摘要:

Flow Injection On-line Sorption Separation and Preconcentration With a Knotted Reactor for Electrothermal Atomic Absorption Spectrometric Determination of Lead in Biological and Environmental Samples XIU-PING YAN AND FREDDY ADAMS* Department of Chemistry, University of Antwerp (UIA), B-2610 Antwerp (W ilrijk), Belgium A selective and robust flow injection on-line sorption preconcentration ETAAS systems, including sorbent extracseparation and preconcentration procedure was developed for tion,3–7 ion exchange,8,9 solvent extraction,10,11 coprecipithe ETAAS determination of (ultra)trace amounts of lead in tation,12,13 and sorption with a knotted reactor (KR).14,15 biological and environmental samples. With the use of The majority of work conducted in this area has been diethyldithiophosphate as complexing agent and citric acid as associated with preconcentration on packed columns either by masking agent, the analyte complex was selectively formed ion exchange or by adsorption.1–9 Compared with on-line ionand adsorbed onto the inner walls of a PTFE knotted reactor exchange separation and preconcentration techniques, those at pH 0.5–3.5 with removal of the high-content salts in saline based on sorbent extraction using bonded-silica with octadecyl water and most of the other metals in digested biological and functional groups (RP C18) as sorbent and diethyldithiocarbaenvironmental samples. The collected analyte complex was mate (DDC) as complexing agent have been shown to be more eluted quantitatively with 35 ml of ethanol and all the eluate selective for the determination of trace amounts of heavy was directly introduced into the graphite tube without a L’vov metals in sea-water since the most common matrices in seaplatform by an air flow.Neither a pre-heating step nor precise water, viz., alkali and alkaline earth elements, do not form timing was needed during elution and eluate introduction. complexes with DDC and can be quantitatively separated ETAAS determination of the concentrated analyte was carried from the other metals.1–3 However, the determination of trace out in parallel with the next preconcentration cycle.An amounts of metals using the DDC–RP C18 sorbent extraction enhancement factor of 125 and a detection limit (3s) of 4.8 ng system suffers interferences from high concentrations of other l-1 Pb, along with a sampling frequency of 31 h-1, were transition metals in the sample due to the competition of their achieved with a 60 s loading time at a sample flow rate of DDC complexes for active sites at the column packing. In 5.4 ml min-1.The precision for 11 replicate determinations at comparison with DDC, diethyldithiophosphate (DDP) has the 0.5 mg l-1 Pb level was 2.1% (RSD). The analytical results been found to be a much more selective complexing agent for for a number of biological and environmental certified the RP C18 column preconcentration and separation of cad- reference materials were in good agreement with the certified mium, lead and copper in the presence of other heavy metals.7 values.Compared with FI on-line preconcentration systems using sorbent columns, those based on the sorption of Keywords: Flow injection ; on-line separation and metal–DDC15–17 or PDC (pyrrolidine dithiocarbamate)14 com- preconcentration; electrothermal atomic absorption plexes on the inner walls of a KR permit the use of higher spectrometry ; knotted reactor; lead; biological and environmental samples sample loading rates, thus achieving higher concentration efficiencies due to low hydrodynamic impedance in the KR, and allow the analysis to be conducted at low cost, owing to ETAAS is a widely used sensitive technique for trace element the long lifetime of the KR and its ease of construction with analysis; however, the direct determination of (ultra)trace no need for packing materials.These novel systems have been amounts of elements in complicated matrices by ETAAS is successfully applied to flame AAS for the determination of usually difficult owing to matrix interferences and/or cadmium in biological samples, and of copper in water and insufficient detection power. Consequently, separation and rice samples by Fang and co-workers.16,17 Difficulties in the preconcentration procedures, such as ion exchange, adsorption, direct coupling of FI on-line KR sorption separation and solvent extraction and coprecipitation, are often needed before preconcentration to ETAAS are due to the discrete non-flow- the ETAAS determination.Conventional off-line procedures through nature of ETAAS, the limited capacity of the graphite for the separation and preconcentration, although effective, tube and the relatively large dead volume of the KR. are usually time consuming and tedious, require large amounts Nevertheless, the applications have been successfully extended of sample and reagents and are vulnerable to contamination to ETAAS for the automated determination of lead in sea- and analyte loss.water by Sperling et al.15 and antimony(III ) in water samples Recently, on-line separation and preconcentration using flow by Yan et al.14 The developed FI manifold and the operation injection (FI) techniques have been shown to be efficient and sequences described by Yan et al.14 are the key to the successful effective in enhancing the sensitivity and specificity of on-line coupling of FI KR sorption separation and preconcen- ETAAS.1–3 With the on-line operation, the drawbacks of the tration systems to ETAAS.The results14–17 indicate that the off-line (batchwise) operation can be overcome to a great KR sorption system is a promising alternative with many extent while the benefits of separation and preconcentration advantages in comparison with the previously used column are further enhanced. To date, various kinds of separation techniques have been adapted to FI on-line separation and systems.3–7 However, the use of DDC or PDC as the com- Journal of Analytical Atomic Spectrometry, April 1997, Vol. 12 (459–464) 459plexing agent in FI on-line KR sorption separation and Germany) at pH 2 was used as the complexing agent. Working standard solutions were prepared by stepwise dilution of preconcentration ETAAS systems14,15 could suffer similar interferences from other heavy metals to those encountered in the 1000 mg l-1 Pb stock solution (Merck) immediately prior to use.Suprapur HNO3 (65% m/m) (Merck), HF (40% m/m) DDC-RP C18 column system.3 In this work, an attempt is made to develop a more selective (Merck) and HClO4 (70% m/m) (Merck) were used for the digestion of biological and environmental samples. FI on-line KR sorption separation and preconcentration system for ETAAS determination of lead in the presence of alkali, alkaline earth and heavy metals using DDP as complexing agent. The feasibility and accuracy of the proposed Certified Reference Materials method are demonstrated by the analysis of a number of biological and environmental CRMs.The National Research Council of Canada (NRCC) CRMs CASS-3 (Coastal Sea-water) and SLRS-3 (River Water), and the Community Bureau of Reference (BCR) CRM 141 EXPERIMENTAL (Calcareous Loam Soil), CRM 320 (River Sediment) and CRM Apparatus 422 (Cod Muscle) were used to check the accuracy of the method.A Perkin-Elmer (Norwalk, CT, USA) Model 3030 atomic absorption spectrometer equipped with a deuterium arc background corrector and a Model HGA-500 graphite furnace was employed throughout this work. A lead hollow cathode lamp Sample Pre-treatment (Perkin-Elmer) was used as the radiation source at a wave- No pre-treatment of CASS-3 and SLSR-3 was needed as the length of 283.3 nm with a 5 mA lamp current and a 0.7 nm acidity of these samples (pH 1.6) adjusted by the supplier for slit-width (low).Pyrolytically coated graphite tubes without conversion purposes fell in the optimum pH range for the platforms (Z-tek, Amsterdam, The Netherlands) were used. preconcentration. The graphite furnace temperature programme used is given in A 0.1 g amount of Cod Muscle (CRM 422) was weighed Table 1. Maximum power heating and internal gas stop were accurately into the PTFE beaker of a standard digestion bomb. used during atomization. Peak height (absorbance), peak area Then, 2 ml of 65% m/m HNO3 were added, and the open (integrated absorbance) and statistical data were printed out bomb, which was covered by a cap with a small hole to prevent with a Perkin-Elmer PR-100 printer. Integrated absorbance contamination, was heated gently on a hot-plate until fumes (Aint) with an integration time of 2 s was used for evaluating appeared. After the solution had cooled, 0.5 ml of 70% m/m results throughout this work.HClO4 was added.The bomb was sealed and heated in an A Perkin-Elmer FIAS-200 FI accessory was used for the FI oven for 8 h at 150°C. The contents of the opened bomb were on-line sorption preconcentration. The standard valve of the evaporated to near dryness on a hot-plate (warning: care FIAS-200 was replaced by a prototype 8-channel 16-port should be taken because of the use of HClO4 in a pressure multifunctional injector valve (Tecator, Ho�gana�s, Sweden). The bomb and the subsequent evaporation to near dryness!).The injector valve is of double-layer rotary design, with eight residual solution was diluted to 50 ml with doubly de-ionized channels on the rotor, separated by 45° from each other. When water. actuated, the rotor turns through 45°, and all the channels on A 0.1 g amount of Calcareous Loam Soil (CRM 141) or the rotor are shifted one channel position in relation to those River Sediment (CRM 320) was weighed accurately into the on the stator.The rotation speed of the two peristaltic pumps, PTFE beaker of the digestion bomb, and then 3 ml of 40% their stop and go intervals, and the actuation of the injector m/m HF and 1 ml of 65% m/m HNO3 were added. The open valve were programmed and controlled by a separate computer bomb with the cap was heated gently on a hot-plate until (DECstation, 316sx), independent of the spectrometer. A labor- fumes appeared. After the solution had cooled, 0.5 ml of 70% atory-made KR consisting of 100 cm×0.5 mm id PTFE tubing m/m HClO4 was added, the bomb was sealed and heated in was used for the collection of the analyte complex.Ismaprene the oven for 8 h at 150 °C. The contents of the beaker were pump tubes (Ismatec, Wertheim, Germany) were employed to transferred into a 100 ml flask and diluted with doubly propel the sample, reagent, eluent and air. All connections de-ionized water. were made with 0.35 mm id PTFE tubing. The confluence point of the sample and the complexing agent solution was positioned about 2 cm upstream of the valve.Procedures Reagents and Standard Solutions The FI manifold for the two different valve positions is shown in Fig. 1. Details for the duration and function of each sequence All reagents used were of the highest purity available, and of at least analytical-reagent grade. Doubly de-ionized water (18 are given in Table 2. The preconcentration was carried out in parallel with the ETAAS determination of the previous sample MV cm-1), prepared from a Milli-Q water-purification system (Millipore, Bedford, MA, USA), was used throughout this concentrated.A complete cycle of preconcentration and eluate introduction without a pre-fill stage lasts 112 s with a sample work. Ammonium diethyldithiophosphate (DDPA) (Aldrich, Steinheim, Germany) in 2% m/v citric acid (Merck, Darmstadt, loading period of 60 s. Table 1 Graphite furnace temperature programme for the determination of lead in the ethanolic eluate with the use of pyrolytically coated graphite tubes without platforms Argon flow rate/ Step Temperature/°C Ramp time/s Hold time/s ml min-1 Drying 100 5 30 300 Pyrolysis 200 5 20 300 Atomization 2000 0 2 0 Cleaning 2600 1 2 300 460 Journal of Analytical Atomic Spectrometry, April 1997, Vol. 12Method Development The FI manifold and operation sequences used for method development were identical with those proposed in recent work.14 A univariate approach was used for optimization starting with conditions based on those used previously for FI column preconcentration.7 The integrated absorbance (peak area) was taken as the main figure of merit, but with simultaneous considerations of precision and efficiency.Two or three cycles of univariate studies were needed to establish the fixed parameters for each univariate study so that they approached optimum values. RESULTS AND DISCUSSION Optimization of Experimental Parameters Chemical parameters including sample acidity, DDPA concentration, composition of rinsing solution, FI variables (KR length, the volume of the eluent loop, the flow rate for elution and eluate introduction, etc.) and ETAAS parameters were optimized according to the procedures described under Method Development.Complexing agent, its concentration and flow rate DDPA was chosen as the complexing agent since it is more selective for lead in the presence of other heavy metals than the dithiocarbamates.7 The influence of the DDPA concentration on the complexing of lead is shown in Fig. 2.The integrated absorbance was constant in the DDPA concentration range 0.6–1.5% m/v. Without DDPA, no analyte could be collected on the KR. The results indicated that the Fig. 1 Schematic diagram of FI on-line sorption preconcentration for ETAAS: P1, P2, peristaltic pumps; S, sample; CR, complexing reagent (0.8% m/v DDPA+2% m/v citric acid, pH 2); WL, wash liquid (0.2% m/v DDPA+0.4% m/v citric acid, pH 2); E, eluent (ethanol); KR, knotted reactor (100 cm×0.5 mm id PTFE tubing); W, waste; EL, eluent loop (35 ml); EC, eluent container; V, injector valve; DT, delivery tube; and ETA, electrothermal atomizer.Valve position: Fig. 2 Effect of DDPA concentration in the complexing agent (a) inject; (b) fill. solution on the integrated absorbance of 0.5 mg l-1 Pb with 30 s preconcentration. All other conditions as in Fig. 1, and Tables 1 and 2. Table 2 Operation sequence of the FI on-line KR sorption separation and preconcentration system for the determination of lead by ETAAS Step Function Time/s Pump active Flow rate/ Medium Valve ml min-1 pumped position 1 Insert DT into ETA 2 — — — Inject 2 [Fig. 1(a)] Elute and introduce 25 2 3 Air Inject eluate into ETA 3 Withdraw DT 2 — — — Inject 4 [Fig. 1(a)] Pre-fill the sampling 10 1 5.4 Sample Inject tubes 1.4 Complexing agent 5 [Fig. 1(b)] Load sample 60 1 5.4 Sample Fill 1.4 Complexing agent 6 [Fig. 1(a)] Rinse KR 10 1 3 Rinsing Inject solution 7 [Fig. 1(a)] Remove residual 10 2 3 Air Inject solution 8 [Fig. 1(b)] Fill EL 3 2 3 Ethanol Fill Journal of Analytical Atomic Spectrometry, April 1997, Vol. 12 461metal–DDP complex could be collected on the inner walls of complex. Thus, the influence of the tube length of this part was examined. Both sensitivity and precision decreased as the a PTFE