ISSN:0144-557X    

DOI:10.1039/AP98623FX001

出版商:RSC    

年代:1986

数据来源: RSC

摘要:

ANPRDI 23( 1 ) 1-44 (1 986) Ana I yt ica I Proceedings Proceedings of the Analytical Division of The Royal Society of Chemistry Brief Contents: CONTENTS Ronald Belcher Memorial Lecture Summaries of Papers 'Research and Development Topics in Analytical Chemistry' Hardcover approx 720pp ISBN 0 85186 797 9 Price f55.00 ($99.00) RSC Members f36.00 Publication Date: Early 7986 Still available: The Chemical Analysis of Water General Principles and Techniques 1st Edition (1974) by A. L. Wilson Analytical Sciences Monographs No. 2 Microfiche edition only Price f 12.00 ($23.00) VAT extra in the U.K. only RSC Members f5.50 VAT extra in U.K. only ORDERING 37 Equipment News 40 SAC 86/3rd BNASS 40 Publications Received 43 Conferences and Meetings 43 Courses 44 Analytical Division Diary General Principles and Techniques by D.T. E. Hunt and A. L. Wilson, Water Research Centre This new edition covers the considerable developments which have taken place in the eleven years since the first edition was published, in the measurement of water quality with particular reference to methods for estimating and controlling possible errors in analytical results. Information Requirements of Measurement Programmes Sampling; The Nature and Importance of Errors in Analytical Results; Estimation and Control of the Bias of Analytical Results; Estimation and Control of the Precision of Analytical Results; Achievement of Specified Accuracy by a Group of Laboratories; Reporting Anaiyical Results; The Selection of Analytical Methods; General Precautions in Water-Analysis Laboratories; Glassware and Other Apparatus; Analytical Techniques; Computers in Water Analysis. RSC Members should send their orders to: The Membership Officer, The Royal Society of Chemistry, 30 Russell Square, London WClB 5DT. Non-RSC Members should send their orders to: The Royal Society of Chemistry, Distribution Centre, Blackhorse Road, Letchworth, Herts SG6 1 HN, Enaland. Electronically typeset and printed by Heffers Printers Ltd, Cambridge, England Services ROYAL CHEMISTRY Info%ation January 1986

ISSN:0144-557X    

DOI:10.1039/AP98623BX003

出版商:RSC    

年代:1986

数据来源: RSC

摘要:

