ISSN:1359-7337    

DOI:10.1039/AC99633FX040

出版商:RSC    

年代:1996

数据来源: RSC

摘要:

PRODUCTS DESIHED AND PRODUEED VALID ANMYTICAL MEASUREMENT BY AMAVSTS FOR AMAlYSTS RAISING QUALITY THROUGH THE VAM INITIATIVE VAM (Valid Analytical Measurement) is a DTI initiative, spearheaded by the Laboratory of the Government Chemist (LGC), in conjunction with the DTI in order to improve the quality of analytical measurements made in the UK, and to facilitate the mutual acceptance of analytical data across international boundaries.This is done by: defining and disseminating best analytical practice which will enable laboratories to deliver reliable developing the tools which enable laboratories to implement best analytical practice; working with analysts in other countries to ensure the comparability of analytical measurements The Royal Society of Chemistry is pleased to be working with the LGC to distribute VAM products - which include books, guidelines, videos and software - that are designed to be used by practising analysts, laboratory managers, analytical quality managers, and university lecturers and their students.results every time; across international boundaries. GUIDELINES FOR ACHIEVING , , QUALITY IN TRACE ANALYSIS M. Sargent & G. MacKay Presents a concise protocol that acts as a check-list of key issues for those involved in carrying out trace analysis, or who rely on the results.Softback book; 1995; ISBN 0-85404-402-7; €12.50 / $22 GENERAL PRINCIPLES OF GOOD SAMPLING PRACTICE N.T. Crosby & /.Pate/ A welcome reference for researchers and professionals who need to access the important information on how to sample, in the form of an instant, thorough and reliable guide.Softback book; 1995; ISBN 0-65404-412-4; €17.50 / $32 TRACE ANALYSIS: A STRUCTURED APPROACH Edited by E. Pritchard, with G. MacKay & J. Points , TO OBTAINING RELIABLE RESULTS L; ' A highly practical and systematic bench guide which deals with the science rather than the paperwork of quality systems, taking the analyst step by step through the stages of any trace analysis.Softback book; 1996; ISBN 0-85404-417-5; €69.50 / $125 --:---., INTERNATIONAL GUIDE TO QUALITY IN L- , ANALYTICAL CHEMISTRV: AN AID TO ' ACCREDITATION A CITAC* document, published by the LGC Provides laboratories with guidance on best practice for improving quality of the analytical operations they perform, and is cross referenced throughout to the related parts of IS0 Guide 25, IS0 9000 and OECD GLP Principles. Cooperation on International Traceability in Analytical Chemistry Softback book; 1996; ISBN 0-94892-609-0; f30 / $55 PROFICIENCY TESTING IN ANALYTICAL CHEMISTRY R.E.Lawn, M. Thompson & R.F. Walker Deals exclusively and comprehensively with the role of proficiency testing in the quality assurance of analytical data, addressing topics such as the organisation, effectiveness and costs and benefits of proficiency testing.Softback book; Due autumn 1996; €22.50 / $39 TRACING THE CHEMICAL LINKS An educational package considering the concepts of calibration and traceability in the context of analytical chemistry, and explaining how these are an essential part of achieving comparability for analytical measurements.Video plus illustrated booklet; 1995; €35 (inc VAT) / $75 / Rest of World f40 CONFIDENCE IN ANALYSIS An educational package looking at the way - analysts establish confidence in the validity of their results. It follows through a complex multi-stage analysis from sampling to reporting and shows how the uncertainty can be estimated.Video plus illustrated booklet; 1995; f35 (inc VAT) / $75 / Rest of World f40 VAMSTAT A computer-aided learning package for the analytical chemist. Includes techniques, practical examples and questions and covers: basic concepts; significance tests; handling non-normal data and outliers; ANalysis Of VAriance (ANOVA); quality of analytical data; regression and applications.Software package (IBM or compatible); Standard single user licence €90 (+VAT) / $160; Multi-user site licence €290 (+VAT) / $520; Discounts available for academic institutions. To order any of these products, or for more detailed information, please contact: Jenny McCluskey, Sales & Promotion Executive Royal Society of Chemistry Thomas Graham House, Science Park Milton Road, Cambridge CB4 4WF, UK Tel +44 (0) 1223 420066 Fax +44 (0) 1223 423623 E-mail: sales@rsc.org wwweb: http://chemistry.rsc .org/rsc/ Information ServicesNOMINATIONS FOR THE 1997 BENEDETTI-PICHLER AWARD The American Microchemical Society is soliciting nominations for the prestigious 1 997 Benedetti-Pichler Award. The award, established in 1966, is given annually to recognise mtstanding achievements in microanalytical chemistry.The award consists of a plaque and txpenses to attend the Eastern Analytical Symposium in Somerset, New Jersey, in November 1997 to receive the award at a session to honour the awardee. Nominations or fkther information, including at least two supporting letters should be sent no later than October 30, 1996 to: Dr Robert G. Michel Department of Chemistry University of Connecticut Storrs, CT 06269 USA Tel : +1 203 486 3143; Fax : +1 203 486 2981; E-mail : MICHEL@UCONNVM.UCONN.EDU Analytical Abstracts The premier source of current awareness information in analytical chemistry is now available on a single SilverPlatter CD-ROM.Analytical Abstracts on CD-ROM features: Approximately 200,000 items from 1980 onwards Easy to use SilverPlatter software (WindowsTM, Quarterly updates with more than 3,000 items 0 Unlimited searching - no additional costs DOS and MacintoshTM formats) Special Discount for Hardcopy Subscribers! Contact us today for further information and a FREE 30-day trial. Judith Barnsby, The Royal Society of Chemistry, Thomas Graham House, Science Park, Milton Road, Cambridge CB4 4WF, United Kingdom Tel: +44 (0) 1223 420066. Fax: +44 (0) 1223 423429 E-mail (Internet): marketing @ rsc.org Information Services

ISSN:1359-7337    

DOI:10.1039/AC996330X042

出版商:RSC    

年代:1996

数据来源: RSC

摘要:

ANPRDI 33(8) 255-292 (1 996) AUGUST 1996I IAnalv t i calIL ommu n icat I onsFormerly Analytical ProceedingsCONTENTSCO M MU N I CAT1 0 N S 255 Some Observations on the Chemiluminescent Reactions of Tris(2,2’-bipyridyl)ruthenium(111)With Certain Papaver somniferum Alkaloids and Their Derivatives-Neil W. Barnett,Richard D. Gerardi, Deborah L. Hampson, Richard A. RussellDifferential Pulse Polarographic Determination of [R-(31-a-(methyimin0)-I -azabicyclo[2.2.2]-octane-3-acetonitrile Monohydrochloride in Tablets Following Derivatization WithDimethyldioxirane-4. N. Ennis, P. C. BuxtonSimultaneous Determination of the Concentrations of Each Enantiomer of 1 -PhenylethylamineUsing Their Quenching of the Fluorescence of Two Chiral Fluorophores-Keith S. Parker, AlanTownshend, Simon J.BaleElectrogenerated Chemiluminescence Using Platinum Electrodes With Hydrogen PeroxidePre-treatment-Neil J. Kearney, Carl E. Hall, Roger A. Jewsbury, Stuart G. TimmisSol-Gel-based Amperometric Glucose Biosensor Incorporating an Osmium Redox Polymer asMediator-Tae-Myung Park, Emmanuel I . Iwuoha, Malcolm R. Smyth, Brian D. MacCraithAdsorptive Stripping Voltammetric Determination of Tryptophan at an ElectrochemicallyPre-treated Carbon-paste Electrode With Solid Paraffin as a Binder-Huaisheng Wang, HuiCui, Aimei Zhang, Renmin LiuIdentification of Selenium Species in Selenium-enriched Garlic, Onion and Broccoli UsingHigh-Performance Ion Chromatography With Inductively Coupled Plasma Mass SpectrometryDetection-Honghong Ge, Xiao-Jia Cai, Julian F. Tyson, Peter C. Uden, Eric R. Denoyer, EricBlockIn-situ Boric Acid Subtractive Process in the Determinations of Some Transition Metals inReactor Coolant Water by Electrothermal Atomic Absorption Spectrometry-Kimmo Tompuri,Jouni Tummavuori261265269271275279283287 Cumulative Author Index288 Technical Abbreviations and Acronyms289 Conference DiaryTHE ROYALSOCIETY OFC H EM I STRYinformationServices Cambridge, EnglandTypeset and printed by Black Bear Press Limited

ISSN:1359-7337    

DOI:10.1039/AC99633BX044

出版商:RSC    

年代:1996

数据来源: RSC

摘要:

Analytical Communications, August 1996, Vol33 (255-260) 255 Some 0 bservations on the C hemi luminescent Reactions of Tris(2,2’-bipyridyl)ruthenium(111) With Certain Papaver somniferum Alkaloids and Their Derivatives Neil W. Barnett, Richard D. Gerardi, Deborah L. Hampson and Richard A. Russell School of Biological and Chemical Sciences, Deakin University, Geelong, Victoria, 321 7, Australia A series of qualitative tests were performed to evaluate the chemiluminescent activity resulting from the reactions of tris(2,2’-bipyridyl)ruthenium(111) with certain Papaver somniferum alkaloids and their derivatives.The tests were conducted over the range pH 2 to 12 with both emission intensities and lifetimes being qualitatively monitored. The results of this preliminary investigation have revealed an emerging pattern, relating certain aspects of molecular structure (within a family of analogous compounds), to chemiluminescent activity. The presence of phenolic moieties, aromatic or quaternary nitrogens quenched the chemiluminescence.Of potential analytical utility was the facility to manipulate the kinetics of chemiluminescence generation with changes in molecular structure and variation of reaction pH The use of tris(2,2’-bipyridyl)ruthenium(11) [R~(bipy)3~+] as an analytical chemiluminescence (CL) reagent originated from observations made by Hercules and Lytle in 1966.’ An acidic solution R~(bipy)~2+ was oxidized by solid lead dioxide and upon addition to concentrated aqueous sodium hydroxide produced an emanation of orange light.The reaction was tentatively described as: PbO, (oxidation) OH- (reduction) Ru( bipy),*+ > Ru( b i ~ y ) , ~ + - [Ru( bipy),2+]* +R~(bipy),~+ + hv with [Ru(bipy)32+]* as the excited state responsible for the emission (about 600 nm).2 Since this initial observation, R~(bipy)~2+ CL has progressively emerged as a successful detection method for both flow injection (FI) and HPLC.3-34 The sensitive and selective determination of a number of analyte types including amino acid^;^-^ amine~,~O-l~ phar- maceuticals with an amine functionality; 15-23 oxalate,24-30 glucose,31 pyruvate,3* amino alcohols33 and ascorbic acid34 has been demonstrated.With only a few exceptions most analytes contain some kind of amine moiety. We have successfully utilized this CL reaction for the determination of oxalate in Bayer liquor29 and codeine in process streams.35 During the latter study considerable variation in the measured CL intensity was observed between codeine and other structurally related corn pound^.^^ While these results were fortuitous with regard to the selective determination of codeine in a complex process stream matrix, the observations could not be easily rationalized given the subtle differences in molecular structure between many of the compounds tested.During the same investigation,35 we quantified the significantly different CL responses for codeine, thebaine, 6-methoxycodeine, laudanosine, oripavine and morphine in the range pH 4 to 6. Similar effects of pH on CL response for other analytes has also been noted by several earlier workers.3,7.8.10,19,20 This investigation was conducted to establish if any relationships existed between either molecular structure or pH and the CL activity upon reaction with Ru(bipy)33+.The work was conducted qualitatively, without instrumentation, to ascertain the viability of pursuing the structure-activity relationship on a more quantitative basis, necessitating the development of purpose built instrumentation to monitor pH effects on Ru(bipy)33+ CL.Experimental Instrumentation Preliminary CL intensity versus time profiles were monitored using a simple stopped flow chemiluminometer. Buffered alkaloid solutions were delivered to a microscale reaction vessel (5 ml) using a peristaltic pump [Gilson (Worthington, OH, USA) Minipuls 31.Ru(bipy)33+ solutions were injected directly into the reaction vessel via a syringe which simultaneously triggered the data collection software. The CL signals were monitored using a bifurcated fibre optic cable (Alitea Medina, WA, USA) and a modified spectrofluorimeter [Hitachi (Tokyo, Japan) F10001. A personal computer (486-DX) fitted with an analogue-to-digital converter (PC-74LA 12 Bit A/D card, Boston Technologies, Australia) running control/data software (FCS version 1.1; A-Chem Technologies, La Trobe, Victoria, Australia) was used for data collection.A Corning 120 pH meter (Halstead, Essex, UK), was used for all pH measurements. Reagents Buffer solutions (10 mmol 1-I), from pH 2 to 12 were prepared as follows from analytical-grade reagents.Glycine, pH 2 (Merck, Poole, Dorset, UK) citric acid, pH 3 (Ajax Auburn, New South Wales, Australia)-sodium citrate (Merck), anhy- drous sodium acetate, pH 4, 5 and 6 (Merck), sodium dihydrogen orthophosphate, pH 7 and 8 (Ajax)-di-sodium hydrogen orthophosphate (Ajax), di-sodium tetraborate, pH 9 and 10 (Merck) and di-sodium hydrogen orthophosphate, pH 11 and 12 (Ajax).All the acids and bases used for pH adjustment were analytical-reagent grade or better. Tris(2,2’-bipyridyl) ruthenium(I1r) chloride hexahydrate (Strem Chemicals, New- buryport, MA, USA) in 0.05 moll-’ H2S04 was oxidized with solid lead dioxide (May and Baker, Dagenham, Essex, UK) and filtered prior to use. Pure samples of Papaver somniferum alkaloids and their derivatives were obtained from Glaxo Wellcome Australia as either free bases or salts.Buffer and sample solutions were prepared with purified and organic256 Analytical Communications, August 1996, Vol33 filtered water (Service Deionisation System, Continental Water Systems, Victoria, Australia). CL Responses From the Buffer Solutions As can be seen from Table 1, a blank response was associated with the buffer solutions from pH 7 to 12, this observation has also been made by other workers3 and was initially assumed to be due to the presence of hydroxide ions.In order to test this assumption, solutions of sodium hydroxide were prepared at concentrations of to 10-1 mol 1-1 (approximate pH range 8 to 14) and reacted with R~(bipy)3~+. Chemiluminescence was only observed for solutions at pH 12.5 or above which represents hydroxide ion concentrations in excess of those present within the buffered sample solutions.This result would appear to rule out any significant interference from hydroxide ions alone, within the pH range employed for the spot tests. However, it is possible that some CL signal may have been recorded using photometric rather than visual detection.The positive CL response recorded for the blank buffer solutions between pH 7 and 12 may have resulted from either concomi- tant impurities or the reagents themselves. The buffer solutions were made at double the original concentrations (20 mmol l-1) and the blank CL response experiments performed again. An appreciable increase in the respective CL emission intensities was observed for the solutions between pH 7 and 12.Given the nature of the compounds used to produce the buffers, the most probable explanation for these observations was the presence of certain, as yet, unknown concomitant impurities capable of producing CL upon reaction with Ru(bipy)33+. It should be remembered that all these observations were made visually, therefore in an instrumental situation such as FI or HPLC operating at pH 7 and above; serious background problems would occur, as reported by Brune and Bobbitt.3 Procedures Aqueous stock solutions (200 ml) of twelve Papaver somni- ferum alkaloids were prepared at various concentrations depending upon availability and solubility.The use of ultra- sonication and addition of a few drops of concentrated hydrochloric acid were often necessary to facilitate dissolution.Subsequently the analyte test solutions were prepared via dilution of the stock solutions with the eleven different pH adjusted buffers in the range pH 2 to 12. The pH of each analyte test solution was monitored before and after the addition of acidified Ru(bipy)33+ to evaluate the capacity of the buffer systems.All qualitative CL tests were performed (five times) in a darkened room under static conditions by the rapid injection of R~(bipy)~3+ (1 .O mmol 1-I, 0.5 ml) by calibrated pasteur pipette into the sample aliquots (20 ml). The absence or presence of CL was then recorded at each pH (see Table l), together with a visual estimation of the emission lifetime and relative intensity compared to that for codeine at a concentration of 5 X 10-5 mmol l-1.A further series of twelve compounds (derivatives of codeine and morphine) were also reacted with Ru(bipy)33+ and any resultant CL noted (see Table 2). Each of the compounds listed in Table 2 was made up in sodium acetate-acetic acid buffer ( = 20 mmol 1-1, pH 6) at approximately 10-4 mol 1-1 (&lo%) and reacted with R~(bipy)~3+ (10-3 mol 1-l, 0.5 ml).Results and Discussion It was noteworthy that, during the qualitative procedures outlined above, the green R~(bipy)~3+ was invariably reduced to give the orange Ru(bipy)32+ upon addition to the alkaloid solutions. This reduction of the CL reagent was observed irrespective of whether light was produced or not. The implication being, therefore, that only certain oxidative reaction pathways involving Ru(bipy)33+ and the alkaloids tested would result in the generation of CL.The data assembled in Table 1 shows clearly that the efficiencies and rates of the CL pathways were markedly affected by both reaction pH and alkaloid molecular structure. In order to best highlight any patterns of molecular structure and CL activity resulting from our observa- tions (see Tables 1 and 2) they will be discussed under three sub-headings.CL Production (Emission Intensity) In the first instance we shall consider only the relative emission intensities from the reactions conducted between pH 2 and 6. An examination of the molecular structures of the alkaloids (see Figs. 1 and 2) revealed a common sub-unit present in the majority of compounds tested (see Fig.3). In all instances where significant CL was observed two structural criteria were obeyed: ( i ) -RO, R’O and R”O were all either alkoxy or acetoxy and (ii) N- was either an acyclic or cyclic tertiary aliphatic amine. It has been suggested that the presence of either carbonyl, ether, hydroxyl or phenyl groups on an analyte molecule has a quenching effect upon Ru(bipy)32.+ CL.4,7,18,’9,36 Table 1 Qualitative CL spot-tests performed on Papaver Somniferum alkaloids reacted with tris(2,2’-bipyridyl)ruthenium(111) at various pH values PH Compound (concentration mol 1-1) 2 3 4 5 6 7 8 9 10 11 12 Buffer blank (2 x 10-2) Codeine (5 X 10-5) BC* 6-methoxy-codeine (7 x 10-5) BC A* Morphine ( 2 X A (1 x 10-4) A Oripavine (3 x 10-4) A Pseudomorphine Thebaine (2 X BF Laudanidine (6 x 10-5) A Laudanosine(6 X 10-5) A Narceine (1 x A Narcotine (1 X DC Papaveraldine (1 X A Papaverine (6 X 10-4) A A DW A DC A EW A EC Bt EC B EFll B EF B EF B G** B EF DC A DC A EC A EF G EF H+t EF DF EF DF EF EF EF G EF EF DF A EF A DC EF EC A A DF A EF A DC EF EC A A DF A EF A DC EC EF A A EF A EF A EF EC EF A A G A EF A G G G A A EF A EF A DC EF EF A A A A BF A B BC DC A A A A DC A B BC DC A A A A DC A DF DC DC A A A A G A DC DC DC A A * A = no CL observed.+ B = weak CL observed. * C = CL visible for >3 s. 5 D = moderate CL observed. 1 E = strong CL observed. 11 F = CL visible for < 3 s. * * G = poor buffer capacity-no result. ++ H = discernible CL response above that of the buffer solution.Analytical Communications, August 1996, Vol33 257 In contrast our observations imply only that the presence of a phenolic moiety (i.e., R, R' or R" = H) will prohibit the generation of CL with Ru(bipy)$+ (see morphine, pseudomor- phine, oripavine and laudanidine). Indeed the presence of carbonyls or non-phenolic hydroxyl groups at various positions Table 2 Qualitative CL spot-tests performed on Papaver Somniferum alkaloid derivatives reacted with tris(2,2'-bipyridyl)rutheniurn(m) at pH 6 Compound CL response Isocodeine A*Bt 2,2'-biscodeine AB 6-acetylcodeine AB 6-oxocodeine CTB 6-0x0- 14-hydroxy-7,8-dihydrocodeine AB 6-oxo-7,8-dihydrocodeine AB 1 0-oxocodeine AB 1 0-hydroxycodeine AD§ 3-ethoxymorphine CD 3,6-diacetylmorphine AB codeine-N-oxide En * A = strong CL observed.+ B = CL visible for > 3 s. * C = moderate CL observed. D = CL visible for < 3 s. 7 E = no CL observed. codeine-N-methyliodide E H3c0Y? CH3 17 H0'""'6 7 6 -methoxycodehe on codeine did not appear to inhibit the production of CL from this reaction (see codeine, isocodeine, 2,2'-biscodeine, 6-oxocodeine, 1 0-hydroxycodeine, 1 0-oxocodeine and 6-0x0- 14-hydroxy-7,8-dihydro-codeine, 6-0~0-7,8-dihydrocodeine). Also, narceine and narcotine contain carbonyl and cyclic ether groups and both these alkaloids gave significant CL with R~(bipy)~3+ between pH 2 and 6.It should also be noted that intense CL was generated when R~(bipy)~3+ was reacted with those alkaloids whose phenolic moieties had been converted to either an alkoxy (see codeine, 3-ethoxymorphine and laudan- osine), or an acetyl group (see 3,6-diacetylmorphine).The results in Table 2 appear to suggest that the oxidation of codeine at positions 6,7,8, 10 or 14 seems to have little or no effect upon the resultant products ability to produce CL with R~(bipy)3~+. However, reduction in the same region does appear to inhibit CL, since we noted that the analytically useful CL obtained from the reaction of thebaine with R~(bipy)~3+ was significantly lower in intensity than that seen with either codeine or 6-methoxycodeine (see Table 1).This was confirmed quantita- tively in a previous study.35 Clearly the second double bond present in thebaine has a major CL quenching effect (cf. 6-methoxycodeine) which may have arisen from some prefer- ential oxidation of the diene via a non-light producing pathway.With regard to the second of our structural criteria for the generation of CL, when the amine was either incorporated into an aromatic ring (as in papaverine and papaveraldine) or morphine papaverine codeine OH NCH3 p seudomorp hine oripavine thebaine laudanidine 1 audanosine narceine narcotine papaveraldine Fig.1 the same basic structure. Molecular structure of Papaver somniferum alkaloids listed in Table 1. The numbering system shown for codeine is also used for the alkaloids having258 Analytical Communications, August 1996, Vol33 H3C 0 HO g N C H 3 / isocodeine H3C 0 0 g N C H 3 / 6-oxocodeine H3C 0 1 0-hydroxycodeine Fig. 2 NCH3 0 NCH: H3 CO H3C-C-O(’” II 2,2’-biscodeine 6-acetylcodeine 0 0 %NcH3 0 H3C 0 NCH, 0 0 3-ethoxymorphe 3,6-&acetylmorphine codeine-N-oxide codeine-N -methylidde Molecular structure of Papaver sornniferurn alkaloid derivative listed in Table 2.quaternized (see codeine-N-oxide or codeine-N-methyliodide) OR’ no CL resulted from the reaction of these compounds with R~(bipy)~3+. The lack of production of CL from the reactions of either codeine-N-oxide or codeine-N-methyliodide with Ru- (bipy)33+ was somewhat inconclusive as only a single reaction pH was evaluated.The results from both the papaverine and papaveraldine experiments, however, seem to implicate aro- matic amines as quenchers of R~(bipy)3~+ CL for these particular substrate molecules. This postulate is supported by the extensive CL activity exhibited by the structurally related compound laudanosine (see Table 1 and Fig.1) which contains a cyclic tertiary amine rather than an aromatic amine. Holeman and Danielsonlg reported good sensitivity using this type of CL for the detection of five antihistamines after separation by HPLC. While four of their analytes exhibited aromatic amine RO I Fig. 3 Common structural subunit present in all the Papaver somniferurn alkaloids studied.Analytical Communications, August 1996, Vol33 259 - - - - codeine (pH 3) - codeine (pH 10) - .. . thebaine (pH 3) thebaine (pH 10) 0 1500 2000 2500 3000 Time/ms Emission intensity versus time profiles for CL reactions of Ru(bipy)33+ with codeine and thebaine at various pH values. Fig. 4 functionality (substituted pyridine ring) each compound also contained an acyclic aliphatic tertiary amine as well.Bock eful.37 have carried out extensive studies on the oxidative quenching of [R~(bipy)~Z+J* by a series of bipyridinium cations such as paraquat, where the amine is both aromatic and quaternized.37 More recently He et a1.22 reported the determina- tion of 6-mercaptopurene using Ru(bipy)32+ CL, the molecular structure of this analyte has an aromatic system containing four heterocyclic nitrogens.Clearly the type of aromatic amine present in the substrate dictates the extent of CL activity. Consider now the results in Table 1 obtained between pH 7 and 12. Two previously inactive compounds (morphine and pseudomorphine) both realised useful CL under the more alkaline conditions and also the emission intensities, for several species, increased with the rise in reaction pH.Most notable were the reactions with morphine, pseudomorphine, thebaine, laudanosine, narceine and narcotine. Noffsinger and Danielson’O attempted to correlate CL emission intensity with analyte pK, concluding that such a correlation acould not be made. Brune and Babbitt? have shown that maximum CL intensity can be observed at pH ranges above the analyte pK, (for certain amino acids).As pK, values were only available for codeine (6.05), morphine (9.85) and narceine (9.3),38 we could not fully evaluate their proposal. While the CL intensity versus reaction pH data for morphine and codeine (see Table 1) appears to support the thesis of Brune and Bobbitt,3 those for narceine were less conclusive.This may reflect the multi- functional nature of the Papaver somnifer-um alkaloids com- pared to the structural simplicity of simple amino acids.’ Of equal importance were those alkaloids which showed no CL activity namely, oripavine, laudanidine, papaverine and papa- veraldine. The latter two compounds both possess an aromatic nitrogen which appears to prevent the generation of CL at any pH.Oripavine has a phenolic group and a diene, both of which severely inhibit CL production. However, like morphine, laudanidine has only a single phenolic moiety so its complete lack of CL activity is difficult to understand especially at pH 7 and above. Also, laudanidine differs from laudanosine only in that the phenolic hydroxyl group on the former is converted to a methoxy on the latter which gave CL with R~(bipy)~3+ from PH 3 to 12.Kinetics of CL Production For those compounds which gave a significant CL response a qualitative evaluation of reaction kinetics was also made. It should be borne in mind that any discussion of either ‘fast’ or ‘slow’ kinetics is related to the production of the excited state [Ru(bipy)32+]* and not simply to either the reduction of Ru(bipy)33+ or the oxidation of a particular analyte.The results in Table 1 show no obvious relationship between molecular structure or pH and the kinetics of CL emission. Table 1 does show a minor trend towards faster kinetics with increased pH, there are, however, too many exceptions to draw any mean- ingful conclusions at this time.This is hardly surprising given the number of potential sites for oxidation on these molecules. Of more importance analytically is that the kinetics (and thermodynamics) of the CL reactions, for certain compounds of commercial interest, can be manipulated by changing the pH. Recently we have been able to demonstrate this in a series of preliminary experiments where the CL intensity versus time profiles for the reactions of codeine and thebaine with Ru(bipy)33+ were recorded at various pH values (see Fig.4). The selective nature of this CL has been utilized for the determination of codeine in process streams containing structur- ally similar alkaloids.35 However, this methodology35 was developed on a purely empirical basis prior to obtaining the data shown in Fig.4. Given the structural similarities of codeine and thebaine, the observed differences in both the kinetics and thermodynamics of their CL reactions with Ru(bipy)’3+ strongly suggests that there may be various light producing pathways clearly dependent upon subtle differences in molec- ular structure and/or chemical environment. Further detailed investigations are currently underway to further quantify and utilize such phenomena.While the current set of experiments has revealed an emerging pattern relating substrate molecular structure, reaction pH and CL activity, the model is far from perfect. We have, however, shown that the rate of CL production can be successfully manipulated for analytical purposes. The authors gratefully acknowledge the help and support from Glaxo Wellcome Australia, in particular their provision of the compounds tested.We should also like to thank our colleague Dr. S. Lewis for helpful discussions during the preparation of this manuscript. References 1 2 3 4 5 6 7 8 9 10 11 Hercules, D. M., and Lytle, F. E., J . Am. Chem. Soc., 1966, 88, 4745. Lytle, F. E., and Hercules, D. M., J .Am. Chem. Soc., 1969, 91, 253. Brune, S. N., and Bobbitt, D. R., Talanra, 1991, 38, 419. He, L., Cox, K. A., and Danielson, N. D., Anal. Lett., 1990, 23, 195. Uchikura, K., and Kirisawa, M., Anal. Sci., 1991, 7, 971. Uchikura, K., and Kirisawa, M., Chem. Lett., 1991, 1373. Brune, S. N., and Bobbitt, D. R., Anal. Chem., 1992, 64, 66. Lee, W.-Y., and Nieman, T. A., J . Chromatogr.A, 1994, 659, 11 1. Jackson, W. A., and Bobbitt, D. R., Anal. Chim. Acta, 1994, 285, 309. Noffsinger, J. B., and Danielson, N. D., Anal. Chem., 1987, 59, 865. Noffsinger, J. B., and Danielson, N. D., J . Chromatogr., 1987, 387, 520.260 Analytical Communications, August 1996, Vol33 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 Uchikura, K., and Kirisawa, M., Anal. Sci., 1991, 7, 803.Downey, T. M., and Nieman, T. A., Anal. Chem., 1992, 64, 261. Uchikura, K., Kirisawa, M., and Sugii, A., Anal. Sci., 1993, 9, 121. Nonidez, W. K., and Leydon, D. E., Anal. Chim. Acta, 1978, 96, 401. Danielson, N. D., He, L., Noffsinger, J. B., and Trelli, L., J . Pharm. Biomed. Anal., 1989, 7, 1281. Targove, M. A., and Danielson, N. D., J . Chronz. Sci., 1990, 28, 50.5. Holeman, J.A., and Danielson, N. D., Anal. Chim. Acta, 1993, 277, 55. Holeman, J. A., and Danielson, N. D., J . Chromatogr. A , 1994, 679, 277. Holeman, J. A., and Danielson, N. D., J . Chrom. Sci., 199.5, 33, 297. Greenway, G. M., and Knight, P. J., Anal. Proc., 1995, 32, 251. He, Z., Liu, X., Luo, Q., Yu, X., and Zeng, Y., Anal. Sci., 1995, 11, 415. Greenway, G. M., Knight, A. W., and Knight, P. J., Analyst, 1995, 120,2549. Rubinstein, I., Martin, C. A., and Bard, A. J., Anal. Chem., 1983,55, 1580. Uchikura, K., Bunseki Kagaku, 1990,39, 323. Egashira, N., Kumasako, H., and Ohga, K., Anal. Sci., 1990, 6, 903. 27 28 29 30 31 32 33 34 35 36 37 38 Egashira, N., Kumasako, H., Kurauchi, Y., and Ohga, K., Proc. Electrochem. Soc., 1993, 93, 674. Jirka, G. P., and Nieman, T. A., Mikrochim Acta, 1994, 113, 339. Barnett, N. W., Bowser, T. A., and Russell, R. A., Anal. Proc., 1995, 32, 57. Skotty, D. R., and Nieman, T. A., J . Chromatogr. B , 1995, 665, 27. Martin, A. F., and Nieman, T. A., Anal. Chim. Acta, 1993, 281, 475. Knight, A. W., and Greenway, G. M., Analyst, 1995, 120, 2543. Yamazaki, S., Ozaki, K., Saito, K., and Tanimura, T., J . High Res. Chrom., 1995, 18, 68. Chen, X., and Sato, M., Anal. Sci., 1995, 11, 749. Barnett, N. W., Bowser, T. A., Gerardi, R. D., and Smith, B., Anal. Chim. Acta., 1996, 318, 309. Vera, D. M., Arguello, G. A., and Arguello, G. A., J . Photochern. Photobiol. A. Chem., 1993, 76, 13. Bock, C. R., Connor, J. A,, Gutierrez, A. R., Meyer, T.J., Whiten, D. G., Sullivan, B. P., and Nagle, J. K., J . Am. Chem. Soc., 1979,101, 4815. The Merck Index: An Encyclopaedia of Chemicals and Drugs, Merck, Rahway, NJ, 1983, 10th edn.. Paper 6102667B Received April i7, 1996 Accepted June 13,1996