KR. In further experiments, 0.8% m/v DDPA in 2% m/v citric acid (pH 2) was employed as the complexing agent.tube length increased if no pre-fill step was used before preconcentration. However, after a sufficient time for a pre-fill The citric acid in the complexing agent solution acted as a masking agent, as used in previouswork on sorbent extraction.7 step, no significant changes in sensitivity and precision could be seen. This is probably due to the saturation of the active The influence of the flow rate of the complexing agent solution (0.8% m/v DDPA+2% m/v citric acid) was examined sites on the tubing walls by the Pb–DDP complex formed during a pre-fill step.The duration needed in a pre-fill step for at a sample flow rate of 5.4 ml min-1. No significant changes in sensitivity and precision were observed within the flow rate this purpose was found to increase with an increase in the tube length. Therefore, the connection between the confluence range 1.0–2.0 ml min-1. A flow rate of 1.4 ml min-1 was used for the delivery of the complexing agent solution for further point of the sample with the complexing agent solution and the valve was kept as short as possible (about 2 cm).experiments. Sample acidity KR rinsing The importance of rinsing the KR carefully before elution was The effect of sample acidity on the integrated absorbance is shown in Fig. 3. The optimH of the sample solution discussed in detail previously.14 An efficient KR rinsing step should effectively remove the non-adsorbed constituents of the ranged from 0.5 to 3.5 when the flow rates of the sample and the complexing agent solution (0.8% m/v DDPA+2% m/v matrix and/or even those weakly bound concomitant elements, but not strip the adsorbed analyte from the KR.Depending citric acid, pH 2) were kept at 5.4 and 1.4 ml min-1, respectively. For further experiments a pH value of 2 was used as the on the pH of the rinsing solution and the stability of the complexes formed, the more weakly bound complexes could acidity of the sample solution.be removed while the more strongly bound complexes could still be retained on the KR. The most important factor in this Sample flow rate and loading time context is the kinetic stability of the analyte–DDP complex and the strength of sorptive bonding to the KR in acid solution. Studies on the effect of sample flow rate showed that the In this work, the adsorbed Pb–DDP complex was found to integrated absorbance of 0.5 mg l-1 Pb with a 30 s preconcenbe easily removed with dilute HNO3 or even de-ionized water tration time increased almost linearly as sample flow rate as a rinsing solution, probably due to the instability of the increased to 4.3 ml min-1, and then gradually levelled off with Pb-DDP complex. It was found to be necessary to add further increase, implying insufficient contact time for achieving appropriate amounts of DDPA and citric acid to the rinsing complete sorption.However,a sample flow rate of 5.4 ml min-1 solution in order to prevent analyte loss.The influences of was chosen to obtain a better sensitivity while maintaining a DDPA and citric acid concentration in the rinsing solution on reasonable degree of adsorption efficiency to ensure good integrated absorbance are shown in Figs. 4 and 5, respectively. reproducibility. The minimum concentrations of DDPA and citric acid for The effect of sample loading time on the integrated preventing analyte loss were found to be 0.2 and 0.4% m/v, absorbance of 0.5 mg l-1 Pb was tested at a sample flow rate respectively, which are close to the concentrations in the of 5.4 ml min-1.With the use of a 100 cm long KR, the mixture of sample and chelating agent solution during sample integrated absorbance increased almost linearly up to a 50 s loading. Furthermore, the use of a mixture of 0.2% m/v DDPA preconcentration time, after which the slope decreased graduand 0.4% m/v citric acid as the rinsing solution at a flow rate ally, presumably as a result of an insufficient capacity of the of 3 ml min-1 for a 10 s rinsing of the KR was found to be KR and/or a chromatographic effect with the adsorbed analyte efficient for the removal of the residual matrix, giving back- being partially leached by further sample.14,16 However, about ground signals of the sample solutions as low as those of 81% of the analyte was collected with a 60 s preconcentration standard solutions. time at a sample flow rate of 5.4 ml min-1.Removal of residual solution T ube length between the confluence point of sample with complexing agent solution and the valve As discussed previously,14,15 the minimization of dispersion and the eluent volume needed for quantitative elution is critical Analyte loss on the tubing walls between the confluence point to the successful adaptation of the KR sorption separation of the sample with the complexing agent solution and the and preconcentration system to the requirements of ETAAS.valve might occur owing to the adsorption of the Pb–DDP Fig. 4 Influence of DDPA concentration in the rinsing solution on Fig. 3 Influence of sample acidity on the integrated absorbance of 0.5 mg l-1 Pb with a preconcentration time of 30 s. All other conditions the integrated absorbance of 0.5 mg l-1 Pb with 30 s preconcentration. All other conditions as in Fig. 1, and Tables 1 and 2. as in Fig. 1, and Tables 1 and 2. 462 Journal of Analytical Atomic Spectrometry, April 1997, Vol. 12(0.8–2.6% RSD) were obtained in the range 2.7–3.4 ml min-1. Thus, a flow rate of 3 ml min-1 was used for elution and eluate introduction. ETAAS parameters In this work, neither a pre-heating step nor a L’vov platform was found to be necessary for successful eluate introduction as only 35 ml of eluate were delivered to the electrothermal atomizer. Moreover, there were no significant influences on sensitivity and precision at drying temperatures of 80–120 °C.The experiments on the effect of pyrolysis temperature showed that the optimum pyrolysis temperature ranged from 200 to 400 °C. Therefore, a drying temperature of 100 °C and a pyrolysis temperature of 200 °C were used for the optimization Fig. 5 Influence of citric acid concentration in the rinsing solution on the integrated absorbance of 0.5 mg l-1 Pb. Other conditions as in of other parameters and for all analytical applications.Fig. 1, and Tables 1 and 2. Interference Studies To achieve this, the residual rinsing solution in the KR and in Interferences from co-existing ions were evaluated under the the eluate delivery tube was completely removed by an air conditions shown in Tables 1 and 2. No interferences from the flow at 3 ml min-1 before elution. most common matrix constituents, viz., alkali and alkaline earth element chlorides in sea-water, were observed due to the Elution and eluate introduction group-specific character of the complexing agent and the efficient rinsing step. As shown in Table 3, heavy metal ions, To minimize the eluent volume needed for quantitative elution viz., FeIII, NiII, ZnII, MnII, CuII and CdII, were tolerable up to and to facilitate eluate delivery, dispersion during elution and at least 800, 100, 100, 10, 0.2 and 0.2 mg l-1, respectively.eluate introduction must be reduced, while a highly efficient Compared with the use of PDC or DDC as complexing agent eluent should be used.Moreover, the KR tube length, and the without masking agent for the KR sorption system, the present flow rate for elution and eluate introduction have to be system gave much better selectivity for the separation and optimized. As shown in previous work,14,15 the use of an eluent preconcentration of lead in the presence of FeIII, NiII, ZnII loop and air flow-driving elution and eluate introduction were and MnII, probably due to the selectivity of DDPA and the effective for this purpose.Ethanol was chosen as the eluent masking effect of citric acid. However, the interferences from due to its effective elution of the adsorbed analyte complex, its CuII and CdII are similar in both systems owing to the easy direct delivery through peristaltic pump tubing and its competition of these ions for complexing agents and of the relatively low toxicity. formed complexes for the active sites on the inner walls of The effects of eluent volume and KR tube length on the the KR.integrated absorbance of 0.5 mg l-1 Pb are shown in Fig. 6. For a 100 cm long KR, the integrated absorbance increased as the eluent volume increasedup to 35 ml, then remainedconstant Analytical Performance of the FI On-line KR Sorption ETAAS with further increase in the eluent volume. As the KR tube System length increased, both the eluent volume needed for quantitat- Characteristic data on the performance of the FI on-line KR ive elution and sensitivity increased.Although a longer KR separation and preconcentration system for ETAAS with the could offer higher sensitivity, the resulting larger volume of use of a 60 s preconcentration time and a sample flow rate of eluate is less favourable for the introduction of the entire eluate 5.4 ml min-1 are summarized in Table 4. In comparison with into the graphite tube due to its limited capacity. Hence, an the direct injection of 35 ml of an aqueous solution, an enhance- eluent volume of 35 ml along with a 100 cm long KR was ment factor of 125 was achieved with a consumption of 6.3 ml chosen as a compromise.of sample solution. The adsorption efficiency was evaluated to The influences of the flow rate of elution and eluate be 81% compared with the total analyte mass loaded onto the introduction on sensitivity and precision for triplicate KR. Despite the incomplete adsorption, good reproducibility determinations of 0.5 mg l-1 Pb were investigated in the range was obtained.The RSD was found to be 2.1% at the 0.5 mg 1.7–5.1 ml min-1. Maximum sensitivity and high precision l-1 level. Furthermore, the linear range of the calibration graph spans at least two decades. The accuracy of the developed method was tested by analys- Table 3 Effect of co-existing ions on the integrated absorbance of 0.5 mg l-1 Pb under the conditions listed in Fig. 1, and Tables 1 and 2 Co-existing ion Concentration/mg l-1 Relative Aint FeIII 800 0.96 1000 0.75 NiII 100 0.98 ZnII 100 0.97 MnII 10 0.96 100 0.85 CuII 0.2 0.97 Fig. 6 Influences of KR tube length and eluent volume on the 0.5 0.55 integrated absorbance of 0.5 mg l-1 Pb with 30 s preconcentration. KR CdII 0.2 0.98 tube length (cm): A, 50; B, 100; C, 150; and D, 200. All other conditions 1 0.64 as in Fig. 1, and Tables 1 and 2. Journal of Analytical Atomic Spectrometry, April 1997, Vol. 12 463Table 4 Performance of the FI on-line KR sorption preconcentration- CONCLUSIONS ETAAS system for the determination of lead.Conditions as in Fig. 1, The results presented here have shown that the analyte–DDP and Tables 1 and 2 complex can be efficiently collected on the inner walls of a Enhancement factor* 125 PTFE KR. The combination of a more selective complexing Adsorption efficiency (%)† 81 agent (DDP) with a masking agent (citric acid) and a newly Sampling frequency/h-1 31 developed FI on-line KR separation and preconcentration Sample consumption/ml 6.3 ETAAS system provides a selective and robust method for the Reagent consumption/ml: determination of lead in biological and environmental samples Ethanol 0.035 0.8% m/v DDPA+2% m/v citric acid 1.6 with relatively complicated matrices at low analyte concen- 0.2% m/v DDPA+0.4% m/v citric acid 4 trations without the need for chemical modification in the Detection limit (3s)/ng l-1 4.8 graphite furnace.Precision (%RSD) (n=11) 2.1 (0.5 mg l-1) Calibration function Aint=-0.001+0.436CPb The authors thank W.Van Mol for useful discussions and (six standards in the range 0.01–1 mg l-1, E. Ivanova for the digestion of CRM 422. X.-P. Y. also CPb in mg l-1) Regression coefficient 0.9998 acknowledges funding for a Postdoctoral Research Fellowship from UIA (University Research Fund) and NFWO (Belgium). * Compared with direct injection of 35 ml aqueous solution. † Compared with total mass loaded onto the KR. REFERENCES Table 5 Results for the determination of lead in biological and 1 Fang, Z.-L., Flow Injection Separation and Preconcentration, VCH, environmental samples Weinheim, 1993. 2 Fang, Z.-L., Flow Injection Atomic Absorption Spectrometry, Sample* Certified Determined† Wiley, Chichester, 1995.CASS-3 0.012±0.004 mg l-1 0.011±0.002 mg l-1 3 Welz, B., Microchem. J., 1992, 45, 163. (Coastal Sea-water) 4 Sperling, M., Yin, X.-F., and Welz, B., J. Anal. At. Spectrom., SLRS-3 0.068±0.007 mg l-1 0.062±0.004 mg l-1 1991, 6, 295.(River Water) 5 Porta, V., Abollino, O., Mentasti, E., and Sarzanini, C., J. Anal. BCR CRM 422 0.085±0.015 mg kg-1 0.081±0.008 mg kg-1 At. Spectrom., 1991, 6, 119. (Cod Muscle) 6 Welz, B., Sperling, M., and Sun, X.-J., Fresenius’ J. Anal. Chem., BCR CRM 141 29.4±2.6 mg kg-1 29.0±0.6 mg kg-1 1993, 346, 550. (Calcareous Loam Soil) 7 Ma, R.-L., Van Mol, W., and Adams, F., Anal. Chim. Acta, 1994, BCR CRM 320 42.3±1.6 mg kg-1 40.7±1.0 mg kg-1 293, 251. (River Sediment) 8 Beinrohr, E., Cakrt, M., Rapta, M., and Tarapci, P., Fresenius’ Z. Anal. Chem., 1989, 335, 1005. * 60 s loading time for water samples and 30 s for the digests of 9 Azeredo, L. C., Sturgeon, R. E., and Curtius, A. J., Spectrochim. biological and environmental samples. Acta, Part B, 1993, 48, 91. † The±terms represent 95% confidence intervals (n=5). 10 Backstrom, K., and Danielsson, L. G., Anal. Chim. Acta, 1990, 232, 301. 11 Tao, G., and Fang, Z., Spectrochim. Acta, Part B, 1995, 50, 1747. 12 Fang, Z.-L., and Dong, L.-P., J. Anal. At. Spectrom., 1992, 7, 439. ing a number of biological and environmental CRMs. The 13 Chen, H.-W., Xu, S.-K., and Fang, Z.-L., J. Anal. At. Spectrom., 1995, 10, 533. determination of lead in these materials was performed using 14 Yan, X.-P., Van Mol, W., and Adams, F., Analyst, 1996, 121, 1061. simple aqueous standards for calibration. The digests of River 15 Sperling, M., Yan, X.-P., and Welz, B., Spectrochim. Acta, Part B, Sediment and Loam Soil had to be diluted in order to bring 1996, 51, 1891. the concentrations within the linear calibration range, although 16 Fang, Z.-L., Xu, S.-K., Dong, L.-P., and Li, W.-Q., T alanta, 1994, a determination could have been made without pre- 41, 2165. concentration. However, the purpose of the test was only to 17 Chen, H.-W., Xu, S.-K., and Fang, Z.-L., Anal. Chim. Acta, 1994, 298, 167. demonstrate the selectivity of the developed procedure for the determination of lead in relatively complicated matrices. The results are shown in Table 5. As can be seen, all the results Paper 6/07430H Received October 31, 1996 obtained by the proposed method are in good agreement with the certified concentrations. Accepted January 13, 1997 464 Journal of Analytical Atomic Spectrometry, April 1997, Vol. 12

