ISSN:0003-2654    

DOI:10.1039/AN99116BP027

出版商:RSC    

年代:1991

数据来源: RSC

ISSN:0003-2654    

DOI:10.1039/AN99116FX029

出版商:RSC    

年代:1991

数据来源: RSC

摘要:

ANALAO 1 16(8) 773-880 (1 991 )The AnalystAugust 1991The Analytical Journal of The Royal Society of ChemistryCONTENTS773 Evaluation of a Rapid Technique for the Determination of Precious Metals in Geological Samples Based on a Selective781 Compensation for Particle Size Effects in Near Infrared ReflectanceChristine R. Bull787 Immunocomplex-immobilization Technique-Sven Oscarsson, Jan Carlsson793 Immobilized-enzyme Electrode for Nicotinamide Adenine Dinucleotide (Reduced Form) (NADH) Sensing andApplication t o the Kinetic Studies of NADH Dependent Dehydrogenases-Hsien-Chang Chang, Akinori Ueno,Hiroshi Yamada, Tomokazu Matsue, lsamu Uchida797 Amperometric Monitoring of Sulphur Dioxide in Liquid and Air Samples of Low Conductivity by Electrodes Supportedon lon-exchange Membranes-Gilbert0 Schiavon, Gianni Zotti, Rosanna Toniolo, Gin0 Bontempelli803 Voltammetric Behaviour of Salicylic Acid at a Glassy Carbon Electrode and Its Determination in Serum Using LiquidChromatography With Amperometric Detection-Denley Evans, John P.Hart, Glan Rees807 Improvement on the Microdiffusion Technique for the Determination of Ionic and Ionizable Fluoride in Cows'Milk-Jacobus F. van Staden, Sophia D. Janse van Rensburg81 1 lndomethacin lon-selective Electrode Based on a Bis(triphenylphosphoranylidene)ammonium-lndomethacinComplex--Ralph Aubeck, Christoph Brauchle, Norbert Hampp815 Column Preconcentration of Trace Metals From Sea-water With Macroporous Resins Impregnated With LipophilicTetraaza Macrocycles-Stephane Blain, Pierre Appriou, Henri Handel821 Determination of Trace Amounts of Tellurium by the Sub-superequivalence Method of Isotope DilutionAnalysis-H i roe Yos h io ka, Yos h i no ri Mi ya ki, Ku ni h i ko H asegawa825 Separation and Determination of Trace Amounts of Cadmium by On-line Enrichment in Flow Injection Flame AtomicAbsorption Spectrometry-Rajesh Pu roh it, Su rekha Devi831 Flow Injection Sample-to-standard Additions Method Using Atomic Absorption Spectrometry Applicable t oSlurries-I g n a ci o Lopez G a rcia, Fra nci sca 0 rt iz So beja n 0, Ma n u e I Her n a n d ez Co rdo ba835 Flow Analysis Method for the Determination of Silicic Acid in Highly Purified Water by Gel-phase Absorptiometry WithMolybdate and Malachite Green-Kazuhisa Yoshimura, Ushio Hase841 Differential Conductimetry in Flow Injection.Determination of Ammonia in Kjeldahl Digests-Jarbas Jose RodriguesRohwedder, Celio Pasquini847 Determination of the Platinum, Rhenium and Chlorine Contents of Alumina-based Catalysts by X-ray FluorescenceSpectrometry-Rao V. C. Peddy, G. Kalpana, ValSamma J. Koshy, N. V. R. Apparao, M. C. Jain, R. V. Patel851 Determination of the Insecticide Promecarb by Fluorogenic Labelling With Dansyl Chloride-F. Garcia Sanchez, C.Cruces Blanco857 Determination of Epinephrine, Norepinephrine, Dopamine and L-Dopa in Pharmaceuticals by a PhotokineticMethod-Carmen Martinez-Lozano, Tomas Perez-Ruiz, Virginia Tomas, Otilia Val861 Selective Spectrophotometric Method for the Determination of Ascorbic Acid in Pharmaceutical Preparations andFresh Fruit Juices-Enaam Y. Backheet, Kamla M. Emara, Hassan F. Askal, Gamal A. Saleh867 Colorimetric Method for the Determination of Captafol (Difolatan) in Commercial Formulations and Residues on Grainsand Apples-Balbir C. Verma, Devender K. Sharma, Hari K. Thaku, Bh. Gopal Rao, Narendra K. Sharma871 Determination of Atmospheric Sulphur Dioxide by Solid-phase Spectrophotometry-Juan M. Bosque-Sendra,Francisca Molina, Eugenia Lopez875 BOOK REVIEWS879 CUMULATIVE AUTHOR INDEXAqua Regia Leach-Charles J. 6. Gowing, Philip J. PottsTypeset and printed by Black Bear Press Limited, Cambridge, Englan

ISSN:0003-2654    

DOI:10.1039/AN99116BX031

出版商:RSC    

年代:1991

数据来源: RSC

摘要:

ANALYST, AUGUST 1991, VOL. 116 773 Evaluation of a Rapid Technique for the Determination of Precious Metals in Geological Samples Based on a Selective Aqua Regia Leach Charles J. B. Gowing and Philip J. Potts" Department of Earth Sciences, The Open University, Walton Hall, Milton Keynes MK7 6AA, UK Results are described for optimizing the rapid determination of precious metals in geological samples (principally of ophiolitic origin, including chromitites) using an aqua regia [HCI + HN03 (3 + I)] leach followed by inductively coupled plasma mass spectrometry. Little improvement in extraction efficiency is afforded by aqua regia attack for longer than 30 min and heating the extraction mixture appears t o have a detrimental effect on the extraction efficiency of the precious metals.The procedure was evaluated by analysis of a suite of standard reference materials and some independently analysed ophiolitic rock samples. Generally, Au and Pd are quantitatively extracted, Pt, Rh, Ru and 0 s are extracted t o a lower but significant extent (2040% recovery) and Ir is poorly extracted (typically I-10% recovery). The insolubility of selected platinum group element minerals in aqua regia is considered t o be the dominant effect for the non-quantitative recoveries. Keywords: Aqua regia leach; geological sample; platinum group elements and gold determination; inductively coupled plasma mass spectrometry; mineral insolubility In recent years, considerable interest has been shown in the development of techniques for the determination of Au and the platinum group elements (PGEs) in geological samples.This has been stimulated not only by. the extensive exploration interest in these metals but also by the development of instrumentation of high sensitivity [particularly inductively coupled plasma mass spectrometry (ICP-MS)] capable of extending applications to samples containing parts per billion (ppb) levels of these elements. Despite these advances in instrumentation, most analytical techniques require a preconcentration step so that these ppb detection limits, essential in geological applications, can be achieved. Many current procedures are based on fire assay. The traditional lead fire assay procedure'-3 is normally restricted to the quantitative recovery of Au, Pt and Pd.435 Over the last decade, there has also been considerable interest in the nickel sulphide fire assay procedure,6,7 not the least because this procedure is capable of recovering quantitatively all the precious metals from a range of sample types.Indeed, the nickel sulphide procedure is effectively the analytical 'bench mark' against which other techniques are normally evaluated and has been subjected to useful further refine- ments in reducing the mass of the button necessary to recover the PGE from 20 g to 1 g.8 Despite the widespread use of fire assay techniques, these procedures have some drawbacks. One reservation is the low recoveries that can be encountered from some sample types, a second fusion then being necessary to recover precious metal residues from the silicate glass.3.4 Another reservation is that the success of the fire assay procedure depends to some extent on the skill and judgment of the operator in optimizing both the flux composition and fusion conditions.With this back- ground, it was decided to evaluate an alternative, very simple, extraction procedure based on a selective aqua regia [HCl + HN03 (3 + l)] leach. This procedure requires only standard laboratory equipment and can be undertaken without diffi- culty by non-specialized laboratory personnel. Furthermore, the technique has the potential of being adapted for operation in a field laboratory. The aim of this paper is, therefore, to report preliminary data on investigations of the aqua regia leach procedure for recovering the PGEs. Aqua regia leach is currently used widely for the determina- tion of Au in geological samples.9-12 The success of the * To whom correspondence should be addressed.technique depends on the ability of aqua regia to dissolve native Au and associated alloys with Cu and Ag. This property arises from the reactivity of nitrosyl chloride (NOCl) and/or free chlorine formed in freshly prepared aqua regia solution (see for example Latimer and Hildebrand13). In some procedures, hydrofluoric acid is added to the reaction mixture to attack silicate phases and facilitate the liberation of Au grains that might not otherwise be wetted by aqua regia alone .14-16 Although there are some reports in the literature concern- ing the use of aqua regia in the determination of the PGEs,17-*4 this preconcentration procedure is not extensively used in practical applications, in part because of uncertainties in the quantitative recovery efficiencies and how this is influenced by sample matrix and PGE mineralogy. One of the aims of the present study is to investigate further the influence of PGE mineralogy on the recovery efficiency by using aqua regia leach.However, as little information is available on the detailed PGE mineralogy of the geological reference materials that are currently available, these materials are not ideal candidates for this study. Following an extensive project to characterize the PGE mineralogy of the Unst ophiolite (Shetland Isles, N. Scotland), detailed minera- logical descriptions of several samples are available .25-*7 One of the samples described as part of this project was a serpentinized chromitite (CHR-C) from Cliff, Unst , contain- ing exceptionally high abundances of the PGEs (in the ppm range).This sample was an obvious candidate for the present investigation, partly because of the availability of a detailed PGE mineralogical description, and partly because one of the secondary aims of the present work was to develop an analytical method suitable for chromitites, recognizing that such samples are not always easy to analyse by conventional fire assay techniques.2.28 Complementary measurements were undertaken on other chromitite samples from Unst (more detailed descriptions of which can be found in the references cited above) and on a reference sample as follows: CHR-A, a chromitite from Harold's Grave, Unst, containing enhanced PGE abundances, particularly Os, Ru and Ir; CHR-B, a 'barren' chromitite from Unst ; CHR-C, an exceptional PGE mineralized chromitite from Cliff, Unst (referred to above); and SARM 7 (previously PTO-l), a platinum ore Certified Reference Material (CRM), available from the South African Bureau of Standards (SABS) for which certified values of all the PGEs are available.29 This sample consists of a gabbroic rock and is therefore representative of silicate materials.The samples CHR-B and CHR-C are sub-samples of the774 ANALYST, AUGUST 1991, VOL. 116 proposed chromitite reference samples CHR-Bkg and CHR- Pt+ , respectively, which are, at the time of writing, the subject of an international cooperative study.30 These sub-samples were separated after jaw-crushing the bulk material used to prepare the reference materials.Although never fully homogenized at the milled powder stage, the mineralogies (though not necessarily the abundances) of CHR-B and CHR-C are expected to be virtually identical to those of CHR-Bkg and CHR-Pt+ , respectively. Experimental Aqua regia was freshly prepared before each experiment by mixing AnalaR or Aristar grade concentrated nitric and hydrochloric acids (BDH Chemicals). Although experimental conditions were varied in evaluating the optimum extraction conditions, the standard procedure was the same. Standard Procedure for the Aqua Regia Leach Freshly prepared aqua regia (20 ml) was added to the rock powder (10.0 f 0.1 g) in borosilicate glass beakers (150 ml) and mixed using a polytetrafluoroethylene rod until com- pletely wetted. The beaker was covered with a watch-glass and the mixture stirred on a magnetic stirring table (together with other samples in the batch) for 1 h at room temperature.The contents of the beaker were then poured into a filter funnel fitted with a Whatman 40 filter-paper (55 mm) together with washings from both beaker and watch-glass cover, rinsed with the minimum volume of de-ionized water. After filtration, the filter-paper and residue were transferred into a polythene bag and sealed for subsequent analysis, as appropriate. The filtrate was transferred into a graduated flask (100 ml) and made up to volume with de-ionized water. Beakers and storage flasks were cleaned before use by soaking in laboratory cleaner overnight and then rinsing with de-ionized water.Between runs, other glassware was rinsed with 50% nitric acid (AnalaR) and then de-ionized water. Each batch of samples included a blank aqua regia control (processed in the same way as the samples) and all leach experiments were performed in duplicate. Analysis of Solutions The aqua regia leach solutions were analysed by ICP-MS (VG Elemental PlasmaQuad) using the Natural Environmental Research Council (NERC) facility then based at the Univer- sity of Surrey or the British Geological Survey (BGS) instrument then based in London. The operating conditions broadly followed those described by Jarvis31 and Gray and Williams32 with some minor amendments to take into account the aggressive nature of the sample solutions (20% aqua regia, and >O.1% total dissolved solids). The auxiliary gas flow rate was set at 0.5 1 min-l, the sample uptake rate being about 0.5 ml min-1. The instrument was set up using a nine element standard solution and the ion lens settings were optimized in order to maximize the 115In+ signal and minimize the doubly charged and oxide ion interferences as monitored by '40Ce2+ and 140CeO+. In order to match the matrix of the sample solutions, the standard solution was prepared with elemental concentrations of 100 pg 1-1 in 20% aqua regia (Aristar grade). This standard solution was run after a calibration blank (20% aqua regia) at the beginning of each batch and then repeated after at least every fifth sample. These data from the standard solution were used to apply a correction for drift in instrument sensitivity by interpolation of count data from adjacent standard solutions.In the course of this work, it was realized that additional interference effects were occurring owing to the suppression of signals in sample solutions containing particularly high contents of dissolved salt. An additional correction was then applied by adding Re and In (both at 100 pg I-* in the analysed solution) as internal standards to all sample and standard solutions. Data for Rh, Ru and Pd were normalized against the In signal and Os, Ir, Pt and Au against the Re internal standard signal, thus minimizing any mass-dependent effects. A blank solution (20% aqua regia) was nebulized for 70 s between sample solutions and 90 s between a drift correction standard (containing higher concentrations of the PGE) and a sample solution in order to avoid memory effects.Analysis of Residues A further series of measurements was carried out on the residue after the aqua regia leach, in order to determine the proportion of concomitant elements co-extracted by aqua regia and therefore present in the analyte solutions as potential interferences. After drying, aliquots of the residue were prepared as powder pellets and analysed by energy dispersive X-ray fluorescence (ED-XRF) , generally following the procedures described by Potts et aZ.33 The proportion of the element leached by aqua regia was calculated from the ratio of measured intensity data between the leached and unleached sample. No formal correction to these data was applied for matrix effects because of the very small fraction of matrix elements leached by the aqua regia.However, some compensation for drift and any residual matrix effects was applied by normalizing the intensity data against Cr counts for the current sample. Other aliquots of the residue were analysed by instrumen- tal neutron activation analysis (INAA) following the proce- dures described by Potts et ~ 1 . 3 ~ A calibration was undertaken relative to a sample of Ailsa Craig microgranite that had been spiked with standard solutions, mixed as a slurry, dried and lightly ground in order to disaggregate the resultant cake. The spiked concentrations in this standard were 1.02 pg g-1 Ir, 10.4 pg 8-1 Pt and Sb, 4.17 pg 8-1 Au and 104 pg g-l As.Neutron flux corrections were made by interpolating the response from two samples of chromitite CHR-C placed at either end of the irradiation can, each of which contained a total of 11 samples. Results and Discussion Optimizing the Aqua Regiu Leach A series of experiments was carried out to test the influence of various parameters on the aqua regia extraction efficiency. The parameters tested were contact time, heating and aqua regia composition. In order to simplify procedures, all experiments were carried out on 10 g of rock powder and 20 ml of aqua regia. Results were obtained both by the analysis of filtrate solutions (ICP-MS) and of residues (ED-XRF and INAA). Experiments were performed on chromitites CHR- A, CHR-B, CHR-C and the platinum ore SARM 7, represent- ing a silicate matrix. Varying the aqua regia composition between the limits of 1 + 3 and 3 + 1 HCl + HN03 appeared to have little influence on the extraction data and all subsequent measurements were made with a standard aqua regia mixture (3 + 1, HCl + HN03).The effect of contact time and heating on the aqua regia-sample reaction mixture are shown in Figs. 1-3. In Fig. 1, data are plotted for the proportion of PGE extracted from the mineralized chromitite CHR-C and the platimum ore SARM 7 at room temperature for a period from 15 to 120 min and also for a separate experiment in which the extraction mixture was heated for 120 min. The results are calculated as the proportion (%) of the element extracted, arbitrarily normalized to the maximum value in order to simplify the visual interpretation of the diagrams.It must be emphasized,ANALYST, AUGUST 1991, VOL. 116 cn 775 50 therefore, that the results do not represent the absolute extraction efficiency, which for several elements is substan- tially less than 100%. The data in Fig. 1 indicate that, for the samples invest- igated, the reaction with aqua regia proceeds rapidly at room temperature and that prolonging the leach time beyond about 30 min does not appear to improve significantly the proportion recovered. The data also indicate that significant proportions of 0 s and Ru are recovered by an aqua regia leach at room temperature. However, for these elements, there is some indication that the corresponding ICP-MS signal is reduced as the contact time is extended.This trend may be due to losses of 0 s and Ru from the analyte or an increased ICP-MS suppression effect if the dissolved salt content of the leachate solutions increases with aqua regia contact time. None of the elements measured showed an improved recovery efficiency, based on the ICP-MS signal obtained in experiments where the extraction mixture was heated for 120 min. The reason for this was not established unambiguously. However, one