ANALYTICAL PROCEEDINGS. JANUARY 1986. VOL 23 5 Research and Development Topics in Analytical Chemistry The following are summaries of eighteen of the papers and posters presented at a Meeting of the Analytical Division held on June 26th-28tht 1985, in the Queen's University of Belfast. Use of Activated Alumina as a Column Packing Material for Adsorption of Oxyanions in Flow Injection Analysis with ICP - AES Detection I. G. Cook and C. W. McLeod Department of Chemistry, Sheffield City Polytechnic, Pond Street, Sheffield S1 1 WB P. J. Worsfold Department of Chemistry, University of Hull, Cottingham Road, Hull HU6 7RX The use of activated alumina as an ion-exchange medium in flow injection analysis - inductively coupled plasma atomic emission spectrometry (FIA - ICP - AES) is currently under investigation, and its application to the determination of phosphorus in steel' and the speciation of Cr(II1) - Cr(V1) in natural waters' has been demonstrated.The procedures offer a novel approach to rapid analyte pre-concentration and matrix removal, by virtue of the fact that activated alumina (acidic form) has a high affinity for the anionic phosphate and chromate species. This study is an attempt to extend the approach to a range of oxyanions, and preliminary results for the deposition and elution of arsenate, borate, molybdate, selenate and vanadate are presented. Experimental A complete description of the instrumentation and operating procedures is given elsewhere. 1.2 Oxyanions were deposited on the microcolumn by injection of a multi-element solution (200 yl of 1 yg ml-1; 0.01 M HN03) into the FIA manifold.Next, a solution of alkali (200 yl of 1 M potassium or ammonium hydroxide) was injected to elute analyte species into the ICP. Analyte emission wavelengths (As 193.7 nm, B 249.7 nm, Cr 267.7 nm. Mo 202.0 nm. P 213.6 nm, Se 196.0 nm, V 292.4 nm) were monitored to provide a continuous record of the emission - time response during deposition and elution. Deposition and elution efficiency for each oxyanion species are expressed (see Results) as a percentage of the peak area obtained by directly injecting standard solutions into the FIA manifold, with no column present [i.e., elution (Yo) = (area of eluted peak/area of peak from direct injection) x 100; deposition (YO) = 100 - (area of breakthrough peaklarea of peak from direct injection) x loo)].Results The deposition - elution characteristics of the oxyanions on activated alumina are summarised in Table 1, and it can be seen that with the exception of borate, all the oxyanion species undergo quantitative deposition for the experimental condi- tions used. The behaviour of the species may be sub-divided into three classes according t o their elution characteristics: (l), those whose elution is less than 60% with 1 M KOH solution- arsenate and vanadate; (2), those whose elution requires strong alkali (KOH) solution-selenate and phosphate; and (3), those where elution is achieved with strong (KOH) or weak (NH40H) alkali-molybdate and chromate. Table 1. FIA - ICP - AES deposition and elution characteristics of oxyanions on activated alumina Elution, "10 Species Deposition, % 1 M NHjOH 1 M KOH Arsenate .. . . >95 8 60 Borate . . . . . . 60 22 33 Chromate . . . . >95 84 79 Molybdate . . . . >95 84 84 Selenate . . . . >95 60 79 Vanadate . . . . >95 41 58 Phosphate . . . . >95 44 80 The species in class 1 must be regarded as irreversibly bound for our purposes as, although they may be desorbed by a stronger eluent ( e . g . , 5~ KOH solution), this would be unacceptable, as it would result in the rapid destruction of the alumina column, and also the high dissolved solids concen- tration would not be compatible with the ICP nebuliser - torch system. Within classes 2 and 3, a tentative series for the desorption of oxyanion species from activated alumina on the basis of the percentage eluted by 1 M ammonia may be proposed as follows: chromate < molybdate c selenate c phosphate.Various mechanisms have been postulate'd for the adsorp- tion/desorption of ions from alumina and similar materials, most of which propose a Langmuirs6 or modified Langmuir7.8 isotherm; however, the reversibility of the adsorption appears to depend on a multitude of factors, a major one of which is time, and all the studies in the literature have been performed under equilibirium conditions. In FIA work, the contact time between solution and adsorbent is very short (< 30 s), and thus equilibrium is unlikely to be attained. The empirical measure- ments of reversibility obtained in this work indicate recoveries of approximately 80% in one deposition - elution cycle for the four elements discussed above, which is useful (though not ideal) for analytical purposes.A comparison has been made between activated alumina and conventional ion-exchange resins (strongly and weakly basic polystyrene - divinylbenzene resins) for the deposition -6 ANALYTICAL PROCEEDINGS, JANUARY 1986. VOL 23 elution of the oxyanions in the FIA system; alumina has been found to have the following advantages: no significant volume changes, this is important in FIA work as the microcolumn may become blocked by the resin swelling; sharp peak profiles are obtained on elution, this is vital for the integration of peak areas and for high sample throughput; and long-term stability under acidic and basic conditions. This work has been carried out with the support of the Procurement Executive.Ministry of Defence. I. G . Cook thanks the Ministry of Defence for the award of a studentship. 1. 2. 3. 4. 5. 6. 7. 8. References McLeod, C. W., Cook, I. G.. Worsfold, P. J., Davies, J . E., and Queay, J.. Spectrochim. Acta, Part B, 1985. 40. 57. Cox, A. G., Cook, I . G.. and McLeod, C. W.. Analysr, 1985, 110, 331. Muljadi, D., Posner, A. M.. and Quirk, J. P., J. SoilSci., 1966. 17, 212. Rajan, S. S. S . , Perrott, K . W., and Saunders. W. M. H., J . Soil Sci., 1974, 25, 438. Ryden, J . C., and Syers, J. K . . Nature, 1975. 255, 51. Rajan, S. S . S . , Nature, 1975. 253. 434. Hingston, F. J.. Posner. A. M.. and Quirk, J . P . . J. Soil Sci., 1974, 25, 16. Rajan, S. S . S . . Nature, 1976, 262, 45.Novel Approaches to Directly Coupled Hig h-performance Liquid Chromatography - Flame Atomic Absorption Spectrometry for Trace Metal Speciation Steve Hill, Les Ebdon and Philip Jones Department of Environmental Sciences, Plymouth Polytechnic, Drake Circus, Plymouth PL4 8AA With the increasing demand for quantitative information on the form of trace metals (speciation) in a variety of samples, new techniques have been developed that allow selective detection at very low levels, typically in the nanogram and picogram range. One of the more successful approaches has been the coupling of the capability of chromatography for species separation, to the selectivity and sensitivity of atomic spectroscopy for detection. Coupled gas chromatography - flame atomic absorption spectrometry (GC - FAAS) tech- niques' are now established and used routinely in a number of laboratories.However, they are limited to compounds with favourable gas solution partition coefficients and thermal stability. High performance liquid chromatography (HPLC) is a better separation technique for many compounds although direct couplings with atomic absorption instruments can give rise to a number of problems. Several methods for coupling HPLC to graphite furnace instruments have been reported.2-4 although they invariably use some form of fraction collection device, and so do not allow continuous, on-line monitoring. Three different interfacing techniques are described here for directly coupled HPLC - FAAS based on conventional instru- mentation, each being simple, cheap to construct, and readily decoupled.Direct Nebuliser Couplings The first, and perhaps most obvious form of coupling is to attach the end of the HPLC column directly to the nebuliser. In practice, however, if this connection is made by a simple length of capillary tubing, problems are encountered in balancing the eluent flow-rate with the uptake rate of the nebuliser. To overcome this, discrete volume nebulisation may be used, either by using a simple funnel arrangement connected to the uptake capillary, or, as in the approach we favour, by the use of a vented capillary tube (Fig. 1). In this way blocking of the funnel by particulates such as dust is prevented, and there is greater flexibility in the positioning of each instrument. In addition, we also incorporate a slotted tube atom trap (STAT) above the flame, to achieve an increase in sensitivity for those elements readily dissociated to their ground state atoms.Although simple in design, this interface lends itself to many applications. and has been used successfully in our laboratories for the determination of tributyltin at sub-parts per billion levels in natural waters,' a species that has proved problemat- ical by other techniques. Coupled Hydride Systems As nebulisation is acknowledged as the most inefficient process in analytical atomic spectrometry, an interface system elim- inating this stage will obviously help increase the amount of column Fig. 1. Approaches to direct nebuliser couplings sample reaching the atom cell and hence improve the sensitiv- ity.One common way of achieving this is by the use of hydride generation, i . e . , formation of the volatile hydride by chemical reduction of the sample, which is then entrained in a stream of inert gas and carried to the observation zone, where it is decomposed by heat to form the atomic vapour (Fig. 2). The elements that form volatile hydrides-As, Bi. Ge, Pb, Se, Sn and Te, include many of direct interest in speciation studies, and so such a technique is of great value. One common application of this method is the determination of reducible arsenic species after separation by HPLC. The system uses a continuous flow hydride system6 and a pre- column packed with Zipax, a silica-based anion-exchange material, prior to separation on the analytical column. Separa- tion of dimethylarsinic acid, monomethylarsonic acid, As(V)ANALYTICAL PROCEEDINGS.JANUARY 1986, VOL 23 7 and As(II1) is achieved within 8 min. detection limits of about 1 ng being obtained for each species (Fig. 3). Not all species of the above elements form volatile hydrides, however. For such compounds the technique may be modified to incorporate an on-line ultraviolet photolysis coil, so that the HPLC eluate passes into the photolysis unit before passing into 4 HPLC column burner Mixingcoil Gas - liquid separator Fig. 2. Schematic diagram of HPLC - hydride - FAAS coupling the continuous flow hydride generator (Fig. 4). This approach has been used to determine tributyltin ions, the hydride of which is a non-volatile liquid that is normally retained in the gas - liquid separator.After optimisation, photolysis has been shown to degrade over 80% of tributyltin chloride into reducible species which may then be detected in the conven- tional way using a small hydrogen diffusion flame. Sample integrity is maintained within the photolysis coil using an air segregation system, the chromatograms produced having well defined peaks (Fig. 5). A detection limit of 2 ng for tributyltin has been obtained using the system. As(l I I) t a c 0 [1 a [r 0 4 8 Time min Fig. 3. - hydride - FAAS Typical chromatogram obtained using directly coupled HPLC by a simple microprocessor which governs both the speed of rotation and the time that the spirals spend in each location. Obviously, each spiral has a maximum loading, and so to remove the constraints on the flow-rate and hence chromato- graphy, a minibore (2 mm i.d.) HPLC column is used.facilitating flow-rates below 1 ml min-1, which prove ideal for this system. Once loaded on to the spirals samples may also be t N2 carrier gas t I uv coil I-cooling water Air I n HPLC column Y c3 Injector t Solvent Fig. 4. interface incorporating the UV irradiation coil Schematic diagram of the complete HPLC - hydride - FAAS de-solvated using a gentle flame from a micro-burner prior to moving into the flame, although radiation heating from the flame is often sufficient for this purpose, The interface has been successfully used for the direct determination of organ- olead compounds in the nanogram and picogram range, including di- and trialkyllead compounds that are normally only determined following derivatisation.A typical chromato- gram obtained for alkyllead compounds in petrol is shown in Fig. 7 . the response from each spiral being clearly observed by increasing the chart speed. Work is now continuing to couple fast-protein liquid chromatography (FPLC) to FAAS using this interface in order to investigate Zn and Cu associated with various protein fractions in plasma. Direct Sample Transport Systems The third interface uses a novel sample transport system to take the eluate from the HPLC directly into the flame. The system consists of a series of platinum wire spirals on to which the eluate is collected as it drops from the end of the HPLC column. Eight such spirals are arranged at 45" to each other on a rotating disc.The disc is then mounted on a small stepper motor and located above the spray chamber so that each spiral may rotate in turn to occupy a position in the flame directly below a small orifice in a quartz tube, which acts as an atom trap (Fig. 6). The movement of the stepper motor is controlled L I , I L I 0 4 8 12 16 20 24 Time( min Fig. 5. hydride generation after UV photolysis Chromatogram obtained from harbour water sample using Conclusions The three systems outlined above, each different in approach, demonstrate the potential and versatility of directly coupled HPLC - FAAS sytems. The use of a flame atom cell as an atomic absorption detector for liquid chromatography has the8 ANALYTICAL PROCEEDINGS. JANUARY 1986. VOL 23 advantage of simplicity and a well understood operation. The interfaces described are simple, reliable, enable real-time analysis and produce continuous chromatograms.Although only a few examples are given for each technique, there are numerous applications in many fields including clinical, indus- trial, forensic and environmental analysis. The authors thank the Science and Engineering Research Council for a studentship (to S. J. H.) and Pye Unicam Ltd. for their support. Burner head 1 1 I I I I : -- - _- - -&L - - - - - - : 1 Quartz tube over microburner Heat shield t v) c 0 n a (I 3 . 4. 5 . 6. Vickrey, T . M., Howell, H. E.. and Paradise, M. T.. Anal. Chem., 1979, 51, 1880. Brown, L., Haswell, S. J . , Rhead, M. M.. O'Neill, P., and Bancroft, K. C. C., Analyst, 1983, 108, 1511.Ebdon, L., Hill, S. J . , and Jones, P . , Analyst, 1985, 110, 515. Ebdon, L . , Wilkinson, J . R . , and Jackson, K. W., Anal. Chim. A m . , 1982, 136, 191. Ribbon to control box HPLC column Fig. 6. Plan view of rotating spirals interface References 1. 2. Ebdon, L . , Ward, R. W.. and Leathard, D. A . , Analysr, 1982, 107, 129. Brinckman, F. E . , Blair, W. R . , Jewett, K. L., and Inverson, W. P., J . Chromatogr. Sci., 1977, 15, 493. TEL \ L 0 4 8 TEL \ L TE L \ L J I 1 I I L 0 4 8 1 2 0 2 4 6 8 1( Ti me:min Fig. 7. Chromatograms of tetraalkyllead compounds in petrol (showing the response from individual spirals) Direct Determination of Chromium in Gallium Arsenide by Electrothermal Atomisation Atomic Absorption Spectrometry with Smith - Hieftje Background Correction D.Johnson and J. 8. Headridge Department of Chemistry, University of Sheffield, Sheffield S3 7HF C. W. McLeod Department of Chemistry, Sheffield City Polytechnic, Sheffield S 1 1 WB K. W. Jackson Department of Chemistry, University of Saskatchewan, Saskatoon, Canada and J. A. Roberts Philips Research Laboratories, Redhill, Surrey RH 1 5HA The determination of trace elements in semiconductor material is an important requirement in the electronics industry and recently much attention has been given to the determination of Cr in gallium arsenide. Various techniques have been used for this particular determination including spark-source mass spectrometry and secondary ion mass spectrometry, 1 neutron activation analysis ,2 differential-pulse polarography ,3 and electrothermal atomisation atomic absorption spectrometry4-5 (ETA - AAS).Much of the previous work on ETA - AAS has been based on the acid dissolution of the sample and, although relatively good precision data have been reported, such procedures are time-consuming, prone to contamination and lacking in sensitivity. Electrothermal atomisation - atomic absorption spectrometry with solid sample introduction, in contrast, is an extremely sensitive technique as demonstrated by recent applications to diverse sample t ~ p e s . 6 ~ A solid- sampling technique has been applied to gallium arsenide by Guenais et al. ,5 but non-specific background interference at the 357.9-nm line was reported and this resulted in poor agreement with data for solution samples.The problem of accurate background correction at the 357.9-nm line using a deuterium are is well known10 and the investigation of alternative background correction systems for this analysis is warranted. Hadeishi and Kimura" have used the Zeeman AASANALYTICAL PROCEEDINGS. JANUARY 1986, VOL 23 9 technique for the determination of Mn, Gn, Ag and Cr in gallium arsenide and, although only limited results were presented, a detection capability in the nanogram per gram range was achieved. An attractive alternative to the Zeeman technique is the recently developed Smith - Hieftje background correction systemlz-13 and this study assesses the suitability of this latter technique for the direct determination of Cr in solid gallium arsenide. Experimental Gallium arsenide slices (4 cm diameter X 500 pm thickness), previously analysed for Cr, were etched in dilute aqua regia to remove the surface impurities.The samples were then rinsed thoroughly in de-ionised water. The brittle slices were then shattered into small fragments (< 10 mg), weighed accurately to 0.01 mg and introduced into the graphite furnace. Standard solutions (10 pl) of Cr were directly introduced into the furnace. An Instrumentation Laboratory (IL) Video 11 atomic absorption spectrometer, equipped with the Smith - Hieftje background correction facility and an IL655 graphite furnace, in conjunction with pyrolytically coated graphite cuvettes and microboats, were used. The furnace operating parameters were as follows: a 15 s ramp to 1300 "C and a 15-s hold (ash); and a 5 s ramp to 2850 "C and a 10-s hold (atomise).For aqueous solutions a drying stage was utilised. Results and Discussion It was established in preliminary studies that when gallium arsenide was heated during the ash stage to 1300 "C decompo- sition yielded elemental As and a molten globule of Ga. Furnace operating conditions were therefore designed (1) to remove the As during the char stage and (2) to release effectively the Cr from the gallium matrix during atomisation. Various heating programmes were investigated and a typical absorbance - time response under fast ramp (see under Experimental) atomisation conditions is given in Fig. 1. It can be seen that at the commencement of the atomisation stage, but prior to the onset of Cr atomic absorption, there is a background absorption of about 0.3.It was considered that the initial background signal was due to absorption - scatter by 0 4 8 12 Time s Fig. 1. Absorbance - time graph for Cr (357.9 nm). A, Total absorbance; B, background-corrected absorbance. (Cr] = 0.46 pg g- elemental As that had condensed in the furnace during the ash stage and had then re-vaporised into the light path on atomisation. At the onset of atomic absorption a relatively large background signal (total absorbance 1.5) was obtained but the effectiveness of the Smith - Hieftje method of background correction is apparent. The interference during the latter stages of atomisation was probably due to light scatter by elemental gallium, as the final atomisation tempera- ture (2850 "C) was greater than the boiling-point of gallium (2400 "C) .An important aim of the study was to compare the analytical responses for previously analysed solid samples with aqueous standards in order to establish whether or not aqueous synthetic standard solutions could be used for the calibration. A plot of integrated peak area absorbance versus mass of chromium for aqueous Cr standard solutions (10 pl aliquots in the range 0.03-0.12 pg ml-I), and solid samples of Cr-doped gallium arsenide (6 data points from each of 3 samples: low doped, 0.23 pg g-1, medium doped, 0.43 pg g-1 and high doped, 0.83 pg g-I), gave a relatively good fit of data (y = 1 . 0 6 ~ + 0.04, correlation coefficient 0.955) and suggests that a calibration based on aqueous synthetic standard solutions may be used for the direct determination of Cr in solid gallium arsenide.Representative data are presented in Table 1 for a calibration based on both aqueous and solid samples (medium- doped material served as a calibration standard). Table 1. Cr concentrations in gallium arsenide samples Cr concentration/pg g- Sample Solid Aqueous Comparative type calibration calibration data* Low doped . . 0.23 k 0.04 0.23 ? 0.04 0.23 High doped . . 1.03 k 0.20 0.99 ? 0.19 0.83 dissolution. Medium doped - 0.46 k 0.07 0.43 * Data based on analytical values by SSMS and ETA-AAS after acid D. Johnson wishes to thank the Science and Engineering Research Council and Phillips Research Laboratories for the award of a studentship. We are grateful to Instrumentation Laboratory for provision of the Video 11 atomic absorption spectrometer.1. 2. 3. 4. 5. 6. 7. 8. 9. 10. 11. 12. References Clegg, J . B., SIA Surf. Interface Anal., 1980, 2 . 91. Brozel. M. R., Tuck, B., Rumsby, D., and Ware. R. M.. J. Cryst. Growth, 1982, 60, 113. Lanza. P., and Taddia, M., Anal. Chim. Acra, 1984, 157, 37. Taddia, M., and Lanza, P.. Anal. Chim. Acta, 1984, 159, 375. Guenais, B., Pondoulec, A.. and Minier, M., Analusis, 1982, 10, 78. Langmyhr, F. J., Stubergh, J. R., Thomassen, Y . . h'anssen, J . E . , and Dolezal, J., Anal. Chim. Acta, 1973, 71, 35. Gong, H . , and Suhr. N. H . , Anal. Chim. Acra, 1976, 81, 297. Headridge. J. B., and Nicholson, R . A., Analyst, 1982, 107. 1200. Headridge, J. B., and Riddington, I.. Analvst, 1984. 109, 113. Guthrie. B., Wolf, W. R., and Veillon, C., Anal.Chem., 1978. 50, 1900. Hadeishi, T., and Kimura. H.. J . Electrochem. Soc., 1979,126, 1988. Smith, S. B . , and Hieftje, G. M., Appl. Spectrosc., 1983, 37. 419.10 ANALYTICAL PROCEEDINGS. JANlJARY 1986. VOL 23 The Chemistry of a Precambrian Soil Profile Fidel A. Cardenas S., 6. G. Cooksey and J. M. Ottaway Department of Pure and Applied Chemistry, University of Strathclyde, Cathedral Street, Glasgow G 1 7x1 and lain Allison and M. J. Russell Department of Applied Geology, University of Strathclyde, Cathedral Street, Glasgow G 1 7XL Rocks on the surface of the earth are continuously undergoing modification owing to the physical and chemical action of agents such as carbon dioxide, water, air, oxygen and micro-organisms. The result of this complex process called weathering is a variety of soils.Of particular importance is the reaction that takes place between water and the mineral components of the rock. This is because carbon dioxide, which is produced mainly by both degradation of organic matter and bacterial action, dissolves in water. Thus, the water becomes acidic and attacks rocks more easily than pure water would do. In the Precambrian period around 1800 million years ago.' plants had not evolved. In the absence of plants the amount of carbon dioxide present in the soil atmosphere was less. Therefore, the rate at which weathering occurred was probably less than it is today. It is the objective of this paper to compare weathering processes in the Precambrian period, when plants were absent, with weathering processes of today.Collection of the Samples Rocks of the Lewisian complex and of the Cambrian system crop out in the north west of Scotland in the neighbourhood of Durness. The Lewisian complex includes the oldest rocks of the UK and consists of highly metamorphosed gneisses with lesser amounts of metasedimentary rocks, e . g . , schists and marbles with ages of between 2600 and 1800 million years (Ma).' Made up chiefly of sandstones, siltstones and carbon- ates, the much younger Cambrian rocks were deposited in a sea, which gradually transgressed north westwards over older Torredonian and Lewisian rocks on which they rest.ZThus, the weathered Precambrian soil was preserved under the basal quartzite of the Cambrian transgression. Altered and un- altered Lewisian gneisses at various distances from the Cambrian - Lewisian unconformity were taken for this investigation in two localities around Durness.The first locality was at a telegraph hut (NC 445 643) south west of Rispond and the second on the beach close to Rispond (NC 438 661). Samples from locality one are granites and gneisses, samples from locality two are mainly gneisses and pegmatites. 0.15 F Experimental Major element analysis on these samples was carried out using a Philips PW 1450/2O automatic X-ray fluorescence spec- trometer equipped with a 60 position sample changer and a microcomputer for data processing. The major elements were analysed from fused glass discs prepared according to the method of Harvey et al.3 Mineralogical analysis was performed using a Philips PW1050/25 automatic X-ray powder diffrac- tometer with a PW1120/00/60 generator and a Carl Zeiss (Jena) Amplival Pol-U polarising microscope.Specimens for the diffractometer were prepared according to the method sug- gested by Hutchinson.-' Thin sections of the rocks wzre pre- pared for microscopic analysis. Results and Discussion Kronberg and Nesbit have proposed a diagram for the quantification of present day weathering5 in terms of two ratios: the ratio (CaO + Na10 + K20)/(A1201 + CaO + Na20 + KzO) as a measure of the degree of breakdown of feldspars; and the ratio (Si02 + CaO + Na20 + K20)/(A1201 + Si02 + NazO + CaO + K 2 0 ) as a measure of the enrichment during the weathering of silicon oxide and aluminium in phases such as quartz, gibbsite and kaolin.It can be seen from Fig. 1 that values for these two ratios, calculated from the chemical analysis of our samples of Lewisian gneisses, fall along the chemical weathering path. showing that they represent a weathering profile. The chemical composition of weathered gneisses taken at 5 . 3.5.2 and 1 m from the unconformity is given in Table 1. I t can be seen that the amount of aluminium oxide present in the rocks increases from the sample at 5 m (the parent material) to the sample closer to the unconformity (the weathered material). In contrast to the aluminium oxide, the potassium oxide decreases from the parent material to the weathered rock. This is exactly what happens in present day weathering and is exactly what is happening in these Precambrian rocks.Table 1. Chemical composition of the weathered gneisses taken at various distances from the Cambrian - Lewisian unconformity S i O . . . . . . K,O . . . . . . H 2 0 . . . . . . Total . . . . A1201 . . . . Mass of oxides at different distances from the uncon for m i t y . % 5.0 m 3.5 m 2.0 in 1.0 rn 66.03 6 1 .ox 60.13 57.57 16.52 '1.45 25.89 30.73 12.41 7.74 7.60 4.55 1.08 3.43 3.30 4.77 96.04 93.70 07.12 97 62 I I Y 1 0 0.25 0.5 0.75 1 .o X Fig. 1. Kronberg weathering diagram (SiO, + CaO + Na,O + K,O) (Al,03 + Si02 + CaO + Na20 + K,O) (CaO + Na,O + K,0) (A1203 + CaO + Na20 + KzO) X = and \'= . The samples mainly consist of aluminium oxide, silicon oxide, potassium oxide and water. Such a four-component system can be represented in three dimensions and projected on a triangular diagram6 as is shown in Fig.2. The mineralogical analysis is summarised in Table 2. The parent material is mainly made of quartz (Qtz). microcline (Mcr) and small amounts of muscovite (Msc). There is very little change in the mineral composition of the samples between 2 and 3.5 m below the unconformity, which is mainly made ofANALYTICAL PROCEEDINGS. JANUARY 1986. VOL 23 11 SiOz form a stable association, which means that pyrophyllite cannot be produced by the reaction between muscovite and quartz that the samples suggest is happening. Fig. 2. Chemical composition of rocks and minerals from the weatherins profile quartz and muscovite. Potassium feldspar (microline) dis- appears in the sample closest to the unconformity and a large amount of pyrophyllite is present instead.By looking at the thin sections under the microscope. it can he appreciated that pyrophyllite appears in contact with muscovite and quartz as if it were produced by a reaction between muscovite and quartz. It will be seen later in this paper that this is not in agreement with the thermodynamic predictions. Weathering profiles are thermodynamic open systems. which are normally represented in terms of activity diagrams.' An activity diagram showing stability fields for potassium feldspar( microcline). muscovite. pyrophyllite. kaolin and gibbsite in equilibrium with natural waters is shown in Fig. 3. Thermodynamic data for this diagram has been taken from reference 8. Table 2. Mineralogical cornposition of the weathered gneisses taken at various distances from the Cambrian - Lewisian unconformity.