ISSN:0267-9477    

DOI:10.1039/a607430h

出版商:RSC    

年代:1997

数据来源: RSC

摘要:

Comparison of Palladium Chemical Modifiers for the Determination of Selenium in Plasma by Zeeman-effect Background Corrected Electrothermal Atomic Absorption Spectrometry BENTE GAMMELGAARD* AND OLE JØNS T he Royal Danish School of Pharmacy, Department of Analytical and Pharmaceutical Chemistry, Universitetsparken 2, DK-2100 Copenhagen, Denmark The thermal stabilities of selenite, selenate, selenomethionine Furthermore, palladium alone has been proposed as a and trimethylselenonium in aqueous solution and plasma were chemical modifier, eventually in its reduced form, reduced compared.The stabilities of the four selenium forms were either by hydrogen, hydroxylamine hydrochloride or ascorbic different in aqueous solution and plasma. The nitric acid acid. Non-reduced palladium has been proposed for the deterconcentration had little influence on the sensitivity of the mination of selenium in urine and whole blood,11 in bovine species. The presence of chloride affected the stability of the liver9 and in serum and plasma using matrix matched stanselenium forms differently in aqueous solution, while there was dards.5 Reduced palladium has been proposed for the deterno pronounced effect on the stabilization in plasma.Different mination of selenium in whole blood14,16 and serum or amounts of palladium, varying from the application of 2 to plasma.14,17 40 mg into the graphite tube, were compared. The application Some confusion exists as to the exact amounts of palladium of 20 mg of palladium showed the best result in terms of and especially magnesium nitrate present in the graphite sensitivity and equal stabilization of the selenium species when furnace as some workers calculate the amounts in micrograms analysing plasma.Different amounts of magnesium nitrate, of the magnesium ion and others calculate the amount as varying from the application of 0.6 to 24.3 mg of magnesium magnesium nitrate. To aid comparison, the amounts of palinto the graphite tube, were examined.The addition of ladium and magnesium used in the above-mentioned papers magnesium nitrate did not improve the results, either in terms have therefore been calculated as the absolute amounts introof sensitivity or equal stabilization, but only increased the duced into the graphite furnace. These are shown in Table 1. background signal. On adding 20 mg of palladium, selenite and In biological samples, selenium occurs as different comselenate were equally stabilized, while the sensitivity of pounds and in different oxidation states.In human plasma, selenomethionine was about 80% of that of the former species. selenium is present as selenocysteine in selenoprotein P and It was not possible to stabilize trimethylselenonium to the glutathione peroxidase, while selenium in albumin is present same extent with this modifier. Peak shapes and appearance as selenomethionine.21 In human urine, selenium has been times of the atomization signals were equal for the four identified as selenite and the trimethylselenonium ion.22 selenium species with this modifier.The addition of 20 mg of Previous studies have shown that different selenium forms palladium was used for the analysis of the serum reference (SeVI, SeIV and Se-II) are not always stabilized to the same material Seronorm Trace Elements Serum, resulting in a mean extent.1–3 When analysing body fluids, erroneous results could of 85.4 mg l-1 (n=16) and a standard deviation of 6.6 mg l-1, therefore occur if the calibration standard was stabilized corresponding to a precision of 3.9% (recommended value differently from the selenium compound in the sample. 86 mg l-1). This modifier is proposed as the best choice for the In a previous study,15 it was shown that palladium in determination of selenium in plasma. combination with magnesium nitrate was a promising modifier Keywords: Zeeman-effect background corrected electrothermal for the equal stabilization of selenite, selenate, selenomethionatomic absorption spectrometry; chemical modification; ine and trimethylselenonium iodide in aqueous solutions.selenium determination; palladium; magnesium nitrate; plasma However, the largest amount of palladium as a modifier alone examined was 2.7 mg. With increasing amounts of palladium and simultaneous addition of magnesium nitrate the stabiliz- Electrothermal atomic absorption spectrometry using Zeeman- ation was improved, but it was not clear if the improvement effect background correction is one of the most widely used was due to the increased amount of palladium or the addition methods for the determination of selenium in biological mate- of magnesium nitrate.rials. However, selenium is not an easy element to determine Martin and Williams9 examined palladium alone, in combi- as it volatilizes at temperatures below 700 °C.1–3 A large nation with magnesium nitrate, and copper in combination number of different chemical modifiers have been proposed to with magnesium nitrate in aqueous solution and digested overcome this problem, the most common being nickel, copper bovine liver and concluded that a combination of copper and and palladium.These have either been used alone or in palladium was the best choice as peak shapes and appearance combination with magnesium nitrate.1–15 times for selenium atomization signals were the same in In recent years, palladium has attracted increasing interest inorganic and biological solutions for this modifier.as a chemical modifier. Palladium in combination with mag- Deaker and Maher4 used a combination of palladium and nesium nitrate has even been proposed as a universal modifier magnesium for the determination of selenium in digested for a large number of elements.10 This modifier has been marine biological tissues (reference materials). They showed proposed for the determination of selenium in digested environthat the same molar levels of palladium and magnesium are mental samples,6,12 natural waters,13 whole blood after a 25-fold dilution8 and plasma or serum.3,10 effective in stabilizing selenite, selenate, selenomethionine, Journal of Analytical Atomic Spectrometry, April 1997, Vol. 12 (465–470) 465Table 1 Absolute amounts of palladium and magnesium used for the determination of selenium Sample Modifier Mg/mg Mg/mmol Ref.Oyster tissue, serum 15 mg Pd+10 mg Mg(NO3)2 1.6 0.067 10 Digested environmental 15 mg Pd+10 mg Mg(NO3)2 1.6 0.067 12 Digested environmental 2 mg Pd+10 mg Mg(NO3)2 1.6 0.067 6 Natural waters 15 mg Pd+10 mg Mg(NO3)2 1.6 0.067 13 Whole blood 11 mg Pd+7.5 mg Mg(NO3)2.6H2O 0.7 0.029 8 Digested serum 15 mg Pd+10 mg Mg(NO3)2.6H2O 0.9 0.039 18 Digested marine tissue 10.6 mg Pd+9.7 mg Mg 9.7 0.40 4 Urine, whole blood 10–500 mg Pd 11 Digested bovine liver 7.5 mg Pd [+5 mg Mg(NO3]2) (0.8) (0.034) 9 Serum, plasma 4 mg Pd 5 Aqueous solution 1 mg Pd 7 Digested geological 2 mg Pd 19 Freeze-dried serum 1 mg Pd+albumin 20 Reduced palladium Plasma, blood 6.7 mg Pd+ascorbic acid 14 Blood 10 mg Pd+ascorbic acid 16 Serum 5 mg Pd+ascorbic acid 17 selenocysteine, selenocystamine and trimethylselenonium trations on the stabilization of different selenium species was examined.iodide in dilute nitric acid. Voth-Beach and Shrader14 examined the recovery of seleno- Selenate (SeVI) and selenite (SeIV) were chosen as inorganic model substances and selenomethionine and trimethylselenon- ethionine, selenocystine and selenomethionine using palladium reduced with ascorbic acid as a chemical modifier.They ium iodide (Se-II) as organic compounds. Although it is not very likely that selenate or trimethylselenonium would be obtained recoveries between 0.95 and 1.11 relative to the absorbance of selenite. They used the modifier for the determi- present in plasma, they were included to make comparison with previous studies possible.nation of selenium in a few plasma and blood samples. The peak shapes and appearance times for selenium in the samples were the same as when inorganic selenium standards were added. EXPERIMENTAL Laborda et al.7 compared non-reduced and reduced palladium, thermally reduced and reduced by hydroxylamine Apparatus and Operating Conditions hydrochloric acid or urea, as chemical modifiers for selenite A Perkin-Elmer (Norwalk, CT, USA) Zeeman 5000 atomic and trimethylselenonium with the purpose of determining absorption spectrometer equipped with an HGA-500 graphite selenium in urine.Non-reduced palladium stabilized the two furnace and an AS-40 autosampler was used for all measure- selenium forms equally, while the sensitivity for trimethyl- ments. Pyrolytic graphite coated tubes with inserted platforms selenonium was only about half the sensitivity for selenite. It (Perkin-Elmer) were used throughout.The tubes were con- was concluded that nickel or non-reduced palladium were ditioned prior to use. Argon was used as the purge gas. The suitable modifiers for the equal stabilization of the two species instrumental conditions and temperature programme are in urine. shown in Table 2. Doc¡ekalova� et al.1 investigated charring curves for selenite, selenate, selenomethionine and trimethylselenonium iodide in aqueous solution with palladium as a modifier in the presence of different ratios of magnesium nitrate.They found that Reagents and Samples atomization signals showed the same form and size for the All reagents were of analytical-reagent grade. Nitric acid (65% four selenium species if the molar ratio was 15300 for palladium m/v, Merck, Darmstadt, Germany), hydrochloric acid (30% and 1530 000 for magnesium nitrate. m/v, Merck) and sulfuric acid (95–97% m/v, Merck) were When trying to compare commercially available modifiers further purified by sub-boiling distillation in an all-quartz with ‘house modifiers’, it was observed that the stabilization apparatus (Hans Ku�rner, Rosenheim, Germany).Milli-Q water of different selenium forms was different when different pal- from a Milli-Q de-ionization unit (Millipore, Bedford, MA, ladium modifiers with the same palladium concentration were USA) was used throughout. used. This inconsistency seemed to be due to different chloride concentrations in the modifier.As the concentration of chloride in plasma and serum is about 0.1 mol l-1, standardization of the samples would cause problems if different selenium species Table 2 Graphite furnace programme and instrumental conditions* were present in the sample and these were not stabilized to for the determination of selenium the same extent. The stabilization of different selenium species in plasma has not previously been examined. It has been Step 1 2 3 4 5 6 examined in different digested environmental samples and reference materials, but it has previously been stated that the Temperature/°C 80 130 1100 2200 2600 20 need for a modifier is highly dependent on the type