摘要:

THE ANALYSTTHE ANALYTICAL JOURNAL OF THE CHEMICAL SOCIETYEDITORIAL ADVISORY BOARD"Chairman: J. M. Ottaway (Glasgow)R. Belcher (Birmingham)L. J. Bellamy, C.B.E. (Waltham Abbey)L. S. Birks (U.S.A.)E. Bishop (Exeter)L. R. P. Butler (South Africa)E. A. M. F. Dahmen (The Netherlands)A. C. Docherty (BJlingham)D. Dyrssen (Sweden)W. T. Elwell (Birmingham)J. Hoste (Belgium)H. M. N. H. Irving (Leeds)M. T. Kelley ( U S A . )W. Kemula (Poland)'J. H. Knox (Edinburgh)G . W. C. Milner (Harwell)*H. J. Cluley (Wembley)'P. Gray (Leeds)G. H. Morrison (U.S.A.)H. W. Nurnberg (West Germany)E. Pungor (Hungary)D. I. Rees (London)'R. Sawyer (London)P. H. Scholes (Sheffield)"W. H. C. Shaw (Greenford)S. Siggia (U.S.A.)'D. Simpson (Thorpe-le-Soken)A. A.Smales, O.B.E. (Harwell)"A. Townshend (Birmingham)A. Walsh (Australia)T. S. West (Aberdeen)"J. Whitehead (Stockton-on- Tees)A. L. Wilson (Medmenham)P. Zuman (U.S.A.)"G. E. Penketh (Billingham)*Members of the Board serving on The Analyst Puhlications CommitteeREGIONAL ADVISORY EDITORSDr. J . Aggett, Department of Chemistry, University of Auckland, Private Bag, Auckland, NEW ZEALAND.Professor G. Ghersini, Laboratori CISE, Casella Postale 3986, 201 00 Milano, ITALY.Professor L. Gierst, Universitk Libre de Bruxelles, Facult6 des Sciences, Avenue F.-D. Roosevelt 50,Professor R. Herrmann, Abteilung fur Med. Physik., 63 Giessen, Schlangenzahl 29, W. GERMANY.Professor W. A. E. McBryde, Faculty of Science, University of Waterloo, Waterloo, Ontario, CANADA.Dr.W. Wayne Meinke, KMS Fusion Inc., 3941 Research Park Drive, P.O. Box 1567, Ann Arbor,Dr. 1. Rubeika, Geological Survey of Czechoslovakia, Kostelni 26, Praha 7, CZECHOSLOVAKIA.Professor J. Riii6ka. Chemistry Department A, Technical University of Denmark, 2800 Lyngby,Professor K. Saito, Department of Chemistry, Tohoku University, Sendai, JAPAN.Dr. A. Strasheim, National Physical Research Laboratory, P.O. Box 395, Pretoria, SOUTH AFRICA.Bruxelles, B ELG I U M.Mich. 481 06, U.S.A.DENMARK.Published by The Chemical SocietyEditorial: The Director of Publications, The Chemical Society, Burlington House,London, W1V OBN. Telephone 01 -734 9864. Telex No. 268001Advertisements: Advertisement Department, The Chemical Society, Burlington House, Piccadilly,London, W1 V OBN. Telephone 01 -734 9864Subscriptions (non-members): The Chemical Society, Distribution Centre, Blackhorse Road,Letchworth, Herts., SG6 1 HNVolume 104 No 1235 February 19790 The Chemical Society 197

ISSN:0003-2654    

DOI:10.1039/AN97904FX005

出版商:RSC    

年代:1979

数据来源: RSC

摘要:

ANALAO 104 (1 235) 97-1 76 (1 979)ISSN 0003-2654February 197997106111117124136143146148151154156160164167172174176THE ANALYSTTHE ANALYTICAL JOURNAL OF THE CHEMICAL SOCIETYCONTENTSSub-micrometre Particle Size Characterisation and Distribution by MercuryPenetration-Nayland G. Stanley-WoodDetermination o f Acrylonitrile Monomer in Plastic Packaging and Beveragesby Headspace Gas Chromatography-G. B.-M. GawellDetermination o f Ethylenethiourea in Ethylenebisdithiocarbamate Fungicides:Comparison o f High-performance Liquid Chromatography and Gas - LiquidChromatography-D. S. Farrington and R. G. HopkinsFluorimetric Determination o f Acetohexamide in Plasma and Tablet Formula-tions Using 1 -Methylnicotinamide-Pamela Girgis-Takla and loannis ChroneosPolarographic Determination o f Trace Elements in Food from a Single Digest-M.Kapel and M. E. KomaitisApparatus f o r the Automatic Preparation of Soil Extracts for Mineral-nitrogenDetermination-J. A. P. Marsh, R. Kibble-White and C. J. StentSHORT PAPERSSpectrophotometric Determination o f Dequalinium Chloride in PharmaceuticalPreparations-C. P. Leung and S. Y. KwanSpectrophotometric Determination o f lsoprenaline Sulphate and MethyldopaUsing Chloranil-Mohamed A. Korany and Abdel-Aziz M. WahbiSpectrophotometric Determination of Microgram Amounts o f Hydroquinone,Pyrogallol and Resorcinol-Q. S. Usmani, M. M. Beg and I. C. ShuklaSimple Procedure for the Determination o f Total Carbon and i t s Radioactivityin Soils and Plant Materials-R.C. DalalDetermination of Lead in Columbite Concentrates by Atomic-absorptionSpectrometry After Sulphide Separation-C. ChowDirect Determination of Calcium, Magnesium and Zinc in Lubricating Oils andAdditives by Atomic-absorption Spectrometry Using a Mixed SolventSystem-Zsuzsa WittmannSpectrophotometric Determination of Iron by Synergistic Extraction withlsonitrosobenzoylacetone and Pyridine-B. J. Desai and V. M. ShindePotentiometric Method for the Rapid Determination o f Sulphate in the Presenceo f Chromi um (VI)-R. PrasadDetermination of Thiocyanates by Thermal Decomposition o f Silver Thio-cyanate-A. Cygahski and T. MajewskiCO M M U N ICATIONSelective Determination o f Arsenic(ll1) and Arsenic(V) by Atomic-absorptionBook ReviewsErratumSpectrophotometry Following Arsine Generation-Susumu NakashimaSummaries of Papers in this Issue-Pages iii, iv, v, v iPrinted by Heffers Printers Ltd Cambridge EnglandEntered as Second Class at New York, USA, Post Offic

ISSN:0003-2654    

DOI:10.1039/AN97904BX007

出版商:RSC    

年代:1979

数据来源: RSC

摘要:

February, 1979 SUMMARIES OF PAPERS I N THIS ISSUE iiiSummaries of Papers in thisSub -micrometre Particle Size Characterisation and Distribution byMercury PenetrationThe technique of mercury porosimetry is usually regarded as a method for thedetermination of surface areas and the evaluation of pore size distributions inporous solids. With fine non-porous or microporous materials the initial low-pressure region of a mercury penetration graph can be used to determine theinter-particle spaces or voids in an assembly of discrete particles.A determination of the particle size and distribution of three powders, in theone micrometre and sub-micrometre size range, has been obtained frommercury porosimetry breakthrough and intrusion pressures. The mercuryintrusion particle diameters and distributions are compared with valuesobtained by gravitational and centrifugal sedimentation methods and electronmicroscopy counts for particle size measurement.Keywords : Pavticle size chavactevisation ; mevcuvy fienetvationNAYLAND G.STANLEY-WOODPostgraduate School of Studies in Powder Technology, University of Bradford, GreatHorton Road, Bradford, BD7 1DP.Analyst, 1979, 104, 97-105.Determination of Acrylonitrile Monomer in Plastic Packaging andBeverages by Headspace Gas ChromatographyA gas - liquid chromatographic method for determining trace amounts ofacrylonitrile in plastic containers and carbonated beverages using a nitrogen-sensitive detector and headspace injection technique is described. The methodis suitable for the determination of acrylonitrile a t concentrations down to0.1 mg kg-l in plastics and 0.005 mg kg-l in beverages.Keywovds : A cvylonitvile deterwination ; headspace gas chvomatogvafihy ;plastic packaging; foodstuflsG. B.-M.GAWELLNational Food Administration, Food Research Department, Box 622, S-751 26Uppsala, Sweden.Analyst, 1979, 104, 106-110.Determination of Ethylenethiourea in EthylenebisdithiocarbamateFungicides : Comparison of High-performance LiquidChromatography and Gas - Liquid ChromatographyA rapid, sensitive method is described for the determination of ethylene-thiourea (imidazolidine-2-thione) in ethylenebisdithiocarbamate fungicides.High-performance liquid chromatography is used with an ultraviolet spectro-photometric detector.The results are compared with those obtained usinggas - liquid Chromatography. All fungicide samples assayed containedethylenethiourea, and gas - liquid chromatography indicated higher con-centrations than high-performance liquid chromatography.Keywords 1 Ethylenethiouvea determinatiow ; ethylenebisdithiocavbamate fungi-cides ; high-pevfovmance liquid chvomatogvaphy ; gas - tiquid chvomato-P P h YD. S. FARRINGTON and R. G. HOPKINSDepartment of Industry, Laboratory of the Government Chemist, Cornwall House,Stamford Street, London, SE1 9NQ.Analyst, 1979, 104, 111-116iv SUMMARIES OF PAPERS I N THIS ISSUEFluorimetric Determination of Acetohexamide in Plasma andTablet Formulations Using 1 - MethylnicotinamideFebruary, 1979A sensitive method is described.for the fluorimetric determination of aceto-hexamide in plasma or in tablets by means of its reaction with l-methyl-nicotinamide, which is shown to be a useful reagent for the determination ofketonic compounds. The limit of detection is approximately 0.2 p g ml-land the relative standard deviation is 3.1% for 2 pg ml-l in plasma.Acetoacetic acid usually does not interfere, but can be separated, if necessary,from acetohexamide by means of a .washing technique. No interference iscaused by the presence of insulin, other (non-ketonic) oral hypoglycaemicdrugs, acetone or pyruvic or cc-ketoglutaric acid.Keywords A cetohexamide determination ; plasma ; tablets ; l-methylnicotin-amide reagent ; spectro$uorimetryPAMELA GIRGIS-TAKLA and IOANNIS CHRONEOSWelsh School of Pharmacy, University of Wales Institute of Science and Tech-nology, King Edward VII Avenue, Cardiff, CF1 3NU.Analyst, 1979, 104, 117-123.Polarographic Determination of Trace Elements in Foodfrom a Single DigestThe determination of 12 trace elements, namely copper, zinc, mercury, lead,cadmium, iron, tin, chromium, arsenic, antimony, selenium and tellurium, inadmixture by means of a cathode-ray polarograph is described. The elementswere investigated in concentrations ranging from 0.1 to 20 p.p.m. by meansof normal, reverse-sweep or resistance - capacitance derivative techniques.The last technique could not be used for mercury, although it was used forall the other elements in concentrations less than 1 p.p.m.The completedetermination took 4-7 h, and was applied to various kinds of food, such asbread, meat and vegetables.Keywords : Trace element determination ; food analysis ; polarography ; singledigestM. KAPEL and M. E. KOMAITISProcter Department of Food Science, The University of Leeds, Leeds, LS2 9JT.Analyst, 1979, 104, 124-135.Apparatus for the Automatic Preparation of Soil Extracts forMineral-nitrogen DeterminationAn apparatus is described that automatically prepares samples and feeds anAutoAnalyzer system. It consists of a reagent adder, which adds the correctvolume of extractant for an approximately weighed amount of soil, and asample preparation unit, which mixes, filters, dilutes and loads samples on toan AutoAnalyzer sampler. The results obtained using the apparatus were ingood agreement with those obtained by manual sample preparation.Keywords Mineral-nitrogen deternzination ; soil analysis ; automatic extvactionJ. A. P. MARSH, R. KIBBLE-WHITE and C. J. STENTAgricultural Research Council Weed Research Organization, Begbroke Hill, Yarnton,Oxford, OX6 1PF.Analyst, 1979, 104, 136-142

ISSN:0003-2654    

DOI:10.1039/AN97904FP009

出版商:RSC    

年代:1979

数据来源: RSC

摘要:

February, 1979 SUMMARIES OF PAPERS I N THIS ISSUESpectrophotometric Determination of Dequalinium Chloride inPharmaceutical PreparationsShort PaperKeywords : Dequalinium chloride determination ; spectrophotometry ; picricacidC. P. LEUNG and S. Y. KWANGovernment Laboratory, Oil Street, North Point, Hong Kong.Analyst, 1979, 104, 143-146.VSpectrophotometric Determination of Isoprenaline Sulphate andMethyldopa Using ChloranilShort PaperKeywords : Isoprenaline sulphate determination ; methyldopa deterinination ;chloranil reagent ; spectrophotometry ; charge-transfer complexMOHAMED A. KORANY and ABDEL-A212 M. WAHBIFaculty of Pharmacy, University of Alexandria, Alexandria, Egypt.Analyst, 1979, 104, 146-148.Spectrophotometric Determination of Microgram Amounts ofHydroquinone, Pyrogallol and ResorcinolShort PaperKeywords : Hydroquinone determination ; pyrogallol determination ; resorcinoldetermination ; spectropJzotometvic microdetermination ; sodium carbonateQ.S. USMANI, M. M. BEG and I. C. SHUKLADepartment of Chemistry, TJniversity of Allahabad, Allahabad, India.Analyst, 1979, 104, 148-151.Simple Procedure for the Determination of Total Carbon and itsRadioactivity in Soils and Plant MaterialsShort PapevKeywords ; Soil organic carbon determination ; plant carbon determination ;I4C radioactivity nzeasuvement ; chromic acid digestionR. C. DALALUniversity of New England, Armidale, N.S.W. 2351, Australia.Analyst, 1979, 104, 151-154.Determination of Lead in Columbite Concentrates byAtomic- absorption Spectrometry After Sulphide SeparationSJzort PaperKeywords : Lead determination ; columbite concentrates ; atomic-absorptionspectrometry ; sulphide separationC.CHOWGeological Survey Laboratory, P.O. Box 1015, Ipoh, West Malaysia.Analyst, 1979, 104, 154-156vi SUMMARIES OF PAPERS I N THIS ISSUEDirect Determination of Calcium, Magnesium and Zinc inLubricating Oils and Additives by Atomic-absorptionSpectrometry Using ii Mixed Solvent SystemShort PaperFcbvuary, 1979Keywords : Lubricating oil analysis ; calcium determination ; magnesiumdeterwination ; zinc determination ; atomic-absovption spectvometvyZSUZSA WITTMANNHungarian Oil and Gas Research Institute, VeszprCm, Hungary.Analyst, 1979, 104, 156-160.Spectrophotometric Determination of Iron by Synergistic Extractionwith Isonitrosobenzoylacetone and PyridineShort PaperKeywords : Synergistic ivon extvactiosa ; iron determination ; spectrophotometry ;alloy analysis ; asonitrosobenzo3,lacetoneB. J.DESAI and V. M. SHINDEDepartment of Chemistry, Shivaj i Univcrsity, Kolhapur 41 6 004, India.Analyst, 1979, 104, 160-163.Potentiometric Method for the Rapid Determination of Sulphatein the Presence of Chromium(V1)Short PaperKeywords : Sulphate determinatzon ; barium ion-selective electrode ;chromium( V I ) ; potentiowefryR. PRASADlnco Europe Limited, European Research and Development Centre, Wiggin Street,Birmingham, B 16 OA J.Analyst, 1979, 104, 164-167.Determination of Thiocyanates by Thermal Decomposition ofSilver ThiocyanateSJzort PapevKeywords : Thiocyanate determinaticn ; tliermal decomposition of silver thio-cyanate ; halide and thiocyanate determinationA. CYGANSKI and T. MAJEWSKIInstitute of General Chemistry, Technical University, ul. Zwirki 36, 90-924 L6di,Poland.Analyst, 1979, 104, 167-171.Selective Determination of Arsenic( 111) and Arsenic(V) byAtomic- absorption Spectrophotometry Following Arsine GenerationCommu nicationKeywovds: Arsenic (111) and arsenic ( V ) selective determination; atomic-absorption spectvophotometry ; hydvide generationSUSUMU NAKASHIMAInstitute for Agricultural and Biological Sciences, Okayama University,Kurashiki-shi, Okayama 710, Japan.Analyst, 1979, 104, 172-173

ISSN:0003-2654    

DOI:10.1039/AN97904BP011

出版商:RSC    

年代:1979

数据来源: RSC

摘要:

