at these concentrations. However, clean-up of 0.2540 g of a chloroform extract of Piper methysticum roots on aluminium oxide showed an average adsorption of 0.7% m/m (based on the dry mass of the sample) of polar “tarry” material ; thin-layer chromatography of the cleaned-up extract showed the absence of polar “tarry” material.However, the amount of the major constituents in the cleaned-up extract was the same as that in the unpurified extract. Hence, although the clean-up procedure provides a method for the determination of polar “tarry” material, it is not necessary for the deter- mination of active constituents. Interferences Using the concentration ranges of standards and samples mentioned under Clean-up of Extract, it was established that heating for 6 h under reflux during extraction and drying for 2 h at 100 “C had no significant effects on the analysis. There was no interference by164 DUVE: MAJOR CONSTITUENTS OF PIPER Analyst, Vol. 106 T v) 0 9. 0) a 2 6 0 5 10 15 20 25 30 Timeimin Fig.2. Gas - liquid chromatogram obtained from a 3-1.11 injection of a 0.7% m/V solution of chloroform extract of Piper methysticum. Other conditions are as stated in text. Peaks : 1 = 7,s-dihydrokawain ; 2 = kawain ; 3 = methysticin; 4 = yangonin; 5 = 7,s- dihydromethysticin : 6 = 5,6-dehydrokawain ; and 7 = 5,6,7,8-tetrahydroyangonin. the trace constituents, as these have different retention times (see Table I). In addition, the comparatively low concentration of trace constituents in the samples did not give rise to any peaks under the given conditions of analysis. The analysis was complete within 0.5 h. A column temperature of 210 “C was used to distinguish the methysticin (3) peak from the yangonin (4) peak within a reasonable time. An increase in temperature resulted in the collapse of these two peaks, whereas a decrease in temperature only increased retention times without any improvement in the separation of the two peaks.I t is clear from Table I that the proposed method is not satisfactory for the determination of trace constituents , owing mainly to the similar and relatively high retention times of some of the constituents. We are currently studying methods of analysis of trace constituents. There is also no interference by polar “tarry” material or non-polar low-boiling compounds, as the latter elute before any of the active constituents. Precision Satisfactory reproducibilities were obtained by the proposed method (Table 11). TABLE I1 CONTENTS OF MAJOR CONSTITUENTS, CHLOROFORM EXTRACT AND POLAR “TARRY’, MATERIAL FROM 10 RUNS ON A RHIZOME SAMPLE OF Piper methysticunz Content, yo mlm Constituent 7,8-Dihydrokawain .. Kawain . . . . hlethysticin . . . . Yangonin . . . . 7,s-Dihydromethysticin 5,6-Dehydrokawain . . 5,6,7,8-Tetrahydroyangon Polar “tarry” material Chloroform extract . . . . . . 1.05 1.14 1.11 1.06 1.11 . . .. 1.22 1.22 1.25 1.25 1.28 . . . . 1.15 1.10 1.12 1.21 1.17 . . . . 0.82 0.82 0.87 0.84 0.87 . . . . 0.68 0.66 0.66 0.65 0.71 . . . . 0.28 0.30 0.30 0.30 0.31 .. .. 0.71 0.73 0.66 0.61 0.67 . . .. 6.56 7.02 6.78 6.92 6.56 iin . . 0.128 0.127 0.127 0.129 0.130 S tandaid Average deviation 1.05 1.08 1.08 1.08 1.09 1.08 2.78 1.31 1.21 1.27 1.29 1.28 1.26 2.38 1.17 1.13 1.17 1.19 1.17 1.16 2.59 0.79 0.81 0.83 0.84 0.86 4.84 3.57 0.62 0.66 0.68 0.68 0.69 0.67 2.99 0.29 0.29 0.31 0.30 0.30 0.30 3.33 0.121 0.121 0.128 0.125 0.126 0.127 1.57 0.64 0.72 0.65 0.71 0.63 0.67 5.97 6.54 6.79 6.66 6.96 7.00 6.78 2.80February, 1981 METHYSTICUM BY GLC Conclusion 165 A simple and rapid gas-liquid chromatographic technique has been developed for the determination of the major constituents of Piper methysticum.This initial study has also shown considerable