of matrix Ramp time/s 1 20 10 0 1 1 Hold time/s 15 15 20 5 5 10 analysed.6 Internal gas flow/ml min-1 300 300 300 0 300 300 The main aim of this work was to examine how different Recorder -5 selenium species behave in plasma in terms of stabilization Read -1 with different combinations of palladium and magnesium nitrate as chemical modifiers with the purpose of proposing a * Instrumental conditions: light source, EDL; lamp current, 10 mA; suitable modifier for the determination of selenium in plasma.power supply, 5W; wavelength, 196.0 nm; slit-width, 0.7 nm; sample volume, 10 ml; chemical modifier volume, 5 ml. Furthermore, the influence of chloride ions and acid concen- 466 Journal of Analytical Atomic Spectrometry, April 1997, Vol. 12Selenium standards Dependence on amount of magnesium The magnesium concentrations were varied from 0.005 mol Four stock solutions (100 mg l-1) of the selenium model compounds were prepared by dissolving the appropriate l-1 [corresponding to the addition of 0.025 mmol or 0.6 mg Mg or 3.7 mg Mg(NO3)2 into the furnace] to 0.2 mol l-1 [corre- amount of selenium compound in water.The selenite solution was diluted from a Titrisol selenium sponding to the addition of 1.0 mmol or 24.3 mgMg or 148.3 mg Mg(NO3 )2].The palladium concentration was kept constant dioxide solution containing 1.000 g l-1 Se (Merck). The selenate solution was prepared from sodium selenate (Na2SeO4) at 4.0 g l-1. (Fluka, Buchs, Switzerland) and the selenomethionine solution was prepared from selenomethionine (Sigma, St. Louis, MO, Variation of palladium content in modifiers with a constant USA). The trimethylselenonium solution was prepared from content of magnesium trimethylselenonium iodide, which was synthesized according to Foster and Ganther,23 as previously described.15 The palladium concentrations were varied from 1 to 8 g l-1 in A solution that was 1×10-3 mol l-1 with respect to nitric a series of modifiers where the magnesium concentration was acid and 0.1% with respect to Triton X-100 (Merck) was used kept constant at 0.10 mol l-1.for further dilution of the standards. RESULTS AND DISCUSSION Modifiers Prior to analysis, all reagents and modifiers were examined and found to be free of selenium contamination.The modifiers The palladium modifiers used were prepared from palladium black (Sigma) dissolved in nitric acid or from palladium nitrate were analysed for chloride by the addition of silver nitrate to the concentrated solution. No chloride was observed. [Pd(NO3)2·2H2O, Fluka] dissolved in Milli-Q water. In addition, commercially available modifiers containing When comparing the modifiers prepared from palladium metal powder or palladium nitrate with the commercially 10±0.2 g l-1 Pd in 15% nitric acid (Merck, Perkin-Elmer) were used.available modifiers, no difference in sensitivity or equal stabilization was observed. Welz et al.24 did not find any difference When preparing dilutions of the palladium modifier, the nitric acid concentration was kept constant to a final concen- when comparing the modifier from Merck with a modifier prepared from palladium metal powder. However, they found tration of 7.5% (1.2 mol l-1) nitric acid (except for final concentrations larger than 4 g l-1 Pd).a difference in stabilization efficiency when they compared the Merck modifier with another commercially available modifier Modifiers containing magnesium were prepared from a 1 mol l-1 solution of Mg(NO3)2·6H2O (Merck, suprapur) (Heraeus). This particular modifier was not tried in this study. The dissolution of the metal powder is laborious and demands dissolved in water. the addition of hydrochloric acid and heating; in our experience it is difficult to prepare adequately concentrated solutions from Plasma standards palladium powder in nitric acid alone.This has also been experienced by others.25 As there was no difference in the A ‘house’ pool of plasma was diluted 1+3 with the solution performance of the modifiers, the modifier from Merck was containing 1×10-3 mol l-1 nitric acid and 0.1% Triton X-100 used in further experiments. and spiked with selenium standards to contain 100 mg l-1 of The sensitivities of the individual graphite tubes can be very the different selenium species in the final solution.different. It has been shown that the same selenite solution The serum reference material, Seronorm Trace Elements analysed on six different tubes on the same day resulted in Serum, was from Nycomed Pharma (Oslo, Norway). integrated absorbances varying between 0.185 and 0.232.26 Before each experiment, selenium standards were analysed Procedure with a standard modifier.If the sensitivity of the graphite tube was different from the normally observed level, the difference In all experiments, 10 ml of 100mg l-1 solutions of the four in sensitivity was corrected for in order to harmonize the selenium species, corresponding to the application of 1 ng of figures and make them directly comparable. selenium into the furnace, were analysed by the addition of 5 ml of chemical modifier. All solutions were analysed in triplicate. Each series of experiments was performed on a new Dependence on Nitric Acid Concentration graphite tube.The effect of nitric acid was not very pronounced, either in The following experiments were performed: aqueous solution or in plasma. The nitric acid concentration of the undiluted palladium modifier is equal to 2.6 mol l-1. Hence, commercially available modifiers will always contain Ddence on nitric acid concentration nitric acid. However, the influence of nitric acid concentration Nitric acid concentrations were varied between 0.5 and 2.6 mol seems to be of minor importance.The acid concentration was, l-1 in palladium modifiers containing a fixed concentration of nevertheless, kept constant during further experiments. 2 g l-1 Pd. Similar results were obtained on adding increasing amounts of sulfuric acid. At 1 mol l-1 sulfuric acid concentration, the decrease in sensitivity was less than 10% for all four Dependence on amount of palladium selenium forms.Palladium concentrations of the modifiers were varied between 0.4 and 8 g l-1 Pd. Dependence on Amount of Palladium In aqueous solution, the sensitivity increased for all four Dependence on chloride concentration selenium forms when the amount of palladium added was increased from 2 to 30 mg. However, the increase in sensitivity Hydrochloric acid concentrations were varied between 0.1 and 2 mol l-1 in modifiers containing 4 g l-1 of Pd in 1.2 mol l-1 was relatively small.It did not exceed 10%, except for trimethylselenonium iodide, for which it increased by about 20% nitric acid. Journal of Analytical Atomic Spectrometry, April 1997, Vol. 12 467and was further increased on adding up to 50 mg of palladium. This is shown in Fig. 1. In plasma, the influence of the amount of palladium was more pronounced. The signals increased by 35% for selenite, 22% for selenate and 72% for selenomethionine when the amount of palladium was increased from 2 to 20 mg. Further increase in the amount of palladium did not increase the sensitivity for these species.When the amount of palladium was increased above 20 mg, atomization signals were broadened but integrated absorbances did not increase. This has also been observed by others.25 For trimethylselenonium iodide, however, the sensitivity was still increasing, the signal being four times larger on adding 50 mg of palladium, corresponding to undiluted modifier.This is shown in Fig. 2. It was not possible to find an amount of palladium that resulted in equal Fig. 3 Dependence of stabilization of selenium species on hydro- sensitivity for the four selenium species. Addition of 20 mg of chloricacid concentration in aqueous solution. Selenium concentration palladium was chosen as the optimum amount for the following of each species, 100 mg l-1. Amount of palladium added, 20 mg. $, experiments. Selenite; +, selenate; 0, selenomethionine; and 2, trimethylselenonium iodide.Dependence on Hydrochloric Acid Concentration of the selenium species in water and plasma may be related to The result of adding hydrochloric acid to the modifier was the chloride content of plasma. Furthermore, the presence of different in aqueous solution and plasma. In aqueous solution, chloride in a palladium modifier may lead to different results, the most pronounced effect was seen when the hydrochloric in terms of equal stabilization, when a comparison is made acid concentration was increased from 0 to 0.1 mol l-1.The with a modifier not containing hydrochloric acid. four selenium forms acted differently. While the sensitivity of When hydrochloric acid was added to plasma, only small selenomethionine was not affected, the sensitivities of selenite increases in sensitivity were observed on adding a concen- and selenate were decreased by about 10%, and the sensitivity tration of 0.1 mol l-1 chloride.This is shown in Fig. 4. As the of trimethylselenonium iodide decreased by almost 30%. This intrinsic chloride concentration of plasma is about 0.1 mol l-1, is shown in Fig. 3. This indicates that the different behaviour a pronounced effect from adding further chloride should not be expected. The same experiment was performed by adding chloride ions as sodium chloride. However, this caused distortion of the atomization signals, as chloride in this form is less volatile than when present as hydrochloric acid.Dependence on Amount of Magnesium In aqueous solution, the sensitivity seemed to decrease for all four selenium species on adding magnesium nitrate. This is shown in Fig. 5. There was no improvement in terms of equal stabilization of the selenium species by adding increasing small amounts of magnesium nitrate. However, the stabilities were almost equal for all four selenium forms on adding 24.3 mg of magnesium. At this concentration, however, a white layer built up inside the tube during analysis after only about ten injections.This has been observed in previous studies.15 Fig. 1 Dependence of stabilization of selenium species on amount of Also, in plasma, a decrease in sensitivity was seen on adding palladium in aqueous solution. Selenium concentration of each species, magnesium nitrate. There was no improvement when consider- 100 mg l-1. $, Selenite; +, selenate; 0, selenomethionine; and 2, ing the equal stabilization of the different selenium forms.This trimethylselenonium iodide. Fig. 4 Dependence of stabilization of selenium species on hydro- Fig. 2 Dependence of stabilization of selenium species on amount of palladium in plasma. Selenium concentration of each species, 100 mg chloric acid concentration in plasma. Selenium concentration of each species, 100 mg l-1. Amount of palladium added, 20 mg. $, Selenite; l-1. $, Selenite, +, selenate; 0, selenomethionine; and 2, trimethylselenonium iodide.