FEBRUARY 1979 The Analyst Vol. 104 No. 1235 Sub-micrometre Particle Size Characterisation and Distribution by Mercury Penetration Nayiand G. Stanley-Wood Postgraduate School of Studies in Powder Technology, University of Bradford, Great Horton Road, Bradford, BD7 1DP The technique of inercury porosimetry is usually regarded as a method for the determination of surface areas and the evaluation of pore size distributions in porous solids. With fine non-porous or microporous materials the initial low- pressure region of a mercury penetration graph can be used to determine the inter-particle spaces or voids in an assembly of discrete particles. A determination of the particle size and distribution of three powders, in the one micrometre and sub-micrometre size range, has been obtained from mercury porosimetry breakthrough and intrusion pressures.The mercury intrusion particle diameters and distributions are compared with values obtained by gravitational and centrifugal sedimentation methods and electron microscopy counts for particle size measurement. Keywords Particle size characterisation ; mercury penetration The high-pressure mercury intrusion technique is commonly used to characterise, in terms of the volume, number and distribution, the void or pore spaces in porous materials. The evaluation and interpretation of the volume of mercury which penetrates, under pressure, into porous solids was initially proposed by Washburn,l,2 who applied the Young - Laplace equation for capillary rise in cylindrical tubes to the measurement of the size of pores in solids: P = (--2yLccos e)lY .. .. .. .. - * (1) where P is the applied intrusion pressure, yLa the surface tension of mercury, 8 the contact angle of mercury and Y the radius of the cylindrical tube. In reality, however, the results of pore size distribution analysis obtained by use of this equation give only an indistinct image of the real situation, as the model chosen is one of a collection of perfectly cylindrical tubes open at both ends.3 The limitations of the Washburn model were recognised by de Boer,* Frevel and Kre~sley,~ Kruyers and Mayer and S t o ~ e . ~ * * Frevel and Kressley proposed an alternative model to describe the penetration of mercury or fluid under pressure into void spaces within a solid sample composed of a collection of non-porous uniform spheres.The derived mathematical relationship allowed the deter- mination and direct comparison, from the initial penetration or breakthrough pressure of mercury, P*, into this assembly of non-porous spheres, of a surface-area equivalent spherical radius, yS, with that of a mercury-porosimeter equivalent particle radius, Ym, over the porosity range 39.54-25.95y0. Later, Mayer and Stowe described, in more general terms, the breakthrough pressure required for the penetration of a fluid into a collection of non-porous, uniform, solid spheres over the extended porosity range of 47.64-25.95y0. This model was subsequently modified to evaluate the toroidal void volume between touching spheres P = -Y&'/A)/Yg . . .. .. .. * (2) where the function (L'/A) was regarded as equivalent to the breakthrough pressure P* and was calculated for all degrees of packing between the two porosities of hexagonal close- packed and cubic-packed spheres (Tables I1 and I11 in reference 7).9798 STANLEY-WOOD : SUB-MICROMETRE PARTICLE SIZE A?"ySt, VOJ. 104 Many attempts3,9,10 have been made to measure the size of irregularly shaped non-porous particles by mercury intrusion. This technique has, however, been criticised because the Frevel and Kressley and Mayer and Stowe models of an assembly of regular monosized spheres are used to characterise a random polysized assembly of particles. Agreement of the mercury particle radius, rm, calculated from the regular monosized sphere model of Mayer and Stowe, with the radii determined from other independent particle characterisation techniques is, according to Van Brake1,ll fortuitous. Svata and ZabranskylO have shown experimentally, however, that the regular sphere model of Mayer and Stowe can be used to evaluate the particle size and the size distribution of spherical and non-spherical particles.The particle size measurement by mercury porosimetry of various sized and shaped bodies of poly(methy1 methacrylate), carbonyl iron and nickel and fritted glass, all non-porous solids, showed close agreement with the particle size determined by sedimentation and microscope- count techniques. When the mercury intrusion technique is used with porous particles, there is a difficulty in separating the effect of the volume of mercury that penetrates the spaces between particles (voids) from the effect of the volume of mercury that penetrates the spaces within particles.The purpose of this investigation was to show whether mercury intrusion could be used to indicate the presence, and measure the shape of the size distribution, of micro- metre and sub-micrometre particles in porous and non-porous powders. Experiment a1 Powders Steel shot This is a plastic-coated steel shot with a mean particle size of 112 pm, as determined by sieve analysis with Endecott test sieves and shaker. The density, as determined with an air pycnometer (Beckman, Model 930) or by mercury displacement, is 7.952 x 103 or 7.615 x lo3 kg m-3, respectively. The theoretical density from the literature12 is 7.750 x 103 kg m-3 (Fig. 1, Table I).I I I I I I I I 1 10 30 50 70 90 99 99.9 99.99 P roba bi I i ty Fig. 1. Sieve and mercury intrusion size distributions for steel shot. 0, Mercury penetration 118 pm; , sieve size 112 pm. Magnesium trisilicate This is a hydrated magnesium silicate. The mean particle size by wide-angle scanning photosedimentation (WASP) is 15.1 pm by mass, with a size range of 3.0-30.0 pm. The number - length particle size obtained from electron microscope photographs at 700 x and 2600 x magnification is 1.9 pm (Fig. 2, Table I), with a size range of 0.38-19.2 pm. The density, as determined by air pycnometry, is 2.17 x lo3 kg m-3.Febmary , 1979 CHARACTERISATION AND DISTRIBUTION BY MERCURY PENETRATION 99 Barium sul$hate scanning photosedimentation technique, of 10.7 pm and a size range of 1.8541.0 pm.This fine, white, insoluble powder has a mean particle size, by mass, with the wide-angle The TABLE I PHYSICAL CHARACTERISTICS OF POWDERS Characteristic Density x 103/kg m-3 . . .. .. Specific volume of solid/cms 8-1 . . .. .. .. Mercury volume/cm3 g-l in pores Structure . . .. .. {voids .. .. .. Porosity (c) . . .. .. .. Packing anglelangular degrees . . .. (L'/A)min. = P* p.s.i.a. . . .. .. Mean particle diameter/pm . . . . WASP d,t .. .. .. .. Centrifugal d,t . . .. .. .. Microscope d, . . .. .. .. Mercury d, .. .. .. .. Steel shot 7.952 0.125 5 0.082 9 Non-porous 0.398 4.214 - 7 1-72 112 (sieve) 119.8 - - 118.0 Barium sulphate 4.36 0.229 4 0.845 0.132 Microporous 0.700 90 3.35 17.0 0.57 1.10 4.8 Magnesium trisilicate 2.17 0.460 1.596 0.066 Microporous 0.752 90 3.35 15.1 40% < 2.3 1.85 17.8 Dicalcium phosphate 2.31 0.432 9 } 0.777 Mesoporous 0.642 90 3.35 18.0 15% < 3.0 2.60 6.8 number - length particle size from electron microscope photographs at 650 x and 2600 x magnification is 1.1 pm, with a size range of 36.2-0.54 pm (Fig.3). The density, as deter- mined by air pycnometry, is 4.36 x lo3 kg m--3. Dicalcium phosphate dihydrate Dicalcium phosphate dihydrate is a white, crystalline, water-insoluble powder with a mean particle size, as determined by photosedimentation, of 18.0pm and a size range of 5.3-27.5 pm. The number - length particle size from electron microscope photographs at 650 x and 2600 x magnification is 2.5 pm with a size range of 0.57-39.1 pm (Fig. 4, Table I).100 r 0 0.1 1 10 30 50 70 90 99 99.9 99.99 Probability Fig. 2. Size distributions for magnesium trisilicate. ., WASP by mass distribution; 0. WASP by surface distribution; A, centrifuge by mass; A, centrifuge by mass (different sample mass) ; 0, mercury penetration (e = 0.75); and 0, electron microscope number count. Adsorption Isotherms Adsorption isotherms of all powders were obtained by low-temperature nitrogen adsorption. The apparatus used was similar to that described in British Standard 4359, Part I.l3 All samples were de-gassed at room.temperature (24 5 1 "C) for 16 h at a vacuum of less than100 STANLEY-WOOD : SUB-MICROMETRE PARTICLE SIZE Analyst, VOZ. IOP Torr prior to adsorption measurements being made. The temperature of adsorption was 77 K and the nitrogen gas used was research grade XX from the British Oxygen Company, Wembley, Middlesex.0 0.1 1 10 30 50 70 90 99 99.9 9 I99 Proba bi I i ty Fig. 3. Size distributions for barium sulphate. A, Mercury penetration (c = 0.70); B, WASP by mass distribution; C, electron microscope number count; D, WASP by surface distribution; and E, centrifuge by mass. Brunauer, Emmett and Teller surface area The specific surface area of the powders was calculated from the monomolecular volume of nitrogen adsorbed between the relative pressure range of 0.05-0.35 and the Brunauer, Emmett and Teller equation.14 Mesopore size range (2.0-100 nm) The nitrogen adsorption isotherms of magnesium trisilicate, barium sulphate and dicatcium phosphate dihydrate, measured over the relative pressure range 0.08-0.98, were used to E 3 10.E : FJ a - + . 5 . 0 0 - .- 1.0 L 0.1 0 0.1 1 10 30 50 70 90 99 99.9 99.99 Probability Fig. 4. Size distributions for dicalcium phosphate. A, Mercury penetration (e = 0.64) ; B. WASP by mass distribution; C, WASP by surface distribution; D, electron microscope number count: and E. centrifuge by mass.February, 1979 CHARACTERISATION AND DISTRIBUTION BY MERCURY PENETRATION 101 characterise the porous or non-porous structure of the powders in the pore size range 2.0- 100 nm (0.1 pm). From the adsorption isotherm 40 values of the amount of nitrogen adsorbed into or on to the solid surface at specific relative pressures were taken. The specific relative pressure values were taken in known, positive incremental steps and these, together with the appropriate nitrogen volumes adsorbed per gram of powder, formed the data input to a computer.The pore size and number distribution were calculated from a computer program of the modified mathematical porous model of Barrett, Joyner and Halenda.15 The computer program was written in FORTRAN for use on an ICL 1904 com- puter. The values of the Kelvin radius were calculated from the Kelvin equation and the statistical thickness of the adsorbed layer was calculated from the data of Schull.16 Micropore surface area (pore radii 1.6-2.0 nm or less) The volume of nitrogen adsorbed at specific relative pressures obtained from the experi- mental adsorption isotherms was compared with the thickness of the unimpeded adsorbed nitrogen layer at the same specific relative pressure as obtained from the Lippens and de Boer t curve.17 The resultant Va - t graph was used to measure the amount of micro- porous area within the powders.When a straight line passes through the origin of the V , - t graph the slope of the line is a measure of the non-microporous surface area : .. .. * (3) .. vi3 St = 1.547- .. t where St is the non-microporous area in m2g-l, V , is the volume of nitrogen adsorbed at specific relative pressures and a specific thickness measured in cm3 g-l and t is the statistical thickness of an unimpeded adsorbed nitrogen layer in nanometres. A convex deviation of the line with the statistical thickness axis indicates a microporous powder. A curve concave to the statistical thickness axis, at large values of layer thickness, indicates a solid with mesoporous structure (pore radii of 2.0-100 nm).Mercury Intrusion Measurement of the volume of mercury penetrating the voids and pores within the four samples were made by use of a high-pressure Micromeritics Mercury Porosimeter, Model 905-1. All of the powder samples were de-gassed at room temperature (24 & 1 "C) for at least 16 h at a vacuum of less than Torr. The steel shot was de-gassed until a vacuum of less than 10-2Torr was maintained for 1 h. The pressure on the mercury was increased incrementally from below atmospheric, 1.11 p.s.i.a. (7.5 kPa), up to 47800 p.s.i.a. (328 MPa). The particle sizes of the collection of particles were calculated from the Mayer and Stowe equation, assuming a surface tension of 0.474 N m-l and a contact angle of 130" for mercury.The particle diameter from mercury intrusion (&) in pm was calculated from .. .. - - (4) .. .. 137.5 x P* P dm = where P is the experimental pressure in p.s.i.a. and P* the reduced breakthrough pressure obtained from Table I1 of reference 7 at various powder sample porosities and mercury contact angles. Electron Microscopy Number - Length Counts Scanning electron photographs of the powders were obtained from a Cambridge Stereoscan S4-10 after coating the particles with pure gold. A total of more than 300 particles for each powder were individually measured by the Feret diameter and the particle sizes classified into different size classes in order to obtain a number - length distribution similar to that specified in British Standard 3406.18102 STANLEY-WOOD SUB-MICROMETRE PARTICLE SIZE Analyst, VOZ.I04 Pipette Centrifuge for Sub-micron Powders The size analysis of particles below approximately 5 pm is carried out in a centrifugal field by using a modified pipette centrifuge.lg The centrifugal head is a 16 cm diameter, horizontally mounted bowl in which six narrow-bore tubes, 7 cm in length, are radially attached to a hollow central shaft. The centrifugal bowl is driven by a constant-speed motor at either 750 or 1500 rev s-l and it contains an initial volume of 150 ml of a dilute suspension of powder (less than 0.1% V / V ) in 0.1% m/V Calgon dispersant. Aliquots of 10 ml are extracted from the spinning bowl through the tubes via the hollow central shaft at various time intervals.The amount of powder extracted at these known time intervals is determined gravi- metrically after being dried in a hot oven. The particle size is determined from the Stokes equation, modified for centrifugal force, and the percentage undersize of the powder evaluated for the Kamak equation.20 Results and Discussion Spherical Steel Shot The sieve size distribution of these spherical, non-porous solids is shown in Fig. 1. The mean sieve size was 112 pm. From the mercury intrusion data the total volume of mercury that penetrated the assembly of steel shot when in the high-pressure mercury sample tube was 0.0829 cm3 g-1 (Fig. 5 and Table I). The porosity of this collection of spheres, or any assembly of particles, can be calculated from the relationship Total volume of mercury penetrating the assembly of particles per gram Total volume of mercury per gram +l/density of solid € = Thus, for steel shot, the porosity of the sample in the mercury sample tube was 0.0829/ (0.0829 + 0.1255) = 0.398.The single acute angle for this packing arrangement and porosity, from Table I in reference 7, is between 71" and 72". Interpolation of the Mayer and Stowe general function (L.'/A)min. or reduced breakthrough pressure P* for "square" access openings for the above acute packing angle gives a value of 4.214 p.s.i.a. for break- through when mercury has a contact angle of 130". The particle size and distribution of non-porous spherical steel shot determined from equation (4) and the percentage of mercury penetrating the packed spherical particles is shown in Figs. 1 and 5.The over-all shape of the distribution graph determined by mercury intrusion is similar to that determined by sieve analysis. 1 .o c b, (? 0.9 5 0.8 5 0.7 2 0.6 0.5 - f 0.4 0.3 + 0.2 F g 0.1 O i l + 0 .- + 0, 1 5 10 50 100 500 1000 500010000 50000 Intrusion pressure/p.s.i .a. Fig. 5. Mercury intrusion graphs: A, Steel shot; B, barium sulphate ; C , magnesium trisilicate ; and D, dicalcium phosphate.February, 1979 CHARACTERISATION AND DISTRIBUTION BY MERCURY PENETRATION 103 Barium Sulphate The mass and surface distributions obtained by photo- and centrifugal sedimentation are shown in Fig. 3, together with the distribution obtained by a number - length electron- microscope count and the mercury intrusion technique.Nitrogen adsorption isotherm analysis by the Barrett, Joyner and Halenda method and the Lippens and de Boer V , - t method shows that barium sulphate contains a large number of pores in the size range up to 2.0nm radius. The number of pores then decreases rapidly so that barium sulphate can be regarded as having no meso- or macropores (Fig. 6). This characterisation of the non-mesoporous structure of barium sulphate is substantiated by the linearity of the Va versus t graph (Fig. 7) at nitrogen layer thicknesses greater than 1.4 nm. 0 10 20 30 40 50 60 70 350 400 450 500 Pore radius/nm Fig. 6. Nitrogen adsorption pore size distributions : A, magnesium trisilicate ; B, dicalcium phosphate; and C, barium sulphate. The mercury intrusion graph (Fig.5) indicates, however, the presence of pores, using the Washburn model, in the diameter range 160-0.0036 pm (160000-3.6 nm). The barium sulphate mercury intrusion graph can be readily divided into regions, one region in which mercury is being forced between particles and has a volume of 0.845 cm3 g-l, and another in which mercury is being forced into particles and has a volume of 0.132 cm3 g-l (Table I, Fig. 5). 120 110 100 90 80 70 0 0.2 0.4 0.6 0.8 1.0 1.2 1.4 1.6 1.8 2.0 2.2 2.4 Statistical thickness of nitrogen layer ( t ) /nrn Fig. 7. Micropore Va v e m u t graphs. A, Magnesium trisilicate; R, barium sulphate ; and C, dicalcium phosphate.104 STANLEY-WOOD : SUB-MICROMETRE PARTICLE SIZE Analyst, VoZ. 104 The porosity of the collection of particles assembled in the mercury sample tube can be evaluated in a similar manner to that shown under Spherical Steel Shot.The porosity of barium sulphate is thus 0.845/(0.845 + 0.132 + 0.2294) = 0.700. This porosity value is beyond the range of porosity versus acute angle of packing studied by Mayer and Stowe. Their relationship of breakthrough pressure versus packing angle (Fig. 9 in reference 7) does show, however, a smooth function over the higher packing angles, which tends to a constant value. A measure of the distribution of sizes, although not the surface diameter size predicted by theory, can be obtained when a breakthrough pressure value of 3.35 p.s.i.a. at a 90" packing angle is taken and substituted into equation (4). The size distribution deter- mined from the percentage of mercury penetrating between these irregularly shaped poly- sized particles follows a similar distribution to that determined by a number - length count.The sub-micrometre tail of the barium sulphate is coincidental with the sub-micrometre particle size distribution determined by centrifugal sedimentation, but is not detected by the gravitational photosedimentation technique. Magnesium Trisilicate The mass and surface size distributions obtained by photosedimentation and the mass distribution from centrifugal sedimentation are shown together with the nuniber - length and mercury intrusion size distributions in Fig. 2. The nitrogen adsorption isotherm analyses by the Barrett, Joyner and Halenda and V , - t methods (Figs. 6 and 7) show that magnesium trisilicate is solely a microporous solid.The mercury volume intrusion graph (Fig. 5) falsely indicates that, like barium sulphate, magnesium trisilicate has pores in the diameter size range 160000-3.6nm. The intrusion graph can be readily divided into two regions, one in which the volume of mercury can be attributed to the filling of voids between particles (a volume of 1.596 cm3g-l), and the second attributable to the filling of pores in particles (a volume of 0.066 cm3g1). The porosity of the collection of magnesium trisilicate particles in the mercury sample tube is 0.752 and the breakthrough pressure can be taken as being 3.35 p.s.i.a. The size distribution calculated from equation (4) shows a similarly shaped distribution to that obtained by a combination of the micrometre photo- sedimentation and the sub-micrometre centrifugal sedimentation techniques, as well as that of the number - length microscope count.The particle radius or diameter predicted by the mercury intrusion theory is a surface diameter but the diameter measured by intrusion,with both magnesium trisilicate and barium sulphate, shows a closer correlation with a mass Stokes diameter than with a surface diameter. Adjustment of the experimental mercury diameter distribution to that of a surface diameter distribution would necessitate a breakthrough pressure value, at a packing range of 90" and mercury contact angle of 130°, in the region of 1.52 p.s.i.a. This breakthrough pressure value could only be achieved at a 90" pack if the contact angle between mercury and solid was in the range 110-100".Dicalcium Phosphate Dihydrate The particle size distributions obtained by photosedimentation and centrifugal sedimenta- tion techniques are shown in Fig. 4. The nitrogen adsorption isotherm analysis (Figs. 6 and 7) shows that dicalcium phosphate is non-microporous but has a mesoporous structure. The mercury volume intrusion graph (Fig. 5) cannot readily be divided into two regions; the volume of mercury filling both the voids between particles and the pores in particles is 0.777 cm3 g-l. The porosity of the collection of dicalcium phosphate particles in the mercury sample tube has been calculated to be 0.642, with a breakthrough pressure of 3.35 p.s.i.a. The shape of the calculated mercury intrusion size distribution graph bears little resemblance, at large diameters, to the sedimentation or number - length distributions. The presence of mesopores in dicalcium phosphate has destroyed the physical model upon which the mercury intrusion diameter calculations are based.After mercury has filled the mesopores within the solid, with the result that the solid can be regarded as non-porous at higher mercury pressures, the mercury size distribution obtained from the Mayer and Stowe physical model is similar to that of the centrifugal and number - length distributions.Febmary, 1979 CHARACTERISATION AND DISTRIBUTION BY MERCURY PENETRATION Conclusions 105 With non-porous spheres the mercury intrusion technique evaluates a similarly shaped distribution of particle sizes to that determined by other, more conventional, particle sizing techniques.The diameter (or radius) measured is not, however, the surface diameter (or radius) predicted by the Mayer and Stowe model. The particle sizes measured by the mercury technique are greater than the sieve sizes of spherical shot. With microporous, irregularly shaped particles, the mercury intrusion particle size tech- nique detects both micrometre and sub-micrometre particles, which can usually only be sized by two separate characterisation methods. The mercury diameter evaluated from the microporous magnesium trisilicate powder correlates with the mass Stokes diameter rather than a surface diameter. The correlation between the mercury diameter and the Stokes diameter is, however, dubious for barium sulphate, except for sub-micrometre particles. With meso- or macroporous material little correlation exists between the mercury diameter and the Stokes diameter.This result supports the observations of Van Brakel and Mayer and Stowe that the spherical model does not perfectly represent a real solid. The mercury particle size technique can, however, be used, with additional adsorption information, in order to obtain an over-all size distribution characterisation in the micrometre and sub- micrometre particle size ranges. 1. 2. 3. 4. 6. 6. 7. 8. 9. 10. 11. 12. 13. 14. 15. 16. 17. 18. 19. 20. References Washburn, E. W., Phys. Rev., 1926, 17, 273. Washburn, E. W., Proc. Nutn. Acud. Sci. U.S.A., 1921, 7 , 116. Scholten, J. J. F., Beers, A. M., and Kiel, A. M., J . Catulysis, 1976, 36, 23. de Boer, J. H., Adv. Catalysis, 1969, 9, 139. Frevel, L. K., and Kressley, L. J., Anulyt. Chem., 1963, 35, 1492. Kruyer, S., Trans. Faraduy SOC., 1968, 54, 1768. Mayer, R. P., and Stowe, R. A., J . Colloid Sci., 1966, 20, 893. Mayer, R. P., and Stowe, R. A., J . Phys. Chem., 1960, 70, 3867. Orr, C., Powder Technol., 1969/70, 3, 117. Svata, M., and Zabransky, Z., Powder Technol., 1970, 3, 296. Van Brakel, J., Powder Technol., 1976, 11, 206. Weast, R. C., Editor, “Handbook of Chemistry and Physics,” Fifty-seventh Edition, CRC Press, British Standard 4369 : Part I : 1969. Brunauer, S., Pure A#@. Chem., 1976, 48, 401. Barrett, E. P., Joyner, L. G., and Halenda, P. O., J . Am. Chew. Soc., 1961, 73, 373. Stanley-Wood, N. G., and Johansson, M. E., Drug Dev. Ind. Pharmacy, 1978,4, 69. Lippens, B. C., and de Boer, J. H., J . Catalysis, 1966, 4, 319. British Standard 3406 : 1963. Allen, T., “Particle Size Measurement,” Second Edition, Chapman and Hall, London, 1970. Kamak, H. J., Ind. Engng Chem., Analyt. Edn, 1961, 23, 044. Cleveland, Ohio, 1976. Received July 6th, 1978 Accepted Sefltember 8th, 1978

