ISSN:0144-557X    

DOI:10.1039/AP98522FX025

出版商:RSC    

年代:1985

数据来源: RSC

ISSN:0144-557X    

DOI:10.1039/AP98522BX027

出版商:RSC    

年代:1985

数据来源: RSC

摘要:

ANALYTICAL PROCEEDINGS, JULY 1985, VOL 22 189 Editorial A Land of Analytical Electrochemical and Bio-sensors Just seventeen hours away by jet from London is Tokyo, the world’s second largest city and capital of the country with the world’s second largest gross national product. A land of mountains and active volcanoes, short and fast-flowing rivers, Japan is about half as big again as the United Kingdom and has a population of 120 million. It is also a land of electro- chemical and bio-sensors: at least so it seemed to one who had been invited to spend 40 days there on a Visiting Profes- sorial Fellowship of the Japan Society for the Promotion of Science (JSPS) relating to analytical electrochemical and bio- sensors. Centres of intense research activity exist throughout Japan on all manner of analytical electrochemical and bio- sensors.It was stimulating to have visited many of these in relation to ion sensing and the synthesis and calibration of ion sensors, gas and humidity sensing based on solid electrolyte, ceramic, semicon- ductor and fuel cell systems, and the application of potentiometry and amper- ometry to bio-sensing. However, most of the time was spent at the hub of bio- sensor research in Japan, that is, at the Laboratory of Resources Utilization of Tokyo Institute of Technology. Here, microbial and enzyme systems are hybri- dised with various electrochemical and other sensors to give an ever increasing range of selective and effective bio- sensors.’ It was illuminating to observe the progress made in the application of integrated circuit systems in “biochips” in relation to views expressed in 1982 in a feature article of the New Scientist.2 With regard to bio-sensing, graduates in chemistry readily adapt to practising the principles and methods of microbiol- ogy, electronics and microcomputerisa- tion, thus emphasising chemistry as an essentially central scientific discipline.Perhaps the practice of requiring fourth year undergraduates to spend most of this year on a research project helps to single out those who are best at broadening their horizons. At any rate, such final year research provide a basis for selecting entrants into the subsequent 2 year mas- ters’ research courses. Also, it helps to develop the kind of enquiring mind required, even of first degree graduates who enter employment and become a mainstay of Japan’s technological strength, which depends so much on aptitude in combining and applying tech- niques.The master’s course can be topped by a 3 year research course for a doctorate. In Japan, university research is based on the koza system, that is, a grouping consisting of a professor, associate profes- sor, research associates, research students and frequently “company personnel.” This does not give quite the freedom of some other countries for younger people to develop the imaginative, individual research initiative that might be expected in universities. Nevertheless, this is a country claiming to have about 400000 personnel on scientific and technological research work to rank third, following only the Soviet Union and the United States of America.There is an effective division of effort between the universities and industry and there is no coyness on the part of Japanese industrial organisations about paying for selected younger employees to spend 1, 2, 3 or more years attached to selected research groups in universities. There appears to be no obligation for such “company personnel” to attend courses or to produce theses, but the practice boosts university research and catalyses the rapid transfer of technology from university to industry to promote rapid product deve- lopment and improvement. “A man must make his opportunity, as oft as find it.” Advancement of Learning, 2 FRANCIS BACON References 1. Karube, I., and Suzuki, S., Anal. 2. Yanchinski, S., New Scientist, 1982, Proc., 1983, 20, 556.January 28, 236. J. D. R. Thomas Department of Applied Chem is tr y, UWET, P.O. Box 73, Cardiff CF13XF RSC ANALYTICAL DIVISION PSA 85 September 1619, 1985, Bradford The fifth Particle Size Analysis Conference to be held under the auspices of The Analytical Division of the Royal Society of Chemistry will be held at the University of Bradford. The Conference will be concerned with measurement of the particle size distribution of powders, suspensions and aerosols, the determination of surface area, particle shape and pore size distribution, and will cover ancillary techniques such as sampling, dispersion and particle density measurement, together with the application of particle characterisation techniques to industrial problems. The plenary lecturers will be Dr. N. G. Stanley-Wood, Dr. T. Allen, Professor E. Heidenreich, Dr. M. Martin and Professor B. Scarlett. For further information and registration forms please contact Dr. N. G . Stanley-Wood, School of Powder Technology, University of Bradford, Bradford, West Yorkshire BD7 1DP.

ISSN:0144-557X    

DOI:10.1039/AP9852200189

出版商:RSC    

年代:1985

数据来源: RSC

摘要:

190 ANALYTICAL PROCEEDINGS, JULY 1985, VOL 22 Research and Development Topics in Analytical Chemistry The following is a summary of one of the papers presented at a Meeting of the Analytical Division held on June 26th and 27th, 1984, in the University of Manchester Institute of Science and Technology. Summaries of sixteen other papers presented at the Meeting were published in the January issue, p. 3. Applications of LESTEQ in Automated Spectrophotometric Titrimetry G. J. Thomson, B. G. Cooksey and J. M. Ottaway Department of Pure and Applied Chemistry, University of Strathclyde, 295 Cathedral Street, Glasgow G I IXL W. C. Campbell Research and Technology Department, Petrochemicals and Plastics Division, ICl plc, P. 0. Box 90, Wilton, M iddlesb ro ug h, Cleveland TS6 810 In many instances there is a gradual change in indicator colour over a substantial volume range in practical titration systems, making accurate visual estimation of the equivalence volume impossible.This may be due to incomplete reaction of the titration reagents, an unfavourable choice of indicator or the presence of another coloured species in the titrand solution. Overall, a decrease in the concentration of the titrant reagents usually results in a less complete change in colour of the indicator and a more gradual slope across the end-point region of the plotted spectrophotometric titration curve. It is clear that if the colour change is measured instrumen- tally, simple visual examination of the plotted titration curve will be hoplessly inadequate in many instances. Therefore, some kind of numerical analysis of the titration data must be employed to produce reliable, accurate estimates of the equivalence volume.However, some of the most commonly used techniques, such as the derivative method,’ the extrapola- tion method2 and the Gran plot,3 whilst performing well when the end-point is fairly sharp, produce serious systematic errors when there is a broader colour change. Anfalt and Jagner4 first suggested that a non-linear least- squares fit of the titration data to an accurate model equation should produce the most accurate estimation of the titration equivalence volume. The advent of cheap microcomputers has made this and similar techniques feasible. This technique has been used in both potentiometric and spectrophotometric titrimetry and has been shown to be superior to any other available method.5 A BASIC computer program, LESTEQ, has been developed for the calculation of the exact equivalence point volume for spectrophotometric titrations using ancillary indicators.A non-linear least-squares fit of the titration data to a model equation by means of a modification of the Gauss - Newton multi-parameter optimisation method forms the basis of the program, which also calculates the indicator and titration conditional stability cons tan ts . The Program The form of the model equation used in LESTEQ is dependent upon the titration reaction stoicheiometry and the nature of the indicator reaction. The following equation is valid for the case where the titrant and titrand species form a 1:l complex, and the indicator forms a 1:l complex with the titrand species before the equivalence point and exists in the “free” indicator form after equivalence. This is analogous to complexometric titrimetry using metallochromic indicators. VKT VK1 (DT- AV) VEKT ( A V - O D ) cT CT (AV-DD) KI (DT-AV) +- - VT = v, - - - - X The erroneous variable A (absorbance) is a function of: firstly, the error-free variable VT (volume of titrant added at any stage of the titration); secondly, two known constants, CT (the concentration of the titrant solution) and VI (the initial volume of titrand solution) (V = VI + VT); thirdly, five unknown system parameters, V , (the equivalence-point volume), KI and KT (the titrand - indicator and titrand - titrant conditional instability products) and DD and D T (constants proportional to the absorptivity of the indicator species in its pre- and post-end-point forms, respectively).It is apparent that the best possible estimates of the system parameters occur when the sum of squares S S = ( A - A c ) ~ is minimised. Ac is the absorbance calculated from the model equation by using estimates of the system parameters. Hart- ley’s modification6 of the iterative Gauss - Newton7 multi- parameter optimisation method provides the means of search- ing for the values of the five system parameters which produce the minimum sum of squares. The convergence of this method on the minimum sum of squares is dependent upon a good initial approximation being input to the algorithm. Assuming (wrongly) that it is A which is the error free variable and VT which is the erroneous variable, the model equation becomes a linear equation with two independent variables.Let Y = VT + (VKT)/CT be the erroneous variable, and X1 = V(DT - AV) / [(AV - DD)CT] and X2 = (AV - OD) / (DT - AV) be the error free variables. By performing a linear regression with two independent variables it is possible to obtain estimates of VE, KT and KI for known values of DD and DT. In LESTEQ the simplex search8 technique provides the means of searching for the values of DD and D T which minimise the sum of squares due to the regression and so produce good initial estimates of the system parameters for the Gauss - Newton algorithm. The major advantage of the simplex search technique is that it will converge towards theANALYTICAL PROCEEDINGS, JULY 1985, VOL 22 191 ~~ ~ Table 1.Accuracy and precision of LESTEQ Keller - Richter LESTEQ Expected Mean RSD, Deviation Mean RSD, Deviation resulth resulth YO from ER, YO resulth YO from ER, % 9.974 x 10-3 9.837 x 10-3 1.86 -1.17 1.000 x 10- 2 0.19 +0.30 4.987 X 10-3 4.998 X 10-3 0.36 +0.22 5.017 x 10-3 0.25 +0.60 9.974 x 10-4 9.790 X 10-4 0.31 -1.84 9.799 x 10-4 0.12 -1.74 9.974 x 10-5 8.839 X 10-5 2.02 -11.3 9.820 x 10- 5 0.51 -1.54 4.987 X 10-5 4.290 X 10-5 5.68 - 13.9 4.914 X 10-5 1.65 -1.64 2.494 x 10-5 2.932 X 10-5 8.13 +17.5 2.410 x 10-5 2.62 -3.33 1.995 X 10-5 2.382 X 10-5 4.64 +22.7 1.964 x 10-5 4.13 -1.34 RSD = Relative standard deviation. ER = Expected result. region of optimum response from a very poor initial approxi- mation.Experimental LESTEQ has been used in conjunction with two automated, microcomputer controlled titrators. Initial development work was performed by using an inexpensive home-built systen- consisting of a Radiometer autoburette, a Pye Unicanl single-beam spectrophotometer, an interface and a Commo- dore PET microcomputer. A 100-ml stirred titration vessel was situated in the sample compartment of the spectrophotometer and regular increments of titrant added throughout the course of a titration. The titration times were typically 15 min and the typical calculation times for LESTEQ 5-20 min. Later investigations into the performance of the program were carried out by using the Hamilton AMICA modular- titration system. A dual-syringe stopped-flow system enabled any ratio of titrant to titrand to be mixed in a static mixing chamber.Measurements were then made in a 1-cm flow- through spectrophotometer cell. This arrangement, coupled with an autosampler capable of holding 50 samples, enabled titration times to be reduced to about 3 min. A dedicated Kontron microcomputer with compiled Microsoft BASIC soft- ware resulted in calculation times being lowered to about 1 min. Thus, a typical sample throughout for the AMICA system would be around 15 samples h-l. The existing AMICA software utilised the Keller - Richter method of equivalence-point estimation. In this method a correction to the point of inflection calculated from four data points around the maximum absorbance change is the approxi- mation to the equivalence point.Results and Conclusions Table 1 shows a comparison between the Keller - Richter and LESTEQ methods in terms of the accuracy and the precision of equivalence point volume estimation carried out by using the AMICA modular titrator. Magnesium was titrated with EDTA at pH 10 (ammonia solution - ammonium chloride solution) buffer using Eriochrome Black-T indicator over a range of concentrations. A series of 10 titrations was carried out at each concentration and for each method. The expected result was calculated from the previous standardisation of both titration reagents. A 0.1 M stock solution of magnesium was standardised by a sulphated-ash procedure and a 0.1 M EDTA stock solution by titration against lead nitrate solution in a hexamine buffer using xylenol orange as the indicator.Subsequent dilutions were assumed to be perfect. With LESTEQ there is good agreement between the observed (computer calculated) results and the standardised results over the whole concentration range. However, it is evident that the performance of the Keller - Richter method is unsatisfactory at the lower concentrations, where the end-point becomes broader. This is typical of inflection point methods. It is also noticeable that the standard deviations for the series of titrations carried out using LESTEQ, where all the titration data is used in the calculation, tend to be lower than those of the Keller - Richter method, where only 4 data points are used. Similar trends were observed for the calcium - EDTA - pH 12 - murexide and sodium acetate - hydrochloric acid - methyl orange titration systems. The use of LESTEQ has been found to be limited to cases where over three measurements are made in the colour change region.Thus in such situations as higher concentrations, where there is an extremely sharp colour change, LESTEQ will fail to converge on the correct answer due to insufficient data in the end-point region. The program’s use is also limited at lower concentrations by long calculation times (e.g., 15 min), which make it inapprop- riate for high volume, routine analyses. In conclusion, LESTEQ has been shown to be an accurate and precise method of calculating the equivalence-point volume for spectrophotometric titrations utilising ancillary indicators. The authors would like to thank Pye Unicam Ltd. for the loan of the SP6 spectrophotometer used in this work, Hamilton Bonaduz AG for the loan of the AMICA system, and Nancy Rent of Hamilton Bonaduz AG for assistance with the modification of the AMICA software which allowed the use of the LESTEQ program. 1. 2. 3. 4. 5. 6. 7. 8. References Meites, L., and Goldman, J. A., Anal. Chim. Acta., 1964,30, 28. Carr, P. W., Anal. Lett., 1971, 4, 893. Gran, G., Analyst, 1952, 77, 661. Anfalt, Y., and Jagner, D., Anal. Chim. Acta., 1971, 57, 165. Goode, S. R., Anal. Chem., 1977,49, 1408. Hartley, H. O., Technometrics, 1961, 3, 269. Adby, P. R., and Dempster, M. A. H., “Introduction to Optimisation Methods,” Chapman and Hall, 1974. Walsh, G. R., “Methods of Optimisation,” John Wiley, 1975.

ISSN:0144-557X    

DOI:10.1039/AP9852200190

出版商:RSC    

年代:1985

数据来源: RSC

摘要:

192 ANALYTICAL PROCEEDINGS, JULY 1985, VOL 22 Trace Analysis in the Food Chain ~ ~~ The following are summaries of three of the papers presented at a Joint Meeting of the East Anglia Region, the Atomic Spectroscopy Group and the East Anglia Section of the RSC held on November 7th, 1984, in the Food Research Institute, Norwich. Some Novel Approaches to Trace Element Analysis John M. Ottaway, John Carroll, Stephen Cook, David Littlejohn, John Marshall and Sharon C. Stephen Department of Pure and Applied Chemistry, University of Strathclyde, Cathedral Street, Glasgow G 1 IXL The demand for the analysis of trace elements in many materials, such as foodstuffs, biological samples, semiconduc- tors, etc., is continually becoming more exacting. Scientists concerned with human and animal nutrition, medicine, modern materials and from many other disciplines find that they require the determination of trace elements at lower and lower concentrations and often expect levels of accuracy and precision, and analytical speed, to be maintained.The analy- tical chemist who is striving to find the answers to some of these problems may himself contribute to this need for improved sensitivity. In the clinical field, a milestone paper by Versieck and Cornelis1 demonstrated this very clearly in the determi- nation of trace elements in blood serum. In many instances, notably chromium, the accepted normal levels of elements have dropped over the past 20-30 years as analytical techniques and expertise have improved. Thus, it is now necessary to have a technique that is capable of determining chromium at the 0.1 pg 1-1 level, whereas 20 years ago 10 pg 1-1 might have been considered adequate.Obviously, techniques of high or higher sensitivity are required. One of the traditional approaches employed to overcome this problem has been to use pre-concentration procedures. However, at the trace element concentration levels of current interest it has become very difficult to avoid contamination during any form of sample pre-treatment, and pre-concentration procedures must be carried out under rigorously controlled clean conditions with ultra-pure reagents. Such conditions are beyond the scope of many laboratories, and it is therefore necessary to adopt an alternative approach. The development of techniques that are highly sensitive and permit the direct analysis of a sample with little or no sample pre-treatment is an obvious choice for many research and routine analytical laboratories. The time saved by avoiding the need for lengthy pre-concentration procedures, and for train- ing chemists in the skills required for them, can represent a significant cost benefit.In the application of atomic-spectrometric methods to trace element analysis, flame atomic absorption spectrometry (FAAS) and, more recently, inductively coupled plasma emission spectrometry (ICPES) provide satisfactory sensitivity and accuracy for many routine requirements. In particular, the ease with which the ICP can provide rapid multi-element analysis on either a sequential or simultaneous basis makes it increasingly attractive to many laboratories.When the highest sensitivity is needed, electrothermal atomic absorption spec- trometry (ETA - AAS) is more likely to be the technique of choice. Certainly, neither FAAS nor ICPES will allow analyte determinations at the pg 1-1 or ng g-1 (in solid samples) levels without pre-concentration procedures with their inherent difficulties. A recent paper in this journal2 covered some of the recent developments in electrothermal atomisation. The technique has undoubtedly gained an undeserved reputation for being prone to serious interference effects, and these have been extensively reviewed elsewhere.3 It is worth re-stating that this is partly caused by the high sensitivity of electrothermal AAS methods. Because analyte determinations are carried out at much lower concentrations in a sample matrix compared with other techniques, the analyte to matrix ratio is often one or two orders lower than that used, or even contemplated, with alternative techniques.Nevertheless, in order to avoid the need for separation of the analyte from the matrix before ETA - AAS analysis, attempts have been made to circumvent established interference phenomena, and matrix modification and standard additions procedures have become routine. Apart from these and other procedures mentioned previously,2 the introduction of oxygen during the ash step of the atomisation cycle is now becoming increasingly popular .495 Perhaps the use of oxygen could be interpreted as a form of matrix modification, but it certainly aids the oxidation of an organic based matrix by converting a simple pyrolysis step into an oxidative decomposition. Besides helping to remove the organic matrix more completely before the atomisation stage, use of oxygen ashing prevents the normal build-up of carbon residues which occurs when charring many organic based matrices in an inert gas atmosphere.We have recently made use of this facility in the development of a simple slurry method for the determination of lead in spinach.6 Freeze dried and powdered spinach is suspended in a thixotropic thickening agent, Viscalex HV30. The slurry is injected directly into the graphite furnace and oxygen ashing is used to remove most of the organic part of the solution matrix by oxidation. Besides preventing the build-up of carbonaceous residue and allowing a higher ashing temperature, the introduction of oxygen ashing reduces the background signal at the atomisation stage and allows accurate results to be achieved by standardisation against aqueous standards.In the earlier paper,2 the importance of atomisation at a high temperature under isothermal conditions was emphasised, and the different approaches devised to achieve this were outlined. At the present time, approximately 18 months later, the situation remains almost the same. The platform method remains the simplest means of increasing the temperature of the vapour phase of the atomiser during atomisation. Platform technology is now becoming more widely available, and its advantages more widely appreciated.7 A technique such as the probe method,8-11 in which the sample is introduced on a separate atomisation surface when the tube and its internal atmosphere have reached isothermal conditions, appears to be fundamentally more attractive and closer to ideal conditions.Although probe systems are not yet commercially available, automatic systems fully compatible with commercial types of autosampler have now been described for both the Perkin- Elmer HGA 50011 and Pye Unicam SP9 systems.12 In bothANALYTICAL PROCEEDINGS, JULY 1985, VOL 22 193 instances, injection of the sample on to the probe is carried out whilst it is positioned inside the graphite tube. Drying and ashing are carried out in this position, and the probe is then removed while the tube is heated to the pre-selected constant temperature conditions.As soon as constant temperature is reached, the probe is rapidly reintroduced into the tube for sample atomisation. In one instance movement of the probe is solenoid operated," in the other it is motor driven.12 In both instances, the probe is introduced via an extra slot cut in the side wall of the graphite tube at its centre. The presence of the extra slot in the tube itself causes a loss in sensitivity.12 Although this sensitivity loss is partially compensated for by the increase in sensitivity of probe atomisation, experiments have been carried out with an automatic version of the end-entry probe,13 in which the probe is introduced from the open end of the atomiser tube. Optical alignment in the smaller commercial atomiser tubes is more difficult in this configu- ration but is relatively easy in tubes of 7.5 mm internal diameter.In order to increase the sensitivity, and the capability of the method to deal with samples containing high acid concentrations, a probe configured in the form of a tube has been designed. Sample volumes up to 70 p1, containing 8% hydrochloric acid, could be tolerated satisfactorily, but at the cost of a reduction in final temperature achieved by the probe and its heating rate, because of the larger mass of graphite. Nevertheless, the tube probe will be particularly useful for volatile elements. One of the primary advantages of probe atomisation is that similar atomisation temperatures are optimum or convenient for all elements. The adoption of acceptable compromise conditions for a range of elements which would be required in a simultaneous multi-element analysis procedure is thus more feasible with the probe method than with tube wall atomisa- tion.As was described previously ,* simultaneous multi- element analysis is feasible with electrothermal atomisation using either atomic absorption with a continuum s0urce~~J5 or atomic-emission spectrometry. 16317 Preliminary experiments18 have demonstrated encouraging performance of the probe with the continuum-source atomic-absorption system developed by O'Haverlg and his colleagues. Were it commercially available, the combination of a simultaneous multi-element atomic-absorption system with an electrothermal atomiser might be an attractive rival to ICPES for trace element analysis in the food chain.It is interesting to note that the system/software developed by O'Haver and Harnly15.19 is now used routinely for exactly this purpose at the US Department of Agriculture laboratories in Beltsville, Maryland. The major optical components that make continuum-source atomic-absorption measurements feasible are the high resolution echelle spectrometer and the high intensity xenon arc light source. The high resolution of the echelle allows absorption measurements to be made at the centre of the line profile (ensuring good sensitivity), and the light source provides sufficient intensity over the narrow band pass to give a good signal to noise ratio. The heart of the system is, however, the wavelength modulation facility, which not only offers highly efficient background correction but also provides a linear calibration range of atomic-absorption measurements over 4-6 orders of magnitude.15 In the original design, the extended calibration range was achieved by the measurement of six background corrected absorbances at different positions over the line profile.The most sensitive absorption signal was obtained at the line centre, but less sensitive .absorption measurements suitable for higher analyte concentrations were made in the wings of the line profile.15-19 The need to collect and store data relating to six absorbances over 16 separate analyte channels at a high sampling rate considerably complicated the computer software and increased the cost of the original system. The software requirements for long-range calibration have recently been greatly reduced by using only two absorbance measurements, obtained by means of a staircase modulation wa~eform.19~20 A high sensitivity calibration graph is obtained as before by measurement of the absorbance at the line centre, and the less sensitive graph is obtained from two small groups of selected wavelengths in the wings of the line profile.Apart from advantages in analytical speed, the major step forward provided by this approach is undoubtedly the reduction in the requirement of computer power. At Strathclyde, a single-channel continuum-source atomic- absorption - atomic-emission spectrometer has recently been constructed on the same principles. Details of this system are being published.21.22 The operation of the system is based on, and under the control of, an Apple IIe microcomputer.The Apple microcomputer performs a number of functions inclu- ding data acquisition and processing. In hardware terms the system is analogous to the Harnly and O'Haver system15 (echelle monochromator, xenon arc lamp, refractor plate for wavelength modulation, etc.), except that the monochromator has only a single output channel. The Apple supplies the modulation waveform through a DAC to the scanner con- troller, which operates the refractor plate. Programs are available for sine, three-stage square wave, or staircase modulation waveforms. The modulated signal from the photo- multiplier tube is passed via a pre-amplifier and ADC directly to the computer. Data acquisition is accomplished using an assembly language programme, but the rest of the software is in compiled BASIC.Program options for flame (continuous signal response) or furnace (transient response) are available in both atomic-absorption and atomic-emission modes. As three different wavelength modulation waveforms are available for each of the four measurement modes, a total of twelve operating systems is available. In each instance, highly efficient background correction is available, and in the four staircase waveform options two calibration graphs are obtained, which give an extended range of linear calibration. The system is clearly highly flexible, and although based on single-channel operation, it is more flexible than typical commercial atomic- absorption systems based on line source operation.The absorbance or emission intensity versus time signal, together with the relevant background absorbance or emission profiles, is automatically displayed on the video screen and can be dumped to a printer if required. Peak area and peak height measurements are printed on the VDU at the same time for either one or two (staircase waveform) forms of measurement. Although this system is still under evaluation and develop- ment, the design principles would offer a future route to simultaneous multi-element analysis by atomic absorption at relatively low cost. The low temperature atom cells used in atomic absorption can be seen as an advantage compared with the ICP, as the complexity of the observed atomic spectra is reduced, with a consequent reduction in spectral interferences.Wavelength modulation is also more easily applied for background correction, etc., in systems with low spectral complexity. Financial support from many sources, including particularly Pye Unicam Ltd., The Pye Foundation, Perkin-Elmer Corpor- ation, MAFF and SERC is gratefully acknowledged. The authors also thank T. C. O'Haver and J. M. Harnly for their collaborative participation in parts of this work. 1. 2. 3. 4. 5. 6. References Versieck, J . , and Cornelis, R., Anal. Chim. Acta, 1980, 116, 217. Ottaway, J. M., Anal. Proc., 1984,21,55. Slavin, W., and Manning, D. C., Prog. Anal. At. Spectrosc., 1982, 5,243. Salmon, S. G., and Holcombe, J. A., Anal. Chem., 1982,54, 630. Eaton, D. K., and Holcombe, J. A., Anal. Chem., 1983, 55, 946.Stephen, S. C., Littlejohn, D., and Ottaway, J. M., Analyst, in the press194 7. 8. 9. 10. 11. 12. 13. 14. Slavin, W., Carnrick, G. R., Manning, D. C., and Prusz- kowzka, E., At. Spectrosc., 1983, 4, 69. L‘vov, B. V., and Pelieva, L. A., Zh. Anal. Khim., 1978, 33, 1572. Slavin, W., and Manning, D. C., Spectrochirn. Acta, Part B, 1982,37,955. Giri, S. K . , Littlejohn, D., and Ottaway, J. M., Analyst, 1982, 107, 1095. Littlejohn, D., Marshall, J., Carroll, J., Corrnack, W., and Ottaway, J. M., Analyst, 1983, 108, 893. Littlejohn, D., Cook, S., Durie, D., and Ottaway, J. M., Spectrochim. Acta, Part B, 1984, 39, 295. Marshall, J., Giri, S. K., Littlejohn, D., and Ottaway, J. M., Anal. Chim. Acta, 1983, 147, 173. Harnly, J. M., O’Haver, T. C., Golden, B., and Wolf, W.R., Anal. Chem., 1979,51, 2007. 15. 16. 17. 18. 19. 20. 21. 22. ANALYTICAL PROCEEDINGS, JULY 1985, VOL 22 Harnly, J. M., and O’Haver, T. C., Anal. Chem., 1981, 53, 1291. Ottaway, J. M., Bezur, L., and Marshall, J., Analyst, 1980, 105, 1130. Marshall, J . , Littlejohn, D., Ottaway, J. M., Harnly, J. M., Miller-Ihli, N. J., and O’Haver, T.C., Analyst, 1983, 108, 178. Carroll, J., Miller-Ihli, N, J., Harnly, J. M., O’Haver, T. C., Littlejohn, D., and Ottaway, J. M., Analyst, in the press. O’Haver, T. C., Analyst, 1984, 109, 211. Miller-Ihli, N. J., O’Haver, T. C., and Harnly, J. M., Anal. Chem., 1984, 56, 176. Marshall, J., Carroll, J., Littlejohn, D., Ottaway, J. M., O’Haver, T. C., and Harnly, J. M., Anal. Proc., 1985,22,67. O’Haver, T. C., Harnly, J.M., Marshall, J., Carroll, J., Littlejohn, D., and Ottaway, J. M., Analyst, 1985, 110, in the press.ANALYTICAL PROCEEDINGS, JULY 1985, VOL 22 195 Table 3. Comparison of NAA, AAS, single and double spike MS. Relative standard deviations are shown in parentheses Total 58Fe/pg by- Total Fe/pg by- Sample NAA MS (SS 58) MS (DS 58) AAS MS (DS 54) 9K 10.84 10.25 10.99 494 534 5N 6.80 11.60 9.78 592 478 4A 4.34 4.36 5.88 450 517 2Y 9.00 12.33 12.85 930 964 8D 17.71 18.80 20.22 1130 1186 (1.6%) (1.7%) ( 1.1 Yo) (2.4% ) (1.4%) (0.5 yo) (1.0%) (0.7 yo) (0.7 yo) (1.5%) (1.5%) (1.7%) (1.5%) (2.0%) (0.5% ) tion analysis (NAA) on aliquots of the same samples (58Fe was first chosen because it was suitable for NAA). To Measure Total Iron Content (of Standard Reference Materials) In the second experiment, four standard reference materials (SRM) chosen to cover a wide range of iron contents were “spiked” with a different isotope, 54Fe, to test its use as an internal standard from which to measure the total iron in biological materials.The SRM and the added 54Fe spike were ashed by using a mixture of concentrated sulphuric and nitric acids, small volumes of 30% hydrogen peroxide being added carefully to aid oxidation. The remaining clear solution was taken up in 2 ml of 6 M hydrochloric acid and extracted with diethyl ether as before, and the 54FePFe ratio determined. The results for the iron content of the four SRM’s were found to lie within the error values set for the standard by the National Bureau of Standards and had the following relative standard deviations: spinach (SRM 1570), 1 .O% ; oyster tissue (SRM 1566), 6.5%; wheat flour (SRM 1567), 5.3%; and bovine liver (SRM 1577a), 3.5%.Double Isotope Labelling to Measure Total Faecal Iron and Iron Bioavailability Having proved the suitability of the two isotopes separately, a double isotope experiment was performed. Aliquots of faecal ash labelled with 58Fe were spiked with 54Fe dissolved in dilute hydrochloric acid and extracted as before. The 58FePFe and 54Fe/56Fe ratios were measured consecutively. The 58Fe enrichment and hence the absorption of iron (bioavailability) and the total iron (via the 54Fe internal standard) were determined. The results from the single and double isotope experiments and those from NAA and AAS are compared in Table 3.We conclude that the over-all accuracy and precision of the results described here compare favourably with those obtained by NAA and AAS and also with those recently reported for the newer resonance ionisation technique. Work continues to consolidate these developments, and multi-isotope studies on calcium and zinc are envisaged. We thank Miss M. Minski for the neutron activation analysis and Miss R. Girdlestone for the atomic-absorption spectro- photometry. 1. 2. 3. 4. 5. 6. 7. 8. 9. 10. 11. References Aston, F. W., “Mass Spectra and Isotopes,” Edward Arnold, London, 1933. Craig, R. D., Errock, G. A., and Waldron, J. D., Adv. Mass Spectrom., 1959, 1, 136. Adams, F., Spectrochim. Acta, Part B, 1983,38, 1379. Kelly, W. R., and Fassett, J. D., Anal. Chem., 1983,55,1040. Habfast, K., Int.J. Mass Spectrom. Ion Phys., 1983,51, 165. Fassett, J. D., Powell, L. J., and Moore, L. J., Anal. Chem., 1984,56,2228. Garner, E. L., and Dunstan, L. P., Adv. Mass Spectrom., 1978,7A, 481. Turnlund, J. R., Michel, M. C., Keyes, B. A., King, J. C., and Margen, S., Am. J. Clin. Nutr., 1982, 35, 1033. Miller, D. D., and Van Campen, D., Am. J . Clin. Nutr., 1979, 32, 2354. Johnson, P. E., J. Nutr., 1982, 112, 1414. Smith, D. L., Anal. Chem., 1983, 55, 2391.196 ANALYTICAL PROCEEDINGS, JULY 1985, VOL 22 sewage effluent in Broadland, and about 75% comes from this source; only 10% of the nitrogen compounds come from effluent, the rest from land-leachage. In particular cases other sources may be important; in the mid-1970s most of the phosphorus supply to Hickling Broad came from the excreta of a large flock of roosting, migrating black-headed gulls (Larus ridibundus L.).Three problems emerge for a scientist advising management agencies on how to solve the problems caused by this nutrient input, or eutrophication. First, it must be established that nutrient concentrations have indeed increased concomitantly with the onset of the problems; secondly, the exact sources of nutrients must be established to create a plan for their most economic removal, as the obvious solution to the problem is to remove its cause; and thirdly, it must be shown that removal of the cause does indeed bring about restoration of the original state aqd that the problems are not irreversible. Evidence for an increase in nutrient concentration comes from analyses obtained over a long period by the local East Anglian Water Company, who have monitored nitrate since the 1940s.Although there are problems in interpreting the data because of changes in the method used, there is a clear trend upwards from maximum winter concentrations below 1 .mg 1-1 (as nitrate-nitrogen) in the 1950s to above 4 mg 1-1 in the 1980s at a site on the River Bure. The value of long-term monitoring cannot be underestimated. Analyses for phosphorus have not usually been a part of routine monitoring and the only clue to changes comes from analyses of the phosphorus concentrations in cores of sediments taken from the bottoms of the Broads and dated by radiometric (2lOPb) methods. Phosphorus is to some extent mobile in anaerobic sediments, but the rate of sedimen- tation is so high in Broadland that chemical conditions are soon stabilised by burial of old sediment by new.The general pattern is of a post-war increase, following the connection of a greater proportion of an increasing population to mains sewerage. The establishment of nutrient budgets poses many problems, but at least a crude budget is necessary for management. Most problems devolve on phosphorus. It is easy to measure (after the digestion of unfiltered water with acid persulphate) the total phosphorus concentration washed from the land, but difficult to know how much of the phosphorus is available to algae, and how much will be fixed on to soil particles and sedimented out before it reaches the main river and Broads.Organic phosphate compounds are washed out of soils, but an unknown fraction of them also is available to algae. Changes in land usage during the 1 or 2 year minimal period needed to establish a budget also add complications. The phosphorus discharged by sewage treatment works is easy to measure and most of it is in the inorganic (available) form. Most problems arise from an estimation of how much is fixed into river sediments and how much passes down river. Reliance on established flow-gauges can also be misplaced. The distri- bution of nutrients to the various Broads is also difficult to determine because of low flow-rates and effects of (freshwater) tides. Dye-tracing of water masses is needed, but this process is time consuming and expensive.In practice, because of all these problems, only an approximate budget is possible and a pragmatic approach of “try it and see” has to be used to find the best way of reducing the nutrient concentrations progressively. The situation cannot be improved by refining analytical methods or by increasing the number of analyses made in the laboratory. Sometimes, desk studies with few analyses can reveal enough for the first practical steps to be taken. It is not possible to remove much nitrate from the water; denitrification by bacteria in the soils of riverine fens and marshes is effective, and corridors of these still existing along the rivers should be rigorously conserved; where they have been drained they should be allowed to re-establish. Nitrate is so soluble that chemical precipitation before the water enters the main rivers is not feasible, but agricultural practices minimising the use of highly soluble nitrogen compounds should be encouraged.Phosphorus can be precipitated with iron(I1) compounds at the sewage treatment works (phosphate stripping). Highly successful restoration has been achieved in two Broads by damming them from the nutrient-rich rivers, but this is not generally possible because it would interfere with navigation rights and perhaps fish migrations. Phosphate stripping has been partly successful. It has restored Cromes Broad on the R. Ant, but as yet not the more seriously eutrophicated Barton Broad. Stripping can only bring the phosphorus concentrations to around the threshold at which the major changes in the Broads took place, because of technical problems in removing the last 5 or 10% of phosphate from the effluent and because of the increased background concentration from the land, which is not controllable. In this situation, various biological techniques, the husbandry of animals like water fleas which graze algae, for example, may be needed to bring about full restoration of the ecosystem.Recent Developments in the History of Chemistry Edited by C. A. Russell This book is intended primarily to inform chemists of recent progress in the history of chemistry. It originated from an initiative of the Historical Group of the Royal Society of Chemistry who for some considerable time had been aware of a rising surge of interest amongst chemists in the history of their subject. Yet there was also considerable frustration in obtaining reliable and up-to-date information, in understanding recent trends and in perceiving the relevance to specific problems of some of the less obviously ’chemical‘ writing of the last few years. Those with whom the Group was in touch included chemistry teachers wishing to introduc-perhaps only occasionally-historical elements into their school curricula. Others who expressed both interest and frustration were members of university and polytechnic chemistry departments, chemists in industrial research and those who had taken early or normal retirement. These are the readers for whom this book has been published, although it is hoped that professional historians of science may also find it to be of interest and value in its general surveys of the literature. ~ ~ ~~ Brief Contents: Introduction; Chemical Biographies; Chemical Education and Chemical Institutions; Chemistry to 1800; General and Inorganic Chemistry; Organic Chemistry; Physical Chemistry; Analytical Chemistry; Biochemistry; Instruments and Apparatus; Industrial Chemistry; Chemistry by Location in Western and Central Europe; Appendix I Periodicals for the History of Chemistry; Appendix II Some Useful Addresses; Author Index; Subject Index: People; Subject Index: Themes. Hardcover 344pp. ISBN 0 85186 917 3 Price f27.50 ($36.00) RSC Members f 12.00 Ordering : Non-RSC Members should send their orders to: The Royal Society of Chemistry, Distribution Centre, Blackhorse Road, Letchworth, Herts. SG6 lHN, U.K. RSC Members should send their orders to: The Royal Society of Chemistry, Assistant Membership Officer, 30 Russell Sauare, London, WC1 B 5DT.

ISSN:0144-557X    

DOI:10.1039/AP9852200192

出版商:RSC    

年代:1985

数据来源: RSC

摘要:

ANALYTICAL PROCEEDINGS, JULY 1985, VOL 22 197 Training Requirements for Atomic Spectroscopy The following is a summary of one of the papers presented at a Joint Meeting of the Atomic Spectrometry and Education and Training Groups held on December 4th, 1984, at the Geological Society, Burlington House, London, W. 1. Atomic Spectroscopy-The Short Course Organiser’s Approach Julian Tyson Department of Chemistry, The University of Technology, Loughborough, Leicestershire, LEI I 3TU The current situation in the UK for flame, furnace and ICP atomic-absorption methods appears to be that short courses are run by the water industry, various instrument manufac- turers and some academic institutions. The Water Industry Training Association (WITA) run about twelve courses each year on AAS.These consist of about five basic courses (application of AAS to analyses in the water industry) and a total of seven courses on the analysis of potable waters and the analysis of effluents and sludges, all by AAS. The courses are 3-4 days in length. About five of the major manufacturers between them run some 30 basic AAS courses per year (2-34 days), with about 20 to 25 ETA courses. Courses on “top of the range” AA and ICP can be conducted at the purchaser’s premises. The Polytechnic of North London puts on three courses ( 3 4 days), a general applied atomic spectroscopy course (every 3 or 4 years), an AAS course (every 1 or 2 years) and an ICP course (every 2 or 3 years). UMIST runs an ICP course (2 days) every year and Loughborough University of Technology (LUT) runs an AAS course (43 days) every year.Aims of the Courses Atomic Absorption Instrument manufacturers The aims of these courses are as follows. 1. The primary aim is to teach participants (mostly purchasers) how to use the instrument, the view being taken that effective use will only be made of the instrument (thereby showing it off in the best possible light) if the user understands how the instrument works and what the problems are going to be. 2. To teach the basic theory of atomic absorption. 3. To teach the basic principles of instrument design. 4. To teach the mechanisms of common interference effects and how they can be overcome. 5. To teach the principles of various atomisation devices (flames, furnaces, hydride generation, etc.). Academic courses In addition to the above aims the academic based courses also aim to achieve the following. 6.To put AA in perspective with other atomic-spectroscopy techniques. 7. To provide in-depth evaluation of instrument performance. 8. To discuss recent advances and likely future developments. 9. To provide examples of the applications of AAS. WITA courses These courses cover the basic aims 1 4 above, together with detailed consideration of the application of the technique to the analysis of clean and dirty water. Instrument manufacturers’ ETA courses These courses tend not to teach comparative atomisation methods but have the other basic aims, the assumption being that participants will have already attended a basic (flame) course. ICP courses Instrument manufacturers do not run courses as such but provide on-site training (in instrument operation) for pur- chasers. The UMIST and PNL courses aim to provide an extensive introduction but do not provide operator training.The courses cover basic principles and theory, instrument design, interferences (in nebulisation), applications, other plasmas (MIP, DCP) , recent developments and future trends. North London Polytechnic gives perhaps slightly more emphasis to the fundamental aspects of plasma processes and spectrometer design, with UMIST giving more emphasis to sample introduction and attendent problems. The aims of the various courses can be summarised as: instrument manufacturers, training in how to operate and optimise; WITA, as above plus how to analyse waters and sludges; academics, as above plus an understanding of the underlying principles, comparison and evaluation.Format of the Courses Flames and ETA Despite the differences in aims, the formats of the various courses are remarkably similar; namely a mixture of lectures (with the occasional tutorial) and practical work, with the allocated time split about 50 : 50 between the two activities. Extensive documentation is also supplied. Instrument manu- facturers use their own instruments and personnel, whereas the other courses use a mixture of this and external lecturers, with additional instruments on loan for the course. There appears to be excellent co-operation between the relevant companies and WITA and the various academic institutions. ICP Courses Despite valiant efforts by the organisers involved, it does not appear possible to adopt the same format for ICP as for AA.The reasons appear to be: ( a ) , the institutions do not have enough of their own equipment to provide useful hands-on training; (b), not enough space can be made available to accommodate a large number of instruments; and ( c ) , instru- ments are large and difficult to transport, install and get running reliably, and therefore manufacturers are reluctant to loan them. Two formats have been tried: firstly, a mixture of lectures plus trips to nearby instrument manufacturers or ICP users for demonstration sessions; and secondly, intensive198 ANALYTICAL PROCEEDINGS, JULY 1985, VOL 22 lectures, tutorials and discussion sessions supplemented by extensive documentation.Content of Courses Instrument Manufacturers’ Basic Course These consider principles, instrument design, calibration, optimisation, interferences, background correction, basic ETA, sample preparation, hydride generation, cold vapour mercury and discrete nebulisation. Instrument Manufacturers’ ETA Course Instrument design, method development, interference effects, background correction, minimising contamination and trouble shooting are covered. WITA These are as for the instrument manufacturers’ basic course, minus the specialised atomisation techniques but plus some relevant applications. North London Polytechnic AA Course This course is as for the instrument manufacturers’ courses but with less practical work. ICP Courses On these courses the origin of spectra, local thermal equilib- rium, line shape, temperature, plasma properties, torch design, spectrometers, nebulisers, optimisation, nebuliser interferences, spectral interferences, calibration, evaluation of performances, other sample introduction methods, other plasmas, ICP - MS, other recent developments, applications, limitations and costs are discussed. The LUT AA Short Course The general philosophy behind all of the short courses that we run in the Chemistry Department at LUT is: firstly, to promote links with industry through contact with course members, instrument manufacturers and related companies; secondly, to keep abreast of instrumental developments; and thirdly, to make money! The money does not go into the organisers’ pockets, but is used to employ much needed support staff, send academic staff and research students to meetings and confer- ences, buy small items of equipment, books, subscriptions to journals, etc.In running these courses we recognise that it is not possible to please all of the participants all of the time, given the wide differences in backgrounds and reasons for attending the course. However, the aims (see earlier) are clearly set out in the descriptive brochures, and there should not be anyone on the courses who will not be pleased at least some of the time. Within the basic lecture - practical format there are plenty of opportunities for discussion between course members and external tutors. The sequence of topics covered in the lectures is basic theory, basic instrument, flames, interferences, calibration, ETA, background correction, hydride generation, instrument design, optimisation, instrument evaluation, ICP, recent de- velopments in flame techniques, atomic fluorescence, compari- son of atomic-spectrometric techniques, applications.As far as is possible, acknowledged experts are invited to give the lectures; thus, in addition to LUT staff, there are some ten or eleven external lecturers. Inevitably there is some overlap between speakers, but this is considered to provide reinforce- ment of ideas rather than as boring repetition. The lecture - tutorial hours total 17. Just as there is considerable agreement between course organisers on the basic format and content of lectures, there is also agreement on the nature of the basic experiments.The LUT course has set experiments on the effect of operating parameters, interferences, calibration and detection limit, background correction, use of the dinitrogen oxide - acetylene flame (in AA and AE), ETA, flow injection and hydride generation. Other courses include experiments on solvent extraction, fault finding, use of an autosampler, the Delves cup, reducing sensitivity and nebuliser adjustments. The LUT practical timetable is highly structured to start with and then becomes more flexible to allow participants (a) to use a particular manufacturer’s equipment, ( b ) extended use of a particular technique, (c) to run their own samples, ( d ) discussion with course tutors. The practical work occupies some 14 hours of which 4 are optional. The course runs under a number of constraints.The timing is set by the availability of the laboratory space (only during vacations) and the requirements to run about 6 courses on a variety of analytical techniques during the year. No allocation of time is given to organisers; the organisation and running has to be fitted in with other demands. There is a limit to the number of instruments that can be made available and hence the numbers of participants (not more than three per instru- ment) and the external speakers are not always available either. Conclusions With regard to AA, there is obviously a considerable demand for basic operator training courses as purchasers prefer to have their operators trained by the manufacturers rather than do it themselves. This makes sense, as the manufacturers do a good job and the minimum number of the purchaser’s personnel are involved in non-productive work. There is a more limited demand for a broader (and usually longer) course, which provides education as well as training. As far as ICP is concerned, the education and training aspects are separated. There is a limited demand for the intense education course. It is a pleasure to acknowledge the help that I have received in preparing this material from Dr. A. Brown (WITA, Burn Hall, Huby, Sutton on Forest, York YO6 lJB), Dr. E. Steers (School of Applied Physics, Polytechnic of North London, Holloway Road, London N7 SDB), Mr. N. Barnett (DIAS, UMIST, P.O. Box 88, Manchester M80 1QD) and from my various contacts at Pye Unicam, Instrumentation Laboratory, Perkin-Elmer, Baird Atomic, Varian Associates, ARL, Beck- man and V. A. Howe.

ISSN:0144-557X    

DOI:10.1039/AP9852200197

出版商:RSC    

年代:1985

数据来源: RSC

摘要:

ANALYTICAL PROCEEDINGS, JULY 1985, VOL 22 199 Modified Electrodes The following are summaries of three of the papers presented at a Joint Meeting of the Analytical Division and the Electroanalytical Group held on February 6th, 1985, in the Scientific Societies Lecture Theatre, London, W.1. Trace Metal Analysis with no Mercury and no Deoxygenation D. M. T. O‘Riordan and G. G. Wallace Chemistry Department, University College Cork, Cork, Ireland Along with the fact that the area of chemically modified electrodes is an interesting and fascinating field of research let me offer some practical justification, as an (electro) analytical chemist, for working in this area. Voltammetric methods of analysis have proved extremely useful over the last two decades,ll2 and the number of applications has grown as a result of the introduction of “push-button, do-all” instrumentation.In our laboratories we are primarily concerned with the analysis of metal ions using electroanalytical methods, either combined with HPLC374 or using the more conventional technique of stripping voltam- metry (see relevant publications cited in references 1 and 2). Work in this area has brought to light two problems, which, despite the recent advances in voltammetric instrumen- tation, still remain with us. They are, the limitations often imposed by the use of a mercury-drop electrode and the complications caused by the presence of dissolved oxygen in samples. The mercury-drop electrode, although an excellent general- purpose, suffers several drawbacks. Amongst them is the instability of the electrode in flowing solutions (particularly troublesome when using HPLC or on-line electroanalysis) .Furthermore, the lack of selectivity of the mercury drop electrode, in that it is prone to interference from adsorbed species and the possibility of interference from the formation of intermetallic compounds during stripping voltammetry, can cause problems. The use of solid electrodes as alternative working electrodes for the voltammetric determination of metals has proved relatively unsuccessful. Oxide layer formation on gold and platinum5 and high residual currents on glassy carbon,6 along with the irreproducible nature of solid electrode surfaces, has limited their deployment. A compromise electrode in the form of a mercury thin film electrode (MTFE) has been relatively successfu1,7 but still the range of working electrodes available to the electroanalytical chemist is severely limited.Dissolved oxygen, as a result of the electrochemical reduc- tion processes it undergoes, causes several problems during voltammetric determinations of metals.8 The presence of dissolved oxygen in normal aqueous solutions at the 10-3 M level ensures that it is present in a substantial excess compared with the trace levels (generally less than 10-7 M) of the metal analytes. Despite the efforts of many workers to devise a method for its remova1,s none has succeeded in providing a truly efficient system. The difficulties encountered in trying to remove dissolved oxygen from solution are accentuated under dynamic conditions (as for HPLC and on-line electroanalysis) or with techniques that involve long analysis times (as is the case with stripping voltammetry, because of the relatively long times required for deposition).A voltammetric method of analysis, employing a working electrode that enables the determination of metals with no interference from dissolved oxygen, is required. An Alternative Technique The area of chemically modified electrodes has matured to such an extent that the application of these devices as working electrodes’ in electroanalytical methods is now a feasible alternati~e.9~10 Research at present in progress in our labora- tories is addressing some of the aforementioned problems, which are associated with the limited choice of working electrode. We are interested in the development of mercury free working electrodes which can be used for trace metal analysis without interference from dissolved oxygen (it does not sound like much if you say it quickly!).We are encouraged by the fact that other workers11 have employed chemically modified electrodes for the analysis of certain metal ions at potentials free from oxygen interference by catalysing reduc- tions at less negative potentials. It is also well known that the redox properties of metal centres can be controlled using complexation. 12 Takata and Fujita13 have used this property to advantage in devising methods of analysis for metal species free from oxygen interference. Various workers have described electrode surfaces on (or in) which metal species may be trapped.14-17 One problem with those previously described is that uptake appears relatively slow and in fact is often only induced under rather severe reaction conditions.Other considerations, such as saturation and selectivity of modified electrodes as well as matrix effects, have been discussed in a recent article dealing with the use of chemically modified electrodes for electroanalysis. 18 A (Po1y)pyrrole-N-carbodithioate Electrode We have recently been involved with the determination of metal ions by HPLC using dithiocarbamate ligands as derivatis- ing agents.3.4 Several properties of the dithiocarbamate ligand 4 s ~s - R = alkyl R\ R/N-c Dithiocarbamate ligand (dtc) make the design and development of chemically modified electrodes containing this species attractive.Firstly, the species forms stable complexes with a wide range of metals374J9-21; secondly, selectivity can be achieved either by pH control or by using specific masking agents19 ; and thirdly, metal dithiocarb- mate complexes are amenable to both electrochemical reduc- tion and oxidation.20 For example, [C~(II)(dtc)~l + e S [Cu(I)(dtc)*]- . . . . [C~(II)(dtc)~1= [Cu(III)(dtc)2]+ + e . . . . [21200 ANALYTICAL PROCEEDINGS, JULY 1985, VOL 22 As dithiocarbamate ligands are readily \ynthesised by the reaction of amines (either primary or secondary) with carbon disulphide,20 we have employed modified electrodes with well defined amine containing polymers as a backbone for further derivatisation. Polypyrrole is one such polymer and it is readily plated on to an electrode surface by oxidation of pyrrole in acetonitrile solution22.23 according to the following reaction: r 1 l H H i x H (However, there appears to be some variation in the number of pyrrole groups the positive charge is spread over.23) We have recently derivatised this electrode further accord- ing to I H This process has been monitored using analytical techniques such as: Fourier transform infrared spectrometry (FTIR), to monitor the “thioureide” band due to the v(C-N) vibration.as well as the v(C-S) vibrations on the derivatised electrode; electron probe microanalysis (EPMA) to gain information on the elemental composition of the electrode surface; scanning electron microscopy (SEM) to monitor any changes in mor- phology of the electrode surface; cyclic voltammetry to monitor (-N-H) and (-N=C<S-) as well as metal complex oxidation and reduction.25 To date, only the uptake and voltammetric responses of the copper ion have been con- sidered in detail. Other metals are currently under investiga- tion.The (po1y)pyrrole-N-carbodithioate electrode is capable of trapping metal ions from solution to provide a surface concentration of -6 x 10-3 rnol cm-3 of copper ions. This compares with concentrations of 10-i rnol cm-3 for poly- pyrrole25 and 2 x 10-7 mol cm--7 for lead at a mercury drop electrode.26 The rate of uptake of mol min-I on the dithiocarbamate electrode compares with uptake of lead on the mercury drop electrode (0.4 x 1 0 - 7 mol min-’).26 Polypyrrole itself has a rate of uptake of 3 x 10-12 rnol min-l;’5 this is extremely slow.A well defined reduction response for the copper dithiocar- bamate complex attached to the electrode surface was obtained.25 This response corresponds to reduction of the complex dissolved in solution, indicating that although the pyrrole group is polymerised this has little effect on the electrochemistry of the dithiocarbamate complex. Unfortu- nately, the oxidation of copper complex trapped on the surface is not so well defined. This is presumably because of the fact that metal complexes containing the pyrrole-N-carbodithioate ligand are relatively difficult to oxidise27 and the polypyrrole electrode itself is involved in an electrochemical oxidation process at these positive potentials.25 Further work is being carried out to elucidate these problems.Conclusions and Future Developments Mercury has been, and probably will be for several more years, the electroanalytical chemist’s first choice for the determina- tion of metal species in solution. However, an alternative is emerging. Chemically modified electrodes can be designed and developed with a specific application in mind. which should make their use for particular analyses very attractive. In this work we have described a (po1y)pyrrole-N- carbodithioate electrode for the trapping and voltammetric analysis of copper ions. Determination of other ions using this electrode is currently under investigation and more quantita- tive work on all metals is being carried out. Through the development of electrodes such as are des- cribed in this paper electroanalytical chemists may soon become aware that the old faithful mercury electrode is not the be all and end all of voltammetric analysis. The authors would like to thank M.A. McKervey (Chemistry Department, University College Cork) for invaluable discus- sions and advice on the derivatisation of polypyrrole. 1. 2. 3. 4. 5 . 6. 7. 8. 9. 10. 11. 12. 13. 14. 15. 16. 17. 18. 19. 20. 21. 22, 23. 24. 25. 26. 27. References Vydra, F.. Stalik, K., and Julakova, E.. “Electrochemical Stripping Analysis,” Ellis Horwood, New York, 1976. Bond, A . M., “Modern Polarographic Methods in Analytical Chemistry.” Marcel Dekker, New York, 1980. Heneghan. G., O’Riordan. K., and Wallace. G. G . . Anal. Chem., 1985, in the press. Bond, A. M.. and Wallace. G . G . .Anal. Chem., 1984.56.2085 (and references cited therein). Adams. R. N., “Electrochemistry at Solid Electrodes.” Marcel Dekker. New York, 1969. Van der Linden, W. E., and Dieker. J. W., Anal. Chim. Acta. 1980, 119, 1. Perone. S . P.. and Davenport, K. K . , J . Electroanal. Chem.. 1966, 12, 269 (and references cited therein). Wallace, G. G . , Trend5 Anal. Chem., submitted for publica- tion (and references cited therein). Murray, R . W., Acc. Chem. Res.. 1980. 13. 135. Faulkner, L. R.. Chem. Eng. News, 1984, Feb. 27. Larochelle, J . H., and Johnson, D. C., Anal. Chem., 1978. 50, 240. Bailar. J . C . , “The Chemistry of the Coordination Com- pounds,” Reinhold, New York, 1956. Takata, Y., and Fujita, K., J. Chromatogr., 1975. 108, 255. Cheek. G. T . , and Nelson, R .F., Anal. Lett., 1978, 11. 393. Oyama, N., Sato, K., and Matsuda, H., J . Electroanul. Chem., 1980, 115, 149. Oyama, N.. and Anson. F. C . , J . Electrochem. SOC., 1980,127, 247 and 249. Martin, C. R., Rubinstein. I., and Bard, A. J., J . A m . Chem. Soc., 1982, 104, 3817. Guadalupe, A . R . , and Abruna. H. D., Anal. Chem.. 1985,57, 142. Hulanicki. A . , Talanta, 1967, 14, 1371. Concouvanis, D . . Znorg. Chem., 1979, 26, 301. Heneghan, G., and Wallace, G. G., unpublished work, University College Cork, 1984. Kanazawa, K. K . , Diaz, A . F., Gardini, W. D., Gill, W. D., Grant. P. M., Kwak, J . F., and Sgreet. G . B., Synth. Met., 1980, 1 , 329. Diaz, A . F.,Mart~nez, A.. Kanazawa, K. K., and Salmon, M., J . Electroanal. Chem., 1981, 130, 181. Asavapiriyanont, S ., Chandler, G . K., Gunawardena, G . A., and Fletcher, D . , J. Electroanal. Chem., 1984, 177, 229. O’Riordan, D . M. T., and Wallace, G. G., unpublished work, University College Cork, 1984. Batley, G. E . , and Florence, T. M., J. Electroanal. Chem., 1974, 55, 23. El A’rnma, A. G . , and Diago, R . S . , Inorg. Chem., 1977, 16, 2975.ANALYTICAL PROCEEDINGS, JULY 1985, VOL 22 Ferrocene-based Enzyme Electrodes 201 H. Allen 0. Hill Inorganic Chemistry Laboratory, South Parks Road, Oxford OX1 3QR Most interesting redox enzymes require attendant proteins that act as mediators of electron transfer. The most ubiquitous example is cytochrome oxidase , which receives electrons from cytochrome c and catalyses the reduction of dioxygen to water. Other important oxidoreductases such as the cytochrome P-450 monooxygenases use either ferredoxins or flavoproteins as the immediate electron donor.Although it had long been thought that rapid, reversible and direct electron transfer of redox proteins at electrode surfaces was not possible, we have shown'-12 that, providing the surface of the electrode is electrostatically compatible with the surface of the protein and, in particular, that part of the surface of the latter which is conducive to electron transfer, relatively rapid electrode kinetics ensue. Thus, the electrochemistry of many cyto- chromes c is well behaved at gold, upon which 4,4'-bipyridyl or related compounds is adsorbed, pyrolytic graphite , which has the edge plane exposed, or the metallic oxide ruthenium dioxide.Having achieved such facile electron transfer, how can one make use of it? We have shown13-17 that it is possible to couple the electrode reaction to enzymes for which the redox proteins act as co-factors. These include cytochrome oxidasel3 (nitrite reductase) from Pseudomonas aeruginosa, a carbon monoxide oxidase ,I6 methanol dehydrogenasel7 and even mammalian cytochrome oxidasel5 in situ, i.e., in the isolated mitochondria. Obviously, in a properly coupled system the current passed reflects quantitatively the presence of the substrate, be it dioxygen, carbon monoxide or methanol, as in the above examples. Although cytochrome c is a stable, well behaved protein, we sought an even more stable alternative capable of more readily accessible variations. We thus arrived at fer- rocenes as suitable analogues of cytochromes c.Ferrocene, q5-bis-cyclopentadienyliron(II), has the following properties in common with the cytochromes c: the two accessible oxidation states are iron(I1) and iron(II1); both of these are low-spin; the low oxidation state forms are only slowly autoxidised; the intermolecular electron exchange rates are fast. Ferrocenes have additional advantages: many substituted ferrocenes are available, conferring on the compounds different over-all charges and a wide range of solubilities in different solvents; most are heat stable; they can be polymerised; they can be used to modify other molecules including proteins. The most important characteristic is the separation of functions. Thus, it is possible to introduce substituents on either or both of the cyclopentadienyl rings and yet retain the properties of a simple one-electron redox couple.Of course, the formal potential is responsive to the substituent(s) , but the electron transfer reactions retain their desirable characteristics of rapidity and reversibility . The first enzyme to which it was shown17 that ferrocene acted as an effective mediator was methanol dehydrogenase. This non-NAD dependent enzyme could therefore be employed in an assay method for methanol and other primary alcohols. The major exploitation, to date,lg is in a new type of glucose electrode. Previous electrodes used for this purpose had actually analysed the dioxygen consumed or the hydrogen peroxide produced as a consequence of the enzyme-catalysed reaction: glucose + 0 2 + gluconolactone + H202 The ferricinium ion acts as a competitive electron acceptor and so the following sequence occurs: glucose + glucose oxidase(,,) -+ gluconolactone + glucose oxidase(,,d) glucose oxidase(,,d) + 2Fc+ + glucose oxidase(,,> + 2Fc + 2H+ 2Fc = 2Fc+ + 2e- The electrode that results gives a linear response from 0-40 mM of glucose, is relatively insensitive to oxygen, is pH- independent within the relevant range, functions in plasma or whole, heparinised blood and is quite insensitive to interfer- ence.Of course, it is possible to couple glucose to other substrates in the presence of the appropriate enzymes. Thus, assays for adenosine triphosphate and creatine kinasel9 have been developed using the ferrocene-based glucose electrode.It has recently been shown20 that ferrocene will act as a mediator to a wide range of enzymes including pyruvate oxidase, amino-acid oxidase and lactate oxidase. Developments of assays based on these enzymes, used in conjunction with ferrocene, can be expected. The synthetic versatility of ferrocene is exemplified21 in a new form of immunoelectrode. In this, the ferrocene is conjugated to the antigen to give a derivative Fc-Ag. The assay rests on the observation that this modified ferrocene still acts as a mediator to glucose oxidase. However, its ability to act as a mediator is greatly impaired when it is bound to the appro- priate antibody (Ab). If Fc-Ag can be displaced from Ab by the unmodified antigen, Ag, then a competition assay for Ag results with the amount reflected in the difference current.This has been applied with success to a new assay for theophylline. Other applications are likely to be forthcoming, although more sensitive amperometric methods will need to be developed before these immunoelectrodes will be competitive with the more traditional radioimmunoassays. I thank my many collaborators whose names are given in the references. The collaboration with Professor I. J. Higgins at the Cranfield Institute of Technology has been most produc- tive. I am grateful to Drs. A. E. G. Cass, G. Davis, M. J. Green and D. Scott for permission to use unpublished work. I am grateful to the Science and Engineering Research Council and Genetics International Inc. for support. 1. 2. 3. 4. 5. 6. 7. 8. 9.10. 11. 12. 13. References Eddowes, M. J . , and Hill, H. A . O., J. Chem. SOC. Chem. Commun., 1977,771. Eddowes, M. J . , and Hill, H. A. O., J . Am. Chem. SOC., 1979, 101,4461. Albery, W. J., Eddowes, M. J . , Hill, H. A. O., and Hillman, A. R., J . Am. Chem. SOC., 1981, 103, 3904. Eddowes, M. J., and Hill, H. A . O., Biosci. Rep., 1981, 1,521. Eddowes, M. J , and Hill, H. A. O., ACS Adv. Chem. Ser., 1982,201, 173. Armstrong, F. A., Hill, H. A. O., and Walton, N. J . , FEBS Lett., 1982, 145, 241. Eddowes, M. J . , and Hill, H. A. O., Faraday Disc. Chern. SOC., 1982,74, 331. Hill, H. A . O., Biochem. SOC. Trans., 1983, 11,453. Armstrong, F. A . , Hill, H. A. O., Oliver, B. N., and Walton, N. J . , J . Am. Chem. SOC., 1984, 106,921. Harmer, M. A., and Hill, H. A. O., J .Electroanal. Chem., 1984, 170, 369. Armstrong, F. A., Hill, H. A. O., and Oliver, B. N., J . Chem. SOC. Chem. Commun., 1984, 976. Allen, P. M., Hill, H. A. O., and Walton, N. J . , J. Electroanal. Chem., 1984, 178, 69. Hill, H. A. O., Walton, N. J . , and Higgins, I. J., FEBS Lett., 1981, 126,282.202 14. Hill, H. A. O., and Walton, N. J., J. Am. Chem. SOC., 1982, 104, 6515. 15. Coleman, J. 0. D., Hill, H. A. O., Walton, N. J., and Whatley, F. R., FEBS Lett., 1983, 154, 319. 16. Turner, A. P. F., Aston, W. J., Bell, J., Colby, J., Davis, G., Higgins, I. J., and Hill, H. A. O., Anal. ChYn. Acta, in the press. Davis, G., Hill, H. A. O., Aston, W. J., Higgins, I. J., and Turner, A. P. F., Enzyme Microbial Technol., 1983,5, 383. 17. ANALYTICAL PROCEEDINGS, JULY 1985, VOL 22 18.Cass, A. E. G., Davis, G., Francis, G. D., Hill, H. A. O., Higgins, I. J., Plotkin, E. V., Scott, L. D. L., and Turner, A. P. F., Anal. Chem., 1984, 56, 667. Green, M. J., Davis, G., and Hill, H. A. O., J. Biomed. Eng., 1984, 6 , 176. Cass, A. E. G., Davis, G., Green, M. J., and Hill, H. A. O., J. Electroanal. Chem., in the press. Green, M. J., Hill, H. A. O., McNeil, C. J., and Scott, D., unpublished work. 19. 20. 21.ANALYTICAL PROCEEDINGS, JULY 1985, VOL 22 203 Kauffmann, J.-M., Laudet, A., Patriarche, G. J., and Christian, G. D., Talanta, 1982, 29, 1077. Kauffmann, J.-M., Patriarche, G . J., and Genies, E. M., Electrochim. Acta, 1982, 27, 721. Kauffmann, J.-M., Vire, J.-C., and Patriarche, G. J., Bioelectrochem. Bionerg., 1984, 12, 413.Kauffmann, J--M., Laudet, A., Vire, J.-c-, and Patriarche, G. J.9 Microchem. J., 1983, 28,357. Prete, M. p., Kauffmann, J.-M., Vire, J--c., GeuSkenS, G-, Debye, B., and Patriarche, G. J., Anal. Lett., 1984, 17, 1391. Patriarche, G. J., Kauffmann, J.-M., and Vire, J.-c., J . Pharm. Biochem. Anal., 1983, 1, 469. Three of the World’s leading. Analytical Journals The Analyst An international journal of high repute containing original research papers on the theory and practice of all aspects of analytical chemistry drawn from a wide range of sources. It also publishes regular critical reviews of important techniques and their applications, short papers and urgent communications (which are published in 5-8 weeks) on important new work, and book reviews. 12 issues per annum (plus index) Analytical Proceedings Analytical Proceedings is the news and information journal of the Analytical Division.It contains special articles, reports of meetings, extended summaries of original papers, safety articles, details of recent legislation, surveys of equipment, and many other items of general interest to analytical chemists both in Britain and overseas. 12 issues per annum (plus index) Analytical Abstracts Analytical Abstracts endeavours to cover the whole field of analytical chemistry, providing more than 12,000 abstracts per annum of papers and books considered to be of importance and interest to analytical chemists. Indexes are published annually. 12 issues per annum (plus indexes) SUBSCRIPTION RATES 1985: The Analyst UK f134.50, USA $260.00, Rest of World f141.00 Analytical Abstracts UK f 199.50, USA $385.00, Rest of World f209.50 Analytical Proceedings UK f 63.50, USA $1 23.00, Rest of World f 67.00 The Analyst, Analytical Abstracts, and Proceedings UK f330.00, USA $638.00, Rest of World f347.00 As above with Abstracts on one side of the paper UK f345.00, USA $666.00, Rest of World f362.00 The Analyst and Analytical Abstracts UK f293.00, USA $566.00, Rest of World f308.00 As above with Abstracts on one side of the paper UK f312.00, USA $603.00, Rest of World f327.00 The Analyst and Proceedings UK f169.00, USA $327.00, Rest of World f 178.W Analytical Abstracts on one side of the paper UK f217.50, USA $420.00, Rest of World f228.00 ORDERING: Orders should be sent to: The Royal Society of Chemistry, Distribution Centre, Blackhorse Road, Letchworth, Herts SG6 lHN, England. The Royal Society of Chemistry Burlington House Piccadilly London W1V OBN