+ + . Abundant; +. relatively large amounts; (+).just distinguishable under microscopy; -- . not ohserved under microscope Distance from the unconformitym Otz P y r Msc Mcr + + - - - 5 + + ( + ) 3.5 + + + + -7 - - + + + + 1 + + + f + - - The chemical composition of oceanic waters falls in both the gibbsite and muscovite fields. The surface ocean waters are depleted in silicic acid by the action of diatoms. which remove silica t o form their skeletons, whereas waters from the bottom of the oceans fall in the muscovite field, showing that in the bottom of the oceans muscovite is being precipitated. The composition of the rivers falls in the field of kaolin. At present. kaolin is being formed in soils along the rivers: note, however.that mountain rivers lie at higher concentrations of silicic acid; they have been in less contact with organic matter and. consequently. the concentration of carbonates is lower. The compositions of desert rivers fall across the pyrophyllite stability field. suggesting that this mineral is being precipitated in such environments. However. this has not been confirmed by any modern investigation. The compositions of acid waters and geysers fall in the pyrophyllite field. indicating that pyrophyllite may be deposited under these conditions. In alkaline lakes, potassium feldspar (microline) is deposited. According to this diagram. muscovite and pyrophvllite do not 108 106 - I - - Y L 104 10' .~ _ _ _ _ _ _ ~ --- I I I Alkaline lakes Gbs I F .- 1 a I 10 10 I I I Mcr I I [H4Si04] M Minerals in equilibrium with natural waters Fig.3. Conclusions The rocks are a fossil weathering profile and contain pyrophvl- lite. The potassium feldspar microcline was first weathered to muscovite. This muscovite was then further weathered to kaolin. The soil containing kaolin was buried deeper and deeper and eventually. when the temperature was high enough (around 300 "C), the reaction between kaolin and quartz took place to produce the pyrophyllite that is present today in this fossil soil. 1. 1 i. 3 . 4 5 . 6. 7. 8. References Lovell. J . P. B.. "The British Isles through Geoloyical Time. A Northward Drift." Wyman. London. 1Y77. Craig. G. Y.. "Geology o f Scotland." Second Edition. Academic Press.Edinburgh, 1983. Harvey. P. K.. Taylor. D. M.. Henry. K. D.. and Bancroft. F . . X-ra?, Spectrom., 1973. 2. 33. Hutchinson. C . S . . "Laboratory Handbook o f Petrographic l'ec h n i q u e s . *' W i I e y . Chic h e\ t e r . 1 9 74. Kronberg, B. I.. and Nesbit. H . W.. J . Soil S c i . . 1981.32.453, Garrels. R . M.. and Mackenzie. F. T.. "Evolution of Sedimen- tary Rocks." Norton. New York. 1971. Garrels. R. M.. and Christ. C. L.. "Solutions. Minerals and Equilibria," Freeman. Cooper and Company. New York. 1965. Helgeson. H. C.. Delani. J . M.. Nesbit. H . W.. and Bird. D. K.. Am. J . Sci., 1978. 37XA.ANALYTICAL PROCEEDINGS, JANUARY 1986. VOL 23 12 Voltammetry of Copper Dialkyldithiophosphates Miles J. Hutchings, G. J. Moody and J. D. R. Thomas Department of Applied Chemistry, Redwood Building, UWIST, PO Box 13, Cardiff CF1 3XF Metal dialkyldithiophosphates (0, 0'-dialkylphosphorodi- thioates) (DDPs) have received particular interest with regard to their anti-oxidant and load-carrying properties when incor- porated into a lubricating oil.1 Zinc dialkyldithiophosphates (ZnDDPs) have received the most interest in this context.1 The mechanism for the action of lubricating oil oxidation involves two processes: the decomposition of hydroperoxides (peroxide destroying): and the prevention of the propagation steps involved in the free-radical oxidation of hydrocarbons (chain-breaking). 1 ZnDPPs possess both types of activity and model studies have been carried out to establish mechanistic information.2-5 Copper(1) DDPs have been shown to exhibit exceptional chain breaking antioxidant activity2 and their peroxide destroying characteristics have recently6 been stud- ied.-0.2 -0.4 -0.6 PotentialIV vs. Ag - AgCl Fig. 1. Differential-pulse polarograms of copper( I) diisopropyl- dithiophosphate at the static mercury-drop electrode. A. B , C, D, E and F = [Cu'DDP]/lO-J M of 1.2, 2.4, 3.6, 4.7, 5.9 and 7.1. respectively. Supporting electrolyte 0.1 M sodium perchlorate in ethanol; drop time: 0.5 s; modulation amplitude: 50 mV; and scan rate: 10 mV s- 1 The formation of copper(I1) dimethyldithiophosphate has been used7 as the basis for a spectrophotometric method for the analysis of spray residues from plant material of S(1.2- dicarbethoxyethy1)-0, 0-dimethyldithiophosphate, commonly known by the generic name, malathion.However, the cop- per(I1) DDP complex is unstable and dissociates to form the copper(1) DDP and bis(dimethoxyphosphorothiono)disul- phide .u This type of behaviour is observed in aqueous media, but in weak or non-polar solvents copper(1) DDPs are reported to form copper(I1) DDP complexes.9 This paper is devoted to the voltammetric behaviour of copper(1) diisopropyldithiophosphate (DPDP) in an ethanolic medium. ~ [ ( R O ) ~ P S ~ ] ~ C U * [(RO)~PSZ]~CU~ + 2[(RO)zPS?]Z Differential-pulse Polarography The differential-pulse polarogram of (Cu'DPDP) in an ethan- olic medium containing 0.1 M sodium perchlorate solution as a supporting electrolyte indicates two distinct reduction processes. The first occurs at low concentrations of Cu'DPDP, while a second reduction process manifests itself at higher concentrations (> ~ 0 .3 mM) with the gradual and eventual disappearance of the former reduction wave (Fig. 1). Table 1. Variation of peak current, I,, with scan rate for 0.11 mM CuIDPDP in 0.1 M sodium perchlorate in ethanol Scan rate (v) -E, IP 1,Iv' 2 Idv ImV s- I IV IpA IpAV-I's IvAV-Is 50 0.603 0.50 2.24 10.00 100 0.615 0.77 2.44 7.70 200 0.630 1.09 2.44 5.45 500 0.655 1.45 2.05 2.90 D .c. Polarography Measurements of the limiting currents resulting from the reduction of Cu'DPDP in solution at the static mercury-drop electrode (SMDE) suggest that the oxidation state of the copper for the processes occurring at high and low concentra- tions of Cu'DPDP is the same for both electroreductions.In addition, diffusion coefficients calculated from the Cottrell equation, assuming a one or two electron reduction process, indicate that the copper species undergoing reduction is probably present as copper(I1). Independent studies involving the chronoamperometric response of CuIDPDP in solution at carbon-fibre and glass carbon microelectrodes, and also controlled-potential coulometry, did not provide credible information with regard to the oxidation state of the copper in the complex. However, in solution CuIDPDP does exhibit an e.s.r. signal indicating that some CUIIDPDP is formed on dissolving Cu'DPDP in ethanol. Table 2. Variation of peak current, I,, with scan rate for 0.55 mM Cu'DPDP in 0.1 M sodium perchlorate in ethanol Scan rate (v) - ErJ 'P I p 2* I,lv ImVs-1 IV IpA 1 p A V - l ' ~ IvAV-Is 50 0.622 1.40 6.26 28.00 100 0.652 2.00 6.32 20.00 200 0.665 2.65 5.93 13.25 500 0.690 3.05 4.21 6.10 * Mean = 5.71 and standard deviation (both in pA V-1 2 s); and coefficient of variation = 16.6.ANALYTICAL PROCEEDINGS.JANUARY 1986. VOL 23 t 4.- C F 13 Linear-sweep Voltammetry The linear-sweep voltammograms of CuIDPDP in solution are shown in Fig. 2 at three different scan rates. The peak potential, Ep, shifts in a negative direction with increasing scan rate. indicating that the reduction process is irreversible. Tables 1 and 2 summarise the linear-sweep voltammetric behaviour of CulDPDP in solution at the hanging mercury- drop electrode (HMDE). The constancy of Zdv''2 at both high and low concentrations of CulDPDP leads to the conclusion that both reduction processes are diffusion controlled. I 1 I 6 \ Cyclic Voltammetry Cyclic voltammograms of CuIDPDP at different delay times ( t d ) are shown in Fig.3. At a delay time of about 60 s, a pre-peak manifests itself and becomes more pronounced on increasing the delay time. This type of behaviour has been observed by other workers where the adsorption of an electroactive species shows a time dependency at the HMDE. 10 t C F 3 u -0.2 -0.3 -0.4 -0.5 -0.6 -0.7 -0.8 -0.9 PotentialN vs. Ag - AgCl Fig. 3. Cyclic voltammograms of 0.11 mM copper(1) diisopropyldi- thiophosphate at the hanging mercury-drop electrode (area = 0.020 cm') at different delay times. Supporting electrolyte. 0.1 M sodium perchlorate in ethanol; scan rate, 100 mV s-*; and sensitivity, 2 pA f.s.d.Analytical Significance A graph of response against concentration of CuIDPDP in an ethanolic medium containing 0.1 M sodium perchlorate solution exhibits two linear regions within a concentration range from 0.07 to 1.00 mM (Fig. 4). However, on considering the determination of copper dialkyldithiophosphates in lubri- cating oils, it can be envisaged that additive interactions and the high resistance of the medium will restrict the analytical usefulness of voltammetry as a means for determining such additives. [CU'DPDP]:~O-~ M Fig. 4. Plot of peak current versus concentration of copper(1) diisopropyldithiophosphate; differential-pulse polarographic mode used. Modulation amplitude, 50 mV; scan rate, 10 mV s-I; SMDE drop time, 0.5 s; and supporting electrolyte, 0.1 M sodium perchlorate in ethanol Conclusions The voltammetric behaviour of copper(1) diisopropyldithio- phosphate in an ethanolic medium is complex.Reduction processes observed at high and low concentrations of the complex are diffusion controlled and involve a common oxidation number for the copper species undergoing reduction at the mercury electrode. In addition, the reduction process is complicated by the adsorption of an electro-reducible species on to the surface of the mercury drop. The Science and Engineering Research Council is thanked for a research studentship (to M.J.H.) within its CASE scheme in association with Esso Chemicals Company. Dr. Lynne Grif- fiths, Dr. T. Colclough and Dr. J. Marsh are thanked for their encouragement and many helpful suggestions.Professor J. E. Simio, Universidade do Minho, Braga, Portugal, is also thanked for discussions made possible by a travel grant (069/84) from the North Atlantic Treaty Organization. References 1. 2. 3. Ford, J. F., J . Znsr. Pet., 1968, 54, 535. Burn, A. J., Tetrahedron, 1966, 22. 2153. Burn. A. J., in Mayo, F. R.. Editor, "Oxidation of Organic Compounds," Volume I , American Chemical Society, Wash- ington, DC, 1968, p. 323. Colclough, T., and Cunneen, J. I., J . Chem. SOC., 1964,4790. Bridgewater, A. J., Dever, J. R., and Sexton, M. D . , J . Chem. Soc., Perkin Trans., 1980, 2, 1006. Sexton, M. D., J. Chem. Soc., Perkin Trans., 1984, 2 , 1771. Norris, M. V . , Vail, W. A., and Averell, P. R.,J. Agric. Food Chem., 1954,2, 570.Hill, A. C., J . Sci. Food Agric., 1969, 20, 4. Yordanov, N. D . , Alexiev. V., Macicek, J., Glowiak, T., and Russell, D. R., Transition Met. Chem., 1983, 8, 257. Webber, A., Shah, M., and Osteryoung, J., Anal. Chim. Acta, 1983, 154, 105. 4. 5. 6. 7. 8. 9. 10.14 ANALYTICAL PROCEEDINGS, JANUARY Chemically Modified Electrodes Containing Complexing Groups the Determination of Trace Metals D. M. T. O'Riordan and G. G. Wallace* Department of Chemistry, University College Cork, Cork, Ireland We have previously justified our interest in the area of chemically modified electrodes and in particular the develop- ment of new electrodes for trace metal analysis.1-3 In recent publications2.3 a method for the preparation of a polypyrrole- N-carbodithioate electrode, capable of concentrating metal ions [Cu(II)] from solution, was described.In this work we have investigated in more detail the electrochemistry of copper ions trapped on the electrode surface and the limitations imposed by the polypyrrole backbone are discussed. The possibility of trapping other metal ions [Hg(II), Co(II), Ni(II), Pb(II), Fe(II), Mn(I1) and Zn(II)] subsequent to voltammetric analysis is considered and the effect of the electrode potential on uptake investigated. Finally, preliminary work on an alternative dithiocarbamate electrode will be presented. Electrochemistry of the Copper Complex on the Electrode Surface Previou~ly2~3, we reported that a reduction response was obtained following the trapping of copper ions on a thio- containing electrode surface where the polypyrrole backbone had retained some of its original conductivity following derivatisation. Consequently, the time spent by the poly- pyrrole electrode in the carbon disulphide derivatising solution had been reduced from 72 to 48 h.Such electrodes are still capable of trapping metal ions from solution, although the electrode surface is saturated at lower copper coverage. (In fact the saturation level was found to be directly proportional to the derivatisation time). More importantly, the shorter derivati- sation time ensures that the polypyrrole backbone maintains some of its original conductivity and a well defined, repro- ducible, voltammetric response for the reduction of copper on the electrode surface is obtained. Also, an oxidation response (EP = + 0.40 V vs.Ag - AgCI), which is presumably due to the copper ions trapped on the electrode surface, is obtained. From other work on dithiocarbamatesh-8 it appears unlikely that the Cu(I1)-Cu(II1) oxidation process would be observed at these potentials, prior to oxidation of the ligand. The oxidation response is therefore most likely due to the Cu(1)-Cu(I1) process. Monitoring of this response, at potentials free from dissolved oxygen interference, should prove analytically useful. Uptake of Other Metals Other metals were investigated to see if they could be trapped on the polypyrrole thio-containing electrode. Of those con- sidered [Hg(II), Co(II), Ni(II), Pb(II), Fe(II), Mn(I1) and Zn(II)] only Hg(I1) gave an additional voltammetric response (+0.60 V versus Ag - AgCI) after uptake.It is possible that some of the other metal ions were bound to the electrode surface and their voltammetric responses were masked by the polypyrrole electrode. Electrode Potential versus Uptake Experiments For the copper and mercury ions, uptake from a 1 M NaN03 solution upon application of different potentials was con- sidered.' On application of a negative potential (-0.60 V versus Ag - AgCI) the rate of uptake was improved by a factor of 4. Present address: Chemistry Department, Wollangong University. Wollangong, NSW, Australia. 986, VOL 23 for An Alternative Electrode In previous work we pointed out that one of the limitations of the polypyrrole backbone is the fact that polypyrrole itself is irreversibly oxidised in the potential range of interest.2.3 Subsequently, we have commenced investigations into the development of another dithiocarbamate-containing elec- trode.Other workers") have shown that tyramine (see below) can be electrodeposited on to a platinum substrate. In these laboratories we have devised a synthesis for N-ethyltyramine, which may be subsequently plated on to a platinum substrate." The secondary amine on the electrode surface may then be derivatised further with carbon disulphide to produce a dithiocarbamate electrode. With this electrode the polymer backbone is electroinactive in the potential range of interest. Further, the presence of the ethyl group should encourage the oxidation of metal centres at less positive potentials than would be expected on the polypyrrole-based electrode.Preliminary work indicates that this electrode is capable of trapping copper, nickel, lead and mercury ions from solution, and well defined voltammetric responses are obtained. Conclusion and Future Developments A polypyrrole-N-carbodithioate electrode suitable for the voltammetric analysis of copper and mercury ions in aqueous solution has been described. The major limitations in the use of this electrode arise because of the voltammetric behaviour of the polypyrrole backbone. More recently, we have commenced investigations into the development of a poly-N-ethyltyraminedithiocarbamate elec- trode. This electrode has a much cleaner electrochemical background and the chemical environment, in which the metals are trapped, should encourage oxidation processes to occur at analytically useful positive potentials. The authors would like to acknowledge the contribution of M.A. McKervey, Chemistry Department, University College Cork. in the development of a synthesis for N-ethyltyramine. We thank the Institute for Industrial Research and Standards (Ireland) and the State Laboratory (Ireland) for continued technical support. References 1. Wallace, G. G., Trends Anal. Chern.. 1985. 4. 145. 2. O'Riordan. D. M. T.. and Wallace, G. G.. Anal. Proc.. 1985. 22. 199. 3. Wallace, G. G.. and O'Riordan. D. M. T.. A n d . Chem., submitted for publication. 4. Bond, A. M.. and Wallace. G. G.. Anal. Chem.. 1982. 54. 1706. 5. Bond, A. M.. and Wallace, G. G., Anal. Chern.. 1984, 56. 2085. 6. Bond, A. M., and Wallace. G .G.. Inorg. Chem.. 1984. 23. 1858. 7. Concouvanis, D.. Prog. Inorg. Chem.. 1979. 26. 301. 8. Hulanicki, A.. Talarita, 1967. 14, 1371.ANALYTICAL PROCEEDINGS, JANUARY 1986, VOL 23 15 9. Wallace, G. G.. and O'Riordan, D. M. T.. unpublished work. 11. Wallace, G. G.. O'Riordan. D. M. T., Fenton. I., and McKervey. M. A , , unpublished work, University College Cork, 1985. University College Cork, 1985. Pham. M. C., Lacaze. P. C.. and Dubois. J. E.,J. Electrochem. SOC.. 1984. 131, 777. 10. Direct Determination of Volatile Trace Elements in Nickel-base Alloys by Electrothermal Vaporisation - Inductively Coupled Plasma - Atomic Emission Spectrometry P. A. Clarke, C. W. McLeod and D. J. Mowthorpe Department of Chemistry, Sheffield City Polytechnic, Pond Street, Sheffield S7 7 WB and D.J. Lee Ross and Catherall Ltd., Forge Lane, Killamarsh, Sheffield Inductively coupled plasma - atomic emission spectrometry (ICP - AES) has been developed extensively as a rapid multi-element technique for the analysis of solutions via pneumatic nebulisation. The progress on direct solids analysis, in contrast, has been slow, although some success has been achieved by the use of spark.l.2 laser.3.4 electrothermal"6 and direct insertion7.8 techniques. In recent work the direct insertion technique has been successfully applied to the simultaneous determination of volatile trace elements (Cd, Mg, Pb and Zn) in Ni-base alloys.9 The method has not been applied to the determination of As and Se as it was considered that the probe temperature available was not sufficiently high to effect the release of these elements from the involatile host matrix.It is known that As and Se form relatively stable compounds with the molten metal and in the related work of electrothermal atomisation - atomic absorption spectrometry (ETA - AAS) relatively high furnace temperatures are utilised for such determinands. 10 The aim of this study was to assess the suitability of electrothermal vaporisation (ETV) - ICP - AES for the simultaneous determination of trace elements in Ni-base alloys. Precise control of the furnace temperature was used to achieve a selective volatilisation of the trace elements from the involatile host matrix. Relatively high furnace temperatures were necessary for the efficient release of As and Se.Experimental The purpose-built ETV unit consisted of a vaporisation cell (small volume, approximately 25 ml), fitted with a two-way valve. allowing the switching of the injector gas either through the vaporisation cell or directly to the plasma torch. A graphite cup of maximum volume 50 pl was held between two graphite electrodes and a Shandon - Southern 3470A low-voltage high-current supply was used to power the assembly. Solid chips of Ni-base material (approximately 5 mg) were placed in the graphite cup, the ETV device was sealed and the heating cycle started. Typical plasma operating conditions were as follows: forward power 1.1 kW, coolant gas flow-rate 16 1 min-1; intermediate gas flow-rate 1.2 1 min-1; injector gas flow-rate, 0.5 1 min-1; and integration time 20 s (100 x 0.2 s).Results and Discussion It was found that by increasing the cup temperature to a point at which the solid Ni-base sample melted completely (1600-1800 "C) the volatile trace metals Cd, Mg, Pb and Zn were released from the molten matrix. The response from the matrix elements at this temperature was minimal and spectral interferences were not observed. The metalloid elements. As and Se, were not vaporised under these conditions. Similar results were noted using the direct insertion techniquey; however, with the ETV unit it was possible to increase further the temperature and so release As and Se from the matrix. Signals recorded for the matrix elements Al, Co, Cr and Ni were very intense and the possibility of spectral interference on the As and Se channels was investigated.In order to clarify this point certified reference materials (Bureau of Analysed Samples, BAS 345 and BAS 346) were analysed. These alloys have the same nominal matrix, but BAS 346 has enhanced trace element concentrations (As: BAS 345, no data; BAS 346, 50 pg g-1. Se: BAS 345, < 0.5 pg g-I; BAS 346, 9 pg g-I). Emission versus time profiles were recorded and an inspection of the overlay for Se (Fig. 1) shows that the greater signal results from the alloy with the higher Se content (BAS 346). This suggests that the signal is a true analyte response and not a spectral interference. t > m al C C 0 m c .- c .- ._ ._ E u1 -Ash -+-Atomise- I I ' ' lBAS 346 I I I I 10 Timeis 0 Fig. 1. Ni-base alloys BAS 345 and BAS 346 Overlay of emission - time profiles for selenium from the This work has been carried out with the support of the Procurement Executive, Ministry of Defence.P. A. Clarke thanks the Ministry of Defence for the award of a studentship. References 1. 2. 3. 4. Human, H. G. C., Scott. R. H., Oakes, A. R.. and West, C. D.. Analyst, 1976, 101, 265. Beaty, J., and Belmore, R., Paper presented at the 1981 Pittsburgh Conference, Atlantic City, NJ. Thompson, M.. Coulter, J . E., and Sieper. F., Analyst. 1981, 106. 32. Carr. J. W.. and Horlick. G.. Spectrochim. Acta Part B , 1982, 37, 1.ANALYTICAL PROCEEDINGS. JANUARY 1986, VOL 23 16 5. 6. 7. 8. Nixon, D. E . , Fassel, V. A., and Kniseley, R. N., Anal. Chem. 1974,46,210. Gunn, A. M., Millard, D. L.. and Kirkbright, G. F., Analyst, Salin, E.D.. and Horlick. G., Anal. Chem., 1979, 51, 2282. Sommer, D., and Ohls, K., Fresenius Z . Anal. Chem.. 1980. 97, 304. 9. 10. McLeod, C . W . , Mowthorpe. D. J . , and Clarke, P. A., Spectrochim. Acta, in the press. Headridge, J. B., and Nicholson, R. A., Analysr. 1982, 107, 1978, 103, 1066. 1200. Novel Method for the Determination of Arsenic, Antimony and Selenium in Single-Cell Protein (Pruteen) S. McCabe and J. M. Ottaway Department of Pure and Applied Chemistry, University of Strathclyde, Cathedral Street, Glasgow G 1 1x1 Pruteen, a single-cell protein animal foodstuff produced by ICI plc, Agricultural Division, is routinely analysed for low levels of arsenic, antimony and selenium in the range 10-500 ng g-1. In production control a highly sensitive analytical technique is required for the determination of these elements.Of the various procedures available, hydride generation atomic absorption spectrometry (AAS) is particularly suitable for the measure- ment of As, Sb and Se at sub-microgram levels.1-3 Recently, improvements in the sensitivity and the precision of hydride generation AAS procedures have been obtained through the use of a liquid nitrogen trap, in order to pre-concentrate the hydrides,k7 which is operated in conjunction with a quartz tube atomiser containing an internal fuel-rich air - hydrogen flame. The increase in sensitivity, obtained with these systems, is the result of a number of factors: (1) the pre-concentration of the hydride prior to atomisation; (2) the high concentration of hydrogen radicals in the air - hydrogen flame, which improves the atomisation efficiencys; and (3) the increased atom residence time owing to the retention of the atomic vapour in the quartz tube.In this study, this adaptation of the procedure has been developed further and applied to the determination of arsenic, antimony and selenium in Pruteen. 12 v + - I Liquid nitrogen Fig. 1. the reaction vessel and heating wire Hydride generation and liquid nitrogen trap system, showing Experimental The apparatus for the reduction of arsenic, antimony and selenium to the corresponding hydrides, their collection in a cold trap, and subsequent determination by atomic absorption spectrometry, is shown in Figs. 1 and 2. All glass parts are constructed from Pyrex glass. The system consists of a reaction vessel (50 ml) that contains the sample and any reagents. Sodium tetrahydroborate solution is injected (with a plastic syringe) into the reaction vessel through a side-arm, which is closed off by a septum held in a Teflon Swagelok union (6.4 mm 0.d.).Helium enters the solution through a fritted bubbler and transports the hydride from the reaction vessel into the cold trap. n Hydride helium + Fig. 2. The quartz tube atomiser, which is positioned on the burner mount of the spectrometer with a brass bracket The hydrides are collected in a cold trap immersed in liquid nitrogen. This consists of a 6 mm 0.d. Pyrex U-tube, the length of which is 30 cm. Approximately two-thirds of the U is filled with a chromatographic packing (15% OV-3 on Chromosorb WAWDMCS 60-80 mesh).Approximately 2 m of Nichrome wire (3Q) is wound around the outside of this trap - U-tube and connected to a variable transformer. This permits the trap to be heated at a controlled rate after the removal of the liquid nitrogen Dewar. The outlet of the cold trap is connected to the atomic absorption detection system. The internal surfaces of the system, with the exception of the reaction vessel, are deactivated by treatment with a silylation reagent. The chromosorb was silanised by treatment with 2% dimethyldichlorosilane in 1 ,1, 1-trichloroethane, and washed with methanol.9 This deactivation process minimises peak tailing. The atomiser consists of a quartz glass tube, 9 mm i.d. and 7 cm long, which is mounted in the EDL beam path of the atomic absorption spectrometer.The helium carrier gas is mixed with air, and the mixture then enters the burner cuvette from the front. Hydrogen is introduced from the opposite side. A hydrogen-rich flame burns inside the burner cuvette and Table 1. Optimum conditions for the hydride generation AAS - liquid nitrogen trap technique H2 flow-rate . . . . . . . . . . . . . . 200 ml min-1 Air flow-rate . . . . . . . . . . . . 200 ml min- 1 He flow-rate . . . . . . . . . . . . . . 100 ml min- I Injection time . . . . . . . . . . . . 1 rnin Reaction time . . . . . . . . . . . . 2 min NaBH, volume . . . . . . . . . . . . 2 ml of 1% mlV in 1 Yo NaOH HCI concentration . . . . . . . . . . 1MANALYTICAL PROCEEDINGS, JANUARY 1986, VOL 23 17 Table 2. Results from the analysis of digested Pruteen Element Dry masshg g - ' Recovery, YO Precision, '10 As 267 94 4.6 Sb 306 97 5.6 Se 347 91 5.0 atomises the hydrides, generating a transient analyte atom population, which is recorded as an AAS peak on the Linseis chart recorder. The atomiser is mounted in a Perkin-Elmer 360 atomic absorption spectrometer, and Perkin-Elmer electrodeless dis- charge lamps were used as excitation sources.The wavelength and band pass settings selected were as prescribed in the manufacturers handbook (As 193.7 nm, Se 196.0 nm and Sb 217.6 nm). Results and Discussion A 1-g sample of Pruteen was placed in a dry 250-ml Pyrex beaker. A 20-ml volume of 50% V/V nitric acid was added to the sample and the mixture digested at 200 "C. The resultant solution was reduced in volume to 5 ml and allowed to cool.A 10-ml volume of a nitric acid - perchloric acid (2 + 1) mixture was then added and the sample solution heated until fumes of perchloric acid were produced. A watch glass was then placed over the beaker and the heating continued for a further 20 min. For the determination of arsenic and antimony, 5 ml of 50% V/V sulphuric acid were then added to the solution, which was heated for a further 30 min. For the determination of selenium, 5 ml of 50% V/V hydrochloric acid were added to the digest and the solution boiled for 30 min. After cooling, the sample solutions were filtered into 25-ml calibrated flasks and made up to the mark with distilled water. Aliquots of these solutions were transferred into the reaction vessel and made up to 40 ml with distilled water and 4 ml of 50% V/V hydrochloric acid were added. In the analysis of Pruteen for antimony, 1 ml of 5 M KI solution was added to the reaction vessel sample to convert antimony into the +3 oxidation state necessary for hydride generation. 1 al C m e s a Q: 0.5 Y I 1 1 I I 0 10 20 30 40 50 Arnountlng Fig. 3. selenium Typical standard calibration graphs for arsenic, antimony and As no chemical interferences were present, digested samples were analysed by comparison with an aqueous standard calibration graph (Fig. 3). Typical aqueous solution signals are shown in Fig. 4 for 8 ng of arsenic. Detection limits ( 2 0 ) for standard aqueous solutions were 0.1 ng for arsenic and antimony and 0.5 ng for selenium. The precision with standard aqueous solutions was calculated as approximately 4% (at the 0.1 absorbance level). T TA Y 4 d Time- Fig. 4. Typical atomic absorption signals obtained with 8 ng of arsenic in a 40-ml sample volume Results for a Pruteen analysis are shown in Table 2 using the optimised parameters listed in Table 1. The precision quoted in Table 2 refers to the standard deviation attained from the analysis of six digests from one sample. Conclusion The hydride collection technique, using a liquid nitrogen trap, is simple and inexpensive and offers good sensitivity and precision in AAS analysis. No background correction is necessary, and no chemical interferences were observed in the determination of As, Sb and Se in a single-cell protein. Automation of this technique will further simplify the proce- dure and eliminate the current rather intensive manual operation. References 1. 2. 3. 4. 5. 6. 7. 8. 9. Vijan, P. N., and Wood. G. R.. At. Absorpr. Newsl.. 1974, 13, 33. Thompson, K. C.. and Thomerson. D. R.. Analvsr, 1974, 99, 595. Goulden. P. D.. and Brooksbank, P., Anal. Chem., 1974, 46, 1431. Holak. W., Anal. Chem., 1969. 41, 1712. Andrae, M. O., Anal. Chem.. 1977, 49, 820. Andrae, M. O., Anal. Chem.. 1981. 53, 1766. Cutter. G. A., Anal. Chim. Acta. 1978, 98, 59. Dedina, J.. and Rubeska. I., Specrrochim. Acra, Parr B , 1980. 35. 119. Couper, A. R.. and Johnson, J. F.. J. Appl. Polym. Sci., 1969, 13. 1487.

