摘要:

Anulytical Communications, December 1996, Vol33 (23H-26H) 23H Highlight Organic-phase Amperometric Biosensors Emmanuel I. Iwuoha and Malcolm R. Smyth* Biomedical and Environmental Sensor Technology (BEST) Centre and School ($ Chemical Sciences, Dublin City University, Dublin 9, Ireland Amperometric biosensors, which are designed for the detection of analytes in non-aqueous media or in organic media containing small amounts of water (microaqeuous systems), have been developed in the past 10 Though there is now some interest in organic-phase immunological sensors, most organic-phase biosensors are enzyme-based.Organic- phase enzyme electrodes (OPEEs) contain redox enzymes as the biological sensing element; these are capable of entering into a redox-catalytic reaction with the analyte in an organic solvent environment.The organic phase may be encapsulated with the enzyme on the electrode surface such that the analyte partitions into the organic phase to react with the enzyme. Alternatively, the analyte may be contained in an external organic medium from which it partitions into the immobilized enzyme film for reaction to take place. Most of the studies so far involving OPEEs are based on the analyte being contained in an external organic medium.2.3 The ability of an OPEE to operate in an organic medium depends on a number of factors, including the nature of the enzyme, the solvent hydrophobicity, the substrate hydrophobicity, the immobilization matrix and the interactions between enzyme and solvent.Several redox enzymes, including tyrosinase, horseradish peroxidase (HRP) and glucose oxidase (GOx), have been used as biocomponents in OPEEs.Factors That Determine the Stability of OPEEs Structural Properties of the Enzyme Some redox enzymes used in OPEEs, such as GOx and HRP, contain active sites that are imbedded deep in the protein molecule. This protects the active sites from direct influence of the organic solvent, thereby ensuring that the flexibility and polarity of the redox-catalytic centres of the biosensor are maintaineG.8 For example, GOx has an FAD redox centre located 8 A from the surface of the e n ~ y m e .~ in addition to the protective location of the redox active site of the protein, an ideal enzyme such as GOx has a surface covered by a carbohydrate shell which affords greater stability to the enzyme in organic media.The effect of these structural features of the * To whom correspondence should be addressed. enzyme is that the biosensor is stable in both hydrophilic polar solvents and non-polar solvents. The former class of solvents are known to denature enzymes by desorbing the essential water of hydration necessary for the maintenance of the structural features for enzyme reaction to take place.In addition to HRP and GOx-based electrodes, some other enzyme-based bio- sensors are also known10 to be stable in both polar and non- polar solvent media. These include systems based on alcohol dehydrogenase, tyrosinase, cholesterol oxidase, laccase and acetylcholine esterase. Recently, it has been shown that the stability of the biosensor in organic media can be enhanced by protein engineering.’(),* In its simplest form, the enzyme component of the biosensor can be chemically modified by targetting specific amino acids on the protein surface without tampering with the active site. For example, HRP electrodes prepared with HRP, in which the &-amino acids have been modified with N-hydroxysuccinic acid esters, show greater stability in organic solvents than electrodes prepared with the native enzyme.Table 1 contains the kinetic parameters of biosensors prepared with native and suberic acid bis(N- hydroxysuccinimide ester) (SA-NHS) HRPs. In both acetoni- trile and phosphate buffer, SA-NHS HRP exhibited higher values for the biosensor turnover rate constant, kcat’, and biosensor substrate specificity parameter, kCat’/KM’, than native HRP.The results show that it is possible to customize an OPEE to make it, not only very stable in a chosen organic phase, but also to amplify its response and substrate specificity. Solvent Properties Another factor that greatly determines the stability of the biosensor in an organic medium is the polarity of the solvent.A polarity index known as Laane’s log P valueL2 is normally used to classify organic phases as either hydrophilic or hydrophobic. Hydrophobic solvents have log P values > 4, while hydrophilic solvents have log P < -2. The former are predicted to be suitable for enzyme assays, while the latter are not, on account of their abiIity to dehydrate both the enzyme active site and the enzyme surface.The effect of this linkage between enzyme reactivity and solvent polarity is that most of the initial studies on organic-phase biosensors were based on hydrophobic Table 1 Table of the kinetic parameters of 0s-polymer-HRP-GCE amperometric responses to H202, for native and suberic acid bis(N-hydroxysuccinimide ester), SA-NHS, HRPs. Each biosensor contained 8.3 X mol cm-2 0~[(byp)~(PVP)~~Cl]Cl and 5 x 10-8 mol cm-? HRP cross-linked with 1.4% v/v glutaraldehyde on a 0.07 1 cm2 glassy carbon electrode (GCE).The electrodes were polarised at 50 mV versus Ag/AgCI at 20 “C and 800 rpm. The acetonitrile reaction medium contained 0.1 mol dm-3 tetraethylammoniurn p-toluenesulfonate as supporting electrolyte Solvent HRP KM’lmmol dm-3 k,,,‘/pmol cm-2 s-I lo4 (kcat’/KM’)/cm s-1 Buffer Native 0.10 64.22 6.42 SA-NHS 0.30 24.22 8.08 90% CH3CN (H20 is co-solvent) Native 0.01 1 2 1.90 20.0 SA-NHS 0.042 109.5 26.0724H Analytical Communications, December 1996, Vol33 solvents such as chloroform and hexane, in which the enzyme components of the sensor were expected to be stable.However, OPEEs based on HRP and GOx have been shown4J3,14 to be stable in essentially hydrophilic solvents such as acetonitrile, butanol, acetone, tetrahydrofuran and methanol.It is noteworthy that the stability of the biosensor in hydrophilic solvents can be enhanced by including some amount of water in the polar organic phase. This is because the immobilized enzyme surface amino acid residues and the redox-catalytic active sites require essential water of hydration for the maintenance of the appropriate configuration of the enzyme, necessary for its binding with substrate molecules in the non- aqueous phase.The amount of water required for an enzyme biosensor to give response in a polar organic-phase medium depends on the enzyme. For some enzymes, the small amount of water retained during biosensor preparation is sufficient for the reactivity of the biosensor.For example, GOx- and HRP-based biosensors have been shown to exhibit measurable responses in 100% CH3CN, while tyrosinase electrodes require up to 5% water to show activity in polar solvents. Hydrophobic solvents also require some amount of water for enzyme biosensing. These solvents are usually saturated with water before being used as media for electroanalytical detection of relevant compounds.15 The amount of water contained in the solvent is usually insignificant compared with the amount of solvent, and it depends on the ability of the solvent to absorb water. The result is a microaqueous solvent system that concentrates its water at the more polar enzyme surface and active site.In this way the essential reactive structures of the enzyme remain intact . Solubility of the Immobilization Matrix A major aspect of biosensor research and technology, is the search for immobilization matrices that are stable in the background solution. Many immobilization matrices for en- zyme-based biosensors are organic in nature, and are therefore bound to dissolve in organic phases.However, some polymeric materials, such as the polyester sulfonic acid anionomers, i.e., Eastman AQ polymers, produced by Eastman Kodak, exhibit exceptional stability in organic solvents. The Eastman AQ 55 polymer is the most widely used immobilization matrix for O P E E S . ~ , ~ ~ , ~ ~ The main drawback of the Eastman AQ polymer is its instability in aqueous phase.This makes it difficult to compare the aqueous- and organic-phase performance of Eastman AQ polymer-based biosensors. Another immobiliza- tion technique which has been found to be suitable for organic- phase biosensing is cross-linking with glutaraldehyde. The glutaraldehyde immobilization matrix has been used to im- mobilize enzyme or enzyme-redox polymer electrostatic com- plexes.The cyclic voltammetric diffusion controlled (no H202) and kinetic-controlled (0.3 mol dm-3 H202) responses of a glutaraldehyde-based HRP-0s-polymer electrode in acetoni- trile medium containing 10% v/v water is shown in Fig. 1. GOx electrodes containing up to 1.5% v/v glutaraldehyde have been shown to be stable in organic solvents such as acetonitrile, tetrahydrofuran, methanol, acetone and butanol, as well as in phosphate buffer.438.10 The advantages of glutaraldehyde im- mobilization over the Eastman AQ polymer is that the sensor can be used in both aqueous and organic phases.Other techniques that have been used to prepare stable OPEEs include entrapment in dialysis membranes, kappa-carrageenan gel, cellulose acetate, preparation of graphite-Teflon and epoxy resin electrodes, cross-linking with poly(ethy1ene glycol) and adsorption on alumina.Many of the immobilization techniques used so far in OPEE fabrication entail drop-coating of the enzyme onto the electrode surface. This does not ensure good reproducibility of the electrode. In a recent application, OPEEs have been produced by entrapping HRP within an electrosynthesized poly(ani1ine) (Pani) film.*O The Pani-based OPEE was found to be stable in a number of solvents, including acetonitrile, acetone, tetra- hydrofuran, propan-2-01 and phosphate buffer.The Pani film has the advantage that its thickness can be controlled, and being a redox conducting polymer, additional electron transfer mediator may not be necessary in the biosensor. The Pani film is grown under acidic conditions, then reduced in buffer medium at neutral pH and then oxidized before being doped with HRP.For example, a Pt-Pani-HRP electrode was prepared in 0.1 mol dm-3 HCl and 2.5% sodium poly(viny1 sulfonate), reduced in buffer at -500 mV for 10 min, then oxidized for 20 min at +650 mV in the presence of 10 mg dm-3 HRP. The Pani- based peroxide sensor prepared in this way was electroactive and did not require an additional electron transfer mediator. Response Characteristics of OPEEs Spectroscopic studies have shown that although organic solvents may affect the ability of an enzyme to catalyse reactions, the enzyme reaction mechanism is not altered.17 Thus, an OPEE reaction scheme can be described by the Michaelis-Menten kinetics in which, within the sensing film, the immobilized enzyme undergoes a redox reaction with the substrate, and the oxidized/reduced enzyme is reconverted to its native form by electron exchange with an electron transfer species.In this ping-pong reaction profile, the substrate and the electron transfer mediator undergo bimolecular reactions with the immobilized enzyme variants.The observed current is related to the enzyme kinetics inside the sensor film, the diffusion coefficient of the substrate in solution and within the immobilization matrix, and the partition coefficient of the substrate into the film. In addition, the homogeneity of the kinetic reaction within the film affects the reaction rate, just as the physical characteristics of the solvent medium affect the diffusion and partition coefficients of the substrate.OPEEs exhibit cyclic voltammetric and steady-state amper- ometric responses similar to those obtained in the aqueous phase. However, the magnitudes of the responses are large in the aqueous medium compared with the organic phase. The CV responses of OPEEs in the presence and absence of substrates are generally lower in the organic phase than that in the aqueous phase.However, the Zk:ZD values are usually higher for the organic phase than the aqueous phase. (Zk and ZD are the catalytic and the diffusion-controlled currents, respectively, measured at the plateau region of the voltammograms; see Fig. 1.) Thus, +630.0 +554.a +479.7 +404.5 2 +329.4 % +254.2 3 +179.0 +0.45 +0.35 +0.25 +O.15 +0.05 PotentialN Fig. 1 Cyclic voltammograms of GCE-HRP-[Os(bpy)2 (PVP)l&l]Cl biosensor in 90% v/v CH3CN (co-solvent water) in the absence (lower voltammogram) and presence (upper voltammogram) of 0.3 mmol dm-3 H202. Surface coverage of Os2+/3+ centres = 5.85 X mol cm-*; [HRP],leCtrode = 20 U cm-2; supporting electrolyte = 0.10 mol dm-3 tetraethyiammonium p-toluenesulfonate (TEATS).Potential sweep rate = 5 mV s-’ versus AgfAgC1.Analytical Communications, December I996, Vol33 25H enzyme-based biosensors are expected to be more efficient and exhibit lower background currents in organic solvents com- pared with aqueous systems. The low electrochemical responses of OPEEs are not unconnected with the differential stabilization of the ground states of the biosensor and the substrate molecule vis a vis the transition state.In addition, the partitioning and diffusion characteristics of the substrate depend on the polarity of the solvent molecule as well as its kinematic viscosity. The partitioning effect of the solvent medium is reflected in the apparent Michaelis-Menten constant, KM’, which is KM/K~, where Ks is the partition coefficient of the substrate into the enzyme film.KM’ values can be used to assess the ease with which the substrate molecule can partition out of the solution into the enzyme. Higher KM’ values mean that the substrate stays more in the solvent medium than within the enzyme film, with the result that higher concentrations of the substrate will be required to saturate the activity of the biosensor.Another kinetic parameter that is used to access biosensor organic-phase reactivity is the kca,‘/KM’ ratio. It is the apparent second-order rate constant of the reaction of immobilized enzyme with the analyte molecule. Whereas kcat’, the apparent turn-over rate constant of the biosensor, refers to the decomposition of the enzyme-substrate intermediate, kcat’/KM’ refers to the reactivity of the enzyme with the substrate.The term k,,,’ depends on the maximum response obtainable from the sensor and may be used to evaluate the solvent dependence of the activation energy of the sensor, but kcat’/KM’ is a measure of the specificity of the biosensor for the substrate. It is therefore possible to use the k,,,‘/KM’ values to evaluate how specific the sensor is for the analyte as the biosensor is transferred from one solvent to another. The higher the kcat’/KM’ value for a substrate in a solvent, the more specific the biosensor is for the analyte.The kc,,’/KM’ value depends on whether the electron transfer mediator is immobilized or in solution. OPEE kcat’/KM’ values for a GOx-based sensor have been shown to be larger for assays in hydrophilic solvents than in aqueous buffer solution, if the mediator is in s ~ l u t i o n .~ However, for systems in which the mediator is immobilized, both the response signal and the kcat’/ KM’ values are smaller for hydrophilic solvents than for phosphate buffer. l o Actually, kca,‘/KM’ is related to the activa- tion free energy of the biosensing system by -AG#/ RT = ln[(kca,‘/KM’)h/kB~, where h and kB are Planck constant and Boltzmann’s constant, respectively, while AGf is the difference between the Gibbs energy of the transition and ground states of the electro-enzymatic reaction.Consequently, kcat’/KM’ values reflect the role of the solvent media in differential stabilization of the ground and transition states of the biosensing system.Application of OPEEs OPEEs have been employed in the detection of compounds of analytical and environmental interest. These include phe- nols, 1-3.5-7 peroxide^'^,^*.*^ and pesticide materia1~.1~,20,21 Phenols, for example, are employed in the preparation of detergents and some pharmaceutical products. The toxicity of these compounds require that their concentration in products and wastes be carefully monitored.OPEEs containing tyrosi- nase provide a fast and efficient method of detecting phenols. Peroxides are used in a very large scale in the industry because of their catalytic and bleaching properties. They are applied as the main bleaching agent in detergent and domestic bleach and as a mild antiseptic/disinfectant in cosmetics. HRP-based OPEEs constitute very efficient tools for monitoring both H202 and organic peroxides in various sample matrices with minimal or no pre-treatment, the only requirement being to find the solvent in which the analyte is soluble.Other traditional techniques for peroxide determination, such as polarographic and photometric methods, are usually time consuming. There are other toxic compounds of environmental interest which act as HRP inhibitors.Examples of these are thiourea and ethylenethiourea, which are the parent compounds (and degra- dation products) of pesticides.14.20 These compounds have deleterious effects if they enter the food chain. Consequently, an efficient method of monitoring them is necessary. OPEEs based on HRP have been applied for the detection of these organic pesticide compounds, which act as HRP inhibitors.The compounds detected include thiourea and ethylenethiourea, which are related to phenylthiourea, and a-naphthylthiourea pesticides. Methylisothiocyanate (MeSCN) is another toxic pesticide degradation compound that has been monitored with peroxide sensors. Fig. 2 shows the calibration curves for MeSNC in 90% acetonitrile, for GCE-Os-polymer-glutar- aldehyde-based organic phase biosensors containing native HRP, acetic acid N-hydroxysuccinimide ester (AA-NHS) HRP or suberic acid bis(N-hydroxysuccinimide ester) (SA-NHS) HRP.The results show that SA-NHS HRP is a more efficient biocomponent for the MeSCN sensor than AA-NHS or native HRP. OPEEs have also been applied in the detection of alcohols in petrol, glucose in organic solvents and trace amounts of water in various materials. This means that it should be possible to apply OPEEs in on-line process control in the alcoholic beverage and petroleum industries.The success of GOx-based organic-phase biosensors shows that it may also be possible to employ such an enzyme electrode in the determination of sugar in highly alcoholic media, such as fermentation products.Conclusions Biosensors that operate in the aqueous phase are limited by the fact that they are only applicable to analytes that are soluble in aqueous media. On the other hand, OPEEs can be applied to analytes that are insoluble (or only partially soluble) in aqueous media but are completely soluble in organic media. Conse- quently, OPEEs have increased the scope of application of biosensors, as almost any compound can now be analysed with an OPEE, provided that an appropriate redox enzyme is available for assays in suitable organic solvents.Moreover, the stability of OPEEs in the organic phase can be enhanced and the response amplified by controlled chemical modification of the enzyme component. It is now possible to use the biosensor substrate specificity parameter, k,,,‘/KM’, to determine the 0.3 1 YSl I s 5 0.2- 3 L NATIVE 0.0 1 I I I 1 0.0 0.3 0.6 0.9 1.2 [MeSNC]/mmol dm-3 Fig.2 Calibration curves of methylisothiocyanate in 90% acetonitrile (10% H20) for HRP-[Os(byp)2(PVP)1&1]C1-GCE biosensors. Curves are shown for native HRP, HRP modified with acetic acid N-hydroxy- succinimide ester, AA-NHS, and suberic acid bis(N-hydroxysuccinimide ester), SA-NHS.Immobilization mahix contained 1.5% v/v glutaraldehyde. [HRPIelectrode = 5 X 10-8 mol cm-2; the acetonitrile solution contained 0.3 mmol dm-3 H202, and 0.1 mol dm-3 TEATS supporting electrolyte. Operational voltage was -100 mV versus Ag/AgC1 at 25 “C and 700 ‘pm.26H Analytical Communications, December 1996, Vol33 solvent that is most suitable for monitoring a particular substrate.References 1 2 3 4 5 6 7 8 9 10 Hall, G. F., Best, D. J., and Turner, A. P. F., Enzyme Microb. Technol., 1988, 10, 543. Saini, S., Hall, G. F., Downs, M. E. A., and Turner, A. P. F., Anal. Chim. Acta, 1991, 241, 1. Wang, J., in Uses of Immobilised Biological Compounds, ed. Guilbault, G. G., and Mascini, M., Kluwer Academic Publication, Dordrecht, 1993, pp.255-262. Iwuoha, E. I., Smyth, M. R., and Lyons, M. E. G., J . Electroanal, Chem., 1995,30, 81. Iwuoha, E. I., Adeyoju, O., Dempsey, E., Smyth, M. R., Liu, J., and Wang, J., Biosens. Bioelectron., 1995, 10, 661. Hall, G. F., and Turner, A. P. F., Anal. Lett., 1991, 24, 1375. Campanella, L., Fortuney, A., Sammartino, M. P., and Tomassetti, M., Talanta, 1994, 41, 1397. Iwuoha, E. I., Smyth, M. R., and Vos, J. G., Electroanalysis, 1994, 6, 982. Hecht, H. J., Scomburger, D., Kalis, H., and Schmid, R. D., Biosen. Bioelectron., 1993, 8, 197. Iwuoha, E. I., Lyons, M. E. G., and Smyth, M. R., Biosens. Bioelecrron., 1996, in the press. 11 12 13 14 15 16 17 18 19 20 21 Ryan, O., Smyth, M. R., and O’Fagain, C., Enzyme Microh. Technol., 1994, 16, 501. Laane, C., Boeren, S., Vos, K., and Veeger, C., Biotechnol. Bioeng., 1987, 30, 81. Iwuoha, E. I., and Smyth, M. R., in Electroactive Polymer Electrochemistry: Part 2 , Methods and Applications, ed. Lyons, M. E. G., Plenum Press, New York, 1996, pp. 297-332. Adeyoju, O., Iwuoha, E. I., and Smyth, M. R., Anal. Chim. Acta, 1995, 305, 57. Hall, G. F., Best, D. J., and Turner, A. P. F., Anal. Chim. Acta, 1988, 213, 113. Wang, J., and Golden, T., Anal. Chem., 1989, 61, 1397. Chartterjee, S., and Russel, A., Biotechnol. Bioeng., 1992, 40, 1069. Schubert, F., Saini, S., Turner, A. P. F., and Scheller, F., Sens. Actuators, B, 1992, 7, 408. Wang, J., Lin, Y., and Chen, L., Analyst, 1993, 118, 277. Adeyoju, O., Iwuoha, E. I., and Smyth, M. R., Electroanalysis, 1995, 7, 924. Wang, J., Dempsey, E., Eremenko, A., and Smyth, M. R., Anal. Chim. Acta, 1993, 279, 203. Paper 6105270C Received July 29, 1996 Accepted September 25, 1996