cause - 50s 0 150 100 50 0 I 0 s I I I L 30 60 90 120 Ti mehi n z 20 60 Tern peratu rePC Fig. 1 Proportion of ( a ) Pd, (b) Pt and (c) Rh extracted from 1, CHR-C and 2, SARM 7 and (d) proportion of 0 s and Ru extracted from CHR-C by aqua regia extraction at room temperature for specified times and for 120 min at specified temperatures.Data have been arbitrarilv normalized to the maximum measured value in order to simplify visial interpretation of the graphs. Analyses were obtained by ICP-MS 0 ' I I I I /+ 3 1 30 60 90 120 240 Time of aqua regia attacwmin Fig. 2 Proportion of ( a ) Ni, (b) Cu, (c) As, (d) Mn, (e) Fe, and (j) Zn remaining in the residue after aqua regia leach for specified extraction times at room temperature of chromitites 1, CHR-A; 2, CHR-B; and 3, CHR-C. Cu in CHR-A and As in CHR-A and CHR-B were below the detection limits. Extraction data are normalized to intensity data obtained from unleached samples. Analyses were made by XRF776 100 4 ANALYST, AUGUST 1991, VOL. 116 1 4 3 could be the reduced activity of aqua regia on heating due to the loss of volatile nitrosyl chloride or chlorine components or, alternatively, the suppression effect due to dissolution of other elements mentioned. Results from the analysis of the residue after aqua regia attack for the elements Ni, Cu, As, Mn, Fe and Zn (by ED-XRF) and Au, As, Sb, Ir, Fe and Co (by INAA) are plotted in Figs.2 and 3, respectively. The true proportion of the element remaining in the residue is plotted as a function of contact time. The data indicate that a high (85-95%) proportion of Au, As and Sb in CHR-C is leached by aqua regia. For Cu, Ni and Co significant proportions are extracted that vary in magnitude when data from different chromitites are compared. Only a very small (if not negligible) proportion of Mn, Fe, Zn and Ir is extracted. As can be concluded from the data in Fig.1, the extraction reaction appears to be completed within 15-30 min. loo[ 50 50 t 1 2 50 t 0 4 u 30 60 90 120 240 Time of aqua regia attacklmin Fig. 3 Proportion of (a) Au, As and Sb in CHR-C; ( b ) Ir in 1, CHR-A and 3, CHR-C; (c) Fe in 1, CHRA, 2, CHR-B and 3, CHR-C; and (d) Co in 1, CHR-A, 2, CHR-B and 3, CHR-C remaining in the residue after aqua regia leach for specified extraction times at room temperature. Au, As and Sb in CHR-A and CHR-B and Ir in CHR-B were below the detection limits. Extraction data represent the true proportion of element extracted. Analyses were made by INAA Extraction Efficiency An estimate of the true proportion of PGE and Au extracted by aqua regia was made by comparing the concentrations of elements leached from chromitite OU-CX (a sub-sample of CHR-C), with the best estimates of results obtained by fire assay from five commercial laboratories.The results listed in Table 1 show the average fire assay concentration and the corresponding proportion measured by aqua regia leach-ICP- MS. They indicate that the aqua regia leach of Pd and Au is about 100%; Pt and Rh, 18%; Ir, 1%; and Ru and Os, uncertain. This uncertainty in the 0 s and Ru data arises from the variability and paucity of results for these elements obtained from commercial laboratories. Application to Other Samples In order to evaluate the more general application of this procedure, the aqua regia leach technique was applied to a suite of samples that had been independently analysed by nickel sulphide fire assay with ICP-MS detection by a commercial laboratory.This suite comprised ophiolitic rocks, including 14 chromitites, three pyroxenites and two wehrlites; the results for which are listed in Table 2. Further comparisons are made in Table 3 in which aqua regia leach data are compared with certifiedrecommended values for a range of PGE reference materials including: PTA-1 (Platiniferous Black Sand), PTC-1 (Sulphide Concentrate), PTM-1 (Ni-Cu Matte), SARM 7 (Platinum Ore), GXR-1 (Jasperoid Soil), and GXR-4 (Copper Mill-head) and SU-la (Ni-Cu-Co Ore). Reviewing these data reveals significant variability for a particular element over a range of different samples. However, some general observations on trends in these data can be made. (i) Although a few samples show high recoveries of Pt by aqua regia leach, the proportion extracted varies significantly from sample to sample and is usually less than 2040%.(ii) The proportion of acid-leachable Pd appears to be high, often near-quantitative, based on an over-all evalua- tion of the data listed in Table 2 (ophiolitic samples). The results of Pd determinations on reference samples are more variable, particularly low recoveries being obtained for PTC-1 and PTM-1. Very high recoveries (>> 100%) were measured from GXR-1 and GXR-4, although there is some uncertainty as to the reliability of the compiled values with which the measured data are compared. (iii) Significant proportions of Rh and Ru were recovered from samples, although in general the proportion is less than 30% of the expected composition.(iv) Although some 0 s is extracted, it is difficult to judge the over-all trend for this element owing in part to determinations falling below the detection limit of the technique used here and in part to the absence of 0 s data for reference samples. (v) The proportion of Ir extracted by aqua regia is low, usually significantly less than 10%. (vi) In general recoveries of Au are high and often near-quantitative, a notable exception being data for PTM-1. (vii) Over-all recoveries are very low for sulphur-rich samples and clearly the extraction technique used here is not suitable for such materials without extensive modification. It may be that this problem is associated with the oxidation of sulphide sulphur by aqua regia leading to a loss in the potency of the acid andor high base metal concentrations in solutions, leading to interferences in the ICP-MS measure- ments.Interpretation of Results Excluding analytical discrepancies, two factors, both related to sample composition, might influence the proportion of PGE extracted by aqua regia leach. The first is whether a proportion of the elements is distributed in the solid solution or is present as discrete mineral grains embedded within the resistant matrix phases (such as chromite) and are notANALYST, AUGUST 1991, VOL. 116 777 ~ ~ ~ ~ ~ ~ ~~~~~~ ~~~~~ ~ Table 1 Composition of mineralized chromitite (OU-CX) in ppm and % recovery of PGE by aqua regia leach of this sample. Bulk composition data are the best estimates derived from five commercial laboratories, obtained from a round robin analytical test organized by Rio Tinto Zinc.Aqua regia recovery is the proportion (%) of the element recovered in the filtrate after a 60 min aqua regia extraction on a 10 g sample at room temperature with measurements made by ICP-MS Sample Pt Pd Rh Ru 0 s Ir Au Bulk composition 40.0 53.0 3.7* - - 6.0* 2.7 Aqua regia recovery 18 104 18 uncertain? uncertain? 1 110 * Additional analytical uncertainty due to the variability of reported results. I Recovery efficiencies of Ru and 0 s are uncertain owing to the lack of data reported by commercial laboratories. Table 2 Comparison of the extraction efficiency of the aqua regiu leach (this work) with samples independently analysed by nickel sulphide fire assay-ICP-MS; all data listed in ppb (whole rock) Pt Pd Rh Ru 0 s Ir Au Sample Chromitite 1" Chromitite 2* Chromitite 39 Chromitite 4§ Chromitite 54 Chromitite 64 Chromitite 7§ Chromitite 8§ Chromitite 94 Chromitite 107 Chromitite 117 Chromitite 127 Chromitite 134 Chromitite 149 Pyroxenite 19 Pyroxenite 29 Pyroxenite 34 Wehrlite 19 Wehrlite 28 This Fire work assay 200 800 110 110 30 83 200 510 17 52 1390 4050 --$ 16 8 51 130 500 230 300 7290 40000 210 280 1140 870 7 9 120 110 29 44 65 100 --$ 48 22 22 This Fire work assay 730 1300 490 1670 36 66 300 340 50 83 3600 3150 --$ 11 --$ 18 450 520 51000 53000 226 220 2200 2100 8 10 --$ 4 40 62 180 220 22 18 37 48 --t - This Fire work assay 216 470 191 610 6 20 79 120 3 13 159 605 1 9 4 46 6 55 --t 662 3700 -1- 87 110 --$ 6 --$ 3 3 9.5 --$ 4 --$ 2.5 - - 1 5.5 This Fire work assay 85 960 76 1200 29 110 460 390 --$ 69 250 1150 --$ 45 --$ 310 35 190 -1- --t --t 290 190 74 44 --$ 26 --$ 4 12 13 --$ 19 --$ 5.5 - - - This Fire work assay --t -t --$ 42 9 88 --$ 34 --$ 675 --$ 24 --$ 160 6 50 -? --t --t 45 98 --$ 12 --$ 4 --$ 4 --4 6 --$ 2 --f 4 - - - - - This Fire work assay 30 380 16 --$ 9 57 18 250 5 50 29 1550 --$ 25 2 180 3 97 -? 70 12000 18 3000 17 120 --$ 14 --$ 4.5 --$ 2.5 4 5.5 --$ 5 --$ 4 - * Samples and analyses for fire assay from R.V. D. Robert, MINTEK, Randberg, South Africa. 7 Not analysed by NiS fire assay. -$ Below the detection limit. 4 Fire assay analyses by Sheen Analytical, Perth, Australia. fl in-house reference samples (Open University); best estimate data derived from five commercial laboratory analyses.This Fire work assay 30 70 13 50 -t --t --t --t -? --t 2 6 2 30 2960 2700 --t 155 130 --$ 2 16 8 --f 6 34 28 -? -4 2 - - - - - - - - Table 3 Comparison of aqua regia leach-ICP-MS data (this work) with expected compositions (given value) of selected reference samples; all data listed in ppb (whole rock) Sample Pt Pd Rh Ru 0 s Ir Au This Given This Given This Given This Given This Given This Given This Given work value work value work value work value work value work value work value Platinum Ore, Platiniferous Black Sulphide Concentrate, Ni-Cu Matte, Ni-Cu-Co Ore, Jasperoid Soil, Copper Mill-head, SARM7(PTO-l)* 1400 3740 1180 1530 174 240 115 430 9 63 19 74 302 310 Sand, PTA-1-t 3320 3050 16 --$ 5 --$ -3 --$ -$ --$ 15 -$ 363 --$ PTC- 17 210 3000 6010 12700 71 620 171 65011 49 2401) -4 17011 136 650 PTM- 1 * * 45 5800 5 8100 132 900 294 59011 31 14011 44 47011 -4 1800 SU-lat-t 59 410 265 370 16 8011 29 5611 -§ 1111 -4 2511 105 160(( GXR-l-$-$ -9 <lo 181 <0.1 12 --$ -§ --$ -5 --$ -§ --$ 2830 3300 GXR-4$$ -9 <10 55 0.2 2 --$ -5 --$ -9 --$ -9 --$ 618 470 * SARM 7: SABS certified values from Steele et al.29 t PTA-1: Canadian Certified Reference Material Project recommended values from McAdam et af.35 $ No reported data.9 Below the detection limit. 7 PTC-1: Canadian Certified Reference Material Project recommended values from McAdam et uf.36 11 Data are additional non-certified values. ** PTM-1: Canadian Certified Reference Material Project recommended values from McAdam et al.37 I-? SU-la: Canadian Certified Reference Material Project recommended values from Steger and B0wman.3~ $$ GXR-1 and GXR-4: compiled values from Gladney and R ~ e l a n d t s .~ ~778 ANALYST, AUGUST 1991, VOL. 116 Table 4 PGE mineralogy of a chromitite from the Cliff area, Unst, Shetland, summarizing aqua regia solubility data Conclusion The results obtained in this work indicate that high, and for ophiolitic rocks, near-quantitative recoveries of Pd and Au are achieved by a room temperature aqua regia leach of l o g samples using 20 ml of acid. Heating the aqua regia leach mixtures led to a reduction in the PGE signal in ICP-MS measurements from sample solutions. Significant but lower recoveries (usually 2040%) of Pt, Rh, Ru and 0 s were also observed but the recovery of Ir was much lower (generally 1-10%).Low recoveries of all the elements were observed from sulphur-rich samples. In assessing the analytical perfor- mance, account must be taken of other elements that are co-extracted (including As, Sb, Ni and Cu from chromitites), which might cause interference effects in analyte solutions. Comparison of the recovery efficiency data with a detailed PGE mineralogical description available for one mineralized chromitite sample indicated that the chemical solubility of individual PGE mineral types could be the dominant factor in dictating recovery efficiencies from ophiolitic samples. Dominant element Mineral* Pt Sperrylite Genkinite Hongshiite Alloy Alloy Stibiopalladinite Potarite Alloy Rh Hollingworthite Unidentified Unidentified Unidentified Rut henian Pd Mertieite Ir Irarsite Ru Laurite pentlandite 0 s Native metal Iridosmine Formula PtAs, (Pt,Pd),Sb3 PtCuAs P t-Pd-Cu Pt-Pd-Au-Cu (Pd,Cu)g(Sb,As)3 PdHg AuPd (Rh,Pt,Pd)AsS Rh(Sb,S) (Rh ,Ni)Sb (Ir,Ru,Rh,Pt)AsS I r (S b , S) (Ru,0s,Ir)S2 (Ni ,Fe, Ru)$& (Pd,Cu)dSb,Ash 0 s Os,Ir Solubility in aqua regia No - - - - - Yes - - - - - - - No - - No * Other minerals reported to be soluble in aqua regia: native gold, platinum and palladium. Minerals reported to be insoluble in aqua regia: cooperite (Pt,Pd,Ni)S; braggite (Pt,Pd,Ni)S; osmiridium (Ir,Os); and platiniridium (Ir,Pt).therefore effectively wetted by aqua regia. The second factor relates to the chemical solubility of individual PGE mineral species in aqua regia. In respect of chromitite CHR-C, the detailed mineralogical descriptions of Prichard et aZ.25 have characterized a variety of PGE mineral types found in this sample.These mineral types are listed within each element category in decreasing fre- quency of observation in Table 4. These data indicate, for example, that the dominant Pt minerals in CHR-C are sperrylite and genkinite, with hongshiite and alloy phases being observed less frequently. A literature survey has revealed some data on the solubility of specified mineral types in aqua regia40.41 and these are also listed in Table 4. Evaluation of the data indicates that there is some correla- tion between the recovery efficiencies listed in Table 1 and the aqua regia solubility data in Table 4. Thus, high recovery efficiencies of Pd appear to correlate directly with the dominant presence of this element in the mineral stibiopalladi- nite and mertieite.The former mineral is reported to be soluble in aqua regia, as might be the latter in view of the similarity in chemical composition. Conversely, the dominant mineral containing Pt in chromitite CHR-C is sperrylite, a mineral that is insoluble in aqua regia, an observation that is consistent with the much lower recovery efficiency of this element (18% in OU-CX). Finally, although information is not available for irarsite, it appears significant that other Ir bearing minerals including osmiridium, iridosmine, platiniri- dium and laurite are all insoluble in aqua regia, so accounting for the very low recovery (1%) of Ir by aqua regia leach of this chromitite. Although these observations relate only to one sample, and discrepancies due to non-wetting by aqua regia of PGE source minerals cannot be discounted, the results described here suggest that the solubility of individual PGE mineral types may be the limiting factor in the more general application of the aqua regia leach method for the determination of the PGEs.The authors are grateful to NERC, Alan Gray, Kym Jarvis (Surrey) and Yuk Cheung (BGS) for the use of the ICP-MS facilities and John Williams and Ed McCurdy for assistance in running the instrumentation, and to Hazel Prichard, Richard Lord, John Bridges of the Open University (OU), Chris Morrissey (Rio Tinto Zinc) and Rob RobCrt (MINTEK) for the loan of samples and comparative data. Support from the Minerals Industry Research Organisation (MIRO) is also gratefully acknowledged.1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 References Smith, E. A., The Sampling and Assay of the Precious Metals, Revised 2nd Ed., Met-Chem Research, Boulder, Colorado, 1987. Haffty, J., Riley, L. B., and Goss, W. D., U.S. Geol. Surv. Bull., 1977, No. 1445. Van Loon, J. C., Pure Appl. Chem., 1977,49, 1495. Bowditch, D. C., Aust. Min. Dev. Lab. Bull. (AMDEL), No. 15, 1973, p. 71. Hall, G. E. M., and Bonham-Carter, G. F., J . Geochem. Explor., 1988,30, 255. RobCrt, R. V. D., Van Wyk, E., and Palmer, R., Nut. Znst. Metall. Repub. S. Afr. Rep., 1971, No. 1371. Hoffman, E. L., Naldrett, A. J., VanLoon, J. C., Hancock, R. G. V., and Manson, A., Anal. Chim.Acta, 1978, 102, 157. Asif, M., and Parry, S. J., Analyst, 1989, 114, 1057. Beevers, J. R., Econ. Geol., 1967, 62, 426. Strong, M. B., and Murray-Smith, R., Talanta, 1974,21, 1253. Terashima, S., Geostand. Newsl., 1988, 12,57. Rivoldini, A., and Haile, T., At. Spectrosc., 1989, 10, 89. Latimer, W. M., and Hildebrand, J. H., Reference Book of Inorganic Chemistry, Revised edn., The Macmillan Company, New York, 1940, p. 206. Sen Gupta, J. G., Talanta, 1989, 36, 651. Trancoso, M. A., and Barros, J. S., Analyst, 1989, 114, 1053. Alvarado, J., and Petrola, A,, J. Anal. At. Spectrom., 1989, 4, 411. Grimaldi, F. S., and Schnepfe, M. M., U.S. Geol. Surv. prof. pap. 1967,575-C, C141. Grimaldi, F. S., and Schnepfe, M. M., U.S. Geol. Surv. prof. pap. 1968,600-B, B99. Palmer, I., Streichert, G., and Wilson, A., Nut.Znst. Metall. Repub. S. Afr. Rep., 1971, No. 1218. Palmer, I., Palmer, R., and Steele, T. W., J. S. Afr. Chem. Znst., 1972, 25, 190. Fryer, B. J., and Kerrich. R., At. Absorpt. Newsl., 1978, 17,4. Sighinolfi, G. P., Gorgoni, C., and Mohamed, A. H., Geostand. Newsl., 1984, 8, 25. Sen Gupta, J. G., and Gregoire, D. C., Geostand. Newsl., 1989, 13, 197. Kontas, E., Niskavaara, H., and Virtasalo, J., Geostand. Newsl., 1990, 14, 477.ANALYST, AUGUST 1991, VOL. 116 779 2.5 26 27 28 29 30 31 32 33 34 35 Prichard, H. M., Potts, P. J., Neary, C. R., Lord, R. A., and Ward, G. R., European Communities Commission, Luxem- bourg Rep., 1989, No. CD-NA-11631-EN-C. Tarkian, M., and Prichard, H. M., Miner. Deposita, 1987, 22, 178. Prichard, H. M., and Tarkian, M., Can. Mineral., 1988,26,979. Lenahan, W. C., and Murray-Smith, R. de L., Assay and Analytical Practice in the South African Mining Industry, The South African Institute of Mining and Metallurgy, Johannes- burg, 1986, p. 228. Steele. T. W.. Levin, J . , and Copelowitz, I . , Nat. Inst. Metall., Repub. S. Afr. Rep., 1975, No. 1696. Potts, P. J . , and Govindaraju, K., Geostand. Newsl., 1989, 13, 193. Jarvis, K. E., Chem. Geol., 1988, 68, 31. Gray, A. L., and Williams, J . G., J. Anal. At. Spectrom., 1987, 2, 599. Potts. P. J., Webb, P. C., and Watson, J . S . , X-ray Spectrom., 1984, 13, 2. Potts, P. J . . Thorpe, 0. W., and Watson, J . S . , Chem. Geol., 1981,34, 331. McAdam, R. C., Sutarno, and Moloughney, P. E., Dept. of Energy, Mines and Resources, Mines Branch, Ottawa Tech. Bull., 1971, No. TB138. 36 McAdam, R. C., Sutarno, and Moloughney, P. E.. Dept. of Energy, Mines and Resources, Mines Branch, Ottawa Tech. Bull., 1973, No. TB176. 37 McAdam, R. C., Sutarno, and Moloughney, P. E., Dept. of Energy, Mines and Resources, Mines Branch, Ottawa Tech. Bull., 1973, No. TB182. 38 Steger, H. F., and Bowman, W. S., Canadian Certified Reference Materials Project, CANMET Rep. , 1980, No. 80-9E. 39 Gladney, E. S . , and Roelandts, I . , Geostand. Newsl., 1990, 14, 21. 40 Schoeller, W. R., and Powell, A. R., The Analysis of Minerals and Ores of the Rarer Elements, 3rd edn., Charles Griffin, London, 1955, p. 32.5. Powell, A. R., in Comprehensive Analytical Chemistry, Volume I c, Classical Analysis: Gravimetric and Titrimetric Determina- tion of the Elements, eds. Wilson, C. L., and Wilson, D. W., Elsevier, Amsterdam, 1962, p. 702. 41 Paper 1 l00989C Received March 4th, 1991 Accepted April 5th, 1991