ISSN:0144-557X    

DOI:10.1039/AP9862300005

出版商:RSC    

年代:1986

数据来源: RSC

摘要:

18 ANALYTICAL PROCEEDINGS, JANUARY 1986. VOL 23 A Software-Controlled System for Automatic Background Correction in Inductively Coupled Plasma - Optical Emission Spectrometry 0. H. Hall, D. Littlejohn and J. M Ottaway Department of Pure and Applied Chemistry, University of Strathclyde, 295 Cathedral Street, Glasgow G 1 IXL and T. C. O'Haver Department of Chemistry, University of Maryland, College Park, Maryland 20742, USA In inductively coupled plasma optical emission spectrometry (ICP - OES) it is widely acknowledged that the problem of spectral interference is a severely limiting factor in the analysis of matrices containing major constituents with line-rich spec- tra, such as titanium pigments, rare earth ores and steel (iron). The extent of this problem has led to many publications detailing specific interferencesl.2 and it is evident that line selection plays a crucial role in analysis by ICP - OES.Spectral interferences may be overcome by extraction of the analyte or interfering element3 or by matrix-matching of the standards.j These methods can be both time consuming and difficult and any simple alternative to solution pre-treatment would appear to be advantageous. It may be considered that any signal measured at the photomultiplier tube (PMT) that is not due to analyte emission constitutes a spectral interference. In the all-argon ICP, the "blank" spectrum consists of a broad-band continuum, a number of atomic argon lines and bands derived from water.5 A matrix may change the plasma background level at a given wavelength by any combination of three ways.These are a shift in the intensity of the continuum background, an enhancement due to the overlap of the wings of a nearby line and direct line overlap. The last of these presents the most serious case and cannot be directly solved by the proposed method. The problems posed by the first two, and a system capable of overcoming these, will be discussed. Method It is possible that for a linearly sloping background, the intensity at one or both sides of the analyte line provides an accurate estimate of the intensity at the analyte wavelength. This fact has been well used to compensate for background shifts, and various methods of making "off-line" measurements have been developed. These include vibrating slits,h scanning gratings' and photodiode arrays.* One popular and simple method employs a quartz refractor plate placed behind the monochromator entrance slit.By altering the angle of the quartz plate in the plane of the incident light the radiation is refracted and displaced at the exit slit. In this way the wavelength viewed at the exit slit may be altered within a small interval, dependent on the thickness of the plate, the angle employed and the dispersion of the monochromator. This technique has been used to cope with background interference in flame atomic emission ,9 graphite furnace atomic emission'(' and continuum source atomic absorption spectrometry11 and, to a lesser extent, in ICP - OES.12 A background correction system for ICP - OES has been constructed that employs the technique of refractor plate wavelength displacement.By means of a relatively simple mechanical modification and BASIC software control, a versa- tile system has been derived from a previously restrictive se t-up. The instrumentation is described in Table 1 and is shown schematically in Fig. 1. Both wavelength positioning (via the plate) and data aquisition are achieved by the computer. The programs were written in Applesoft BASIC and converted into a compiled version for routine use. Table 1. Instrumentation ICP Source Philips PV8490 50 MHz Babington-type nebuliser Peristaltic pump (1.5 ml min- 1 ) Spectraspan IIIB (single channel) Echelle grating, 79 gr mm - I Slit widths, 100 pm (O.OOH~.OlS nm) General scanning GP3OOD torque motor with compatible scanner controller 20 x 20 x 5 mm quartz plate (Suprasil) Apple IIe 64k.twin disk drives Kaga monochrome monitor Apple printer ADC - Interactive Structures A1-13 (12-bit) DAC - Interactive Structures AO-03 (#-bit) Spectrometer Modulation Computer Interfaces Software Routines The program is divided into two main sections. To allow (a) scanning of the region near an analyte line and investigation of the spectrum; and (b) choice of measurement point> and the subsequent measurement routine. Scan Display Before it is possible to scan the spectrum around the analyte line, it is necessary to position the monochromator at or near the desired wavelength as the scan interval available via the - t Motor plate Monochromator ADC filter Fig. 1. Schematic diagram of imtrumentationANALYTICAL PROCEEDINGS.JANUARY 1986. VOL 23 .-- 19 -- ---- plate is about 0.15 nm. In order to facilitate wavelength positioning. the signal from the PMT is plotted in real time on the computer monitor screen. This visual signal displav. o r intensity - time graph. allows manual peaking of the wavelength while aspirating a solution containing the analyte. During this procedure the plate is stationary in its central position. Fig. 2 illustrates this routine. The lower right-hand corner has three rolling scroll columns, which are the intensity. '/o relative standard deviation (RSD) and the ADC gain. respectively. There are four available ADC gains with ranges of 0-100 mV. 500 mV. 1 V or 5 V. which are selected via the software. This provides a large dynamic range (40 000).and allows autoranging to be built into the measurement sequence - - (Fig. 2). r ' ! i 4918 1 . 3 1 5162 1 . 1 1 Fig. 2. Visual display of the intensity change observed during positioning of the spectrometer wavelength. The signal illustrate\ the auto-ranging facility. which reduces automatically the ADC gain if the intensity goes out of scale. The numbers in the lower right corner represent three "rolling scroll" column\ of (reading left t o right) inten\ity. RSD(%) and the ADC gain Once the wavelength has been positioned, a scan of the spectrum can be obtained. This is achieved by moving ("stepping") the motor through its angular rotation and measuring the intensity at each point. The 8-bit DAC permits 256 possible motor position\ and the program permits a display of S O .125. o r 250 points in the full scan. The total wavelength interval remains fixed and o n l y the interval between the points is altered. At each position of the motor, during the scan, ;I measurement is made at the ADC, corresponding to that position. The delay between positioning and measurement can be varied, dependent on the electronic filter (time constant) used externally (Fig. 1 ) . The intensity - position graph is di\played o n the \creen. reprownting the spectrum. Fig. 3 presents a typical screen display o f a scan. The vertical line is a movable cursor. The position and intensity o f the cursor point are shown in the lower right-hand corners. Manipulation and use o f the spectrum is made through Menu 3 (Fig. 4 ) . Options 1-6 execute functions permitting a closer inspection o f the scan. As 4096 ADC units are plotted o n a 160-point vertical display, intensity resolution is lost.This i\ overcome by the "Subtract Baseline" and "Expand" option\. The former allows any point o n the spectrum t o be chosen as the base line for subtraction o f all o f the lower part o f the intensity scan. The "Expand" option permits expansion o f the scan in the intensity axis and is often used after implementation of the "Subtract Baseline" routines. The "Smooth Scan" facility consists of a simple \liding average where each point is reduced to the average o f a \pecified number around it. Thi\ Fig. 3. Spectral scan display with cursor line on the vanadium(V 11) peak at 309.31 1 nm. The numbers at the lower right corner are (from left to right) the cursor position and the intensity at that point.respectively allows a clearer appraisal of spectral structure and assists peak finding in noisy spectra. Measurement After consideration of the spectrum around the analyte line the measurement section may be entered by selection of Menu 3 options 7-9 (Fig. 4). These offer direct line measurement. one-point and two-point background correction. respectively. In all instances each point is selected by the use of the cursor (Fig. 3) and is stored in the computer memory. The integration measurement time per point (1-10 s ) and the number of repeat integrations (1-9) are variable. Fig. 4. di\play ( 1-6) and \election of mc;i\urcment routine\ 17-9) Menu option\ that allow manipul;ition of ;I \pcctr;rl 4c;in I n the measurement mode the visual signal display (Fig.2) is again adopted, the plate being placed at the chosen cursor line position. This allows the user to decide when the analyte signal has steadied and when measurement 5hould commence. At each measurement position the AD(' is sampled at around 250 Hz, each reading being entered i n t o :I cumulative array. the average o f which constitutes the integrated intensity. Each result may be included in ;I c;~libration file. after heing identified as either a standard o r si~mple. Fig. 5 displays ii t y pi ca I pri n t ou t fc )r an an ;I 1 ys i s , i n c I 11 d i n g e iic h i n d i v I d u a 1 measurement and the calibration data that ;ire fitted by ;I linear least-square\ routine.20 ANALYTICAL PROCEEDINGS, JANUARY 1986.VOL 23 TWO POINT CORRECTION : 3 * 5 SECOND INTEGRATIONS. STANDARD1 LINE BGD.1 BGD.2 CORR. INTO GAIN=1 4051 425 370 3655 GAIN=l 4104 434 379 3699 GAIN=l 4105 423 37 1 3709 4086(.7%) 427(1.3%) 373(1.3%) 3687(.7%) STANDARD1 GAIN=2 2299 425 353 1912 GAIN=2 GAIN=2 SAMPLE1 GAIN=3 GAIN-3 GAIN=3 SAMPLE2 GAIN=2 GAIN=2 GAIN=2 SAMPLE3 GAIN=l GAIN-1 GAIN=l 21 99 408 357 1818 2224 425 355 1836 2240(2.3%) 419(2.3%) 355(.5%) 1855(2.6%) 440 41 0 359 57 439 41 7 372 45 442 41 8 36 1 54 440( -3%) 415(1%) 364(1.9%) 52(12%) 1307 1313 1239 33 1334 1366 1245 32 1337 1358 1247 38 1326(1.2%) 1345(2.1%) 1243( -3%) 34 ( 9 4%) 3943 401 5 3775 55 3984 4057 381 6 55 3932 4007 3744 65 3953(.6%) 4026(.6%) 3778(.9%) 58( 9 . 9%) CALIBRATION DATA CONC . I-EXPTL I-CAE 01 3687 3687 -05 1855 4 1855 SLOPE= 36640 INTERCEPT= 22.9999971 SCATTER= 0 SAMPLE I-EXPTL CONC .-CALC 1 2 3 52 34 58 7-91 484794E-04 3 0021 841 8E-04 9 55240252E-04 Fig. 5. Typical printout of data based on two-point background correction. Measurements at the Mn I1 257.610 nm line (100 pm slit width). Standards 1 and 2 are 0.1 and 0.05 pg ml-l of Mn, respectively. Sample matrices: (1) dilute HCl; (2) 0.4% mlV Ti; and (3) 4% mlV Ti. A11 solutions contained 20% V/V HCI. Sample concentrations in pg ml-I Application The samples of particular interest were titanium-based matrices, supplied by Tioxide UK Ltd., which were analysed for minor and trace element constituents. Brief studies showed that in the concentration range 1 4 % mlV of titanium, many analyte lines were rendered useless owing to severe direct line overlap problems.Detection limits also deteriorated rapidly as a result of vast background continuum shifts. Analyte recoveries in these solutions ranged from 25-60%, hence standard additions procedures were required for analysis. One example of an analyte wavelength that suffers a large continuum shift due to titanium is the Mn I1 257.610 nm line. The results in Fig. 5 exhibit the capability of the system in handling such enhancements. Niobium and zirconium (2040 pg g-1) have been deter- mined in titanium pigment after HF dissolution; the final solution contained 8% mlV titanium. The analysis was possible because the more sensitive lines of these elements (Nb I1 269.706-nm and Zr I1 343.823-nm lines) are free from spectral overlap and relatively large concentrations of the analytes wereANALYTICAL PROCEEDINGS, JANUARY 1986, VOL 23 21 present.Determinations of lower levels (1-10 pg g-1) of Fe, Sn and P proved impossible owing to the insensitivity of the usable lines, although at higher concentrations measurements would be possible. Conclusions The system is capable of fast, easy spectral display and versatile off-line correction. For both qualitative and quantitative assessment it has proved excellent. In practice it has proved useful for the identification of usable lines and reliable in analyte measurements. However, in the analysis of titanium matrices, the requirements of finding an analyte line that is both free from structured spectral overlap and sensitive make determinations frequently impossible, even at high analyte concentrations (>1 pg g-1). In these situations the system described above, like most routines currently available, could not be applied.The authors are grateful for the active support and interest provided by R. C. Hutton (now with V. G. Isotopes Ltd.) and other staff of Tioxide International Ltd. We would also like to thank The Pye Foundation (for D.L.) and the SERC (for D.H.H.) for financial support of this work. 1. 2. 3. 4. 5 . 6. 7. 8. 9. 10. 11. 12. References Boumans. P. W. J. M., “Line Coincidence Tables for Inductively Coupled Plasma Atomic Emission Spectrometry,” Permagon Press, Oxford, 1980. McLaren, J. W . , and Bermann, S. S . , Spectrochim. Acta Part B. 1985, 40, 217. Ando, J., Uchida, H .. Iwasaki. K., and Tanaka, K., Anal. Lett., 1981, 14. 1143. Wallace, G. F.. Hoult, D. W., and Ediger, R. D., At. Spectrosc., 1980, 1, 120. Forster, A. R., Anderson, T. A., and Parsons. M. L., Appl. Spectrosc., 1982, 36, 499. Visser, K., Hamm, F. M.. and Zeeman, P. B.. Appl. Spectrosc., 1976. 30, 72. Ediger, R. D., and Fernadez, F. J., Atom Spectrosc., 1980, 1, 1. Betty, K . R., and Horlick, G., Appl. Spectrosc., 1978, 32, 31. Snelleman, W . , Rains, T. C., Yee, K. W., Cook. H. O., and Menis, O., Anal. Chem., 1970, 42, 394. Epstein, M. S . , Rains, T. C., and O’Haver, T. C., Appl. Spectrosc., 1976, 30, 324. Messman, J. D.. Epstein, M. S . , Rains, T. C., and O’Haver, T. C., Anal. Chem., 1983. 55, 1055. McLaren, J . W., and Bermann, S. S . , Appl. Spectrosc., 1981.35. 403. Comparison of Curve-fitting Algorithms for Atomic-absorption Spectrophotometry Stephen R. Bysouth and Julian F. Tyson Department of Chemistry, Lo ug h bo ro ug h University of Techno logy, L oug h boro ug h, L eicestersh ire LE113TU With the introduction of automatic data handling techniques, several curve-fitting algorithms have appeared for calibration in AAS.1 As the processes that cause the bending of the calibration graph are numerous,’ many different models have been used. In this paper, a comparison of the existing algorithms that are commercially available is presented. The algorithms were compared with each other and with (a) linear interpolations3 and (b) graphical plots by three colleagues who used various manual curve-fitting methods.Experimental Nickel, chromimum and magnesium were used as test ele- ments. These three elements give different calibration graph shapes for the concentration ranges used. Eight standards, including a blank, were prepared for each element and these were presented to either a Shandon- Southern A3300 or a Pye Unicam SP90A atomic-absorption spectrometer. The response was recorded on a chart recorder and the blank levels were subtracted from the standard absorbances to account for any drift. The calibration data obtained are given in Table 1. All seven standards (the blank being excluded) for each element were used to calculate the test parameter for “goodness of fit.” Test Parameters The sum of squares of the percentage deviations (SSPD), equation (l), and its root mean square (RMSPD), equation (2), were used to assess the “goodness of fit.”3 N c;- c;, SSPD= Z (-X .. . . (1) i = l c;, SSPD 4 RMSPD= (7) . . . . . . where Ci is the ith concentration calculated by the algorithm, CZ is the ith actual concentration and N is the number of data Table 1. Calibration and test data Data used for calibration Nickel Concentration/ 4-point 5 -point mg 1 - I X X 0 X X 5 10 X 20 X 30 X 40 50 X X 60 Absorbance 0 0.198 0.317 0.472 0.560 0.617 0.662 0.701 Chromium Concentration/ mg 1 - I 0 5 10 15 20 30 40 50 Absorbance 0 0.296 0.510 0.696 0.824 1.002 1.140 1.198 Magnesium Concentration/ mg 1-1 Absorbance 0 0.1 0.3 0.4 0.5 0.7 1 .o 1.3 0 0.092 0.283 0.384 0.484 0.656 0.926 1.1.5622 ANALYTICAL PROCEEDINGS, JANUARY 1986, VOL 23 Table 2.Summary of algorithms Manufacturers who have adopted the algorithm Baird Atomic (Data-Comp system) Name assigned to algorithm in Table 3 Baird Quadratic Equations* used and method of curve fitting C = a + bA + cA2, solved for three data points or fitted by least-squares for more data points. Coefficients reduced for less than three data points AIC = a + bA + cA2, fitted as above C = aA + bA2 + cA3, fitted as above C = k,A + k3A2)/(k2A - 1). If the top standard absorbance is within 15% of that predicted by the bottom standard, k 3 is set to zero. When the number of standards is three or less, the equation is solved with the appropriate number of coefficients. Otherwise the equation is fitted by least-squares AIC = a + bA + cA2, solved for each set of three calibration points C = a + bA + cA2.A straight line is calculated between the blank and lowest standard. A quadratic is then fitted between every two data points, a third point being calculated using extrapolated slopes C = a + bA, solved for every two points Baird Atomic (Alpha-Star system) Instrumentation Laboratory IL Cubic Perkin-Elmer Baird Rational PE3 Coefficient PE2 Coefficient Varian Rational PU Quadratics Linear Interpolations - * Where C is concentration, A is absorbance and a, b, c , k , , k 2 , and k3 are coefficients to be found during the fitting procedure. Varian Pye Unicam points tested; in these instances N = 7 as the percentage deviation for a blank is infinite. To give an indication of any significant differences in the test parameters for each algorithm, their standard deviations (SD) were estimated using equations (3) and (4), derived from the rules for the propaga- tion of random errors.4 A 1 % RSD was assumed for Ci and 0% for Ck.N 0.02c:(c:.- CL) (C$ SDSSPD = lO4 { i = l [ (4) Algorithms and Programs The eight different algorithms and calibration methods that were tested are summarised in Table 2. Programs were written in BASIC for a Sharp MZ700 microcomputer. These programs, where possible, were tested against manufacturers' own instruments. These comparisons showed that the program that modelled the algorithm of the Pye Unicam SP9 computer was not the same and therefore the SP9 computer was used for subsequent experiments. Adjust- ment of the burner height was used to produce the required absorbance values.The other programs written for the microcomputer showed no significant differences from those used in the commercial instruments. Results and Discussion The results are presented in Table 3. Values of the associated SSPDs are not given as they vary by powers of ten and can be calculated from the RMSPD values. The large value for the Perkin-Elmer three-coefficient fit to the four-point magnesium calibration is due to the curve becoming discontinuous and asymptotic to absorbance 0.6 (Fig. 1). In this instance, the commercial algorithm would choose the two-coefficient version of the function. The quadratic function does not provide a satisfactory model for calibration graphs as, although a parabola can produce the required asymptote, there is a high degree of curvature at the blank level where, in practice, the calibration is often linear.Only when the calibration is virtually linear is a good fit obtained. The results for the manual plots are the average RMSPD and SDRMSpD values for the three analysts involved. These values compare favourably with the computer-fitted algorithms but individual results varied and no one person was consistently better. Linear interpolation only appears to be useful if the curvature is slight or if many calibration points are employed. As might be expected, all the algorithms performed similarly for a nearly linear calibration. Some differences are apparent for the curved calibrations although, except in the instances discussed above, these differences are only slight. There are several methods of assessing "goodness of fit." A commonly used parameter is the correlation coefficient.However, this tends to give values close to unity if the fit is Table 3. Results of calibrations Nickel 5-point 4-point Algorithm RMSPD SDRMSPD Baird Quadratic 28.77 0.170 Baird Rational 1.323 0.373 ILCubic . . . . 4.890 0.341 PE3 Coefficient 1.660 0.373 PE2 Coefficient 2.684 0.390 Varian Rational 1.134 0.367 PU Quadratics 2.110 0.365 Linear Interpolations 6.335 0.439 Manual . . . . 5.107 0.403 RMSPD SDRMSpD 26.58 0.194 1.344 0.386 7.102 0.317 1.413 0.371 2.789 0.386 1.687 0.368 4.621 0.381 14.45 0.485 3.415 0.430 Chromium 5-point 4-point RMSPD SDRMSPD 20.47 0.242 2.672 0.392 6.270 0.385 2.630 0.391 4.733 0.367 2.528 0.396 2.619 0.374 RMSPD SDRMSpD 21.37 0.247 2.646 0.392 5.958 0.372 2.472 0.392 4.692 0.372 2.478 0.397 3.336 0.397 4.734 0.411 9.298 0.427 2.880 0.395 2.420 0.387 Magnesium 5-point 4-point RMSPD SDRMSpD 2.158 0.367 0.968 0.381 1.832 0.369 1.859 0.366 2.874 0.358 0.694 0.383 0.734 0.383 RMSPD SD R M s PD 2.888 0.359 1.336 0.368 1.290 0.368 11.85 0.491 3.693 0.350 1.300 0.367 0.672 0.377 1.288 0.385 1.223 0.383 3.330 0.367 3.211 0.369ANALYTICAL PROCEEDINGS, JANUARY 1986.VOL 23 23 good for higher absorbances and concentrations, even if the fit is poor further down the calibration graph. In addition, they cannot be used for algorithms that either solve equations explicitly or use interpolation methods. // ,/ 0 1 ,;/ /' . . , - CONCENTRATION 0+1.3pprn Fig. 1. Discontinuity observed when the three-coefficient version of the Perkin-Elmer algorithm is fitted to the four-point magnesium calibration The sum of squares of the percentage deviations and its root mean square are more appropriate measures of "goodness of fit" as they can be used for all algorithms if intermediate data points are included between calibration points.As these parameters are based on the percentage deviations in concen- trations, they give a measure of fit, unbiased toward any part of the graph. The RMSPD represents the possible percentage error in the concentrations, due to the curve-fitting algorithm. Conclusion Although the calibration algorithms vary greatly. there is no evidence that any of the commercially available algorithms perform significantly better than any others. Often, the fit is dependent on the number of data points and degree of curvature.For example, if the fit by a quadratic is poor, increasing the number of calibration points will make little difference, but the converse is true for linear interpolations. Given the difficulty of predicting what the shape of a particular element's calibration graph will be, it seems likely that whatever algorithm is used, there will always be errors due to an lack of fit. The authors thank the Trustees of the Analytical Chemistry Trust Fund of the Royal Society of Chemistry for the award of an SAC Research Studentship. References 1. Tyson, J. F.. Analyst, 1984, 109. 313. 2. Price, W. J . . "Spectrochemical Analysis by Atomic Absorp- tion," Heyden, London, 1979. 3. Miller-Ihli. N. J . , O'Haver.T. C.. and Harnly. J . M.. Spectrochim. Acta, Part B . 1984. 39. 1603. 4. Miller, J. C., and Miller. J . N . , "Statistics for Analytical Chemistry,'. Ellis Horwood. Chichester. 1984, p. 46. Effect of Flow Cell on Dispersion in Flow Injection Analysis David C. Stone and Julian F. Tyson Department of Chemistry, University of Technology, Loughborough, Leicestershire LE 7 7 3TU Since the introduction of flow injection analysis (FIA) by Rhiitka and Hansen.' numerous methods have been used to quantify dispersion in FIA systems. These include the use of numerical techniques for solving the diffusion - convection equation' and various flow models from the chemical engineer- ing literature.3 All of these methods involve the concept of an ideal or theoretical manifold, which arises from the assump- tions that (i) conditions of laminar flow exist; (ii) the flow is undisturbed by the injection process; (iii) flow occurs down a long, straight tube of circular cross-section; and (iv) the solute concentration is measured in a plane at right-angles to the direction of flow.In practice, it is found that valves and connectors cause disruptions of the laminar flow profile, as does the detector (the nebuliser of a flame AAS, for example), which may well measure concentration within a finite volume (spectro- photometric flow cells, for instance). Further, manifolds for real analyses are often more complicated than the simple. single line example envisaged in theoretical treatments of dispersion. For these reasons, various new flow models are being examined, and an extensive investigation of the factors affecting physical dispersion has been undertaken.In this paper, the results of studies of the contribution of the flow cell used with spectrophotometers to the dispersion are presented. Experimental Computer Modelling The effect of measuring solute concentration within a finite volume, rather than a plane, was investigated using computer calculations based on the well stirred mixing tank model of Tyson and Idris.4 If a volume V, of a sample of unit concentration is allowed to enter a mixing tank of a volume V,. at a flow-rate u , it can be shown that the resulting concentration - time profile at the tank outlet consists of exponential rise and fall curves described by the following equations: C = 1 -exp[-utlV,].Odr~t, .. . . ( I ) C = exp[-u(t - tp)/Vm], t > t , . . . . (2) where C is concentration and tp is the time for the peak maximum to be obtained, equal to VJu. If it is assumed that the cell volume consists of a large number of plane perpendicular detectors in series, and that no dispersion occurs within the cell volume. then the effect of cell volume can be computed using a simple numerical integration technique. This was done for a range of conditions for the model, using a six-strip Simpson's rule procedure. Effect of the Flow Cell The effect of the flow cell on the observed dispersion was investigated using a single line manifold, consisting of a 110 cm length of 0.58 mm i.d. PTFE tubing, a Rheodyne 5020 sample valve and a Gilson Minipuls 2 peristaltic pump.A water-soluble dye (tartrazine) was used as the tracer, and detected using a Pye Unicam SP6-250 visible spectro- photometer. The four flow cells that were used are illustrated24 ANALYTICAL PROCEEDINGS, JANUARY 1986, VOL 23 in Fig. 1. Two cells were conventional flow cells with volumes of 8.0 and 60 p1 (Pye Unicam). The third was a combined fluorescence - absorption cell with a volume of 25 pl (Hellma). The effective volume was much smaller, however, as only a part of the cell volume coincided with the light patch of the spectrophotometer. The fourth cell was constructed in-house from a length of drawn glass capillary tubing mounted in a black Perspex block, and had a volume of approximately 0.6 pl. L' L f W (dl Fig. 1. Flow cells used in this work.F indicates the flow path; L indicates the light path; and shading indicates the optical volume. Cell volume: ( a ) 0.6; ( b ) 8.0; ( c ) 25; and ( d ) 60 pl The peak shapes obtained using each cell in turn were recorded using a chart recorder connected to the detector. Steady-state absorbance values were also obtained for each cell, so that the peak shapes could be normalised for direct comparison. The effect of the flow cell on observed dispersion was further investigated using 8.0 and 35 pl cells, the latter being of similar construction to the 60 pl cell. The peak dispersion, D, (defined as the ratio of injected to peak concentrations) was measured as a function of flow-rate for a range of manifold parameters for both cells. Results and Discussion Results obtained from the theoretical calculations are shown in Fig.2. These results predict that, in the absence of any additional mixing within the cell, there will be an increasing distortion of the observed peak shape for increasing cell volume. This involves a decrease in peak height, an increase in peak width and a shift in the time taken for the peak maximum to appear. 