ISSN:0144-557X    

DOI:10.1039/AP9852200199

出版商:RSC    

年代:1985

数据来源: RSC

摘要:

204 ANALYTICAL PROCEEDINGS, JULY 1985, VOL 22 Fibre Optics for Chemical Sensing The following are summaries of seven of the papers presented at Joint Meeting of the Special Techniques Group and the Institute of Physics held on February 20th, 1985, at Imperial College, London S.W.7. Optical Fibre Sensors in Chemical Analysis R. Narayanaswamy Department of Instrumentation and Analytical Science, UMIST, P.O. Box 88, Manchester M60 IQD The development of optical fibre sensors has been a growing new technology. Much of the work has been concerned with the measurement of physical parameters, such as pressure, temperature, flow, level, etc., and interest in the development of chemical sensors based on optical fibres has been recent. These devices, based on a simple concept, involve a reagent phase coupled to an optical fibre, and measure concentrations, often at trace level, through a change in the optical properties of the reagent. Light from a suitable source travels along an optical fibre from one end to the other, where it is returned to the same or another optical fibre by reflection or scattering or by luminescence stimulated by source radiation, and finally to a light measuring system.Optical fibre chemical sensors offer several advantages over the conventional sensors. These optical sensors can measure concentrations without significantly perturbing the sample, and they can be used for continuous sensing. Because the signal is optical, these sensors are not subject to electrical interfer- ence and do not require a “reference” like the potentiometric sensors.The referencing is carried out optically by the use of the ratio method, in which part of the conducted light not affected by the measurement variable can be used to correct for other optical variations. It is possible to achieve high degrees of stability by this method. Optical fibre chemical sensors can offer significant c a t advantages over other conventional sensors, and will be particularly useful in clinical and other biomedical applications because they involve no electrical connection, transmit only low-energy radiation and are cap- able of miniaturisation. These sensors are also subject to several limitations. For example, ambient light will interfere with the sensors; therefore, they must be used in a dark environment or the optical signal should be modulated.Because the reagent and the analyte are normally in different phases there is necessarily a mass-transfer step before constant response is reached which, in turn, limits the response times of the optical fibre sensors. In these sensors, the measurement is equilibrium based rather than rate or diffusion dependent, and the specificity of the measurement is achieved by chemical instead of physical means. Reversible and specific spectropho- tometric reactions are available for most chemical and bio- chemical constituents of interest, and sensors can be developed to respond to analytes for which other types of sensors are not available. It is possible to use two or more probe detection wavelengths and have sensors that respond simultaneously to two or more analytes.In general, optical fibre chemical sensors will have a limited dynamic range. They are still too new to be of proven value in most applications, but their potential is considerable. Two general types of optical fibre sensors have been developed for chemical species. The first type is the photo- metric or bare ended fibre. Here the optical fibres are used as light carriers in techniques such as reflectometry, spectropho- tometry and fluorimetry, and do not contain a transducer of any sort at the end of the fibre. The second type of chemical sensor is the most recent type and will be the subject of this paper. The techniques employed in these sensors have normally been based on the immobilisation of chemical reagents and integrated with optical fibres. These devices may involve either bifurcated optical fibre bundles or a single optical fibre (Fig.1). In bifurcated optical fibre sensors [Fig. l(a)] separate fibres transmit incident and detected radiation and observe only that part of the reagent phase that falls within both the cone of incident radiation and the cone of detected radiation. When single optical fibres are employed [Fig. l(b)], it will be necessary to distinguish between the incident and detected radiation either temporally or by wavelength to avoid problems due to scattered radiation. Optical fibres can be coated on the outside with reagent phase [Fig. l(c)], which modifies the transmission characteristics of the fibre by the change in refractive index of the coating resulting from chemical reaction of the reagent.The diameter of the optical fibres employed can be of the order of a few micrometres and sensors in the sub-millimetre size range have already been developed. The instrumentation associated with optical fibre chemical sensors is optical as well as electronic and many of these items are well developed. Hence, progress in the field of chemical sensors based on optical fibres is limited only by the sensor development. These sensors may be classified as reversible, in which the reagent phase is not consumed by its (a) Incident - incident light - To detector ( C) - To detector - Incident light - Reagent phase Fig. 1. Sensor based: (a) on bifurcated optical fibre bundle; ( b ) , on a single optical fibre with reagent phase at the end of the fibre; ( c ) , with reagent phase coated on the outside of the single optical fibreANALYTICAL PROCEEDINGS, JULY 1985, VOL 22 205 interaction with analyte, or non-reversible, in which the reagent phase is consumed.The latter type of sensor can be useful for sensing if the relative consumption of the reagent phase is small or if it should be possible to renew the reagent phase. A variety of optical sensors have been developed for a number of chemical species (ions and molecules) that are based on the use of reagents immobilised on a stable solid support. Although optical fibres are employed to transmit optical radiation, only a few of the sensors reported have the configuration of a probe. Depending on the particular device, the optical property measured is absorbance, reflectance or luminescence.A pH sensor based on absorbance of phenol red indicator dyestuff immobilised by covalent bonding on to poly- acrylamide microspheres has been developed to measure in the physiological pH range of 7.0-7.4, with an accuracy and precision of kO.01 pH and a temperature coefficient of 0.017 pH OC-1.1 The probe design is similar to that in Fig. l(a) but uses two plastic optical fibres (diameter = 0.15 mm) instead of a bifurcated bundle of optical fibres to transmit incident and detected radiation. When solid support matrices are employed in sensors, it will be difficult to measure the transmitted light. In these instances the intensity of the light reflected may be used as a measure of the change in colour of an immobilised reagent phase.As the measurement of absorbance may be difficult, reflectance measurement will be attractive for sensors that involve colorimetric reactions. An optical fibre pH sensor based on reflectance measurement has been developed.2 The probe utilises a plastic optical fibre bundle of nominal diameter 1 mm, transparent in the visible region of the electromagnetic spectrum. At the end of the fibre a sensitive tip is fabricated consisting of a styrene - divinylbenzene copolymer on to which the indicator dyestuff bromothymol blue is adsorbed. The indicator matrix is retained in position at the end of the optical fibre using a membrane of PTFE (Fig. 2). Changes in pH in the vicinity of the sensitive tip cause a variation in the attenuation of specific reflectance bands.The measured response corre- sponds to the attenuation of the reflected radiation at 593 nm with a reference wavelength of 800 nm, to provide the optimum dynamic range with respect to pH. The probe is capable of measuring pH in the range 7.0-9.0 with a temperature coefficient of 0.015 pH "C. - 1 Stable response for a step change in pH of 1 is obtained from the probe within 5 min. However, when the pH probe is subject to a large, abrupt step change in pH, only 50 s are required to achieve a stable response. An important contributing factor to the response time is thought to be the rate of diffusion of solvent and ions across the PTFE membrane and through the reagent - polymer matrix packing of the probe. A reversible optical waveguide sensor based on reflectance has been reported for ammonia vapours.3 A small glass capillary tube (1.1 mm 0.d.and 0.8 mm i.d.) is fitted with a pulse modulated LED (560 nm) at one end and a phototransis- tor detector at the other end to detect the colour change. The tube is coated with a thin solid film composed of an oxazine perchlorate dye [as in Fig. l(c)], which changes its eolour rapidly from blue to red when exposed to ammonia vapour. The colour of the dye is reversed when the ammonia vapour is Sheathing Sealing tube removed. With this device it was possible to detect ammonia vapour concentrations down to values as low as 60 p.p.m. in ambient air at a relative humidity of 40%. Measurement of fluorescence is an extremely sensitive technique capable of measuring low analyte concentrations.It is particularly well suited for optical sensing and the wavelengths of detected and incident radiations are different. Hence, single optical fibres [as in Fig. l(b)J can be employed in these sensors. The geometry of an optical fibre fluorescence sensor would correspond to front-surface detection [Fig. l(a) and (b)] and the response from such a sensor would be linear at very low analyte concentrations. The simplest type of fluores- cence sensor involves measurement of fluorescence at a single wavelength. For example, a pH sensor based on the fluores- cence of cellulose immobilised fluoresceinamine has been reported,4 and in this sensor the base form of the reagent is fluorescent. Therefore, as pH increases there is an increase of fluorescence as the acid form of the dye is converted to its base form.The useful working pH range of this sensor is 3-6 and the sensor reaches a steady fluorescent intensity in about 15-30 s. Another type of fluorescence based sensor would be the use of non-fluorescent reagents that form fluorescent complexes with metal ions and these reactions can be used for sensing metal ions. For example, the reagent morin (3,5,7,2',4'-penta- hydroxyflavone) immobilised on cellulose by covalent bonding is only weakly fluorescent by itself. However, by reaction with aluminium or beryllium ions, it forms intense fluorescent complexes. The detection limits reported with these sensors are 27 p.p.b. for aluminiums and 9 p.p.b. for beryllium.6 An optical fibre glucose sensor based on fluorescence measurements and on the principle of competitive binding of the analyte (glucose) on a substrate with a fluorescence labelled indicator has been developed by Schultz et aZ.7 In this affinity sensor, concanavalin A (Con A) was used as the substrate, immobilised on the inside of a dialysis hollow fibre.Con A has a specific binding character for glucose. Fluorescein labelled dextran (Fl-Dextran) is the competitive ligand. In the sensor element, the following reversible reactions take place Glucose + Con A e Con A-glucose . . (1) F1-Dextran + Con A Con A-Fl-Dextran . . (2) When the sensor element is exposed to higher concentrations of glucose, reaction (1) is displaced to the right and reaction (2) to the left, which results in an increase in free F1-Dextran, the luminescence of which is measured.However, a reduction in glucose concentration results in a decrease in the fluorescence signal. The approach can be used for a wide variety of organic species provided that suitable specific affinity substrates and competing ligands can be found. Sensors can also be based on a decrease in fluorescence intensity of the reagent phase upon reaction with the analyte. Although this system is less desirable, the detection of certain analytes becomes possible. For example, a sensor for measuring physiological oxygen partial pressure has been developed,s which is based on the quenching of the fluorescence of perylene dibutyrate dye adsorbed on to polystyrene. The matrix polystyrene is hydro- phobic and highly permeable to oxygen.The fluorescence of the dye is excited by incident radiation through one optical fibre and observed through the other. Fibre core Indicator support polymer I / / / / / / ' / / / / / I W Fibre suppoh polymer Guide tube Membrane Fig. 2. Cross-sectional diagram of the optical fibre probe design based on reflectance (from reference 2)206 ANALYTICAL PROCEEDINGS, JULY 1985, VOL 22 Conclusion Optical fibre sensors for chemical analysis are clearly attractive in concept and they have great potential. Their success requires the development of appropriate reagent phases. These sensors are mostly in a development stage; however, the devices studied to date have illustrated the possibility of a wide variety of approaches and systems. Interest in optical fibre sensors for chemical species is certain to intensify in the next few years.There is substantial interest in the development of new sensors for various environmental, agricultural, clinical and other biomedical applications. Although this technology may lead to improvements in traditional sensors, the most important application would be in the development of entirely new devices. 1. 2. 3. 4. 5. 6. 7. 8. ~ References Peterson, J. I., Goldstein, S. R., Fitzgerald, R. V., and Buckhold, D. K., Anal. Chem., 1980, 52, 864. Kirkbright, G. F., Narayanaswamy R., and Welti, N. A., Analyst, 1984, 109, 1025. Giuliani, J . F., Wohltjen, H., and Jarvis, N. L., Optics Lett., 1983, 8, 54. Saari, L. A., and Seitz, W. R., Anal. Chem., 1982, 54, 821. Saari, L. A., and Seitz, W.R., Anal.Chem., 1983, 55, 667. Saari, L. A., and Seitz, W. R., Analyst, 1984, 109, 655. Schultz, J . S., Mansouri, S., and Goldstein, I. J., Diabetes Care, 1982, 5 , 245. Peterson, J. I., Fitzgerald, R. V., and Buckhold, D. K., Anal. Chem., 1984, 56,62. Low-cost Fibre Optic Chemical Sensors T. E. Edmonds and 1. D. Ross Department of Chemistry, Loughborough University of Technology, Loughborough, Leicestershire LEI I 3TU Introduction In the last few years there has been increasing interest in the field of chemical sensors; so much so that even the casual observer can identify the rumbling of a “band-wagon”! In spite of the fact that some chemical sensors have been with us for many decades (e.g., the pH electrode and the oxygen electrode), it is only recently that the unique advantage of these devices has been recognised, viz., the expertise of the analytical chemist and the chemistry of the analytical method may be encoded in a simple rugged device.Such devices have coinsiderable potential for use in medical, industrial and hazardous environments. Fibre-optic probes are chemical sensors, but compared with, say, piezoelectric crystals or ion-sensitive electrodes, the over-all chemical transducer is rather complicated. Most sensors can be considered as consisting of a relatively simple transducer that is rendered selective by a coating or membrane of some form or another. The fibre-optic probe, on the other hand, has a complex transducer consisting of two optical - electronic arrangements associated with the light source and the signal detector.It is the intention of this paper to demonstrate that a workable fibre-optic probe can be constructed using much simpler light source and detector systems than have been employed hitherto. Apparatus In keeping with our objective of simplicity and low-cost, we selected step-index polymer cable consisting of a polymethyl methacrylate core with a fluorinated polymer cladding. The core - cladding diameter (1.00 mm), numerical aperture (0.47) and attenuation (150 dB km at 600 nm) render this cable unsuitable for transmission of data over long distances (on account of both the attenuation and the modal dispersion), but for short lengths of cable (in our case less than 50 cm) the cable is adequate. An obvious source to use was a light emitting diode (LED).LEDs exhibit a virtually linear relationship between radiated power and forward current, and this, coupled with their rapid rise time (typically 100 ns), simplifies the design of the power-supply (an appropriate voltage source and limiting resistor will do) and obviates the need for mechanically chopping the light should a pulsed system be desirable. In this work a Honeywell “Sweetspot” diode was used with a peak emission wavelength of 665 nm and a band width of 22 nm. The LED was run at a continuous forward current of 50 mA. The band width is typical of surface emitting LEDs; edge-emitting devices may have a narrower band pass. The “Sweetspot” range of optoelectronic devices are designed to couple with a fibre optic system via “DNP” connectors. The relatively narrow band width and ease of coupling to fibre-optic cables are two further advantages of LEDs.On the debit side, the emitted power of an LED is subject to variation due to both ageing effects and temperature, and the restricted range of wavelengths available is a severe limitation too. The detector was a “Sweetspot” PIN photodiode. In common with most devices of this type, the diode is sensitive to light throughout the visible region and the near infrared, but is relatively more sensitive at longer wavelengths. When the diode is reverse- biased, it acts as a high impedance variable current source, the current being a linear function of the intensity of light incident on its active surface. The circuitry for amplifying this signal is straightforward, consisting of a JFET operational amplifier configured as a current to voltage converter with a gain of 106.Probe Design All of the probes that we used consisted of independent source and detector optical fibres, brought into close physical contact at the sensor end. The sensor consisted of an immobilised zone of coloured indicator. Two methods were used for immobilisa- tion: sorption of the indicator on to XAD-2 beads (cross-linked styrene divinylbenzene copolymer)l32; incorporation of the indicator into a porous membrane such as “collodion” (cellu- lose nitrate), usually with 5 pm beads of XAD-2 to improve the light scattering in the membrane. Our target in this work was the fabrication of a device that would respond to pH in the “environmental” range 4.0-8.0. Bromothymol blue was chosen as the indicator, having a h,,,, at 620 nm and a pH transition interval from 6.0 to 7.6.Typical responses to pH for collodion membrane sensors are shown in Fig. 1. Two facts are immediately obvious from these curves. First, the pH transi- tion range is not in the 6.0-7.5 region, but takes place some 1.5 pH units later. This is not unexpected: a shift in the value of the pH transition region and/or a shift in the absorption spectrum due to absorption of indicator on to colloids is well known,3 and has been deliberately used to improve spectrophotometric methods.4 The variation in response for the two probes also comes as no surprise.Conservative calculations indicate that less than 0.5% of the incident light is reflected back along the detector fibre, and this inefficiency, coupled with the rather rudimentary probe design, is bound to lead to poor probe-to- probe reproducibility.The time taken to reach an equilibrium response is a function of membrane thickness (see Table 1).ANALYTICAL PROCEEDINGS, ULY 1985, VOL 22 207 being the probe design both from the point of view of efficiency and reproducibility. The source and detector electronics should be pulsed and tuned, respectively, to enable the discrimination of the chemical signal from the ambient light. (All of the work reported here was carried out with the sensor in a light-tight box.) Reproducibility and drift in the system should be investigated more rigorously, but a reference channel ( i e . , a green LED) should improve the system performance here. Perhaps the most interesting conclusion from this work is that there is every possibility of modifying the chemistry of the sensor to suit the rather restricted wavelength range available from commercial LEDs.Thus, acid - base indicators, the pH transition of which seemed too low for the production of an “environmental” pH probe, may yet be of use because of the 1.5 pH unit shift arising from immobilisation in a membrane. A concomitant shift in the Lax, would also make the choice more appropriate. We also noted a more extended pH response when the indicator was incorporated in a membrane in which amine groups were known to be present in significant numbers. Our interpretation of this phenomenon is that some form of buffering occurs within the membrane.If this is so, then other exciting possibilities are opened up for tuning the chemistry of the system. 5.0 7.0 9.0 5.0 7.0 9.0 PH Fig. 1. Response of two collodion probes to pH Table 1. Variation of response time with membrane thickness Membrane Response thicknesdmm timeimin 3 30 2 15 1 7.5 One of the probes was tested for reproducibility by taking readings in buffers of pH 4.00, 7.00 and 10.00, respectively. The results are shown in Table 2. Conclusions The simple low cost system described here has demonstrated that low cost fibre-optic chemical sensors can be constructed. Clearly a number of features need to be improved, not the least Table 2. Variation of response with pH Response to pH/mV pH of buffer Maximum solution 1 2 3 Mean deviation 4.00 138.0 140.0 139.0 139.2 1.2 7.00 137.8 137.7 137.5 137.6 0.2 10.00 66.0 67.6 67.8 67.1 1.1 References 1.2. 3. 4. Kirkbright, G. F., Narayanaswamy, R., and Welti, N. A., Analyst, 1984, 109, 15. Kirkbright, G. F., Narayanaswamy, R., and Welti, N. A., Analyst, 1984, 109, 1025. Klotz, I. M., Chem. Rev., 1947, 41, 373. Dagnall, R. M., West, T. S., and Young, P., Analyst, 1967,92, 27. Monomode Fibre Optic Temperature Sensors J. D. C. Jones and D. A. Jackson Physics Laboratory, The University, Canterbury, Kent CT2 7NR In recent years, advances in optical technology, notably in fibre optics, have led to renewed consideration of optical sensors for a wide range of applications.1 Potential advantages include intrinsic safety, freedom from electromagnetic interference and compatibility with optical communication systems; the sensing element is chemically inert, and the sensor system is dielectric. Amongst optical fibre sensors, monomode devices have been shown to possess the greatest measurement resolu- tion, and the subject of this paper is the application of these devices to the measurement of temperature.The basic transduction mechanism is the temperature induced modula- tion in the phase2 and polarisation state3 of a fibre guided coherent optical beam. In this paper, phase modulation sensors are termed interferometric and polarisation state modulation sensors are termed polarimetric; it should, however, be noted that in each instance the operation of the sensor is dependent on the interference of light. It will be shown that interfero- metric sensors (discussed in the next section) possess extreme measurement resolution, with minimum detectable temperat- ure changes in the range 1 pK-1 mK for sensing elements of realistic dimensions. Polarimetric devices (see below) offer greater dynamic range with simpler optical configurations, but Sensing element Directional coupler Signal arm I 1 Laser diode source I Input arm ‘ I Photodiode Referencearm I 4 I t , + out pu’t arm Fig.1. Interferometric fibre optic sensors (Michelson configuration)ANALYTICAL PROCEEDINGS, JULY 1985, VOL 22 with reduced resolution. We have recently developed optical configurations which recover both phase and polarisation state information, and hence effectively combine the advantages of the interferometric and polarimetric devices; these are also described below.Interferometric Sensors Many classical interferometer configurations can be adapted to form interferometric fibre optic sensors. Fig. 1 shows a fibre-optic Michelson interferometer for temperature sensing, which has been developed in these laboratories.4 A coherent optical beam from a diode laser is launched into the input fibre arm. The guided beam is amplitude divided by a directional coupler into the signal and reference arms. The guided beams are reflected from the silvered distal end faces of the fibre arms, and recombine at the directional coupler. The irradiance of the recombined beam is monitored by a photodiode situated at the end of the interferometer output arm. In our experi- ments, the interferometer was fabricated from low bire- fringence (circular core) fibre; polarisation effects are hence insignificant.Denoting optical phase delays for the signal and reference arms by and @*, respectively, we may write E’ 0~ (ei@l + eiq2)E . . . . . . (1) where E and E‘ are the electric field vectors for input and output beams, respectively. The output irradiance, 1‘, is hence given by I’ = E’.E’* 1 + cos(+I-@*) . . . . (2) The signal and reference arms were placed in good thermal and mechanical contact throughout their mutual length. Hence, relative phase changes are restricted to the small additional length, I, of the signal arm. Therefore: 4nln h * * (3) 4) = 0 2 =- . . . . where n is the effective refractive index of the fibre and h is the vacuum wavelength of the source used.The temperature sensitivity of the system arises from thermal expansion, and from the temperature dependence of the refractive index: 4n n ar !?