ISSN:0003-2654    

DOI:10.1039/AN97499FX029

出版商:RSC    

年代:1974

数据来源: RSC

ISSN:0003-2654    

DOI:10.1039/AN97499BX031

出版商:RSC    

年代:1974

数据来源: RSC

摘要:

August, 19741 SUMMARIES OF PAPERS IN THIS ISSUE iiiSummaries of Papers in this IssueThe Determination of Trace Amounts of Lead in Steel byAnodic Stripping VoltammetryA procedure is described for the direct determination of lead in steelsamples by anodic stripping voltammetry. The method is both simple andrapid, about 1 hour being needed to complete an analysis. No chemicaloperations are required other than sample dissolution, which minimises therisk of contamination or losses of lead. Interferences to the method are few,only copper (more than 0.9 per cent.) and molybdenum (more than 0.1 percent.) being of any significance.The detection limit of the method was calculated to be 0.0001 per cent.of lead and the accuracy and precision have been established by the analysisof a wide range of standard steel samples.B.METTERS and B. G. COOKSEYDepartment of Pure and Applied Chemistry, University of Strathclyde, CathedralStreet, Glasgow, G1 1XL.A~zaZyst, 1974, 99, 467-465.-4 Modification to the Extraction - Atomic-absorption Method for theDetermination of Antimony, Bismuth, Lead and TinIt has been shown that the use of organometallic compounds for thecalibration procedure used in the extraction - atomic-absorption determinationof antimony, bismuth, lead and tin in metallurgical materials leads to erroneousresults a t low concentration levels. Results are presented, which demonstratethat the technique is satisfactory when calibrated with solutions that havebeen taken through the extraction procedure.K. THORNTONHenry Wiggin and Co.Ltd., Holmer Road, Hereford.and KEITH E. BURKEThe International Nickel Company, Inc., Paul D. Rlerica Research Laboratory,Sterling Forest, Suffern, New York 10901, U.S.A.Analyst, 1974, 99, 469-470.Spectrophotometric Determination of Magnesium in TobaccoLeaves with Eriochrome Black BA spectrophotometric method for the determination of magnesium intobacco leaves with Eriochrome black B is described. After destruction ofthe tobacco leaves with a mixture of nitric, perchloric and sulphuric acids, themagnesium is separated from interfering ions by ion-exchange chromatographyon Dowex 5OW-X12, 100 to 200 mesh, and determined spectrophotometricallyafter coupling it with Eriochrome black B. The accuracy of the method wastested against a reference material.G.SLEGERS and A. CLAEYSLaboratory of Analytical Chemistry, Faculty of Pharmaceutical Sciences, StateUniversity of Ghent, 9000 Ghent, Belgium.Analyst, 1974, 99, 471-475iv THE ANALYST [August, 1974THE ANALYSTED IT0 RIAL ADVISORY BOARDChairman: H. J. Cluley (Wembley)*L. S. Bark (Solford)R. Belcher (Birmingham)L. J. Bellamy, C.B.E. (Waltham Abbey)L. S. Birks (U.S.A.)E. Bishop (Exeter)'R. M. Dagnall (Huntingdon)E. A. M. F. Dahmen (The Netherlands)*J. B. Dawson (Leeds)A, C. Docherty (Billingham)D. Dyrssen (Sweden)*W. T. Elwell (Birmingham)*D. C. Garratt (London)J. Hoste (Belgium)D. N. Hume (U.S.A.)H. M. N. H. Irving (Leeds)M. T. Kelley (U.S.A.)*J. A. Hunter (Edinburgh)W.Kemula (Poland)*G. F. Kirkbright (London)G. W. C. Milner (Harwell)G. H. Morrison (U.S.A.)*J. M. Ottaway (Glasgow)*G. E. Penketh (Billingham)S. A. Price (Tadworth)D. 1. Rees (London)E. B. Sandell (U.S.A.)*R. Sawyer (London)A. A. Smales, O.B.E. (Harwell)H. E. Stagg (Manchester)E. Stahl (Germany)A. Walsh (Australia)T. S. West (London)P. Zuman (U.S.A.)*A. Townshend (Birmingham)* Members of the Board serving on the Executive Committee.NOTICE TO SUBSCRIBERSSubscriptions for The Analyst, Analytical Abstracts and Proceedings should beThe Chemical Society, Publications Sales Omce,Black horse Road, Letch wort h, He rts.Rates for 1974(other than Members of the Society)sent to:(a) The Analyst, Analytical Abstracts, and Proceedings, with indexes .. . . f37.00(b) The Analyst, Analytical Abstracts printed on one side of the paper (withoutindex), and Proceedings . . . . . . . . . . . . . . f38.00(c) The Analyst, Analytical Abstracts printed on one side of the paper (withindex), and Proceedings . . . . . . . . . . . . . . f45-00The Analyst and Analytical Abstracts without Proceedings-(d) The Analyst and Analytical Abstracts, with indexes . . . . . . . . f34.00(e) The Analyst, and Analytical Abstracts printed on one side of the paper (without(9 The Analyst, and Analytical Abstracts printed on one side of the paper (withindex) . . . . . . . . . . . . . . . . . . €35.00index) . . . . . . . . . . . . . . . . . . €4240(Subscriptions are NOT accepted for The Analyst and/or for Proceedings aloneAugust, 19741 SUMMARIES OF PAPERS I N THIS ISSUEThe Spectrophotometric Determination of Ampicillin and Cloxacillinin Combined InjectionsA method is described for the determination of ampicillin and cloxacillinin injections, Ampicillin is determined by an absorbance difference techniquebased on the higher absorbance of ampicillin at 268 nm in a solution a t pH 5than in one a t pH 9.Cloxacillin is determined by measurement of the absorb-ance a t 275 nm and the application of a small correction for the absorbance ofampicillin. The accuracy, precision and specificity of the method are dis-cussed. The analytical results obtained for commercial samples of ampicillin -cloxacillin (1 +- 1 and 2 + 1) mixtures are compared with those obtained bymicrobiological assay.A.G. DAVIDSON and J. B. STENLAKEDepartment of Pharmaceutical Chemistry, University of Strathclyde, Glasgow,G1 1XWAnalyst, 1974, 99, 476-481.VColorimetric Determination of Piperazine in PharmaceuticalFormulationsPiperazine can be satisfactorily determined in pharmaceutical prepara-tions or formulations such as effervescent granules and elixirs containinghexamine, colchicine, atropine sulphate, sodium benzoate, lithium benzoate,lithium citrate, sodium citrate, sodium hydrogen carbonate, tartaric acid,citric acid, lactose, sucrose and Tinct. ammi visnaga. The diluted samplesolution is treated with a 0-6 per cent. aqueous 1,2-naphthoquinone-4-sulphon-ate solution in the presence of acetate - citrate buffer a t pH 7.5.The tempera-ture of the reaction should be between 10 and 15 "C and the colour produced ismeasured a t 490nm.YEHIA M. DESSOUKYPharmaceutical Chemistry Department, Faculty of Pharmacy, Cairo University,Cairo, Egypt.and SAAD A. ISMAIELResearch Department, Socidte' Misr pour 1'Industrie Pharmaceutique, 92 El MatariaStreet, Post El Zeitoun, Cairo, Egypt.Analyst, 1974, 99, 482-486.Colorimetric Determination of Antazoline in Some PharmaceuticalPreparations with Sodium NitriteA colorimetric method for the determination of antazoline in pharma-ceutical preparations with sodium nitrite has been developed. The methodinvolves treatment of a cooled and acidified dilute aqueous solution of thesample with sodium nitrite. The yellow colour produced is stabilised by theaddition of propan-2-01 or ethanol and the absorbance then measured a t410 nm.Naphazoline, tolazoline, clemizole, diphenhydramine, chlorpheniramine,ephedrine, cetrimide, benzalkonium chloride and zinc salts, even if present inamounts ten times greater than that of antazoline, do not interfere.M. M. AMER, M. S. TAWAKKOLDepartment of Analytical Chemistry, Faculty of Pharmacy, Cairo University, Cairo,and S. A. ISMAIELResearch Department, Socie'td Misr pour 1'Industrie Pharmaceutique, 92 El MatariaStreet, Post El Zeitoun, Cairo, Egypt.Analyst, 1974, 99, 487-490.Egypt

ISSN:0003-2654    

DOI:10.1039/AN97499FP085

出版商:RSC    

年代:1974

数据来源: RSC

摘要:

August, 19741 THE ANALYST viiCLASSIFIED ADVERTISEMENTSThe rate for classified advertisements is 35p a line (or spaceequivalent of a line) with an extra charqe of ~ o p for theuse of a Box Number. Semi-displayed classifiedadvertisements are L4 for single-column inch.Copy for claSsi$ed advertisements required not later thanthe r8th of the month preceding the date of Publication whichi s on the r6th of each month. Advertisements should beaddressed to J. Arthur Cook, 9 Lloyd Square, London,WCrX 9BA. Tel.: or-837 6315APPOINTMENTS VACANTLady chemist, graduate or non-graduate, required for wcrlis laboratoryin south-east London. Please apply to Box No. 236, c/o J. ArthurCook, 9 Lloyd Square, London WClX 9BA.IPlease mention THE ANALYSTwhen replying to advertisementsBOOKSMONOGRAPHSREPRINTSorders for all publications ofthe Society (except journals)should be sent direct or througha bookseller to-THE SOCIETY FORANALYTICAL CHEMISTRYBook Deportment9/10 Savile Row,London, WIX IAFANALOlD compressed chemicalreagents offer a saving in the use oflaboratory chemicals.The range of over 50 chemicals in tabletform includes Oxidizing and ReducingAgents, Reagents for Colorimetric Analysisand Indicators for Complexometric titra-tions.Full details of all Analoid preparations freeon request from:RIDSDALE & CO.LTD.Newham Hall, Newby,Middlesbrough, Teesside TS8 9EATelephone: 0642 372 16m)I kq RANKHILGERHOLLOW 11 I/ CATHODELAMPS For 4Long LifeReliability11 and 11High Energydelivery service11 ILong LifeCompletely sealed lamp with windowfused t o the glass envelope, preciselyfilled t o the appropriate gas pressure.High EnergyRank Hilger lamps achieve strong emis-sion by concentrating the discharge ina cavity of the small diameter cathode,which is fully shielded.This also elimi-nates spurious discharges and ensuresexcellent short term stability.ReliabilityManufactured on an advanced produc-tion plant from materials carefullyselected for their high purity. Duringprocessing lamps are baked at a pres.sure of lo+* Torr to orovide maxi.mum outgassing giving rise t o lowbackground and clean line spectrum.UniversalDesigned to fit not only Rank HilgerH1170 and H1550 but almost everyAtomic Absorption Spectrophoto-meter, being suitable for 3 modeoperation: D.C., Pulsed or Modulated.ValueBeing at the lower end of the pricerange coupled with Long Life, HighEnergy and Reliability offer a remark-able value for such a high quality lamp.RANKHILGERWESTWOOD.MARGATE mL KENTmCT94JL- ENQLAN...v111 SUMMARIES OF PAPERS I N THIS ISSUEColorimetric Determination of Small Amounts of Cs to GI,-,Alcohols in Their Phthalate Esters[August, 1974Methods were examined for the determination of small amounts of C,to C,, alcohols in their phthalate esters involving the use of vanadium 8-hydroxyquinolinate in benzene or toluene and 3,5-dinitrobenzoyl chloride inpyridine as colorimetric reagents. Two modified procedures are presented thatwill reliably determine free alcohols in the 0.01 to 0.4 per cent.m/m range.S. HARRISON, H. HINCHCLIFFE and G. L. WOODROFFEResearch and Development Department, Imperial Chemical Industries Limited,Petrochemicals Division, Billingham, Teesside, TS23 1 JB.Analyst, 1974, 99, 491-497.Determination of Primary and Secondary Amines Alone and inMixtures with Tertiary AminesAn iodatometric method has been developed for the determination ofprimary and secondary amines that is based on their quantitative reactionwith phenyl isothiocyanate in dimethylformamide to form substituted thio-ureas. These are titrated with potassium iodate in an acidic medium a troom temperature. The end-point is detected visually by the yellow colourimparted to the solution by the first drop of the iodate solution in excess,and potentiometrically by using a bright platinum-wire indicator electrodeand a saturated calomel reference electrode. Methods have also been de-veloped for the determination of primary (or secondary) amines and tertiaryamines in the presence of each other.An excess of phenyl isothiocyanate,added to the mixture in solution in dimethylformamide, converts the primary(or secondary) amines into the corresponding di- (or tri-) substituted thioureas,whereas the tertiary amines are left unreacted. The conductimetric titrationof tertiary amines with trichloroacetic acid, followed by the iodatometrictitration of thioureas formed, enables the mixture to be analysed for bothcomponents.BALBIR CHAND VERMA and SWATANTAR KUMARDepartment of Chemistry, Punjabi University, Patiala, India.The methods described are simple, accurate and reliable.Analyst, 1974, 99, 498-502.A Technique for the Determination of Trace Anions by theCombination of a Potentiometric Sensor and Liquid Chromatography,with Particular Reference to the Determination of HalidesA technique for the determination of trace amounts of halides in thepresence of other ions is described.The species are separated by means ofliquid chromatography and detected potentiometrically by a silver - silverchloride micro-electrode. The technique readily lends itself to automationand an apparatus for rapid, repetitive analyses has been designed. By carefulchoice of the eluting agent and stationary phase it is possible to achieve avariety of separations, e.g., the determination of chloride in the presence ofexcess of sulphide, the separation and determination of nanogram amounts ofchloride, bromide and iodide in mixed halide solutions and the determinationof chloride in boiler waters.In addition, conditions for the extension of the technique to the separationand determination of other anions are proposed.M.C. FRANKS and D. L. PULLENBP Research Centre, Chertsey Road, Sunbury-on-Thames, Middlesex, TW16 7LN.Analyst, 1974, 99, 503-514August, 19741 SUMMARIES OF PAPERS I N THIS ISSUEA Method for the Determination of Total Sulphur inSilicate RocksAfter oxidising sulphides to sulphates with a mixture of sodium chlorateand hydrochloric acid, the sample is refluxed with a reducing mixture ofsodium iodide, red phosphorus, hypophosphorous acid, orthophosphoric acidand propionic acid.The hydrogen sulphide generated is absorbed in potas-sium hydroxide solution and titrated against 2- (hydroxymercuri) benzoic acidsolution with dithizone as indicator. The approximate range covered is 5 to2000 mg kg-l of sulphur in the rock.J. M. MURPHY and G. A. SERGEANTDepartment of Trade and Industry, Laboratory of the Government Chemist,Cornwall House, Stamford Street, London, SE1 9NQ.Analyst, 1974, 99, 515-518.A Titrimetric Method for the Determination of Sulphate in FertilisersA method is described for the determination of sulphate in fertilisers inwhich the sulphate is precipitated with barium chloride from an acidifiedEDTA solution.The precipitate is filtered off by using membrane filters andis dissolved in ammoniacal EDTA ; the excess of EDTA is then titrated againsta solution of magnesium ions with Eriochrome black T as the indicator. Nointerference is encountered from iron, aluminium, fluoride or phosphate ions.A. D. CAMPBELL, D. P. HUBBARD and N. H. TIOHDepartment of Chemistry, University of Otago, Box 56, Dunedin, New Zealand.Analyst, 1974, 99, 519-522.Loss of Zinc and Cobalt During Dry Ashing of Biological MaterialLoss of zinc and cobalt during dry ashing of marine mussels (Mytilusedulis) and brown seaweed (Fucus spiralis) has been studied by using materiallabelled by exposure of living organisms to sea water spiked with zinc-65 orcobalt-60.Even after ashing in porcelain crucibles at temperatures of up to1000 OC, no significant loss of zinc or cobalt by volatilisation was observed.After ashing a t 450 and 550 "C, both radionuclides could be removed quantita-tively from the crucibles by leaching with hydrochloric acid. Adsorption onthe crucible after ashing at 1000 "C was measured only for cobalt-60. Asignificant proportion of the cobalt tracer could not be removed from thecrucibles by treatment of the latter with acid. From these results it is con-cluded that dry ashing is a reliable method of sample destruction for thedetermination of zinc and cobalt in M . edulis and F. spiralis.J. G. van RAAPHORST, A. W. van WEERS and H. M. HAREMAKERReactor Centrum Nederland, Petten, The Netherlands.Analyst, 1974, 99, 523-527.Determination of Copper(1) with N- BromosuccinimideIt has been found that N-bromosuccinimide readily and quantitativelyoxidises aqueous solutions of copper(1) at room temperature and in thepresence of dilute hydrochloric acid, the oxidising agent being irreversiblyreduced to succinimide.A procedure is suggested for the determination of copper(1) by titrationwith standard N-bromosuccinimide solution ; the results obtained werefound to be equally precise but more accurate than those given by thepermanganate and complexometric methods.Copper(II), after preliminary reduction, can also be determined by thesuggested procedure, whether alone or admixed with copper (I).A. ABOU EL KHEIR, M.AYAD and M. M. AMERDepartment of Analytical Chemistry, Faculty of Pharmacy, Cairo University, Cairo,Analyst, 1974, 99, 528-532.Egypt.iX THE ANALYST [August, 1974Reprints of Review PapersREPRINTS of the following Review Papers published in The Analyst since 1963 are available fromthe Book Department, Society for Analytical Chemistry, 9/10 Savile Row, London, W1X 1AF(not through Trade Agents). A complete list of all reprints available from earlier years can beobtained on request.The price per reprint is 50p; orders for four or more reprints of the same or different Reviewsare subject to a discount of 25 per cent. Remittance with order, made out to “Society forAnalytical Chemistry, ” will prevent delays.“Classification of Methods for Determining Particle Size,” by the Particle Size Analysis“Methods of Separation of Long-chain Unsaturated Fatty Acids,” by A.T. James (August,“Beer’s Law and its Use in Analysis,” by G. F. Lothian (September, 1963).“A Review of the Methods Available for the Detection and Determination of Small Amountsof Cyanide,” by L. S. Bark and H. G. Higson (October, 1963).“Circular Dichroism,” by R. D. Gillard (November, 1963).“Information Retrieval in the Analytical Laboratory,” by D. R. Curry (November, 1963).“Thermogravimetric Analysis,” by A. W. Coats and J . P. Redfern (December, 1963).“Some Analytical Problems Involved in Determining the Structure of Proteins and Peptides,”“The Faraday Effect, Magnetic Rotatory Dispersion and Magnetic Circular Dichroism, ” by“Electrophoresis in Stabilizing Media,” by D.Gross (July, 1965).“Recent Developments in the Measurement of Nucleic Acids in Biological Materials, ” by“Radioisotope X-ray Spectrometry,” by J . R. Rhodes (November, 1966).“The Determination of Iron(I1) Oxide in Silicate and Refractory Materials,” by H. N. S.“Activation Analysis,” by R. F. Coleman and T. B. Pierce (January, 1967).“Techniques in Gas Chromatography. Part I. Choice of Solid Supports,” by F. J. Palframan“Heterocyclic Azo Dyestuffs in Analytical Chemistry,” by R. G. Anderson and G. Nickless“Determination of Residues of Organophosphorus Pesticides in Food,” by D. C. Abbott and“Radioactive Tracer Methods in Inorganic Trace Analysis : Recent Advances, ” by J. W.“Gamma-activation Analysis,” by C.A. Baker (October, 1967).“Precipitation from Homogeneous Solution,” by P. F. S. Cartwright, E. J. Newman and“Industrial Gas Analysis,” by (the late) H. N. Wilson and G. M. S. Duff (December, 1967).“The Application of Atomic-absorption Spectrophotometry to the Analysis of Iron and“Inorganic Ion Exchange in Organic and Aqueous - Organic Solvents,” by G. J. Moody and“Radiometric Methods for the Determination of Fluorine,” by J. K. Foreman (June, 1969).“Techniques in Gas Chromatography. Part 11. Developments in the van Deemter RateTheory of Column Performance,” by E. A. Walker and J. F. Palframan (August, 1969).“Techniques in Gas Chromatography. Part 111. Choice of Detectors,” by T. A. Gough andE. A. Walker (January, 1970).“Laser Raman Spectroscopy,” by P. J. Hendra and C. J. Vear (April, 1970).“Ion-selective Membrane Electrodes,” by Ern6 Pungor and KlAra T6th (July, 1970).“X-ray Fluorescence Analysis,” by K. G. Carr-Brion and K. W. Payne (December, 1970).“Mass Spectrometry for the Analysis of Organic Compounds,” by A. E. Williams and H. E.“The Application of Non-flame Atom Cells in Atomic-absorption and Atomic-fluorescence“Liquid Scintillation Counting as an Analytical Tool,” by J. A. B. Gibson and A. E. Lally“The Determination of Some 1,4-Benzodiazepines and Their Metabolites in Body Fluids,”Sub-committee of the Analytical Methods Committee (March, 1963).1963).by Derek G. Smyth and D. F. Elliott (February, 1964).J. G. Dawber (December, 1964).H. N. Munro and A. Fleck (February, 1966).Schafer (December, 1966).and E. A. Walker (February, 1967).(April, 1967).H. Egan (August, 1967).McMillan (September, 1967).D. W. Wilson (November, 1967).Steel,” by P. H. Scholes (April, 1968).J. D. R. Thomas (September, 1968).Stagg (January, 1971).Spectroscopy,” by G. F. Kirkbright (September, 1971).(October, 1971).by J. M. Clifford and W. Franklin Smyth (May, 1974)

ISSN:0003-2654    

DOI:10.1039/AN97499BP089

出版商:RSC    

年代:1974

数据来源: RSC

摘要:

AUGUST, 1974 THE ANALYST Vol. 99, No. 1181 The Determination of Trace Amounts of Lead in Steel by Anodic Stripping Voltammetry BY B. METTERS AND B. G. COOKSEY (Department of Pure and Applied Chemistry, University of Strathclyde, Cathedral Street, Glasgow, G1 IXL) A procedure is described for the direct determination of lead in steel samples by anodic stripping voltammetry. The method is both simple and rapid, about 1 hour being needed to complete an analysis. No chemical operations are required other than sample dissolution, which minimises the risk of contamination or losses of lead. Interferences to the method are few, only copper (more than 0.9 per cent.) and molybdenum (more than 0.I per cent.) being of any significance. The detection limit of the method was calculated to be 0.0001 per cent.of lead and the accuracy and precision have been established by the analysis of a wide range of standard steel samples. POLAROGRAPHIC techniques, in particular anodic stripping voltammetry, are well suited to the determination of lead in a wide variety of materials. In recent years, however, d.c. polarography has seldom been used as an analytical tool for trace-metal analysis following the advent of the more convenient technique of atomic-absorption spectroscopy. This latter technique and spectrophotometry1*2 are the main methods currently employed for deter- mining lead in steel samples. Both techniques are limited to deteminations of lead in the range 0.1 to 0.01 per cent. in solutions containing 1.0 g of steel per *OO ml and solvent extrac- tion3-5 is required for the determination of lower levels.The solvent-extraction step is necessary either to provide a sufficiently concentrated solution for the final measurement or to remove interfering ions, particularly iron(II1) and copper(II), as described by Hofton and H ~ b b a r d . ~ Such procedures are extremely lengthy and frequently involve more than one extraction step as well as the necessary washing of the solvent. In addition, determinations of lead at concentrations below 0.002 per cent. in the steel are unreliable because at this level, losses of lead in the extraction steps, and contamination from the large surface area of the glass- ware and variety of chemicals used, become significant. Therefore, any direct procedure in which solvent extraction is eliminated would be advantageous.Such a method, however, must be extremely sensitive and relatively free from interferences. Anodic stripping voltam- metry partially satisfies both of these demands as it combines the selectivity of polarography with great sensitivity (e.g., lead can be determined down to a concentration of 1 x l o - 9 ~ in pure water). Many elements have been determined by this method and a review of the literature has been given.6 Initially, the technique was used only for the determination of zinc, cad- mium, lead and copper in water sample^^-^ because of the relatively poor selectivity caused either by overlap of stripping peaks or by a large background current that resulted from a high concentration of components in the sample solution, e.g., acids and iron(II1) in solutions of steel samples.These difficulties have been overcome by several workers,1°-12 who included a “medium exchange’’ step in the method. The electrolytic concentration step is carried out in the usual way in the sample solution, but the stripping or oxidation procedure is carried out in an alternative medium, usually a pure electrolyte, e.g., potassium chloride, perchloric acid, etc. In this way, large residual currents are eliminated from the stripping polarogram with little or no effect on the oxidation currents due to the metal ion under test, and by careful choice of this medium a good separation of oxidation peaks can be obtained. By using medium exchange, complex solutions can be analysed and Ariel, Eisner and Gottesfieldlo applied the technique to the determination of copper in Dead Sea brine and 0 SAC and the authors.467458 METTERS AND COOKSEY: THE DETERMINATION OF TRACE AMOUNTS OF [Analyst, VOl. 99 of copper in the presence of manganese and bismuth. This work was followed by application of a method for determining copper in steel samples.12 Phillips and Shainll also used medium exchange in the determination of tin in steel samples, but anodic stripping voltammetry does not appear to have been used for a similar determination of lead. The method proposed in this paper is direct, rapid and relatively simple, medium exchange being necessary only when molybdenum and fairly large amounts of copper are present. Interferences are few and the method has been applied to the determination of lead in a wide variety of steel and cast iron samples with a detection limit of 1 x per cent.of lead. EXPERIMENTAL APPARATUS- The instrument used was a Radiometer, Copenhagen, PO4 polarograph. The experi- ments were carried out in a water-jacketed 40-ml glass electrolysis cell at 25 "C with a three-electrode system that consisted of a saturated calomel reference electrode, a platinum- wire counter electrode and a hanging mercury drop electrode (H.M.D.E.). The solution was stirred by means of a rotating glass paddle stirrer operating at 600 r.p.m. and was de- oxygenated with a stream of nitrogen from an inlet tube. The cell arrangement is shown in Fig. 1. stir re \ 1 NZ Saturated calomel 1 r , reference electrode .M. Platinum counter Glass cell WateJ 1 vAIJ; ii\ 0 0 7 - Fig.1. Electrode assembly (H.M.D.E. denotes - \ hanging mercury drop electrode) The two leads from the polarograph were connected to the three electrodes via the (This electronic unit is not specially constructed electronic system represented in Fig. 2. required if a polarograph suitable for use with three electrodes is used.) RECOMMENDED METHOD FOR THE PREPARATION OF THE HANGING MERCURY DROP ELECTRODE- Seal about 2.0 cm of 22-gauge platinum wire into a 10-cm length of 7 mm 0.d. soda- glass tube in a bunsen flame. Cut off the protruding wire, grind it flush with the glass on a suitable rotating grinder and polish the end of the electrode with successively finer grades of emery cloth. Then etch the tip of the electrode in hot aqua regia for several minutes and rinse it well with distilled water.In order to achieve electrical contact, partially fillAugust, 19741 LEAD I N STEEL BY ANODIC STRIPPING VOLTAMMETRY 459 Counter S 2 A P Dummy 1 d + 1 5 V ov i Refer en ce TT f 3kn 1Mn G \ f - Balance b 0 T4 To calomel electrode terminal on polarograph s3B 0 ~ 5 To dropping - mercury S2 B electrode terminal on Dummy A T3 Use polarograph H.M.D.E. graph to a three-electrode system. Fig. 2. Circuit diagram of the electronic unit for the conversion of a two-electrode polaro- S1, on - off; S2, dummy - cell; and S3, balance - use the tube with mercury and place a length of copper wire in the tube as shown in Fig. 3. The electrode is now ready to be electroplated with mercury. ELECTROPLATING THE H .M .D .E .- Place the H.M.D.E.and a platinum-wire counter electrode in a beaker containing 10 per cent. V/V nitric acid. Connect the two electrodes to a constant-current device and alternately anodise then cathodise the H.M.D.E. for 15 s at 25 mA, finishing with cathodic electrolysis for 15 s. Rinse the electrodes with distilled water and place them in a beaker containing concentrated mercury(I1) nitrate solution that is 1 M in nitric acid. Now cathodise the H.M.D.E. for several minutes at 25mA until the exposed platinum tip is completely covered with mercury. Rinse it well with distilled water, shaking off any excess of mercury globules that may have formed during the electrolysis. Distil led Fig. 3. The hanging mercury drop electrode (D.M.E. denotes dropping-mercury electrode)460 METTERS AND COOKSEY: THE DETERMINATION OF TRACE AMOUNTS OF [Analyst, Vol.99 The electrode should now be capable of picking up individual drops of mercury repro- ducibly delivered from a dropping-mercury electrode, as shown in Fig. 3. (If a constant- current device is not available, the electroplating procedure can be carried out in the elec- trolysis cell by using a three-electrode system, ie., H.M.D.E., platinum counter electrode and reference electrode. The H.M.D.E. should be cleaned with nitric acid, anodised and cathodised at h1.5 V versm S.C.E., then electroplated in the mercury(I1) nitrate solution at -1.5 V versus S.C.E.) The electroplating procedure needs to be carried out only once a week provided the electrode is stored in distilled water when not in use.It is necessary, however, to use a fresh drop of mercury for each determination of lead. Throughout this work two drops of mercury were collected in the glass cup from a 72-cm head of mercury, which resulted in a single drop weighing approximately 11.0 mg. TREATMENT OF GLASSWARE IN ORDER TO REDUCE ADSORPTION- The electrolysis cell, calibrated flasks and pipettes were thoroughly cleaned in dilute nitric acid and allowed to stand in distilled water for 24 hours. Following a thorough rinsing with distilled water, the flasks and electrolysis cell were treated with a 2 per cent. solution of dimethyldichlorosilane in carbon tetrachloride for 24 hours. (A commercial form of this solution, Repelcote, is available from Hopkin & Williams Ltd.) The apparatus was then rinsed with methanol and finally several times with distilled water.(For the analysis of steels a 150-ml PTFE beaker was used during the dissolution of the samples.) REAGENTS- All chemicals were of analytical-reagent grade purity and solutions of them were prepared with distilled water from an all-glass distillation apparatus13 and stored in polythene bottles so as to minimise contamination. A stock 1.0 x 1 0 A 3 ~ solution of lead was prepared by dissolving analytical-reagent grade lead nitrate in distilled water, from which a standard 1.0 x M solution was prepared daily. Oxygen-free nitrogen was used throughout this work for de-oxygenating test solutions. RECOMMENDED METHOD FOR THE ANALYSIS OF STEEL SAMPLES- Procedure A-Weigh 1.0 g of steel sample into a PTFE beaker and add 10.0 ml of 40 per cent.nitric acid (for stainless-steel samples, add 5-0 ml of 60 per cent. perchloric acid); heat the mixture until the sample has dissolved, then dilute and filter the resulting solution into a 100-ml calibrated flask, wash the insoluble residue and make the combined filtrate and washings up to volume (solution 1). To a 10-ml aliquot of solution 1, add 5-0 ml of 10 per cent. ascorbic acid solution and 1.0 ml of 0.1 M zinc nitrate solution and dilute the mixture to 100.0 ml in a calibrated flask (solution 2). Carefully rinse the electrolysis cell twice with a few millilitres of solution 2 and then fill the cell to within about 1 cm of the top of the cell (i.e., about 30.0 ml). De-oxygenate the sample solution by bubbling nitrogen through the solution for 5 minutes and then raise the gas bubbler above the solution so as to allow nitrogen to flow over the surface of the solution throughout the remainder of the test in order to maintain an inert atmosphere.Catch a drop of mercury of suitable size on the H.M.D.E. (Fig. 3), place the latter in the cell and connect all electrodes to the relevant terminals. Switch on the rotating glass paddle stirrer (600 r.p.m.) and apply a potential of -0.6 V versus S.C.E. to the H.M.D.E., simultaneously starting a stop-watch in order to time the electrolysis. At the end of the required electrolysis time (15 minutes for steels containing 0.0001 to 0.002 per cent. of lead and 3 minutes for those containing 0.002 to 0.02 per cent. of lead), switch off the stirrer and allow 30 s for the solution to become quiescent.Now record the current - voltage stripping curves, using either of the procedures described below. Procedure B (for steels containing less than 0.2 per cent. of copper and less than 0.02 per cent. ofmolybdenum)-Scan the voltage of the H.M.D.E. from -0.6 to +O-4 V veysus S.C.E. at the rate of 0.4 V min-1, recording the current - voltage curve. Measure the peak height at -0.3 V veyms S.C.E. Procedure C (for steels containing 0.2 to 0.95 per cent. of copper andlor 0.02 to 0.10 per cent. of molybdenum)-Remove the electrolysis cell from under the electrode assembly, discard the test solution, refill the cell with a solution that is 5 X l W 3 ~ in sodium acetate andAugust, 19741 LEAD I N STEEL BY ANODIC STRIPPING VOLTAMMETRY 461 1 x 1 0 - 3 ~ in Zn2+ and contains 0.5 per cent.of ascorbic acid, which has previously been de-oxygenated by bubbling nitrogen through it, and replace the cell under the electrodes (this operation should take not more than 10s). Scan the voltage of the H.M.D.E. from -0.6 to +0.4 V versus S.C.E. at the rate of 0.4 V min-1, recording the current - voltage curve. RESULTS AND DISCUSSION Calibration graphs were prepared for lead in acetate buffer medium but straight lines were not obtained with the use of either distilled or de-ionised water (Fig. 4). It was thought that the curvature was due to the existence of a lead complex, and all solutions were acidified in the hope that such a complex would not be formed in acidic media.Although this treat- ment had no effect, the addition of both acid and zinc ions did result in the production of straight-line calibration graphs (Fig. 4) (typical stripping peaks are shown in Fig. 5). Lead/M x Fig. 4. Calibration for lead pure aqueous solutions. T d (electrolysis time), 3 minutes; E , (electrolysis potential), - 0.6 V versus S.C.E. ; and VB (voltage scan rate), 0.4V min-1. A, HC10, + Zn2+; B, distilled water + CH,COONa or HC10,; and C, de-ionised water + CH,COONa or HC10,. I, is peak current This effect is not fully understood but there are two possible explanations: (i) in distilled water and certainly in de-ionised water, minute amounts of complexing agents are present, which are capable of forming strong complexes with lead even in acidic solution.The lead complex, once formed, is not reduced at the H.M.D.E. in the available potential range, and only when there is an excess of free lead ions in solution is a peak observed at -0.3 V versus S.C.E. This result explains why the effect is more marked in de-ionised water, which contains additional complexing agents washed over from the ion-exchange resin during purification. It is assumed that zinc forms a stronger complex than lead with the impurity and that it is an effective releasing agent. Therefore, provided sufficient zinc is added to the test solutions, any lead present will exist as free ions and a straight-line calibration graph will be obtained. (ii) The solutions contain a surface-active impurity, which is adsorbed on to the surface of the H.M.D.E.and thus prevents effective reduction of lead ions until a certain concentration of lead is reached. This inference could also explain the more marked effect in de-ionised water, which is more likely to contain organic material from the ion-exchange resin that may be adsorbed on to the H.M.D.E. In this instance it is assumed that the zinc ion is preferentially adsorbed on to the H.M.D.E. and does not prevent efficient electro-reduction of lead ions.462 METTERS AND COOKSEY: THE DETERMINATION OF TRACE AMOUNTS OF [A%i!&jst, VOl. 99 E versus S.C.E./V Fig. 5. Typical stripping peaks for lead. Electrolysis time, 15 minutes, electrolysis potential, - 0-6 V ueysus S.C.E. and voltage scanning rate, 0.4 V min-l. Solutions contained 1.0 g 1-l of iron (prepared from pure iron powder), 1.0 X M of Zn2+ and 0.5 per cent.of ascorbic acid, together with the equivalent of lead, per cent. : 1, 0.0; 2, 0.0004; 3, 0.0008; 4, 0.0012; and 5, 0.0016; the peak a t S O - 1 V vew'sus S.C.E. is due to copper contaminationAugust, 19741 LEAD I N STEEL BY ANODIC STRIPPING VOLTAMMETRY 463 In view of the above results, both acidification of and the addition of zinc ions to all solutions are recommended as a precautionary measure in the determination of lead. CALIBRATION GRAPHS IN THE PRESENCE OF IRON- Polarographic analysis of steel samples is not possible with solutions containing iron( 111) because of the large cathodic residual current that begins at zero applied voltage from the reaction, FeS+ + e- + Fez+ Iron(II), however, is not reduced at a mercury electrode until a potential of -1.4 V vmus S.C.E.in acidic solution is reached and, therefore, the interference of iron(II1) is easily overcome by chemical reduction with ascorbic acid. Calibration graphs for two ranges of lead concentration were obtained in solutions containing 1.0 g 1-1 of iron to which ascorbic acid had been added [Fig. 6 (a and b ) ] . The results in Fig. 6 (b) are similar to those in Fig. 4 and it is apparent that the presence of iron and ascorbic acid has no effect on the electro-deposition or subsequent stripping of lead at the H.M.D.E. .. .. * (1) .. 0.1 0 0.08 a 2 0.06 \a 0-04 0.02 0 Lead/M x 10-8 0.20 0.1 6 0.1 2 0.08 0.04 C Lead/M x lo-* 1 i ” I I I I I I I I 0.0004 0.0012 0.0020 0.004 0,012 0.020 Lead, per cent.Lead, per cent. Fig. 6. M of Zna+ and 0-5 per cent. of ascorbic acid. Calibrations for lead in the presence of 1.0 g 1-1 of iron. (a) Solutions contained 1.Og 1-1 of iron, 1.0 x Electro- lysis time, 15 minutes. (b) Solutions as for (a). Electrolysis time, 3 minutes INTERFERENCES- The first step in most analytical methods is the preparation of a suitable sample solution and, in steel analysis, it is accomplished by the addition of an acid or a mixture of acids. The choice of acids in a procedure for the determination of lead by anodic stripping voltam- metry is restricted to either nitric acid or perchloric acid. Hydrochloric acid cannot be used because in chloride media the stripping peak for lead is considerably less than that in other media* and the sensitivity is reduced owing to the merging of the lead peak with the oxidation current of mercury.1° A similar effect has been observed in this laboratory when using sulphate media.The complex composition of steel also indicates that many other metal ions will be present in the sample solution. Additions of manganese, cobalt, aluminium, barium and calcium up to a concentration of 1 x M, and of nickel and chromium up to 5 x M, to a solution that is 4 x ~O-’M in lead and contains 1.Og1-1 of iron and 0.5 per cent. of ascorbic acid had no effect on the stripping peak for lead. The elements that would be expected to interfere are those which are reduced at the H.M.D.E. at -0.6 V versus S.C.E. or before this value is reached, resulting in peaks that merge with the peak for lead or con- tribute significantly to the background current.The most important of these elements that occur in steel samples are molybdenum, tin and copper, and the interference caused by each of them was examined in detail.464 METTERS AND COOKSEY: THE DETERMINATION OF TRACE AMOUNTS OF [Autalyst, Vol. 99 Interference due to co$$er-The electro-reduction of Cu2+ ions at the H.M.D.E. occurs at about +0.05 V zlersus S.C.E., indicating that at -0.6 V versus S.C.E. any copper present in the sample solution will contribute to the background current. When the same medium was used for both electrolysis and stripping, copper at a concentration of only 3 x 1 0 - 5 ~ (0.19 per cent. of copper in a solution containing 1.0 g 1-1 of steel) could be tolerated (Fig.7). However, as previously mentioned, the technique of medium exchange has been used before to overcome such problems and Fig. 8 shows the effect of increasing copper concen- tration on the stripping voltammogram of a 1 x 1 0 + ~ solution of lead (0.0002 per cent. of lead), using a de-oxygenated 1 x M solution of zinc that was 1 x 10-1 M in hydrogen ions and contained 0.5 per cent. of ascorbic acid as the stripping medium. The tolerance limit for copper concentration has now been raised to 1 x l o - 4 ~ (0-63 per cent. of copper). I I I I I 1 I E versus S.C.E./V 0 -0.1 -0.2 -0.3 -0.4 -0.5 -0.6 Y,,,,, 0 -0.1 -0.2 -0.3 -0.4 -0.5 -C E versus S.C.E./V Fig. 7. Interference due to copper (direct). Stripping peaks for a solution containing the equivalent of 0.004 per cent.of lead in a steel in the presence of increasing amounts of copper. The solutions contained 1.0 g 1-1 of iron, 0-5 per cent. of ascorbic acid and 1.0 x M of ha+. Electrolysis time, 15 minutes, electrolysis potential, - 0.6 V versus S.C.E. and scanning rate, 0.4 V min-l. All four solutions contained 0.004 per cent. of lead, with B, +0.063; C, +0.189; and D, + 0.252 per cent. of copper Fig. 8. Interference due to copper (medium exchange into 0.1 M nitric acid). Stripping peaks for a solution containing the equivalent of 0.0004 per cent. of lead in a steel in the presence of increasing amounts of copper. The solutions con- tained 1.0 g 1-1 of iron, 0.5 per cent. of ascorbic acid and 1.0 x M of h a + . Electrolysis time, 15 minutes, electrolysis potential, -0.6 V versus S.C.E.and scanning rate, 0.4 V min-l. All four solutions contained 0.0004 per cent. of lead, with B, +0.063; C, $0.63; and D, +1.26 per cent. of copper The sharp rise in anodic current at about 0.0 V veysus S.C.E. is due to the stripping peak of copper, but the large cathodic current at -0.4 V versus S.C.E., which increases with copper concentration, is not so easily explained. An increase of this magnitude in the background current is often attributed to the electro-reduction of hydrogen ions, but in these solutions with a concentration of these ions of only 0.1 M this reaction does not usually occur until about -1.0 V versus S.C.E. is reached. As a test of this hypothesis, stripping voltammograms were obtained for solutions that were 1 x 10-8 M in lead (0.0002 per cent.of lead) with a single concentration of copper (3.5 x l o - 4 ~ ; 2.2 per cent. of copper), using a stripping medium of different pH in each instance. The results of this test are shown in Fig. 9, and it can be clearly seen that the cathodic background current is reduced as the pH of the stripping medium is increased.August, 19741 LEAD I N STEEL BY ANODIC STRIPPING VOLTAMMETRY 465 This large increase in the cathodic background current can therefore be ascribed to the reaction- .. - - (2) H++e--+iJH, .. .. .. The overpotential for this reaction is about -0.8 V at a mercury electrode, but is considerably less, about -0.4 V, at a copper electrode. In solutions containing large concentrations of copper that are electrolysed at -0.6V versus S.C.E., the copper concentration inside the mercury drop will be high because of the reaction- ..* - (3) Cu2+ + 2e- -+ (CuOHg) . . .. Therefore, it is likely that the H.M.D.E. now behaves as a copper amalgam electrode and reaction (2) is catalysed by the presence of copper. In order to determine any loss in peak current for lead as a result of the presence of copper, stripping peaks were measured for a solution that was 1 x ~ O - ' M in lead and con- tained copper at various concentrations, using an acetate buffer medium for the stripping stage. The extent of copper interference on the stripping peak for lead, which is shown in Fig. 10, becomes significant at concentrations of copper above 1.5 x M (approximately 0.95 per cent. of copper in a solution containing 1.0 g 1-I of steel).This level is a considerable improvement on the previous tolerance limits for copper, but solutions containing copper above this concentration can be analysed if a new calibration graph is prepared with copper present. However, it can be concluded that steel samples that contain up to 0.95 per cent. of copper, which is sufficiently high for most types of steels, can be analysed successfully. 0 -0.1 -0.2 -0.3 -0.4 -0.5 -0.6 E versus S.C,E./V Interference due to copper (medium exchange pH variation). Stripping peaks for a solution containing the equivalent of 0.0004 per cent. of lead and 2.2 per cent. of copper in a steel. The stripping was carried out in solutions at various pH values. Sample solutions and electrolysis conditions as described in Fig.7 Fig. 9. 0 0.63 1.26 1.89 2.52 3.15 Copper, per cent. Fig. 10. Interference due to copper (medium exchange). The effect of copper content on the stripping peak of a steel sample containing 0.002 per cent. of lead. The stripping was carried out in a sodium acetate solution (pH 4.3) by using procedure C . Sample solutions and electrolysis conditions as described in Fig. 7) Interference due to molybdenum-Polarographic waves due to the reduction and oxidation of various ionic states of molybdenum also occur before -0-6 V versus S.C.E. is reached and therefore contribute to the background current. Stripping peaks should not be obtained, however, as molybdenum cannot be plated into a mercury cathode and in acidic media gives soluble reduced species.14 However, according to Lagrange and Schwing,14 molybdenum( VI) at pH 5 is reduced at a mercury electrode to a solid product, Mo02.2H20, which can then be determined by anodic stripping.466 METTERS AND COOKSEY: THE DETERMINATION OF TRACE AMOUNTS OF [Analyst, Vol.99 When molybdenum was added to solutions containing iron, zinc and ascorbic acid, and anodic stripping analysis carried out for lead, interference occurred at 2 x M molybdenum concentration (0.02 per cent. of molybdenum), which consisted of both polarographic waves and stripping peaks (Fig. 11). When medium exchange was used, the limit of interference was raised to 1 x 1 0 - 5 ~ concentration (Fig. 12), which is equivalent to approximately 0.1 per cent. of molybdenum in solutions containing 1.0 g 1-1 of steel, and this concentration of molybdenum is therefore the maximum that can be tolerated in a steel sample; it should be sufficient in most instances except for stainless steels, which often contain more molybdenum.I l l -0.1 -0.2 -0.3 -0.4 -0.5 -( E versus S.C.E./V Fig. 11. Interference due to Fig. 12. Interference due to molybdenum molybdenum (direct). Stripping (medium exchange). Stripping peaks for a solution peaks for a solution containing the containing the equivalent of 0.0004 per cent. of equivalent of 0.0004 per cent. of lead in a steel in the presence of increasing amounts lead in a steel in the presence of of molybdenum. The stripping was carried out increasing amounts of molybdenum. in a sodium acetate solution (pH 4.3) by using Sample solutions and electrolysis procedure C.Sample solutions and electrolysis conditions as described in Fig. 7. conditions as described in Fig. 7. All three solu- All three solutions contained 0.0004 tions contained 0.0004 per cent. of lead, with B, per cent. of lead, with B, $0.02; +0.1; and C, $0.2 per cent. of molybdenum and C, $0.05 per cent. of molyb- denum Interference due to ti%-Tin is reduced at a mercury electrode in acidic media at about -0-45 V versus S.C.E. and has been determined in steel samples by Phillips and Shainll who used anodic stripping voltammetry. The presence of tin in sample solutions, by increasing the background current and giving a stripping peak that merges with the peak for lead, will therefore interfere. This interference was found to occur when tin at the low concentration of 5 x ~ O - ' M was added to test solutions just prior to electrolysis, a large peak resulting at -0.42 V veysus S.C.E.even when medium exchange was employed. However, when the same amount of tin was added to the steel sample at the dissolution stage and the usual analysis carried out, no such peak occurred. Additions of tin in the same manner at concentrations up to 1 x 1 0 - 5 ~ (equivalent to 0.1 per cent. of tin on a 1-0 g 1-1 steel sample) caused no interference. The most likely explanation of this behaviour is that the added tin is precipitated in the nitric acid medium at the dissolution stage, together with any silicon and carbon present in the steel sample, which is then filtered off. Thus, it can be concluded that tin does not interfere. ANALYSIS OF STEEL SAMPLES- A wide variety of steel and cast-iron samples were analysed by using procedure A and the results are shown in Table I.The electrolysis time and mode of stripping (procedureAugust, 19741 LEAD IN STEEL BY ANODIC STRIPPING VOLTAMMETRY TABLE I ANALYSIS OF STANDARD STEEL SAMPLES 467 Lead, per cent. 1 Certificate Sample value Found BCS 330 . . .. .. .. 0.003 0.0022 BCS 328 . . .. .. .. 0-015 0.014 BCS 334 . . .. .. * . 0*0011 0*0010 BCS 335 . . .. ,. .. 0.0015 0-00 13 Cast iron 1 . . .. .. .. 0*0028* 0.0022 2 .. . . .. . . 0-014* 0.013 3 .. .. .. . , 0.0022* 0.0024 4 .. .. .. .. 0.0038* 0-0038 5 .. .. . . .. 0.0018* 0.00 16 6 .. .. .. .. 0*007* 0.0073 BCS 328 . . . . .. . . 0.0 15 0-015 BCS 326 . . .. .. .. 0.014 0-013 Cast iron (A) .... .. 0*0018* 0.0019 .. .. .. 0.0004* 0-0004 0.0017 0*0004 (B) Cast iron (A) + 0.1 per cent. of tin Cast iron (B) + 0-1 per cent. of tin 0.0018 0.0004 * BCIRA values. Procedure Direct T d l minutes 15 3 15 15 15 3 15 15 15 3 3 16 16 16 16 Medium exchange 3 B or C) chosen depended on the sample composition. As can be seen from Table I the results compare satisfactorily with the certificate values or, for the cast-irons, with the results obtained in the BCIRA laboratory. The time taken for a single analysis depends on the lead content of the sample, but even at the lowest levels of lead, it is possible to complete the analysis together with a standard addition, if required, in 1 hour, a considerable saving in time over the methods in current use, which can take several hours to complete.A further advantage is that the method is direct, no solvent extractions or chemical operations being necessary other than sample dissolution. The risk of contamination and losses by adsorption or in the solvent extraction step is therefore virtually eliminated, resulting in greater accuracy. The precision of the method is demonstrated in Table I1 and the measured coefficient of variation (8 per cent.) should be acceptable for such a low level of lead. The detection limit of the method (Table 111) was calculated to be 1 p.p.m. (1 x per cent. of lead) in the steel sample. TABLE I1 REPRODUCIBILITY TEST Lead, per cent. BCS 328 (0.015 per cent. of lead) 0.013 0.0125 0.012 0.0125 0.0145 0-014 0.0115 0.01 3 0.0135 0.015 0.014 0-0115 0.0135 0.0135 0.0125 Mean .. . . . . 0.013 Coefficient of variation Standard deviation . . 0.001 8 per cent. Cast iron (A) (0.0018 per cent. of lead) 0.00 19 0.0020 0.0019 0.0017 0.0019 - - - - - - - - - - 0.0019 6 per cent. 0~0001468 METTERS AND COOKSEY Interferences to the method are few, only those of copper and molybdenum being of any significance. The maximum allowable level of molybdenum in the steel sample was found to be 0.1 per cent. With copper, up to about 0.2 per cent. in the steel can be tolerated, but this level is raised to 0.9 per cent. when medium exchange is used. The medium exchange step is not difficult to apply, but some care is necessary in ensuring that the solution transfer is carried out as quickly as possible. Anodic stripping voltammetry has been applied pre- viously t o the determination of copper12 and tinll in steels and there is no reason why steel could not also be analysed for other amalgam-forming elements, e.g., bismuth, antimony, etc.If these elements were present in the sample at suitable concentrations it may even be possible to determine them simultaneously. TABLE I11 Test solutions contained 1.0 g 1-1 of iron and 0.5 per cent. of ascorbic acid, Electrolysis potential, -0.6 V veYsus S.C.E. ; electrolysis time, 15 minutes; and rate of voltage scan, 0-4 V min-1 DETECTION LIMIT FOR THE DETERMINATION OF LEAD IN STEEL and were 0.1 M in H+ and 1 x M in Zn2+ ions Lead added, per cent. Peak current/pA 0.0 0.0025, 0.0035, 0*0030 0.5 x 10-4 0.0050, 0.0040, 0*0060 1.0 x 10-4 0.0050, 0.0060, 0.0060 1.5 x 10-4 0.0070, 0.0085, 0.0060 2.0 x 10-4 0*0090, 0.0110, 0~0100 Variation about the calibration graph corresponds to a standard deviation of 0.29 x per cent. of lead. :. Detection limit = 4s = 1.16 x loe4 per cent. of lead (1.16 p.p.m.) One of us (B.M.) thanks the Shotton Works of the British Steel Corporation for financial support during this work, as well as Mr. A. E. Burgess, Glasgow College of Technology, for some useful discussion. 1. 2. 3. 4. 5. 6. 7. 8. 9. 10. 11. 12. 13. 14. REFERENCES Sandell, E. B., “Colorimetric Determination of Traces of Metals,” Volume 111, Interscience Postlethwaite, R. T., Kidman, L., Bagshawe, B., Bills, K. M., Harrison, T. S., and Wattsmith, Chakrabarti, C. L., Robinson, J. TV., and West, P. W., Analytica Chim. Acta, 1966, 34, 269. Dagnall, R. M., and West, T. S., Talanta, 1964, 11, 1553. Hofton, M. E., and Hubbard, D. P., Analytica Chim. A d a , 1970, 52, 425. Barendrecht, E., “Stripping Voltammetry,” in Bard, A. J ., Editor, “Electroanalytical Chemistry,” Kublik, Z., Acta Chim. Hung., 1961, 27, 79. Sinko, I., and Dolezal, J., J . Electroanalyt. Chem., 1970, 25, 299. Ariel, M., and Eisner, U., Ibid., 1963, 5, 362. Ariel, M., Eisner, U., and Gottesfield, S., Ibid., 1964, 7, 307. Phillips, S. L., and Shain, I., Analyt. Chem., 1962, 34, 262. Gottesfield, S., and Ariel, M., J . Electroanalyt. Chem., 1965, 9, 112. Bishop, E., and Sutton, J. R. B., Analytica Chim. Acta, 1960, 23, 8. Lagrange, P., and Schwing, J. P., Analyt. Cheuvt., 1970, 42, 1844. Publishers, New York, 1965, p. 555. J. A., J . Iron Steel Inst., 1970, 208, 500. Volume 2, Marcel Dekker Inc., New York, 1967, p. 53. Received November 27th, 1973 Accepted February 5th, 1974