ISSN:1359-7337    

DOI:10.1039/AC9963300255

出版商:RSC    

年代:1996

数据来源: RSC

摘要:

Analytical Communications, August 1996, Vol33 (261-263) 26 1 Differential Pulse Polarographic Determination of [ R-( Z)] -a- ( m et h ox y i m i n 0 ) - 1 -aza b i c y c I o [ 2.2.21- octane-3-acetonitri le Mono hydrochloride in Tablets Following Derivatization With Dimet hyldioxi rane J. N. Ennis and P. C. Buxton Pharmaceutical Development, SmithKline Beecham Pharmaceuticals, Yew Tree Bottom Road, Burgh Heath, Epsom, Surrey, UK KT18 5XQ Dimethyldioxirane was used as a derivatizing agent to convert [R-(Z)]-a-(methoxyimino)-l-azabicyclo[2.2.2]- octane-3-acetonitrile monohydrochloride (I) to the corresponding quinuclidine N-oxide which was then determined by differential pulse polarography in the presence of common excipients.The polarographic method yielded an LOD of 1.2 pg ml-1 calculated from the linearity plot by conventional statistical means. However, by increasing the pulse amplitude to 100 mV, a 5.0 ng ml-1 solution could be detected.This constituted an approximate eight-fold improvement over conventional HPLC procedures. The method was suitable for dosage forms containing I at levels as low as 100 ppm. Samples were extracted with phosphate buffer solution of pH 7.0, diluted into Britton Robinson buffer solution of pH 3.0 and the polarographic response recorded at a peak potential of -0.97 V.The method was linear over the sample concentration range of 0.05 to 25 pg ml-1 with r = 0.9993. The precision for the determination of I at the 100 ppm level was 0.5%. New drug candidates are increasingly of high potency and low dose.When combined with poor spectroscopic properties the analysis of formulated products containing such drugs presents a significant challenge. Typical of this class of compound is [R-(a]-a-(methoxyimino)- 1 -azabicyclo[2.2.2]octane-3-aceto- nitrile monohydrochloride (I; Fig. 1) which has a modest UV response (A: = 260, h = 232 nm) and requires formulated product dosages in the microgram range. Good sensitivity can be attained by HPLC but large injection volumes are required. Furthermore HPLC currently provides the only means of product identification and there is, therefore, a requirement for an independent means of routine qualitative analysis.It is common practice in analytical chemistry to increase sensitivity or specificity by derivatization and this has been particularly successful for primary and secondary amines.Tertiary amines such as I are not so readily derivatized but it is conceptually possible to oxidize the quinuclidine moiety to the corresponding N-oxide and render the molecule susceptible to reductive electrochemical analysis. Such a derivatization was WCN Fig. 1 [2.2.2]octane-3-acetonitrile monohydrochloride (I).Chemical Structure of [R-(a]-mu-(methoxyimino)- 1 -azabicyclo- considered desirable following the failure to determine the compound directly by anodic oxidation. In principle the electron pair on the quinuclidine nitrogen atom is available for anodic oxidation using a solid electrode such as glassy carbon and applied potentials in the range 0-1 .O V. In practice the peak oxidation potential of I proved to be close to the electrode limit and severely limited the sensitivity of the technique. Reductive polarography of the N-oxide of I would be expected to occur in a potential region well clear of electrode or supporting electrolyte responses. Dimethyldioxirane (DMD) possesses advantageous proper- ties as a derivatizing agent in that it is volatile and hence easily removed from the sample matrix. It is also highly reactive and has been shown to carry out a wide variety of oxygen transfer reactions frequently in high yield.132 It was the aim of this work to examine the feasibility of this concept using the highly strained peroxide DMD as the derivatizing agent and a differential pulse polarographic method to analyse the resultant N-oxide (Fig.2). Differential pulse polarography was selected in preference to other forms of polarography due to its high sensitivity and ease of quantita- tion. Experimental Reagents and Instrumentation All reagents were of analytical-reagent grade. The polaro- graphic equipment consisted of a Princeton Applied Research (PAR) polarographic analyser in conjunction with a PAR 303 polarographic detector.The polarographic detector was oper- ated under the following conditions: initial potential, -0.5 V; final potential, -1.4 V; scan rate, 5 mV s-1; large drop size; scan increment, 4 mV; stepdrop time, 0.8 s; pulse height, 20 mV. The electrolyte used was Britton Robinson buffer (0.04 mol 1-1 in acetic acid, orthophosphoric acid and boric acid), adjusted to pH 3.0 with 1 moll-’ sodium hydroxide.Phosphate buffer solutions were pH 7.0 with a molarity of 0.05 mol 1-l. Preparation of Dimethyldioxirane A round-bottom flask was equipped with an efficient mechan- ical stirrer and a solids addition funnel, connected by means of F;1’OCH3 DMD, RT 10 MIN Y + 0 - Fig. 2 Derivatization of I with DMD.262 Analytical Communications, August 1996, Vol33 -0.45 -0.40 -0.35 -0.30 -0.25 d 3 -0.20 -0.15 -0.10 -0.05 -0.00 --0.05 -.'_ a U-tube to a receiving flask which was cooled to -78 "C by means of a dry ice-acetone bath. The reaction flask was charged with an aqueous EDTA solution, 4.3 mmol 1-l (300 ml), distilled acetone (210 ml) and sodium hydrogen carbonate (160 g). While vigorously stirring, solid oxone (potassium caroate; 320 g) was added slowly over a period of approximately 1 h.The DMD-acetone distillate (60 ml) was collected in the cooled (-78 "C) receiving flask. The molarity of the DMD solution was determined by titration. Acetic acid-acetone solution (3+2; 3 ml) and KI solution (10% m/v; 5 ml) were charged to a flask, followed by the DMD solution (0.2 ml). This solution was titrated against a sodium thiosulfate solution (0.01 moll-l).The DMD molarities obtained were in the range 0.08-0.1 mol 1-1. - -- -- -- -- -- -- -- -- -- I I I 1 I I 1 1 I Derivatization Efficiency Duplicate samples of a stock solution of I at a concentration of 0.252 mg ml-l, were transferred by pipette (15, 20 and 25 ml) into 100 ml calibrated flasks to achieve concentrations of 8.2, 10.9 and 13.6 pg ml-I.The mass of I in each flask corresponded to 3.75, 5.00 and 6.25 mg. Acetone (20 ml) and dimethyldi- oxirane solution (5 ml) were then added to each flask and the volume made up with distilled water. The solutions were allowed to stand for 10 min, 20 ml portions were then transferred by pipette into 100 ml calibrated flasks and diluted to volume with Britton Robinson buffer of pH 3.0. The solutions were analysed by polarography against an N-oxide of I standard prepared at the Chemical Development Laboratories of SmithKline Beecham Pharmaceuticals. Method Performance Conditions System repeatability Standard solutions of an N-oxide of I in pH 3.0 buffer solution (10.0 pg ml-1) were measured six times and the peak current recorded.Method repeatability Five tablets containing nominally 125 pg of I, in a lactose-based formulation, were placed in a calibrated flask (50 ml) and dissolved in phosphate buffer solution of pH 7.0 (20 ml).Acetone (20 ml) and DMD solution (5 ml) were added and the solution was set aside for 15 min. The flasks were made up to volume with further phosphate buffer of pH 7.0 and the resultant stock solution (5 ml) was transferred into a calibrated flask (25 ml) and diluted to volume with electrolyte solution of pH 3.0.Six replicates were prepared in this way and analysed. Quantification Sample solutions were quantified by reference to an externally prepared standard solution of the N-oxide of I at a concentration of 2.5 pg ml-1 in electrolyte solution. The content of I was calculated from the following formula: is v1 v3 Content = -.CR,-.- pg per tablet Where: is = sample peak current; iR = reference peak current; CR = reference concentration; V1 = working sample volume (25); V2 = dilution factor (5); V3 = stock sample volume (50); and N = number of tablets (5).1R v2 N Results and Discussion Optimization of Detection Conditions In order to determine the optimum conditions for reductive polarographic analysis the variation of electrode potential and peak response with pH of the N-oxide of I was measured (Table 1).The maximum response and separation from the reduction of the supporting electrolyte was observed at pH 3.0. This pH was, therefore, chosen as the optimum for the analysis and yielded well defined polarograms (Fig.3). Derivatization Efficiency The quantitative nature of the conversion (Fig. 2) was illustrated by derivatizing I and analysing the resulting solutions against a reference standard of the N-oxide of I. The percentage conversions are recorded in Table 2 with an overall mean of 101%. ~~~ Table 1 Variation of electrode potential and peak height with pH for the polarographic reduction of I N-oxide PH 2 3 4 5 6 7 8 9 Peak potential/V - 0.85 -0.97 -1.36 - 1.34 -1.31 -1.34 -1.34 - 1.34 Peak c urrent/nA 435 490 435 185 150 255 285 285 Table 2 Conversion of I to the N-oxide using DMD Working Concentration concentration of of I N-oxide Mean Mass of I N-oxidel determined/ Recovery recovery 3.75 8.2 8.2 100 101 8.2 8.3 101 5.00 10.9 11.0 101 101 10.9 11.0 101 6.25 13.6 13.6 100 101 13.6 13.8 101 Ilmg pgmml-1 pg ml-' (%I (%)Analytical Communications, August 1996, Vol33 263 -0.60 -- -0.50 d 3..- -0.40 \ -0.30 -0.20 The high conversions indicate that the derivatization may be regarded as quantitative on analytical solutions containing milligram quantities of I. -- -- -- -- Method Performance Linearity and limit of detection For solutions containing I in the absence of excipients good linearity was observed over the concentration range of 0.05 to 25.0 pg ml-I with r = 0.9993.The LOD calculated from the linearity plot by conventional statistical means3 was found to be 1.2 pg ml-l reflecting the uncertainties at the lower end of the calibration plot. By increasing the pulse amplitude to 100 mV, a 5 ng ml-l solution could be detected with a peak current of 13 nA representing an -0.70 T I.-0.00 1 I I I I I t -1.20 -1.00 -0.80 -0.60 -0.40 EIV Fig. 4 pulse height). Differential pulse polarogram of I (N-oxide (5 ng ml-I; 100 mV Table 3 Method performance for the dedrivatization of I precision System repeatability Current (n A) Concentration of I 0.5 pg ml-1 10.0 pg ml-I 74 345 75 343 71 344 73 345 75 346 72 342 Mean 73 344 S 1.6 1.47 RSD (%) 2.2 0.43 Method repeatability yg I tablet-' 2.5 pg ml-1 (nominally 125 pg per tablet) 123.3 122.6 122.5 121.8 123.3 123.4 122.8 0.63 0.50 approximately eight-fold improvement over conventional HPLC procedures with respect to the LOD (Fig.4). Precision Precision measurements conducted on solutions with concen- trations in the anticipated range were good.The system repeatability improved with increasing concentration as ex- pected and the method repeatability carried out by analysing tablets yielding a nominal assay concentration of 2.5 pg ml-l was 0.5% (Table 3). This compared favourably with HPLC procedures which yielded a method repeatability of 2.4%. Accuracy The tablets analysed for method repeatability contained an accurately known quantity of I of 125 pg per tablet.The mean result of 122.8 pg yielded an accuracy of 98.2%. Scope of Analysis Analysis of tertiary amines such as I can be achieved at low analytical concentrations. Limiting factors for the analysis of I were mainly matrix effects from excipient material rather than inherent lack of sensitivity. Simple carbohydrates such as lactose appear to be unreactive to DMD but other excipients that contain oxidizable material can limit the scope of the analysis.The tablet formula analysed in this work displayed large backgrounds sufficient to make the quantification impossible if the dose of I was reduced below 100 pg. The LODs reported here are those derived from the drug alone and in practice usable limits will depend more on the nature and level of excipients present in the samples.In addition to improvements in sensitivity and precision compared to conventional HPLC the generation of a polarogram characteristic of the drug provided an alternative means of identification. This is useful in satisfying regulatory require- ments with an experimentally straightforward procedure. Further benefits of this derivatization procedure might also be anticipated in the analysis of biological fluids where sensitivity and specificity are key requirements. For such procedures the ability to remove the volatile derivatizing agent could be a useful advantage. We would like to thank Professor B. A. Marples at Loughbor- ough University of Technology for his help in the discussions of this project. References Murray, R. W., Chem. Rev. (Washington, D.C.), 1989, 89, 1187. Adam, W., Curci, R., and Edwards, J. O., Acc. Chem. Res., 1989,22, 205. Miller, J. C., and Miller, J. N., Statistics for Analytical Chemistry, Ellis Horwood, Chichester, 1986, pp. 96-100. Paper 6/03330J Received May 13, I996 Accepted June 17,1996

