ISSN:0003-2654    

DOI:10.1039/AN98712BP005

出版商:RSC    

年代:1987

数据来源: RSC

摘要:

The AnalystThe Analytical Journal of The Royal Society of ChemistryAdvisory Board*Chairman: J. D. R. Thomas (Cardiff, UK)J. F. Alder (Manchester, UK)D. Betteridge (Sunbury-on-Thames, UK)E. Bishop (Exeter, UK)*C. Burgess (Ware, UK)D. T. Burns (Belfast, UK)G. D. Christian (USA)*M. S. Cresser (Aberdeen, UK)L. de Galan (The Netherlands)*A. G. Fogg (Loughborough, UK)*C. W. Fuller (Nottingham, UK)V. D. Goldberg (London, UK)T. P. Hadjiioannou (Greece)A. Hulanicki (Poland)*C. J. Jackson (London, UK)*P. M. Maitlis (Sheffield, UK)E. J. Newman (Poole, UK)T. B. Pierce (Harwell, UK)E. Pungor (Hungary)J. Rhcka (Denmark)R. M. Smith (Loughborough, UK)W. I. Stephen (Birmingham, UK)M. Stoeppler (Federal Republic of Germany)K. C. Thompson (Shefiield, UK)*A.M. Ure (Aberdeen, UK)A. Walsh, K.B. (Australia)G. Werner (German Democratic Republic)T. S. West (Aberdeen, UK)*P. C. Weston (London, UK)J. D. Winefordner (USA)Yu. A. Zolotov (USSR)P. Zuman (USA)*Members of the Board serving on the Analytical Editorial BoardRegional Advisory EditorsFor advice and help to authors outside the UKDr. J. Aggett, Department of Chemistry, University of Auckland, Private Bag, Auckland, NEWDoz. Dr. sc. K. Dittrich, Analytisches Zentrum, Sektion Chemie, Karl-Marx-Universitat, Talstr.Professor L. Gierst, Universite Libre de Bruxelles, Faculte des Sciences, Avenue F.-D.Professor H. M. N. H. Irving, Department of Analytical Science, University of Cape Town,Dr. 0. Osibanjo, Department of Chemistry, University of Ibadan, Ibadan, NIGERIA.Dr.G. Rossi, Chemistry Division, Spectroscopy Sector, CEC Joint Research Centre,Dr. 1. RubeSka, Geological Survey of Czechoslovakia, Malostranske 19, 118 21 Prague 1,Professor K. Saito, Coordination Chemistry Laboratories, Institute for Molecular Science,Professor M. Thompson, Department of Chemistry, University of Toronto, 80 St. GeorgeProfessor P. C. Uden, Department of Chemistry, University of Massachusetts, Amherst,Professor Dr. M. Valcarcel, Departamento de Quimica Analitica, Facultad de Ciencias,Professor Yu Ru-Qin, Department of Chemistry and Chemical Engineering, Hunan University,ZEALAND.35, DDR-7010 Leipzig, GERMAN DEMOCRATIC REPUBLIC.Roosevelt 50, Bruxelles, BELGIUM.Rondebosch 7700, SOUTH AFRICA.EURATOM, lspra Establishment, 21020 lspra (Varesel, ITALY.CZECHOSLOVAKIA.Myodaiji, Okazaki 444, JAPAN.Street, Toronto, Ontario M5S I A I , CANADA.MA 01003, USA.Universidad de Cbrdoba, 14005 Cordoba, SPAIN.Changsha, PEOPLES REPUBLIC OF CHINA.Editor, The Analyst:Philip C.WestonSenior Assistant Editors:Judith Brew, Roger A. YoungAssistant Editors:Anne Horscroft, Pamil SehmiEditorial Office: The Royal Society of Chemistry, Burlington House,Piccadilly, London, WIV OBN. Telephone 01-734 9864. Telex No. 268001Advertisements: Advertisement Department, The Royal Society of Chemistry, BurlingtonHouse, Piccadilly, London, W1V OBN. Telephone 01-437 8656. Telex No. 268001The Analyst (ISSN 0003-2654) is published monthly by The Royal Society of ChemistryBurlington House, London WIV OBN, England.All orders accompanied with payment shoulcbe sent directly to The Royal Society of Chemistry, The Distribution Centre, Blackhorse RoadLetchworth, Herts. SG6 IHN, England. 1987 Annual subscription rate UK f160.00, Rest oWorld €179.00, USA $315.00. Purchased with Analytical Abstracts UK f364.00, Rest of Workf403.00, USA $709.00. Purchased with Analytical Abstracts plus Analytical Proceedings Ukf411 .OO, Rest of World €455.00, USA $801 .OO. Purchased with Analytical Proceedings Uk€200.00, Rest of World f224.00, USA $394.00. Air freight and mailing in the USA b)Publications Expediting Inc., 200 Meacham Avenue, Elmont, NY 11003.USA Postmaster: Send address changes to: The Analyst, Publications Expediting Inc., 20[Meacham Avenue, Elmont, NY 11003.Second class postage paid at Jamaica, NY 11431. Alother despatches outside the UK by Bulk Airmail within Europe, Accelerated Surface Posioutside Europe. PRINTED IN THE UK.Information for AuthorsFull details of how to submit material forpublication in The Analyst are given in theInstructions to Authors in the January issue.Separate copies are available on request.The Analyst publishes papers on all aspects ofthe theory and practice of analytical chemistry,fundamental and applied, inorganic andorganic, including chemical, physical, biochem-ical, clinical, pharmaceutical, biological, auto-matic and computer-based methods. Papers onnew approaches to existing methods, newtechniques and instrumentation, detectors andsensors, and new areas of application with dueattention to overcoming limitations and to un-derlying principles are all equally welcome.There is no page charge.The following types of papers will be con-sidered:full papers, describing original work.Short papers: the criteria regarding origin-ality are the same as for full papers, but shortpapers generally report less extensive investi-gations or are of limited breadth of subjectmatter.Communications, which must be on anurgent matter and be of obvious scientificimportance. Rapidity of publication is enhancedif diagrams are omitted, but tables and formulaecan be included.Communications receive pri-ority and are usually published within 5 8weeks of receipt.They are intended for briefdescriptions of work that has progressed to astage at which it is likely to be valuable toworkers faced with similar problems. A fullerpaper may be offered subsequently, if justifiedby later work.Reviews, which must be a critical evaluationof the existing state of knowledge on a par-ticular facet of analytical chemistry.Every paper (except Communications) will besubmitted to at least two referees, by whoseadvice the Editorial Board of The Analyst will beguided as to its acceptance or rejection. Papersthat are accepted must not be published else-where except by permission. Submission of amanuscript will be regarded as an undertakingthat the same material is not being consideredfor publication by another journal.Regional Advisory Editors.For the benefit ofpotential contributors outside the United King-dom, a Panel of Regional Advisory Editorsexists. Requests for help or advice on anymatter related to the preparation of papers andtheir submission for publication in The Analystcan be sent to the nearest member of the Panel.Currently serving Regional Advisory Editors arelisted in each issue of The AnalystManuscripts (three copies typed in double spac-ing) should be addressed to:The Editor, The Analyst,Royal Society of Chemistry,Burlington House,Piccadilly,LONDON W1V OBN, UKParticular attention should be paid to the use ofstandard methods of literature citation, includingthe journal abbreviations defined in ChemicalAbstracts Service Source Index. Wherever pos-sible, the nomenclature employed should fol-low IUPAC recommendations, and units andsymbols should bethose associated with SI.All queries relating to the presentation andsubmission of papers, and any correspondenceregarding accepted papers and proofs, shouldbe directed to the Editor, The Analyst (addressas above). Members of the Analytical EditorialBoard (who may be contacted directly or via theEditorial Office) would welcome comments,suggestions and advice on general policy mat-ters concerning The Analyst.Fifty reprints of each published contribution aresupplied free of charge, and further copies canbe purchased.0 The Royal Society of Chemistry, 1987. Allrights reserved. No part of this publication maybe reproduced, stored in a retrieval system, ortransmitted in any form, or by any means,electronic, mechanical, photographic, record-ing, or otherwise, without the prior permissionof the publishers

ISSN:0003-2654    

DOI:10.1039/AN98712FX009

出版商:RSC    

年代:1987

数据来源: RSC

摘要:

ANALAO 112(3) 217-344 (1987)The AnalystMarch 1987The Analytical Journal of The Royal Society of Chemistry21722122723 123724324725325926326727 127728328729 129530130530931 331732 1325329CONTENTSA Solid-state Ion-selective Electrode Membrane Preparation Based on Benzyldimethyltetradecylammonium Dithio-SulphatoargenateTitus N. Nwosu, William A. StrawDetermination of Ultra-trace Amounts of Cadmium in Natural Waters by the Combination of a Solvent ExtractionProcedure and Anodic Stripping Voltammetry-Samuel B. Adeloju, Kathryn A. BrownDetermination of Lead and Cadmium in Urine by Differential-pulse Anodic Stripping Voltammetry-A. Nur Onar,Aytekin TemizerPolarographic Study of Solutions of Cadmium(l1) in Acetylacetone-Jesus Hernandez Mendez, Fernando BecerroDominguezDetermination of B2 Vitamers in Pharmaceutical Preparations, Foods and Animal Tissues by a PhotokineticMethod-Tomas Perez-Ruiz, M.C. Martinez-Lozano, Virginia TomasCathodic Stripping Voltammetry of 5-Fluorouracil-A. J. Miranda Ordieres, M. J. Garcia Gutierrez, A. Costa Garcia,P. Tufion Blanco, W. Franklin SmythInvestigation of the Adsorptive Stripping Voltammetric Behaviour of the Anticancer Drugs Chlorambucil and5-Fluorouracil-Joseph Wang, Meng Shan Lin, Vince Villalodimetric and Bromimetric Flow Injection Amperometric Methods of Determining Sulphite in which Iodine orBromine 1s Monitored at a Glassy Carbon ElectrodeArnold G. Fogg, Christopher W. Guta, Antoine Y. ChamsiFlow Injection Analysis-Use of lmmobilised Enzymes for the Determination of Ethanol in Serum-Juan Ruz, MariaDolores Luque de Castro, Miguel ValcarcelDetermination of Analytical Parameters in Drinking Water by Flow Injection Analysis.Part 1. SimultaneousDetermination of pH, Alkalinity and Total Ionic Concentration-Francisco Cafiete, Angel Rios, Maria Dolores Luquede Castro, Miguel ValcarcelDetermination of Analytical Parameters in Drinking Water by Flow Injection Analysis. Part 2. SimultaneousDetermination of Calcium and Magnesium-Francisco Cafiete, Angel Rios, Maria Dolores Luque de Castro, MiguelVa Ica rcelFlow Injection Atomic Absorption Spectrometry with Air Compensation-lgnacio Lopez Garcia, Manuel HernandezCordoba, Concepcion Sa nchez-Ped refioObservation on the Interference by Copper(ll), Cobalt(l1) and Nickel(l1) on the Determination of Arsenic by ArsineGeneration Atomic Absorption Spectrometry-John Aggett, Yasu hisa HayashiDetermination of Copper by Electrothermal Atomisation Atomic Absorption Spectrometry following Coprecipitationwith Hafnium HydroxideJoichi Ueda, Natsuko YamazakiDetermination of Urinary Cobalt Using Matrix Modification and Graphite Furnace Atomic Absorption Spectrometrywith Zeeman-effect Background Correction-Mary M.Kimberly, George G. Bailey, Daniel C. PaschalEvaluation of the Stability of Some Elements During Lyophilisation of Rat Liver Using Atomic AbsorptionSpectrometry-Eiji Uchino, Kazuo Jin, Toshibumi Tsuzuki, Katsuhiro lnoueSpectrophotometric Determination of Phosphorus as Orthophosphate Based on Solvent Extraction of the IonAssociate of Molybdophosphate with Malachite Green Using Flow Injection-Shoji Motomizu, Mitsuko OshimaSpectrophotometric Study of the Reaction of Zinc with Cadion 2B in the Presence of Triton X-100-Nai-kui Shen,Wen-tien Chu, Zu-yue Chen, Wan-long Ziao, Yu-rui ZhuSpectrophotometric Determination of Tungstate Using an Iron Salt, a Thiol and Oxygen-Anatol E berhard, George L.Ke nyo nDetermination of Vitamin BI2 as Cobalt by Use of a Catalytic-Spectrophotometric Method-J.Medina-Escriche,M. L. Hernandez-Llorens, M. Llobat-Estelles, A. Sevillano-CabezaDetermination of Nicotinamide in Pre-mixes by Near-infrared Reflectance Spectrometry-Brian G. OsborneDetermination of Nitrogen in Cheese Using Copper Sulphate as a Kjeldahl Catalyst: Results of an Inter-laboratoryExtraction Method for the Determination of Total Chlorine in Coal-James A.Cox, Raaidah SaariPoly(Hydroxamic Acid) Chelating Resin. Part II. Separation of Zinc from Cadmium and of Cobalt from Copper andDetermination of Zinc in the Presence of Copper, Cadmium and Lead by Photometric EDTA Titration with SelectiveStudy-Eric Florence, Wendy M. HarrisNickeCAjay Shah, Surekha DeviMasking-Colin G. Halliday, Michael A. Leonardcontinued inside back coverTypeset and printed by Black Bear Press Limited, Cambridge, EnglanSHORT PAPERS333 Identification of Divalent Sulphur Compounds by an Enthalpimetric Approach t o the Iodine - Azide Reaction-Miechw335 Photometric Determination of Gallium in Silicate Rocks After Separation by Crown Ether Extraction-Hideko Koshima,337 Application of Microwave Digestion t o the Analysis of Peat-Clara S.E. Papp, Lynn 6. Fischer339 Determination of Methanol in Gasoline by Gas Chromatography-Stanford L. Tackett341 Inhibitor-treated Microbial Sensor for the Selective Determination of Glutamic Acid-Klaus Riedel, Frieder SchellerWronski, Whdystaw GoworekHiroshi OnishiI Classified Advertisements, Burlington House, Piccadilly, London WIV OBNTel: 01437 8656.343 BOOK REVIEWSCLASSIFIED ADVERTISINGDisplay & Semi-Display f12 per SCCPhysical Organic Chemistc.EZ1,OOO + bonusThis major multinational pharmaceutical company wishesto strengthen further their well established UKdevelopment group.Consequently our client now intendsto ap oint a physical organic chemist to make asignilcant impact upon the analytical development ofprospective drug candidates.Keportin at Director level and responsible for a team offive peode, you will apply all the latest analyticaltechniques. Prospects for early promotion within this USthe pharmaceutical/fine chemical industries or otherorganisations employing high precision analyticaltechniques. As you would expect technical excellence isvital, combined with the ersonality and ability toperform and achieve at a gigh level. The strategicimportance of this appointment means that for the rightperson salary is unlikely to be a limiting factor and,naturally, generous assistance is available to relocate tothe Home Counties.based multinational are excellent, and just as important,scientific staff are able to retain detailed professionalinvolvement rather than becoming administrators, even atthe most senior levels.You must have a PhD and at least five years experience inInitially, please contact Larryor write enclosin a CV and current salarydetails at Mac.ilfm Davies, Salisbury House,Bluecoats, Hertford, SG14 1PU.on (0992INTERNATIONAL SEARCH EXECUTIVE552552

ISSN:0003-2654    

DOI:10.1039/AN98712BX011

出版商:RSC    

年代:1987

数据来源: RSC

摘要:

ANALYST, MARCH 1987, VOL. 112 217 A Solid-state Ion-selective Electrode Membrane Preparation Based on Benzyldimethyltetradecylammonium Dithiosulphatoargentate Titus N. Nwosu University of Ife, Chemistry Department, Ile-lfe, Oyo State, Nigeria and William A. Straw 5 Cheynies Court, Arundel Way, Highcliffe, Christchurch, Dorset BH23 5DX, UK A technique for preparing a solid-state ion-selective electrode membrane based on benzyldimethyltetradecyl- ammonium dithiosulphatoargentate has been investigated and the performance of the Ag(S203)23--sensitive electrode constructed from this membrane has been assessed. Microscopic examination of the material showed that when freshly prepared, it was amorphous and insensitive to Ag(S203)23- ions. Over a period of time, extensive crystal growth occurred and the membrane subsequently responded selectively to Ag(S203)23- ions between 10-5 and 10-2 M activity.Keywords: Solid-state membrane; dithiosulphatoargentate ion-selective electrode; crystalline organo-silver ion pair Investigations into solid-state membrane systems for ion- selective electrodes began with the introduction of glass membrane systems for pH electrodes. Such systems are now well established for other cations, e.g., NH4+, K+ and Na+. Other forms of solid-state membrane systems include the europium-doped single crystals of the LaF3 type,l quartz crystals of piezoelectric detector systems,2 compressed discs of microcrystalline materials (e.g., Ag& PbS) embedded in suitably inert matrices and pressed pellets of polycrystalline salts such as Ag2(Hg14)3 and Ag2(Hg14) - Ag2S.4 The most frequently reported methods of membrane preparation with polycrystalline materials involve pellet formation at high pressure and subsequent heating (annealing) of the formed membrane to very high temperatures (200-600 OC).5 Although these membrane systems work satisfactorily, problems that do not appear to have been fully resolved are associated with the need for high pelleting pressures, the provision of a suitable sealing adhesive, overcoming poor mechanical strength and leakage associated with pressed pellet membrane systems.In this paper, a technique of solid-state membrane prepara- tion based on benzyldimethyltetradecylammonium dithio- sulphatoargentate, and an evaluation of the performance of the Ag(Sz03)23--selective electrode formed from this mem- brane, is described.Experimental Reagents All chemicals used were of analytical-reagent grade, except for the Zephiramine (benzyldimethyltetradecylammonium chloride; Fluka) which was purified by recrystallisation from acetone containing a small amount of methanol. Sodium dithiosulphatoargentate [Nadg(S203)2] solution (0.005 M). A method of preparation based on that of Pourad- ier and Rigola6 was used. A 0.01 M silver nitrate solution (50 ml) was run slowly from a burette into a three-necked borosilicate glass reaction vessel containing an equal volume of 0.02 M sodium thiosulphate solution. The reactants were continuously stirred by means of a stream of nitrogen bubbles, which also maintained the reaction system under an inert atmosphere. Stirring was continued for about 2 min after the reactants had been added. The solution was then transferred into a borosilicate glass flask, stoppered and stored at room temperature in the dark until required.Benzyldimethyltetradecylammonium dithiosulphatoargen- tate. The preparation involved precipitation from an aqueous phase by mixing freshly prepared 0.005 M sodium dithiosul- phatoargentate (100 ml) and 0.005 M Zephiramine (300 ml). The precipitate was recovered by decantation and subse- quently washed several times with de-ionised, distilled water. The ion pair was further purified by extraction into chloro- form. After recovery by low-temperature removal of the solvent, the material was transferred into a clean, light-proof sample bottle and used immediately as prepared.Found: C, 60.29; H, 9.16; N, 3.08; S, 8.96; Ag, 7.48%. C69H12606N3S4Ag requires c, 62.34; H, 9.55; N, 3.16; S, 9.64; Samples were prepared for scanning in KBr disc format. The principal IR absorption bands were found to occur at the following wavelengths: 3100, 2920, 2870, 1505, 1430, 1325, 1120,1105,670 and 550 cm-1. Ag, 8.11%. Membrane Preparation The physical form of the precipitated benzyldimethyltetra- decylammonium dithiosulphatoargentate (a sticky, semi- solid) was such that it could not be pressed into a pellet. Therefore, 0.1-0.15 g of the freshly prepared material was weighed out and used to seal a hole (ca. 6 mm diameter) bored into the plastic (polyethylene) cover of a glass sample tube. A micro-spatula was used to spread and smooth the material over the hole.It was easier to do this by holding the cover (membrane support) against a clean glass slide at the same time. The process was continued until both sides of the membrane became polished and transparent. Thereafter, it was stored in a clean, dust-free cabinet in the absence of light, for a minimum period of 4 weeks, in order to allow crystallisation to occur. Electrode Assembly Two ends of a glass sample tube (7 cm diameter) were cut off. One end was fitted with a matching cover on to which the membrane had been previously sealed such that a water-tight joint was obtained. The tube was then filled to the two- thirds level with the internal reference solution [lO-3 M Na3Ag(S203),]. The other end of the tube was then fitted with a cover similar to that used above, but through which a saturated calomel reference electrode was inserted. The whole unit constituted the ion-selective electrode assembly.218 v) E > - 7 0 - .- .;r E L I j 50 ANALYST, MARCH 1987, VOL.112 2 x 10-4 M ' i I ' , ' ( I \ I \ '--- 7 5 x 10-4 M \I x 10-3 M I I I Apparatus An Orion Research Ionalyser 901 millivolt - pH meter was used for e.m.f. measurements in conjunction with a calomel reference electrode. Similarly, a glass pH combination elec- trode was used for all pH measurements with the same instrument. A Servoscribe- IS (RE541.20, Smith Industries) strip-chart recorder was used to follow the potential - time behaviour of the electrode. Results and Discussion Nature of the Membrane The ion pair was a sticky, semi-solid material when freshly prepared.When it was used in this form for membrane preparation, the latter showed no response to changes in the dithiosulphatoargentate ion activity of test samples. Micro- scopic examination of the membrane (using polarised light) revealed that when freshly prepared, it was wholly amor- phous. Crystal growth was observed during the third week, and by the fourth week, the semi-solid, amorphous material had been transformed into substantial arrays of needle crystals. Also, the ionic conductivity increased and the membrane showed a Nernstian response to changes in activity of the primary (dithiosulphatoargentate) ion. Calibration Graph Calibration graphs were prepared by plotting measured potentials (mV) against various activities of dithiosulphato- argentate test solutions using the electrode assembly de- scribed above.Within the range 1 x 10-5-1 x 1 0 - 2 ~ , the electrode exhibited Nernstian behaviour with a slope charac- teristic of 20 mV decade-'. The limit of detection of dithiosulphatoargentate was found to be 9.2 x 1 0 - 6 ~ . From interference studies conducted on electrode performance, it appeared that competition from free thiosulphate and sul- phide ions present in the sample solution through side- reactions such as dissolution, oxidation or reduction occurring on the membrane surface could shorten the linear-response range and adversely affect the limit of detection. Response Time and Potential Stability The practical response time (ts5) of the electrode measured in constantly stirred primary ion solutions of various activities was found to be ca.15 s. However, this response time became longer as time progressed, especially beyond the second week of the life of the electrode. Potential stabilities measured in l o - 4 ~ dithiosulphatoargentate over a period of 2 h had a mean value of 8 mV d-1. The reference electrode was checked for drift at regular intervals, as this could be mistaken for or added to electrode instability. Towards the end of the useful life of an electrode, appreciable drift was observed in the electrode potential. Effect of Stirring This was investigated at various activities of primary ion (10-L10-6 M dithiosulphatoargentate). At a stirring rate of about 20 rev min-1 using a magnetic stirrer (Tecam MSI) the developed potentials were higher relative to those obtained in unstirred solutions over the range of activities investigated.Practical response times (ts5) of the electrode were observed to be affected by stirring, being generally of shorter duration in stirred solutions. Interference Effects The possible effects of Ag(S203)-, S2032-, S2- and C1- as interferents on the electrode performance were studied by the fixed interference method.7 Selectivity coefficients calculated using the appropriate equation7 are presented in Table 1. Free thiosulphate ions (S2032-) showed some competition in addition to sulphide ions (Sz-) whereas monothiosulphato- argentate ions interfered significantly with the electrode response. Chloride ions had no adverse effect on the electrode response, except for a seemingly higher potential stability which was observed in more dilute solutions (lo-5-lo-6~).When it was apparent that the dynamic response time of the electrode was affected by interferents, notably Ag(S203)- and S2O32-, a strip-chart recorder was coupled to the millivolt- meter so that the potential - time behaviour could be closely studied. By comparison of typical outputs presented in Fig. 1, the sluggish response of an electrode in the presence of Ag(S203)- is evident. Whereas typical response times of 12 s represent normal behaviour, these are increased about three-fold (to 35 s) in the presence of monothiosulphatoargen- tate. pH Effects Standard sample solutions were prepared, having been adjusted to various pH values by the addition of dilute sodium hydroxide solution or hydrochloric acid. Changes in potential of the electrode as a function of pH at two activity levels of dithiosulphatoargentate ion, viz., 10-3 and 10-4 M, are shown in Fig.2. At very low pH values, free thiosulphate ions and ionic silver were detected in the sample solutions. The membrane is thought to be the probable source of these ions, through dissociation of the membrane surface in an acidic medium. At higher pH values, particularly between 7 and 12, the electrode response was found to be more reliable. 1-A Concentration (B) ~~ . . . . . . . . . . 1 x 1 0 - 4 ~ $ 9 0 1 , 0 v) Time/s Fig. 1. Dynamic response behaviour of dithiosulphatoargentate electrode in the presence of monothiosulphatoargentate ions (inter- ferent).A, (solid line) normal potential - time behaviour of Ag(S203)23--responsive electrode (dynamic response times fn) and B , (broken line) potential - time behaviour of A in the presence of AgS203--interferent between 10-4 and 10-3 M (dynamic response times ti) Table 1. Selectivity coefficients Interferent , X . . . . . . AgS2032- s2032- S2- C1- Molar concentration of X 1 x 10-4 1 x 10-4 1 x 10-4 1 x 10-4 E;(s203)2-,X * . . . 52 3.8 1 x 10-1 No interferenceANALYST, MARCH 1987, VOL. 112 219 I I I I I 1 I I 1 1 3 5 7 9 11 13 PH Fig. 2. Effect of H on electrode response. A, 10-3 M Ag(S203)*3- and B, M Ag!S2O3),3- Application The silver content of a photographic fixing bath is a good preliminary guide to its degree of exhaustion, in addition to indicating how much silver may be recovered from it.8 Trivalent dithiosulphatoargentate ions have been found to be the predominant species in freshly constituted liquors.9 In a simulated photographic fixing bath, preliminary investigations with the currently described electrode have shown that it is possible to monitor silver as dithiosulphatoargentate at levels of up to 1.5 X 1 0 - 2 ~ with a mean error of <+2.58% using a pre-calibrated electrode and the Orion 901 Ionalyser set to provide concentration output directly, having been pre- programmed with the relevant calibration data, including the slope characteristic derived for that particular electrode.Using a standard additions technique and sodium sulphate as an ionic strength adjuster, typical correlation plots reveal good linearity with correlation coefficients r > 0.991.Conclusion A solid-state membrane system based on benzyldimethyl- tetradecylammonium dithiosulphatoargentate, prepared as described above, has been found to respond selectively to Ag(S203)23- ions in a Nerstian manner between 10-2 and 1 0 - 5 ~ . Advantages are the ease and simplicity of the preparation of a reponsive membrane, its liquid leak-proof nature and its improved mechanical strength over contempor- ary membranes. The wider applicability of this membrane preparation technique depends on finding electroactive materials capable of being converted from a sticky, semi-solid state to that of a crystalline, ionically conducting solid of appreciable mechanical strength. 1. 2. 3. 4. 5. 6. 7. 8. 9. References Frant, M. S., and Ross, J. W., Science, 1966, 154, 1553. Guilbault, G. G., Zon-Sel. Electrode Rev., 1980, 2, 4. Gordievski, A. V., Zhukov, A. F., Shterman, V. S., Sarvin, N. I., and Vrusov, Y. I., Zh. Anal. Khim., 1974,29, 1414. Sekerka, I., and Lechner, J. F., Analyst, 1981, 106, 323. Mascini, M., and Liberti, A., Anal. Chim. Acta, 1973, 64, 63. Pouradier, M. J., and Rigola, J., C. R. Acad. Sci., 1972, 275, 515. IUPAC Recommendations, Pure Appl. Chem., 1976,48,127. Russell, G., “Chemical Analysis in Photography,” Focal Press, London, 1965, p. 226. Mees, C. E., in James, T. H., Editor, “The Theory of the Photographic Process,” Fourth Edition, Macmillan, New York, 1977, p. 441. Paper A61254 Received July 28th, 1986 Accepted November 5th, 1986

ISSN:0003-2654    

DOI:10.1039/AN9871200217

出版商:RSC    

年代:1987

数据来源: RSC

摘要:

ANALYST, MARCH 1987, VOL. 112 221 Determination of Ultra-trace Amounts of Cadmium in Natural Waters by the Combination of a Solvent Extraction Procedure and Anodic Stripping Voltammetry Samuel B. Adeloju and Kathryn A. Brown Trace Analysis Unit, Division of Chemical and Physical Sciences, Deakin University, Victoria 32 7 7, Australia A method combining a solvent extraction procedure with anodic stripping voltammetry for the reliable determination of ultra-trace amounts of cadmium in natural waters is reported. The element is pre-concentrated as cadmium dithiocarbamate into Freon and subsequently back-extracted into an acidic aqueous medium for the voltammetric determination. The utilisation of the pre-concentration procedure prior to the determination substantially reduces the required plating time for the electrochemical technique and hence permits a considerable reduction in the over-all analysis time.The separation of the analyte from the bulk of the natural water matrix permits the direct determination of 90.1 pg 1-1 of the element from a calibration graph, whereas for lower concentrations the use of the standard additions method is necessary for reliable determination. The minimum amount of cadmium that can be pre-concentrated and determined reliably by the voltammetric technique is 0.025 yg I-'. The application of the over-all approach to an NBS Standard Reference Water (fresh water) and local sea water samples was successful. Keywords: Cadmium determination; anodic stripping voltammetry; solvent extraction; natural waters The determination of trace elements in aquatic environments has become increasingly important in the understanding of some aspects of chemical oceanography and limnology.1 However, most of the key elements are present at ultra-trace levels and can only be reliably determined by relatively few analytical techniques. Also, in some instances the adequacy of some of the available techniques for the direct determination of these elements is hampered by the nature of the sample matrix.For example, the use of atomic absorption spec- trometry (AAS) for the direct determination of trace elements in sea water is limited by the high salt content, which interferes significantly with the technique, and separation - pre-concen- tration procedures are therefore required for adequate determination. However, the sea water matrix does not interfere with voltammetric determinations but rather acts as a suitable supporting electrolyte for these techniques.As a result, anodic stripping voltammetry (ASV) is often con- sidered the best analytical technique for the direct determina- tion of the key elements in sea water and is ideal for chemical speciation in natural waters as no preliminary separation - pre- concentration is required.l-5 Unfortunately, however, other analytical techniques that employ a separation - pre-concentration procedure are often preferred over the direct ASV method for the determination of the total concentration of trace elements in sea water because of the considerable reduction in analysis time. Typically, the direct simultaneous determination of four elements such as zinc, cadmium, lead and copper in sea water by ASV at a hanging mercury drop electrode (HMDE), usually based on triplicate measurements with deposition times of 2&30 min each and determination by three standard additions, takes more than 6 h per sample (or approximately 90 min per element).This is considerably longer than the 25-40 min per element that can be achieved for a large number of samples by other analytical techniques based on the separation - pre-concentration of the analyte.6-9 Although it may also be argued that the pre-concentration procedure employed in the latter approach is time consuming, the distinct advantage in this instance is that the time contribution is considerably reduced with an increasing number of samples, whereas the opposite is true for the electrochemical pre- concentration used in direct ASV determinations.The main time-consuming step in the ASV method is the time required to plate a suffient amount of the ultra-trace levels of the analytes into the mercury electrode (deposition step). Hence, a substantial reduction in the over-all analysis time can be accomplished either by increasing the electrode surface area or by pre-concentration of the analyte prior to the ASV determination, as is presently carried out in AAS and several other analytical techniques .6-9 It has already been demonstrated that the use of electrodes with larger surface areas than the HMDE, such as glassy carbon and mercury film electrodes, and the use of a rapid staircase stripping waveform result in some reduction of the analysis time.lOJ1 However, it is still of interest to establish whether a similar reduction in the analysis time can be accomplished by use of a pre-concentra- tion - separation procedure with the ASV determination, providing that the problem of contamination can be readily overcome. Also, if possible, the approach will be useful in making direct comparisons with other analytical techniques for trace element determinations in natural waters.Generally, whereas various pre-concentration - separation procedures such as solvent extraction, ion exchange and coprecipitation or co-crystallisation~~9J2 are commonly used with other analytical techniques for the determination of trace and ultra-trace levels of elements in natural waters, their use with voltammetric techniques has been limited.13-15 More significantly, in instances where such pre-concentration - separation procedures have been employed, none seem to have resolved the problem of the long analysis time associated with the deposition step of the ASV method. For example, a solvent extraction procedure employed by Joyner et al. 14 for the ASV determination of some trace elements requires the evaporation of the organic solvent and subsequent wet ashing of the residue, both of which also contribute to the over-all analysis time as does the long deposition time employed for the direct determination. It is evident, therefore, that some work still needs to be carried out in order to accomplish a similar reduction in the ASV analysis time by use of a suitable pre-concentration procedure as those presently employed with the other analytical techniques.This paper reports a method that combines a solvent extraction procedure with ASV for the reasonably rapid and reliable determination of the total concentration of cadmium222 ANALYST, MARCH 1987, VOL. 112 in natural waters. The method employs an acid back-extrac- tion step for the removal of the analyte from the organic phase and thus eliminates the need for the evaporation of the organic solvent and subsequent wet ashing, as was necessary in the work reported by Joyner et a1.14 The effectiveness of the pre-concentration method in reducing the required plating time for the ASV determinations in natural waters and the possibility of direct determination of cadmium from a calibra- tion graph rather than by the time-consuming standard additions method were carefully examined.Cadmium was chosen for this study as it has recently been demonstrated that the atmospheric contribution of the element is minimal under most laboratory conditions16 and the likelihood of contamina- tion is therefore minimised. The adequacy of the analytical approach for the determination of ultra-trace amounts of cadmium was assessed by use of an NBS Standard Reference Water, prior to its application to some local sea water samples. Experimental Reagents and Standard Solutions All acids used were Aristar grade purity (BDH Chemicals) and all other reagents and organic solvents were of analytical- reagent grade purity.Acetic acid - acetate buffer solution (2 M) was prepared by mixing equal volumes of 8 M acetic acid and 4 M ammonia solution. A 4 M buffer solution was similarly prepared from 16 M acetic acid and 8 M ammonia solution. A stock solution (1 g 1-1) of cadmium(I1) was obtained from BDH Chemicals. The required standard (1 mg 1-1) was prepared weekly by the appropriate dilution of this stock solution with 0.1 M hydrochloric acid. A chloride solution was prepared by dissolving 3.5 g of sodium chloride (Merck, Suprapur) in 100 ml of water and this was used to examine the possible effect of the salt content of sea water on the extraction procedure. Distilled, de-ionised water prepared as previously described16 was used for all sample and solution preparations.Instrumentation An EG and G Princeton Applied Research (PAR) micro- processor-based polarographic analyser (PAR Model 384) equipped with a PAR Model 303 static mercury drop electrode and a PAR Model 305 stirrer was used to record all stripping voltammograms. The electrode compartment consisted of a hanging mercury drop electrode, a silver - silver chloride (saturated KC1) electrode and a platinum wire electrode as its working, reference and auxiliary electrodes, respectively. Solution pH measurements were made on an Activon portable pH meter. Standard solutions of cadmium(I1) were added to the polarographic cell with fixed volume Soccorex micro- pipettes with disposable tips. Glassware All glassware, Teflon and polyethylene bottles were soaked in 2 M nitric acid for at least 7 d and rinsed several times with distilled, de-ionised water prior to use.Between experiments, the used glassware and bottles were soaked in 2 M nitric acid for at least 12 h and again rinsed several times with distilled, de-ionised water before use. Pudkation of Organic Solvents, Chelating Agent and Buffer Organic solvents Chloroform, Freon and trichloroethylene were purified by shaking 100 ml of each solvent with 50 ml of 2 M nitric acid in a separating funnel for 10 min. The organic layer was collected and stored in a pre-cleaned bottle and the aqueous layer was discarded. Chelating agent A solution of sodium diethyldithiocarbamate (1% m/v) was prepared by dissolving 1 g of the salt in 100 ml of distilled, de-ionised water and filtered through a Whatman No.541 hardened ashless filter-paper. The solution was then purified by shaking with 20 ml of chloroform for 2 min and was used within approximately 1 h of preparation. Beyond this period, it was observed that the recovery efficiency for cadmium was reduced to 4 0 % owing to the rapid decomposition of sodium diethyldithiocarbamate in aqueous solution. However, attempts to use the more stable ammonium tetramethylene- dithiocarbamate proved to be unsuccessful, possibly owing to its higher solubility in the organic solvents, which resulted in serious interference problems with the voltammetric determi- nation of cadmium. It is therefore recommended that the sodium diethyldithiocarbamate solution be used for the pre-concentration as described in this work within 1 h of preparation.Buffer solutions The buffer solutions were purified by adding 12.5 ml of the purified chelating agent into 500-ml portions and shaking for 2 min with three successive 20-ml portions of purified chloro- form. After each shaking, the organic layer is discarded and a fresh solution of chloroform is added. Natural Water Samples A fresh water sample (NBS Standard Reference Water) was obtained from the US National Bureau of Standards, Wash- ington, DC. The sample was preserved as recommended by the supplier. Also, two sea water samples were obtained from an industrialised area (Corio Bay) in Geelong, Victoria, Australia. One of these samples was immediately acidified to pH 2 with nitric acid whereas the other was retained at the natural pH of 8.2.Both samples were analysed immediately after collection. A sea water sample was also obtained from an unpolluted source in Queenscliff, Victoria, Australia. Solvent Extraction Procedure A 50-ml aliquot of each sample solution (blank, standard or unknown) was transferred, by pipette, together with 5 ml of 2 M buffer and 2 ml of chelating agent into a 250-ml separating funnel. The mixture was shaken gently to mix, then 20 ml of organic solvent was added and the funnel was shaken continuously for 2 min in order to enable the extraction of metal chelate(s) into the organic layer. The funnel was then left to stand for 2 min to permit phase separation and, after this period, most of the organic layer (lower layer) was run off into a 50-ml separating funnel.A fresh 10-ml volume of organic solvent was then added into the original 250-ml separating funnel and again shaken for 1 min. After phase separation, the organic layer was combined with the previous portion in the small separating funnel and 400 pl of concen- trated hydrochloric acid were added. The mixture was then shaken for 20 s and after a period of 10 min, 9.6 ml of distilled, de-ionised water were added. The funnel was again shaken for another 20 s and, after 10 min, the organic layer was run off and discarded. The retained aqueous layer was collected into a 10-ml polyethylene bottle and an aliquot of this was taken for the ASV determination. Stripping Voltammetric Determinations An aliquot (5 ml) of the acid extract was transferred into the polarographic cell, de-gassed for 5 min and maintained under a flow of nitrogen during the experiment.Cadmium was then determined by ASV at the HMDE using the following conditions: operating mode, differential-pulse stripping (DPASV); deposition potential, -0.80 V vs. Ag - AgClANALYST, MARCH 1987, VOL. 112 223 (saturated KC1); final potential, -0.45 V; scan rate, 8 mV s-1; duration between pulses, 0.5 s; modulation amplitude, 50 mV; deposition time, 285 s (stirred); equilibration period, 15 s (unstirred). The deposition of cadmium into the mercury electrode was achieved by using a fast stirring rate and a medium-sized drop with a surface area of 0.015 cm2. In some instances, the large-sized mercury drop with a surface area of 0.024 cm2 may be used, but this can be readily dislodged if a long deposition time (>120 s) is employed, prior to the stripping step, and is therefore not recommended. The amount of cadmium present in the natural water samples was determined from a calibration graph prepared under the same conditions using the stripping peak which appeared at about -0.6 V vs.Ag - AgCl. In some instances, the recovery efficiency of the extraction procedure was determined by the standard additions method. The solution was de-oxygenated after each addition for 30 s prior to the ASV measurement. Working Area All reagent preparations, purifications and extractions were carried out in a Class 100 clean room controlled at a temperature of 22.5 _+ 0.5 “C. All stripping voltammetric determinations were carried out in a Class 1000 clean room under similar temperature control. Both of these laboratories form part of the Deakin University Trace Analysis Unit.Results and Discussion Choice of Organic Solvent A very important consideration in the employment of a solvent extraction procedure for the voltammetric determina- tion of trace elements is the choice of a suitable organic solvent. It is well known that such voltammetric techniques are generally prone to interference by organic constituents in aqueous media,lJ7 and its avoidance or minimisation in the final acid extract is therefore critical for reliable determination by this approach, The two possible sources of organic contamination of the final acid extract are (i) from the chosen organic solvent and (ii) from the chelating agent.Whereas the organic contamination from the chelating agent may be readily overcome or minimised by avoiding the use of an excessive amount of chelating agent, the choice of a suitable solvent is more critical in this regard. Consequently, an investigation of the suitability of three solvents for the combined solvent extraction - voltammetric determination of cadmium was undertaken. The three solvents considered were chloroform, Freon and trichloroethylene, which have been used by other workers in conjunction with AAS. 1t3-20 Being “heavy” solvents (specific gravity >1.0) they all form the lower extraction phase, thus permitting minimum sample handling and consequently less likelihood of sample contamination or loss of analyte. Fig. 1 shows that the three solvents were suitable for the five-fold pre-concentration and reliable voltammetric determination of 1 pg 1-1 of cadmium.However, the sensitivity and resolution of the voltammetric peaks were greatly influenced by the different organic solvents employed in the extraction pro- cedure. The resolution of the cadmium peak was only affected when chloroform was employed as the organic phase for the extraction. The cause of this is not exactly known, but it is thought to be associated with the carry-over of either some of the chloroform or the chelating agent during the acid back- extraction. The resulting broadness and slanting of the cadmium peak in this instance made it difficult to obtain reliable peak-height measurements. For Freon and tri- chloroethylene, the resolution and the base line of the peaks were excellent for accurate measurements. Although the extraction of cadmium (as dithiocarbamate) into Freon produced the most sensitive peak, the associated I I 0.8 0.7 0.5 - 1.8 0.7 0.5 - E N - 0.8 0.7 0 Fig.1. Typical stripping peaks for cadmium extracted with different organic solvents: (a) chloroform; ( b ) Freon; and (c) trichloroethylene. Volume of chelating agent = 5 ml of 1% mlV; 1 pg 1-1 of Cd; concentration factor = 5 reproducibility, from sample to sample, is not as good as those with chloroform and trichloroethylene. This clearly reflects the greater tendency for the solvent or chelating agent to interfere with the voltammetric determinations when Freon is employed in the extraction procedure. Surprisingly, the best recovery efficiency, based on the extraction of six separate samples at the 1 pg 1-1 level for each solvent, was also obtained with Freon (100 k 1%) compared with chloroform (80%) and trichloroethylene (95 k 5%).These results indicate that in spite of the observed peak current variation with Freon, complete recovery of cadmium was still accomplished with the use of 5 ml of the chelating agent. These observations seem to suggest that the presence (or carry-over) of some of the chelating agent in the acid extract may be reponsible for the variation. Evidently, the volume (5 ml of 1% mlV) of the chelating agent used in the extraction procedure may be far in excess of the required amount. However, it is conclusive that the variations caused by the excess amount of the chelating agent in the example of Freon does not have any direct influence on the quantitative recovery of the analyte.There- fore, on the basis of the sensitivity of the cadmium peak and recovery efficiency, Freon was chosen as the ideal solvent for the combined solvent extraction - voltammetric determination of cadmium and the factor responsible for the poor reproduci- bility with this solvent was further investigated. Other beneficial factors that influence the choice of organic solvent are the very low mutual solubilities of Freon and water, rapid phase separation and the relatively low human toxicity of this solvent. Influence of Chelating Agent Concentration The only other factor that may be responsible for the peak current variations observed in the cadmium extractions with Freon is the amount of the chelating agent used.The presence of an excessive amount and/or the higher solubility of the chelating agent in the organic solvent may result in its carry-over into the acid extract. This, in turn, may result in some interference with the voltammetric determination and could affect the peak-current measurements to various degrees, from one sample to the other, depending on the amount of the chelating agent present in the final acid extract. As a result, the possible effect of an excessive amount of chelating agent on the voltammetric determination was examined by varying the amount used for the solvent extraction.224 ANALYST, MARCH 1987, VOL. 112 Fig. 2 shows the precision obtained for cadmium peak measurements using various amounts of chelating agent for the extraction procedure.Evidently, the smaller volumes (<3 ml) gave much better reproducibility than volumes of 5 ml of the chelating agent and this indicates that the organic solvent (Freon) may not be responsible for the interference with the voltammetric determination of cadmium. It is conclusive, therefore, from the results in Fig. 2, that the excess amount of chelating agent is not necessary, as reflected by the identical mean peak currents and the fact that the carry-over of the excess in the aqueous phase during the acid back-extraction is possible. The use of S3 ml of 1% mlV sodium diethyldithio- carbamate in the extraction procedure was therefore recom- mended for the reliable determination of cadmium.Determination of the amount of cadmium extracted with the different amounts of chelating agent by the method of standard additions gave recovery efficiencies of 100 If: 1% for 5 ml and 95 h lo/' for 1-3 ml of 1% mlVsodium diethyldithio- carbamate. It is interesting to note that such high recoveries can be obtained with considerably less chelating agent than is usually employed in other analytical techniques. On the basis of sensitivity and reproducibility, it was decided that the use of 2 ml of the chelating agent is adequate for the reliable determination of ultra-trace amounts of cadmium in some natural waters and, hence, may enable determination by direct calibration. Also, Fig. 2 shows that the peak current variations were considerably reduced when smaller volumes of the chelating agent were used.Direct Calibration The usual method for the determination of trace elements in natural waters is by standard additions which, ideally, compensates for both the influence and variation of the matrix composition on the measurements. However, when employed for the determination of ultra-trace amounts of elements with ASV, the approach becomes rather time consuming, as discussed earlier. In some instances, the use of a single addition of the standard is employed in determining the analyte, but, whereas this approach may reduce the time required for the ASV analysis, it may also yield misleading results depending on the influence of the matrix components on the linearity of the analytical response with increasing concentration and is therefore not recommended. 15 The pre-concentration of cadmium from the bulk of the sample matrix by the solvent extraction procedure provides an alternative and reliable means of matching both the final sample and standard matrices, thus enabling determination by use of a calibration graph prepared under identical conditions.Such an approach has been successfully used for the determi- nation of selenium in complex biological matrices by the use of an ion-exchange procedure with cathodic stripping voltam- metry.15 However, the reliable utilisation of the direct calibration approach requires very good reproducibility from the measurement of one sample to the other, typically of the order of d +5%. In this work, the use of 2 ml of the chelating agent in the extraction procedure gave an acceptable repro- ducibility and is therefore adequate for the reliable determina- tion of ultra-trace amounts of cadmium in natural waters by the direct calibration approach.Fig. 3 shows that, with the use of 2 ml of the chelating agent for the extraction of the analyte from the chloride solution (3.5% mlV) a linear response was obtained up to 1.00 yg 1-1. At concentrations S0.05 yg 1-1, the reproducibility of the sample measurements was poor and was, therefore, not included in the calibration graph. Evidently, the solvent extraction procedure can only be used with ASV for the determination of cadmium in natural waters by the direct calibration approach at concentrations 20.10 yg 1-1. Hence, for lower concentrations, the use of the standard additions method is necessary for reliable determinations.As little as T I I I I I 1 I I 1 0 1 2 3 4 5 6 Volume of chelating agentiml Fig. 2. Influence of chelating agent volume on the reproducibility of cadmium stripping peaks. Solvent, Freon; other conditions as in Fig. 1 0.0 0.5 1 .O Before extraction (0.0) (2.5) (5.0) After extraction Cd/pg I-' Fig. 3. Calibration graph for direct quantification of cadmium in natural waters 0.025 yg 1-l of cadmium can be pre-concentrated by the solvent extraction procedure and be adequately measured by ASV with only a 5-min plating time. This represents a considerable reduction in the required deposition time for the determination of cadmium in most unpolluted natural waters and can readily be used with the standard additions method for reasonably rapid analysis.It may also be possible to extend the use of the direct calibration approach to these levels either by use of a longer deposition time or a higher pre-concentra- tion factor. These possibilities are currently being investi- gated. The use of the direct calibration method for >1 pg 1-1 of the element was not considered as these concentrations can readily be determined by direct ASV (without extraction) with no more than a 5-min deposition time. Alternatively, such samples can be made amenable for determination with the direct calibration method by dilution of the cadmium concen- tration to within 0.10-1.00 yg 1-1, without any serious contamination problem. Application to Natural Water Samples The calibration graph obtained in this work (Fig.3) by the solvent extraction procedure can be used for the reliable determination of cadmium concentrations in some natural waters. The adequacy of such a direct calibration approachANALYST, MARCH 1987, VOL. 112 225 Table 1. Concentrations of cadmium found in some natural water samples No. of Direct calibration Standard additions Certified value/ Sample determinations concentratiodpg 1-1 concentratiodpg 1-1 I.18 I-' NBS fresh water* . . . . 6 10.1 k 0.3 9.5 2 0.4 10.0 k 1.0 Extracted sea water . . 6 0.35 k 0.01 0.34 k 0.02 Unextracted sea water . . 3 0.33 f 0.04 * Diluted by a factor of 25 and extracted using a 4 M buffer solution. Table 2. Effect of the solvent extraction procedure on the ASV determination of natural levels of cadmium in sea water.Results obtained for the same sea water sample from an unpolluted source (Queenscliff, Victoria, Australia) Direct ASV ASV with extraction Deposition Concentration Concentration time/s idnA of Cd*/pg 1-1 idnA of Cd*/pg 1-1 300 1.14 0.036 10.29 0.053 600 2.29 0.038 22.86 0.055 1200 5.71 0.051 57.14 0.054 1800 10.86 0.055 89.29 0.055 * Quantified by standard additions method for separate portions of sea water sample. was verified by the application of the over-all method to some diluted aliquots of the NBS fresh water sample. Table 1 shows that the results obtained by the direct calibration approach for this water agreed favourably with the certified value. Simi- larly, the determination of the extracted samples by the standard additions method gave results comparable to both the direct calibration and certified values.Undoubtedly, the direct calibration approach is adequate for the reliable determination of ultra-trace amounts of cadmium in some fresh water samples. Moreover, the successful application of the calibration graph (prepared with the chloride solution) for the precise and accurate determination of cadmium in the fresh water sample suggest that the proposed direct calibration method is adequate for both saline and non-saline waters. The adequacy of the direct calibration method was also assessed by the application of the method to sea water samples collected from an industrial area in Victoria (Corio Bay). The results in Table 1 again indicate that both the direct calibration and standard additions values agreed favourably.However, an interesting observation in this instance was that the results obtained for both acidified and non-acidified sea water samples were identical. It appears therefore that there is no need for the acidification of the sea water samples, provided that the determination is carried out immediately after collection. The use of sodium diethyldithiocarbamate as the chelating agent in the extraction procedure is thought to be responsible for this observation. This powerful chelating agent can readily displace most natural complexing agents that may be bound to cadmium in both the acidified and non-acidified sea water. It is recommended, however, that the sample be acidified if it is to be kept over a period of days before analysis in order to prevent any loss of analyte by adsorption on to the container walls at the natural pH.Generally, the use of the solvent extraction procedure prior to the ASV determination of cadmium at the ultra-trace levels enabled a considerable reduction in the required deposition (or plating) time. In addition, the use of the calibration graph for the quantification of 20.1 pg 1-1 of the element enabled a further reduction in the over-all analysis time. The results in Table 2 show that at least a 20-30 min deposition time was required for the reliable determination of natural levels of cadmium in sea water by the conventional ASV method, whereas only a 5 min deposition time was adequate in achieving a similar sensitivity with the use of the solvent extraction procedure. The use of the longer deposition time with the solvent extraction procedure was useful only in enabling a more sensitive determination of the element than by the direct ASV method (without extraction), but was not necessary as indicated by the results in Table 2.Although it could be argued that the solvent extraction procedure is long and tedious, the required time for the pre-concentration diminished considerably with an increase in the number of samples. On its own, the time required by the solvent extraction procedure is minimal. At least six or more samples can be extracted by this procedure with the aid of an automatic sample shaker in 90 min. Therefore, if used in this way, this step will contribute no more than 15 rnin per sample in the over-all analysis time. Even if the determination of cadmium in a single sample is considered, which rarely occurs, the extraction and determination by three standard additions with triplicate measurements may be completed in <2.5 h.This represents a considerable time saving of at least 2 h over the conventional ASV method which requires about 4.5 h for the determination of natural levels of the element in sea water at the HMDE with a 20 min deposition time. A similar reduction in analysis time is applicable to other electrodes with larger surface areas. A more substantial reduction in analysis time may be accomplished by the proposed method for samples containing 30.1 yg 1-1 of the element by use of the calibration graph. In this instance, the over-all analysis time is reduced to 35 rnin per sample, based on the extraction of six samples.It is conclusive, therefore, that with the use of the direct calibra- tion approach a similar reduction in the over-all analysis time to those obtained with the other preferred analytical tech- niques is also possible with ASV. The possibility of lowering the applicable range of the direct calibration approach is currently being investigated. Conclusion The combined solvent extraction - stripping voltammetric method provides a reliable and reasonably rapid approach for the determination of ultra-trace amounts of cadmium in natural waters. The method gives a reproducibility with a relative standard deviation of 2 4 % for the determination of 20.1 pg 1-1 of the element. The only evident limitation at present is that the proposed direct calibration approach is only applicable to some natural waters where the cadmium concentration is 20.1 pg 1-1.In this regard, the direct calibration approach is best suited for the determination of the element in polluted natural waters. Nevertheless, with the use of the standard additions method, naturally low levels of cadmium may be determined by this approach with some saving of time over the direct ASV method. The minimum amount that can be determined in this way is 0.025 yg 1-1. The total analysis time of 35 min required per sample by the ASV226 direct calibration approach compares favourably with those of other preferred analytical techniques. However, when the method of standard additions is employed with the extraction procedure, the total analysis time (based on the extraction of six samples) is increased to 75 min per sample, although this is still substantially faster than the required analysis time (270 min) for the direct ASV determination of the natural level of cadmium in a single sea water sample. More work is being undertaken to lower the applicable concentration range of the direct calibration method reported in this study. 1. 2. 3. 4. 5. 6. 7. References Nurnberg, H. W., Anal. Chim. Acta, 1984, 164, 1. Batley, G. E., Mar. Chem., 1983, 12, 107. Mart, L., Nurnberg, H. W., and Rutzel, H., Fresensius Z. Anal. Chem., 1984,317, 201. Scarponi, G., Capodaglio, G., Cescon, P., Cosma, K., and Frache, R., Anal. Chim. Acta, 1982, 135,263. Batley, G. E., andmorence, T. M., Electroanal. Chem., 1976, 72, 121. Kinrade, J. D., and Van Loon, J. C., Anal. Chem., 1974,46, 1894. Danielsson, L. G., Magnusson, B., and Westerlund, S., Anal. Chim. Acta, 1978, 98, 47. 8. 9. 10. 11. 12. 13. 14. 15. 16. 17. 18. 19. 20. ANALYST, MARCH 1987, VOL. 112 Armannsson, H., Anal. Chim. Acta, 1979, 110,21. Yu, J. C., Hutchison, F. I., and Wai, C. M., Anal. Chem., 1982,54,2536. Batley, G . E., and Florence, T. M., J. Electroanal. Chem., 1974, 55,23. Florence, T. M., Anal. Chim. Acta, 1980, 119,217. Jan, T. K., and Young, D. R., Anal. Chem., 1978, 50, 1250. Zirino, A., “Marine Electrochemistry,” Wiley, New York, 1981, Chapter 10. Joyner, T., Healy, M. L., Chakravarti, D., and Koyanagi, T., Environ. Sci. Technol., 1967, 1, 417. Adeloju, S. B., Bond, A. M., Briggs, M. H., and Hughes, H. C., Anal. Chem., 1983,55, 2076. Adeloju, S. B., and Bond, A. M., Anal. Chem., 1985,57,1728. Adeloju, S . B., Bond, A. M., and Noble, M. L., Anal. Chim. Acta, 1984, 161, 303. Rasmussen, L., Anal. Chim. Acta, 1981, 125, 117. Magnusson, B., and Westeriund, S., Anal. Chim. Acta, 1981, 131, 63. Smith, R. G. Jr., and Windom, H. L., Anal. Chim. Acta, 1980, 113,39. Paper A61186 Received June 6th, 1986 Accepted September 29th, 1986

ISSN:0003-2654    

DOI:10.1039/AN9871200221

出版商:RSC    

年代:1987

数据来源: RSC

摘要:

ANALYST, MARCH 1987, VOL. 112 227 Determination of Lead and Cadmium in Urine by Differential-pulse Anodic Stripping Voltammetry* A. Nur Onar and Aytekin Temizer Department of Analytical Chemistry, Faculty of Pharmacy, Hacettepe University, 06100 Ankara, Turkey Differential-pulse anodic stripping voltammetry was used to determine Cd and Pb in urine. A hanging mercury drop electrode was chosen as the working electrode. Cd and Pb were determined separately after a digestion procedure and an additional pre-electrolysis step was used prior to the determination of Cd. Two deposition times and two standard additions were used to determine the concentrations of the metal ions. Keywords: Differential-pulse anodic stripping voltammetry; biological material; cadmium determination; lead determination; urine analysis The growing realisation of the biochemical importance of toxic metals has given rise to numerous investigations of these metals in biological materials.Lead and cadmium are two toxic metals deserving particular attention. Although there does not appear to be a simple relationship between exposure to toxic metals and their urinary excretion, the analysis of urine is a convenient first step in confirming exposure to heavy metals. 1 Differential-pulse anodic stripping voltammetry (DPASV) is relatively inexpensive and is one of the most sensitive and selective techniques used in the determination of trace amounts of metals at natural levels.2 In DPASV metal ions are reduced and amalgamated at a hanging mercury drop elec- trode (HMDE) or a mercury-film electrode (MFE) during pre-electrolysis at a suitable applied potential.The reduced metals are then re-oxidised by means of a potential ramp imposed between the working electrode and a platinum counter electrode.3 This technique is, however, affected by water-soluble proteins present in urine samples and the possibility of intermetallic interferences also needs to be considered. Intermetallic compound formation predomi- nantly occurs when using an MFE, but is also observed when using an HMDE.4 The HMDE is widely used as a working electrode in anodic stripping voltammetry (ASV). It was one of the first elec- trodes to be developed for this purpose and is still often considered to be the best electrode in such determinations.5 In 1961 Kemula and Kublik6 used ASV to determine lead in the freshly voided urine.Several workers have since reported polarographic or voltammetric procedures for the determina- tion of Pb, Cd and T1 in urine.7 Although direct determination in urine has been performed, it is accepted that digestion of the urine sample is necessary because of the presence of water-soluble proteins. A disadvantage of the decomposition procedure is the risk of contamination of the sample by acids.1 Freeze-drying of the urine sample before digestion decreases the contamination and also decreases the amount of acid required for the digestion.7 Potentiometric stripping analysis has been shown to be applicable to the direct determination of metals in urine and this technique has been further developed using microcomputers .3,8 In this study Cd and Pb were determined in urine samples using the DPASV technique with the HMDE as a working electrode.The application of an additional pre-electrolysis step before the determination of Cd was introduced in order to decrease the effects of water-soluble proteins and intermetal- lic interferences in the voltammetric determination. A modi- fied standard additions technique was also introduced; this * This work is taken partly from the PhD Thesis of A. N. Onar and was presented at the 2nd National Chemistry Symposium at METU, Ankara, Turkey, 18-20 September, 1985. technique was recently developed for the determination of metals by ASV.9 The accuracy and precision were greatly increased by the use of this technique. Experimental Apparatus The DPASV was performed with a PAR Polarographic Analyzer (174A) fitted with an X - Y recorder (Omnigraphic Model 2000).The working electrode was a Metrohm 410E HMDE and a standard calomel electrode (SCE) was used as the reference electrode; a platinum coil served as the counter electrode. A chronometer (Sprint, Switzerland) was used in the control of the different stages of the stripping voltam- metric procedure. A magnetic stirrer with a synchronous motor and a Teflon-covered bar were also used. The solutions in the electrochemical cell (PAR 6062) were de-aerated prior to the voltammetric determination by passing purified nit- rogen through for 15 min. During the experiments nitrogen was passed over the solution. The interior of the HMDE capillary required treatment with dimethyldichlorosilane at regular intervals, as recommended by the manufacturer.The remaining apparatus was as previously described. 10 The urine samples were decomposed in the acid digestion bombs (Parr 4745) and all glass equipment was thoroughly cleaned before each series of experiments.11 A variable micropipette (Vitopet) was used for adding the standard metal solutions to the electrochemical cell. Reagents and Solutions The CH3COOH, Pb(N03)2, Cd(N03)2 and CHCOONa used were from BDH Chemicals (AnalaR) and the HN03 was from Merck (pro analysi grade). The metal stock solutions were prepared in the millimole range in 0.005 M HN03. The acetate buffer (1 M CH3COONa and 1 M CH3COOH, pH 4.96), which served as a supporting electrolyte, was electrolysed con- tinuously at a mercury cathode for 5 d before use.12 The water used for the preparation of solutions was triply distilled in a Pyrex still.Urine Samples Early morning urine samples were collected from urban and rural inhabitants of Ankara, Turkey. Although the city centre of Ankara is heavily polluted, the rural area is relatively free from pollution. The collected urine samples were frozen shortly after acidification with HCl (to give a total concentra- tion of 0.2 M HC1) and were stored at -20 "C in polyethylene bottles. A 1-ml aliquot of urine was digested with 2 ml of concentrated HN03 in an acid digestion bomb at 120 "C for 5 h.228 ANALYST, MARCH 1987, VOL. 112 Stripping Procedure A 0.5-ml aliquot of the digested urine sample was added to 4.5 ml of acetate buffer in an electrochemical cell.The solution was de-aerated for 15 min with pure nitrogen and pre- electrolysis was carried out for 3 min at -0.74 V vs. SCE with stirring. The size of the mercury drop was 3 divisions per drop for the HMDE. After the solution had been allowed to stand for 30 s, an anodic scan was applied at 2 mV s-1. The Pb anodic peak was obtained at -0.60 V. This procedure was repeated using an electrolysis time of 1.5 min. Measurements obtained with two different pre-electrolysis times were rep- eated at least once before adding small volumes of the standard metal solutions twice and repeating the whole procedure to determine the concentration of Pb by the modified method of standard additions. A 0.5-ml aliquot of digested urine sample was added to 4.5 ml of acetate buffer in the electrochemical cell.After de-aeration of the solution, the potential was set at -0.74 V, and the solution was electrolysed for 4 min with stirring. The mercury drop, which was twice as large as that used previously (6 divisions per drop), was carefully taken out of the electrochemical cell and then thrown away. This procedure was repeated once more with a new mercury drop. The potential was then increased to -0.90 V and Cd was determined using the same procedure as that described for Pb. The anodic stripping peak of Cd was obtained at -0.77 V vs. SCE. The stripping peak current values were measured and plotted versus accumulation times in order to obtain two least-squares calibration lines (Fig.1). These two lines intersected the x-axis at the same point, and this intersection point defined the analyte concentration. 70 50 P . c E 3 * 30 10 410 6 2 0 2 6 10 Concentration of Pb2+, p.p.b. Fig. 1. Least-squares calibration graphs obtained from the voltam- metric determinations Results and Discussion One of the problems encountered in DPASV is the interaction between materials that have been pre-concentrated into or on to the electrode. One such interaction is the formation of intermetallic compounds by metals deposited into the mer- cury. Such intermetallic interactions usually cause one of the stripping peaks to be depressed in comparison with the peak height obtained in the absence of the second metal. A shift in the stripping peaks of the constituent metals or the appearance of a new peak may also be observed.Frequently noted examples are the interactions between Cu and Zn, Cu and Cd or Zn and Ni that result in depressions of the Zn or Cd stripping peaks.13J4 Under the experimental conditions investigated, we obser- ved a depression in the Cd peak during the simultaneous determination of Cd and Pb. It can be seen from Fig. 2 that, in subsequent voltammograms, the Cd peak diminished whereas the Pb peak increased. The depression of the Cd peak also occurred when Pb was spiked into the analyte solution, as shown in Fig. 3. To overcome this problem, it was decided to determine Pb and Cd in separate analyses. A pre-electrolysis step was included before the determination of Cd at a potential at which Cd was not affected.After the deposition of these ions on to the mercury drop, the drop was removed. In this way the concentration of the interacting ions in the analyte solution was decreased and we observed that the Cd stripping peaks obtained after this additional pre-electrolysis were more reproducible. The recorded peak heights are proportional to the bulk concentrations of the corresponding trace metals in the sample solution. The actual peak heights are sensitive to traces of organic surface-active material that might still be present in the solution after digestion. These surfactants tend to be adsorbed at the electrode surface and this affects the rate of the electrode -( process and consequently the corresponding 1 I I I 9 -0.8 -0.7 -0.6 -0.5 Applied potentialiV vs.SCE Fig. 2. Simultaneous determination of Cd2+ and Pb2+. Voltammo- grams were taken sequentially from A to D with a deposition time of 1.5 min ~ ~~~ ~~ Table 1. Precision of the DPASV method Mean and standard deviation of method Urine Concentrafion of Concentration of Pb2+, Cd2+, sample Pb2+,p.p.b. Cdz+,p.p.b. p.p.b. p.p.b. 1 28.71 2.46 x =28.68 X = 2.46 28.65 2.46 S.d. = f0.02 S.d. = fO.01 28.68 2.46 15.83 2.72 S.d. = k0.02 15.78 2.72 2 15.78 2.72 2 = 15.80 =2.72 S.d. = kO.01ANALYST, MARCH 1987, VOL. 112 229 Table 2. Results of recovery study. The concentration of Pb2+ and Cd2+ in urine sample 1 without the spike added were 9.49 k 0.06 and 1.82 f 0.01 p.p.b., respectively Spike Metal Concentration added, Recovery, Pb2+ 18.54 8.55 105.97 18.06 8.55 100.35 18.00 8.55 99.65 Cd2+ 6.64 5.18 93.11 7.23 5.18 104.55 7.18 5.18 103.55 ion found, p.p.b. p.p.b.O/O T -0.9 -0.8 -0.7 Applied potentialiv vs. SCE Fig. 3. Depression of the Cd2+ peak when spiked with Pb2+. A and C, Cd2+ peak before Pb2+ was spiked; B and D, Cd2+ peak after Pb2+ was spiked. In voltammograms A and B the deposition time was 1.5 min and in C and D it was 3 min current. This prohibits the use of calibration graphs for the evaluation of the recorded peak currents and therefore the standard additions technique was used.’ Whang et aZ.9 developed a variant of the standard additions calibration procedure for determinations by KSV in order to eliminate the background current. In this method, two deposition times and two standard additions were used in the determination of the unknown concentration.In DPASV, the background current is eliminated electronically. By employing this modified standard additions calibration, we obtained two calibration lines with different slopes. These lines intersected on the x-axis at the unknown concentration of the analyte. The intersection at the same point on the x-axis of these lines indicates the accuracy of the determinations. The functioning of the HMDE is of vital importance for the precision and accuracy of the ASV analysis, and therefore the filling procedure and other handling hints given by the manufacturer should be strictly followed. It has been reported 1. 2. 3. 4. 5. 6. 7. 8. 9. 10. 11. 12. 13. 14. that irreproducible results are frequently observed even if the filling is carefully carried 0 ~ t .5 By employing the modified standard additions calibration technique, we succeeded in compensating for the problem of irreproducible results from the HMDE. The precision was evaluated by analysing two different urine samples three times each. The means and standard deviations of the determinations were calculated and the results are summarised in Table 1. The results obtained are acceptable for both Cd and Pb. In a recovery study, the urine sample was spiked with a known amount of each analyte prior to analysis. This urine sample was then subjected to the digestion and analysis procedures. Comparison of the results obtained using the spiked samples showed the losses and the contamination of the analytes during the various steps of the analysis. The recovery data for the spiked urine sample is shown in Table 2. The proposed method based on stripping voltammetry at an HMDE provides a simple approach for the determination of Pb and Cd in urine down to p.p.b. levels. References Lund, W., and Eriksen, R., Anal. Chim. Acta, 1979, 107, 37. Adeloju, S. B . , Bond, A. M., and Hughes, H. C., Anal. Chim. Acta, 1983, 148, 59. Jagner, D., and Graneli, A., Anal. Chim. Acta, 1976,83, 19. Palrecha, M. M., Kulkarni, A. V., and Dhaneshwar, R. G., Analyst, 1986, 111, 375. Sagberg, P., and Lund, W., Anal. Chim. Acta, 1977,94,457. Kemula, W., and Kublik, Z., Nature (London), 1961, 189,57. Golimowski, J., Valenta, P., Stoeppler, M., and Nurnberg, H. W., Talanta, 1979, 26, 649. Graneli, A., Jagner, D., and Josefson, M., Anal. Chem., 1980, 52,2220. Wen Whang, C., Page, J. A., van Loon, G., and Griffin, M. P., Anal. Chem., 1984,56,539. Temizer, A., and Ozaltin, N., J. Assoc. Off Anal. Chem., 1986,69, 192. Mitchell, J. W., Talanta, 1982, 29, 993. Laxen, D. P. H., and Harrison, R. M., Anal. Chem., 1981,53, 345. De Angelis, T. P., Bond, R. E., Brooks, E. E., and Heineman, W. R., Anal. Chem., 1977,49, 1794. Wang, J., Farias, P. A. M., and Luo, D. B., Anal. Chem., 1984, 56, 2379. Paper A61240 Received July 22nd, 1986 Accepted October 15th, 1986

ISSN:0003-2654    

DOI:10.1039/AN9871200227

出版商:RSC    

年代:1987

数据来源: RSC

摘要:

231 ANALYST, MARCH 1987, VOL. 112 Polarographic Study of Solutions of Cadmium(l1) in Acetylacetone Jesus Hernandez Mendez and Fernando Becerro Dominguez Department of Analytical Chemistry, Faculty of Chemistry, University of Salamanca, Salamanca, Spain The polarographic behaviour of Cd" solutions in acetylacetone with NaCIO4 as a supporting electrolyte was studied using the following polarographic techniques: conventional d.c. polarography, normal, direct pulse polarography, normal reverse pulse polarography and differential-pulse polarography. The influence of various instrumental parameters was studied and the reversibility of the electrode process was evaluated. The heterogeneous rate constant, k,, and the electrode transfer coefficient, a, were determined. The determination of these parameters by differential-pulse polarography was carried out by optimising the fit of the experimental data to the theoretical equation according to the modified simplex method.Keywords: Cadmium; polarography; acetylacetone solution; non-aqueous solvents; kinetic parameters Acetylacetone is an aprotic polar solvent, the dielectric constant of which has an intermediate value (23) at 25 "C. The solvent is only sparingly miscible with water and has been used as an extracting and chelating reagent in the determination of several metallic ions,l although it has rarely been employed as an electrochemical solvent. It was used by Nelson and Iwarnoto2 in the experimental evaluation of liquid junction potentials, and by Fujinaga and Lee3 in the polarographic determination of FeIII and UVI, previously extracted as acetylace tonates.Pulse polarographic techniques have usually been used to improve the sensitivity of analytical determinations, and they have not received much attention in the field of electrode kinetic processes. At present, however, interest in such areas is growing, as reflected in the work of Fonds et aL,4 Gfilvez et aZ.5 and Birke et aZ.6 This paper reports some polarographic studies [conven- tional (d.c.) polarography , normal, direct pulse polarography (NPP), normal, reverse pulse polarography (RPP) and differential-pulse polarography] of solutions of CdII in acetyl- acetone with NaC104 as the supporting electrolyte. CautiowPerchlorate solutions in organic media are ex- plosively hazardous and appropriate precautions should be taken.Experimental In the d.c. polarographic study, an Amel electroanalytical system consisting of Modules 563 (multipurpose unit), 564 (function generator) and 551 (potentiostat) was employed. In this system the Model 460 mechanical drop timer and the 7040 A x - y recorder (Hewlett Packard) were coupled together. In the pulse techniques a Metrohm Polarecord E-506, equipped with an E-505 stand, was employed. A three-electrode system was used, the working electrode being a dropping-mercury electrode (DME) . The reference electrodes were an aqueous saturated calomel electrode (conventional polarography) and an Ag - AgCl - KC1 refer- ence electrode (pulse techniques) and the counter electrode Table 1. Polarographic behaviour of Cd" in acetylacetone with different supporting electrolytes.Concentrations: Cd", 1 mM; elec- trolyte, 0.1 M was a platinum wire. Except when the influence of the temperature was being studied, all polarographic measure- ments were carried out at 25 "C. CdII solutions were prepared by dissolving Cd(NO& in acetylacetone. Highly purified nitrogen was passed through the solution to remove dissolved oxygen. Results and Discussion In acetylacetone media, the supporting electrolytes employed, NaC104, LiC104 or TEAP (tetraethylammonium perchlor- ate), exhibit the electroactivity ranges shown in Table 1. Different reference electrodes were tested and it was found that the saturated calomel electrode (SCE) could be used with no apparent drawbacks. NaC104 was chosen as the supporting electrolyte because the reduction waves of CdII show a better morphology in this medium than in the others tested (LiC104 or ZEAP).Under these conditions, CdII gives a simple and well defined reduction wave at a potential close to -0.4 V. Plots of log(& - i) against Ed,,. (where i = current intensity, id = diffusion current intensity and = dropping electrode potential) showed that the electrode process is quasi-rever- sible and coulometric determinations (Table 2), at constant potential, of the number of electrons, n, led to a value of 2, such that the electrode reaction may be expressed as CdII + Hg + 2e Cd(Hg) The limiting current of the wave decreases with increasing ionic strength of the medium; at the supporting electrolyte concentrations employed (0.01 - 1.00 M), the ratio fits the equation F ~ 1 3 .1 4 - 9.72 pi . . . . . . (1) where Z = diffusion current constant and p = ionic strength. This equation yields a value for the diffusion current at infinite dilution of I, = 3.6 pA mmol-1 1 mg-js-4. The temperature coefficent (0) of the wave was determined as recommended by Meites,7 and proved to be 1.30% "C-1 for a temperature range of 1040 "C. Using the graph of log il vs. t , which is linear, a value of o = 1.29% "C was obtained. Supporting EdV electrolyte Potential rangeN i,JpA vs. SCE NaC104 . . . . +0.5to-1.2 4.96 -0.406 LiC104 . . . . +0.5to-1.0 3.75 -0.440 TEAP . . . . +0.6t0-1.4 3.86 -0.450 Table 2. Coulometric determination of the number of electrons, n. Applied potential: -0.800 V vs. SCE n .. . . . . . . . . . . 1.98 2.01 0.5 1.83 0.1 Concentration of CdIVmM . . 1 .O232 ANALYST, MARCH 1987, VOL. 112 Influence of Instrumental Parameters D.c. polarography The influence of the height of the mercury reservoir was evaluated according to the expression and the more rigorous expression i1/@,,, = K~ + K ~ / @ ~ ~ ~ . . . . . . (3) where h,,,, is the height of the mercury reservoir after correcting for the effects of retropressure and interfacial tension. In both instances a good correlation was found; for equation (2), r2 = 0.9999, and for equation (3), r2 = 0.9960, for a height range of 17.5-75 cm. Pulse techniques The study of the relationship between the current intensity and the surface area of the mercury drop electrode was split between the study of the height of the mercury column and the study of the drop lifetime.Previously, the linearity between the flow-rate, rn, and the corrected height was determined by using a 0.1 M solution of NaC104 in acetylacetone. The data obtained agree with the equation rn = 13.83 x 10-3hfbz7, with r2 = 0.9992. Tables 3 and 4 show the data obtained in the study of the relationship between intensity and h,,,, and t. Good linearity between the current intensities and the two-thirds power of the respective parameter was found, and hence between the limiting currents (NPP, RPP) and peak current (DPP) of the polarographic waves and the surface area of the mercury drop. Table 3. Influence of the surface area of the electrode. Variation of the height of the mercury column 35 40 45 50 55 60 65 70 75 4.60 5.10 5.55 5.95 6.30 6.70 7.10 7.35 7.65 iNpp = 0.51 + 0.42 h!om r2 = 0.9980 iRpp = 0.05 + 0.24 iDpp = 1.41 + 0.28 htorr r' = 0.9929 r2 = 0.9988 2.45 2.75 2.90 3.20 3.35 3.50 3.80 4.05 4.25 4.14 4.53 4.80 5.01 5.27 5.52 5.76 5.99 6.22 Table 4.Influence of the area of the electrode. Variation of the drop lifetime tls 0.4 0.6 0.8 1 .o 1.2 1.4 2.0 3.0 4.0 5.0 6.0 I"PP/ PA 2.90 3.60 4.18 4.76 5.40 5.95 7.60 10.00 12.15 14.10 15.68 I'RPPJ PA 2.94 3.08 4.05 4.64 5.32 6.02 7.50 8.00 9.05 9.80 11.40 I'DPPI PA 2.12 2.42 2.90 3.20 3.46 3.66 4.22 5.13 5.94 - - As expected, when the sweeps were performed in normal pulse polarography in the anodic direction, and especially when studying the influence of the drop lifetime (different pre-electrolysis times of the solution under study), a deviation from linearity occurred (lower regression coefficients) and high values of the ordinate at the origin could be observed.Concentration of Electroactive Species In d.c. polarography, the limiting current of the reduction wave of CdII proved to be proportional to the concentration of CdI1 (C) (id.,. = -0.37 + 5.12c, r2 = 0.9998), such that from the simultaneous consideration of the effects of temperature, the height of the mercury reservoir and the concentration of Cd", it may be inferred that the polarographic wave of CdII in acetylacetone with 0.1 M NaC104 as the supporting electrolyte is diffusion controlled. Fig. 1 shows the values obtained for the current intensities in the three pulse techniques for a concentration range of electroactive species of 0.01-1.0 M.In normal-pulse polarography , linearity may be observed (iNpp = 0.25 + 7.72c, r2 = 0.9953), but in reverse-pulse polarography linear regression analysis of the limiting current again points to the limited use of this technique for the determination of species (iRpp = -0.03 + 4.60c, r2 = 0.9886). In differential-pulse polarography , good linearity can be observed in the concentration range studied (iDpp = 0.01 + 5.37c, r2 = 0.9984), although for CdII concentrations lower than 4 x 10-5 M, deviations from linearity begin to appear; this is probably due to the presence of a residual polarographic component due to oxygen, and to the difficulties of elim- inating this. Influence of Pulse Amplitude In differential-pulse polarography , peak current and peak potential are functions of the amplitude of the pulse applied.The following equation is applicable to the currents: Ai =f[(l - a)/(l + a)] . . . . . . (4) where AE a = exp (s x 1) Ai being the peak current and AE the pulse amplitude applied. 6 0.2 0.4 0.6 0.8 1.0 Concentration of cadmium/mnn iNpp = 0.15 + 4.73 B iDpp = 0.184 + 0.649 t3 r2 = 0.9993 r2 = 0.9994 iRpp = 1.72 + 2.95 B r2 = 0.9661 Fig. 1. Variation of current with concentration of cadmiumANALYST, MARCH 1987, VOL. 112 233 Table 5 shows the values of the peak currents of the polarograms obtained with a 1.0 mM solution of CdII and by modifying the amplitude of the pulse from +0.050 to -0.050 V. A perfect correlation may be seen between the respective peak currents and the theoretical values of the function[(l- a)/(l+ a)]: Ai = -0.168 + 6.950[(1- a)/(l + a)], r2 = 0.9983 for AE<O; and Ai = -0.055 - 5.776[(1 - a)/ (1 + a)], r2 = 0.9987 for AE>O.Moreover, in reversible processes the peak potential must shift towards less cathodic values than the half-wave potential when the amplitude of the pulse is increased (if the pulses are of negative amplitude) according to the expression Ep = Eh - AEl2 . . . , . . * ( 5 ) where Ep = peak potential, Eh = half-wave potential and A E = pulse amplitude. The values found (Table 6) for the same series of experiments as reported above show a logical sequence shifting towards progressively less negative values as the amplitude decreases (both with respect to the series of equal sign and to amplitudes with the same absolute value but with different signs).However, they do not fit the theoretical equation if the process is reversible, as on applying regression criteria the fit is not good and, more importantly, the values of the slope are appreciably separated from what would be expected (-0.5). This fact clearly indicates that the process is not reversible. Reversibility of the Process The reversiblity of the system was studied, both qualitatively and quantitatively, by all the polarographic techniques. In order to qualify the process, the criteria based on the re- lationship between the current intensities (iNPPIiRPP or 1 AiyAz; I) and the difference in potentials ( E ~ , N P ~ - E4,Rpp; Ep - Ei) were employed. In the determination of the characteristic kinetic parameters, however, use was made of the “logarithmic” plots for conventional polarography and the two techniques of normal-pulse polarography , and optimisa- tion and curve fitting for differential-pulse polarography .Table 5. Differential-pulse amplitude on peak current A E N A ilpA 0.050 4.32 0.045 3.94 0.040 3.80 0.035 3.34 0.030 2.97 0.02s 2.52 0.020 2.05 0.015 1.57 0.010 1.07 0.005 0.53 polarography: influence of pulse A EIV A ilpA -0.005 0.59 -0.010 1.19 -0.015 1.76 -0.020 2.35 -0.02s 2.90 -0.030 3.46 -0.035 3.92 -0.040 4.40 -0.045 4.68 -0.050 5.16 Table 6. Differential-pulse polarography: influence of pulse amplitude on peak potential A EIV 0.050 0.045 0.040 0.035 0.030 0.025 0.020 0.015 0.010 0.005 EdV -0.371 -0.370 -0.366 -0.364 -0.358 -0.356 -0.353 -0.351 -0.349 -0.346 A EIV -0.005 -0.010 -0.015 -0.020 -0.025 -0.030 -0.035 -0.040 -0.045 -0.050 E,IV -0.342 -0.339 -0.337 -0.335 -0.337 -0.334 -0.332 -0.330 -0.328 -0.325 Qualification of the Reversibility of the Process The expression of the ratio between the limiting currents in NPP and RPP is always a function of the drop lifetime, t, and the time of measurement, tm (200 ms in the apparatus used), employed and adopts different forms according to whether the process is reversible or irreversible.In the former instance, the ratio is and in the latter (7) Considering the difference in half-wave potentials in the different sweeps, it is clear that if the process is reversible, the two waves will have the same half-wave potential value, whereas for a totally irreversible process the half-wave potential in an anodic sweep will be a few millivolts lower than in the cathodic sweep.From the value of the difference it is possible to calculate the expression for the product na (a = electrode transfer coefficient): RT naF Et, ~pp-Eh, Rpp = - [0.57 + 1.49 (tm/t)] . . (8) Polarograms were recorded at all the drop times available on the apparatus employed. The results are shown in Tables 7 and 8. The results do not coincide with those deduced theoretically either for the instance in which the process is reversible or for that in which it is irreversible. From these results, it would seem that, by exclusion, the process is quasi-reversible. The study of the second criterion confirms this aspect as although the half-wave potentials in anodic Table 7.Normal-pulse polarography: qualification of reversibility of the process by ratio of currents I’NPPII‘RPP tls 0.4 0.6 0.8 1.0 1.2 1.4 2.0 3.0 4.0 5.0 6.0 Reversible 0.76 0.84 0.88 0.94 0.92 0.94 0.96 0.98 0.98 0.99 0.99 Irreversible 0.98 1.20 1.38 1.54 1.69 1.83 2.18 2.67 3.09 3.45 3.78 Experimental 0.99 1.17 1.03 1.03 1.02 0.99 1.01 1.25 1.34 1.44 1.38 Table 8. Normal-pulse polarography: qualification of the reversibility of the process by difference in half-wave potentials tls 0.4 0.6 0.8 1 .o 1.2 1.4 2.0 3.0 4.0 5.0 6.0 EtNPP - E i R P P 0.018 0.020 0.024 0.024 0.024 0.024 0.028 0.026 0.026 0.026 0.026 na 1.78 1.31 1.06 0.90 0.85 0.82 0.65 0.66 0.63 0.62 0.61234 ANALYST, MARCH 1987, VOL. 112 sweeps are a few millivolts more negative than those of the cathodic sweep (and hence not reversible), the values of the expression na deduced from them are not constant, indicating that the process in not wholly irreversible either.In DPP, the criteria that allow the qualification of reversi- bility of the process are based on the influence of the so-called kinetic parameter in the ratio (in absolute values) between the “anodic” and “cathodic” peak currents and on the difference between the anodic and cathodic peak potentials. These can be expressed as follows: Reversible charge transfer: E“p - E; = IAEI Quasi-reversible charge transfer: ~ “ p - E; < I A E ~ IAi;/A$ < 1 for a b 0.5 ]AE?Az;l> 1 or <1 for a < 0.5 Irreversible charge transfer: lAiE/Ai;( < 1 By digital simulation it is possible to determine the variation of the above two expressions with the absolute value of the amplitude of the pulse applied.As the difference in the peak potentials in the studies performed with identical solutions of Cd” in acetylacetone is nearly always lower than the value of the amplitude of the pulse applied and the absolute value of the peak current ratio is always less than unity, the process can be qualified as quasi-reversible (Table 9). Determination of Kinetic Parameters D.c. polarography For the evaluation of the values of na and kf, the hetero- geneous rate constant, a logarithmic analysis was performed of the waves, recording them as described by Meites.8 The equation 0.05915 where kf = heterogeneous rate constant and D = diffusion coefficient, was employed, which is valid in the range 0.1 Si/& S0.94 for maximum values of id (end of drop time).9 The value of na was also determined using the relationship between E2 and Ei E2 - Ei = - 0.0517/na .. . . . . (10) The use of the equations of Ilkovic, Koutecky and Matsuda allows an estimation of the diffusion coefficients as D = 3.9 x 10-6cm2s-1. Table 10 shows the results obtained. The values of na are almost identical when obtained either by equation (10) or from the slope of the logarithmic plot of equation (9). The values of the heterogeneous rate constant, kf, allowed us to establish the reduction process of CdII in acetylacetone, with NaC104 as the supporting electrolyte, as being quasi- reversible, at least from the viewpoint of conventional polarography .The variation of the values of na and the gradual decrease in the heterogeneous rate constant on increasing the concentra- tion of the supporting electrolyte seems to point to some kind of interaction between the electroactive species and the anion of the supporting electrolyte. Pulse tee hn iq ues The equation used is similar to that in conventional polaro- graphy, although slightly more complex. E is plotted versus the expression where x = iNPdiCott. which includes the ratio, x , between the current of normal pulse polarography and the current of Cottrell. Table 9. Differential-pulse polarography: qualification of reversiblity of the process )AElImV 10 20 30 40 50 60 70 80 90 100 EE- E; 10 16 28 30 32 36 40 46 57 70 I Ai7Aigl 0.84 0.82 0.83 0.79 0.75 0.81 0.83 0.84 0.84 0.91 Table 10.Determination of kinetic parameters (d.c. polarography). Concentration of CdII = 1 mM NaC1O4/M 0.01 0.02 0.03 0.04 0.05 0.06 0.08 0.10 0.20 0.30 0.40 0.50 0.60 0.80 1.00 E f l vs. SCE -0.422 -0.428 -0.416 -0.414 -0.428 -0.430 -0.426 -0.406 -0.420 -0.417 -0.404 -0.404 -0.368 -0.355 -0.349 na* 0.55 0.50 0.59 0.53 0.52 0.45 0.52 0.78 0.68 0.66 0.67 1.10 1.15 1.23 1.40 * Values obtained by equation (10). t Values obtained by a logarithmic plot. nat 0.55 0.50 0.59 0.53 0.52 0.46 0.52 0.78 0.68 0.67 0.67 1.10 1.15 1.23 1.39 k&m s - 1 1.0 x 10-2 1.0 x 10-2 9.3 x 10-3 9.2 x 10-3 8.0 x 10-3 6.3 x 10-3 5.8 x 10-3 4.3 x 10-3 3.5 x 10-3 3.2 x 10-3 2.8 x 10-3 2.7 x 10-3 2.5 x 10-3 2.2 x 10-3 6.8 x 10-3 Table 11. Determination of kinetic parameters (NPP, RPP).Concentrations: Cd”, 1.0 mM; NaClO,, 0.1 M k x 103 na tls NPP RPP NPP RPP 0.4 0.6 0.8 1 .o 1.2 1.4 2.0 3.0 4.0 5 .O 6.0 0.21 0.21 0.22 0.23 0.23 0.25 0.23 0.24 0.23 0.23 0.24 0.41 0.27 0.22 0.16 0.27 0.30 0.25 0.27 0.31 0.23 0.30 1.6 1.7 1.7 1.8 2.5 2.6 2.6 2.4 2.1 2.1 2.1 3.9 3.0 2.6 1.4 1.8 1.6 1 .o 0.9 0.8 0.8 2.1ANALYST, MARCH 1987, VOL. 112 235 ~~ Table 12. Determination of kinetic parameters (DPP) . Modified simplex method A EImV (Y Pk - 5 0.37 3.50 +5 0.39 3.59 - 25 0.48 3.86 + 25 0.50 3.92 - 50 0.65 3.89 + 50 0.63 4.10 ~~ ~ _ _ _ ~ ~ ~ ~ ~ ~ The results are presented in Table 11; from these it is possible to qualify the process as quasi-reversible from the point of view of normal-pulse polarography . For quasi-reversible and/or totally irreversible processes, the equation for the intensity of a mercury drop electrode, giving a solution for the current of pulses in a plane electrode in an expansion and including a term corresponding to the convection effect, is, according to Birke et a1.6 i = (z + 6); Air,=., (x6)i P2Q2Fl (1;) .. . . (11) where p2 = exp {Q$:[(T+ 6); - ~ ~ ] } e r f c ( Q ~ { $[(T+ 6)’ - z ’ ]}A) PI = exp (Q: $ z 3 ) erfc[Ql( $T ;)&I k( 1 + E~) Dt Qi = - Ei = exp [ (Ei - Equation (1 1) clearly overcomes the difficulty involved in the use of plotting techniques for the determination of kinetic parameters. It is the technique of curve fitting to the experimental data that allows the determination of the exact values of K and na. Here we use the simulation of the intensity function for some assumed values of k and na by means of optimisation processes, the modified simplex method, and the values leading to a better fit of the experimental data to the theoretical equation are determined.We use the normalised current function Aiplid.c., where id.c. is the diffusion current in conventional polarography , which cam be determined experimentally. The influence in the search procedure of the initial values, the size of the simplex and the number of iterations was considered; however, the most decisive factor is the number of experimental points employed in the fitting. Having once confirmed the viability of the method, the experimental curves corresponding to similar sweep potentials are fitted for the pulse amplitudes shown in Table 12. The values obtained allowed us to qualify the reduction process of the species CdII in acetylacetone as quasi-reversible. The decrease in the values of the heterogeneous rate constant are lower than those found with conventional polarography , in agreement with the well known “loss” of reversiblity on the passage from the conventional technique to differential-pulse polarograph y . 1. 2. 3. 4. 5. 6. 7. 8. 9. References Burger, K., “Organic Reagents in Metal Analyses,” Pergamon Press., Oxford, 1973, p. 112. Nelson, I. V., and Iwamoto, R. T., Anal. Chem., 1963,35,867. Fujinaga, T., and Lee, H. L., Talanta, 1977, 24, 395. Fonds, A. W., Brinkman, A. A. A. M., and Los, J. M., J. Electronal Chem., 1964, 7 , 171. GAlvez, J., Serna, A., and Molina, A., J. Electroanal. Chem., 1981, 124, 201. Birke, R. L., Kim, M. H., and Strassfeld, M., Anal. Chem., 1981, 53, 852. Meites, L., “Polarographic Techniques,” Second Edition, Wiley, New York, 1965, p. 138. Meites, L., “Polarographic Techniques,” Second Edition, Wiley, New York, 1965, p. 219. Meites, L., and Israel, Y., J. Am. Chem. SOC., 1961, 83,4903. Paper A6/111 Received April 9th, 1986 Accepted July lst, 1986

ISSN:0003-2654    

DOI:10.1039/AN9871200231

出版商:RSC    

年代:1987

数据来源: RSC

摘要:

ANALYST, MARCH 1987, VOL. 112 237 Determination of B2 and Animal Tissues Vitamers in Pharmaceutical Preparations, Foods by a Photokinetic Method Tomas Perez-Ruiz,* M. C. Martinez-Lozano and Virginia Tomas Department of Analytical Chemistry, Faculty of Sciences, University of Murcia, Murcia, Spain A fast and sensitive photokinetic method for the determination of riboflavin (RF) and riboflavin 5’-phosphate (FMN) is described. It is based on the rate of photoreduction of RF and FMN by EDTA in the absence of oxygen. The rate of photoreduction is a linear function of the concentration of both B2 vitamers at very low concent rations. The rate of these photochemical reactions is monitored polarographically by recording the limiting current of p-benzoquinone, which is reduced by the 1,5-dihydro forms of RF and FMN that are generated in the photochemical reactions.The results obtained by the application of the fixed time, fixed concentration and initial rate kinetic methods have been evaluated. An alternative method for monitoring the rate of the process is by measuring the time necessary for the total reduction of p-benzoquinone. The end-point is detected with two platinum electrodes at an applied voltage of 100 mV. The procedure has been successfully applied to the determination of B2 vitamers in pharmaceuticals, foods and animal tissues. Keywords : Riboflavin and riboflavin 5 ’-phosphate determination; pharmaceuticals, foods and animal tissues; photo kinetics; pola rograp h y; B2 vitam ers The most widespread flavins of biological importance are riboflavin (RF) and the so-called nucleotides, riboflavin 5’-phosphate (FMN) and the intramolecular complex of FMN with adenosine 5’-phosphate (FAD) These three flavins appear in biological tissues in the free state (mainly riboflavin) or as complexes with proteins (both nucleotides) and other components of living cells.The determination of riboflavin (vitamin B2) using bio- assay, microbiological and chemical methods has been reviewed previously. 1~ Bioassays, although time consuming and with a number of inherent factors that can lead to error, are still the ultimate reference standard for other methods of determination because they measure the activity of the vitamin rather than the total amount of vitamin present. However, because of the time and error factors in these biological methods, wet chemical methods are the most popular procedures used today.The manual chemical method most used for the determina- tion of vitamin B2 is fluorimetry, which can be carried out either by the measurement of the intensity of the natural fluorescence of riboflavin or by measurement of the fluor- escence of lumiflavin obtained by the photochemical pre- treatment of riboflavin.l.3 Spectrophotometric methods have been developed using the absorbance of vitamin B2 at 444 nm.1 Electrochemical methods have also been utilised, including d.c. and a.c. polarography, differential-pulse polar- ography , linear-sweep polarography and adsorptive stripping voltametry.l+ll High-performance liquid chromatography with UV and fluorescence detection has been used to determine riboflavin and other B2 vitamers in food products, beverages, blood, meat extracts and the body fluid of plankton.12-20 Semi-automated systems based on the above methods can run from 40 to 60 samples per hour and have shown good agreement with manual methods.21-23 Table 1 summarises some of the more important methods for the determination of riboflavin. The facile photoreduction of free flavins by a variety of amino acids, carboxylic acids and amines has been known for many years.2c28 One of the most effective photoreductants is ethylenediaminetetraacetic acid (EDTA).2+34 Under anaer- * To whom correspondence should be addressed. obic conditions and over a wide pH range, the photoreduction of RF and FMN by EDTA is accomplished in a matter of seconds with a source of white light, according to reactions 1 and 2: RF + EDTA + H2O hv\ RFH2 + ED-triacetic + CH20 + C02 (1) FMN + EDTA + H20 FMNH2 + ED-triacetic + CH20 + C02 ( 2 ) Employing these photochemical reactions a sensitive kinetic method has been developed for the determination of both B2 vitamers.It is based on the rate of reactions (1) and ( 2 ) , which is a linear function of the concentration of RF and FM at very low concentrations. Theory The basic rate equation for the photochemical reaction (1) or (2) is (3) where A represents riboflavin or riboflavin 5’-phosphate. The intensity of absorbed radiation can be obtained by application of the Beer - Lambert law: --- dEA1 - 2 +kIOk [l - ex~(-2.3 E ~ ~ [ A ] ) ] . . (4) dt k where h refers to each of any photochemically active wavelengths incident on the sample, is the quantum yield at a given wavelength, IOk is the incident intensity of radiation, ~1 the molar absorptivity of RF or FMN and b the depth of solution containing RF or FMN.When the absorbance due to the photoactive component is small, of the order of 0.05 or less, a simplified form is obtained by a Taylor-series expansion of the exponential term: --- d[A1 - Ekb[A] . . . . (5) dt This means that the rate of the photochemical reaction is a linear function of the instantaneous concentration of RF or238 ANALYST, MARCH 1987, VOL. 112 Table 1. Methods for the determination of riboflavin Method Range/pM Anodic adsorptive stripping voltammetry Cathodic adsorptive stripping Differential-pulse polarography .. . . 0.26-2.6 5 x 10-4-1 x 10-2 voltammetry . . . . . . . . . . 0.01-0.1 D.c. polarography . . . . . . . . 50-400 A.c. polarography . . . . . . . . 1-50 Photometry . . . . . . . . . . . . 50-400 Fluorimetry . . . . . . . . . . . . 0.13-1.3 Laser fluorimetry . . . . . . . . . . High-performance liquid chromatography Detection limit/pM 2.5 x 10-5 1 x 10-2 50 0.2 6 x 10-6 13 Reference 10,ll 9,11 8 1,4,5 7 1,3,24 25 1 19 n I I 121 Fig. 1. Illumination device. 1 = Lamp; 2 = cooling system; 3 = lens; 4 = reaction cell; 5 = magnetic stirrer; and 6 = polarograph. RE = reference electrode; WE = working electrode FMN. This is the basis for photochemical kinetic determina- tion.35 Just as in fluorescence,36 the sensitivity of the determination can be increased by increasing the intensity of the photolytic radiation.Experimental Apparatus The arrangement of the apparatus used is shown in Fig. 1. An electronic voltage regulator was used to obtain a close voltage control for a stable radiation source. A Sylvania 250 W, 24 V halogen lamp was used as source of visible radiation. The light produced was passed through a small water-cooled chamber which was arranged so that several filters could be used. A lens system was used to focus the light on the reaction cell, which was thermostated at 25 k 0.5 "C. A magnetic stirrer was used to stir the solution in the cell. A Radiometer PO4 polarograph was used to record the current - time curves. Reagents All chemicals used were of analytical-reagent grade and doubly distilled water was used throughout.Rboflavin and riboflavin 5'-phosphate were recrystallised twice from 1 M acetic acid.37 Riboflavin standard solution, 10-4 M. Dissolve the ribo- flavin, previously dried and stored in the dark in a desiccator over P205, in 0.02 M acetic acid to give 1 1. Store under toluene at ca. 10 "C. Working standards should be prepared daily from this solution by diluting with doubly distilled water. Riboflavin 5'-phosphate standard solution, 10-4 M. Dissolve the riboflavin 5'-phosphate, dried and stored as above, in doubly distilled water to give 1000 ml. Store under toluene at ca. 10°C. Working standards should be prepared daily as required. The RF and FMN solutions were stored and handled either in the dark or under red safelights in order to prevent photochemical degradation.General Procedure The sample must always be prepared under diffuse light. To the reaction cell, add 1 ml of 2 M acetate buffer solution (pH 5.5), 5 ml of 0.1 M EDTA, 1 ml of 0.001 Mp-benzoquinone and an appropriate volume of standard or sample riboflavin or riboflavin 5'-phosphate solution to give a final concentration between 0.2 and 5 WM. Dilute to exactly 30ml with doubly distilled water. Keep the cell at 25 f 0.5"C by thermostatic control. Remove oxygen from the solution by bubbling through with pure nitrogen (99.97%) for 10 min. Apply a potential of -0.25 V vs. SCE at the dropping-mercury electrode, then switch on the halogen lamp and the polaro- graphic recorder simultaneously and record the current - time curve.When two platinum electrodes are used, apply an e.m.f. of 100 mV between them and record the curve corresponding to the "titration" of p-benzoquinone with photogenerated leucoriboflavin or leucoriboflavin 5'-phosphate. The analyte (RF or FMN) concentration is calculated from the corresponding equation for the calibration graph. Determination of RF or FMN in pharmaceutical preparations Pharmaceutical preparations containing water-soluble vit- amins are dissolved by warming with doubly distilled water and are diluted to volume in a 100-ml calibrated flask. A suitable aliquot is analysed by general procedure. Determination of riboflavin in foods Milk. Add 50 ml of 0.1 M hydrochloric acid to 10 g of milk in a 100-ml Erlenmeyer flask and heat on a steam-bath nearly to dryness.Repeat this operation twice more. Cool, adjust to pH 5.5 with sodium hydroxide solution, dilute to volume with doubly distilled water in a 50-ml calibrated flask and analyse a suitable aliquot. Bread. To 5 g of bread (cut into small pieces) add 25 ml of 0 . 1 ~ hydrochloric acid and heat. Cool the suspension, filter through a cloth and then through a filter-paper. Collect the filtrate in a 100-ml calibrated flask and proceed as for milk. Enriched corn flour. To 2 g of corn flour add 25 ml of 0.1 M hydrochloric acid and heat. Cool, filter and proceed as for bread. Determination of total flavins in animal tissues The tissue must be excised fresh from animals, immediately weighed and cut into small pieces. A 3 g mass of tissue is placed in a few millilitres of doubly distilled water, previously heated to 80 "C, and kept at 80 "C for 3-5 min.The tissue is ground in a glass homogeniser and then the resulting suspension is transferred quantitatively into a 50-ml beaker, diluted with doubly distilled water to a volume of over 30 ml and heated at 80 "C for 15 min with occasional stirring. After cooling at room temperature, the suspension is diluted to volume with doubly distilled water before heating, then stirred and filtered. The solution is placed in a small brown bottle andANALYST, MARCH 1987, VOL. 112 I 239 I 100 0 b I I I I 4 6 8 1 0 PH Fig. 2. riboflavin; broken line, riboflavin 5’-phosphate Rate of photoreduction, V,, as a function of pH. Solid line, 100 s 0 1 2 3 [EDTA]/10-* M Fig. 3. Dependence of the rate of photoreduction, VR, on the initial concentration of EDTA.Conditions: pH, 5.5 (acetate buffer); RF (or FMN) concentration, 3.3 x 1 0 - 6 ~ . Solid line, riboflavin; broken line, riboflavin 5‘-phosphate an equal volume of 0.2 M hydrochloric acid added. After the bottle has been sealed, the sample solution is hydrolysed at 120 “C in an autoclave for 2 h. The solution is then transferred into a 100-ml calibrated flask and diluted to volume with doubly distilled water. An aliquot is used for the determina- tion of total flavins by applying the general procedure. Results and Discussion Photoreduction of Riboflavin and Riboflavin 5’-Phosphate by EDTA In the absence of oxygen RF and FMN are photodecolorised by EDTA, whereas FAD is not. The rate of photoreduction (VR) of both B2 vitamins is pH dependent, as shown in Fig.2. The dependence of VR on the initial concentration of EDTA is shown in Fig. 3. Reproducible results of VR for RF and FMN are obtained by working in a buffered medium and at EDTA concentra- tions higher than 1 0 - 2 ~ . For ten determinations of the amount of reduced form of RF and FMN generated at pH 5.5 (acetate buffer) and [EDTA] = 1.6 x 1 0 - 2 ~ , the relative standard deviation found was 0.9% for RF and 0.98% for FMN. Determination of Riboflavin The product of photoreduction of riboflavin by EDTA, 1,5-dihydroriboflavin (RFH2), is a strong reductant4J8Jg and reacts rapidly with oxidants, to give riboflavin. When an oxidant such as p-benzoquinone (Q) is added to a solution of RF and EDTA at pH 5.5 and illuminated, the photoreduced RFH2 is oxidised; RF is then photoreduced again by EDTA and the cycle is repeated until all of the p-benzoquinone is transformed into hydroquinone: RF + EDTAB RF’H2 .. . . (6) RFH,+Q+RF+QH, . . . . (7) 0 1 2 3 4 Time/min Fig. 4. Current- time curves: 1.6 x 1 0 - 2 ~ EDTA, 0 . 1 2 ~ acetate buffer (pH 5 . 9 , 3.3 x l o - S M p-benzoquinone, 25°C. Riboflavin concentration: (1) 0.0; (2) 0.33; (3) 0.67; (4) 1.67; (5) 2.66; (6) 4.0; and (7) 5.3 VM I 5 4 3 0 5 Time/m in 10 Fig. 5. Amperometric detection: 1.6 X 1 0 - 2 ~ EDTA, 0 . 1 2 ~ acet- ate buffer (pH 5.5), 3.3 x 10-5 M p-benzo uinone, 25 “C. Riboflavin concentration: (1) 0.0; (2) 0.33; (3) 0.67; &) 2.66; and (5) 4.0 pM The rate of the photochemical reaction can be followed in two ways, (1) by recording polarographically the limiting current of p-benzoquinone (Fig.4) and (2) by measuring the time necessary to reduce p-benzoquinone, which involves the “photochemical titration” of benzoquinone with photogener- ated RFH2 employing amperometric detection (two platinum electrodes at 100 mV applied voltage (Fig. 5).4Oy4I No change in the initial rate of the photochemical reaction was detected with a variation in the p-benzoquinone concen- tration in the range 1 x 10-5-2 x 10-4 M. From the analytical point of view, ap-benzoquinone concentration should be used that provides a current that is easily measured, and for this purpose 3.3 X 10-5 Mp-benzoquinone was chosen as a suitable concentration. The rate of the over-all redox process is affected by temperature and the temperature was therefore kept at 25 k 0.5 “C.To summarise, the best experimental conditions for the photokinetic determination of riboflavin are an EDTA concentration of 1.6 x 10-2 M, a p-benzoquinone concentra- tion of 3.3 x l o - 5 ~ and a pH of 5.5. Determination of Riboflavin 5’-Phosphate The product of photoreduction of FMN by EDTA under anaerobic conditions is also a strong reductant39 and reacts rapidly with oxidants. The rate of this photochemical reaction can therefore be followed in the same two ways that were described for riboflavin by coupling an appropriate oxidant such as p-benzoquinone. The working conditions (EDTA and p-benzoquinone con- centrations, pH and temperature) are the same as those used for the determination of riboflavin.240 ANALYST, MARCH 1987, VOL.112 Table 2. Calibration graphs for the fixed time method. Conditions: EDTA, 1.6 X 1 0 - 2 ~ ; acetate buffer, 0 . 1 2 ~ (pH 5.5); p-benzo- quinone, 1.6 X M; and temperature, 25 “C Equation of calibration graphs 0.5 i = 2.08 - 0.20C 0.9963 1.0 i = 2.06 - 0.21C 0.9979 2.0 i = 2.02 - 0.24C 0.9988 3.0 i = 2.04 - 0.22C 0.9951 5.0 i = 2.03 - 0.25C 0.9927 Time/min (0.2-5 p~ of RF or FMN)* r * C, p ~ ; i, @. ~~ Table 3. Photokinetic determination of riboflavin and ribo- flavin-5’-phosphate R.s.d.,* Yo ~~ ~ Concentration Riboflavin range/pm Riboflavin 5’-phosphate Tangent . . . . . . 0.5-5 1.96 1.53 Fixed time . . . . . . 0.2-5 1.02 0.97 Fixed concentration . . 0.5-5 1.74 1.67 * Relative standard deviation at 1.67 p~ level.Table 4. Tolerance to foreign vitamins in the determination of riboflavin Vitamin added Thiamine (BJ . . Biotin(H) . . . . Pyridoxine(B6) . . Cobalamin(B,,) . . Ascorbic acid (C)* Nicotinamide . . Pantothenicacid . . Molar ratio, [vitamin added]: [RF] . . 1000 . . 1000 . . 100 . . 100 . . 200 . . 50 . . 50 * Previously destroyed by oxidation. Table 5. Photokinetic determination of riboflavin and riboflavin 5’-phosphate in pharmaceutical preparations Riboflavin found/mg Photokinetic Fluorimetric Sample * Source method? method$ Protovit . . . . . . . . Roche 1.02 f 0.04 1.01 Apiroserum dextrobergone Ibys 0.59 k 0.02 0.58 Polybion C . . . . . . Merck 14.6 k 0.04 14.9 Becozyme . . . . . . Roche 3.96 f 0.05 4.02 Riboflavin-5-phosphate foundmg Photokinetic Fluorimetric method? method$ Organzyme B .. . . . . Bama 7.75 k 0.03 7.68 * Composition of samples. Protovit: riboflavin, 1 mg; thiamine, 2, mg; nicotinamide, 10 mg; pyridoxine, 1 mg; biotin, 0.1 mg; ascorbic acid, 50 mg; dexpanthenol, 10 mg; retinol, 5000 I.U.; ergocalciferol, 1000 I.U.; saccharin; 5 mg; and water, 1 g. Apiroserum dextrober- gone: riboflavin, 0.6 mg; thiamine, 0.5 mg; nicotinamide, 0.5 mg; pyridoxine, 0.3 mg; glucose, 5 g; sodium chloride, 0.85 g; and water, 100 g. Polybion C: riboflavin, 15 mg; thiamine, 15 mg; nicotinamide, 50 mg; calcium pantothenate, 25 rng; pyridoxine, 10 mg; biotin, 0.15 mg; cobalamin, 10 pg; ascorbic acid, 300 mg; saccharin, 15 mg; and excipient, 5 g. Becozyme: riboflavin, 4 mg; thiamine, 10 mg; pyridoxine, 4 mg; nicotinamide, 40 mg; biotin, 0.5 mg; cobalamin, 8 pg; panthenol, 6 mg; and water 2 mg.Organzyme B: riboflavin phosphate, 8 mg; thiamine, 20 mg; pyridoxine, 10 mg; calcium pantothenate, 25 mg; nicotinamide, 50 mg; cyanocobalamin, 25 pg; biotin, 0.2 mg; and excipient, 0.5 g. t Average of four determinations f s.d. $ Single determination by fluorimetry. From the results obtained above, it can be inferred that the fixed time method is the best method as it gave the best sensitivity, linear range and correlation coefficient (Table 3). Calibration Graphs In order to obtain the calibration graphs, three methods were used, the initial rate, fixed time and fixed concentration methods. The best method was chosen on the criteria of sensitivity (i. e., the slope of the calibration graph), linear range and correlation coefficient (r).Fixed concentration method Two procedures were used. In the first procedure the reciprocal of the time needed for the current of p-benzoquin- one to decrease to 50% of the initial value was plotted against RF and FMN concentration in the range 0.5-5p~. This calibration graph has the equation llt = 0.18Cwith r = 0.9970, where the time ( t ) is measured in seconds and the concentra- tion ( c ) in micromoles. In the second procedure, two platinum electrodes at an applied voltage of 100 mV were used to measure the reduction of all of the p-benzoquinone. The reciprocal of this time was plotted against the concentration of RF and FMN. Fixed time method The calibration graphs of current versus RF and FMN concentrations at times of 0.5, 1, 2, 3 and 5 min are shown in Tables 2 and 3.The best correlation coefficient was obtained for a fixed time of 2 min and this was chosen as the most suitable measuring time. Initial rate method A plot of the reaction rate (first 2 min) versus RF or FMN concentrations yields the equation V = 0.25C with r = 0.9900. Limit of Detection, Limit of Quantitation, Precision and Accuracy The limit of detection42 is 0.06 p~ and the limit of quantita- tion42 is 0.18 p~ for both RF and FMN. A study of the precision was performed by carrying out 11 independent measurements on solutions of various concentra- tions of RF and FMN and fixed concentrations of EDTA, p-benzoquinone and acetate buffer. The relative standard deviation was in the range 0.97-1.37% for RF and 0.59-1.46% for FMN at a concentration of 0.2-0.5 p ~ .The accuracy was studied in the range 0.2-5p~ RF and FMN. The linear regression of the values obtained for each analysis of each sample and the corresponding real values were obtained. The statistical t-test was applied to the study of the slope, and the intercept of the straight line was obtained. From this study it is concluded that this method does not present a systematic error (a slope equal to unity) and it does not need a blank correction (an intercept equal to zero). Selectivity The effect of the presence of other vitamins is shown in Table 4. The limiting value of the concentration of a foreign vitamin was taken as that value which caused an error of not more than 2.5%. The presence of oxidants or reductants that under the recommended experimental conditions can oxidise to hydro- quinone or reduce to benzoquinone is not compatible with this photokinetic method and they must be previously destroyed.Hence, for example, the interference of ascorbic acid is eliminated by oxidation with permanganate and the excess of permanganate is removed by adding nitrite; the excess of nitrite is then destroyed by adding urea.ANALYST, MARCH 1987, VOL. 112 241 ~~ Table 6. Determination of B2 vitamins in foods Labelled amount of vitamid mg Per Sample 100 g Wholemilk . . . . . . - Baby milk 1 . . . . . . 0.60 Baby milk 2 . . . . . . 0.47 Whitebread . . . . . . - Wholemeal bread (fortified) 0.79 Enrichedcornflour . . . . 1.50 Amount of Vitamin determined/mg per 100 g photokine tic method* 0.15 f 0.02 0.62 f 0.05 0.46 f 0.03 0.17 -C 0.02 0.77 k 0.04 1.52 f 0.02 * Average of four determinations f s.d.t Single determination by fluorimetry. Fluorimetric method? 0.14 0.61 0.47 0.18 0.76 1.46 Table 7. Determination of total flavins in animal tissues (rat) Total flavins found/yg g-1 Lumiflavin Photokinetic fluorescence Tissue method* methodt Heart . . . . . . . . 12.5 f 0.05 12.3 Kidney . . . . . . 27.8 k 0.04 28.1 Liver . . . . . . . . 24.6 f 0.06 24.9 * Average of four determinations f s.d. t Single determination by lumiflavin method. Applications The method has been applied to the determination of RF and FMN in pharmaceutical preparations. As can be seen in Table 5, the results were in good agreement with the label specifications of the pharmaceutical preparations and with those obtained by the official spectrofluorimetric meth0d.2~ Other applications have been the determination of vitamin B2 in foods (milk, baby milk, wholemeal bread, white bread and enriched corn flour) and in animal tissues. Tables 6 and 7 show the results obtained.Conclusions The results obtained demonstrate the suitability of photo- kinetic analysis as a simple method for the determination of riboflavin and riboflavin 5’-phosphate. The proposed method is sensitive and rapid and allows the determination of both B2 vitamers with good accuracy and precision, and is virtually free from interferences from many associated foreign vita- mins. From a review of the chemical methods for the determination of riboflavin (Table 1) the photokinetic method is seen to rank among the sensitive methods.References Strohecker, R., and Henning, H. M., “Vitamin Assay Tested Methods,” Verlag Chemie, Weinheim, 1965. Sebrell, W. H., and Harriss, R. S., “The Vitamins,” Volume 5, Academic Press, New York, 1967. Koziol, J., Methods Enzymol., 1971, 18B, 253. Zuman, P., “Polarography in Medicine, Biochemistry and Pharmacy,” Interscience, London, 1958, pp. 389-394. 5 . 6. 7. 8. 9. 10. 11. 12. 13. 14. 15. 16. 17. 18. 19. 20. 21. 22. 23. 24. 25. 26. 27. 28. 29. 30. 31. 32. 33. 34. 35. 36. 37. 38. 39. 40. 41. 42. Davidson, I. E., in Smith, W. F., Editor, “Polarography of Molecules of Biological Significance ,” Academic Press, Lon- don, 1979. Kalinowska, Z. E., Acta Polon. Pharm., 1964,21,365. Breyer, B., and Biegler, T., J.Electroanal. Chem., 1959/60, 1, 453. Lindquist, J., and Farroha, S. M., Analyst, 1975, 100, 377. Florence, T. M., J. Electroanal. Chem., 1979, 97, 219. Wang, J., Luo, D. , Farias, P., and Mahmond, J., Anal. Chem., 1985, 57, 158. Sawamoto, H., J. Electroanal. Chem., 1985, 186,257. Ang, C. Y. W., and Moseley, F. A., J. Agric. Food. Chem., 1980, 28, 483. Kwok, R. P., Rose, W. P., Tabor, R., and Pattison, T. S., J. Pharm. Sci., 1981, 70, 1014. Kamman, J. F., Labuza, T. P., and Warthesen, J. P., J. Food Sci., 1980,45, 1497. Bognar, A., Dtsch. Lebensm. Rundsch., 1981, 77, 431. Skurray, G. R., Food Chem., 1981,7, 77. Woodcock, E. A., Warthesen, J. J., and Labuza, T. P., J. Food Sci., 1982, 47, 545. Mauro, D. J., and Wetzel, D. I., J. Chromatogr., 1984, 299, 281.Ichinose, N., Adachi, K., and Schwedt, G., Analyst, 1985,110, 1505. Ichinose, N., and Adachi, K., Analyst, 1986, 111, 391. Egberg, D. C., and Potter, R. H., J. Agric. Food Chem., 1975, 23, 815. Pelletier, O., and Madere, R., J. Assoc. Off, Anal. Chem., 1977,60, 140. Egberg, D. C., J. Assoc. Off. Anal. Chem., 1979, 62, 1041. “Official Methods of Analysis of the Association of Official Analytical Chemists,” Thirteenth Edition, Association of Official Analytical Chemists, Washington, DC, 1980. Ishibashi, N., Ogawa, T., Imasaka, T., and Kunitake, M., Anal. Chem., 1979, 51, 2096. Frisell, W. R., Chung, C. W., and Mackenzie, C. G., J. Biol. Chem., 1959,234, 1297. Hemmerich, P., Massey, V., and Weber, G., Nature (London), 1967,213,728. Walker, W., Hemmerich, P., and Massey, V., Helv. Chim. Acta, 1967, 50, 2269. Merkel, J. R., andNickerson, W. J., Biochim. Biophys. Acta, 1954, 14, 303. Holmstrom, B., Photochem. Photobiol., 1964, 3, 97. Enns, K., and Burgess, W., J. Am. Chem. SOC., 1965,87,5766. Hass, W., and Hemmerich, P., Z. Naturforsch., Teil B, 1972, 27, 1035. Elliot, L., and Bruice, T., J. Am. Chem. SOC., 1973,95,7901. Massey, V., Stankovich, M., and Hemmerich, P., Biochem- iMy, 1978, 17, 1. Lukasiewicz, R. J., and Fitzgerald, J. M., Anal. Lett., 1969,2, 159. Skoog, A., and West, D., “Principles of Instrumental Analy- sis,” Holt, Rinehart and Winston, New York, 1971. Smith, E. C., and Metzler, D. E., J. Am. Chem. SOC., 1963,85, 3285. Straus, G., and Nickerson, W., J. Am. Chem. SOC., 1961,83, 3187. Knobloch, E., Methods Enzymol., 1971, 18B, 305. Sierra, F., Sanchez-Pedreiio, C., Perez Ruiz, T., and Martinez Lozano, C., Anal. Chim. Acta, 1977, 94, 129. Perez-Ruiz, T., Martinez Lozano, C. , and Tom& V., Mikro- chim. Acta (Wien), 1985,II, 367. ACS Committee on Environmental Improvement, Anal. Chem., 1980, 52, 2242. Paper A61378 Received October 1 Oth, 1986 Accepted November 14th, 1986