ISSN:0003-2654    

DOI:10.1039/AN9790400097

出版商:RSC    

年代:1979

数据来源: RSC

摘要:

106 Analyst, February, 1979, Vol. 104, p p . 106-110 Determination of Acrylonitrile Monomer in Plastic Packaging and Beverages by Headspace Gas Chromatography G. 9.-M. Gawell" National Food Administration, Food Research Department, Box 622, S-761 26 Uppsala, Sweden A gas - liquid chromatographic method for determining trace amounts of acrylonitrile in plastic containers and carbonated beverages using a nitrogen- sensitive detector and headspace injection technique is described. The method is suitable for the determination of acrylonitrile at concentrations down to 0.1 mg kg-l in plastics and 0.005 mg kg-l in beverages. Keywords A crylonitrile determination ; headsfiace gas chromatography ; piastic packaging; foodstufls Polymers with acrylonitrile monomer as a component are used for packages and household articles designed for foodstuffs.During the manufacture of acrylonitrile copolymers, a small fraction of unreacted acrylonitrile monomer becomes physically entrapped in the polymer and can migrate slowly during storage and when in contact with food or other materials. Recently, experiments with rats in the USA1 have indicated that acrylonitrile might be carcinogenic. This has led to a proposed ban in the USA on the use of acrylonitrile co- polymer materials for beverage packaging. These packages are also used in Sweden, and this work was initiated to develop a sensitive gas-chromatographic method for the determination of residual acrylonitrile in carbonated beverages and food packages. The methods for the determination of acrylonitrile in food and beverages published so far are often time consuming, require special laboratory equip- ment and are not very sensitive.Jones and Smith2 described a headspace gas-chromatographic method for acrylonitrile in fat with a sensitivity of 0.2 mg k g l . The Nederlandse Vereniging-federatie voor Kunst- stoffen3 has recommended a method that involves distillation into xylene followed by gas chromatography. This method gives a sensitivity of 0.4 mg k g l in aqueous simulants and about 1 mg k g l in fat. The Food and Drug Administration* has recommended a method for the determination of acrylonitrile in aqueous extracts, which involves an azeotropic distillation with methanol, followed by differential-pulse polarography. The sensitivity of this method is reported to be 0.01 mg k g l in water and 0.03 mg k g l in beer.The deter- mination of acrylonitrile in food processing plants using spectrophotometry involving photochemical bromination and the formation of a red pyridine - benzidine complex was described by Kroller.5 The detection limit was reported to be 0.01 mg kg-1. Several methods have been published on the determination of acrylonitrile in plastics,6-8 involving various solvents, gas-chromatographic columns and techniques, The method proposed here is applicable to plastics, carbonated beverages and simulating solvents. The headspace technique and the nitrogen-sensitive detector make the determination of acrylo- nitrile monomer rapid, accurate and sensitive (Fig. 1). Experimental Apparatus Normal laboratory equipment, 20-ml glass vials equipped with natural rubber seals and aluminium caps, a Fermpress H 207 for sealing the bottles and a l-ml gas-tight syringe were used.Reagents All reagents were of analytical-reagent grade. A cry lonitrile. * Present address : Astra Pharmaceuticals AB, S-161 86 Sodertalje, Sweden.GAWELL 107 The propylene carbonate was purified prior to use by the following The propylene carbonate was heated to 120 "C with magnetic stirring and Pvopionitrile. Profiylene carbonate. procedure. nitrogen bubbling for 12 h,* then stored under nitrogen. Gas Chromatography The gas chromatograph was a Varian 2700 with an alkali flame-ionisation detector (rubidium sulphate). Nitrogen was used as the carrier gas at a flow-rate of 23 ml min-1. Air and hydrogen flow-rates were optimised to give the best detector response.The chart speed of the 1-mV recorder was 0.8 cm min-l and the amplifier ranges were 4 x 10-l2 and 8 x 10-12AmV-l. The main column used was a 3 m x 2 mm i.d. glass column packed with 0.2:/, Carbowax 1500 on 60-SO-mesh Carbopack C. The injector temperature was 200 O C , the detector temperature 200 "C and the oven temperature 70 "C, isothermal. In order to confirm the presence of acrylonitrile in a sample, another glass column of 3 m x 2 mm i.d. was used, packed with ZOyo Carbowax 20M on Chromosorb W, 60-80 mesh. The instrumental con- ditions were the same except for the oven temperature, which was 50 "C, and the nitrogen flow-rate, which was 15 ml min-l. The retention times for acrylonitrile and propionitrile on the alternative column were 1.7 min and 2.3 min, respectively.Procedure Preparatiovt of samples of the carbonic acid had been evolved. pieces. Samples of carbonated beverages were shaken in a stoppered Erlenmeyer flask until most Samples of the plastic materials were cut into small Determination of acrylonitrile in beverages A 3-1111 volume of the homogenised sample was weighed into a glass vial, 2.0 ml of d 0 1 2 3 4 5 6 Ti me/m i n Fig. 1. Beer from a glass bottle spiked with 0.03 mg of acrylonitrile per kilo- gram. Propionitrile was added as internal standard. Attenuation 4 x 10-12 A mV-l. I R I I 1 1 I 0 . 1 2 3 4 5 6 Time/m i n Fig. 2. Beer sample from Barex bottle with propionitrile added as internal standard. Attenuation 8 x 10-l2 A mV-l.0 1 2 3 4 5 Time/m in Fig. 3. Plastic sample from Barex bottle with propio- nitrile added as internal standard. Attenuation 8 x 10-l2 A mV-l.108 GAWELL : DETERMINATION OF ACRYLONITRILE MONOMER IN PLASTIC Analyst, VoZ. 104 propionitrile (the internal standard), at a concentration of about 0.5 pg ml-l in distilled water, and 3.0 ml of distilled water were added and the bottle was closed with a rubber seal and cap. The sample was then ready for gas chromatography (Fig. 2). The vial was shaken well and placed in an oven at 90 "C for at least 30 min. Determination of acrylonitrile in plastic containers The plastic sample (0.3-0.5g) was weighed into a glass vial, 4.0ml of propionitrile in propylene carbonate (at a concentration of about 1.5 pgml-l) and 4.0ml of propylene carbonate were added and the bottle was then closed with a rubber seal and cap.The vial was placed in an oven at 90 "C until all the plastic had dissolved arid then left at room temperature overnight. Prior to the gas-chromatographic determination, the glass vial containing the polymer sample was placed in an oven at 90 "C for at least 30 min (Fig. 3). Gas-chromatographic determination (90 "C) gas-tight syringe, and injected into the chromatograph. A l-ml volume of gas was drawn from the gaseous phase in the glass vial using a warm Blank determinations The solutions and solvents were tested by headspace gas chromatography prior to use to ensure freedom from peaks that would interfere with the determination of acrylonitrile (Figs. 4, 5 and 6). Preparation of calibration graphs Acrylonitrile standard solutions of concentrations 0.01, 0.03, 0.05, 0.10 and 0.15 pg ml-l i I I I d 4 I 0 1 2 3 4 5 0 1 2 3 4 5 6 0 1 2 3 4 5 6 0 1 2 3 4 5 Tirne/rn i n Ti me/m i n Tirnehin Time/rnin Fig.4. Beer Fig. 6 . Carbonated Fig. 6. Blankdeter- Fig. 7. Purified sample from a glass soft drink from a glass mination of purified propylene carbonate bottle. Attenuation bottle. Attenuation propylene carbonate. spiked with 0.12 pg 8 x 10-12 A mV-'. 8 x 10-12 A mV-l. Attenuation 8 x 10-12 of acrylonitrile per A mV-l. millilitre, correspond- ing t o 0.96mg kg-l in plastic. Propio- nitrile was added as internal standard. Attenuation 8 x 10-la A mV-l.February, 1979 PACKAGING AND BEVERAGES BY HEADSPACE GAS CHROMATOGRAPHY 109 were prepared in distilled water.A 3.0-ml volume of acrylonitrile standard solution, 2.0 ml of propionitrile solution in distilled water at a concentration of about 0.5 pg ml-l and 3.0 ml of liquid (similar to the sample, but which has never been in contact with any plastic, for example, beer from a glass bottle) were pipetted into a vial. The vial was then sealed, shaken and placed in an oven at 90 "C for at least 30 min. The chromatographic deter- mination was then carried out as described above (Fig. 1). Acrylonitrile standard solutions of concentrations 0.2, 0.4, 0.6, 0.8 and 1.0 pg ml-l were prepared in purified propylene carbonate. A 4.0-ml volume of acrylonitrile standard solution and 4.0ml of propionitrile solution in propylene carbonate, at a concentration of about 1.5 pg ml-1, were pipetted into a vial.The vial was then sealed, shaken and heated in the oven at 90 "C for 30 min. A 1-ml volume of the gaseous phase from the headspace was injected into the chromatograph as described above (Fig. 7). Although the peak heights in Fig. 7 are approximately the same as in Fig. 1, the amounts of acrylonitrile and propionitrile vary and proportionally represent substantially different amounts, because the solubilities of acsylonitrile and propionitrile in propylene carbonate are different from the solubilities in water. Duplicate determinations from each acrylonitrile standard solution were made in addition to duplicate injections from each vial. The concentrations of the standard solutions in the vials correspond to 0.01-0.15mgkg1 in the beverages and 2-10mgkg-l in the plastics.For each pair of injections of the standard solutions, the mean of the peak-height ratios for acrylonitrile and propionitrile were calculated (y) and plotted against the mass ratio of acrylonitrile to propionitrile of each standard solution in the same vial (x) in order to con- struct calibration graphs. Quanti$cation and plastics. The linear equation y = kx + I was used to calculate the acrylonitrile content in beverages (y - J)wis WSk Amount of acrylonitrile in sample (mg kg-l) = where y = mean of the two acrylonitrile to propionitrile peak-height ratios from the gas chromatograms of two headspace injections of the sample solution; I = intercept on the y-axis; Wi, = amount of internal standard added (pg); Ws = amount of sample weighed out (g); and k = slope of the straight line.Results and Discussion The proposed method was used to examine twelve samples of beer and four samples of carbonated soft drinks packed in Barex. The residual acrylonitrile in the packages ranged from 2 to 5 mg kg-1 and in some samples of beer and soft drinks trace amounts of acrylonitrile (<0.005 mg k g l ) were found (Fig. 2). The equations used to calculate the acrylonitrile content in beverages and plastics were y = 2.434% + 0.001 and y = 2.254~ - 0.017, respectively. The acrylonitrile standard solutions in distilled water remained stable for approximately 1 week in a refrigerator. The relative standard deviation calculated on ten determinations on Barex plastic from beer bottles was 2.9% with an acrylonitrile concentration of 2.8 mg k g l and 3.5% with an acrylonitrile concentration of 1.8 mg k g l .The method is also suitable for the determination of acrylonitrile in simulating solvents such as water, 3% acetic acid and 10% ethanol, which are often used in specific migration tests. The determination of acrylonitrile with this method is easy, rapid and, due to the use of the nitrogen-sensitive detector, very sensitive. The linear regression was 0.9999 for both calibration graphs. The accuracy of the determinations in plastics was tested. The author thanks Mrs. Margaretha Adolfsson-Erici for her skilful contribution to the experimental work, Miss Marie Kusters for assistance with the English translation and Mr. Bonny Larsson for valuable comments on the manuscript.110 GAWELL References Fd Chem. News, 1977, Jan. 24th, 20. Jones, N. X I . , and Smith, F. J., “Special Report on Monomer Migration from Lustropac 1010 ABS into Soft Margarine and Vegetable Oil,” Monsanto Petrochemicals and Polymers Company, Newport Research, St. Louis, Mo., 1974. Draft Packaging and Utensils Regulation (Food Law), Chapter 12, Section 3, “3.4. Specific Migration. Migration of Acrylonitrile and Methacrylonitrile. 3.4.1.1. Migration in Watery Simulants. 3.4.1.2. Migration in Fat,” Nederlandse Vereniging-federatie voor Kunststoffen , Haarlem, The Netherlands, 1975. Personal communication, Food and Drugs Administration, Washington, D.C., 1976. Kroller, E., Dt. LebensmittRdsch., 1970, 66, 11. Steichen, R. J., Analyt. Chern., 1976, 48, 1398. Bundesgesundheitsblatt, 1977, 20, 162. Personal communication, Lonza AG, Basle, Switzerland. 1. 2. 3. 4. 5. 6. 7. 8. Received August 7th, 1978 Accepted September 1 lth, 1978

ISSN:0003-2654    

DOI:10.1039/AN9790400106

出版商:RSC    

年代:1979

数据来源: RSC

摘要:

Analyst, February, 1979, Vol. 104, PP. 111-116 111 Determination of Ethylenethiourea in Ethylenebisdithiocarbamate Fungicides: Comparison of High-performance Liquid Chromatography and Gas - Liquid Chromatography D. S. Farrington and R. G. Hopkins Department of Industry, Laboratory of the Government Chemist, Cornwall House, Stamford Street, London, SE1 9NQ A rapid, sensitive method is described for the determination of ethylene- thiourea (imidazolidine-2-thione) in ethylenebisdithiocarbamate fungicides. High-performance liquid chromatography is used with an ultraviolet spectro- photometric detector. The results are compared with those obtained using gas - liquid chromatography. All fungicide samples assayed contained ethylenethiourea, and gas - liquid chromatography indicated higher con- centrations than high-performance liquid chromatography.Keywords : Ethylenethiourea determination ; ethylenebisdithiocarbamate fungi- cides ; high-performance liquid chromatography ; gas - liquid chromato- graphy Ethylenethiourea (ETU) (imidazolidine-2-thione) may occur in formulations both as a manufacturing impurity and as a product of degradation during storage.132 The presence of ETU in commercial formulations is of concern because of its possible carcinogenic and teratogenic proper tie^.^^^ However, deficiencies in available data have inhibited any firm conclusions being reached in respect of these properties. The routine method of analysis for the active ingredient in ethylenebisdithiocarbamate (EBDC) formulations is based on the determination of carbon disulphide after acid hydro- lysis.536 This technique is not applicable in the presence of copper and the Dubosq method' is then used, The latter method is based on hydrolytic cleavage using hydriodic acid but ETU interferes. As ETU may be present in formulations to the order of several per cent., this interference can be significant and a correction must be made for the ETU content. Johnson and Tyler* determined ETU in aqueous extracts from a number of EBDC fungicides using paper chromatography.All fungicides based on ethylenebisdithiocarbamic acid were reported to contain ETU. Thin-layer chromatography was used by Czegledi- Janko and Hollo9 to examine the degradation of zineb and maneb. Fishbein and FawkeslO also reported a thin- layer chromatographic method for the determination of the degradation products of EBDC fungicides.Thin-layer chromatography was used by Bontoyan and Looker2 as a screening method for detecting ETU prior to gas - liquid chromatographic analysis of EBDC fungicide formulations maintained under conditions of elevated temperature and humidity. It was thought that high-performance liquid chromatography would give high specificity, speed and sensitivity, and form the basis of a method by which the results obtained by gas - liquid chromatography could be critically assessed. A number of workers have reported methods for the assay of ETU. Experiment a1 Liquid Chromatograph A Waters Associates, Model 6000, constant-volume solvent-delivery system was used. A variable-wavelength ultraviolet monitor (Cecil Instruments, Model CE 212), fitted with a 10-pl flow cell and set at 240 nm, was used as a detector.Preparation of Column A stainless-steel column tube, 180 x 4.6 mm i.d., was washed with chloroform and Crown Copyright.112 FARRINGTON AND HOPIUWS : DETERMINATION OF ETHYLENETHIOUREA Analyst, VoZ. 104 methanol and polished on the inner surface. One end was fitted with a 3 x & in Crawford Patent column end fitting, and the other end was coupled to a 400-mm pre-column through a Spherisorb CN, 5 p m (Phase Separations Ltd.), was packed into the column from a slurry in ethanol under a pressure of 5000 lb k2. The pre-column was removed and the analytical column was prepared for stop-flow injection by removing the top few millimetres of packing and inserting a disc of stainless-steel fine-mesh gauze, of 8 pm nominal porosity, a plug of silanised glass-wool and a top plug of porous PTFE.Sample Injection A Varian Associates stop-flow injector was used and samples were injected on to the stainless-steel fine-mesh gauze fitted on top of the packing. A needle guide was incorporated in the injector to ensure that the samples were introduced on to the centre of the top of the column. x $in Swagelok union. Conditions for Gas - Liquid Chromatography ionisation detector, under the following operating conditions : Analyses were carried out on a Pye 104 gas - liquid chromatograph, fitted with a flame- Column . . .. .. .. . . . . 1000 x 4 mm i.d., glass Column packing . . .. .. . . Chromosorb WHP, 2% of Carbowax 20M TPA on Chromosorb W, 80-100 mesh Column oven temperature .. . . .. 210°C Detector oven temperature . . . . . . 250 "C Injection zone temperature . . .. . . 230°C Carrier gas . . . . .. .. . . Nitrogen, free from oxygen Flow-rates: nitrogen . . .. . . . . 50 cm3 min-l hydrogen . . .. .. . . 50 cm3 min-l air .. .. .. . . 600 cm3 min-l Reagents Ethylenethiourea. ETU standard solution. Ethanol, absolute. Methanol. Analytical-reagent grade. Hexane. Spectrograde (Fisons). MobiZe phase for liquid chromatografihy. Recryst allised, obtained from Robinson Brothers Lt d. Prepare a standard solution in methanol, containing 0.7 g 1-1 of ETU. Prepare a solution containing 35% VjV of absolute ethanol in hexane. Procedure Weigh accurately about 0.5 g of EBDC, or a mass equivalent to 0.5 g of active ingredient, into a 15-cm3 centrifuge tube.Add 10.0 cm3 of methanol from a pipette, stopper the tube and shake it vigorously for 15min on a wrist-action shaker. Filter the resulting mixture through a filter-paper that retains particles of at least 5 pm, for example, Whatman No. 42, and collect the filtrate. Inject 4 pl of the filtrate on to the top of the liquid chromatograph using a flow-rate of 0.8 ml min-l. Complete the analysis as soon as possible. Calculate the ETU content of the sample by comparing the peak height with that obtained from a 4-pl injection of standard solution. To ensure that the response of the detector is within the linear range, dilute the sample until the peak height obtained is equal to or less than that obtained with the standard.Gas - liquid chromatographic analyses are undertaken using the same solution and the same injection volume as are used for liquid-chromatographic determinations. Results and Discussion A number of procedures were investigated for extracting ETU from EBDC fungicides. A cold-extraction technique was adopted in order to reduce the risk of production of ETU. Optimum speed and repeatability were obtained by shaking the fungicides with solvent forFebrunvy, 1979 I N ETHYLENEBISDITHIOCARBAMATE FUNGICIDES 113 15 min. Duplicate analyses of a sample of ziram fortified with ETU (3%) gave recoveries of approximately 90%. The use of water has been reported; however, it was found that this extraction solvent resulted in un- acceptably broad peaks on the liquid Chromatograph.Methanol and methanol - chloro- form have also been used for extraction and these were investigated. Both systems resulted in good peaks on the liquid chromatograph and were shown to have similar extraction efficiencies with respect to ETU. However, methanol - chloroform co-extracted greater amounts of other substances and therefore methanol was adopted as the extraction solvent. ETU has a sharp absorbance at 240 nm and this was adopted as the wavelength setting on the UV monitoring system. ETU was readily eluted from a Spherisorb ODS system, but adequate retention could not be obtained. A bonded stationary phase exhibiting polar characteristics appeared likely to be suitable and Spherisorb CN was selected. With hexane - ethanol (65 + 35) as the mobile phase, ETU is eluted from a Spherisorb CN packed column with a good peak shape and optimum resolution.Ethylenethiuram monosulphide (ETRI) elutes shortly before ETU, but is not extracted efficiently by methanol and does not absorb strongly at 240nm. The procedure was applied to both technical samples and formulations of zineb, maneb, mancozeb and Vondozeb (a co-ordination product of EBDC with zinc and manganese ions). Figs. 1 and 2 show typical chromatographic traces obtained from analyses of these compounds. With the procedure described, 0.01% of ETU can be determined in the four active ingredients investigated. The choice of the solvent was limited by the low solubility of ETU. The results obtained are shown in Table I. TABLE I ETHYLENETHIOUREA CONTENT OF ETHYLENEBISDITHIOCARBAMATE FUNGICIDES Fungicide Maneb technical 1 ..Maneb technical 2 .. Maneb formulation 1 . . Maneb formulation 2 . . Maneb formulation 3 . . Zineb technical 1 .. Zineb technical 2 .. Zineb formulation .. Maneb - zineb technical . . Maneb - zineb formulation Mancozeb technical 1 . . Mancozeb technical 2 . . Mancozeb technical 3 . . Mancozeb technical 4 . . Mancozeb formulation . . Vondozeb technical 1 . . Vondozeb technical 2 . . Vondozeb technical 3 . . Vondozeb formulation . . Figures are expressed as percentages. Results by high-performance liquid chromatography .. 0.1’ 0.1 .. 1.2, 1.3 .. 0.7, 0.7 .. 1.1, 1.1 .. 1.0, 1.1 .. .. .. 1.1, 1.2 1.5, 1.4 2.1, 2.0 0.4, 0.4 1.0, 1.0 .. 0.1, 0.1 . . 0.2, 0.2 . . 0.2, 0.2 .. 0.2, 0.2 . . 0.2, 0.2 ..1.0 .. 1.3, 1.3 . . 1.4, 1.5 .. 0.9, 1.0 Results by gas - liquid chromatography 0.2, 0.2 2.8, 2.9 1.2, 1.3 2.0, 2.0 1.8, 1.9 1.5, 1.5 2.5, 2.4 2.1, 2.2 0.6, 0.6 1.2, 1.3 0.1, 0.1 0.4, 0.4 0.4, 0.4 0.4, 0.4 0.2, 0.3 1.6 2.1, 2.1 2.9, 2.6 1.5, 1.3 High-performance liquid chromatographic traces obtained from analyses of extracts of maneb samples showed that maneb yields greater concentrations of co-extractives than the other EBDC fungicides examined (Fig. 1). This was also observed by Czegledi- Janko and Hollo9 when analysing a number of EBDC fungicides by thin-layer chromatography. Mancozeb samples contained comparatively little ETU and less co-extractives (Fig. 1) , which agrees with the findings of Bontoyan and Looker.2 Gas - liquid chromatographic analysis of all samples gave two peaks (Fig.2), the second eluting shortly after ETU. A correlation between the area of this peak and the rate of114 FARRINGTON AND HOPKINS : DETERMINATION OF ETHYLENETHIOUREA Analyst, VoZ. 104 formation of ETU from EBDCs was noted by Bontoyan and Looker.2 They concluded that ETU may arise as a result of two decomposition routes, a direct route from the parent EBDC and formation via intermediate degradation products. A mechanism for the formation of ETU through degradation products, but not directly from the parent EBDC, was proposed by Marshall.ll This was based on studies of the thermal decomposition of EBDCs in aqueous media. a ) 1 c u 0 5 10 0 5 10 C ) It L - 0 5 10 Ti me/m i n 11 0.2 unit L 0 5 1 0 0 5 1 0 Fig. 1. Typical liquid chromatograms obtained from 4-1.11 injections of: (a) ETU standard; (b) maneb technical; (c) zineb technical ; (d) mancozeb technical ; and (e) Vondozeb technical. The analysis of an aqueous solution of nabam and an aqueous suspension of zineb, by injection on to a gas chromatograph, was investigated by Zielinski and Fishbein.12 These experiments gave rise to peaks attributable to ETU rather than peaks directly attributable to the EBDCs.They concluded that this was due to the formation of ETU on sample injection, although ETU already present in the sample could have contributed to the peaks observed. A sample of ETM, which high-performance liquid chromatographic analysis indicated contained no apparent ETU, when dissolved in chloroform - methanol and injected on to the gas chromatograph yielded a significant peak with the same retention time as ETU.It was observed by Marshallll that ETM can give rise to ETU at elevated temperatures and our findings are in agreement with that observation. It is apparent that ETU is formed from one or more precursors, present as degradation products of EBDCk or as impurities, on injection in methanolic solutions on to a gas chromatograph. It is significant that those samples which contained the greatest concentration of co-extractives, based on high- performance liquid chromatographic analyses, gave rise to the largest difference between the gas - liquid and high-performance liquid chromatographic results for ETU content. Samples extracted with chloroform -methanol (1 + 1) gave the same result for ETU content , when assayed by high-performance liquid chromatography, as those obtained following extraction with methanol (Table 11).However, on injection on to the gas- liquid chromatograph, higher results for ETU were observed than those obtained followingFebrimry, 1979 I N ETHYLENEBISDITHIOCARBAMATE FUNGICIDES ‘b) c i 116 9 0 5 10 15 G 5 10 15 0 5 10 1 5 0 5 10 1 5 0 5 10 15 T ime/’m i n Fig. 2. Typical gas - liquid chromatogranis obtained from 4-pl injections of: (a) ETU standard; (b) maneb technical; (G) zineb technical; ( d ) mancozeb technical ; and ( 6 ) Vondozeb technical. comparative methanol extractions. This serves to illustrate further that degradation of ETU precursors is occurring during gas - liquid chromatography, with possible contributions from unresolved extraneous peaks.TABLE I1 ETHYLENETHIOUREA CONTENT OF ETHYLENEBISDITHIOCARBAMATE FUNGICIDES USING CHLOROFORM - METHANOL AS THE EXTRACTION SOLVENT Figures are expressed as percentages. Results by high-performance Fungicide liquid chromatography chromatography Results by gas - liquid Maneb technical 1 .. .. 0.1, 0.1 0.2, 0.2 Maneb technical 2 .. .. 1.2, 1.2 4.3, 4.2 Maneb formulation 2 . . .. 1.1, 1.1 3.3, 3.1 Maneb formulation 3 . . . . 1.1, 1.1 2.9, 2.6 Zineb technical 1 .. .. 1.2, 1.2 1.4, 1.4 Maneb - zineb formulation . . 0.9, 1.0 1.4, 1.4 Mancozeb technical 1 . . . . Vondozeb technical 1 . . . . Vondozeb technical 2 . . .. 0.1, 0.1 0.9 1.2 0.1, 0.1 2.6 2.7116 FARRINGTON AND HOPKINS Conclusion If ETU analyses are carried out by gas - liquid chromatography, results will be obtained that reflect both the sample content of ETU and the ETU that may be formed by thermal decomposition from other compounds present.The use of high-performance liquid chromato- graphy offers a rapid analytical technique that yields a truer estimate of the ETU content. The authors thank Robinson Brothers Ltd. for supplying ETM and ETU. They also thank the Government Chemist for permission to publish this paper. Some of this work was carried out as part of the programme of the Dithiocarbamates Panel of the Pesticides Analysis Advisory Committee. 1. 2. 3. 4. 5. 6. 7. 8. 9. 10. 11. 12. References Bontoyan, W. R., Looker, J. B., Kaiser, T. E., Giang, P., and Olive, B. M., J . A s s . Off. Analyt. Bontoyan, W. R., and Looker, J. B., J . Agric. Fd Chem., 1973, 21, 338. Graham, S. L., Davis, K. J., and Hansen, W. H., Fd Cosmet. Toxicol., 1975, 13, 493. Pure Appl. Chem., 1977, 49, 675. Clarke, D. G., Baum, H., Stanley, E. L., and Hester, W. F., Analyt. Clzem., 1951, 23, 1842. Raw, C. R., Editor, “CIPAC Handbook,” Volume I, Heffer, Cambridge, for Collaborative Inter- Dubosq, F., Chim. Analyt., 1967, 49, 68. Johnson, E. I., and Tyler, J . F. C., Chemy Ind., 1962, 304. Czegledi-Janko, G.. and Hollo, A., J . Chromat., 1967, 31, 89. Fishbein, L., and Fawkes, J., J . Chromat., 1965, 19, 364. Marshall, W. D., J . Agric. Fd Chem., 1977, 25, 357. Zielinski, W. L., and Fishbein, L., J . Chromat., 1966, 23, 302. Chem., 1972, 55, 923. national Pesticides Analytical Council, 1970, p. 463. Iicceived Jzme 21st, 1!378 Accepted September 4th, 1978

ISSN:0003-2654    

DOI:10.1039/AN9790400111

出版商:RSC    

年代:1979

数据来源: RSC

摘要:

Anahst, February, 1979, Vol. 104, pp. 117-123 117 Fluorimetric Determination of Acetohexamide in Plasma and Tablet Formulations Using 1-Methylnicotinamide Pamela Girgis-Takla and loannis Chroneos Welsh School of Pharmacy, University of Wales Imtitztte of Science and Technology, King Edward V I I Avenue, Cardifl, CF1 3NU A sensitive method is described for the fluorimetric determination of aceto- hexamide in plasma or in tablets by means of its reaction with l-methyl- nicotinamide, which is shown t o be a useful reagent for the determination of ketonic compounds. The limit of detection is approximately 0.2 pg ml-l and the relative standard deviation is 3.1% for 2 pg ml-l in plasma. Acetoacetic acid usually does not interfere, but can be separated, if necessary, from acetohexamide by means of a washing technique.No interference is caused by the presence of insulin, other (non-ketonic) oral hypoglycaemic drugs, acetone or pyruvic or a-ketoglutaric acid, Keywords : A cetolaexamide determination ; plasma ; tablets ; 1- fiaetlzylnicotin- amide reagent ; spectrofluorimetry The finding1 that the oral hypoglycaemic compound acetohexamide can exist in at least two polymorphic forms made it of interest to develop a sensitive and relatively specific method for the routine determination of plasma levels of the unchanged drug, especially in view of a recent report2 that marked variations in bioavailability can exist between different batches of tablets containing the related sulphonylurea tolbutamide. Serum concentrations of acetohexamide can be measured colorimetrically by a modification3 of Spingler's pro- ~ e d u r e . ~ Both the drug and its metabolite, hydroxyhexamide, can be determined in serum or plasma by isotope dilution analysis,5 or by a two-component spectrophotometric pro- cedure6 utilising measurements at 247 and 228 nm, a disadvantage of the last method being that blank absorbance values are high.A more sensitive gas - liquid chromatographic method has been proposed by Kleber et aL7 for which the calibration concentrations of the drug and its metabolite range from 5 to 40 pg ml-l in plasma. The method described here is suitable for plasma samples containing 0.5-2.5 pg ml-l of acetohexamide. It introduces a new use for 1-methylnicotinamide as a reagent for the fluorimetric determination of ketones, with which it is known8 to react readily, after treatment with an alkali in order to convert it into the highly reactive a-carbinol.The reaction has previously been applied only to determining 1-methylnicotinamide by condensation in the presence of an alkali with a variety of different methyl ketones. Huff and Perlzweig8 and Carpenter and Kodiceks have described similar procedures for 1-methylnicotinamide, which involve condensation in aqueous alkali, followed by acidification to about pH 0.5 with hydrochloric acid, a short period of heating and finally the addition of potassium dihydrogen orthophosphate in order to buffer the mixture at about pH 2. In another modification, Clark et a1.l0 converted the 1-methylnicotinamide into a fluorescent derivative by treatment with acetophenone in alcoholic potassium hydroxide, and then acidified the solution with 99% formic acid.The acidification in each procedure reversibly changed the fluorescence from greenish blue to blue and enhanced it. The method developed for acetohexamide in plasma can also be applied to the determina- tion of the drug in tablets, and is simpler and less time consuming than the high-performance liquid chromatographic method of Beyer,ll or the Salim and Hilty spectrophotometric assay,12 which is the official method of the United States Pharma~opeia.1~ Other techniques that have been proposed for tablet preparations include non-aqueous titration,l4>l5 colorimetry using either cobalt acetate16 or 2 ,4-dinitrophenylhydrazine17 and polarography.18118 GIRGIS-TAKLA AND CHRONEOS : FLUORIMETRIC DETERMINATION OF Apparatus Fluorimetric measurements were carried out on a Perkin-Elmer, Model 1000, fluorescence spectrophotometer in 1 x 1 cm silica cells with a 371-nm filter at an angle of 20" for excita- tion and an emission wavelength setting of 437 nm, slit width N and scale expansion x 1 or x 2.The infrared spectra were recorded from potassium bromide discs using a Perkin-Elmer, Model 357, grating spectrometer. Spectrophotometric measurements were made using a Unicam SP500 Series 2 spectro- photometer and 1-cm silica cells. Ultrafiltration of plasma solutions of acetohexamide was carried out using Amicon Centri- flo Membrane Cones CF 50A with conical supports and 50-ml centrifuge tubes. Analyst, Vol.104 Experimental Reagents of fluorescent contaminants. All reagents were of analytical-reagent grade, and were checked before use for the presence Acetohexamide. 1-Methylnicotinamide iodide reagent, 3% m/V in lo-* M hydrochloric acid. This solution should be freshly prepared. Dissolve 9 g of nicotinamide in 50 ml of dimethylformamide and add 10ml of methyl iodide. Allow the mixture to stand, preferably overnight, then separate the product by filtration, wash it with about 50 ml of dimethylformamide and dry it in air (melting-point 205-208" C). Formic acid solution, 50% V/V. Hydrochloric acid, 2 M and 10% m/m. Sodium hydroxide solutiozzs, 0.01, 0.1 and 5 M. Chloroform. De-ionised water was used in the preparation of all solutions. Kindly supplied by Eli Lilly and Company Ltd.Prepared from 98-100% m/m formic acid. Preparation of hydroxyhexamide A solution of 1 g of acetohexamide in 60 ml of 2yo m/V sodium hydroxide solution was prepared and 150 mg of sodium tetrahydroborate( 111) were added and dissolved by shaking. The solution was left to stand for 1 h at room temperature and then acidified with 40 ml of 10% m/m hydrochloric acid. The product was separated immediately by filtration, washed with approximately 150 ml of water and recrystallised from dilute ethanol] to give 0.7 g of h ydroxyhexamide ( N - [ (cyclohexylamino) carbonyll-4- (1 -hydroxyethyl) benzenesulphon- amide), melting at 146-149 "C. The compound was identified by thin-layer chromatography on pre-coated silica gel 60 F,,, plates with chloroform - formic acid (97 + 3) as the solvent system6 and by nuclear magnetic resonance and infrared absorption spectrometry.Thin- layer chromatography showed a trace amount of acetohexamide in the product that was difficult to remove by further recrystallisation. The infrared spectrum (Fig. 1) showed the presence of an OH band at 3520 cm-l. The slight shoulder a t 1690 cm-1 provided evidence that a carbonyl group had been reduced when the spectrum was compared with that of Wave len g t h/p m 2.5 3.0 4.0 5.0 6.0 7.0 8.0 9.0 10 12 14 16 20 30 40 100 -7- 80 a- & 60 E 40 4-J + .- c p 20 0 4000 3500 3000 2500 2000 1800 1600 1400 1200 1000 800 600 400 200 W avenumber/cm-' Fig. 1. Infrared spectrum of hydroxyhexamide.February, 1979 ACETOHEXAMIDE IN PLASMA AND TABLETS USING 1-METHYLNICOTINAMIDE 119 acetohexamide polymorph B,l as this shows a clear doublet with peaks a t 1690 and 1665 cm-1.Attempts to prepare hydroxyhexamide on a small scale by catalytic hydro- genation, which has been used by previous workers,lSJO were not successful as the reduction proceeded too far, and the product was N- [ (cyclohexylamino)carbonyl]-4-ethylbenzene- sulphonamide. Procedures Assay of acetohexamide in plasma Mix 2.0 ml of plasma with 1.5 ml of 2 M hydrochloric acid in a glass- stoppered tube. Add 20.0ml of chloroform, shake thoroughly and allow to stand for at least 30 min in order to allow the phases to separate. Transfer the bulk of the chloroform layer into a stoppered centrifuge tube and centrifuge. Measure 5.0ml of the chloroform extract into a 10-ml calibrated flask and evaporate to dryness in a stream of nitrogen at room temperature.(Up to three 5-ml portions of the chloroform extract can be evaporated to dryness, and subjected separately to the fluorimetric procedure to improve the confidence limit of the assay.) Dissolve the residue in the 10-ml flask in 0.10 ml of 0.01 M sodium hydroxide solution, add 0.10 ml of 1-methylnicotinamide reagent, followed immediately by 0.10 ml of 5 M sodium hydroxide solution and mix thoroughly. Exactly 2 min after addition of the 5 M sodium hydroxide solution make up to volume with 50% V/V formic acid. Measure the fluorescence intensity (maximum value) after 1.5-2 h. Correct the observed fluorescence by subtracting the fluorescence intensity measured using the same procedure on a drug-free plasma sample taken from the same subject prior to drug administration.Determine the concentration of acetohexamide in the plasma by reference to a calibration graph obtained by carrying out the assay procedure using 2.0-ml aliquots of standard solutions of acetohexamide in 0.01 M sodium hydroxide solution or in plasma containing 0, 0.5, 1.0, 1.5, 2.0 and 2.5 pg ml-l of drug. [Fluorescence intensity values after subtraction of the appropriate blank (zero drug concentration) are the same, irrespective of whether the drug is dissolved in plasma or in aqueous alkali.] If necessary, dilute the test fluorescent solution with 50% V/V formic acid, in order to bring the fluorescence intensity within the range of the calibration graph. Determine the approximate concentra- tion of acetohexamide in the plasma by reference to the calibration graph.Repeat the assay using a dilution of the plasma in 0.01 M sodium hydroxide solution accurately prepared to contain about 2.0 pg ml-l of acetohexamide. Carry out a plasma blank determination using an equivalent dilution in 0.01 M sodium hydroxide solution of the drug-free plasma sample. Plasma extraction. Fluorimetric procedure. Calculation. Assay of acetohexamide in tablets Accurately weigh an amount of the powder equivalent to 500 mg of acetohexamide into a 100-ml calibrated flask, add 60 ml of 0.1 M sodium hydroxide solution and shake for 30 min. Make up to volume with 0.1 M sodium hydroxide solution, mix and filter, and discard the first 20 ml of filtrate. Dilute 5.0 ml of filtrate to 50.0 ml with water.Transfer 1.0 ml of this solution into a 100-ml calibrated flask, add 0.01 M sodium hydroxide solution to volume and mix. To 1.0 ml of the final dilute solution in a 100-ml calibrated flask, add 1.0 ml of 1-methylnicotinamide reagent followed immediately by 1.0 ml of 5 M sodium hydroxide solution and mix thoroughly. Exactly 2 min after addition of the 5 M sodium hydroxide solution, make up to volume with 50% V/V formic acid. Measure the fluorescence intensity (maximum value) after 1.5-2 h. Determine the concentration of acetohexamide in the final dilute solution by reference to a calibration graph, obtained by carrying out the fluorimetric procedure on 1 .O-ml aliquots of standard solutions of acetohexamide in 0.01 M sodium hydroxide solution containing 0, 2.0, 4.0, 6.0, 8.0 and 10.0 pg ml-l of drug.Calculate the amount of aceto- hexamide in milligrams per tablet using the expression Sample preparation. Weigh and powder 20 tablets. Fluorimetric psocedure. Calculatioiz. myytl x c, x 100 Mass per tabletlmg = m2120 GIRGIS-TAKLA AND CHRONEOS : FLUORIMETRIC DETERMINATION OF Analyst, VoZ. 104 where C, pg ml-1 is the concentration of acetohexamide in the final sample solution and m, and m2 g are the average mass of the tablets and mass of sample taken, respectively. Results and Discussion Factors Affecting the Fluorimetric Procedure The significance of the reagent concentrations and reaction times selected for the recom- mended method was shown by carrying out the fluorimetric procedure described for the assay of acetohexamide in tablets, using a 10 pg ml-l solution of acetohexamide, and varying the concentration of the different reagents individually. Changing the concentration of 1-methylnicotinamide iodide in the reagent solution over a range from 0.5 to 10% m/V showed that a concentration of at least 2-3% was necessary for the highest fluorescence intensity to be obtained [Fig.2(a)], and that when reagent concentrations were reduced to 1 or o.5y0, fluorescence readings decreased to 78 or 52% of the maximum, respectively. When using a 3% reagent concentration, maximum fluorescence develops within 1.5 h and remains stable, decreasing by not more than about 2% of its value over 24 h. When the reagent concentration is increased to 5 or loyo, maximum fluorescence is measured after 1 h, after which time readings decrease gradually because of quenching caused by a yellow colour that slowly develops in the solution after acidification.This quenching effect can be avoided by extracting the aqueous solution with dichloromethane immediately after acidifi- cation and measuring the fluorescence in the organic layer at the usual wavelength settings. I I I I I 2 4 6 8 1 0 : I g 200 E 100 - LL 0 1 2 3 4 5 Concentration of 'I-methylnicotinamide, % m/V Time/min 600 =1 500 2 E cn *.' .- C + 400 m >: z 300 + 0, C + .- . 8 200 2 C g 100 - 0 2 4 6 8 1 0 Sodium hydroxide concentration/N Fig. 2. Influence on fluorescence intensity developed from 10 pg of acetohexamide of : ( a ) , 1-methylnicotinamide iodide reagent concentration ; ( b ) , duration of reaction with 2, 3 or 5 N sodium hydroxide solution; and (c), normality of sodium hydroxide soh tion.February, 1979 ACETOHEXAMIDE IN PLASMA AND TABLETS USING 1-METHYLNICOTINAMIDE 121 By following this procedure it was shown that there is no significant change in fluorescence intensity when the l-methylnicotinamide concentration is increased from 3 to 5 or 10%.Substituting l-methylnicotinamide chloride for the iodide does not result in any enhancement of fluorescence, and the iodide was therefore used throughout the work as it is easier to prepare. The fluorescence intensity is also influenced by the duration of the reaction in alkaline solution, as well as by the concentration of alkali used. Fig. 2(b) shows that the fluorescence intensity reaches a maximum when the reaction in alkali is allowed to proceed for 2 min, but that it is reduced if a longer reaction time is used.This observation is the same irrespec- tive of whether the fluorescence is measured in aqueous solution, or after extraction into dichloromethane. The highest fluorescence readings are obtained when the sodium hydroxide reagent concentration is 5 M or above [Fig. 2(c)]. With the higher alkali concentrations (8 or 10 M), however, the fluorescence is slightly less stable because of the quenching effect already mentioned. The fluorescence intensity is about ten times higher in acidic than in alkaline solution , and the final fluorimetric measurement is therefore made after acidification with formic acid.This acid produced a higher fluorescence than hydrochloric acid, possibly because with the latter a heating step is necessary in order to develop the fluorescence. The optimum concentration of formic acid in the final solution is approximately 50% VlV. With concentrations higher than this a yellow colour again develops and the fluorescence is less stable; with lower concentrations fluorescence develops to the same extent but more slowly. Precision and Accuracy The calibration graph for the assay in plasma was linear, and fluorescence intensity measurements (arbitrary units), after subtraction of the blank value, ranged from 130 for 0.5 pg ml-l to 653 for 2.5 pg ml-l of acetohexamide in human plasma or in 0.01 M sodium hydroxide solution. In order to test the precision of the procedure, eight replicate deter- minations were made on each of two plasma solutions containing 1.0 and 2.0 pg ml-l of acetohexamide.The relative standard deviations of the assay were calculated to be 3.7 and 3.1y0, respectively. The accuracy of the procedure was tested by assaying a sample of plasma to which had been added 40.0 pg ml-l of acetohexamide and comparing the results obtained by using the fluorimetric method and the spectrophotometric procedure of Smith et aL6 The average recovery (see Table I) was 99.2% by the fluorimetric method, while by the spectrophotometric procedure it was 93.0% calculated by means of the Smith equation, or 90.0% calculated directly using an value of 538.9 at 247 nm, which had been deter- mined for the acetohexamide sample used.The lower recovery by the spectrophotometric procedure, which does not apply a correction for the losses in the extraction steps involved, is in accordance with recoveries of 91 and 94% reported by Smith et al. The spectrophoto- metric procedure is not sufficiently sensitive to allow a comparison of the two methods of assay at lower levels of drug concentration in plasma. Defining the detection limit as the amount giving twice the background (blank) fluorescence, the limit calculated for aceto- hexamide in plasma is 0.2 pg ml-l, The calibration graph was also linear for the assay of tablets, and fluorescence intensity measurements (arbitrary units), after subtraction of the blank value, ranged from 113 for 2.0 pg ml-l to 564 for 10.0 pg ml-l of acetohexamide in TABLE I COMPARISON OF THE FLUORIMETRIC AND SPECTROPHOTOMETRIC METHODS FOR THE ASSAY OF PLASMA CONTAINING 40 pg ml-1 OF ADDED ACETOHEXAMIDE Fluorimetric method Spectrophotometric methods Acetohexamide Recovery, Acetohexamide Recovery, found/pg ml-1 % found/pg ml-l Yo 39.6 99.0 37.3 93.3 39.7 99.3 37.6 94.0 39.7 99.3 36.8 92.0 36.1 90.0 37.6 94.0 Mean: 99.2 Mean: 93.0122 GIRGIS-TAKLA AND CHRONEOS : FLUORIMETRIC DETERMINATION OF Analyst, VoZ.104 0.01 M sodium hydroxide solution. In order to test the precision and accuracy of the fluorimetric procedure, results were compared with those obtained by the spectrophotometric method of the United States Pharmacopeia.l3 Five portions, each equivalent to about 500 mg of acetohexamide, were taken from the same powdered sample of tablets, accurately weighed, and made up to 100 ml in 0.1 M sodium hydroxide solution in the usual way.Duplicate aliquots of the filtered solution were assayed by each method. The results are shown in Table 11. TABLE I1 COMPARISON OF THE FLUORIMETRIC AND USP METHODS FOR THE ASSAY OF A COMMERCIAL SAMPLE OF ACETOHEXAMIDE TABLETS Average mass per tablet, 0.653 g; nominal content of acetohexamide, 500 mg. Fluonmetric method USP method r \ r Acetohexamide found Percentage of Acetohexamide found Percentage of Sample mass/g per tabletlmg stated amount per tabletlmg stated amount A A > 0.669 0 500 508 0.650 9 493 495 0.6444 509 503 0.648 1 505 611 0.666 5 614 506 Mean .. .. 504 Relative standard deviation .. . . 1.3% 100.0 101.6 98.6 99.0 101.8 100.6 101.0 102.2 102.8 101.2 500 509 503 512 605 504 504 503 501 502 100.0 101.8 100.6 102.4 101.0 100.8 100.8 100.6 100.2 100.4 100.9 f 1.1 504 100.9 f 0.5 0.7% Specificity The procedure can be used for the determination of acetohexamide in the presence of insulin and other commercially available oral hypoglycaemic drugs, as none of these contains a ketone grouping.The fluorimetric reaction is also not shown by lactic or fl-hydroxy- butyric acids, or by hydroxyhexamide, which is the main route of metabolism for aceto- hexamide in man.21 Acetone and acetoacetic, pyruvic and a-ketoglutaric acids produce fluorescent derivatives in the reaction with 1-methylnicotinamide. When, however, solutions of these compounds, in the concentrations shown in Table 111, were tested by the assay procedure for aceto- hexamide in plasma, only acetoacetic acid was extracted with chloroform and remained behind after evaporation of the solvent in an amount sufficient to make it liable to interfere TABLE I11 FLUORESCENCE MEASUREMENTS MADE BY SUBJECTING SOLUTIONS OF POSSiBLE INTERFERING COMPOUNDS TO THE ASSAY PROCEDURE FOR ACETOHEXAMIDE I N PLASMA Compound Acetoacetic acid .... Acetone . . .. .. .. Hydroxyhexamide . . . . Lactic acid , . .. .. fl-Hydroxybutyric acid .. a-Ketoglutaric acid . . .. Pyruvic acid . . .. .. Concentration in solution/ mg ml-l 5.2 8 6 0.1 0.1 0.36 0.01 Fluorescence intensity, * arbitrary units 474 0 0 0 0 0 10 * Blank fluorescence subtracted.Febt“ZdUYy, 1979 ACETOHEXAMIDE I N PLASMA AND TABLETS USING 1-METHYLNICOTINAMIDE 123 with the assay.The acetoacetic acid can be separated from the acetohexamide by washing an aliquot of the chloroform extract twice, each time with an equal volume of distilled water. Repeated assays using this washing technique showed that, on average, 93% of the chloroform-extracted acetohexamide will remain in the chloroform layer, the remainder passing into the aqueous washings. Experience has shown that interference, if any, from acetoacetic acid in the assay procedure for plasma is likely to be slight. Plasma samples were taken from ten diabetic patients who were not receiving any acetohexamide, but were being controlled either by diet alone, or by treatment with insulin, chlorpropamide or metformin. The average plasma blank reading (arbitrary units) was 27 (range 23-30) by the normal assay procedure, 23 (range 20-25) when the chloroform was washed twice with water and 18 (range 16-19) when aliquots of the chloroform extracts were washed once with an equal volume of 0.5 M sodium carbonate solution before evaporation.The mean blank fluorescence (reagent blank) when the procedure was carried out using 0.01 M sodium hydroxide solution in place of plasma was also 18 (range 17-19). Blanks obtained using plasma samples from non-diabetic patients were in the same ranges. Binding of Acetohexamide to Plasma Attempts to separate acetohexamide from plasma proteins by ultrafiltration showed that the drug is largely bound by these proteins. Solutions of the drug in human plasma con- taining 10 and 100 pg ml-l of acetohexamide were assayed by the fluorimetric procedure.Samples (about 5ml) from each solution were also subjected to ultrafiltration by centri- fuging for about 30 min in Centriflo membrane cones. The ultrafiltrate collected from each solution was then subjected to fluorimetric assay. It was found that the binding of aceto- hexamide was 96.6y0 for the 10 pg ml-l and 95.7% for the 100 pg ml-l solution. These figures support the measurements of binding of the same drug to various human proteins made by JudisZ2 by means of equilibrium dialysis, which showed that acetohexamide can be bound S8yo to albumin, and to a lesser extent to fibrinogen I and a-globulin IV-4. We are grateful to Dr. T. M. Hayes of the Welsh National School of Medicine, Cardiff, and to his colleagues who kindly helped us to obtain plasma samples from diabetic subjects.1. 2. 3. 4. 5. 6. 7. 8. 9. 10. 11. 12. 13. 14. 15. 16. 17. 18. 19. 20. 21. 2 2 . References Girgis-Takla, P., and Chroneos, I., J. Phavm. Pharmac., 1977, 29, 640. Simmons, D. L., Legore, ,4. A., Picotte, P., Chbnier, &I., and Jasmin, K., Can. J . Pharm. Sci., 1977, Maha, G. E., Kirtley, W. R., Root, M. A., and Anderson, R. C., Diabetes, 1962, 11, 83. Spingler, H., Klin. Wschr., 1957, 35, 533. Galloway, J. A., McMahon, R. E., Culp, H. W., Marshall, F. J., and Young, E. C., Diabetes, 1967, Smith, D. L., Vecchio, T. J., and Forist, A. A., Metabolism, 1965, 14, 229. Kleber, J. W., Galloway, J . A., and Rodda, B. E., J . Pharm. Sci., 1977, 66, 635. Huff, J . W., and Perlzweig, W. A., J , Biol. Chew., 1947, 167, 157. Carpenter, K. J., and Kodicek, E., Biochem. J . , 1950, 46, 421. Clark, B. R., Halpern, R. M., and Smith, R. A., Analyt. Biochem., 1975, 68, 54. Beyer, W. F., Analyt. Chem., 1972, 44, 1312. Salim, E. F., and Hilty, W. W., J . Pharm. Sci., 1967, 56, 385. “United States Pharmacopeia,” Nineteenth Revision, United States Pharmacopeial Convention Baltazar, J., and Ferreira Braga, M. M., Revta Port. Farm., 1966, 16, 169. Agarwal, S. P., and Walash, M. I., Indian J . Pharm., 1972, 34, 109. Meier, G. N., Kuhn, 0. S., Pierart, F. O., and Cortes, S. J. S., Revta R. Acad. Cienc. Exact. Fis. Nut. .4mer, M. M., and Walash, M. I., Bull. Fac. PharPn. Cairo Univ., 1973, 12, 399. Tammilehto, S., Favnz. Aikak., 1973, 82, 39. Eli Lilly and Co., BY. Pat. 912789, 1962. Marshall, F. J., Sigal, M. V., Jr., Sullivan, H. R., Cesnik, C . , and Root, 34. A., J . Med. Chem., 1963, Welles, J . S., Root, M. A., and Anderson, R. C., Proc. SOC. Exp. Bid. Med., 1961, 107, 583. Judis, J . , J. Pharm. Sci., 1972, 61, 89. 12, 85. 16, 118. Inc., Rockville, Md., 1975, p. 15. Madr., 1971, 65, 653. 6, 60. Received July loth, 1978 Accepted August 15th, 1978