0.80 0.60 u 0.40 0.20 0 2.00 4.00 6.00 8.00 10.00 r:s Fig. 2. Concentration - time profiles for the well stirred tank model. Cell volume: A, 0; B, 10; C, 25; D. 40; E. 60; and F, 80 pI ( V , = 60 pl; V , = 10 vl; and flow-rate ( u ) = 2.0 ml min-1) The experimental results for peak shape show similar effects between the 60 pl cell and the other cells (Fig. 3). The slight variations between the curves for the 0.6, 8.0 and 25 pl (with lower effective volume) cells may reflect different contri- butions to dispersion, but could equally well be attributable to experimental error.0.60 0.50 0.40 0.30 0.20 0.10 $ 0 2 0.50 0.40 0.3C 0.21 0.1 t: s Fig. 3. ( b ) 2.00 ml min-1. Cell volumes: A, 0.6; B. 8.0; C. 25; and D. 60 pl Peak shapes obtained at flow-rates of ( a ) 6.00 ml min- I ; and Some typical results for the variation of D with flow-rate are shown in Fig. 4. Clearly, the cell is contributing to the dispersion, and in a manner which shows some flow-rate dependence. This has been attributed to edges, sharp bends and bore changes within the cell, which will all give rise to regions of turbulence and "dead water". The extent of such regions may be expected to show some flow-rate dependence.Conclusions The results show clearly that the flow cell used with any FIA manifold cannot be ignored as a source of dispersion within the system. The flow cell contributes to the over-all dispersion inANALYTICAL PROCEEDINGS. JANUARY 1986. VOL 23 2.00 F 0.00 1.00 2.00 3.00 4.00 5.00 6.00 7.00 8.00 9.00 Flow-rateiml min-l Fig. 4. Variation of the dispersion with flow-rate. 0 = 35-pl cell; A = 8.0 11-cell. Sample volumes: A. 113; B, 180; C, 222; and D, 287 11, ( L = 30.1 cm; 1.d. = 58 mm; 0.01 M NO.; at 300 nm) 25 two ways. Firstly, there is an effect due to the effective measuring volume of the cell. Therefore, this should be kept to the smallest size possible consistent with the requirements of sensitivity and light throughput.Secondly, the flow path through the cell is important. To minimise dispersion, cells should be used that produce the least possible disturbance of the carrier stream from the manifold, in accordance with LC practice. Financial support for this work form the SERC and Pye Unicam Ltd. is gratefully acknowledged. References 1. RGfi€ka, J . , and Hansen. E. H . , Anal. Chirn. Acta, 1975, 78, 145. 2. Vanderslice. J. T.. Stewart, K. K.. Rosenfeld, A. G., and Higgs, D., Talanra, 1981, 28, 1 1. 3. R%i€ka, J., and Hansen. E. H . . Anal. Chirn. Acra, 1978,W. 37. 4. Tyson, J . F . , and Idris. A. B . , Analysr, 1981. 106. 1125. The Chemiluminescence Determination of Drugs Richard W. Abbott and Alan Townshend Chemistry Department, University of Hull, Hull HU6 7RX Chemiluminescence (CL) is the emission of light produced in a chemical reaction.Electronically excited molecules formed in the reaction decay to the ground state. emitting photons: A + B -+ P* + P + hv Oxidations are the most frequent source of CL, as they involve large free energy changes. CL and bioluminescence reactions have been interesting phenomena for many centuries,' but until recently their potential as analytical detection techniques had not been greatly appreciated. However, their remarkable advantages in terms of sensitivity and selectivity over established techniques such as UV - visible and fluorescence spectrometry, together with the development of extremely sensitive instrumentation,' has generated much interest recently, and several excellent applications have been reported?-5 This has stimulated the present research, which uses CL for the detection of trace levels of drugs.Instrumen tation Flow injection analysis ( FIA)h has been successfully combined with CL detection.' CL emissions are short-lived, typically a few seconds, and the intensity depends on the rate of mixing of analyte and reagent. The FIA system has to be designed, therefore, to achieve rapid, reproducible mixing and very rapid transfer to the detector cell. Sample injection (25 yl) Flow-rate = Recorder Flow rate = Waste 1.3 ml min- Fig. 1. Manifold and detector for FIA with chemiluminescence detection Flow-through CL detectors previously built in this labora- tory have been described by Wheatley8 and Faizullah and Townshend.' The present detector (Fig.1) is a logical evolution of these. The housing contains a Perspex T-piece which allows the required efficient mixing of the sample with the oxidant, a coiled glass flow cell, backed by a mirror, and a sensitive photomultiplier tube (Thorn EM1 97890B), attached to a stable high-voltage power supply (Thorn EM1 3000R). An indication of the extreme sensitivity of the detector was obtained by the determination of hydrogen peroxide based on its CL reaction with luminol in the presence of cobalt ions as a catalyst.' The detection limit was about 1 fmol of Hz02. Applications A number of drugs and other organic compounds have been screened for CL using the system shown in Fig. 1. Drug samples, including antibiotics, monoamine oxidase inhibitors and opium alkaloids, were injected into a suitable buffer carrier stream, and tested for CL emission on mixing with various oxidants, including hypochlorite, peroxide and per- manganate.Morphine is typical of a compound which is found to give CL on oxidation. Highly sensitive methods for the fluorimetric HO Q-f) OH determination of morphine by oxidation with alkaline hexa- cyanoferrate(II1) to form the fluorescent dimer pseudo- morphine have previously been described,l0.11 in which less than 10 ng of morphine could be detected. The present assay was developed using permanganate (6 x 10-4 M ) , which mixes with morphine in a carrier stream of 0.1 M polyphosphoric acid, converting morphine into pseudo- morphine and producing CL.1' The flow-rate was 2.6 ml26 ANALYTICAL PROCEEDINGS.JANUARY 1986. VOL 23 min-1. Fig. 2 shows a typical recorder output for a morphine calibration. Under the optimum conditions, the detection limit for morphine is 2 fmol (0.7 pg or 1 x lo-'" M ) . A logarithmic calibration plot for morphine is shown in Fig. 3. The relative standard deviation of ten replicate sample injections of 1 x 10-5 M morphine was 1.8%. A throughput of 150 samples per hour is possible. Structurally similar opiates, e . g . , nor- morphine, morphine N-oxide and dihydromorphine. also give CL in this reaction. The method is being applied for post- column HPLC detection, so these potential interferences can be separated and resolved by careful selection of the mobile phase. The authors thank the SERC and the Home Office for a studentship (to R.W. A.) under the CASE scheme, and Dr. R. Gill for useful discussions. 6oo r------ 500 t 1 400 > E '$ 300 - E c 0 .- w 200 I 100 E 1 min H D I - Time Fig. 2. Typical recorder output for standard morphine solutions: A, 1 x 10-6 M; B. 4 x 10-6 M; C. 5 x 10-6 M; D, 8 x M; and E. 1 X M 1. 2. 3. 4. 5 . 6. 7. 8. 9. 10. 11. 12. 3.0. > E 5- 'v, 2.0. c C Q) c c 0 v) c. .- .- .- 1.0 ' Q) 0, 0 1 0.0 ' -1.0' ' 1 -10.0 -9.0 -8.0 -7.0 --6.0 -5.0 --4.0 Log[morphinel M Fig. 3. Log - log calibration plot for morphine References Harvey. E.N., "A History of Luminescence from the Earliest Times to 1900," American Philosophical Society. Philadelphia. 1957. Stanley, P. E.. Trends Anal. Chern.. 1983. 2. 248. Burguera, J. L., and Townshend, A.. Takunra, 1980. 27.309. Kobayashi, S . , Sekino, J.. Honda. K . . and Imai. K.. Anal. Biochem.. 1981, 112. 99. Weeks, I., and Woodhead. J . S . . J . Clin. Immicnoassay. 1984. 7, 82. RGiiCka. J., and Hansen. E., "Flow Injection Analysis." Wiley. New York, 1981. Townshend, A.. Anal. Proc.. 1985. 22. 370. Wheatley, A.. PhD Thesis. University of Hull. 1983. Faizullah. A. T., and Townshend. A . . And. Proc., 1985. 22. 15. Jane I., and Taylor. J . F., J . C'tzromatogr. 1975. 22. 15. Nelson, P. E . . Nolan. S. L.. and Bedford. K. R . . J . Chromatogr, 1982, 234, 407. Abbott, R. W..Townshend, A., and Gill. R.. Analyst, in the press. Indirect Amperometric Detection of Metal Ions Following Ion Chromatog rap hic Separations H. Hojabri, A. C. Lavin and G. G. Wallace" Chemistry Department, University College Cork, Cork, Ireland and J.M. Riviello Dionex Corporation, Sunnyvale, CA, USA The separation of metal species using liquid chromatography and their subsequent detection remains a challenging field of research.'-20 In recent times, various methods involving the separation and detection of metal dithiocarbamate complexes using normal- or reversed-phase chromatography, and their subsequent detection, have been described.Is20 The use of the dithiocarbamate ligand as a derivatising agent is particularly attractive when using amperometric detection, as it allows monitoring of the oxidation process at potentials free from Present address: Chemistry Department, Wollangong University, Wollangong, NSW, Australia. dissolved oxygen interference. 1 + x However, the previously described methods are limited in that, in some instances, the instability of metal complexes, on-column, mitigates against development of a successful method of analysis.Further. the free dithiocarbamate ligand is. itself. electroactive at relatively low positive potentials. Therefore, with any method involving the use of an excess of ligand, such as in sirir complexation chromatography.1-F-2" the metal oxidation responses must be monitored on top of a high background. In this work, the fact that the free dithiocarbamate ligand is oxidised at relatively low positive potentials was used to advantage. A method involving separation o f metal ions usingANALYTICAL PROCEEDINGS, JANUARY 1986. VOL 23 27 ion chromatography fo 1 1 owed by post - co 1 u m n d e ri v a t i sa t i on using the dithiocarbamate ligand and indirect amperometric detection has been developed.Basis of the Method The ligand employed in this work was the pyrrolidine dithiocarbamate ligand ( I ) . This ligand is oxidised at relatively low positive potentials (Ep = +0.38 V versus Ag - AgCl) at a glassy carbon electrode. By monitoring the decrease in this oxidation response in the presence of metal ions, an indirect amperometric detection scheme was devised. The ligand was added post column using a conventional tee mixer and reacted with the metal ions as they eluted from an ion chromatography column. A Dionex CS-2 column was employed using the following chromatographic eluents: eluent 1: eluent 2: 0.010 M oxalic acid - 0.0075 M citric acid, pH 4.3 (adjusted with NaOH) ; 0.040 M tartaric acid - 0.012 M citric acid, pH 3.3 (adjusted with NaOH).The reagent was prepared in the same solution as was used for the chromatographic eluent. Responses were detected using a BAS LC-17 amperometric detection package. with a glassy carbon electrode. Results and Discussion Preliminary experiments were carried out in a flow-through system without the chromatographic column. Various parameters were investigated, as follows. Reagent concentration. This was found to affect the responses obtained on injection of metal ions. Optimum results were obtained using M pyrrolidine. Reagent flow-rare. Increased flow-rates improved the responses observed for metals which form more labile com- plexes (e.g.. Cd2+). However. an increased flow-rate had an adverse effect on the responses observed for metals that form stable complexes (e.g.. Cu'+).owing to the increased dilution with more reagent present. Coriiposirion of reagent soliction. The addition of small amounts of organic solvent** (up to 20% ViV) to the reagent solution increased the linear range of the calibration graphs owing to the increased solubility of the metal - dithiocarbamate complexes. Reactor coil temperature. The rate of increase of signal with temperature was more marked for metals that form less stable complexes (e.g., Ni'+, Pb'+). Potential ~t'ui~eforni. D.c., differential-pulse and normal- pulse amperometry were investigated. The pulse waveforms gave slightly better signal to noise ratios as they are less dependent on flow-rate." Following investigation of these parameters, the column was included in the system and metal ions chromatographed and detected using the post-column reaction - indirect ampero- metry system.Results for the metals investigated are sum- marised in Table 1. For Co?+ injections no response is observed. presumably because Co:+ forms relatively stable complexes with oxalate and tartrate. Similarly. it appears that the Cd2+ forms a relatively stable complex with oxalate. The cadmium dithiocarbamate complex is known to have a low stability constant." With Hg" the ~~~~~~~~~~ "Acetonitrile and methanol were investigated. metal ion is not eluted from the column under the chromato- graphic conditions employed. When the column is omitted from the system a response with similar sensitivity to C U ~ + is obtained.The Zn2+ and Fe'+ dithiocarbamate complexes are known to have low stability constants22 and once again no response is observed. For Fe2+ injections the response observed is not due to a decrease in oxidation current of the free ligand as it is of opposite sign. The response is presumably due to a positive oxidation process involving oxidation of the metal from either Fe(1I) oxalate or Fe(1I) tartrate to Fe(II1). Table 1. Minimum detectable levels (p.p.m.) (signal = four times the noise level). For details of elunets see Basis of the Method. Dionex CS-2 column Eluent 2 Metal injected Eluent 1 Cu'+ Co' + Pb'+ Cd'+ Hg'+ Zn2+ Fez+ Fez + Ni2 + "NR = no response. 0.10 0.40 NR" 0.50 NR NR NR 2.00 NR 0.25 0.40 NR 0.50 0.20 NR NR 1 .oo NR Conclusion and Future Developments The ability to detect metal ions using ion chromatography and indirect amperometric detection has been demonstrated.Other chromatographic systems that employ different elu- ents and other dithiocarbamate reagents (which form water- soluble species) are currently under investigation. The authors gratefully acknowledge the financial assistance of the Dionex Corporation, Sunnyvale. CA. References 1 . 7 i. 3. 4. 5 . 6. 7. 8. 9. 10. 11. 12. 13. 14. 1s. 16. 17. 18. 19. 20. 21. 22. Cassidy, R. M., in "Trace Analysis." Volume 1. Academic Press. New York. 1981. Cassidy, R . M.. and Elchuck. S . . Anal. Chem., 1982.54. 1558. Cassidy. R . M.. and Elchuck. S . . Anal. Chem.. 1982. 54. 772. Roston. D . A.. Anal. Chem., 1984.56. 241 Schwedt, G.. and Budde. R.. C'hromcirographia. 1982. 15,527. Curira, R . C.. and Carr. P. W.. J . Chromarogr. Sci.. 1982. 20. 461. O'Laughlin. J . W.. Anal. Chem.. 1982. 54. 178. Hoffmann, B. W., and Schwedt. G . , J . High Resolur. Chromarogr. Chromatogr. Comniun., 1982. 5 , 439. Hobbs. P. J.. Jones, P.. and Ebdon. L.. Anal. Proc.. 1983.20, 613. Eggers. H.. and Russel. H. A.. Chromarographia. 1983. 17. 486. Gaetini. E.. Laureri, C. F.. and Manpia. A.. Ann. Chim. (Rome), 1979. 69, 181. Edward-Inatimi, E. B.. J . Chromarogr.. 1983. 256. 253. Smith, R. M . . and Yankey, L. E.. Analyst, 1982. 107. 774. Smith, R. M.. Anal. Proc.. 1984, 21. 73. Smith, R. M.. Butt. A. M.. and Thakur. A.. Anulysf. 1985. 110,35. Bond. A. M.. and Wallace. G . G.. Anal. C'hem..1981. 53. 1209. Bond. A. M.. and Wallace. G . G . . And. Chem.. 1082. 54. 1706. Bond. A. M.. and Wallace, G . G.,Anul. C'hem.. 1983.55.718. Bond, A. M.. and Wallace. G . G . . J . Liy. Chrornafogr.. 1983. 6. 1799. Bond. A. M.. and Wallace. G . G.. Anal. Chem.. 1984. 56. 2085. Swartzfager. D. G., And. C'hem., 1976. 48. 2180. Hulanicki. A.. Talurira. 1967. 14. 1371.28 ANALYTICAL PROCEEDINGS, JANUARY 1986, VOL 23 On-line Voltammetric Analysis of Aniline M. G. Taylor and T. E. Edmonds Department of Chemistry, Loughborough University of Technology, Loughborough, Leicestershire LE7 7 3TU- The amperometric detector is one of the most sensitive HPLC or FIA detectors for the direct determination of aromatic amines, from a few nanograms down to a few picograms of amine.1 With ultraviolet photometric detection, a detection limit of 5 x 10-6 M has been achieved.' Aromatic amines are readily oxidised at solid electrodes at potentials close to + 1 .O V (versus S.C.E.).The amperometric oxidation of aniline is particularly troubled by poisoning of the electrode surface, owing to a build-up of a polymeric film, which is an accumulation of electrode reaction products. Various sugges- tions for overcoming electrode poisoning are to be found in the literature,2-4 and the methods need to be evaluated individu- ally for the problem at hand. In this paper, various attempts at overcoming the electrode poisoning associated with the detec- tion of aniline are described, and their suitability for on-line monitoring is discussed. The electrochemical oxidation of aniline5 follows the classic ECE mechanism, i.e., in dilute sulphuric acid solution: HNH HNG H (1) + (2) -2H' - Ph- N- Ph-NH2 C (4) p Aminodiphenylamine Significant amounts of benzidine are formed in more acidic solutions.Polymerisation is said to be by a succession of head to tail dimerizations: 2 5 - 4e - Tetramer -4H' - 2e Te t rame r( ox) -2, -2H+ - Octamer (emeraldine) } -2H+ Four methods of overcoming poisoning were tried, as follows. 1. Solvent effects. The effects of pH and solvent on the electrode poisoning were investigated using linear sweep voltammetry. The percentage decrease in peak height for consecutive sweeps was used as a measure of the electrode poisoning. 2. Voltammetry at microelectrodes. The voltammetry of aniline at a carbon fibre microelectrode was investigated, because if the chemical follow-up reactions are of a moderate rate, then they will tend to occur away from the electrode surface.This is because the mean distance that an oxidised molecule diffuses away from the electrode is greater than its radius. 3. Pulse amperometry. Electrochemical detectors are normally operated in the d.c. mode, where good rectilinearity and sensitivity are obtained, but electrode poisoning may be a problem. Pulse amperometry is suggested as a suitable alternative when the poisoning becomes troublesome,6 but the method tends to be less sensitive and less reproducible owing to the charging current generated. The advantage over the d.c. mode is that at correctly chosen initial potentials the adsorbed species may be stripped off the electrode surface.Also. the electrode only experiences the working potential for short periods of time, and so less electrode reaction products are formed. 4. Modification of electrode. If an electrode is held at + 1 .O V in aniline solution the faradaic current decays exponentially, corresponding to the build-up of film on the electrode. A glassy carbon electrode was coated with a film in this way, to try to prevent poisoning during amperometric analysis of aniline solutions. Experimental The voltammograms were generated using a Metrohm VA 61 1 scanner and VA 612 detector. Electrodes used were glassy carbon (Metrohm), carbon paste7 and microelectrodes. The microelectrodes were prepared by cementing a carbon fibre into a drawn-out glass capillary using epoxy adhesive cured with 40% HF,7 and the fibre was cut back to form a disc.To make an electrical connection to the fibre, the capillary was filled with mercury and a platinum wire was in contact with the latter. Epoxy cured using the normal hardener was found to be unsuitable, because a high background current was observed owing to residual traces of amines which are a constituent of the hardener. D.c. and pulse hydrodynamic amperometry were performed using flow injection analysis (FIA). The operating potential for the d.c. mode was +1.1 V and for the differential pulse mode +0.8 V with a pulse amplitude of +0.2 V. The pulse width was 160 ms, the current being sampled for the last 20 ms. The electrochemical detector was of the wall jet design8 and the working electrode was glassy carbon.The FIA experiments were carried out by injecting a 75-pI sample into a carrier stream of 0.1 M sulphuric acid, flowing at 6 ml min-1 through 0.8 mm i.d. PTFE tubing. The glassy carbon electrode was modified by applying a potential of + 1.1 V for 5 min in a solution of 10-3 M aniline in dilute sulphuric acid. Results and Discussion The lowest electrode poisoning was observed in a solvent system of 0.1 M pyridine in acetonitrile, with 0.1 M sodium perchlorate as background electrolyte and a glassy carbon electrode. The concentration of aniline was 5 x 10-3 M . Less polymerisation is observed in this solvent because azobenzene is the major product of the electrode reactions and it is stable to further coupling reactions.The oxidation mechanism is known to be two consecutive one-electron transfers resulting in a neutral radical. Two of these radicals can couple together to form azobenzene. The yield of azobenzene is increased by adding pyridine because its basic properties assist the removal of the second electron from the initial radical, and hence more azobenzene is produced. The minor products of the electrode reactions account for the polymerisation observed. In aqueous solution the electrode was poisoned least at pH values less than 1 using a glassy carbon electrode or greater than 11 using a carbon paste electrode. Presumably the formation of the moreANALYTICAL PROCEEDINGS, JANUARY 1986, VOL 23 29 easily polymerised products is suppressed under these condi- tions.Linear sweep voltammetry at a carbon fibre microelectrode for M aniline in dilute sulphuric acid showed a marked decrease in peak current for consecutive sweeps. This indicates that the follow-up chemical reactions are rapid, and these microelectrodes offer no advantage over conventional ones in this respect. Although acetonitrile gave less filming of the electrode, this solvent was not used in FIA because organic solvents tend to corrode the electrode and pump tubing, leading to poor long-term stability. FIA of aniline using a detector operated in the d.c. amperometric mode gave a good detection limit of lo-’ M, but the electrode was poisoned even at this level. Less poisoning was observed when the detector was operated in the pulse amperometric mode, but the detection limit was poorer by one order of magnitude.This may be attributed to the large charging current generated by the potential pulses. In this instance glassy carbon proved to be unsuitable as an electrode for the pulse amperometric detection of aniline at the low concentration range required, owing to the large residual currents generated. An investigation of other types of elec- trodes that might be more suitable for this method is required, as the long-term stability of the pulse mode of detection is good. The chtmically modified glassy carbon electrode was less susceptible to poisoning than a conventional one, but it was still significant and also the sensitivity was decreased by about 50%. The decrease in sensitivity would suggest that the ionic conductivity of the polymer film is fairly low. Aniline mole- cules must either diffuse through the film or through “pin-hole” imperfections to reach the electrode surface where electron transfer occurs. It is reasonable to assume that the predomi- nant mode of diffusion cannot be via the pin-holes, as the faradaic current would continue to decrease rapidly as poly- merisation occurred at the free electrode sites.This method is not suitable for application to on-line monitoring because the response is decreased to a similar degree as a conventional electrode over a period of time. Conclusion None of the methods investigated was completely successful at limiting the electrode poisoning. Pulse amperometry was the most promising method and a study of electrodes suitable for this technique is now in progress.Financial support from the Ministry of Defence is gratefully acknowledged. 1. 2. 3. 4. 5. 6. 7. 8. References Barek, J., Pacakova, V., Stulik, K., and Zima, J., Talanta, 1985, 32, 279. Koile, R. C., and Johnson, D. C., Anal. Chem., 1979,51,741. van Rooijen, H. W., and Poppe, H., Anal. Chirn. Acra, 1981, 130, 9. Hui, B. S.. and Huber, C. 0.. Anal. Chirn. Acra, 1982, 1 3 4 , 2 1 1 . Adams, R. N., “Electrochemistry at Solid Electrodes,” Marcel Dekker, New York, 1969. Dieker, J. W., van der Linden. W. E., and Poppe, H.. Talanta, 1979, 26. 511. Henriques. P. H., and Fogg, A. G., Analyst, 1984, 109. 1195. Fogg, A. G.. and Summan, A. M., Analyst, 1984, 109, 1029. In situ Complexation Chromatography for the Determination of Metal Ions Gerard Heneghan and Gordon G.Wallace” Department of Chemistry, University College Cork, Cork, Ireland In sifu complexation chromatography has proved useful for the determination of metal ions in recent years.’-10 With this method, a ligand is included in the chromatographic eluent. On injection of metal ions they form complexes in situ and are subsequently chromatographed. Some interesting work using the dithiocarbamate ligand and reversed-phase chromato- graphy has been reported.2-9 This ligand is particularly useful as it enables sensitive detection of the metal complexes to be carried out using either spectrophotometric or amperometric detection. Use of these thio-containing ligands allows monitor- ing of electrochemical oxidation processes for the metal complexes at potentials free from dissolved oxygen inter- ference. 2-6 However, the above method is limited in the number of species that can be determined owing to: (i) instability of some complexes on the chromatographic column; (ii) lack of chro- matographic selectivity for some metal complexes; (iii) excess ligand interference at the detection stage using either spectro- photome try or amperome try.In the course of this work we have been investigating ways to overcome the above problems. Results and Discussion Use of Other Derivatising Reagents Two other thio-containing ligands were considered: morpho- line dithiocarbamate (I) and ethyl xanthate (11). Present address: Chemistry Department, Wollangong University, Wollangong, NSW. Australia. I II The morph ligand exhibited no advantages over the pre- viously investigated diethyl (dedtc) or pyrrolidine (pyrr) dithiocarbamates.2-6 The long-term instability of morph in the chromatographic eluent is a disadvantage.The etxan ligand displays interesting electrochemical properties in that the dixanthogen dimer (a product of free ligand oxidation) is oxidised at more positive potentials than required to oxidise metal complexes such as those containing Co, Cd, Hg or Zn.” This should allow more efficient detection of these metals than when using dedtc or pyrr, whose dimers are oxidised at similar potentials to the Co, Cd, Hg or Zn complexes. To date, however, this advantage has not been realised because the metal xanthate complexes are very unstable in water- containing mobile phases.’] This instability problem may be overcome using in situ adduct formation (see later) or separations that involve non-aqueous solvents, such as aceto- nitrile on a suitable stationary phase.In Sifu Adduct Formation Other workers have demonstrated the ability of thio- containing metal complexes to form adducts with nitrogen- containing ligands.12-15 This phenomenon has been used to increase solubility,16 to improve stability17 and to adjust the chromatography18-19 of thio-containing ligands. Use of the30 1 ,lo-phenanthroline (phen) ligand was considered in this work. In the course of this work it was found that by including equimolar phen - pyrr in the chromatographic eluent the chromatography could be affected. In selected instances, for example the separation of Cu*+ and Ni2+, the selectivity factor was improved by including the phen ligand.11 Investigations into another system in which the chromato- graphic eluent contained phen and etxan in a 1 : 2 molar ratio indicated that the presence of phen markedly improved the on-column stability of the metal complexes. l 1 The presence of phen also affected the oxidation processes for some complexes, making them easier to oxidise.This in turn improved the amperometric detection levels for these metals. 11 In Situ Exchange Chromatography In solvent extraction with the dithiocarbamate ligand, previous workers have occasionally employed a labile metal complex as the reagent rather than the free ligand.20.21 This can be used to improve the stability of the reagent and/or the selectivity of the method.In the course of this work the Z n ( ~ y r r ) ~ complex was considered as a reagent for in situ complexation chromato- graphy. In addition to the above advantages the Zn(pyrr)* complex is oxidised at much more positive potentials than the free ligand and so the detection of species such as those containing copper, nickel, cobalt and lead could be achieved on a low background. Some problems were encountered with this method in that Z n ( ~ y r r ) ~ is retained on the stationary phase employed for separations. 1 1 Conclusions and Future Developments The concepts of in situ adduct formation and in situ exchange chromatography have been demonstrated to be feasible. These methods should prove useful in the development of new ANALYTICAL PROCEEDINGS, JANUARY 1986.VOL 23 chromatographic separations and should also prove useful in improving the detection of metal ions using in situ complexa- tion chromatography. 1. 2. 3. 4. 5. 6. 7. 8. 9. 10. 11. 12. 13. 14. 15. 16. 17. 18. 19. 20. 21. References Berthod. A., Kolosky, M.. Rocca. J . L.. and Vittori, O., Analusis, 1979, 7, 395. Bond, A. M., and Wallace, G . G.. Anal. Chem., 1981, 53. 1209. Bond, A. M., and Wallace, G. G., Anal. Chem.. 1982, 54, 1706. Bond, A. M.. and Wallace. G . G.. Anal. Chem., 1983.55.718. Bond, A. M . , and Wallace, G . G.. J. Liq. Chromatogr.. 1983, 6 , 1799. Bond, A. M., and Wallace. G. G., Anal. Chem., 1984, 56. 2085. Smith, R. M . , and Yankey, L. E.. Analvst. 1982. 107, 744. Smith, R. M., Anal. Proc., 1984, 21, 73.Smith, R. M.. Analyst, 1985. 110. 35. Hoffmann, B. W., and Schwedt. G., J. High Resolur. Chro- matogr. Chromatogr. Commun., 1982. 5 . 