ISSN:1359-7337    

DOI:10.1039/AC9963300261

出版商:RSC    

年代:1996

数据来源: RSC

摘要:

Analytical Communications, August 1996, Vol33 (265-267) 265 Simultaneous Determination of the Concentrations of Each Enantiomer of I-Phenylethylamine Using Their Quenching of the Fluorescence of Two Chiral Fluorophores Keith S. Parkera, Alan Townshenda,* and Simon J. Baleb uSchool of Chemistry, University of Hull, Hull. UK HU6 7RX Kent. UK CT13 9NJ Analyticul Research and Development, Pfzer Central Research, Sandwich, Quenching of the fluorescence of two chiral fluorophores, 2,2’-dihydroxy-l,l’-binaphthalene and 2,2’-diamino-l,l’- binaphthalene, has been used successfully to determine the concentrations of each of the two enantiomers of the chiral quencher 1-phenylethylamine in a mixture of both enantiomers.From the concentrations of each enantiomer, the enantiomeric excess may be calculated.The method has been shown to be accurate with an LOD of 10% of one enantiomer in the presence of the other. The need to determine the levels of each enantiomer in a sample of a chiral drug compound is becoming increasingly important to the pharmaceutical industry. There are many examples of differences in behaviour within the body of each enantiomer of a drug compound in terms of metabolism, excretion and effect.’-9 There is a need, therefore, to know the enantiomeric composition of a sample so that effects, both therapeutic and adverse, may be attributed to one or other enantiomer.The propensity of the compound to racemize or epimerize under the conditions in the body or on manufacture or storage also needs to be determined.Several methods are in use for the determina- tion of chiral purity including optical methods such as optical rotation measurement and separation methods including LC and GC. This paper is one of a series in which interactions between fluorophores and quenchers are used to discriminate between chiral forms of compounds. In this paper, a method involving the formation of a complex between a chiral analyte and a single enantiomer of a chiral fluorophore in its excited state, resulting in the quenching of the fluorescent emission of the fluorophore, is described. The three-dimensional arrangements of the groups in both fluorophore and analyte result in the formation of complexes of different stability for each enantiomer of the quencher analyte.The more stable the complex, the greater the likelihood that it will form, the greater the likelihood of transfer of energy from the excited state species to the analyte, and so the greater the efficiency of quenching.Differences in the effi- ciency of the quenching of a single enantiomer of a chiral fluorophore by each enantiomer of a chiral quencher have been seen in several cases, lO-15 most notably with binaphthalene or one of its derivatives as the fluorophore.16-19 We have shown previously that this interaction may be used to measure enantiomeric purity.2”JI Under appropriate conditions, fluorescence quenching obeys the Stern-Volmer relationship: * To whom correspondence should be addressed where: Q,o is the fluorescent intensity in the absence of quencher; Q, is the fluorescent emission intensity in the presence of quencher; [Q] is the concentration of quencher; and Ksv is the Stern-Volmer constant, a measure of the efficiency of quenching.A plot of @o/@ versus [Q] (a Stern-Volmer plot) will give a straight line with a gradient Ksv and an intercept of I. Where two quenchers (in this case the two enantiomers) are present this equation may be extended to take into account the effect of the second quencher.22 where Ksvl and Ksv2 are the Stern-Volmer constants for the quenching by quencher 1 at concentration [QI] and by quencher 2 at concentration [Q2], respectively.The use of a different fluorophore will produce a similar equation, with different Ksv values. The values of KSV, and KSV2 for each fluorophore may be determined by quenching with a pure sample of each enantiomer. Thus the values of [Q1] and [Q2], the concentrations of each enantiomeric quencher, may be determined by solution of the two quenching equations.This method has been used successfully to measure the levels of different halides in a mixture by Wolfbeis and Urbano.22 In order to test the validity of this general principle, 1 -phenylethylamine has been used as a quencher.Each enantiomer of 1 -phenylethylamine may be viewed as a separate quencher, with its own quenching efficiency, for the quenching of the fluorescence of a chiral fluorophore. A single enantiomer of 2,2’-dihydroxy- 1,l ’-binaphthalene was used as one fluor- ophore; the quenching of its fluorescence has already been used in the calculation of the enantiomeric purity of 1 -phenyl- ethylamine.2I A single enantiomer of 2,2’-diamino- 1 , 1’-bina- phthalene was used as the second fluorophore. Experimental Apparatus and Reagents The binaphthalene derivatives used were of puriss grade and samples of the enantiomers of 1 -phenylethylamine used were of purum grade, all obtained from Fluka, Gillingham, Dorset, UK.The solvents used were HPLC grade obtained from Merck, Poole, Dorset, UK.All measurements were made on a Perkin-Elmer Model 3000 fluorescence spectrophotometer equipped with a 1 cm quartz cell, with excitation and emission slits set at 15 and 5 nm, respectively. Procedure A stock solution containing (S)-2,2’-dihydroxy- 1 , 1’-binaph- thalene at a concentration of 80 pmol 1-1 was prepared as266 Analytical Communications, August 1996, Vol33 follows.The compound (23 mg) was dissolved in 100 ml of acetonitrile. A 10 ml aliquot was diluted ten-fold in acetonitrile to give the stock solution. A 60 pmol 1-1 stock solution of (S)- 2,2’-diamino- 1,l ’-binaphthalene was prepared similarly from 17 mg of the compound. Stock solutions of each enantiomer of 1 -phenylethylamine were prepared at a concentration of 25 mmol 1-1 by dissolving 1.6 g of the appropriate enantiomer in 50 ml of acetonitrile and diluting a 10 ml aliquot of this solution ten-fold with acetonitrile.Calibration solutions containing 8 pmol 1- I 2,2’-dihydroxy- 1,l‘-binaphthalene and 0-1 3 mmol 1-1 1-phenylethylamine were prepared by placing 5 ml of 2,2’-dihydroxy- 1,1 ’-binaph- thalene stock solution in a 50 ml calibrated flask, adding 0-25 ml of l-phenylethylamine stock solution, and diluting to volume with acetonitrile.The fluorescence emission intensity of each solution was measured at an excitation wavelength of 330 nm and an emission wavelength of 360 nm, the excitation and emission maxima, respectively, of 2,2’-dihydroxy- 1,l’- binaphthalene.Calibration solutions containing 6 pmol 1- 1 2,2’-diamino- 1,l’-binaphthalene and 0-1 3 mmol 1 - 1 1 -phenylethylamine were prepared in an identical manner. The fluorescence emission intensity of each solution was measured at an excitation wavelength of 350 nm and an emission wavelength of 405 nm, the excitation and emission maxima, respectively, of 2,2’-diamino- 1,l ’-binaphthalene.Test mixtures of known enantiomeric composition were prepared by mixing appropriate amounts of pure samples of each enantiomer. Test stock solutions were prepared by dissolving 1.6 g of the mixture in 50 ml of acetonitrile. A 10 ml aliquot of this solution was diluted ten-fold in acetonitrile to give the stock solution. Two test solutions containing a known amount of each enantiomer were prepared by volume of the test stock solution to a 5 ml diamino- 1 , 1’-binaphthalene stock solution and 1.1 - (0) , I adding a known sample of 2,2’- to a 5 ml sample [ I-phenylethylamine]/mmol I-’ Fig.1 Stem-Volmer plots showing the quenching of; (a) (R)-2,2’dihy- droxy- 1 ,l’-binaphthalene; (b) (R)-2,2’-diamino- 1 - 1‘-binaphthalene fluores- cence by (R) and (S)- I -phenylethylamine.Table 1 Determination of the composition of two test samples of 1 -phenylethylamine Test 1 Test 2 Test Actual Test Actual value value value value mmoll-1 5.76 5.26 1.56 1.83 mmoll-1 2.21 2.70 9.70 9.30 mmoll-1 7.97 7.96 11.3 11.1 composition 72 66 14 17 Concentration (R)/ Concentration (S)/ Overall concentration/ Enantiomeric [% (R)-enantiomer] of 2,2’-dihydroxy- I , 1 ’-binaphthalene stock solution, each in 50 ml calibrated flasks.These test solutions were diluted to volume with acetonitrile and the fluorescent emission intensity meas- ured under the appropriate conditions for the particular binaphthalene derivative. Results and Discussion Stern-Volmer calibration plots were prepared for each fluoro- phore based on quenching by pure samples of each enantiomer of 1 -phenylethylamine [Figs.1 ( a ) and (b)]. The differences in slopes for the two enantiomers are small, but measurable, that for the 2,2’-diamino compound being greater than for the 2,2’- dihydroxy compound. The Stern-Volmer constants for each plot were calculated and used to determine the concentrations of each enantiomer of 1 -phenylethylamine present in two test solutions.From these concentrations the enantiomeric composi- tion of the sample of 1 -phenylethylamine could be calculated. The results are compared to the actual values in Table. 1 They show that the concentrations of enantiomer can be measured in both cases, although the accuracy varies from 5 to 22%. The overall concentration of quencher is determined with much greater accuracy ( < 2%).Student’s t-test was used to compare the results obtained for 100% of the (R) enantiomer of 1 -phenylethylamine and 90% (R) + 10% (S), and showed that they were significantly different, whereas the comparison of the results for 100% (R) and 95% (R) + 5% (S) were not. This indicated an LOD of 10% of one enantiomer in the presence of the other. Conclusions The investigation has shown that the principle of the procedure for determination of the concentration of 1 -phenylethylamine in solution, while at the same time giving the concentrations of each individual enantiomer, thus allowing the enantiomeric excess of the sample to be determined, is valid.However, because of the small differences between the Stern-Volmer constants for the two enantiomers, the accuracy of the procedure using the reagents selected is inadequate for application to real samples.Attempts will be made to seek fluorophores with greater differences in their Ksv values, in order to identify useful reagents. Where more than two quenchers are present in solution the Stern-Volmer relationship may be extended further to include additional terms for additional quenchers.The use of an increased number of chiral fluorophores may then allow this method to be used to measure the concentrations of additional interfering quenchers, or, alternatively, all four stereoisomers of a compound with two chiral centres. The authors thank the EPSRC and Pfizer Central Research for their support of this project through a CASE award.References 1 2 3 4 5 6 7 8 9 10 11 12 Hacksell, U., and Ahlenius, S., Trends Biotechnol., 1993, 11, 73. Witte, D. T., Ensing, K., Franke, J.-P., and De Zeeuw, R. A., Pharm. World Sci., 1993, 15, 10. Beckett, A. H., Biochem. Soc. Trans., 1991, 19, 443. Kerremans, A. L. M., Neth. J . Med., 1993, 42, 80. Stinsen, S. C., Chem. Eng. News, 1992, 70, 46. Borman, S., Chem. Eng. News, 1990, 68, 9.Evans, A. M., Eur. J . Clin. Phurm., 1992, 42, 237. Knihinicki, R. D., Williams, K. M., and Day, R. O., Biochem. Pharmucol., 1989, 8,4389. Scott. A. K., Drug Saf, 1993, 8, 149. Fox. M. A., and Singletary, N. J., Tetrahedron Lett., 1979, 74, 21 89. Corradini, R., Sartor, G., Marchelli, R., Dossena, A., and Spisni, A., J. Chem. Soc., Perkin Trans. 2 , 1992, 1979. Gafni, A,, J . Am. Chem. Soc., 1980, 102, 7367.13 Tundo. P., and Fendler, J. H., J . Am. Chem. Soc., 1980, 102, 1760. 14 Metcalf, D. H., Stewart, J. M. M., Snyder, S. W., Grisham, C . M., and Richardson, F. S., Inorg. Chem., 1992, 31, 2445. 15 Yang, H., and Bohne, C., J . Plzotochem. Photohio(. A : , 1995, 86, 209. 16 James, T. D., Sandanayake, K. R. A. S., and Shinkai, S., Nature (London), 1995,379, 345. 17 Iwanek, W., and Mattay, J., J . Photochem. Photohiol. A : , 1992, 67, 209. 18 hie, M., Yoroxu, T., and Hayashi, K., J . An?. Chen7. Soc., 1978, 100, 2236. Analytical Communications, August 1996, Vol33 267 19 20 21 22 Yorozu, T. Hayashi, K., and hie, M., J . At77. Chem. Soc., 1981, 103, 5480. Parker, K. S., Townshend, A., and Bale, S. J., Anal. Proc., 1995, 32, 183. Parker, K. S., Townshend, A.. and Bale, S. J., Anul. PI.oc., 1995, 32, 329. Wolfbeis, 0. S., and Urbano, E., And. Cheni., 1983, 55, 1904. Papei- 61042380 Accepted June 17, 1996