++?) l a T A l a T a T . . . . (4) where T is the absolute temperature. Typically, lll(a1laT) -5 x 10-7 K-1 and anBT - 10-5 K-1; hence the tempera- ture dependence of the refractive index dominates, and lll(a+laT) -200 rad K-* rn-1. A number of signal processing schemes for the fibre interferometer have been developed (see, for example, references 1 and 5-7), which yield phase resolutions in the 1 prad to 1 mrad range. Hence, for a sensing element of length 1 = 10 mm, temperature resolutions in the range 0.5 pK-O.5 mK can be realised. It may be noted from equation (2) that the interferometer irradiance is periodic, and for the above system parameters the period is - 3 K.This is effectively the unambiguous dynamic range of the system, unless recourse is made to long range phase tracking tech- niques4; however, even these can fail when the system is initialised. Polarimetric Sensors The basis of operation of the polarimetric sensor is the measureand induced change in state of polarisation of the guided beam.3?8 This effect is only significant in specially manufactured fibres, which exhibit a high degree of linear birefringence. This last property can be attained using a number of techniques, but the present experiments were conducted using York Technology “bow-tie” fibre.9 The cladding of this fibre contains elements which possess different coefficients of thermal expansion. Hence, as the fibre cools following manufacture, considerable anisotropic stress is set up across the fibre core, with resulting stress induced bire- fringence. It is also clear that the degree of birefringence is a function of temperature. The waveguiding properties of a birefringent fibre can be characterised in terms of its polarisa- tion eigenmodes: there are the two orthogonal linear states which propagate without change.The refractive index is effectively different for the two eigenmodes, and their phase velocities are hence unequal. It is therefore logical to define a “fast” and a “slow” eigenmode with associated modal phase delays of @f and &, respectively. The fibre can be conveniently represented in terms of a Jones matrix ( 0 ei+f ei@s O ) where the eigenaxes (the azimuths of the eigenmodes) coincide with the axes of the coordinate system.An optical arrangement which determines this polarisation state information is illustrated in Fig. 2.10 The sensing element consists of a short length of birefringent fibre, which is interrogated via a similar fibre of arbitrary length. The two fibre sections are fusion spliced such that their eigenaxes are at 45” to one another. A linearly polarised coherent source is orientated to populate only a single eigenmode of the input fibre. The second mode is used to carry the output signal to the detector. It may thus be shown that the system operation is insensitive to temperature changes in the input lead. The electric field vector of the light measured at the output is given by E’ F R L ( R - S R+)L FT E‘ .. . . where S and are the Jones matrices-for the s_ensing element and input lead as described above, R+ and R- matrices for rotations of +45” and -45”, PR and PT are Jones matrices for reflection and transmission by the polarising beam splitter, Fig. 2. Polarimetric fibre optic sensor, incorporating temperature insensitive input lead. Polarisation states and eigenmodes shown thus: -. Sensing element and input Lead are high birefringence monomode fibresANALYTICAL PROCEEDINGS, JULY 1985, VOL 22 BS 209 M Sensing element BS I -+ Fig. 3. Combined Michelson interferometric and polarimetric fibre optic sensor: HeNe, single longitudinal mode helium neon laser; BS, (non- polarising) beam splitter; Al, A2, linear polarisation analysers; D1, DZ, photodiodes; M, mirror.Polarisation states and eigenmodes shown thus: -. Sensing element is a high birefringence monomode fibre and E the electric field vector for the input light where E a (1 ,O). We may thus show that the output irradiance is given by I’ a 1 - COS(@-@,) . . . . . . (6) This expression is analogous to equation (2) for the Michelson interferometer, and shows that the sensor output is dependent on the modal phase delay between the eigenmodes for the sensing element only. To determine the thermal response of the sensor, equation (4) is applied for each of the eignmodes separately. Therefore, 1 a (A@) 4n An a1 1 i3T h li3T i3T - -- - [ - + s c a n ) ] . . . . (7) where A@ = @f - GS and An = nf - n,. From the physical properties of the fibre, it may be shown that the sensitivity 111 a(A@)/aT - 10 rad K-1 6-1.Therefore, as is shown in Table 1, the temperature resolution of the polarimetric sensor is much less than for the interferometric sensor, but the unambiguous dynamic range is concomitantly increased. Table 1. Performance summary for monomode fibre optic temperature sensors. Data refer to a 10 mm sensing element of York Technology “bow-tie” structure monomode fibre, used in reflection with a source wavelength in the range 633-850 nm. Temperature and energy resolutions have been calculated assuming 1 mrad phase resolution in the signal recovery technique employed. Energy resolution is shown both for the complete sensing element and per unit area of sensing element surface. The dynamic range is defined as the temperature change producing a phase change of 2n radians in the sensor output Interferometric Polarimetric Sensitivit y/rad K - 1 m - 1 200 10 Temperature resolution/mK 0.5 10 Dynamic range/K 3 60 Energy resolution/@ 0.08 1.6 @ cm-2 1.3 26 Combined Sensors We have recently developed optical configurations which simultaneously recover phase and polarisation information.One of these is illustrated in Fig. 3, and can be considered as a combined Michelson interferometer and polarimetric sensor. 11 A linearly polarised coherent source excites both eigenmodes of the birefringent sensing fibre. Analyser Al is orientated at 45” to the fibre eigenmodes (but orthogonal to the azimuth of the input light); the output at D1 is hence of the polarimetric form.Analyser A2 is orientated parallel to an eigenmode: light from this eigenmode hence mixes coherently with that from the mirror to produce a Michelson-like output at D2. The electric field vector of the light at D1 can be written E’1 =A& Sd+ E’ . . . . . . (8) where A. is the Jones matrix for a horizontal linear analyser and E cc (1,O) as before. The detected irradiance is hence given by 1’1 a 1 - COS(@f - @,) . . . . (9) as expected. For the output at D2 E~~ a A,, (R- S R + + M ) E . . . . (10) where M = ieie and represents the reference beam reflected by the mirror, I is the identity matrix and 6 is the phase difference arising from air paths. A,, is the Jones matrix for a linear analyser at 45” to the horizontal. The detected irradiance is thus 1‘2 1 + COS(@f - 0) .. . . (11) which is of the Michelson form, as in equation (2). An alternative configuration of reduced complexity is shown in Fig. 4. This shows a fibre Fabry-Perot interferometer incorporating a birefringent fibre sensing element.12 The “mirrors” of the Fabry-Perot cavity are formed by the normally Sensing element BS I Fig. 4. Dual fibre Fabry-Perot sensor: HeNe, single longitudinal mode helium neon laser; PBS, polarising beam splitter; BS, (non-polarising) beam splitter; D,, DZ, photodiodes. Polarisation states and eigenmodes are shown thus: -. Sensing element is a high birefringence monomode fibre210 ANALYTICAL PROCEEDINGS, JULY 1985, VOL 22 cleaved end faces of the fibre, and interference between light reflected from the input and distal fibre end faces occurs.The finesse of the device is low, unlike the conventional Fabry- Perot, and higher order reflections can be neglected; the transfer function is hence Michelson-like. This device can be understood as consisting of two Fabry-Perot cavities, each corresponding to one of the fibre eigenmodes. A polarising beam splitter aligned with the fibre eigenaxes is then used to determine the phase delay for each of the eigenmodes separately. The electric field vectors for the light at D1 and D2 are given, respectively, by: E‘, = P R E ’ , E’2 = FTE’ ; E’ = (c$ + f ) E” . . . . (12) E’ is the electric field vector of the input light, whose state of polarisation is such as to excite both fibre eigenmodes, ex., E” 0~ (1,l). We hence find the detected irradiances 1‘1 1 + cos @f , 1’2 0~ 1 + cos QS .. (13) Thus, each output gives “interferometric” information, where- as the phase difference between the outputs yields the “polarimetric” information. In the present experiments, the differential phase was determined by using a simple pseudo- heterodyne signal processing technique. Conclusions A performance summary for the sensors described is shown in the table. It is seen that interferometric fibre-optic sensors offer extreme resolution in temperature measurement, and that by exploiting the properties of highly birefringent fibre, measurements can be made within a wide unambiguous dynamic range. APPENDIX The matrices used in the preceding discussion are given below: where $q(Of) and GS(8,) are the phase delays for the fast and slow eigenmodes, respectively, for the fibre sensing element (or input lead).R’ 1. 2. 3. 4. 5. 6. 7. 8. 9. 10. 11. 12. 1 A 4 5 = - (I ’> f i l l References Giallorenzi, T. G., Bucaro, J. A., Dandridge, A., Sigel, G. H., Cole, J. H., Rashleigh, S. C., and Priest, R. G., ZEEE J. Quantum Electron, 1982, QE18,626. Hocker, G. B., Appl. Opt., 1979, 18, 1445. Eickhoff, W., Opt. Lett., 1981, 6 , 204. Corke, M., Kersey, A. D., Jackson, D. A., and Jones, J. D. C., Electron. Lett., 1983, 19, 471. Cole, J. H., Danver, B. A., and Bucaro, J. A., IEEE J. Quantum Electron, 1982, QE18, 694. Dandridge, A., and Tventen, A. B., Opt. Lett., 1982, 7 , 279. Jackson, D. A., Kersey, A. D., Corke, M., and Jones, J. D. C., Electron Lett., 1982, 18, 1081. Rashleigh, S. C., “IEE 1st Conference on Optical Fibre Sensors (London),” Conference Publication 221,1983, p.210. Birch, R. D., Payne, D. N., Varnham, M. P., and Tarbox, E. S., “IEE 1st Conference on Optical Fibre Sensors (Lon- don),” Conference Publication 221, 1983, p. 83. Corke, M., Kersey, A. D., Liu, K., and Jackson, D. A., Electron. Lett., 1984, 20, 67. Corke, M., Jones, J. D. C., Kersey, A. D., and Jackson, D. A., Electron. Lett., 1985, 21, 148. Leilabady, P. A., Jones, J. D. C., Kersey, A. D., and Jackson, D. A., J. Phys. E: Sci. Instrum., 1985, submitted for publication. Application of Single Optical Fibres to Remote Absorption Measurements L. A. Hilliard BP Research Centre, Sunbury-on-Thames, Middlesex Recently there has been much interest in remote spectroscopic sensing using optical fibres.1-4 The advantages for monitoring chemical samples in remote or hazardous environments are obvious. Remote chemical analysis can be achieved by linking a spectrophotometer to the remote environment by optical fibres. By using multiplexing, one instrument may monitor several different samples in different places in quick suc- cession. Alternatively a “dedicated” instrument can be con- structed for monitoring one chemical in a binary mixture. This instrument may use single multi-mode fibres and simple analogue electronics with cheap detectors and sources, giving tolerable fibre lengths of a few hundred metres. The accuracy in measuring the concentration of a given chemical will depend on the strength of the absorption and on the components of the liquid mixture, but will typically be of the order of k 1 % or less.In this paper it is reported that spectroscopic measurements are possible using single optical fibres (100-140 pm) and a white light source. A brief description of a prototype instru- ment is also given. Applications of this technique are envisaged in on-line analysis where, in addition to the well known advantages of optical fibres, the method offers potential improvements in reliability together with reduced complexity over existing techniques. Near Infrared Absorption The use of optical fibres for such measurements limits the choice of wavelengths to between 0.6 and 2 pm. Absorption bands in this part of the near infrared spectrum correspond to overtones and combination bands of the fundamental mole- cular vibrations occurring at longer wavelengths in the infrared.Only those overtones arising from stretching vibrations of polar groups are strong enough to cause significant absorption in this region, however. Although the degree of absorption is an order of magnitude lower than that for the fundamental vibrations, the availability of higher energy radiation sources and more sensitive detectors largely compensates for this. If there are no interactions between solvent and solute then the absorption spectrum of the resulting mixture is a super- position of the two individual spectra and varies with the concentration of solute, as shown below. Transmission is given by T = Z/&, where I is the transmitted intensity and Z, the incident intensity, Now, absorbance is given by: A = ax = -loglo T (the Lambert - Beer Law) where a = absorptivity and x = path length.ANALYTICAL PROCEEDINGS, JULY 1985, VOL 22 21 1 For a binary mixture the absorbance is given by: where al, cl, a2 and c2 are the absorptivity and concentration of solute and solvent, respectively, and A = x [cI (a1 - a2) + a21 because c2 = 1 - cl.Hence, for a given path length, the absorbance should be a linear function of concentration cl. In practice, it is found to be necessary to compensate for variations in the light source, variable connector losses, dirt on windows, etc., by introducing a reference wavelength that is not absorbed by either the solute or the solvent. The absorbance is then defined as: where TI = transmittance at absorbing wavelength and T2 = transmittance at reference wavelength.Multi-com- ponent liquids can be analysed in the same way by a careful choice of more than one reference wavelength. Results Measurements were made by using a monochromator driven by computer control. The source was a tungsten lamp illuminating the entrance slit of the monochromator. Light was carried from the monochromator to the sample cell via lo& 140 pm fibre, a collimated beam being produced through the cell by means of two crown glass spherical lenses. Signal detection was achieved using a 1 mm diameter germanium detector incorporating an amplifier with offset to enable adjustment for dark current to be made. All silica (AS) optical fibres were employed as plastic clad silica (PCS) fibres have a very strong absorption at 1400 nm due to the presence of OH groups; 100-140 pm fibres gave perfectly adequate signal intensity.Initial measurements were made of alcohols in gasoline because of the interest in this mixture as a fuel. It was found that there is strong absorption due to alcohol at 1550 nm, a wavelength most convenient for fibre optic transmission, this absorption being caused by a combination band of OH and CH vibrations. Measurements were made for the concentration ranges &lo% and 0-100%. The experimental results show absorbance to be linear with concentration to within 1% of span. It was found that neither dissolved water nor differing gasoline compositions had any appreciable effect on the measurement. Temperature has a small effect, resulting in a calibration error of 1% of span per 10 "C temperature rise.Similar results were obtained for mixtures of ethanol in water despite the fact that both liquids are polar so that there will be complex changes in the interactions as the ethanol concentra- tion changes. Another application which has been proved to be feasible is that of interface detection between dissimilar liquids. Clearly, such measurements as are described above may be useful to the chemical industry for on-stream composition measurements. For example, the measurement of trace water in ethylene dichloride or concentrated sulphuric acid at 1.4 pm.1 To the author's knowledge there exists no com- prehensive set of near infrared spectra, so that the best way to find if a chemical species of interest is measurable in this way is to go out and look for it! When selecting wavelengths for measurement there may be a choice between the first overtone and a second (weaker) overtone at a shorter wavelength. * The factors governing this choice will be the requirements for measurement range and path length, the availability of light sources (cheap LEDs are currently only available for operation below 1 pm) and of a suitable reference wavelength close to the measurement wavelength.Other factors to consider include the sensitivity and signal to noise ratio of detectors and the transmission of the optical fibres at the wavelengths of interest. Instrument Design The choice of light source (as stated above) is obviously dependent on the wavelengths employed. LEDs are to be preferred because they lend themselves to multiplexing. Some simple analogue electronics have been developed for process- ing such signals.Light from two LEDs is alternately pulsed into a single fibre by using an optical fibre Y coupler. After being amplified, filtered and clamped, the multiplexed signal from the sensor is then taken to a sample and hold circuit in order to derive separate signals for each wavelength; the absorbance is then calculated using a log ratio amplifier. In practice, it was found necessary to include a separate reference channel by tapping light from the input signal using a 1 : 10 fibre optic X coupler. This compensates for ageing of the LEDs and drift in the electronics. For measurements at wavelengths greater than 1 pm white light sources must be used as LEDs are at present prohibitively expensive.Interference filters can then be used to select the measurement wavelengths. As these filters may introduce as much as 5 dB of loss, the signal to noise ratio will be lower than for the other system; however, experiments indicate that this can be tolerated without any loss in accuracy using hundreds of metres of fibre. Conclusion Near infrared spectroscopy is a useful technique for chemical analysis and lends itself well to remote measurement using fibre optics and relatively unsophisticated electronics. Its successful implementation for process stream analysis would provide further impetus for the development of other fibre optic sensors for process control. References 1. Jones, C., Instrum.Technol., 1982 (August), 51. 2. Maugh 11, T. H., Science, 1982, 218, 875. 3. Newby, K., Reichert, W. M., Andrade, J. D., and Benner, R. E., Appl. Opt., 1984, 23, 1812. 4. Borman, S. A., Anal. Chem., 1981,53, 1616A. L, = - n + l * LO where h, = wavelength of fundamental; L, = wavelength of nth overtone (n = 0,1,2 . . .).212 ANALYTICAL PROCEEDINGS, JULY 1985, VOL 22 Biomedical Sensing Using Optical Fibres A. M. Smith Unilever Research, Sharnbrook, Bedford MK44 7 LQ Fibre optics and medicine have been linked since the early days of fibre production. The use of flexible light guides for endoscopy was one of the first commercial applications of the technology, Fibres were also exploited for sensing in an oximeter utilising haemoreflectometry 25 years ago. 1 There now exists a range of different fibre sensing concepts which have been used for in-vivo and in-vitro measurements of a number of different medical parameters. The main emphasis of this review paper is the use of fibres in chemical or biochemical sensors.In-vivo Sensors Potential advantages of using fibres for in-vivo sensing (usually arterial or vascular catheterisation procedures) include small physical size, flexibility, absence of risk of electric shock hazard and relatively good biocompatibility . The simplest use of a fibre sensor is for remote spectroscopic measurements. The spectral reflectivity of haemoglobin in red blood cells varies markedly with oxygenation at wavelengths around 660 nm, whilst at other wavelengths (e.g., 805 nm) very little variation occurs.By taking the ratio of the reflected light at two wavelengths the oxygenation state (oxygen saturation) can be determined. Commercial oximeters using colour chopper-wheels or LEDs are available. Typically, separate fibres are used to transmit and receive the light and these are mounted within a lumen of a catheter to allow measurements to be taken within the arteriovascular system.2 The pH can be a very useful measure of patient condition during and after major operations. A device which is designed for in-vivo use has been developed at the National Institute of Health, Bethesda, Maryland.3 The sensor consists of trans- mitting and receiving fibres, which allow monitoring of the change in colour of an indicator dye (phenol red) by using a two wavelength ratio method. The dye is immobilised on 5-10 pm polyacrylamide beads within a thin dialysis tube (which allows a free flow of hydrogen ions).The back-scattering of light into the receiving fibre is enhanced further by the addition of 1-pm diameter polystyrene microspheres. The basic pH sensor concept can be used in modified form to measure other blood parameters. In many clinical situations the partial pressure of oxygen (P02)is a useful measure of patient condition in addition to oxygen saturation. The oxygen quenching behaviour of fluorescent dyes can be used for PO2 measurements.4 Here, the outer membrane of the transducer cell is a hydrophobic polymer (Celgard) and the dye is immobilised on an organic support to avoid humidity sensi- tivity problems. Current sensors are still experimental and suffer from fibrin build-up when used in the bloodstream. The pH sensor can also be adapted5 for the measurement of carbon dioxide (pC02) by suspending the particles coated with indicator dye in a solution of potassium hydrogen carbonate and potassium chloride (buffer).Initial animal experiments have been encouraging. Glucose concentration has been measured6 by using a dialysis tubing based cell interrogated by fibres. A lectin that binds carbohydrate is immobilised on the inner wall of the tube, which is filled with a solution of fluorescently labelled dextran. The dextran binds to the lectin and relatively low fluorescence is observed. If the sensor is placed in blood, glucose traverses the membrane and displaces some of the dextran (which is too large to go through the wall) into the cell, hence increasing the measured fluorescence. In principle this technique could be extended to other metabolites by substitut- ing an alternative specific binding agent in place of the lectin. Fibres have also found medical application in the measure- ment of physical parameters, such as blood flow (laser Doppler anemometry and dye dilution measurements), intracranial pressure (diaphragm sensor) and temperature.Relatively few commercial sensor systems are available at present but a number of companies are known to be developing units. The possibility of carrying out sophisticated signal processing and the simultaneous measurement of more than one parameter will greatly extend the usefulness of the sensors.Immunological Sensors Recently work has begun on the development of sensors for large relative molecular mass immunochemicals which make use of optical waveguides. At present these devices have, for the most part, been demonstrated on planar substrates,’ but could in principle use cylindrical fibres. The key feature of the sensors is that a layer of specific biochemical (e.g., an antibody specific to a single antigen) is immobilised on the surface of a waveguide. The evanescent field of the radiation travelling within the waveguide extends through the specific layer. As a result of the evanescent field interaction, any immunochemical binding will affect the amount of light transmitted in or lost from the guide. These changes are monitored to quantify small concentrations (nM range) of specific antigens.The interaction can be enhanced by using the surface plasmon resonance in thin silver films,* and in some instances increased sensitivity is obtained by allowing competition between unlabelled antigens in an unknown concentration sample and known concentra- tions of labelled antigen. Current results indicate that immu- nological sensors using optical methods may well be easier to use than conventional diagnostic tests and yield results in shorter time periods with fewer incubations. Conclusions The use of optical fibres and waveguiding principles for measuring parameters of biomedical interest is still in its infancy. Current results are, however, encouraging and it is to be expected that the area will grow.References 1. Polanyi, M. L., and Hehir, R. M., Rev. Sci. Instrum., 1960,31, 401. 