ISSN:0003-2654    

DOI:10.1039/AN9749900457

出版商:RSC    

年代:1974

数据来源: RSC

摘要:

Analyst, August, 1974, Vol. 99, @. 469470 469 A Modification to the Extraction = Atomic- absorption Method for the Determination of Antimony, Bismuth, Lead and Tin BY K. THORNTON (Henry Wiggin and Co. Ltd., Holmer Road, Hereford) AND KEITH E. BURKE (The International Nickel Company, Inc., Paul D. Merica Research Laboratory, Sterling Forest, Suffern, New York 10901, U.S.A.) It has been shown that the use of organometallic compounds for the calibration procedure used in the extraction - atomic-absorption determination of antimony, bismuth, lead and tin in metallurgical materials leads to erroneous results at low concentration levels. Results are presented, which demonstrate that the technique is satisfactory when calibrated with solutions that have been taken through the extraction procedure, THE previously published analytical method for the determination of trace amounts of antimony, bismuth, lead and tin1 depends on the extraction of the iodides of these elements into a solution of tri-n-octylphosphine oxide (TOPO) in 4-methylpentan-2-one.The extract is then nebulised directly into the atomic-absorption flame. The calibration procedure involves the use of solutions of organometallic compounds in 4-methylpentan-2-one. It is shown in Table V of the original publication1 that agreement between this procedure and two other atomic-absorption techniques is generally acceptable at higher concentration levels of the elements determined. Furthermore, the accuracy is shown to be satisfactory at these higher levels for the analysis of standardised samples (Table VI of the original publication).At lower concentration levels, particularly below 10 p.p.m., the agreement between the three techniques is poor and it is difficult to assess the accuracy of the extraction procedure as no certified standards are available at these levels. Subsequent work has shown that the calibration procedure that makes use of organo- metallic compounds dissolved in 4-methylpentan-%one gives rise to erroneous results, the values being, in general, higher than the true values. However, a simple modification, in which aqueous solutions of the elements determined are carried through the extraction pro- cedure and treated in the same way as the sample solutions, overcomes this difficulty. This modification avoids any error that might be introduced by variation in the extraction con- ditions or errors due to light scattering, molecular absorption or organometallic bonding.The revised procedure has been applied to several of the samples included in Table VI of the original publication. Originally, two alternative atomic-absorption techniques were used for comparison purposes. The comparison procedure with a simple 4-methylpentan-2-one extraction is subject to the same problems that have been encountered with the proposed method, with TOPO in solution in 4-methylpentan-2-one, whereas the separation technique based on co-precipitation with manganese dioxide is inherently less sensitive and consequently inaccurate at levels below about 10 p.p.m. In an effort to overcome these difficulties alter- native analytical procedures have been developed that involve solvent-extraction separations followed by square-wave polarographic determination.Lead was extracted as the iodide complex in a method similar to that proposed by Luke,3 and bismuth, tin and antimony were extracted as their complexes with isoocty1thiog1ycollate.* The results obtained by using the modified extraction - atomic-absorption procedure and the polarographic methods are given in Table I. @ SAC and the authors.470 THORNTON AND BURKE TABLE I DETERMINATION OF ANTIMONY, BISMUTH, LEAD AND TIN IN COMMERCIALLY AVAILABLE MATERIALS Found, p.p.m. A r 3 Designation HW E-3923 HW E-3924 HW E-3925 HW E-3926 HW E-3927 HW E-3928 HW E-3929 HW E-3930 HW E-3931 HW E-3932 HW B-7047 HW B-7048 HW B-7049 HW B-7050 HW B-7051 BCS 310 BCS 310/1 BCS 387 BCS 371 NBS 671 NBS 672 HW F-292 HW F-293 HW F-294 HW F-295 HW F-296 HW F-297 Matrix TINCOLOY alloy 800 INCOLOY alloy 800 INCOLOY alloy 800 INCOLOY alloy 800 INCOLOY alloy 800 INCOLOY alloy DS INCOLOY alloy DS INCOLOY alloy DS INCOLOY alloy DS INCOLOY alloy DS TINCONEL alloy X750 INCONEL alloy X750 INCONEL alloy X750 INCONEL alloy X750 INCONEL alloy X750 TNIMONIC alloy 90 NIMONIC alloy 90 NIMONIC alloy 901 Commercial nickel Nickel oxide Nickel oxide Nickel Nickel Nickel Nickel Nickel Nickel Antimony 4+ A* 6 13 11 12 13 12 9 7 9 9 1 <1 <1 <1 <1 4 4 3 1 <l <1 2 <1 <1 <1 <1 <1 B* 4 13 12 12 13 12 11 9 10 11 1 1 1 1 1 5 5 1 <1 <1 <1 <1 (1 <1 <1 - - Bismuth - A B <0.5 <Om5 <Om5 <0*5 (0.5 <0*5 <0.5 <0.5 <0*5 <0*5 <Om5 <0.5 <0.5 <0.5 <O-5 <0*5 (0-5 ( 0 .5 t 0 . 5 <0.5 <0.5 <0*5 <0*5 <0*5 <0*5 <0*5 <0.5 ( 0 . 5 <0.5 <Om5 (0.5 ( 0 . 5 (0.5 <0*5 <O-5 <0*5 <On5 (0.5 (0.5 <0.5 (0.5 <0*5 <O-5 <On5 <Om5 <0*5 <0*5 <0*5 <0*5 (0.5 ( 0 . 5 (0.5 (0.5 - Lead 4+ A 1.0 1.2 1.0 0.8 1.0 0.9 1.0 1.0 1.0 1.0 1.5 1.2 1.2 1.3 1-3 5.7 0.5 15 14 14 35 0.8 0.8 1.4 0.9 1.1 0.3 B 1.7 1.2 1.8 1.4 1.4 1.4 0.9 0.9 1.1 1.1 1-5 1.3 1-2 1.6 1-3 5.5 0.7 15 - 16 39 0.8 0.8 1.7 0.5 1.0 0-5 Tin + A 17 20 23 24 20 20 19 19 13 19 7 7 3 3 3 21 35 19 <1 <2 <2 <1 <1 <1 <1 <1 <1 B 25 24 24 23 21 22 17 20 19 19 6 6 6 4 4 24 34 18 3 4 3 <1 <1 <1 <1 <1 - * Laboratory A used the modified TOP0 - atomic-absorption procedure, while laboratory B used t INCOLOY, INCONEL and NIMONIC are registered trademarks of The International Nickel solvent extraction - polarographic methods. Company, Inc. REFERENCES 1. 2. 3. 4. Burke, K. E., AnaZyst, 1972, 97, 19. Leonard, M. A., and Swindall, W. J., Ibid., 1973, 98, 133. Luke, C. L., Analytica Chim. Acta., 1967, 39, 447. Fritz, J. S., Gillette, R. K., and Mishmash, H. E., Analyt. Chem., 1966, 38, 1869. Received July 2nd, 1973 Accepted February 20th, 1974