ISSN:1359-7337    

DOI:10.1039/AC9963300265

出版商:RSC    

年代:1996

数据来源: RSC

摘要:

Analytical Conzrnunirations, August 1996, Vol33 (269-270) 269 Elect rogenerated Chem i lum i nescence Using Platinum Electrodes With Hydrogen Peroxide Pre-t reat ment Neil J. Kearney, Carl E. Hall, Roger A. Jewsbury,* and Stuart G. Timmis Department of Chemical and Biological Sciences, University of Huddersfield, Huddersfeld, West Yorkshire, UK HDI 3DH A novel hydrogen peroxide pre-treatment of platinum electrodes is described.The method is applied to the electrogenerated chemiluminescence of luminol using air-segmented flow and an improved response and precision are reported. A rectilinear calibration for luminol is reported from 10-10 mol dm-3 to 10-5 rnol dm-3 with a LOD of 1 x 10-10 mol dm-3. An explanation in terms of surface effects on the electrode surface is given, supported by cyclic voltammetry results.Electrogenerated chemiluminescence as a basis for detection of analytes in flowing streams has the potential advantage over chemiluminescence of in situ generation of reagents, greater control over the oxidation and more efficient cell design.' In practice these advantages have rarely been successfully ex- ploited owing to the problems of competing reactions at the electrode surface, poor reproducibility and response. Control over the redox reactions can be achieved both by varying the potential of the electrodes and by their composition.This work describes a pre-treatment procedure that enhances the response and improves the reproducibility of the electro- generated chemiluminescence of luminol, 3-aminophthal- hydrazide, using platinum electrodes in a flowing stream.Electrogenerated chemiluminescence from luminol was ob- served2 shortly after the chemiluminescence of luminol was reported. The mechanism of the electrically generated chemi- luminescence of luminol, whilst not categorically established, is thought to be similar to that of chemiluminescence.3 The luminol anion (LH-) undergoes one electron electrochemical oxidation to a diazaquinone, which is then further oxidized by hydrogen peroxide or superoxide to give an excited 3-aminophthalate species which emits light (Scheme 1).A comparative study using a rotating disc ring electrode showed that for luminol in air-saturated alkaline solution, Au, Pt and PbO2 electrodes gave a greater response than glassy carbon.4 A glassy carbon electrode has been used for electro- oxidation of luminol in the presence of hydrogen peroxide at pH 7.4 at potentials of between 0.5 V versus SCE to oxidize luminol and below 1 V to avoid oxidation of peroxide.5 At a clean Pt electrode the oxidation potential of hydrogen peroxide was HO; L - LOOH h v Scheme 1 Simplified mechanism for the luminescent oxidation of LH- * To whom correspondence should be addressed.similar but in use an oxide layer formed and the oxidation potential was 0.44 V. It has also been suggested that hydrogen peroxide can be formed at the cathode by reduction of oxygen6 and that this can react with the oxidized luminol radical produced at the anode. In flow systems this method has been exploited by using a glassy carbon electrode to generate peroxide which then reacts with the oxidized luminol at a gold electrode giving a response over the range 10-10 to 10-5 rnol dm-3 lurnin01.~ Using Pt electrodes in a static system a response over the range 2 X 10 -* to 1 X mol dm-3 has been reported.x It is generally believed that larger responses can be obtained in cells with laminar flows and small electrode gaps.A miniature cell with a 50 pm gap between cleaned Pt electrodes has given a linear response over the range 10-8 mol dm--? to 10-5 mol dm-3 with an LOD in the region of 1 X 10-8 rnol dm-3.9 In general Pt electrodes have not given consistent results due to ill-defined electro-oxidation reactions, therefore pre-treat- ment of the electrodes has been investigated.The benefits are demonstrated using a cell with some turbulent flow but with air- segmentation to reduce dispersion. Experimental Flow experiments were carried out using systems constructed in house. A dual electrode glass flow-through cell of volume 0.5 cm3 was constructed with two Pt wires, one as a coil and the other a central straight wire. The electrodes were connected to a constant voltage power supply (Dl00, Farnell, Leeds, UK).Tygon tubing (1.02 mm id) was used for the flowing streams which were pumped using a four channel peristaltic pump (H0733 1, Ismatec, Carshalton, Surrey, UK). Injection was by a six-way mechanical injection valve and the detector comprised the electrode flow-through cell adjacent to a photomultiplier tube (98 13B, Thorn EMI, Ruislip, Middlesex, UK) operating at 1 100 V with output directly to a chart recorder (BD40, Kipp and Zonen, Delft, Netherlands).For the luminol experiments, the injection volume was 85 x 10-f~ dm-i and the flow rate 4.5 cm-' min-1. Cyclic voltammetry was carried out using 0.25 cmz Pt working electrode, a 2 cm2 Pt flag counter electrode and a SCE reference using 1 X rnol dm--? luminol solution in pH 10 carbonate buffer or chloride (0.2 rnol dm-3) at pH 12.5.The voltage sweeps were generated using a scanning potentiostat (Sycopel Scientific, Boldon, Tyne and Wear, UK) and recorded on an -x-y chart recorder. All the scans shown were recorded at 100 mV s-l in an anodic direction. Chemicals were used without further purification. Solutions of luminol(98%, Avocado, Heysham, UK), hydrogen peroxide (30% m/m stabilized, Sigma, Poole, UK) and sodium carbonate and bicarbonate (AnalaR, BDH, Poole, UK) were freshly prepared daily using dei-onized water.270 Analytical Communications, August 1996, Vol33 Results and Discussion The analytical system consisted of injecting luminol at pH 12.5 into a chloride carrier stream at the same pH, with a potential applied between the electrodes.The change in current output from the photomultiplier was recorded as a voltage peak on the chart recorder. Increasing the potential between the electrodes indicated that luminescence was observed at 0.6 & 0.1 V and a larger signal at 2.0 ? 0.1 V. A potential of 2.0 V was thus selected. Even with cleaning with several mineral acids or iron(II1) chloride reproducibility was poor and the method was not useful.As one reported source of variability in Pt electrodes is oxide formation on the surfacel0 hydrogen peroxide was tried as a pre-treatment. This led to an enhancement of the response by 1000 >E 100 g .- 10 a, L Y ' a" 0.1 0.01 io-I0 1 0 - ~ 10" 1 0 - ~ l o 6 1 0 - ~ Luminol concentration/mol dm-3 Fig.1 tion. Conditions as in text. Change in chemiluminescence emission with luminol concentra- 11 00 pA cm-' 0.0 0.7 1.7 EN Fig. 2 buffer at untreated Pt electrodes. Conditions as in text. Cyclic voltammogram (repeat scans) of luminol in pH 10 carbonate 0.0 0.7 1.7 EN Fig. 3 untreated (I) and treated (11) Pt electrodes. Conditions as in text. Cyclic voltammogram of luminol in pH 10 carbonate buffer at two orders of magnitude and considerably improved reproduci- bility.A pre-treatment of 60 s was sufficient and no additional benefit was observed on longer pre-treatment of up to 180 s. To remove free hydrogen peroxide, the system was briefly rinsed with carrier. Longer washing caused the peroxide enhancement to be lost. The procedure used was thus: ( i ) wash with nitric acid (2 mol dm-3, 180 s); (ii) pass hydrogen peroxide solution (0.18 mol dm-3, 60 s); (iii) wash with carrier (10 s); (iv) inject luminol solution in carrier; and (v) wash with carrier. The change in response with luminol concentration (Fig.1) showed linearity from 10-10 mol dm-3 to 10-5 mol dm-3 with an LOD (3s) of 1 x 10-10 mol dm-3. The repeatability of the response for 1 x mol dm-3 luminol showed an RSD of 3%.In contrast to the chemiluminescent oxidation which requires catalysts such as metal ions or enzymes, metal ions had little effect on the signal, although the repeatability deteriorated. Luminol chemiluminescence requires the presence of oxygen or hydrogen peroxide in addition to the luminol radical which can be produced electrochemically.It is unlikely that sufficient hydrogen peroxide remains from the pre-treatment or that peroxide is likely to be stable under the conditions at the electrode surfaces. It has been suggested3 that the superoxide will be formed by the reaction of the luminol radical and oxygen and that reaction of the superoxide with the luminol radical will then form the hydroperoxide which proceeds to the light emitting step.A major problem with electro-oxidation reactions of organic compounds is deposition on the electrode surfaces, particularly of polymeric material. This can be seen as a decay in the measured electrochemical signal if a 1 X 10-3 mol dm-3 luminol solution is scanned continuously (Fig. 2). It is believed that the improved performance of the Pt electrodes after pre- treatment is a consequence of reduced deposition of luminol derived material.The cyclic voltammogram of luminol in pH 10 carbonate buffer (Fig. 3) shows the effect of peroxide treatment of Pt electrodes. This appears to be the suppression of the electro-oxidation reaction leading to polymeric species. A similar suppression was observed in pH 10 chloride buffer (Fig.4). In this instance both the oxidation of luminol and chloride have been affected. Similar results were observed for other species liable to radical formation on oxidation such as pyrrole. This technique of pre-treatment may have general applica- bility in electrogenerated chemiluminescence and in electro- chemical sensors and is being further investigated. References 1 2 3 4 5 6 7 8 9 10 11 Knight, A.W., and Greenway, G. M., Analyst, 1994, 119, 879. Harvey, N., J . Phys. Chem., 1929, 33, 1456. Merenyi, G., Lind, J., and Eriksen, T. E., J. Biolumin. Chemilumin., 1990, 5, 53. Vitt, J. E., Johnson, D. C., and Engstrom, R. C., J . Electrochem. SOC., 1991,138, 1637. Sakura, S., Anal. Chim. Acta, 1992, 262, 49. Haapakka, K. E., and Kankare, J., Anal. Chim. Acta, 1982, 138, 263. Nieman, T. A., J . Res. Natl. Bur. Stand. ( U S . ) , 1988, 93, 501. An, J., Chen, X., and Chen, H., Fenxi Huaxue, 1988,16, 127. Chen, X., and Sato, M., J . Flow Injection Anal., 1995, 12, 54. Hoare, J. P., The Electrochemistry of Oxygen, Wiley, New York, 1968. Vitt, J. E., Johnson, D. C . , and Engstrom, R. C . , J . Electrochem. Soc., 1991, 138, 1637. Fig. 4 untreated (I) and treated (11) Pt electrodes. Conditions as in text. Cyclic voltammogram of luminol in chloride at pH 12.5 at Paper 6J03887E Received June 4,1996 Accepted July 2,1996

ISSN:1359-7337    

DOI:10.1039/AC9963300269

出版商:RSC    

年代:1996

数据来源: RSC

摘要:

Analytical Communications, August 1996, Vol33 (271-273) 27 1 Sol-Gel-based Am peromet ric GI ucose Biosensor Incorporating an Osmium Redox Polymer as Mediator Tae-Myung ParkaJ’, Emmanuel I. Iwuohaa, Malcolm R. Smythal* and Brian D. MacCraithc a Biomedical and Environmental Sensors Technology (BEST) Centre, School of Chemical Sciences, Dublin City University, Dublin 9, Ireland b Department of Environmental Engineering, Suncheon Technical Junior College, Suncheon 540-744, Korea c BEST Centre, School of Physical Sciences, Dublin City University, Dublin 9, Ireland A novel amperometric biosensor for the determination of glucose was constructed by first immobilizing glucose oxidase and an osmium redox polymer, [0~(bpy)2(PVP)~~Cl]Cl, on the surface of a glassy-carbon electrode, followed by coating with a sol-gel film derived from methyltriethoxysilane. The redox potential of this electrode was found to be +247 mV versus Ag/AgCl, and glucose could therefore be determined amperometrically at +400 mV versus Ag/AgCl.The concentration range of linear response, slope of linear response and LOD were 2.0-31 mmol 1-1, 44.5 nA mmol-1 1 and 0.5 mmol 1-l, respectively. Although L-ascorbate was electrooxidizable at this potential, uric acid and paracetamol were both found not to interfere.mediator, but no information was given in their paper about the determination of glucose. Glezer and Lev,26 however, prepared Pt electrodes with GOD entrapped within a vanadium penta- oxide gel and demonstrated the possibility of detection of glucose by cyclic voltammetry. Pankratov and Lev2’ have also demonstrated the use of a silica-carbon-based glucose sensor containing tetrathiafulvalene as the mediator.This communication describes the development of a sol-gel- based amperometric biosensor based on GOD and an osmium redox polymer as mediator. We have investigated the electro- chemical behaviour of this electrode and shown that glucose can be determined amperometrically with good analytical charac- teristics.An important area of biosensor research is concerned with the immobilization of enzymes at transducer surfaces. Conven- tional methods of enzyme immobilization include covalent binding, physical adsorption, encapsulation in polymers, or cross-linking to a suitable supporting matrix. Generally, supporting matrices are constructed of conducting and non- conducting polymers,1-3 conducting organic salts,4 and carbon- based composite materials,s-s or layers of enzymes adsorbed onto electrode surface^.^ Recent research has demonstrated that silicate glasses obtained by the sol-gel method can provide such a supporting matrix, and that biomolecules can be immobilized by this method.Isl7 These bioceramic materials have been further applied to produce silica-based photometric l8-21 and FI detectors.22 More recently, low temperature processing condi- tions, chemical inertness, negligible swelling effects, tunable porosity, improved thermal stability and the high purity of sol- gel-derived glasses have opened up many more possibilities for novel sensor appli~ations.~3 Although most applications to date have been based on spectroscopic techniques, a few papers have emerged recently on the use of electroanalytical procedures.For instance, Narang et al.24 have described a series of prototype tetraethylorthosilicate-derived sol-gel thin films for immobili- zation of glucose oxidase (GOD). Their multiple-layer, ‘sand- wich’-type construction for the determination of glucose relied on the direct detection of hydrogen peroxide.But the lack of electronic conduction in silica gel means that electrochemical methods generally rely on incorporation of an electron transfer mediator. For instance, Audebert et ~ 1 . ~ 5 have demonstrated the activity of GOD doped within a sol-gel derived matrix (tetramethylorthosilicate) using hydroxymethylferrocene as * To whom correspondence should be addressed.Experimental Materials The source of all chemicals used for experiments are given in parentheses; methyltriethoxysilane (MTEOS, Aldrich, Mil- waukee, WI, USA); glucose oxidase (EC l. 1.3.4 type VII-S, 183 000 U g-l from Aspergillus, Sigma, St. Louis, MO, USA); uric acid (Aldrich); paracetamol (Merck, Poole, Dorset, UK); ascorbic acid (Aldrich); Na2HP04 and NaH2P04 (Merck). Aqueous solutions were prepared in de-ionized water (Easypure RF, Branstead water purification system).The osmium poly- mer, [Os(bpy)2(PVP)l&l]Cl, used as an electron transfer mediator, was prepared as described elsewhere.28 Instrumentation Cyclic voltammetry experiments were carried out using a BAS (West Lafayette, USA) CV- 1 OOW voltammetric analyser interfaced to a PC-IBM compatible computer.Amperometric measurements were performed with an EG&G Princeton Applied Research (Princeton, NJ, USA) Model 400 electro- chemical detector connected to a WPA CQ95 x-t recorder. Preparation of Sol-Gel-den’ved Electrode To prepare the modified electrode, 3 pl of 1% 0 s polymer (in methanol) was coated on the surface of a glassy-carbon electrode and dried at room temperature to evaporate off the methanol, and then a 3 pl aliquot of GOD solution (1000 U ml-l; prepared in phosphate buffer, pH 6.4) was coated onto the surface of the modified electrode and allowed to air dry in a covered beaker.After 24 h, a sol-gel layer was spin cast on to the surface. The silica sol was prepared by mixing 1.6 ml of de- ionized water and 0.03 ml of 0.05 moll-’ HC1 (which was used272 Analytical Comniurzications.August 1996, Vol33 _. to catalyse the reaction), followed by the addition of 4.5 ml of MTEOS. The resulting molar ratio of MTEOS to water was 1 :4. Procedures All measurements were carried out in a three-electrode system using a platinum wire and Ag/AgCl (saturated KCI) as the auxiliary and reference electrode, respectively.All steady state measurements were performed at +400 mV versus Ag/AgCl in 0.1 mol 1-1 phosphate buffer (pH 6.4). The argon gas was passed through the cell to prohibit the competition of oxygen with the osmium redox centre.29 The solution was stirred at 500 rpm using a magnetic stirrer. Results and Discussion Electrochemical Behaviour of Sol-Gel-derived Electrode Cyclic voltammograms of the 0s-polymer-GOD-sol-gel elec- trode in 0.1 mol I--' phosphate buffer (pH 6.4) solution at different scan rates is shown in Fig.1. In the absence of glucose, the enzyme gives no response and only the 0s-polymer electrochemistry is observed. The relationship between the peak anodic current, il, and, the square root of the scan rate, was then investigated.The function (i,Jvl'2) had a constant value at different scan rates. This means that the catalytic electrooxida- tion current of the sol-gel derived electrode is diffusion controlled current, i.e., diffusion of electrons along the polymer chain. On addition of glucose, a catalytic oxidation wave was observed (Fig.2). This means that GOD is reduced by glucose diffusing into the sol-gel film, electrons are transferred from the GOD-FADH2 to the Os3+ sites, and the electrons are then transferred to the electrode surface. The elimination of a reduction peak means that the reduced state of the sol-gel film is maintained. The pH dependence of the electrode attained at a concentration of 10 mmol 1-1 showed a constant response between pH 6.4 to 7.2 using 0.1 mol 1- 1 phosphate buffer.We selected a pH of 6.4 for all further experiments in consideration of analytical sensitivity. Fig. 3 shows the steady-state current- time response of the sol-gel-derived glucose biosensor. The electrode response is linear within the concentration range 2.0-3 1 mmol 1- I with I' = 0.9877. The LOD and slope of linear response were 0.5 mmol 1-l and 44.5 nA mmol 1-' with 3.6% RSD, respectively.The LOD was calculated based on twice the +2.2001 +I ,710 +1.220 +0.730 +0.240 -0.250 5 -0.740 3. 2 -1.230 -1.720 -2.21 0 +0.55 +0.45 +0.35 +0.25 +0.15 +0.05 PotentialN versus Ag/AgCI Fig. 1 Cyclic voltammograms of the sol-gel derived electrode in absence of glucose. Conditions: 0.1 mol 1-1 phosphate buffer, pH 6.4; scan rates (mV \ - I ) : a, 5; b.10: c, 20; d, 50; e. 100; f, 200. value of the background current. The electrode response time was about 60 s until steady-state values were obtained. The slow response time is due to diffusion of glucose into the sol-gel film. When the sol-gel derived electrode was stored dry in the laboratory at room temperature, the response of the electrode showed no change over the short term (about 1 week).Interferences Interference effects were investigated by testing the responses of the sol-gel derived electrode to L-ascorbic acid, paracetamol (4-acetamidophenol) and uric acid. The three compounds are the substances that interfere with the detection of glucose in blood samples. The response of the sol-gel-based glucose electrode to 0.1 and 2 mmol 1-1 concentrations of each of the interferents, in the absence of glucose, are given in Table 1.The results show strong interference effect from L-ascorbic acid, which increase as the L-ascorbic acid concentration increases. Paracetamol does not show interference effects on the electrode at concentrations lower than 2 mmol 1-1. The biosensor completely blocks the interference effect of uric acid within the concentration range investigated.+0.300 +0.010 - -0.280 - -0.570- -0.860- 3 s -1.150- L L 5 -1.440- -1.730 - -2.600 I +0.55 +0.45 +0.35 +0.25 +0.15 +0.05 PotentialN versus Ag/AgCI Fig. 2 Cyclic voltammogram of the electrode shown in Fig. 1 under Ar after addition of 31 mmol 1-1 glucose. Conditions: 0.1 mol 1-1 phosphate buffer, pH 6.4; scan rate 5 mV s-1; stirring speed 500 rpm.Fig. 3 Amperometric responses of sol-gel derived glucose biosensor on successive addition of glucose. Conditions: 0.1 mol 1- phosphate buffer, pH 6.4; EdPP1 = +400 mV I'ersus Ag/AgCl; stirring speed 500 rpm.Analytical Communications, August 1996, Vol33 273 Table 1 Responses of the sol-gel-based amperometric sensor to inter- ferents.(The response of each interferent is measured in the absence of glucose). Concentration/ Interferent mmoll-1 ResponsehA L-Ascorbic acid 0.1 21 1 .0 127 2.0 24 1 Paracetamol 0.1 1 .o 2.0 Uric acid 0.1 1 .o 2.0 Conclusions We have developed a novel sol-gel-based amperometric glucose biosensor using an osmium redox polymer as mediator. Our initial results demonstrate that a sol-gel film coated onto a modified GOD-Os-polymer-glassy-carbon electrode can be used as an amperometric glucose biosensor that provides a linear response to glucose within the concentration range of 2.0-31.0 mmol 1-l, which was stable for at least 1 week.In previous studies we have used the osmium polymer in biosensor matrices employing carbon paste, glutaraldehyde, polyethylene glycol and conducting polymers.30-33 These other matrices are characterized by much faster response times, but smaller linear ranges than the sol-gel system described in this communication.As such, the use of sol-gel immobilization may be useful for situations in which there are high glucose concentrations. It may also prove to be a more robust method of immobilization, with possible applications in bioprocess monitoring.References I Foulds, N. C., and Lowe, C. R., Anal. Chem., 1988,60,2473. 2 Bartlett, P. N., and Whitaker, R. G., J . Electroanal. Chem., 1987,224, 37. 3 Malitesta, C., Palmisano, F., Torsi, L., and Zambonin, P. G., Anal. Chem., 1990,62,24. 4 Kawagoe, J. L., Niehaus, D. E., and Wightman, R. M., Anal. Chem., 1991,63,2961.5 Frew, J. E., Harmer, M. A., Hill, H. A. O., and Libor, S. I., J . Electroanal. Chem., 1986, 201, 1. 6 Chen, L., Lin, M. S., Hara, M., and Rechnitz, G. A., Anal. Lett., 1991, 24, 1. 7 8 9 10 11 12 I3 14 15 16 17 I8 19 20 21 22 23 24 Hale, P. D., Inagaki, T., Lee, H. S., Karan, H. I., Okamoto, Y., and Skotheim, T. A., Anal. Chim. Acta., 1990, 228, 31. Wring, S. A., and Hart, J.P., Analyst, 1992, 117, 1215. Schuhmann, W., Ohara, T. J., Schmidt, H. L., and Heller, A., J. Am. Chem. Soc., 1991,63, 677. Avnir, D., Levy, D., and Reisfeld, R., J . Phys. Chem., 1984, 88, 5956. Braun, S., Rappoport, S., Zusman, R., Avnir, D., and Ottolenghi, M., Mater. Lett., 1990, 10, 1. Kuselman, I., Kuyavskaya, B. I., and Lev, O., Anal. Chim. Acta., 1992, 256, 65.Schwok, A., Avnir, D., and Ottolenghi, M., J . Am. Chem. SOC., 199 1, 113, 3984. Gun, G., Tsionsky, M., and Lev, O., Muter. Res. Soc. Symp. Proc., 1994,346, 1011. Gun, J., Tsionsky, M., Golan, Y., Rubinson, I., and Lev, O., J . Electroanal. Chem., 1995, 395, 57. Gun, G., Tsionsky, M., and Lev, O., Anal. Chim. Acta, 1994, 294, 261. Barreau, S., and Miller, J. N., Anal. Commun., 1996, 33, 5H.Ellerby, L. M., Nishida, C. R., Nishida, F., Yamanaka, S. A., Dunn, B., Valentine, J. S., and Zink, J. I., Science, 1992, 255, 1 1 13. Dunuwila, D. D., Torgerson, B. A., Chang, C. K., and Berglund, K. A., Anal. Chem., 1994,66, 2739. Schwok, A., Ottolenghi, M., and Avnir, D., Nature, 1992, 355, 240. Lee, J. E., and Saavedra, S. S., Anal. Chim. Acta., 1994, 285, 265. Tatsu, Y., Yamashita, K., Yamaguchi, M., Yamamura, S., Yama- moto, H., and Yoshikawa, S., Chem.Lett., 1992, 1619. MacCraith, B. D., McDonagh, C., O’Keeffe, G., McEvoy, A. K., Butler, T., and Sheridan, F. R., Sens. Actuators, B , 1995, 29, 5 1. Narang, U., Prasad, P. N., Bright, F. V., Ramanathan, K., Kumar, N. D., Malhotra, B. D., Kamalasanan, M. N., and Chandra, S., Anal. Chem., 1994,66, 3139. Audebert, P., Demaille, C., and Sanchez, C., Chem. Mater., 1993, 5. 911. Glezer, V., and Lev, O., J . Am. Chem. Soc., 1993, 115, 2533. Pankratov, I., and Lev, O., J . Electroanal. Chem., 1995, 393, 35. Forster, R. J., and Vos, J. G., Macromolecules, 1990, 59,4372. Gregg, B. A., and Heller, A., Anal. Chem., 1990, 62, 258. Iwuoha, E. I., Smyth, M. R., and Vos, J. G., Electroanalysis, 1994,6, 982. Pravda, M., Adeyoju, O., Iwuoha, E. I., Vos, J. G., Smyth, M. R., and Vytras, K., Electroanalysis, 1995, 7, 619. Pravda, M., Jungar, C. M., Iwuoha, E. I., Smyth, M. R., Vytras, K., and Ivaska, A., Anal. Chim. Acta, 1995, 304, 127. Adeyoju, O., Iwuoha, E. I., Smyth, M. R., and Leech, D., Analyst, 1996, 121, in the press. 25 26 27 28 29 30 31 32 33 Paper 6103149H Received May 7,1996 Accepted June 28, 1996