439. Heneghan, G.. and Wallace, G. G.. in preparation. Toropova, V. F.. Budnikov, G. K., Ulakovich. N. A.. and Viter. I. P., Zhr. Obshch. Khim., 1975. 45, 1359. Fayyaz, M. U.. and Grant. M. W.. Aust. J . Chem., 1977, 30, 285. Budnikov, H . C., Ulakovich, N. A.. and Postnova, L. V.. J . Electroanal. Chem., 1983, 154, 171. Malik, W. U., Bembi. R.. and Bhardwaj. V. K., J. Indian Chem. SOC. 1984, 61, 379. Coates, E., Rigg. B., Saville. B., and Skeltan. D., J. Chem. SOC. 1965, 5613. Schunck, W . , and Schwedt, G., Freseniris Z . Anal. Chem.. 1984, 318, 47. Soundararajan, G.. and Subbaigan, M.. Indian J. Chem., 1983. 22A, 402.Higgins, G. M. C., and Saville, B.,J. Chem. SOC., 1963, 2812. Wyttenbach. A., and Bajo. S . . Anal. Chem., 1975.47. 1813. Bajo. S . , and Wyttenbach, A.. Anal. Chem., 1979. 51. 376. Reversed-phase High-performance Liquid Chromatography of Plasma Melphalan and Chlorambucil: Comparison of Three Detection Methods C. G. Adair Department of Haematology, The Queen's University of Belfast, Institute of Pathology, Grosvenor Road, Belfast BT72 6BL and D. Thorburn Burns and M. Harriott Department of Analytical Chemistry, The Queen's University of Belfast, David Keir Building, Stranmillis Road, Belfast BT9 5AG Melphalan and chlorambucil are two cytotoxic alkylating agents currently used in the therapeutic management of several haematological malignancies. 1.2 Although the effectiveness of these drugs is unquestionable, several potentially serious side effects, primarily related to bone marrow toxicity, may occur, and therefore a knowledge of the pharmacokinetics of melpha- lan and chlorambucil may enable therapy to be optimised whilst safeguarding the patient from toxicity.To facilitate pharmacokinetic studies of these alkylating agents, a high- performance liquid chromatographic (HPLC) assay has been developed for melphalan3 and chlorambuci14 using UV detec- tion (Amax, 260 nm) with limits of detection of 5 and 10 ng ml-l, respectively. In an attempt to improve assay sensitivity fluorescence (Aex 260 nm: kern 350 nm) and electrochemical (oxidation potential 0.92 V) detection systems have been evaluated and the results compared with those obtained using UV detection.Experimental Instrumentation HPLC equipment from the Pye Unicam PU 4000 system (Pye Unicam, Cambridge) incorporated a dual reciprocating pump used with the following detectors: (a) PU 4020 variable- wavelength UV detector; (b) PU 4022 electrochemical detec- tor with wall jet flow cell consisting of a silver chloride reference electrode, glassy carbon working electrode and a stainless-steel counter electrode; and (c) Kratos FS 970 fluorescence detector. Chromatograms were recorded on a Pye Unicam CDP4 computing integrator with 100 mV input f.s.d. The column was Spherisorb ODS 5 pm (250 x 4.6 mm i.d.) (Phase Separations, Queensferry). The column and Rheodyne 7125 injection valve (Cotati, CA. USA) with a 200-p1 sample loop were mounted in a block heater (Jones Chromatography.Cardiff) and main-ANALYTICAL PROCEEDINGS, JANUARY 1986, VOL 23 31 tained at 40 "C. The guard column (3 cm) packed with Spherisorb ODS ( 5 pm) was also enclosed in the block heater. Chromatographic Conditions For melphalan determination. the mobile phase was a mixture of 80% methanol (Fisons, Loughborough), 20% water and 0.0135% (mlV) sodium dodecylsulphate (BDH Chemicals. Poole, Dorset) as counter ion. This was adjusted to pH 3.11 using dilute sulphuric acid. Chlorambucil resolution was unaffected by ion-pair HPLC. so the mobile phase could be simplified to a mixture of 80% V/V methanol and 20% VIV water. All materials and reagents were of HPLC grade and were filtered where appropriate with a 0.22-pm Millipore filter (Millipore.Bedford, MA, USA) prior to use. For both drugs the rate of flow of mobile phase was 1.3 ml min-1, with retention times of 3.2 min (chlorambucil) and 9.S rnin (melphalan). Sample Preparation and Extraction The sample preparation and extraction procedures were the same for both drugs. Whole blood (6 ml) was taken from a patient with an indwelling venous catheter and stored (2 "C) in blood collection tubes with lithium heparin anticoagulant. After centrifugation (1500 g for 10 min at -6 "C), 3-ml aliquots of plasma were removed and the macromolecular components precipitated with 132 pl of cold perchloric acid (2 "C). The mixture was vortex mixed for 5 min prior to centrifugation (1300 g for 15 min at -6 "C) and the clear supernatant was removed and passed through a Sep-Pak Clx reversed-phase column (Waters Assoc..Taunton, MA. USA), which was subsequently washed with 10 ml of 15% (VIV) methanol in water (2 "C). Melphalan and chlorambucil were eluted from the Sep-Pak with 2 ml of cold methanol (-20 "C). The eluate was stored at -20 "C prior to injection of 200-p1 aliquots on to the column. Results and Discussion Using this chromatographic and extraction system. ultraviolet. fluorescence and electrochemical detectors were compared for sensitivity. The observed limits of detection (2a) are given in Table 1. It is evident that UV detection is the system of choice for both alkylating agents. owing to the low level of base-line noise associated with the UV detector. In contrast. both fluorescence and electrochemical detection had high levels of background noise.The primary cause of background noise associated with the latter system was pulsation from the pump. Calibration and recovery data for the two drugs based on UV detection were as follows. For melphalan. the coefficient of variation for ten concentra- tion duplicates (10-2000 ng ml-1) extracted from plasma was 2.67% and the correlation coefficient for the calibration graph was 0.992. The within-batch variability for standard solutions of 20,200 and 2000 ng ml-1. each chromatographed ten times. was 6.7%. 1.5% and 0.5%. respectively. The recovery of the drug from plasma was approximately 60% with a limit of detection of 5 ng mi-I. -1 I I L 2 4 6 8 1 0 Time min Fig. 1. melphalan administration Chromatogram from a patient with multiple myeloma before For chlorambucil, the coefficient of variation for ten concentration duplicates (20-1200 ng ml- I ) extracted from plasma was 0.71% and the correlation coefficient for the calibration graph was 0.998.The within-batch variability for standard solutions of 100, 500 and 1000 ng ml-1. each chromatographed ten times, was 1.67%. 0.96% and 0.18"/0. respectively. The recovery of the drug from plasma was 6(J'% with a limit of detection of 10 ng ml-1. Typical chromatograms are shown in Figs. 1-4. Table 2. Variation in patient plasma levels of melphelan and chlorambucil Drug plasma level (mean k s.d. ling ml I Time after drug administrationih Melphalan Chloram bucil 3 4 5 3 5 k 17 197 2 73 20 k 1 0 1 1 Y 2 4 7 I2k-l 67 2 33 Pharmacokinetic data on melphalan and chlorambucil were obtained during clinical studies.Log drug plasma concentra- tion L'ersiis time profiles were obtained for 1 1 patients receiving melphalan and 8 patients receiving chlorambucil. A typical ~ Table 1. Summary of detection limits Instrumentation Limit of detectioning ml I Detection system 0 p t i m i s a t i o n PU 8 100 PU 4020 SFM 25 FS 070 i.,,, 3.50 nm polarograph PU 3022 Detection Con d i t i on s Me I p h a 1 an C h I o r a m b u ci I Ultraviolet . . . . Pye Unicam Pye Unicam i.,,,,,, 260 nm 5 10 Fluorescence . . Kontron Kratos LCx 760 nm: 1' 21 Electrochemical . . Tacussel Pye Unicam Oxidation. 0.92 V 30 2332 ANALYTICAL PROCEEDINGS, JANUARY 1986, VOL 23 graph is shown in Fig. 5 . From such a graph the elimination rate constant for the drug in a given patient can be calculated by regression analysis of the terminal data points (i.e., the plasma concentrations at 2 , 3 , 4 and 5 h).The elimination rate constant can be accurately estimated only if plasma concentrations are v 1, c m m c (1 - - d I I I I I 0 2 4 6 8 1 0 Fig. 2. Chromatogram from Fig. 1 patient 90 min after oral administration of 12 mg of melphalan (peak height = 100 ng ml- l ) Time/m in Lt 0 2 4 6 Timeimi n Fig. 3. leukaemia before chlorambucil administration Chromatogram from a patient with chronic lymphocytic determined over a sufficient time period, which necessitates accurate measurement of very low levels of the drugs. The inter-patient variation in plasma levels of melphalan and chlorambucil measured at the terminal points is show'n in Table 2.From this clinical study it is evident that only UV detection permits the measurement of melphalan over a sufficient time period to quantify the elimination rate constant in all patients. 0 2 4 6 Ti mei mi n Fig. 4. Chromatogram from Fig. 3 patient 70 min after oral administration of 30 mg of chlorambucil (peak height = 272 ng ml-I) In these clinical studies we have attempted to quantify the relationship between drug elimination and renal function by regression analysis of the elimination rate constant with renal function. Chlorambucil elimination was shown to be indepen- dent of renal function ( N = 8, r = 0.03, P = 0.96). However, melphalan elimination correlated positively with renal function ( N = 11, r = 0.83, P = 0.003).Consequently, higher and 0 60 120 180 240 300 Time after administrationimin Fig. 5. receiving 12 mg of melphalan by oral administration Log (plasma concentration) versus time for a patient afterANALYTICAL PROCEEDINGS. JANUARY 1986, VOL 23 33 potentially toxic levels of melphalan were obtained in patients with poor renal function. This gives a pharmacokinetic basis to previous clinical observations5 that have shown that a patient receiving melphalan, and who had a poor renal function, suffered bone marrow toxicity more frequently than patients with normal renal function. 3. 1. 2. We thank Professor J. M. Bridges, professor P. F. D'Arcy, Dr. work. This work was supported by the Northern Ireland 4. 5 . A. D. Crockard and Dr. 2. R. Desai for their interest in this Leukaemia Research Fund.References Dorr, R. T., and Fritz. W. L., "Cancer Chemotherapy Handbook," Kimpton. London, 1980. Chapter 10. p. 507. Dorr, R. T.. and Fritz, W. L., "Cancer Chemotherapy Handbook," Kimpton, London, 1980, Chapter 10. p. 303. Adair, C. G., Burns. D. T., Crockard, A. D., Desai, Z. R., and Harriott, M., J . Chromatogr., Biomed. Appl., 1984, 336. 429. Adair, C. G.. Burns. D. T., Crockard. A. D., and Harriott, M., J. Chromatogr., Biomed. Appl., 1985, 342. 447. Cornwell, G. G., Pajak. T. F., McIntyre, 0. R., Kochwa, S.. and Dosik. D., Cancer Treat. Rep., 1982, 66. 475. Modified Sample Preparation and Chromatography for the Separation of Human Bile Acid Conjugates G. R. Campbell Department of Surgery, The Queen's University of Belfast, Belfast BT9 5AG and M.Harriott and D. Thorburn Burns Department of Analytical Chemistry, The Queen's University of Belfast, Belfast BT9 5AG Bile acids are synthesised in the liver, where they are converted into bile salts by reactions with nitrogenous bases such as glycine and taurine. When abnormalities occur in the hepato- bilary system, the levels and patterns of the bile salts may be affected. A method has been developed for the determination of the conjugated bile acids in human bile without the previously necessary complicated hydrolysis. 1 Past procedures have been time consuming and in some instances have produced artefacts.1-3 In the method described here, the bile sample, prior to chromatography, is divided into unconjugated free bile acids, glycine and taurine conjugated groups using an ion-exchange column that incorporates a lipophylic gel, piperidinohydroxypropyl-Sephadex LH-20. Experimental High-performance Liquid Chromatography The chromatographic system consisted of a Waters Model M-45 pump, a Waters pBondapak CIS column (25 cm X 4 mm i.d.stainless steel) (particle size 10 pm), a Rheodyne Model 1257 loop injection value (100 pl) and a Uvicord-S fixed- wavelength detector set at 206 nm, absorbance range 0.05 absorbance unit f.s.d. Peaks were quantified using a Hewlett- Packard 3390A computing integrator. Chemicals The standard sample consisted of glycine and taurine conju- gates of cholic, chenodeoxycholic, deoxycholic and lithocholic acids (Sigma). All other chemicals, unless specified otherwise, were of analytical-reagent grade.Piperidinohydroxypropyl- Sephadex LH-20 (PHP-LH-20) was prepared form Sephadex LH-20 using epichlorohydrin (BDH Chemicals) and boron trifluoride etherate (Sigma) as catalyst .4 Piperidine (BDH Chemicals) residues are bonded to this matrix and thus provide the ion-exchange characteristics.5 The functional groups of this residue react with the hydroxy entities of the bile acids. The mobile phase consisted of methanol (HiPerSolv, BDH Chem- icals) - water mixtures. Octanesulphonic acid (Sigma) was used as a counter ion. Establishment of Chromatographic Conditions for Glycine and Taurine Conjugates This method was developed in order to provide a simple, accurate and inexpensive technique for the determination of the levels of glycine and taurine conjugated bile acids.Experiments were aimed at obtaining optimum separations of both glycine and taurine conjugates of chenodeoxycholic and deoxycholic acids, which were known from previous work to be the most difficult bile acid conjugates to resolve. Detection was achived by UV monitoring at 206 nm. At higher wavelengths little if any response to bile acid structures is obtained. The sensitivity of detection improves as wavelength is reduced.6 The effects of aqueous phase concentration (25-33% WV), pH range (2-6) and counter ion all affected the resolution of the bile acids. The optimum mobile phase was found to be 72.5% V/V methanol - 27.5% V/V aqueous phase, the aqueous phase being buffered to pH 6 using McIlvaine's buffer.' The mobile phase contained an ion-pairing reagent; octanesulphonic acid appeared to be most suitable (150 mg per 500 cm3 of mobile phase).The optimum flow-rate (0.6 cm3 min-1) was deter- mined from plotting the height equivalent to a theoretical plate (HETP) against flow-rate (cm3 min-1). The mobile phase was vacuum filtered using a 0.45-pm filter (Millipore), degassed by ultrasonication and thereafter gently purged with helium. Separation of Bile Salts into Unconjugated Free Acid, Glycine and Taurine Conjugated Groups Before evaluation of the conjugated bile acid pattern by HPLC, the bile sample was subjected to ion-exchange separa- tion on PHP-LH-20, which interacts with the hydroxy groups on the bile acids. A 100-pl volume of the bile sample was diluted in 2 cm3 of phosphate buffer (pH 7.0) and the mixture was washed through a CI8 Bond-Elut cartridge using 8 cm3 of water followed by 2 cm3 of 1.5% V/V methanol.The bile fraction was eluted by washing through with 4 cm3 of 100% methanol. An aliquot (1 cm3) was placed on a column of PHP-LH-20 (36 x 6 mm i.d.).7 Group separation was achieved by eluting the column first with 8 cm3 of 0.1 M acetic acid in 90% methanol, secondly with 8 cm3 of 0.2 M formic acid in 90% methanol and finally with 8 cm3 of 0.3 M acetic acid - potassium acetate buffer (pH 6.5).4 These solutions eluted the free acids, glycine conjugated acids and taurine conjugated acids, respectively. Each fraction was collected and evaporated to dryness. The glycine conjugated residue was taken up in 1 cm3 of mobile phase and 100 pl were injected into the HPLC system.The taurine conjugated fraction was taken up in 1 cm3 of water and passed through a CI8 Bond-Elut cartridge to remove inorganic cations, especially potassium, from the eluent. The taurine fraction was washed through using 2 cm3 of 90% methanol and 100 p1 were injected into the HPLC system. The free acid fraction was discarded. Results and Discussion It was found that the recovery of glycine conjugated bile acids improved from 53% to 92 k 1.7% (n = 4) and the recovery of34 ANALYTICAL PROCEEDINGS, JANUARY 1986. VOL 23 Cholate I lo.oo' A Cholate I Chenodeoxycholate Chenodeoxycholate Time - Fig. 1. flow-rate, 0.6 cm3 min- 1 : wavelength. 206 nm; concentration. 10 pg per 100 PI: and chart speed, 1 crn per 5 rnin Typical chromatograms of ( a ) taurine and ( b ) glycine conjugates.Conditions: mobile phase, 72.5'10 of aqueous buffer (pH 6.0): taurine conjugates increased from 67% to 90.5 k 3.04% (n = 4) as the length of the PHP-LH-20 column was increased from 17 to 36 mm. Calibration graphs were linear over the range 0-10 g per 100 p1 for each conjugate examined. Typical chromatograms of prepared mixtures of bile acid conjugates are shown in Fig. 1. Separation of bile components into unconjugated free acids, glycine and taurine conjugated bile acids appears to be a suitable method for determining the level and profile of these compounds in biological samples. The results obtained com- pare favourably with those reported by other workers in this field'-*; if differentiation into these fractions is not carried out overlap occurs, resulting in poor resolution.Alternative procedures rely on ion pairing, micro-HPLC and gradient chromatography.* These methods often require low pH, which can be detrimental to column efficiency.9 The determination of bile acids in biological fluids is important as abnormalities in their levels may reflect a functional disorder of the liver. In the colon bile acid transformation may be associated with cancer of this organ. Similarly, bile acid reflux into the stomach may lead to gastric cancer .6.8 References 1. 2. 3. 4. 5 . 6. 7. 8. 9. Goto. J.. Kato, H.. Saruta. Y.. and Nambara. T.. J. Chro- rnarogr.. 1981. 226, 13. Linnet, K.. Scand. J. Clin. Lab. Invest.. 1082. 42, 455. Mingrone. G . . Greco, A. V . . and Passi.S . . J. Ciiromutogr., 1980, 183, 277. Alme, B., and Nystrom, E., J. Chrornatogr., 1971. 59, 45. Goto, J.. Hasegawa. M., Kato, H.. and Narnbara. T.. Clin. Chim. Acta, 1978. 87, 141. Parris. N. A., J. Chrornatogr., 1977. 133. 273. Goto, J.. Kato, H.. Saruta. Y.. and Nambara. T.. J. Liy. Chrornatogr., 1980, 3. 991. Miyagi. H., Miura. J., Takata, Y . . Kamitake. S . . Ganno. S . . and Yamagata, Y . , J . Chrornatogr., 1982. 239. 733. Ishii, D.. Murata, S . . and Takeuchi. T.. J. Chroniritogr.. 1983. 282, 569. Voltammetric Determination of Inorganic Lead and Dimethyl- and Trimethyllead Species in Mixtures Patrick J. Hayes and Malcolm R. Smyth School of Chemical Sciences, National Institute for Higher Education, Glasnevin, Dublin 9, Republic of Ireland The presence of lead in the environment is well documented.'-3 One of the main sources of lead comes from exhaust emissions from cars, and most of the lead emitted from this source is in the form of inorganic salts, although some is in the alkylated form.On release into the environment, these alkylated compounds may degrade by various pathways to inorganic lead.4 A serious consequence of this degradation is the possible dissolution of dialkyl- and trialkyllead species (in addition to inorganic lead) in aquatic systems. In this way they can cause deleterious effects in both freshwater and marine biota.5 In general, alkyllead species are more toxic than inorganic lead. Obviously, analytical procedures are required in order to quantify these various lead species. Methods for the determi- nation of tetraalkylleads are well established.68 but fewer methods exist for the determination of dialkyl- and trialkyllead species.Some of the existing methods are based on complexa- tion and spectrophotometric determination,Y-ll but are not applicable to trace metal studies owing to their lack of sensitivity. More sensitive methods involving graphite furnace atomic absorption spectrometry12 and gas chromatography13 have been developed. In this paper we demonstrate how voltammetric methods may be applied to the speciation of inorganic lead and dialkyl- and trialkyllead compounds. Experimental Dimethyllead dichloride and trimethyllead chloride were supplied by Associated Octel Ltd. and were used without further purification. All other reagents used were of analytical- reagent grade.Stock solutions of the various lead species were prepared in Millipore-grade water and stored in the dark. An acetate buffer (pH 3.30). composed of 0.2 M glacial acetic acid - 0.2 M sodium acetate in Millipore-grade water. was used. Differential-pulse polarography (DPP) and differential- pulse anodic-stripping voltammetry (DPASV) studies were performed using an EG & G Princeton Applied Research Corp. (PARC) Model 174A polarographic analyser. This was used in conjunction with an EG & G PARC Model 303 SMDE. The cell consisted of a dropping-mercury working electrode. a platinum counter electrode and an Ag - AgCl reference electrode. For DPASV studies, a hanging mercury drop electrode (HMDE) was used in conjunction with an EG & GANALYTICAL PROCEEDINGS.JANUARY 1986. VOL 33 35 PAKC Model 305 stirrer. Prior to polarographic investiga- tions. solutions were degassed with oxygen-free nitrogen for 4 min. All pI4 measurements were made using a Philips PW 9410 digital pH meter with an Orion combination pH electrode. Results and Discussion Preliminary DPP studies of inorganic lead and dimethyl- and trimethyllead species were undertaken in acetate buffer (pH 3.30). Inorganic lead is characterised by a single reduction peak at -0.40 V, which represents a single two-electron reduction step. For dimethyllead dichloride the reduction process is more complicated. Overall. the reduction involves two electrons, although the differential-pulse polarogram contains four peaks (see Fig. 1). The peaks at -0.34 and -0.59 V appear to be adsorption-controlled processes. It has previously been shown that for trimethyllead chloride the first peak at -0.65 V represents a single electron reduction step whereas the second wave at - 1.12 V involves a three-electron reduction.1J (Epl) over the range -0.6 to -1.1 V.However, the peak current for these species increased at more negative E,, values. Trimethyllead chloride was not detected below -0.7 V. Beyond this value. the peak current increased with more negative Epl values. By choosing a plating potential of -0.6 V, 2.0 4 c t g 1.0 3 0 I 1 Fig. 2. Dependence of current on plating potential ( € p , ) for A. 1 x 1U-' M Pb(NO,),; and B. 1 x 10 5 M (CH,), PbCI, in 0.2 M acetic acid - sodium acetate buffer. pH 3.30. containing 1.45 x 10-5 M EDTA.Conditions: plating time 4 rnin; scan rate 2 mV 5 1: modulation amplitude 100 mV J 0 -0.5 -1.0 E V VS. Ag - AgCl Fig. 1. Differential-pulse polarographic behaviour of ( A ) 1 x 10 A hf Pb(N03),; (B) 1 X 1 0 -I M (CH,),PhCI,: and (C) 1 X 1 0 - ' h! (CH),PbCI in 0.3 hi acetic acid - sodium acetate buffer, pH 3.30. Conditions: scan rate. 3 mVs 1 ; and modulation amplitude 25 rnV Studies i n binary mixtures were carried out using DPP and reveal the possibility of measuring inorganic lead in the presence of trimethyllead chloride. In contrast, when dimethyllead dichloride is present with inorganic lead or trimethyllead chloride, overlapping of peaks occurs. making it impossible to measure dimethyllead dichloride in the presence of the other two lead species by DPP.Because of the aforementioned interference problem. the use of complexing agents as a means of separating the species was investigated. EDTA was found to be a suitable complexing agent, addition of EDTA to a solution containing inorganic lead resulting in a second peak at -0.88 V, arising from reduction of the Pb - EDTA complex. Addition of sufficient EDTA shifts the reduction peak for inorganic lead completely to -0.88 V. In contrast, reduction of the dimethyllead - EDTA complex takes place at -0.45 V. Here the previous over- lapping of these species has been eliminated by using EDTA. Trimethyllead chloride does not complex with EDTA. Although DPP is a sensitive analytical technique, it is not sufficiently sensitive to determine trace metals at their environ- mental concentrations.Therefore, DPASV was used in the further development of the above speciation scheme because of its enhanced sensitivity over DPP. Studies by DPASV revealed that the peak potential for the re-oxidation step is the same for all three lead species. The oxidation step is as follows: Pb( Hg)-+Pb'+ + 2e- ( + Hg) The peak currents for inorganic lead and dimethyllead dichloride did not show any dependence on plating potential I -/ 1- j &/ -1- 0 -0.6 - 0.7 - 0.8 -0.9 -1.0 E,, V VS. Ag - AgCl only inorganic lead and dimethyllead dichloride can be measured. With an EPI value greater than -0.7 V all three species can be measured. Addition of EDTA to solutions containing mixtures of inorganic lead and dimethyllead dich- loride effectively precludes the measurement of inorganic lead when a plating potential of -0.6 V is used (see Fig.2 ) . Based on these results, a speciation scheme is proposed in Table 1. Table 1. Proposed scheme for simultaneous voltammetric deterrnina- tion of inorganic lead. dimethyllead and trimethyllead species Initial plating Step potential - 1 1 .-I Lead species measured Inorganic lead. dimethyllead dichloride. trimethyllead chloride 9 L Inorganic lead, I dimethyllead dichloride -0.6 Addition of EDTA 3 -0.6 4 Dimethyllead dichloride 1 . 2. 3 . 4. 5 . 6 . 7. References Bryce-Smith, D.. Trends Anal. Chem.. 1982. 1. 199. Chau, Y . K.. and Wong. P. T. S . . NRS Spec. Publ., 1981. No. 618. 65. Harrison. R. M.. and Laxen. D. P. H.. €iit.iron. Sci. Technol.. 1978. 12. 1383. Jarvie. A. W. P.. Markall. R. N . . and Potter. €1. R.. Oiviron. Res.,1981.25.241. Chau. Y . K.. Wong, P. T. S . . Kramar. 0.. Bengert. G. A.. Crue. R. B.. Kinrade. J . 0.. Lye. J . , and Van Loon. J . C.. Bull. Environ. Contam. Toxicol.. 1980. 24, 265. Chau. Y. K., Wong. P. T. S . . Bengert, G. A . . and Krarnar. 0.. Anal. Chem.. 1979. 51. 186. Sirota, G. R . . and Uthe. J . F . . Anul. Chem., 1977. 49. 833.ANALYTICAL PROCEEDINGS, JANUARY 1986. VOL 23 36 8. 9. 10. 11. Chau, Y. K . , Wong, P. T. S . , and Goulden. P. D., Anal. Chem. Acta, 1976, 85, 421. Pilloni, G., and Plazzogna, G., Anal. Chim. Acta, 1966, 35, 325. 1981,53, 1336. Schmidt, U . , and Huber, F., Anal. Chim. Acta, 1978,98, 147. Aldridge, W. N., and Street, B . . Analyst, 1981. 106, 60. 12. 13. 14. De Jonghe, W. R. A., Van Mol. W. E . , and Adams, F. C., Anal. Chem., 1983, 55, 1050. Estes, S . A., Uden, P. C.. and Barnes. R. M.. Anal. Chem., Colombine, M. P., Fuoco, R., and Papoff. P., Ann. C'him. (Rome), 1982, 7 2 , 547. ROYAL SOCIETY OF CHEMISTRY ANNUALCHEMICALCONGRESS University of Wawick April 8-11,1986 Call For Posters The Analytical Division's Symposium at the Annual Congress is entitled "New Spectroscopic Sensors and Techniques." It is hoped to run two poster sessions that reflect this theme, one dealing with atomic and the other with molecular spectrometry. Titles of the oral presentations include "Fibre Optic Probes for Chemical Sensing", "Atomiser Source lCPs in Atomic Fluorescence Spectrometry," "ICP - MS" and "Models, Measurements, Methods and Machines in Analytical Spectrometry." Other topics include, FTUV, NIRR, OTTER and LIDAR. Please send a title and a brief abstract of your poster (around 100 words) to Dr. D. I. Coomber, c/o Analytical Division, The Royal Society of Chemistry, Burlington House, Piccadilly, London W1V OBN. Please make your submissions before February 14th, 1986. ~ Nitrification Inhibition in the ~ Treatment of Sewage ~ Compiled by M. Richardson, Thames Water The purpose of this book is to provide a rapid reference to information on those chemicals which have been reported in the literature to produce inhibition of nitrification to sewage works processes, etc. Some information is also included where chemicals are known to inhibit anaerobic processes. The first section of this book following the introduction consists of data relevant to nearly 300 chemical compounds including information relating to substance names, molecular weights, molecular formulae, degree of nitrification inhibition, concen- tration and reference. This information has been collated from sources indicated in the reference section. The final section of this book lists in ascending mass number the chemicals described in the earlier section. It will be of particular value when gas chromatography - mass spec- trometry is undertaken on trader's discharge, sewage, river or other water samples. This book should be an invaluable source of information for scientists and trade effluent officers within Water Undertak- ings, Government Departments, etc., and equally for environ- ment and safety officers within manufacturing industry who may not have ready access to, or knowledge of, the potential risks associated within the discharge of their waste chemicals. Contents: Chemicals with effects on nitrification in sewage treatment works; References; List of compounds in ascending mass number. Softcover 108pp ISBN 0 85186 596 8 Price f 15.00 ($26.00) RSC Members f 10.00 ORDERING RSC Members should send their orders to: The Membership Officer, The Royal Society of Chemistry, 30 Russell Square, London WClB 5DT. Non-RSC Members should send their orders to: The Royal Society of Chemistry, Distribution Centre, Blackhorse Road, Letchworth, Herts SG6 lHN, England. Infoeation Semces