ISSN:0003-2654    

DOI:10.1039/AN9871200237

出版商:RSC    

年代:1987

数据来源: RSC

摘要:

ANALYST, MARCH 1987, VOL. 112 243 Cathodic Stripping Voltammetry of 5-Fluorouracil A. J. Miranda Ordieres, M. J. Garcia Gutierrez, A. Costa Garcia and P. Tuiion Blanco Department of Analytical Chemistry, University of Oviedo, Oviedo, Spain and W. Franklin Smyth Department of Pharmacy, The Queen's University of Belfast, Medical Biology Centre, 97 Lisburn Road, Belfast BT9 7BL, UK An investigation has been carried out into the anodic differential-pulse polarographic and cathodic stripping voltammetric behaviour of the anticancer drug 5-fluorouracil at mercury indicator electrodes. The anodic peak, corresponding to the formation of a mercury salt, has been used for the determination of the drug in formulations with a relative standard deviation of 0.5%. The cathodic stripping voltammetric behaviour of this mercury salt can be used to monitor I O - 7 - I O - 8 ~ concentrations of 5-fluorouracil in a borax - HCIO4 buffer, pH 7.8. The interference of CI- with the cathodic stripping signals is only evident at concentrations of CI- four orders of magnitude higher than the concentration of the drug and uric acid was found not to interfere at up to equal concentrations of 5-fluorouracil.The cathodic stripping method has been applied to the determination of 5-fluorouracil in human serum. Keywords : 5- Fluorouracil determination; cathodic stripping voltammetry; anticancer drug 5-Fluorouracil is frequently used for the treatment of gastro- intestinal, head and neck carcinomas. The cytotoxic effect of 5-fluorouracil requires metabolic activation to 5-fluoro-2'- deoxyuridine-5-monophosphate, which inhibits the enzyme thymidylate synthetase.The cytotoxic effect is also assumed to be due to the formation of 5-fluorouridine triphosphate, which is incorporated in cells to form fraudulent RNA. 5-Fluorour- acil is metabolised in the liver to dihydro-5-fluorouracil, which is devoid of cytotoxic activity. The active 5-fluorouracil nucleotides are trapped within the cell as they cannot readily cross the cell membranes owing to their highly polar character. Microbiological assays1J have high sensitivities (ca. 10 ng ml-1) but may be unsuitable methods of determination when patients are also receiving antibiotics. Their accuracy at low concentrations of 5-fluorouracil has been questioned.3 The spectrophotometric determination of 5-fluorouracil in biological samples can only be used for the measurement of pg ml-1 concentrations.4~5 The method of Morimoto et aZ.5 utilises ion-pair extraction of the drug into methylene chloride followed by back-extraction into an aqueous solution.Gas chromatography of 5-fluorouracil is hampered by its polar character resulting in its adsorption on the column. Chromatography of the underivatised compound on packed columns requires the deactivation of the glass column by silanisation followed by treatment with Carbowax 20M and the use of a polar stationary phase (Versamide 900) coated on a re-silanised packing material.6 The use of capillary columns has been reported to minimise the adsorption phenomenon.7 5-Fluorouracil has in most instances been chromatographed after prior derivatisation and a multitude of reactions have been used for this purpose, e .g , methylation by flash alkylationg and reaction with dimomethane9 or methyl- iodide.10 Silylation has been used to convert the drug to its corresponding trimethylsilyl derivative. Gas chromatography with single ion monitoring12 seems to offer the highest sensitivity and under optimum conditions can have a limit of detection of the order of 1 ng ml-1. Detection by a nitrogen - phosphorus detector7 and an electron-capture detector13 also offers low ng ml-1 limits of detection, whereas detection using a flame-ionisation detector is only applicable to plasma levels exceeding 200 ng ml- 1.11 Reversed-phase liquid column chromatography using an acidic mobile phase and pBondapak c18 packing material has been widely used for the separation of 5-fluorouracil from endogenous material and metabolites formed by anabolic and catabolic activity.14715 Determination of the drug has been carried out using photometric detection at 254-280 nm.Limits of detection are in the range 10-100 ng ml-1 for plasma samples. The highest sensitivity was obtained using an extensive work-up procedure consisting of ion-exchange chromatography and solvent extraction, which minimise interferences from plasma constituents and allow large plasma samples and high detector sensitivities to be used.16 Dryhurst17 studied the d.c. polarographic reduction behavi- our of 5-halogenouracils. A four-electron process was obser- ved for 5-fluorouracil, resulting in the formation of dihydrour- acil and F-.The E4 value occurred at a very negative potential, -1.85 V vs. SCE, at pH 7.4 and was therefore of little value for the reliable determination of low concentra- tions of the drug. Palacek et aZ.18 studied the reaction of 30 purine and pyrimidine derivatives with the mercury electrode and noted that 5-fluorouracil gives a cathodic stripping voltammetric (CSV) response. This paper is concerned with a rapid, accurate and precise formulation assay of 5-fluorouracil using its anodic differential-pulse polarographic peak at a dropping- mercury electrode and differential-pulse cathodic stripping voltammetry at the hanging mercury drop electrode at low ng ml-1 concentrations for use in the determination of the drug in human serum samples.Experimental Instrumentation and Reagents Pulse polarographic measurements were performed using a Metrohm E-506 polarograph coupled with an E-505 polaro- graphic stand. The dropping-mercury electrode (DME) had mass of mercury rn = 2.97 mg s-1 and droptime t = 2.20 s measured in an open circuit in water under a mercury head of 63.8 cm. For cyclic and cathodic stripping voltammetric experiments a Metrohm E-611 potentiostat coupled with a Metrohm E-612 scanner was used. The curves were recorded244 ANALYST, MARCH 1987, VOL. 112 on a Graphtec WX-4421 X - Y plotter. A hanging mercury drop electrode (Metrohm EA-410) was used as a working electrode, with a drop area of 2.20 mm* (four divisions of the micrometer dial of this instrument). The potentials were measured vs.SCE. The reference electrode was separated from the assay solution by a salt bridge filled with pure background electrolyte. Britton - Robinson and sodium tetraborate - perchloric acid buffers were used as background electrolytes. All reagents were of analytical-reagent grade (Carlo Erba RPE) and the water was purified by using a Milli-Q (Millipore) system. Procedures Fluorouracil assay in formulations The content of a fluorouracil ampoule is transferred quantita- tively into a calibrated flask and diluted to a final volume of 1 1 with water. A 1.0-ml aliquot of this solution is diluted to 25 ml with background electrolyte (borax - HC104 buffer, pH 7.8) and transferred into the polarographic cell. After a 5-min purge with argon the polarogram is recorded between -0.20 and +0.15 V, using a pulse amplitude of 50 mV, a drop time of 1 s and a scan rate of 5 mV s-1.A calibration graph is prepared under the same conditions and used for the quantitative evaluation of the polarograms. Fluorouracil assay in serum Equal volumes (1.0 ml) of serum and saturated ammonium sulphate solution are mixed and then centrifuged. This solution is extracted with 12 ml of a mixture of diethyl ether and propanol (4 + 1). The organic layer is evaporated on a water-bath at 40 "C under an inert gas flow. The dry extract is dissolved in 10 ml of background electrolyte, transferred into the cell and after 5 min de-aeration, the CSV curve is recorded using a pre-concentration potential of +O. 17 V, a pre-concen- tration time of 30 s in quiescent solution and a potential scan rate of 30 mV s-1.The standard addition method is used for the determination of 5-fluorouracil in serum. Results Differential-pulse Polarography of Anodic Process at the Dropping-mercury Electrode Using the optimum operating conditions (drop time, 1 s; modulation amplitude, 50 mV; scan rate, 5 mV s-1; and a supporting electrolyte of borax - HN03, pH 7.8) 5-fluorour- acil gave an anodic peak at +0.03 V vs. SCE, corresponding to the formation of a mercury salt. The peak current was proportional to the concentration of the drug in the range 2 x l o - 5 - l o - 4 ~ (Fig. 1) and the technique has a detection limit (signal to noise ratio = 2) of l o - 5 ~ using these conditions. The slope of the calibration graph was 0.62 x l o 7 nA M-1.A two-peak pattern with a new wave forming at more positive potentials was observed at concentrations >2 X 10-4 M (Fig. 1) and this is consistent with the multilayer formation of mercury salts around the dropping-mercury electrode during the drop's lifetime. Application to Formulation Analysis An ampoule containing 250 mg of 5-fluorouracil in solution was taken and treated as under Procedures. The resulting anodic peak, when recorded ten times on the same solution, gave a mean peak current of 515 nA and relative standard deviation of 0.5%. This mean current corresponded to a mean concentration of the drug of 7.82 x ~ O - S M , which in turn corresponded to 254.3 mg of 5-fluorouracil in the original ampoule, a relative error of 1.6%. Seven different ampoules were then taken and the contents of the drug determined in each instance by this method.The differential-pulse method yielded a relative error of 4% and a relative standard deviation of 1.6%. Cyclic Voltammetry of 5-Fluorouracil at the Hanging Mercury Drop Electrode The effect of scan rate on the observed anodic (formation of the mercury salt of 5-fluorouracil) and cathodic (reduction of this mercury salt) cyclic voltammetric peaks was studied in the borax - HC104 supporting electrolyte, pH 7.8 (Fig. 2). The potential range -0.50 to +0.20 V was scanned at a hanging mercury drop electrode using a concentration of 5-fluorouracil of 5 X 1 0 - 6 ~ . The anodic peak height was linearly related to (scan rate)$ for scan rates between 5 and 100 mV s-1, whereas the cathodic peak only showed such linearity between 5 and 20 mV s-1. At higher scan rates the cathodic peak current fell away from linearity owing to the partial re-dissolution of the partially insoluble mercury salt on the hanging mercury drop electrode.The effect of the potential at which the scan was reversed from the anodic to the cathodic direction on the resulting A- -0.2 0.0 +( E N vs. SCE Fig. 1. Effect of concentration of fluorouracil on the peak of differential-pulse polarograms in borax - HC104 buffer (pH 7.8). Conditions: scan rate, 5 mV s-1; pulse amplitude, 50 mV; drop time; 1 s. Concentration of fluorouracil: A, 0; B, 2 X 10-5; C, 6 x 10-5; D, and E, 2 x 10-4 M. A' I -0.500 ENvs. SCE Fig. 2. Effect of sweep rate on the cyclic voltammogram of 10-5 M fluorouracil in borax - HC104 buffer (pH 7.8).Scan rate: A, 5; B, 10; C, 20; D, 50 and E, 100 mV s-1ANALYST, MARCH 1987, VOL. 112 100 80 P 2 60- 245 - - / / ( a ) 12.5 nA I Y ! 40tj ( b ) .’P I a L--LzG-y , , , , , , 0 20 40 60 80 100 120 140 160 180 Time/s Fig. 3. Effect of re-concentration time on stripping peak current of fluorouracil (pH f 8 ) . Fluorouracil concentration: A, 5 X lo-’; B, 10-6; and C, 5 x 1 0 - 6 ~ -0.03 +0.17 -0.03 +0.17 E N vs. SCE Fig. 4. Uric acid interference on the cathodic stripping peak current of fluorouracil. Sweep rate: 10 mV s-1. Pre-concentration potential: +0.170 V. Pre-concentration time: 60 s in stirred solution. (a) 4 x M uric acid M fluorouracil and (b) 4 x M fluorouracil and 2 x Fig. 5. Cathodic strip ing voltammetric determination of fluoro- uracil in human serum ager liquid - liquid extraction. (a) Serum blank; ( b ) serum spiked with 10-5 M fluorouracil; (c) standard addition of 100 p1 of l o - 4 ~ fluorouracil solution; and ( d ) standard addition of 200 p1 of M fluorouracil solution.For conditions see text cathodic peak height increased with increasing positive switching potential as increasing amounts of the mercury salt of 5-fluorouracil accumulated on the electrode surface. This effect was observed up to a +0.22 V switching potential. At potentials more positive than this the cathodic peak was increasingly distorted by the reduction of mercury ions formed by the anodic dissolution of mercury in the anodic scan. Cathodic Stripping Voltammetry of 5-Fluorouracil at the Hanging Mercury Drop Electrode The pre-concentration of the mercury salt of 5-fluorouracil was carried out in quiescent solution at a concentration of 5 X 1 0 - 6 ~ in the borax - HC104 buffer (pH 7.8) using the hanging mercury drop electrode, a scan rate of 50 mV s-1 and a pre-concentration time of 15 s.The resulting cathodic strip- ping voltammograms showed an optimum pre-concentration potential of +0.16 to +0.17 V. The peak currents varied in a linear fashion with pre-concentration time in the range 15-180 s for concentrations of the drug less than 10-6 M using a stirred solution (Fig. 3) The calibration graph of cathodic stripping peak current w. concentration in the range 2 x 10-8-1.2 x 1 0 - 7 ~ was then constructed using an accumulation time of 5 min in stirred solution plus 15 s in quiescent solution, an accumulation potential of +0.17 V, a final potential of -0.50 V and a scan rate of 50 mV s-1.Linearity was observed in this concentra- tion range with negative deviations from linearity at concen- trations >1.2 x 10-7 M due to completion of the monolayer of the mercury salt on the electrode surface with resulting losses in adsorption for non-monolayer species at higher concentra- tions of 5-fluorouracil in solution. A molar excess of C1- of up to four orders of magnitude had no effect on the cathodic stripping signal of the mercury salt of the drug, e.g., 10-3 M C1- could be tolerated in a cell solution containing 10-7 M 5-fluorouracil. It would appear that under the operating conditions chosen for the cathodic stripping voltammetry, the formation, and more importantly the adsorption, of Hg2C12 is insignificant, a valuable factor in the analysis of certain biological fluids by this method.Uric acid, a potential interferent in urine analysis, is a more serious problem in that only an equimolar concentration can be tolerated without alteration of the peak height of the cathodic stripping signal of the mercury salt of 5-fluorouracil (Fig. 4). At higher molar excesses of uric acid, the adsorption of the mercury salt of uric acid results in a competition with the mercury salt of the drug for sites on the hanging mercury drop electrode. Both these mercury salts subsequently give reduc- tion signals during the cathodic stripping step. Determination of 5-Fluorouracil in Human Serum Fig.5 illustrates the cathodic stripping voltammetric determi- nation of 5-fluorouracil determined in human serum following the solvent extraction procedure outlined under Procedures. Because uric acid is a serious interferent, the use of solvent extraction is absolutely necessary. The 4 + 1 ratio of diethyl ether and n-propanol as solvent is also critical. If the ratio is lower, electroactive species are extracted giving interfering peaks close to the DPCSV signal of the drug. If the ratio is higher the recovery of 5-fluorouracil drops. The organic phase volume must be at least six times higher than that of the aqueous phase. The limit of detection was found to be 5 X 10-6 M and the mean recovery was 43%. Dr. P. Tuiion Blanco and his co-workers in Oviedo acknow- ledge the support of the CAICYT organisation in Madrid for research project No.1182/81, of which the work detailed in this paper forms a part. References 1. 2. Garrett, E. R., Hurst, G. H., and Green, J. R., Jr., J. Pharm. Sci., 1977, 66, 1422. Brandberg, A., Almersjo, O., Falsen, A., Gustavsson, B., Hafstrom, L., and Lindblom, G. B., Acta. Pathol. Microbiol. Scand., 1977,85,227. 3. Kawabata, N., Sugiyama, S . , Kuwarmura, T., Odaka, Y., and Satoh, T., J. Pharm. Sci., 1983, 72, 1162. 4. de Leenheer, A. P., Cosyns-Duyck, M. C., and van Vaeren- bergh, P. M., J. Pharm. Sci., 1977, 66, 1190. 5. Morimoto, Y., Akimoto, M., Sugibayashi, K., Nadai, T., and Kato, Y., Pharmazie, 1981, 36, 155. 6. Driessen, O., De Vos, D., and Timmermans, P. J. A., J. Chromatogr., 1979, 162, 451.246 ANALYST, MARCH 1987, VOL. 112 7. de Bruijn, E. A., Driessen, O., van den Bosch, N., van Strijen, E., Slee, P. H. Th. J., Van Oosterom, A. T., and Tjaden, U. R., J. Chromatogr., 1983,278,283. 8. Panarotto, C., Martini, A., Belvedere, G., Bossi, A., Donelli, M. G., and Frigerio, A., J. Chromatogr., 1974,99,519. 9. Min, B. H., and Garland, W. A., Res. Commun. Chem. Pathol. Pharmacol., 1978, 22, 145. 10. Cano, J. P., Rigault, J. P., Aubert, C., Carcasson, Y., and Seitz, J. F., Bull. Cancer, 1979, 66, 67. 11. Windheuser, J. J., Sutter, J. L., and Auen, E., J. Pharm. Sci., 1972, 61, 301. 12. Aubert, C., Sommadossi, J. P., Coassolo, P., Cano, J. P., and Rigault, J. P., Biomed. Mass Spectrom., 1982, 9, 336. 13. Pantarotto, C., Fanelli, R., Fillippeschi, S., Facchinetti, T., Spreafico, F., and Salmona, M., Anal. Biochem., 1979,97,232. 14. Miller, A. A,, Benvenuto, J. A., and Loo, T. L., J. Chromatogr., 1982,228, 165. 15. Quebbeman, E. J., Hoffman, N. E., Hamid, A. A. R., and Ausman, R. K., J. Liq. Chromatogr., 1984,7, 1489. 16. Buckpitt, A. L., and Boyd, A. R., Anal. Biochem., 1980,106, 432. 17. Dryhurst , G., “Electrochemistry of Biological Molecules,” Academic Press, London, 1977, pp . 238-241. 18. Palacek, E., Jelen, F., MacAnh, H., and Lasovsky, J., Bioelectrochem. Bioenerg., 1981, 8, 62. Paper A61309 Received September 4th, 1986 Accepted October loth, 1986

ISSN:0003-2654    

DOI:10.1039/AN9871200243

出版商:RSC    

年代:1987

数据来源: RSC

摘要:

ANALYST, MARCH 1987, VOL. 112 247 Investigation of the Adsorptive Stripping Voltammetric Behaviour of the Anticancer Drugs Chlorambucil and 5-Fluorouracil Joseph Wang, Meng Shan Lin and Vince Villa Department of Chemistry, New Mexico State University, Las Cruces, NM 88003, USA Adsorptive stripping voltammetry provides highly sensitive determinations of the amounts of the anticancer drugs chlorambucil and 5-fluorouracil. A static mercury drop electrode is immersed in a stirred solution of the drug for a fixed time (60-300 s) at a suitable potential and the adsorbed species is then stripped in the linear scan mode. The pre-concentration potentials and stripping peak potentials (versus Ag - AgCI) are -0.6 and - 1.30 V, respectively, for chlorambucil, and -0.25 and -0.42 V, respectively, for 5-fluorouracil.Cyclic voltammetry is used to explore the interfacial and redox behaviours. Short pre-concentration periods suffice to quantify chlorambucil and 5-fluorouracil down to the 3 x 10-8 and 3 x 10-9 M levels, respectively. The effects of possible interferences indicate that routine clinical applications would require an appropriate sample pre-treatment procedure. Keywords: Chlorambucil; 5-fluorouracil; anticancer drugs; stripping voltammetry, adsorptive pre-con- cen tra tion It has been shown recently that highly sensitive measurements of various pharmaceutical compounds can be achieved by means of adsorptive stripping voltammetry. ~2 This technique utilises controlled interfacial accumulation of the analyte on to the working electrode as an effective pre-concentration step prior to the voltammetric measurement of the surface-bound species.Procedures for quantifying important drugs such as diazepam,2 digoxin ,3 chlorpromazine,4 tetracyclines,5 tri- cyclic antidepressants6 or streptomycin7 down to the 10-9- 10-10 M level, have been reported. Recent activity in this field is being aimed at exploiting the inherent sensitivity and simplicity of the method for trace measurements of potent anticancer drugs. Anticancer drugs have become increasingly important in the treatment of neoplastic diseases. In view of their inherent risk of toxicity, the actual dosage must be carefully controlled. ‘This requires detailed knowledge of plasma levels, tissue distribution or excretion of anticancer drugs. The development of highly sensitive and selective analytical methods for measuring anticancer drugs is therefore essential to obvious clinical problems.Recent adsorptive stripping procedures for measuring trace levels of the anti- tumour agents methotrexate,g adriamycing or ciplatinlo have illustrated the potential of the method in this direction. This paper describes highly sensitive adsorptive stripping voltammetric procedures for quantifying trace amounts of chlorambucil and 5-fluorouracil. Chlorambucil is an alkylating agent which is widely used to treat chronic lymphocytic leukaemia, malignant lymphomas, ovarian carcinomas and Hodgkin’s disease. The fluorinated pyrimidine 5-fluorouracil is frequently used in the treatment of a wide variety of solid tumours. A number of spectrophotometric, chromatographic and radioimmunoassay methods have been described for the determination of chlorambucil and 5-fluorouracil.11-14 Despite the inherent sensitivity of voltammetric techniques, electroanalysis has rarely been attempted for trace measure- ments of these anticancer drugs. Whereas no reports are available on the voltammetric quantification of chlorambucil, a brief study describing the flow injection analysis of 5-flu- orouracil with voltammetric detection was published on completion of the present work.15 In that paper, an enhanced response due to interfacial accumulation was reported, allowing quantification of micromolar concentrations of the drug. The interaction between various purine and pyrimidine derivatives with mercury electrodes was first described by Palecek et al.l 6 A sensitive stripping procedure, based on this interaction, was exploited recently for trace measurements of the uracil derivative pseudouridine .17 As illustrated in the present study, nanomolar and submicromolar levels of 5-flu- orouracil and chlorambucil, respectively, can be measured following their controlled accumulation at the hanging mer- cury drop electrode. Conditions for enhancing the surface concentration of these drugs have been carefully optimised. In addition to its analytical utility, the reported interfacial and redox behaviours can offer a better understanding of the antineoplastic activity of 5-fluorouracil and chlorambucil. Experimental A PAR 264A voltammetric analyser with a PAR 303A static mercury drop electrode and other equipment were used as described previously .33 Stock solutions (5 x 10-4 M) of 5-fluorouracil and chlor- ambucil (Sigma) were prepared daily by dissolution in water and diethyl ether, respectively.Supporting electrolytes were 0.025 M boric acid solution (adjusted to pH 10 with sodium hydroxide) for 5-fluorouracil and acetate buffer (pH 5.4) for chlorambucil. The supporting electrolyte solution (10 ml) was added to the cell and de-gassed with nitrogen for 8 min. The pre-concentra- tion potential was applied to the working electrode for a selected time, while the solution was stirred at 400 rev min-1. The stirring was then stopped and, after a 15-s rest period, a negative-going scan was initiated, with simultaneous record- ing of the resulting voltammogram.After background voltam- mograms had been recorded, aliquots of the drug standard were introduced and the adsorptive stripping cycle was repeated with a new drop. (A 4-min nitrogen purge was required after the addition of chlorambucil to minimise changes in the background current attributed to diethyl ether, used for the drug dissolution.) All data were obtained at room temperature. Results and Discussion Determination of 5-Fluorouracil Fig. l ( a ) , B illustrates repetitive cyclic voltammograms for 5 X 10-7 M 5-fluorouracil recorded after stirring for 1 min at -0.25 V. A large and defined cathodic peak, due to the reduction of the adsorbed drug, is observed at the first scan (designated as 1) at ca. -0.42 V. A small oxidation peak (at -0.34 V) is observed on scanning in the anodic direction.ANALYST, MARCH 1987, VOL.112 248 .- (a) I200 nA 1 4.0 3.0 .- FI) -I 2.0 1 .o 0.40 0.45 3 0.50 0.55 -0.25 -0.45 -0.65 -0.85 -1.05 -2.5 -1.5 -0.5 EN Log VN s-1 Fig. 1. (a) Re etitive cyclic voltammograms for 5 x 10-7 M 5-fluorourad in M borate buffer (pH 10) solution, after stirring for 1 min (400 rev min-I) at -0.25 V (B). (A) An analogous voltammogram without accumulation. Scan rate, 50 mV s-1. (b) Dependence of the logarithm of the peak current (A) and of peak potential (B) on the logarithm of the potential scan rate r I(a’ T -0.25 -0.40 -0.55 EN ( b ) 60 180 300 Time/s Fig. 2. (a) Effect of re-concentration period on the stripping voltammogram for 1 X 1g-l M 5-fluorouracil in 0.025 M borate buffer (pH 10) solution.Pre-concentration period: (A) 0; (B) 30; (C) 60; (D) 120; and (E) 180 s. (b) Current - time graphs at (A) the 5 X 10-8 M and (B) the 1 x M levels. Scan rate, 50 mV s-l Subsequent scans exhibit substantially smaller cathodic and anodic peaks, indicating rapid desorption from the surface. Voltammogram A represents the analogous response without accumulation; the voltammetric peaks are substantially smaller than those obtained following accumulation. The effects of the potential scan rate ( v ) on the peak current (ip) and potential were evaluated for the surface-bound 5-fluo- rouracil [Fig. l ( b ) ] . A log i, verms log v graph was linear over the 5-100 mV s-1 range (A), with a slope of 1.06 (correlation coefficient, 0.999). A slope of 1.0 is expected for an ideal reaction of surface species.The peak potential shifts nega- tively on increasing the scan rate (B), first slowly (ca. 11 mV between 5 and 20 mV s-1) and then more rapidly (ca. 142 mV between 20 and 500 mV s-1). By the use of 5 X 10-7 M solutions, surface saturation was observed after 60 s. Under these conditions, the charge transferred in the reduction step corresponds to 0.63 pC (as was calculated by integration of the peak). A surface coverage of 2.2 x 10-10 mol cm-2 can be calculated by division of the charge by the conversion factor nFA. Each adsorbed 5-fluorouracil molecule therefore occu- pies an area of 0.76 nm2. - -0.25 -0.40 -0.55 E N 200 P 100 0 5 10 15 20 25 Concentration/l 0-8 M Fig. 3. (a) Stripping voltammograms obtained for solutions of increasing 5-fluorouracil concentration (from 2.5 x 10-8 to 12.5 x 10-8 M (A-E).Pre-concentration period, 90 s. Other conditions as in Fi . 1. Broken lines represent the response without accumulation. (bf Current - concentration graphs after different pre-concentration times: (A) 0; (B) 45; and (C) 90 s The spontaneous adsorption of 5-fluorouracil can be used as an effective pre-concentration step prior to the voltammetric determination of the drug. Fig. 2(a) shows linear-scan voltammograms for 1 X 10-7 M 5-fluorouracil after different pre-concentration times. Although quantification at this level is not feasible without pre-concentration (graph A), well defined peaks are observed following pre-concentration. The longer the pre-concentration time, the more 5-fluorouracil is adsorbed, and the larger the peak current.For example, a 27-fold enhancement of the peak is observed following 180-s pre-concentration. Fig. 2(b) shows graphs of peak current versus pre-concentration time at two levels of 5-fluorouracil. As expected for this type of interfacial pre-concentration, deviations from linearity are observed at periods longer than 120 s (1 X The redox and interfacial behaviours of 5-fluorouracil, and hence the resulting adsorptive stripping response, are affected by the solution conditions. The response was examined in the presence of various supporting electrolytes, e.g. , ammonium chloride, sodium hydroxide, sodium acetate, potassium ni- trate, potassium chloride and boric acid - sodium hydroxide solutions.The best results (with respect to peak enhancement and shape, base-line current and reproducibility) were obtained with a boric acid - sodium hydroxide solution (pH 10). A pre-concentration potential of -0.25 V offered the best signal to background characteristics. The linear scan stripping waveform yielded improved response characteristics and a speed advantage over differential-pulse excitation, and was used in all subsequent work. Mass transport during the pre-concentration step influences the 5-fluorouracil response, indicating a fast rate of adsorption. For example, a stirring rate of 400 rev min-1 resulted in a 4.7-fold enhancement of the response over that obtained in a quiescent solution [conditions as in Fig. 2(a), C]. Quantitative evaluation is based on the dependence of the peak current on the 5-fluorouracil concentration.Fig. 3(a) shows voltammograms for solutions of increasing 5-fluorour- acil concentration, from 2.5 X 10-8 to 1.25 X M, after 90-s M) and 180 s (5 X 10-8 M).ANALYST, MARCH 1987, VOL. 112 249 pre-concentration. Well defined stripping peaks are observed over this concentration range. In contrast, the corresponding solution-phase response (broken line) is not useful for quantitative work at this level. Fig. 3(b) shows the resulting calibration graphs for the 2.5 X 10-8 to 2.5 X M range, obtained at different pre-concentration times. The response is linear up to 2.25 x 10-7 M (45s pre-concentration, B) and 1.5 X 10-7 M (90-s pre-concentration, C). Such deviations from linearity are expected for processes limited by adsorption of the analyte, where the calibration graph reflects the corre- sponding adsorption isotherm.Least-squares treatment of the initial linear portions yielded slopes of 6.6 (B) and 11.2 (C) nA per 10-8 M (correlation coefficients, 0.999 and 0.997, respec- tively). The detection limit was determined from the 2.5 X 10-8 M 5-fluorouracil response shown in Fig. 3(a), A. Based on a signal to noise ratio of 3, these data corresponded to a detection limit of 3 x 10-9 M (ie., 3.9 ng in the 10 ml of solution used). This value represents a lowering of detection limit by two orders of magnitude, compared with the recently reported voltammetric procedure for 5-fluorouracil. 15 The reproducibility was determined by eight successive measure- ments on a stirred 1 x 10-7 M 5-fluorouracil solution (60-s pre-concentration); the mean peak current was 47 nA, in the range 4 W 9 nA, with a relative standard deviation of 2%. The effect of several possible interferences (including non-electroactive surfactants and endogenous and exogenous electroactive species) on the 2 x 10-7 M 5-fluorouracil strip- ping response was evaluated [conditions as in Fig.2(a), C]. The competitive surface coverage of the model surfactants gelatin, albumin and camphor is illustrated in Fig. 4. Succes- sive 2 p.p.m. increments of the gelatin and albumin level yielded a gradual depression of the 5-fluorouracil response (up to 35 and 15% of its original magnitude at 10 p.p.m. albumin and gelatin, respectively). In contrast, similar additions of camphor yielded only a slight increase (up to 9%) in the 5-fluorouracil peak.The effect of other anticancer drugs, which can be co-administered with 5-fluorouracil, was also investigated. The addition of 6 x 10-7 M cyclophosphamide did not interfere in the quantification of 2 X 10-7 M 5-fluoro- uracil. In contrast, a 6 x 10-7 M addition of methotrexate yielded a large peak at ca. 0.15 V negative to the 2 X M 5-fluorouracil peak, hence severely distorting the response of interest. This antineoplastic agent is known to yield a defined adsorptive stripping response .g Ascorbic acid and copper ions were tested as model endogenous electroactive substances. Their presence at the 5 x 10-6 M level had no effect on the 2 x M 5-fluorouracil response. Reducible endogenous spe- cies, with similar redox potentials, are expected to interfere.Determination of Chlorambucil The interfacial accumulation of the alkylating agent chloram- bucil is indicated from repetitive cyclic voltammograms recorded following stirring for 1 min at -0.6 V (Fig. 5). The drug exhibits an irreversible reduction peak at -1.30 V. The first scan (designated as 1) yields a large current response due to the reduction of the adsorbed species. Subsequent voltam- mograms, recorded on continued scanning, yielded smaller and smaller peaks, indicating gradual desorption of the product. By use of a 5 x 10-7 M chlorambucil solution, surface saturation was observed after 60 s. The amount of charge consumed during the cyclic voltammetry experiment by the redox process at saturation is 0.23 pC.A graph of log (peak current) against log (scan rate) for the surface-adsorbed chlorambucil over the range 5-500 mV s-1 was linear with a slope of 0.79. A 140 mV negative shift in the peak potential was observed on increasing the scan rate from 5 to 500 mV s-1. The current enhancement, associated with the interfacial accumulation, allows trace measurements of chlorambucil based on the adsorptive stripping approach. Fig. 6 shows linear scan voltammograms for 6 X 10-7 M chlorambucil after \ 0 5 10 Concentration, p.p.m. Fig. 4. Effect of (A) gelatin, (B) albumin and (C) camphor on the 5-fluorouracil(2 x 10-7 M) stripping peak. Other conditions as in Flg. 2(4, c .- I I I I I -0.6 -0.9 -1.2 -1.5 EN Fig. 5. Repetitive c clic voltammograms for 5 X M chlor- ambucil in acetate b d e r ( H 5.4) solution after stirring for 1 min at -0.60 V.Scan rate, 50 m$s-1 -1.0 -1.2 -1.4 -1.6 EN Fig. 6. Effect of pre-concentration period on the stripping voltam- mogram for 6 x 10-7 M chlorambucil. Pre-concentration period: (A) 0; (B) 30; (c) 60; and (D) 90 s. Other conditions as in Fig. 5250 E a c a -0.2 -0.6 -1.0 40 P * 20 0 I PH Fig. 7. Effect of (A) pH and (B) pre-concentration potential on the chlorambucil peak current. Chlorambucil concentration, (A) 3 x 10-7 M and (B) 4 x 10-7 M; pre-concentration time, 120 s; pre-concentration potential, (A) -0.6 V. Other conditions as in Fig. 5 different pre-concentration periods. Quantification at this level is not feasible without pre-concentration (A). The peak height increases rapidly with an increase in pre-concentration time, indicating enhancement of the chlorambucil concentra- tion on the mercury surface.For example, a 90-s pre- concentration period yields a 45-fold enhancement of the peak current , relative to that attained without pre-concentration. Among the various electrolytes tested, the best results were obtained with an acetate buffer solution. The pH dependence of the chlorambucil stripping peak was evaluated over the pH range 4.6-6.5 (Fig. 7, graph A). The peak was well developed over the pH range 5.0-6.5, yielding its maximum height at pH 5.4. The effect of the pre-concentration potential on the peak current was evaluated over the range -0.15 to -0.9 V (Fig. 7, graph B). The stripping current was low at -0.15 V, then increased rapidly to a maximum value (using potentials ranging from -0.40 to -0.65 V), and finally decreased.Acetate buffer (pH 5.4) and a pre-concentration potential of -0.60 V were used in all subsequent work. The linear scan and differential-pulse waveforms yielded similar response charac- teristics; the former was used throughout because of its speed advantage. A series of eight concentration increments, from 2 X 10-7 to 1.6 x 10-6 M chlorambucil, were used to evaluate the linearity and sensitivity. Fig. 8(a) shows five voltammograms from this series, over the range 2 X 10-7-1 X M (A-E) following 90-s pre-concentration. Well defined stripping peaks are observed. In contrast, the corresponding response without pre-concentration (broken lines) is not useful for quantitative work at this level.Fig. 8(b) shows the resulting calibration graphs for the range 2 x 10-7-1.6 x 10-6 M obtained at different pre-concentration times [(A) 0, (B) 45 and (C) 90 s]. With 45-s pre-concentration, the response is linear for the entire range (slope, 5.4 nA per 10-7 M; correlation coefficient 0.983). In contrast, the 90-s data exhibit linearity up to 1 X 10-6 M (slope of the initial linear portion, 9.1 nA per M; correlation coefficient, 0.998). A detection limit of 3 X 10-8 M of chlorambucil can be calculated based on the signal to noise ratio (SIN = 3) of the voltammogram shown in Fig. 8(a), A. ANALYST, MARCH 1987, VOL. 112 EN 120 Ib’ P 80 .a 40 0 4 8 12 16 Concentration/lO-7 M Fig. 8. (a) Strip ing voltammograms obtained for solutions of increasing chloramgucil concentration from 2 x 10-7 to 10 x 10-7 M (A-E).Pre-concentration period, 90 s. Other conditions as in Fig. 5. Broken lines represent the response without pre-concentration. ( b ) Resulting calibration ra hs after different pre-concentration times: (A) 0; (B) 45; and (Cf9fs The reproducibility was determined from ten successive measurements on a stirred 5 x 10-7 M chlorambucil solution (60-s pre-concentration). The mean peak current was 21.5 nA, with a range of 18.3-23.2 nA, and a relative standard deviation of 7%. The effects of various species that are likely to be in biological samples on the 5 x 10-7 M chlorambucil response were evaluated. A 1 x 10-6 M addition of the endogenous compounds uric or ascorbic acid had no effect on the chlorambucil peak.Similarly, no interference was observed in the presence of 1 x 10-6 M 5-fluorouracil. In contrast, severe interference was observed in the presence of 1 x 1 0 - 6 ~ methotrexate (that yielded a large overlapping response) and 2 p.p.m. gelatin (that caused a 40% diminution of the chlorambucil peak). In conclusion, highly sensitive voltammetric measurements of the anticancer drugs chlorambucil and 5-fluorouracil are feasible following their interfacial accumulation and reduction at the hanging mercury drop electrode. The above data suggest that sample pre-treatment (clean-up) procedures, e.g., selective extraction, common in the clinical laboratory, should be used for the practical determination of these drugs, particularly for separating interfering surfactants present in clinical samples.A separation step will be required for improving the specificity of the method, e.g., for differentiat- ing between the parent compound and its major metabolites. This work was supported by the National Institutes of Health, grants No. GM30913-03 and RR08136-13. References 1. Wang, J . , Am. Lab., 1985, 17, No. 5, 41. 2. Kalvoda, R., Anal. Chim. Acta, 1984, 162, 197. 3. Wang, J., Mahmoud, J. S., andFarias, P. A. M., Analyst, 1985, 110, 855. 4. Jarbawi, T. R., and Heineman, W. R., Anal. Chim. Acta, 1982, 154,359. 5. Wang, J., Peng, T., and Lin, M. S., Bioelectrochem. Bioenerg., 1986, 15, 147.ANALYST, MARCH 1987, VOL. 112 251 6. Wang, J., Bonakdar, M., and Morgan, C., Anal. Chem., 1986, 58, 1024. 7. Wang, J . , and Mahmoud, J. S., A n d . Chim. Actu, 1986, 186, 31. 8. Wang, J . , Tuzhi, P., Lin, M. S., and Tapia, T., Talanta, 1986, 3,707. 9. Chaney, C. E., and Baldwin, R. P., Anal. Chem., 1982, 54, 2556. 10. Wang, J., Peng, T., and Lin, M. S., Bioelectrochem. Bioenerg., in the press. 11. Adair, C. C., Burns, D. T., Crockard, A. D., and Harriot, M., J. Chromatogr., 1985, 342, 447. 12. Ahmed, A. E., Koenig, M., and Farrish, H. H., J. Chro- matogr., 1982, 233,392. 13. Windheuser, J. J., Sutter, J. L., and Auen, E. J. Pharm. Sci., 1972, 61, 301. 14. Stetson, P. T., Shukla, V. A., and Ensminger, W. D., J. Chromatogr., 1985,344, 385. 15. Bouzid, B., and Macdonald, A. M. G., Anal. Proc. 1986,23, 295. 16. Palecek, E., Jelen, F., Hung, M. A., and Asovsky, J., Bioelectrochem. Bioenerg., 1981, 8, 621. 17. Palecek, E., Anal. Chim. Acta, 1985, 174, 103. Paper A61372 Received October 6th, 1986 Accepted November 12th, 1986

ISSN:0003-2654    

DOI:10.1039/AN9871200247

出版商:RSC    

年代:1987

数据来源: RSC