+, selenate; 0, selenomethionine; and 2, trimethylselenonium iodide. 468 Journal of Analytical Atomic Spectrometry, April 1997, Vol. 12trimethylselenonium iodide in aqueous solution were equal if the ratio of selenium to palladium was 15300 and the ratio of selenium to magnesium nitrate was 1530 000. For an injection of 1 ng of selenium this would correspond to the addition of 0.3 mg of palladium and 30 mg of magnesium nitrate (or 4.8 mg of Mg). These modifier concentrations are fairly low compared with the concentrations used by others and the result is not in accordance with our results in aqueous solution, although equal stabilization was obtained in aqueous solution with the combination of 20 mg of palladium and 24.3 mg of magnesium.The intrinsic magnesium content of human plasma is about 0.7–0.9 mmol l-1 (corresponding to an addition of 0.2 mg Mg when applying a plasma sample of 10 ml). The absence of an effect of the addition of magnesium nitrate cannot, therefore, be ascribed to the intrinsic presence of magnesium in plasma.Fig. 5 Dependence of stabilization of selenium species on amount of Adding the modifier prior to the sample instead of after the magnesium in aqueous solution. Selenium concentration of each sample did not have any influence on the results. species, 100 mg l-1. Amount of palladium added, 20 mg. $, Selenite; Atomization signals with and without background correc- +, selenate; 0, selenomethionine; and 2, trimethylselenonium iodide.tion were recorded for the selenium compounds in water and plasma using a modifier containing 4 g l-1 of palladium alone is shown in Fig. 6. This shows that even though an equal and with modifiers containing 4 g l-1 of palladium and increas- stabilization is apparently obtainable by adding large amounts ing amounts of magnesium nitrate. The shape and appearance of magnesium nitrate in aqueous solution, this is not possible times for the atomization signals were similar for all four in plasma.selenium forms in plasma and aqueous solution. Martin and Williams9 observed that the peak appearance time and shape was different for selenite and the selenium form Dependence on Amount of Palladium When the Amount of present in digested bovine liver when using palladium and Magnesium Was Constant palladium in combination with magnesium nitrate as chemical Ideally, concentrations of palladium and magnesium should modifiers. be varied simultaneously to find the optimum concentrations The effect of adding increasing amounts of magnesium of these reagents.Such an experiment, however, is difficult to nitrate to plasma containing selenite standard is shown in perform as it demands a large number of injections on the Fig. 7. It appears that the atomization signal was not increased, same graphite tube. The sensitivity of the individual graphite and that only the background absorbance increased.This is tubes can be different and decreases after a certain number of in accordance with the results of Xiao-Quan and Bei,6 who injections. Also, the lifetimes of the tubes are very different. In compared the performances of palladium, palladium–ascorbic practice, such an experiment is, therefore, not possible to acid and palladium–magnesium modifiers in the analysis of perform as it would involve too many corrections for sensi- digested environmental samples and concluded that the need tivity.Instead, one magnesium concentration was chosen to for the addition of ascorbic acid was matrix-dependent, while examine whether variation of the amount of palladium in the addition of magnesium nitrate was not recommended as plasma would improve the performance. When the magnesium the background signal for the palladium–magnesium modifier concentration was kept constant at 12.2 mg of magnesium, no was much larger than for the palladium modifier alone.improvement in equal stabilization was observed for any of According to our results, the behaviour of the four selenium the palladium concentrations. species is different in plasma and aqueous solution. The The most commonly used combination of 15 mg of palladium chemical modifier should, therefore, be optimized for each and 10 mg of magnesium nitrate (corresponding to 1.6 mg matrix. This is supported by the results of Deaker and Maher,4 Mg)10,12,13 was examined separately for the determination of who observed that the optimum level of modifier was depen- the four selenium species in plasma, with no improvement as dent on the sample type; in digested marine tissue, more regards the equal stabilization of the selenium forms.modifier was required to obtain a maximum signal than in Doc¡ekalova� et al.1 stated that the shape and size of the aqueous solution. atomization signals of selenite, selenate, selenomethionine and Fig. 6 Dependence of stabilization of selenium species on amount of Fig. 7 Dependence of atomization signal on amount of magnesium magnesium in plasma. Selenium concentration of each species, 100 mg l-1. Amount of palladium added, 20 mg. $, Selenite, +, selenate; 0, for 100 mg l-1 of selenite in plasma. Amount of palladium added, 20 mg. $, Corrected signal; and 0, signal+background. selenomethionine; and 2, trimethylselenonium iodide. Journal of Analytical Atomic Spectrometry, April 1997, Vol. 12 469The sensitivity of selenomethionine was less than that of selenite in plasma. However, this is considered to be acceptable taking the general uncertainty on the analysis of biological materials at this concentration level into account. The stabilization of trimethylselenonium iodide was not successful but this is of minor importance as this metabolite is hardly present in plasma and the compound was only examined for comparison purposes. A general observation was that the sensitivity for selenate was poorer than for selenite in aqueous solution, whereas the sensitivities of these two species were equal in plasma.This indicates that selenate is reduced to selenite in plasma. In conclusion, the addition of magnesium nitrate did not improve the determination of selenium in terms of equal stabilization of the different selenium compounds or improve Fig. 8 Dependence of stabilization of selenium species on pre- the sensitivity. According to our experiments, the addition of treatment temperature in aqueous solution.Selenium concentration of 20 mg of palladium alone is just as effective as the combination each species, 100 mg l.1. Amount of palladium added, 20 mg. $, of palladium and magnesium nitrate. Selenite; +, selenate; 0, selenomethionine; and 2, trimethylselenonium iodide. REFERENCES 1 Doc¢§ekalova¢¥, H., Doc¢§ekal, B., Komarek, J., and Novotny¢¥, I., J. Anal. At. Spectrom., 1991, 6, 661. 2 Welz, B., Schlemmer, G., and Voellkopf, U., Spectrochim. Acta, Part B, 1984, 39, 501. 3 Ne` ve, J., Chamart, S., and Molle, L., T race Elem. Anal. Chem. Med. Biol., 1987, 4, 349. 4 Deaker, M., and Maher, W., J. Anal. At. Spectrom., 1995, 10, 423. 5 Gardiner, P. H. E., Littlejohn, D., Halls, D. J., and Fell, G. S., J. T race Elem. Med. Biol., 1995, 9, 74. 6 Shan, X.-Q., and Wen, B., J. Anal. At. Spectrom., 1995, 10, 791. 7 Laborda, F., Vin.uales, J. M., Mir, J. M., and Castillo, J.R., J. Anal. At. Spectrom., 1993, 8, 737. 8 Van Cauwenberg, R., Robberecht, H., Van Dael, P., and Deelstra, H., J. T race Elem. Electrolytes Health Dis., 1990, 4, 127. 9 Martin, C. K., and Williams, J. C., J. Anal. At. Spectrom., 1989, 4, 691. 10 Schlemmer, G., and Welz, B., Spectrochim Acta, Part B, 1986, Fig. 9 Dependence of stabilization of selenium species on pre- 41, 1157. treatment temperature in plasma. Selenium concentration of each 11 Radziuk, B., and Thomassen, Y., J.Anal. At. Spectrom., 1992, species, 100 mg l.1. Amount of palladium added, 20 mg. $, Selenite; 7, 397. +, selenate; 0, selenomethionine; and 2, trimethylselenonium iodide. 12 Welz, B., Schlemmer, G., and Mudakavi, J. R., J. Anal. At. Spectrom., 1988, 3, 93. Thermal pre-treatment curves for the four selenium com- 13 Welz, B., Schlemmer, G., and Mudakavi, J. R., J. Anal. At. pounds in water and plasma were obtained from 600 to 1500 ¡ÆC Spectrom., 1988, 3, 695. 14 Voth-Beach, L.M., and Shrader, D. E., Spectroscopy, 1986, 1, 49. with the addition of 20 mg of palladium. These are shown in 15 Johannessen, J. K., Gammelgaard, B., J©ªns, O., and Hansen, Figs. 8 and 9, respectively, from which it can be seen that the S. H., J. Anal. At. Spectrom., 1993, 8, 999. optimum pre-treatment temperature is 1100 ¡ÆC in aqueous 16 Knowles, M. B., and Brodie, K. G., J. Anal. At. Spectrom., 1988, solution as well as in plasma. The peak shapes of the atomiz- 3, 511.ation signals were highest and narrowest at 2200 ¡ÆC, which is 17 Knowles, M. B., and Brodie, K. G., J. Anal. At. Spectrom., 1989, the atomization temperature used in this work. Beyond this 4, 305. 18 Van Dael, P., Van Cauwenbergh, R., Robberecht, H., Deelstra, H., temperature, the signals were broadened. Again, the different and Calomme, M., At. Spectrosc., 1995, 16, 251. sensitivity of the individual species in the two matrices is 19 Xiao-Quan, S., and Kai-Jing, H., T alanta, 1985, 32, 23. observed. 20 Teague-Nishimura, J. E., Tominaga, T., Katsura, T., and Calibration graphs for all four selenium forms were linear Matsumoto, K., Anal Chem., 1987, 59, 1647. from 50 to 250 mg l.1 in aqueous solution as well as in plasma 21 Deagen, J. T., Beilstein, M. A., and Whanger, P. D., J. Inorg. (concentrations in final solutions). Correlation coefficients were Biochem., 1991, 41, 261. 22 Hasunuma, R., Ogawa, T., and Kawanishi, Y., Bull. Environ. better than 0.999 in all instances. Thus, the difference in Contam. T oxicol., 1993, 50, 19. sensitivity is not due to the linear range being exceeded for 23 Foster, S. J., and Ganther, H. E., Anal. Biochem., 1984, 137, 205. any of the compounds. The detection limits were calculated as 24 Welz, B., Schlemmer, G., and Mudakavi, J.R., J. Anal. At. three times the standard deviation after 30 measurements on Spectrom., 1992, 7, 1257. diluted plasma by use of the modifier containing 4 g l.1 of 25 Beach, L. M., Spectroscopy, 1987, 2, 21. palladium and were found to be 4.2 mg l.1 for selenite, 4.1 mg 26 Johannessen, J. K., PhD Thesis, The Royal Danish School of Pharmacy, Copenhagen, Denmark, 1996. l.1 for selenate and 4.7 mg l.1 for selenomethionine. The accuracy of the method when adding 20 mg of palladium Paper 6/07167H as a chemical modifier was examined by analysing the serum Received October 21, 1996 reference material Seronorm Trace Elements Serum by means Accepted January 20, 1997 of three standard additions of selenite. The mean value was 85.4 mg l.1 with a standard deviation of 3.3 mg l.1 (n=16), corresponding to a precision of 3.9%. This is in good agreement with the recommended value of 86 mg l.1. 470 Journal of Analytical Atomic Spectrometry, April 1997, Vol.