ISSN:0003-2654    

DOI:10.1039/AN9911600773

出版商:RSC    

年代:1991

数据来源: RSC

摘要:

ANALYST, AUGUST 1991, VOL. 116 781 Compensation for Particle Size Effects in Near Infrared Reflectance Christine R. Bull Silsoe Research Institute, Wrest Park, Silsoe, Bedford MK45 4HS, UK The interpretation of near infrared (NIR) reflectance in terms of material composition is complicated by particle size variation. An investigation is reported into the relationship between NIR reflectance and particle size using data from ground wheat samples. It is shown that the sensitivity of reflectance to particle size is dependent on the absorption properties of the material at the sampling wavelength. A theory relating a function of reflectance, known as the Kubelka-Munk function, to particle size suggests that the effects due to changing wheat composition can be isolated from the particle size effects using normalized Kubelka-Munk functions.Analysis of the spectral data from the ground wheat samples supports this theory. The ratio of Ku belka-Munk functions at appropriate absorption and reference wavelengths is subsequently manipulated to yield single term calibration equationsfor the protein and water in the wheat samples. Comments are made on the implication of the reported findings for various calibration techniques. Keywords: Near infrared reflection; particle size; Kubelka-Munk function In order to use near infrared (NIR) reflectance as an analytical tool it is necessary to relate the function of the sample reflectance at a given wavelength to the concentration of a sample constituent via a known calibration equation. For convenience this equation is usually chosen to be linear.Kubelka and Munkl developed a function, F(R), of the reflectance, R , which is linearly related to the absorption (and hence the concentration) of the sample. Their function is as follows: (1 - R)2 K F(R) = ~ -- 2R S - where K is the absorption coefficient and S is the scatter coefficient. However, most calibrations use the simpler log (1/R) signal treatment which is approximately proportional to the concen- tration of the absorbing material. For either mathematical treatment, the function value will depend not only on the concentration of the constituent to be measured but also on the physical properties of the sample. Consequently it is necessary to compensate for the interfering effects by measur- ing the reflectance at a number of absorption and reference wavelengths.In general, the calibration is then obtained from a linear combination of these terms. In some instances, the appropriate sampling wavelengths can be selected by looking at a series of reflectance scans from samples in which the component of interest is varied. The wavelength at which there is maximum variation is then identifiable as an absorp- tion wavelength whereas a wavelength at which there is minimal variation can be used as a reference. However, this simple selection procedure is often impossible because of interfering factors, such as particle size. This is illustrated in Fig. 1 which shows the Kubelka-Munk function at a water absorption band of 1944 nm as a function of the water content of ground wheat samples of varying particle size.When the wavelengths cannot be visually selected they are determined numerically using multiple linear regression. This 'black box' approach is not entirely satisfactory because the selection is not based on the absorption features and hence composition of the sample material. A technique that is used to emphasize the absorption features of a spectral scan is to look at the first and second differential of the log (1/R) plot. This will be sensitive to the change in shape of spectral scans taken from a series of samples. There are two disadvantages with this calibration technique. Firstly, one has to determine the reflectance of the sample over a segment of the spectrum. This makes the method inappropriate for simple analytical devices which determine the sample reflectance at a discrete number of fixed wavelengths.Secondly, the differences in the derivative scans are difficult to interpret. Norris and Williams2 compared these calibration techniques for samples of ground wheat with particular reference to the effects of varying particle size. They concluded that the most satisfactory mathematical treatment was to take second derivatives of the log (1/R) functions, although there was little difference in the performance of mathematical treatments provided equivalent numbers of wavelengths were used. Cowe and McNicoP used principal components analysis (PCA) to analyse log (UR) reflection scans of wheat flour samples. The principle of this analysis technique is that each of the multiple wavelength scans can be specified as a single point in the multi-dimensional space defined by the log (l/Ri) axes where i is the number of sampling wavelengths.The first component is a vector drawn through this space which accounts for the majority of the variation in the scans and will take the form where the coefficients C l j are the weights or loadings of the first component. The second component is the axis orthogonal to the first, along which there is maximum residual variation. Subsequent components are defined as the axes at right angles to all preceding principal components, which in turn exhibit the greatest amount of unexplained variation in the data. There are several advantages in using the PCA approach to analyse NIR data. Firstly, the spectra are reduced to a small number of computed values which can be easily used in regression modelling.Secondly, the relative loadings of each 0.9 E l + .- Y O.* t + + 11 12 13 14 15 Water content (%I Kubelka-Munk function at a water absorption band (1944 Fig. 1 nm) as a function of the water content of ground wheat samples782 ANALYST, AUGUST 1991, VOL. 116 of the principal components give some indication of the location and strength of the absorption bands and hence chemical composition of the samples. This paper shows how PCA can be developed to isolate the changes in reflectance due to compositional change from those due to particle size effects. The influence of particle size on reflectance is discussed and it is shown that the displacement in the log 1/R spectra due to particle size is greater at absorption bands.A theory relating the Kubelka-Munk function to particle size is introduced and is used to show how the variations in normalized Kubelka-Munk functions can be used to highlight areas in which changes in sample composi- tion give rise to variation in reflectance. This is then used to determine suitable absorption and reference wavelengths for use in a calibration based on the ratio of two Kubelka-Munk functions. A discussion on the influence of particle size variations on various calibration techniques is given in the final section. Experimental Thirty-nine samples of wheat flour were scanned at the Flour Milling and Baking Research Association (FMBRA), Chor- leywood, Hertfordshire, UK, with a Neotec 6350 analyser.Spectra were recorded as log (UR) at 2 nm intervals from 1100 to 2500 nm. In order to reduce the effect of electrical noise, the spectrum for each sample was calculated as the average of 50 scans and was smoothed with the use of a five point moving average algorithm. To eliminate edge effects, data points below 1120 and above 2480 nm were discarded. For the purpose of this paper a reduced data set of 171 data points, evenly spaced between 1120 and 2472 nm, which were taken from the original data, was used. Reference analytical values for protein and moisture were also supplied by FMBRA. These samples have previously been used by Osborne4 in investigations into the use of NIR as a tool for assessing potential for bread making and by Cowe and McNicoP in the PCA described earlier.The data presented by Norris and Williams2 have also been used in the present analysis. Results Effect of Particle Size on Near Infrared Reflectance The analysis of the wheat data by PCA (Cowe and McNicol) showed that 98.6% of the variation in the log (1/R) spectral scans could be accounted for by the first component. The loadings Cli [eqn. (2)] for each of the wavelengths of the first component are illustrated in Fig. 2. This profile closely resembles the original log 1/R spectra. The weights show a general trend to increase with increasing wavelength but are peaked at wavelengths usually associated with the water 1000 1500 2000 2500 Wavelengthhm Fig. 2 Loadings of the first component from the principal component analysis of the reciprocal log reflectance spectra for ground wheat samples absorption bands (approximately 1200, 1450 and 1944 nm).However, the analysis showed that there was little correlation (0.16) between the water content of the wheat and the first component, which led Cowe and McNicol to conclude that there was bound water in the sample. In practice, as will be shown below, the increased variation at the absorption bands is consistent with the particle size effects. The total reflection for a particulate material consists of reflections of the incident light at each of the refractive index discontinuities, particularly at the air-particle interfaces. In addition, there will be some absorption of the radiation within each particle which will reduce the radiation penetrating the sample and hence reduce the back-reflection from subsequent interfaces. Consequently, for two samples of the same composition ground to different particle sizes, there is more back-reflection from the finer particles because the radiation encounters more discontinuities per unit length of the sample material traversed.This effect becomes more pronounced when the light is absorbed more strongly. Consequently, there is a stronger relationship between back-reflection and particle size for wavelengths that are more strongly absorbed than those that are only weakly absorbed. If the majority of the variation in the wheat samples is due to particle size differences, the loadings of the first principal component, illustrated in Fig. 2, will be strongly correlated with the sensitivity of the log (UR) measurements to particle size.A recent paper5 related the reflectance from an infinitely thick powdered sample to the reflectance and absorption of the average particulate. By expressing the average properties of the individual particle in terms of the mean particle size it was shown that the relationship between the Kubelka-Munk function, F(Rk), and the mean particle size, x , is (3) X F(R1) = - 2dkrx where dk, the l/e penetration depth, is the depth at which the radiation into a ‘solid block’ of the material has reduced by a factor l/e (e = 2.718) and rk is the reflectance at a single air-particle discontinuity. The assumptions made in the derivation of this equation are: that the dimensions of the particle size are greater than half the coherence length of the radiation within the sample; that the absorption within each particle is small; and that the ratio between particle, absorp- tion and reflectance is not large.It has been shown5 that these approximations are applicable to milled grain. Manipulation of eqns. (3) and (1) enables us to explore the relationship between the log (1/R) function and the particle size (in units of dirk). This is illustrated for a series of samples of varying concentration in Fig. 3. Over a small range of Kubelka-Munk values there is a near linear relationship between log (UR) and mean particle size. This was observed by Norris and Wil- hams2 They presented data for spectral scans of ground wheat 0.5 0.334 s c 0) -I 0.167 1 I I 0 0.25 0.50 0.75 Kubelka-Munk function (particle size in units of 2dhrh) Fig.3 Theoretical relationship between the reciprocal log reflec- tance, log 1/R, and particle size for samples with different exponential penetration depths, dANALYST, AUGUST 1991, VOL. 116 Table 1 Theoretically derived values of reciprocal log reflectances (log l/R), at a number of wavelengths, for grain particles of various milled sizes; the gradient of the log 1/R versus mean particle size (MPS) response graph, and the loadings of the first principal component of the log 1/R data set 783 Log 1/R for various MPS Gradient of Loading of CLm 150 pm 200 pm 250 pm 300 pm MPS/10-4 pm-l component Wavelength/ log 1/R versus first principal 5 - 4 - 3 - 2 - 1.200 1.300 1.400 1 so0 1.600 1.700 1.800 1.950 2.000 2.100 2.200 2.300 2.400 01.2 '1.3 1 - 0.0606 0.0519 0.0866 0.1265 0.1132 0.1085 0.1065 0.1870 0.1540 0.1904 0.1598 0.1971 0.1998 0.0699 0.0600 0.0999 0.1459 0.1306 0.1252 0.1229 0.2154 0.1775 0.2193 0.1842 0.2269 0.2300 0.0782 0.0670 0.1116 0.1629 0.1458 0.1399 0.1373 0.2402 0.1981 0.2446 0.2056 0.2350 0.2564 0.0856 0.0734 0.1222 0.1783 0.1596 0.1531 0.1503 0.2620 0.2167 0.2672 0.2248 0.2764 0.2801 1.666 1.430 2.370 3.448 3.088 2.970 2.916 4.996 4.174 5.114 4.328 5.280 5.346 0.01841 0.0 1253 0.03809 0.06569 0.05259 0.05239 0.05240 0.11544 0.09664 0.09825 0.09265 0.10808 0.11095 61 1 01.5 1.8 01.4 2.3 2-4 0 2.1 1% 2.2 a0 2.0 I I I I I I I 0.02 0.04 0.06 0.08 0.10 0.12 Loadings of first principal component Fig.4 Plot of the gradient of the reciprocal log reflectance, log 1/R, versus mean particle size (MPS) relationship plotted against the first component from the principal component analysis of the log 1/R spectra for ground wheat samples.The sampling wavelength in micrometres is indicated beside each point samples with mean particle size varying from 150 to 335 pm. The reflectance of the sample with the smallest mean particle size has been ascertained at a number of sampling wavelengths, taken at 100 nm intervals from 1.2 to 2.4 pm. The reflectance value at 1.9 pm was unclear from their data so a closely adjacent value at 1.95 pm has been taken. These reflectance values were substituted into eqn. (1) to calculate the Kubelka-Munk functions, which were then placed into eqn. (3) in order to obtain values for the dkrh product at each of the sampling wavelengths.Once the dhrh product is known it is possible to calculate the Kubelka-Munk and log (l/R) functions for a range of particle sizes from 150 to 300pm (Table 1). This approach has been used previously5 on ground wheat samples and it has been shown that the theoretically predicted sensitivity of the reflectance to particle size is in close agreement with the experimentally determined values (Norris and Williams) at a number of wavelengths. Table 1 shows that there is a tendency for the reflectance from a sample to be more sensitive to particle size effects at longer wavelengths. However, the increase is not uniform but is peaked at wavelengths at which there is high sample absorp- tion. For example, the gradient of the response at 1950 nm is greater than that at 2200 nm.Fig. 4 shows the calculated gradients of the log (l/R) versus particle size graphs plotted against the appropriate loadings Cli of the first principal component. The correlation is good (0.986) bearing in mind that the wheat samples used to determine the sensitivity of the log (URJ function to particle size are unrelated to those used to determine the loadings. This linearity indicates that most of the structure shown in the first principal component loadings, and hence most of the variation in the spectra of the 39 samples, is due to differences in the sensitivity of each wavelength to particle size variations. The importance of this observation is that spectral scans of two powdered samples which are markedly different from each other, especially at the absorption bands, may have the same chemical composi- tion with the differences arising solely from particle size effects.This explains why the first principal component from the analysis of the log (l/R) spectra shows little correlation (0.16) with water (Cowe and McNicol3) despite peaks in the loadings at the water absorption bands. The ability of eqn. (3) to explain most of the spectral variations of the 39 samples in terms of particle size effects is a strong indication of the validity of the theory applied to ground wheat. The deviations from linearity of the points in Fig. 4 is due either to limitation in the theory or to interference effects resulting from changes in the sample composition. However, variations in sample composition will only affect the loading of the first component if they are positively correlated with changes in the particle size, otherwise these variations will appear in the loadings of the higher order (and orthogonal) principal components.To isolate the changes in the loading of the first principal component that are not due to particle size variation we can look at the variation in the spectra given by the logarithm to the base ten of the Kubelka-Munk function, log F(Rh), at each wavelength A. Eqn. (3) states that this wavelength dependent function is equally sensitive to changes in the sample particle size at all wavelengths. Consequently, if the major variation in the sample reflectance is due to particle size effects and these effects are not correlated to any other sample variable, one would expect the loadings of the first principal component, which essentially show the sensitivity of each wavelength to particle size, to be uniform over the whole wavelength range.If, however, the protein content is correlated to particle size, features of the protein absorption bands would be super- imposed on the flat topped particle size response. The loadings of the first principal component of the log F(R1) scans are shown in Fig. 5. Overall, the response is much flatter than that obtained for the log (UR) data but there are features corresponding to protein at 1200, 1500, 1700, 2200 and 2370 nm (Mohsenin6) which implies that the protein content of the samples is correlated to the particle size. This supposition is supported by the relatively strong correlation between the first principal component and the protein content of the wheat samples (0.71) (Cowe and McNicol). This observation high- lights a limitation of the PCA technique, i.e., if the sample set is not defined relative to an appropriate set of axes then the effects of one source of sample variation may mask another in the loadings of the principal axes.Having shown that approximately 98.0% of the variability in the spectral scans of the wheat samples (the variance accounted for by the first principal component) is due to784 ANALYST, AUGUST 1991, VOL. 116 E E 0.10 00 h +, 0.05 r? b- I I 1 1000 1500 2000 2500 Wavelengthlnm Fig. 5 Loadings of the first component from the principal component analysis of log Kubelka-Munk spectra for ground wheat samples 1000 1500 2000 2500 Wavelengthlnm Fig.6 Loadings of the first component from the principal component analysis of the normalized Kubelka-Munk spectra for ground wheat samples particle size it is clear that in order to use NIR as an analytical tool, it is imperative to compensate effectively for particle size effects. Compensation for Particle Size Effects In order to compensate effectively for particle size effects, one needs to determine some function of the sample reflectance at one or more wavelengths, which is correlated to the analyte to be measured but which is independent of the physical properties of the sample. Eqn. (3) states that the Kubelka- Munk function is directly proportional to the particle size, x . Consequently, the ratio of two Kubelka-Munk functions should be independent of particle size. Furthermore, if the Kubelka-Munk functions are taken at appropriate absorp- tion, F(RA) and reference F(RR), wavelengths, it should be possible to obtain a calibration, F , given by (4) which is particularly sensitive to a given constituent, where RA and RR are the reflection of the sample at the absorption and reference wavelengths, respectively.The problem now lies in the selection of the appropriate wavelengths because the correlation between the Kubelka- Munk function at an absorption band and the concentration of the absorbing constituent is masked by particle size effects (Fig. 1). This difficulty can be overcome by looking at the normalized Kubelka-Munk functions. The mean particle size of a sample affects the Kubelka-Munk functions at each of the sampling wavelengths by the same multiplicative factor [eqn.(3)] so the normalized Kubelka-Munk function, defined by + * 0.011 1 1 12 13 14 15 Water content (%) Fig. 7 Normalized Kubelka-Munk function at a water absorption band (1944 nm) as a function of the water content of ground wheat samples is independent of particle size where h is one of the sampling wavelengths. Furthermore, this normalized function will only vary strongly at wavelengths corresponding to the absorption bands of the varying constituents. The identification of the wavelength at which the variation is maximum can be achieved by looking at the loadings of the first principal component of the normalized Kubelka-Munk data set.These loadings will be peaked at wavelengths which are particularly sensitive to compositional changes. The loadings of the first principal component for the grain data set are shown in Fig. 6. Identifying which of these peaks is sensitive to a particular varying constituent is not a trivial matter and absorption bands associated with various constituents will often overlap. However, the peaks in Fig. 6 at 1450 and 1944 nm have been identified in the literature as water absorption bands (Soc- rates’) and those at 2200 and 2370 nm are associated with protein (Mohsenin6). Clearly, this approach is more successful for isolating the effects of the various chemical absorbers of a material than using PCA on log (1/R) data as the majority of the particle size effects have been removed. The normalized Kubelka-Munk function at 1944 nm is shown plotted against the moisture content of the samples in Fig.7. The correlation coefficient of the graph is 0.972 which corresponds to a standard error in calibration of 0.22%. In comparison with Fig. 1, the normalization procedure has largely corrected for particle size effects. This again gives support to the assumption [eqn. (3)J that the Kubelka-Munk function is proportional to the mean particle size of the sample. Basing a calibration on a normalized Kubelka-Munk function of this sort limits the technique to instruments that can determine the reflectance over a scan of wavelengths. This renders the method inappropriate for instruments that measure the reflectance at a number of fixed wavelengths.In this instance, it is necessary to use a Kubelka-Munk ratio of the form given in eqn. (4). Fig. 7 shows that 1944 nm is a suitable wavelength value in the numerator function F(RA). A suitable reference function, F(R,), can be determined by maximizing the correlation between the ratio F(RA)/F(RR) and the moisture content for the reference wavelength. This was found to be at 2064 nm which corresponds to a wavelength value that is comparatively insensitive to compositional changes in the sample (Fig. 6) although its selection will also have been optimized to compensate for the influence of other variable constituents on the 1944 nm absorption peak. The calibration ratio is shown plotted against the moisture content of the wheat samples in Fig.8. The correlation coefficient of this graph is 0.989 which corresponds to a standard error of 0.12%. The same approach can be used for protein. From Fig. 6 and the literature6 it is assumed that the Kubelka-Munk function which is most sensitive to protein variation is at 2200 nm. This function has therefore to be used as the numerator in a ratio ofANALYST, AUGUST 1991, VOL. 116 785 1.5 i 1 c 0 C .- +- 0 a 1.2 1 I I I I 11 12 13 14 15 Ratio of the Kubelka-Munk functions at 1944 and 2064 nm as Water content (%) Fig. 8 a function of the water content of ground wheat samples 0.81 r I .- 0.80 1 .- 0.75 1 U 0.74 L . . 6 8 10 12 14 Protein content (%) Fig. 9 a function of the protein content of ground wheat samples Ratio of the Kubelka-Munk functions at 2200 and 2120 nm as the form given in eqn.(4). The best denominator function F(RR) determined as before has been found to be at 2120 nm. The relationship between the ratio of these two Kubelka- Munk functions and the protein content is shown in Fig. 9. The correlation coefficient of this graph is 0.971 which corresponds to a standard error in calibration of 0.42%. Clearly, as one would expect from the theory, the ratio of two Kubelka-Munk functions compensates for most particle size effects and it is possible to obtain a reasonably accurate single term calibration for the water and protein content of the samples. Unfortunately, however, the accuracy and stability of such a calibration depends not only on the effective compensation for particle size effects but also on the optimum selection of the sampling wavelengths.Absorption features of various constituents often overlap which means that a particular absorption peak cannot be uniquely identified with one absorber and it is necessary to compensate for the effects of interfering constituents. This often requires more than two sampling wavelengths. However, the basic principle of using a number of terms based on the division of Kubelka-Munk functions (or appropriate summations of Kubelka-Munk functions) to compensate for particle size effects is still valid. The ability of the theory to relate the majority of the variation in the 39 wheat spectra to the sensitivity of each wavelength to particle size effects and to compensate effectively for particle size effects using the normalized Kubelka-Munk function and Kubelka-Munk ratio gives some confidence in the validity of eqn.(3). This understanding of the relationship between reflectance and particle size has some implications on the applicability of various calibration algorithms which are discussed in the following section. Influence of Particle Size Variations on Various Calibration Techniques The simplest procedure that corrects for some of the physical variability in the samples is to take the ratio of reflectance at an absorption and reference wavelength. This ratio can then be linearly related to absorber concentration (Stafford et ~ 1 . 8 ) . It was shown above that this is a fairly crude correction because the relationship between particle size and reflectance is wavelength dependent.However, if the reference and absorption wavelengths are chosen to be suitably close together, their sensitivity to particle size will be sufficiently similar to derive a reasonable calibration (Bull9). A calibration in which the measured constituent is related to a linear combination of the reflectance or log (UR) functions at paired wavelengths is well established and, over a small range of mean particle sizes, it can compensate for particle size effects as well as (and sometimes better than) the Kubelka- Munk ratio (Norris and Williams*). The success of these calibrations can be explained by making reference to Fig. 3 which shows how the log (1/R) function at the reference and absorption wavelength will vary as a function of particle size. Over a small particle size range, the relationship between log (1/R) and particle size is approximately linear.Consequently, if the log (1/R) functions at the absorption and reference wavelengths are multiplied by appropriate calibration con- stants they will have similar sensitivity to particle size over a range of absorber concentrations. Similar reasoning can be applied to the calibration based on a linear combination of the reflectance at two wavelengths because, over a small range, the reflectance will vary approximately linearly with particle size. The difference between the functions will therefore be largely independent of particle size. However, the gradient and intercept of the log (1/R) versus particle size response graph, at the reference and absorption wavelength, will depend on the range of particle sizes covered (Fig.3). Consequently, one would expect a calibration based on the linear combination of log (l/R) or reflection functions to be dependent on the range of mean particle sizes in the calibration set and for the calibration accuracy to deteriorate when used on a sample set with dissimilar particle size distributions. This was observed by Norris and Williams2 who showed that the standard error of prediction of protein, in ground wheat samples, for a calibration using three log (1/R) terms, varied between 0.2 and 0.4% with a variation in mean particle size of 50 ym in the test data sets. This shows the importance of ensuring that the test samples are milled in the same way as the calibration set. It will also be noted that, for wheat samples, it is necessary for the calibration set to have similar protein contents as the test set because of the correlation between particle size and protein content.A further disadvantage of these calibration techniques is that the measured quantities do not give a feel for the relative absorber concentrations in the samples prior to calibration. It is anticipated, from the theory presented above, that a calibration based on the ratio of paired Kubelka-Munk functions will be independent of the mean particle sizes of the test and calibration set because the particle size dependence of the Kubelka-Munk functions cancel each other out. This function will therefore be the basis of a calibration which is more robust with respect to changes in the average particle size of the sample sets.Conclusions A theory which relates the effects of particle size to the reflectance of ground wheat flour has been used to show that the majority of the variation in the spectra from the ground wheat samples is due to particle size effects. The effects of particle size changes are particularly pronounced at the absorption bands of the sample which masks the effects of changing sample composition. A technique which highlights the changes in NIR reflectance due to compositional changes is presented. This paper shows that the normalized Kubelka-Munk function and the ratio of two Kubelka-Munk functions is786 ANALYST, AUGUST 1991, VOL. 116 largely independent of particle size. Consequently, by careful selection of the denominator and numerator functions, it is possible to obtain a function that is linearly related to the abundance of a given sample constituent. A linear calibration based on a Kubelka-Munk ratio has been determined for water and protein. A discussion on the influence of particle size variations on various calibration techniques draws a number of conclusions. Firstly, the ratio of reflectance at two wavelengths will always be particle size dependent because the relationship between particle size and reflectance is wavelength dependent. Secondly, a calibration based on a linear combination of reflectance functions will only be able to compensate for particle size effects over a limited range of particle sizes. It is argued that a calibration based on a combination of Kubelka- Munk ratios will be insensitive to changes in the mean particle size of the test and calibration set. The author thanks Dr. B. Osborne of the FMBRA for providing the reflectance scans and analytical values for the samples, and Dr. J. McNicol of the Scottish Agricultural Statistics Service for his helpful advice. References 1 Kubelka, P., and Munk, F., 2. Tech. Phys., 1931,12, 593. 2 Norris, K. H., and Williams, P. C., Cereal Chem. , 1984,61,159. 3 Cowe, I. A., and McNicol, J. W., Appl. Spectrosc., 1985, 39, 257. 4 Osborne, B. G., J. Sci. Food Agric., 1984,35, 106. 5 Bull, C. R., J. Mod. Opt., 1990, 37, 1955. 6 Mohsenin, N. N. , Electromagnetic Radiation Properties of Food and Agricultural Products, Gorden and Breach, London, 1984, 7 Socrates, G. , Infrared Characteristic Group Frequencies, Wiley, Chichester, 1980, p. 137. 8 Stafford, J. V., Weaving, G. S., and Bull, C. R., J. Agric. Eng. Res., 1989, 43, 57. 9 Bull, C. R., J . Agric. Eng. Res., 1991, 49, 113. p. 300. Paper 1100941 I Received February 2nd, 1991 Accepted April 18th, I991

ISSN:0003-2654    

DOI:10.1039/AN9911600781

出版商:RSC    

年代:1991

数据来源: RSC

摘要:

ANALYST, AUGUST 1991, VOL. 116 787 Immunocomplex-immobilization Technique Sven Oscarsson Biochemical Separation Centre, Uppsala Biomedical Center, Uppsala University, Box 577, 751 23 Uppsala, Sweden Jan Carlsson Pharmacia Diagnostics AB, 75 1 82 Uppsala, Sweden An immunoassay technique is presented, which works equally well with either a radiaoctive isotope or an enzyme as the label. It is based on the competition between a labelled antigen of known concentration and an unlabelled antigen (concentration to be determined) for binding to an antibody, which previously has been thiolated by the heterobifunctional reagent N-[3-(2-pyridyldithio)propionyl]succinamide [marketed as N-succinimidyl-3-(2-pyridyldithio)propionate; SPDP)]. The soluble immunocomplex formed is then immobi- lized onto agarose beads containing reactive disulphide groups.After washing, to remove any unbound and non-specifically adsorbed material, the radioactivity or the enzymic activity on the agarose beads is determined and the concentration of the antigen in the test sample calculated according to the conventional procedure for a competitive immunoassay. In order to demonstrate the applicability of the technique a test procedure for the determination of total serum immunoglobulin E is presented. Keywords: lmmunoassay technique; immunocomplex immobilization; radioimmunoassay; enzyme immunoassay A large number of immunoassay methods based on the biospecific reaction of an antibody with its corresponding antigen have been developed. With these methods it has been possible to detect low concentrations of many types of antigens and antibodies in complex media such as plasma and urine.One of the most widely used immunochemical techniques is the solid phase method presented by Wide.' In this method one of the immuno-components is bound to a solid phase, e . g . , polysaccharide beads, which can be separated from the liquid phase when the immunochemical reaction has reached equi- librium. One disadvantage of this method is the slow rate of transport of the immuno-components from the solution onto the immuno-components in the solid phase owing to poor convection and steric hindrance in the solid phase, thus leading to long reaction times. A system in which the immunocomplex is formed in solution rather than on the solid phase is the double antibody solid phase (DASP) system described by van Weemen and Schuurs.2 In this system the soluble immunocomplex formed is subsequently immobilized via a second antibody, covalently coupled to the solid phase; the second antibody is directed against the first.Unlike conventional competitive immunoassay , the DASP system allows the use of a large excess of immobilized antibody. This partly compensates for the slow rate of migration of the soluble immunocomplex in the solid phase and affords relatively rapid formation of the immobilized immunocomplex. For the DASP system to work properly however, a second antibody, specific for the primary antibody must be available. Non-specific reactions can otherwise occur between the antibody on the solid phase and, for instance, immunoglobulin (Ig) G from the sample or labelled analyte.This leads firstly, to a decreased capacity of the solid phase to bind the soluble immunocomplex, and secondly, to a high background. In order to make the second antibody specific, the first antibody used for the immunization step must be pure, however, an additional immuno-adsorption step might also be necessary. In the proposed technique a competitive immunochemical reaction occurring in solution is performed first, as in the DASP method, but in contrast to DASP, the second step is not accomplished by a subsequent immunochemical reaction but rather through a chemical reaction involving the thiol groups in the first antibody (introduced prior to the immunochemical reaction) and the reactive disulphide groups on the solid phase.Experimental Materials Sephadex G-25M, Activated Thiol-Sepharose 4B, Sephacryl S-300, N-[3-(2-pyridyldithio)propionyl]succinamide [mar- keted as N-succinimidyl-3-(2-pyridyldithio)propionate; SPDP] and Dextran T-70 were obtained from Pharmacia Fine Chemicals (Uppsala, Sweden). Anti-IgE (anti-DE1) of IgG- class (obtained from immunized-rabbit serum) purified by immuno-adsorption; 1251 labelled IgE (100 ng -1.6 @); 1251 labelled anti-IgE (0.4 pg - 5 pCi) labelled by the chloramine- T method; human IgE sera from patient Nils Danielson (ND); and components from the Prist IgE kit, namely, paper discs with covalently coupled IgE and the horse serum diluent, were kindly supplied from Pharmacia Diagnostics (Uppsala, Sweden). The polystyrene test-tubes were from Kebo AB (Stock- holm, Sweden).The ethanol (99.5%) used to dissolve the SPDP was purchased from Svensk Sprit AB (Stockholm, Sweden). The buffer salts: NaH2P04-2H20; KH2P04; and disodium dihydrogen ethylenediaminetetraacetate (EDTA) (C10H14N2Na208-2H20), were all purchased from Merck (Darmstadt , Germany) and were of analytical-reagent grade. The NgCl2.6H2O, NaN3 and Na2C03 used were also pur- chased from Merck. p-Galactosidase (G-5635 VIII), glutathione, dithiothreitol (DTT), human serum albumin (HSA) and o-nitrophenylp-D- galactopyranoside were purchased from Sigma (St. Louis, MO, USA), NaCl and Tween 20 from Kebo AB and methanol from May and Baker (Rhone Poulenc, Dagenham, Essex, UK) . Di-2-pyridyl disulphide (2,2'-dithiopyridine; 2-PDS), iodoacetamide and 2-mercaptoethanol were purchased from Fluka (Buchs, Switzerland).Methods Thiolation of IgG For the thiolation of IgG, the heterobifunctional reagent SPDP was used. The IgG (2.6 x 10-5 mol dm-3,4.3 mg ml-1)788 ANALYST, AUGUST 1991, VOL. 116 was dissolved in 0.1 mol dm-3 sodium phosphate buffer, pH 7.5, and the SPDP in the ethanol (to give a 1-8 mmol dm-3 concentration range). A 30 p1 aliquot of SPDP solution (3 mmol dm-3) was rapidly added to 0.3 ml of the protein solution (2.6 x 10-5 mol dm-3) with stirring. The final ethanol concentration should be less than 10% v/v in order to avoid precipitation of the protein. The reaction was carried out for 30 rnin at room temperature (21-25 "C). Any excess of reagent and reaction products of low relative molecular mass (M,) were removed by gel filtration on a Sephadex G-25 column (16 x 52 mm), which had been equilibrated with 0.1 mol dm-3 sodium phosphate buffer, pH 6.0, containing 0.15 mol dm-3 NaCl and 5 mmol dm-3 EDTA.The void fractions, containing the modified antibody (with the 2-pyridyl disulphide groups) were collected and pooled. To the pool was added DTT dissolved in 0.1 mol dm-3 sodium phosphate buffer, pH 6.0, containing 0.15 mol dm-3 NaCl and 5 mmol dm-3 EDTA (5 mmol dm-3 final concentration) and the reduction was allowed to proceed for 15 rnin at room temperature. Excess of reducing agent and 2-thiopyridone were removed by passing the reaction mixture through a Sephadex G-25 column. The number of thiol groups introduced into the antibody was estimated by titration with 2-PDS as described by Grazetti and Murray.3 Derivatives with 3-6 thiol groups per IgG molecule were prepared.The concentration of SPDP was estimated by a previously described method.4 Preparation of freeze-dried thiolated IgG To 1 ml of thiolated IgG (1 X 10-6 mol dm-3,0.16 mg ml-I), prepared as described above, 9 ml of 0.1 mol dm-3 sodium phosphate buffer (pH 6.0) containing 0.1% Tween 20,0.2% HSA, 2% Dextran T-70, 5 mmol dm-3 EDTA and 1 x 10-6 mol dm-3 glutathione, was added. The solution obtained was transferred into a glass bottle and freeze-dried. The lyophi- lized material was stored at +4 "C until used. Solid phase Activated Thiol-Sepharose 4B has previously been used as the solid phase.5 It has a content of 1 pmol of reactive disulphide groups per millilitre of swollen gel. The agarose derivative is delivered as freeze-dried powder and, therefore, had to be re-swollen before use.This was done by suspending the powder in distilled water or in 0.1 mol dm-3 sodium phosphate, pH 7.0, on a glass filter-funnel. The gel thus formed was washed with 0.1 mol dm-3 sodium phosphate buffer containing 5 mmol dm-3 EDTA, pH 7.5, to remove the lyophilizing additives. After washing, the gel was dried on the glass filter-funnel by use of a vacuum for 5 min in order to eliminate interstitial liquid. A 250 mg portion of the dried gel was then suspended in 1 ml of 0.1 mol dm-3 sodium phosphate buffer containing 5 mmol dm-3 EDTA, pH 7.5. This suspension was used in the test procedure. Preliminary experiments have also been performed with agarose of small particle size (2-4 pm) with a substitution degree of 377 pmol g-1 of dried product (kindly supplied by Pharmacia Fine Chemicals).Preparation of the IgE-/3-galactosidase conjugate P-Galactosidase (50 mg) was dissolved in 2 ml of 0.02 mol dm-3 sodium phosphate buffer, pH 7.5, containing 0.1 rnol dm-3 mercaptoethanol. After reaction for 60 rnin at room temperature, the enzyme solution was passed through a Sephacryl S-300 gel-filtration column. The column was previously equilibrated with 0.1 mol dm-3 sodium phosphate buffer, pH 7.0, containing 2 mmol dm-3 and 0.02% NaN3. After separation, the fractions that showed 6-galactosidase activity were pooled and the enzyme solution was concentrated using an Amicon cell (Amicon, Danvers, MA) to a volume of 2 ml.The concentrated enzyme solution was passed through a Sephadex G-25 column equilibrated with 0.1 mol dm-3 sodium phosphate buffer containing 2 mmol dm-3 M g Q , pH 8.0. The thiol content of the P-galactosidase preparation obtained, estimated by titration with 2-PDS73 was 10 mol of SH-groups per mol of enzyme. A 0.2 ml aliquot of human IgE(ND) (10 mg ml-1) dissolved in 0.1 mol dm-3 sodium phosphate buffer, pH 7.5, was mixed with 8 pl of SPDP (1.25 mmol dm-3, dissolved in ethanol). The reaction was carried out for 30 min at room temperature. The excess of reagent, and reaction products of low M , were removed by gel filtration on a Sephadex G-25 column (16 X 52 mm) equilibrated with 0.1 mol dm-3 sodium phosphate buffer, pH 6.0.The void fractions containing the modified IgE (substituted with 2-pyridyl disulphide groups) were collected and pooled. The degree of substitution was found to be 0.66 reactive disulphide structures per IgE molecule, as estimated photometrically by reaction with DTT.4 A 1.4 ml aliquot of modified IgE (3.48 pmol dm-3) was mixed with 0.3 ml of 6-galactosidase (prepared as described above) (8.04 pmol dm-3). The reaction was allowed to proceed for 5-6 d at 4 "C. A 100 mol dm-3 excess of iodoacetamide was added to the solution (in order to block any unreacted sulphydryl groups). The excess of reagent was removed by gel filtration on a Sephadex G-25 column (16 x 42 mm) that had been equilibrated with 0.1 mol dm-3 sodium phosphate buffer, pH 6.7. A 1.7 ml aliquot of 0.3 mol dm-3 sodium phosphate buffer (pH 6.7) containing 0.04% NaN3, 0.2% HSA and 0.2% Tween 20 was then added to the reaction mixture.This preparation of IgE-P-galactosidase conjugate was stable for at least one month at 4 "C. Application of the immunocomplex immobilization technique to the determination of IgE using a radioisotope (1251) as the label Thiolated anti-IgE (either lyophilized and reconstituted with 10 ml of distilled water or freshly prepared as described above) was diluted to a final concentration of 1 x 10-9 mol dm-3 (0.16 pg ml-1) with 0.1 mol dm-3 sodium phosphate buffer, pH 7.5, containing 0.1% Tween 20 and 5 mol dm-3 EDTA. A 50 pl aliquot of this solution and 50 pl of patient serum were added to polystyrene test-tubes. After 30 min 50 pl of IgE labelled with 1251 and having an IgE concentration of 1.4 x 10-10 mol dm-3 (27 ng ml-1) were added to the test-tubes and the reaction was carried out for 90 rnin at room temperature (21-25 "C).A 100 p1 portion of a suspension of Activated Thiol-Sepharose 4B (corresponding to 25 mg of dried gel per tube), was then added to each test-tube and the mixture agitated for 60 min on an Ika-Wibrax-WXR shaker (Janke & Kunkel, IKA-Labortechnik, Staufen, Germany) at 200 rev min-1. Two millilitres of 0.1 mol dm-3 sodium phosphate buffer, pH 7.5, containing 0.15 mol dm-3 NaCl and 0.1% Tween 20 was then added to each test-tube. The tubes were centrifuged and the supernatant was sucked off; the procedure was repeated twice. The radioactivity on the solid phase was then measured with a y-counter.All stages were performed at room temperature. A calibration graph was constructed using standards with known IgE concentrations, where 1 unit = 2.42 ng of IgE. The standard points were obtained by diluting the IgE in IgE-free horse serum diluent. This serum diluent was also used as the negative control. The count values obtained for the adsorbed immune complex with thiolated anti-IgE, diluted in IgE-free horse serum, and 1251-labelled IgE as the only constituents during the immuno- logical reaction, is called BO. The count values for the standard points of known IgE concentrations were expressed as percentages of Bo. Application of the immunocomplex-immobilization technique on IgE determination with P-galactosidase as a marker Thiolated anti-IgE (either lyophilized and then reconstituted with 10 ml of distilled water, or freshly prepared as described above) was diluted to a final concentration of 2 x 10-9ANALYST, AUGUST 1991, VOL.116 789 rnol dm-3 (0.32 pg ml-1) with 0.1 mol dm-3 sodium phosphate buffer, pH 7.5, containing 0.1% Tween 20 and 5 mmol dm-3 EDTA; 50 p1 of the diluted solution and 50 pl of patient serum were added to each test-tube. After 30 min, 50 pl of IgE-P-galactosidase conjugate with an IgE concentration of 2 x 10-9 mol dm-3 (0.32 pg ml-1) were added to the test-tubes and the reaction was carried out for 90 rnin at room temperature. A 100 p1 portion of the stirred suspension of Activated Thiol-Sepharose 4B was then added, which corre- sponds to 25 mg of dried gel per test-tube, and the tubes were incubated for 60 rnin on the shaker (200 rev min-1) at room temperature; 2 ml of 0.1 mol dm-3 sodium phosphate buffer containing 0.1% Tween 20 was then added to each test-tube.the tubes were centrifuged and the supernatant was sucked off, this procedure was repeated twice. After the last washing and rinsing step a residue of 500 pl remained in all of the tubes, 400 pl of the substrate solution was then added to each tube and the contents were mixed. The substrate solution consisted of 0.2 mol dm-3 sodium phosphate buffer containing 2 mmol dm-3 MgC12, 0.25% Tween 20 and 13.2 mmol dm-3 o-nitrophenyl-P-D-galactopyranoside at pH 7.2. After 30 min incubation at 37 "C the enzyme reaction was stopped by adding 0.8 ml of 1 rnol dm-3 Na2C03 to each test-tube.The liberated o-nitrophenolate was measured at 405 nm on an LKB Ultralab System 7400 filter-photometer (LKB, Bromma, Sweden). A calibration graph was obtained by diluting IgE in IgE-free horse serum diluent. This serum diluent was also used as the negative control. The count value obtained for the adsorbed immune complex with thiolated anti-IgE diluted in IgE-free horse serum and 1251 labelled IgE as the only constituents during the immunological reaction is called Bo. The count values for the standard points of known IgE concentrations were expressed as percentages of Bo. Phadebas IgE-Prist method One paper disc with covalently coupled anti-IgE was added to each of a series of test-tubes, together with 100 p1 of patient serum. Incubation was performed (without shaking) at room temperature for 3 h, then 2 ml of 0.1 rnol dm-3 sodium phosphate buffer, pH 7.5, containing 0.15 mol dm-3 NaCl and 0.1% Tween 20 were added.After 5 rnin the wash solution was quantitatively aspirated. This procedure was repeated twice. The 125I-labelled IgE (77 ng ml-1) was added to each test-tube and the mixture incubated without shaking for 17 h at room temperature, after which time 2 ml of 0.1 mol dm-3 sodium phosphate buffer, pH 7.5, containing 0.15 rnol dm-3 NaCl and 0.1% Tween 20 were added. After 5 rnin the wash solution was sucked off; the wash procedure was repeated twice. The radioactivity of the paper discs was then measured with a y-counter and the IgE concentration of the different samples was determined by using a calibration graph obtained with standards of known IgE concentration.Results and Discussion The principle of the immunocomplex-immobilization tech- nique is shown in Fig. 1. Thiolated antibody is mixed with a sample containing a known (standard) or unknown (patient sample) amount of corresponding antigen. Isotope or enzyme labelled antigen (Ag*) is then added and incubation is performed for a fixed period of time during which a mixture of labelled and unlabelled immunocomplexes is formed, both types containing thiol groups. The immunochemical reaction is carried out until equilibrium has been reached. A suspension of beaded agarose containing reactive disul- phide groups (Activated Thiol-Sepharose 4B) is then added and the final mixture is incubated for 60 min in order to effect the immobilization of the immunocomplexes onto the agarose beads.The unbound material, as free labelled antigen, is finally removed from the agarose gel and the amount of bound material is determined by conventional techniques (isotope measurement for radioimmunoassays and enzyme activity measurements for enzyme linked immunosorbent assays (see also under Methods). Step 1 Step 2 Step 3 Step 4 Ab-SH 2 Ag (Competitive reaction) %pg* I -S-S-Ab-Ag* + unbound Ag -S-S-Ab-Ag + unbound AS* I I I Wash solution I 3 x 2 m l t I Determination of solid phase bound radioactivitv or enzymic activity Fig. 1 Test principle (schematic representation. Step 1. (a) Thio- lated antibody (Ab-SH) and sample containing antigen (Ag) to be determined are added to each test-tube and incubated for 30 min.(b) Labelled antigen (Ag* = 1251 or P-galactosidase) is then added to each test-tube. The reaction must then proceed for another 60 min. Step 2. Activated agarose is added to each test-tube and the tubes are incubated for 60 rnin with mixing. Step 3. The solid phase is washed three times to remove unbound and non-specifically bound Ag*. Step 4. The tubes containing the washed agarose beads are placed in a y-counter to determine the radioactivity. The enzymic activity is determined by adding substrate to the test-tubes Step 1 Step 2 Step 3 Step 4 0. DDT (ox) DDT (red) B Anti body-N H-C-CH,-CH,-SH + A 0,H OF SH-CH2-CH-CH-CH2-S H DDT = Di t h i ot h rei to I Fig. 2 Scheme for thiolation of antibody with SPDP. Step 1. SPDP is added to the antibody solution.Step 2. After 30 min of reaction, the excess of reagent and products of low M, are removed by de-salting on a Sephadex G-25 column. Step 3. The modified antibody is treated with DIT. Step 4. Excess of reducing agent and 2-thiopyridone formed are removed by de-salting on a Sephadex G-25 column790 ANALYST, AUGUST 1991, VOL. 116 Immunoglobulins (of IgG type) normally do not contain free sulphydryl groups and thus have to be provided with such groups; this can be achieved using a number of methods. In this work the two-step procedure outlined in Fig. 2 has been used. Reactive disulphide groups (2-pyridyl disulphide groups) are first introduced using the heterobifunctional reagent, SPDP, which reacts with exposed amino groups on the antibodies. By varying the amount of the excess of reagent, the desired degree of substitution can be obtained.The protein-bound 2-pyridyl disulphide groups are then converted into thiol groups by reaction with a thiol of low M, such as DTT. As the 2-pyridyl disulphide groups are very reactive, this reduction can be performed by using an equimolar concentration of the thiol of low M,, thus avoiding simultaneous reduction of the native disulphide groups. It is important to provide the antibodies with a sufficient number of thiol groups so that binding of the immunocomplex formed to the agarose beads can occur even if some of the thiol groups become sterically hindered as a result of the formation of the immunocomplex. Antibodies have been prepared with 2.9, 3.5,6.0,12.9 and 18.2 rnol of thiol groups per rnol of IgG.The amount of immunocomplexes immobilized, with a substitu- tion degree of 18.2 mol of thiol groups per IgG molecule, was 98.6% of the amount immobilized when an antibody with a substitution degree of 2.9 was used. The highly substituted antibodies (18.2 mol of thiol groups per mol of IgG) were found to precipitate into the solution after 3 d of storage at 4 "C. It was also found that 3-6 mol of thiol groups per mol of IgG was sufficient to ensure effective immobilization. More- over, within this range the immunochemical reactivity of the thiolated antibody was the same as that for the native antibody. The thiolated antibody preparations (with a substitution degree of 3-6 rnol of thiol groups per mol of IgG), when kept in 0.1 rnol dm-3 sodium phosphate buffer, pH 7.5 (containing 5 mmol dm-3 EDTA), at 4 "C for 1 month, showed no decrease in immunochemical or chemical reactivity. After the addition of HSA, Dextran T-70, EDTA and glutathione (see under Experimental), it was also possible to freeze-dry the thiolated antibodies.Such preparations appeared to be stable for at least 2-3 months. The rate of the immunochemical reaction, i. e . , the competition between sample and labelled antigen for binding sites on the thiolated antibody, depends, among other things, on the relative concentrations of the participating immunocomponents and their affinities for each other. As the concentration in a 100 80 - s $ 60 >. > 1 -0 c .- .- c 2 40 20 0 1 1 I I l l 1 2.5 5 25 lY0 200400 IgE/U rnl- Fig. 3 Calibration graph for determination of IgE with 1251 as a marker.The calibration graph for determination of IgE in unknown samples is established according to the procedure presented under Methods. The count min-1 values for the different standard concen- trations are plotted as a percentage of the counts min-* value obtained for the negative control (IgE-free horse serum diluent) competitive system is fixed for the antibody, faster reaction can be obtained, for example, by exchanging the antibody for one with higher affinity. With the IgE used in this work, equilibrium in the immunochemical reaction was obtained within 90 min. For the immobilization of the immunocomplex, any solid phase can, in principle, be used, provided that it can be substituted with reactive disulphide groups.In this study Activated Thiol-Sepharose 4B, which contains 1 pmol of reactive disulphide groups per ml of swollen gel, was used. This agarose gel shows low non-specific adsorption and is commercially available. With the optimal mass of gel (12.5-25 mg of dried gel per test-tube) maximum binding of immunocomplex was reached after about 60 min of incubation, with mixing. The calibration graph for IgE with 1251 as a marker, with a total reaction time of 2.5 h (1.5 h of immunochemical reaction and 1 h of immobilization reaction) is shown in Fig. 3. With this calibration graph it is possible to estimate the concentra- tion of IgE in the interval between 8 and 400 U of IgE ml-1. The non-specific adsorption to the matrix was 4.5% of the total added activity and BdT was 22%.The calibration graph for IgE with P-galactosidase as the label is shown in Fig. 4. Under the conditions specified it would be possible to estimate the IgE concentration over the range 5400 U of IgE ml-1 and with a total reaction time of 3 h (1.5 h of immunochemical reaction, 1 h of immobilization 100 80 s 7J 5 60 0 n >. > '0 40 Q I c .- .- 0 1 2.5 5 25 100200400 IgE/U ml-' Fig. 4 Calibration graph for determination of IgE with the enzyme as marker. The calibration graph for determination of IgE in unknown samples is established according to the test procedure presented under Methods. Every absorbance value at 405 nm for the standard points with known IgE concentrations is taken as a percentage of the absorbance value at 405 nm for the negative control (IgE-free horse serum diluent) 3.5 3.0 2.5 2.0 3 ; 1.5 0, 0, 1.0 0.5 m .- I - I - I J 0 0.5 1.0 1.5 2.0 2.5 3 3.5 Log(lgE/U ml-1) ICI Fig.5 Scattergram obtained after a comparative study of the IgE content estimated in 41 serum samples using the commercial Phadebas IgE-Prist technique and the immunocomplex-immobilization (ICI) techniqueANALYST, AUGUST 1991, VOL. 116 791 reaction and 0.5 h of substrate reaction time). The non- specific adsorption to the matrix was 0.5% of the total added activity and BdT was 10%. When the commercial product Phadebas IgE-Prist and the immunocomplex-immobilization-technique with 1251 as the marker were compared by estimating the IgE content in serum samples from 41 persons, the correlation coefficient between the two techniques was 0.955 (see Fig.5 ) . The concentration of IgE in the 41 different patient samples was 2.8-2890 U of IgE ml-1. All of the serum samples with IgE values >400 U ml-1 were diluted in IgE-free horse serum diluent. In an attempt to increase the speed of the reaction in the immunocomplex-immobilization step, agarose derivatives of smaller particle sizes were investigated. Preliminary experi- ments demonstrated that the use of 2-pyridyl disulphide- containing agarose derivatives with small particle diameter (2-4 pm) instead of the commercially available Activated Thiol-Sepharose 4B with a particle diameter of 50-250 vm, shortened the time required for immobilization by at least 30 min. Moreover, the need to agitate the suspension during the immobilization reaction was also eliminated with this prepara- tion because the small agarose beads did not settle out. Preliminary experiments with thyroid stimulating hormone, nortriptylin, human IgG, insulin and 62-microglobulin indi- cate that the immunocomplex-immobilization technique is a fast and easy method with a wide range of applications and thus, should be an interesting alternative to the conventionally used immunoassay techniques. We thank Dr. R. Brandt for valuable discussions and Dr. D. Eaker for the linguistic revision. References Wide, L., in Radioimmunoassay Methods, eds., Kirkham, K. E . , and Hunter, M. M., Livingstone, London, 1970, p. 199. van Weemen, B. K., and Schuurs, A. H., FEBS Lett., 1971,15, 232. Grazetti, D. R., and Murray, J. F., Arch. Biochem. Biophys., 1967, 119, 41. SPDP; Heterobifunctional Reagent. Pharmacia Fine Chemicals AB, Uppsala, Sweden, 1978. Affinity Chromatography; Principles and Methods, Pharmacia Fine Chemicals AB, Uppsala, Sweden, 1988. Paper 1 I00491 C Received February 4th, 1991 Accepted March 26th, 1991