ISSN:0144-557X    

DOI:10.1039/AP9862300018

出版商:RSC    

年代:1986

数据来源: RSC

摘要:

ANALYTICAL PROCEEDINGS, JANUARY 1986, VOL 23 37 Equipment News Spectrometer Options and Accessories Several new options are announced for the makers’ 1700 Series single beam Fourier transform infrared spectromet- ers. The standard Model 1750 now scans to 5000 cm-1 in the near infrared and the wavelength range of the Model 1710, which scans to 4400 cm-1, is extended to 5000 cm-1 when this instrument is used with the Model 3600 Data Station. A caesium iodide beam splitter is now avail- able as an alternative to the standard potassium bromide beam splitter; it allows the Model 1710 to scan from 4300 cm-1 to 220 cm-1 and the Model 1750 to scan from 5000 cm-1 to 220 cm-1. An accessory is available to permit emission measurements to be made on samples placed in the sample compartment of 1700 Series instruments.Options already avail- able include a GC - IR accessory for on-the-fly capability, a sample shuttle which gives double beam operation, liquid nitrogen cooled MCT detectors and diffuse reflectance and microscope acces- sories. Perkin-Elmer Ltd., Post Office Lane, Beaconsfield, Buckinghamshire HP9 1QA. Gas Chromatography - Infrared System The Digilab FTS-60 system has been designed for high-performance capillary gas chromatography - Fourier transform infrared work, especially with the Hew- lett-Packard 5890A gas chromatograph, but interfaces are available for other instruments. The FTS-60 offers the same on-the-fly sensitivity as the makers’ FTS- 80 system and features a permanently mounted GC accessory using the external collimated beam, minimum length fused silica transfer lines, no loss of GC resolu- tion through the “all-glass” light pipe, a high sensitivity MCT detector and return of the GC eluents to the GC flame ionisation detector.Polaron Equipment Ltd., 53-63 Green- hill Crescent, Watford Business Park, Watford. Hertfordshire WDl 8QS. Gas Chromatographs Two new members of the 8000 Series of instruments, based on the Model 8300, are announced. Standard features of the 8000 Series include automatic bleed com- pensation for temperature programmed operation; this single column - detector system provides correction for column bleed after a single calibration run. At the end of an analysis the data handling system displays the result as a fully edited report. The Model 8400 is a single detec- tor type instrument and is available as 8410, essentially for packed column oper- ation, and as 8420, for dedicated capillary column operation.the Model 8500 is for single or dual detector operation and features an extensive range of detector combinations from the six types available, packed and capillary injectors installed simultaneously, the ability to switch between two detector signals with auto- mated bleed compensation applied to each signal independently, and a range of pneumatics modules for optimum control of carrier and detector gases. Perkin-Elmer Ltd., Post Office Lane, Beaconsfield, Buckinghamshire HP9 1QA. Chromatography Accessories Sci-Vi, a system which uses a specially designed crimp top vial holding up to 300 pl of sample and fitting into a sleeve designed to fit specific auto-sampler vial holders, can now be used with the Hew- lett-Packard Model 1090 autosampler.Also available are 20 mm steel crimpers for standard HS vials and a 20-mm de- capper. Chromacol Ltd., Glen Ross House, Summers Row, London N12 OLD. Detector for HPLC AMOR is an amperometric detector for HPLC offering high sensitivity (sub- picogram) and a short stabilisation time. The working electrode is constructed so that the edge of the glassy carbon material is extraneous to the cell, thus reducing noise and drift. The cell is insulated and electrically shielded and its volume can be varied by applying different poly- fluoroethylene spacers (0.4-1 .O pl). There is an auto-zero function and a sensitivity range of 0.1-1000 nA.The potential setting may be from -2 to +2 V. Spark Holland BV, P.O. Box 388,7800 AJ Emmen, The Netherlands. Accessories for Capillary Chromatography Capillary starter kits for on-column injec- tion allow chromatographers to convert packed gas chromatographs within 2 h and to go back to packed columns within 5 min. Two kits are offered, both contain- ing an on-column injector, a fused silica chemically bonded column, a make-up tee, a set of ferrules, a syringe for on-column injection, a set of wrenches and instructions. Kit A includes a basic flow unit 101 while Kit B contains a 102 unit. In addition to the established 10 m fused silica columns with an inside diameter of 0.53 mm, two additional lengths have been introduced, 25 m and 50 m, with plate numbers up to 37 000 and 75 000, respectively. All three columns are available with chemically bonded phases CP-Si1 5 CB, CP-Si1 8 CB, CP-Sil 19 CB and CP-Wax 52 CB.Chrompack UK Ltd., Unit 4, Indescon Court, Millharbour, London E 14 9TN. Electrophoresis Systems DNA/RNA electrophoresis preparation packages are available and can be tailored to suit individual users’ needs. They can include a variety of transilluminators, camera systems, films, safety equipment, gel-producing apparatus, gel driers and power supplies. Ultraviolet Products, Science Park, Milton Road, Cambridge CB4 4BN. Liquid Scintillation Vials -- Lumac Click Vials feature a click-grip cap closure ensuring a tight cap-to-vial seal, thus eliminating leakage problems and providing a safer counting environment. They are compatible with all liquid scintil- lation counting systems currently in use.May and Baker Ltd., Liverpool Road, Eccles, Manchester M30 7RT. Titrators The DL20 CompactTitrator is a multi- purpose titrator. In the standard mode titration stops when an equivalence point is detected. In more complex reactions, such as the determination of sulphuric acid and citric acid in cider, another mode can be selected to allow all equivalence points up to the maximum volume to be determined. TAN and TBN can be deter- mined according to the kinetic conditions defined in ASTM Standard Methods D664 and D2896, using special programs. A special program is also available for the determination of p- and m-values. The DL40RC MemoTitrator is designed for use where there is a requirement to switch frequently between several unrelated methods.It offers storage of up to 2538 ANALYTICAL PROCEEDINGS, JANUARY 1986, VOL 23 complete titration methods based on up to 12 defined reagents. The DL18 Karl Fischer Titrator gives precise results €or water concentrations, ranging from low p.p.m. up to 100%. MSE Scientific Instruments Ltd., Manor Royal, Crawley, West Sussex RHlO 2QQ. Infrared Analyser The IRGA 320 non-dispersive infrared analyser continuously measures the con- centration of one gas in a mixture and provides wide measuring ranges for many gases and vapours including CO, COz, CH4, N 2 0 , hydrocarbons and refrigerant gases. It can be calibrated for two independent measuring ranges with a dual linear scale meter. Repeatability is +I% of full scale deflection or better, linearity is k2Y0 or better and response time is typically less than 10 s for most samples.May and Baker Ltd., Liverpool Road, Eccles, Manchester M30 7RT. Water Analysis By using existing instruments and inter- facing them with the SAC80 autosampler via an interface box (Water Measuring Box) Radiometer have extended the number of automated measurements that may be made at one time. Thus, the direct measurements which can be carried out are conductivity and pH, or conductivity and direct concentration. Indirect methods offered are conductivity and titration to one end-point, or conductivity and standard addition - Gran plot, or conductivity, pH and titrations. A new analysis system, part of the Radiometer 800 Series, is capable of measuring pH and conductivity, and performing titra- tions automatically on up to 20 samples in a batch, all simultaneously.A typical Sieger IRGA320infrared analyser The IRGA 320 is packaged in sealed, dust-proof units, which can be purged for electrically hazardous atmospheres. Options include a digital display with auto zero unit, a dual level alarm and full microprocessor control through which sample information can be converted to customer defined parameters. Sieger TPA Ltd., 34 Tresham Road, Orton Southgate, Peterborough PE2 OSE. Volumetric and Buffer Concentrates The Volucon volumetric range of 23 concentrates are supplied in six-ampoule cartons and produce standard solutions, usually 0.1 N , accurate to within +0.2%. The Volucon buffer range of 12 concen- trates, covering pH values 1-12, are supplied in single ampoule cartons and produce standard solutions accurate to within k0.05 pH unit, with the majority at k0.02 pH units.application in water analysis is the simul- taneous determination of alkalinity, pH and conductivity . V. A. Howe and Co. Ltd., 12-14 St. Ann's Crescent, London SW18 2LS. Industrial Analyser Hamilton's AMICA automatic industrial analyser is of modular design and each unit can be interfaced with other makers' equipment through an RS232C link. The autosampler and liquid processing unit can, for example, pick up samples, dilute them with reagent and dispense them prior to measurement by some type of instrument. V. A. Howe and Co. Ltd., 12-14 St. Ann's Crescent, London SW18 2LS. Precision Thermometer The F25 gives a stable measurement for one or two PTlOO resistance probes, displaying both absolute and differential temperatures.Its absolute accuracy is k0.025 "C with user-selectable resolu- tions of 0.01 and 0.001 "C over the full temperature range of -100 to +450 "C. This range may be extended by using alternative probes. Automatic Systems Laboratories, Saxon Street, Linford Wood, Milton Keynes, Buckinghamshire MK14 6LD. Heating Mantle Available in sizes from 50 ml to 3 1, Labsafe is a flask heater in which the electrical connections and element are totally encapsulated in ceramic material which has a special coating to render the mantle spill-proof. The outer surface can- not overheat or become too hot to touch. The mantle, which carries a 5-year gua- rantee, can be rinsed under clean water.Isopad Ltd., Isopad House, Shenley Road, Borehamwood, Hertfordshire WD16 1TE. Centrifuge The Heraeus-Christ Minifuge T is a ref- rigerated, bench-top centrifuge with mi- croprocessor control and serial interface for connection to a personal computer. It features a memory for 32 user programs as well as 9 selectable acceleration profiles and 10 braking profiles. It has a tempera- ture range from -19 to +39 "C. V. A. Howe and Co. Ltd., 12-14 St. Ann's Crescent, London SW18 2LS. Clinical Chemistry System The DU Clinical Chemistry System is a low reagent volume flow cell instrument, computer based for automatic operation and final answer calculation for more than 50 chemistries and the makers' recently introduced Epsilon series of immunoas- says.A single unit contains analyser, flow cell, temperature control facilities and keyboard, and a printer which generates large format printouts sits on top. TheANALYTICAL PROCEEDINGS, JANUARY 1986, VOL 23 39 TECHNIQUES FOR THE AUTOMATED OPTlMlSATlON OF HPLC SEPARATIONS by I.C. Berridge, Pfizer Central Research, Sandwich, UK High-performance liquid chromatography (HPLC) is now a widely applied and established analytical technique used throughout industry and research. As a tool which the analyst counts as a standard method, it is appropriate that with the increasing management of laboratory methods via microcomputers, the linking of HPLC to such systemsshould be considered. The author does not intend to provide a fully comprehen- sive review of all the possible options which are available; but rather to assess the different approaches and explore the potential for further improvements.The aim i s to improve the efficiency of HPLC in producing analytical results by optimising the sensitivity, selectivity and speed of separation - a selection of procedures and techniques are drawn together and a wide range of approaches are considered. “HPLC” will enable readers wishing to purchase equipment and/or software for automated optimisation to consult this single reference source with confidence for comparison of the available approaches. 0471 90861 4 208PP December’85 f 22.00/$37 .OO SELECTED METHODS OF TRACE METAL ANALYSIS Biological and Environmental Samples by J.C. Van loon, University of Toronto This monograph focuses on the analysis of trace metals in biological tissues and environmental materials. in most cases, procedures are so completely covered that no other source is necessary to finish an analysis.included are chapters on such important topics as determination techniques, basic materials, sample preparation, botanical and zoological samples, and cli n ica I samples. Series: Chemical Analysis, Volume 80 0471 89634 9 38OPP October’85 €56.25/$73.15 MASS SPECTROMETRY OF HETEROCYCLIC COMPOUNDS Second Edition by Q.N. Porter, Department of Organic Chemistry, University of Melbourne, Parkville, Victoria, Australia Covering the important mass spectrometry literature since 1981, this volume aids practicing organic chemists in their interpretations of the mass spectra of the major classes of heterocyclic compounds.it contains six new chapters and nearly twice as much material as the previous edition, including discussions of chemical ionization, field ionization and desorption and negative ions. Also featured are more than 1500 fragmentation reaction schemes. Series: General Heterocyclic Chemistry 0471 09901 5 992PP June’85 €256.30/$333.25 METHODS OF BIOCHEMICAL A Pd A LY S I S Volume 31 by D. Glick An encyclopedia of methods in biochemical analysis, this modern series keeps biochemists and analytical chemists abreast of experimental innovations and improvements in biochemical techniques and instrumentation. Volume 31 treats such analytical methods as Rapid-Flow-Quench Method in fast reactions, circular dichroism, continuous measurement of dissolved gases by quadrupole mass spectrometer and high- performance liquid chromatography. Series: Methods of Biochemical Analysis, Volume 31 0471 82177 2 552PP J u ne’85 €66.95/$87.15 ORGANIC ELECTRONIC SPECTRAL DATA Volume 21,1979 edited by J.P.Phillips, University of Louisville, D. Bates, Michigan Tech University, H. Feuer, Purdue University and B.S. Thyagarajan, University of Texas This annual series continues the cooperative effort to abstract and publish in formula order all the ultraviolet-visible spectra of organic compounds presented in the journal literature. Over 50 chemists have searched over 100 titles during the course of this project to assemble over 350 000 spectra through these twenty vo I u mes . Series: Organic Electronic Spectral Datra, Volume 21 0471 83047 X 1088pp November’85 €1 22.75/$159.60 I i These books are available through good bookshops or, in case of difficulty, direct from Wiley.If ordering direct please make chequespayable to JOHN WILEY & SONS LTD. You may also telephone yourcredit cardorder - dial lOO(UK only) andask for FREEFONE 3477.40 ANALYTICAL PROCEEDINGS, JANUARY 1986, VOL 23 makers’ Soft-Pac and Memory-Pac soft- ware modules are employed, the former containing software for chemistries and immunoassays and the instructions for automatic instrument set-up, and the latter providing back-up data storage and test parameter storage. Beckman Ltd., Progress Road, Sands Industrial Estate, High Wycombe, Buck- inghamshire . Literature Six Fourier transform infrared applica- tions leaflets highlight specific perfor- mance characteristics of the makers’ 1700 Series FTIR spectrometers. They deal with the following aspects: No.1, com- pensation of atmospheric water and car- bon dioxide levels, No. 2, rapid measure- ment of highly scattering samples, No. 3, quantitative infrared analysis, No. 4, quantitative infrared analysis up to 5000 cm-1, No. 5 , emission measurements and No. 6, wavelength extension. Perkin-Elmer Ltd., Post Office Lane , Beaconsfield, Buckinghamshire HP9 1QA. A brochure gives details of the ChromSep HPLC column system and 16 different packing materials, including the new ChromSpher 5 pm spherical packing. It also gives information on accessories, such as guard columns, pre-concentration columns and solvent saturation columns, and on applications, including drugs and acids. Chrompack UK Ltd. , Unit 4, Indescon Court, Millharbour, London E l 4 9TN. An application leaflet describes the latest technology for sample preparation in food analysis. It explains how sample prepara- tion can be simplified by using sorbent extraction, for which there are available a range of Bond Elut bonded silica extrac- tion columns; these can be used with a Vac Elut sample processing station to extract up to 10 samples simultaneously. Analytichem International, P.O. Box 158, Cottenham, Cambridge CB4 4UT. A booklet, “An Introduction to Fluores- cence in Biological Analysis” by A. T. Rhys Williams, is available. Reference is made to the value of the combination of the separative power of liquid chromato- graphy and the sensitivity of fluorescence, together with the increased flexibility afforded by microprocessors. Perkin-Elmer Ltd., Post Office Lane, Beaconsfield , Buckinghamshire HP9 1QA. A catalogue of analytical standards con- tains descriptions of over 1000 standards as well as several related reference books. The chemicals offered are supplied in compact kits. Applications include the identification of unknown chromato- graphic peaks, the teaching of gas and thin film chromatography, infrared and ultraviolet spectrometry, and melting- point determinations. PolyScience Corporation, P.O. Box 48312 , Niles, Illinois 60648, USA.