ISSN:0003-2654    

DOI:10.1039/AN9749900469

出版商:RSC    

年代:1974

数据来源: RSC

摘要:

Analyst, August, 1974, Vol. 99, p$. 471-475 47 1 Spectrophotometric Determination of Magnesium in Tobacco Leaves with Eriochrome Black B BY G. SLEGERS AND A. CLAEYS (Laboratory of Analytical Chemistry, Faculty of Pharmaceutical Sciences, State University of Ghent, 9000 Ghent, Belgium) A spectrophotometric method for the determination of magnesium in tobacco leaves with Eriochrome black B is described. After destruction of the tobacco leaves with a mixture of nitric, perchloric and sulphuric acids, the magnesium is separated from interfering ions by ion-exchange chromatography on Dowex 5OW-X12, 100 to 200 mesh, and determined spectrophotometrically after coupling it with Eriochrome black B. The accuracy of the method was tested against a reference material. VARIOUS methods, such as atomic-absorption spectrophotometryl and spark-source mass spectrometry,2 are applied to the determination of magnesium in tobacco products.In spectrophotometry, several lake-f orming sensitive reagents such as 4- (4-nit rophenylazo) -1 - naphthol3 and thiazol yellow dyes4 have been described for the direct determination of magnesium. It is difficult, however, to obtain reproducible results with these dyes, because of variations in experimental conditions. Eriochrome black T [l-(l-hydroxy-2-naphthylazo)-2-hydroxy-5-nitro-4-naphthalene- sulphonic acid], used by Harvey, Komarmy and Wyatt,5 Gasser6 and Young and Sweet,' is a more suitable reagent. This reagent is sensitive but not as specific as those mentioned above and interferences from calcium, copper, manganese, iron, aluminium, cobalt and nickel have been reported by Diehl, Goetz and Hach.Eriochrome black B [ 1-( l-hydroxy-2-naphthy1azo)- 2-hydroxy-4-naphthalenesulphonic acid] does not have any disadvantages compared with Eriochrome black T and was used in the method proposed in this paper for the determination of magnesium because of its greater sensitivity. EXPERIMENTAL APPARATUS- meter equipped with l-cm glass cells. REAGENTS AND MATERIALS- Standard stock magnesium solution, 46.6 mg I -l-This solution was prepared with analytical-reagent grade magnesium chloride (MgC1,. 6 H,O) and was standardised by the gravimetric pyrophosphate procedure. A 1 + 9 dilution of the stock solution was used in the experiments. Bufer solutions-Buffer solutions (0.1 M) in the pH range 8.5 to 10 were prepared with ammonia solution and ammonium chloride, and a 0.1 M buffer solution with a pH of 10.5 was prepared with piperidine and hydrochloric acid, Dye solution, 0.5 per cent.m/V-Eriochrome black B (0-5 g) was dissolved in methanol and the solution diluted to 100 ml with methanol. Dye solutions were freshly prepared each day and stored in the dark. EFFECT OF pH STUDY- Absorption spectra of the dye - magnesium complex were determined in the pH range 8.5 to 10.5 on solutions prepared by mixing 10 ml of 1 + 9 magnesium stock solution with 10 ml of buffer and 2 ml of dye solution in a 100-ml calibrated flask and diluting the mixture to the mark with water. Blanks were prepared in a similar way, omitting the magnesium solution. The absorption spectra, which were measured immediately, showed that the highest absorbance was obtained at A,,, 559 nm with the solution at pH 10.5.All absorbance measurements were carried out with a Beckman, Acta V, spectrophoto- @ SAC and the authors.472 SLEGERS AND CLAEYS : SPECTROPHOTOMETRIC DETERMINATION OF [Analyst, Vol. 99 The colour intensity also depends on the amount of dye present in the solution. This aspect was investigated by adding increasing amounts of 0-5 per cent. m/V dye solution to a mixture of 10ml of 1 + 9 magnesium stock solution and 10 ml of buffer of pH 10-5 and diluting the mixture to 100 ml. Blanks were prepared in a similar way, with the magnesium solution omitted. The colour intensity remained constant with concentrations of dye above a fifteen-fold excess (Fig. 1).INFLUENCE OF DYE CONCENTRATION- 0.3 L n Lo w W 0.2 2 L m -e 0.1 9 2 C d I I 1 I I 6.4 12.8 19.2 25.6 32.1 Mole ratio, M9 Fig. 1. Influence of dye concentration on the absorption SEQUENCE OF ADDITION OF REAGENTS- As it is possible that magnesium may form different complexes with Eriochrome black T, the order of addition of the reagents is important. Four sequences were therefore investigated for Eriochrome black B at pH 10.5, with mixing of the solution after the addition of each component. For the first sequence 10 ml of 1 + 9 magnesium stock solution, 10 ml of buffer solution and 2 ml of 0.5 per cent. m/V dye solution in methanol were transferred into a 100-ml cali- brated flask and the mixture was diluted to the mark with water; for the second, the order was buffer solution, dye solution, then 1 + 9 magnesium stock solution followed by dilution to 100 ml; for the third, 1 + 9 magnesium stock solution, dye solution, then buffer solution and dilution to 100ml; and for the fourth, 1 + 9 magnesium stock solution, buffer solution, followed by dilution to nearly SO ml, then dye solution and dilution to 100ml.In each instance a blank was prepared in a similar way. From the results (Fig. 2), it can be seen that the third sequence gave the highest absorbance and stability and was used in the following experiments. Absorbances of samples and blanks were measured at intervals at 559 nm. COMPOSITION OF THE COMPLEX- Only one complex is formed between magnesium and Eriochrome black B at pH 10.5, and it was investigated by measuring the absorbance of a series of solutions in which the mole ratio of magnesium to dye reagent was varied between 1 : 10 and 10: 1.For the continuous varia- tion experiments (Fig. 3), the concentrations of both the magnesium and dye solutions were 1.8 x 10 -3 M. Appropriate volumes of the two solutions, totalling 10 ml, were transferred by pipette into 100-ml calibrated flasks, 10 ml of buffer of pH 10.5 were added and the mixtures diluted to 100 ml. Blanks con- taining 1 to 10 ml of dye solution were treated in a similar way. From the absorption curves, the absorbance at Amax. was deducted, as shown in Fig. 3. The value of n, the number of ligands bound per cation, can be obtained from the relationship a = where Xm,,. represents the molar fraction of the dye at the point where the difference curve is at a maximum.With Xma,. equal to 0.66, the value of n is 1.94 or approximately 2 mol of ligand per mole of magnesium. The solutions were then stored in the dark for 1 hour. Xmax. 1 - Xmax.Auwst, 19741 MAGNESIUM I N TOBACCO LEAVES WITH ERIOCHROME BLACK B 473 -I--- 0.2 ~ 120 240 360 480 T irne/minutes Fig. 2. Development and stability of the colour. Sequence of addition of reagents : 1, first sequence; 2, second sequence; 3, third sequence; and 4, fourth sequence Molar fraction of the dye Fig. 3. Composition of the complex. A, Total curve obtained with aliquots of 1.8 x M magnesium and dye solutions totalling 10 ml plus 10 ml of buffer solu- tion (pH 10.5) made up to 100ml and solution stored for 1 hour in the dark (absorbance measured a t 559 nm against water) ; B, curve for 10 ml of 1.8 x lo-' M dye solution plus 10 ml of buffer solution made up to 100 ml and treated as for A; and C, curve obtained by subtracting B from A, giving absorbance of the mag- nesium - dye complex STUDY OF CALIBRATION GRAPH- The analysis was carried out by use of a calibration graph.Because several factors, such as pH and sequence of addition of reagents, influence the absorbance, values for the calibration graph must be measured under constant experimental conditions. From previous experi- ments it was decided to construct the curve by using the third sequence of additions, i.e., 10 ml of 1 + 9 magnesium stock solution followed by addition of 2 ml of 0.5 per cent.m/V dye reagent in methanol, 10 ml of buffer of pH 10.5, and final dilution to 100 ml. The effect of time of storage of the reaction solution in the dark was investigated by measuring the absorbance at various intervals. Because Beer's law is followed in the con- centration range studied (Table I), the relationship y = ax + b can be applied to the calibra- tion graph. The factors a and b are calculated by the method of regression analysis (Table I) and the variation about the fitted curve is calculated from the equation- C(2 - Y)2 0 v, = where CD is the degree of freedom (number of points on the graph - 2) ; 2 the measured absorb- ance; and Y the absorbance calculated from the regression curve. From Fig. 4 it is deduced that the variation is at a minimum after 4 hours' storage in the dark.PROCEDURE FOR THE SEPARATION OF MAGNESIUM- After wet destruction of 1 g of lyophilised tobacco leaves with a mixture of strong oxidis- ing acids consisting of 10 ml of 70 per cent. m/V perchloric acid, 10 ml of 35 N sulphuric acid and 10 ml of 14 N nitric acid, the residual acid mixture was transferred quantitatively into a474 SLEGERS AND CLAEYS SPECTROPHOTOMETRIC DETERMINATION OF [Andyd, VOl. 99 250-ml calibrated flask and diluted to the mark with water; 3 ml of this solution were neutral- ised with ammonia solution and poured into a column, 1 cm in diameter and 14 cm in length, loaded with cation exchanger Dowex 5OW-X12, 100 to 200 mesh, After washing the column with water until the eluate was free from acid, it was eluted with 0.7 11 hydrochloric acid in order to remove sodium, the first 28 ml of eluate being rejected and the sodium collected in the next 100 ml.The column was then eluted with 0.7 M nitric acid, potassium being collected in the next 190 ml of eluate and magnesium in the following 275 ml. Finally, calcium was eluted with 2 M nitric acid and was collected in 150 ml of eluate. Preliminary experiments had shown that the recovery of these elements was quantitative under these conditions. TABLE I CALIBRATION GRAPH WITH MEASUREMENTS AT DIFFERENT STORAGE TIMES Absorbance at 559 nm after Magnesium r A 1 1440 minutes takenlg 1-1 0 minutes 60 minutes 255 minutes 480 minutes 1-36 x 10-4 2-27 x 10-4 4.55 x 10-4 5-91 x 10-4 9-10 x 10-4 1.14 x 10-3 1-36 x 10-3 a b V Y 3.64 x 6.82 x 0.198 0.150 0.390 0.357 0.568 0.512 0-730 0.885 1.073 0.033 0.062 3.0 x 10- 0.128 0,158 0.324 0.365 0.500 0-519 0.738 0.893 1.081 0.035 0.013 3 4.9 x 10- 0.097 0-167 0.310 0.375 0.498 0.552 0.775 0.950 1-126 0.038 -0*010 1.8 x 10- 4 0.085 0.194 0.308 0.403 0.498 0.588 0.805 1.000 1.160 0.040 -0.012 - 4 3.0 x 10-4 0.060 0.246 0.300 0.468 0.568 0.708 0.935 1.176 1.382 0.048 - 0.043 1.2 x 10-3 The eluate fraction containing magnesium was evaporated to dryness, the residue dissolved in 5 ml of 0.1 M hydrochloric acid and the solution transferred quantitatively into a 100-ml calibrated flask and diluted to the mark with water. A 10-ml aliquot of this solution was then mixed with 2 ml of 0.5 per cent.m/V dye solution and 10 ml of buffer solution of pH 10.5 and the mixture diluted to 100 ml.After storing the solution and a blank for 4 hours in the dark, their absorbances were measured at 559 nm. " 300 900 1500 Time/minutes Fig. 4. Slope of the variation about the fitted curve RESULTS AND DISCUSSION Various samples of tobacco leaves were subjected to the described procedure and the The magnesium contents were read off from the calibration graph and corrected for dilution. results are given in Table 11.August, 19741 MAGNESIUM IN TOBACCO LEAVES WITH ERIOCHROME BLACK B 475 In order to check the accuracy of the method, the magnesium content of orchard leaves, a standard reference material SRM 1571 , supplied by the National Bureau of Standards, was determined by the described method. The result, shown in Table 11, agreed with the value given by the National Bureau of Standards.TABLE I1 DETERMINATION OF MAGNESIUM IN TOBACCO LEAVES IMPORTED INTO THE BELGIAN - LUXEMBURG ECONOMIC UNION Brand U. S. A., burley U.S.A., fire cured U.S.A., air cured U.S.A., flue cured Indonesia, air cured Greece, sun cured Rhodesia, flue cured Magnesium content, per cent. mjm 0.41 0.43 0.48 0.38 0.45 0.46 0.46 Brand Magnesium content, per cent. m/m India, flue cured 0.45 Turkey, sun cured 0.43 Malawi, burley, air cured 0.46 Malawi, fire cured 0.46 Belgium, air cured 0.41 Orchard leaves SRM 1571 0.65 NBS value 0.62 f 0-02 Mean error, per cent. 2-0 CONCLUSION A spectrophotometric method has been established for the determination of magnesium in tobacco leaves with an error of &Z per cent. The calibration graph has been studied in the concentration range 1.4 x 10 -4 to 1.3 x 10 -3 g 1-1 of magnesium. However, the method is sensitive enough to permit determinations to be made in the microgram range. The method has been applied to the determination of magnesium in tobacco leaves imported into the Belgian - Luxemburg Economic Union. REFERENCES 1. 2. 3. 4. 5. 6. 7. 8. Pinkerton, A., Aust. J . Exp. Agr. Anim. Husb., 1971, 11, 99. Jones, R. M., Kuhn, W. F., and Varsel, C., Analyt. Chem., 1968, 40, 10. Farhan, F., Mikrochemie Mikrochem. Acta, 1950, 35, 560. Mitchell, T. A., Analyst, 1954, 79, 280. Harvey, A., Komarmy, J., and Wyatt, G., Analyt. Chem., 1953, 25, 498. Gasser, J. K. R., Analyst, 1955, 80, 482. Young, A., and Sweet, T. R., Analyt. Chem., 1955, 27, 419. Diehl, H., Goetz, C., and Hach, C., J . Amer. Wat. WAS Ass., 1950, 42, 40. Received January 29th, 1974 Accepted March 7th, 1974