ISSN:0003-2654    

DOI:10.1039/AN9911600787

出版商:RSC    

年代:1991

数据来源: RSC

摘要:

ANALYST, AUGUST 1991, VOL. 116 793 Immobilized-enzyme Electrode for Nicotinamide Adenine Dinucleotide (Reduced form) (NADH) Sensing and Application to the Kinetic Studies of NADH Dependent Dehydrogenases Hsien-Chang Chang, Akinori Ueno, Hiroshi Yamada, Tornokazu Matsue" and lsamu Uchida Department of Molecular Chemistry and Engineering, Faculty of Engineering, Tohoku University, Sendai 980, Japan Amperometric determination of nicotinamide adenine dinucleotide (reduced form) (NADH) at an immobilized-diaphorase (Dp) electrode is described. The measurement was conducted using ferrocenylmethanol as a mediator in a stirred solution at 0.20 V versusa saturated calomel electrode. A linear relationship between the steady-state current and the concentration of NADH was found over the range 0.005-0.1 25 mmol dm-3.The immobilized-Dp electrode showed outstanding stability and the current response reached a steady state within 2-3 seconds upon addition of NADH. The proposed electrode was used to follow the reactions of pig heart lactate dehydrogenase and horse liver alcohol dehydrogenase. The kinetic investigation using the immobilized-Dp electrode gave the kinetic parameters (Michaelis constants, Km values, and maximum velocities, Vm values), which were in satisfactory agreement with those determined by a conventional spectrophotometric method. Keywords: Diaphorase; immobilized-enzyme electrode; nicotinamide adenine dinucleotide sensor Recently much attention has been focused on enzyme sensors based on various electrochemical methods of detection. * Many of the enzyme sensors proposed so far are based on amperometric or potentiometric detection coupled with oxi- dases .2 Dehydrogenases requiring nicotinamide adenine di- nucleotide (reduced form, NADH; oxidized form, NAD+) have not frequently been used for enzyme sensors in spite of their range of catalytic capabilities.More than 250 NADH dependent dehydrogenases3 have so far been purified from various sources, and many attempts have been made to use dehydrogenases for selective organic syntheses.4 The major problem in the practical application of dehydrogenases is the difficulty in regeneration of the coenzyme. Although the formal potential of the NADH-NAD+ couple is -0.56V versus a saturated calomel electrode (SCE) at pH 7.0,s large overpotentials are necessary for the direct electrochemical oxidation of NADH (about 1.1 V at a glassy carbon6 and 1.3 V at a platinum electrode7).The overpotentials are mainly ascribed to the very positive formal potential for the NADH- NADH.+ couple (0.78 V versus SCE).8 Thus, it is practically impossible to reduce the overpotential if NADH*+ is formed as the intermediate. However, it has been found that quinones,g 3-P-naphthoyl- Nile Blue,lo phenazine methosulphate,11 Meldola Blue 12 and a hexacyanoferrate film on an electrode13 effectively mediate the oxidation of NADH. Generally, these approaches are limited by their low stability or slow response. Another method used to accelerate the oxidation is the use of an enzyme reaction coupled with an electron transfer mediator such as a ferrocenel4 or a ferrocyanide complex.15 Cass et al.16 found that lipoamide dehydrogenase catalyses the electro- chemical oxidation of NADH. Recently, Miki et al. 17 demon- strated that diaphorase (Dp) pasted in carbon powder exhibits catalytic activity for the oxidation of NADH. The determination of NADH using an immobilized-Dp electrode in the presence of ferrocenylmethanol (FMA) as a mediator is reported here (Fig. 1). The proposed Dp, purified from Bacillus stearothermophilus, showed excellent stability. The immobilized-Dp electrode was also used for the determi- nation of the kinetic parameters of NADH dependent de h y drogenases. * To whom correspondence should be addressed. Experimental Materials Diaphorase I (E.C. 1.6.99-; relative molecular mass, -30 000; activity, 2030 U mg-1 for NADH) (1 U = 16.67 nkat) from Bacillus stearothermophilus was obtained from Unitika (Kyoto, Japan).The enzyme was purified according to a method reported previously. 18 The enzyme concentration was determined from the absorbance at 460 nm using ~ ~ 6 0 = 12 OOO dm3 mol-1 cm-1 .I8 Lactate dehydrogenase (LDH, (S)-lactate: NAD+ oxidoreductase, E.C. 1.1.1.27, relative molecular mass, 140 000; activity, -5000 U ml-1 for pyruvate) from pig heart was purchased from Oriental (Osaka, Japan). Alcohol dehydrogenase from horse liver (HLADH, ethanol: NAD+ oxidoreductase, E.C. 1.1.1.1, relative molecular mass, 84 0oO; activity, 27 U ml-1 for ethanol) was purchased from Boehr- inger, Mannheim, Germany. As the LDH and HLADH have not been purified, concentrations of these enzymes are expressed as U ml-1.The NADH was obtained from Sigma (St. Louis, MO, USA) and its concentration was determined based on &340 = 6220 dm3 mol-1 cm-1.19 Ferrocenylmethanol was obtained from Tokyo Kasei (Tokyo, Japan) and recrystal- lized from hexane. Sodium pyruvate and glutaraldehyde (25% solution) were purchased from Wako Chemicals (Osaka, Japan) and used as received. Cyclohexanone, 2-methyl- and 3-methylcyclohexanone were obtained from Wako and dis- tilled before use. All solutions were prepared with water purified by using a Milli-QII system (Millipore, Milford, MA, USA). Electrode FMA NADH FMA+ NAD+ Fig. 1 diaphorase (Dp) [reduced form, Dp (red); oxidized form, Dp (ox)] Scheme for electrocatalytic oxidation of NADH catalysed by794 ANALYST, AUGUST 1991, VOL.116 Instrumentation and Measurements Cyclic voltammetry and amperometry were performed with a potentiostat (Model HAB-151, Hokuto Denko, Tokyo, Japan) connected to an x-y recorder (Model WX43096, Graphtec, Tokyo, Japan) in a 4 ml solution of 0.05 mol dm-3 phosphate-NaOH buffer (pH = 7.5). The working electrode was a glassy carbon disc (3 mm in diameter) (Tokai Carbon, Tokyo, Japan, GC-20) mounted in a polytetrafluoroethylene rod. The electrode surface was polished to a mirror-like finish with 300 nm alumina. Preparation of the immobilized-Dp electrode was carried out by mixing 4 x 10-4 ml of a 1 mmol dm-3 solution of Dp with 2 x 10-4 ml of a 2% v/v glutaraldehyde solution at the electrode surface. The surface concentration of the immobilized Dp was 5.7 nmol cm-2.The electrode was kept at room temperature (25 "C) for 2 h in order to allow polymerization at the surfaces. A measurement made using a scanning electron microscope indicated that the thickness of the enzyme film was approximately 0.01 mm. The immobilized-enzyme electrode was kept in a buffer solution at 4 "C for at least 2 d before the measurements were made. The counter electrode was a platinum wire and the potentials were referenced to an SCE. All measurements were carried out using a water-jacketed cell kept at 30 "C under an atmosphere of nitrogen. The amperometric measurements for enzyme assays and kinetic studies were performed with stirring to avoid the influence of diffusion. The activity of LDH for pyruvate and that of HLADH for cyclohexanones were measured by monitoring the decrease in the concentration of NADH.The concentration of NADH was determined by the oxidation current for NADH observed at the immobilized-Dp electrode. The electrode potential was set at +0.20V versus SCE. The LDH assay solutions containing 0.1 mmol dm-3 NADH and various concentrations of pyruvate ranged from 0.06 to 0.50 mmol dm-3. The concentration of LDH was fixed at 250 U ml-1. The initial rate of the enzyme reaction was determined by the decrease in the concentration of NADH after substrate was added to the solution. In order to confirm the validity of the electrochemical method for the kinetic study described above, a spectro- photometric method was also examined. The concentration of NADH was determined by the absorbance at 340 nm using a multichannel spectroscopic system Model MCPD-110A (Ohtsuka Electric, Osaka, Japan).Results and Discussion Basic kinetic studies for the mediated electro-oxidation of NADH catalysed by Dp in solution have been reported by Matsue et aZ.18 It was found that FMA and 1-ferrocenylethanol showed smaller K , values and higher molecular activities than -0.4 -0.2 0 0.2 0.4 0.6 E N versus SCE Fig. 2 Cyclic voltammograms for 0.20 mmol dm-3 FMA on the immobilized-Dp electrode in 0.05 mol dm-3 phosphate buffer (pH 7.5). Solid line, without addition; broken line, with addition of 0.15 mmol dm-3 NADH; scan rate, 10 mV s-l; and temperature, 30 "C Co(phen)32+ (phen = 1,lO-phenanthroline) and Fe(CN)&-, indicating that these ferrocene derivatives act effectively as the mediators in accelerating the enzymic oxidation of NADH.In this paper, FMA, has been used mainly, because of its electrochemical reversibility, stability and commercial availability. Cyclic Voltammetric Behaviour of Immobilized-Dp Electrode Fig. 2 shows the cyclic voltammograms for 0.2 mmol dm-3 FMA at the immobilized-Dp electrode. In the absence of NADH, the voltammogram shows a well defined, reversible peak at about +0.20V versus SCE. The addition of 0.15 mol dm-3 NADH to the solution results in the appearance of a pre-wave at about +O.O5V and an obvious increase in the oxidation peak. The appearance of the pre-peak is caused by depletion of NADH in the vicinity of the electrode surface.20 This phenomenon is observed when catalytic reactions proceed rapidly at the electrode surface. The findings de- scribed above demonstrate that the Dp immobilized on the electrode surface effectively catalyses the oxidation of NADH by the oxidized form of FMA (Fig.1). Dependence of Current Response on the Concentration of NADH Fig. 3 shows the current response at the immobilized-Dp electrode upon the addition of NADH. The current response was extremely rapid and reached a steady-state current (&) within 2-3 seconds. The mediator rapidly shuttles between the 2.0 2 1.6 2 $ 1.2 2 (3 0.8 0.4 0 1 2 3 4 Ti me/m in Fig. 3 Current response at the immobilized-Dp electrode for successive additions of NADH (indicated by arrows) to the 0.05 mmol dm-3 phosphate buffer containing 0.2 mmol dm-3 FMA.Potential, 0.20 V versus SCE; temperature, 30 "C, [FMA], 0.02 mmol dm-3. [NADH]: A, 0; B, 12.5 x 10-6; C, 25 x 10-6; D, 50 x 10-6; E, 75 x 10-6; and F. 100 X 10-6 mol dm-3 3.5 Q 2.5 (D I s ? k 1.5 3 0 0.5 0 0.05 0.10 0.15 [NADH]/mmol dm-3 Fig. 4 Relationship between the oxidation current and concentra- tion of NADH observed at the immobilized-Dp electrode in buffer solutions containing: A, 0.05; B, 0.01; C, 0.15; and D, 0.20 mmol dm-3 FMA. Temperature, 30 "C; E = 0.2 V versus SCEANALYST, AUGUST 1991, VOL. 116 r I m u (? r - E 0.015 E E 0.010 . 0.005 795 - - - ,"' . ..' " electrode and the active centre of D p to oxidize NADH to NAD+ efficiently. The is, values increased linearly with the concentration of NADH (slope, 0.0235 mA dm3 mmol-1). The linear range depended on the concentration of FMA as shown in Fig.4. At an FMA concentration of 0.05 mmol dm-3, the is, value deviated from linearity in the range above 0.04 mmol dm-3. However, linearity was observed over a relatively wide range of concentrations (0.005-0.125 mmol dm-3) when the concentration of FMA was 0.20 mmol dm-3. The immobilized-Dp electrode showed outstand- ing stability; the variation of the is, value for repeated measurements (160 samples) over 3 months was +5%. The Dp-catalysed oxidation of NADH by FMA+ is com- posed of two basic reactions:20 2FMA+ + Dp(red) -+ 2FMA + Dp(ox) (1) Dp(ox) + NADH + Dp(red) + NAD+ + H+ (2) where Dp(red) is the reduced form of diaphorase and Dp(ox) is the oxidized form. When the concentration of NADH is small, the over-all reaction is controlled by reaction (2) and the is, value increases with the concentration of NADH.As the concentration of NADH increases, the relative contribution of reaction (1) to the over-all reaction becomes important. The above explains the deviation of the is, values from linearity at high concentra- tions of NADH. The relative contribution of reactions (1) and (2) to the over-all reaction is also governed by the concentra- tion of FMA+. The linear range of the is, versus the concentration of NADH plot becomes wider with increasing concentration of FMA. ,' Determination of LDH and HLADH Activity The determination of LDH activity for pyruvate was carried out in the presence of a specific amount of LDH (20-fold dilution of the enzyme as purchased), 0.2 mmol dm-3 FMA, 0.1 mmol dm-3 NADH and various concentration of pyru- vate.The initial concentration of NADH was large compared with the K , value of LDH for NADH (described later). Therefore, the over-all enzyme reaction should be controlled by the reaction between LDH and pyruvate. The decrease in the concentration of NADH along with the enzyme catalysed t c C 2 3 0 LDH NaP Time - Fig. 5 Time dependence of the current responses at immobilized-Dp electrode. Sodium pyruvate (Nap) A, 0.06, B. 0.16 and C, 0.50 mmol dm-3, was added to the buffer solution containing 0.20 mmol dm-3 FMA, 0.10 mmol dm-3 NADH and LDH (20-fold dilution of the original bottle). Temperature, 30 "C Table 1 HLADH ass;y tor cqclohexanone derivatives. All values in U m l 1 Method Substrate* Electrochemical Spectrophotometric C yclohexanone 3.680 4.820 2-Methylcyclohexanone 0.083 0.072 3-Methylcyclohexanone 1.960 1.970 * All substrates at a concentration of 0.1 mmol dm--3.reduction of pyruvate was monitored from the oxidation current for NADH at the immobilized-Dp electrode. Fig. 5 shows the decay in the oxidation current immediately after the addition of pyruvate. From the initial velocity (vg) of the decay in the concentration of NADH, the activity of the original LDH suspension €or pyruvate was calculated to be about 5000 U ml-1, which is in good agreement with the values deter- mined by the spectrophotometric method. The HLADH activities for three cyclohexanone derivatives (cyclohexanone, 2-methyl- and 3-methylcyclohexanone) were also determined by the proposed electrochemical methods.The results are summarized in Table 1. The values determined by the present procedure are in good agreement with those obtained by the spectrophotometric method. This table also shows large differences in activity among the substrates. These differences show the specificity of HLADH for cyclohexanone derivatives .zl Kinetic Study of LDH The K , and V, values for the oxidation of pyruvate catalysed by LDH were also determined by the proposed electrochem- ical method. First, the dependence of v0 on the initial concentration of NADH in the presence of an excess of pyruvate (0.5 mmol dm-3) (Fig. 6) was investigated. The value of vo was found to be constant when the concentration of NADH was >0.075 mmol dm-3. At such concentrations of NADH, the binding sites of Dp for NADH are saturated and thus the enzyme reaction is controlled by the reaction between Dp and pyruvate.Therefore, the kinetic studies of LDH for pyruvate were carried out in the presence of 0.1 mmol dm-3 NADH, which was in the region of saturation. The kinetic measurement for LDH was carried out at +0.20 V versus SCE. The immobilized-Dp electrode was first immersed in a solution containing FMA, NADH and LDH. The electrode was held for 1 min, then various amounts of > 0 . 1 5 1 T O I I I I I 1 1 [NADH]/mmol dm-3 0 0.05 0.1 0.15 0.2 0.25 Fig. 6 Relationship between initial rate (vo) and the concentration of NADH measured by spectrophotometric method. Measurements were carried out in a phosphate buffer solution (2 ml) containing 0.5 mmol dm-3 pyruvate and LDH (20-fold dilution of the original bottle).Temperature, 30 "C 0.020 ~ Fig. 7 Double reciprocal plot for kinetic studies of LDH observed by 0, the immobilized-Dp electrode and A, the spectrophotometric method. Concentrations of FMA and NADH were 0.20 and 0.10 mmol dm-3, respectively. Temperature, 30 "C796 ANALYST, AUGUST 1991, VOL. 116 pyruvate were injected into the solution. The rate of enzyme activity of LDH on pyruvate was monitored by the decrease in the concentration of NADH by using the immobilized-Dp electrode. The vo value at the various concentrations of pyruvate can be determined from the slope of the initial decay curve. The Lineweaver-Burk plot for vo and concentration of pyruvate is linear as shown in Fig.7. The K, and V , values determined from the slope and intercept were 0.097 and 0.161 mmol dm-3 min-1, respectively. These values are in good agreement with those obtained from the ordinary spectro- scopic method. In conclusion, the above results demonstrate that the immobilized-Dp electrode has excellent capability of sensing NADH in a solution. The Dp immobilized at the electrode surface showed a good stability and no obvious decrease in the activity was found for at least 3 months. Optimization of the immobilization conditions (e.g., the ratio of Dp to glutar- aldehyde etc.) would improve the sensitivity of the immobi- lized-Dp electrode. A variety of chemicals can be detected by using immobilized-enzyme electrodes co-immobilized with other NADH dependent enzymes.The electrochemical be- haviour of the immobilized-enzyme electrodes co-immobi- lized with an L-amino acid dehydrogenase is now being investigated for the detection of a specific L-amino acid.22 The present system can also be widely applied to assays and kinetic studies of various types of NADH dependent enzyme reactions. References 1 Johnson, D. C., Ryan, M. C., and Wilson, G. S., Anal. Chem., 1986.58.33R. 2 Yokoyama, K., Tamiya, E., and Karube, I., J. Electroanal. Chem., 1989,273, 107. 3 You, K.-s., CRC Crit. Rev. Biochem., 1985, 17,313. 4 Chenault, H. K., and Whitesides, G. M., Appl. Biochem. Biotech., 1987, 14, 147. 5 Lehninger, A. L., in Principles of Biochemistry, North Pub- lisher, New York., 1982, ch. 17. 6 Moiroux, J., and Elving, P. J., Anal. Chem., 1978, 50, 1056. 7 Jaegfeldt, H., J. Electroanal. Chem., 1980, 110, 295. 8 Matsue, T., Suda, M., Uchida, I., Kato, T., Akiba, U., and Osa, T., J. Electroanal. Chem., 1987, 234, 163 and references cited therein. 9 Tse, D. C.-S., and Kuwana, T., Anal. Chem., 1978,50, 1315. 10 Schelter-Graf, A., Schmidt, H. L., and Huck, H., Anal. Chim. Acta, 1984, 163,299. 11 Torstensson, A., and Gorton, L., J. Electroanal. Chem., 1981, 130, 199. 12 Gorton, L., J. Chem. Soc., Faraday Trans. I , 1986, 82, 1245. 13 Yon Hin, B. F. Y., and Lowe, C. R., Anal. Chern., 1987, 59, 2111. 14 Green, M. J., and Hill, H. A. O., J. Chem. SOC., Faraday Trans. 1, 1986, 82, 1237. 15 Yao, T., and Wasa, T., Anal. Chirn. Acta, 1985, 175,301. 16 Cass, A. E. G., Davis, G., Green, M. J., and Hill, H. A. O., J. Electroanal. Chem., 1985, 190, 117. 17 Miki, K., Ikeda, T., Todoriki, S., and Senda, M., Anal. Sci., 1989, 5 , 269. 18 Matsue, T., Yamada, H., Chang, H.-c., Uchida, I., Nagata, K., and Tomita, K., Biochim. Biophys. Acta, 1990, 1038, 29. 19 Winer, A. D., J. Biol. Chem., 1964, 239,3598. 20 Matsue, T., Yamada, H., Chang, H.-c., and Uchida, I., Bioelectrochern. Bioenerg., 1990, 24, 347. 21 Dulton, H., and Branden, C.-I., Bioorg. Chem., 1981, 10, 1. 22 Chang, H.-c., Yamada, H., Ueno, A., Matsue, T., and Uchida, I . , Denki Kagaku, 1990,58, 1211. Paper 0/03562I Received August 6th, 1990 Accepted January 28th, I991