variations in the amounts of major constituents in roots, rhizomes and commercial powders. It is expected that the proposed method of analysis will find use in the column chromatographic separation and purification of the major constituents of Piper methysticum. The procedure described here is also being used for a study of variations in the amounts of constituents in roots, rhizomes and commercial powders of local Piper methysticum as part of a general survey of the quality of the local product.I am indebted to Professor R. Hansel of Frieie Universitat, Germany, for providing a number of pure constituents of Piper methysticum, useful information and reprints of papers. I also thank Mr. A. Singh of Ministry of Agriculture and Fisheries, Fiji, for a critical review of the manuscript and the Fiji Government for permission to publish this paper. 1. 2. 3. 4. 5. 6. 7. 8. 9. 10. 11. 12. 13. 14. 15. 16. 17. 18. 19. References Cuzent, C. R. Acad. Sci., 1860, 50, 436. Gobley, J. Pharm. Chim., 1860, 37, 19. O’Rorke, C. R. Acad. Sci., 1860, 50, 598. Duve, R. N., Fiji Agric. J , . 1977, 38, 81, and reference cited therein. Hansel, R., Pac. Sci., 1968, 22, 293, and reference cited therein. Keller, F., and Klohs, M. W., Lloydia, 1963, 26 (l), 1, and references cited therein. Hansel, R., personal communication, 1978. Klohs, M. W., Keller, F., Williams, R. E., Toekes, M. I., and Cronheim, G. E., J . Med. Pharm. Meyer, H. J., Arch. Int. Pharmacodyn., 1962, 138, 505; Chem. Abstr., 1963, 58, 1827. Hansel, R., Weiss, D., and Schmidt, B., Planta Med., 1966, 1. Meyer, H. J., and May, H. U., Klin. Wochenschr., 1964, 42 407; Chem. Abstr., 1964, 61, 9932. Meyer, H. J., and Meyer-Burg, J., Arch. Int. Pharmacodyn., 1964, 148, 97; Chem. Abstr., 1964, 60, Meyer, H. J . , Arch. Int. Pharmacodyn., 1964, 150, 118; Chem. Abstr., 1964, 61, 13770. Meyer, H. J., and Kretzschmar, R., Arch. Exp. Pathol. Pharmakol., 1965, 250, 267; Chem. Abstr., Meyer, H. J., Arch. Int. Pharmacodyn., 1965, 154, 448. Hansel, R., and Beiersdorff, H. U., Arzneim.-Forsch., 1955, 9, 581; Chem. Abstr., 1960, 54, 2665. Borshe, W., and Peitzsch, W., Chem. Ber., 1930, 63, 2414. Borshe, W., and Peitzsch, W., Chem. Ber., 1929, 62, 360. Borshe, W., and Bodenstein, C. K., Chem. Bey., 1929, 62, 2515. Chem., 1959, 1 (l), 95. 15013. 1965, 65, 12750. Received July 18th, 1980 Accepted August 131h, 1980

ISSN:0003-2654    

DOI:10.1039/AN9810600160

出版商:RSC    

年代:1981

数据来源: RSC

摘要:

166 Analyst, February, 1981, VoL. 106, pp. 166-171 Gas-chromatographic Determination of Trace Amounts of Lower Fatty Acids in Ambient Air Near and in Exhaust Gases of Some Odour Sources Yasuyuki Hoshika Aichi Environmental Research Centre. 7-6, Tsuji-machi, Kita-ku, Nagoya-shi, Aichi, 462, Japan The gas-chromatographic determination of trace amounts of the lower fatty acids (C,-C,) in ambient air near and in the exhaust gases of some odour sources was investigated. The sample for the gas-chromatographic deter- mination was prepared by trapping in a pre-column packed with FFAP + orthophosphoric acid (H,PO,) on Carbopack C a t 25 and 30 "C. The lower fatty acids were identified and quantitated from the difference between the chromatograms obtained using the FFAP + H,PO, pre-column and that obtained using the FFAP + H,PO, plus an alkaline pre-column. The method has been applied to the analysis of lower fatty acids in practical specimens, namely the ambient air near accumulated poultry manure, in a pig pen and a fish meal factory, and in the exhaust gases from a corn starch manufacturing factory and from a poultry manure dryer.The sample volume is as low as about 0.4 1 and the method is sensitive (detection limit about 0.5 p.p.b.) and rapid (including the concentration and analysis of one sample, about 15 min are required). This sensitivity and precision are adequate for use in odour pollution analysis. . The coefficient of variation is less than 6%. Keywords : Lower fatty acid determination ; air analysis ; pre-column con- centration ; odour sources Organic acids and their derivatives are very important in industry; however, several of these compounds are primary irritants1 and the lower fatty acids are frequently implicated in odour pollution.In this context, their identification and quantitation are extremely important, as is their analysis in foods, in cigarettes and cigarette smoke and in drugs. Because of their low odour threshold (below 1 part per lo9 in air) , the determination of trace concentrations of these compounds in air presents a formidable task. The direct gas-chromatographic determination of these free lower fatty acids at low con- centrations has been limited by adsorption and decomposition in the analytical column, "ghosting" phenomena and tailing of eluting peaks. Kuksis2 reviewed recent developments in the methodology and application of gas - liquid chromatography to the analysis of free fatty acids in the underivatised form.Two special issues of the Journal of Chromatographic Science were devoted to a comparison of techniques for the analysis of fatty acids and fatty acid esters.3 DiCorcia and co-~orkers~-~ reported the complete separation of C,-C, lower fatty acids at the nanogram level using gas - liquid - solid chromatography (GLSC) without adsorp- tion and tailing. In this study, a simple and rapid procedure for the accurate determination of trace amounts of lower fatty acids (C2-C5) in ambient air near and in the exhaust gases of some odour sources is described. The acids were collected in a pre-column using FFAP (0.1%) + ortho- phosphoric acid (H3PO4) (0.1%) on Carbopack C (80-100 mesh) a t 25 and 30 "C.For desorp- tion the pre-column was heated from room temperature to 200 "C in 30 s, then held at that temperature for 30 s. The sample was subsequently introduced into the GLSC system with FFAP (0.3%) + H,PO, (0.3%) on Carbopack B (60-80 mesh) as the packing in the analytical column. The packing of the main analytical column and the trapping pre-column produced peaks free from tailing. The identification of the peaks of the acids was carried out by the disappearance method using an alkaline pre-column. The peaks obtained were generally sharp. Experimental Reagents and Materials The lower fatty acids (C2X5) and other reagents used were obtained from PolyScience (Niles, Ill., USA), Wako Pure Chemical Industries (Osaka, Japan), Katayama ChemicalHOSHIKA 167 All reagents were of A standard solution of lower fatty acids was prepared by dissolving the acids in distilled Calibration graphs were Industries (Osaka, Japan) and Tokyo Kasei Kogyo (Tokyo, Japan).guaranteed or analytical-reagent grade. water to give a concentration of each acid of 0.1 mmol per 10 ml. obtained using the standard solution diluted 1 + 10 or 1 + 100. The column packings were purchased from Wako. Apparatus The gas chromatograph used was a Shimadzu, Model GC5AP5F, equipped with on-column injection, a flame-ionisation detector (FID) and a digital integrator (Shimadzu, Model ITG- 2A), for the determination of retention times and quantitative analysis. The detector signal was recorded at 10 mV full scale simultaneously on a Shimadzu, Model R-201, double-pen recorder.GLSC Conditions The main analytical column was glass (1.5 m x 3 mm i.d.), packed with 0.3% FFAP + 0.3% €€,PO, on Carbopack B (60-80 mesh). This column was pre-conditioned at 250 "C for 10 h with a constant flow of nitrogen (60 ml min-l) before