ISSN:1359-7337    

DOI:10.1039/AC9963300271

出版商:RSC    

年代:1996

数据来源: RSC

摘要:

Analytical Communications, August 1996, Vol33 (275-277) 275 Adsorptive Strip p i ng Vo I t a m met r ic Determination of Tryptophan at an Electrochemically Pre-treated Carbon-paste Electrode With Solid Paraffin as a Binder Huaisheng Wang, Hui Cui, Aimei Zhang and Renmin Liu Department of Chemistry, Liaocheng Teacher's College, Liaocheng, Shandong, Chinu An adsorptive stripping voltammetric method for the determination of tryptophan at an electrochemically pre-treated carbon paste electrode has been developed.Solid paraffin was used as the binder of the carbon paste. The electrode was pre-treated in a solution of HAc-NaAc (pH 3.5) by holding it at +1.8 V versus an Ag/AgCl electrode for 10 min. On the pre-treated electrode the adsorption of tryptophan was greatly enhanced. Tryptophan was accumulated in HAc-NaAc solution (pH 3.5) at a potential of +0.3 V versus an Ag/AgCl electrode for a certain time, and then determined by differential anodic stripping voltammetry.An oxidative peak was observed at +1.01 V. The relationship between peak current and tryptophan concentration was linear in the range 8-100 ng ml-l of tryptophan, and the detection limit was 2 ng ml-l.The method has been applied to the determination of tryptophan in amino-acid compound injection and synthetic serum samples. The relative standard deviation (6 determinations) was less than 2%. Tryptophan is one of the essential amino acids for human bodies. The amino acid and its metabolites are of biochemical importance, so the determination of tryptophan in the human body and foods is of great interest in life science research.Methods for the determination of tryptophan are mainly based on HPLC14 and spectrophotometry.s-8 Few methods are based on electroanalytical Wang et al. I 1 developed a differential pulse voltammetric method for the determination of tryptophan by using a carbon paste electrode modified with 10% (m/m) montmorillonite.This method needs a medium exchange procedure and the sensitivity is not high (linear range 2.0-10.0 kg ml-1, detection limit 70 ng ml-1). Wu et ~ 1 . 1 2 developed another differential pulse voltammetric method for the determination of tryptophan by using a potato-juice modified carbon paste electrode. It can detect tryptophan in the presence of tyrosine. The linear range of the method is 9.8 X 10-6-9.8 X 10-5 mol 1-1 of tryptophan with a detection limit of 1.5 X 10-6 mol I-'. An indirect adsorptive stripping voltammetric method for tryptophan was developed by Zoulis, which is based on the enzymic cleavage of the amino acid to indole and the pre-concentration of indole at the carbon paste electrode.lO The linear range of this method is 20-80 pmol l-1.This method also needs a medium exchange procedure, and a 140 min incubation time is needed. A solid carbon paste electrode employing solid paraffin as a binder has many advantages over the common carbon paste electrode.I3 Electrochemical pre-treatment of glassy carbon electrodes has been shown to be advantageous. Some discussion of the literature on the electrochemical pre-treatment of glassy carbon electrodes was presented previou~ly.~~17 In the present work, a solid carbon paste electrode has been made and the effect of electrochemical pre-treatment studied.Electrochem- ical pre-treatment of the electrode surface before it is used to determine tryptophan has subsequently been shown to be advantageous. An adsorptive stripping voltammetric method for the direct determination of tryptophan has been developed. The proposed method has been applied to the determination of tryptophan in serum and amino acid compound injection samples.Good results have been obtained. Experimental Apparatus An MP- 1 Stripping Analyser (The Seventh Communication Plant of Shandong) and a 79-1 Voltammetric Analyser equipped with an LZ3-104 X-Y recorder were used for the determination.The three-electrode system used in this study contained a laboratory-made carbon paste working electrode, a platinum wire counter electrode, and an Ag/AgCl reference electrode. A PHS-3C pH meter (Shanghai Leici Instrument Plant) was used for the measurement of the pH of the supporting electrolyte s o h tion. Reagents The amino acids used were of biochemical-reagent grade.Other reagents were of analytical-reagent grade. All solutions were prepared with sub-boiling doubly distilled water. L-Tryptophan stock solution (2.0 mg ml-I). This solution was prepared by dissolving 0.2000 g of L-tryptophan in 10 ml of 0.15 moll-' HC1 and diluting to 100 ml with water. It remains stable for at least 2 weeks if it is kept in a refrigerator. The working solutions used for calibration and standard addition were prepared daily by diluting the stock solution with water.HAc-NaAc solution (pH 3.5). A 1.64 g amount of sodium acetate and 18.4 ml of acetic acid were dissolved in 100 ml of water. The pH of the solution was adjusted to 3.5 with 3 moll-' HCl and 3 moll-' sodium hydroxide solution, and the solution was then diluted to 200 ml with water.Graphite powder. Graphite powder (1 80 mesh) of spectro- scopic grade (Shanghai Graphite Plant) was used for the preparation of the electrode. ParafSin. This had a melting-point of 56-62 "C. Preparation of the Electrode A 1 .OO g amount of graphite powder and 0.18 g of paraffin were put into a 10 ml beaker and heated until the paraffin had melted completely. After thorough mixing with a glass bar, the mixture was filled into a glass tube (50 mrn long, 6 mm od, 4 mm id) and pressed tightly with a steel bar (3.5 mm od).A copper bar (1.7 mm od) was inserted into one end of the mixture to perform the electrical connection while the mixture was hot. After the276 Analytical Communications, August 1996, Vol33 mixture had become cool, the surface of the other end was polished with emery cloth until it had a shiny appearance. Chemical Pre-treatment of the Electrode The three electrodes were immersed in 20 ml of solution containing 0.16 mol 1-1 acetic acid and 0.01 mol 1-1 sodium acetate.The potential of the electrode was maintained at +I .8 V versus an Ag/AgCl reference electrode for 10 min with stirring.Then, the electrode was scanned between +0.3 V and +1.4 V until a steady-state current-voltage profile was obtained. Measurement Procedure The three electrodes were immersed in a 20 ml solution containing 0.16 moll-’ acetic acid, 0.01 moll-’ sodium acetate and L-tryptophan, and an initial potential (accumulation poten- tial), +0.3 V versus an Ag/AgCl reference electrode (saturated KCl), was applied to the working electrode.The stirrer was activated for a fixed period. After a 30 s resting period, scanning was started and continued up to the final potential +1.4 V with the scan rate of 490 mV s-1. The second order differential anodic stripping voltammogram was recorded. The oxidative peak of L-tryptophan appeared at +1.01 V, and the peak current was measured by measuring the distance of the two peaks.After each scan, the electrode was maintained at the final potential for 20 s, with stirring, for oxidative cleaning of the remaining accumulated species. The electrode was then suitable for the next determination. Determination of L-tryptophan in Samples The double standard additions method was used throughout for the determinations.The three electrodes were immersed in 20 ml of supporting electrolyte. The second differential anodic stripping voltammogram was recorded and the current peak of the blank solution was measured as described above. Then, 50 1-11 of sample were added to the electrolytic cell, and the peak current was measured similarly. Two 50 yl portions of L- .tryptophan standard solutions were added to the cell and the peak currents were again measured.All currents were corrected for the blank. The concentration of L-tryptophan in each sample was calculated by extrapolating to zero current the plot of the corresponding three corrected currents versus the concentration of added L-tryptophan. Results and Discussion L-Tryptophan can be oxidized at a solid carbon paste electrode.The electrochemical pre-treated electrode gives about a 200-fold increase in signal over the untreated electrode. High sensitivity can be obtained when a second order differential anodic stripping voltammetric method is used. Fig. 1 shows the second order differential anodic stripping voltammograms of standard solutions of L-tryptophan on the electrochemical pre- treated electrode.The peak at 1.01 V is due to the oxidation of L-tryptophan, and the peak current is proportional to the concentration of L-tryptophan. The peak at 0.4 V is due to the charging current. Relationship between peak current (e”), peak potential (E,) and scan rate A linear relationship was observed between the logarithm of peak current (log e”) and the logarithm of scan rate (log Y).The linear regression equation in 0.04 yg ml-1 L-tryptophan solution is: log eN = -l.lO+log Y r = 0.996 This indicates that the peak current is controlled by adsorption and dispersion. The peak potential E, has a linear relation with scan rate. The linear regression equation is r = 1.00 E , = 0.74 + 0.1 log Y Effect of pH on Ep Under a given scan rate, the pH of the solution has a significant effect on the oxidative peak potential.This indicates that H+ ions take part in the oxidation of tryptophan. In the pH range 2.0-9.2, E, becomes more negative with the increase of the pH. A linear relation between E, and pH was obtained. The linear regression equation is: E, = 1.14 - 0.036 pH Y = 0.986 Optimization of Measurement Conditions Supporting electrolyte and pH Solutions of HCI-KCl, HAc-NaAc and NaH2PO4-Na2HPO4 were used as the supporting electrolyte.The results indicate that when the HAc-NaAc solution was used, the baseline was steady and high sensitivity could be obtained. The effect of the pH of HAc-NaAc solution was also investigated. It was found that when the HAc-NaAc solution with a pH of 3.5 was used as the supporting electrolyte, the maximum current could be ob- tained.EfSect of amount of graphite powder The electrodes with the graphite content of 75,80 and 85% were tested for the determination. The results indicate that the peak current increases with the increase in the amount of graphite. The glassy carbon electrodes were also used for the determina- tion, but no stripping peak was found.All of these results indicate that it is the graphite powder that has an adsorptive effect on L-tryptophan. When the amount of graphite powder is higher than 85%, the amount of paraffin is too small to hold it together and the electrode cannot work. So the electrodes with the graphite powder of 85% were used for all measurements. Effect of accumulation potential At the given scan rate, no significant difference in peak current was found with the accumulation potential of 0-+0.5 V.However, the peak current became small when the accumula- ~ ~ 0.3 0.5 0.7 0.9 1.1 1.3 E N Fig. 1 Second order differential anodic stripping voltammograms of L- tryptophan on the pre-treated electrode. Supporting electrolyte HAc-NaAc solution (pH 3.5); L-tryptophan concentration (ng ml-1): l,O.O; 2, 10.0; 3, 20.0; 4, 40.0; 5 , 50.0.Scan rate, 490 mV s-1; accumulation time, 60 s.Analytical Communications, August 1996, Vol33 277 tion potential was more negative. This may have been due to the reduction of the activated groups on the surface of the electrode and the adsorption effect of the electrode being decreased. The accumulation potential was +0.3 V for all determinations.Effect of accumulation time When the concentration of L-tryptophan is 80 ng ml-1, e" has a linear relationship with the square root of the accumulation time if the accumulation time is less than 50 s. The linear regression equation is eN = 1.18 + 10.34 t1/2 Y = 0.997 (e" in pA s - ~ ) If the accumulation time is longer than 50 s, the graph becomes curved.This may be due to saturated adsorption. The bigger the concentration of L-tryptophan, the shorter the time needed to achieve saturated adsorption. Calibration Under the optimum analytical conditions, the second order differential stripping peak current has a linear relationship with the concentration of L-tryptophan in the range of 8-100 ng ml-1. When the accumulation time is 60 s and the scan rate is 490 mV s-1, the linear regression equation is: eN = 0.71 +0.95 C (n = 5, r = 0.995) (e" in pA s - ~ ; C in ng rn1-l) The precision and detection limit of the method were also investigated. The precision of the method was calculated on the basis of the results obtained from ten replicate analyses of 80 ng ml-1 of L-tryptophan.The relative standard deviation was 1.3%.Under the optimum analytical conditions a detection limit of 2 ng ml-l for L-tryptophan can be obtained with the accumulation time of 2 min. If the electrode was washed by keeping the electrode potential at +1.40 V for 20 s after the determination of 80 ng ml- L-tryptophan (accumulation time, 60 s), no peak current was found in the blank solution. This indicates that the determination of high concentration solutions T 0.3 0.5 0.7 0.9 1.1 1.3 E N Fig.2 Second order differential anodic stripping voltammograms of samples. 1, Synthetic serum; 2. amino acid compounds injections. Other conditions are as in Fig. 1. has no influence on the determination of low concentration solutions. Interference Nineteen other amino acids were tested by use of the proposed method.No interference was found except with tyrosine. Tyrosine has an overlapping peak with tryptophan, but the sensitivity for tyrosine is very low. More than a two-fold excess of tyrosine interferes with the determination of tryptophan. Metal ions also have no influence. Therefore, tryptophan in samples can be determined directly. Applications The proposed method has been applied to the determination of tryptophan in amino acid compound injections and synthetic serum samples, Serum samples containing from 6.8 to 24.4 pg ml-1 tryptophan were measured directly according to the procedure given in the Experimental section.Amino acid compound injection samples were diluted and then analysed. The voltammograms of the samples were shown in Fig. 2.The recovery of the method was also studied by the addition of a standard solution of tryptophan to the samples. The recoveries from spiked samples were 97-102% for amino acid injections and 96-103% for synthetic serum samples when 20.0 pg ml-1 of tryptophan were added. The precision of the method was also investigated by analysing the samples 7 times. The relative standard deviations were 1.1-1.7%. The relative errors of the proposed method were less than +6% compared with the results obtained by a HPLC method.18 References 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 Bednar, I., Sodersten, P., and Qureshi, G.A., J . Liq. Chromatogr., 1992, 15, 3087. Paradoyannis, I., and Samanidou, V., J . Liq. Chromatogr., 1991, 14, 1409. Kojima, E., and Kai, M., J .Chromatogr. B, Biomed. Appl., 1993,123, 187. Molnar-Perl, I., and Pinter-Szakace, M., J . Chromatogr., 1993, 632, 57. Kojima, E., and Kai, M., Anal. Sci., 1993, 9, 25. Fletouris, D. J., and Botsoglou, N. A., J . AOAC Znt., 1993, 76, 1168. Iizuka, H., and Yajima, T., Biol. Pharm. Bull., 1993, 16, 103. Bertini, J., Mannucci, C., Noferini, R., Perico, A., and Rovero, P., J . Pharm. Sci., 1993, 82, 179. Tucker, D. J., Bond, A. M., and Zhang, Q., J . Electroanal. Chem. Interfacial Electrochem., 1989, 26, 127. Zoulis, N. E., Nikolelis, D. P., and Efstathiou, C. E., Analyst, 1990, 115, 291. Wang, G., Peng, T., Shen, B., Zhu, P., and Qu, L., Fenxi Huaxue, 1993, 21, 779. Wu, J., Shi, Q., Wang, G., and Peng, T., Fenxi Hauxue, 1994, 22, 599. Mao, Q., Wu, S., and Zhang, H., Fenxi Huaxue, 1995,23, 648. Fogg, A. G., Fernhdez-Arciniega, M. A., and Alonso, R. M., Analyst, 1985, 110, 851. Fogg, A. G., Femandez-Arciniega, M. A., and Alonso, R. M., Analyst, 1985, 110, 1201. Fogg, A. G., Alonso, R. M., and Femandez-Arciniega, M. A., Analyst, 1986, 111, 249. Fogg, A. G., and Alonso, R. M., Analyst, 1987,112, 1071. Liao, W., and Lu, X., Chinese J . Chromatogr., 1992, 10(2), 118. Paper a03251 F Received May 9, 1996 Accepted June 25, 1996