ISSN:0003-2654    

DOI:10.1039/AN9790400117

出版商:RSC    

年代:1979

数据来源: RSC

摘要:

124 Analyst, February, 1979, Vol. 104, $$. 124-135 Polarographic Determination of Trace Elements in Food from a Single Digest M. Kapel and M. E. Komaitis" Procter Department of Food Science, University of Leeds, Leeds, LS2 9JT The determination of 12 trace elements, namely copper, zinc, mercury, lead, cadmium, iron, tin, chromium, arsenic, antimony, selenium and tellurium, in admixture by means of a cathode-ray polarograph is described. The elements were investigated in concentrations ranging from 0.1 to 20 p.p.m. by means of normal, reverse-sweep or resistance - capacitance derivative techniques. The last technique could not be used for mercury, although it was used for all the other elements in concentrations less than 1 p.p.m. The complete determination took 4-7 h, and was applied to various kinds of food, such as bread, meat and vegetables. Keywords : Trace element determination; food analysis ; 9olarography ; single digest The determination of trace elements in food is of great importance, as some of them have nutritional significance, whilst others are toxic.The aim of this work was to establish a method that is accurate at an adequate level, is not very expensive, is rapid in comparison with other available techniques and covers a wide area of applicability. Solvent extraction has been used for a long time for the separation of trace elements. The technique is not characterised by specificity, a fact that can be turned to advantage when the method is combined with polarography. In the method recommended here, the sample is first digested with sulphuric and nitric acids and the pH then adjusted to a suitable value.Mercury(II), copper(II), zinc(II), cadmium(I1) and lead( 11) are removed in the form of dithizone (diphenylthiocarbazone) complexes.1 Concentrated hydrochloric acid is added to the aqueous phase, after which iron(II1) tin(1V) and antimony(III), if present, can be extracted in the form of salts of cupferron (ammonium N-nitrosophenylhydroxylamine) .2 The pH of the resulting aqueous phase is adjusted with ammonia solution, after which chromium(V1) is extracted with sodium diethyl- dithiocarbamate in chl~roform.~ The aqueous phase remaining is digested with sulphuric and nitric acids and, after complete destruction of the organic material, tellurium(IV), selenium(IV), antimony(V) and arsenic(II1) are determined.A flow diagram of the pro- cedure is shown in Fig. 1. Apparatus manufactured by Southern Analytical Limited. Corning Scientific Instruments. Reagents Experimental The work was carried out with an A 1660 Davis differential cathode-ray polarograph The pH was measured with a Digital 110 Expanded Scale pH meter manufactured by Nitric acid, concentrated relative density 1.42 (AnalaR) . Sulphuric acid, concentrated, relative density 1.84 (AnalaR) . Hydrochloric acid, concentrated, relative density 1.18 (AnalaR). Potassium hydroxide solution, 3 M. Ammonia solution, 6 M. Hydrochloric acid, 4 M. Sulphuric acid, 2 M. Dithizone solution. Cupfewon solzdion, 10% m/V. Prepared by solution of 50 mg of dithizone in 1 1 of chloroform. Prepared by solution of l o g of cupferron in 1 O O m l of distilled water.Greece. * Present address : Department of Food Chemistry, University of Athens, Navarinou 13, Athens( 144),KAPEL AND KOMAITlS 125 Dithizone in CHC13 pH 5.0-5.5 Sodium diethyldithiocarbamate solution, 5% m/V. diethyldithiocarbamate in 100 ml of distilled water. Ammorcium citrate solution, 20% m/V. Potassium iodide solution, 2% mlV. Potassiwn disulphite solution, 5% m/ V . Potassium chloride solution, 1 M. Hydrogefi peroxide, 30%. Ethyl acetate. Prepared by solution of 5 g of sodium r Zn(ll) Digest Sam p I e 10-159 - Dithizone in CHCl3 pH 8.5 Dithizone in CHCI:, pH 1.5-2.0 I Pb(ll), Cd(ll) I I CHC13 Aqueous solution A and ethyl acetate 1 Fe(lll), Sn(lV), Sb(lll) HCI or H2S04 + cupferron I Sodium diethyldithiocarbamate and CHC13 4 PH 4-6 Cr(VI) H Acid digestion Te(lV), Se(lV) KI + Kfi205 Fig. 1.Flow diagram of the procedure. Procedure The sample (10-15 g) was weighed accurately and transferred into a 600-ml Kjeldahl flask. Concentrated sulphuric acid (20-25 ml) and concentrated nitric acid (20-25 ml) were added and the mixture was heated in a fume cupboard (see Note 1). When the liquid turned brown, small amounts of concentrated nitric acid were added dropwise until complete decolorisation occurred. After decolorisation, heating was continued until the fumes of nitric acid had disappeared. This usually happened in 10-20 min according to the mass of sample. When the flask had cooled, the solution was transferred into a beaker with 40-50ml of water and 6 M ammonia solution was added, with stirring, until the pH value became 1.5-2.0.The solution was cooled and transferred into a 500-ml separating funnel, after which extrac- tion with 50-ml portions of dithizone in chloroform (extract 1) was used for the removal of126 KAPEL AND KOMAITIS : POLAROGRAPHIC DETERMINATION OF Analyst, Vol. 104 mercury(I1) and copper(I1). The extraction was continued until the organic layer did not change in colour. To the aqueous phase, 1 ml of 20% m/V ammonium citrate solution was added (see Note 2), the pH then being adjusted to between 5 and 5.5 with 6 M ammonia solution. The solution was transferred into a separating funnel and zinc(I1) was extracted with 50-ml portions of the solution of dithizone in chloroform (extract 2).Once again, the colour of the organic layer was used as the criterion for the completeness of the extraction. Subsequently, 1 ml of 20% m/V ammonium citrate solution was added and the pH was adjusted to 8.5 with 6 M ammonia solution. Lead(I1) and cadmium(I1) were then similarly extracted with dithizone in chloroform (extract 3). The chloroforni extracts, 1 and 2, were mixed and, after evaporation of the chloroform, digested with a concentrated nitric acid - sulphuric acid mixture (1 + 1). After cooling, dilution and adjustment to pH 5.0 with 6 M ammonia solution, mercury(II), copper(I1) and zinc(I1) were determined polarographically in 1 or 2 M ammonium sulphate solution. Alternatively, the digestion could be carried out with concentrated sulphuric acid and 30% hydrogen peroxide.For the determination of cadmium(I1) and lead(II), extract 3 was evaporated and the residue was digested with the nitric acid - sulphuric acid mixture (1 + 1) as above, after which the solution was diluted, and 1 M potassium chloride solution and 6 M ammonia solution were added until a solution 0.1 M in potassium chloride, 0.1 M in sulphuric acid and 0.8 M in ammonium sulphate was obtained with a pH of 1.6. This solution was then examined polarographically according to the details in the manufacturer’s manual. Once again, concentrated sulphuric acid and 30% hydrogen peroxide could be used as a digestion mixture. The chloroform extracts 1 , 2 and 3 were mixed and, after evaporation of the chloroform, digested with the concentrated nitric acid - sulphuric acid mixture (1 + 1).After cooling, dilution and adjustment to pH 5.0 with 6 M ammonia solution, mercury(II), copper(II), zinc(I1) and cadmium(I1) were determined polarographically in 1 or 2 M ammonium sulphate solution. For the determination of lead(I1) in a portion of the above solution, 1 M potassium chloride solution was added together with enough sulphuric acid to give a solution 0.1 M in potassium chloride, 0.1 M in sulphuric acid and 0.8 M in ammonium sulphate with a pH of 1.6. This solution was then examined polarographically for lead( 11). After the extraction of dithizonates, the aqueous phase was treated with either concentrated hydrochloric acid or concentrated sulphuric acid, so that the latter constituted 10% by volume of the resulting mixture.An excess of 10% w/V aqueous cupferron solution was added, after which iron(III), tin(1V) and antimony(II1) were extracted with two 50-ml portions of chloroform followed by similar portions of ethyl acetate (see Note 3). From the combined organic layers, the solvents were evaporated; then 25 ml of con- centrated nitric acid and 30ml of 30% hydrogen peroxide were added. The mixture was boiled until only a small volume remained. If a brown colour developed, a few drops of 30% hydrogen peroxide were added cautiously. The digestion was continued until the mixture became colourless. After the destruction of the organic material, 50 ml of 4 M hydrochloric acid were added and the solution was partly neutralised with 33.3 ml of 3 M potassium hydroxide solution.The mixture was transferred into a 100-ml calibrated flask and diluted to the mark with water, so that it became 1 M in potassium chloride and 1 M in hydrochloric acid. This solution was examined polarographically. The acidic aqueous phase remaining after the extraction of the cupferronates was adjusted with 6 M ammonia solution to a pH of 4-6, and an excess of 5% m/V aqueous sodium diethyl- dithiocarbamate solution was added. The mixture was shaken and chromium(V1) was extracted with several 50-ml portions of chloroform until the aqueous layer was clear and colourless. The organic layer was digested with concentrated nitric acid and 30% hydrogen peroxide in the manner described above. Twenty-five millilitres of 4~ hydrochloric acid were added, and the mixture was neutralised with 3~ potassium hydroxide solution.It was then transferred into a 100-ml calibrated flask and diluted to the mark with water, so that it became 1 M with respect to potassium chloride. This solution was investigated polaro- graphically, the peak potential, E,, being -1.25 V. Hence, all metals forming dithizonates were removed from the solution. Alternatively, the determination could be carried out as follows.Februayy, 1979 127 Alternatively, a mixture of concentrated nitric and sulphuric acids in equal proportions by volume could be used. In this instance, the neutralisation was carried out with 6~ ammonia solution and the solution was diluted until it became 1-2 M with respect to ammonium sulphate. The aqueous phase remaining from the extraction of chromium was evaporated to a small volume and then digested with a concentrated nitric acid - sulphuric acid mixture (1 + 1).After this digestion, a convenient amount (10-50 m1) of hydrochloric acid (1 + 1) was added and the solution was split into two portions of equal volume. The first of these portions was evaporated until fumes appeared, cooled, diluted and neutralised with 6~ ammonia solution to pH 8.5. It was then suitably diluted to yield a solution 1 M with respect to ammonium sulphate. Tellurium(1V) and selenium(1V) were determined polarographically in this mixture. To the second portion, 2% m/V potassium iodide and 5% m/V potassium disulphite solutions were added in excess and the mixture was heated until white fumes appeared. The flask was cooled and the solution, suitably diluted to a known volume, was divided into two equal parts. To the first part, distilled water was added until the solution was 1 M in sulphuric acid.The second part was diluted and neutralised with 6~ ammonia solution to pH 8.5. After suitable dilution, antimony (111), originally present as antimony(V) , was determined polarographically in a solution 1 M in ammonia and 1 M in ammonium sulphate. TRACE ELEMENTS I N FOOD FROM A SINGLE DIGEST Arsenic( 111) was then determined polarographically. NOTES- solution is recommended. of some metals. tin(1V) a t high pH. ammonium citrate solution. not be achieved with chloroform alone. 1. 2. In the presence of chromium(III), the addition of 2 ml of 3% m/V ammonium peroxodisulphate The addition of 20% m/V ammonium citrate solution was necessary to prevent the precipitation Ammonium oxalate solution was also tried, but was not effective in the presence of It was found that the extraction of tin(1V) was unaffected by the presence of The use of ethyl acetate for the extraction of tin(1V) was necessary, as a complete recovery could 3.Results A sample solution containing all the elements under consideration was digested according to the method described above. The solution was diluted to 1 1, and the method was applied to 10 ml of the diluted solution. The determination was carried out twice, and the results are shown in Table I. TABLE I DETERMINATION OF ELEMENTS IN ADMIXTURE Peak height, scale divisions Concentra- Peak r h \ Metal ion tion, p.p.m. Supporting electrolyte potential/V Found Theoretical Recovery, % Cu(I1) Hg(W Zn(I1) Pb(I1) Cd(I1) Fe(II1) Sn(1V) Cr(V1) As(1II) Se(1V) Sb(II1) Te( IV) ..7.21 . . 16.69 . . 15.82 , . 10.26 . . 11.37 . . 11.20 . . 9.47 . . 15.64 . . 9.23 .. 17.44 .. 10.00 . . 13.07 0.8 M 0.8 M - 0.23 -0.05 -1.30 - 0.52 - 0.73 -1.05 -0.60 - 1.85 -0.75* -1.50 - 0.617 1.10, 1.00 1.10, 1.05 2.60, 2.60 1.10.1.10 1.65: 1.65 1.16, 1.00 0.55, 0.52 2.50, 2.50 14.20, 14.00 0.60, 0.60 2.20; 2.00 3.75, 3.80 1 1.10 1.16 2.63 1.80 1.70 ~. . 1.07 0.50 2.50 15.00 0.60 2.20 4.00 100.00, 90.91 95.65, 91.30 98.86, 98.86 61.10, 61.10 97.06, 97.06 10841 9346 110’00’ 1ok 00 100:oo: 100:00 100.00, 100.00 100.00, 90.91 93.75, 95.00 94.67, 93.33 * Resistance - capacitance derivative circuit used. t Tellurium(1V) was determined by means of the peak at -1.04 V.In the determination of antimony(II1) a correction was made for the ovedapping tellurium peak. From this table, it can be seen that all the elements can be determined accurately with recoveries ranging from 90.9 to lOOyo, except for lead(II), where the recovery was 55.565!40. This discrepancy was due to the fact that some lead(I1) was lost by precipitation in the form of sulphate. Lead must be present at a concentration less than 7 p.p.m. in order to prevent this loss.TABLE I1 DETERMINATION OF THE ELEMENTS IN BEEF SAUSAGE WITH PORK Concentration, Concentration, Concentration, Concentration, Concentration, p.p.m.