being connected to the FID. The chromatographic conditions for the FID were as follows : main analytical column temperature, 200 or 220 "C; injection port and detector temperature, 250 "C; carrier gas (nitrogen) flow-rate, 50 ml min-l; hydrogen and air flow-rates, 50 ml min-I and 1.0 1 min-l, respectively. Pre-column Collection and Injection Methods for Sample Gases The pre-column consisted of a 16 cm x 4 mm i d . glass tube packed with 0.1% FFAP + 0.1% H3PO4 on Carbopack C (80-100 mesh).The pre-column was pre-conditioned at 250 "C for 2 h with a constant flow of nitrogen (60 ml min-l) before being used for the collection of the sample gases. The sample gases were collected directly in the pre-column at 25 and 30 "C using a vacuum pump (Mini-Vac, Model PS-05, Yamato, Japan, maximum rate 5 1 min-I) and a gas meter (T-3 Dry Test gas meter, Chubushinagawa Seisakusho, Japan, 1 1 rev-I). Pre- concentration of the lower fatty acids (C2-C,) in sample gases was carried out by using the FFAP + H3P04 adsorption pre-column and the FFAP + H3P04 pre-column plus an alkaline pre-column. The FFAP + H3P0, pre-column and the alkaline pre-column were connected with a PTFE tube (2 cm x 1 mm i.d.) and the sampling rate was about 1 1 per 4 min.Then the FFAP + H3P0, pre-column was connected to the carrier gas line of the gas chromato- graph. After about 5 min, a constant flow-rate of the carrier gas was maintained, and the FFAP + H,PO, pre-column was heated from room temperature (21 "C) to 200 "C in 24 s, this tempera- ture was maintained for 30 s, then the column was cooled to room temperature. For the preparation of the calibration graphs, the volume of the standard solution and the diluted solution of the lower fatty acids (C2-C5) injected into the FFAP + H3PO4 pre-column was usually 1-5 p1. Alkaline Pre-column The alkaline pre-column consisted of 2% sodium hydroxide on glass beads (30-60 mesh) (Wako), packed into a glass tube (2 cm x 5 mm i.d.). This column was pre-conditioned a t 150 "C for 2 h with a constant flow of nitrogen (60 ml min-l) before being used for the collection of the sample gases.This column was placed in the front port of the FFAP + &PO, pre- column, and used for the identification of the lower fatty acid peaks by means of the dis- appearance method. Results and Discussion Separability of the Lower Fatty Acids and Other Compounds on the Main Analytical Column (FFAP + HsP04) by the Direct Injection Method The retention times of about 130 compounds, including eight lower fatty acids, on the main analytical column (FFAP + H,PO,) were determined by the direct injectionmethod at a column temperature of 200 "C.8 According to previous work: the following peaks overlapped : acetic168 HOSHIKA: TRACE AMOUNTS OF LOWER FATTY ACIDS IN Analyst, I/'d.106 acid with cis-but-2-ene, tralzs-but-2-ene, buta-l,3-diene and isopropyl chloride ; propionic acid with 2,2-dimethylbutane, cyclohexane, sec-butyl chloride and isobutyl chloride ; isobutyric acid with isobutyl formate; butyric acid with ethyl acrylate; isovaleric acid with hept-l-ene; and valeric acid with amyl formate; however, cis-but-2-ene, trans-but-2-ene, buta-l,3-diene and 2,2-dimethylbutane passed through the FFAP + H3PO4 pre-column during the pre- concentration of the sample gases at 25 or 30 "C, because these compounds have low boiling- points, and the other compounds also passed through the alkaline pre-column. Calibration Graphs The FID response produced a straight line in the approximate range 10-1 000 ng of the six lower fatty acids (Fig.l ) , and the detection limit at twice the noise level was about 0.5 ng. Therefore, when the pre-concentration volume is 1 1, the minimum detectable concentration is