ISSN:0003-2654    

DOI:10.1039/AN99217FX005

出版商:RSC    

年代:1992

数据来源: RSC

ISSN:0003-2654    

DOI:10.1039/AN99217BX007

出版商:RSC    

年代:1992

数据来源: RSC

摘要:

ANALYST, FEBRUARY 1992, VOL. 117 117 On-line Microwave Digestion of Slurry Samples With Direct Flame Atomic Absorption Spectrometric Elemental Detection Stephen J. Haswell and David Barclay School of Chemistry, University of Hull, Hull HU6 7RX, UK A flow injection (FI) system for on-line microwave digestion of slurried samples with direct elemental determinations by flame atomic absorption spectrometry is described. Organically based elemental reference samples were prepared as slurries in 5% v/v HN03 and the system was optimized for slurry mass, acid strength and tube and microwave cavity geometry. Bubble formation during digestion was controlled by post-digestion cooling and pressure regulation. Comparison of direct and FI calibrations indicated no apparent loss in sensitivity. Various samples were examined and elemental recoveries for Car Fe, Mg and Zn were typically found to be in the range 94-107% with precisions of less than 4.5% relative standard deviation.The major source of error was found t o be in the dispersion of solids (<I80 pm) as slurries in dilute HN03. The throughput of samples in the system developed was found t o be 1-2 min per sample. Keywords: Microwave digestion; on-line digestion; flame atomic absorption spectrometry; elemental analysis; flow injection The decomposition of samples by microwave digestion prior to trace elemental determination has now become a popular method with many analysts for a wide range of matrix types.' To date, the majority of digestions described have been based upon a batch technique, in which the microwave digestion replaces traditional wet or dry ashing methodology.2.3 The move to a microwave digestion approach offers many advan- tages over the conventional methods including reduction in digestion time, digestion of difficult matrices and dissolution in what is essentially a closed environment , which reduces volatile analyte loss and atmospheric Contamination.' Despite these obvious advantages, the batch type of approach to sample digestion is still prone to contamination problems associated with the sample, reagent and containment together with analyte loss and potential errors from volumetric transfers.Many of these problems can be overcome or controlled by adopting a flow injection (FI) methodology.4 It would seem therefore that the advantages of a microwave digestion approach to sample dissolution could be further improved by incorporating the digestion into an FI manifold.There have been only a few reports of systems based upon this approach54 to date, and none of the methodologies describes a continuous-flow system with on-line microwave digestion. This paper reports the development of such an FI system with on-line microwave digestion coupled directly to an atomic absorption (AA) spectrometer for elemental determination. Experimental Apparatus A schematic diagram of the on-line digestion system is shown in Fig. 1 and was assembled as follows: (i) an Ismatec pump MV-Z; (ii) a Rheodyne injection valve (Anachem 5020) with a 1 ml sample injection loop; (iii) a CEM MDS81 microwave oven containing 20 m of 0.8 mm i.d.poly(tetrafluoroethy1ene) (PTFE) tubing; (iv) a 5 m cooling loop in an antifreeze bath cooled by six Peltier devices (MI 1069T-O3AC, Marlow Industries, Tadworth, UK); (v) a Rheodyne injection valve (Anachem 5020) with an in-line back-flush filter fitted in the place of the injection loop; (vi) a CEM pressure sensor; and (vii) a 516.75 kPa (75 psi) back-pressure regulator (Anachem P736). The outlet from the back-pressure regulator was coupled directly to the nebulizer of an AA spectrometer (Thermoelectron 357) using 0.8 mm i.d. PTFE tubing (Anachem 33-1331). The same tubing was used throughout the system for all couplings and loops. A magnetic stirrer for slurry preparation and a chart recorder (Linseis LS52) for data collection were also used.Reagents Concentrated nitric acid and all 1000 ppm standards were of AnalaR quality supplied by Merck (Poole, Dorset, UK) and the water used was distilled, de-ionized. The following Certified Reference Materials (CRMs) were used: Chlorella, Mussel, Sargasso and Pepperbush supplied by National Institute for Environmental Studies (NIES), Japan, together with a National Institute of Standards and Technology (NIST) Standard Reference Material (SRM) 1577a Bovine Liver. Procedure Calibration was carried out by filling the 1 ml sample loop of the injector with standards prepared in 5% v/v HN03 within the linear calibration range for each of the elements studied as follows: Ca, 0-2 ppm; Fe, 0-4 ppm; Mg, 0-0.4 ppm; and Zn, 0-0.8 ppm. In addition, calibration was performed directly reservoir Sample loop (1 mi) Injection valve Microwave Back-flush filter Injection valve Spectrometer Fig.1 system Schematic diagram of the on-line microwave digestion FI118 ANALYST, FEBRUARY 1992, VOL. 117 with the AA spectrometer, i.e., without the FI system in place. Slurries of the reference materials were prepared by accurately weighing an appropriate mass (50-500 mg) of the solid (t180 pm) into a beaker and adding dilute HN03 (5% v/v; 25-250 ml) using a pipette to obtain a slurry in the range 0.005-0.5% d v . The selection of a suitable percentage slurry was dictated by the elemental concentration in the particular sample but was chosen to give an acceptable minimum mass (not less than 25 mg) without producing a slurry of greater than 1%.It was observed that for some samples a slurry of >1% could cause blockages in the small apertures within the switching plate of the Rheodyne valve. The slurries were agitated by a magnetic stirrer for approximately 30 s, and an aliquot of approximately 2-3 ml was taken using a syringe (10 ml) to fill the 1 ml sample loop. The slurry was found to be stable for at least 10 h. The flow rate throughout the system was adjusted to match the optimum nebulizer uptake rate of the AA spectrometer. The operating parameters of the AA spectrometer were in accordance with the manufacturer’s recommended instrument conditions for each element using an air-acetylene flame. Absorption signals for both standards and slurries were recorded on a chart recorder and peak height measurements were taken.Replicate injections were perfor- med for standards and samples in the same experimental run for the appropriate element. Results and Discussion Calibration Calibration using a series of five standards in 5% v/v HN03 was carried out within the linear calibration range for each element by using the AA spectrometer in a conventional manner. The results from this direct nebulization were compared with those obtained by processing the same standards through the FI system with the microwave oven at 90% power. This established that no loss in instrumental performance occurred from analyte dispersion or mixing owing to the total length of tubing used and heating in the microwave cavity. The results in Table 1 indicate that, for standards in dilute acid (5% v/v HN03), no apparent loss in linearity or sensitivity occurs following the passage of a 1 ml slug of standard through the F1 system compared with direct nebulization.During the passage of the standards through the microwave cavity, out-gassing was observed as bubbles but these completely recondensed in the cooling loop. The flow rate through the system was maintained during this out- gassing period. The relevance of gas evolution is discussed in detail in the following section. Optimization of Digestion Conditions Given that the flow rate for the FI system was more or less limited to a narrow range (4-6 ml min-1) by the nebulizer uptake rate of the AA spectrometer, the digestion conditions or degree of dissolution in the microwave cavity were governed by three variables, namely, microwave power, tube length and acid slurry strength.As one of the objectives of this work was to minimize the time taken for the digestion whilst maintaining sensitivity, power ratings from 0 to 90% were evaluated for both signal sensitivity and digestion. Signal sensitivity was evaluated by comparing the absorbance response for a standard solution of 4 ppm of Cu in 5% v/v Table 1 Comparison of the calibration for Mg of the AA spectrometer with nebulization and an FI system Correlation System Slope Intercept coefficient Nebulizer 32.7 0.95 0.9994 FI 32.02 1.04 0.9999 HN03, over a range of microwave powers (Fig. 2). At microwave powers of below lo%, very little or no out-gassing occurred in the 20 m digestion loop and this led subsequently to a high degree of sample dispersion and a corresponding loss in sensitivity.In the range 10-55% power (stage 2), an increase in gas formation was observed that, by stage 3, led to the sample slug undergoing optimum fragmentation into discrete cells or sample fractions, which on cooling recom- bined to produce the original sample with minimal dispersion (Table 1). This rather unexpected consequence of bubble formation together with the efficiency of digestion observed for solid samples during the out-gassing process clearly represents an important process in the method described (a point returned to later). The results for slurry samples indicated that, as expected, a high power rating was preferable and that at 90% power (525 W) maximum digestion and sensitivity were achieved for the samples investigated.Selec- tion of 100% power was not found to give any improvement in results over those for 90% and so, in order to minimize heating of the oven components by continuous use, 90% power was selected for the work described. Having selected a fixed microwave power, the geometry of the tubing in the micro- wave cavity was evaluated. Various lengths (10,20 and 30 m) and internal diameters (0.3, 0.5 and 0.8 mm) of tubing were examined and these were either knotted into a ball or wrapped around the conventional 12 bomb holder as supplied by CEM for conventional microwave digestion. A satisfactory diges- tion was identified by observing the passage of the digested 1 ml slug of slurry (for this chlorella was used) for particulates following centrifugation, collected after the back-pressure 160 E 140 E 2 120 E 100 0, .- Y ca 80 2 60 40 ca - - -- - - Stage Stage * 2 3 Stage 1 1 I I I I I I I 0 10 20 30 40 50 60 70 80 90 Microwave power (%) Fig. 2 Plot of microwave power versus signal response for a Cu standard (4 ppm). Stage 1, insufficient power to cause out-gassing therefore little signal improvement; stage 2, increasing power gives increasing out-gassing therefore signal improves; and stage 3, no further increase in out-gassing therefore no further signal improve- ment, i.e., maximum sensitivity above 55% I m* Stage Stage ’ 1 0 5 10 15 20 25 30 Length of tubingh Fig.3 Percentage recoveries of Fe (chlorella 0.2% slurry in 5% v/v HN03) versus length of digestion tubing (0.8 mm i.d.) at a flow rate of 6 ml min-I.Stage 1, increasing time in microwave up to a point where maximum out-gassing occurs and digestion is fully underway; stage 2, increasing time in cavity and hence digestion without increasing dispersion; and stage 3, full digestion is accomplished for sampleANALYST, FEBRUARY 1992, VOL. 117 119 regulator with no in-line filter in place. The 0.3 mm i.d. tubing was found to block readily with slurries. The 0.8 mm i.d. tubing was found to give a longer residence time in the cavity over the 0.5 mm i.d. tubing for the same length with no obvious blocking effects or loss in sensitivity. By using 0.8 mm i.d. tubing it was found that, at lengths greater than 20 m and flow rates of up to 6 ml min-1, total digestion was achieved, i.e., negligible particulates remained (Fig. 3). Some particu- lates (2-5, visible by eye) were always present probably owing to poor dissolution associated with an unavoidable dilution or 3 4 5 6 7 HN03 in slurry (% v/v) Fig. 4 Percentage recoveries of Fe (chlorella 0.2% slurry) versus acid strength of the slurry. Stage 1, increasing digestion as out-gassing develops; stage 2, optimum digestion with out-gassing sufficient for digestion without disrupting the flow; and stage 3, flow disrupted and erratic owing to excessively violent out-gassing giving loss in repro- ducibility and increase in dispersion and sampling time dispersion effect of the 1 ml slug of 5% v/v HN03 at the trailing interface with the water carrier, characteristic of FI peaks. It was decided to place a back-flush filter in the system to protect the back-pressure regulator from possible block- ages, and to prepare for the likelihood of residual material with matrices that might be studied at a later date.This filter was constructed in-house from 4 mm diameter stainless-steel mesh (50 pm) housed in a modified zero dead volume coupling packed with acid-washed glass wool to minimize dispersion. As a 20 x 0.8 mm i.d. length of tubing was found to be adequate for digestion of all the samples analysed in this work, it only remained necessary to establish the best geometry for the tubing in the cavity. Knotting of the tubing, in an attempt to minimize dispersion, was found not to be advantageous because it led to local heating of the solid material and eventual blockage of the tubing, as it became physically stuck in the tight curves of the knot.The best geometry for the 20 m tubing was found to be when it was wound around the carousel designed conventionally for holding 12 digestion bombs; a modified version of the carousel was used in subsequent work. During the digestion of slurries various acid strengths were evaluated ((1-70% v/v HN03) by investigating the percentage recoveries of Fe from a 0.2% m/v slurry of chlorella in 5% v/v HN03 (Fig. 4). At acid strengths below 5% v/v HN03, the production of bubbles or a gas phase in the digestion loop was low which led to poor digestion and consequently lower signal recoveries. As indicated previously, the presence of a gas phase has been found to be essential for the digestion of samples and early attempts to remove the bubbles completely by increasing the back-pressure of the system was found to give poorer digestions of the slurry material.As yet it is not clear what the actual mechanism of digestion is but the Table 2 Percentage recoveries and precisions for reference materials, n = 10 Mg Ca Recovery RSD Recovery RSD (Yo) (Yo) (Yo) (Yo) Chlorella 96.3 2.3 99.0 3.1 Reference value 0.33% 0.49% CRM Mussel 97.2 1.4 100 7.6 Reference value 0.21% 0.13% CRM Sargasso 102 0.89 93.8 1.9 Reference value 0.67% 1.41% CRM Reference value 0.408% 1.38% SRM 1577a Pepperbush 96.5 0.8 103 1.9 - Bovine Liver 94.0 4.5 ND Reference value 600 pg g- 1 120 pg g- 1 *ND = Not detectable in a slurry of S0.5%. Zn Fe Recovery RSD Recovery RSD ND* - 97.5 1.5 20.5 pg g-1 0.185% (Yo) (%) ( Y O ) (Yo) 102 7.0 96.8 4.0 106 pg g- 158 pg g- ND - 95.4 3.3 16.4 pg g- 187 pgg-l 93.6 0.7 98.0 3.0 340 pg g- * 205 pg g- 1 95.9 0.7 107 4.3 123 pg g-1 194 pg g- Table 3 Relative standard deviations and percentage slurries for reference materials, n = 10 Mg Ca Zn RSD Slurry RSD Slurry RSD Slurry (%) (% m/v) (YO) ('7; m/v) (%) (YO m/v) CRM CRM CRM No.9 CRM SRM 1577a - - Chlorella 2.3 0.01 3.1 0.01 Mussel 1.4 0.01 7.6 0.01 7.0 0.5 Sargasso 0.89 0.005 1.9 0.005 - - Pepperbush 0.8 0.01 1.9 0.01 0.7 0.2 Bovine Liver 4.5 0.01 - - 0.7 0.5 Fe RSD Slurry (%) (YO m/v) 1.5 0.2 4.0 0.5 3.3 0.4 3.0 0.5 4.3 0.5120 ANALYST, FEBRUARY 1992, VOL. 117 presence of gas, liquid and solid phases in narrow bore tubing under the influence of heat from microwave power is significant and should form the basis of a more fundamental study.If, however, gas production is allowed to become too excessive then controlling the bubble size and flow charac- teristics through the digestion loop will lead to erratic flows and dispersion effects. These problems, which can be asso- ciated with increasing acid strength, led to a loss in signal sensitivity and as a consequence the apparent low recoveries observed in Fig. 4 for the higher acid concentrations. Obviously the degree of gassing will be a function of sample type and volume; the experimental conditions described in this paper were found to be adequate for the range of sample types studied. It may be necessary, however, to review the acid slurry strength, microwave power and pressure and length of the digestion loop in the microwave cavity for alternative sample types.Analysis of Samples The results, expressed as percentage recoveries and precision [relative standard deviation (RSD)], for the four elements determined by atomic absorption spectrometry (AAS) are summarized in Table 2. It was necessary to select different slurry masses or dilution factors for the various samples to facilitate obtaining a signal in the linear working range of the spectrometer. It was considered that sample masses below 25 mg would lead to unacceptable precision, and the percentage slurries (m/v) that were selected are summarized in Table 3. The elemental concentrations in some samples, for example Zn in chlorella/Sargasso and Ca in Bovine Liver, were found to be too low, using the maximum slurry concentration of 0.5%, to give a measurable signal.What is apparent from observations taken during experimental work is that it is not just the mass of the sample taken that influences precision but also the wettability or dispersion of the solid as a slurry that affects the results. For example Bovine Liver and Mussel were difficult to disperse in the 5% v/v HN03 and gave correspond- ingly higher RSD values. Surfactants were not used in this work but this is one possible area to investigate further. In general the results obtained were found to be acceptable in terms of recoveries and precision. Conclusion The method described offers rapid and efficient sample preparation using on-line microwave digestion of slurries with direct elemental detection by flame AAS. The sample preparation and analysis time for ten replicate samples was approximately 0.5 h for the method described compared with 2 h for the same number of samples prepared by the microwave bomb digestion technique.The elemental levels that can be determined are at present governed by the limited calibration range of the AA spectrometer and the current percentage slurry range used. Clearly this limitation could be overcome by the use of a wider dynamic calibration range such as that offered by inductively coupled plasma atomic emission spectrometry. Problems were experienced with the dispersion or wettability of some of the powdered (<180 pm) organic reference material used but at worst these gave precisions of 6 5 % RSD. Samples were processed in approximately 2 min and elemental recoveries for the samples studied were typically in the range 96107%. The authors gratefully acknowledge the contribution of Dr. P. Riby and L. Neville to the progress of this work. References Introduction to Microwave Sample Preparation, eds. Kingston, H. M., and Jassie, L. B., American Chemical Society, Washing- ton, DC, 1988. Fisher, L. B., Anal. Chem., 1986,58, 261. Millward, C. G., and Kluckner, P. O., J. Anal. At. Spectrom., 1989, 4, 709. RfiiiEka, J., and Hansen, E. H., Flow injection Analysis, Wiley, New York, 2nd edn., 1988. de la Guardia, M., Salvador, A., Burguera, J. L.. and Burguera, M., J. Flow Injection Anal., 1988, 5 , 121. Burguera, M., Burguera, J. L., and Alarcon, 0. M., Anal. Chim. Acta, 1986, 179,351. Paper ll03046I Received June 20, 1991 Accepted September 10, 1991

ISSN:0003-2654    

DOI:10.1039/AN9921700117

出版商:RSC    

年代:1992

数据来源: RSC

摘要:

ANALYST, FEBRUARY 1992, VOL. 117 121 High-pressure Microwave Digestion for the Determination of Arsenic, Antimony, Selenium and Mercury in Oily Wastes Milford B. Campbell and George A. Kanert Laboratory Services Branch, Ministry of the Environment, 125 Resources Road, Rexdale, Ontario, Canada M9W5Ll The determination of arsenic, antimony and selenium by hydride generation and mercury by the cold vapour technique, following acid digestion of oily waste samples, can be very difficult if there is any residual organic matter. The preparation stage usually requires prolonged digestion to oxidize the organic matter. The situation is exacerbated if these metals are present as the organometallic derivatives. A method has been developed in which the organic matrices of oily waste samples, after solvent extraction, are completely oxidized.This has been achieved through the use of a newly designed microwave system that uses special high-pressure vessels capable of withstanding internal pressures in excess of 82.2 bar (1200 psi) (1 bar = 105 Pa). The relative standard deviation using organometallic standards ranged from 5.4 to 6.6% and the recoveries for four spiked oily waste samples ranged from 89 to 105%. Keywords: Petroleum-based oily waste; microwave digestion; mercury, arsenic, antimony and selenium determination; cold vapour and h ydride-generation atomic absorption spectrometry The US Environmental Protection Agency Method 1330 'Extraction Procedure for Oily Wastes' is used to determine the mobile metal concentration in petroleum-based wastes' and hence provides a method for screening these types of wastes prior to landfilling.Owing to the expected low concentrations in these types of wastes, arsenic, antimony and selenium are determined by the hydride generation method whereas mercury is determined by the cold vapour method. Both of these methods usually require perchloric acid diges- tion of the samples in order to destroy any organic matter.273 Microwave digestion techniques have been used success- fully to digest different types of samples, ranging from geological4 to biological5 matrices. Fisher4 found that by using a low-pressure system, a decrease in the volatility of the more volatile elements led to improved sample to acid contact and hence a more complete digestion of the samples.Very little work, however, has been performed with these types of samples at internal vessel pressures in excess of 13.7 bar (200 psi). In this study, a microwave technique employing specially designed high-pressure digestion vessels was used to digest oily waste samples in a non-perchloric acid medium at pressures in excess of 82.2 bar (12 psi) (1 bar = 105 Pa). Experimental Microwave Digestion System The Milestone microwave system (Mandell Scientific, Guelph, Ontario) consists of three separate units: oven, fume extraction module and a capping station. The MLS-1200 oven is capable of producing a maximum power of 1200 W. An actual value of 1100 W was found by monitoring the change in temperature of 1 kg of water. The power can be varied from 1 to 50% in 1% increments and then by 25% increments to 100%.A microprocessor is used to control the power. For the acid digestions, a maximum of 50% power was used. High-pressure Vessels The Milestone high-pressure vessels (HPV 80) are constructed of a patented polymer derivative that allows the use of high boiling-point acids such as sulfuric acid at temperatures up to 350 "C. The vessels used had special seal rings and burst discs rated at 100 bar (1460 psi). Four vessels at a time were placed in the oven. Instrumentation A Varian Techtron Model 70 atomic absorption spectrometer capable of operating at 193.70, 217.58 and 196.03 nm for the determination of arsenic, antimony and selenium, respec- tively, was used. The atomizing chamber is a 10 cm long X 6 mm i.d. quartz tube wound with Chrome1 C insulated resistance wire (approximately 1 5;2 per 0.3 m).The operating temperature of 850 "C was controlled by a Variac transformer. The sample flow together with various reagents was maintained by use of a Gilson Minipuls 2 proportional pump (Varian Techtron). Tygon tubing of various inner diameters was used to connect the different units. The sample and reagents were mixed in the two mixing coils prior to introduction into the atomizer in a stream of argon via a centrally placed 2 mm i.d. quartz tube. The output from the atomic absorption spectrometer was fed into a strip-chart recorder set at an input voltage of 10 mV. A schematic diagram of the analytical system is shown in Fig. 1. An LCD/Milton Roy mercury monitor set at 254 nm was used to determine mercury.Reagents All inorganic acids used were of Baker analysed ACS grade, except sulfuric acid which was of BDH analytical-reagent grade. Doubly distilled water was used as a diluent. Toluene and tetrahydrofuran used for the extractions were from Caledon Laboratories (distilled in glass). Organometallic Standards No reference materials for metals in oily wastes were available. As the primary area of interest was in the metals that were present as the organometallic derivatives and not simply as particulates in oil, the following organometallic standards obtained from Conostan Division, Conoco (Ponca City, OK, USA) were used: antimony alkyl sulfonate in white mineral oil (5000 pprn); selenium amine sulfonate in white mineral oil (100 pprn); arsenic amine sulfonate in white mineral oil (100 pprn); and mercury alkyl dithiocarbomate in white mineral oil (100 pprn), A working mixed standard solution of 1 pprn was prepared by diluting the standards appropriately with toluene.122 Proportioning f Pump i L L - I ANALYST, FEBRUARY 1992, VOL.117 Argon 300 cm3 min-' r Heated quartz cell Recorder and atomic absorption spectrometer Sampler = 1:2 Fig. 1 Atomic absorption spectrometer manifold for the determination of arsenic and antimony Inorganic Standards The aqueous inorganic standards for arsenic, antimony and selenium were prepared by diluting lo00 mg dm-3 certified atomic absorption stock solutions (Delta Scientific). The mercury stock standard solution (100 mg dm-3) was obtained from J. T. Baker and diluted appropriately.Oily Wastes The following oily wastes were used: OW-1, oil-saturated clay; OW-2, tank bottom from oil refinery (lumpy solid); OW-3, sludge from oil refining (soft, oily solid); and OW-4, aqueous oily sludge (aqueous brown upper portion and solid lower layer). Owing to the high incidence of volatile pet- roleum products such as diesel fuel in some of the oily waste samples, the solids were air dried for 24 h in a fume hood prior to use. Procedure Duplicate oily wastes were Soxhlet extracted according to the US Environmental Protection Agency Method 1330.1 These extracts were composed of a mixture of toluene and tetra- hydrofuran together with the organics solubilized in them. The combined solvent extracts were filtered through a 0.46 pm glass-fibre filter to remove any large particulates that would pass through the extraction thimble.A 2 ml aliquot of the mixed solvent extracts was pipetted on to an 8.5 x 5 cm piece of Whatman QA-M quartz microfibre filter horizontally supported at one end. The filter was left at room temperature in a fume hood for approximately 20 rnin to allow for the bulk of the solvents to evaporate. It is important that blank determinations are performed on these filters prior to use; they were found to produce fairly high blanks for selenium and arsenic. Therefore, the filters were pre-treated by boiling in 6 mol dm-3 hydrochloric acid for approximately 15 min. The excess of acid solution was drained and the filters were then dried in the microwave oven followed by heating in a muffle furnace at 450 "C for at least 5 h.This treatment should produce blank values below 0.1 ppb for these elements. Each filter was cut into small pieces and then placed in a high-pressure digestion vessel. A 5 ml mixture of sulfuric and nitric acid (3 + 1) was carefully added to each of the vessels. The burst disc was placed in position, then the cap was screwed in place and tightened in the capping station to 10 N m. The vessels were inserted in the rotating safety shield and the whole arrangement was placed in the microwave oven. The samples were digested according to program 1: 20% power for 2 min, 30% power for 3 rnin and 50% power for 6 min. [Caution: It is important with these types of samples that a gradual rate of heating is used, otherwise the pressure build-up may be too rapid.It is also important that, if the waste sample is suspected to contain unstable compounds, the bombs are left in their safety shield in the oven long enough to allow adequate cooling and cessation of any chemical reac- tions. If this is not done, it is possible for one or more of the bombs to vent on removal from the oven.] At the end of the digestion period, the assembly was removed from the oven and immersed in cold water until the vessels had cooled sufficiently that they could be removed from the safety shield. The vessels were opened carefully with the venting tube in place (according to the manufacturer's instructions). In order to ensure complete digestion, a further 3 ml of the sulfuric-nitric acid mixture were added cautiously to each vessel and they were resealed as above.The samples were further digested according to program 2: 25% power for 2 min and 50% power for 10 min. At the end of this time, the vessels were cooled and opened as described above. Approximately 5-10 ml of distilled water were added slowly to each of the digests; care was taken at this stage as there was a vigorous reaction with the evolution of brown fumes. The solutions were allowed to stand for 5 rnin and then gently swirled. The filter pieces were then washed into clean 150 ml glass beakers with doubly distilled water. These solutions and the filters were completely colourless once all the nitrogen dioxide had escaped. The solutions were thoroughly stirred with a glass rod and then filtered through Whatman No.541 filter-paper into 50 ml calibrated flasks. After the initial solutions had drained from the filters, they were rinsed with small volumes of doubly distilled water. The filters were pressed with a glass rod to remove as much of the rinse solution as possible and the contents of the flasks were then diluted to the mark with doubly distilled water. It has been found from this work that the nitric acid remaining within the final solution suppresses the arsenic and antimony signals. In order to counteract this, 0.2-0.3 g of urea was added slowly to a 10 ml aliquot of each of the above solutions contained in 15 ml acid-rinsed sample vials. After the effervescence had subsided, the vials were capped and inverted several times to ensure complete mixing. These samples were then analysed for arsenic, antimony and selenium.ANALYST, FEBRUARY 1992, VOL. 117 123 For the determination of arsenic, antimony and selenium by the hydride generation atomic absorption method, it is important that these metals are in the correct oxidation state.The results obtained from the work performed by Sinemus et a1.6 show that much better sensitivities are obtained for these elements when they are in their lower oxidation states. In the present study, the reduction to the lower oxidation state was achieved by using 10% m/v potassium iodide for arsenic and antimony and 6 mol dm-3 hydrochloric acid for selenium. In each instance these solutions were automatically added by using a proportioning pump and mixed in a ten-turn glass coil as indicated in Fig.1. Reduction to arsine, stibine and hydrogen selenide was accomplished by the introduction of 2% sodium tetrahydroborate(i1i) into the flow system prior to the second mixing coil. Certain transition metals, such as nickel, copper and iron, are known to interfere in the determination of arsenic, antimony and selenium . 7 3 According to Kirkbright and Taddia,g the interference effect could be due to the reduction of these transition metals to the free metals; these could then absorb the hydride product, thus reducing the signal. It has been foundg-10 that 5 mol dm-3 hydrochloric acid extends the analytical range in which these analytes can be determined in the presence of these interfering metals. Inductively coupled plasma spectrometric studies have shown that one or more of these transition metals can be present, in excess of the concentrations known to cause interference problems,g in oily waste extract samples, especially if the particulate matter is not removed from the extract.For this reason, 6 mol dm-3 hydrochloric acid was mixed with the sample solution prior to the reduction stage. Mercury was determined by pouring the remaining diges- tate solutions into small glass beakers. Saturated potassium permanganate was added dropwise until a persistent pink colour was obtained (Landi et al.11 noted the merits of having a strongly oxidizing environment for the determination of mercury). The beakers were covered with watch-glasses, placed on a rotating platform and then inserted in the microwave oven. These samples were slowly heated at 15% power for 15 min.Samples that became colourless during this time were separately treated with potassium permanganate as above and re-heated. The manganese dioxide formed in this process was dissolved by careful treatment with 35% v/v hydrogen peroxide to the point where a pink colour was just obtained. If the hydrated manganese dioxide is not removed, it can hinder the oxidation of the mercury ions." The final solution for the determination of mercury was treated with a 20% m/v solution of hydroxylamine sulfate to reduce the potassium permanganate. The ionic mercury was then reduced by tin(ii) chloride to its elemental form in an automatic mixing system similar to that shown in Fig. 1. Digestion recovery studies were carried out in two stages.The first was by evaluating the recovery of the mixed standards only, and the second by evaluating the effects of the different oily waste samples under investigation on the added standards. In each instance, 0.5 ml of the 1 ppm mixed organometallic standard in toluene was pipetted on to the support and left to evaporate. In the second method, the 0.5 ml volume was added to the support after the solvent from the sample had evaporated. These samples were microwave digested in the same manner as the oily waste extracts. Results and Discussion The recoveries obtained from the digestion and analysis of mixed 10 ppb organometallic standards are shown in Table 1. The lower values found in the 8 ppb range were attributed to a slight loss of pressure in these particular vessels.The short-term precisions (relative standard deviations) are fairly consis tent. Results of the oily waste spiking study are presented in Table 2. The recoveries ranged from 90 to loo%, indicating Table 1 Recoveries (%) obtained from the digestion of the mixed 10 ppb organometallic standards Mercury 9.6 10.4 9.1 8.5 9.4 Average 9.4 Standard Relative deviation (YO) 0.622 standard deviation (Yo) 6.6 Arsenic Antimony Selenium 10.0 10.1 10.1 9.6 10.3 8.8 8.9 8.6 9.5 10.8 9.9 10.3 9.6 10.1 9.8 9.8 9.8 9.7 0.621 0.613 0.523 6.3 6.2 5.4 Table 2 Recovery data from the analysis of spiked duplicate oily waste samples Element Mercury Arsenic Antimony Selenium Parameter Concentration in extract + spike (PPb) (PPb) Average extract concentration Recovery at the 10 ppb spike level (Yo) Concentration in extract + spike (PPb) (PPb) Average extract concentration Recovery at the 10 ppb spike level (Yo) Concentration in extract + spike (PPW (PPb) Average extract concentration Recovery at the 10 ppb spike level (%) Concentration in extract + spike (PPb) (PPb) Average extract concentration Recovery at the 10 ppb spike level (%) Sample OW4 OW-2 OW-3 O W 4 9.7 11.3 10.3 9.6 0.6 1.3 0.8 0.6 91 100 95 90 9.6 9.4 11.4 10.3 0.4 0.5 0.9 0.8 92 89 105 95 10.1 10.5 11.8 9.8 0.7 0.8 2.1 0.3 94 97 97 9s 9.6 10.1 10.5 9.8 4 .3 " 0.5 0.7 <0.3* 96 96 98 98 * Estimated detection limit for selenium made from 30 ( n = S), where o is the standard deviation of the results obtained from the support and solvent blanks. that the digestion method was acceptable for the sample types investigated. Cross-contamination between samples was not a problem owing to the material used in the construction of the high-pressure vessels.Rinsing each vessel with 5% v/v nitric acid followed by rinsing with distilled and doubly distilled water was all that was required. Conclusions Petroleum-based oily waste samples are extremely difficult to oxidize completely via acid digestion. It has been found from earlier studies carried out under normal atmospheric condi-124 ANALYST, FEBRUARY 1992, VOL. 117 tions that residual organic matter in solution will result in numerous problems in the determination of mercury, arsenic, antimony and selenium. With the automated method of analysis used, coating of the mixing tube by an oily film occurs, which results in a gradual decrease in sensitivity.This film also results in drift of the recorder baseline and irregularly shaped peaks. With the high-pressure microwave system and using the method described, all of these problems were eradicated. One of the main disadvantages of this high-pressure technique is that there is no way of accurately monitoring the internal pressure and temperature inside the high-pressure vessels. It was established, however, that the internal pres- sures developed in the vessels were between 82.2 and 95.9 bar (1400 psi) by carefully monitoring the conditions under which a burst disc would rupture. This microwave technique, and other high-pressure systems that are becoming commercially available, will be increasingly used to digest samples with difficult matrices. The authors thank all those who assisted in this project, particularly T. McIntosh and H. Konstantinou for carrying out the hydride measurements and D. Russell for carrying out the mercury determinations. Special thanks are extended to J. Pimenta for technical support and providing different samples. References 1 Extraction Procedure for Oily Wastes, Method 1330, Test Methods for Evaluating Solid Waste, SW-846, United States Environmental Protection Agency, Washington, DC, 1986. May, K., and Stoeppler, M., Fresenius Z. Anal. Chem., 1984, 317, 127. Official Methods of Analysis of the Association of Official Analytical Chemists, Association of Official Analytical Chem- ists, Washington, DC, 13th edn., 1980, sect. 25, p. 110. 4 Fisher, L. B., Anal. Chem., 1986, 58, 261. 5 Abu-Samra, A., Morris, J. S., and Koirtyohann, S. R., Anal. Chem., 1975,47, 1475. 6 Sinemus, H. W., Melcher, M., and Welz, B., At. Spectrosc., 1981, 2, No. 3, 81. 7 Welz, B., and Melcher, M., Analyst, 1984, 109, 569. 8 Welz, B., and Melcher, M., Analyst, 1984, 109, 573. 9 Kirkbright, G. F., and Taddia, M., Anal. Chim. Acta, 1978, 100, 145. 10 Welz, B., and Melcher, M., Analyst, 1984, 109, 577. 11 Landi, S., Fagioli, F., Locatelli, C., and Vecchietti, R., Analyst, 1990, 115, 173. 12 Stainton, M. P., Anal. Chem., 1971,43,625. 2 3 Paper 1102546E Received May 30, 1991 Accepted August 6, 1991

ISSN:0003-2654    

DOI:10.1039/AN9921700121

出版商:RSC    

年代:1992

数据来源: RSC

摘要:

ANALYST, FEBRUARY 1992, VOL. 117 125 Critical Evaluation of Three Analytical Techniques for the Determination of Chromium(vi) in Soil Extracts Radmila MilaW, Janez Stupar, Nevenka Kofuh and Janez KoroSin Joief Stefan Institute, University of Ljubljana, 61000 Ljubljana, Jamova 39, Slovenia, Yugoslavia Three different analytical techniques [Ir5-diphenylcarbazide spectrophotometry, chelating ion-exchange chromatography (Chelex-IOO), and ion-pairing reversed-phase high-performance liquid chromatography (RP-HPLC) combined with electrothermal atomic absorption spectrometry (ETAAS)] were critically evaluated for the determination of CrV1 in soil extracts. Spectrophotometry was not applicable t o the analysis of most soil extract samples owing t o its high limit of detection (LOD = 30 ng cm-3), and the possibility of the instantaneous reduction of Crvl under the acidic conditions employed.A Chelex 100 column, although adequately sensitive (LOD = 1.5 ng cm-3), is inclined t o give higher results as inert and moderately labile Cr"l complexes partially passed through the resin together with CrVt. In addition, very small particles (€0.45 pm) carrying chromium can produce severe positive systematic errors. In order t o avoid this, filtration employing a 0.1 pm filter is recommended. Ion-pairing RP-HPLC was found t o be the most sensitive technique (LOD = 0.3 ng cm-3). It might also give high chromate results if negatively charged Cr1I1 complexes form ion pairs with tetrabutylammonium phosphate and their elution partially coincided with that of CrV1. Fulvate ligands showed this type o f interference.Reversed-phase HPLC is not suitable for analysis of extracts obtained from soils with freshly added tannery waste owing t o the effects of the undestroyed tannery waste matrix. This study showed that each method investigated was vulnerable t o some type of interference. Keywords: Chromate; spectrophotometry; chromatography; electrothermal atomic absorption spectrometry; soil extracts Chromium appears frequently as a pollutant in terrestrial systems. The hexavalent form is of prime concern because of its high toxicity. Studies by several workers of the formation and fate of Crvl in soil have produced controversial results. This may be owing to the lack of reliable analytical data on Crvl in soil extracts.A number of different analytical techniques are available for the determination of CrV1;1-26 however, most of them cannot be applied efficiently to complex samples such as soil extracts. Samples containing Crvl and Cr1I1 at nanogram levels have been concentrated on anion- and/or cation-exchange resins. 1-4 Investigations based on various extraction techniques5.6 combined with atomic absorption spectrometry (AAS) or inductively coupled plasma atomic emission spectrometry (ICP-AES) have also been carried out. Coprecipitation of CrV1 with lead sulfate7 is satisfactory for water and waste water samples which are low in chloride and sulfate ions (found to produce severe negative interferences). A flow injection method for the determina- tion8 of Cr042- and Cr"' was also reported.The 1,5-diphenyl- carbazide spectrophotometric method was used for the determination of Crvl in soil extract samples,9-14 but high positive interferences were also observed. 14 Several papers have been published on the determination of trace amounts of Crvl and Crlll in water samples using reversed-phase high- performance liquid chromatography (RP-HPLC) combined with AAS.ls-I9 A method has been developed20 for the determination of Crvl in aqueous samples or sample extracts using ion chroma- tography coupled with ICP mass spectrometry or colorimetry. Trace amounts of Crv' and Crrr1 in water samples have also been determined using chelating ion-exchange resins.21-2s The main advantage of these resins is their high purity. A comparative study26 of the lead sulfate coprecipitation, the 1,5-diphenyIcarbazide spectrophotometric and the chelation- extraction methods was carried out for the determination of Crvl in the presence of a large excess of Cr1I1.The aim of this work was to develop a reliable analytical technique for the determination of Crvl in soil extract samples. The 1,5-diphenylcarbazide spectrophotometric method and chelating ion-exchange chromatography (Chelex 100) and ion-pairing RP-HPLC separation, both combined off-line with electrothermal AAS (ETAAS) as a Cr specific detector, were critically evaluated. In addition, various interferences asso- ciated with the sample matrix were carefully investigated for each method examined. Experimental Apparatus and Procedures Detection systems A Varian Cary Model 16 spectrophotometer adjusted to a wavelength of 540 nm was used for the determination of Crvl by the 1,5-diphenylcarbazide spectrophotometric method .9 Separated chromium species (chelating ion exchange, RP-HPLC) were determined on a Varian AA 575 atomic absorption spectrometer with a Perkin-Elmer HGA 76B graphite furnace.The spectral bandpass of the monochroma- tor was 0.2 nm. A Varian Techtron hollow cathode lamp was operated at a current of 5 mA. The integrated absorbance of chromium was measured at the 357.9 nm line by injecting 15 mm3 of the sample. The atomization temperature was 2773 K. Background absorption was largely eliminated by careful control of the ashing conditions (ashing temperature between 1673 and 1873K, time 10-30s) and/or by a deuterium background corrector.Platform atomization was found to be superior to wall atomization. High chromate concentrations, employed in some RP-HPLC separations (interference studies), were measured by flame AAS (Varian AA-5, N20-acetylene flame). Separation system All soil extract samples were centrifuged with a Heraeus Model 17s Sepatech Biofuge at 10 000 rev min-1 for 20 min and filtered through 0.1 pm membrane filters, unless stated otherwise. Chelating ion exchange. A system was developed for the separation of chromium species employing Chelex 100 (50- 100 mesh) resin. The batch procedure of Isozaki et al.23 was modified into a column technique to obtain a better separation and lower limits of detection (LODs). Chelex 100 (Na form) resin was first transformed to the NH4 form23 and approxi- mately 0.3 g of the resin was slurried with water and transferred into the column (using a 1 cm3 plastic pipette tip) without prior purification.Quartz wool was placed at each end of the column, and the column was connected to a peristaltic126 0 ANALYST, FEBRUARY 1992, VOL. 117 II A B C 1 pump (Ismatec MS4 Reglo) through which the flow rates could be varied between 0.2 and 4.5 cm3 min-1. The column resin was first treated with 10 cm3 of buffer solution,27 followed by passage of a buffered sample (5 cm3 of sample and 5 cm3 of buffer) at a flow rate of 1 cm3 min-1 and then washed with 15 cm3 of buffer at a flow rate of 4.5 cm3 min-1. The sample eluent was collected in a beaker, acidified with 0.2 cm3 of nitric acid (1 + l ) , evaporated to approximately 2 cm3, transferred into a 5 cm3 calibrated flask and diluted with water to the calibration mark.Chromium not retained by the column was then measured by ETAAS. The same procedure was applied to the blank solution. Although the column capacity was not exceeded by performing 20 consecutive separations (solutions containing 80 ng cm-3 of Cr3+ and 40 ng cm-3 of C++), the multiple use of the same column for the separation of chromium in soil extracts resulted in an increased blank value. Therefore, the column was refilled with fresh resin for each sample analysis. Ion-pairing RP-HPLC. Separation of chromium species was accomplished on a column (150 x 4.6 mm i.d.) packed in-house with LiChrosorb RP CIS particles (5 pm). The eluent was pumped through the column at a rate of 1.0 cm3 min-1 by a gradient Merck-Hitachi 6200 intelligent pump.The sample was introduced onto the column by a Rheodyne injector equipped with a laboratory-built 5 cm3 loop. The minimum volume for reproducible injection of samples on the loop was found to be 12 cm3. Chromium(v1) was separated from other chromium species by the formation of an ion pair with the reagent, tetrabutylammonium phosphate (PIC A). The soil extract (2-10cm3) was mixed with the reagent and diluted with water to 15 cm3 so that the final reagent concentration was 1 x 10-2 mol dm-3. Standard and blank solutions were prepared in the same way and injected into the loop. The pump (operated at a flow rate of 1 cm3 min-1) was programmed in the following steps, which were determined experimentally to ensure quantitative and reproducible sepa- ration.( i ) The prepared sample (15 cm3) was injected into the loop and washed with water onto the column for 15 min (inject position). (ii) The injector was then turned to the load position to ensure the direct elution of the retained chromate by 50% v/v methanol-water. This phase was completed in 12 min. (iii) The column was rinsed with water for 33 min and was then ready for the next separation. The Crvt peak appeared 19 min after injection in a volume of 0.2-0.4 cm3. The. retention time depended on the pH and sample matrix and could vary; therefore, ten fractions of 0.2 cm3 were collected from the eighteenth minute after injection. Chro- mium(v1) was then determined in collected fractions ‘off-line’ by ETAAS.A typical chromatogram for Crvl is presented in Fig. 1. Reagents Merck Suprapur acids and doubly distilled water were used for the preparation of samples and standard solutions. All other chemicals were of analytical-reagent grade. 0.300 A standard Crlll stock solution (1000 pg cm-1) was prepared by dissolving 1.0 g of Cr metal powder (99.99%) in 20 cm3 of HCI (6 mol dm-3) followed by dilution to 1OOOcm3 with water. Chromium(1ir) citrate, maleate, oxalate and fulvate com- plexes (1000 pg cm-3) were prepared by mixing CrCI3 stock solution with an appropriate amount (3 : 1 ligand to Cr ratio) of citric, maleic and oxalic acid, respectively. Solid fulvic acid isolated from peat soil (relative molecular mass about 1000) was used to form chromium(rI1) fulvate.The concentration of fulvic acid in the solution (145 pg (3111-3) was more than adequate for complexation of added Crttl. A standard Crvl stock solution (1000 pg cm-3) was prepared by dissolving 2.828 g of potassium dichromate in 1000 cm3 of water. Chelating resin Chelex 100, Na form, 50-100 mesh was obtained from Sigma. Tetrabutylammonium phosphate (PIC A) (0.5 mol dm-3, Supelco) stock solution was used as an ion-pairing reagent. Methanol (Merck) for chromatography was employed in HPLC measurements. Formic acid-potassium hydroxide buffer solutions were applied27 for the pH range 2.5-5.0 and cellulose nitrate membrane filters, 0.45-0.05 pm, and 25 mm diameter (Sar- torius), were used in the filtration procedure. Sample Preparation Soil samples were prepared by shaking 2.00 g of moist soil for 2 h with 20cm3 of KH2P04 (1.5 x 10-2moldm-3), then centrifuging and decanting.The KH2P04 solution (1.5 X 10-2 mol dm-3) was used to extract efficiently water-soluble chromate and chromate sorbed on various oxides and clay particles.9 Samples were then filtered through 0.1 pm mem- brane filters and the concentration of total soluble chromium was determined by ETAAS. Aliquots of these solutions were used for the determination of different chromium species. It was found experimentally that filtering through 0.45 pm membrane filters did not remove particles from the solution efficiently. These fine particles containing chromium might produce large positive systematic errors if separation is performed on ion-exchange columns.This problem is virtually eliminated by filtering through 0.1 pm filters. Results and Discussion Parameters Influencing Chromium Speciation and the Study of Interferences Spectrophotometry Two similar procedures were tested: addition of acidified 1,5-diphenylcarbazide reagent to the sample9 (Procedure I), and addition of 1,5-diphenylcarbazide reagent to the sample before acidification14 (Procedure 11). Procedure I: a lcm3 aliquot of azide reagent was added to 10 cm3 of soil extract and the magenta colour was compared with standard Crv’ solutions at 540 nm after 20 min. For preparation of the azide reagent, 120cm3 of 85% v/v H3P04 were diluted with 280 cm3 of distilled water and added to 0.4 g of 1,5-diphenylcarbazide dissolved in 100 cm3 of 95% v/v ethanol.Procedure 11: a 0.4 cm3 aliquot of 1,5-diphenylcarbazide solution (0.1 g of 1,5- diphenylcarbazide dissolved in 10 cm3 of acetone) was added to 10 cm3 of soil extract. Then, 0.2 cm3 of H2S04 (20 g of 95% H2S04 dissolved in 80 cm3 of distilled water) was added to the sample. The magenta colour was compared with standard Crvl solutions at 540nm after 20 min. Analysis of soil extract matrices to which Crvl had been added indicated that both procedures produced similar results. The reproducibility of measurement for six parallel deter- minations of Crv’ (50 ng cm-3) was found to be k2%. The LOD (30) for Crv’ in aqueous standard solutions was 10ngcm-3. In soil extracts, the matrix influenced the sensitivity of the measurements. The LOD in soil extracts was found to be 30 ng cm-3, and therefore analysis of most soil extracts was not possible employing this technique. TheANALYST, FEBRUARY 1992, VOL.117 method could only be applied in particular situations where soils heavily polluted with Cr were investigated. Parameters influencing the determination of Cr"' in soil extracts. The influence of organically complexed Cr1I1 present in soil extract samples on the determination of Crvl was studied. To 1 pg cm-3 of Crvl, 2.5 pg cm-3 of Cr"' complexes were added and Crv' was determined by Procedure I and Procedure 11. The results are presented in Table 1. It can be seen from Table 1 that interferences from organically complexed Cr"' species were almost negligible in Procedure 11. The slightly lower results obtained by Procedure I can be explained by partial reduction of chromate in the presence of reductants under low pH conditions. Influence of particles and soil matrix on the determination of Cr"l in soil extracts.Clay and peat soil extracts were prepared as described previously. After centrifugation, the supernatant was separated from the solid residue either by decanting, decanting and filtering through a 0.45 pm filter or filtering through a 0.1 pm filter. The total concentration of soluble chromium in these soil extracts was below 1 ng cm-3. A 500 ngcm-3 concentration of Crvl was added to each soil extract and Crvl was determined by spectrophotometry (Procedure I). The influence of particles in the soil extracts and the influence of the soil matrix were studied. The results are presented in Table 2.Two effects influencing the results in Table 2 could be observed. The first is light scattering caused by particles in solution, which contributes to severe positive interferences, particularly in soils with a high clay fraction. It is therefore reasonable to expect that the results previously reported by Bartlett and James9 overestimate the oxidation of Cr"' to chromate in soils with a high content of MnOz. The formation of chromate from soluble Crrl' species in soils due to the presence of Mn02 was actually confirmed in our laboratory but the levels of chromate found in the soils were significantly lower. In addition, Heringer Donmez and Kalenberger14 by not performing filtration actually reported anomalously high Crvl levels in soil (28.7% clay) leachates, which is consistent with our observation.The second effect observed from Table 2 is a slightly lower recovery in filtered extracts. This is probably due to the partial reduction of chromate in the acidic medium of the 1,5-diphenyIcarbazide reagent by reducing substances in the soil extract. Chelating ion exchange-ETAAS Influence of pH on the sorption of chromium. The resin and standard solutions of Crl" (CrCI3) (200 ng cm-3) and Crvl (40 ng cm-3) were prepared at pH 3-5 in formic acid-potas- Table 1 Influence of organically complexed Cr"' on the determination of CrV' by the 1,5-diphenylcarbazide spectrophotometric method Procedure I Procedure 11 Addedpg ~ m - ~ CrV' CrV1 Cr"' CrV' pgcm-3 (%) pgcm-3 (%) 2.5 (Citrate) 1 .o 0.94 94 0.94 94 2.5 (Oxalate) 1.0 0.85 85 0.94 94 2.5 (Maleate) 1.0 0.98 98 1.00 100 2.5 (Fulvate) 1.0 0.89 89 0.95 95 found/ Recovery found/ Recovery Table 2 Influence of particles in soil extracts and soil matrix on the spectrophotometric determination of CrV' (Procedure I) Particle Sample CrV' added/ CrV1 found/ Recovery size/pm characteristic ng cm-3 ng cm-3 (Yo) >0.45 Clay soil 500 907 181 <0.45 Clay soil 500 47 1 94 (0.1 Clay soil 500 469 94 Peat soil 500 503 101 Peat soil 500 472 94 Peat soil 500 463 93 127 sium hydroxide buffer solutions.27 The efficiency of sorption as a function of pH is shown in Fig.2. A quantitative sorption of Crrrr was obtained in the pH range examined, while most of the Crvl passed through the column resin. Optimum con- ditions for the separation of Cr"' and Crvl were found at pH 3.5-4.5.A pH of 4.0 was chosen for further work, owing to the increased possibility of Crvl reduction in the lower pH range. Even at this pH, about 15% of added Crvl (40 ng ~ m - ~ ) was retained on the resin column. Separation of Cr"' and Crvl ions in synthetic mixtures. Synthetic mixtures of Cr"' (CrCI3) and Crvl (K2Cr207) solutions were prepared at pH 4.0 in various concentration ratios. Chromium(v1) was measured in the eluate. The results are presented in Table 3. The proposed separation of Cr"' from Crvl is satisfactory for CrV1 concentrations not exceeding 20 ng cm-3, probably because of the efficient elution from the resin column. Most of the soil extracts analysed were in this concentration range; at higher concentrations of Crvl, samples should be diluted prior to separation.The reproducibility of measurement, tested for six parallel determinations of Crvl (10 ng cm-3, was found to be +_5.5%. The LOD (30) for CrV1 in aqueous standard solutions and soil extracts was 1.5 ng cm-3. Influence of Cr"' complexes on the separation and determi- nation of Crv' in soil extracts. The existence of negatively charged low relative molecular mass organic complexes of chromium has been demonstrated in soil pore waters.28 Similarly, water-soluble Crlll in the soil solution is expected to be bound to some of the soil borne organic ligands. The presence of these Cr"' complexes in soil extracts might affect the determination of chromate. For this reason, the influence of negatively charged low and high relative molecular mass organic complexes of Crl'l on the determination of Crv' was studied in the concentration range expected in soil extracts. The results are presented in Table 4.It is evident that moderately labile and inert Crlll organic complexes passed partially through the resin column and apparently yielded higher Crvl values. Chromium(II1) citrate and fulvate produced the most severe positive interferences. As these ligands can be found in most soils and waste materials .- ; I s I I 3 4 5 PH Fig. 2 Efficiency of sorption of: A, chromium(m) (200 ng cm-3 of Cr3+) and B, chromium(v1) (40 ng cm-3 of @+), on Chelex 100 resin (50-100 mesh) as a function of pH Table 3 Separation of Cr"' (CrCI3) and CrV' (K2Cr207) ions in synthetic mixtures on Chelex 100 chelating resin (50-100 mesh) at pH 4.0 and determination of CrV1 by ETAAS Addedng cm-3 Cr"' CrV1 200 200 200 100 200 40 200 20 200 4 200 2 CrV' found/ ng cm-3 172 86 35.8 19.9 4.1 1.9 Recovery of CrV1 (% ) 86.0 86.0 89.5 99.6 103.0 95 .O128 ANALYST, FEBRUARY 1992, VOL.117 Table 4 Influence of organically complexed Cr"' on the determination of CrV1 by chelating ion exchange-ETAAS Addedng~rn-~ CrV' found Recovery Cr"' CrV1 ng cm-3 (%) 70 (Citrate) 30 75 .O 250 70 (Oxalate) 30 42.5 142 70 (Maleate) 30 32.5 108 70 (Fulvate) 30 58.2 194 in measurable amounts, analysis of soil extracts employing the separation technique described here will tend to give higher chromate results. Influence of particles on the determination of Crvl in soil extracts. Extracts of clay soils contain considerable numbers of particles below 0.45 pm, which pass through normal liquid chromatography columns.Therefore, the presence of such particles would lead to serious errors in chromate results if such columns were employed for separation. This effect was carefully examined. After centrifugation, soil extracts were filtered through 0.45,0.2 and 0.1 pm membrane filters and the concentration of chromate was determined in these filtrates. Decreasing Crvl values were observed. With KH2P04 (1.5 x 10-2 rnol dm-3) extracts, no further change in the Crvl results was observed by the use of a 0.05 pm filter. Thus, filtering through a 0.1 pm membrane filter was found to be satisfactory for efficient removal of particles from KH2P04 (1.5 x 10-2 rnol dm-3) soil extracts. On the other hand, aqueous soil extracts in the absence of electrolytes contain substantially more fine particles (light yellow colour).A 0.05 pm mem- brane filter should be employed in this instance if accurate Crvl results are to be obtained. Ion-pairing reversed-phase HPLC-E TAAS Influence of p H on the separation of CrVi. The variation of pH between 4.0 and 7.0 in aqueous standard solutions of chromate resulted in a slightly shifted retention time of Crvl during elution, but had no influence on the separation of Crvl. At pH values lower than 3, a substantial reduction of chromate ion concentration in the presence of electron donors was observed. Influence of C P on the separation of Crvl. A standard solution of chromium(n1) chloride (400 ng cm-3) was pre- pared alone or in synthetic mixtures containing 200- 400 ng cm-3 of Crvl.Positively charged Cr"' species did not form ion pairs with the PIC A reagent and showed no influence on the determination of Crvl. In addition, the calibration graph for Crv' in the presence of 400 ng cm-3 of chromium(II1) chloride was found to be linear in the concen- tration range 200-400 ng cm-3. The reproducibility of measurement tested for six parallel determinations of CrV' (1 ng cm-3) was found to be +4.5%. The LOD (30) for aqueous standard solutions was 0.2 ng cm-3 and for soil extracts, 0.3 ng cm-3. Influence of C P on the separation and determination of Crv' in soil extracts. Negatively charged complex species of Cr"' might form ion pairs with the PIC A reagent analogous to chromate.If the elution of these species coincides with that of chromate, an interference in the determination of the latter would result. These effects were therefore investigated carefully by employing several organic complexes of Crrrr typical of the soil environment. The results are presented in Table 5. It is evident that citrate, oxalate and maleate complexes of Cr1I1 had no influence on the determination of Crvl. In contrast, fulvate complexes formed ion pairs with the PIC A reagent that partially coincided with Crvl elution. The severe positive interference of chromium(Ir1) fulvate on the determination of Crvl (Table 5) was similar to that observed in the use of the chelating ion-exchange separation technique (Table 4). The complexation of Cr'Il with fulvic acid ligands in Table 5 Influence of organically cornplexed Cr"' on the determination of CrV' by ion-pairing RP-HPLC-ETAAS Addedng ~ m - ~ CrV1 found Recovery Cr"' CrV1 ng cm-3 (Yo) 70 (Citrate) 30 29.4 98.0 70 (Oxalate) 30 29.5 98.3 70 (Maleate) 30 29.3 97.6 70 (Fulvate) 30 55.7 185 Table 6 Effect of PIC A reagent concentration on the determination of CrV' in soil extracts (KH2P04, 1.5 x 10-2 mol dm-3) by ion-pairing RP-HPLC-ETAAS Peat soil Clay soil Concentration ofPICA CrV1 CrV1 CrV1 reagent/ added/ found Recovery found Recovery mol dm-3 ng cm-3 ng cm-3 (YO) ng cm-3 (%) 5 x 10-4 500 19 3.7 50 6.3 2 x 10-3 500 210 41.1 330 65.4 1 x 10-2 500 580 116.0 490 98.0 2 x 10-2 500 580 116.0 491 98.2 soil solution, therefore, leads to an overestimate of the chromate content of these soils.Influence of particles on the determination of CrV1 in soil extracts. It was found experimentally that small particles present in soil extracts did not influence the results of the determination of Crv' when separation was performed on the HPLC column. Obviously, the dense packing of the HPLC column, in contrast to the Chelex 100 column, retained particles of <0.45 pm. However, deposition of these particles in the column during consecutive determinations results in a continuous increase of the blank values and in column pressure. Consequently, the lifetime of the column was drastically reduced. The lifetime of the HPLC column was prolonged when soil extracts were filtered through 0.1 pm membrane filters. Influence of the concentration of PIC A reagent on the determination of Crvi in soil extracts.The PIC A reagent was used to form an ion pair with the chromate ions for preconcentration from natural pond-water samples. 18 The optimum concentration of the reagent in the sample solution was reported to be 5 x lO-4mol dm-3. Extraction of water-soluble chromate from a soil sample was performed using 1.5 x 10-2 mol dm-3 KH2P04. Owing to the similarity of Cr042- and H2PO4- ions, the latter should also form an ion pair with the PIC A reagent. Additionally, some similar ions that could react with the reagent might be present in the soil solution. Thus, the optimum concentration of PIC A reagent in soil extracts for quantitative formation of an ion pair with the chromate ion in soil should exceed 1 X 10-2 rnol dm-3.In order to prove this assumption, the following experiment was carried out. Chromium(v1) (500 ng cm-3) was added to each of two soil extracts (peat and clay soil extracted with KH2P04, 1.5 x 10-2 rnol dm-3) in which the contents of total soluble chromium were below 1 ng cm-3. Various concentrations of PIC A reagent were added to 10 cm3 of these solutions and the samples were diluted to 15 cm3 with water. The concentra- tions of PIC A in the final solutions for HPLC separation were between 5 x 10-4 and 2 X 10-2 rnol dm-3. Aqueous standard solutions of 500 ng cm-3 of Crvl were prepared with these concentrations of PIC A reagent. Recovery for the aqueous standard solutions of CrV1 was found to be between 98 and 100% for all concentrations of PIC A examined.The effect of the PIC A reagent concentration on the determination of Crvl in soil extracts (extracted with KH2P04, 1.5 X 10-2 mol dm-3) is presented in Table 6. An additional experiment with aqueous extracts of the same soils indicated about 90% recovery for added Crvl at a concentration of PIC A of 5 x 10-4 mol dm-3. From theseANALYST, FEBRUARY 1992, VOL. 117 129 observations and the data in Table 6, it was concluded that KH2P04 formed ion pairs with the PIC A reagent. The optimum concentration of the reagent depends on the KH2P04 concentration in the solution. A concentration of 1 x 10-2 rnol dm-3 of PIC A reagent was found to be optimum when 1.5 X 10-2 rnol dm-3 KH2P04 is used as the extractant solution. Analysis of Soil Extracts In order to evaluate the capability of the methods for the determination of chromium in soil extracts, various types of soil samples were selected and analysed for total soluble and hexavalent soluble chromium.The selection of soils was such as to provide a wide variety of sample matrices characterized either by the physico-chemical nature of the soil or by the soil and waste material together. The first group represented natural soils of different characteristics (clay, sandy, peat and acid soils) with a low and medium concentration of total chromium (40-120 pg g-1) and three natural serpentine soil samples with a high concentration of total chromium (480- 1300 pg g-1). The total chromium was determined by the procedure described previously.29 The second group consti- tuted some natural soils mixed with tannery waste, which were left to settle for 6 months under atmospheric conditions.The total chromium contents of these soils were 2300-3800 pg g-l. Finally, a clay field soil, which had been treated continuously Table 7 Determination of total Cr, total soluble Cr and soluble CrV' in KH2P04 (1.5 x 10-2 rnol dm-3) extracts of various natural soils by RP-HPLC-ETAAS, ~t = 3 Soil sample No. 1 2 3 4 5 6 7 8 9 10 11 Sample characteristic Sandy soil Clay soil Peat soil Clay soil Sandy soil Clay soil Serpentine soil Serpentine soil Serpentine soil Acid soil Acid soil Total soluble TotalCr/ Cr/ 45 2.4 115 4.5 40 2.9 93 17.0 75 5.6 92 16.5 1324 75.4 485 46.1 825 43.4 127 35.3 85 30.5 Mg-' ngg-' Soluble CrV1 ETA AS)/ <5 <5 <8 10 <5 8.8 53.9 27.2 36.9 (RP-HPLC- ngg-' pH of extract 6.0 5.5 6.3 5.7 5.6 5.7 5.4 5.6 5.9 3.8 4.2 for 17 years with tannery waste, was sampled 4 years after the last waste application.The total chromium concentration of this field was between 1000 and 1400 pgg-1, while the concentration on a nearby meadow, which was indirectly contaminated by tannery waste (wind), was 170 pg g-1. Samples were prepared in triplicate as described under Experimental and analysed by the appropriate method. All of the samples were extracted with KH2P04 (1.5 x 10-2 rnol dm-3), but some of the contaminated soils were also extracted with water in order to demonstrate the effects of chromate adsorption and the presence of colloidal particles in the extract solution on the results of the determination of chromium.The results of these measurements are sum- marized in Tables 7-9. The concentration of soluble (KH2P04, 1.5 X 10-2 mol dm-3) hexavalent chromium in natural soils was generally below the detection limit for spectrophotometry and chelating ion exchange-ETAAS. The RP-HPLC-ETA AS technique, being the most sensitive, was therefore used (Table 7). It is evident that the concentrations of soluble (KH2P04, 1.5 x 10-2 mol dm-3) chromate in natural soil samples are very low, with the exception of the serpentine soil samples. Despite the sensitive technique used (LOD = 0.3 ng cm-3) some of the concentrations were below the LOD. The inconsistency in the LOD shown in Table 7 reflects the differences in moisture contents of the soil samples considered.The pH of all the soil extracts was between 5.5 and 6.3 with the exception of the acid soil samples No. 10 (pH = 3.8) and No. 11 (pH = 4.2). When these two samples were analysed by RP-HPLC- ETAAS, the Crvl peak did not appear either in the soil extracts or in the soil extracts to which Crvl was added due to a low pH and the presence of reducing substances in these soils. Chromium(v1) added to the sample extract was reduced immediately. Despite the relatively high total soluble chro- mium concentration in acid soils (Table 7) the expected chromate concentration should be extremely low owing to the nature of these soils. Chromium(v1) was also added to the extracts of other soil matrices. Recoveries obtained were between 98 and 115%, which indicated that RP-HPLC-ETAAS was a suitable technique for the determination of soluble (KH2P04, 1.5 x 10-2 rnol dm-3) hexavalent chromium in natural soils with normal pH values.Table 8 Determination of total Cr, total soluble Cr and soluble CrV1 in KH2P04 (1.5 X by tannery waste using spectrophotometry, chelating ion exchange-ETAAS and RP-HPLC-ETAAS, n = 3 rnol dm-3) extracts of various soils contaminated Soluble Crv'/ng g-1 Soil sample No. I I1 111 IV V VI Sample characteristic Sandy soil* Clay soil* Peat soil* Clay soil? Clay soil? Clay soil? Total Cr/ 2360 2470 3730 1400 1050 169 Pg g-' Total soluble Cr/ 776 1621 960 384 233 37 ngg-' Chelating ion Spectro- exchange- photometry ETAAS 482 517 1245 1389 <750 658 <450 290 ~ 4 5 0 169 33 - RP-HPLC- ETAAS 434 909 822 297 184 34 * Tannery waste treated, analysed 6 months after waste application.t. Seventeen years of continuous tannery waste application, analysed in the fourth year after the last application. pH of extract 6.1 5.3 6.4 6.0 5.8 5.4 Table 9 Determination of total Cr, total soluble Cr and soluble CrV1 in aqueous soil extracts of various soils contaminated by tannery waste using chelating ion exchange-ETAAS and RP-HPLC-ETAAS, n = 3 Soluble Crv'/ng g-1 Soil Sample Total Total soluble Chelating ion RP-HPLC- pH of sample No. characteristic Cr/pg g- * Crhg g- exchange-ETAAS ETAAS extract IV Clay soil* 1400 V Clay soil* 1050 VI Clay soil* 169 238 140 48 183 92 38 143 6.7 62 6.8 <5 5.7 * Seventeen years of continuous tannery waste application, analysed in the fourth year after the last application.130 ANALYST, FEBRUARY 1992, VOL.117 Contaminated soil samples were analysed in triplicate by all three techniques (Table 8). Spectrophotometry could not be applied generally to the analysis of some of the contaminated soil extracts because of the poor LODs (30 ng cm-3 for Crvl). Results for the determination of Crvl obtained by RP-HPLC- ETAAS and chelating ion exchange-ETAAS agreed very well for soil samples treated with tannery waste, analysed 4 years after the last application (samples IV, V and VI). It was found experimentally that after consecutive RP-HPLC-ETAAS analyses of these samples, the blank value was constant but a slight broadening of the chromate peak appeared. When extracts of soil samples freshly treated with tannery waste are analysed (samples I, I1 and III), the influence of the waste matrix should be taken into account.Tannery waste is a protein-based matrix with a high content of CrlI1, organic polymers and reducing substances which could produce interferences in the determination of Crvl. In order to suppress these effects, the soil extracts were diluted 1 + 7.5 prior to RP-HPLC determinations and 1 + 2 prior to chelating ion exchange. The samples were not diluted for spectrophoto- metric analyses because of the poor LOD of the technique. It should be emphasized that with these types of soil extracts excessively high and variable blanks appeared when using RP-HPLC-ETAAS, making the results uncertain and in disagreement with those given by the other two techniques.In addition, the lifetime of the RP-HPLC columns was drastically reduced. Results using spectrophotometry and chelating ion exchange-ETAAS correlated well for these samples, pro- vided that the concentration of chromate in the soil extracts was >30ngcm-3. This good agreement indicates that the concentrations of organic ligands forming relatively inert Cr complexes, present in the samples should be low, otherwise much higher results would be obtained by chelating ion exchange-ETAAS. The accuracy of the result for the freshly treated peat soil sample (sample 111) obtained by chelating ion exchange-ETAAS is questionable as no reliable comparison could be made using the other two techniques. According to the expected interference effects of chromium(I1r) fulvate, the reported value might be too high.A comparison of the results from samples IV, V and VI (Tables 8 and 9) obtained by RP-HPLC-ETAAS reflected the effect of chromate adsorption on clay minerals, which was reported by Bartlett and James.9 On the other hand, from the results obtained by chelating ion exchange-ETAAS two effects were superimposed: the effect due to adsorption and the effect produced by the colloidal particles present in the aqueous extract. Conclusions Light scattering on colloidal particles can produce severe positive systematic errors in the spectrophotometric determi- nation of Crvl in soils containing a high clay fraction. Similarly, particles passing through the ion-exchange column contribute to higher results for Crvl. In HPLC separation, particles had no direct influence but reduced the lifetime of the RP columns. A KH2P04 (1.5 X mol dm-3) extraction solution efficiently released adsorbed chromate9 and pre- vented formation of colloidal solutions.Nevertheless, filtering through a 0.1 pm membrane filter was found to be necessary. A low LOD for Crvl in soil extracts (0.3 ng cm-3) using RP-HPLC-ETAAS enabled the determination of soluble hexavalent chromium in most natural soils. For soil samples contaminated by tannery waste, spectrophotometry was found to be suitable only for heavily contaminated soils. This technique may produce lower chromate results probably owing to partial reduction of Crvl during the measuring procedure in acidic media and a reducing environment. Analysis of samples treated with tannery waste, 4 years after the last application, showed very good agreement in the determination of soluble (KH2P04, 1.5 x 10-2 mol dm-3) Crvl between RP-HPLC-ETAAS and chelating ion exchange -ETAAS techniques, taking into account soil characteristics.Analysis of freshly treated tannery waste soil samples employ- ing RP-HPLC-ETAAS gave uncertain results because of the influence of the waste matrix and the variable blank. Spectrophotometry and chelating ion exchange-ETAAS results for these samples agreed well when the chromate concentration in the soil extracts was above 30 ng cm-3. Despite the considerable experimental efforts associated with the preparation of this paper, the reliable determination of chromium in soil extracts still remains a problem at least for some particular situations.Nevertheless, the experimental evidence presented demonstrates the complex nature of the effects of soil and tannery waste. This work was supported by the Research Council of Slovenia and US Environmental Protection Agency (project JF 908). The authors thank Professor S. A. Katz for valuable sugges- tions in preparing this manuscript and Dr. A. R. Byrne for assistance with linguistic correction. The authors also thank Marko Zupan, for providing some soil samples and their physical characterization. 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 References Fajgelj, A., and Kosta, L., Vestn. Slov. Kern. Drus., 1987, 34, 175. Pankow, J. F., and Januer, G. E., Anal. Chim. Acta, 1974,69, 97. Naranjit, D., Thomassen, Y., and Van Loon, J. C., Anal. Chim. Acta, 1979, 110,307. Minoia, C., Mazzucotelli, A., Cavalleri, A., and Minganti, V., Analyst, 1983, 108, 481. Bergmann, H., and Hardt, K., Fresenius 2. Anal. Chem., 1979, 297, 381. Donaldson, E. M., Talanta, 1980, 27, 779. Vos, G., Fresenius Z. Anal. Chem., 1985. 320, 556. de Andrade, J. C., Rocha, J. C.. and Baccan, N., Analyst, 1985, 110, 197. Bartlett, R., and James, B., J. Environ. Qual., 1979, 8, 31. James, B. R., and Bartlett, R. J., J. Environ. Qual., 1983, 12, 169. James, B. R., and Bartlett, R. J., J. Environ. Qual., 1983, 12, 173. James, B. R., and Bartlett, R. J., J. Environ. Qual., 1983, 12, 177. James, B. R., and Bartlett, R. J., J. Environ. Qual., 1984, 13, 67. Heringer Donmez, L. A., and Kalenberger, W. E., J. Am. Leather Chern. Assoc., 1989, 84, 110. Krull, I. S., Bushee, D., Savage, R. N., Schleicher, R. G., and Smith, S. B., Anal. Lett., Part A , 1982, 15, 267. Krull, I. S., Panaro, K. W., and Gershmann, L. L., J. Chromatogr. Sci., 1983, 21, 460. Lawrence, K. E., Rice, G. W., and Fassel, W. A., Anal. Chern., 1984, 56, 292. Syty, A., Christensen, R. G., and Rains, T. C., At. Spectrosc., 1986, 7 , 89. Syty, A., Christensen, R. G., and Rains, T. C., J. Anal. At. Spectrorn., 1988,3, 193. Roehl, R., and Alforque, M. M., At. Spectrosc., 1990, 11,210. Mayazaki, A., and Barnes, R. M., Anal. Chem., 1981,53,364. Colella, M. B., Siggia, S., and Barnes, R. M., Anal. Chem., 1980, 52, 967. Isozaki, A., Kumagai, K., and Utsumi, S., Anal. Chim. Acta, 1983, 153, 15. Florence, T. M., and Batley, G. E., Talanta, 1976. 23, 179. Knudtsen, K., and O’Connor, G. A., J. Environ. Qual., 1987, 16, 85. Nazario, C. L., and Menden, E. E., J. Am. Leather Chem. Assoc., 1990,85, 212. Perrin, D. D., Aust. J. Chem., 1963, 16, 572. Brown, L., Haswell, S. J., Rhead, M. M., O’Neill, P., and Bancroft, K. C. C., A n a l p , 1983, 108, 1511. Ajlec, R., Cop, M.. and Stupar, J., Analyst, 1988, 113, 585. Paper 1 lOI559A Received April 3, 1991 Accepted September 2, 1991