ISSN:0003-2654    

DOI:10.1039/AN9749900471

出版商:RSC    

年代:1974

数据来源: RSC

摘要:

476 Analyst, August, 1974, Vol. 99, ~59. 476-481 The Spectrophotometric Determination of Ampicillin and Cloxacillin in Combined Injections* BY A. G. DAVIDSON AND J. B. STENLAKE (Department of Pharmaceutical Chemistry, University of Strathclyde, Glasgow, G1 1X W ) A method is described for the determination of ampicillin and cloxacillin in injections. Ampicillin is determined by an absorbance difference technique based on the higher absorbance of ampicillin a t 268 nm in a solution a t pH 5 than in one at pH 9. Cloxacillin is determined by measurement of the absorb- ance a t 275 nm and the application of a small correction for the absorbance of ampicillin. The accuracy, precision and specificity of the method are dis- cussed. The analytical results obtained for commercial samples of ampicillin - cloxacillin (I + 1 and 2 + 1) mixtures are compared with those obtained by microbiological assay.THE methods that are currently used to determine penicillins in pharmaceutical preparations have been reviewed.l They include measurement of the absorbance at 320 to 360 nm of the copper-stabilised penicillenic acid, formed by isomerisation of the penicillin in acidic solution, iodimetric assay after alkaline cleavage of the p-lactam ring and microbiological assay. The non-specificity of these methods, however, renders them unsuitable for the analysis of mixtures of different penicillins, and separation of the individual penicillins is usually necessary prior to their determination. One mixture of penicillins available commercially contains ampicillin sodium and cloxa- cillin sodium in 1 : 1 and 2: 1 proportions, as the free acids.These preparations are supplied in vials as a dry powder for injection after reconstitution with water. The component penicillins have been determined microbiologically following separation on a column of basic acetate resin2 and after separation by a gel-electrophoretic method based on that of Lightbown and de Rossi3 (D. Sykes, personal communication). A chemical determination* of ampicillin and cloxacillin has been described in which ampicillin is determined colorirnetrically after reaction with ninhydrin, and cloxacillin is determined by measurement of the absorbance at 346 nm, 12 minutes after dissolving the sample in hydrochloric acid solution. In the present work we have determined both penicillins by the spectrophotometric analysis of solutions prepared from a single sample weighing.Ampicillin is determined by difference spectrophotometry because at 250 to 275 nm the absorbance of ampicillin is greater in solution at pH 5 than at pH 9, while the absorbance of cloxacillin is unchanged by variation in pH. Thus, the absorbance at 268 nm of a solution of the penicillin mixture in pH 5 buffer, read against a solution of the mixture of equal concentration in pH 9 buffer, is proportional to the concentration of ampicillin. Cloxacillin is determined by measurement of the absorb- ance of the sample at 275 nm and correction for the absorbance of ampicillin. EXPERIMENTAL SPECTROPHOTOMETER- Absorbance values were measured in l-cm cells that were matched for equal path length on a Unicam SP1800 spectrophotometer working in the double-beam mode.The most sensitive absorbance range, 0 to 0.2, was used for measurement of the absorbance differences of ampicillin, and the range 0 to 1.0 was used for the measurement of cloxacillin absorbance. The slit-width, 0.5 mm, remained constant throughout the assay. REAGENTS- Ampicillin sodium and cloxacillin sodium reference compounds-These were kindly supplied by Beecham Pharmaceuticals, Worthing, Sussex. The British Pharmacopoeia1 assay results, calculated as the free acids, supplied by the manufacturer were 91.3 per cent. (C,,H,,N,O,S) and 89.8 per cent. (C,,H,,ClN,O,S), respectively. * The preliminary results of this work were presented a t the British Pharmaceutical Conference, @ SAC and the authors.September 1973, London.DAVIDSON AND STENLAKE 477 “Double-strength” pH 9 bNfer-Dissolve potassium chloride (7-46 g), boric acid (6.18 g) “Double-strength” pH 5 bufer-Dissolve sodium acetate trihydrate (38.10 g) and glacial “Sin@-strength” pH 5 bufer-Dilute double-strength pH 5 buffer with an equal volume and 1 N sodium hydroxide solution (41.6 ml) in water and dilute the solution to 1 litre. acetic acid (7.2 g) in water and dilute the solution to 1 litre. of water. PROCEDURE AMPICILLIN- Dissolve the ampicillin - cloxacillin mixture (about 65 mg of the 1 + 1 mixture or about 48 mg of the 2 + 1 mixture, accurately weighed) in water and make the volume up to 25 ml. By use of a pipette transfer 10-ml volumes into two flasks, one containing double-strength pH 5 buffer (10 ml) and the other double-strength pH 9 buffer (10 ml). Immediately measure the absorbance of the pH 5 buffered solution with the pH 9 solution in the reference cell at the difference maximum, which is at a wavelength of about 268 nm.Measure the absorbance difference of a similarly prepared solution of ampicillin sodium (about 32 mg, accurately weighed). CLOXACILLIN- Measure the absorbance of the pH 5 buffered solution of the sample (prepared above) at 275 nm with single-strength pH 5 buffer in the reference cell and, similarly, measure the absorbance of the standard solution of ampicillin sodium in pH 5 buffer (also prepared above). Then measure the absorbance of a standard solution of cloxacillin sodium (about 32 mg in 50 ml for the 1 + 1 mixture or 32 mg in 100 ml for the 2 + 1 mixture) in single-strength pH 5 buffer. CALCULATION- The concentration (c) of ampicillin (C,,H,,N,O,S) in the sample (per cent m/m) is given by AA,,,(sample) c = X AA 268( st andard) amount of ampicillin sodium standard (mg) amount of sample (mg) x purity of ampicillin sodium standard (per cent.) where AA,,, is the absorbance difference at 268 nm, and the percentage purity of the standard is expressed in terms of the free acid. By using the calculated concentration of ampicillin in the sample and the absorbance at 275 nm of the standard ampicillin solution, calculate the contribution of ampicillin to the absorbance at 275 nm of the sample solution.Then calculate the cloxacillin concentration from the corrected absorbance at 275 nm of the sample solution, using the appropriate ratios and amount of standard in a similar equation.RE s u LTS EFFECT OF pH ON THE ABSORBANCE OF AMPICILLIN AND CLOXACILLIN- The ultraviolet absorption spectra of ampicillin and cloxacillin in pH 5 and pH 9 buffers are shown in Fig. l(a) and (c). Fig. l ( b ) shows the absorbance difference spectrum of ampi- cillin, obtained for the solution buffered a t pH 5 with a solution of identical concentration in pH 9 buffer in the reference cell. Peaks occur in the difference spectrum at wavelengths of 256, 262 and 268 nm. The effect of pH on the absorbance of ampicillin and cloxacillin was further investigated by measuring the absorbance at 268 nm of standard solutions of the penicillins in various buffers in the pH range 3 to 10 (Fig.2). The absorbance of cloxacillin is constant in this pH range, while ampicillin shows maximum absorbance below pH 5 and minimum absorbance above pH 9. The mid-point of the sigmoid curve for ampicillin occurs at a pH of 7.30, which is in agreement with the pK, value of ampicillin of 7 ~ 2 5 . ~478 DAVIDSON AND STENLAKE : THE SPECTROPHOTOMETRIC DETERMINATION [ANaZyst, Vol. 99 ACCURACY OF THE METHOD- The accuracy of the method was investigated by analysing several standard mixtures of ampicillin sodium and cloxacillin sodium. The results are given in Table I. PRECISION- In order to determine the precision of the method ten replicate analyses were made on each of a 1 + 1 and a 2 + 1 ampicillin - cloxacillin mixture.The relative standard deviations and limits of error for a single determination (P = 0-95) for each mixture are recorded in Table 11. 250 260 270 280 250 2 60 270 Wavelength/n m 1 .o 0.8 a, c m 0.6 =e v) ' 0.4 0.2 0 I 260 270 280 290 Fig. 1. Ultraviolet spectra of ampicillin and cloxacillin. (a) Ultraviolet spectra of ampicillin (0.06 per cent. m/V) in pH 5 (- - -) and pH 9 (-) buffer. (6) Difference spectrum of ampicillin (0.06 per cent. m/ V ) . (c) Ultraviolet spectra of cloxacillin (0.08 per cent. m/V) in pH 5 and pH 9 buffer. At these pH values the graphs are superimposed STABILITY OF THE PENICILLINS- The extent of hydrolysis of the penicillins under the conditions of the assay was investi- gated. Aliquots of the standard solutions of ampicillin and cloxaxillin at pH 5 and pH 9, alone and combined, were examined for unhydrolysed penicillins, 30 minutes after preparation of the solutions, by the iodimetric method of Finholt, Jurgensen and Kristiansen.6 The results are given in Table 111.3 4 5 6 7 8 9 10 PH Fig. 2. The effect of pH on the absorb- ance (1 per cent., 1 cm) of ampicillin (A) and cloxacillin (B) a t 268 nmAugust, 19741 OF AMPICILLIN AND CLOXACILLIN IN COMBINED INJECTIONS 479 TABLE I DETERMINATION OF AMPICILLIN AND CLOXACILLIN, AS THE FREE ACIDS, IN STANDARD MIXTURES OF AMPICILLIN SODIUM AND CLOXACILLIN SODIUM Added, per cent. mlm 45.6 44.5 45.6 46.4 35.1 85.3 61.8 59.6 60.3 61.0 50.1 41.0 Found, per cent. mlm 45.2 43.9 45-2 45.8 35.8 55.3 62.1 59.2 60.7 60.8 50.2 40.4 Ampicillin 7 Recovery, per cent.99.1 98.7 99.1 98.7 102.0 100.0 100.5 99.3 100.7 99.7 100*2 98.5 Added, per cent. mlm 44.9 46.0 44.9 44.2 55.3 35.4 28.9 31.1 30.4 29-7 40.5 49.5 Cloxacillin Found, per cent. m/m 44-9 45.4 44.4 44.3 55-1 35.8 28.6 31.5 30-2 29.3 40-5 49.6 A 1 Recovery, per cent. 100.0 98.7 98.9 100.2 99.6 101.1 99.0 101.3 99.3 98-7 100~0 100.2 ADHERENCE TO BEER’S LAW- In one series the amount of cloxacillin sodium was constant at 32 5 1 mg and that of ampicillin sodium ranged from 0 to 40 mg. In the other series, the amount of ampicillin sodium was constant at 32 1 mg and the cloxacillin sodium ranged from 0 to 40 mg. In the first series AA,,, was proportional to the ampicillin concentration, and in the second series the corrected A2,5 value was proportional to the concentration of cloxacillin, thus confirming that Beer’s law is obeyed for both penicillins.Two series of mixtures were prepared. TABLE I1 RELATIVE STANDARD DEVIATIONS (RSD) AND LIMITS OF ERROR (LE) OF A SINGLE ASSAY (P = 0.95) FOR TEN REPLICATE ASSAYS OF 1 + 1 AND 2+ 1 AMPICILLIN - CLOXACILLIN MIXTURES Ampicillin Cloxacillin 7- 1 RSD, LE, RSD, LE, per cent. per cent. per cent. per cent. Ampicillin - cloxacillin (1 + 1) 0.84 f 1.90 1.03 f2.33 Ampicillin - cloxacillin (2+ 1) 0.80 f 1.81 1.62 f 3.66 SPECIFICITY- buffered at pH 5 and 9. acid), are recorded in Table IV. A number of penicillins were examined for absorbance differences at 268 nm in solutions The results, expressed as A€.,,, for the anhydrous penicillin (free ASSAY RESULTS- Commercial batches of ampicillin - cloxacillin (1 + 1 and 2 + 1) injections, supplied by Beecham Pharmaceuticals, were assayed.The results, in Table V, are compared with those TABLE I11 LEVELS OF UNHYDROLYSED PENICILLINS REMAINING 30 MINUTES AFTER PREPARATION OF THE SOLUTIONS Amount per cent. Penicillin pH remaining unhydrolysed, Ampicillin 5 100.0 Ampicillin 9 99.8 Cloxacillin 5 99.4 Cloxacillin 9 98.9 Ampicillin - cloxacillin (1 + 1) 5 99.8 Ampicillin - cloxacillin (1 + 1) 9 99.5480 DAVIDSON AND STENLAKE : THE SPECTROPHOTOMETRIC DETERMINATION [A .nabst, Vol. 99 obtained by the manufacturer by using a microbiological method of assay after separation of the penicillins by gel electrophoresis. TABLE IV ABSORBANCE DIFFERENCE AT 268 nm OF SEVERAL PENICILLINS UNDER THE CONDITIONS OF The results THE ASSAY FOR AMPICILLIN are expressed as for the anhydrous penicillin (free acid) Penicillin Ampicillin sodium Ampicillin trihydrate Penicillinase-hydrolysed ampicillin Amox ycillin Pivampicillin Cloxacillin Flucloxacillin Benz ylpenicillin Phenoxymethylpenicillin Phenethicillin Methicillin Carbenicillin Propicillin 105.6 105-2 104.7 Precipitated 0 0 0 0 0 0 0 0 - 576.4 DISCUSSION The higher absorbance of ampicillin in an acidic solution compared with that in an alkaline solution, upon which the determination of ampicillin depends, is caused by protona- tion of the aminobenzyl side chain substituent in acidic solution. This protonation abolishes the interaction with the r-electron ring resonance, which occurs in alkaline solution, and which results in a blurring of the spectrum [Fig.l ( a ) ] . A similar effect has been observed for cyclizine . 7 Although the peak in the difference spectrum of ampicillin [Fig. l ( b ) ] at 262 nm is larger than that at 268 nm, the noise associated with the absorbance measurement of the sample at 262 nm is also increased, owing to the higher absorbance of cloxacillin at this wavelength. Measurements of absorbance differences were therefore made at 268 nm. TABLE V CONCENTRATION OF AMPICILLIN AND CLOXACILLIN, AS THE FREE ACIDS, IN COMMERCIAL MIXTURES OF AMPICILLIN SODIUM AND CLOXACILLIN SODIUM Batch A B C D E F G H I J Ampicillin, per cent. m/m h I 3 Spectrophotometric* Microbiologicalt 43.6 43.8 43.7 45.1 43.7 45.7 43.6 44.7 44.6 44.2 44.0 43.4 59.9 57.7 58.7 5'7.0 59.0 58.4 59.0 57.7 Cloxacillin, per cent. m/m I 1 Spectrophotometric* Microbiologicalt 46.2 43.3 45.9 44.1 46.2 46.0 46.6 44.7 46.9 46.8 45.8 43.5 29.8 29.1 30.8 30.5 31.3 30-4 30.9 30.3 * Each result is the mean of four determinations.7 Results supplied by Beecham Pharmaceuticals. The maximum difference in absorbance of ampicillin is obtained by using buffers of pH less than 5 and greater than 9 (Fig. 2). Buffers of pH 5 and pH 9 were chosen for the assay of ampicillin because of the known instability of penicillins in highly acidic or alkaline so1utions.l Provided that the assay is carried out within 30 minutes of the preparation of the solutions , the hydrolysis of the penicillins is negligible (Table 111). It is important that measurement of absorbance differences at 268 nm is carried out in cells of matched path lengths.Even a slight difference in path length will produce an error in the absorbance difference of the sample solution due to the high absorbance of the cloxa- cillin. Thus, there should be zero absorbance at 268 nm when a solution of cloxacillin sodium in water (0.08 per cent. m/v) is placed in both the reference and test cells.August, 19741 OF AMPICILLIN AND CLOXACILLIN IN COMBINED INJECTIONS 481 The concentration of cloxacillin is obtained after correction of the total absorbance of the sample at 275 nm for the absorbance of the ampicillin. The contribution of ampicillin to the total absorbance at this wavelength is small, and it has been calculated that, for a mixture of equal parts of ampicillin and cloxacillin, an error of 10 per cent.in the calculated ampicillin concentration results in an error of only 0.5 per cent. in the calculated concentration of cloxacillin. In the microbiological assay of ampicillin and cloxacillin, the limits of error for a single determination are about &7-0 per cent. for ampicillin and h5.5 per cent. for cloxacillin (D. Sykes, personal communication). As the limits of error for a single determination (P = 0.95) in the present work ranged only from h1.81 per cent. for ampicillin to h3.66 per cent. for cloxacillin (Table 11) greater precision is obtained with the spectrophotometric method. Of the penicillins examined (Table IV), only ampicillin and amoxycillin exhibit absorbance differences at 268 nm. Amoxycillin (2- ( -) -amino-~-hydrox~benz~lpenicillin) shows a large, negative absorbance difference at 268 nm, probably due to ionisation of the phenolic group in alkaline solution.Pivampicillin, the pivaloyloxymethyl ester of ampicillin, precipitates in pH 9 buffer and therefore cannot be determined by the procedure. Other penicillins, lacking an a-aminobenzyl side chain substituent, gave a zero absorbance difference at 268 nm and it should be possible to determine ampicillin in the presence of these penicillins. However, the absorbances (1 per cent., 1 cm) of phenethicillin, phenoxymethylpenicillin, methicillin, propicillin and flucloxacillin at 268 nm are two to four times higher than that of cloxacillin. In order to avoid the high noise level in the measurement of the absorbance difference of ampicillin, the amount of mixed penicillins used in the assay would have to be reduced, with consequent loss in precision.Ampicillin sodium and ampicillin trihydrate (dissolved by the addition of a few drops of pH 9 buffer solution) show equal absorbance differences on an equimolar basis (Table IV). Standard mixtures of ampicillin trihydrate and cloxacillin sodium analysed by the procedure gave good results; however, commercial mixtures of these penicillins, formulated as a dry powder for suspension, gave good results for ampicillin but high results for cloxacillin, due to absorption at 275 nm by other components of the suspension. Ampicillin and the penicilloic acid formed by penicillinase hydrolysis of ampicillin gave equal absorbance differences at 268 nm (Table IV).The results for ampicillin will therefore include any ampicillin penicilloic acid present in the mixture. The levels of ampicillin penicilloic acid are low at the time of manufacture but may increase on storage. The extent of degradation in old samples may be quickly determined by measurement of their iodine consumption6 and samples containing unacceptable levels of penicilloic acids should be assayed by the more specific, but less precise, gel electrophoretic method. The results obtained by use of the present method for standard mixtures are in good agreement with the theoretical values (Table I). The results obtained for commercial samples are in reasonable agreement with those obtained by the manufacturer using microbiological assay after separation of the penicillins by gel electrophoresis (Table V). The principal advantages of the present method are that it is rapid, as preliminary separation of the penicillins is unnecessary, and that it is more precise than the microbiological method of assay. We thank Beecham Pharmaceuticals for supplying materials and the microbiological results and Mrs. M. Tweedie for technical assistance. 1. 2. 3. 4. 5. 6. 7. REFERENCES Garratt, D. C., “The Quantitative Analysis of Drugs,” Third Edition, Chapman and Hall Ltd., Saccani, F., Neri, C., and Sudano, F., Boll. Chim.-Farm., 1969, 108, 777. Lightbown, J. W., and de Rossi, P., Analyst, 1965, 90, 89. Celletti, P., Moretti, G. P., and Petrangeli, B., Farmaco, Ed. Prat., 1972, 27, 688. Rapson, H. D. C., and Bird, A. E., J . Pharm. Pharmac., 1963, 15, Suppl., 222. Finholt, P., Jurgensen, G., and Kristiansen, H., J . Pharm. Sci., 1965, 54, 387. Furman, W. B., J . Ass. Off. Analyt. Chem., 1968, 51, 1111. London, 1964, p. 50. Received August 20th, 1973 Accepted March 7th, 1974