ISSN:0003-2654    

DOI:10.1039/AN9911600793

出版商:RSC    

年代:1991

数据来源: RSC

摘要:

ANALYST, AUGUST 1991, VOL. 116 797 Amperometric Monitoring of Sulphur Dioxide in Liquid and Air Samples of Low Conductivity by Electrodes Supported on Ion-exchange Membranes Gilbert0 Schiavon" and Gianni Zotti National Research Council, Institute of Polarograph y and Preparative Electrochemistry, Corso Stati Uniti 4, 1-35100 Padua, Italy Rosanna Toniolo and Gin0 Bontempelli" Institute of Chemistry, University of Udine, Hale Ungheria 43, 1-33100 Udine, Italy An amperometric sensor is described for the determination of sulphur dioxide in both gaseous atmospheres and solutions of low conductivity. It consists of a porous Pt electrode (facing the sample) supported on one face of an ion-exchange membrane (Nafion 417) which serves as a solid polymer electrolyte. The other side of the membrane faces an internal electrolyte solution (1 mol dm-3 aqueous perchloric acid) containing the counter and reference electrodes.This sensor is inserted into a flow cell in which gaseous or electrolyte-free aqueous samples are fed by a peristaltic pump placed in a closed-loop path and SO2 is oxidized at an applied potential of 0.65 V versus Ag-AgCI. The device is found t o be characterized by a high current sensitivity and a short response time, 24 A cm-2 mol-1 dm3 and 1 s respectively for gaseous samples; 0.4 A cm-2 mol-1 dm-3 and 4s, respectively, for water solutions), and by good stability and low background noise. The dynamic range extends up t o 2 x mol dm-3 (gaseous samples) and 1 x 10-3 mol dm-3 (water samples) with good linearity, and detection limits of 8 x 10-9 mol dm-3 (gaseous samples) and 4 x 10-7 mol dm-3 (water samples) are predicted for a signal-to-noise ratio of 3.The advantages offered by this type of sensor over conventional gas-permeation membrane electrodes are discussed. Keywords: Sulphur dioxide; amperometric sensor; ion exchange membrane-supported electrode; elec- trolyte-free sample; electroanalysis Assessment of ecosystem acidification resulting from air pollution necessitates monitoring of the airborne concentra- tions of the contributing acidifying gaseous species, including the trace gas S02. Similarly, monitoring of the SO2 contam- inating various environmental liquid media is of essential interest. Moreover, such a species is used as a reducing agent or as a reactant in a number of industrial processes where the effective use of SO2 is largely dependent on the availability of in situ and instantaneous SO2 sensors suitable for its control, particularly in gaseous media.Several methods and instruments are available to monitor SO2 levels both in gaseous and liquid media using a great variety of systems.1.2 As every instrument has its own characteristics and limitations, new approaches to the detec- tion and analysis of SO2 appear to be desirable. In principle, analysers based on the direct electrochemical oxidation of this polluting species should be characterized by excellent results and should possess the required attributes for providing in situ measurements of SO2 and for its continuous monitoring. Unfortunately, these electroanalytical procedures cannot be applied directly to air samples or to liquid media with a low conductivity in which a sufficient concentration of supporting electrolyte is not present. The addition of electrolyte would cause contamination or perturbation of the natural solution equilibria in polar media, and such an addition is impractic- able in media that have very low dielectric constants because they are unable to dissolve ionic species.In order to overcome this drawback, membrane electrodes3-7 may be adopted but their performance is strongly affected by the transfer of analyte through the gas-permeable membrane, which proves to be a critical step. A fairly low sensitivity and fairly long response time are peculiar to these electrodes and their response also exhibits a high temperature dependence result- ing from the consequent change in the permeability of the membrane. * To whom correspondence should be addressed.The monitoring of SO2 by using an electroanalytical sensor suitable for the determination of electroactive analytes present in gaseous media or in solvents of high resistivity, described previously by Schiavon and co-workers,8-10 is reported. It is based on the use of a porous working electrode supported on one surface of an ion-exchange membrane whose other surface is in contact with an electrolyte solution containing the counter and reference electrodes. The porous working electrode faces the analyte sample in which no supporting electrolyte is needed because electroneutrality in the neighbourhood of the electrode surface is maintained by ionic migration through the ion-exchange membrane.Thus any membrane-permeation step is avoided and the analyte is monitored directly in its natural medium as the membrane separating the sample from the internal electrolyte does not act as a filter for gaseous molecules but serves to ensure the transfer of charged species from the working to the counter electrode. A similar approach has also been adopted by other workers. 11-13 Experimental Chemicals and Instrumentation All the chemicals used were of analytical-reagent grade (obtained from Carlo Erba, RPE) and were employed without further purification. Nitrogen atmospheres containing known amounts of SO2 and stock solutions of SO2 in water obtained from a Milli-Q system (Millipore) (de-aerated preliminarily with nitrogen) were prepared by diluting suitable known amounts of the pure gas.They were standardized by the iodine method14 and diluted further to the desired concentration with nitrogen or Milli-Q water. Voltammetric and amperometric measurements were per- formed by using a three-electrode unit equipped with an Amel 551 potentiostat driven by an Amel 568 digital logic function generator. The recording device was a Hewlett-Packard798 ANALYST, AUGUST 1991, VOL. 116 7080A measurement plotting system. All the tests were conducted at room temperature. Coating Procedure The ion-exchange material used as the solid polymer elec- trolyte (SPE) was a porous Nafion 417 cationic perfluorinated membrane 0.425 mm thick (Aldrich), reinforced with poly- tetrafluoroethylene (PTFE), which was cleaned by boiling in concentrated HN03 for 1 h and then in Milli-Q water for 1 h.The membrane was cut into discs of 1 cm in diameter which were equilibrated in aqueous 1 mol dm-3 NaC104 for 3 h and then in aqueous 1 X 10-3 rnol dm-3 [Pt(NH3)4]C12 (Johnson Matthey) (5 ml of solution for each disc). Following the Pt" loading, these discs were exposed to air on one side and on the other side to a hot (50-60 "C) aqueous alkaline solution (pH = 10) of NaBH4 (1 x 10-3 rnol dm-3) for 3-4h.15716 The concentration and volume of the PtIr solution was adopted in order that a uniform Pt film, 0.5 pm in thickness, could be formed upon total reduction by NaB&. The concentration of this last reducing solution, together with its temperature and the contact time were chosen so as to lead to the formation of a smooth and strongly adherent film located mainly on the surface of the membrane (thus enabling good contact with a Pt wire permitting electrical connection), with a minor portion of the deposited Pt embedded inside the membrane, in order to ensure good adherence of the electrode even under high mechanical tension.Such a Pt distribution can be accom- plished because once the Pt exchange membrane is soaked in the reducing solution, Pt" cations located near the surface are reduced to the metallic state, thus giving rise to a concentra- tion gradient between the bulk of the membrane and its surface which causes further Pt" cations to diffuse towards the surface of the membrane where they are in turn reduced.Different conditions for the Pt deposition step led to less effective electrode films characterized by lower current densities. A decrease of the Pt loading caused poor inter- particle contact, while thicker films in less intimate contact with the ion-exchange sites of the membrane were formed at higher Pt loadings. Electrode Assembly After completion of the coating procedure, Nafion discs were equilibrated in aqueous 1 rnol dm-3 HC104 and then installed in the electrode assembly shown schematically in Fig. 1. The cells were constructed as follows. A disc of Pt-coated ion-exchange membrane was clamped, with the porous conductive layer directed downward, at the bottom of a Pyrex cylinder, the end of which was threaded to a drilled PTFE holder sealing the assembly by means of an elastic O-ring resistant to acids and bases.Electrical contact with the Pt working electrode was made by placing a platinum ring on the inner side of the drilled holder, onto which the border of the Pt film deposited on the ion-exchange membrane was pressed. A Pt wire welded to this Pt ring and piercing through the holder permitted electrical connection. Such an assembly allowed about 0.4 cm2 of Pt film to be exposed to external media. The uncoated side of the membrane faced an internal compartment containing the 'internal electrolyte' (3 ml of aqueous 1 mol dm-3 HC104 solution) and equipped with an aqueous Ag-AgC1 reference electrode and a Pt counter electrode. This compartment was obtained by sealing the Pyrex cylinder to the bottom of a standardized hollow glass stopper in order to make possible the use of this electro- analytical device as a gas-tight stopper for the standardized glass flow cell (shown schematically in Fig.Z), in which all the experiments were conducted. This device enables electrochemical processes to occur in media of high resistivity, without the need for supporting electrolytes, according to the principle illustrated in Fig. 1 C R . a -b - -d Fig. 1 Schematic view of the Pt covered Nafion electrode. a, Internal compartment filled with aqueous 1 mol dm-3 HC104 and equipped with C, a Pt counter electrode and R, a Ag-AgC1 reference electrode; b, Nafion ion-exchange membrane; c, porous Pt coating; d, Pt-ring collector; and e, sample Counter Reference Sa m p I e -- stream Fig.2 Schematic view of the flow cell. Sample volume = 20 ml (enlarged view). When the platinum electrode is held at an appropriate potential, the substrate (SO2) in the working sample is oxidized to yield sulphate and hydrogen ions (the hydrogen ions being trapped in the membrane), coupled with an ionic migration through the membrane to maintain electroneutrality. Concurrently, hydrogen ions in the counter solution are reduced at the counter electrode, thus restoring the ionic content of the internal electrolyte. Hydrogen produced in this cathodic reaction is periodically removed by purging with nitrogen . The flow cell was fed with N2-S02 atmospheres or SO2 solutions in Milli-Q water whose flow rate was kept constant, unless stated otherwise, at 100 ml min-1 by a peristaltic pump placed in a closed-loop path.A flow meter was inserted in the stream in order to monitor this flow rate.799 ANALYST, AUGUST 1991, VOL. 116 4.0 I N I Eu a 2.0 E a 0 0.5 E N 1 .o Fig. 3 Voltammogram recorded at a Pt-Nafion electrode, with aqueous 1 mol dm-3 HC104 as the internal electrolyte, with a nitrogen stream containing SO2 (0.13 mmol dm-3; 8.3 mg 1-1; 2900 ppm v/v). Scan rate, 0.05 V s-1; and flow rate, 100 ml min-1. Current densities refer to the geometric area of the electrode Results and Discussion Anodic Behaviour of SOz at PtSPE Electrodes Platinum porous films have been selected as working elec- trodes because such an electrode material assures an over- voltage which is as small as possible for the SO2 oxidation process.17 A typical steady state linear sweep voltammogram recorded at a Pt-covered Nafion electrode with 1 mol dm-3 aqueous HC104 as the internal electrolyte and a nitrogen stream containing SO2 flowing into the closed-loop flow cell is shown in Fig.3. A single well-formed and reproducible anodic wave is observed, whose limiting current extends over a fairly wide range of potentials. This wave, whose irreversible character has been confirmed by applying the usual criteria to cyclic voltammograms recorded in stationary N2-SO2 atmos- pheres, resembles closely that found at conventional platinum electrodes for the oxidation of SO2 dissolved in aqueous HC1045.17 (the medium used as the internal electrolyte for wetting the rear side of the Nafion membrane). Fairly similar voltammograms, located at the same poten- tials, were also recorded at the same Pt-SPE electrode in electrolyte-free water in which SO2 was dissolved, regardless of the concentration.This last result is rather surprising as a progressive anodic shift of the SO2 oxidation process is expected with a decrease in the amount of SO2 present, owing to the concomitant increase in pH (from about 2.3 up to 7.0 for concentrations decreasing from 1 x 10-2 to 1 x 10-9 rnol dm-3) which causes increasing fractions of SO2 to be present in the solution as the conjugate base HS03-. This apparent contradiction however can be easily explained by considering the irreversibility affecting the electron transfer involved in this anodic oxidation17 which is able to mask the dependence of potential on pH, thus leading to the coin- cidence of voltammograms recorded on strongly acidic and basic buffered SO2 solutions.5 Alternatively, it can be accounted for by admitting that the pH governing the anodic reaction is that of the internal electrolytes rather than the pH of the sample.For both aqueous and gaseous samples, no progressive decrease in the over-all voltammogram was observed in the second and subsequent scans, contrary to the results obtained at conventional platinum electrodes for SO2 aqueous solutions containing supporting electrolytes.'* The ability of ion- exchange membranes to prevent the passivation of Pt elec- trodes, on which they are present as a film, has been previously observed in the surface-conditioned oxidation of methanol occurring at Nafion-covered Pt electrodes.19 On the basis of these voltammograms, a potential of 0.65 V versus an Ag-AgCI electrode was applied to the Pt-SPE electrode when performing the subsequent amperometric measurements of SO2 in both gaseous imd liquid samples. 0.5 N I Eu a E 0.25 0 2 4 6 Time/min Fig.4 Current-time profile recorded at a Pt-Nafion electrode, with aqueous 1 rnol dm-3 HC104 as the internal electyrolyte, on nitrogen streams containing SO2 in the following concentrations: A, 4.7 X mol dm-3 (105 pm); B, 9.4 x 10-6 rnol dm-3 (210 pm); C, 15.1 x mol dm-3 6 3 8 ppm); D, 19.5 x 10-6mol dm-3 637 ppm); and E 23.7 X 10-6 mol dm-3 (531 pprn). Applied potential, 0.65V; flow rate. 100 ml min-1 Table 1 Performance of the amperometric SPE sensor Gaseous Water Feature Unit* samples samples Sensitivity A cm-2 mol-1 dm3 24 0.4 pA cm-2 ppm-l 1.07 6.25 Response Detection time S 1 4 limit mol dm-3 8 x 10-9 4 x 10-7 I.18 I-' 0.6 26 PPb 180 26 Dynamic range mol dm-3 Up to 2 x 10-4 Up to 1 x 10-3 mg 1-1 Up to 12.8 Up to 64 PPm u p to 4500 Up to 64 * For gaseous samples, ppb and ppm (v/v) are employed.Such a value for the potential was selected in order to maximize the current detection sensitivity of the sensor and, at the same time, to minimize its dependence on the working potential. Performance of the Amperometric Sensor In order to test the performance of the Pt-covered Nafion electrode as an amperometric sensor for SO2 in gaseous atmospheres, nitrogen streams containing known and increas- ing concentrations of SO2 were fed into the flow cell shown in Fig.2, typically at a flow rate of 100 ml min-1, and the SO2 content was monitored by measuring the current flowing when a potential of 0.65 V was applied to the working Pt film. The results obtained are summarized in Fig. 4 which shows a typical current-time response recorded during these measure- ments. Each addition of SO2 causes a rapid rise in the current which attains a satisfactory constant value in a fairly short time (about 15 s). This constant current was found to be reprodu- cible within +3% and to depend linearly on the concentration of SO2 over a fairly wide range which extends up to about 2 X 10-4 rnol dm-3 (12.8 mg 1-1,4500 ppm v/v). From a plot of this steady state current density versus SO2 concentration obtainedANALYST, AUGUST 1991, VOL.116 in the range 5 X 10-8-5 x 10-6 mol dm-3, a sensitivity of 24 A cm-2 mol-l dm3 (1.1 PA cm-2 ppm-l) was obtained with a correlation coefficient of 0.998. As the residual current density at the working potential was about 1 FA cm-2 with a standard deviation of about 0.06 pA cm-2 (background noise), a detection limit of approximately 8 x 10-9 rnol dm-3 (0.6 pg 1-1, 180 ppb v/v) could be achieved for a signal-to-noise ratio of 3. These results are summarized in Table 1 where they are compared with those obtained for electrolyte-free water samples containing SO2. Table 1 also gives the response time which cannot be inferred correctly from the current-time profiles shown in Fig. 4, as the time required to attain a steady-state current is due mainly to the inertia opposing the flowing system required in order to achieve equilibrium conditions after each increase in the concentration of S02.Consequently, the response time was investigated by carrying out suitable experiments. A 95% response was observed in 1 s when the electrode was transferred rapidly from air to the cell, shown in Fig. 2, fed with a gas stream containing 1 X 10-5 mol dm-3 (0.64 mg 1-1, 224 ppm v/v) of SO2. This response time remains practically unaffected when the electrode is transferred from air to gas streams with higher or lower contents of S02. The performance of the proposed Pt-SPE sensor with reference to the determination of SO2 dissolved in electrolyte- free water was evaluated after preliminary tests aimed at verifying the effect of the possible transfer of ionic species from the internal electrolyte to samples of high resistance by either permeation through the ion-exchange membrane or incidental leaks caused by defective sealing of the internal electrolyte compartment.With this purpose, a series of Pt-SPE electrodes, all containing aqueous 1 mol dm-3 HC104 as the internal electrolyte and prepared following the pro- cedure reported above, were soaked in Milli-Q water samples (20 ml) in which different amounts of SO2 were dissolved, so as to attain concentrations ranging from 1 x 10-6 to 1 x 10-3 rnol dm-3. The pH due to the presence of SO2 remained virtually unchanged over time (1 h) when each of the samples was monitored using a glass electrode. Only after longer times (about 5 h) could a slight decrease of pH be detected in the more diluted samples (for 1 X 10-6 mol dm-3 solutions a pH decrease from 6 to 4.8 was found), thus suggesting that any electrolyte leakage can be ruled out and no appreciable ionic transfer due to permeation through ion-exchange membranes can be expected to occur during the time typically required for the analysis of a sample (about 2 min).The results concerning solutions containing SO2 in elec- trolyte-free water, summarized in Table 1, were obtained by the same tests described above for nitrogen atmospheres containing S02. In all instances, current responses were practically unaffec- ted by the temperature (1040 "C) as expected on the basis of the absence of gas-permeation steps. Moreover these responses were independent of moderate variations in both the applied potential (k0.05 V) and the sample flow rate, provided that this last parameter was higher than a minimum value (about 50 ml min-1 for gaseous streams and about 80 ml min-1 for aqueous streams).The attainment of constant currents for flow rates above a threshold value can be accounted for by considering that SO2 molecules from the gaseous or liquid stream must diffuse across the thin porous Pt film before reaching the Pt-Nafion interphase where they undergo oxidation. This fact implies that the SO2 oxidation rate is controlled by a diffusion-permeation Pt layer, the thickness of which is independent of the flow rate. Conse- quently, no increase in the current is expected for flow rates above the threshold required to reduce the thickness of the stagnant layer near the electrode surface to a value that allows a more rapid diffusion within such a layer than within the porous Pt film.The long-term stability of these Pt-covered Nafion elec- trodes in gaseous or liquid streams containing SO2 appeared to be totally satisfactory in that no appreciable change in the current response was observed even after 2 months of continuous use. Different Pt-Nafion electrodes, constructed following an identical procedure, led to very similar responses (k4%) when tested on the same SO2 samples, thus indicating that quite reproducible electrode surfaces can be obtained by the coating procedure adopted here. In conclusion, SPE electrodes appear to be useful alterna- tives to conventional amperometric membrane electrodes from which they differ in their inherent conceptual approach.The polymeric membrane in the latter sensors serves as a gas-permeable interphase between the sample and the indica- tor electrode, while Nafion in SPE electrodes acts as an 'ion-permeable, membrane separating the internal electrolyte from the working electrode, which is in direct contact with the sample. Thus, any gas-permeation step is avoided and it is this feature that makes SPE electrodes preferable to conventional membrane electrodes. Thus, sensitivity and response time for the detection of SO2 in gaseous atmospheres, as reported in Table 1 for Pt-covered Nafion electrodes, are undoubtedly better than those found at conventional membrane electrodes [0.3 pA cm-2 ppm-1 and 30s (reference 4) or 0.04 pA cm-2 ppm-1 (reference 7)].In addition the lower detection limit (ppb rather than ~pm4.7)~ the wider dynamic range (up to 4500 ppm as opposed to 1500 ppm7) and the negligible effect of temperature, must be considered among the advantageous characteristics deriving from the absence of a gas-permeation step. Moreover, the electroanalytical sensor proposed in the present paper is advantageous even when compared with a device proposed recently which permits the amperometric determination of SO2 in air samples by a closed-loop flow injection system containing a regenerable chemical probe.20 These advantages notwithstanding, the proposed SPE electrodes cannot be adopted for direct measurements of SO2 in ambient air as a lower limit of detection (of about one order of magnitude) is required.However, they can be used successfully for the direct determination of SO2 in power plant plumes and for industrial hygiene measurements, without the need for preconcentration. All of the characteristics mentioned above make this type of electrode particularly attractive for the monitoring of any electroactive species present in gaseous phase or highly resistive solutions, provided that an appropriate working potential is chosen. Selectivity, however, may be, in principle, unsatisfactory in some instances, owing almost exclusively to the value of the working potential. Thus, for the monitoring of SO2, it has been found that the oxidation of two possible interfering species, NO and H2S, can occur at the high positive potential applied (0.65 V).It must be remarked however that such an insufficient selectivity is not conditioned by the electroanalytical SPE sensor proposed here, but is peculiar to amperometric measurements. Nevertheless, the interferences mentioned are not a real drawback in this instance as much lower concentrations are usually expected for NO and H2S with respect to SO2 so that their effect on the measurement of SO2 can be neglected. Moreover, the chemical nature of these interfering species is such as to permit their easy removal prior to SO2 detection by passing the sample through suitable reaction columns. We thank S. Sitran of the National Research Council, Institute of Polarography and Preparative Electrochemistry (Padova) for skilful experimental assistance.The financial aid of the Italian National Research Council and of the Ministry of the University and of the Scientific and Technological Research is gratefully acknowledged.ANALYST, AUGUST 1991, VOL. 116 801 1 2 3 4 5 6 7 8 9 10 References Hollowell, G. D., Gee, G. Y., McLaughlin, R. D., Anal. Chem., 1973,45,63A. Sickles, J. E., and Grohse, P. M., Sampling and Analysis Methods for Sulphur Dioxide and Nitrogen Dioxide: A Litera- ture Review, Research Triangle Institute Report no. RT1/2823/ 00-011, Research Triangle Institute, Research Triangle Park, NC, 1984. Shaw, M., and Thatcher, I . , US Patent 40 17 373. Chem. Abstr., 1977,87.055005g. Bruckenstein, S . , Tucker, K. A., and Gifford, P. R., Anal. Chem., 1980,52, 2396. Bergman, I., J. Electroanal. Chem., 1983, 157, 59. Langmaier, J., Opekar, F., and Pacakova, V., Talanta, 1987, 34,453. Langmaier, J., Polak, J., and Opekar. F., Analyst, 1988, 113, 501. Schiavon, G., Zotti, G., and Bontempelli, G., Anal. Chim. Acta, 1989, 221,27. Schiavon, G., Zotti, G.. Bontempelli, G., Farnia, G., and Sandona, G., Anal. Chem., 1990, 62, 293. Schiavon. G., Zotti, G., Toniolo, R., and Bonternpelli, G., Efectroanalysis, 1991, in the press. 11 12 13 14 15 16 17 18 19 20 Kaaret, T. W., and Evans, D. H., Anal. Chem., 1988,60,657. De Wulf, D. W., and Bard, A. J., J. Electrochem. SOC., 1988, 135, 1977. Harth, R., Mor, U., Ozer, D., and Bettelheim, A., J. Electrochem. SOC., 1989,136,3863. Jeffery, G. H., Bassett, J., Mendham, J., and Denney, R. C., Vogef’s Textbook of Quantitative Chemical Analysis, Longman, Harlow, 1989, p. 398. Millet, P., Pineri, M., and Durand, R., J. Appl. Electrochem., 1989, 19, 162. Fedkiw, P. S., and Her, K. W., J. Electrochem. Soc., 1989,136, 899. Zhdanov, S. I., in Encyclopedia of Electrochemistry of the Elements, ed. Bard, A. J., Marcel Dekker, New York, 1975, Spotnitz, R. M., Colucci, J. A., and Langer, S. H., Electrochim. Acta, 1983,28, 1053. Enea, O., J. Electroanal Chem., 1987, 235, 393. Rios, A., Luque de Castro, M. D., Valcarcel, M., and Mottola, H. A., Anal. Chem., 1987, 59, 666. Paper 1 /00349F Received January 23rd, 1991 Accepted April 4th, 1991 vol. IV, pp. 314-335.