ISSN:0003-2654    

DOI:10.1039/AN9921700125

出版商:RSC    

年代:1992

数据来源: RSC

摘要:

ANALYST. FEBRUARY 1992. VOL. 117 131 Determination of Lithium, Beryllium, Cobalt, Nickel, Copper, Rubidium, Caesium, Lead and Bismuth in Silicate Rocks by Direct Atomization Atomic Absorption Spectrometry Toshihiro Nakamura, Hideyuki Oka, Hidehiro Morikawa and Jun Sat0 Department of Industrial Chemistry, Meiji University Higashimita, Tama-ku, Kawasaki 2 14, Japan Optimum experimental conditions for the atomic absorption spectrometric determination of Li, Be, Co, Ni, Cu, Rb, Cs, Pb and Bi by direct atomization of solid silicate rock samples were investigated. A 1.0 mg portion of powdered sample (particle size 0.3-25 pm) was mixed with the same amount of graphite powder and atomized in a miniature graphite cup placed within a graphite furnace according to the heating programme that was established.Calibration was effected using aqueous standard solutions in which 500 ng of K were added for Rb and Cs as an ionization buffer. Results for nine geochemical standard reference rock samples showed good agreement with the recommended values; the correlation coefficient for the found and recommended values was 0.9993. The relative standard deviation ( n = 5) was less than 10%. Keywords: Silicate rock; minor element; direct atomization; electrothermal atomic absorption spectrometry; suppression of ionization interference The concentrations of minor and trace elements in rocks are conventionally determined by d.c. arc emission spectrometry, atomic absorption spectrometry (AAS) or inductively coupled plasma atomic emission spectrometry (ICP-AES) .Usually AAS and ICP-AES require complicated and tedious methods for the dissolution of rock samples. Also, these processes are sometimes accompanied by accidental contamination. Hence, direct atomization of these elements has an advantage over the conventional method and the electrothermal atomization of solid samples has been investigated owing to its high sensitiv- ity, low determination limits and short analytical time requirement. Since the 1940s, d.c. arc emission spectrometry has domi- nated the determination of trace components in rock sam- ples.' This technique is, however, still of low sensitivity for alkali and alkaline earth elements.' Although several attempts have been made to atomize solid rock samples in flame AAS,3-5 applications are limited owing to insufficient sensitivity.This is not only because the temperature of the flame is not sufficiently high but also because maintenance of the flame at a constant temperature is difficult. The combination of ablation with a laser beam and flame atomization was tried for the determination of copper in rocks,h.7 but the sensitivity was not sufficiently high and the reproducibility was not acceptable. Katskov and L'VOV~ first employed a furnace for the direct atomization of Pb. Leucke et al.9 accurately determined Co, Cu and Pb using a heating programme for the atomization based on the boiling-points of the analytes. Calibration was effected by using solid materials whose elemental compo- sitions were similar to those of the samples. Fuller and Thompson10 and Isozaki et al.11 successfully determined Cu and Zn in rock powders by injecting the sample powders as a slurry into a furnace. This technique involved simplified analytical processes and the reproducibility was improved. Langmyhr et aZ.12 and Siemer and Wei13 determined Pb by mixing rock powders with graphite powder to enhance the heat conductivity between the graphite furnace and the sample powder. The choice of calibrating materials is a serious concern when a solid sample is atomized directly. Few investigations have been made on the use of aqueous solutions for calibration except those by Siemer and Wei*3 and Isozaki et al.11 This investigation was concerned with further applications of direct solid atomization AAS to determine Li, Be, Co, Ni, Cu, Rb, Cs, Pb and Bi in silicates by using a graphite furnace and graphite cups.The following aspects are discussed: (1) particle size of sample powder, (2) mixing ratio of graphite powder, (3) temperatures for drying, pyrolysis and atomiza- tion, (4) amount of sample powder and ( 5 ) suppression of ionization interference. The effective use of aqueous standard solutions for obtaining calibration graphs is also discussed. Experimental Apparatus A Hitachi 2-8000 Zeeman atomic absorption spectrometer was used in conjunction with monatomic hollow cathode lamps and a data-processing module which compensates for background absorbance. The operating conditions are given in Table 1. Pure argon was used as the inert gas. The atomization was conducted with a graphite crucible furnace of a cup type with a custom-made miniature cup (3.0 mm external diameter x 3.8 mm external height, 2.5 mm internal diameter x 3.0 mm internal height) after Atsuya and Itoh14 from a spectroscopic graphite rod (Nippon Carbon).The Table 1 Operating conditions for the determination of Li. Be, Co. Ni, Cu, Rb, Cs. Pb and Bi in silicate rock samples Parameter Li Be co Ni c u Rb cs Pb Bi Analytical line/nm 670.8 234.9 240.7 232.0 324.8 794.8 852.1 283.3 306.8 SI i t-wid t h/nm 0.4 1.3 0.2 0.2 1.3 0.4 1.3 1.3 0.2 Lamp current/mA 10.0 10.0 10.0 10.0 7.5 10.0 6.0 7.5 6.0 Absorbance Ar sheath gas flow rate/ml min-I Ar carrier gas flow rate/ml min-1 Peak area 30 200132 0.15 0.10 ANALYST, FEBRUARY 1992, VOL. 117 (a) - - miniature cups were heated to 2600 "C before use to remove any contamination.The powder mill was an Ishikawa Model AGA grinder with an agate mortar and pestle. The powder mixer was an Iwaki Model MA-1 mill with a polyethylene cylinder (24 mm X 12 mm i.d.) and a single polyethylene ball (9 mm diameter). Powdered samples (0.5-1 .O mg) weighed on a Mettler Model M3 microbalance were introduced into the miniature cup using paper. The miniature cup was handled with titanium or brass tweezers. Liquid samples were injected into the cup within the graphite furnace with a Model 4700 Eppendorf micropipette (10 yl). Particle size distributions were measured with a Shimadzu Model SA-CP3L centrifugal particle size analyser. U I I Reagents and Samples Graphite powder of about 50 pm particle size was prepared from spectroscopic graphite (Nippon Carbon) by grinding in an agate mortar.Standard solutions for calibration were prepared from commercially available 1000 ppm standard solutions (Junsei Kagaku) by dilution before use. All reagents used were of analytical-reagent grade and water was de- ionized. The samples were nine geochemical standard reference rock samples issued by the Geological Survey of Japan: JG-1, JGb-1, JP-1, JR-1, JR-2, JA-1, JB-la, JB-2 and JB-3. Procedure Rock samples were ground in order to make the average particle radius less than 2 pm (with no particles exceeding 10 pm). A portion of the powdered sample was mixed with the same amount of graphite powder. A 1.0 mg amount of the 0'11 ~ 0.35 0 0.09 0.08 $ TI 0 $ 0.07 L 0.06 0.05 + - I , I I 10.15 0 50 100 150 Fig.1 Variation in the absorbance of A, Cu; B, Rb; and C, Pb in silicate rock samples with grinding time. A 0.50 mg amount of rock powder (JG-1) mixed with 0.50 mg of graphite powder was atomized. The rock powder (3 g) was ground with an Ishikawa grinder (Type AGA) 0.04 Time/min -0 -0 C & 7 4- m 6 H R\\ B 0 50 100 150 Ti me/m i n Fig. 2 Variation in relative standard deviation (n = 5) of the absorbance of A, Cu; B, Rb; and C, Pb in silicate rock powder (JG-1) with grinding time mixture was weighed in the tared miniature graphite cup and inserted into the graphite furnace. The mixture was dried, pyrolysed and atomized according to the heating programme described later. Absorbance was determined by integration of the spectral lines in the absorbance-time spectrum.Analyte concentrations were determined by comparison with calibra- tion graphs for standard solutions. For Rb and Cs, 20 ppm of K were added to each calibration standard solution to suppress the ionization interference. Results and Discussion Particle Size and Rock Sample Minor elements are present in rocks forming solid solutions with rock-forming minerals or at the boundary of the minerals as oxides, sulfides, etc. They are heterogeneously distributed in rocks. Therefore, sufficient grinding of samples to fine powders is required to make the sample powders homo- geneous for better reproducibility. Wilson15 reported that it is desirable for the number of particles to exceed 1 x 105 to obtain favourable analytical repeatability (i.e., relative standard deviations of less than 10%) in the analysis of solid samples.His calculation showed that 1 mg of powder contains 7 x 103 particles for a particle size of 50 pm and 8 x 105 particles for a particle size of 10 ym. At a size of 5 ym, the number of particles in 1 mg of powder is 7 x 106, and sampling errors become less important. Fig. 1 shows the variation in absorbance of Cu, Rb and Pb with grinding time for a 3.0 g rock sample (JG-1). The variation in the associated relative standard deviations for each element is shown in Fig. 2. The absorbance of Be, Co, Ni, Cu, Pb and Bi increased abruptly in the first 20 min of grinding, which was followed by a gradual increase up to 160 min. For Li, Rb and Cs, the absorbance was constant and essentially independent of grinding time.The relative stan- dard deviation decreased slightly for the first 20 min, then levelled off. Both the median and the mode of the particle size distribution also decreased abruptly in the first 5 min and levelled off up to 160 min. The variation in the relative standard deviations of the absorbance showed a similar trend. The optimum grinding time was longer than 20 min for Be, 1.0 1 ( b ) I I 1 I I 1 I 0.4 I I I 1 I I 1 I I 0 10 20 0 10 20 0 10 20 Time/s Fig. 3 Absorbance-time profiles for (a) Cu; (b) Rb; and (c) Pb in A , aqueous solution; B, silicate rock powder (JG-1); and C, a mixture of rock powder and graphite powderANALYST, FEBRUARY 1992. VOL. 117 133 Co, Ni, Cu, Pb and Bi, although 20 min was sufficient for Li, Rb and Cs for good reproducibility. Based o n data obtained so far, the following grinding conditions are considered to be applicable: a 3.0 mg rock sample is ground for 20 min and the particle size range of the powder is 0.3-25 pm (median 2.4 pm; mode 2.0 pm), 12% of the particles being larger than 10 pm.According to Wilson,l-5 this size distribution is suitable for obtaining sufficien ti y accurate results. Addition of Graphite Powder Mixing o f graphite powder with solid powders in the graphite furnace gives a sharp spectral line12.13 owing to the improved atomization. This is because the total effective surface area of bulk samples is made larger and the heat conductivity is improved. Fig. 3 shows the change in the profiles of spectral lines with the addition of graphite powder. Fig. 4 shows the variation in the absorbance of Cu, Rb and Pb with amount of graphite powder added. Stable and maximum absorbance was obtained with 0.5 mg of rock powder mixed with 0.5 mg of graphite powder.Fig. 5 shows the secondary electron images of rock powder (JR-1) after heating at 2400 "C with and without graphite powder. The presence of graphite powder gives a sharp spectral line and enhances the intensity of the spectral lines for Li, Be, Ni, Cs, Pb and Bi. No change in the half-width of the spectral lines of Co, Cu and Rb was observed, although the absorbances of these elements were intensified. The mixing ratio of graphite powder to sample powder for maximum absorbance for each element is as follows: Li, >0.8; Be, 1.5-3.0; Co, 1.0-1.5: Ni, 1.0-1.5; Cu, 0.0-1.0; Rb, 0.8-1.0; Cs, 0.8-2.0; Pb, 1.G-1.5; and Bi, 0.8-1.5.A mixing ratio of graphite powder of 1.0 appeared to be suitable, although for Be the value was 2.0. The proposed method is effective even for a sample with a concentration that is too high for the dynamic range; a sample containing elements at levels as high as a few hundred ppm can be analysed simply by increasing the amount of graphie powder. 0.05 1 I I 1 Heating Programme Optimum temperatures for the drying, pyrolysis and atom- ization steps for each element were determined with JG-1 and for Pb with JB-la. Fig. 6 shows the variation in absorbance for Cu, Rb and Pb with temperature. The heating programme established is given in Table 2. Sample Amount Fig. 7 shows the variation in absorbance with amount of sample.The absorbances for Li, Be, Ni, Cu, Cs, Pb and Bi are proportional to sample amount from 0.2 to 1.0 mg and for Co and Rb up to 0.7 and 1.5 mg, respectively. When the sample amount exceeds the upper limit or the content of the element is more than 2-3 ng, the proportionality no longer holds. However, when aqueous standard solutions are atomized, proportionality holds up to 10-30 ng. This result implies that an increase in sample amount inhibits the diffusion of atomic vapour. Hence, the optimum sample amounts for Li, Ni, Cu, Rb, Cs, Pb and Bi are 1.0 mg and those for Be and Co, which are more sensitive than other elements, are 0.2 and 0.5 mg, respectively. Suppression of Ionization Interference For an element with an ionization potential higher than 4.6 eV, the ionization interference can be ignored.16 The ioniza- tion potentials of Rb and Cs are 4.2 and 3.9 eV, respectively, hence suppression of the interference is required.Potassium, with a low ionization potential, is usually added to suppress the interference. Rock samples usually contain significant amounts of Na and K which suppress the interference, but those elements are not present in standard solutions.134 ANALYST, FEBRUARY 1992, VOL. 117 0 500 1000 1500 TemperaturePC 2000 2500 Fig. 6 Variation in the absorbance of (a) Cu; (b) Rb; and (c) Pb in silicate rock samples (JG-1) with A, drying; B, pyrolysis; and C, atomization temperature Table 2 Analytical conditions for the determination of Li, Be, Co, Ni, Cu, Rb, Cs and Pb in silicate rock samples Parameter Li Be c o Ni c u Rb c s Pb Bi Sampleamount/mg 1.0 0.2 0.5 1 .o 1 .o 1 .o 1 .o 1.0 1.0 Grinding time*/min 20 Particle sizet/pm 2.0 (0.3-25) Mixing ratio of sample Atomization conditions: to graphite powder 1:l 1:2 1: 1 1 : l 1 : l 1:l 1:l 1 : l 1: 1 Drying 120,30$ 120,30 150,30 150,30 200,30 120,30 120,30 150,30 120,30 Pyrolysis 1400,30 1900,30 900,30 900,30 700,30 1300,30 1600,30 1000,30 1500,30 Atomizing 2600,20 2600,20 2600,20 2600,20 2600,20 2400,30 2600,lO 2600,15 2600,lO * 3.0 g of silicate rock samples ground with an Ishikawa type AGA grinder.t Mode diameter. $ The first value in each pair is temperature in "C and the second is time in seconds. (u C n 0.6 p n (0 U 0.4 CJ, C 4- - 0.2 0 0 0.5 1.0 1.5 2.0 Arnountlmg Fig. 7 Variation in the absorbance of A, Cu; B, Rb; and C, Pb in silicate rock powder with amount of sample Fig.8 shows the variation in absorbance for 5 ng of Rb and Cs with the amount of K. Maximum absorbance was obtained for Rb and Cs by adding 40 and 100 ng of K (eight times the amount of Rb and 20 times the amount of Cs) or more, respectively. For more than 500 ng, the data processor cannot compensate for the background absorbance. Application to Geochemical Standard Rock Samples The results for nine geochemical standard rock samples 0 1 2 3 4 Amountlmg Fig. 8 Variation in the absorbance of 5 ng of A, Rb; and B, Cs in 10 pl of aqueous solution with amount of K added as an ionization suppressor obtained under the proposed conditions are given in Table 3. The calibration standard solutions for Rb and Cs contain K in a 30-fold excess.Each value in Table 3 is the average of five measurements; the relative standard deviations are less than 10%. Values recommended by Ando et al. 17 are also given in Table 3 for comparison. Fig. 9 shows the correlation between the found and the recommended values. The correlation coefficient of 0.9993 indicates that the present results agree well with the recommended values, and that an aqueous standard solution is effective for the calibration in the direct atomization AAS of solid rock samples.ANALYST, FEBRUARY 1992, VOL. 117 135 Table 3 values17 Results for the determination of Li, Be, Co, Ni, Cu, Rb, Cs, Pb and Bi in GSJ standard reference rock samples (in ppm), and literature Li Be co Ni c u This RSD* Ref.This RSD* Ref. This RSD* Ref. This RSD* Ref. This RSD* Ref. Sample work (Yo) 17 work (Yo) 17 work (YO) 17 work (YO) 17 work (%) 17 JG-1 84.1 2 85.9 3.3 4 3.1 4.2 5 4.0 6.1 6 6.0 5.2 3 5.6 JGb-1 4.6 4 4.3 0.37 8 0.36 60.3 11 61.6 25.9 6 25.4 85.6 2 86.8 JR-1 63.6 6 62.3 3.2 4 3.1 0.70 5 0.65 0.60 5 0.66 1.4 8 1.4 JR-2 87.5 3 83 3.8 3 3.4 0.36 8 0.4 0.93 6 0.84 1.4 12 1.4 JA-1 10.8 2 10.5 0.57 7 0.50 10.2 8 11.8 5.9 2 5 42.4 10 42.4 JB-la 11.9 7 11.5 1.5 5 1.4 39.5 9 39.5 128.9 0.3 135 56.5 5 55.5 JB-2 8.3 3 8.0 0.28 5 0.27 36.1 8 39.8 19.9 6 19 222 5 227 JB-3 6.8 0.4 7.2 0.74 5 0.74 33.4 5 36.3 38.6 9 38.8 195 1 198 JP-1 1.5 4 1.8 N.D.? <1 127 9 116 - - 2460 4.4 6 5.7 Rb c s This RSD* Ref. This RSD* Sample work (YO) 17 work (Yo) JG-1 173.2 0.3 181 11.8 2.0 JGb-1 5.1 6 4 N.D.JP-1 N.D. <1 N.D. JR-1 260 0.7 257 23.1 2.4 JR-2 286 3 297 26.6 4.1 JA-1 11.6 3 11.8 N.D. JB-la 41.2 1 41 1.8 12 JB-2 6.2 9 6.2 N.D. JB-3 13.7 5 13 N.D. * Relative standard deviation (YO), n = 5. t N.D. = not detected. Pb Bi Ref. 17 0.52 0.014 0.51 0.65 0.009 0.033 0.020 - - Ref. 17 10.2 0.27 >o. 1 20.2 26 0.64 1.2 0.90 1.1 This work 25.1 2.2 N.D. 17.9 20.4 6.1 7.2 5.1 5.0 RSD* Ref. 3 26.2 7 1.9 6 19.1 1 21.9 4 5.8 4 7.2 6 5.4 4 5.5 (Yo) 17 0.11 This RSD* work (YO) 0.48 19 N.D. N.D. 0.48 11 0.65 8.9 0.95 5.9 N.D. N.D. N.D. 0 4 8 12 16 20 24 Reported value (ppm) Fig. 9 Correlation between reported values and those obtained with the proposed method for X , Li; +, Be; A, Co; 0, Ni; 0, Cu; 0, Rb; 0, Cs; A, Pb; and H, Bi in GSJ standard rock samples.The correlation coefficient is 0.9993 The authors thank S. Morita, H. Moriya and S. Miyanomoto for technical assistance, and the Machine Shop, Meiji University, for making the miniature cup. 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 Reeves, R. D., and Brooks, R. R., Trace Element Analysis of Geological Materials, Wiley-Interscience, New York, 1978, Govindaraju, K., Mevelle, G., and Chouard, C., Anal. Chem., 1974,46, 1972. Katskov, D. A., Kruglikova, L. P., L’vov, B. V., and Polzik, L. K., Zh. Prikl. Spektrosk., 1974, 20, 739. Katskov, D. A., Kruglikova, L. P., and L’vov, B. V., Zh. Anal. Khim., 1975, 30, 238. Karyakin, A. V., Pchelinsev, A. M., Shidlovskii, A. I., Vul’fson, E. K., and Tsimgarelli, M. N., Zh. Prikl. Spectrosk., 1973, 18, 610. Vul’fson, E. K., Karyakin, A. V., and Shidlovskii, A. I., Zh. Anal. Khim., 1973, 28, 1253. Katskov, D. A., and L’vov, B. V., Zh. Prikl. Spektrosk., 1969, 10, 382. Leucke, W., Eschermann, F., Lennartz, U., and Papastamat- aki, A. J., Neues Jahrb. Mineral. Abh., 1974, 120, 178. Fuller, C. W., and Thompson, I., Analyst, 1977, 102, 141. Isozaki, A., Morita, Y ., and Utsumi, S., Bunseki Kagaku, 1990, 39, 605. Langmyhr, F. J., Stubergh, J. R., Thomassen, Y., Hassen, J. E., and Dolezal, J., Anal. Chim. Acta, 1974, 71, 35. Siemer, D. D., and Wei, H. Y., Anal. Chem., 1978, 50, 147. Atsuya, L., and Itoh, K., Bunseki Kagaku, 1982,31, 708. Wilson, A. D., Analyst, 1964, 89, 416. Suzuki, M., and Ohta, K., Bunseki, 1984, 125,416. Ando, A., Mita, N., and Terashima, S., Geostand. Newsl., 1987, 11, 159. p. 151. References 1 Brooks, R. R., and Boswell. C. R., Anal. Chim. Acta, 1965,32, 339. Paper 1 I03891 E Received July 29, 1991 Accepted September 25, 1991