* p.p.ni.* p.p.m.* p.p.m.* p.p.m.* ion Cr(V1) Fe (111) Pb(I1) Cd(I1) Sn(IV) HdII) Cu(I1) Zn(I1) As (I1 I) Sb(II1) WIV) SelIV) Ft 20.93 13.40 14.98 9.08 13.54s 14.08 21.50 8.99 19.74 12.39: 12.89: 15.28 25.28t Et 20.15 12.20 14.42 13.22 14.64 21.50 9.28 20.38 11.88 12.89 16.84 22.47 Ft 12.13: 6.78 8.041 7.58 5.668 8.47: 7.97: 5.37: 14.20 5.86s 11.49: 11.35 6.32: 6.09: 8.28 8.85$7 9.49$ 12.99 Et 11.65 7.05 8.34 7.64 8.47 12.43 5.37 11.79 6.87 7.45 9.74 11.99 .r 12.35: * The concentrations are expressed in p.p.m. of sample mass. t F = found; E = expected. : Resistance - capacitance derivative circuit was used. 5 Reverse voltage was used. T[ Calculated a t different values of E, (peak potential). Ft 16.96: 18.96 10.30 13.58: 8.60: 12.875 12.06 17.80 8.15s 8.55: 17.25: 17.47; 9.81:. 11.28: 13.62; 20.32; Et 17.70 10.72 12.67 11.61 12.87 18.88 8.15 17.91 10.44 11.32 14.79 19.74 Ft 3.65: 3.55 4.35: 4.03: 4.10: 5.95 2.90; 6.15; 3.42; 4.151 4.75: 6.40; 6.10 3.70 4.35 4.00 4.35 6.50 2.80 6.15 3.60 3.80 5.10 6.80 5.80 3.50 4.15: 4.065: 4.50: 6.25 3.00: 6.65: 3.52$ 4.101 4.853: 6.45: Et 6.25 3.80 4.40 4.10 4.60 6.65 2.90 6.33 3.70 4.00 5.12 6.95 Ft 1.09: 0.57 0.83: 0.67: 0.74: 1.03 0.54: 1.16: 0.72: 0.59: 0.90: 1.22: Et 1.02 0.62 0.73 0.67 0.74 1.09 0.47 1.03 0.65 0.65 0.86 1.14 Concentration, Concentration, Concentration, p.p.m.* p.p.m.* p.p.m.* -7 & Ft 1.04: 0.70; 0.77: 0.76: 0.881 1.47 0.56: 1.37 0.63: 0.73; 1.07; 1.24: Et 1.21 0.73 0.86 0.79 0.88 1.29 0.56 1.22 0.71 0.77 1.01 1.34 Ft 1.11: 0.80: 0.68: 0.76; 0.87: 1.46 0.52: 1.29: 0.63: 0.70; 1.10: 1.26: Et 1.20 0.72 0.86 0.78 0.87 1.28 0.55 1.22 0.71 0.77 1.00 1.34TABLE I11 DETERMINATION OF THE ELEMENTS IN BEEF Metal ion Cr(V1) Sn(1V) Fe(II1) Pb(I1) Cd(I1) Hg(I1) Cu(I1) Zn(I1) As( 111) Sb (I I I) Te(1V) Se(1V) Concentration, p.p.m.* - Ft Et 12.64 11.96 11.061 6.54 7.24 7.37 8.11 5.81: 7.84 8.69: 8.69 7.905 10.93 12.76 5.85; 5.51 5.61s 12.55; 12.10 12.32 7.2817 7.05 6.651 8.50 7.65 10.65 9.94 13.88 13.33 8.88: Concentration, p.p.m.* * Ft Et 13.57jV 15.27 13.261 9.77 9.25 10.541 10.93 11.13 10.09g 12.68:: 15.13 7.36s 7.481 14.00: 14.75 9.01:v 8.861 11.63 15.19 20.43: 15.76 10.02 11.10 16.29 7.04 15.45 9.01 9.77 12.76 17.03 Concentration, p.p.m.* * Ft Et 10.0517 10.49 9.83 5.80 6.35 6.23 7.51 8.23; 4.59: 6.88 6.93s 7.62 7.84: 12.42 11.19 4.83s 4.83 Footnotes as in Table 11.Concentration, p.p.m.* - I Ft Et 5.55: 5.90 3.40 3.60 4.051 4.20 3.851 3.85 4.20: 4.20 5.95 6.30 2.60: 2.70 Concentration, p.p.m.* - Ft Et 6.10: 6.50 3.80 3.95 4.65: 4.675 4.25: 4.30 4.702 4.75 6.75 6.95 2.90: 3.00 Concentration, p.p.m.* -7 I.'t Et 1.68: 1.57 1.78 0.88 0.95 1.19: 1.12 0.84 1.03 1.231 1.14 1.20 2.14 1.67 0.72: 0.72 5.11 10.20 10.61 5.951 5.95 6.90: 6.60 1.49: 1.59 9.61; 6.70 6.19 3.10: 3.45 3.65: 3.85 1.02: 0.93 6.68: 8.04 6.71 3.50: 3.75 3.93: 4.17 1.091 1.00 9.97 8.77 6.20: 6.55 6.90; 7.30 1.471 1.31 13.19 11.69 4.75: 4.90 5.40: 5.45 1.63: 1.76 12.12; Concentration, p.p.m.* * Ft Et 0.97: 1.09 0.60 0.65 0.73: 0.78 0.68: 0.71 0.77: 0.79 1.09 1.16 0.50: 0.50 1.14: 1.10 0.601 0.64 0.64: 0.69 0.83: 0.91 1.21: 1.21 Concentration, p.p.m.* r--J--7 Ft Et 1.082 1.12 0.64 0.68 0.77: 0.80 0.701 0.73 0.81: 0.81 1.08 1.19 0.49: 0.52 1.18: 1.13 0.66: 0.66 0.64: 0.72 0.86; 0.94 1.17: 1.25w 0 TABLE I V DETERMINATION OF THE ELEMENTS IN PORK LUNCHEON MEAT Metal ion Cr(V1) Sn(1V) Fe(II1) Pb(I1) Cd(I1) Hg (11) Cu(I1) Zn(I1) As ( 111) Sb ( I I I) Te (IV) Se(1V) Concentration, p.p.m.* * Ft Et 12.02 11.54 12.02; 6.72 6.99 5.85: 8.26 5.60: 7.54 7.70: 8.39 8.399 11.25 12.31 5.06: 5.32 19.11 11.68 12.55: 6.48'; 6.81 7.139 7.38 9.12:fi 9.65 8.69: 13.42 12.87 12.42: Concentration, p.p.m.* - Ft Et 12.30 12.73 12.30: 7.41 7.71 9.76: 9.11 9.11 6.19 8.35 9.25 9.25 9.253 12.80 13.58 5.54 5.86 5.863 13.58 12.88 13.63: 7.26; 7.51 7.73: 8.14 9.80 10.64 15.96 14.19 14.94: t Concentration, p.p.m.* F r Ft Et 12.998 13.65 7.95 8.27 8.88 9.77 9.63 : 6.63 8.95 9.03: 13.32 14.57 6.86 6.29 6.29: 14.67 13.81 14.32: 7.89: 8.05 7.94: 8.73 10.38: 11.42 11.421 16.90 15.22 17.90: 9.939 9.93 Footnotes as in Table 11.Concentration, p.p.m.* - Ft Et 5.85; 5.45 3.15 3.30 3.80$ 3.90 3.45: 3.60 3.95; 3.95 6.00 5.85 2.455 2.50 5.953 5.55 3.20: 3.20 3.30; 3.50 4.30; 4.60 6.25: 6.10 I Concentration, p.p.m.* ----h--7 Ft Et 5.45$ 5.60 3.10 3.32 3.85: 4.00 3.50: 3.65 4.00: 4.05 6.10 5.95 2.55: 2.55 5.93: 5.75 3.15: 3.30 3.38; 3.60 4.25: 4.60 6.20; 6.20 f Concentration, p.p.m.* - Ft Et 1.01: 1.09 0.63 0.66 0.651 0.78 0.72; 0.72 0.77: 0.79 1.08 1.16 0.54: 0.50 1.17: 1.10 0.60: 0.64 0.76: 0.70 0.91: 0.91 1.31: 1.22 Concentration, p.p.m.* b r Ft Et 0.91: 0.98 0.54 0.59 0.65: 0.70 0.61: 0.64 0.71; 0.71 0.96 1.05 0.42: 0.45 0.99: 0.99 0.521 0.58 0.59: 0.63 0.96: 0.82 1.23: 1.09 * Z U Concentration, t, 1;; z p.p.m.* 0, Ft Et 0.87: 0.94 N 0.56 0.57 8 0.65: 0.67 W 0.61: 0.61 % 0.68: 0.68 U 0.94 1.00 0.43; 0.43 kl 0.93; 0.95 Z 0.65: 0.56 * 0.57; 0.60 8 0.86; 0.78 Z 0 1.08: 1.04 bTABLE V c3 w * 0 M Concentration, Concentration, Concentration, Concentration, Concentration, Concentration, Concentration, Concentration, M r p.p.m.* p.p.m.* p.p.m.* p.p.m.* p.p.m. * p.p.m.* p.p.m.* p.p.m.* M DETERMINATION OF THE ELEMENTS I N BREAD Footnotes as in Table 11.Metal - - & & & & & & Et Ft Et Ff Et Ft Et Ft Et F1. Et Ft Et ion Ft Cr(V1) 14.67$T[ 14.67: Sn(1V) 7.75 Fe(II1) 10.06: Pb(I1) 5.24 Cd(I1) 10.67: Hg(I1) 15.78 6.46 Zn(I1) 12.50 13.183 As(II1) 7.34: Sb(II1) 9.20 8.07 Te(1V) 12.811 10.40 Se(1V) 14.38 cqIq 5.86: .. Et Ft 13.57 13.74 13.74: 8.22 9.77 9.72 11.10: 8.90 7.18: 9.87 10.835 10.03 14.48 18.27 6.10 16.24 9.72 6.25 7.203 13.73 14.405 8.01 8.25: 8.68 10.46; 11.35 14.32 11.27; 15.14 19.65: 14.99 9.07 10.73 9.83 10.89 15.99 6.90 15.16 8.84 9.59 12.53 16.71 13.57 15.42'; 8.11 9.64 11.61: 7.20: 10.77: 18.40 7.67: 6.945 15.55: 15.26 9.02: 8.25: 9.795 10.53 13.26 17.20 14.81 8.97 10.60 9.72 10.77 15.80 6.82 14.99 8.74 9.48 12.38 16.52 7.35: 5.25 5.90: 5.10; 5.80: 7.60 3.40: 3.495 7.55s 7.45: 4.50: 5.30: 6.20: 9.45: 7.80 7.30: 4.70 5.10 5.55 5.95: 5.10 4.95: 5.65 5.65: 8.30 7.90 3.60 3.40 7.85 7.95: 4.60 4.55: 4.98 5.12$ 6.50 5.95f 8.70 10.00: 7.85 4.75 5.60 5.15 5.70 8.40 3.60 7.95 4.65 5.00 6.55 9.25 1.02: 0.56 0.629 0.61: 0.68: 0.69: 1.16 0.41 : 0.90: 0.53: 0.62: 0.90: 1.06: 0.95 0.58 0.68 0.63 0.69 1.02 0.44 0.96 0.56 0.61 0.80 1.06 1.16: 0.79 0.86: 0.84: 0.922 1.36 0.7 1 : 1.28: 0.79: 0.80: 1.12; 1.68: 1.35 0.82 0.96 0.88 0.98 1.44 0.62 1.36 0.79 0.86 1.12 1.50 1.191 0.776 0.92: 0.87: 0.97: 1.40 0.641 1.41: 0.76; 0.82: 1.06: 1.55: 1.39 2 0.84 z 0 0.91 u Y 0.99 2 1.01 3 1.48 0.64 + 1.41 $ s 0.82 pj U 0.89 ij 1.16 5 1.55TABLE VI DETERMINATION OF THE ELEMENTS IN MIXED VEGETABLES Metal ion Cr(V1) Sn(1V) Fe (111) Pb(I1) Cd(I1) Hg(I1) Cu(I1) Zn(I1) As(II1) Sb (I I I) Te (IV) Se(1V) Concentration, p.p.m.* * Ft Et 7.11 7.11 6.807 6.80: 3.91 4.30 4.855 5.09 4.63: 3.58: 4.66 5.17: 5.17 4.765 4.71 8.55 7.58 2.975 3.27 3.84 7.525 7.19 6.361 4.11: 4.19 5.15 4.55 5.54 5.94 8.325 7.92 7.60: Footnotes as in Table 11.Concentration, Concentration, Concentration, Concentration, Concentration, p.p.m.* p.p.m.* p.p.m.* p.p.m.* p.p.m.* - - - - r - - - - - l Ft Et Ft Et Ft Et Ft Et Ft Et 19.73: 21.14 22.39 20.15 4.20: 4.50 4.25: 4.40 0.75$ 0.88 19.37:7 18.42: 12.30 14.40s 13.75: 9.70: 14.15 13.965 15.36: 20.94 9.83': 8.855 20.23 12.79 11.09 12.20 2.55 2.75 2.55 2.65 0.55 0.53 15.13 13.11: 14.42 3.05: 3.25 2.96: 3.15 0.59; 0.63 13.86 9.25: 13.22 2.85$ 2.95 5.75: 5.77 0.53: 0.58 15.36 15.42 14.65 3.30: 3.30 3.15: 3.20 0.64: 0.64 22.55 24.26 21.50 4.95 4.83 4.85 4.70 0.94 0.94 9.74 8.62 9.28 1.90: 2.10 1.80: 2.00 0.44: 0.41 21.38 18.54 20.39 4.851 4.60 4.45: 4.45 0.83: 0.89 13.36: 13.301 12.47 11.89 11.89 2.40; 2.60 2.45: 2.60 0.52: 0.52 14.02 13.52 12.89 12.89 2.60: 2.90 2.651 2.80 0.53: 0.56 19.63 17.67 16.84: 16.84 3.80: 3.80 3.55; 3.65 0.92$ 0.98 21.63; 23.67 22.6317 23.57 23.97: 22.47 4.75: 5.05 4.65: 4.90 0.79: 0.74 Concentration, p.p.m.* - 0.69: 0.77 Ft Et 0.43 0.46 0.53: 0.55 0.48: 0.50 0.60: 0.57 0.75 0.82 0.33: 0.35 0.77: 0.77 0.47; 0.45 0.45: 0.49 0.59: 0.68 0.75 0.85 * Z U Concentration, !$ Ft Et ..p.p.m.* k (------A---72 0.66: 0.71 5: w 0.41 0.43 0 0.49: 0.51 8 0.45: 0.47 ' 0.54: 0.52 $ U 0.69 0.76 3 0.31: 0.33 pj 0.75: 0.72 z 0.44: 0.42 5 0.43: 0.49 z 0.75: 0.80 V 5 5 0.56: 0.60 0TABLE VII DETERMINATION OF THE ELEMENTS IN JELLY Footnotes as in Table 11. Concentration, p.p.m.* Metal ( - A - , ion Cr(V1) Sn(1V) Fe (I1 I) Pb(I1) Cd(I1) 2(Y# Zn(I1) As (I 11) Sb(II1) Te(1V) Se(1V) Ft Et 7.987 8.78 4.81 5.32 6.29: 6.29 4.43: 5.76 4.03 6.19: 6.38 5,805 7.36 9.37 3.69 4.05 8.545 8.88 9.43 5.54 5.18 6.31 5.62 6.857 7.34 7.01 9.367 9.79 9.41 10.11: Concentration, p.p.m.* * Ft Et 10.94 10.50 5.87 6.36 6.71: 7.52 4.77 6.89 7.21s 8.03: 9.60 4.60: 5.22 11.39 6.93: 7.63 7.98:7 7.90: 11.36: 11.71 11.12: 7.63 11.21 4.84 10.63 6.19 6.72 8.78 11.71 Concentration, Concentration, Concentration, Concentration, p.p.m.* p.p.m.* p.p.m.* p.p.m.* ---- Ft Et Ft Et Ft Et Ft Et 13.25: 12.72 4.05: 4.30 4.10: 4.40 1.93: 2.075 13.25 7.40 7.70 2.45 2.60 2.47 2.65 1.33 1.26 8.28 9.11 2.70; 3.05 2.90: 3.15 1.38: 1.48 9.76: 6.49: 8.34 2.66: 2.65 2.875: 2.90 1.41: 1.36 8.43: 9.25 3.025: 3.10 3.15: 3.20 1.42: 1.51 8.41s 15.50 13.57 4.30 4.60 4.45 4.70 2.52 2.21 6.08: 5.86 1.90: 1.97 2.00: 2.05 1.03: 0.96 5.865 12.06; 12.87 4.50; 4.35 4.45: 4.45 2.23: 2.10 12.16 7.20: 7.50 2.55: 2.55 2.85: 2.60 1.15: 1.22 7.40 8.14 2.50: 2.75 2.65; 2.80 1.21: 1.33 9.30S7 10.63 3.25: 3.50 3.50: 3.70 1.64: 1.73 9.30: 16.36 14.18 4.30: 4.80 4.601 4.90 2.38: 2.31 16.68: Concentration, p,p.m.* * Ft Et 0.78: 0.88 0.49 0.54 0.65: 0.63 0.56; 0.58 0.68: 0.64 0.89 0.94 0.41: 0.41 0.98: 0.89 0.51: 0.52 0.52: 0.56 0.68: 0.74 1.10: 0.98 Concentration, p.p.m.* * Ft Et 0.85: 0.92 0.62 0.56 0.64: 0.66 0.58: 0.60 0.67: 0.67 0.98 0.98 0.42: 0.42 0.97: 0.93 0.53: 0.54 0.66: 0.59 0.71: 0.77 1.03: 1.03 c1 w w134 Analyst, Vo,?.I04 For this purpose, a portion of food (10-20 g) was weighed accurately, and known amounts of the elements concerned were added. After treatment in accordance with the method described above, the metals were determined polarographically.The foods used were beef sausage with pork, fresh meat (beef), pork luncheon meat, bread, mixed vegetables, jelly, orange and lemon drink. The results are shown in Tables 11, 111, IV, V, VI, VII and VIII, respectively. KAPEL AND KOMAITIS : POLAROGRAPHIC DETERMINATION OF The method was applied to various kinds of food. TABLE VIII DETERMINATION OF THE ELEMENTS I N ORANGE - LEMON DRINK Footnotes as in Table 11. Metal ion Cr(V1) .. . . Sn(1V) . . .. Fe(II1). . .. Pb(I1) . . .. Cd(I1) . . .. Hg(I1) .. .. Cu(I1) . . .. Zn(I1) . . .. As(II1). . . . Sb(III).. .. Te(1V) .. .. Se(1V) .. .. Concentration, p.p.m.* * =t Et 0.17: 0.16 0.10 0.10 0.11: 0.11 0.09; 0.10 0.11: 0.11 0.18 0.17 0.08: 0.07 0.15: 0.16 0.10: 0.09 0.11 0.10 0.12: 0.13 0.15 0.17 Concentration, p.p.m.* F f.Et ' 0.86: 0.78 0.49 0.47 0.55: 0.56 0.49: 0.51 0.57: 0.57 0.855 0.83 0.37: 0.36 0.72: 0.78 0.47: 0.46 0.52: 0.50 0.62: 0.65 0.83: 0.87 Concentration, p.p.m.* - Ft Et 0.15: 0.16 0.09 0.10 0.12; 0.11 0.10: 0.10 0.11: 0.11 0.18 0.17 0.07: 0.07 0.17: 0.16 0.09: 0.09 0.09$ 0.10 0.14: 0.13 0.18: 0.17 Discussion The elements were investigated in a range of concentrations varying from 21 to 0.1 p.p.m. Mercury( 11) was determined by the normal polarographic process, because the mercury peak is in the region of zero voltage and the resistance - capacitance derivative circuit does not give accurate results. Tin(1V) was also determined by the normal process. The elements most accurately determined were cadmium(II), lead(I1) (for concentrations less than 7 p.p.m.), antimony(III), arsenic(III), zinc(I1) and tellurium(1V).Copper(I1) was determined with less accuracy, whilst chromium(VI), selenium(IV), iron(III), mercury(I1) and tin(1V) gave results which were even less accurate. The errors in the determinations of chromium(VI), selenium(IV), iron(II1) and tellurium(1V) could be attributed in part to the fact that the peaks of these elements were close to the hydrogen peak, with which they partly overlapped. Mercury(I1) gave a peak near the point of zero voltage (E, = -0.05 to -0.07 V) and, because of this, it could not be deter- mined accurately. The same was true of copper(I1) ( E , = -0.20 V), although to a lesser degree. The time required for the complete analysis varied from 4 to 7 h, according to the con- centrations of the elements. The reason for this was that the time required for the extraction of copper(II), mercury(II), lead(II), zinc(I1) and cadmium(I1) with dithizone increased considerably when the concentrations were greater than 10 p.p.m. The use of the resistance - capacitance derivative graph was necessary for most of the elements at concentrations below 1.2 p.p.m. Conclusion The procedure possesses considerable advantages of time and cost over most other methods designed for the simultaneous determination of 12 elements.February, 1979 TRACE ELEMENTS I N FOOD FROM A SINGLE DIGEST 135 Our thanks are due to Professor D. S. Robinson and Professor A. G. Ward for their interest and encouragement and to J. Sainsbury Limited for financial support. One of us (M.E.K.) thanks Professor D. S. Galanos of the University of Athens for his encouragement, the Greek State Scholarship Foundation for a maintenance grant and the University of Athens for leave of absence. References 1. 2. 3. Fischer, H., Angew. Chew., 1937, 50, 919. Baudisch, O., Chemikerzeitung, 1909, 33, 1298. Lacoste, R. J., Earing, M. H., and Wiberly, S. E., Analyt. Chew., 1951, 23, 871. Received March 3rd, 1978 Accepted Augu.st 23rd, 1978