2 . Krovetz, J. L., Brenner, J. I., Polanyi, M., and Ostrowski, D., Brit. Heart J . , 1978,40, 1010. 3. Peterson, J. I., Goldstein, S. R., and Fitzgerald, R. V., Anal. Chem., 1980,52,864. 4. Peterson, J. I., and Fitzgerald, R. V., Anal. Chem., 1984, 56, 62. 5. Vurek, G. G., and Feustel, P. J., “Proceedings of the 35th Annual Conference in Engineering in Medicine and Bio- chemistry, Philadelphia, PA, USA, 1982.” 6. Schultz, J. S., Mansouri, S., and Goldstein, I. J., Diabetes Care (U.S.A.), 1982, 5 , 245. 7. Sutherland, R. M., Dahne, C., Place, J . F., and Ringrose, A. S., Clin. Chem., 1984, 30, 1533. 8. Liedberg, B., Nylander, C., and Lundstrom, I., Sensors Actuators, 1983, 4, 299.ANALYTICAL PROCEEDINGS, JULY 1985, VOL 22 Raman Thermometry Using Optical Fibres G.W. Bibby and J. N. Ross Central Electricity Generating Board, CERL, Leatherhead, Surrey J. P. Dakin and D. J. Pratt Plessey Electronic Systems Research Ltd., Roke Manor, Romsey, Hampshire 20 > - I n - c 2 1 5 - c - a - tl .- .- .- t l - m 10- C T - 5 - A novel temperature sensor is described which enables the temperature distribution along an optical fibre to be measured, using the Raman scattering in the fibre. A short pulse of light is launched into the fibre and the backscattered light is collected. The intensities of the Stokes and antistokes bands are measured as a function of time after the launch of the pulse. From the ratio of the antistokes to Stokes intensities the temperature is deduced as a function of time delay and hence distance along the fibre.1.2 - - Theory The ratio of the intensities of the antistokes to Stokes lines for wavenumber shifts +Y is given by3 -1.4 R(7') = (h,/ha)4 exp(-hcv/kT) .. . . (1) - 0 v, ; c - where h, and ha are the Stokes and antistokes wavelengths, h is Planck's constant, c the speed of light, k Boltzmann's constant and T the absolute temperature. This ratio is independent of light intensity and fibre geometry or materials and thus, in principle, permits the absolute temperature to be measured without any detailed knowledge of the optical fibre itself. In practice, if the ratio of the Stokes and antistokes lines is measured using a detector with either filters or a spectrometer to separate the Raman bands, then the observed ratio R' (7') will differ from the true ratio R(7') by a factor of F, which will arise from differences in the sensitivity of the detector or transmission of the filters at the two wavelengths. This factor can, of course, be measured, but this is unnecessary if a section of fibre at a known temperature 8 is used as a reference.With a little manipulation (1) can be expressed in the form 1/T = 1/8 - (kfhcv) [In R'(7') - In R'(8)l . . (2) The absolute temperature can then be calculated from R' (0, R'(8), 8, Y and a few fundamental constants. Experimental Procedure In order to demonstrate that this method can be used to measure the distribution of temperature along a fibre a system has been set up using an argon ion laser as a source.The laser produces pulses of around 15 ns duration at a repetition rate of 40 kHz, with a peak power of about 5 W. The wavelength can be selected from a number of lines, but is usually 514 or 488 nm. The light passes through a beam splitter before being launched into the fibre. The backscattered light is directed by the beam splitter through a monochromator on to a photomul- tiplier detector. The monochromator has a band width of about 0.5 nm (20 cm-1) and is tuned to collect Stokes or antistokes light at a wavenumber shift of between 200 and 600 cm-1. The peak Stokes scattering occurs at around 400 cm-1 for the doped silica glasses used in most optical fibres. The signals from the photomultiplier were averaged using a boxcar integrator, and recorded and processed with a micro- computer.The temporal resolution was about 20 ns, giving a resolution length of 2 m in the fibre. Results A standard optical fibre with a 50 vm core and a numerical aperture of 0.2 was used as the sensor. A length of 5 m was immersed in an ice-bath to provide a reference temperature 213 and one, or sometimes two, further sections of the fibre were immersed in water-baths at different temperatures. Typical examples of Stokes and antistokes intensity records are shown in Fig. 1. The time taken to scan the boxcar through the range of delay 0-1 ps was 100 s and the integration time constant 0.2 s. The hot and cold regions of the fibre can readily be seen on the antistokes record. The corresponding plot of ln[R'( T ) ] is shown in Fig. 2. The scatter of points arises from the shot noise of the photomultiplier and the corresponding uncertainty in the temperature of each point is about +5 "C.By averaging over several points this uncertainty can easily be reduced to +2 "C. The temperatures corresponding to the room temperature part of the fibre and the two hot regions, calculated using equation (2) with Y = 300 cm-1, are 22, 71 and 52 "C compared with temperatures measured with mercury in glass thermometers of 20,72 and 50 "C. The temperatures calculated are not always in such close agreement. Changing the excitation wavelength from 514.5 to 488 nm, and with the same wavenumber shift, the calculated temperatures were about 10% low. Also, with a larger Raman shift the calculated temperature tends to decrease.The reasons for these calibration errors has not been established. 25 I Fig. 1. The intensity of the antistokes and Stokes light at wavenumber shifts of f300 cm-1 -1.0 80 ,:,+% Fig. 2. Stokes intensity The logarithm of the ratio of the antistokes intensity to the214 ANALYTICAL PROCEEDINGS, JULY 1985, VOL 22 Conclusions References The technique of distributed Raman thermometry has been demonstrated and the results are encouraging. There is still some uncertainty about the calibration which needs to be clarified, but a sensitivity of a few degrees has been demon- strated. 1. D a k h J . p.9 UK Pat. GB2140554Ay 1984- 2- Dakiny J - p*, Pratty D* J - , Bibby, G* w - y and J * N*, “Conference on ‘Optical Fibre Sensors’, San Diego, CA, USA, February 13-14, 1985.” 3, Long, D, A., c‘~aman ~pectroscopy,~7 McGraw-H[ifl, London, 1977.Spectral Filtering Optical Fibre Sensors J. P. Dakin Plesse y Electronic Systems Research Limited, Roke Manor, Romse y, Hampshire Optical fibre sensors can conveniently be divided into two main types. The first type, the intrinsic sensor, is constructed using the fibre as the sensing element, and arranging for the physical parameter to be monitored to modify the transmission proper- ties. The second type, the extrinsic sensor, uses more conven- tional optical techniques to modify coupling between an input and output fibre, and therefore merely uses the fibre as a means of guiding optical energy to and from a remote, passive sensing head. To date, all commercial sensors have used the extrinsic sensing approach because of the ease of tailoring the design of the transduction system to best suit the measurement. By far the simplest optical parameter to detect is the intensity of the transmitted light, as this requires only a simple photodiode and amplifier. However, solid-state sources show significant variations in output with temperature, and environ- mental effects, in particular mechanical disturbances, will cause significant transmission changes in optical fibre leads and interconnections. Thus, although convenient, intensity modu- lation is in many respects one of the least desirable methods of transduction.The spectral filtering technique uses a broad band light source, and the transducer modifies the spectrum of the transmitted light in a manner amenable to measurement by a spectrometer arrangement.In general, the spectral filtering should exceed that normally occurring in short optical links so that the output is essentidlly unaffected by environmentai effects on cables and connectors. In view of the attractiveness of the “frequency” output of spectral filtering sensors, com- bined with reasonably straightforward incoherent detection, this method of sensing has been the basis of many sensors developed in the author’s laboratory. The remainder of this paper will review the types of spectral filtering sensors which have been devised, both within our laboratories and elsewhere. Many of these show practical potential for low cost chemical process industry requirements. Analogue Sensors Based on Spectral Filtering Techniques All of the following sensors enable analogue monitoring of a physical parameter from a spectral measurement of the transmitted optical signal.Graded inteqerence filter sensor Perhaps the simplest form of analogue spectral filtering sensor for the measurement of displacement is shown in Fig. 1. The filter is a graded interference filter, which at any one point on its surface has a narrow-band filter characteristic. The wavelength of peak transmission, however, changes monoton- ically along its length due to a gradual variation in the interlayer thickness in this direction. Thus, the peak trans- mitted wavelength at the output end of the system is a function of the position of the filter, which may of course be mechanically coupled to an actuator arm or similar mechanical arrangement.Graded interference filter * White light source X X Output to wavelength analyser Fig. 1. Displacement sensor using graded interference filter Rotating interference filter sensor A modification of the previous idea using rotation of a conventional (i. e., uniform layer thickness) interference filter to sense, for example, shaft rotation is shown in Fig. 2. The peak transmitted wavelength changes as rotation occurs, but the output wavelength is, in general, a highly non-linear function of angle. * White light source + - Outputto O wavelength Angle of filter an a I yser _ - (a) (6) Fig. 2. Sensor using rotating interference filter. (a), Sensor; (b), sensor response Fabry - Perot filter sensor The Fabry - Perot resonator provides an inherently sensitive method of measuring small changes in the gap between two highly reflective (but slightly transmissive) mirrors (Fig.3). In general, for mirror spacings of several pm, there are usually many fringe orders possible in the resonator (see Fig. 3 lower curves) and ambiguity of position is possible. By measuring the output accurately over a wide range of wavelengths the ambiguity can be resolved for closely spaced Fabry - Perot cavities. With spacings of the order of less than one wavelength, the ambiguity ceases (but at the expense of some degree of sensitivity) as only the lowest order fringe is present. Work on this sensor was first published by Glen, Morey and Snitzer. 1ANALYTICAL PROCEEDINGS, JULY 1985, VOL 22 215 Semiconductor band edge filter * White 1 output to light wavelength source sensor I I I I 7 8 9 10 Spacing (t)lpm Fig.3. Fabry-Perot sensor (reproduced from reference 1) Rotating grating (or zone plate) sensor A convenient analogue method of measuring shaft angle rotation is to attach a dispersive element such as a diffraction grating (Fig. 4).2 As the grating rotates, the peak wavelength coupled from the input to the output fibre varies in a precise and predictable manner, determined by the usual diffraction relationships. The sensor head is merely a miniature mono- chromator. The UK National Physical Laboratory has had a number of samples of a very compact sensor head made for experimental evaluation.3 In addition, M. Hutley of NPL has devised a linear version of the encoder based on moving holographic zone plates which exhibit wavelength dependent focusing properties.Diffraction grafting Fibre on rotating shaft Broad band I ig ht source It J- To wavelength analyser Fig. 4. Spectral filtering sensor using diffraction grating on rotating shaft Semiconductor edge filter sensors All of the sensors described so far have measured displace- ment. The semiconductor edge filter sensor, however, in common with the following four types, allows temperature to be monitored. This sensing method relies on the shift in the position of the edge of the absorption band of a semiconductor material as its temperature is changed. In addition to normal semiconductor materials, there is a range of commercial glass filters with similar optical properties. Monitoring the spectrum of the transmitted light allows the temperature of the sensor head to be deduced from the band edge position (Fig.5). In addition to the monitoring of a single discrete sensor, Theocharous4 has succeeded in monitoring several such sensors, linked in tandem with optical fibre cable. This was achieved using optical time domain reflectometry to measure the loss of a splice containing a thin section of temperature dependent glass filter. The ratio of the loss at two wavelengths was used to avoid errors due to non-wavelength-dependent splice losses. Broad band light source Wavelength analyser Fig. 5. Spectral filtering temperature sensor using semiconductor band edge filter Liquid crystal filter Certain types of cholesteric liquid crystal exhibit a significant change in their scattering characteristics over a typical tem- perature range of 20-30 "C.This change results in a marked change in the colour of the scattered light over this same temperature range. Johnson and Rozzells have used the effect in order to measure temperature. Normally, the liquid crystal will be used in a back-scattering configuration. Rare-earth doped fibre sensor This sensor is the only intrinsic sensor using spectral filtering. Its operation (Fig. 6) relies on the temperature dependence of the absorption spectrum of certain rare earth dopants when incorporated into the core of a glass fibre. Both europium and neodymium have been investigated experimentally, and both dopants may be used for sensor fabrication.6 Temperature measurement up to 800 "C is reported with a prototype system using a neodymium dopant.r--.-. -. -! Hot zone Normal fibre doped-silica doped-si lica Rare-earth Normal fibre I I * Broad band light source doped fibre I To spectrum anal yser Fig. 6. Spectral filtering sensor using rare earth doped fibre Christiansen filter sensor This sensor uses the well-known Christiansen effect, where a mixture of, for example, glass and liquid is chosen to have intersecting refractive index versus wavelength graphs. At the wavelength of intersection, the mixture appears to be optically homogeneous, while away from this wavelength the scattering loss increases. The normal configuration of this filter is with the glass - liquid mixture placed in a collimated white light beam and followed by a focusing lens and spatial filter to reject scattered light.The author's configuration,7 however, uses the mixture as the core of an optical wave guide in order to simplify construction and improve compatibility with an optical fibre system (Fig. 7). However, the normal configuration can still be used with the same optics as in Fig. 5, but replacing the filter by a Christiansen cell. The temperature dependence of the sensor arises because the liquid refractive index changes significantly with tempera- ture, whereas that of normal glasses changes by at least an order of magnitude less. Thus, the wavelength of minimum scattering (or peak transmission) changes with the temperature of the sensor head.216 ANALYTICAL PROCEEDINGS, JULY 1985, VOL 22 Fibre optic compatible Christiansen filter in waveguide configuration * J.To wavelength analyser A Broad band light source Input fibre Output fibre / / / / / / / / A 0 0 0 0 0 0 0 0 V / / / / / / / / / d 0 0 0 0 0 0 0 . B Liquid (hot) Liquid (cold) 'c Glass a r' 100 1 I I I I D Fig. 7. ( a ) , Spectral filtering sensor using a waveguide Christiansen filter: A, sensor arrangement; B, attachment to fibre; C, ray paths in filter, N glass > N liquid; D, ray paths in filter, N glass = N filter. ( b ) , Christiansen filter, temperature response Birefringent crystal sensor This temperature sensor relies on the combined wavelength and temperature dependence of the birefringence shown by most types of anisotropic crystal. The arrangement (Fig. 8), devised by Mezzetti, was reported by Harmer in a recent review.8 The birefringent crystal is contained between the polariser and analyser in a reflective configuration.The retro-reflected signal from the cell varies with both wavelength and temperature. By varying the monitoring wavelength to produce cyclic variations in transmission, the temperature can be deduced from the wavelength of peak transmission. Birefringent crystal Tunable light source 0- Detector \ Potahser Mirrored end Fig. 8. Remote temperature sensor using birefringent crystal Photoelastic sensor with wavelength characterisation This pressure sensor is similar in concept to the preceding sensor. However, it utilises the photoelastic effect, which is a pressure induced anisotropy , causing birefringence in a trans- parent material such as glass or plastic.As with the birefringent crystal sensor, the birefringence can be monitored as a function of wavelength rather than merely observing the intensity of light emerging from the sensor cell (Fig. 9). This method has been reported by Jones and Spooncerg for use as an optical sensor for pressure. * Photoelastic Broad band material light source (plastic or glass) To wavelength analyser Fig. 9. Photoelastic pressure sensor with wavelength characterisa- tion Expanded beam sphere lens connectors Sensed volume / 1.0 light source Receiver 0- 800 900 1000 1 Wavelengthhm Fig. 10. Liquid level sensor using 0-H or C-H absorption 100ANALYTICAL PROCEEDINGS, JULY 1985, VOL 22 217 Liquid level sensor A simple liquid level sensor using spectral filtering techniques has been reported.10This sensor utilised the absorption from a 980 nm 0-H vibrational band to sense the presence of water in an optical path.The relatively broad emission from an LED with a peak spectral response at 940 nm was used to provide source power and a simple receiver dichromator using a dichroic mirror and two silicon photodiodes were used to detect spectral changes due to the water absorption (Fig. 10). The ratio of photocurrent from the two detectors was used to make a threshold decision as to whether water was present in the measurement region. By using a vertical beam, an analogue indication of water level can be made. Conclusions It may be seen from the above review that a large number of types of spectral filtering sensor can be constructed with relatively simple hardware yet with potentially reliable and accurate operation. It is the author’s belief that this is one of the most promising methods of remote optical sensing for applications such as the process industry, in view of the absolute nature of the “frequency” output, combined with easy detection of the output signal. 1. 2. 3. 4. 5 . 6. 7. 8. 9. 10. References Glen, W. H., Morey, W., and Snitzer, E. N., NASA Report CR159468. Dakin, J. P., Br. Put., Appl. No. 7918056; Am. Put., No. 4355898. Priority date May 23rd, 1979. Hutley, M., personal communication. Theocharous, E., “Proceedings of the 1st Conference on Optical Fibre Sensors, April 26-28th, 1983, London,” ZEE Conf. Pub. No. 221, Institution of Electrical Engineers, London, pp. 10-13. Johnson, C., and Rozzell, T. C., Microwave J . , 1975,44. Snitzer, E., Morey, W., and Glen, W. H., “Proceedings of the 1st International Conference on Optical Fibre Sensors, April 26-28th, 1983, London,” IEE conf. Pub. No. 221, Institution of Electrical Engineers, London, pp. 79-82. Dakin, J. P., Br. Put., No. 1558404. Harmer, A. L., “Proceedings of Institute of Measurement Symposium, November 1981, London,” Institute of Measure- ment and Control, London, pp. 1-15. Jones, B. E., and Spooncer, R. C., “Proceedings of the 1st International Conference on Optical Fibre Sensors, April 2&28th, 1983, London,” IEE Con& Pub. No. 221, Inst‘itut‘ion of Electrical Engineers, London, pp. 173-177. Dakin, J. P., and Holliday, M. G., “Proceedings of the 1st International Conference on Optical Fibre Sensors, April 26-28th, 1983, London,” IEE Conf. Pub. No. 221, Institution of Electrical Engineers, London, pp. 91-95. Health and Safety in the Chemical Laboratory - Where do we go from here? This publication provides an overview of health and safety developments in the chemical laboratory and workplace, and will provide essential reading for anyone involved in these areas. Brief Contents: Accident and Dangerous Occurrence Statistics in the United Kingdom; Morbidity and Mortality Studies; Economics of Health and Safety Measures; Procedures and Statistics in France; Professional Negligence, Liability and Indemnity; The System in the United States of America; The System in the United Kingdom; The System in the Federal Republic of Germany; Hazards of Handling Chemicals; Hazards of Apparatus, Equipment and Services; Managing People; What Standards Should We Use? Conflict of Safety Interests with Legislation; The Protection of Workers Exposed to Chemicals: the European Community Approach; Recommendations Arising from the Symposium. Special Publication No. 57 Softcover 206pp 0 85186 945 9 Price f16.50 ($30.00). RSC Members f12.00 Ordering: Non-RSC Members should send their orders to: The Royal Society of Chemistry, Distribution Centre, Blackhorse Road, Letchworth, Herts SG6 1 HN, England. RSC Members should send their orders to: The Royal Society of Chemistry, Membership Officer, 30 Russell Square, London WC1B 5DT. The Royal Society of Chemistry Burlington House, Piccadilly London W1V OBN