ISSN:0003-2654    

DOI:10.1039/AN9921700131

出版商:RSC    

年代:1992

数据来源: RSC

摘要:

ANALYST, FEBRUARY 1992, VOL. 117 137 Determination of Lanthanum in Food and Water Samples by Zeeman-effect Atomic Absorption Spectrometry Using a Graphite Tube Lined With Tungsten Foil Shen Miao-Kang Hangzhou Health and Anti-Epidemic Station, Hangzhou 310006, People's Republic of China Shi Yin-Yu Department of Chemistry, Zhejiang University, Hangzhou 310027, People's Republic of China A sensitive, selective method for the determination of lanthanum in food and water samples by atomic absorption spectrometry using a graphite tube lined with tungsten foil is described. The atomization of lanthanum from the tungsten surface gives better analytical sensitivity, a lower atomization temperature and negligible memory effects. The characteristic mass and detection limit of the method were 8.1 X 10-9 and 7.85 x 10-9 g, respectively.The precision (relative standard deviation in the range 5.9-9.9%), accuracy and interferences of the method were also investigated. The method can be used directly for the determination of trace amounts of lanthanum in food and water without pre-dissociation of the matrices. The results obtained by this method are in good agreement with those obtained from inductively coupled plasma atomic emission spectrometry. Keywords: lanthanum determination; food and water; atomic absorption spectrometry; tungsten atomizer Atomic absorption analysis with electrically heated graphite atomizers has found widespread acceptance as a routine method in many research and application laboratories. Atomization of the sample in a graphite tube heated to 3000 "C gives a method of good sensitivity, capable of determining a large number of trace elements directly in diverse sample matrices. However, the determination of lanthanum by atomic absorption spectrometry (AAS) has always been difficult because of the poor sensitivity and strong memory effect for the element.These problems are generally attri- buted to carbide formation resulting from the interaction of lanthanum with carbon from the atomizer. The graphite itself has been improved to overcome these problems. Although a pyrolytic graphite coated graphite tube has been used,*-5 this atomizer does not give adequate sensitivity for lanthanum which forms refractory carbides with graphite at high temper- atures resulting in incomplete vaporization of the lanthanum.Attempts have also been made to use a graphite tube pre-coated with a salt of tantalum, zirconium or tungsten.68 Other attempts have been made to use a tantalum or tungsten-tantalum lining inserted inside the graphite fur- nace.9-17 These atomizers can eliminate physical contact and, hence, reaction between the graphite surface and analytes, and significantly increase the analytical sensitivity. However, the lifetime of the atomization surfaces is not very long and the reproducibilities are not satisfactory because of the deforma- tion of the metal at high temperatures. This paper describes the determination of lanthanum in food and water samples by Zeeman-effect background-correc- ted AAS using a graphite tube lined with tungsten foil.Experimental Apparatus A Perkin-Elmer Zeeman 5000 AA spectrometer with a hollow cathode lamp operating at 30 mA and with a slit-width of 0.4 nm was used. Measurements were made at 550.1 nm. Zeeman-effect background correction was used. A Perkin- Elmer HGA-500 graphite furnace atomizer was operated under the conditions given in Table 1. Samples were injected using a Perkin-Elmer AS-40 autosampler. The volume in- jected was 20 pl. Signals were recorded on a Perkin-Elmer Model 100 recorder. The height of the absorbance peak was measured. Preparation of the Furnace Lined With Tungsten Foil The tungsten lining was prepared from a rectangular strip (12 x 15 mm) of tungsten foil of purity 99.967% and thickness 0.1 mm and formed by winding a tungsten strip around a glass rod of diameter slightly less than that of the graphite tube.The lining was inserted into the centre of a new pyrolytic graphite coated graphite tube and a metal rod was placed into the tube and firmly rolled inside it to attach the tungsten foil smoothly to the inner lining of the graphite tube. In order to prevent distortion of the foil on heating, the tube lined with foil was pre-treated in the HGA 500 atomizer, by heating several times to a high temperature in a current of argon according to the following programme: (i) the tube was heated to 500 "C in 10 s, held at that temperature for 5 s, and then cooled down in 10 s; and (ii) the operation was repeated for temperatures of 1000, 1500,2000 and 2500 "C, followed by blank firings according to the heating programme in Table 1.The lifetime of the tungsten-surface atomizer was about 130 firings. Reagents All reagents for the interference study were of analytical- reagent grade and all solutions were prepared with doubly distilled, de-ionized water. Lanthanum oxide stock standard solution, 1000 pg ml-l. Accurately weigh 0.1000 g of lanthanum oxide (99.99%) and dissolve by gently heating in 10 ml of concentrated hydro- chloric acid. Wash the solution into a 100 ml calibrated flask and dilute to the mark with water. Store the solution in a stoppered polyethylene bottle. Lanthanum working standard solution. Prepare working solutions to cover a wide concentration range by mixing an Table 1 Graphite furnace conditions for the determination of lanthanum Argon Temper- Ramp Hold flow rate/ Step aturePC time/s time/s ml min-1 Dry 100 5 10 300 200 5 10 300 300 Dry Ash 1300 5 10 Atomize 2400 0 4 0 Clean 2500 1 3 300138 2 e 8 0.1 ANALYST, FEBRUARY 1992, VOL.117 :;-:-/ A B ' aliquot of the stock solution with water (final acidity approximately 4% hydrochloric acid) and store in polyethylene bottles. Concentrated hydrochloric acid, Suprapur grade. Concentrated nitric acid, Suprapur grade. Procedure The lanthanum content of food and water samples is usually small and therefore it is often necessary to concentrate the element in a small volume. Preparation of food samples for analysis Samples of food e.g., rice, wheat, maize, milk powder, vegetable and tea were oven-dried and ground. An accurately weighed sample (5-10 g) was placed in a porcelain crucible and covered.The crucible was heated for 1 h on a hot-plate, then several drops of concentrated nitric acid were added to aid charring and the heating was continued. The temperature was increased gradually in order to avoid sputtering of the sample. After charring was complete, the crucible was heated in a muffle furnace at 400 "C for 30 min and 600 "C for a further 6 h. The crucible was removed, cooled and, after the addition of 0.5 ml of concentrated nitric acid, heated in the same manner as described above for 6 h. The sample ash was treated with 0.2 ml of concentrated hydrochloric acid and 0.2 ml of water (warming may be necessary). The solution was trans- ferred quantitatively into a 5 ml calibrated flask and diluted to the mark with water.In this ashing method, the preparation of a blank was necessary. Preparation of water samples for analysis A suitable volume of water sample (250-500 ml) was transferred into a 500 ml beaker and 0.2 ml of concentrated hydrochloric acid was added. The beaker was heated in a water-bath and the solution was evaporated to 2-3 ml. The concentrated solution was transferred quantitatively into a 5 ml calibrated flask and diluted to the mark with water. Determination The treated solutions were decanted into the sampling cups of the autoanalyser and measurements of lanthanum were made using the conditions given in Table 1. The absorbances were recorded on chart paper. Results and Discussion Optimization of the Graphite Furnace Programme Experiments were carried out to ascertain the best tempera- ture and time for the drying, ashing and atomization steps.A mirror was used to observe the drying of the sample in the 200 1000 1800 2600 TemperaturePC Fig. 1 A, Ashing and B, atomization graphs for lanthanum furnace. It was found that a temperature of >150"C allowed uniform drying with no sputtering, and that the drying time was dependent on the volume of solution injected into the furnace; a time of 20 s was required for 20 p1 of a sample. In order to optimize the ashing and atomization temperatures, graphs were constructed for an aqueous solution containing 10 pg ml-1 of lanthanum and the optimum concentration of hydrochloric acid; the results are shown in Fig. 1. The optimum ashing temperature was 1300°C (the dip in the ashing curve probably related to the volatilization of lan- thanum) and the optimum atomization temperature was 2400°C.The internal gas-stop mode and maximum power were used during the atomization stage. The use of the gas-stop mode in the atomization step can produce tube memory effects, therefore, in order to overcome this, the tube was fired at 2500°C for 3 s with the internal gas at the maximum flow rate of 300 ml min-1. The ashing time for the sample in the hydrochloric acid matrix was 10 s. The atomization temperature was maintained for only 4 s, as longer atomization times did not effectively reduce the peak height of the blank solution further, but only reduced the usable life span of the graphite tube. Effect of Hydrochloric Acid Concentration In a hydrochloric acid matrix, the chloride complex is usually present before atomization. However, lanthanum chloride hydrolyzes to oxychlorides on evaporation and, at a higher temperature, these decompose to the oxide.Therefore, the lanthanum oxide intermediate can be expected to form before atomization in a hydrochloric acid medium. The effect of hydrochloric acid concentration on the determination of lanthanum is shown in Fig. 2. The peak height absorbance increased slightly as the concentration of hydrochloric acid increased from 0.5 to 1.0% v/v, but remained constant at concentrations of >1.0%. Therefore, a hydrochloric acid concentration of 4.0% v/v was selected. Calibration and Standard Additions Graphs In order to obtain a calibration graph, standard solutions containing 0-10 pg ml-1 of lanthanum with the optimum concentration of hydrochloric acid were subjected to the furnace programme.When using the standard additions method, 0, 1, 2, 4 and 8 pg ml-1 of lanthanum and the optimum concentration of hydrochloric acid were added to a sample. The correlation coefficient of the calibration graph for absorbance versus concentration of lanthanum was 0.9991. The slopes of the calibration and standard additions graphs were in close agreement, which illustrated that the matrix effect was negligible. Sensitivity and Detection Limit The sensitivity can be conveniently measured in terms of 'characteristic mass'. In this work, the proposed method has a characteristic mass of 8.1 x 10-9 g, which is similar to that given by L'vov and Pelieva.9 The detection limit, i.e., the I I 8 0.2 V V ] 2 0.1 0 2 - 4 6 8 Concentration( %) Fig.2 Effect of concentration of hydrochloric acid on absorbance of lanthanumANALYST, FEBRUARY 1992, VOL. 117 139 ~~ ~ Table 2 Precision of the results for the determination of lanthanum Relative Concen- Standard standard Sample tratiod Mean deviation/ deviation 1 2.0 0.0238 0.00237 9.9 2 4.0 0.0426 0.00374 8.8 3 20.0 0.214 0.0 1270 5.9 No. pg ml-1 absorbance pg ml-l ( Y o ) Table 3 Recovery of lanthanum from food and water samples Sample - No. 1 2 3 4 5 6 7 8 9 10 Lanthanudpg ml- - Present* 1.55 1.60 2.10 2.30 2.85 1.10 1.85 1.95 2.70 1.70 Added 2.5 2.5 2.5 2.5 2.5 3.0 3.0 3.0 3.0 5.0 * Preconcentrated samples. Found 3.90 3.65 4.30 4.40 4.70 3.80 4.95 4.40 4.85 5.87 - Recovery ( Y o ) 94.0 82.0 88.0 84.0 74.0 90.0 103.0 82.0 73.0 83.0 Table 4 Interference of other ions on the determination of lanthanum Ion None Ca2+ Mg2+ K+ Na+ cu2+ Fe3+ Zn2+ Pb2+ C@+ Cd2+ Mn2+ Sn4+ Si2+ Se4+ c1- N03- s o p ~ 1 3 + Amount added (PPm) - 500 200 500 500 10 100 10 10 10 10 10 5 5 100 2 300 80 100 Relative absorbance 1 .oo 0.97 0.97 0.89 1.05 1 .oo 0.89 0.97 0.98 0.91 0.90 0.99 0.93 0.94 1.09 1 .OO 1.05 0.89 0.99 lowest concentration level that can be determined to be statistically different from a blank, is defined as three times the within-batch standard deviation of a single blank determi- nation ( i e ., the standard deviation was obtained from replicate analyses of a single blank sample), corresponding to a 99% confidence level. In this study, the detection limit of lanthanum was 7.85 x 10-9 g.Precision and Accuracy The precision (relative standard deviation) of the method was obtained for replicate analyses of one sample during the same run and the results are shown in Table 2. The within-batch precision of the method, obtained for replicates of three synthetic samples with 2.0, 4.0 and 20.0 pg ml-1 lanthanum added, varied over the range 5.9-9.9%. In order to study the accuracy of the method, the recovery of standard additions of lanthanum to the samples was investigated over the entire sample preparation procedure. Each addition was preformed in duplicate. The results obtained gave recoveries ranging from 73 to 103% of lanthanum (Table 3). 0.2 a C (D e 2 2 0.1 0 1 2 3 4 5 6 No. of firings Fig.3 tube Memory effects of atomizers: A, lined tube; and B, unlined Table 5 Comparison of results for the determination of lanthanum Lanthanum concentration/yg ml- Sample - No. 1 2 3 4 5 6 7 8 9 10 Proposed method 0.155 0.210 0.230 0.285 0.110 0.185 0.195 0.085 0.063 0.050 ICP-AES method 0.142 0.221 0.213 0.298 0.123 0.213 0.198 0.104 0.078 0,052 Interference Study Different amounts of other ions were added to a test solution containing 10 pg ml-1 of lanthanum and lanthanum was determined using the proposed procedure. The absorbance data were compared with the value obtained for 20 pl of a 10 pg ml-1 pure lanthanum standard solution. The results obtained are given in Table 4, which shows that not all the elements tested interfered with the determination of lan- thanum.The interference caused by other rare earth elements was also investigated and appeared negligible. This illustrates that the selectivity of the method is very good. Memory Effect Lanthanum in the presence of carbon is considered to form a non-volatile carbide that may result in a memory effect. This memory effect was reduced when a pyrolytic graphite coated graphite tube lined with tungsten foil was used (Fig. 3). After injection of a 20 p1 aliquot of a 10 pg ml-1 solution into a graphite tube lined with tungsten foil, a return to background levels was obtained without any blank firings. However, after injection of a 20 aliquot of a 10 pg ml-1 solution into an unlined graphite tube, a significant memory effect remained even after five blank firings. Therefore, the atomization surface plays an important role in the determination of lanthanum.Comparison of Two Analytical Methods A comparison of the results obtained by the proposed method with those obtained using inductively coupled plasma atomic emission spectrometry (ICP-AES) is shown in Table 5. The results show that there is satisfactory agreement and no significant difference ( P > 0.05) between the two methods.140 ANALYST. FEBRUARY 1992. VOL. 117 6 7 8 9 10 References Zvonimir, G . , Fresenius' Z. Anal. Chem., 1978, 289, 337. Horsky, S. J., At. Spectrosc., 1980, 4, 129. Horsky, S. J . , and Fletcher, W. K . , Chem. Geol., 1981,32,335. Sen Gupta, J . G . , Anal, Chim. Acta, 1982, 138, 295. Sicinska, P . , and Michalewska, M., Fresenius' 2. Anal. Chem., 1982, 312, 530. Sastry, M. D., Bhide, M. K . , Savitri, K., Babu, Y., and Joshi, B . D., Fresenius' Z. Anal. Chem., 1979, 298, 367. Sneddon, J., and Fuavao, V. A., Anal. Chim. Acta, 1985, 167, 317. Zatk, V. J., Anal. Chem., 1978, 50, 538. L'vov, B . V., and Pelieva, L. V., Zavod. Lab., 1978. 44, 173. Chen, Y.-W., and Li, J.-X., Fenxi Huaxue, 1979, 7, 7. 11 12 13 14 15 16 17 Wahab, H. S., and Chakrabarti, C. L., Spectrochim. Acta, Part B , 1981,36, 475. Wahab, H . S . , and Chakrabarti. C. L., Spectrochim. Acta. Part B, 1981.36.463. Sen Gupta, J. G.. Talanta. 1984, 31, 1053. L'vov, B. V., and Pelieva. L. V., Zh. Anal. Khim., 1979, 34, 1 744. Sen Gupta. J . G . , Talanta, 1985, 32, 1. Sen Gupta, J . G . , Talantu, 1987, 34, 1043. Ma. Y.-Z., and Wu, Z.-K., Huaxue Tongbao. 1982, 1, 22. Paper I I04 144 D Received August 8, 1991 Accepted October 2, 1991

ISSN:0003-2654    

DOI:10.1039/AN9921700137

出版商:RSC    

年代:1992

数据来源: RSC

摘要:

ANALYST, FEBRUARY 1992, VOL. 117 141 Rapid and Simple Method for the Determination of Copper, Manganese and Zinc in Rat Liver by Direct Flame Atomic Absorption Spectrometry Svjetlana Luterotti Department of Chemistry, Faculty of Pharmacy and Biochemistry, University of Zagreb, I, A. KovaCiCa, Zag re b, Croatia Tihana Zanic-GrubiSic and Dubravka Juretic Department of Medical Biochemistry, Faculty of Pharmacy and Biochemistry, University of Zagreb, I, A. Kovac'ica, Zagreb, Croatia An acidic homogenate method, which includes simple homogenization pre-treatment of tissue material and direct nebulization flame atomic absorption spectrometry (FAAS), is successfully applied t o the simultaneous determination of copper, manganese and zinc in rat liver. The proposed method involves only a few steps for sample pre-treatment at room temperature, making the risk of systematic errors very small.Because recoveries of 101% for copper, 98% for manganese and 100% for zinc could be achieved using aqueous standards, matrix-matched standards were redundant. Favourable results obtained in biological media, including limits of detection of 0.04, 0.03 and 0.04 m g 1-1 for Cu, Mn and Zn, respectively, together with accuracies of 0-3%, and relative standard deviations ranging from 2 t o 10% are further evidence of the suitability of the method. Keywords: Copper, manganese and zinc determination; rat liver; direct flame atomic absorption spec- tro m etr y; sample pooling and homogenization ; direct standardization Tissue analyses are generally time-consuming and prone to error, mainly because sample preparation is tedious and includes the risk of contamination or loss of analyte.Dry and wet decomposition methods have been used frequently as procedures for the treatment of tissue samples' prior to determinations using atomic absorption spectrometry (AAS).2 Wet ashing is generally quicker than dry ashing and has been applied to various types of tissue at room or elevated temperature.3-10 However, the hazardous nature of the reagents requires constant vigilance, and high purity reagents must be used in order to avoid high blank values. Digestion time was, however, reduced to as little as 1 h by solubilizing the tissue with a quaternary ammonium hydroxide.11.12 Chelation of vanadium(v) with hydroxamic acid and subse- quent extraction into an organic solvent has recently been applied to wet digested samples of sera and urine.13 Dry ashing of samples is time-consuming, and some metals can be volatilized.1714 Low-temperature dry ashingls gives less likelihood of loss of volatile metals but the method is slow and requires expensive apparatus. However, Failta and Kiser,16 and Parker4 reported the successful dry ashing pre-treatment of tissues, plasma and feed prior to measurements using AAS of copper, zinc, manganese, iron and magnesium. Pre-treatment requiring a minimum of handling and glass- ware has been described by Jackson and Mitchell,17 who simply homogenized the sample material with water produc- ing a uniform suspension; suitable dilutions of the homogen- ate were introduced into the Delves cups for direct determina- tion of cadmium by AAS.The necessity for a simple method of hepatic metal quantification has served as a stimulus for a detailed analytical validation of a method where final measurement by flame AAS (FAAS) of trace metals in the liver of experimental animals is preceded by a simple homogenization procedure. Experimental Reagents Analytical-reagent grade chemicals were used. Ultrapure HCI was purchased from J . T . Baker. All-glass apparatus was used to produce doubly distilled water, which was used throughout the study. Standard solutions. In addition to aqueous standards, solutions matched to corresponding samples, with respect to biological matrix, concentration of acid and major inorganic salts content i.e., synthetic samples, were also prepared.Matrix modifier. As liver is a complex matrix, synthetic samples corresponding to 1 + 5 (by mass) liver homogenates were prepared as 1.1% m/v solutions of human albumin, enriched with inorganic salts to contain 163 mg 1-1 Na+, 489 mg I-* K+, 226 mg 1-1 C1- and 1390 mg 1 - 1 H2P04- in the final solution. For liver homogenates diluted 1 + 29 synthetic samples diluted similarly were used. A solution of human albumin for intravenous application, stabilized with 0.16 mmol sodium octanoate per gram of protein was used. Standard reference material (SRM) . Lyophilized Bovine Liver (SRM 1577a) purchased from The National Institute of Standards and Technology ~ (NIST), was used. Apparatus Atomic absorption spectrometer, Perkin-Elmer 305B, with a Hitachi 56 recorder, was used throughout for the determina- tion of metal ions.Liver samples were homogenized using an U1 tra-Turrax homogenizer, type Janke and Kunkel GK, and the homogen- ates were centrifuged using a Sorwall S-34 centrifuge. A Radiometer Model PHM 85 digital pH meter with a combination glass electrode (Radiometer GK 2322 C) was used for pH readings. Acid-washed polyethylene laboratory accessories were used where possible, such as spatulas, spoons, bottles for storing samples and standard solutions. Tubes with tight-fitting caps were used for centrifugation of samples and for storing the centrifuged samples. Analytical Procedures Sampling Several female Wistar rats were used in each experimental group. The animals were anaesthesized and killed by decapita-142 ANALYST, FEBRUARY 1992, VOL.117 Table 1 Instrumental parameters Lamp Lamp Slit setting Scale expansion Element Lamp type Wavelengthlnm current/mA power/W (band pass)/nm factor c u HCL* 325.2 15 - 4 (0.7) 1 Mn HCL 280.3 20 - 4 (0.7) 3 Zn EDLt 214.6 - 6 4 (0.7) 1 * Hollow cathode lamp. t Electrodeless discharge lamp. Table 2 Internal quality control of AHM; values given for dry mass of sample SRM 1577a Bovine Liver Certified value/ Value found, RSD Accuracy Analyte pg g-1 CI/pg g-1 (%) (%) Cu 158 k 7 154.3 k 3.0 (12)* 2.0 3 Mn 9.9 t- 0.8 9.9 k 0.5 (13) 5.1 0 Zn 123 k 8 124.6 % 5.5 (13) 4.6 2 * Number of individual samples in parentheses, p = 0.995. tion. The livers were rapidly removed, rinsed with ice-cold saline (0.15 mol dm-3 NaCI), gently blotted dry, cut into small pieces using stainless-steel surgical scissors and mixed with a plastic spoon.The pooled material was then randomly divided into several sample portions each of them being weighed accurately (1 x 10-4 g). The remaining portion was used for the determination of the dry mass content. The pooled material was immediately frozen at -20 "C in polyethylene bottles and thawed prior to use. Acidic homogenate method (AHM) In order to obtain a homogeneous tissue sample, a portion of pooled rat liver was homogenized with an exact five-fold amount (by mass) of water for 3 x 1 min. The aqueous homogenate was then adjusted to be 1 mol dm-3 in HC1. The suspension obtained was shaken for 30 min, and centrifuged at 12 OOOg. The original supernatant (1 + 5 by mass), served for the direct measurement of copper and manganese; an additional dilution to give a final dilution of 1 + 29 was prepared for the measurement of zinc.The lyophilized NIST Bovine Liver sample required a hydration treatment (370 h) prior to homogenization. The same dilution ratios were applied for the determination of manganese and zinc in the SRM as for native liver; however, for the measurement of copper a 1 + 29 dilution was needed. FAAS measurements Liquid samples were aspirated directly into the nebulizer system of the instrument. The analytical signals were processed in the peak height mode. The deuterium arc background corrector was used throughout. Instrumental parameters are given in Table 1. All the measurements were performed under standard pressure and flow rate conditions for both air (2.1 x 105 Pa, 22.5 1 min-1) and acetylene (5.5 X 104 Pa, 3.9 1 min-I). Estimation of dry mass content An accurately weighed portion of the pooled tissue material was dried for 24 h at 20-25 "C at a pressure of <30 Pa.Evaluation of analytical results Analytical data are presented as confidence intervals, CI = i 1 talnl/2, at the stated probability levelp (X = mean value, t = tabulated t-value, o = standard deviation and n = number of results) . 0.16 I 1 0.12 a, C t-0 0.08 z 2 0.04 0 0.4 0.8 1.2 1.6 2.0 Metal ion concentration/mg I-' Fig. 1 Regression analysis applied to the total calibration data for: A, zinc: y = 0.099574~ - 0.004428, r = 0.9995, n = 6 (7); B, copper: y = 0.023521~ + O.ooOo80; r = 0.9996, n = 6 (6); and C, manganese: y = 0.019102~ - 0.00037; r = 0.9998, n = 6 (6).The individual points refer to mean absorbance zk o. Number of parallels denoted in parentheses Limits of detection (cL) are expressed as the concentration derived from the smallest measure xL that can be detected with a reasonable certainty for a given analytical procedure: xL = Xb + 3ab; cL = f(xL) (x,, = mean of the blank measurements and q., = standard deviation of the blank measurements), according to IUPAC recommendations. 1s Results and Discussion It is known6 that procedural errors are very much smaller than the biological variations that can arise if a tissue sample is not representative of a whole organ. Any objective validation of an analytical method applied to tissue material requires that the analytical results be free from the influence of biological variations.Therefore, the present investigations were per- formed on pooled rat liver material treated as several independent samples, thus fluctuations of results indicate only the imperfections of the method. By limiting the number of manipulations and reagents employed, the risk of contamination is considerably reduced. From this point of view the proposed simple and direct determination of metal-ions from liver homogenates appears to be very promising. Acidic Homogenate Method Comprehensive preliminary studies directed us to propose an acidic extraction of metal ions, applicable to the aqueous homogenates of both the native rat liver and the lyophilized SRM Bovine Liver.Internal quality control and analytical performances The results of the AHM internal quality control presented in Table 2, indicate that the application of deuterium arc background correction assures interference-free results and excellent accuracy of the method for all three analytes.ANALYST, FEBRUARY 1992, VOL. 117 143 Table 3 Analytical performances of AHM Calibration sensitivity*/ 1 mmol-l Limit of Characteristic Aqueous ASA Recovery* detection$/ concentration/ Analyte standards standardst (Yo.> mg 1-1 mg 1-l c u 1.49 k 0.08 1.50 f 0.01 101.30 k 5.21 0.04 0.15 Mn 1.27 rf: 0.05 1.18 k 0.07 98.29 k 2.42 0.03 0.21 Zn 6.51 k 0.19 6.45 k 0.13 99.57 rf: 3.0 0.04 0.06 (97.36107.20)O (96.83-101.09) (96.56103.03) *x k (5. t p = 0.996. $ Matrix-matched standards (acid-salt-albumin).§ Range of values given in parentheses. Table 4 AHM in analysis of rat liver; values given for dry mass of sample Sample Performance group characteristic c u Mn Zn I CI/pg g-1 12.4 k 1.2(9)* 5.4 f 0.5 (12) 105.8 rf: 4.6(11) dpgg-1 1.0 0.5 4.0 SEMt/pg g-1 0.3 0.1 1.3 RSD(%) 7.6 8.8 3.8 I1 CI/pg g-' 19.8 f 2.0 (6) 6.2 f 1.0 (7) 122.7 rf: 5.8 (6) o/pg g-1 1.0 0.6 3.0 RSD(%) 5.1 9.8 2.4 I11 CI/pg g;I 20.4 k 2.4 (6) 9.0 f 1.0 (7) 136.9 rf: 5.0 (6) (54% g- 1.2 0.6 2.6 RSD(%) 6.1 6.5 1.9 SEM/pg g-1 0.4 0.2 1.2 SEM/pg g-1 0.5 0.2 1.1 * Number of individual samples in parentheses; p = 0.995. t Standard error of the mean. Table 5 Day-to-day precision of AHM; values given for dry mass of sample SRM 1577a Bovine Liver Analyte CI/pg g-1 d p g g-' SEM*/pg g-' RSD (%) c u 148.8 k 4.2 (25)t 6.7 1.4 4.5 Mn 10.3 * 0.4 (25) 0.6 0-1 5.8 Zn 126.6 k 4.0 (25) 6.4 1.3 5.1 * Standard error of the mean.t Number of individual samples in parentheses, p = 0.995. The high conformity of calibration graphs for aqueous and matrix-matched standards together with the mean recoveries (Fig. 1 and Table 3) make the use of aqueous standards sufficient, and is further confirmation of the selectivity of the method. The results for aqueous and matrix-matched stan- dards were treated together, resulting in the mean values and standard deviations shown in Fig. 1. Values obtained for the limits of detection and characteristic concentrations, indicating acceptable sensitivity of the method are summarized in Table 3. Homogenization of the pooled livers was found to be critical.Repeatability of the results obtained through the application of the method to the analysis of native rat liver (Table 4) confirm that effective homogenization of the tissue material has been achieved. Precision studies were completed by calculation of day-to-day relative standard deviations (RSDs) (Table 5). The precision seems to be independent of run type (within-run, Table 2; day-to-day run, Table 5) for manganese and zinc. These results, together with those from Table 4 and nearly 100% recoveries for all three metals, indicated that both the precision and accuracy of the proposed method were satisfactory. Conclusion An acidic homogenate method for the direct quantification of copper, manganese and zinc in rat liver, with negligible risk of contamination or analyte loss, has been developed.The method is simple and rapid, as no time-consuming procedures are employed for either standardization or sample prepara- tion. Accurate and precise analyses can be performed using 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 aqueous standards. Reliability together with simplicity and speed make the proposed method suitable for use in clinical chemistry research, where the relationship between hepatic trace metal status and metabolic disorders are dealt with. References Analytical Methods Committee, Analyst, 1960, 85, 643. Analytical Methods for Atomic Absorption Spectrophotometry , Perkin-Elmer, Norwalk, CT, USA, March 1971, BC-13. Jacob, R. A., Klevay, L. M., and Logan, G. M., Jr., Am. J. Clin. Nutr., 1978, 31, 477. Parker, H. E., At. Absorpt. Newsl., 1963, 2,23. Cheek, D. B., Graystone, J. E., Willis, J. B., and Holt, A. B., Clin. Sci., 1962, 23, 169. Parker, M. M., Humoller, F. L., and Mahler, D. J., Clin. Chem., 1967, 13, 40. Kahnke, M. J., At. Absorpt. Newsl., 1966, 5 , 7. Clegg, M. S., Keen, C. L., Loennerdal, B., and Hurley, L. S., Biol. Trace Element Res., 1981, 3, 107. Uriu-Hare, J. Y., Stem, J. S., Raeven, G. M., and Keen, C. L., Diabetes, 1985, 34, 1031. Sprenger, K. B. G., and Franz, H. E., Clin. Chem., 1983, 29, 1522. Murthy, L., Menden, E. E., Eller, P. M., and Petering, H. G., Anal. Bidchem., 1973,53, 365. Gross, S. B., and Parkinson, E. S., A&. Absorpt. Newsl., 1974, 13, 107. Ishida, O., Kihira, K., Tsukamoto, Y., and Marumo, F., Clin. Chem., 1989,35, 127. Thiers, R. E., in Trace Analysis. Symposium on Trace Analysis N. Y. Acad. Med. 1955, eds. Yoe, J. H., and Koch, H. J., Wiley, New York, 1957, pp. 637-666. Sanui, H., Anal. Biochem., 1971, 42, 21. Failla, M. L., and Kiser, R. A., J . Nutr., 1981, 111, 1900. Jackson, K. W., and Mitchell, D. G., Anal. Chim. Acta, 1975, 80, 39. IUPAC, Pure Appl. Chem., 1976,45, 99. Paper 1 l03455C Received July 9, 1991 Accepted September 2, 1991

ISSN:0003-2654    

DOI:10.1039/AN9921700141

出版商:RSC    

年代:1992

数据来源: RSC

摘要:

ANALYST, FEBRUARY 1992, VOL. 117 145 Synthesis of a Morin Chelating Resin and Enrichment of Trace Amounts of Molybdenum and Tungsten Prior to Their Determination by Inductively Coupled Plasma Optical Emission Spectrometry Xing-yin Luo,* Zhi-Xing Su,* Wen-yun Gao, Guang-yao Zhan and Xi-jun Chang Department of Chemistry, Lanzhou University, Lanzhou 730000, People's Republic of China A morin chelating resin was synthesized using aminated poly(viny1 chloride) as the starting material, and the optimum conditions for the synthesis were established. The parameters governing the characteristics of the resin for the adsorption of MoVi and Wvi including acidity, flow rate, rate constant, saturated capacity of adsorption, effect of re-use, interfering ions and desorption were investigated. The MoV1 and Wvl concentrations in standard samples were determined by using inductively coupled plasma optical emission spectrometry, with satisfactory results.For concentrations of MoVt and W'" of 0.4 mg 1-1, the relative standard deviation was 2.8% for MeV' and 2.6% for WI. The structure of the chelating resin was deduced by infrared spectrometry, and the mechanism of the enrichment of MoV1 and Wi is discussed. Keywords: Morin chelating resin; synthesis; enrichment; molybdenum and tungsten determination; inductively coupled plasma optical emission spectrometry Morin (2' ,3,4' ,5,7-pentahydroxyflavone) is amongst the most sensitive reagents used for the spectrophotometric and spec- trofluorimetric determination of a number of metal ions,'J particularly M0.3~4 In the present work this reagent was attached to an aminated macroporous poly(viny1 chloride) resin by means of the Mannich reaction to give a chelating resin containing morin as the functional group.This chelating resin is resistant to the action of strong acid or base. The conditions for the synthesis of the resin and its ability to absorb trace amounts of MoV1 and Wvl were studied. The resin was used in the analysis of several standard samples with satisfac- tory results. Experimental Apparatus and Instruments An ICP 6500 inductively coupled plasma optical emission spectrometer (Perkin-Elmer) , a Nicolet 170-sx Fourier trans- form infrared (FTIR) spectrometer, a Model pHs-3A digital pH meter and an electrodynamic oscillator were used. The adsorption column consisted of 0.2 g of chelating resin, which was kept in distilled water for 8-10 h before use, in a glass tube (20 cm long, 0.46 cm i.d., 0.15 cm i.d.at the lower end), in which a small pad of cotton-wool had been placed at the lower end beforehand. Materials and Reagents The aminated macroporous poly(viny1 chloride) resin (N content: 10%; particle diameter: 0.36-0.45 mm) was synthe- sized by the procedure described in ref. 5 . All the reagents used were of analytical-reagent grade and distilled water was used throughout. The stock solution of MoV1 or Wvl, at a concentration of 1.0 g I-*, was prepared by dissolving 0.4610 g of ( N H ~ ) ~ M o ~ O ~ ~ - ~ H ~ O or 0.4486 g of Na2W04.2H20 in dis- tilled water and diluting to 250 ml with distilled water. The mixed standard solution of MoV1 and Wvl, each at a concentra- tion of 20 mg 1-1, was prepared by diluting the stock solution with distilled water.Synthesis of the Morin Chelating Resin A 0.5 g amount of morin was placed in a three-necked 250 ml flask fitted with a condenser and agitator. Then, 100.0 ml of * Authors to whom correspondence should be addressed. 95% ethanol, 20.0 ml of formaldehyde solution (40%), 2.5 ml of concentrated HCI and 1.0 g of aminated poly(viny1 chloride) resin were added. After heating the mixture at 72 "C for 10 h, the chelating resin was washed with distilled water until the washings were of neutral pH and dried under IR radiation. The dried resin was then placed in an extractor and extracted with 95% ethanol in order to wash off the free morin.The resin was then dried for later use. The procedure described in ref. 6 was used to determine whether there was any morin in the extracting agent. The synthesis procedure can be described as follows: 72 "C, HCI + CH20 + morin- 95% C,H,OH -CH=CH-C H 2-C H- I YH 7% CH2 -NH-CH2-mOrin Procedure A 0.2 g amount of the morin chelating resin, which had been stored in distilled water for 8-10 h, was loaded onto the adsorption column. Then, 2.0 ml of the mixed stock standard MoV1 and Wvl solution were transferred by pipette into beakers and diluted with distilled water to 100.0 ml. The solutions were adjusted to pH 2.0 and then passed through the adsorption column at a flow rate of 1.0 ml min-1. Molyb- denum(v1) and Wv' were desorbed quantitatively with 25 ml of 0.1 moll-1 NaOH, and the desorption solutions were evapor- ated to about 5 ml, then transferred into 10 ml flasks and diluted to the mark with distilled water.Molybdenum(v1) and Wvl were determined by inductively coupled plasma optical emission spectrometry (ICP-OES) using the following instru- ment settings: forward power, 1100 W; viewing height, 14 mm; plasma gas (argon) flow rate, 14 1 min-1; auxiliary gas (argon) flow rate, 0.6 1 min-1; nebulizer gas (argon) flow rate, 1.0 1 min-1; and wavelengths, 204.598 nm for MoV1 and 209.475 nm for Wvl. Results Effect of Acidity on Adsorption Equal concentrations of mixed MoV1 and Wvl standards were diluted to equal volumes, and the solutions concentrated using146 1.0 0.9 0.8 0" 0.7 0.6 0.5 0.4 \ ANALYST, FEBRUARY 1992, VOL.117 - - - - - - - 100.0 g 97.5 - c 0 .- c 95.0 2 92.5 90.0 100.0 - 97.5 s - C .- c 95.0 P a U 92.5 90.0 2 1 0.5 cHcl/mol I-' I 1 I I " J 1 2 3 4 5 6 7 PH - - - - - I 1 I I I Fig. 1 Effect of acidity on adsorption for: A, Mo; and B, W r 1 the adsorption column. The acidity was maintained in the range from 2.0 moll-' HCl to pH 7.0. The results (Fig. 1) show that MoV1 is adsorbed quantitatively at all the acidities, whereas Wvl is not adsorbed completely in 2.0 moll-1 HCl. As Wvl forms W03-H20 at pH 61.0,' the adsorption of MoV1 and Wvl at pH 2.0 is considerable. Adsorption Rate By using the procedure described above, the flow rates of the MoV1 and Wvl solutions through the column were varied from 0.5 to 2.5 ml min-1. The results, given in Fig. 2, show that MoV1 and Wvl are adsorbed quantitatively at a flow rate of 0.5-1.0mlmin-1.Hence a flow rate of 1.0ml min-1 was selected for the adsorption of MoV1 and Wvl. Dynamic Saturated Capacity of Adsorption A 0.05 g amount of chelating resin was weighed accurately and, after storing in distilled water for 8-10 h, it was loaded onto the adsorption column. Then, a 100 mg 1-1 MoV1 solution (pH2.0) was passed through the column at a flow rate of 1.0 ml min-1 and the eluate was collected in 10 ml fractions in order to determine the concentration of MoV1, until c = co, where co is the initial concentration of MoV1 in the solution and c is the concentration of MoV1 in the eluate. The results, given in Fig. 3, show that the dynamic saturated capacity of adsorption of the resin for MoV1 was 4.17 mmol per gram of dry resin.A 0.1OOO g amount of resin was weighed accurately and, using a similar method, the dynamic saturated capacity of adsorption of the resin for W1 was determined. The value obtained was 0.762 mmol per gram of dry resin; the results are given in Fig. 3. 0.3 ' I I I 1 0.1 0.3 0.5 0.7 [W]/mmol per gram of dry resin 1 I I I I 1 0 1.0 2.0 3.0 4.0 [Mo]/mmol per gram of dry resin Fig. 3 Dynamic saturated capacity of adsorption for: A, Mo; and B, W 5.0 4.0 CI, 3.0 0 E 2.0 1 .o 3.0 F - C 2.0 1 .o n 0 40 80 120 160 200 tlmin Fig. 4 Rate constant of adsorption for: A, Mo; and B, W Determination of Rate Constant A 0.1OOO g portion of resin was placed in each of two 100 ml conical flasks. A 50.0 ml volume of a MoV1 solution of pH 2.0 (Mo = 100 mg 1-1) was added to one of the flasks and 50.0 ml of a Wvl solution of pH 2.0 (W = 100 mg 1-l) were added to the other. Both flasks were attached to the electrodynamic oscillator and shaken at normal speed (100 cycles min-1).The metal uptake was determined at intervals of 20 min until equilibrium was reached (about 160 min). The results are shown in Fig. 4. According to Brykina et aZ.,g the isothermal adsorption equation for a low concentration of ions can be expressed as -ln(l - F) = kt, where F = QJQm, and t is the reaction time, Qt is the adsorption capacity at reaction time t, Qm is the adsorption capacity at equilibrium and k is the rate constant. The values of k obtained from the slope of a linear calibration graph were 4.58 x 10-4 s-1 for MoV1 and 3.88 x 10-4 s-1 for Wvl.The results are given in Fig. 4. Desorption Conditions and Desorption Curves After MoV1 and Wvl had been adsorbed by the resin following the above procedures, the columns were desorbed with 0.01, 0.05,O. 10,0.25 and 0.50 moll-1 NaOH solution, respectively. The results, given in Fig. 5, show that MoV1 and Wvl are desorbed quantitatively when the concentration of NaOH solution is higher than 0.1 moll-'. Hence 0.1 moll-1 NaOH solution was selected for the desorption of MoV1 and Wvl. By using the eluent selected above, the desorption curves of MoV1 and Wvl were obtained. The results, given in Fig. 6, show that 25 ml of eluent were sufficient for desorption.ANALYST, FEBRUARY 1992, VOL. 117 147 100 80 20 0 0.1 0.2 0.3 0.4 0.5 [NaOH]/mol I-' Fig.5 Effect of concentration of NaOH (25 ml) on desorption efficiency of: A, Mo; and B, W I I L 0 5 10 15 20 25 Amount of NaOH added/ml Fig. 6 Desorption curves of: A, Mo; and B, W, using 0.1 moll-1 NaOH as eluent Table 1 Results of the re-use of the resin Adsorption (YO) No. of times resin used Mo"' W V ' 1 99.8 100.0 2 100.0 100.5 3 100.5 100.0 4 99.3 99.9 5 99.6 99.3 6 99.5 100.1 Stability and Re-use of the Resin After the resin has been treated with strong acid or base, it can still be used for the adsorption of MoV1 and Wvl, with recoveries in the range 95.4-100%. Experiments were also carried out to determine the number of times the resin could be used. A column containing adsorbed MoV1 and Wvl was eluted with 3 mol 1-1 NaOH or 6mol1-1 HCI and then washed with 20 ml of 0.1 moll-1 HCI and distilled water until the washings were of neutral pH.This enrichment, desorption and neutralization procedure was repeated six times and the absorbing ability of the resin for MoV1 and Wvl was virtually unchanged. The results are given in Table 1. Interfering Ions Various interfering ions were added, respectively, to the diluted standard solutions of MoV1 and Wvl (0.4 mg 1-1) at a 100-fold excess over MoV1 and Wvl. The MoV1 and Wvl were then enriched and determined. Table 2 shows that a 100-fold excess of concomitant ions causes little interference and that a 50-fold excess of Fellr interferes seriously. However, if 0.1 g of ascorbate is used to reduce a 250-fold excess of Fell1 to Fe", the Table 2 Effect of interfering ions on MoV1 and Wv' at a concentration of 0.4 mg 1-l Recovery (%) Interfering Concentra- ion tion/mg 1-l MoV1 WV' 40.0 40.0 40.0 20.0 40.0 40.0 40.0 40.0 40.0 40.0 40.0 40.0 250.0 250.0 100.6 100.1 98.8 20.7 98.9 97.8 96.6 94.0 92.7 91.1 97.2 93.3 101.2 98.6* 97.7 97.6 200.5 21.3 91.7 97.7 97.1 96.6 97.7 106.3 105.1 91.4 100.2 96.7* * The recoveries were obtained after 250 mg 1-1 of Fellr had been masked with 5000 mg I - * of ascorbate.Table 3 Lowest limits of adsorption Amount Concentration Recovery (YO ) MoV1 Wvl VolumeA MoVi Wvl MoV1 Wvl 0.010 0.0125 0.50 20.0 25.0 95.7 73.3 0.010 0.025 0.50 20.0 50.0 95.3 91.8 addedmg (PPb) interference from Fell1 can be eliminated and the recoveries of MoV1 and Wvl are 98.1 and 96.3%, respectively. Lowest Limits of the Resin for Adsorption of MoV1 and Wv' Mixed solutions of MoV1 and Wvl of lower concentration (ppb level) were determined as described under Procedure.The results, given in Table 3, show that the morin chelating resin possesses a fairly strong absorbing ability for MoV1 and Wvl: 20ppb of MoV1 and 50 ppb of Wvl can be absorbed quantitatively. Precision and Analysis of Standard Samples Eight portions of a standard solution containing equal concentrations of MoV1 and Wvl were accurately transferred by pipette into beakers and diluted with distilled water to give a concentration of 0.4 mg 1-1. The diluted solutions were analysed as described under Procedure; the average results for eight determinations were 0.393 mg 1-1 for MoV1 and 0.398 mg 1-1 for Wvl and the relative standard deviations were 2.8% for MoV1 and 2.6% for Wvl.The four standard samples were dissolved as follows: a 0.1000 g amount of a standard sample was weighed accurately and dissolved in 10 ml of an acid mixture [HCI-HN03 (3 + l)] with heating (the undis- solved residue was removed by filtration). The solution was then transferred into a 100 ml flask and diluted to the mark with distilled water. A 5.0 ml aliquot of the solution was transferred by pipette into a 100 ml beaker and the contents of Mo and W in the standard sample were determined as described under Procedure. The results are given in Table 4. It can be seen that the contents of Mo and W determined are in agreement with the certified values. Discussion Conditions for the Synthesis of the Morin Chelating Resin The effects of the conditions used for the synthesis on the morin content of the chelating resin were investigated.The148 ANALYST, FEBRUARY 1992, VOL. 117 Table 4 Results of the analysis of standard samples using the WpGsed method Concentration (%) Relative Certified standard value Found deviation Sample Element (YO) Average (%) 1. 20Cr3MoWV (No. 192)* Mo 0.64 0.63 0.61 0.62 0.62 3.0 W 0.38 0.40 0.40 0.38 0.39 2.6 W 0.43 0.41 0.43 0.43 0.42 2.3 2. CrMoWV (No. H34)* Mo 0.46 0.45 0.44 0.45 0.45 2.2 3. 38CrWVAI (No. 169-2)* W 0.32 0.32 0.33 0.30 0.32 0 4. Standard steel? Mo 0.26 0.25 0.27 0.25 0.26 0 * Certified reference materials supplied by the Central Iron and Steel Research Institute of the Department of Metallurgical Industry of China.f Certified reference material supplied by the Iron and Steel Factory at Chongqin, Sichan province. Table 5 Effect of the conditions of the synthesis on the morin content of the resin Reaction Morin Reaction Morin Morin Morin temperature1 content* time1 contentt HCHOI content$ Morinl contents "C ("/I h (%) ml (% 1 g (Yo) 30 9.1 4.0 21.9 1.0 9.6 0.05 25.6 50 21.9 6.0 28.6 2.0 13.8 0.1 35.5 72 32.6 8.0 31 .O 4.0 35.5 0.2 21.9 10.0 32.6 6.0 25.9 0.3 16.7 * Aminated resin, 0.2 g; HCHO, 2.0 ml; morin, 0.1 g; 95% C2H50H, 20 ml; concentrated HCl, 0.5 ml; and reaction time, 10 h. f Aminated resin, 0.2 g; HCHO, 2.0 ml; morin, 0.1 g; 95% C2H50H, 20 ml; concentrated HCl, 0.5 ml; and reaction temperature, 72 "C. $ Aminated resin, 0.2 g; morin, 0.1 g; 95% C2H50H, 20 ml; reaction time, 10 h; reaction temperature, 72 "C; and concentrated HCI, 0.5 ml.§ Aminated resin, 0.2 g; HCHO, 4.0 ml; 95% C2H50H, 20 ml; reaction time, 10 h; reaction temperature, 72 "C; and concentrated HCI, 0.5 ml. 4000 3333 2666 2000 1666 1333 1000 666 333 Wave nu r n berlcm - Fig. 7 IR spectra of aminated poly(viny1 chloride) resin (spectrum l), morin chelating resin (spectrum 2), morin chelating resin saturated with MoV1 (spectrum 3) and morin chelating resin saturated with Wvl (spectrum 4) results, shown in Table 5, indicate that a higher content of morin can be obtained. However, if the proportion of materials used is incorrect and the reaction time is too long, the mechanical strength of the chelating resin obtained is affected; moreover, the catalytic effect of acid in the Mannich reaction must be taken into account.Hence, the following optimum conditions were selected for the synthesis: aminated poly(viny1 chloride) resin, 0.2 g; formaldehyde, 4.0 ml; morin, 0.1 g; ethanol, 20.0 ml; concentrated HCI, 0.5 ml; reaction time, 10 h; and reaction temperature, 72 "C. Structure of the Morin Chelating Resin and Adsorption Mechanism Fig. 7 shows the IR spectra of the aminated resin (spectrum 1), the morin chelating resin (spectrum 2) and the morin resin saturated with MoV1 (spectrum 3) or Wvr (spectrum 4). The peaks in Fig. 7 can be analysed as follows9~*0 (vmax/cm-*). Spectrum 1: 3340.5 (YN-H); 2908.7 and 2826.8 (6CH2 and YC-H); 1632.4 [vC=C of the highly conjugated system resulting from the dehydrochlorination reaction in the course of the amination of poly(viny1 chloride)5 and 6N-H]; 1571.6 (8N-H); and 1114.9 (pN-H).Spectrum 2: 3345.9 (YN-H); 2928.8 (vC-H); 1627.2 (vC=C + vC=O); 1440.2 (YC=C of the aromatic rings); the two peaks that were present in spectrum 1, at 1571.6 and 1114.9 cm-1, were not visible and this showed that the reaction was essentially complete. The additional peak at 1173.0 cm-1 in spectrum 2 may be due to the YC-0 of phenol and that at 1068.9 cm-1 to pN-H. Hence, according to the principle of the Mannich reaction and the IR spectra, the structure of the resin might be as shown below: -CH=CH-CHZ-CH- I or -CH =CH-CHZ-CH- I NH FH2 HO 0 OHANALYST, FEBRUARY 1992, VOL. 117 149 On comparing spectrum 3 with spectrum 2 in Fig. 7, it can be seen that the peaks at 3345.9 and 2928.8 cm-1 change slightly.At 1525.9 cm-1 a new but weak peak appears; this is due to the reaction of MoV1 with the >C=O function of the resin, which causes the peak for >C=O to be shifted from 1627.2 to 1525.9 cm-1. Two other peaks at 1173.0 and 1068.9 cm-l become weaker and are shifted to 1190.5 and 1098.1 cm-1, respectively; this shows that MoV1 can also react with the -OH and -NH- groups of the resin. Some obvious changes can be seen below 1000cm-1 and the peaks can be assigned as followsll (vmax/cm-l): 945.9 and 909.0 (vasymMo=O); 845.8 (vaSymMo=O); 711 .O and 664.3 (vasymMo-O-Mo). Therefore, it can be concluded that the mechanism of the adsorption of MoV1 onto the resin consists of two arts: the first is a chelating 3 or 5 to form a chelating structure with a five- or six-membered ring; the second part is an association mechan- ism, i.e., MoV1 reacts with the -NH- group to form an ion associate.On comparing spectrum 4 with spectrum 2 in Fig. 7, it can also be seen that there is only a slight change in the peaks at 3345.9 and 2928.8 cm-1; no peak appears at 1525 cm-1 and the peak at 1173.3 cm-1 is smaller and is shifted to 1178.3 cm-1. This is a small change and might indicate that Wvl reacts weakly with >C=O and -OH to form a chelating structure. That the peak at 1068.9 cm-* becomes weaker and is shifted to 1099.3cm-1 shows that the reaction between Wvl and the -NH- group of the resin is fairly strong. Compared with spectrum 3, some obvious changes can be seen in spectrum 4 below 1100 cm-1 and the peaks can be assigned as follows12 (vmaX/cm-l): 974.9 (vaSymW=O); 887.9 (vaSymW=O); 805.3 (60-W-0).Therefore, it can be concluded that the mechan- ism of the adsorption of Wvl onto the resin also consists of two parts: the main part is an association mechanism, i.e., Wvl reacts with the -NH- group to form an ion associate; the minor part is a chelating mechanism, i e . , Wvl reacts with the >C=O function and the OH group in position 3 or 5 of the resin to form a chelating structure with a five- or six- membered ring. mechanism, i.e., MoV1 reacts with P ,C=O and -OH in position Conclusion The proposed method using a morin chelating resin to adsorb trace amounts of MoV1 and Wvl from sample solutions selectively is satisfactory. The proposed method was also found to be efficient for the determination of trace amounts of MoV1 and Wvl in standard samples. The method is rapid, accurate and convenient. In addition, the chelating resin is stable and its synthesis is also simple and rapid. 1 2 3 4 5 6 7 8 9 10 11 12 References Goppelsroeder, F., Fresenius 2. Anal. Chem., 1868, 7 , 195. Sanz-Medel, A., and Garcia Alonso, J. I., Anal. Chim. Acta, 1984, 165, 159. Almassy, G., and Viyvari, M., Magy. Kem. Foly., 1956,62,332. Murata, A., and Yume Muchi, F., Shizuoka Daigaku Koga- kubu, Kenkyu Hokoku, 1958,9,97. Guo, X . W., Zhang, J. F., Su, Z. X., and Cao, D. R., Guangpuxue Yu Guangpu Fenxi, 1990, 10(3), 48. Spot Tests in Organic Analysis, Publishing House of Fuel Chemistry Industry, Beijing, 1972, pp. 315-316. Cotton, F. A., and Wilkinson, G., Advanced Inorganic Chemistry, Wiley, New York, 3rd edn., 1972, pp. 545-583. Brykina, G. D., Marchak, T. V., Krysina, L. S., and Velyavskaya, T. A., Zh. Anal. Khim., 1980, 35, 2294. Zhong, H. Q., Elementary IR Spectral Methods, Publishing House of Chemistry Industry, Beijing, 1984, pp. 118-132. Ning, Y. C., Structure Identification of Organic Compounds and Organic Spectroscopy, Qin Hua University Press, Beijing, 1989, Cousius, M., and Green, M. L. H., J. Chem. SOC., 1964, 1567. Sengupta, A. K., and Nath, S. K., Indian J . Chem., Secr. A , 1981, 20(2), 203. pp. 329-351. Paper 1104722A Received September 11, 1991 Accepted September 26, I991

ISSN:0003-2654    

DOI:10.1039/AN9921700145

出版商:RSC    

年代:1992

数据来源: RSC

摘要:

ANALYST, FEBRUARY 1992, VOL. 117 15 1 High-performance Liquid Chromatographic Study of Nickel Complexation With Humic and Fulvic Acids in an Environmental Water Peter Warwick and Tony Hall Department of Chemistry, L ough bo roug h University of Tech nolog y, Loug h borough, L eicestershire LE77 3TU, UK A high-performance liquid chromatographic method was developed to determine cation-exchange capacities and conditional association constants for metal interactions with humic and fulvic materials present in environmental waters. The method does not require prior extraction of the humic and/or fulvic compounds. A salt-gradient is used, which exploits the size-exclusion and adsorption properties of a coated porous silica stationary phase, in order t o separate and permit the measurement of the free and complexed metal concentrations.The results are subjected t o a weak and strong binding site interpretation. Keywords : Hum ic acid; fulvic acid; nickel co mplexa tion ; high -performance size-exclusion chromatograph y; water There is increasing interest in the fate and behaviour of trace metals in the environment and humic and fulvic acids, present in natural waters, play an important role. Humic and fulvic acids bind (complex) with metal pollutants and thereby affect such diverse phenomena as transport mechanisms, toxicity, bioavailability and the effectiveness of recovery and clean-up procedures.’ The acids generally occur as complex het- erogeneous mixtures of polymeric anions showing local and seasonal variations in composition.Probably every structural analytical technique, from classical elemental and functional group analysis to advanced instrumental analytical methods, including infrared (IR) , nuclear magnetic resonance (NMR) and mass spectrometry, have been applied to these com- pounds but owing to their complexity full structural elucida- tion has not been achieved. However, much information has been gained;2,3 for example, it has been discovered that humic and fulvic acids often contain aromatic backbones carrying a variety of functional groups, e.g., phthalate, salicylate and amine functions. With metals, humic and fulvic acids form anionic complexes, whereas with organic pollutants, mol- ecular association or covalently bound species can be pro- duced. One particular environmental concern is the interac- tion of humic and fulvic acids with radionuclides.Radionu- clides can enter the environment as a result of accidental or controlled releases of effluent. In addition, in the future it may be envisaged that over prolonged time periods eventual ingress of groundwater into planned radioactive waste reposi- tories is likely. Consequently, a full understanding of possible complexation and transport mechanisms is highly desirable. The speciation of a metal ( i . e . , distribution amongst the various possible physico-chemical forms) determines its over- all geochemical and biological behaviour. In order to assess the amount of ‘free’ metal species present, both the inorganic and organic complex speciation must be known, which requires a knowledge of the relevant stability constants.Also with humic and fulvic acids it is desirable to have a measure of the maximum amount of metal uptake that is likely under any given set of environmental conditions. This requires the determination of the maximum cation complexing capacity (C,) of the material. In order to explain the mechanism of metal binding with humic and fulvic acids, several models have been developed. In a review, Falck4 categorizes the models into two main types, either discrete ligand models or continuous distribution models. A survey of the literature suggests that the discrete ligand model is accepted by most workers. The model proposes that major binding sites (L), such as salicylate and phthalate, act as ligands towards the metals (M).The stoichiometry at a ligand site is 1 : 1, but of course the denticity may be higher. Hence the complexation reaction may be simply represented as M + L = M L for which the conditional association value (site binding constant) is given by where the brackets denote concentrations and in particular [L] is the concentration of free binding sites, i.e., not the concentration of humic or fulvic compounds. The number of ligand sites present determines the maxi- mum complexing capacity of the substance, i.e., (3) The mathematical interpretation of experimental data depends, of course, on the model adopted and no model has escaped criticism. A large number of analytical techniques have been applied to the determination of association and capacity values and are described in several texts.”6 However, in general, in order to simplify the chemistry involved investigations have been carried out using extracted humic and fulvic materials.Diethylaminoethyl (DEAE)-cellulose or XAD resins are commonly used as the extraction media.’ The adsorbed organics are eluted from these extractants using alkaline solutions in which they are readily soluble. However, such extraction procedures can be expected to change the proper- ties and structures of the organic compounds. States of aggregation, stereochemistry, inherent metal content and over-all purity are likely to be affected. Measured values of K and C, could well then be different to those applicable to the in situ material. Accordingly, this paper describes a high-perfor- mance liquid chromatographic technique which has been developed primarily to study in situ materials, i.e., direct investigation of the environmental water, without any prior treatment, apart from 0.45 pm filtration, which is employed to remove most of the colloidal clays and micro-organisms which may be present and mild rotary evaporation when preconcen- tration is necessary.The technique involves high-performance size-exclusion chromatography (HPSEC) and by means of152 - i --a .. . 4- c- Injection 0.05 mol I-' - NaCl Gradient ANALYST, FEBRUARY 1992, VOL. 117 A salt-gradient elution exploits cationic absorption, which is normally considered to be a disadvantage of HPSEC. A separation of anionically complexed metal from free metal is achieved, so that the relative amounts in an equilibrium mixture can be measured.During the time of the separation (<lo min), the dissociation of the previously formed complex must be negligible. The method was developed from work previously reported using Sephadex gels,8 but the gel tech- nique was rejected in favour of the higher speed and resolution of the high-performance technique. Accordingly, this investigation was conducted using nickel, a typical divalent transition metal, which is known to form complexes with humic and fulvic materials, and for which the kinetics of association and dissociation are slow , several days being required for the attainment of equilibrium. Nickel complexation has been studied extensively by various workers using extracted humic and fulvic materials.9-11 Nickel43 (t4 = 100 years, P,E,,, = 66 keV) was used to label the nickel mixtures so that the chromatographically separated com- plexed and free nickel could be assayed using liquid scintilla- tion counting.Currently the technique is being modified to study europium complexation using europium-152 (t4 = 13 years, y, 0.122 MeV, 62%) and solid-state counting. Pump Port Experimental Apparatus The HPSEC experiments were carried out using a Philips PU 4000 Series liquid chromatograph fitted with a PU 4100 gradient pumping system, a PU 4021 diode-array detector (DAD), a Rheodyne injection valve with a 100 pl loop and PU 6000 and PU 6003 integration and control software. SynChro- Pak GPC-60 guard (SO x 4.6 mm i.d.) and analytical (250 x 4.6 mm i.d.) size-exclusion columns were used, which con- tained a 5 pm porous silica stationary phase, coated with a glyceryl propyl bonded phase.This material has a stated linear relative molecular mass separation range of 300-20 000 for dextrans and 300-30 000 for proteins and is suitable for humic and fulvic acid fractionation. In exploratory experiments, the eluate emerging from the DAD was passed into a Waters Model 420 fluorescence detector, which was used to help to confirm the identity of the humic and fulvic fractions. In the complexation experiments the eluate emerging from the DAD passed through a Canberra Packard Flo-One/Beta (A140) radioactivity detec- tor, fitted with an 800 pl flow cell. A scintillation cocktail (Ecoscint A; National Diagnostics) was mixed with the column eluate before the flow-through cell.The experimental arrangement is shown schematically in Fig. 1. He . de-gassing . Environmental Water Sample A surface water was taken from moorland in the Derbyshire Peak District, near the village of Moscar. The water was subjected to 0.45 pm filtration, which was commenced 3 h after collection, and this was followed by the rotary evapora- tion of a 1000 ml sample to 250 ml at 30 "C, the resulting sample henceforth being referred to as ~4 moorland water. Rotary evaporation is a mild process and the 4-fold increase in concentration was undertaken to increase the ease of detec- tion of the humic and fulvic material. The rotary evaporation lowered the pH from the original in situ value of 3.8 to 3.5. As the extent of nickel complexation increases with increase in pH, a very small amount of concentrated NaOH solution was added to adjust the pH to 6.3.The estimated concentration of the humic and fulvic acid species from both the TOC (total organic carbon) and ultraviolet (UV) absorption data was about 52 mg 1-1. The results of the analysis of the X4 moorland water are given in Table 1. Guard column Fluorescence detector - Diode t- Waste Fig. 1 HPSEC experimental arrangement a+ Analytical column array detector Table 1 Analyses of working solutions Radioactivity Parameter PH Chloride Bromide Alkalinity as CaC03 Ammonia as N Nitrite as N Calcium Magnesium Sodium Potassium Total hardness as CaC03 Sulfate Phosphate as P Silica Fluoride Total oxidized nitrogen Total organic carbon Total inorganic carbon Boron Molybdenum Uranium Lithium Strontium Iron Manganese Aluminium Vanadium Lead Chromium Copper Nickel Zinc Cadmium Barium Cobalt A Units mg 1-1 mg 1-1 mg 1-1 mg 1-1 mg 1-1 mgl-1 mg 1-1 mg I-' mg 1-1 mg 1-1 mg 1-1 mg 1-1 mg 1-1 mg 1-1 mg 1-1 mg 1-1 mg 1-1 mg I-' mgl-1 mg I-' mgl-1 mg 1-1 - Pg I-' MI-' Pg 1- CLg1-l I% 1-' M1-l I % - ' Pg I-' CLg I-' Pg I-' Pg 1- Pg I-' - detector - Electrical conductivity at 20 "C pS cm-1 x4 Water 6.3 86 - - 0.130 0.013 21 14 48 107 122 <0.05 40 0.33 0.5 4.1 26 51 <0.01 <0.5 <o.1 - 0.08 3200 1780 330 < 10 <5 <2 63 10.3 65 300 < 10 420 0.53 Sodium humate 6.3 33 10 - 0.090 0.057 <1 <o. 1 25 <0.1 <1 123 <0.05 <o. 1 <0.05 <0.5 25 .o - < 10 <0.01 <o. 1 <0.01 510 <10 204 < 10 <5 <2 20 <5 <lo 4 .5 (10 < 10 108 - Purified Humic Acid Sample For comparison purposes, duplicate experiments were con- ducted using humic acid (HA) prepared from the semi- reference material sodium humate purchased from Aldrich. This material is well characterized, with capacity and complex- ation data readily available.Ql3 The purified HA wasANALYST, FEBRUARY 1992, VOL. 117 1- 153 produced by lowering the pH of a solution of Aldrich sodium humate to below 1 and then filtering off the precipitated HA. After washing and drying, 57.7mg of the precipitated HA were dissolved in 1 1 of high-performance liquid chromato- graphy (HPLC)-grade water. Finally, the pH was adjusted to 6.3 by the addition of a small amount of NaOH solution. The results of the analysis are given in Table 1.Preparation of Sample Mixtures Solutions of Ni(N03)2 were prepared in the range from 1 to 1 x 10-5 moll-1 from the analytical-reagent grade salt and then either 25 or 50 pl aliquots of these solutions were added to 5 ml samples of the ~4 moorland water and the purified humic acid solution. Mixtures were produced containing total nickel concentrations ranging from about 5.0 X 10-8 to 1.0 X 10-2 moll-'. Also, to each mixture either 25 or 50 pl of nickel-63 solution were added. The larger amount of nickel-63 was necessary for the more concentrated nickel solutions to permit detection of the complex in the presence of a large excess of free nickel. The nickel-63 solution was prepared by adding 135 pl of stock NiC12 solution (37 MBq ml-1; supplied by Amersham International) to 5 ml of HPLC-grade water.The total nickel concentrations of the mixtures were corrected for dilution effects, added nickel-63 and, for the ~4 water, the original nickel content. Characterization of the Dissolved Organic Matter The presence of humic and fulvic compounds in the moorland water was established as follows. A chromatographic separation of the filtered ~4 water was carried out using the GPC-60 column and the chromatogram obtained at 230 nm is shown in Fig. 2 (A). The elution volumes of the early peaks demonstrated the presence of large organic molecules. The UV absorption spectra of peaks 1, 2 and 3 showed a gradual increase in absorption, with decreas- ing wavelength, and the absence of specific absorption bands, properties which are typical of humic and fulvic materials.The narrow UV absorption spectrum of peak 4 indicated the presence of inorganic species, e.g. , NO3-, whereas the UV absorption spectrum of the low-intensity final peak (peak 5) suggested that small organic species were also present. A sample of the filtered ~4 moorland water was treated with DEAE-cellulose, which, as stated above, is known to extract humic and fulvic anions, and the chromatographic separation was repeated. The effect on the UV absorption monitored at 230 nm is shown in Fig. 2 (chromatogram B). The macromolecular organic species have been removed. Humic and fulvic compounds are generally fluorescent, hence the effect of the DEAE-cellulose treatment on the 2 I I I l l I l l I 0 2 4 6 8 10 12 14 16 18 Time/min Fig.2 UV chromatogram of x 4 moorland water monitored at 230nm: A, before and B, after treatment with DEAE-cellulose. (Peaks 1, 2 and 3 are due to large organic species, peak 4 is attributable to inorganic nitrate and peak 5 results from small organic species) fluorescence of the eluted species was determined. The results are shown in Fig. 3 (chromatograms A1 and Bl). Surprisingly, the first UV peak with a retention time of 9.3 min was not associated with fluorescence (cf., Figs. 2 and 3); however, the major UV peak exhibited fluorescence, which was removed by the DEAE-cellulose treatment. The fluorescent low relative molecular mass organic peak 5 was not completely removed. The fluorescence associated with the low relative molecular mass organic species may have indicated either fragmentation of the humic/fulvic materials or the presence of precursors. The lack of fluorescence associated with the largest molecules may be attributable to either the absence of appropriate aromatic groups or quenching caused by impurities and/or aggregation.Humic acid is by definition insoluble in very acidic solutions, i.e., pH tl. Accordingly, a sample was acidified and precipitation was observed. From measurements of the decrease in UV absorption a 70% fulvic-30% humic compo- sition was deduced (the precipitated humic material was separately tested and found to complex nickel in another series of experiments). From the above evidence, a knowledge of the source of the water, its light-brown colour and its acidity (pH = 3.8) when collected and the TOC content, it was concluded that the presence of humic and fulvic compounds had been estab- lished.Further corroborating evidence resulted from the nickel complexation experiments which, taking into account the time delay between the DAD and the radioactivity detector, 1 I I 1 1 I I I 1 I 0 2 4 6 8 10 12 14 16 18 Ti me/mi n Fig. 3 Fluorescence chromatogram of x 4 moorland water: Al, before and B1, after treatment with DEAE-cellulose. (For designa- tion of peaks see Fig. 2 and text) c-- Time -c Fig. 4 Simultaneous UV absorbance (230nm) and 63Ni activity (counts min-1) chromatograms versus time (min) for x 4 moorland water containing labelled Ni(N03)* solution154 ANALYST, FEBRUARY 1992, VOL. 117 0.40 8 0.30 C m e $j 0.20 Q 0.10 0 5 10 15 20 Ti me/m in Fig.5 UV absorbance chromatogram (230 nm) of purified Aldrich HA containing labelled Ni(N03)? solution. (Peak 1 is due to large organic species, peak 2 is due to nitrate and peak 3 results from small organic species) 45 min Run Equilibration start I I 1 35 min 15 min 5 min A (%) 100 100 100 0 B(%) 0 0 0 100 0 100 100 100 0 0 Fig. 6 Salt-gradient elution profile. Flow rate, 0.230 ml min-l; A, 0.05 moll-l NaCl; and B, 0.5 moll-' NaCl showed that nickel complexed with the species responsible for the early peaks (see Fig. 4). Chromatographic separation of the purified Aldrich HA yielded a simpler chromatogram. A typical result is shown in Fig. 5. The species producing peaks 1 and 3 are organic and fluorescent whereas peak 2 is due to added nitrate.Again, peak 3 is considered to be due to either humic fragments or precursors. Complexation Experiments After allowing 9 d for the mixtures to reach equilibrium, at 20°C, 100 yl samples of the mixtures were subjected to chromatographic analysis, using the salt-gradient technique, detailed in Fig. 6. Results In all instances the nickel activity eluted in two main fractions, as shown in Figs. 7 and 8, which show typical ~4 moorland water and purified HA chromatograms. The nickel humate/ fulvate complexes being partially excluded were eluted first with the 0.05 moll-1 NaCl and were identifiable with the early peaks in the corresponding DAD spectra, whereas the free Ni2+(aq) which suffered adsorption was eluted much later by the 0.5 moll-' NaCI.The nickel-63 chromatographic peak areas were used in conjunction with the known total nickel concentration to calculate the amounts of free and complexed nickel in each mixture. The assumption was made that the contribution of other nickel species to either peak was negligible. The results are given in Tables 2 and 3. A control experiment employing quench correction was conducted. Sample quenching was found to be insignificant, as peak area calculations employing disintegrations min-1 instead of counts min-1 gave identical results. Maximum Complexing Capacities In order to determine the maximum complexing capacities (C, values) of the ~4 moorland water and the purified HA, logarithmic plots of complexed versus free nickel concentra- tions were constructed (Figs.9 and 10). By using curve-fitting software (Macintosh Plus computer; Cricket Graph, polynomial order 2), best fit equations were 0 5 10 15 20 25 30 35 40 45 50 55 60 Ti melmin Fig. 7 Activity chromatogram of a typical x4 moorland water sample containing labelled Ni(N03)2 solution 0 5 10 15 20 25 30 35 40 45 50 55 60 Time/m in Fig. 8 Activity chromatogram of a typical HA sample containing labelled Ni(N03)2 solution obtained and differentiated to calculate the maximum com- plex concentrations theoretically achievable under the con- ditions used if precipitation is ignored. In this way the X4 moorland water was determined to have a C, value of 3.70 X 10-6mol 1-1 at a free nickel concentration of 3.18 X 10-3 moll-1, whereas the purified Aldrich HA yielded a C, value of 4.20 x 10-6 moll-1 at a free nickel concentration of 5.14 x 10-3 moll-'. In both instances precipitation occurred at about 1 x 10-3 moll-' of added nickel, which precluded precise experimental location of the maxima.Conditional Association Constant Determinations By using the appropriate maximum complexing capacity and eqn. (3), the individual values of the conditional association constants were determined for each mixture. The log Kvalues are included in Tables 2 and 3. For the moorland water the average log K was 4.40 [standard deviation (SD) = 0.41 (n = 9)] and for the purified Aldrich HA log K = 4.1 [SD = 0.25 (n = lo)]. Discussion This investigation was conducted at pH 6.3 to ensure a significant degree of complexation, as the exact nickel speciation of the original water was not an objective of this study.The results given in Table 4 cover a range of pH values. Generally, stability constants and maximum complexing capacities are found to decrease with decreasing pH, which is attributable to increased competition from H+ ions for binding sites. However, these experiments were conducted below pH 7 to avoid complications arising from the formation of hydroxy species. It can be seen that the log K values found are similar t o those reported by other workers for similar systems. Preconcentration results in pH changes, hence assessment of the nickel speciation in an original in situ water would need to take this into account. However, the inherent sensitivity ofANALYST, FEBRUARY 1992, VOL. 117 155 Table 2 Nickel complexation with x4 moorland water (C, = 3.72 x rnol 1-1) [Ni complex]/ [Nil/ moll-' mol 1-1 [Ni total]/ [Ni complex] [Nil moll-' (YO area) (% area) 2.619 x 10-7 16.01 83.99 4.193 x 10-8 2.200 x 3.107 x 10-7 15.69 84.31 4.875 x 10-8 2.620 x 10-7 7.075 x 10-7 13.41 86.59 9.488 x 10-8 6.126 x 1.197 x 10-6 16.05 83.95 1.921 x 10-6 1.005 x 5.163 x 10-6 10.26 89.74 5.297 X 10-7 4.633 x 10-6 1.006 x 10-5 5.85 94.15 5.885 x 10-7 9.472 x 10-6 4.973 x 10-5 2.86 97.14 1.422 X 10-6 4.831 x 10-5 9.829 x 10-5 2.67 97.33 2.624 x 10-6 9.567 x 10-5 4.927 x 10-4 0.443 99.557 2.183 x 10-6 4.905 x 9.805 x 10-4 0.443 99.557 4.344 X 10-6 9.762 X * Average log K = 4.40 [SD = 0.41 (n = 9)].Log [Ni complex] -7.378 -7.312 -7.023 -6.716 -6.276 -6.230 -5.847 -5.581 -5.661 -5.362 Log "il -6.658 -6.582 -6.213 -5.998 -5.334 -5.024 -4.316 -4.019 -3.309 -3.01 li Log K* 4.72 4.71 4.63 4.74 4.56 4.30 4.11 4.41 3.46 - Table 3 Nickel complexation with purified Aldrich HA (C, = 4.2 X mol I-l) [Ni complex]/ [Nil/ moll-1 moll-' [Ni total]/ [Ni complex] [Nil moll-' (YO area) (YO area) 8.692 x 10-8 7.40 92.60 6.432 x 10-9 8,049 x 10-8 1.357 x 8.99 91.01 1.220 X 10-8 1.235 X 5.325 x 10-7 7.32 92.68 3.898 x 10-8 4.935 x 1.022 x 10-6 7.11 92.89 7.266 X 10-8 9.493 X 4.988 x 6.46 93.54 3.222 x 4.666 x 9.889 x lo-" 5.61 94.39 5.548 x 10-7 9.334 x 10-6 4.955 x 10-5 2.60 97.40 1.288 X 10-6 4.826 X 9.811 x 10-5 2.15 97.85 2.109 x 10-6 9.600 X 4.927 x 10-4 0.58 99.43 2.833 x 10-6 4.899 x 9.805 x 0.37 99.63 3.628 X 10-6 9.769 X * Average log K = 4.10 [SD = 0.25 (n = lo)].Log [Ni complex] -8.191 -7.914 -7.409 -7.139 -6.492 -6.256 -5.890 -5.676 -5.548 -5.440 Log [Nil -7.094 -6.908 -6.307 -6.023 -5.331 -5.030 -4.316 -4.018 -3.310 -3.010 Log K* 4.28 4.37 4.28 4.27 4.25 4.21 3.96 4.02 3.60 3.81 Fig. 9 Ni-X4 moorland water binding. y = -6.1289 - 0.55856~ - 0.11186~~ and r2 = 0.986. Shaded area shows precipitation region Fig. 10 Ni-humate binding. y = -6.0161 - 0.55820~ - 0.12192~~ and r2 = 0.998. Shaded area shows precipitation region the technique can be exploited for direct measurements on in situ waters when the level of dissolved organic material is appropriately high, i.e., about 10 mg 1-1 TOC or higher. It is worth noting that the high SD of the current results is due in part to the wide concentration range studied, i.e., six orders of magnitude (see later), and the use of individual experimental results rather than the averaging of the means of several replicate series of experiments.Table 4 Over-all log K values Sample PH Log K Ref. Soil fulvic acid 5.0 4.20 2 Ground water fulvic acid 6.5 5.2 13 Stream fulvic acid 7.0 4.63 3 Lake water fulvic acid 8.0 5.14 4 x4 Moorland water 6.3 4.40 This work Purified Aldrich HA 6.3 4.10 This work ~~~~~~ ~~~~~ The apparent difference between the average log K values obtained for the moorland water and purified HA is not statistically significant. The capacities show about a 10% difference. The increased capacity of the purified material is arguably due to the freeing of metal sites during the dissolution and re-acidification stages. The values of K and C, obtained must be regarded only as operational values.They are not thermodynamic constants and cannot even be described as stoichiometric constants. The reasons are easily demonstrated by returning to eqn. (3). Rearrangement of this equation gives Hence a graph of [ML]/[M] against [ML] should be a straight line with a slope of - K and an intercept on the abscissa of The graphs obtained by treating the data in this way are presented in Figs. 11 and 12. The lack of linearity is immediately apparent. However, the C, values obtained from the intercepts on the abscissa are comparable to the previously produced values, i.e., 3.9 x 10-6mol 1-1 for the Ni-HA (previous value 4.20 x 10-6 moll-1) and 4.8 X 10-6 moll-1 for the ~4 moorland water (previous value 3.70 X 10-6 mol 1-1).It should be noted that Perdue14 stated that maximum capacities determined by adding excess of metal ion may be in error, especially at low ligand concentrations, and suggested that the H+ ion capacity should be used as an upper limit for the site capacity. This approach is precluded with an in situ water sample, but Kim et al. ,*5 for example, used one third of the H+ ion capacity as a measure of humic concentration in Am3+ complexation studies. However, the [MLlrnax.156 ANALYST, FEBRUARY 1992, VOL. 117 2000 rf 0 X 7 , log K = 3.8 1 0 1000 2000 3000 4000 5000 [ML]/10-9 mot I-' Fig. 11 Ni-x4 moorland water complexation loo0 ~ 800 K = 4.8 2 i 4 0 0 1 1 1 200 \ - - log K = 3.9 \ m 0 1000 2000 3000 4000 [ML]/10-9 mot 1-1 Fig. 12 Ni-humate complexation lack of linearity is because within humic and fulvic materials, there are a variety of ligand sites with different binding strengths.In more refined treatments of the ligand site model, capacities of each type of site are determined in order to derive individual binding site constants. Methods usually involve the Scatchard analysis technique, an approach much used by biochemists in drug-receptor binding studies.16 Figs. 11 and 12 can be interpreted in this way. If the simplifying assumption is made that only two types of site are present, A-sites, which are strongly binding, and B-sites, weakly binding, then from eqn. (4) Now, during the early stages of metal addition [MLIB = 0 because the added metal binds preferentially to the A-sites, and therefore the initial slope is -KA, and in the latter stages of the titration [MLIA = [ML]A,max, and therefore the final slope is -KB.The log K values thus determined are presented in Table 5 and again over-all similarity is observed. However, this approach can be criticized: the location of the linear sections is arbitrary and co-operativity and stereochemical effects may, in reality, cause the binding strengths and capacities to change continually as complexation progresses, i.e. , the tenets of the continuous distribution model.3 In yet another approach, Klotzl7 recommended the use of so-called stepwise stoi- chiometric constants. Accordingly, it is emphasized that the values given in Tables 2 and 3 are based on total capacities whereas the values in Table 5 are based on the weak and strong site approach.Conclusion Within the 'analytical window' of the technique and under the conditions used, only minor differences have been discovered between the nickel complexation properties of the dissolved humic and fulvic components of the ~4 moorland water and Table 5 Approximate strong and weak site log K values x 4 Moorland Purified Site water Aldrich HA Strong 5.5 4.8 Weak 3.8 3.9 the purified HA material. However, the ~4 water apparently possesses a proportion of stronger sites than the purified HA. The main objective of the study was achieved, namely to demonstrate that, when the rate of dissociation is slow, this form of HPSEC can be used in complexation studies to measure both complexed and free metal concentrations in environmental waters without prior extraction of the humic and fulvic materials. The technique is now being applied to compare extracted humic and fulvic materials from the same moorland water.A full account of the comparison will be reported elsewhere. 18 The authors thank the UK Department of Environment for funding this work. The results of this work will be used in the formulation of government policy, but the views expressed in this paper do not necessarily represent government policy. The Analytical Laboratories of the National Rivers Author- ity, Meadow Lane, Nottingham, are also thanked for the water analysis, as is Christine Bartrop for typing the manu- References script. 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 Patterson, J. W., and Passino, R., Metals Speciation, Separation and Recovery, Lewis, Chelsea, MI, 1987.Flaig, W., Beutelspacher, H., and Rietz, E., in Soil Com- ponents, ed. Gieseking, J. E., Springer-Verlag, New York, Humic Substances in Soil, Sediment and Water, eds. Aiken, G. R., McKnight, D. M., Wershaw, R. L., and MacCarthy, P., Wiley, New York, 1985. Falck, W. E., A Review of Modelling the Interaction Between Natural Organic Matter and Metal Cations, British Geological Technical Report WE/88/49, British Geological Survey, Not- tingham, 1988. Buffle, J., Complexation Reactions in Aquatic Systems, Ellis Horwood, Chichester, 1988. Christman, R. F., and Gjessing, E. T., Aquatic and Terrestrial Humic Materials, Ann Arbor Science Publishers, Ann Arbor, MI, 1983. Miles, C. J., Tuschall, J. R., Jr., and Brezonk, P. L., Anal Chem., 1983,55, 410. Warwick, P., Shaw, P., Williams, G. M., and Hooker, J. P., Radiochim. Acta, 1988, 44/45, 59. Schnitzer, M., and Hansen, E. H., Soil Sci., 1970, 109,333. Mantoura, R. F. C., and Riley, J. P., Anal. Chim. Acta, 1975, 78, 193. Haworth, D. T., Pitluck, M. R., and Pollard, B. P., J. Liq. Chromatogr., 1987, 10, 2877. Kim, J. I., Buckau, G., Li, G. H., Duschner, H., and Psarros, N., Fresenius J. Anal. Chem., 1990, 338, 245. Smith, B., Higgo, J. J. W., Moodie, P., Davis, J., Williams, G. M., and Warwick, P., Comparative Study of Humic Sub- stances in Groundwaters: I . The Extraction of Humic Material from Drigg Groundwater and a Study of its Ability to Form Complexes with Cobalt and Nickel, DOE Report No. DOE/ HMIP/R/90/087. DOE, London, 1990. Perdue, M. E., in Metal Speciation: Theory, Analysis and Application, eds. Kramer, J. R., and Allen, E. H., Lewis, Chelsea, MI, 1988. Kim, J. I., Rhee, D. S., and Buckau, G., Radiochim. Acta, 1991, 52/53, 49. Scatchard, G., Ann. N.Y. Acad. Sci., 1949, 51, 660. Klotz, I. M., Acc. Chem. Res., 1974, 7, 162. Warwick, P., and Hall, A,, Radiochim. Acta, submitted for publication. Paper 1 I04583 K Received September 4, 1991 Accepted September 30, 1991 1975, VOI. 1, pp. 1-211.

ISSN:0003-2654    

DOI:10.1039/AN9921700151

出版商:RSC    

年代:1992

数据来源: RSC