ISSN:0003-2654    

DOI:10.1039/AN9790400124

出版商:RSC    

年代:1979

数据来源: RSC

摘要:

136 Amlyst, February, 1979, Vol, 104, @p. 136-142 Apparatus for the Automatic Preparation of Soil Extracts for Mineral-nitrogen Determination J. A. P. Marsh, R. Kibble-White and C. J. Stent Agricitltural Research Council Weed Research Organizatzon, Begbroke Hill, Yarnton, Oxford, OX5 1PF An apparatus is described that automatically prepares samples and feeds an AutoAnalyzer system. It consists of a reagent adder, which adds the correct volume of extractant for an approximately weighed amount of soil, and a sample preparation unit, which mixes, filters, dilutes and loads samples on to an AutoAnalyzer sampler. The results obtained using the apparatus were in good agreement with those obtained by manual sample preparation. Keywovds : Mineral-nitrogen determination ; soil analysis ; automatic extraction The Microbiology Group at the Weed Research Organization analyses about 3 500 samples for mineral-nitrogen each year.The manual preparation of such large numbers of soil samples for the determination of ammonium- and nitrate- plus nitrite-nitrogen on an Auto- Analyzer is tedious, time consuming and prone to considerable operator error. As accurate weighing of soil samples and addition of a fixed volume of extractant are two of the slowest jobs involved, it was decided that a considerable saving of skilled labour could be made by building a modified version of the reagent adder manufactured by the British Sugar Corporation Ltd. for use in their Tarehouse Laboratories. The remainder of the system was improved by designing and building a sample preparation unit (SPU) that would automatically stir and filter the mixture, dilute the extract and load it on to an AutoAnalyzer sample tray.Apparatus Reagent Adder The reagent adder (Fig. 1) consists of a basic beam balance (Denward Instruments Ltd.) modified to weigh between 19 and 21 g and tared to use standard-mass plastic beakers. One balance pan has been removed and replaced with a palladium probe connected to the equip- ment electronics. This probe is suspended in the neck of a specially manufactured pipette with a precision-bore neck (T. W. Wingent Ltd., Cambridge) designed such that a 1-g variation from 20 g on the balance gives a difference of 2 ml in the of the equipment cs dependent on the use of an electrolytic solution were chosen for the liquid in use (2 M potassium chloride solution).KCI reservoir pipette. The working and component values I Control panel I Inlei. Outlet f- valve Microswitch sv 2 (MS 1) Fill Empty Normal Inhibit Inhibit Fill Fig. 1. Layout of the reagent adder.MARSH, KIBBLE-WHITE AND STENT 137 A schematic drawing of the electronic circuit is shown in Fig. 2. When the mains switch S1 is closed, power is applied to the mains transformer whose nominal output of 20 V is rectified to produce a supply, allowing for losses in the transformer and rectifiers, of approxi- mately 24 V d.c.; S2, the Inhibit empty switch, and S3, the Inhibit fill switch, are shown in the inhibit position, and S4, the Normal - Flush switch, is shown in the flush position. During normal operation these switches are closed. When power is applied to the circuit, the two resistors (R1 and R2) apply sufficient current to the two transistors (in a Darlington pair) to turn on the solenoid valve SV1 and its associated indicator lamp LP1.The potassium chloride solution rises in the pipette until it reaches the mid probe (Fig. l), when the low-impedance path through the liquid removes the drive from the transistors and valve SV1 turns off. Between 19 and 21 g of moist soil are placed in a beaker on the balance pan, then push-button PB1 is operated. This applies power to the relay RA/2 and the mid probe is disconnected by contact RA2, thus re-establishing drive to the transistors and turning SV1 on again. The relay is latched on by contact RA1. The pipette then continues to fill until the liquid reaches the palladium probe, when the potassium chloride solution again removes drive from the transistors, thus turning off SV1 and RA2, which is unlatched. Indicator lamp LP1 also goes out, showing that the liquid has reached the required level.The beaker is removed from the balance pan and placed on a platform, which operates microswitch MS1. This applies power to valve SV2 and removes the transistor drive, via S4, from the Fill valve circuit, allowing the contents of the pipette to drain into the beaker. YT), N Liquid Mid Circuit diagram of the reagent adder. Fig. 2. As soon as MS1 has been operated the next sample is weighed out, while the pipette is draining. If the whole of the system is to be drained or washed out then switch S4 is opened and the microswitch MS1 is closed, thus applying power to SV1 and SV2 at the same time.It is important that the system should be thoroughly washed out after use because if potassium chloride solution is left in the system it crystallises and blocks the valves. If part of the pipework is to be drained, S2 or S3 is operated as appropriate. Sample Preparation Unit (SPU) After addition of the reagent, a magnetic follower (length 40 mm) is placed in each beaker, and the beakers are fitted into cups on the conveyor on the SPU (B in Fig. 3). The beakers are carried over four rows of revolving magnets to give a total of 48-min stirring. Subse- quently, the beakers stand, unstirred, for 12 min and then tip into filter-funnels (C in Fig. 3) around the periphery of a 10-sided Perspex table.The soil extract, filtered through a Whatman No. 1 filter-paper, is collected in a beaker (D in Fig. 3) fitted to a similar table lying below the filter-table. The tables rotate on a common spindle until the sample reaches a sample pick-up arm (E in Fig. 3) and is drawn up, via a peristaltic pump (F in Fig. 3), and is automatically diluted when necessary and dispensed via another arm (G in Fig. 3) into the cups on an AutoAnalyzer tray (H in Fig. 3).138 MARSH et al.: APPARATUS FOR THE AUTOMATIC PREPARATION Analyst, VoZ. 104 T I 470'rnrn 360imm K I Sick view L Fig. 3. Layout of the sample preparation unit. The frame of the conveyor is constructed from angle-iron. The conveyor consists of two lengths of Renolds chain with a 30-mm pitch, joined by 30 aluminium slats (350 x 25 mm) at right-angles to the chain.Each slat carries four plastic cups into which 100-ml plastic beakers fit tightly. The conveyor, in a triangular configuration with gear wheels at the angles (J in Fig. 3), is driven by a Parvalux 9.0 N m torque motor (K in Fig. 3) running at 5 rev min-l. The stirring mechanism consists of 16 horseshoe magnets (Gallenkamp, Cat. No. XJP- 780-T) set in four rows of four (L in Fig. 3) and attached to ball races by spindles, each of which carries a toothed pulley. The magnets are rotated by a toothed drive belt, and their polarities are arranged such that the magnetic field helps the turning and, therefore, only a small motor (Parvalux SD8S, 50 W) is required to drive them.l The two 10-sided Perspex tables are fixed 150 mm apart on a vertical spindle driven by a motor (95 W).The top disc is 0.965 m across the flats, 9 mm thick and each of the 10 sides has four holes drilled to hold 63 mm diameter polypropylene funnels (M in Fig. 3). The funnels are fitted with rubber grommets (9.5 x 6.3 mm), which sit flush on the upper table to prevent unfiltered sample from running down the outside of the funnel and dripping into the beaker below. The lower disc has depressions drilled in it to locate the collection beakers, under the funnels (D in Fig. 3). The collection beakers are 100-ml plastic beakers that have been cut off at approximately 35 mm to facilitate entry of the pick-up arm. At the edge of the upper surface of the lower disc are cams (N in Fig.3), which are positioned to contact a microswitch that controls the position of the tables at each movement. The movements of the Perspex tables and the conveyor are co-ordinated so that as each beaker moves over the end of the conveyor it tips its contents into a funnel aligned with it. A sheet of polyethylene draped from a bar (P in Fig. 3) prevents cross-contamination of the samples as they move round. The sample pick-up unit is situated approximately one third of the way round the table (E in Fig. 3) and is designed to pick up the sample from the table, flush with air, pick up a wash solution and then flush with air again until it picks up the next sample. It consists of an arm connected to a 12-V car windscreen-wiper motor withFebruary, 1979 OF SOIL EXTRACTS FOR MINERAL-NITROGEN DETERMINATION 139 limit switches to stop movement at the end of the track.For sampling or washing, the arm is lowered into the sample or wash solution by means of an electromagnet, and is returned to its upper position after sampling, by means of a spring. The sample is picked up through a small-bore polyethylene tube, the suction being provided by tubing on the peristaltic pump of the AutoAnalyzer. If required, the sample can be diluted at this stage using the pump and AutoAnalyzer coils, and it is then pumped to the AutoAnalyzer sample tray via a dispenser unit (G in Fig. 3). The dispenser unit is similar to the pick-up unit, but has a simple horizontal movement from the AutoAnalyzer sample tray to waste. The timing and switching sequence of the SPU operates on a 3-min cycle, which is started and maintained by the presence of a 110-V a.c.supply, that is switched by the timing cam on a Technicon AutoAnalyzer. Individual timings, which are all given as time after initiation, are controlled by standard plug-in (octal) timers and the auxiliary switching by standard plug-in (octal) relays. Schematic drawings of the circuit are shown in Fig. 4. The two halves of the circuit are drawn separately for clarity only. Fig. 4 (a) is the timing circuit and Fig. 4 ( b ) is the low-voltage switching circuit. The presence of mains voltage at the inlet powers timer TF, which closes the 110-V circuit. When the 110-V supply from the Auto- Analyzer switches on, relay RA operates and connects the mains supply to the timer circuit and the 12-V d.c.supply to the low-voltage circuit. Power is thus supplied to the pick-up solenoid via pin 4 and the pick-up arm drops into the sample on the lower table. After 60 s TA stops and the +12-V supply is transferred from the pick-up solenoid to the pick-up motor via pin 3. This allows the pick-up arm to spring back to its upper position. The motor swings the pick-up arm to the wash position where a double-limit switch removes power from the motor and connects the +12-V supply to relay RD. This relay latches on and removes the +12 V from pin 3 for the remainder of the cycle. At the same time mains voltage is applied to timer TG and via contact TG1 to the table motor. The table rotates and allows microswitch C (MSC) to slide off the table cam (N in Fig.3) and close. This powers RB, which by-passes TG1. TG cuts out while the table is in motion so that when the next cam position is reached MSC removes the +12 V from RB, which in turn removes power from the table motor, thus stopping the table at the appropriate point. At every fourth movement of the table, one corner of the table momentarily closes micro- switch MSB, which powers RC. This applies mains voltage to the conveyor motor and a cam on the conveyor allows microswitch MSA to close before MSB opens again, thus main- taining power to RC. This allows the conveyor to continue to move to its next position when the next cam opens MSA and RC removes power from the conveyor motor. After 75 s TD makes contact TD1, enabling contact TE1 to apply +12 V to the dispenser motor, thereby driving the dispenser arm to the sample position, where it is stopped by a limit switch.Timer TE is initiated by mains power from the other contact on TD. After 90 s TB completes its cycle and contact TB1 is made, enabling TC1 to apply +12 V to the pick-up solenoid via pin 2, thus lowering the pick-up arm into the wash solution. Contact TB2 applies mains voltage to TC. After 115 s TE switches, TE1 applies +12 V to the dispenser motor via pin 3 and the dispenser arm drives to the waste position, where it is stopped by a limit switch. After 120 s TC completes its cycle, TC1 removes +12 V from the pick-up solenoid, allowing the pick-up arm to return to its upper position, and applies +12 V to the pick-up motor, which drives the pick-up arm to its original position over the disc, where a limit switch stops it.The equipment remains in this position for the remainder of the 3 min. The end of the sequence is signified by a short interruption in the 110-V supply from the AutoAnalyzer, which causes a similar break in the 12 V and mains supplies, thus re-setting the timers. When the 110-V supply is re-applied, by the timing cam of the AutoAnalyzer, the sequence re-st arts. Results Tests were carried out to compare the efficiency of extraction results obtained using a manual weighing and sample preparation method, and the reagent adder and sample preparation unit. Extracts of four replicates of 10 soils were prepared by each method and140 MARSH et al. : APPARATUS FOR THE AUTOMATIC PREPARATION Analyst, VoZ.104 analysed for nitrate- plus nitrite-nitrogen by a diazotisation and coupling reaction with sulphanilic acid and N-( 1-naphthy1)ethylenediamine and ammonium-nitrogen by an indo- phenol method. These methods q e described fully by Greaves et aL2 (4 2 s I I I 11ov N 8 TD2 6 L 5 I s 4 6 RB 1 3 1 T l f N r - - - y - - - - I Pickup -7 - 6 I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I t I I I I I -1 I I I I I I I I I I r -.- - - - - - - TE 1 I 1 3 I I ' 2 (- Fig. 4. Circuit diagrams of the sample preparation unit.February, 1979 OF SOIL EXTRACTS FOR MINERAL-NITROGEN DETERMINATION 141 Results for the nitrate plus nitrite determinations (Table I) show that there was very little difference in the amounts extracted with either technique.TABLE I DETERMINATION OF NITRATE- PLUS NITRITE-NITROGEN IN SOIL AFTER EXTRACTION BY MANUAL AND AUTOMATIC TECHNIQUES Sample 4 5 6 7 8 9 10 Manual preparation determinations/ deviation/ Coefficient of pg g-l dry soil pg g-' variation, % 23.15 0.671 2.9 14.00 0.303 2.2 8.38 0.120 1.4 38.00 1.281 3.4 19.08 0.426 2.2 4.80 0.293 6.1 10.08 0.086 0.9 47.45 0.426 0.9 14.80 0.172 1.2 57.94 1.536 2.7 7 A \ Mean of 4 Standard Automatic preparation determinations/ deviation/ Coefficient of pg g-l dry soil pg g1 variation, % 23.26 0.587 2.5 13.54 0.220 1.6 8.18 0.160 2.0 36.82 0.913 2.5 19.04 0.323 1.7 4.82 0.223 4.6 10.54 0.250 2.4 47.66 0.798 1.7 15.28 0.215 1.4 58.49 0.538 0.9 A f \ t Mean of 4 Standard Automatic - manual A \ Difference/ Standard w g - l error/trggl t 0.11 0.446 0.26 - 0.46 0.187 2.46 - 0.20 0.100 2.03 - 1.17 0.786 1.49 - 0.04 0.268 0.15 0.01 0.184 0.07 0.46 0.132 3.49 0.20 0.462 0.45 0.47 0.138 3.44 0.54 0.814 0.67 Ammonium-nitrogen (Table 11) extracted by the automatic method was slightly higher than that extracted manually, and was significantly higher when less than 2 pg of ammonium- nitrogen per gram of dry soil was present.This may be due to the timing for the filtering of the samples and loading the AutoAnalyzer being shorter than with the manual method. The regression coefficient for results of automatic against manual methods is significantly less than 1 at P = 0.01, the relationship between the two being TABLE I1 DETERMINATION OF AMMONIUM-NITROGEN IN SOIL AFTER EXTRACTION BY MANUAL AND AUTOMATIC TECHNIQUES Sample 1 2 3 4 5 6 7 8 9 10 Manual preparation determinations1 deviation/ Coefficient of pg g-l dry soil pg gl variation, % 6.93 0.264 3.8 4.41 0.261 5.9 I * If Mean of 4 Standard 3.10 0.082 2.6 4.28 0.101 2.4 2.49 0.113 4.5 0.48 0.024 5.1 3.18 0.030 0.9 1.46 0.017 1.2 0.55 0.024 4.3 1.48 0.024 1.7 Automatic preparation Mean of 4 Standard determinations/ deviation/ WLg g-l dry soil !J.g g-' 6.53 0.155 4.33 0.088 3.20 0.051 4.37 0.278 2.42 0.118 0.66 0.025 3.37 0.136 1.63 0.062 0.74 0.025 1.61 0.015 Coefficient of variation, yo 2.4 2.0 1.6 6.4 4.9 4.4 4.0 3.8 3.4 0.9 Automatic - manual L I \ Diff erencel w g-l - 0.40 - 0.08 0.09 0.09 - 0.07 0.08 0.18 0.17 0.19 0.13 Standard errorlpg g-1 t 0.153 2.64 0.138 0.69 0.048 1.93 0.148 0.68 0.082 0.82 0.018 4.69 0.070 2.66 0.032 5.20 0.017 11.28 0.014 8.89 Discussion The equipment described in this paper has now been used for the preparation of over 4000 samples for determining ammonium- and nitrate- plus nitrite-nitrogen in soil and has proved reliable.A comparison of the labour-intensive stages in the manual and automatic sample prepara- tions is shown in Table 111. The equipment has reduced the labour requirement by approxi- mately 60%. The SPU has also been used for the extraction of available phosphate from soil, although analytical problems have, at present, made it impossible to automate the loading of the A4utoAnalyzer for this analysis. We thank Mr. D. F. A. Horsley and Mr. P. R. Leaton of the British Sugar Corporation Ltd. for their guidance and advice in building the reagent adder, Mr. R. W. Foddy for constructing the equipment and Mr. R. C. Simmons who designed the prototype circuits and did the original wiring.142 MARSH, KIBBLE-WHITE AND STENT TABLE I11 COMPARISON OF STAGES REQUIRING LABOWR IN THE MANUAL AND AUTOMATIC NITRATE- PLUS NITRITE-NITROGEN ON AN AUTOANALYZER PREPARATION OF SOIL SAMPLES FOR DETERMINATION OF AMMONIUM- AND Stage 1 2 3 4 5 6 7 8 9 10 11 Manual preparation Accurate weighing into bottles Accurate addition of extractant Stoppering bottles Placing on shaker Removing from shaker Transport to filter area Unstoppering bottles Setting up filters Filtering Diluting Loading AutoAnalyzer tray Stage Automatic preparation I Rough weighing into pre-tared beakers 2 Placing on conveyor belt 3 Setting up filters References 1. 2. Baker, K. F., Analyst, 1970, 95, 885. Greaves, M. P., Cooper, S. L., Davies, H. A., Marsh, J. A. P., and Wingfield, G. I., “Technical Report, Received September 13th, 1978 Accepted September 20th, 1978 Agricultural Research Council Weed Research Organization,” 1978, No. 45, p.55.

ISSN:0003-2654    

DOI:10.1039/AN9790400136

出版商:RSC    

年代:1979

数据来源: RSC