ISSN:0144-557X    

DOI:10.1039/AP9852200204

出版商:RSC    

年代:1985

数据来源: RSC

摘要:

218 ANALYTICAL PROCEEDINGS, JULY 1985, VOL 22 Equipment News Mass Spectrometers The Kratos MS80 is a high-performance medium-resolution instrument featuring double focusing forward geometry, which incorporates hexapoles. A comprehen- sive range of inlet systems includes gas chromatography, with facilities for packed and capillary column work. It can be provided with an LC - MS inlet using the Thermospray mode of operation; this inlet technique is especially suitable for the analysis of labile species that have undergone separation by liquid chromat- ography. The MS50 is an ultra-high- resolution instrument with a mass resolu- tion of 150 000 static (10% valley defini- tion) and an option of 80 000 dynamic. As in the MS80 there is a fast scanning magnet especially designed for high reso- lution capillary columns.There is also selected ion monitoring (SIM), which at high resolution will detect down to the low fg level. A range of ionisation tech- niques is available, which includes EI, CI, DCI, FAB, FD and FI. The MS25RF is a GC - MS system in which the mass analyser and data systems are integrated into a single unit. A high performance gas chromatograph is fully engineered into the MS25RF, permitting chromatography under all conditions in both chemical ionisation and electron impact modes. Spectros International plc, Barton Dock Road, Urmston, Manchester M312LD. Atomic-absorption Instruments The System 1000 extends the makers’ atomic-absorption spectrometers. From GBC, of Australia, it consists of a graph- ite furnace flameless atomisation head and power controller, an autosampler and a microprocessor keyboard controller.The graphite furnace uses the makers’ transverse heating, whereby the heat is applied more uniformly than is normally possible, to prevent non-atomic interfer- ences. The controller allows up to ten steps of temperature and temperature rise rates to be programmed, thus allowing complex samples, such as biological mat- ter, to be dried and ashed in stages before final atomisation. The System 1000 is compatible with the GBC 902 double- beam and 903 single-beam spectrometers. EDT Research, 14 Trading Estate Road, London NWlO 7LU. Atomic-absorption Spectrometer The Shimadzu microcomputerised AA- 670 offers complete automation and rep- resentation of all data on an A4 size printer - plotter chart.After calling an element up from a list of 64 stored in R6M memory, the AA-670 will proceed to set the lamp current, slit width, wavelength, type of flame, detector gain and gas flow-rates, all these settings, including peaking the lamp in to the appropriate wavelength and calibration curves, being automatically recorded on the A4 chart. Features include a programmable eight Iamp turret, supplied as standard, a choice of 6 points for the calibration curve, standard addition and a transient memory for accurately capturing the atomised signal when using the graphite furnace. V. A. Howe and Co. Ltd., 12-14 St. Ann’s Crescent, London SW18 2LS. Fourier Transform Infrared Spectrometer The Model 1800 is suitable for both research and routine operation.It offers resolution better than 0.2 cm-1 and the frequency range can extend from the near to the far infrared. The double beam optical system overcomes the problems associated with traditional single beam operation. By taking interleaved scans of, sample and reference beams, excellent cancellation of atmospheric absorption is achieved without delays for purge stabili- sation. The optical design and the use of full double-sided interferograms ensure good photometric accuracy up to high absorbance values. The software pro- vided with the system contains extensive spectral manipulation routines, interac- tive colour graphics and OBEY program- ming. QUANT-3 and SEARCH-3 IR applications software are also available as well as languages such as BASIC and FORTRAN 77.The Model 1800 is designed to accept system expansions. Perkin-Elmer Ltd., Post Office Lane, Beaconsfield, Buckinghamshire HP9 1QA. Cuvette Mixer The Stir and Add mixer is moulded from clear plastic with a straight handle of the same material. Used to add enzymes and other materials to solutions in spectro- photometer cells, the device is moved vertically in the cell for complete mixing. One model is for 10mm light path cells; another is for semi-micro cells. R. B. Radley and Co. Ltd., London Road, Sawbridgeworth, Hertfordshire CM219JH. Spectrofluorimeter The Shimadzu RF540 can monitor chemi- luminescence reactions at constant excita- tion and emission wavelengths with good stability attained through the use of a double optical design for light source monitoring, or reactions can be moni- tored by repetitive overlay scanning.Up to 100 repetitive scans with delays of up to 100min can be programmed. In either mode a parallel head printer draws a frame around the spectrum or trace for presentation, and automatic integration and smoothing gives optimal recording of the spectrum irrespective of scan speeds. GBC System I000 atomic-absorption spectrometerANALYTICAL PROCEEDINGS, JULY 1985, VOL 22 219 Features include the storage of nine methods for ease of use in routine applica- tions and a lamp housing which allows measurements down to 200nm on both monochromators. There are data files for data handling. V. A. Howe and Co. Ltd., 12-14 St. Ann’s Crescent, London SW18 2LS. Spectrofluorimeter Systems The Spex Fluorolog-2 series is made up of nine different spectrofluorimeter systems , each of which is controlled by the new DMlB spectroscopy laboratory coordina- tor.A wide range of accessories for polarisation, phosphorescence and kinet- ics experiments is available. Photon counting is standard for all systems. Glen Creston Instruments Ltd., 16 Dalston Gardens, Stanmore, Middlesex HA7 1DA. Fourier Transform Infrared Spectrometer The Digilab FTS-80 uses the FTS optical bench and the Digilab 3280 data station. Spectra can be manipulated on the colour display using a joystick, with hard copy provided by a four-colour plotter. Avail- able accessories include microtransmit- tance, microreflectance, a photoacoustic cell, diffuse reflectance, ATR, beam con- densers and a dedicated capillary GC - IR option.The use of a standard Unix-type operating system means that a wide range of software is available, including work processing and data base management. Polaron Equipment Ltd., 53-63 Green- hill Crescent, Watford Business Park, Watford, Hertfordshire WD1 8XG. Gas Chromatograph The HachKarle Series 100 instrument is designed for general purpose, routine applications, such as natural, permanent, respiratory and combustion gases and light hydrocarbons from p.p.m. to percentage levels. Manually operated gas sampling and column switching valves are available and can be installed at the factory or by the user. Thermal conductiv- ity and flame ionisation detectors are available. Techmation Ltd., 58 Edgware Way, Edgware, Middlesex HA8 8JP. Chromatography Workstation An agreement provides for the sale of chromatography software and A to D interfaces developed by Nelson Analy- tical Inc.for the HP9000 Series 200 desk-top computer. Up to 10 instruments, each interfaced to up to 2 detectors, can be controlled by the HP3362A or B chromatography system. This system, which consists of the HP9000 Series 200 computer, a Winchester disk drive and the HP ThinkJet printer, together with Nelson’s XtraChrom software and one of four intelligent interfaces, can be used for gas, liquid, capillary and ion chromato- graphy and amino-acid analysis. Hewlett Packard Ltd., Miller House, The Ring, Bracknell, Berkshire RG12 1XN. Au tosampler The AS-8300 programmable multi- sampling system for the Model 8300 Gas Chromatograph consists of a pneumatic- ally operated injection system and an electronically controlled, removable sam- ple tray, which holds up to 100 vials. Up to 10 sampling sequences can be generated for the AS-8300 that are completely inter- active with any of the 10 GC methods which can be stored in the memory of the Model 8300.Sampling and injection parameters are selectable for each group of samples. A sequence can be “paused” at any time to allow the interposition of a single sample without disrupting the origi- nal sequence. Deviations are reported at the end of every sequence. Perkin-Elmer Ltd., Post Office Lane, Beaconsfield, Buckinghamshire HP9 1QA. Automated Headspace Sampler The HP19395A provides an alternative method of preparing solid or viscous liquid samples for syringe injection to any packed or capillary column gas chromato- graphy or GC - MS systems.It also switches quickly to the on-column mode. It operates over the temperature range from 15 to 150°C. Hewlett Packard Ltd., Miller House, The Ring, Bracknell, Berkshire RG12 1XN. Automated On-column Injection for Chromatographs The HP7673A handles an automatic sequence of up to three liquid samples, with injection parameters regulated by the built-in controller or through the HP3392A integrator or the HP5880A gas chromatograph. It may also be configured into a fully automated laboratory system, linking into the makers’ instrument network (INET) integrator. Hewlett Packard Ltd., Miller House, The Ring, Bracknell, Berkshire RG12 1XN. Multiple Splitter This accessory for the makers’ Model ATD 50 automatic thermal desorber splits the sample before and/or after the cold trap of the ATD 50.Advantages include the possibility of analysing wide boiling range samples in one run, easier thermal desorption of high-boiling materials, short analysis times, the ability to use lower desorption temperatures and large split ratios. The multiple splitter can be fitted to existing Model ATD 50 desorbers having a fused silica heated transfer line. Perkin-Elmer Ltd., Post Office Lane, Beaconsfield, Buckinghamshire HP9 1QA. Dual Wavelength Chromatographic Scanner (TLC and Gel Electrophoresis) The Shimadzu CS930 is a microcom- puterised version of the CS910 instru- ment. It eliminates the base-line fluctua- tion caused by local irregularities of the layer thickness of the TLC plate.It incorporates the zig-zag scanning method, which avoids errors caused by the irregular shape of the developed spots. Other features include a working curve linearising function, a background correction function, an automatic wavelength scanning function, a slit- adjustment function and various peak detecting parameters. There is a facsimile parallel line printer - plotter. V. A. Howe and Co. Ltd., 12-14 St. Ann’s Crescent, London SW18 2LS. Equipment for Liquid Chromatography The Spectroflow 783 is an absorbance detector. Under flowing conditions and using the standard flow cell, it provides less than +1% noise at 0.001 AUFS, giving the maximum signal to noise ratio. Drift is held to less than 0.0001 Au h-1. The 783 is programmable and computer compatible.Flow cells are self aligning and easily changed to allow operation in microbore LC, open column LC, conven- tional HPLC, fast LC, semi-preparative LC and low pressure LC. The DS 650 chromatography computer allows real- time data acquisition and system control. Based on Data General’s Model lO/SP desktop microcomputer, it includes a 15 MB Winchester hard disk, one 368 kB diskette drive, 256 kB of main memory, a 12 in green monochrome bit-mapped monitor, graphics printer and a multi- plexed A to D converter capable of accepting data simultaneously from up to 8 detectors utilising four separate time bases. The Spectroflow 400 solvent delivery system for HPLC offers +0.1% flow-rate reproducibility. It has a dual piston design that delivers pulse-free sol- vent metering from 0.010 ml min-1 to 5 ml min-1 without requiring head, piston or electronic modifications.Remote con- trol and automation of flow-rate are possible. A low pressure gradient mixing accessory, the Spectroflow 430 Gradient Former, is available. Spectros International plc, Barton Dock Road, Urmston, Manchester M31 2LD. Pressure Monitor for HPLC This instrument is designed to sense pressure in metering pumps that do not have an integral pressure transducer. It will interrupt the power if either the high or low pre-set limit is exceeded. High and low limits can be set in either of 2 ranges: 0-6 000 lb in-2 for standard HPLC meter- ing pumps and 0-12 000 lb in-2 for high pressure rated pumps.220 ANALYTICAL PROCEEDINGS, JULY 1985, VOL 22 Laboratory Data Control (UK] Ltd., Milton Roy House, High Street, Stone, Staffordshire ST15 8AR.Analysis of Volatiles in Water The Purge and Trap Injector, designed for the analysis of low-concentration vol- atile compounds in water, allows volatile components to be stripped out of the aqueous sample and concentrated in a cold trap. The components are then Perkin-Elmer Ltd., Post Office Lane, Beaconsfield, Buckinghamshire HP9 1QA. Solvent Optimisation Software for Liquid Chromatography PESOS runs on the makers’ 7500 profes- sional computer with the Series 4 liquid chromatography solvent delivery system. It works with any mobile phase system, including normal phase, reversed phase, line on the chromatogram. Chromato- grams can be graphically reprocessed as many times as necessary in order to achieve satisfactory base-line drawing.For example, if several fused peaks exist the CI-1OB can be programmed to draw the base line at all valley points or at selected valley points. Laboratory Data Control (UK) Ltd., Milton Roy House, High Street, Stone, Staffordshire ST15 8AR. Capillary Probe Mixtures Kit 681C contains 15 mixtures for check- ing the performance of most capillary columns, from essentially non-polar to polar (strongly acidic or basic). The com- PolyScience capillary probe mixtures Chrompack purge and trap injector released by rapid heating and directly injected into the capillary column. A special feature is the condenser, which removes the entrained water, thus preventing blocking of the cold trap. Chrompack UK Ltd., Unit 4, Indescon Court, Millharbour, London El4 9TN.Solvent Delivery System for HPLC The Series 400 four-solvent delivery system can be used for all types of HPLC: microbore, analytical, high speed and semi-preparative. Single function com- mand keys simplify operation, including the multi-method programming. A method consists of up to 10 programming steps, any of which can be modified on the fly. Nine user-defined methods can be stored and all stored data are protected by battery back-up. Low conversion volumes allow solvent compositions to be changed in seconds. The Series 400 pump can be programmed for isocratic or gradient analysis. Simultaneous solvent and flow programming is standard. ion pair and ion exchange. The operator simply instructs the system to scout the appropriate variable, i.e., solvent strength, pH, ionic strength, and reviews the result.Features include the possibility of investigating a triangular plane that describes a three-solvent space and a three-dimensional tetrahedron that des- cribes a four-solvent space. By means of the colour triangle on the screen the user can position the cursor over an area of interest and call on to the screen the mobile phase composition and file name for the chromatogram taken at that com- position. Chromatographics 3 graphics can then be used to look at the actual chromatogram so that ideal conditions can be selected for the separation. Perkin-Elmer Ltd., Post Office Lane, Beaconsfield, Buckinghamshire HP9 1QA. Computing Integrator The CI-1OB now features base-line draw- ing as standard.When the run is complete the CI-1OB automatically plots the base position of the capillary probes for the more polar columns are adjusted to have the hydrocarbon peaks in the position where they are most useful. The probes for strongly acid and strongly basic col- umns are useful in checking the complete- ness of deactivation of fused silica col- umns before they are coated. Poly Science Corporation, P. 0. Box 48312, Niles, Illinois 60648, USA. Metering Pump for HPLC New features have been added to the microMetric pump. The microMetric now has the capability of being externally controlled: the run, stop, purge, fill, pre-pressure, and flow-rate functions can be remotely actuated via a rear panel connector. In addition, provision has been made to monitor remotely the pump’s pressure. A partial fill provision has been added to allow a pre-settable level less than the full syringe barrel capacity.Other features include pressure capability to 10 000 lb in-2, a settable overpressure limit with re-set and an LED overpressure signal. Laboratory Data Control (UK] Ltd., Milton Roy House, High Street, Stone, Staffordshire ST15 8AR.ANALYTICAL PROCEEDINGS, JULY 1985, VOL 22 221 Ion Chromatography Columns The Hamilton PRP-X100 is a resin based anion column specifically designed for ion chromatography. It is used exclusively in the non-suppressed mode. Ions such as C1-, NO7.- and SO& can be cleanly separated in 90 s and there is virtually no restraint on the column eluent range (pH 1-13).The PRP-X100 can be used for separations involving borates, fluorides and silicates and three sizes are available: 100, 150 and 250mm. Hamilton HC-40 and HC-75 polysulphonic and cation exchange resin gels are formulated for the separation of complex carbohydrates, polyols and glycols. HC-40 is semi-rigid and useful for higher oligosaccharides analysis or the observation of anomeric flavouring components in beer. HC-75 gel is suitable for the rapid screening of many sugar-containing materials, such as beet sugars and corn syrups. Both HC-40 and HC-75 are stable in the pH range 1-13. V. A. Howe and Co. Ltd., 12-14 St. Ann's Crescent, London SW18 2LS. Surface Analysis Systems The XSAM 800 provides high energy resolution X-ray photoelectron spectro- scopy (XPS) with a high sensitivity scan- ning Auger microprobe (SAM).High performance secondary ion mass spectro- scopy (SIMS) and ion scattering spectro- scopy (ISS) use the makers' Analytical MiniBeam ion gun capabilities to provide complementary information from the outermost layers of a material. Other techniques, such as UPS and LEED, are fully supported within the system. Appli- cations of the system include analyses of fibres, polymers, catalysts and lubricants and the investigation of adhesive prob- lems, discoloration and corrosion. It is also used in materials science for research into embrittlement, fatigue, cracking and wear as well as in semiconductor produc- tion. The Series 800 SIMS system is configured as a combination SIMS/ISS system that features a complete vacuum system of gauges, pumps, controls and a nineteen-port analysis chamber. It is com- pletely computer controlled and it incor- porates a quadrupole mass analyser with an integral energy filter.A choice of ion guns can be made from the range which includes MiniBeam I and 11, MacroBeam, MacroFAB, liquid metal and Duoplas- matron models. Automation is provided by the SIMScan computer system desig- ned for spectrometer control, data acqui- sition and information handling. Spectros International plc, Barton Dock Road, Urmston, Manchester M312LD. Portable Infrared Analyser The PA404 provides direct digital readout of the concentration of a specific com- pound and is suitable for such applications as hazard level C02 measurements in work areas, breweries and greenhouses, solvent leak detection in the hydrocarbon and petrochemical industries, CO analy- sis in garages and ethylene oxide measure- ment in hospitals.The PA404 can be calibrated to measure most organic and Spectros XSAM 800 many inorganic compounds with direct range from 100p.p.m. to loo%, depend- ing upon the particular application. It has been fully tested by the CEGB and is a Recommended Instrument for Hazardous Area Monitoring of C02 in air, particu- larly in nuclear power plants. Servomex Ltd., Crowborough, Sussex TN6 3DU. Titrator The DL20 CompactTitrator is a single- method titrator which ensures exact automatic repetition of a method, even after a change in operator. It can be configured, for example, for the following methods: titrations to a pre-set end-point