ISSN:1359-7337    

DOI:10.1039/AC996330023H

出版商:RSC    

年代:1996

数据来源: RSC

ISSN:1359-7337    

DOI:10.1039/AC99633FX062

出版商:RSC    

年代:1996

数据来源: RSC

摘要:

AVSOFE 33('2) 23'-1-26H 413-440 (1996) DECEMBER 1996Ill I I IAnal y t i callln I I ' Lommu n I cat I onsFormer I y Analytical ProceedingsCONTENTSH IG H LIG HT 23HCOMMUNICATIONS 413417421425Highlight-Organic-phase Amperometric Biosensors-Emmanuel I. Iwuoha, Malcolm R. SmythAccelerated Solvent Extraction of Phenols From Soil-John R. Dean, AraizSantamaria-Rekondo, Edwin LudkinModel for the Sorption of Organic Compounds by Soil From Water-Salwa K. Poole, Colin F.PooleHigh Temperature Water Extraction Combined With Solid Phase Microextraction-HiroyukiDalmon, Janusz PawliszynCapillary Zone Electrophoresis of Human Recombinant Erythropoietin Using C8 CoatedColumns Without Additives in the Running Buffer-Oscar Nieto, Pedro Hernandez, LucasHernandezDetermination of Trace Amounts of Alcohols and Phenols in Complex Mixtures asFerrocenecarboxylic Acid Esters With Gas Chromatography-Atomic EmissionDetection-Jurgen Rolfes, Jan T.AnderssonRacemization of an L-Phenylalanine Residue Catalysed by an Adjacent Cysteine in aBradykinin Peptide Antagonist-Charles Brown, Paul Cutler, Christine Eckers, William Neville,George Okafo, Patrick Camilleri429433437 Cumulative Author Index438 Technical Abbreviations and Acronyms439 Conference DiaryA&- THE ROYALSOCIETY OF f @ ,CHEMISTRY%@ lnfOrmatiOn Typeset and printed by Black Bear Press Limited, /' Services Cambridge, England 1359-733711996112:l-Analytical CommunicationsINSTRUCTIONS TO AUTHORSAnalytical Communications is the analytical communications journal of theRoyal Society of Chemistry and publishes communications on all aspects ofthe theory and practice of analytical chemistry, fundamental and applied,inorganic and organic, including chemical, physical, biochemical, clinical,pharmaceutical, biological, automatic and computer-based methods.Com-munications on new techniques and instrumentation, detectors and sensors,and new areas of application, with due attention to overcoming limitationsand to underlying principles, are all equally welcome.There is no page charge for papers published in Analytical Cornmu-nications.The following types of papers will be considered.Communications, which must describe novel work and be of immediatescientific significance. These articles should be restricted to 2-4 journalpages in length (up to 10 A4 pages of typescript), including figures, tablesand references.Communications are usually published within 10 weeks ofreceipt. They are intended for brief descriptions of work that has progressedto a stage at which it is likely to be valuable to workers faced with similarproblems. A fuller paper may be offered subsequently, if justified by laterwork, to The Analysr or JAAS. Communications will be examined by twoindependent referees.Highlights, these are mini-reviews or summaries of research in a welldefined, specific topic area covering approximately the last 12 months. (i)Given topics should review work no more than approximately 12 monthsold. (ii) Articles should cover only the most interestinghignificantdevelopments in that specific subject area, eg., sample preparation forcapillary HPLC or sensors for pesticides analysis.(iii) The articles shouldbe highly critical and selective in referencing published work (perhapsreferring only to four or five papers if necessary). (iv) A small amount ofspeculation (one or two paragraphs) of possible future developments mayalso be appropriate in the Conclusions section. (v) Articles should be brief2-3 journals pages (up to 7 double-spaced, typed A4 pages), contain nomore than 1 or 2 tables and preferably no figures.Copyright. The whole of the literary matter (including tables, figures,diagrams and photographs) in Analytical Communications is Royal Societyof Chemistry copyright and may not be reproduced without permission fromthe Society or such other owner of the copyright as may be indicated.Papers that are accepted must not be published elsewhere except bypermission.Submission of a manuscript will be regarded as an undertakingthat the same material is not being considered for publication by anotherjournal in any language. All authors submitting work for publication arerequired to sign an exclusive copyright licence. All submissions should beaccompanied by a completed form, without which publication cannotproceed.Manuscripts. Papers should be typewritten in double spacing on one sideonly of the paper. Copies of any related, relevant, unpublished material andraw data should be made available on request.Each table and illustrationshould be on a separate sheet at the end of the text; three copies of text andillustrations should be sent to the Managing Editor, Analytical Cornmu-nications, The Royal Society of Chemistry, Thomas Graham House,Science Park, Milton Road, Cambridge CB4 4WF, and a further copyretained by the author.Administration and Publication Procedure. Receipt of a contributionfor consideration will be acknowledged immediately by the EditorialOffice. The acknowledgement will indicate the paper reference numberassigned to the contribution. Authors are particularly asked to quote thisnumber on all subsequent correspondence.All communications are sent to two referees, whose names are notdisclosed to the authors. On the basis of the referees’ reports, the ManagingEditor decides whether the paper is suitable for publication, eitherunchanged or after appropriate revision.This decision and relevantcomments of the referee are sent to the author.When rejection of a paper is recommended, the Editor informs the author,Authors will receive formal notification when papers are accepted forand returns the top copy of the manuscript.publication.Proofs. The address to which proofs are to be sent should accompany thepaper. Proofs should be carefully checked and returned immediately (byfirst class mail, air mail, express mail or fax). Particular attention should bepaid to numerical data both in the tables and text.Offprints. Fifty offprints of each paper are supplied free of charge.Additional reprints can be purchased.ANALYTICAL JOURNALS 1997Published by The Royal Society of ChemistryThe AnalystISSN 0003-265412 issues a year plus indexSterling f535.00US Dollars $963.00Analytical Communications1 2 issues a year plus indexSterling f210.00US Dollars $378.00ISSN 1359-7337Journal of Analytical AtomicSpectrometry (jAAS)12 issues a year plus indexSterling €657.00US Dollars $1 183.00ISSN 0267-9477Analytical Abstracts12 issues a yearSterling f656.00US Dollars $1 174.00ISSN 0003-2689SPECIAL PACKAGES (Non-RSC members only)The Analyst, Analytical Abstracts and AnalyticalCommunicationsJournal Ref.No. 0000-01 24Sterling f 1260.00US Dollars $2268.00The Analyst and Analytical CommunicationsJournal Ref. No. 0000-0140Sterling f670.00US Dollars $1 206.00The Analyst and Analytical AbstractsJournal Ref. No. 0000-01 32Sterling f 1072.00US Dollars $1 930.00ORDERINGNon-RSC members should send their orders to:The Royal Society of Chemistry, Turpin Distribution Services Limited, Blackhorse Road, Letchworth,Tel: +44 (0) 1462 672555 Fax: +44 (0)1462 480947 E-mail: turpin@rsc.orgRSC Members should obtain members’ prices and send their orders to:Membership Administration Department, The Royal Society of Chemistry, Thomas Graham House,Science Park, Milton Road, Cambridge CB4 4WF, United KingdomTel: +44 (011 223 420066 Fax: +44 (011 223 423623 E-mail: rscl @rsc.orgTurpin Distribution Services Limited is wholly owned by The Royal Society of ChemistryHerts SG6 1 HN, United Kingdom /c- THE ROYALI SOCIETY OF*A Informatio

ISSN:1359-7337    

DOI:10.1039/AC99633BX064

出版商:RSC    

年代:1996

数据来源: RSC

摘要:

Analytical Communications, December I996, Vol33 ( 4 1 3 4 6) 413 Accelerated Solvent Extraction of Phenols From Soil John R. Dean, Araiz Santamaria-Rekondo and Edwin Ludkin Department of Chemical and Life Sciences, University of Northumbria at Newcastle, Ellison Building, Newcastle upon Tyne, UK NEI 8ST Accelerated solvent extraction (ASE) has been investigated as a method of sample preparation in environmental analysis.Using an experimental design approach the influence of operating variables (pressure, temperature and extraction time) on the recovery of seven phenols that have been slurry spiked on soil has been investigated. In every case, except 2-methylphenol, no variables were found to be significant within the limits investigated (pressure, 4-20 MPa; temperature, 30-70 "C; and extraction time, 5-25 min).ASE was compared with shake-flask extraction and found to give similar recoveries. The exception was 2,4-dimethylphenol which was not recovered by the shake-flask approach and was only poorly recovered using ASE (mean recovery, 24.5% with an RSD of 17.6%, based on 15 individual determinations). For shake-flask extraction the mean recoveries ranged from 70.3 to 116.1% with RSDs of 6.5 to 27.2% compared to ASE which gave mean recoveries from 73.8 to 90.2% and RSDs between 8.7 and 21.1%.It is expected that improvements in precision would be obtained using an automated ASE system. A cleaner UV chromatogram was obtained from the ASE extract. The feasibility of ASE in environmental analysis is demonstrated. Phenolic compounds are toxic aryl alcohols widely used in the chemical industry for the manufacture of polymers, textiles, resins, dyes, petroleum refining, pulp processing and coal coking, 1-3 Moreover, chlorinated phenoxy acid herbicides and organophosporus pesticides can degrade yielding chloro and nitrophenols, re~pectively.~-5 The determination of phenols in the environment is therefore indicative of industrial pollution.It is not surprising to find that some phenols are included in the list of priority pollutants of the US Environmental Protection Agency, as well as in some European regulation^.^^^ The analytical characterization of organic contaminants is the first step when considering an efficient strategy for soil remediation. Thus, due to their possible damaging effect on the environment, phenols must be monitored in a rapid and simple way.The sample preparation stage is crucial in any chemical analysis of environmental samples. The analytical procedures applied for extracting phenols from the soil matrix prior to subsequent chromatographic analysis should avoid losses of the analytes and the use of large amounts of toxic solvents.As traditional methods, such as Soxhlet extraction, are time consuming and require large amounts of organic solvents, new alternative approaches are actively being sought. In this context both supercritical fluid extraction7,* and microwave-assisted extraction9 have both been applied to the extraction of phenols from soils. This paper however, considers the use of accelerated solvent extraction (ASE)l0 for the extraction of phenols from soil.As ASE is the newest of the techniques available for extraction from solid matrices there is limited information currently available in the scientific literature. During the work reported in this paper a home-built system was utilized for extraction of seven phenols from a slurry-spiked soil sample. Experimental Instrumentation A schematic diagram of the manual accelerated solvent extraction system is shown in Fig.1. A syringe pump (Carlo Erba SFC 300, capacity 150 ml) was used to pump the organic solvent into the 2 ml stainless steel sample extraction cell (Phase Separations, Clwyd, UK) which was located in a fan controlled oven (temperature range 25-100 "C). A nitrogen cylinder, operated at 40 psi, was used to purge the sample of residual solvent. Strategic positioning of two Rheodyne valves allowed the extraction cell to be filled with solvent, isolated from the rest of the system or purged with either fresh solvent or N2 gas.The phenols were analysed using an isocratic HPLC system (Thermo Separations, Stone, Staffs, UK). An acetonitrile-water (40 : 60) mobile phase containing 1 o/o acetic acid was pumped at 1 ml min-1 by a high pressure pump (model: P4000).Samples and standards were injected (50 pl) into a 25 cm X 4.6 mm ODs2 Spherisorb column (Phase Separations) located in a chromatographic oven maintained at 35 "C. Phenols were detected using a UV/VIS spectrophotometer (Model UV 1000) at 275 nm. Peak analysis was performed using Peak Simple software (SRI Instruments, Torrance, CA, USA).Materials Seven phenols were selected for analysis. Stock solutions of the phenols (phenol, 2,4-dichlorophenol, 4-nitrophenol, 4-chloro- 3-methylphenol, 2-methylphenol, 2-nitrophenol and 2,4-dime- thylphenol) were prepared at the approximate 1000 pg ml-' level. All phenols were supplied by Merck, Poole, Dorset. Soil Syringe pump Pump Purge valve Nitrogen cylinder I ~ $ E i i o n Oven Static valve Collection (Solvent + Analytes) Fig.1 Schematic diagram of manual ASE system.414 Analytical Communications, December 1996, Vol33 (3.8% carbon, 3.4% water and pH of 6.1) was prepared after collection by air-drying and subsequent passing through a 2 mm sieve. Procedure Soil spiking A sample of soil (15 g) was spiked using a slurry spiking procedure.Using this approach the phenols were added to a large volume of solvent (50 ml of acetonitrile) and the solvent evaporated overnight. In the absence of native samples, slurry spiking is the preferred approach.ll Soil samples were spiked such that 100% recoveries were expected, on the basis of complete extraction, for each of the phenols.The concentration of phenols in the extractant solution was calculated to be 100 ng ml-1. Table 1 Experimental variables for the central composite design PressureNPa TemperaturePC Time/min Star points 4 and 20 30 and 70 5 and 25 Mid point 12 50 15 Cube points 7 and 17 38 and 62 9 and 21 Accelerated solvent extraction Approximately 0.01 g of Celite was placed at the outlet of the extraction cell prior to every extraction in order to prevent blockages.To the cell was then added about 0.5 g of the spiked soil sample. The cell was then placed in the oven at a preset temperature for 10 min to allow sample equilibration and the chosen pressure was applied for a period of time (5 to 25 min). Finally, the static valve was opened to allow fresh solvent to pass through the cell for about 10 s.A N2 purge for removal of residual solvent was then applied. The volume of the extraction Table 2 High-performance liquid chromatography performance data* Compound Retention time/ Capacity min %RSD factor/K %RSD phenol 5.40 3.9 0.85 11.8 4-Nitrophenol 6.10 5.7 1.20 14.2 2-Methylphenol 7.08 6.1 1.53 13.1 2-Nitrophenol 9.06 6.8 2.24 12.5 2,4-Dimethylphenol 9.68 8.1 2.44 12.7 4-Chloro-3-methylphenol 1 1.19 9.7 3.02 14.2 2,4-Dichlorophenol 13.09 10.5 3.68 14.9 * n = 10 collected over five months.Table 3 Central composite design results Experi- ment number 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 Pressure/ MPa 12 7 7 17 17 20 12 4 12 12 12 7 17 7 17 Tempera- ture/”C 70 62 62 62 62 50 50 50 50 50 30 38 38 38 38 Extraction time/ min 15 9 21 9 21 15 25 15 5 15 15 21 21 9 9 Percentage Recovery 4-Nitro- 2-Methyl- 2-Nitro- 2P-Dimethyl- 4-Chloro-3- 2,4-Dichloro- Phenol phenol phenol phenol phenol methylphenol phenol 96 98 72 83 89 96 81 91 80 84 87 83 83 79 74 95 91 72 83 86 87 78 88 76 78 88 81 78 76 71 70 69 56 66 129 108 91 98 92 91 90 84 100 108 101 87 88 62 74 77 83 67 81 73 66 77 72 70 68 62 31 23 23 25 32 29 23 29 24 21 25 26 21 19 17 91 1 04 52 68 78 71 68 92 94 64 76 70 69 63 56 83 81 57 72 76 85 74 89 68 66 77 76 73 68 70 Table 4 Statistical data from multiple linear regression of the seven phenols r r2 p-values- intercept P T t P2 TZ t2 PT Pt Tt Phenol 4-Nitrophenol 2-Methylphenol 2-Nitrophenol 0.8524 0.7266 0.3850 0.2583 0.8376 0.2440 0.6329 0.7995 0.2559 0.7216 0.0752 0.1085 0.863 1 0.7450 0.1 166 0.1959 0.449 1 0.2524 0.5826 0.2519 0.3517 0.4232 0.1791 0.1212 0.9404 0.8843 0.0340 0.0255 0.8101 0.0435 0.4408 0.2893 0.8805 0.0663 0.0060 0.0281 0.8523 0.7264 0.1555 0.1251 0.5376 0.5608 0.2824 0.2699 0.7522 0.6690 0.1316 0.1116 2P-Dimethyl- phenol 0.8085 0.6536 0.2456 0.1368 0.4425 0.7185 0.3898 0.4906 0.8142 0.1558 0.7438 0.7438 4-Chloro-3- methylphenol 0.8006 0.6410 0.5294 0.41 86 0.9695 0.9559 0.7643 0.6525 0.761 1 0.9628 0.1283 0.1627 2,4-Dichloro- phenol 0.7 192 0.5 173 0.2962 0.1803 0.7515 0.5697 0.25 16 0.5807 0.7840 0.6888 0.4070 0.2685Analytical Communications, December 1996, Vol33 415 solvent was approximately 10 ml.All extractions were done using acetonitrile only. Shake-flask extraction Approximately 1 .O g of soil was placed into a container and 10 ml of an acetonitrile-water (40 : 60) mixture added.The contents of the container were then shaken for 10 min and the liquid phase removed, filtered through a 0.2 pm filter (Acrodisc, Phase Separations) to remove particulates prior to analysis. Experimental Design An experimental design approach was used to evaluate the influence of the three main operating variables of ASE, i.e., pressure, temperature and extraction time.The design chosen was a central composite design (CCD). In a CCD five levels of each variable are considered, and the number of experiments equals fifteen. The five variable-levels or values correspond to two star points, two cube points and one mid point. The limits of the variables or star points were set according to instrumental limitations.The ranges chosen were as follows: pressure 4-20 MPa (600-3000 psi); temperature 30-70 "C; and, extraction time 5-25 min. The CCD values are shown in Table 1 . Results and Discussion HPLC Chromatographic Data The retention time of the seven phenols studied over the course of this work (5 months) was determined. The results are shown in Table 2.The data shows the reasonable stability of the system. The approximate limit of detection of the seven phenols was as follows: 10-15 ng ml-l for phenol, 4-nitrophenol, 2-methylphenol and 2-nitrophenol and 30 ng ml-1 for 2,4-di- methylphenol, 4-chloro-3-methylphenol and 2,4-dichlo- rophenol. Peak height was chosen as the method of quantitation for the chromatographic peaks as better precision was obtained (%RSD ranged from 4.0 to 21.4 for peak area measurements and 1.0 to 12.8 for peak height measurements, as determined over 2 d and based on 10 determinations).Calibration plots for each phenol were obtained over the concentration range 0-120 ng ml- I . Correlation coefficients for the calibration plots ranged from 0.9990 to 0.9999. Experimental Design The fifteen experiments were done according to the CCD and the recoveries obtained are shown in Table 3.Multiple linear regressionI2 was then applied to the data using a model of the following form: Y = Po + (31P + P*T + (33t + ( 3 2 2 + p p + (36t* + p7PT + (3sPt + p97-t where Y is the percentage recovery of the individual phenol, Po is the intercept, P1...9 are derived coefficients, P = pressure, T = temperature and t is the extraction time.The results of the multiple linear regression are shown in Table 4. Significance at the 95% confidence level is indicated by a p value of < 0.05. The significant values are highlighted in bold in Table 4. It can be seen that only 2-methylphenol has any significant variables at the 95% confidence level.The significant variables are the intercept, pressure and extraction time, together with the interaction terms of pressure X extraction time and temperature X extraction time. It is therefore apparent that the operating limits of the chosen variables have limited influence on the 125C 0 > E \ - 2 0) cn .- Fig. 2 1250 1 I I 5 10 15 1250 0 5 ll 1 I 10 15 0 Ti me/m i n HPLC chromatograms of (a) a 100 ng ml-1 phenol standard mixture; (b) shake-flask extract; and (c) accelerated solvent extract.~~ Table 5 Extraction of seven phenols from a slurry spiked soil: comparison of shake-flask and a manual accelerated solvent extraction 4-Nitro- Phenol phenol Shake-flask* mean 102.4 104.5 %RSD 6.5 8.1 n 5 6 ASE+ mean 85.1 81.9 %RSD 9.1 8.7 n 15 15 * Acetonitrile-water (40 : 60).+ 100% acetonitrile. 2-Methyl- 2-Nitro- 2,4-Dimethyl- 4-Chloro-3- phenol phenol phenol methylphenol 70.3 75.7 not recovered 116.1 15.1 27.2 17.9 5 6 6 6 90.2 73.8 24.5 74.4 21.1 11.3 17.6 19.9 15 15 15 15 2,4-Dichloro- phenol 88.2 19.6 6 74.3 11.0 15416 Analytical Communications, December 1996, Vol33 recovery of the phenols. The exception to this is 2,4-dimethyl- phenol which was only poorly recovered.As the limited recovery of 2,4-dimethylphenol was consistently low (< 32 %) it appears that some irreversible binding occurs to the soil matrix. The lack of dependence upon the operating variables of phenol extraction is shown in Table 5 where all fifteen sets of data are summarized in terms of percentage mean recovery and %RSD. Comparison of ASE With Shake-flask Extraction A comparison of shake-flask extraction and ASE was done to assess the potential of the newer technique.Using shake-flask extraction an acetonitrile-water (40 : 60) mixture was used for extraction. However, this was not possible with ASE. In ASE the presence of water caused blockage of the extraction cell by the soil-water suspension. All subsequent work using ASE used 100% acetonitrile.As can be seen in Table 5 similar recoveries were obtained using shake-flask, under the same conditions, as compared to the manual ASE method. It should be remembered that the ASE data is the mean of 15 experiments during which all three operating variables were altered. No recovery data was obtained for 2,4-dimethylphenol using shake-flask extraction.However, it was noted that the ASE method gave a cleaner HPLC chromatogram than shake-flask extraction (Fig. 2). The importance of this observation may lead to chromatographic column lifetime being extended. Conclusions Data presented demonstrates the potential for ASE in environ- mental analysis. The observed cleaner UV chromatograms obtained using ASE could have implications for column lifetime.P. Hancock is gratefully acknowledged for characterization of the soil used in this study. References 1 2 3 4 5 6 7 8 9 10 11 12 Manahan, S. E., Hazardous Waste Chemistry, Toxicology and Treatment, Lewis Publishers, New York, 1990. Tver, D. F., Dictionary of Dangerous Pollutants, Ecology and Environment, Industrial Press, New York, 1981. Straub, C. P., Practical Handbook of Environmental Control, CRC Press, Boca Raton, FL, 1989. Puig, D., and Barcelo, D., J . Chromatogr. A, 1996, 733, 371. Lacorte, S.. and Barcelo, D., Environ. Sci. Technol., 1994, 28, 1159. DiCorcia, A., Bellioni, A., Madbouly, M. D., and Marchese, S., J . Chromatogr. A, 1996, 733, 383. Llompart, M. P., Lorenzo, R. A., and Cela, R., J. Chromatogr. Sci., 1996, 34 ,43. Llompart, M. P., Lorenzo, R. A., and Cela, R., J . Chromatogr. A, 1996, 723, 123. Lopez-Avila, V., Young, R., and Beckert, W. F., Anal. Chem., 1994, 66, 1097. Dean, J. R., Anal. Commun., 1996, 33, 191. Dean, J. R., Analyst, 1996, 121, 85R. Massart, D. L., Vandeginste, B. G. M., Deming, S. N., Michotte, Y., and Kaufman, L., Chemometrics: A textbook, Elsevier, Amsterdam, 1988. Paper 6f06.5006 Received September 20, I996 Accepted October 21, 1996

ISSN:1359-7337    

DOI:10.1039/AC9963300413

出版商:RSC    

年代:1996

数据来源: RSC

摘要:

Analytical Communications, December 1996, Vol33 (41 7-41 9) 417 Model for the Sorption of Organic Compounds by Soil From Water Salwa K. Poolea and Colin F. Pooleb.* College of Science, Technology and Medicine, South Kensington, London UK SW7 2AY Department of Chemistry, Wayne State University, Detroit, MI 48202, USA E-mail: cfp@chem.wuyne.edu ZenecaISmithKline Beecham Centre for- Analytical Chemistry, Imperial The solvation parameter model successfully accounts for the factors that determine the uptake of organic compounds by soil from water. Wet soil is significantly less cohesive than water, is less dipolar, has roughly similar hydrogen-bond basicity, and substantially weaker hydrogen-bond acidity.By analogy to other distribution systems it has similar extraction power to isobutanol and retention properties that resemble those of sodium taurodeoxycholate micelles in micellar electrokinetic chromatography.The possible use of chromatographic systems for the surrogate estimation of the soil-water sorption coefficient (Koc) is discussed. The soil-water sorption coefficient expressed on an organic carbon basis (Koc) is one of the key input parameters in models used to estimate the mobility and fate of organic contaminants in ground water systems.Since the experimental determination of KO, is difficult and time consuming by standardized methods,' estimated values are often used. Several correlation methods have been developed for this purpose using the aqueous solubility,2,' the octanol-water partition coefficient,24 topological and quantum mechanical descriptor^,^-^ and the retention factor for mixed mobile phases in reversed-phase LC.R.9 Some workers have recommended the use of kw, the value for the retention factor in water obtained by extrapolation from binary mobile phase compositions containing an organic solvent, as being a more realistic parameter for the soil-water system than the retention frzctor at some intermediate water- organic solvent composition.10.1 I However, many of these correlations were developed for specific compound series, such as alkylaromatics, and are not usually applicable to other compound types, especially polar compounds. The defining principle in all the above studies is that the sorption of non-ionic organic compounds by soil from water can be represented by a distribution mechanism between water and the water saturated organic fraction of soil in a manner analogous to liquid-liquid distribution and retention in re- versed-phase LC.The analogy with reversed-phase LC is strongly supported by the notion of water percolating through a somewhat porous inert microstructure (mineral matter in soil and silica in chromatography), supporting an immobilized solvated organic layer (organic matter in soil and the chemically bonded phase in chromatography), which is responsible for the selective retention of organic compounds.This being the case, it should be possible to model the uptake of organic compounds by soil from water using analogous models successfully applied to the retention of organic compounds in reversed-phase LC and solid-phase extraction.The most useful of these models is the * To whom correspondence should be addressed. solvation parameter model, which not only predicts retention, but enables different sorbents (or as proposed here, soils) to be characterized by their capacity for fundamental intermolecular interactions. l2-l9 The solvation parameter model in a form suitable for interpreting retention in reversed-phase LC and the uptake of organic compounds by soil from water is set out below: (1) where SP is some free energy related property of the system, such as the distribution constant in the soil-water system (log Koc) or the retention factor (log k ) in reversed-phase LC; the solute descriptors are identified as the characteristic volume V,, the excess molar refraction R2, the dipole/polarizability parame- ter ny, and a: and (s: parameters characterizing the solute's effective hydrogen-bond acidity and hydrogen-bond basicity, respectively. The system constants in eqn.(1) are defined as follows: the r constant is a measure of the relative importance of interactions with solute n- or n-electron pairs; the s constant determines the relative importance of dipole-type interactions; the a constant is a measure of the difference in hydrogen-bond basicity; the b constant is a measure of the difference in hydrogen-bond acidity; and the m constant is a measure of the relative ease of forming a cavity for the solute in the solvent and solvated sorbent.For any combination of solvent and sorbent the system constants can be obtained by multiple linear regression analysis for a group of solutes with known descriptors that are sufficiently varied to define all interactions in eqn.(1) and of sufficient number to establish the statistical validity of eqn. (1). In this paper it is shown that the solvation parameter model provides an adequate description of the sorption of organic compounds by soil from water and can be used to identify factors that explain the observed correlations within compound families for soil-water sorption and liquid- liquid distribution and reversed-phase LC retention factors. SP = c + mVx/100 + I * R ~ + STC? + acUy + h(37 Experimental The primary set of soil-water distribution constants (log Koc) were taken from Hong et al. 1 1 These proved inadequate in the range of solute descriptors available for modeling (see later) and were supplemented by additional values taken from the literat~re.'~~~"*~'O Values were selected using the following criteria: (i), values from more than one independent study did not vary by more than 0.2 log units; (ii), where only single values were available sufficient checks and precautions were included in the study to ensure the validity of the result; and (iii), that solute descriptors were available for the solute without having to resort to parameter estimates.The solute descriptors used in the model were taken from several ~ o u r c e s . * 5 ~ ~ ~ ~ ~ The fly values are corrected to allow for the variable basicity of certain compounds in solvents containing significant amounts418 Analytical Communications, December 1996, Vol33 of water For the reader’s convenience the distribution constants and solute descriptors used in this study are presented in Table 1.For multiple linear regression analysis and statistical treat- ment of data the program SPSS/PC+ (version 5.0) (SPSS, Chicago, IL, USA) run on a Vectra personal computer (Hewlett- Packard, Stockport, Cheshire, UK) was used.Results and Discussion Solute descriptors were available for 19 of the 23 solutes used by Hong et al.” in their correlation of soil-water distribution constants with chromatographic retention factors (these are Table 1 Solute descriptors and soil-water distribution constants used to construct the model given by eqn.(3) Solute descriptors Solutes Aniline 4-Nitroaniline 4-Bromoaniline 4-Nitrophenol Phenol Nitrobenzene 2,4-Dichlorophenol Acetophenone Benzene Toluene Ethylbenzene Chlorobenzene 1,2-DichIorobenzene 1,3-Dichlorobenzene 1 ,4-Dichlorobenzene 1 ,ZDimethylbenzene 1,3-Dimethylbenzene 1,4-Dimethylbenzene 1,2,4-Trichlorobenzene Naphthalene 1 -Methylnaphthalene 1 -Ethylnaphthalene Anthracene Phenanthrene Pyrene n-Propylbenzene n-Butylbenzene 1,2,3-Trirnethylbenzene 1,3,5-Trimethylbenzene Bromobenzene 1,3,5-Trichlorobenzene 1,2,3,4-Tetrachloro- benzene 1,2,3,5-Tetrachloro- benzene Dichloromethane Chloroform Tetrachloromethane 1,1,2-Trichloroethane l11,2,2-Tetrachloro- 1,2-Dibromoethane 1,1,2-Tnchloroethene 2,4-Dichlorophenol 2,4,6-Trichlorophenol 4-Bromophenol 1-Naphthol 4-Methoxyacetanilide Phenylurea 3-Methylaniline Azobenzene Dibenzothiophene ethane VX/lOO 0.8162 0.9910 0.9910 0.9493 0.775 1 0.8910 1.0200 1.0139 0.7164 0.8573 0.9982 0.8388 0.9612 0.9612 0.9612 0.9982 0.9982 0.9982 1.0836 1.0854 1.2263 1.3672 1.4540 1.4540 1.5850 1.1391 1.2800 1.1391 1.1391 0.8914 1.0836 1.2060 1.2060 0.4943 0.6167 0.7391 0.7576 0.8800 0.7404 0.7146 1.0200 1.1420 0.9501 1.1440 1.3133 1.0726 0.9570 1.4808 1.3790 R2 0.955 1.220 1.130 1.070 0.805 0.87 1 0.960 0.818 0.610 0.60 1 0.613 0.7 18 0.872 0.847 0.825 0.663 0.623 0.613 0.980 1.340 1.344 1.371 2.290 2.055 2.808 0.604 0.600 0.728 0.649 0.882 0.980 1.180 1.160 0.387 0.425 0.458 0.499 0.542 0.747 0.524 0.960 1.010 1.080 1.520 0.970 1.110 0.946 0.680 1.959 7c; a? 0.96 0.26 1.91 0.42 1.25 0.31 1.72 0.82 0.89 0.60 1.11 0 0.99 0.58 1.01 0 0.52 0 0.52 0 0.51 0 0.65 0 0.78 0 0.73 0 0.75 0 0.56 0 0.52 0 0.52 0 0.81 0 0.92 0 0.90 0 0.87 0 1.34 0 1.29 0 1.71 0 0.50 0 0.51 0 0.61 0 0.52 0 0.73 0 0.73 0 0.92 0 0.85 0 0.57 0.10 0.49 0.15 0.38 0 0.68 0.13 0.63 0.10 0.76 0.10 0.40 0.08 0.99 0.58 1.01 0.82 1.17 0.67 1.05 0.61 1.63 0.48 1.40 0.77 0.95 0.23 1.20 0 1.31 0 P,” h K 0 c 0.50 1.41 0.38 1.88 0.39 1.96 0.26 2.32 0.30 1.43 0.28 1.94 0.14 2.75 0.49 1.63 0.14 1.91 0.14 2.18 0.15 2.41 0.07 2.41 0.04 2.50 0.02 2.47 0.02 2.44 0.16 2.34 0.16 2.50 0.16 2.52 0 2.94 0.20 2.96 0.20 3.36 0.20 3.77 0.26 4.42 0.26 4.28 0.29 4.80 0.15 2.87 0.15 3.39 0.19 2.80 0.19 2.82 0.09 2.18 0 2.85 0 3.84 0 3.20 0.05 1.44 0.02 1.65 0 1.85 0.08 1.89 0.08 1.90 0.17 1.64 0.03 2.00 0.14 2.75 0.08 3.02 0.20 2.41 0.37 2.64 0.86 1.40 0.77 1.35 0.55 1.66 0.44 3.03 0.18 4.00 indicated by the first 19 rows in Table 1). Only a modest fit to the solvation parameter model was obtained with the r and s constants being statistically insignificant, eqn.(2): log Koc = (0.58 k 0.32) + (2.17 f 0.34)(Vx/100) + (0.17+0.13)ay - (2.17k0.23) p,” (2) p = 0.957 SE = 0.13 F = 54 n = 19 Here p is the overall correlation coefficient, SE the standard error in the estimate, F the F-statistic, and n the number of solutes. The range of descriptor values is too narrow and significant cross-correlation between descriptors (p [R2/ 3~3 = 0.89; and p[ny/pF] = 0.51) prevents a stable model from being produced.For this data set a two parameter model containing only the volume term and /3F would be just as adequate in describing the data set (rn = 2.16; b = -2.06; c = 0.59; p = 0.952; SE = 0.14; and F = 77).Additional solutes with known values of log Koc were added to the original data set to give a more robust fit. Using all 49 solutes in Table 1 the following model was obtained: log Koc = - (0.07 f 0.12) + (2.59 k O.l5)(V,/lOO) + (0.75_+0.10) R2 - (0.46k0.13) T C ~ - (1.95f0.18) rj,” (3) p = 0.982 SE = 0.17 F = 292 n = 49 The model provided by eqn.(3) is intuitively sensible and quite similar to other transfer models containing water as one phase; the range of values for the solute descriptors is now reasonable and cross-correlation is not a problem; and the model coeffi- cients are stable to random deletion of solutes. Given that determinations of log KOC within and between different labora- tories are rarely better than 0.2 log units it is unlikely that a better fit could be obtained without further refinement in the log KoC values.The driving force for the uptake of non-ionic organic compounds by soil is the relative ease of cavity formation in the wet organic matter compared to the same process in water supplemented by more favorable solute lone pair electron interactions with wet soil than with water.Dipole-type interactions (s constant) and solute hydrogen-bond base inter- actions (b constant) favour retention by water. Solute hydrogen- bond acid interactions are not significant in the uptake of organic compounds by wet soil (a constant is statistically insignificant).This is not to say that wet soil has no hydrogen- bond basicity, but rather that the hydrogen-bond basicity of wet soil and water are about the same. As observed in other distribution systems containing water the model reflects the influential role of the high cohesive energy of water and its strong hydrogen-bond acidity on the distribution process.The solvation parameter model, eqn. (3), provides an explanation for the family behaviour of other common correla- tion methods and their inability to accommodate a wide range of solute types. Those systems which are most useful as surrogate reference systems for estimating log KOC will have similar system constants to those given in eqn. (3) or, if the range of solute properties is restricted (compounds with a weak capacity for polar interactions), then similar values of the system constants for which the descriptors involved are numerically significant.The octanol-water distribution system (rn = 3.81, r = 0.56, s = -1.05, a = 0.03, and b = -3.46)22 is not a good model for the soil-water system, since wet octanol is sig- nificantly less cohesive than wet soil, less dipolar, and a significantly weaker hydrogen-bond acid.It cannot be expected that a single correlation equation between log Kow and log Koc will be found for a wide range of compounds of different polarity without inclusion of additional terms that reflect the difference in the capacity for dipole-type, and particularly,Analytical Communications, December 1996, Vol33 419 solute hydrogen-bond base interactions between the two systems. On the other hand the difference in system constants for wet soil and the isobutanol-water distribution system (rn = 2.76, r = 0.48, s = -0.64, a = -0.05, h = -2.28)22 is much smaller and this distribution system would be a much better model for predicting the behaviour of organic compounds in the soil-water system.There is good agreement between the system constants observed for retention in sodium taurode- oxycholate micelles in micellar electrokinetic chromatography (MEKC) (m = 2.48, r = 0.65, s = -0.46, a = 0.00, b = -2.07)23 and the soil-water distribution system, such that retention factors determined by MEKC could be correlated with log Koc . It has been recommended to use log kw, extrapolated from the retention factors measured in reversed-phase LC with binary mobile phases to zero organic solvent, as a surrogate parameter to correlate with log KoC.1O~ll Here there are two problems.A linear extrapolation from a high organic solvent composition has generally been used whereas actual measure- ments of the retention factor over a wide composition range indicates substantial curvature in the extrapolation for some compounds.This extrapolation error, which is often substantial, can only confound efforts to derive a correlation between log kw and log KOC.l6 Experimental values of log kw for different reversed-phase chromatographic systems would tend to suggest that silica-based sorbents are too weakly hydrogen-bond basic and generally too cohesive to be good surrogate models for the soil-water distribution system.12-'6 Given that the system constants for reversed-phase LC systems can be varied over a wide range by changing the composition of the mobile phase it would seem inevitable that a sorbent mobile phase combination could be identified to mimic the soil-water distribution system.However, it seems likely that these systems would contain a predominant amount of water, quite different to those used previously8-10 for estimating log Koc, or would need to be based on sorbents specially tailored to create the properties required to mimic the system constants for the soil-water system.It should be noted here that there is considerable variation of log KOC values for different soil types,' such as agricultural soils and sediments.Normalizing the soil-water distribution constant by reference to the percent organic carbon of the soil cannot completely account for the heterogeneous composition of the soil organic matter. The system constants given here will best fit the properties of mature agricultural soils. A further use of the solvation parameter model would be the characterization of different soil types by their capacity for different inter- molecular interactions. References 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 von Oepen, B., Kordel, W., and Klein, W., Chemosphere, 1991, 22, 285.Chiou, C. T., Porter, P. E., and Schmedding, D. W., Environ. Sci. Technol., 1983, 17, 227. Briggs, G. G., J . Agric. Food Chem., 1981, 29, 1050.Magee, P. S., Sci. Total Environ., 1991, 109, 155. Sabljic, A., Environ. Sci. Technol., 1987, 21, 358. Bahnick, D. A., and Doucette, W. J., Chemosphere, 1988, 17, 1703. von Oepen, B., Kordel, W., Klein, W., and Schuurmann, G., Sci. Total Environ., 1991, 109, 343. Vowles, P. D., and Mantoura, R. F. C., Chemosphere, 1987, 16, 109. Hodson, J., and Williams, N. A., Chemosphere, 1988, 17, 67.Szabo, G., Prosser, S. L., and Bulman, R. A., Chemosphere, 1990,21, 495. Hong, H., Wang, L., Han, S., and Zou, G., Chemosphere, 1996, 32, 343. Miller, K. G., and Poole, C. F., J . High Resolut. Chromatogr., 1994, 17, 125. Seibert, D. S., and Poole, C. F., Chromatographia, 1995, 41, 51. Poole, S. K., and Poole, C. F., Analyst, 1995, 120, 1733. Seibert, D. S., Poole, C. F., and Abraham, M. H., Analyst, 1996,121, 511. Poole, C. F., Poole, S. K., Seibert, D. S., and Chapman, C. M., J . Chromatogr. B, Biomed. Appl., in the press. Abraham, M. H., Chadha, H. S., and Leo, A. J., J. Chromatogr. A, 1994,685, 203. Abraham, M. H., and Roses, M., J . Phys. Org. Chem., 1994, 7, 672. Abraham, M. H., Roses, M., Poole, C. F., and Poole, S. K., J . Phys. Org. Chem., in the press. Abraham, M. H., Andovian-Haftvan, J., Whiting, G. S., Leo, A., and Taft, R. S., J . Chem. SOC., Perkin Trans. 2 , 1994, 1777. Abraham, M. H., Chadha, H. S., and Mitchell, R. C., J. Pharm. Pharmacol., 1995, 47, 8. Abraham, M. H., and Chada, H. S., in Lipophilicity in Drug Action and Toxicology, ed. Pliska, V., Testa, B., and van de Waterbeemd, H., VCH, Weinheim, 1966, p. 31 1-337. Poole S. K., and Poole, C. F., Analyst, 1997, 122, in the press. Paper 610691 2F Received September 9,1996 Accepted September 22, I996

ISSN:1359-7337    

DOI:10.1039/AC9963300417

出版商:RSC    

年代:1996

数据来源: RSC

摘要:

Analytical Communications, December 1996, Vol33 (421 4 2 4 ) 42 1 High Temperature Water Extraction Combined With Solid Phase Microextraction* Hiroyuki Daimon and Janusz Pawliszynt Department of Chemistry, Univei-sity of Waterloo, Waterloo, Ontario, Canadu N2L3G1 " Solid phase microextraction (SPME) was coupled with high temperature water extraction for the determination of non-polar semivolatile compounds in solid matrices.Two different SPME approaches, dynamic and static, are described. In the dynamic extraction technique, analytes leached by hot water from a matrix were collected in a vial and simultaneously extracted from the water with SPME fibre. The collection vial was cooled using a bath to avoid analyte loss by evaporation. The optimized method using water extraction and SPME allowed quantitation based on external calibration with polycyclic aromatic hydrocarbons (PAHs) extracted from spiked sand.In the static approach, the water extraction of solids was performed using a high pressure cell with an SPME fibre inserted in the vessel. During the extraction process analytes were released from the matrix and then partitioned into the fibre coating as the cell was cooling down.In this technique isotopically labeled analogs of target analytes were used to compensate for the partial re-adsorption of analytes on the solid matrix at lower temperatures. The effects of different experimental conditions on the static high temperature water-SPME fibre extraction of NIST certified reference material, urban air particulates, were investigated. For quantitation the SPME fibre was removed from the collection water or the high pressure cell, and was immediately transferred to a gas chromatograph injection port to analyse extracts without any clean-up or pre-concentration. Both methods enable the determination of PAHs in solid matrices without use of organic solvent.Static high temperature water-SPME has an additional advantage since it does not require a high pressure pumping system.Recently, high temperature water has been demonstrated to be a potentially useful analytical extraction fluid for a range of polar and non-polar organic chemicals in environmental solids. 192 However, several millilitres of organic solvent are used to collect the extracted compounds after the fluid exits from the extraction cell.In the past few years, a new extraction technique, solid phase microextraction (SPME), has been proved to be a relatively simple, time efficient method for sample preparation."-" The poly(dimethylsi1oxane) coating has shown itself to be very effective in extracting aromatic hydrocarbons and polycyclic aromatic hydrocarbons from water or soil .3,5-7 The water extraction was already applied to the extraction of polar analytes from solid matrices.For example, herbicides in a soil matrix have been extracted with a poly(acry1ate) coated fibre after the addition of water.8 Volatile non-polar analytes have * Paper presented in part at the 7th International Symposium on Supercritical Fluid Chromatography and Extraction, March 3 1-April 4, 1996, Indianapolis, IN, USA.+ To whom correspondence should be addressed. also been quantitatively extracted using the headspace SPME technique after addition of water and heating of the matrix.9 In this communication, SPME is used to quantify semivolat- ile non-polar analytes removed from a solid matrix by dynamic and static high temperature water extraction. SPME eliminates organic solvent in the collection step, which is typically used in the hot water extraction technique.1.2 The potential of high temperature water extraction-SPME-GC for the determination of PAHs was evaluated.In the dynamic approach, the sample, placed in an extraction cell, was constantly swept with fresh hot water. The removed analytes were collected in a vial, extracted by SPME fibre from water and then transferred to the analytical instrument for quan titation.In the static approach, the extraction was performed in a high pressure cell with an inserted SPME fibre. The hot water acts as an extraction solvent, facilitating the release of analytes from the solid matrix. SPME fibre, on the other hand, is used to isolate and concentrate the analytes simultaneously from water into the coating.The cool-down period is applied to partition analytes into the fibre coating more efficiently. Experimental Dynamic high temperature water extraction was performed in a procedure similar to conventional supercritical fluid extraction. An ISCO 260D syringe pump (ISCO, Lincoln, NE, USA) was operated in the constant pressure mode to supply water to the system.HPLC stainless steel columns (4.6 mm id, 100 mm length) were used for the extraction cell. A Varian (Mis- sissauga, Ontario, Canada) 6500 GC oven was used to control the extraction temperature. A standard stock solution containing naphthalene, anthracene, fluoranthene, pyrene and ben- zo[a]pyrene was prepared containing 50 pg ml-1 of each hydrocarbon in acetone. This stock solution was spiked to a desired concentration into the sand or water.One gram of the sand (purified by acid, 50-150 mesh, BDH Inc., Toronto, Canada) was loaded into the extraction cell. In order to study collection methods, the outlet of the extraction cell was connected to a flow restrictor constructed from a 30 pm id, 10 cm length of fused silica tubing (Polymicro Technologies, Phoenix, AZ, USA).The flow rate was 1.2 ml min-' at 300 atm (30 397.5 kPa). For the extraction of urban air particulates (SRM 1649, National Institute of Standards and Technology, Gaithersbug, MD, USA), a 50 pm id, 10 cm length, restrictor was used. This restrictor yielded a flow of 1.1 ml min-I at 50 atm. The flow rate was as liquid water measured at the pump.In order to investigate the collection efficiency after dynamic high temperature water extraction, 5 1-11 of standard solution was spiked onto sand loaded in an extraction cell. The water extraction was performed at 300 "C and 300 atm for 15 min. Fig. I shows a schematic diagram of the collection apparatus for dynamic high temperature water extraction. Collection of the extracted analytes was performed by inserting the outlet of the restrictor into a 40 ml glass vial containing 10 mi of water with a stirrer bar.All collection vials were silanized prior to use. The422 Analytical Communications, December 1996, VoE33 collection vial was cooled to less than 5 "C with a cooling bath to avoid analyte evaporation during the extraction. At the same time, a 30 pm poly(dimethylsi1oxane) fibre (Supelco Canada, Mississauga, Ontario, Canada) was exposed to the collection water with rapid stirring to extract the analytes.After the dynamic high temperature water extraction, the volume of collected water was adjusted to 30 ml by adding several ml of water. Then, the extracted analytes in the collected water were extracted with the SPME fibre for 70 min with rapid stirring. To establish an equivalent 100% recovery, an appropriate spike of the same standard mixture was spiked into 30 ml water in the 40 ml glass vial.After SPME the fibre was removed and inserted into a GC injection port, then thermally desorbed at 270 "C for 5 min. For static high temperature water extraction and simul- taneous extraction with SPME fibre, 50-100 mg of urban air particulates was placed in a 20 ml high pressure cell.The high pressure cell was constructed of stainless steel in the Science Machine Shop at the University of Waterloo (Fig. 2). Iso- topically labeled standards, 10[2H] fluoranthene, 10[2H] pyr- ene, 12[2H] benzo[a]pyrene (Cambridge Isotope Laboratories, Andover, MA, USA) were spiked as internal standards into the air particulates.An 18 ml volume of water was added to the cell. SPME was performed with the 30 pm poly(dimethylsi1oxane) fibre. A SLIPFREE fitting (Keystone Scientific, Bellefonte, PA, USA) was closed tightly to seal around the SPME needle. Static high temperature water extraction was performed for 120 rnin at different extraction temperatures. To heat the cell, a hot-plate (Series 400HPS, VWR Scientific, Ontario, Canada) was used.Extraction cell -1 Restrictor I / I Cap with septum I Stirrer plate I Fig. 1 dynamic high temperature water extraction. Schematic diagram of apparatus for collection method after Connector Fig. 2 temperature water extraction with simultaneous SPME. Schematic diagram of high pressure extraction cell for static high After static high temperature water extraction was completed, the cell was cooled down for 120 min by switching off the hot- plate.Once the extraction was completed, the fibre was withdrawn back into the needle, removed from the cell and analysed immediately with the GC-MS. A Varian 6000 gas chromatograph equipped a flame- ionization detector (FID) was used for the separation and analysis of all compounds, except for those in urban air particulates. A Supelco SPB-5 (30 m X 0.25 mm id, 0.25 pm film thickness) was used for separations.The oven temperature was initially set at 50 "C for 5 min; it was programmed at 15 "C min-1 to 310 "C and held for 10 min. The injector and detector temperatures were maintained at 270 "C and 300 "C, re- spectively. Quantitation of the spiked test compounds was based on GC-FID analysis.Analysis of extracts from urban air particulates by GC-MS was performed with a Varian Saturn gas chromatograph-ion trap mass spectrometer (Mississauga, On- tario, Canada). Results and Discussion SPME Following Dynamic High Temperature Water Extraction To characterize the performance of the 30 pm thick poly- (dimethylsiloxane) fibre coating for five PAHs collected in water, several parameters were studied: absorption time profile, linearity and reproducibility.The absorption time profiles for PAHs from directly spiked 30 ml (8.3 ppb) water onto a fibre coated with 30 pm poly(dimethylsi1oxane) were determined. The absorption time profile is a graph of absorbed amount as a function of exposure time.The exposure time must be long enough for equilibrium to occur or for the rate of absorption to have slowed in order to improve precision. Four PAHs (naphthalene, anthracene, fluoranthene and pyrene) reached equilibrium in 60 min. Although the equilibrium was not reached at 60 rnin for benzo[a]pyrene, the absorption time was set at 70 rnin for the subsequent experiments for practical convenience. The RSDs for three replicates of a 70 rnin SPME were very good, ranging from 4 to 9%.The main purpose of the initial investigation was to evaluate SPME after the water extraction and not necessarily to study the high temperature water extraction mechanism. Therefore, spiked sand was utilized to investigate the collection efficiency. The effects of water cooling and simultaneous dynamic high temperature water leaching-SPME on recoveries were investi- gated (Table 1).The amounts extracted corresponding to four different collection conditions were compared to that of directly spiked water. Without use of the cooling bath (Table 1, columns B and D), naphthalene and anthracene were lost during the water extrac- tion because of their volatility.In addition, the amount extracted of benzo[a]pyrene was higher compared with spiked water because the temperature of collected water rose to 40 OC, which increased the mass transfer of analytes between collected water and the fibre coating. Lower relative molecular mass com- pounds were not affected since they had already reached equilibrium. On the other hand, when water was cooled to 5 "C with the cooling bath (C,E), the amounts extracted of naph- thalene and anthracene were almost loo%, but the amount of benzo[a]pyrene extracted was very low without simultaneous extraction with SPME fibre (C).Extracted benzo[a]pyrene might have immediately precipitated from the collection water or adsorbed onto the vial walls. The importance of simultaneous SPME extraction of collection water with dynamic high temperature water extraction, as opposed to after the water extraction was completed, was less significant for the other four compounds.When SPME was performed simultaneously withAnalytical Communications, December 1996, Vol33 423 the water extraction and the cooling bath was used, the extraction amounts of all the tested compounds were good, with a range of 96-1 1 1 % compared with amounts from spiked water at room temperature.The extraction amounts also showed excellent reproducibility, with the relative standard deviations for all investigated analytes being less than 10% for triplicate spike extractions. In simultaneous water-SPME extraction (D,E), the total extraction time of SPME was actually 85 min (simultaneous with water extraction for 15 min and SPME extraction for 70 min).Therefore, to compare the extraction amounts between B,C and D,E, the extraction with SPME on B,C should have been 85 min after water extraction. The simultaneous extraction (D,E) increased the extraction time by SPME after the water extraction compared with B,C. The collected water was cooled to prevent evaporation of naphthalene during the water extraction, and to control the extracted amount of benzo[a]pyr- ene by SPME extraction. The results clearly indicate the importance of simultaneous water-SPME extraction and cool- ing of the system.As an application of the optimized collection method after dynamic high temperature water extraction, the extraction of PAHs from urban air particulates was performed for 15 min at 250 "C and 50 atm (Table 2).The concentration was estimated based on an external calibration curve corresponding to spiked water, and compared with the certified value supplied by NIST. Relative standard deviations were based on triplicate dynamic high temperature water extractions. Under the extraction conditions, water easily carried the particles from the extraction cell, through a 0.5 pm frit, to the restrictor. Therefore, restrictor plugging often occurred.In order to prevent this problem, the extraction cell was filled as followed: a filtering fibre (Pyrex wool), 0.2 g of sand, sample and more sand. The water extraction and optimized collection method provided concentrations above 85 % of the concentrations certified by NIST for fluoranthene and pyrene.Since the extraction conditions were not fully optimized, the estimated concentration of benzo[a]pyrene was lower, and the relative standard deviations were also higher, compared with the certified values supplied by NIST. One of the reasons for higher RSDs is associated with the presence of traces of acids originating from the acid washed sand used for prevention of restrictor plugging, which was detected in the extract and caused poor shapes of the PAH peaks.The results demonstrate that the present collection technique after dynamic high temperature water extraction is potentially useful since it does not use organic solvents and includes sample preparation and convenient introduction for subsequent analysis.Static High Temperature Water Extraction With Simultaneous SPME Usually, 5-15% of water content for 2 g of sample in a 4 ml vial will not cause significant pressure build-up as the sample vial is heated to over 100 "C. On the other hand, a very large amount of water added (> 50 %) into the sample produces a pressure increase after heating of the vessel, which will cause leaking of an ordinary vial design.9 Therefore, a specially constructed high pressure container, which allowed fibre insertion, was used in the static high temperature experiments. For the quantitative analysis of complex real-world samples, the use of isotopically labeled internal standard is a very effective and accurate quantitation method, since the internal standards and target compounds have very similar physio- ~~~~ ~ ~ Table 2 Quantitation of PAHs from Urban Air Particulates (NIST1649) with dynamic high temperature water extraction* Estimated concentration as % of certified Cert.conc./ concentration (%RSD): pg g-1 (%RSD) 250 "C, 50 atm Fluoranthene 7.1 (7) 134.0 (16) 87.5 (15) Pyrene 7.2 (7) Benzo[a]pyrene 2.9 (17) 72.0 (29) * Estimated concentrations versus certified values supplied by NIST.Relative standard deviations are based on triplicate dynamic high temperature water extractions. Table 3 Quantitation of PAHs from Urban Air Particulates (NIST 1649) with static high temperature water extraction* Estimated concentrations as % of certified concentration (% RSD) Cert. conc./ Pi? g-' 100mg, 50mg, 50mg, (%RSD) 270°C 270°C 300°C Fluoranthene 7.1 (7) 137 (6) 149 (8) 128 (2) Pyrene 7.2 (7) 113 (9) 121 (7) 105 (7) Benzo[a]pyrene 2.9 (17) 49 (14) 72 (10) 70 (4) * Estimated concentrations versus certified values supplied by NIST.Relative standard deviations are based on triplicate static high temperature water extractions and simultaneous extraction with SPME fibre at each condition. Table 1 Amount of PAHs extracted by SPME from collection water under different conditions after dynamic high temperature water extraction of spiked sand as % of amount extracted by SPME from spiked water (%RSD)* Direct No cooling, Cooling, No cooling, Cooling, SPME SPME SPME SPME SPME from after after simultaneous simultaneous spiked water water with water with water water extraction extraction extraction extraction A B C D E Naphthalene 100 (4) 19 (18) 110 (8) 19 (18) 96 (7) Anthracene 100 ( 5 ) 70 (11) 99 (10) 76 (8) 97 ( 5 ) Fluoranthene 100 (6) 101 (6) 107 (6) 102 (6) 111 (8) Pyrene 100 (9) 102 (6) 103 (6) 102 (10) 111 (8) Benzo [ a] pyrene 100 (4) 158 (29) 50 (8) 162 (18) 102 (7) * All experiments were performed in triplicate.Amount spiked to sand equals amount spiked to water.A, Extraction with SPME fibre was performed on 30 ml of spiked water, without high temperature water extraction. B, Collection device was without cooling bath and extraction with SPME fibre after the water extraction was complete. C, Collection device was with cooling bath and extraction with SPME fibre after the water extraction was complete. D, Collection device was without cooling bath and simultaneous extraction with SPME fibre.For B, C, D and E, spiking was performed on sand placed in extraction cell. Extraction was performed at 300 "C, 300 atm and 1.2 ml min-1 for 15 min.424 Analytical Communications, December 1996, Vol33 chemical properties. This approach to calibration was used for static hot water extraction combined with simultaneous SPME.The urban air particulates were heated to 270 and 300 "C for 120 min in the presence of water. The analytes released were extracted with SPME fibre while the cell was cooling down for 120 min for the determination of the concentration levels of PAHs. Table 3 shows the effects of different experimental conditions on the relative amount extracted from the air particulates by the static hot water-SPME technique.Concen- trations were estimated based on internal standards, and compared with the certified value supplied by NIST. For the PAHs tested, this method gave good quantitative results. The estimated concentration of benzo[a]pyrene in- creased with decreasing amount of sample. Increasing the extraction temperature to 300 "C did not increase the estimated concentrations significantly.The results indicated that 270 "C and 50 mg were the best conditions for the quantitation of PAHs with this method. The lower temperature is preferable for the method because 300 "C is close to the operational limit temperatures of the fibre and the O-ring seal of the cell. The above results demonstrate that the hot water-SPME method can be successfully applied to the determination of non- polar semivolatile compounds present in solid samples. The advantage of a dynamic technique is associated with simple external calibration and fibre protection, since the coating is exposed to relatively pure extraction water under low tem- perature conditions.Static high temperature water extraction with simultaneous SPME technique, on the other hand, uses only simple apparatus and procedure, and extends the SPME technique for analytes tightly bound to soils. Additional investigations of the extraction time and agitation are expected to improve the estimated concentration of benzo[a]pyrene. The financial support of Supelco, Varian and the Natural Sciences and Engineering Research Council of Canada is gratefully acknowledged. References Hawthorne, S. B., Yang, Y., and Miller, D. J., Anal. Chem., 1994,66, 2912. Yang, Y., Bowadt, S., Hawthorne, S. H., and Miller, D. J., Anal. Chem., 1995, 67,4571. Arthur, C. L., and Pawliszyn, J., Anal. Chem., 1992, 64, 1960. Louch, D., Motlagh, S., and Pawliszyn, J., Anal. Chem., 1992, 64, 1187. Zhang, Z., and Pawliszyn, J., Anal. Chenz., 1993, 65, 1843. Zhang, Z., and Pawliszyn, J., J. High Resolut. Chromatogr., 1993,16, 689. Potter, D. W., and Pawliszyn, J., Environ. Sci. Technol., 1994, 28, 298. Boyd-Boland, A. A., and Pawliszyn, J., J . Chrornutogr. A, 1995,704, 163. Zhang, Z., and Pawliszyn, J., Anal. Chem., 1995, 67, 34. Paper 6/06307A Received September 13,1996 Accepted October 22, 1996

ISSN:1359-7337    

DOI:10.1039/AC9963300421

出版商:RSC    

年代:1996

数据来源: RSC

摘要:

Analytical Communications, December 1996, Vol33 (42.5427) 425 Capillary Zone Electrophoresis of Human Recombinant Erythropoietin Using C8 Coated Columns Without Additives in the Running Buffer Oscar Nieto,* Pedro Hernandez and Lucas Hernandez? Department of Analytical Chemistry and Instrumental Analysis, Faculty of Sciences, Universidud Autbnoma de Madrid, E-28049 Madrid, Spain The use of C8 coated capillary columns enabled the characterization of human recombinant erythropoietin (r-HuEPO) as well as its determination by capillary electrophoresis. Samples were injected at 90 V cm-l during a period of 30 s and the electrophoregrams were recorded at 210 V cm-l.A 0.03 moll-' phosphate solution at pH 7.0 (without additives) was used as the running buffer. A minimum amount of 19 k 1 pg, which corresponds to a concentration in the sample of 0.37 pg ml-1, was found using UV detection at 240 nm.Several experiments were carried out in order to study the behaviour of r-HuEPO in this type of capillary. r-HuEPO was determined in spiked urine samples. The gl ycoprotein hormone erythropoietin (EPO) controls blood oxygen levels via the number of circulating red blood cells.The peptide chain of the hormone bears 166 amino acid residues in mature erythropoietin. The hormone is produced in the liver and kidney of adult mammals and in the liver of foetal and new-born mammals. The cells which respond to erythropoietin in the bone marrow and the spleen of adults as well as foetal liver have been identified. In erythropoietin progenitor cell cultivates, this hormone stimulates proliferation and differentiation of a major number of red blood cells.1-5 A few years after erythropoietin was discovered, the gene of erythropoietin was cloned.6 Recently, the microheterogeneity of human recombinant ery- thropoietin (r-HuEPO) has been characterized by electrospray ionisation mass spectrometry.7 In this work, 52 different structures due to the wide variety of oligosaccharide chains with sialic acid as their terminal moiety have been characterized.The determination of r-HuEPO in blood, and especially in urine (the biological fluid of choice for anti-drugs control), is of great interest for controlling drug abuse in sports events.*-IO Elite athletes have been shown to exhibit increased EPO blood levels that change throughout a day and peak at the end of a training season.11 The high variability of EPO levels in urine and the lack of reliable criteria for determining the origin of the hormone (whether endogenous or externally administrated) preclude reliable conclusions in relation to sports drugs tests, so a dependable method for determining the hormone is much needed.Capillary zone electrophoresis is a separation technique which is characterized by the large number of theoretical plates and the excellent efficiency in the separations which are obtained.12 This technique is very suitable for the separation of peptides and proteins.13,14 At least four different isolation * Present address: European Commission, Institute for Reference Materials and Measurements, Retieseweg, B-2440 Geel, Belgium. t To whom correspondence should be addressed.techniques are used: ion-exchange chromatography; reversed- phase chromatography; gel filtration; and affinity chromat- ography.15 Several papers have been published on the determination of r-HuEPO by capillary electrophoresis. 16,17 In these publica- tions, the phenomenon of microheterogeneity due to a different charge-to-mass ratio, as a consequence of the different length of the carbohydrate chains bound to terminal sialic acid residues, is shown.Hence, the electrophoregrams exhibit a succession of peaks which correspond to the putative isoforms of the glycoprotein. Tran et a1.16 achieved very good resolution of the different glycoforms using 0.1 mol 1- 1 phosphate-acetate buffer at pH 4.0 as the running buffer, after a 10 h equilibration time.Watson and Ya017 used 0.01 mol 1-1 tricine-0.01 moll-' NaC1-2.5 mmol l-1 diaminobutane-7.0 moll-1 urea at pH 6.2 as the running buffer. After a treatment of r-HuEPO with neuraminidase, the sialic acid residues of r-HuEPO were released and the resolution of the different glycoforms was improved.In another recently published work,lX the efficiency of the use of alpha-, omega-bis-quaternary ammonium alkanes as additives in the running buffer have been tested for the separation of several glycoproteins, improving the results obtained with the use of 1,4-diaminobutane. Likewise, Benedek and Thiedelg obtained the calibration plots of glycoproteins by Fergusson analysis when capillary electrophoresis was carried out with poly(ethy1ene oxide) coated columns in the presence of sodium dodecyl sulfate.In this work a Cg coated capillary column was chosen since a simple method of the separation was required, and the use of additives in the running buffer, which could affect the stability of the glycoprotein and the preservation of standards, was avoided.The use of C8 coated columns, a reversed-phase silica which is very suitable for the isolation of hydrophobic species in aqueous medium when less retention than C18 is required, avoids the adsorption of positively charged residues of the silanol groups on the inner walls of the capillary at pH values above 2.0. Therefore, as is shown in this communication, the use of additives in the running buffer is not necessary as is the case for many separations of organic compounds by capillary electrophoresis with uncoated columns.Experimental Apparatus The capillary electrophoresis equipment consisted of a Euro- phore (Toulouse, France) model Prime Vision I manual injector which allowed samples to be injected electrokinetically. A Europhor model Prime Vision V energy source was connected to the manual injector enabling a difference of potentials of up to 30 kV to be applied.A Europhor model Prime Vision 11 UV detector with a D2 lamp was used.426 Analytical Communications, December 1996, Vol33 A Cg coated capillary column CElect-H150 from Supelco (Bellefonte, PA, USA) with an internal diameter of 75 pm and an external diameter of 363 pm was used.The length of the column was 100.0 cm with an effective length of 65.0 cm. The electrophoregrams were recorded using a Varian (Wal- nut Creek, CA, USA) model 4290, integrator. A pH meter Metrohm (Herisau, Switzerland) model 645 equipped with a glass combined Ag/AgCl electrode was used. Reagents Recombinant human erythropoietin, r-HuEPO, from Sigma (St. Louis, MO, USA) was reconstituted with 1.00 ml of water to obtain a solution containing 10 hormone biological activity units (100 000 units per milligram of protein), 0.1 moll- 1 NaCl, 0.01 mol 1-l phosphate buffer at pH 7.0 and 0.1 mg ml-l lactose.The r-HuEPO was used without further purification. The corresponding amount of phosphate buffer was added to this solution in order to obtain a sample solution with a phosphate concentration equal to that in the running buffer, without any modification of the final concentration of r-HuEPO.Clg Sep-Pack cartridges from Millipore (Bedford, MA, USA) were used to separate the substances in urine that could be adsorbed on the inner walls of the capillary. All the reagents were of analytical grade quality and the water was de-ionized by a Milliro-MilliQ system from Waters (Milford, MA, USA).The dissolved oxygen in the running buffer and in the sample solutions was removed by vacuum with continuous stirring for 20 min. Afterwards, the solution was filtered using Nylon filters from DynaGard (Laguna Hills, CA, USA) (0.45 pm pore size). Procedure The capillary column was flushed by applying a vacuum in the cathodic extreme of the capillary for 5 min with each one of the following solvents and solutions: 0.1 moll- NaOH, H20, 0.I mol 1-1 H3P04, H20, methanol and HzO. Then, the capillary was filled with the running buffer and was equilibrated by applying the running potential for 10 min. The sample was electrokinetically injected at 9.0 kV for 30 s in the cathodic extreme of the capillary and the electrophoregrams were recorded, if nothing else was noticed, at a potential of 21 .O kV.The wavelength of the detector was set at 240 nm, at which the best S/N was obtained. The electro-osmotic flow was evaluated by measuring the time it took a phenol sample to pass through the effective length of the capillary column under the given experimental condi- tions. To a series of freshly collected urine samples, the corre- sponding amount of phosphate buffer at pH 7.0 was added in order to obtain a final concentration of 0.03 mol 1-l, and increasing amounts of r-HuEPO from 0.00 to 5.00 mg ml-l were added.The r-HuEPO spiked urine samples were passed through a pre-activated Clg cartridge, removing some sub- stances that could be adsorbed on the inner capillary walls.The electrophoregrams were recorded as described above. Results and Discussion Fig. 1 shows the electrophoregram of 1.08 pg ml-I r-HuEPO recorded at a potential of 21.0 kV by using 0.03 mol 1-1 phosphate at pH 7.0 as the running buffer. The sample was electrokinetically injected at 9.0 kV over a period of 30 s. The electrophoregram exhibits a succession of sharp peaks which are due to the different isoforms of the hormone.This microheterogeneity phenomenon has previously been obser- ved1c19 and is due to the variety of oligosaccharide chains with sialic acid groups bound to the extremes. The electrophoregram of 1.0 pg ml-1 phenol, recorded under the same experimental conditions, exhibits a peak at a migration time of 10.1 min whereas the migration times of peaks attributed to r-HuEPO are between 1 1.1 and 1 1.7 min.This indicates that the r-HuEPO is negatively charged at pH 7.0 and migrates in the opposite direction to the electro-osmotic flow. The use of low concentration running buffers is required for working with these capillary columns. Some interaction between the hormone and the reversed-phase coating on the inner walls of the capillary occurs when the running buffer concentration increases.In fact, the use of running buffers with concentrations from 0.01 to 0.10 moll-' results in an increase in the temperature from 20 to 35 "C. As a result of this increase in temperature, the migration time of each peak increases without any improvement in their resolution. The linear relationship of the efficiencies versus the neperian logarithm of running buffer concentration is adjusted to the following equation: l/t = 1.9 (f0.2) X 10-3-2.7 X In C (r2 = 0.992) where C is the running buffer concentration and is given in mmol 1-l.For further experiments, a 0.03 mol 1-l phosphate buffer at pH 7.0 was chosen as the most suitable running buffer. The migration time of each peak decreases when the applied potential increases and linear relationships of the efficiencies, l/t, versus the applied electric field are obtained.The values of slopes are 1.06 k 0.04 V cm-1 s-1 (r2 = 0.995) from the first to the last observed peak. Therefore, the values of the electrophoretic mobility coefficients of r-HuEPO, are in the range of p = 6.86 k 0.05 cm2 V-l s-I when using Cg coated capillary columns with an internal diameter of 75 pm and 0.03 mol 1-1 phosphate at pH 7.0 as the running buffer.The electro-osmotic flow also increases when the applied electric field increases and a linear relationship of the efficiency versus the applied electric field is also obtained. From the value of the slope, the electro-osmotic coefficient of posm = 6.4 cm2 V-1 s-1 has been estimated.The peak resolution is improved when the applied potential increases, but the optimum S/N is achieved at a potential of 21 .O kV, which is the potential applied for further experiments. When the electrokinetic injection is carried out at different injection times and the injection potential is maintained at 9.0 kV, the peak heights increase linearly up to an injection time of 30 s.Thereafter, no variation in peak heights is observed and the peak widths increase until all peaks overlap. A similar phenomenon is observed when the electrokinetic injection is carried out by maintaining the injection time at 30 s and by 50 pUA I 0.0 2.0 4.0 6.0 8.0 10.0 12.0 14.0 Ti me/min Fig. 1 Electrophoregram of 1.08 pg ml-1 r-HuEPO recorded at 21.0 kV using 0.03 mol 1-1 phosphate buffer at pH 7.0 as the running buffer.Injection time 30 s at 9.0 kV. UV detection at 240 nm.Analytical Communications, December 1996, Vol33 427 increasing the injection potential. From the values of the electrophoretic mobility coefficients for the putative isoforms previously estimated, the total amount of substance injected in the capillary column is Q = 53 f 2 pg.The sum of all peak areas versus concentration shows a linear relationship in the range of concentrations between 0.00 and 5.00 pg ml-1, which correspond to amounts of substance injected between 0 and 245 pg, respectively. This linear relationship is adjusted by least squares to the following equation: A = 23 + 740 Q (r2 = 0.998) where A is the sum of all peak areas, given in pUA s- l , and Q is the amount of injected substance given in pg.The minimum amount of substance that can be detected was calculated using the method of Knoll.20 Thus, the minimum amount detected is 19 k 1 pg which corresponds to a concentration in the sample solution of 0.37 pg ml-1 for the injection conditions. This was tested for a solution of the same concentration in which the signal due to erythropoietin was observed, but no signal was obtained for less concentrated solutions.Once the urine samples spiked with r-HuEPO have been prepared as indicated in the experimental section, the samples are injected at 9.0 kV for 30 s and the electrophoregrams are recorded at 21.0 kV using 0.03 mol 1-1 phosphate at pH 7.0 as the running buffer.Fig. 2 shows the electrophoregrams of the urine sample A, before and B, after the addition of r-HuEPO in order to obtain a concentration in urine of 5.00 pg ml-l. The peaks attributed to r-HuEPO are similar to those obtained in the electrophoregram in Fig. 1. The migration time of all the peaks are in the range of 10.9 to 1 1.6 min and the peak at a migration time of 10.1 min corresponds to the substances migrating with the electro-osmotic flow.The percentage of recovery was monitored by cathodic stripping voltammetry21 and was 95.2%. A linear relationship between the sum of peak areas and the amount of injected substance was adjusted by least squares and the following equation was obtained: eof r-Hu EPO 0.0 2.0 4.0 6.0 8.0 10.0 12.0 14.0 16.0 18.0 20.0 Time/min Fig.2 Electrophoregram of a urine sample containing 0.03 mol 1-I of phosphate buffer at pH 7.0 A, before and B, after spiking with r-HuEPO at a concentration of 5.00 pg ml-l. The electrophoregram was recorded as in Fig. 1. A = 46.0 + 685 Q (r2 = 0.997) where A is the peak areas, given in pUA s-1, and Q is the amount of injected substance in the column given in pg.The natural concentration of EPO contained in the urine sample is under the limit of detection. Hence, the use of a detection system more sensitive than UV spectrophotometry for determination of EPO in urine samples using capillary electrophoresis is suggested. In our team, an electrochemical method for detection and determination of r-HuEPO in urine samples was developed in which a detection limit of 0.059 ng ml-1 was achieved.21 Therefore, future work will involve the implementation of a suitable method for the determination of EPO by capillary electrophoresis using electrochemical detection.The authors thank the CICyT (I + D project no. Dep 103/9 CSD (Spain) for their financial support. References ) and 1 2 3 4 5 6 7 8 9 10 1 1 12 13 14 15 16 17 18 19 20 21 Jacobson, L.O., Goldwasser, E., Fried, W., and Pizak, L. F., Trans. Assoc. Am. Physicians 1957, 10, 305. Hammond, D., and Winnick, S., Ann. N . Y . Acad. Sci. 1974, 230, 219. Sherwood, J. B., and Goldwasser, E., Endocrinology (Baltimore), 1978,103, 866. Miyake, T., Kung, C. K. H., and Goldwasser, E., J . Biol. Chem., 1977, 252( 1 3 , 5558. Wieczorek, L., Hirth, P., Shope, K.B., Scigalla, P., and Kriiger, D., in Molecular Biology of Erithropoietin. Reprint translated from Innovative Aspekte der Klinishen Medezin, ed. Gurland, H., Springer- Verlag, Berlin, 1989. Jacobs, K., Shoemaker, C . , Rudesdorf, R., Neill, S. D., Kaufman, R. J., Mufson, A., Seehra, J., Jones, S. S., Hewick, R., Fritsch, E. F., Kawakita, M., Shimuzu, T., and Miyake, T., Nature (London), 1985, 313, 806.Rush, R. S., Derby, P. L., Smith, D. M., Merry, C . , Rogers, G., Rohde, M. F., and Katta, V., Anal. Chem., 1995, 67(8), 1442. Gareau, R., Audran, M., Baynes, R. D., Flowers, C. H., Duvallet, A., Senecal, L., and Brisson, G. R., Nature (London), 1996, 380, 113. Smith, A. D., and Perry, J. P., Ann. Pharmacother., 1992, 26(5), 653. Badia, R., de la Torre, R., and Segura, J., Biol. Clin. Haematol., 1992, 14(3), 177. Ricci, G., Masotti, M., De Paoli Vitali, E., Vedovato, M., and Zanotti, G., Acta Haematol., 1988, 80(2), 95-8. Jorgenson, J. W., and Lucaks, K. D., Anal. Chem., 1981, 53, 1298. Grossman, P. D., Colburn, J. C . , Lauer, H. H., Nielsen, R. G., Riggin, R. M., Sittainpalain, G. S., and Rickard, E. C., Anal. Chem., 1989,61, 1186. Cohen, A. S., and Karger, B. L., J . Chromatogr., 1987, 397, 409. Newton, P., LC-GC, 1990, 8, 1 16. Tran, A. D., Park, S., Lisi, P. J., Huynh, 0. T., Ryall, R. R., and Lane, P. A., J . Chromatogr. 1991, 542, 459. Watson, E., and Yao, F., Anal. Biochem., 1993, 210(2), 389. Oda, R. P., Madden, B. J., Spelsberg, T. C . , and Landers, J. P., J . Chromatogr., 1994, 680( l), 85. Benedek, K., and Thiede, S., J . Chrornatogr., 1994, 676( I), 209. Knoll, J. E., J . Chromatogr. Sci., 1985, 23, 422. Hemindez, L., Nieto, O., and Hemindez, P., Anal. Chim. Acta., 1995, 305, 340. Paper 6106997E Received October 14, 1996 Accepted November 7,1996

ISSN:1359-7337    

DOI:10.1039/AC9963300425

出版商:RSC    

年代:1996

数据来源: RSC

摘要:

Analytical Communications, December 1996, Vol33 (429432) 429 Determination of Trace Amounts of Alcohols and Phenols in Complex Mixtures as Ferrocenecarboxylic Acid Esters With Gas Chromatography-Atomic Emission Detection Jiirgen Rolfes and Jan T. Andemson* Ahteilung Analytische Chemie, Annrganisch-chemisches Institut, Universitat Miinster, Wilhelm-Klemm-Strasse 8, 0-481 49 Miinster, Germany. E-mail: anderss@uni-muenster.de A method for the derivatization of phenols and alcohols with ferrocenecarboxylic acid chloride to the corresponding esters is described.The esters are separated by capillary GC and detected by atomic emission in the iron-selective mode, thus ensuring detection limits of 150 fmol injected and high selectivity over non-iron-containing compounds.A crude oil (Brent, North Sea) is used as an example, without any sample preparation before the derivatization, and is shown to contain approximately 1 pmol phenols g-1. Experimental Reagents All chemicals came from commercial suppliers. The purity of the compounds used for the preparation of standards was > 97% (manufacturer's data). Aluminium oxide (neutral) was stored for at least 24 h at 155 "C.All solvents were purified by percolation through aluminium oxide. Screw caps G-8 and Teflon septa G-8-PTFE were from Chromatographie Service (Langerwehe, Germany). The atomic emission detector (AED) allows the selective and simultaneous GC detection of a variety of elements. Many applications have been published since its commercial intro- duction in 1989, including attempts to determine the elemental ratios of eluting compounds.For detection of some metallic elements such as mercury and iron the AED offers excellent selectivity and sensitivity. Therefore metal-containing deriva- tization reagents should be useful for the highly selective detection of such chemical classes that can be derivatized. This possibility does not seem to have been explored before. In this work we utilize the AED's high selectivity and sensitivity in the iron selective detection mode at 302 nm for the determination of alcohols and phenols by converting them to their ferrocene- carboxylic acid esters.Each hydroxyl group is labelled with one iron atom. The advantages of this method are the great selectivity (3.5 X loh i'ersus carbon, ref.2) and the low detection limit for iron (150 fmol injected as determined in this work), so that even trace amounts of alcohols and phenols can be determined in complex matrices. In principle, an oxygen-selective detector, e.g., the oxygen flame ionization detector or the AED in the oxygen-selective mode [selectivity for oxygen versus carbon of lo6 (ref. 3) and 25 000 (ref.2) respectively], might also be employed for the detection of alcohols and phenols. However, it would not be possible to differentiate between phenols/alcohols and other oxygen-containing compounds. The higher selectivity displayed by the AED for metals compared to the selectivity for oxygen is also a point in favour of the metal-containing derivatives. Furthermore, in this preliminary communication it is shown that through correct choice of reaction conditions, it should be possible to distinguish between phenols and alcohols in crude oil samples.* To whom correspondence should be addressed. Instrumentation A Hewlett Packard (Avondale, PA, USA) HP 5980 I1 gas chromatograph with a split/splitless injector, an HP 5921 A (AED) and a CTC A200 SE autosampler (Chromtech, Idstein, Germany) were used.The microwave oven was a household Samsung Compact RE 2000 (Samsung Electronics, Steinbach, Germany) with 500 W power. Gas Chromatography A DB-5 column (23 m x 320 pm id, 0.25 pm film thickness, J + W, Cologne, Germany) with a deactivated retention gap (3 m X 320 pm, J + W) was used. Helium carrier gas was at 72 kPa constant pressure, resulting in a linear velocity of 38 cm s-l at 80 "C.Injector and detector temperatures were 300 and 350 "C, respectively. Makeup gas flow (measured at cavity vent, with window purge 'off' and ferrule purge vent uncapped) was 44 ml min-1 for the carbon-selective mode (193 nm) and 140 ml min-1 for the iron-selective mode (302 nm). Cavity pressure was 10 kPa. The temperature programme was: 80 "C for 2 min, 30 "C min-1 to 230 OC, 5 "C min-1 to 350 "C (held for 10 min) .Preparation Ferrocenecarboxylic acid chloride (FCC) was prepared accord- ing to a modification of a previously published pr~cedure.~ At room temperature 750 p1 of oxalyl chloride was added by a microlitre syringe to a stirred suspension of 1.50 g (6.5 mmol) of ferrocenecarboxylic acid and a small amount of 4-dimethyla- minopyridine as catalyst in 70 ml of dry benzene.A Dimroth condenser with a bubbler was attached to the flask and the reaction mixture was stirred for 20 min at room temperature. The colour changed from orange to a dark red and small bubbles of CO and COz could be observed. To complete the reaction, the mixture was heated for 3 min at the boiling point. The benzene was evaporated under slight vacuum at 55-65 "C and the residue was extracted with 3 x 10 ml of dry pentane, which was removed by aspiration with a Pasteur pipette.The combined430 Analytical Communications, December 1996, Vol33 extracts were filtered through a glass frit, which was washed with pentane (10 ml). The solution was concentrated to 20 ml and stored at -5 "C overnight. The FCC precipitated as dark red, cubic crystals.After removal of the pentane layer (Pasteur pipette) the product was dried under vacuum and was used for derivatization without further purification. The yield was 1.10 g (68%). Preparation of standard derivative A 4.5 mg (70 pmol) portion of 2-fluoroethanol and 17 mg (90 pmol) of FCC were stirred for 3 h at 50 "C in 1 ml of pyridine. The reaction mixture was poured into 15 ml of 2 mol 1-1 HC1.The aqueous phase was extracted with 1 X 5 ml and 2 X 1 ml of dichloromethane. The combined organic layers were washed with 2 ml of water, dried over molecular sieves and evaporated to dryness. The residue was dissolved in dichloromethane and purified by passing through a column (10 cm X 1 cm id) of aluminium oxide.Evaluation of reaction conditions Stock solutions each containing 1 mg ml-1 of (2-fluoroethy1)- ferrocenecarboxylic acid ester (FEE), pentan- 1-01 ( 1 -Pent), octan-2-01 (2-0ct), 2-ethylcyclohexanol (EHex), decan- 1-01 ( 1 -Dec), diphenylmethanol (DPM), 1 -phenylpropan-2-01 (P- Pr), 2,4,6-trichlorophenol (TCP) and 4-nitrophenol (4-NP) were separately prepared in toluene.They were combined to give a mixture containing 100 ng pl-1 of each alcohol and the internal standard FEE. The following esterification procedure was a modification of a fluorescence derivatization for HPLC pub- lished elsewhere.5 A 10 pl portion of the sample mixture containing 1 pg of each alcohol was transferred into a small screw-capped vial with two Teflon septa. A 200 pl portion of FCC in dichloromethane (0.085 mol 1-1) and varying amounts of a solution of (4-dimethy1amino)pyridine (DMAP) in di- chloromethane (0.165 moll-') were added.The reaction vessel was tightened securely and heated for 60 s (four times, interrupted by pauses of 1 min) in the microwave oven. After the first minute the screw caps had to be retightened to provide better sealing against the pressure in the vessels.(Despite the high vapour pressure of dichloromethane, no problems were encountered with loss of solvent.) The vials were allowed to cool down and the DMAP and the excess of FCC were removed through separation on a microcolumn (30 mm X 5 mm id of aluminium oxide). The ferrocenecarboxylic acid esters were eluted with 3.5 ml of dichloromethane.The microcolumn was easily prepared from a Pasteur pipette, by inserting a piece of glass wool into the pipette and then heating it in a lighter flame and extruding its tip to form a small capillary. The dichloro- methane was evaporated at 40 "C by a gentle stream of nitrogen to about 1 ml, then 400 pl toluene were added and the volume was reduced to 350 p1. Toluene was added to a final value of 1000 p1.One pl of this solution was injected into the gas chromatograph. Preparation of crude oil samples A 108 mg portion of Brent crude oil (North Sea) was diluted with 200 pl of toluene and spiked with 30 p1 of the FEE stock solution as internal standard. Samples of 33 pl were treated as described above with either 10 pl of DMAP for derivatization of phenols or 300 pl for phenols and alcohols.Results and Discussion The effect of the catalyst concentration on the esterification of alcohols and phenols with FCC (see Fig. 1) was studied. Table 1 shows the conversion to ester of the phenols and the primary and secondary alcohols as a function of the ratio of catalyst to the acid chloride. For reactions without catalyst only traces of esters were formed.If a ratio of DMAP : acid chloride of between 0.1 : 1 and 0.5 : 1 was used, the phenols (2,4,5-trichlorophenol, 4-ni- trophenol) gave the ferrocenyl esters in yields of 75-loo%, but the alcohols were not esterified. An excess of catalyst (DMAP : acid chloride, 2 : 1) provided a yield of around 84% for the primary alcohols (pentan- 1-01 and decan- 1 -ol), but only 1 4 0 % for the secondary alcohols (2-ethylcyclohexanol, 1 -phenylpropan-2-01, octan-2-01). A further increase of the DMAP concentration only slightly improved the yields.These results indicate that it should be possible to differen- tiate between phenols and primary and secondary alcohols in a sample. To test the method on a 'real' sample, a crude oil was treated at two different ratios of FCC and catalyst.Fig. 2(a) shows a chromatogram obtained for the separation of the ferrocenecarboxylic acid esters of the phenolic compounds which were selectively esterified at the low ratio of 0.1 : 1 of DMAP to acid chloride. The amount of phenols present can be estimated to = 1 pmol g-1 of the crude oil, calculated with respect to the peak area of the internal standard.The two large peaks at 7.1 min and 7.4 min correspond to the methyl- and ethylesters of FCA and are ubiquitous contaminants, also present in blank runs. So far we have only identified the phenol ester (co-elution with authentic standard). In Fig. 2(b) the same sample was esterified with an excess of 3 : 1 DMAP : acid chloride. All peaks in this chromatogram that do not occur in Fig.2(a) are assumed to be primary and secondary alcohols. Their total amount is about 7 pmol g-1 of crude oil. There is such a large number of alcoholic compounds that not all of them are resolved in the chromatogram. The high selectivity of the method is emphasized in Fig. 2(c) which DMAP vCooR + HO-R CH,CI,HCI + L q L C O C l FCC Fig. 1 to yield the corresponding ester. Reaction of ferrocenecarboxylic acid chloride (FCC) with alcohol ~ ~ ~~ ~ Table 1 Yield of ester with FCC in a mixture of alcohols and phenols with different ratios of DMAP (values from two independent esterifications, abbreviations in the text) DMAP : FCC 1 -Pent 1 -Dec 2-EHex P-Pr 2-0ct DPM TCP NP 0.0: 1 5 5 5 4 1 1 0 2 2 2 0 0 0 0 0 0 0.1 : 1 0 1 0 0 1 1 0 0 1 1 0 0 94 9 1 7 5 7 4 0.5: 1 2 4 2 4 1 1 0 0 1 1 5 4 1 0 2 1 0 4 7 2 8 1 1 : l 14 9 13 9 4 3 4 4 2 2 34 32 100 99 7 5 7 5 2: 1 84 81 90 82 14 13 37 32 25 22 59 57 111 107 77 72 4: 1 89 83 97 90 19 15 47 37 34 27 5x 51 96 79 72 60Analytical Communications, December 1996, Vol33 43 1 shows the carbon selective chromatogram.It consists mainly of unresolved compounds with an overlaid pattern of n-alkanes.The largest alkanes of the chromatogram correspond to 50 ng per peak, about four orders of magnitude more than the limit of detection for the alcohols. Despite this large amount of alkanes, the selectivity of the detection for iron was high enough so that iron-containing compounds could be detected without spectral interferences. The reagent FCC can be readily prepared from the commer- cially available acid.The ferrocene esters showed excellent chromatographic behaviour and their peaks displayed no broadening or tailing. Initial investigations with ferrocene as an internal standard failed because ferrocene was formed through decarboxylation of FCC under the reaction conditions. Hence, ferrocenecar- boxylic acid 2-fluoroethylester was used as the internal standard.The peak of the fluoroethylester was well resolved from the ethyl- and the isopropylester, and there seem to be no other coeluting compounds with iron-selective detection. The stability and recovery of the esters under the reaction conditions was investigated for three ferrocenecarboxylic acid esters which were taken through the complete procedure. All esters showed recoveries higher than 96%.An advantage of the method is that sample preparation is very simple. Such a complex sample as a crude oil can be used directly. The sample preparation is fast and takes only 20 to 25 min per sample due to the rapid reaction made possible by microwave heating. Conventional heating in a metal block led to considerably longer reaction times and much larger variations in yield.However, the volatility of the esters is low. With the temperature programme used here only up to C34 alcohols can be eluted. For the same reason, dihydroxy compounds probably are not accessible with this method. Another drawback is the restriction to analytes without amino functions which would be retained in the microcolumn separation on aluminium oxide. In spite of these disadvantages the derivatization with FCC is a powerful method for the analysis of small quantities of phenols and alcohols in difficult matrices and it seems to be possible to use the method to distinguish between these two classes of compounds.The influence of other reaction conditions and the derivati- zation of other hydroxyl compounds and other matrices are under active study and will be the subject of a later paper, as well the identification of as many as possible of the derivatized hydroxyl compounds in the crude oil.We thank S. Bobinger for developing software for the editing of AED chromatograms. 350 'C 10 IS 20 25 30 35 (b) 1201 I 7 80 v c o - 0 8 Fc (302 nm) a27 4-71 Catalyst : acid chloride 3 : 1 C (1 93 nm) Catplyst : acid chloride 3 : 1 11111 10 15 20 2s 30 35 Time/min Fig. 2 Iron selective (a) and (b) and carbon selective (c) chromatograms of a crude oil (Brent, North Sea) esterified with FCC.432 Analytical Communications, December 1996, Vol33 References 5 Karlsson, K.-E., Wiesler, D., and Novotny, M., Anal. Chem., 1985, 57, 299. 1 Quimby, B. D., and Sullivan, J. J., Anal. Chem., 1989, 62, 1027. 2 Uden, P. C., in Selective Detectors, ed. Sievers, R. E., Wiley, New York, 1995, pp. 143-170. 3 Gokeler, U. K., in Selectice Detectors, ed. Sievers, R. E., Wiley, New York, 1995, pp. 99-126. 4 Burkhardt, E. R., Doney, J. J., Bergman, R. G., and Heathcock, C. H. J., J . Am. Chem. SOC., 1987, 109, 2022. Paper 6107092B Received October 17, I996 Accepted November 12, I996

ISSN:1359-7337    

DOI:10.1039/AC9963300429

出版商:RSC    

年代:1996

数据来源: RSC

摘要:

Analytical Communications, December 1996, Vol33 (433436) 433 Racemization of an L-Phenylalanine Residue Catalysed by an Adjacent Cysteine in a Bradykinin Peptide Antagonist Charles Brown, Paul Cutler, Christine Eckers, William Neville, George Okafo and Patrick Camilleri* SmithKline Beecham Pharmaceuticals, The Frythe, Welwyn, Hertfordshire, UK AL6 9AR A number of peptides and proteins contain aromatic amino acid residues adjacent to a cysteine group in their primary sequence.Analysis of the ratios of D- to L-phenylalanine residues in some bradykinin antagonists has revealed that racemization occurs during conventional strong acid hydrolysis of the parent peptides. The mechanism for this unusual transformation appears to involve intramolecular catalysis by an adjacent cysteine residue.Amino acid analysis of peptides and proteins, following vapour phase hydrolysis under strong acid conditions (5.5 mol 1-' hydrochloric acid), is routinely carried out in our laboratory. Recently we measured the amino acid content of several bradykinin antagonists. 1 These peptides, which reduce the vasodilation effects caused by bradykinin, contain residues of both D- and L-phenylalanine in their sequence.Peptides 1 and 2 are examples of this class of bradykinin antagonists. The stereochemistry of the phenylalanine residues in these peptides is essential to the potency of these drugs. Thus, an accurate and correct estimation of the ratio of the enantiomers of this aromatic amino acid is important for meaningful structure- activity evaluation.D- Arg-Arg-Pro-H yp-G1 y -Phe-C y s-D-Phe-Leu- Arg 1 D- Arg- Arg-Pro-Hyp-G1 y -Phe-Ser-D-Phe-Leu- Arg 2 In this communication we report the observation that chiral analysis using a number of derivatization and separation procedures revealed the racemization of L-phenylalanine during the acid hydrolysis of compound 1. In contrast, the expected 1 : 1 ratio was obtained for compound 2.The experiments we describe point to intramolecular catalysis by the adjacent cysteine residue. The peptides used in this study showed the expected amino acid composition and molecular mass by electrospray mass spectrometric analysis. Experimental Materials Dansyl chloride was 98% pure from Aldrich (Gillingham, Dorset, UK), whereas naphthalene-2,3-dicarboxaldehyde was purchased from Tokyo Kasei Kogyo (Japan).o-Phthalaldehyde was purchased from Sigma (Poole, Dorset, UK) at > 99% pure. The chiral thiol N-isobutyryl-L-cysteine was purchased from Novabiochem (Nottingham, UK) at > 99% pure. A11 buffer components (sodium dihydrogenphosphate, sodium hydroxide, * To whom correspondence should be addressed. sodium acetate trihydrate, boric acid and acetic acid) were of BDH AR grade from Merck (Poole, Dorset, UK).Chromato- graphic super purity grade eluents, methanol and acetonitrile, were supplied by Romil (Cambridge, UK). Chiral additives (sodium taurodeoxycholic acid and (3-cyclodextrin) were sup- plied by Aldrich. Buffer solutions were prepared by using distilled/de-ionized water (Milli-Q) and pH adjustments were made by the addition of appropriate amounts of dilute sodium hydroxide or acetic acid.Bradykinin and ~-Arg-(Hyp3,D- Phe7,Leu8)-bradykinin were purchased from Sigma. Other peptide samples were available at SmithKline Beecham. Peptide Hydrolysis Procedure Approximately 2 mg ml- 1 solutions of the above materials were prepared in water, following which 4 X 25 pl aliquots of each solution were pipetted into individual Pyrex hydrolysis tubes (6 X 50 mm) and the tube contents evaporated to dryness in a vacuum centrifuge. Sample tubes were placed into a hydrolysis reaction vial; 250 pl of constant boiling HC1-phenol were next pipetted into the bottom of the reaction vial. After de-gassing thoroughly with nitrogen, the vial was sealed and vapour-phase hydrolysis allowed to proceed for 24 h at 110 "C on a heating block.Following hydrolysis the sample tubes were removed from the reaction vial and evaporated to dryness in a vacuum centrifuge. Next, 25 pl of re-drying solution (methanol-water- triethylamine, 2 : 2 : 1 v/v), were added to each sample tube, vortex mixed and tube contents again evaporated to dryness, in order to remove ammonia.Dried hydrolysed peptide samples were re-dissolved in 100 pl of borate buffer with thorough mixing. Appropriate volumes of the hydrolysate were removed for derivatization, as described below. Derivatization Procedures Dansyl chloride derivatization In a typical reaction an aliquot (100 pl) of 100 mmol l-1 borate buffer, pH 9.2, was added directly to the dried peptide residue and agitated until the solid dissolved.An equal volume (100 pl) of dansyl chloride solution (8 mmol 1-l solution prepared in acetonitrile) was then added to the peptide solution and the reaction mixture left at room temperature for 30 min. The resulting reaction solutions were evaporated to dryness. Naphthalene-2,3-dicarboxaldehyde (NDA) derivatization A volume (100 pl) of 100 mmol 1-1 borate buffer, pH 9.2, was added to a hydrolysis tube containing the dried residue. An aliquot of sodium cyanide (10 pl of a 10 mmol 1-1 aqueous solution) followed by NDA (80 pl of a 1 mmol 1-1 solution in methanol) was then added and the reaction solution left at room434 Analytical Communications, December 1996, Vol33 temperature for 15 min.As the reaction proceeded, the colour of the solution changed from clear to green-yellow.The reaction solution was then used for analysis. N-Isobutyryl-L-cysteine-o-phthaldehyde (IBLC-OPA) derivatization A 5 yl volume of amino acid reference standard (200 nmol ml-1 of glycine, L-arginine, D-arginine, L-phenylalanine, D-phenyl- alanine and L-leucine), 5 pl of OPA (1 mg ml-l in borate buffer) and 5 p1 of IBLC (4 mg ml-1 in borate buffer) were added to a volume of 85 p1 of 0.15 moll-1 sodium borate buffer, pH 10.4, in a 600 pl eppendorff tube.The tube was capped and vortexed mixed. Reaction was allowed to proceed for 5 min before injection via an autosampler on to the chromatographic system for analysis. Amino acids from hydrolysed peptide samples were treated in the same way, replacing the 5 pl of amino acid reference standard with 5 pl of reconstituted hydrolysed peptide solution in borate buffer.Capillary Electrophoresis (CE) CE measurements were carried out using a Beckman P/ACE Model 5000. Electrophoresis conditions were as follows: capillary, 57 cm X 50 ym id untreated fused silica (effective length, 50 em); buffer, 30 mmol 1-1 sodium phosphate, 10 mmol 1-1 borate, 40 mmol 1-l p-cyclodextrin (cyclomalto- hertaoce) and 100 mmol 1-1 taurodeoxycholic acid at pH 7.1 with 10% (v/v) methanol; applied voltage, 20 kV; detection, He-Cd laser induced fluorescence, hex 442 nm, hem 470 nm; temperature, 25 "C; injection, 1-3 s high pressure injection followed by 6 s of water.High-performance Liquid Chromatography Measurements were carried out on a Hewlett-Packard 1050 chromatographic system, using the following conditions: col- umn, Waters NovaPak CI8, 60A, 4 pm, 3.9 X 150 mm; column temperature, 40 "C; eluent, (A) 0.023 mol 1-l sodium acetate trihydrate (pH 6.0), (B) methanol-acetonitrile (600 : 50, v/v); gradient, 5-50% B in 45 min; flow rate, 1 ml min-1; detection, ABI Spectroflow 980 fluorescence detector, hex 230 nm, 470 nm; injection volume, 5 pl.Results and Discussion Table 1 gives the measured ratios of D- to L-phenylalanine for five related peptides hydrolysed by 5.5 mol 1-l hydrochloric acid at 110 "C for 24 h. The resulting mixture of amino acids was derivatized by reaction with dansyl chloride (DNS-Cl), naphthalene-2,3-dialdehyde (NDA) or isobutyryl-L-cysteine ortho-phthalaldehyde (IBLC-OPA).In the first two cases chiral discrimination was carried out by micellar electrokinetic capillary chromatography (MECC) using a combination of taurodeoxycholic acid and P-cyclodextrin2 (Fig. 1). High- performance liquid chromatography (HPLC) was used to separate the diastereoisomers formed after derivatization with IBLC-OPA (Fig. 2). More than one derivatization and separa- tion procedure was used in order to increase the accuracy of the ratios in Table 1.From the data given in Table 1 and Figs. 1 and 2 it is clear that when the thiol group in the cysteine residue is free, a 2 : 1 ratio of D- to L-phenylalanine is obtained. This change from the expected 1 : 1 ratio is not due to selective decomposition of one of the phenylalanine residues, as achiral analysis after acid hydrolysis of 1 confirms the presence of two of these aromatic amino acids.Moreover, derivatization of the thiol group by reaction with iodoacetic acid (compound 3), oxidation (com- pound 4) or formation of a cysteine-bridged analogue1 (com- pound 5 ) prior to hydrolysis restores the expected 1 : 1 ratio of the two enantiomers. The role of the free cysteine residue in the catalysis of racemization is again emphasized by the fact that its replacement by serine in a closely related peptide (compound 2) once more gives the expected ratio of D- to L-phenylalanine.The reaction of thiol groups with the carbonyl group in aldehydes is well d~cumented.~ Examples that involve the attack by a thiol on the carbonyl group of an amide are rare.4 Recently, such a reaction has been proposed as the first step in the enzyme catalysed formation of a thiazole ring from a gly-cys dipeptides in the biosynthetic pathway for microcin B17, a 43-residue peptide antibiotic that is ribosomally synthesized.Mention of the racemization of amino acids N-terminally linked to cysteine during acid hydrolysis was also made by Woiwode et a1.6 in their study of peptides related to insulin A and insulin B.The experimental results listed in Table 1 can be rationalized in terms of attack by the thiol group on the carbonyl moiety linking the cysteine and L-phenylalanine residues. The mecha- Table 1 Ratio of L-: D-phenylalanine residues analysed by three different derivatization methods Ratios of L-Phe : D-Phe Separation via capillary Separation electrophoresis via reversed phase HPLC NDA DNS-Cl IBLC-OPA Peptide derivatization derivatization derivatization D-Arg-Arg-Pro-Hyp-Gly-Phe-Ser-D-Phe-Leu-Arg (2) 1.00 : 1.00 1.00 : 1.03 1.00 : 1.03 D-Arg-Arg-Pro-Hyp-Gly-Phe-Cys-D-Phe-Leu-Arg (3) 1.00 : 1.00 1 .OO : 1.03 1.00: 1.14 D- Arg- Arg-Pro-Hyp-G1 y-Phe-C y s-D-Phe-Leu- Arg ( 1) 1.00: 1.98 1 .OO : 1.93 1 .oo : 2.12 I I I S-CH2COOH D-Arg-Arg-Pro-Hyp-Gly-Phe-Cys-D-Phe-Leu-Arg (4) 1.00 : 1.09 1.00: 1.10 - S03H Arg-Leu-D-Phe &f2 (5) 1 .OO : 1.03 1.00: 1.10 1.00 : 1.04 D- Arg- Arg-Pro-H ypGly- Phe-Cys-SAnalytical Communications, December 1996, Vol33 435 ~~ limited extent.It is also possible that preferred partial acid hydrolysis of one of the bonds between cysteine and the two phenylalanine residues does not allow the expected 1 : 3 ratio to be observed; this is difficult to confirm without more detailed experimentation using radiolabelled amino acid residues.In conclusion, from our study it is clear that analysis of biologically active peptides containing both D- and L-amino acid residues linked to cysteine requires derivatization of the thiol moiety before acid hydrolysis.Unless this is done erroneous values of D- to L-ratios are determined which may lead to the misinterpretation of structure-activity data. nism we propose for the racemization of the L-phenylalanine residue in 1 involves as a first step the formation of a five- membered intermediate, as shown in Scheme 1. It is interesting that our results show that D-phenylalanine (on the right hand side of the cysteine residue) does not seem to be racemized to the same extent in the process.Intramolecular attack on the thiol on the carbonyl group of this phenylalanine residue will involve the formation of a seven-membered ring which is less favourable than the five-membered species shown in Scheme 1. The peptide linkage and a high acid concentration are necessary for the racemization to occur.This was confirmed by perform- ing the hydrolysis in two stages, which involved reaction at 110 "C in 0.6 mol 1-1 HC1 for 24 h, followed by incubation at the same temperature in 6 mol 1-1 acid. Analysis after the first stage indicated partial hydrolysis, and showed the formation of the enantiomers of phenylalanine and peptides containing these residues.Following the final stage of hydrolysis the ratio of D- to L-phenylalanine was closer to the expected ratio of 1 : 1. Thus, the L-phenylalanine which was cleaved in the first reaction was not racemized, and the conversion of the L- to D- form only occurs via intramolecular catalysis by the thiol group under strongly acidic conditions. The formation of intermediate 6 to produce the activated iminium system 7 is essential to the racemization process.The structure of 6 is similar to one described by Barnett and J e n k ~ , ~ who found the loss of water in a similar model compound to be fast below pH 2. Finally, studies carried out by McDonald4 have shown that intramolecular attack of a thiol group on an amide occurs with a 107-fold rate acceleration compared with the intermolecular reaction.However, the observed 1 : 2 ratio of L- phe to D-phe cannot be explained only in terms of the mechanism in Scheme I , as this would lead to a ratio of 1 : 3. One possibility for the former ratio is that racemization of the D- residue via a seven-membered intermediate does occur to a I I I I I x 3 i '20,0[ ((I) Time/min Fig.2 HPLC separation of the IBLC-OPA derivatives after acid hydrolysis of ( a ) 1 and (b) 5. - : : , : : . , $ J ! ; : 2 -20.0 Sj 7.0 9.0 11.0 13.0 15.0 17.0 19.0 21.0 23.0 25.0 27.0 29.0 - rc3 0 a .g 25.0 - c2 20.0 15.0 10.0 5.0 0.0 -5.0 I Time/min _t__t_i, : : : : . i i 9.0 11.0 13.0 15.0 17.0 19.0 21.0 23.0 25.0 27.0 29.0 Time/min 18 16 14 12 10 8 6 4 2 0 0 5 10 15 20 25 30 35 +/-H' II 1 I +/-Hi 7 Fig. 1 hydrolysis of ( a ) 1 and (b) 3. MECC separation of NDA-derivatized amino acids after the acid Scheme 1.436 Analytical Communications, December 1996, Vol33 We thank Professor A. J. Kirby for fruitful discussions during the preparation of this manuscript. References 1 Cheronis, J. C., Whalley, E. T., Nguyen, K. T., Eubanks, S. R., Allen, L. G., Duggan, M. J., Loy, S. D., Bonham, K. A., and Blodgett, J. K., J. Med. Chem., 1992,35, 1563. Okafo, G. N., Bintz, C., Clark, S. E., and Camilleri, P., J. Chem. Soc., Chem. Commun., 1992, 1189. Lienhard, G. E., and Jencks, W. P., J. Am. Chem. SOC., 1966, 88, 3982. 2 3 4 5 6 7 McDonald, R., Can. J. Chem., 1983, 61, 1846. Yorgey, P., Lee, J., Kordell, J., Vivas, E., Warner, P., Jebaratnam, D., and Kolter, R., Proc. Natl. Acad. Sci. USA, 1994, 91, 4519. Frank, H., Woiwode, W., Nicholson, G., and Bayer, E., Liebigs Ann. Chem., 1981, 354. Barnett, R. E., and Jencks, W. P., J. Am. Chem. Soc., 1969, 91, 2358. Paper 6107447B Received November 1,1996 Accepted November 8,1996