ISSN:0003-2654    

DOI:10.1039/AN9749900476

出版商:RSC    

年代:1974

数据来源: RSC

摘要:

482 Autalyst, August, 1974, Vol. 99, $9. 482-486 Colorimetric Determination of Piperazine in Pharmaceutical Formulations BY YEHIA M. DESSOUKY AND SAAD A. ISMAIEL (Pharmaceutical Chemistry Department, Faculty of Pharmacy, Cairo University, Caivo, Egypt) (Research Department, SociLtte' Misr pour l'lndustrie Pharmaceutique, 92 El Mataria Street, Post El Zeitoun, Cairo, Egypt) Piperazine can be satisfactorily determined in pharmaceutical prepara- tions or formulations such as effervescent granules and elixirs containing hexamine, colchicine, atropine sulphate, sodium benzoate, lithium benzoate, lithium citrate, sodium citrate, sodium hydrogen carbonate, tartaric acid, citric acid, lactose, sucrose and Tinct. ammi visnaga. The diluted sample solution is treated with a 0.6 per cent.aqueous I ,2-naphthoquinone-4-sulphon- ate solution in the presence of acetate - citrate buffer a t pI3: 7-5. The tempera- ture of the reaction should be between 10 and 15 "C and the colour produced is measured a t 490nm. PIPERAZINE, a pyrazine derivative, is one of the most potent drugs used as an anthelmintic for the treatment of threadworms and roundworms in both children and adults, the worms usually being voided from the host while the drug is still active. Piperazine in effervescent granules cannot be determined by using either the non-aqueous titration method of the United States Pharmacopeial or the gravimetric method of the British Pharmacopoeia.2 Further, the colorimetric methods of Pankrat~,~ who used the reineckate salt, and Perlm~tter,~ who used acidified 9-benzoquinone, could not be applied to this preparation.This inapplicability is because of the presence and interference of hexamine, citric and tartaric acids, sodium citrate, sodium hydrogen carbonate and other interfering ingredients that yield either higher or lower results. The titrimetric f ormol method suggested by Simionovici, Rsianu and Cuculescu5 also cannot be used for an accurate determination of piperazine in coloured pharmaceutical preparations as the colour masks the end-point, giving high results. Gravimetric methods, such as those described by Maynard6 for the diacetyl derivative and by Chemerisskaya' for the dichromate derivative were tried, but accurate results were not obtained as the methods are not sensitive. They are time consuming, however. In the present investigation, a colorimetric method for the determination of piperazine in some pharmaceutical preparations is proposed that involves the use of the red colour formed on treating piperazine with 1,2-naphthoquinone-4-sulphonate in the presence of acetate - citrate buffer at pH 7.5.The proposed method is free from the drawbacks of the above methods.l-' REAGENTS- The pure piperazine used was of pharmaceutical grade and all of the chemicals and reagents were either of analytical-reagent or pharmaceutical grade. 1,2-Naphthoquinone-4-suZ~honate reagent-Dissolve 0.3 g of 1,2-naphthoquinone-4-~~1- phonate sodium salt in 50ml of water. Acetate - citrate bufler, pH 7.5-Dissolve 5.0 g of sodium acetate and 1.0 g of sodium citrate in about 50 ml of water and adjust the pH of the solution to 7.5 with a few drops of 33 per cent.acetic acid, then dilute to a volume of 100 ml with water and filter if necessary. Standard solution-Weigh accurately 100 rng of piperazine citrate (or hexahydrate, depending on the formulation) and dissolve it in sufficient water to produce 100 ml. Dilute 3 ml of this solution to 100 ml with water. Samples-The composition of each of the samples tested is given in Table I. @ SAC and the authors. EXPERIMENTAL This reagent must be freshly prepared.DESSOUKY AND ISMAIEL TABLE I FORMULATION OF ANALYSED SAMPLES Sample 483 Constituent Piperazine hexahydratelg . . Piperazine citratelg . . .. Hexamine/g . . . . .. Colchicine/g . . .. .. Atropine sulphatelg . . .. Sodium benzoatelg .. .. Lithium benzoate/g . . .. Lithium citrate/g . . . . .. Sodium citratelg . . .. .. Tartaric acid/g . . .. . . Citric acid/g . . .. .. Lactose/g . . .. .. .. Sucroselg . . .. .. .. Tinct. ammi visnaga/ml . . Sodium hydrogen carbonatelg . . .. Coloured flavoured syruplml . . Flavoured, coloured effervescent mixturelg . . .. .. Effervescent mixturelg . . * . I 14.0 I1 1.5 5.0 - I11 1.8 4.2 1.8 - - - IV 1.125 1.5 - - 2.25 1.5 - 6.3 46.8 23-4 23.4 4.5 - - 30-65 16.2 10.3 3.0 8.025 - - to 100.0 to 50.0 - - to 70.0 - to 90.0 - to 90.0 Samples I and I1 (Vermizine elixir and coli-urinal effervescent granules) supplied by SociCtC Sample I11 (Urosolvine effervescent granules) supplied by the Nile Company for Pharma- Sample IV (Ciluryl effervescent granules) supplied by ADCO (The Arab Drug Company), Sample V (Urolithine effervescent granules) supplied by Kahira Pharmaceutical and Misr pour 1’Industrie Pharmaceutique, Cairo, Egypt.ceuticals and Chemical Industries, Cairo, Egypt. Cairo, Egypt. Chemical Industries, Cairo, Egypt. PREPARATION OF SAMPLE SOLUTIONS- For effervescent granules, transfer an amount of the sample equivalent to 100 mg of piperazine citrate (or hexahydrate, depending on the formulation) into a 100-ml calibrated flask, add about 50 ml of water, and when the effervescence has ceased, dilute to volume with water. For elixirs, dilute the appropriate volume of the sample with water to give a final concentration equivalent to 3 mg of piperazine citrate (or hexahydrate, depending on the formulation) in 100 ml of water.Dilute 3 ml of this solution to 100 ml with water. TABLE I1 COMPARISON BETWEEN THE RESULTS OF THE PROPOSED METHOD AND THE B.P. 1968 METHOD Amount of piperazine citrate (or hexahydrate) I Recovered: by proposed method & I 64.7 107.8 60-3 100.5 62.5 104.1 60.0 100.0 59.4 99.0 60.8 101.3 56.3 93.8 60.3 100.5 58-2 97.0 61.0 101.6 PFtg per cent. Recovered $ by B.P. 1968 method Standard Sample* Takenlpg addedt/pg - I 60 I1 60 I11 60 IV 60 V 60 60 - 60 - 60 - 60 - 60 - - - - - - Pg per cent. 48.0 80.0 38.4 64.0 50.1 83.5 57.6 96.0 - - - - - - - § The standard deviation is k0.97 per cent. for the proposed method. * See footnote to Table I. 7 Added as a 0.003 per cent. aqueous solution of piperazine citrate (or hexahydrate). 9 The precipitated piperazine picrate was unfilterable.Each value given is the average of three experiments.484 DESSOUKY AND ISMAIEL : COLORIMETRIC DETERMINATION OF [Analyst, VOl. 99 PROCEDURE- To 2 ml of the standard and sample solutions in separate test-tubes, add 5 ml of the acetate - citrate buffer, pH 7.5, mix well and then cool the mixtures in a water-bath at a temp- erature between 10 and 15 "C for 2 minutes. Add to each test-tube 3 ml of 1,Z-naphtho- quinone-4-sulphonate reagent, again mix well and leave them to stand for about 10 minutes. Measure the absorbance of both standard and sample solutions at a wavelength of 490 nm against a blank carried out simultaneously. A Carl Zeiss, Model M4 QII, spectrophotometer was used. Calculate the amount of piperazine citrate (or piperazine hexahydrate) as follows : x 3 x 100 = piperazine salt in sample, per cent.m/m s m2 where T and S are the absorbance values of sample and standard, respectively, m, is the amount of standard present in 2 ml, and m2 is the amount of sample present, theoretically, in 2 ml. The results are given in Table 11. a) 0.4 g 0.3 0.2 0 S B a 0- 1 0 460 500 540 580 620 660 Wave1 engt h/nm Fig. 1. Light absorbance spectrum of the colour formed RESULTS AND DISCUSSION During the study of the reaction of 1,2-naphthoquinone-4-sulphonate with some secondary amines, a red colour was observed on treating piperazine with 1,2-naphthoquinone-4-sulphon- ate at a pH between 6 and 9. 0.5 0.4 - a) f $ 0.3 2 n - a 0.2 - 5 6 7 8 9 1 0 PH Fig. 2. Optimum pH for the reaction (wavelength, 490 nm)August, 19741 PIPERAZINE IN PHARMACEUTICAL FORMULATIONS 485 A studv of the colour formed showed that the maximum absorption occurs at 490 nm (Fig.l), that ihe optimum pH for the reaction is 7.5 (Fig. 2) and that the optimum tration of the reagent is 18 mg per 10 ml (Fig. 3). concen- 0 10 18 20 30 Amount of reagentlmg 40 Fig. 3. Variation of absorbance with amount of 1,2-naphthoquinone-4-sulphonic acid, sodium salt (wavelength, 490 nm) The coloured product of the reaction started to precipitate after about 15 minutes. This time decreased with increase in the concentration of piperazine (Fig. 4). The above precipi- tation accounts for the slight differences in the calibration graphs. The addition of not only ethanol, but also methanol, propan-1-01 and propan-2-01, was tried, without success in reducing the differences.Extraction into an immiscible solvent, such as chloroform, was also tried and gave partial extraction. However, the organic layer changed to a yellowish colour. The stability of the red colour (before the formation of a red precipitate) differs according to the concentration. The most stable colour is obtained when 60 pg of piperazine salt are used (Fig. 4). 0.5 0.4 0) f 0.3 f a 2 0.2 0.1 - 1 I 1 I I I I I I I o 4 a 12 16 20 24 28 32 36 40 44 Ti me/minutes Fig. 4. Colour stability a t different concentrations of piperazine salt (wavelength, 490 nm) : A, 80 pg; B, 60 pg; and C, 100 pg Beer’s law is obeyed for amounts of piperazine citrate from 20 to 120 pg (Fig. 5), although the simultaneous use of a standard, with which to compare the sample, is found to be necessary in order to obtain accurate results, as the slope of the calibration graph differs slightly on repetition.486 DESSOUKY AND ZSMAIEL Piperazine citrate/pg Fig. 5. length, 490 nm) Calibration graph for different amounts of piperazine citrate (wave- REFERENCES 1. 2. 3. 4. 5. 6. 7. “The United States Pharmacopeia,” XVIIIth Revision, Mack Co., Easton, Pa., 1970, p. 507. “British Pharmacopoeia,” The Pharmaceutical Press, London, 1968, p. 776. Pankratz, R. E., J . Pharm. Sci., 1961, 50, 175. Perlmutter, S. H., J . Ass. Off. Agric. Chem., 1958, 41, 506. Simionovici, R., RAianu, J., and Cuculescu, V., Rev. Chim., Bucharest, 1959, 10, 105; Analyt. Maynard, W. R., jun., J , Ass. Off. Agric. Chem., 1959, 42, 610. Chemerisskaya, A. A., Zh. Analit. Khim., 1956, 11, 356; Chem. Abstr., 1956, 50, 15349’. Abstr., 1959, 6, 4154. Received April 19th, 1973 Accepted February 4 t h 1974