or equivalence point; p and m values (acid and base capacities) in water; TAN/TBN (total acid or base number according to DIN/ASTM methods) in petroleum pro- ducts. Provision is made for the inter- facing of the makers' balances and print- ers, sample transports and various com- mercial printer - plotters.The weighing data are reconciled automatically and recorded. The DL20 can be operated by a personal computer via an RS232C inter- face. Mettler Instrumente AG, CH-8606 Greifensee, Switzerland. Diluter - Dispenser The Hamilton MLlOOO is a development from the Microlab 1000. It employs a yedno type of programming and opera- tion. Its microprocessor handles all instru- ment operations and will store up to 50 methods in a battery backed memory. Diluent of between 1 p1 and 25 ml and samples of up to 5ml are used.An RS232C interface is supplied as standard. V. A. Howe and Co. Ltd., 12-14 St. Ann's Crescent, London SW18 2LS. Preparation of Standard Solutions An automatic computer controlled system from Hamilton combines the makers' Microlab M diluter - dispenser with a Mettler or Sartorius balance and Epson HX-20 computer. The user puts an approximate amount of sample on the balance. The computer registers the exact mass and calculates the amount of liquid required to be added and this is automat- ically dispensed by the Microlab M. V. A. Howe and Co. Ltd., 12-14 St. Ann's Crecent, London SW18 2LS. pH Meter The Radiometer PHM85 offers resolution of0.001pH,0.001pX,O.1mVand0.1"C. Buffer values can be selected as required and are stored in the microprocessor's memory.Sensitivity, zero point and ISO- pH point are automatically calculated and displayed during calibration. There is an RS232C interface and the PHM85 can be connected to a sample changer for auto- matic measurement of up to 20 samples in series. When connected to an automatic222 ANALYTICAL PROCEEDINGS, JULY 1985, VOL 22 burette the PHM85 controls automatic end-point titrations. V. A. Howe and Co. Ltd., 12-14 St. Ann's Crescent, London SW18 2LS. Conductivity - Temperature Stick Meter The SMCl meter for measuring the con- ductivity and temperature of liquids can be used with the cell coupled directly to the meter for one-handed operation or with an extension lead for taking readings in awkward places. Its three conductivity ranges cover most applications and tem- perature is measured in the range 0-45 "C.J. Bibby Science Products Ltd., Stone, Staffordshire ST15 OSA. Thermometer A hand-held digital, battery-powered thermometer for measuring temperatures from -50 to +500"C to an accuracy of +1% is available. It is suitable for use in ambient temperatures from -5 to +50 "C. There is indication of low batteries when these approach the end of their life, on average after 600 h. Alexander Controls Ltd., Reddicap Trading Estate, Sutton Coldfield, West Midlands. Digital Thermometer This instrument is capable of absolute and differential temperature measurements with a resolution of 0.001 "C from -270 to +630"C. A wide range of easily inter- changeable sensors is available. The thermometer is microprocessor con- trolled. Linearisation data are held in ROM, and the readout is in "C or O F .RS 232 and IEEE options are available. Automatic Systems Laboratories, Saxon Street, Linford Wood, Milton Keynes, Buckinghamshire MK14 6LD. Gas Flow Meter The measurement range of the FM5 Microflowmeter has been doubled. The new instrument covers the range 0-10 ml min-1. This extension of range has been achieved by means of a flow multiplier, which can be readily fitted or removed. Flows of less than 0.25 ml min-1 can be measured. Cook Variometers Ltd., P.O. Box 36, High Wycombe, Buckinghamshire HP13 6BB. Thermal Analysis System The FP800 Thermosystem can now be enhanced by the connection of a com- puter. A software kit has been designed for the Epson HX20 hand-held computer.As well as controlling the unit, it permits various standard applications to be called up, results to be calculated and data stored for later processing. The FP800 can also be configured for special applica- tions, so that it operates in accordance with national and international standards, e.p.. measuring the melting-Doint of Dlas- tics according to ASTM and the clear melting-point of fats according to AOCS. Mettler Instrumente AG, CH-8606 Greifensee, Switzerland. Ethylene Oxide Personal Monitoring System The Amsco self-scan daily monitoring system consists of a badge holder and badge, a one-step developing solution and an electronic reader. Each badge takes seconds to prepare and is ready to read in 10 min, including developing time.It identifies personal exposure levels accu- rately down to 0.3 p.p.m. ethylene oxide. TSC Medical Co. Ltd., Mediplan House, Market Place, Warwick cv34 4SL. Analysis of Powder Surface Area Monosorb, part of the Quantachrome range of powder characterisation instru- ments, uses a single point Brunauer, Emmett and Teller method, provides reproducibility better than 0.5% and requires only 6min for completion of an analysis. Only small samples are required. Samples can be outgassed prior to analysis on the built-in outgassing station or on the compatible Monotector dual outgassing station and then transferred to the test station without contamination. Techmation Ltd., 58 Edgware Way, Edgware, Middlesex HA8 8JP. Blood Analyser The Counter Model S770 offers a seven parameter blood profile, performed auto- matically from sample aspiration through to the hard-copy printout of results at a Coulter Electronics Ltd., Northwell Drive, Luton, Bedfordshire LU3 3RH.Lactate Measurement L-Lactate measurement capability has been added to the Model 27 Industrial Analyser, permitting the measurement of aerobic capacity in sports medicine research. Other measurement capabilities of the Model 27 are fructose, sucrose, dextrose, lactose, starch and alcohol. The measurement range for lactate is 0-134 mg dl-1 (0-15.00 mmol l-1). Clandon Scientific Ltd., Lysons Ave- nue, Ash Vale, Aldershot, Hampshire GU12 5RQ. Potentiostat - Galvanostat The ECP130 has a current output of 1 A and a potential compliance of 30 V. It can be used with either a 2- or 3-electrode cell and can maintain either a pre-set flow of current through the cell (galvanostat) or maintain a pre-set potential (potentio- stat). It has IR compensation to overcome the high solution resistance often encoun- tered in organic liquids.EDT Research, 14 Trading Estate Road, London NWlO 7LU. Biological Alkali Microanalyser The Model OP-266 measures the concen- tration of sodium and potassium ions in a wide variety of biological fluids, including blood and plasma. The sample is aspir- ated by a vacuum-type pump, the read button is pressed and as soon as a steady reading is obtained the result appears on the display in concentration units. Ion- selective electrodes are employed, which Coulter S770 blood analyser rate of 70 specimens h-1. It is compatible are thermostatically controlled.The sam- with the makers' Haematology Accu- ple requirement is only 50 p1. Comp system for full quality control EDT Research, 14 Trading Estate assurance. Road, London NWlO 7LU.ANALYTICAL PROCEEDINGS, JULY 1985, VOL 22 223 Rifflers The Quantachrome Sieving Riffler takes up to 2500cm3 of powder into its hopper via a sieve which can remove large agglomerations or extra-large particles according to the choice of mesh size. The powder is then vibrated in the hopper and fed gradually into a series of eight collec- tors. For smaller samples the Rotary Micro Riffler can take an initial quantity of up to 150 cm3 of powder and distribute it evenly into a series of test-tubes of 4 or 15 cm3. Techmation Ltd., 58 Edgware Way, Edgware, Middlesex HA8 8JP.Rotary Mixer A new mixer accommodates up to 32 samples per tray, with the option of additional tray assemblies that can be pre-loaded ready for mixing. It rotates samples at any one of three angles, 45", 60" or 75", ensuring flexibility whatever size of tube is used. An efficient mixing process of end over end inversion is provided. Coulter Electronics Ltd., Northwell Drive, Luton, Bedfordshire LU3 3RH. Immersion Probe Coolers Neslab Cryocool probe coolers are an alternative to liquid nitrogen for continu- ous cooling applications, such as cold trapping. They provide temperatures as low as -100°C continuously. A wide range of temperature control accessories is available and low temperature fluid circulation is possible using the Neslab Cryoflow pump.Jencons (Scientific) Ltd., Cherrycourt Way Industrial Estate, Stanbridge Road, Leighton Buzzard, Bedfordshire LU7 8UA. Test-tube mixer The Miximatic has variable speed control and accepts a variety of sizes of test-tubes and flasks. It creates a vortex instantly. Jencons (Scientific) Ltd., Cherrycourt Way Industrial Estate, Stanbridge Road, Leighton Buzzard, Bedfordshire LU7 8UA. Transilluminators A range of 10 modular transilluminators now features stainless-steel tops and smoked plastic tops for eye and filter protection. The range caters for 254, 302 and 365 nm irradiation of all types of gel, from mini-size up to 40cm. Ultra-Violet Products, Science Park, Milton Road, Cambridge CB4 4BN. Digital Storage Oscilloscope The Hitachi VC-6041 instrument, suitable for analysis and observation of high-speed signals, samples at 40 MHz, providing digital storage of 10 MHz single-shot events and 40 MHz repetitive waveforms.It features a memory of 4096 words per channel, enabling a horizontal resolution of 400 steps per division. A portion of the stored waveform can be magnified up to 100 times for detailed observation. The instrument can store up to 4 waveforms simultaneously without any loss of resolu- tion. Features include a ground reference display and an X - Y display function to enable accurate phase shift measure- ments. An analogue output for chart recorders of X - Y plotters is standard and a GPIB IEEE 488 interface is an optional extra. Thurlby Electronics Ltd., New Road, St. Ives, Cambridgeshire PE17 4BG. Water Still The Aqua 4200 automatic still is electron- ically controlled and can be supplied with either mains water or water from an ion-exchange unit to produce distilled water complying with all of the principal pharmacopoeias.The conductivity of the product is up to 1 pS cm-1 (25 "C). It has an output of 4.2 1 h-1 and is suitable for wall mounting or benchtop use. Schott Glass Ltd., Drummond Road, Astonfields Industrial Estate, Stafford ST16 3EL. Graphing Systems A range of systems is available for use with the makers' 4035 and 1421 digital storage oscilloscopes. They provide accu- rate measurement of input voltage levels, from less than 2 mV to 25 V, as well as the ability to plot waveforms, with band- widths up to 20MHz and rise times of as low as 80 ns, on to A4 chart paper over the complete chart size.Features include a video screen which displays the waveform prior to plotting. Gould Electronics Ltd., Roebuck Road, Hainault, Ilford, Essex IG6 3UE. Heated Sample Lines A range of heated sample lines for use with analysers of hydrocarbons and oxides of nitrogen is announced. Pri- marily designed for carrying heated diesel exhaust gas from engine to analyser, the lines provide a constant high temperature to avoid the risk of gas condensation. They are run on low voltage, electrically isolated from the mains power supply, and are free from electrical interference which could disturb sensitive measuring equipment. Signal Instrument Co. Ltd., Standards House, 1 Doman Road, Camberley, Sur- rey GU15 3DW. Glass Lined Metal Tubing The material is available in Yi6, ?A8 or Y4 in 0.d.sizes with a choice of internal diameters. Scientific Glass Engineering (UK) Ltd., 1 Potters Lane, Kiln Farm, Milton Keynes MKll3LA. Fusible Fittings In the event of fire the fittings automatic- ally shut off the flow of flammable or hazardous fluids. For example, a fusible fitting can be installed on the air supply to a normally closed air-to-open valve. If there is a fire the fusible material will melt, causing the loss of air pressure. The fittings are available in brass or 316 stainless steel and are fitted with a eutec- tic material with a choice of melting- points (71, 124 or 138°C). Sizes are $, $ and f in. Swagelok (UK) Ltd., 3 Kelvin Close, Science Park North, Birchwood, War- rington WA3 7PB. Laser Power and Energy Monitors The Series 5000 monitor has improved accuracy and features fast response characteristics, allowing the sensitivity of the radiometer to be fully exploited for both accurate measurement and for the observation of trends when tuning lasers or setting up optical systems.Laser Instrumentation Ltd., Unit 1, Bear Court, Daneshill East, Basingstoke, Hampshire. Literature A 32-page guide, "Troubleshooting Guide-How to Locate Gas Chromato- graphy Problems and Solve Them Your- self," includes sections on isolating the source of a problem, checking the carrier gas system, testing the system for leaks and changing the column and septum. A 16-page troubleshooting table sum- marises many common trouble symptoms, their possible causes and recommended remedies.Supelchem (R. B. Radley and Co. Ltd.), London Road, Sawbridgeworth, Hertfordshire CM21 9JH. A leaflet presents the Ion83 ion meter, which offers pH, pX and mV measure- ment. Calibration data for 1-3 electrode pairs are stored individually. Also featured is multi-point calibration with up to 5 standards. There is an output for connection to a sample handler. Radiometer A/S, Emdrupvej 72, DK- 2400, Copenhagen NV, Denmark. A leaflet describes the 8660 XRF simul- taneous and the 8680 XRF simultaneous - sequential spectrometers. Optimised for multi-channel analyses, the 8600 series224 ANALYTICAL PROCEEDINGS, JULY 1985, VOL 22 complements the 8400 series, which is optimised for sequential applications. The 8660 features up to 31 standard monochromators, while the 8680 offers up to 20 standard monochromators with one goniometer.Applied Research Laboratories SA, En Vallaire, CH-1024 Ecublens, Switzer- land. Two brochures deal with the industrial use of analytical equipment. One, entitled “Analytical Equipment in the Iron and Steel Industry,” covers iron and steel making together with analytical require- ments, techniques, instrumentation, procedures and performance, referring to emission spectroscopy and X-ray fluores- cence spectrometry as well as the use of computers and microprocessors. The other is on X-ray spectrometry in the cement industry and covers the cement process, analytical requirements and the XRF technique and sample preparation procedures. Pye Unicam Ltd., York Street, Cam- bridge CB1 2PX. Two brochures on chromatography are available, One, “Capillary Column Chro- matography,” describes various injection techniques, gives information on selecting columns, column mounting and connec- tions and includes advice on achieving automation and precision through the use of an autosampler.The other, “The Fourth Dimension in Liquid Chromato- graphy,” describes the makers’ modular HPLC products: pumps, injectors, col- umns, detectors and data systems. Perkin-Elmer Ltd., Post Office Lane, Beaconsfield, Buckinghamshire HP9 1QA. Two applications notes, “Assay of Acid Constituents in Mixed Acid and Phosp- horic Acid Etchants” and “Determination of Sulphur Compounds in Kraft Liquors Using Ion Chromatography,” have been issued. The first describes the procedures for separating various acids used as etchants and pickling liquors by ion chro- matography exclusion and subsequent detection by conductivity.The other out- lines the ability of ion chromatography to provide information about the ionic con- tent of the Kraft black, green and white liquors; chromatograms show that sul- phite, sulphate, oxalate and thiosulphate can all be determined in less than 20 min. Dionex (UK) Ltd., Eelmoor Road, Farnborough, Hampshire GU14 7QN. A series of 12 applications sheets, “Gas Adsorption and Catalysis” has been issued. Contained in a folder with general introduction and bibliography on hetero- geneous catalysis and zeolites they are available on request along with previous sheets: “Elastomers,” “Cements and Plasters,” “Thermal Analysis of Fossil Fuels” and “Thermal Hazard Evalua- tion.” Clandon Scientific Ltd., Lysons Avenue, Ash Vale, Aldershot, Hamp- shire GU12 5RQ. A folder contains leaflets on ASYST software for analytical chemists and researchers. Used with the IBM Personal computer, ASYST includes: Module 1 , system - graphics - statistics; Module 2, analyses; and Module 3, data acquisition. The software runs on the IBM PC, PC - XT, PC - AT, Compaq, Columbia and other IBM-compatible computers. Macmillan Software Co., 866 Third Avenue, New York, NY 10022, USA. Leaflets describe a range of two-stage regenerable deionisers. The 12R and 24R units are manual models. They produce up to 250 and 400 1 h-1, respectively, of C02-free water during the first part of the treatment and then distilled quality water during the latter part of the run. The 40R and 60R are semi-automatic and offer 700 and 1000 1 h-1, respectively. They pro- duce a good standard of water and can be linked to end polishing units for extremely high quality water. The 40RA and 60RA are based on the 40R and 60R but incorporate a control module, PMC4, which initiates automatic regeneration, the set point of which can be adjusted between 0.5 and 30pScm-1. The TSA 300 and 400 models offer maximum flow- rates of 1.5 and 2.3 m3 h-1 and the water conductivity is approximately 1.0 pS cm-1. Houseman (Burnham) Ltd., UK Indus- trial Division, Waterglade House, 53-57 High Street, Maidenhead, Berkshire SL6 1JU. The JME magnetic susceptibility balance is the subject of a 4-page leaflet. The balance employs moving magnets and a stationary sample tube, and magnetic susceptibility is calculated from an instan- taneous digital readout by using a simple equation. Only 250mg of sample are required. Johnson Matthey Equipment Ltd., South Way, Exhibition Grounds, Wem- bley HA9 OHW. Literature is available on Genesis 21, a fully automated clinical chemistry ana- lyser offering flexibility in processing of samples. It can carry out over 21 analy- tical procedures on up to 50 patient samples at a rate of 200 tests h-1. Allied Instrumentation Laboratory (UK) Ltd. Leaflets give details of the SR-1OA and Micro (MSR) spinning rifflers. The SR- 10A provides 20 equivalent samples from a single pass. The MSR has a maximum capacity of 25ml before riffling. The makers’ complete range of spinning rif- flers includes several other models. Microscal Ltd., 79 Southern Row, Lon- don W10 5AL.

ISSN:0144-557X    

DOI:10.1039/AP9852200218

出版商:RSC    

年代:1985

数据来源: RSC

摘要:

224 ANALYTICAL PROCEEDINGS, JULY 1985, VOL 22 Analytical Division Distinguished Service Award Nominations are invited for the Division’s Distinguished Service Award, the Rules for which are as follows: 3. The aim of the Award is to recognise exceptional voluntary service over a period of years to the Analytical Division of The Royal Society of Chemistry (including that to the Society for Analytical Chemistry). The Award shall normally be in the form of an illuminated address which may be accompanied by such additional recognition as Council of 4. the Division shall agree. Nominations for the Award will be invited annually from members of Council of the Division, and may be received from any member of the Division. They shall be made in 5 . writing, with supporting evidence, to the President of the Analytical Division, Royal Society of Che- mistry, Burlington House, London, Analytical Division, which shall recommend to Council of the Divi- sion ( a ) to whom an award should be made, (b) the nature of the award or (c) that no award should be made. The Award shall be made by the Council of the Analytical Division, which must approve any alteration of these Rules. W~V-OBN. - Nominations for the Award should be Nominations shall be considered by sent to the President of the Analytical the Honours Committee of the Division before August 31st, 1985.

ISSN:0144-557X    

DOI:10.1039/AP9852200224

出版商:RSC    

年代:1985

数据来源: RSC