ISSN:0003-2654    

DOI:10.1039/AN9911600797

出版商:RSC    

年代:1991

数据来源: RSC

摘要:

ANALYST, AUGUST 1991, VOL. 116 803 Voltammetric Behaviour of Salicylic Acid at a Glassy Carbon Electrode and Its Determination in Serum Using Liquid Chromatography With Amperometric Detection Denley Evans Department of Pathology, West Wales General Hospital, Carmarthen, Dyfed SA3I ZAF, UK John P. Hart* Department of Science, Bristol Polytechnic, Coldharbour Lane, Frenchay, Bristol, Avon BS16 IQY, UK Glan Rees Department of Haematology, West Wales General Hospital, Carmarthen, Dyfed SA31 2A F, UK Cyclic voltammetry was used t o investigate the oxidation of salicylic acid at a planar glassy carbon electrode. The electrode reaction was found t o be dependent on the pH and ionic strength of the acetate buffer, which contained 35% methanol. Under these conditions the maximum electrochemical signal was obtained with a supporting electrolyte of 0.06 mol dm-3 acetate buffer in 35% methanol (pH 5.0).The peak current value (i,) increased by approximately 10% when the methanol concentration was decreased to 8%. The substance was found t o undergo an irreversible reaction involving one electron and possibly two protons in the initial oxidation step with at least one possible quasi-reversible follow-up reaction. The optimum mobile phase for the liquid chromatography, with amperometric detection, of the serum extracts was found to be 0.06 mol dm-3 acetate buffer in 8% methanol (pH 5.0); the acidified serum was extracted with a mixture of chloroform and acetonitrile (60 + 40), prior t o injection onto a reversed-phase column. The peak current was measured at +1.35 V and the calibration graph was found t o be linear in the range 4-200 ng of sample injected.The average recovery from serum was found to be 60% with a relative standard deviation of 5.8%. A pharmacokinetic study was carried out and the results obtained were comparable t o those found in the literature. It was concluded that the method developed had possible application for the measurement of trace levels of salicylic acid in clinical studies. Keywords: Salicylic acid; amperometric detection; cyclic voltammetry; serum Aspirin, the prototype of the salicylates, is a non-steroidal anti-inflammatory agent and is known as acetylsalicylic acid.' Zrz vivo, the drug rapidly hydrolyses to salicylic acid and acetate. The continued interest in the methods of analysis of salicylic acid is undoubtedly due to its consumption as a therapeutic drug.Thus, its therapeutic action and its toxic effects make it a drug that is subjected to continuous research. The techniques described for the determination of salicylic acid are many and varied and include colorimetry, titrimetry, ultraviolet (UV) spectrophotometry, gas chromatography, direct potentiometry and high-performance liquid chromato- graphy (HPLC).2 Of these, the titrimetric methods are subject to interference from reducing agents and, hence, are non- specific. The colorimetric methods, which involve the forma- tion of complexes with iron(IrI), are insensitive for the determination of salicylic acid at sub-nanogram levels. Several chromatographic methods have been developed including liquid chromatography (LC) usually using either UV or fluorescent detectors. To date, there are only a few reports on the use of amperometric detection for the assay of salicylic acid,3 and no systematic studies on the optimization of the electrochemical conditions for liquid chromatography with electrochemical detection (LCEC), or the nature of the oxidation reaction at a glassy carbon electrode, have been reported.The purpose of thc first part of this work was to carry out detailed studies on the electrochemical behaviour of salicyclic acid at a glassy carbon electrode using cyclic voltammetry and a variety of solution conditions. This information was then used in the development of a selective and sensitive assay involving LCEC, which was required to be suitable for use in pharmacokinetic studies following a single 300 mg oral dose of aspirin and for the quantitative measurement of salicylic acid * To whom correspondence should be addressed.in human blood serum in the nanogram range after low doses of aspirin. Experimental Reagents All reagents were of analytical-reagent grade unless stated otherwise. Salicylic acid was purchased from Sigma and the solvents used for HPLC were of HPLC grade and purchased from Merck. The supporting electrolytes used for the cyclic voltammetry studies were prepared from stock solutions containing either 2 mol dm-3 sodium acetate or acetic acid. These were mixed to give solutions of the required pH (a pH meter was used) and the resultant acetate buffers were diluted with methanol and water to the desired concentrations.The methanolic acetate buffer was used as the mobile phase in the LCEC studies and was prepared by mixing 0.6 mol dm-3 sodium acetate-acetic acid buffer solution (pH 5.0) and methanol to give a final electrolyte concentration of 0.06 mol dm-3. Stock dilutions of salicylic acid were prepared in methanol and were protected from the light during all of the investigations. The mobile phase was filtered through glass micro-fibre membranes (Whatman GFE) and subsequently de-gassed with helium. Apparatus An Oxford electrode portable potentiostat, equipped with a Gould Series 60 OOO x-y plotter, was used for recording the cyclic voltammograms. A three-electrode cell was employed incorporating a glassy carbon working electrode, a saturated calomel reference electrode (SCE) and a platinum-wire counter electrode.The LCEC was performed using an LDC Constamatic Model 111 reciprocating pump plus pulse damperANALYST, AUGUST 1991, VOL. 116 and an EDT LCA15 amperometric electrochemical detector together with a Bioanalytical System TL-5 thin-layer cell containing a glassy carbon electrode. Voltammetric Procedure Cyclic voltammetry (CV) was performed on solutions containing 5 X 10-4 mol dm-3 salicylic acid, 35% methanol and 0.05 mol dm-3 (pH 3.0-7.0) acetate buffers, in order to investigate the effect of pH. The voltammetric conditions were as follows: initial potential, 0 V; scan rate 50 mV s-1; and final potential 1.50 V. The effect of the ionic strength of the acetate buffer was studied in the concentration range 0.01-1 .O mol dm-3 using a scan rate of 50 mV s-1.In between the successive runs the working electrode was cleaned by washing it with distilled water, polishing the surface with aluminium oxide, rinsing again with distilled water and finally drying with tissue paper. The effect of the concentration of methanol was investigated by dissolving 5 x 10-4 mol dm-3 salicylic acid in solutions of 8-35% methanol-0.06 mol dm-3 acetate buffer, pH 5.0, and performing CV using the conditions described above. The phenomenon of adsorption was investigated by varying the scan rate for a 5 x 10-4 mol dm-3 salicylic acid solution in 35% methanol-0.06 mol dm-3 acetate buffer (pH 5.0). In order to obtain the hydrodynamic voltammograms, 20 pl sample volumes, containing 5 x 10-4 mol dm-3 salicylic acid in 35% methanol-0.06 mol dm-3 (pH 5.0) acetate buffer were injected onto the column and the potential was varied between 0.50 and 1.60 V.A flow rate of 1 ml min-1 was used. Hydrodynamic voltammograms were constructed by plotting the recorded peak currents against the applied potential. The optimum potential for the determination of salicylic acid was found from the position of the plateau on the hydrodynamic wave. Determination of Salicylic Acid in Human Serum by LCEC Serum samples (0.5 ml) were extracted using the method described by Tebbett and Omile.4 Briefly, this involved acidification of the serum (0.5 ml) with 0.1 ml of 1 mol dm-3 HCl and extraction with 2 ml aliquots of a solvent mixture consisting of chloroform and acetonitrile (60 + 40).The serum sample was extracted three times and the extracts combined. The organic phase was evaporated to dryness under a stream of nitrogen and the residue redissolved in 100 p1 of the mobile phase. Samples of 20 p1 volume were injected onto the column. Calibration, Recovery and Precision of LCEC A calibration graph of peak current versus mass of salicylic acid injected was constructed over the range &lo00 ng per 0.5 ml of serum. The recovery of salicylic acid was determined by spiking the serum, which was known to be free from salicylic acid, with concentrations in the range 30-500 ng per 0.5 ml of serum. In order to keep the volumes constant throughout the experiment, a known concentration of salicylic acid was added to the 100 pl of 1 mol dm-3 HCl used for the acidification step.Further samples containing 50 ng per 0.5 ml of serum were analysed to investigate the precision of the assay. Pharmacokinetic Study Two healthy male volunteers participated in the study. The subjects had not taken aspirin or any other drug for at least 10 d prior to the study. Both subjects fasted overnight and in the morning 5 ml of venous blood was drawn through an in-dwelling polytetrafluoroethylene catheter inserted into the I 4 r I 1 0 0.5 1 .o 1.5 EN versus SCE Fig. 1 Cyclic voltammogram of 5 x rnol dm-3 salicylic acid in 35% methanol4.05 mol dm-3 acetate buffer solution (pH 5.0), using a glassy carbon electrode. Initial potential, 0 V; scan rate, 50 mV s-1. 1F and 2F, first and second forward scan; lR, first reverse scan 45 l5 i I I I 10 1 I 0.8 3 4 5 6 7 Fig.2 Effect of pH on peak current and peak tential for peak Ia (see Fig. 1) using 5 x 10-4 mol dm-3 sacylic acid in 35% methanol-0.05 mol dm-3 acetate buffer PH cephalic vein and collected into a plain serum container. Both subjects were then given 300 mg of aspirin orally, which was swallowed with water. Further blood samples were taken every hour throughout the day for 7 h. The samples were centrifuged within 10 min of collection and the serum separated, quickly frozen and stored at -20 "C. All of the samples were extracted and chromato- graphed the following day, using the same procedure as for the spiked serum. Results and Discussion Voltammetric Behaviour of Salicylic Acid at a Glassy Carbon Electrode and Optimization of Conditions Cyclic voltammograms were recorded for salicylic acid solutions in the pH range 3.0-7.0.In all instances, the first forward scan exhibited a single anodic peak (Fig. 1 Ia) owing to the oxidation of the salicylic acid, probably at the phenolic moiety. The oxidation process is irreversible; however, a reduction reaction does occur, as seen from peak IIc (Fig. 1). The reaction giving rise to this peak is quasi-reversible as indicated by the appearance of peak IIa on the second forward scan (Fig. 1). These results suggest that on the first anodic scan salicylic acid undergoes an electron transfer-chemical reaction-elec- tron transfer reaction pathway at the glassy carbon electrode. The final product(s) of this pathway might be of quinone-typeANALYST, AUGUST 1991, VOL.116 structure , which might be expected to undergo the quasi- reversible redox reactions producing peaks IIa and IIc (Fig. 1). Such reactions have been reported for a structurally- related compound containing a phenolic moiety.5 The variation of peak potential (E,) and peak current (i,) with pH for peak Ia is shown in Fig. 2. A break appears in the E, versus pH graph indicating a pk’ value of 5.0. Below this break the E, value obeys the relationship: E, = 1.50 - 0.11 pH. It should be noted that the pk, values of salicylic acid are 2.97 and 13.40, corresponding to dissociation of carboxylic acid and phenolic groups, respectively.6 Therefore, the pk’ is probably a manifestation of the dissociation of the carboxylic acid group. The resultant anion is stabilized by intramolecular hydrogen bonding with the adjacent phenolic group;7 such stabilization might be expected to alter the electrochemical behaviour of this group (see below).Values of an, (where a is the transfer coefficient and n, is the number of electrons involved in the rate determining step) were calculated at various pH values for peak Ia, as shown in Table 1, according to the following equation: 0.048 an, = E,-(E$) Where E,/2 is the potential at half peak current (442). The an, values calculated for the oxidation reaction producing peak Ia suggest that one electron is involved in the rate determining step. The slope of the Ep versus pH plot below the pk‘ value suggests that two protons might be involved in this step.If this is true, then protons from both the phenolic and carboxylic acid group might be controlling the rate of oxidation. In addition, the oxidation process became independent of pH above the pk’ value; these effects can be associated with intramolecular hydrogen bonding as discussed above. Peak Ia (Fig. 1) achieved maximum current with the methanolic acetate buffer at pH 5.0; therefore, this pH was used throughout the remainder of the study. The effect of the ionic strength of the acetate buffer (at pH 5) on the voltammetric peaks was investigated; the largest signal was obtained with a supporting electrolyte strength of 0.06 rnol dm-3, as shown in Fig. 3. In order to determine whether or not the initial oxidation process was accompanied by adsorption phenomena, plots of Table 1 Values of ena for salicylic acid (peak Ia) PH aria 3.0 0.32 4.0 0.48 5.0 0.80 6.0 0.69 7.0 0.69 t .s 0 0.02 0.04 0.06 0.08 1.0 Ionic strength of acetate bufferhol dm-3 Fig. 3 Effect of ionic strength on peak current for peak Ia (see Fig. 1) using acetate buffer as the supporting electrolyte, containing 35% methanol, and 5 X 10-4 mol dm-3 salicylic acid 805 idCV versus V: were constructed for peak Ia, where C = concentration and V = scan rate. That adsorption did not occur at the electrode surface as no increase of current function (idCV) with V was apparent, is indicated in Fig. 4. The effect of the concentration of methanol on the peak current was examined over the range 8-35% v/v; the acetate buffer composition was fixed at a pH of 5.0 and an ionic strength of 0.06 rnol dm-3.The magnitude of peak Ia was found to increase by about 10% when the methanol concentra- tion was decreased from 35 to 8%. Therefore, over the range studied, the methanol concentration had little effect on the oxidation process; the small increase observed might be because of an increase in the solubility of salicylic acid, or changes in the conductivity of the electrolyte. These observations suggest that the mobile phase compo- sition, for the LCEC assay of salicylic acid, should be 0.06 mol dm-3 acetate buffer (pH 5.0), and that the methanol concentration can be varied between 8 and 35% with only a small change in the magnitude of the signal. Optimization of LCEC Conditions In order to determine the optimum applied potential for electrochemical detection, following HPLC, a hydrodynamic voltammogram was constructed for salicylic acid. The maxi- mum current was obtained at a potential of +1.5 V (versus an Ag-AgC1 electrode) or greater as shown in Fig.5. However, it is apparent that the background current is rather high at 1.5 V, which results in significant noise levels when achieving high sensitivities. It was found that much lower noise levels could be achieved with an applied potential of +1.35 V and only a small decrease in signal was observed. This resulted in improved signal-to-noise (S/N) ratios and consequently lower limits of detection. !2 11 3 10 A s 0 9 3 4 5 6 7 8 9 10 v: Fig. 4 Current function versus Vh for 5 x 10-4 rnol dm-3 salicylic acid in 35% methanol-0.06 rnol dm-3 acetate buffer, pH 5 1000 800 600 2 .3 400 200 0 0.8 1.10 1.2 1.4 1.6 1.8 E a pp I i e d N Fig.5 Hydrodynamic voltammogram for salicylic acid in 35% methanol-0.06 rnol dm-3 acetate buffer, pH 5.0. Concentration of salicylic acid standard, 5 x mol dm-3. Solid line, hydrodynamic wave for salicylic acid; and broken line, background current806 ANALYST. AUGUST 1991, VOL. 116 I 18 10 5 Ti me/m i n 0 Fig. 6 Chromatogram obtained by LCEC for extract from serum spiked with 50 ng of salicylic acid. Serum extract was dissolved in 0.1 ml of the optimum mobile phase consisting of 8% methanol-0.06 mol dm-3 acetate buffer, pH 5.0. Volume injected, 20 pl. SA = salicylic acid 1400 E $1200 c - E q 1000 0 Q) Q L 800 \ 0 2 600 .- 4- 8 400 E m V .- 6 200 .- - m v 0 I I I I I I I 1 2 3 4 5 6.7 8 Time/h Fig. 8 Concentration-time relation after a 300 mg oral administra- tion of aspirin on two volunteers (both men): solid line, volunteer 1; and broken line, volunteer 2 j L ” J l , , , , 1 0 200 400 600 800 1000 Salicylic acid added/ng per 0.5 ml of serum Fig. 7 &loo0 ng per 0.5 ml of serum Recovery of salicylic acid added to serum over the range The variation of retention time of salicylic acid with different percentage methanol composition (8-35%) was examined. The shortest retention time (4 min) occurred with a mobile phase containing 35% methanol-0.06 mol dm-3 acetate buffer, pH 5.0. However, when serum extracts were analysed with this eluent the analyte peak was not resolved from the interference peaks. In order to resolve the salicylic acid peak from interfering peaks, the methanol concentration was reduced to 8% (Fig.6). These conditions were then considered suitable for the assay of salicylic acid following oral doses of aspirin. Calibration, Recovery and Precision of LCEC Assay The calibration graph of peak current versus the mass of salicylic acid injected was linear over the range 4-200 ng of mass injected. The recovery of salicylic acid added (30-500 ng per 0.5 ml of serum) to serum was linear, with a mean recovery of 60% (Fig. 7) and the relative standard deviation for 5 samples spiked with 100 ng per 0.5 ml of serum and for 5 samples spiked with 10 ng per 0.5 ml of serum was 2.2 and 5.9% , respectively. The limit of detection was found to be 4 ng of salicylic acid injected; this value was based on an S/N ratio of 3 : 1 and a full-scale deflection of 10 nA.Analytical Application It was of particular interest to ascertain whether the present method was suitable for studies on the absorption rate of salicylic acid following an oral dose of aspirin. The pharmaco- kinetic curves obtained for two healthy male subjects follow- ing a 300 mg oral dose of aspirin are shown in Fig. 8. These results are in good agreement with previously published reports.8 This indicates the reliability of the present method at the dosage level administered. It has recently become desirable to determine the minimum level of serum salicylic acid able to reduce platelet activity,gJo which becomes raised in coronary heart disease. Thus, a sensitive assay for low levels of this drug could be of major importance in such clinical studies. Bearing in mind that the present assay has a detection limit of 4 ng injected, we intend applying it to patients undergoing treatment with low level doses of aspirin. The authors thank M. Norman and R. Shadwick of Bristol Polytechnic for technical assistance. \ \\ 1 2 3 4 5 6 7 8 9 10 References Jackson, J. V., Moss, M. S., and Widdop, B., Clarke’s Isolation and Identification of Drugs, Pharmaceutical Press, London, 1984, p, 965. Stewart, M. J., and Watson, I. D., Ann. Clin. Biochem., 1987, 24,552. Selinger, K., and Purdy, W. C., Anal. Chim. Actu., 1983, 149, 343. Tebbett, I. R., andOmile,C. I.,J. Chromatogr., 1985,329,196. Hart, J. P., and Hayler, P. J., Anal. Proc., 1986, 23, 439. Handbook of Chemistry and Physics, ed. Weast, R. C., CRC Press, Boca Raton, FL, 1974, p. D-129. Sykes, P. , A Guidebook to Mechanisms in Organic Chemistry, Longman, London, 1970, p. 62. Bochner, F., Williams, D. B., Morris, P. M., Siebert, D. M., and Lloyd, J. V., Eur. J. Clin. Pharmacol., 1988, 35, 287. Hirsch, J., Stroke, 1985, 16, 1. Sinzinger, H., Kaln, J., Fischa, P., O’Grady, J., Prostaglandins Leukotrienes and Essential Fatty Acids, 1988, 34, 89. Paper I I009396 Received February 2nd, I991 Accepted April 2nd, 1991