ISSN:1359-7337    

DOI:10.1039/AC9963300433

出版商:RSC    

年代:1996

数据来源: RSC

摘要:

Analytical Communications, December 1996, Vol33 437 Abe, Shigeki, 137 Adams, M. J., 401 Aexiou, John, 235 Agater, Irena B., 367 Akapo, Samuel O., 3 11 Alder, John F., 61, 323 Amici, Marina, 303 Anderson, Jan T., 429 Askal, Hassan F., 177 Atmaca, Sedef, 19 Awano, Kazutoshi, 137 Backheet, Enaam Y., 177 Bakke, Berit, H7 Bale, Simon J., 265 Bandyopadhyay, Manas, 3 15 Bames, Paul, 245 Bamett, Neil W., 255 Baron, Mark G., 393 Barreau, Stephanie, 5H Bartle, Keith D., 403 Barwick, Vicki, 123 Baxter, Douglas C., H7 Beckett, Philip R., 47 Beer, Paul D., 365 Behlke, Mary Kate, 91 Bellakhal, N., 69 Benmakroha, Yazid, 23 Bertran, Celso A., 227 Best, Markus, 71 Block, Eric, 279 Borisova, L.V., 133 Bozhkov, Ognyan, 133 Bradbury, Wayne H., 357 Braven, Jim, 53 Brennan, Martin C., 375 Brereton, Richard G., 23 1 Bridgeland, Eric S., 241 Brisset, J.L., 69 Brown, Charles, 433 Burgess, Robert J., 327 Buxton, P. C., 261 Cai, Qingyun, 163 Cai, Xiao-Jia, 279 Cambiaso, Andrea, 27 Camilleri, Patrick, 433 Caprais, Jean-Claude, 37 I Careri, Maria, 159 Chalkley, Robert, 2 15 Chen, G. N., 99 Chen, Jie, 343 Christie, Ian, 23, 27 Cilloni, Romina, 159 Cirovic, Dragan A., 231 Cockcroft, Jeremy K., 245 Compton, Richard G., 319 Cook, Kenneth, 51 Cooke, Michael, 93 Copeland, Kenneth C., 47 Creaser, Colin S., 5 Crowther, David, 5 1, 7 1, 93 Cui, Hui, 275 Cumming, Robert H., 21 Cutler, Paul, 433 Dai, Miao, 11 1 Daimon, Hiroyuki, 42 1 Davis, Teresa A., 47 Dean, John R., 191, 413 Delfino, Luca, 27 Deng, Anping, 343 Deng, Jiaqi, 111 Denoyer, Eric R., 279 Diamond, Dermot, 1H Dolman, Sarah J.L., 139 Donachie, Andrew, 293 Draou, K., 69 Driou, Alain, 143 Duan, J. P., 99 Diirbeck, H. W., 107 Ebdon, Les, 53 Eckers, Christine, 215, 433 CUMULATIVE AUTHOR INDEX JANUARY-DECEMBER 1996 Edwards, Jeff, 215 Elson, Steve, 215 Emons, H., 107 Endo, Masatoshi, 137 Ennis, J. N., 261 Ersoy, Lale, 19 Euerby, Melvin R., 403 Fagan, Peter A., 193 Fallick, Anthony E., 331 Feely, Stephen J., 5 Feldmann, Jorg, 11 Feng, Feng, 11 1 Fernandes, Joio R., 397 Fielden, Peter R., 335 Forrest, Alexander R.W., 357 Forster, Robert J., 1H Frampton, Nicholas C., 53 Frech, Wolfgang, H7 Games, D. E., 79 Gao, Zhiqiang, 223 Gara, David, 51 Ge, Honghong, 279 Gerardi, Richard D., 255 Gorecki, Tadeusz, 361 Greenway, Gillian M., 57, 65, 139, 171 Griffin, John, 2 15 Guo, Feng, 361 Guomin, Liu, 385 Gupta, Suman, 239 Haddad, Paul R., 103, 193 Hall, Carl E., 269 Hall, Christopher, 245 Hampson, Deborah L., 255 Handa, S.K., 239 Hanfland, Michael, 245 Harakuwe, Anthony H., 103 Hasebe, Kiyoshi, 75 Haskins, Neville, 215 Hassan, S. S., 299 Haswell, Stephen J., 293 Hausermann, Daniel, 245 Hernandez, Lucas, 425 HernAndez, Pedro, 425 Hewlins, M.J. E., 79 Hill, Shannon, 235 Holcombe, David, 85 Houghton, Edward, 5 Hu, Jingtian, 167 Hu, Xiaoya, 297 Hunt, Anya L., 61, 323 Imai, Kazunori, 75 Imre, Sedat, 19 Irish, Donald, 361 Iwuoha, Emmanuel I., 271,23H Jackson, L. S., 299 Jacobsen, Rebecca M., 231 Jacques, Simon D. M., 245 Jana, Nikhil R., 315 Jeffries, Teresa E., 35 Jewsbury, Roger A., 241, 269, 367 Johnson, Christopher M., 403 Jupe, Andrew C., 245 Kakizaki, Teiji, 75 Kavanagh, Peter E., 235 Kawachi, Elizabete Y., 227 Kearney, Neil J., 241, 269 Kelly, Sara J., 241 Kneafsey, Brendan J., 375 Knight, Andrew W., 171 Koller, Dagmar, 57 Kom, Stewart R., 241 Kubota, Lauro T., 227, 397 Kurochkin, Vladimir D., 381 Kurochkin, Vladimir E., 115 Law, Brian, 15 Lawrence, S.D., 79 Leng, Zongzhou, 297 Lenglet, M., 69 Leonard, Raymond G., 375 Lewis, Rhobert, 203 Li, Ping, 385 Lin, R.E., 99 Linden, Guy, 143 Liu, Haiying, 1 11 Liu, Lin Kwok, 175 Liu, Renmin, 275 Livingstone, Peter C., 235 Lopez Paz, J. L., 31 Love, Gordon D., 331 Ludkin, Edwin, 413 Lugari, Maria Teresa, 159 McClellan, James A., 1 17 MacCraith, Brian D., 271 McCreedy, Tom, 335 McDowall, M. A., 79 McMurrough, Ian, 9 McRae, Carole, 331 Madigan, David, 9 Maines, Andrew, 27 Maiolo, Katherine C., 199 Makarova, Elena D., 115 Malan, Daniel, 339 Manini, Paola, 159 Martin, Neil D., 241 Mecozzi, Mauro, 303 Miclo, Laurent, 143 Miller, James N., 5H Minty, Brian, 203, 307 Monaghan, John M., 51 Monser, Lotfi I., 65 Morgan, E.David, 15 Mortimer, Roger J., 365 Murray, Ian P., 331 Myers, Peter, 403 Narayanaswamy, Ramaier, 393 Nei, Lembit, 319 Nelms, Simon M., 57 Neto, Graciliano de Oliveira, 397 Neville, William, 433 Newbury, Sarah F., 117 Nie, Lihua, 163 Nieto, Oscar, 425 Nitescu, Ioana, 21 Obiols, J., 205 Ogden, Michael W., 197, 199 Okafo, George, 433 Pal, Anjali, 315 Pal, Tarasankar, 3 15 Park, Tae-Myung, 27 1 Parker, Keith S., 265 Parkes, Helen C., 347 Paull, Brett, 193 Pawliszyn, Janusz, 129, 219, 361, Pearce, Nicholas J.G., 35 Perkins, William T., 35 Penin, Emmanuel, 143 Petrucelli, Giovanni C., 227 Pietrantonio, Eva, 303 Poerschmann, Juergen, 2 19 Poole, Colin F., 353, 417, 1 lH, Poole, Salwa K., 353, 417, 1 IH, Pueyo, M., 205 Qi, Deyao, 11 1 Quincey, Paul, 41 Raftery, Declan P., 375 Raith, Angelika, 35 Ramsey, Edward D., 79, 95, 203, Raynor, Mark W., 95 Rees, Anthony T., 307 Rigby, G., 19H Rob, Abdul, 357 Rodger, Alison, 117 Rogers, L.J., 401 Rogerson, Vicky, 93 Rolfes, Jurgen, 429 42 1 15H 15H 307 Roulin, StCphanie C. P., 403 Rowell, Frederick J., 21, 299 Russell, Richard A,, 255 Saglik, Serap, 19 Saleh, Gamal A., 177 Santamaria-Rekondo, Araiz, 41 3 Sargent, Mike, 151 Sarradin, Pierre-Marie, 37 1 Sato, Misao, 137 Sau, Tapan K., 315 Saunders, Ginny C., 347 Schantz, Michele M., 91 Scholefield, David, 53 Seymour, Mark, 5 Shallcross, Jane A., 347 Shalliker, Ross A., 235 Sharma, K.K., 239 Shawky, S., 107 Si, Zhikun, 167 Siow, Kok Siong, 223 Smith, Roger M., 327 Smyth, Malcolm R., 9, 271, 375, Snape, Colin E., 331 Snell, James, H7 Snook, Richard D., 335 Soscic, James, 235 Steffen, Alexandra, 129 Sutton, Peter, 53 Szemes, Fridrich, 365 Teale, Philip, 5 Thomassen, Yngvar, H7 Timmis, Stuart G., 269 Tompuri, Kimmo, 283 Townshend, Alan, 31,265 Treves Brown, Bernard J., 335 Tubino, Matthieu, 397 Tummavuori, Jouni, 283 Turrillas, Xavier, 245 Tyson, Julian F., 279 Uden, Peter C., 91, 279 Vadgama, P., 19H Vadgama, Pankaj, 23, 27 van Staden, Jacobus F., 339 Vaughan, Andrew A., 393 Verreschi, Giovanni, 27 Vilalta, E., 205 Walker, James C., 199 Walker, R.F., 183 Walmsley, Anthony D., 293 Wang, Huaisheng, 275 Wang, Ronghui, 163 Wei, Wei, 163 Weightman, John S., 365 Williams, John R., 15 Williams, Kath, 367 Wohlfarth, Claudia, 7 1 Wray-Cahen, Diane, 47 Wu, Liping, 343 Wu, Lixiang, 38.5 Wu, Xinxin, 11 1 Xu, X. Q., 99 Yang, Hong, 343 Yang, Jinghe, 167 Yang, Xiucen, 343 Yao, Shouzhuo, 163 Yong, S. C. Karen, 323 Zeng, Ying, 343 Zhang, Aimei, 275 Zhang, F., 99 Zhang, Guiling, 167 Zhang, Qikai, 385 Zhang, Xiaolin, 11 1 Zhang, Zhanyen, 1 11 Zhang, Zhouyao, 219 Zhao, Shanlin, 385 Zhou, Guangjun, 167 23H

ISSN:1359-7337    

DOI:10.1039/AC9963300437

出版商:RSC    

年代:1996

数据来源: RSC