ISSN:0144-557X    

DOI:10.1039/AP9862300037

出版商:RSC    

年代:1986

数据来源: RSC

摘要:

40 ANALYTICAL PROCEEDINGS, JANUARY 1986, VOL 23 Publications Received Improved Hollow Cathode Lamps for Atomic Spectroscopy. Edited by Sergio Caroli. Ellis Horwood Series in Analytical Chemistry. Pp. 232. Ellis Horwood. 1985. Price 235. ISBN 0 85312 707 7 (Ellis Horwood); 0 470 20209 2 (Halsted Press). Physico-Chemical Behaviour of Atmos- pheric Pollutants. Proceedings of the Third European Symposium held in Varese, Italy, 10-12 April 1984. Edited by B. Versino and G. Angeletti. Pp. xiv + 666. D. Reidel. 1985. Price Dfl 235; $84; E59.75. ISBN 90 277 1873 3. Annual Reports on NMR Spectroscopy, Volume 16. Edited by G. A. Webb. Pp. x + 373. Academic Press. 1985. Price f91; $100. ISBN 0 12 505316 9. ISBN 0066 4103. Halogenated Benzenes, Toluenes and Phenols with Water. Edited by Ari L.Horvath and Forrest W. Getzen. Solubility Data Series, Volume 20. Pp. xxiv + 266. Pergamon Press. 1985. Price &64; $100. ISBN 0 08 023926 9. Computer Aided Chemical Thermody- namics of Gases and Liquids. Theory, Models, and Programs. Paul Benedek and Francis Olti. Pp. xxx + 731. Wiley. 1985. Z86.95. ISBN 0 471 87825 1. Journal of Micronutrient Analysis. Edited by R. Macrae. A New Journal. Elsevier Applied Science Publishers. f28 (Vol. 1, 2 issues), f60 (Vol. 2, 4 issues). ISSN 0266-349X.ANALYTICAL PROCEEDINGS, JANUARY 1986, VOL 23 41 David B. Williams Practical Analytical Electron Microscopy in Materials Science 1984. VII, 153 pages with 147 figures and 18 tables. Hardcover. DM 120.-/€39.00 ISBN 3-527-26224-5 High-voltage electron beams striking thin specimens in the electron microscope produce many signals.Recent development of methods for detecting and quantifying these signals transformed the technique of transmission electron microscopy (TEM) to analytical electron microscopy (AEM). This book describes the potential and presents examples for the applica- tion of AEM, compares this method with other electron microscope pro- cedures such as TEM and scanning electron microscopy and shows the limitations of the AEM technique. Current and potential users gain a reliable impression of the advances made possible by the exploitation of analytical electron microscopy: 1. the characterization of small particles, either isolated or enclosed in a matrix, 2. the investigation of monolayer-type segregation effects, 3.the identification of small phases and observation of microsegregation effects in multi-component materials, and 4. analyses of the microstructure and microchemistry of beam sensitive materials such as polymers. Explanations of theoretical fundamentals, detailed practical hints and rll numerous references at the end of each chapter make this book a comprehensive introduction to analytical electron microscopy. VCH To obtain this book please contact: The Royal Society of Chemistry, Blackhorse Road, Letchworth SG6 IHN. Phone (04626) 72555 Customers outside the UK and Eire please contact: VCH Verlagsgesellschaft, P.O. Box 1260/1280, D-6940 Weinheim, Federal Republic of Germany42 ANALYTICAL PROCEEDINGS, JANUARY 1986, VOL 23 Applications of Dynamic NMR Spectroscopy to Organic Chemistry by M.Oki Series: Methods in Stereochemical Analysis, Volume 4 1985. XII, 426 pages 8 figures and 109 tables. Hardcover DM 208.4f69.00 ISBN 3-527-261 66-4 Among the physical techniques that are used to solve problems in organic chemistry, nuclear magnetic resonance spectroscopy is probably the most in- fluential. However, dynamic NMR spectroscopy usually receives only little atten- tion as most books emphasize its physical aspects rather than its power for the investigation of conformational processes in organic chemistry. This book was written to familiarize organic chemists with the potentials and applications of dynamic NMR spectroscopy. Consequently, it provides all the knowledge required to gain access to this exciting field.To afford a general idea of areas in which the technique was successfully applied, studies of internal rotation, inversions of amines and topomerization by dissociation are briefly reviewed. This leads to a comparison of the kinetic data obtained by dynamic NMR spectroscopy and other methods and to the presentation of the caveats that must be observed when discussing the data. Nine chapters then deal in detail with the investigation of rotations about various kinds of bonds, of conformational changes in ring compounds, of the stereody- namics of imines and amines, and of degenerate, acid-base, and dissociation reactions. In every case the emphasis is on basic examples rather than sophisticated concepts. Also the book is intended to be instructive rather than exhaustive. For the convenience of the readers, methods by which the free energy of activation is derived from kinetic data as well as specifications of conformations are reported in two appendixes. To obtain this book please contact. The Royal Society of Chemistry, Blackhorse Road, Letchworth SG6 1HN. Phone (04626) 72555 - Customers outside the UK and Eire please contact - VCH Verlagsgesellschaft, P.O. Box 1260/1280, D-6940 Weinheim, Federal Republic of Germany

ISSN:0144-557X    

DOI:10.1039/AP986230040b

出版商:RSC    

年代:1986

数据来源: RSC

摘要:

ANALYTICAL PROCEEDINGS, JANUARY 1986, VOL 23 43 Courses Fibre Optics for Instrumentation March 25-27,1986 and September 16-18, 1986, Manchester This practical t. :e-day course concern- ing basics, sensors and systems will be organised jointly by UMIST and Sira Ltd. and held at UMIST. The course consists of lectures, demonstrations, discussion groups and practical sessions. The aim is to provide participants with a basic know- ledge of fibre optics and an appreciation of the technical and commercial potential of fibre optics as applied to instrumen- tation. An overview of the total subject will be given, but with particular emphasis on multi-mode technology. The course is intended for potential users of fibre optics in conjunction with instrumentation. Participants need not have a detailed prior knowledge of fibre optics.The lecturing team includes speak- ers from UMIST, BICC, GEC, Plessey, Hewlett Packard and ICI. The topics covered include: basic optics and prin- ciples of fibre optics; optical radiation sources and detectors; optical effects of potential use in measurement; fibre optic cables, couplers, connectors and splices; optical sensors, with case studies; chem- ical sensors; systems design; fibre optic connectors between instruments; eco- nomics of fibre optic systems; and the future for fibre optics in instrumentation. For further information and a leaflet please contact the Conference Unit, Sira Ltd., South Hill, Chislehurst, Kent BR7 5EH. Perspectives in Modern Chemical Spectro- scopy April 7-11, 1986, Norwich This course will be held in the University of East Anglia. It will consist of lectures, seminars and practical demonstrations and will cover both basic principles and recent developments. Further details are available from Dr. D. L. Andrews, School of Chemical Sciences, University of East Anglia, Nor- wich NR4 7TJ. Short Courses 1986, Lough borough The following courses will be held in the University of Technology. April 7-11, Gas - Liquid Chromatography; April 21-25, Separations for Biotechnology and Biochemistry; June 23-27, Radioisotope Techniques; July 7-10, Statistics for Analytical Chemistry; July 7-1 1, Flow Injection Analysis; September 15-19, High-performance Liquid Chromato- For further information contact Miss C. D. Newton, Department of Chemistry, Loughborough University of Technology, Loughborough, Leicestershire LE11 3TU. graphy.

ISSN:0144-557X    

DOI:10.1039/AP986230043b

出版商:RSC    

年代:1986

数据来源: RSC

摘要:

44 ANALYTICAL PROCEEDINGS, JANUARY 1986, VOL 23 Analytical Division Diary JANUARY Thursday, 23rd, 5.15 p.m.: Cardiff Western Region Annual General Meeting, followed by the “Seeing is Believing”: The Microscope in Analysis. Speaker: E. B. Reynolds. UWJST, Cardiff, Contact: Mr. F. W. Sweeting, Wessex Water Authority, Water Management Unit, Mead Lane, Saltford, Bristol BS18 3ER. (Tel. 02217-3983). retiring Chairman’s Address. Thursday, 23rd, 10.30 a.m.: London Joint Pharmaceutical Analysis Group. Standards for Veterinary Medicines. “The BP (Veterinary) and Immunologicals,” by I. Davidson. “The Industrial Viewpoint,” by J. Sanford. “The View of the Practising Veterinary Pharmacist,” by J. A. Yacomeni. “The View of the Practising Veterinary Surgeon,” by Professor A. Steele-Bodger. Hall, Pharmaceutical Society of GB, 1 Lambeth High Street, London, S.E.l. Registration is necessary.Cost &15 to members’ of the Group, 220 to non-members. Contact: JPAG Secretariat, Room 403, Pharmaceutical Society of GB, 1 Lambeth High Street, London SE17JN. (Tel. 01-735-9141). Wednesday, 29th, 4 p.m.: Belfast Northern Ireland Region. In-vivo Sensors. Speaker: G. Svehla. Queen’s University, Belfast. There is no registration fee for this meeting. Contact: W. J. Swindall, Department of Chemistry, David Keir Building, The Queen’s University, Belfast BT9 5AG. (Tel. 0232-661111, Ex. 4428). FEBRUARY Wednesday, 5th, 2.30 p.m.: London Analytical Division, in conjunction with the Radiochemical Methods Group. Tracers in Analysis. “Neutron Activation Analysis of Stable Isotopic Tracers for Studies of Mineral Bioavailability,” by J.E. Whitley. “Some Applications at Harwell of Tracers in Environmental Analysis,” by J . S. Hislop. “The Use of Sulphur Hexafluoride Tracer in Power Station Plume Studies,” by A. R. W. Marsh. Scientific Societies Lecture Theatre, 23 Savile Row (entrance in New Burlington Place), London, W. 1. Registration is necessary. Cost f 3 to RSC members, &6 to non-members and gratis to student and retired members. Contact: Miss P. E. Hutchinson, Analytical Division, Royal Society of Chemistry, Burlington House, London W1V OBN. (Tel. 01-437-8656). Friday, 7th, 5.15 p.m.: Plymouth Western Region, jointly with the Peninsula Section of the Snapshot Chromatography-Biomedical Application of New Speaker: A.F. Fell. Department of Chemistry, The Polytechnic, Plymouth. There is no registration fee for this meeting. Contact: Mr. F. W. Sweeting, Wessex Water Authority, Water Management Unit, Mead Lane, Saltford, Bristol BS18 3ER. (Tel. 02217-3983). RSC. Technology in HPLC. Wednesday, 12th, 9.30 a.m.: London Chromatography and Electrophoresis Group, jointly with Biomedical Separations. “HPLC of Proteins and Peptides,” by D. H. Calam. “Trials and Tribulations with the HPLC of Thiols and Disulphides in Physiological Fluids,” by D. Perret. “The Use of Chromatography in the Study of the Absorption and Metabolism of a Cefuroxine Prodrug,” by J. Ayrton. “The Simultaneous Analysis of Theophylline, Caffeine and Their Metabolic Products in Human Plasma by Reversed Phase Gradient Elution HPLC: The Use of a Specially Developed Sample Preparation Column,” by T.E. B. Leakey. “The Chromatographic Problems with a Difficult Analyte ,” by J. Chamberlain. “The Determination of Carbohydrates and Sugar Alcohols in Biological Samples by Capillary GLC,” by D. Stevenson and R. Passas, Royal College of Surgeons, 35-43 Lincoln’s Inn Fields, London, W.C.2. Registration is necessary. Cost $27.60 to members of RSC and Chromatographic Society and f41.40 to non- members. Contact: Mrs. J. A. Challis, Executive Secretary, The Chromatographic Society, Trent Polytechnic, Burton Street, Nottingham NG1 4BU. (Tel. 0602-418248). the Chromatographic Society. Wednesday, 12th, 7 p.m.: Macclesfield North West Region. Novel Approaches to the Analysis of New Drugs. “Spectroscopic,” by P.Hampson. “Chromatographic,” by F. Bailey. ICI Pharmaceutical Division, Macclesfield Works, Maccles- Registration is necessary. Cost 23. Contact: Dr. A. Mathias, IQAD, ICI Pharmaceuticals Division PLC, Alderley House, Alderley Park, Maccles- field, Cheshisre SKlO 4TR. (Tel. 061-740-1460). field. Wednesday, 12th, 11 a.m.: London South East Region. Analysis in Engines. Introduction by J. G. Firth. “Coherent Anti-Stokes Raman Scattering Measurements of Tem- perature and Species in a Firing Petrol Engine,” by R. Williams. [continued inside back cover]ANALYTICAL PROCEEDINGS, JANUARY 1986, VOL 23 ... 111 Analytical Divison Diary, continued February, continued “Flame Speed Measurement in a Spark Ignition Engine,” by Dr. Kalgech. “Laser Measurements of Flame Propagation Rates in Engines,” by 0.Lee. “Testing of Fuel Efficient System by Use of a Distributed Microprocessor Facility.” by A. G. Hayne. Occupational Medicine and Hygiene Laboratories, Health and Safety Executive, 4031405 Edgware Road, London, N.W.2. Registration is necessary. Cost &10 to RSC members, &13 to non-members and €7 for students. Contact: Dr. J. G. Firth, Occupational Medicine and Hygiene Laboratories, Health and Safety Executive, 403/405 Edgware Road, London NW2 6LN. (Tel. 01- 450-891 1). Tuesday, lSth, 4 p.m.: Loughborough Students Chemical Society. Midlands Region, jointly with the Loughborough University Blood, Sweat and Tears, Modern Clinical Analysis. Speaker: L. J . Kricka. Lecture Theatre 5002, Edward Herbert Building, University There is no registration fee for this meeting.Contact: Mr. H. E. Brookes, 35 Dunster Road, West of Technology. Loughborough. Bridgford, Nottingham NG2 6JE. (tel. 0602-231769). Wednesday, 19th, 6.30 p.m.: London South East Region and Microchemical Methods Group. Recent Developments in FIA. Discussion introduced by J . F. Tyson. Savoy Tavern, Savoy Street. London, W.C.2. There is no registration fee for this meeting. Contact: Mr. P. R. W. Baker, 55 Braemar Gardens, West Wickham, Kent. (Tel. 01-777-1225). Wednesday, 19th, 1.30 p.m.: Hull North East Region. Analytical Applications of Diode Array Spectroscopy. ”Fundamental Aspects of Photodiode Array Detectors in Analytical Spectroscopy and HPLC,” by A. F. Fell. “Diode Array Detectors in Pharmaceutical Analysis,” by K.Leiper. “Quantitative Applications of Derivative Spectroscopy,” by T. Owen. “Flow Injection Analysis as a Sample Handling Technique for Diode Array Spectroscopy,” by P. J. Worsfold and K. Wolf. Lecture Theatre A, Chemistry Department, The University, Hull. Registration is necessary. Cost €10 to RSC members, €15 to non-members and &2 to retired and student members. Contact: Dr. P. J. Worsfold, Chemistry Department, The University, Hull HU6 7RX. (Tel. 0482-46311, Ex. 7469). Wednesday, 19th, 10 a.m.: London Special Techniques Group, jointly with the Institute of Acoustic Techniques. “Photoacoustic Fourier Transform Infrared Spectroscopy.” by B. C. “Photoacoustic UV/Vis/NIR Spectroscopy-Depth Profiling,” by “Acoustic Microscopy,” by G. A . D. Briggs. “Acoustic Emission of Paints,” by T. A. Strivens. “Acoustic Testing in Chemical Process Plant ,” by R. Watkins. “Thermosonimetry of Minerals,“ by G. M. Clarke. “Laser Optoacoustic Spectroscopy,” by a speaker to be named. Scientific Societies Lecture Theatre, 23 Savile Row (entrance in New Burlington Place), London, W. 1. Registration is necessary. Cost f35 to RSC and Institute of Physics members, &45 to non-members and €5 to student and retired members. Contact: Mr. A. G. Ferrige, Physical Chemistry Depart- ment, Wellcome Research Laboratories, Langley Court, Beckenham, Kent BR3 3BS. (Tel. 01-658-2211, Ex. Physics. Beadle. R. M. Miller. 5357). Thursday, 20th, 4.15 p.m.: Aberdeen Scotland Section of the RSC. Scottish Region, jointly with the Aberdeen and North of Atoms for Analysis. Speaker: J . M. Ottaway. Department of Chemistry, The University, Aberdeen. There is no registration fee for this meeting. Contact: Dr. J. M. Warren, Department of Biochemistry, Royal Infirmary. Glasgow G4 OSF. (Tel. 041-552-3535, Ex. 5108).

ISSN:0144-557X    

DOI:10.1039/AP9862300044

出版商:RSC    

年代:1986

数据来源: RSC