ISSN:0003-2654    

DOI:10.1039/AN9911600803

出版商:RSC    

年代:1991

数据来源: RSC

摘要:

ANALYST, AUGUST 1991, VOL. 116 807 Improvement on the Microdiff usion Technique for the Determination of Ionic and Ionizable Fluoride in Cows' Milk Jacobus F. van Staden Department of Chemistry, Faculty of Science, University of Pretoria, Pretoria 0002, South Africa Sophia D. Janse van Rensburg Center for Stomatological Research, Faculty of Dentistry, University of Pretoria, Pretoria 0002, South Africa ~~~ ~~~ Different microdiffusion techniques for the determination of ionic and ionizable fluoride in milk have been evaluated for measurement with a fluoride-selective sensor. This work culminated in a modified version of the hexamethyldisiloxane-acid diffusion technique where an increased amount of perchloric acid and sodium hydroxide seemed to be a prerequisite for accurate and precise results.The configuration of the fluoride-selective electrode was also modified by using an adapted microanalytical procedure to determine fluoride in small volumes (50 p1 each) and to avoid contamination between successive milk samples. The resultant procedure is free of interferences and is capable of measuring fluoride in milk at concentrations from 0.02 to 10.00 mg dm-3 with improved accuracy and precision compared with earlier work. Keywords: Milk; fluoride determination; microdiffusion; fluoride-selective electrode The effect of maternal fluoride intake on the fluoride content of breast milk forms an important part of the ultimate fluoride intake of infants. According to the literature,' there is no change in the fluoride content of milk, regardless of the fluoride intake of the mother.Despite this, the documented val~es2~3 for the fluoride content of cow's milk still span a wide range between 0.04 and 0.8 mg dm-3. With a fluoride content of 0.8 mg dm-3, milk can be an important source of fluoride during the mineralization period of teeth in children.2 However fluorosis may occur when additional fluoride is taken and the individual is overdosed. It would therefore be an advantage to have a reliable method available to detect small changes in the concentration of fluoride in milk. According to Dabeka et al. ,4 the variation in the documented values for the fluoride content in milk could be due to the variety of analytical methods that are used. Although the response time of the lanthanum trifluoride ion-selective electrode is relatively fast, giving rapid and reliable fluoride measurements, the pre-treatment of small volumes of biological samples with very low fluoride contents is not easy to perform.A variety of procedures for the determination of fluoride in biological fluids have been described, ranging from direct potentiometry,2,5 hydrolysis and pyrohydrolysis,5 adsorption on calcium phosphate,6 different ashing procedures2-7 and different microdiffusion techniques using hydrochloric,7~8 sulphuric9 and perchloric acids10 for the release of ionic fluoride. Despite the develop- ment of relatively simple, accurate and reliable analytical techniques for the determination of ionic fluoride, the values reported from different laboratories continue to span a wide range and many differences of opinion still exist.This study was initiated because of the controversy sur- rounding the actual concentrations of ionic and ionizable fluoride in milk, measured using the various reported methods, and because of the potential for fluoride in milk to act as an anticariogenic agent. It is possible that the differences in the values obtained for fluoride can be attributed to the pre-treatment of milk samples and an investigation to evaluate different microdiffusion procedures for the determination of free and ionizable fluoride in milk was undertaken. Experimental Solutions All solutions were prepared from analytical-reagent grade chemicals unless otherwise specified, dissolved in doubly distilled, de-ionized water and stored in polyethylene con- tainers. Sodium hydroxide solution, 0.5 rnol dm-3.Dissolve 20.0 g of NaOH carefully in 600 cm3 of distilled water, cool to room temperature and dilute quantitatively to lo00 cm3 with distilled water. A 0.05 mol dm-3 solution of NaOH was prepared by suitable dilution with distilled water. Acetic acid solution, 0.2 mol dm-3. Dissolve 11.6 cm3 of acetic acid (99.5 per cent by mass; relative density = 1.05; and concentration = 17.4 mol dm-3) in 500 cm3 of distilled water and dilute to lo00 cm3 with distilled water. H2S04 solution, 1.5 mol dm-3. Dissolve 84 cm3 of H2S04 (96 per cent by mass; relative density = 1.84; and concentra- tion = 18.00 mol dm-3) carefully in 800 cm3 of distilled water, cool to room temperature and dilute to 1000 cm3 with distilled water.Perchloric acid solution, 6.0 mol dm-3. Add 516 cm3 of 70 per cent by mass HC104 (relative density = 1.66; and concentration = 11.6 mol dm-3) to approximately 600 cm3 of distilled water and dilute to lo00 cm3 with distilled water. A 4.0 mol dm-3 perchloric acid solution is prepared by suitable dilution of the 6.0 rnol dm-3 perchloric acid with distilled water. HCl sohtion, 0.5 rnol dm-3. Add 44.5 cm3 of HCI (35 per cent by mass; relative density = 1.18; and concentration = 11.3 mol dm-3) to approximately 600 cm3 of distilled water and dilute quantitatively to lo00 cm3 with distilled water. Sulphuric (or perchloric) acid saturated with hexamethyl- disiloxane (HMDS). Saturate a 1.5 mol dm-3 H2S04 solution with HMDS by adding 10 cm3 of HMDS to 500 cm3 of 1.5 rnol dm-3 H2SO4 in a separating funnel and shaking it vigorously for 5 min.Prepare an HMDS-saturated HC104 solution by mixing 10 cm3 of HMDS with 500 cm3 of 6.0 mol dm-3 HCI04 (or 500 cm3 of 4.0 rnol dm-3 HC104) in a separating funnel and shaking it vigorously for 5 min. Use these solutions within 24 h. Standard calibration fluoride solutions. Prepare standard calibration fluoride solutions (containing 0.1,0.3,0.5,1.0,2.5 and 5.0 mg dm-3 of F-) by suitable dilution of the stock standard NaF solution containing 100 mg dm-3 of fluoride (Type S3596 from Radiometer, Copenhagen). Add a fixed aliquot of a background ionic solution containing total ionic strength adjustment buffer (TISAB) (TISAB 11, Orion Cat. No. 940909), 0.5 mol dm-3 HCl and 0.5 mol dm-3 NaOH to each fluoride standard solution, and dilute with distilled water.This gives standard calibration fluoride solutions with the same ionic background as is used for sample measurement.808 ANALYST, AUGUST 1991, VOL. 116 Standard fluoride solutions for recovery experiments. Pre- pare standard fluoride solutions (containing 0.02, 0.05, 0.1, 0.2, 0.3, 0.5, 1.0, 5.0 and 10.0 mg dm-3 of F-) by suitable dilution of the stock standard NaF solution containing 100 mg dm-3 of fluoride (Type S3596 from Radiometer) with distilled water for the recovery experiments, in order to evaluate the accuracy and precision of the methods. The standard fluoride solutions are then treated in the same way as the milk samples. The main purpose of these standard solutions is for evaluation of the sample pre-treatment procedures of different diffusion techniques, as discussed later.Sample Collection Milk was collected from the cow’s teat in polyethylene bottles, the udder being pre-washed with distilled water. The fluoride contents of the cows’ drinking water were varied in order to establish a variation in the fluoride contents of the milk from the different cows. All samples were kept refrigerated and analysed within 12 h. Sample Pre-treatment Milk samples were treated using the microdiffusion technique reported by Whitfords and a modified microdiffusion tech- nique described by Spak et al.10 This was done in order to isolate and concentrate the fluoride in the samples. The following modified diffusion sample pre-treatment procedure was used.Pipette 3 cm3 of milk samples into the bottom parts of non-wettable plastic Petri-dishes. Spot the lids of the Petri-dishes carefully with 200 1-11 of 0.5 mol dm-3 NaOH solution, using a micropipetting system (Oxford instamatic solid displacement, Lancer, St. Louis, USA). Seal the lids tightly to the bottom parts of the dishes with vaseline. Add 6 cm3 of the HDMS-saturated 6.0 rnol dm-3 HC104 diffusion reagent through a small hole in the lid of the diffusion dish. Seal the small hole tightly with vaseline directly after addition of the acid. Diffuse the samples for 24 h at room temperature (22°C) in the sealed Petri-dishes. The ionic and ionizable fluoride are thus released from the sample and adsorbed by the NaOH solution on the lids. After diffusion, dry the lids in a desiccator.Dissolve the dried spots by adding 200 1-11 of 0.5 rnol dm-3 HC1 followed by 200 pl of TISAB buffer. The final volume is 400 pl and the pH of the solution is 5.1. Sample Measurement Following sample pre-treatment the final sample solutions (at pH 5.1) are analysed using a microanalytical procedure involving modificationsll-14 of the fluoride-selective elec- trode. The advantages are that not only are the interferences removed, but the fluoride contents of the samples being concentrated are above the detection limit of the fluoride- selective electrode. The amount of sample obtained from the sample pre- treatment is sufficient to allow three successive fluoride analyses of 50 1-11 each to be carried out. In this method the fluoride-selective electrode is filled with a fluoride electrode internal solution (0.1 rnol dm-3 KCl, 1 x 10-3 rnol dm-3 NaF saturated with AgCl and 1% agar gel) via a fine needle into a small hole just above the internal side of the membrane, taking care to avoid the formation of air bubbles.The hole is sealed with bees wax. The fluoride-selective electrode is inverted with the active membrane surface positioned vertically upwards. A liquid solution of KCI-agar gel forms a bridge between the reference electrode and the fluoride-selective electrode.13 A 50 pl aliquot of the pre- treated sample solution is pipetted onto the membrane of the electrode. The full sensing surface of the electrode was used with no adaptor positioned on the active membrane surface. The main reason for not using an adaptor is that the presence of a sleeve is a possible source of contamination.13 Fluoride measurements were performed using a Radiometer F1052F fluoride-selective electrode in conjunction with a Radiometer K4040 calomel reference electrode.The potentials were measured at room temperature with a Radiometer (Model Ion 85) ion analyser. The voltage was recorded after 2 min which was the time at which maximum equilibration at the electrode was achieved. The electrode measurements were linear over the whole concentration range except at the low concentration of 0.02 mg dm-3, where the calibration graph deviates from linearity . Recovery Experiments Recovery experiments were performed on standard solutions of fluoride and on milk samples. Standard Solutions The recovery of standard solutions containing fluoride at concentrations of 0.02, 0.05, 0.1, 0.2, 0.3, 0.5, 1.0, 5.0 and 10.0 mg dm-3 was evaluated by subjecting them to the same sample pre-treatment as the milk samples, followed by sample measurement.The main purpose of this operation was to evaluate the recovery of ionic fluoride using the different microdiffusion techniques described, compared with pure ionic fluoride standards at the primary starting-point , before proceeding to recovery experiments on milk samples for further evaluation. Milk Samples The recovery of ionic fluoride (0.02,0.1,0.2,0.5 and 1.0 pg of fluoride) added to milk samples was also evaluated following the same sample treatment and with the same aim. Results and Discussion Many a~thors~38-10 considered that the use of the HMDS-acid diffusion technique for the determination of ionic fluoride in biological fluids gave very satisfactory ,9 accurate and preciselo results.The development of the initial diffusion technique by TaveslsJ6 marked an important breakthrough for the determi- nation of ionic fluoride in biological samples, because it isolated the ionic fluoride in samples from interferences and concentrated the fluoride in the original sample to a smaller volume for measurement.9 This has the advantage that the analysed solution contains an ionic fluoride concentration above the limit of sensitivity of the fluoride-selective elec- trode, which improves the accuracy and precision.9 In addition to the normal requirements for good analysis (physical operations), it seems that the efficiency of the microdiffusion technique depends on the amount of sample used, the type, concentration and volume of acid added, and the concentration and volume of NaOH spotted on the lids of the Petri-dishes.The efficiency of the technique was evaluated in a series of recovery experiments. The results, using standard solutions with the procedure described by Whitford9 with 24 h diffusion, revealed a recovery of between 68 and 76% for fluoride concentrations ranging from 0.02 to 0.5 mg dm-3. This decreased sharply to 30% for the standard solution containing 1.0 mg dm-3 of F- and about 14% for the solutions containing 5.0 and 10.0 mg dm-3 of F-. The results also showed that the recovery from the same standard fluoride solution (0.02 mg dm-3) varied considerably. This clearly indicated that the dissociation of the HMDS molecule as described by Whitfords in the presence of 1.5 rnol dm-3 H2S04 is not quantitative and repeatable.It was also observed from results obtained when H2S04 was replaced by HCI that the results are again not quantitative and repeatable. Although 50 pl of 0.05 rnol dm-3 NaOH is only sufficient to trap about 0.047 mg of F- theoretically, the same percentage of recovery (68-76%) indicated that up to 0.5 mg of F- couldANALYST, AUGUST 1991, VOL. 116 809 Table 1 Recovery of ionic fluoride from standard solutions with the modified HMDS-acid diffusion technique Standard fluoride Fluoride mg dm-3 mgdm-3 Recovery*( %) solution used/ recovered/ 0.020 0.050 0.100 0.200 0.300 0.500 1 .000 5 .OOO 10.000 0.019 0.048 0.101 0.195 0.303 0.476 0.879 4.300 8.200 95.0 f 5.7 96.0 f 5.7 101.0 k 5.8 97.5 f 5.6 101.0 f 5.6 95.2 k 5.5 87.9 f 5.2 86.0 f 5.0 82.0 f 4.9 * Average of 14 tests in each instance, with the appropriate standard deviation. be trapped experimentally and that the lack of NaOH solution is not the main cause and did not contribute significantly to the low precision and recovery obtained.The NaOH solution however played a significant role for concentrations above 0.5 mg dm-3 of F-. The evaluation of the recovery of ionic fluoride with standard solutions according to the procedure described by Spak et al.10 (50 yl of 0.5 rnol dm-3 NaOH, 2 cm3 of 4.0 rnol dm-3 HC104 and 24 h diffusion) showed a recovery of between 82-85% for fluoride concentrations in the range 0.02-0.5 mg dm-3, which is an improvement on the 68-76% obtained with Whitford’s method.9 The recovery decreased to about 66% for solutions containing 0.5-5.0 mg dm-3 of F- and to 50% for those containing 10.0 mg dm-3 of fluoride.It is clear from these results that Spak’s method10 showed an over-all improvement on the reagents used by Whitford.9 Two factors contributed to the improvement. The HC104 solution seemed to be more efficient in dissociating the HMDS molecule, which contributed significantly to the release of more fluoride. The recovery was also enhanced by increasing the concentration of NaOH on the lid from 0.05 mol dm-3 used by Whitford9 to 0.5 rnol dm-3 as used by Spak et al. 10 The precision also improved, giving a coefficient of variation of 19.5% for 14 tests.However, it was clear that the amount of HC104 was still insufficient for the dissociation reaction to proceed to the extent where all the ionic fluoride was released. Our results also showed that the amount of NaOH solution was insufficient for trapping all the released fluoride. With these two factors in mind, the HMDS-acid diffusion technique was modified by increasing the amount and concentration of HC104 to 6 cm3 and 6.0 rnol dm-3, (twice the volume of the milk sample used) and the NaOH solution to 200 yl (instead of 50 PI) and 0.5 rnol dm-3. The recovery results obtained for fluoride standard solutions with concen- trations from 0.02 to 10.0 mg dm-3 using this method are given in Table 1. It followed from the results that a significantly increased amount of fluoride is transferred to the NaOH, 95.0-101% for solutions containing 0.02-0.20 mg dm-3 of F-.The results also indicate an improvement in recovery for the fluoride solutions with higher concentrations ranging between 0.5 and 10.0 mg dm-3 of F-. The precision also improved giving a coefficient of variation of 6.0% for the 14 tests. The recovery of different amounts of fluoride added to actual milk samples is shown in Table 2, which shows an improvement in the accuracy in fluoride transfer and entrap- ment with the proposed modified diffusion technique. The coefficient of variation was 7.4%. It is also clear from the percentage recovery obtained, that the added fluoride is not covalently bound to such an extent by the protein and Ca2+ ion in milk that it cannot be released sufficiently by the proposed modified diffusion technique.Table 2 Recovery of ionic fluoride from spiked milk samples with the modified HMDS-acid diffusion technique [F-] of milk Fluoride Recovery of used*/mg dm-3 added/pg fluoride added? (%) 0.300 0.037 0.065 0.075 0.067 0.175 0.097 0.067 0.067 0.075 0.067 0.055 1 .o 1 .o 1 .o 1 .o 0.5 0.5 0.1 0.1 0.1 0.1 0.02 0.02 100.0 k 6.8 86.0 k 5.9 100.0 f 6.3 91.0 k 6.5 88.0 f 5.7 88.0 f 5.8 97.0 f 6.6 94.0 & 6.6 99.0 * 6.7 99.0 k 6.7 100.0 k 5.6 100.0 & 5.6 * Mean result as determined by microdiffusion and direct poten- -t Average of 14 tests in each instance with the appropriate standard tiometry (accuracy and precision taken into consideration). deviation.Table 3 Comparison of the results for milk samples obtained with the proposed modified HMDS-acid diffusion technique and direct potentiometry HMDS method [F-]/mg dm-3 0.156 0.241 0.122 0.092 0.061 0.079 0.094 0.098 0.134 0.120 0.079 0.141 0.109 0.060 Direct potentiometry [F-]/mg dm-3 0.162 0.214 0.126 0.094 0.078 0.080 0.074 0.092 0.106 0.104 0.078 0.108 0.090 0.066 The fluoride-selective electrode tends to become unstable, owing to the deposition of organic material from the milk samples on the sensor over a period of time in direct potentiometry.2.5 The time involved depends on the amount and nature of protein in milk. In the present study, after between 2 and 9 determinations the membrane surface had to be cleaned before further analysis, Direct potentiometry is therefore avoided in the determination of fluoride in milk on a routine basis, however, for 1-2 determinations at a time, before cleaning of the membrane surface, it acts as a good comparative method.In the present study the membrane surface was cleaned between individual determinations and the electrode calibrated. The accuracy of the proposed modified HMDS diffusion technique was confirmed by comparing the results obtained with direct potentiometry (Table 3). The results indicate an acceptable level of agree- ment between the two methods. The proposed modified HMDS diffusion technique however offers the following advantages over direct potentiometry. The fluoride contents of the samples are concentrated by the diffusion technique to a level above the detection limit of the fluoride-selective electrode, which is not possible with direct potentiometry; an advantage which increases the accuracy and precision. The active membrane surface of the LaF3 crystal is not contami- nated or blocked, as any interferences are removed by the diffusion technique and therefore contamination between successive samples is avoided.It is also not necessary to prepare different calibration standards for samples with different matrices as the same background ions are present in810 ANALYST, AUGUST 1991, VOL. 116 the diffused standards and samples. The percentage recovery of the added fluoride is higher with the proposed diffusion technique. It can be concluded that the proposed procedure is free of interferences and can be used to measure fluoride in milk in the concentration range 0.02-10.00 mg dm-3 with improved accuracy and precision compared with earlier work.The accuracy in the concentration range 0.022-0.50 mg d ~ n - ~ of F- is superior to that in the 1.00-10.00 mg dm-3 range. This project was supported financially by the Medical Research Council of South Africa, the Foundation for Research Development (FRD), Pretoria and the University of Pretoria. References 1 Ekstrand, J., Boreus, L. O., and de Chateau, P., Br. Med. J. 1981,283,761. 2 Backer Dirks, O., Jongeling-Eijndhoven, J. M. P. A., Flissel- baalje, T. D., and Gedalia, I., Caries Res., 1974, 8, 181. 3 Larsen, M. J . , Senderovitz, F., Kirkegaard, E., Poulsen, S., and Fejerskov, O., J. Dent. Res., 1988, 67, 822. 4 Dabeka, R. W., Karpinsky, K. F., Mckenzie, A. D., and Bajdik, C. D., Food Chem. Toxicol., 1986,24,913. 5 Duff, E. J., Caries Res., 1981, 15, 406. 6 Venkateswarlu, P., Singer, L., and Armstrong, W. D., Anal. Biochem., 1971,42, 350. 7 Esala, S., Vuori, E., and Helle, A., Br. J. Nutr., 1982,48, 201. 8 Fry, B. W., and Taves, D. R., J. Lab. Clin. Med., 1970, 75, 1020. 9 Whitford, G. M., in Monographs in Oral Science, ed. Myers, H. M., Karger, Basle, Switzerland, 1989, pp. 1-30. 10 Spak, C. J., Hardell, L. I., and de Chateau, P., Acta Paediatr. Scand., 1983, 72, 699. 11 Venkateswarlu, P., Clin. Chim. Acta, 1975, 59, 277. 12 Hallsworth, A. S., Weatherell, J. A., and Deutsch, D., Anal. Chem., 1976,48, 1660. 13 Vogel, G. L., Chow, L. C., and Brown, W. E., Caries Res., 1983, 17,23. 14 Retief, D. H., Summerlin, D. J., Harris, B. E., and Bradley, E. L., Caries Res., 1985, 19, 248. 15 Taves, D. R., Talanta, 1968, 15, 969. 16 Taves, D. R., Talanta, 1968, 15, 1015. Paper 0105331 G Received November 27th, 1990 Accepted April 11 th, 1991

ISSN:0003-2654    

DOI:10.1039/AN9911600807

出版商:RSC    

年代:1991

数据来源: RSC