ISSN:0267-9477    

DOI:10.1039/a607167h

出版商:RSC    

年代:1997

数据来源: RSC

摘要:

Combined Nickel and Phosphate Modifier for Lead Determination in Water by Electrothermal Atomic Absorption Spectrometry YUPING XU* AND YANZHONG LIANG Department of Geological Sciences, University of Illinois at Chicago, 845 West T aylor Street, Chicago, IL 60607, USA. Email: yuping@uic.edu The effectiveness of nickel nitrate plus ammonium of a suitable modifier.3,13–20 These criteria include suppression of the interference associated with sample matrices, increase of dihydrogenphosphate as a combined modifier for the determination of lead in a variety of water samples by ETAAS pyrolysis temperature without Pb loss and enhancement of Pb signals.Therefore, further work in this area is deemed necessary. was evaluated. Optimization of the temperature program, modifier mass and pyrolysis hold time for the determination of In this work, we investigated the effectiveness of a combined nickel nitrate and ammonium dihydrogenphosphate modifier for lead was carried out.The results indicate that the combined modifier allows the quantitative stabilization of Pb in water the determination of Pb in natural waters. The objective was to develop a valid and feasible analytical scheme for the direct samples up to 1200 °C during the pyrolysis step. In comparison, the maximum pyrolysis temperature without the measurement of Pb in drinking and surface waters with complex matrices, some of which interfere with Pb determination. modifier is 900 °C or lower.The modifier further reduces the background absorbance caused by sample matrices and significantly enhances the sensitivity of Pb determination. The EXPERIMENTAL observed detection limit is 0.14 mg l-1 with a sample volume of Instrumentation and Reagents 10 ml. The characteristic mass or sensitivity of the proposed method is 7 pg. The tolerable amounts of various interferents A Perkin-Elmer (Norwalk, CT, USA) Model 503 atomic absorption spectrometer equipped with an HGA-2100 graphite such as chloride, sulfate and carbonate in the presence of the modifier are high enough for the determination of lead in a furnace controller and a deuterium arc background corrector was used for all atomic absorption measurements.Collection variety of waters. The recoveries of spiked Pb in tap water and waters from variably contaminated waters from a ditch, a lake of absorbance signals and peak integration were achieved with PeakSimple II software via a data acquisition board (SRI and a river in the Chicago area were 89–101%.The proposed method has several advantages over commonly used methods Instruments, Torrance, CA, USA). The wavelength, spectral bandwidth and lamp current used for Pb determination were with H3PO4 or Mg(NO3)2+NH4H2PO4 as modifiers. set according to the recommendations of the instrument manu- Keywords: L ead; chemical modifier; atomic absorption facturer. Pyrolytic graphite-coated graphite tubes (Perkin- spectrometry ; water; interferences Elmer, Part No.B0135653) with a L’vov graphite platform Lead is toxic to humans and animals, especially to young (Perkin-Elmer, Part No. B0121091) were utilized throughchildren. As a result of worldwide accumulation, Pb presents a out. When chemical modifiers were used, they were injected serious environmental and health hazard.1 In natural waters, separately from sample solutions. Pb concentrations often range from 1 to 30 mg l-1.2 Hence Aliquots used for the samples and modifiers were 10 ml in all highly sensitive analytical techniques are necessary to determine cases except when indicated otherwise.The internal gas flow was the Pb concentrations in these waters. Because of its sensitivity, interrupted during the atomization stage. The graphite furnace versatility, speed and specificity, ETAAS has been extensively temperature program for Pb determination is given in Table 1. utilized in the direct analysis of water for numerous elements.3 Chemicals with the highest available purity but at least of In recent decades, ETAAS has been the US Environmental analytical-reagent grade were used to prepare solutions in deion- Protection Agency (USEPA) method of choice for elemental ized water. Standard solutions were made from a commercial analysis including Pb in water samples.4,5 However, problems stock standard solution of 1000 mg l-1 prepared from Pb(NO3)2 arise in analyzing samples composed of complex matrices.(VWR Scientific, Chicago, IL, USA).Working standard solutions Chemical interference encountered in a pulse-heated electrother- were obtained by dilution to volume with deionized water. mal atomizer frequently causes depression of absorbance signals Chemical modifier solutions of Ni, La and Mg were prepared due to the co-volatilization of the analyte with the matrices.6 from their nitrate salts. Palladium modifier solutions were Currently available ways of reducing such interferences include prepared by diluting a 5% Pd(NO3)2 stock solution with platform atomization,7 probe atomization,8 Zeeman effect back- deionized water. Modifier solutions of ammonium dihydrogenground correction9 and chemical treatment of the sample in the phosphate (NH4H2PO4) were prepared by dissolving the salt graphite furnace.10,11 Chemical modification is preferred as a simple approach to alleviating interferences encountered especi- Table 1 ETAAS temperature program for Pb determination in water samples using the Ni(NO3)2+NH4H2PO4 modifier ally in the direct determination of volatile elements in samples with significant amounts of matrices.The modifiers can act as Step 1 2 3 4 pyrolysis aids and delay the vaporization of the analyte before Temperature/°C 130 1000 2400 2650 the graphite tube is nearly isothermal,12 the ideal condition for Ramp time/s* 2 2 1 1 atomization. To date, a variety of modifiers have been suggested Hold time/s 40 50 5 5 for Pb, including palladium,13 palladium plus magnesium,3 Ar gas flow/ml min-1 300 300 0† 300 ammonium phosphate,14 ascorbic acid,15 lanthanum16 and others.17–20 Despite considerable efforts having been made to * Estimated.search for valid Pb modifiers in the past, only limited success † A continuous flow of 300 ml min-1 was used for the USEPA Method 7421. has been reported for modifiers that meet some or all the criteria Journal of Analytical Atomic Spectrometry, April 1997, Vol. 12 (471–474) 471in deionized water. The above chemicals were obtained from Aldrich (Madison, WI, USA) and had a purity of 99.999%. All the reagents and deionized water were tested for Pb prior to the experiments and no detectable amounts of Pb were found. Water Samples and Their Characterization Three surface water samples were taken from the Lake Calumet area, 15 miles south of downtown Chicago, IL, USA. The surface and ground waters in the area have been substantially contaminated with slag wastes from local steel companies.In addition to the slag wastes, other solid waste materials such as household trash, demolition debris and fly ash were used as fill materials to convert the extensive coastal marshes into Fig. 1 Effects of chemical modifier on the pyrolysis and atomization usable industrial and residential properties. The wastes were temperatures for the analysis of 5 mg l-1 Pb standard solution. also disposed of in several landfills in the area.Interstate Combined modifier, 100 mg ml-1 Ni(NO3 )2+10 mg ml-1NH4H2PO4 . Highway 94 lies immediately to the west of the area, where it Curves A and C show the relationship between integrated Pb peak absorbance and pyrolysis temperature with a fixed atomization tem- receives road deicing agents and automobile exhaust fumes perature of 2400 °C, (A) with and (C) without the modifier. Curves B from the highway. Lake Calumet connects Lake Michigan and D depict the relationship between Pb signal and atomization through the Calumet River. temperature with a fixed pyrolysis temperature of 600 °C, (B) with One sample was taken directly from Lake Calumet, a second and (D) without the modifier. from the Calumet River and the third from a ditch near the highway.The ditch water appeared to have come from surface runoff, although mixing with the groundwater is possible before of the combined modifier, no significant differences for the Pb the water was discharged to the ditch.These samples were signal were observed with atomization temperatures between filtered through 0.2 mm syringe filters (Gelman, Ann Arbor, MI, 1800 and 2400 °C (curve B in Fig. 1). However, the integrated USA) using a syringe filter assembly (VWR Scientific, USA). absorbance of Pb without the modifier increased with increase They were kept at 4 °C for a brief period of time before analysis. in atomization temperature (curve D).Considering the dif- The tap water used in this study was taken from the campus of fusion effect at high atomization temperatures,21 an atomizthe University of Illinois at Chicago. The source of the water is ation temperature of 2400 °C and a pyrolysis temperature of Lake Michigan. The water sample was taken after running the 1000 °C appear to be the optimum for Pb determination. tap for a few minutes and analyzed without further treatment. These temperatures were adopted in subsequent experiments, All the samples were analyzed for their pH values with an including Pb determinations in water samples and recoveries Orion (Cambridge, MA, USA) pH meter and their concen- of spiked Pb from the samples.trations of F-, Cl-, SO42-, PO43- and NO3- with a DX100 Ion Chromatograph (Dionex, Sunnyvale, CA, USA). No phos- Pyrolysis time phate was detectable in the samples. The hold time for pyrolysis in routine analysis is usually set at about 30–40 s.However, complex matrices in some water RESULTS AND DISCUSSION samples produce high background absorption that can be Selection of Chemical Modifiers beyond the correction capacity of a Zeeman effect corrector.17,22 Therefore, it often requires a longer hold time for the determi- Several preliminary experiments were conducted to select a nation of trace volatile elements (e.g., Pb) in these water samples. valid chemical modifier for the measurement of Pb in the water The approach of increasing the pyrolysis time has been found samples.The modifiers tested include commonly used Pd, to be successful in many cases and has overcome the problem Pd+Mg, phosphate, La and Ni. In view of their stabilization of background absorption.5,23 However, the validity of using and the enhancement effects on the Pb signal in standard this approach depends on the effectiveness of chemical modifiers solutions and the water samples, Ni(NO3)2 and NH4H2PO4 in stabilizing the analyte during the pyrolysis.The relationship were selected as a combined modifier for further investigation. between Pb recovery and pyrolysis hold time is depicted in An initial screening of various concentrations of Ni(NO3)2 and Fig. 2 for 2 ng Pb-spiked ditch water (the sample with the most NH4H2PO4 was done to provide an adequate combination for complex matrix) in the presence of Ni(NO3)2 plus NH4H2PO4 the optimization of the temperature program discussed below. as the modifier. The recovery was in the range 85–94% when the holdtime increasedfrom 30 to 110 s.The maximumrecovery Optimization of ETAAS Conditions of 94% occurred at a hold time of 70 s.Therefore, a pyrolysis T emperature program time of 70 s was adopted in subsequent work. The temperatures of pyrolysis and atomization were first optimized using a Pb standard solution of 5 mg l-1 in both the presence and absence of modifiers. The optimization results are shown in Fig. 1. For the optimization of the pyrolysis temperature, an atomization temperature of 2400 °C was used.When 100 