about 0.5 parts per lo9 (p.p.b.). These sensitivities are adequate for use in odour pollution analysis. Repeatability of the FFAP + H3P04 Pre-column Injection Method The repeatability and uniformity of the retention times and peak areas (as counts on the digital integrator) of the six lower fatty acids were evaluated by the FFAP + H3P04 pre- column injection method. As can be seen from Table I, they showed good uniformity and repeatability. The amounts of the lower fatty acids tested were 60-100 ng. TABLE I REPEATABILITY OF RETENTION TIMES AND PEAK AREAS (AS COUNTS ON DIGITAL INTEGRATOR) OF SIX LOWER FATTY ACIDS USING THE FFAP + H,PO, PRE-COLUMN INJECTION METHOD (fi = 6) Retention time Value f standard Coefficient of Acid deviationlmin variation, yo I A \ Acetic acid .. . . . 1.24 f 0.03 2.4 Propionic acid . . . . 1.71 f 0.05 2.9 Butyric acid . . . . 2.88 f 0.09 3.1 Isobutyric acid. . . . 2.47 f 0.08 3.2 Valeric acid . . . . 6.75 f 0.09 1.6 Isovaleric acid . . . . 4.92 f 0.10 2.0 Peak area - Value f standard deviation, counts 6034 f 337 QQV f 413 13602 f 277 10899 f 540 14897 f 548 14190 f 455 1 Coefficient of variation, yo 5.6 4.6 2.0 5.0 3.7 3.2 Relationship Between Recovery of the Lower Fatty Acids on the FFAP + H3P04 Pre-column and the Nitrogen Carrier Gas Volume Passed Through the Pre-column The amounts of each lower fatty acid used were the same as in Table I.Quantitative recovery was obtained up to a sampling gas volume of about 5 1, except that the recovery of acetic and propionic acids decreased slightly (Table 11). This test was carried out at 40 "C and a constant flow-rate of nitrogen of 0.25 1 min-1. Typical Gas Chromatograms of Lower Fatty Acids (c2-C~) in Practical Specimens Fig. 2(A) shows a typical gas chromatogram of acetic (peak l), propionic (2), isobutyric (3), butyric (4) and isovaleric acid (5) in ambient air near accumulated poultry manure. The volume concentrated was 0.4 1, and the sample air was trapped directly in the FFAP + H ,PO4 pre-column. The pre-concentration procedure was as described under Experimental. Fig. 2(B) shows a typical gas chromatogram of sample air collected on the FFAP + &PO4 pre-column, but passed through the alkaline pre-column.In this chromatogram, the peaks of isobutyric, butyric and isovaleric acids have disappeared completely. The lower fatty acids were identified and quantitated from the difference between chromatograms (A) and (B) in Fig. 2. Fig. 2(C) shows a typical gas chromatogram of small amounts of the six lower fatty acids.February, 1981 AIR NEAR ODOUR SOURCES BY GC 169 I I Fig. 1. Calibration graphs for six standard lower fatty acids using FFAP + H,PO, pre-column: C, acetic acid; 0, propionic acid; x , isobutyric acid; A, butyric acid; 0, isovaleric acid; and *, valeric acid. Main analytical column conditions : 0.3% FFAP + 0.3% H,PO, on Carbopack B (60-80 mesh) ; 1.5 m x 3 mm i.d., glass; 220 "C; N,, 50 ml min-l ; FID.Pre-column conditions : 0.1 yo FFAP + 0.1% H,PO, on Carbopack C (80-100 mesh); 16 cm x 4 mm i.d., glass; temperature programmed from room temperature (21 "C) to 200 "C in 24 s and maintained a t 200 "C for 30 s. L' A L B hi6 I Time/m in Fig. 2. Typical gas chromatograms of lower fatty acids (A), (B) in ambient air near an ac- cumulated poultry manure and six standard lower fatty acids (C). (A) Sample volume, 0.4 1; FFAP + H,PO, pre-column. Peaks: 1, acetic acid; 2, propionic acid; 3, isobutyric acid; 4, butyric acid; and 5, isovaleric acid. (B) Sample volume, 0.4 1; FFAP + H,PO, pre- column, but passed through an alkaline pre- column. (C) Six standard lower fatty acids; FFAP + H,PO, pre-column. Peaks: 1, acetic acid (6 ng); 2, propionic acid (7.4 ng); 3, iso- butyric acid (8.8 ng); 4, butyric acid (8.8 ng); 5, isovaleric acid (10.2 ng); 6, valeric acid (10.2 ng).Main analytical column temperature, 220 "C. FID sensitivity, 2 x lo2. Other condi- tions as described under Experimental. The concentrations of the six lower fatty acids detected in practical specimens are listed in Table 111, which also gives the odor characteristics (quality and intensity) of the sample gases. Deibels found butyric acid, ethanol and acetoin to be the chief volatilecompounds in accumu- lated poultry manure. Burnettlo used a combination of gas chromatography and organo- leptic techniques to determine the compounds responsible for the offensive odour. The vola- tile components of liquid manure supernatant were identified, using the cold trapping method with acetone - dry-ice and gas chromatography, as mercaptans, organic acids, indole and skatole.Sulphur compounds , organic acids and skatole were implicated as major malodorous components. The acidic volatile compounds reported in poultry liquid manure were acetic, propionic, isobutyric, butyric, isovaleric and valeric acids. The acidic compounds in swine manure have been reported to be formic, acetic, propionic and butyric acids.l19l2 Yasuhara and Fuwa13 isolated seventeen carboxylic acids and four phenols from liquid swine manure by steam distillation and identified them by gas chromatography - mass spectro- metry. The major components were butyric, isovaleric, benzoic and phenylacetic acids and $-cresol.Only the carboxylic acid fraction has the same odour as liquid swine manure. Butyric, isovaleric and phenylacetic acids have a very strong odour of manure.170 HOSHIKA: TRACE AMOUNTS OF LOWER FATTY ACIDS IN Analyst, vd. 106 TABLE I1 RELATIONSHIP BETWEEN RECOVERY OF SIX LOWER FATTY ACIDS ON THE FFAP + &PO, Values given are percentage recoveries at 40 "C. PRE-COLUMN AND NITROGEN CARRIER GAS VOLUME PASSED THROUGH THE PRE-COLUMN 7 Acid 0.5 Acetic acid . . .. . . 100 Propionic acid . . .. . . 100 Butyric acid . . .. . . 100 Isobutyric acid.. .. . . 100 Valeric acid . . .. . . 100 Isovaleric acid . . .. . . 100 Gas volume/l - 1.5 5.0 98.9 86.5 99.4 94.9 100 100 100 100 100 100 100 100 Okabayashi et aZ.14 analysed trace concentrations (1-15 p.p.b.) of lower fatty acids (C,-C,) in air from a poultry manure farm and exhaust gases from heated domestic animal faeces such as pig, cow and hen, using chemical trapping with 50 ml of a 1% aqueous solution of sodium hydroxide in two impingers, followed by an acidic diethyl ether extraction method.Hoshika and co-worker~1~~~6 reported analyses of sulphur compounds, lower aliphatic amines, ammonia and lower fatty acids in the exhaust gas from a poultry manure dryer. At least four lower fatty acids in the exhaust gas were identified (isovaleric, isobutyric, butyric and valeric acids, 200-15900 p.p.b.) by the use of chemical trapping with 10% sodium hydrox- ide solution and decomposition with concentrated sulphuric acid and extraction with diethyl ether. Tsuji and co-worker~,~~@ reported the detection of lower fatty acids (c,&) in the exhaust gas from a poultry manure dryer (18-770 p.p.b.), in the air of a pig house (2.2-260 p.p.b.) and in ambient air near accumulated cow manure (2.4-190 p.p.b.) using an alkaline filter-paper collection method followed by acidic decomposition and diethyl ether extraction.TABLE I11 DETERMINATION OF CONCENTRATIONS OF LOWER FATTY ACIDS (CZ-C,) IN PRACTICAL SPECIMENS Fatty acid concentration, p.p.b. Acetic Propionic Butyric Isobutyric Valeric Isovaleric * I \ Sample gas Volume/l acid acid acid acid acid acid Ambient air near accumulated poultry manure . . . . 0.4 304 82 6 6 n.d.* 3 Ambient air near a fish meal Air in a pig pen.. . . . . 