ISSN:0003-2654    

DOI:10.1039/AN9749900482

出版商:RSC    

年代:1974

数据来源: RSC

摘要:

Analyst, August, 1974, Vol. 99, pp. 487-490 487 Colorimetric Determination of Antazoline in Some Pharmaceutical Preparations with Sodium Nitrite BY M. M. AMER, M. S. TAWAKKOL (Department of Analytical Chemistry, Faculty of Pharmacy, Cairo University, Cairo, Egypt) AND s. A. ISMAIEL (Research Department, Soci&te' Misr pour l'lndustrie Pharmaceutique, 92 El Mataria Street, Post El Zeitoun, Cairo, Egypt) A colorimetric method for the determination of antazoline in pharma- ceutical preparations with sodium nitrite has been developed. The method involves treatment of a cooled and acidified dilute aqueous solution of the sample with sodium nitrite. The yellow colour produced is stabilised by the addition of propan-2-01 or ethanol and the absorbance then measured at 410 nm. Naphazoline, tolazoline, clemizole, diphenhydramine, chlorpheniramine, ephedrine, cetrimide, benzalkonium chloride and zinc salts, even if present in amounts ten times greater than that of antazoline, do not interfere.ANTAZOLINE can be determined in some pharmaceutical preparations by using the acidimetric methods of the B.P.,l the B.P.C.,2 the non-aqueous titration methods of the U.S.P.,3 Selles and Rodriguez4 and Rink and F3emhofer5 or by using the EiZ& value mentioned by Clarke,6 but none of these methods can be used if antazoline occurs in combination with ephedrine, diphenhydramine or chlorpheniramine. Phenylephrine, if present, interfered in the determination by the last four of the above methods, while naphazoline interfered in the determination by the methods of Slack and Mader,' Horioka and Ishiokas and K ~ m - T a t t , ~ giving high results.In this paper, a sensitive colorimetric method for the determination of antazoline in pharmaceutical preparations is proposed in which use is made of the qualitative colour reaction reported by Auterhoff.lO The proposed method was found to be simple, quick and devoid of the drawbacks of the methods mentioned above. EXPERIMENTAL REAGENTS- Antazoline hydrochloride standard stock solution-Dissolve 100 mg of antazoline hydro- chloride in sufficient water to give a final volume of 100 ml. Antazoline methanesulphonate standard stock solution-Dissolve an amount of an tazoline methanesulphonate (previously determined according to the B.P.C. 1968) in sufficient water to give a final concentration of 100 mg of antazoline methanesulphonate per 100 ml.Sodium nitrite solution-Dissolve 2 g of sodium nitrite (Merck) in sufficient water to give a final volume of 100 ml. Hydrochloric acid, approximately 2.5 N. Hydrochloric acid, concentrated. Ethanol, 95 per cent. CALIBRATION GRAPH- Dilute the appropriate volume of the antazoline salt stock solution with water to give solutions of final concentrations ranging between 0.4 and 2-4mg of the antazoline salt per 100 ml. Transfer 5 ml of the diluted antazoline salt solution (containing 20 to 120 pg of the corresponding antazoline salt) into a glass-stoppered test-tube, cool it in ice, add 2 ml of 2.5 N hydrochloric acid, mix and then add 1 ml of sodium nitrite. Mix the solution and, after @ SAC and the authors.488 [Analyst, Vol.99 2 minutes, add 2 ml of ethanol, mix again, then remove the test-tube from the ice and, after 3 minutes, measure, at 410 nm, the absorbance of the yellow colour produced against a blank carried out simultaneously. AMER et aZ. : COLORIMETRIC DETERMINATION OF ANTAZOLINE PREPARATION OF SAMPLE SOLUTIONS- Tablets-Transfer an amount of the powder containing about 50 mg of antazoline hydro- chloride into a 100-ml calibrated flask, add about 80 ml of water and shake the flask until the powder has dissolved (about 30 minutes). Dilute this solution to volume with water, filter it through a dry Whatman No. 1 filter-paper and dilute 2 ml of it to 100 ml with water. Drops, injectable solutions and syru$s-Dilute an appropriate volume of the sample with water so as to give a final concentration of about 1 mg of the antazoline salt per 100 ml.Creams-Transfer an amount of the cream containing about 100 mg of antazoline hydro- chloride into a beaker and extract it successively with 50, 20 and 20 ml of 0.5 N hydrochloric acid by heating the mixture each time on a boiling water bath, while stirring, until the cream has melted, allowing it to stand for about 5 minutes, then cooling it in ice, and decanting the aqueous extract into a separating funnel. Extract the combined aqueous extracts with about 10 ml of chloroform, make the aqueous layer up to 200 ml with water, then dilute 2 ml of the latter to 100 ml with water. Lotions-Transfer 5 ml of the well shaken sample into a 100-ml calibrated flask, add 5 ml of concentrated hydrochloric acid, mix and, when the effervescence ceases, dilute the mixture to volume with water.Filter it through a dry Whatman No. 1 filter-paper and reject the first turbid 10 ml of the filtrate, then dilute a volume containing about 1 mg of antazoline hydrochloride to 100 ml with water. DEVELOPMENT OF THE COLOUR- Transfer 5 ml of the prepared sample solution into a glass-stoppered test-tube and complete the procedure as directed under Calibration graph, starting with “cool it in ice, add 2 ml of 2.5 N hydrochloric acid. . .” Calculate the concentration of the antazoline salt from the calibration graph, RESULTS AND DISCUSSION The results obtained by applying the proposed method to the determination of antazoline hydrochloride and antazoline methanesulphonate in some pharmaceutical preparations are given in Table I.TABLE I COMPARISON OF RESULTS BY THE PROPOSED METHOD AND SLACK AND MADER’S METHOD Antazoline salt I i Pharmaceutical Indicated Added Recovery, per cent.? preparations* on to f A > label/ labelled by proposed by Slack and Mader’s material / mg method method Sensol tablets 100 - 99.6 f 1.06 103-04 f 1-34 Antistine ampoules 50 - 100.4 f 1.14 101.5 f 1.89 Antistine privine solution 500 - 98.42 f 2.32: 112 f 2.521 Fenozal drops5 500 100-92 f 0.61 101-92 f 0.31 Calazol lotion 1000 - 95.6 f 1 not applicable Calazol cream 2000 - 105 f 2.37 not applicable mg - 50 100.1 f 0.65 25 100 f 0.81 - 250 98.93 f 1.35 - 500 97.4 & 0.82 1000 103.83 f 1.82 - * Sensol tablets, fenozal drops, calazol lotion and calazol cream were supplied by Socidtd Misr pour Antistine ampoules and antistine privine solution were supplied ? Mean of six experiments.$ Calculated as antazoline hydrochloride ; for conversion into antazoline sulphate results were multiplied 4 Specially prepared authentic sample containing the correct amount of ingredients. 1’Industrie Pharmaceutique, Cairo, Egypt. by Ciba Laboratories, Basle, Switzerland. by 1.041.August, 19741 IN SOME PHARMACEUTICAL PREPARATIONS WITH SODIUM NITRITE 489 The qualitative colour reaction of antazoline and sodium nitrite reported by Auterhoff lo was studied and used in order to develop a quantitative method for the determination of antazoline in pharmaceutical preparations. The absorption spectrum of the colour obtained showed a maximum at 410 nm (Fig.1) and the maximum intensity of the colour was obtained by using 20mg of sodium nitrite (Fig. 2) and hydrochloric acid of about 2.5 N concentration (Fig. 3) (both sulphuric and hydrochloric acids at about 2.5 N concentration gave the same colour intensity). The colour reached its maximum intensity after about 10 minutes, remained stable for about 5 minutes and then began to fade (Fig. 4). Wavelength/nrn Fig, 1. Absorption spectrum of the colour produced 0.4 ~ 0.3 2 0.2 Fi 0 Is m 0.1 I I I I I I I J 0 5 10 15 20 25 30 35 40 Sod i u rn nit r it e/mg Fig. 2, Effect of the amount of sodium nitrite on the intensity of the colour The use of ethanol (or propan-2-01) was found to improve the stability of the colour (Fig. 4), the intensity of which obeyed Beer's law for amounts of antazoline hydrochloride between 20 and 120 pg and of antazoline methanesulphonate between 20 and 140 pg.The colour was found to be unstable at room temperature or on heating the solution but cooling the latter in ice or to a temperature below 10 "C improved the stability of the colour (Fig. 4). As shown in Table I, the proposed method gave more accurate results than those obtained by the method of Slack and Mader, especially for samples that contain naphazoline. Diphenhydramine, chlorpheniramine, naphazoline, ephedrine, codeine, benzalkonium chloride, phenylephrine, tolazoline, clemizole, phenol, camphor, esters of p-hydroxybenzoic acid, sodium cyclamate, saccharin sodium, zinc oxide and calamine in amounts ten times greater than that of antazoline did not interfere.0 1 2 3 4 Hydrochloric acid/N Fig. 3. Effect of concentration of hydrochloric acid on the intensity of the coIour490 AMER, TAWAKKOL AND ISMAIEL The reaction is probably due to the aniline moiety of the antazoline molecule, because naphazoline and tolazoline, which contain the same imidazoline ring but no aniline moiety, did not give a colour, whereas phentolamine and diethylaniline, which contain an aniline moiety, gave a yellow colour with a maximum absorption at 360 nm. 0.4 I 1 0 5 15 25 35 Time elapsed/minutes Fig. 4. Improved stability of the colour a t temperatures below 10 “C (solutions cooled in an ice-cold water-bath for both stable and un- stable colours) : A, ethanol - water; and B, water The explanation that the colour results from the formation of a nitroso derivative of antazoline similar to that which occurs in the nitrosation of a dialkylaniline such as diethyl- aniline with nitrous acid, as stated by Finarl11l2 and Mann and Saunders,l39l4 cannot be supported as the maximum absorption of the yellow colour produced with phentolamine and diethylaniline differed greatly from that with antazoline.1. 2. 3. 4. 5. 6. 7. 8. 9. 10. 11. 12. 13. 14. REFERENCES “British Pharmacopoeia 1968,” The Pharmaceutical Press, London, 1968, p. 56. “British Pharmaceutical Codex 1968,” The Pharmaceutical Press, London, 1968, p. 46. “United States Pharmacopeia,” Fifteenth Revision, 1955, Mack Publishing Company, Easton, Pa., Selles, E., and Rodriguez, A. M., Galenica Acta, 1956, 9, 33; Chem. Abstr., 1957, 15, 8375 h. Rink, M., and Riemhofer, M., Mitt. dt. pharm. Ges., 1961, 31, 197. Clarke, E. G. C., assisted by Berle, J., “Isolation and Identification of Drugs in Pharmaceuticals, Slack, S. C., and Mader, W. J., J . Amer. Pharm. Ass., Sci. Edn., 1957, 46, 742. Horioka, M., and Ishioka, H., J . Pharm. SOC. Japan, 1961, 18, 76. Kum-Tatt, L., J . Pharm. Pharmac., 1960, 12, 666. Auterhoff, H., Arch. Pharm., Bed., 1950, 283, 244. Finar, I. L., “Organic Chemistry, The Fundamental Principles,” Volume I, Fifth Edition, The p. 62. Body Fluids and Post Mortem Material,” The Pharmaceutical Press, London, 1969, p. 198. English Language Book Society and Longman Group Limited, London, 1969, p. 621. -, oP. cit., p. 697. Mann, F. G., and Saunders, B. C., “Practical Organic Chemistry,” Longmans, Green and Co. Ltd., Fourth Edition, London, 1960, p. 202. -__ , , op. cit., p. 376. Received December 17th, 1973 Accepted February 25t12, 1974

ISSN:0003-2654    

DOI:10.1039/AN9749900487

出版商:RSC    

年代:1974

数据来源: RSC