mg ml-1 Ni(NO3)2 with 10 mg ml-1 NH4H2PO4 was added to the standard solution, a pyrolysis temperature up to 1200 °C could be used without Pb loss (curve A in Fig. 1). In comparison, pyrolysis temperatures over 900 °C in the absence of the modifier (curve C) resulted in significant Pb loss. In addition, the sensitivity of Pb determination with the addition of Ni(NO3)2 and NH4H2PO4 was enhanced about twofold or more compared with Pb determination without the modifier.In order to optimize the atomization temperature, a Fig. 2 Influence of pyrolysis time on the recovery of 2 ng lead-spiked ditch water. The maximum recovery of 94% occurs at 70 s. fixed pyrolysis temperature of 600 °C was used. In the presence 472 Journal of Analytical Atomic Spectrometry, April 1997, Vol. 12Table 2 Test of interferences in the determination of Pb in 5 mg l-1 Concentration and ratio of modifiers Pb standard solutions spiked with selected interferent ions.The The modifier concentration has been known to influence recovery is the calculated percentage based on the integrated absorbance. The recovery of the interference-free standard solution significantly the sensitivity of metal determinations.22–24 In was 100%. Standard deviations were calculated from four replicates addition to increasing the possibility of contamination of the graphite tube, a high modifier concentration generally Interferent concentration/ Recovery depresses the absorbance signal owing to secondary adsorption Interferent Compound used mg ml-1 (%) at the cooler ends of the graphite tube.25,26 On the other hand, Cl- NaCl 0.91 93±7 a higher modifier concentration stabilizes the analyte to higher SO42- Na2SO4 1.01 89±5 pyrolysis temperatures.27 Therefore, careful optimization of the ClO4- Fe(ClO4)3 0.84 94±5 modifier concentration is essential in overcoming the problems CO32- Na2CO3 0.68 91±8 while achieving higher sensitivity and thermal stability.In this work, a standard Pb solution of 5 mg l-1 was used to examine the influence of increasing concentrations of Ni(NO3 )2 and have the ability to reduce the impact of sulfate anions in NH4H2PO4 on the Pb absorbance signals. The results are natural waters. The reaction mechanism between the modifier shown in Fig. 3. It should be noted that the optimization of and sulfate is unknown.In addition to sulfate, the modifier Ni(NO3)2 and NH4H2PO4 concentrations was accomplished can also effectively tolerate up to 0.91 mg ml-1 Cl-, separately. During the optimization of the Ni(NO3 )2 concen- 0.84 mg ml-1 ClO4- and 0.68 mg ml-1 CO32-. The high tration the NH4H2PO4 concentration was fixed at 10 mg ml-1 recovery of 93±7% for Cl- in Table 2 indicates that the and during the optimization of NH4H2PO4 concentration the modifier would be effective for Pb determination in chloride- Ni(NO3)2 concentration was fixed at 100 mg ml-1.For the rich water samples. purpose of comparison, the Pb absorbance without any modi- fier is also shown in Fig. 3. These results show that the Limit of Detection and Characteristic Mass maximum signal enhancement is achieved at an Ni(NO3)2 concentration of 100 mg ml-1 and an NH4H2PO4 concen- The detection limit for the proposed method is 0.14 mg l-1, tration of 10 mg ml-1. The combined modifier provides much calculated as three times the standard deviation of a 10 ml higher signal enhancement than an Ni(NO3)2 or NH4H2PO4 blank solution for 10 determinations.The characteristic mass modifier alone. The effect of Ni(NO3)2 concentrations greater or sensitivity of Pb detection, defined as the mass of Pb which than 25 mg ml-1 on the integrated Pb absorbance is not very yields an integrated absorbance signal of 0.0044, is 7 pg. significant. Similarly, NH4H2PO4 concentrations in the range 8–20 mg ml-1 have almost an equivalent effect on the signal Sample Analyses and Method Comparison enhancement. In the light of this finding, 100 mg ml-1 Using the proposed method, we determined the Pb concen- Ni(NO3)2 and 10 mg ml-1 NH4H2PO4 were chosen as the trations in and recoveries from tap water and three surface combined modifier for the interference tests below.water samples from the Lake Calumet area in Chicago. Table 3 gives the results for Pb in various water samples and the Non-spectral Interferences concentrations of some major anions in the samples.The The frequently encountered non-spectral interferent ions and samples were found to have Pb concentrations in the range compounds on the signal of the analyte element were investi- 0.78–2.48 mg l-1. The recoveries of 2 ng of Pb added to the gated in this work. Various interferent ions were added to a samples were 89–101%. 5 mg l-1 Pb standard solution, and the Pb recovery was The results obtained from this study were compared with measured for each interferent.The results are given in Table 2, those from two commonly used ETAAS methods for Pb where the RSD was calculated from four replicates. Appar- determination (Table 3). The first is the standard method ently, the combined modifier of 100 mg ml-1 Ni(NO3)2 and recommended by the USEPA.5 It involves the addition of 10 ml 10 mg ml-1 NH4H2PO4 can tolerate sulfate up to 1.01 mg ml-1 in a 1 ml sample of H3PO4 as the modifier and a 20 ml sample with only a 5–10% depression of the signal generated by injection with continuously flowing purge gas in the absence 5 mg l-1 Pb.Sulfate is one of the most troublesome interferent of a platform. The second method uses 2 mgMg(NO3)2+10 mg ions for Pb determination.17,28 The above modifier appears to NH4H2PO4 as a combined modifier.29 The maximum ashing temperature was found to be about 1000 °C under the instrumental conditions used in this study.When used to analyze 5 mg l-1 Pb standard solution at an ashing temperature of 900 °C and an atomization temperature of 2000 °C, the latter method has a detection limit of 0.39 mg l-1 and a characteristic mass of 15 pg. These values were significantly higher than those obtained in this study. The analytical performances of these two methods with the water samples used in this study are shown in Table 3. The Pb concentrations and recoveries measured for the four water samples with the USEPA method were 1.01–2.57 mg l-1 and 79–90%, respectively.For the ditch water with a complex matrix, the recovery with the USEPA method is significantly lower than that with the proposed method. The method using the Mg(NO3)2+NH4H2PO4 modi- fier also gives significantly lower recoveries of 2 ng of Pb added to the water samples. In particular, the method cannot overcome the interference from the complex matrix of the ditch Fig. 3 Signal enhancement in response to the amount of modifier water. As a result, the peaks obtained during the analyses were used for the determination of Pb in 5 mg l-1 standard solution.A fixed not sufficiently well resolved to give meaningful results. This concentration of 10 mg ml-1 NH4H2PO4 was used in Ni(NO3)2 comparison demonstrates the effectiveness and applicability of optimization (left columns) and a fixed concentration of 100 mg ml-1 the proposed modifier for the direct determination of Pb in Ni(NO3)2 was used in NH4H2PO4 optimization (middle columns).The far right column is the Pb signal without any modifier. natural waters with complex matrices. Journal of Analytical Atomic Spectrometry, April 1997, Vol. 12 473CONCLUSION It is reasonable to suggest that Ni(NO3)2 plus NH4H2PO4 can act as an effective chemical modifier for the determination of Pb in several types of waters by ETAAS. Based on the experimental results, the pyrolysis temperature can be set to as high as 1200 °C, which is compatible with or higher than those used with the most common chemical modifiers such as Pd and Pd+Mg.The combined Ni(NO3)2+NH4H2PO4 modifier also enhances the Pb signal and reduces matrix interferences. The recovery data for different water samples appear to verify the effectiveness and applicability of the proposed method for the direct measurement of Pb in water samples with complex matrices. This work was supported by the Campus Research Board of the University of Illinois at Chicago and by the Searle Environmental Health and Safety Fellowship (Illinois, USA).REFERENCES 1 Jaworski, J. F., in L ead, Mercury, Cadmium and Arsenic in the Environment, ed. Hutchinson, T. C., and Meema, K. M., Wiley, New York, 1987, pp. 3–16. 2 Concon, J. M., Food T oxicology : Contaminants and Additives, Marcel Dekker, New York, 1988. 3 Welz, B., Schlemmer, G., and Mudakavi, J. R., J. Anal. At. Spectrom., 1988, 3, 695. 4 Association of Official Analytical Chemists, Official Methods of Analysis of the Association of Official Analytical Chemists, AOAC, Arlington, VA, 15th edn., 1990. 5 USEPA T est Methods for Evaluating Solid Waste: L aboratory Manual—Physical/Chemical Methods, SW-846, USEPA, Washington, DC, 3rd edn., 1986. 6 Ni, Z. M., and Shan, X. Q., Spectrochim. Acta, Part B, 1987, 42, 937. 7 L’vov, B. V., Peliva, L. A., and Sharnopolskii, A. I., Zh. Prikl. Spektrosk., 1977, 27, 395. 8 Gir, S.K., Littlejohn, D., and Ottaway, J. M., Analyst, 1982, 107, 1095. 9 Slavin, W., Carnrick, G. R., Manning, D. C., and Pruszkowska, E., At. Spectrosc., 1983, 4, 69. 10 Ediger, R. D., At. Absorpt. Newsl., 1975, 14, 127. 11 Shan, X. Q., and Ni, Z. M., Acta Chim. Sin., 1981, 39, 575. 12 Hinds, M. W., Katyal, M., and Jackson, K. W., J. Anal. At. Spectrom., 1988, 3, 83. 13 Shan, X. Q., and Ni, Z. M., Can. J. Spectrosc., 1981, 26, 219. 14 Zong, Y. Y., Parsons, P. J., and Slavin, W., Spectrochim.Acta, Part B, 1994, 49, 1667. 15 Ebdon, L., and Lechotycki, A., Microchem. J., 1986, 34, 340. 16 Fletcher, I. J, Anal. Chim. Acta, 1983, 154, 235. 17 He, B., and Ni, Z. M., J. Anal. At. Spectrom., 1996, 11, 165. 18 Bu-Olayan, A. H., Al-Yakoob, S. N., and Alhazeem, S., Sci. T otal Environ., 1996, 181, 209. 19 Cabrera, C., Lopez, M. C., Gallego, C., Lorenzo, M. L., and Lillo, E., Sci. T otal Environ., 1995, 159, 17. 20 Donat, J. R., Lao, K. A., and Bruland, K. W., Anal.Chim. Acta, 1994, 284, 547. 21 Perez-corona, M. T., Beatriz De La Calle�-Guntin�as, M., Madrid, Y., and Camara, C., J. Anal. At. Spectrom., 1995, 10, 321. 22 Liang, Y. Z., Li, M., and Rao, Z., Fresenius’ J. Anal. Chem., 1997, 357, 112. 23 Ni, Z. M., He, B., and Han, H. B., Spectrochim. Acta, Part B, 1994, 49, 245. 24 Liang, Y. Z., Li, M., and Rao, Z., Anal. Sci., 1996, 12, 629. 25 Thomaidis, N. S., Piperaki, E. A., and Efstathiou, C. E., J. Anal. At. Spectrom., 1995, 10, 221. 26 Frech, W., Li, K., Berglund, M., and Baxter, D. C., J. Anal. At. Spectrom., 1992, 7, 141. 27 Mandjukov, P. B., Vassileva, E. T., and Simeonov, V. D., Anal. Chem., 1992, 64, 2596. 28 Welz, B., Schlemmer, G., and Mudakavi, J. R., J. Anal. At. Spectrom., 1992, 7, 1257. 29 Manning, D. C., and Slavin, W., Appl. Spectrosc., 1983, 37, 1. Paper 6/07046I Received October 16, 1996 Table 3 Lead concentrations and recoveries of 2 ng Pb added to natural waters from various sources. Standard deviations were calculated from four replicates. Anion concentrations were obtained from ion chromatographic analyses. No phosphate was detected Pb found*/mg l-1 Recovery* (%) Sample A B C A B C pH SO42-/mg l-1 Cl-/mg l-1 F-/mg l-1 NO3-/mg l-1 Lake Calumet 1.37±0.09 1.39±0.13 1.59±0.21 89±4 84±6 70±9 7.74 56.8 50.4 0.59 6.3 Calumet River 0.78±0.12 1.01±0.17 ND† 101±8 89±5 78±8 7.58 44.8 38.8 0.25 6.4 Ditch water 1.41±0.12 1.29±0.19 —‡ 93±8 79±9 —‡ 7.79 564 216 17.2 3.9 Tap water 2.48±0.04 2.57±0.22 3.82±0.37 95±2 90±6 85±7 7.78 28.1 12.5 1.27 1.5 * A, This method; B, EPA Method 7421; C, 2 mg Mg(NO3)2+10 mg NH4H2PO4.29 † ND=not detected. ‡ Pb peaks too complex to resolve. Accepted December 17, 1996 474 Journal of Analytical Atomic Spectrometry, April 1997, Vol