0.2 1540 990 247 164 20 41 factory . . . . . . 0.4 30 n.d.* 4 3 n.d.* n.d.* Exhaust gas from a corn starch manufacturing factory .. 0.3 2000 32 57 34 182 22 Exhaust gas from a poultry manuredryer .. .. 0.2 1696 1190 546 399 37 250 - 1000 34 1 1.3 0.62 2.7 Odour recognition threshold concentration" . . . . * n.d. = not detected (concentration less than 0.5 p.p.b.). Odour characteristics Rancid, Strong putrid, faecal, rotten eggs, ammoniacal Fishy Moderate Faecal, Strong putrid, pungent pungent, strong rancid, sour Sulphury, Very Ammoniacal, Very rancid, strong rotten eggs, faecal, putrid Horiba et aZ.19 reported analytical results for lower fatty acids (C,-C,) in ambient air of a fish meal factory (2.4-15.9 p.p.b.) and a pig house (1.6-11.3 p.p.b.) using an alkaline filter-paper collection method followed by acidic decomposition and chloroform extraction.February , 1981 AIR NEAR ODOUR SOURCES BY GC 171 However, these methods14-19 are not suitable for the routine analysis of large numbers of samples, because the trace analysis of lower fatty acids (C,-C,) in ambient air and exhaust gas samples requires large sample volumes (about 10-1 000 1).In the method described here, the sample volume was as small as about 0.4 1 and the time required for the whole procedure, including the concentration and analysis of one sample, was about 15 min. As shown in Table 111, the concentrations of the lower fattv acids measured, especially butyric and isobutyric acids, were well above the odour thresholds. Therefore, these compounds may be responsible for the unpleasant odours frequently associated with poultry, fish and pig waste.The coefficient of variation was less than 6%. The author thanks Professor Dr. T. Shono, Osaka University, and Dr. K. Yoshimoto, Aichi Environmental Research Centre, for useful suggestions. 1. 2. 3. 4. 5. 6. 7. 8. 9. 10. 11. 12. 13. 14. 15. 16. 17. 18. 19. References Fasset, D. W., in Patty, F. A., Editor, “Industrial Hygiene and Toxicology, Volume 11, Toxicology,” Kuksis, A., Sep. Purif. Methods, 1977, 6, 353. Ackman, R. G., and Metcalfe, L. D., J . Chromatogr. Sci., 1975, 13, 397 and 453. DiCorcia, A., Fritz, D., and Bruner, F., Anal. Chem., 1970, 42, 1500. DiCorcia, A., and Bruner, F., Anal. Chem., 1971, 43, 1634. DiCorcia, A., Anal. Chem., 1973, 45, 492. DiCorcia, A., and Samperi, R., Anal. Chem., 1974, 46, 140. Hoshika, Y., Bunseki Kagaku, 1978, 27, 381. Deibel, R. H., in Rrady, N. C., Editor, “Agriculture and the Quality of Our Environment,” American Association for the Advancement of Science, Washington, D.C., Pub. No. 85, 1967, pp. 395-399. Burnett, W. E., Environ. Sci. Technol., 1969, 3, 744. Miner, J. R., and Hazen. T. E., Trans. Am. SOC. Agric. Eng., 1969, 12, 772. Merkel, J. A., Hazen, T. E., and Miner, J . R., Trans. Am. SOC. Agric. Eng., 1969, 12, 310 and 315. Yasuhara, A., and Fuwa, K., Bull. Chem. SOC. Jpn., 1977, 50, 731 and 3029; 1979, 52, 114, Okabayashi, M., Ishiguro, T., Hasegawa, T., and Shigeta, Y . , Bunseki Kagaku, 1976, 25, 436. Hoshika, Y., Kadowaki, S., Kozima, I., Koike, K.. and Yoshimoto, K., Bunseki Kagaku, 1974, 23, Hoshika, Y., Yoshida, H., Takai, Y., and Yoshimoto, K., Bull. Aichi Environ. Res. Centre, 1979, Tsuji, M., Yamasaki, T., Okuno, T., and Shintani, Y., Hyogoken Kogai Kenkyu Hokoku, 1977, No. 9, Tsuji, M., Okuno, T., Yamasaki, T., and Shintani, Y . , Hyogoken Kogaz Kenkyu Hokoku, 1978, No. 10, Horiba, H., Yamanaka, S., and Hattori, T., Akusyu no Kenkyu ( J . Odor Control), 1978, 7(31), 35. Second Edition, Interscience, New York, 1962, pp. 1771-1795. 917. No. 7, 96. 71. 51. Received June 20th, 1980 Accepted September Sth, 1980

ISSN:0003-2654    

DOI:10.1039/AN9810600166

出版商:RSC    

年代:1981

数据来源: RSC