ISSN:1359-7337    

DOI:10.1039/AC9963300275

出版商:RSC    

年代:1996

数据来源: RSC

摘要:

Analytical Communications, August 1996, Vol33 (279-281) 279 Identification of Selenium Species in Selenium-enriched Garlic, Onion and Broccoli Using High-Performance Ion Chromatography With Inductively Coupled Plasma Mass Spectrometry Detection Honghong Gea, Xiao-Jia Caia, Julian F. Tysona,* Peter C. Uden," Eric R. Denoyerb, and Eric Block" a Department of Chemistry, Box 3451 0, University of Massachusetts, Amherst, h The Perkin-Elmer Corporation, 761 Main Ave., Norwalk, CT 06897-0215, USA c Department of Chemistry, State University of New York, Albany, NY 12222, USA MA 01 003-451 0, USA Six standard selenium species including selenocystine, methyl selenocysteine, selenite, selenomethionine, allyl selenocysteine and selenate have been separated by high-performance ion chromatography on a Hamilton PRPX-100 column and detected by ICP-MS.Selenium enriched vegetables were analysed. Five selenium species and several unknown peaks were detected. The nutritional effects and possible toxicity of selenium on a biological system depend on different selenium forms and their quantitites.1.2 Selenium compounds are dispersed throughout the environment by a variety of industrial, agricultural and natural processes.There is a growing interest in the speciation of selenium in different matrices, in order to gain an understanding of the selenium cycle and of selenium bio- availability in nature. Chromatographic separation with element specific detection is clearly a candidate method for trace selenium speciation.3 Element specific detectors such as atomic absorption spectrom- eters and atomic emission spectrometers have been coupled to GC for the determination of volatile selenium species or derivatives,@ while ion-exchange, ion-pair and reversed-phase liquid chromatography have been used with derivatization for the separation and quantification of inorganic selenium species and selenoamino acids.'@-19 In this work, an inductively coupled plasma mass spec- trometer has been directly coupled to a high-performance ion chromatographic system.For the first time, six non-volatile selenium species were separated in a relatively short time. The compounds were selenite, selenate, selenocystine, methyl selenocysteine, allyl selenocysteine and selenomethionine. Several selenium-enriched vegetables were analysed, including garlic, onion and broccoli.Allium group vegetables tend to take up inorganic selenium followed by conversion to various organic f0rms,~O-23 possibly including methyl selenocysteine and allyl selenocysteine. Five selenium species were identified in these vegetables and some unknown peaks were also observed. * To whom correspondence should be addressed.Experimental Sample Sources Standard selenomethionine, selenocystine and sodium selenate were purchased from the Sigma (St. Louis, MO, USA). Plasma- grade standard selenite solution was obtained from Spex Industries (Edison, NJ, USA). Standard methyl selenocysteine and allyl selenocysteine were obtained from Professor H. Ganther, University of Wisconsin (Madison, WI, USA). Freeze- dried selenium-enriched plant samples were obtained from Dr.C. Ip, Roswell Park Cancer Institute (Buffalo, NY, USA). The plants were grown under standard nutrition conditions with addition of sodium selenate. Instrumentation The separation system consisted of a Waters M-6000 high pressure pump and a Hamilton PRPX-100 PEEK anion exchange column (4.6 mm X 150 mm). A Perkin-Elmer SCIEX ELAN 5000 inductively coupled plasma mass spectrometer was used as the detector for which the operation parameter settings are shown in Table 1.Polyethylene tubing (30 cm X 0.58 mm id) transferred the LC eluent into the spectrometer cross-flow nebulizer inlet. A Scott-type, double-pass spray chamber was used. Method The optimized mobile phase was 5 mmol 1- ammonium citrate buffer at pH 4.8 containing 2% methanol as an organic modifier. The mobile phase was delivered isocratically at 1 ml min-l.Selenium enriched plant samples studied include garlic samples ~~ Table 1 Parameter settings for ICP-MS Forward rf power/W Plasma Ar flowb min-l Auxiliary Ar flow/l min-l Carrier (nebulizer) Ar flowj min-' Sampler and skimmer cones Scan mode Dwell time/ms Mass of selenium monitored 1000 15 0.8 0.75-0.95 Ni Peak hop 100 82280 Analytical Communications, August 1996, Vol33 7 A and B (containing approximately 1000 and 100 ppm total Se, respectively), an onion sample C (containing approximately 100 ppm total Se), a broccoli sample D (containing approx- imately 300 ppm total Se) and a natural garlic sample (estimated to contain approximately 0.02 ppm total Se).Finely ground plant samples were accurately weighed (about 0.2 g) into 15 ml centrifuge tubes, and extracted with 4 ml of 10% methanol in 0.2 mol 1-l hydrochloric acid ~olution.~3 The mixtures were sonicated for 1 h and centrifuged. The extracts were filtered through a 0.2 pm polysulfone membrane filter after adjusting the pH to about 7.0, and retained for analysis.Spike recovery experiments based on selenate were performed. Comparison of chromatographic peak profiles with those of reference analytes was also carried out. 40000- 30000 I v) v) C 3 - Results and Discussion Fig. 1 shows the chromatogram of six standard selenium species (250 ppb for each standard). Figs. 2, 3, 4 and 5 are chromatograms for garlic samples A and B, and the onion and broccoli samples C and D, respectively.Methyl selenocysteine, selenocystine, selenite, selenomethionine and selenate were observed in garlic sample A (1000 ppm Se), methyl seleno- cysteine being the predominant form of selenium observed. Selenocystine was found to be predominant in garlic sample B (100 ppm Se). At this stage it is not clear whether the peak 2 I - Ti rne/rnin Fig. 1 Separation of six selenium standard compounds (1, selenocystine; 2, methylselenocysteine; 3, selenite; 4, selenomethionine; 5 , allylseleno- cysteine; and 6, selenate; 250 ppb of Se for each standard) by high performance ion chromatography with ICP-MS detection.Fig. 2 Separation of compounds (1, selenocystine; 2, methylseleno- cysteine; 3, selenite; 4, selenomethionine; Y, unknown; and 6 , selenate) extracted from garlic sample A (approximately 1,000 ppm total Se).contribution to selenocystine is completely resolved from any other selenium-containing species eluted in the void volume. Methyl selenocysteine, selenocystine and selenomethionine were found in the onion sample while the two former amino acids together with selenite and selenate were found in the broccoli sample.There were three major unknown peaks observed in these samples, peak X with 5.60 min retention time, peak Y with 10.06 min retention time and peak Z with 12.07 min retention time. Chromatographic comparison experiments 8000 r Y I A 6ooo/ : I . 8 4000 I\ I \ 5 c .- P I 10 12 14 0 2 4 6 8 Ti me/mi n Fig. 3 Separation of compounds (1, selenocystine; X, unknown; and Y, unknown) extracted from garlic sample B (approximately 100 ppm total Se).2 Y 0 2 4 6 8 10 12 14 16 18 Tirne/rnin Fig. 4 Separation of selenium compounds (1, selenocystine; 2, methyl- selenocysteine; 3, selenite; Y, unknown; and Z, unknown) extracted from onion (approximately 100 ppm total Se). - 1 Fig. 5 Separation of selenium compounds (1, selenocystine; 2, methyl- selenocysteine; 3, selenite; X, unknown; Y, unknown; 6, selenate; and Z, unknown) extracted from broccoli (approximately 300 pprn total Se).Analytical Communications, August 1996, Vol33 28 1 showed that none of these unknown peaks could be identified with any of the standards available.Selenate recovery from a freeze-dried garlic sample was 98 f 4%. No selenium species were detected in the natural garlic sample (0.02 ppm Se), as the method does not have sufficient detection power.The provision and maintenance of the ELAN 5000 spectrometer by the Perkin-Elmer Corporation is gratefully acknowledged. References Olivas, R. M., and Donard, 0. F. X., Anal. Chim. Acta, 1994, 286, 357. Kolbl, G., Kalcher, K., Irgolic, K., and Magee, R. J., Appl.Organomet. Chem., 1993, 7, 443. Uden, P. C., Element-Specfic Chromatographic Detection by Atomic Emission Spectroscopy, American Chemical Society Symposium Series No. 479, ACS, Washington DC 1992. Cai, X., Uden, P. C., Block, E., Zhang, X., Quimby, B., and Sullivan, J. J., J . Agric. Food Chem., 1994, 42, 2081. Jiang, G.-b., Ni, Z.-m., Zhang, L., Li, A., Han, H.-b., and Shan, X.-q., J .Anal. At. Spectrom., 1992, 7, 447. D’Ulivo, A., and Papoff, P., J . Anal. At. Spectrom., 1986, 1. 479. Jiang, S. G., Robberecht, H., and Adams, F., Appl. Organomet. Chem., 1989, 3, 99. Cai, X., Uden, P. C., Block, E., Zhang, X., Quimby, B., and Sullivan, J. J., J. Agric. Food Chem, 1995, 43, 1754. Tanzer, D., and Heumann, K. G., Atmos. Environ. 1990, 24A, 3099. 10 11 12 13 14 15 16 17 18 19 20 21 22 23 Laborda, F., de Loos-Vollebregt, M.T. C., and de Galan, L., Spectrochim. Acta, Part B , 1992. 46, 1089. Lafreniere, K. E., Fassel, V. A., and Eckel, D. E., Anal. Chem., 1987, 59, 879. Potin-Gautier, M., Boucharat, C., Astruc, A., and Astruc, M., Appl. Organomet. Chem., 1993, 7, 593. Sanz-Medel, A., Aizpun, B., Marchante, J. M., Segovia, E., Femandez, M.L., and Blanco, E., J . Chromatogr., 1994,683,233. Schlegel, D., Mattusch, J., and Dittrich, K., J . Chromatogr., 1994, 683, 26 1. Matni, G., Azani, R., Van Calsteren, M. R., Bissonnette, M. C., and Blais, J. S., Analyst, 1995, 120, 395. Colon, L. A., and Barry, E. F., J . High Res. Chromatogr., 1991, 14, 609. Huyghues-Despointes, A., Momplaisir, G. M., Blais, J. S., and Marshall, W. D., Chromatographia, 1991, 31, 481. Blaise, J.-S., Huyghues-Despointes, A., Momplaisir, G. M., and Marshall, W. D., J . Anal. At. Spectrom., 1991, 6, 225. Laborda, F., Chakraborti, D., Mir, J. M., and Castillo, J. R., J . Anal. At. Spectrom., 1993,8, 643. Banuelos, G. S., Dyer, D., Ahmad, R., Ismail, S., Raut, R. N., and Dagar, J. C., US. J . Soil Water Conserv., 1993, 48, 530. Wang, W., Tang, J., and Peng, A., Shengwu HuaXue Zazhi, 1989,5, 229; Chem. Abstr., 1989, 111, 95847d. Yang, M., Wang, K., Gao, L., Han, Y., Lu, J., and Zou, T., J. Chin. Pharm. Sci., 1992, 1, 28; Chem. Abstr., 1993, 118, 77092~. Ip, C., and Lisk, D. J., Carcinogenesis (London), 1994, 15, 1881. Paper 6102833K Received April 23, I996 Accepted June 20, I996

ISSN:1359-7337    

DOI:10.1039/AC9963300279

出版商:RSC    

年代:1996

数据来源: RSC