ISSN:0267-9477    

DOI:10.1039/a607046i

出版商:RSC    

年代:1997

数据来源: RSC

摘要:

Determination of Ytterbium in Digesta and Animal Faeces by Electrothermal Atomic Absorption Spectrometry E� DER C. LIMAa, FRANCISCO J. KRUG*b, JOAQUIM A. NO� BREGAa AND ELISABETE A. N. FERNANDESb aDepartamento de Quý�mica, Universidade Federal de Sa�o Carlos, Sa�o Carlos-SP, Brazil bCentro de Energia Nuclear na Agricultura, Universidade de Sa�o Paulo, P. O. Box 96, 13400–970, Piracicaba-SP, Brazil A procedure for ytterbium determination in digesta and faeces surfaces have been reported for the determination of ytterbium, 8,9 but it was pointed out that ytterbium easily forms of animals by ETAAS with a transversally heated graphite atomizer and Zeeman-effect background correction is refractory carbides.Low sensitivity was observed for uncoated graphite tubes and high atomization temperatures for relatively described. Up to 2500 mg l-1 of Na, K, P or Si, 1000 mg l-1 of Ca or Mg and 500 mg l-1 of Fe, Cu, Al, Zn or Mn did not long periods of time were required for minimizing memory effects.On the other hand, it was also demonstrated that interfere significantly with the atomization of 10.0 mg l-1 Yb. The characteristic mass was 3.0 pg and a detection limit of ytterbium can be atomized from pyrolytic graphite, offering a highly sensitive method.10–12 More recently, a method for 0.15 mg l-1 for a 20 ml sample volume was calculated. Digesta and faeces were dry-ashed in a muffle furnace and the resulting ytterbium determination in a mixture of rare earth oxides and crude yttria by ETAAS was published,13 in which europium ash was treated with hydrochloric acid.Application of the ttest to the results obtained by the proposed procedure and by and samarium were utilized as chemical modifiers. The methods reported to date for ytterbium determination by instrumental neutron activation analysis demonstrated that there was no significant difference at the 5% probability level. ETAAS are summarized in Table 1.The aim of this work was to show that ETAAS with a Keywords: Ytterbium; digesta; faeces; transversely heated transversely heated graphite atomizer and Zeeman-effect back- graphite atomizer; atomic absorption spectrometry ground correction can be used for the direct determination of low concentrations of ytterbium in digesta and faeces samples Ytterbium salts have been recommended as particulate-phase for animal nutrition studies, thus permitting a marked markers in studies of animal nutrition.Ytterbium fulfils the reduction in the amount of ytterbium necessary for use as a requirements of a good marker, because it is associated pre- solid-phase marker in field experiments. dominantly with particulate matter, is non-absorbable and can be accurately measured in digesta and faeces when infused at EXPERIMENTAL a rate which does not affect the digestive processes.1–2. The first method described for ytterbium determination in Instrumentation digesta and faeces was based on FAAS with a dinitrogen A Perkin-Elmer 4100ZL atomic absorption spectrometer, fur- oxide–acetylene flame.2 To circumvent interference effects, the nished with an ytterbium hollow cathode lamp and equipped authors recommended a matrix matching approach together witha longitudinal Zeeman-effectbackground correction system with the addition of potassium as ionization buffer.2 and a pyrolytically coated transversely heated graphite atomizer Nevertheless, to achieve suitable levels of ytterbium in the (THGA) with integrated L’vov platform (Part No.Bo 504033), samples, in relation to the sensitivity of FAAS, large amounts was employed. Sample aliquots of 10.0 or 20.0 ml were delivered of ytterbium acetate (about 100–300 mg kg-1 diet dry matter) into the tube by means of an AS-71 autosampler also from have to be added to the daily diet of the animals; however, the Perkin-Elmer. Argon was used as the purge gas. The operating field experiments become extremely expensive.parameters and heating programme are shown in Tables 2 and In the last 12 years, several approaches have been adopted 3, respectively. All measurements were made with at least three for enhancing the sensitivity of methods for rare earth element replicates and based on integrated absorbance values. determination by ETAAS, particularly for ytterbium. To attain this objective, tantalum-lined graphite,3 graphite tubes lined with tungsten foil,4 graphite tubes lined with a tungsten spiral Reagents and Reference Solutions and tantalum foil,5 graphite tubes lined with tungsten or tantalum foil,6 graphite tubes coated with zirconium or lantha- All solutions were prepared by using distilled, de-ionized water. Hydrochloric and nitric acids were distilled in quartz sub- num,6 and graphite tubes coated with tantalum,7 have been proposed for atomization.However, according to Yizai and boiling stills (Ku� rner). A 1000 mg l-1 ytterbium stock solution was prepared from Di-Jun,5 all these methods present some limitations: (i) The metals that formed carbides were lost from the Yb2O3 (Johnson Matthey), previously treated at 850 °C for 6 h, by dissolving 0.1139 g of the oxide in 15 ml of concentrated graphite surface after a number of firings, decreasing the useful lifetime of these atomizers.nitric acid by gentle heating on a hot-plate, and making the volume up to 100 ml with water. The analytical calibration (ii) A thin tantalum and tungsten foil-lined graphite tube suffered a deformation problem: the metal foil gradually graph for ytterbium was obtained with reference solutions within the range 1.00–15.0 mg l-1 Yb prepared by serial dilution became fragile with use and susceptible to cracking, making further determinations impossible because a leak appeared of the stock solution with 0.24 mol l-1 HCl.All solutions were stored in poly(propylene) bottles (Nalgene) as recommended.7 between the metal foil and the inner surfaceof the graphite tube.(iii) The metal surface was smooth after atomization, and a The effect of concomitants was investigated with solutions containing up to 2500 mg l-1 Na (NaCl, Merck), 2500 mg l-1 small sample aliquot could flow out of the surface of the tantalum and tungsten. K (KCl, Merck), 2500 mg l-1 Si (Na2SiO3 5H2O, Hopkin & Williams), 2500 mg l-1 P (KH2PO4, Baker), 1000 mg l-1 Ca A few methods using direct atomization from graphite Journal of Analytical Atomic Spectrometry, April 1997, Vol. 12 (475–478) 475Table 1 Methods for ytterbium determination by ETAAS Atomization Sample Graphite tube atomizer Temperature/°C Time/s m0*/pg Rocks Pyrolytic 2500 5 110 Rocks Pyrolytic 2200 10 3.812 Rocks and sediment Pyrolytic lined with tungsten spiral and tantalum foil 2427 12 1.045 Pyrolytic 2427 22 1.125 Uncoated 2800 10 52.66 Pyrolytic 2800 10 3.66 Reference solutions Zr-treated pyrolytic tube 2800 10 2.26 La-treated pyrolytic tube 2800 10 2.66 Tungsten foil lining 2800 10 2.76 Rare earth oxides Pyrolytic; europium and samarium as modifiers 2000 3 1.313 Food Tungsten foil lining 2300 3 4.274 Reference solutions Uncoated 2600 10 508 Pyrolytic 2600 10 3.68 Sea-water Pyrolytic 2800 5 211 Pyrolytic 2700 10 3.69 Reference solutions Pyrolytic-platform 2800 10 2.89 Pyrolytic-microboat 2800 10 3.79 Faeces and digesta Ta-treated pyrolytic tube 2500 5 —7 * m0 : characteristic mass.Table 2 Operating parameters for ytterbium determination HCl) according to IUPAC.14 Accuracy was assessed by analysing the same sample by instrumental neutron activation analy- Wavelength 398.8 nm sis (INAA). Additions of 100, 200 and 300 ng of Yb, by Slit-width 0.2 nm delivering 100, 200 and 300 ml of a solution containing 1.00 Signal measurement Integrated absorbance mg l-1 Yb to 300 mg of ground cow’s faeces, were also Integration time 5 s performed. Additions were made to triplicate solid samples, Sample volume 20 ml Purge gas Argon which were subsequently dry-ashed by the procedure described above.The contents of ytterbium reported by ETAAS were the average of at least three measurements of each sample Table 3 THGA heating programme digested in triplicate. The heating programme of Table 3 was used throughout and 20 ml of solutions were always employed. Step Temperamp/s Hold/s Argon flow rate/ ml min-1 Instrumental Neutron Activation Analysis Dry 140 3 5 250 Dry 160 10 40 250 Samples of about 150 mg of digesta and faeces were inserted Pyrolysis 1400 10 20 250 into appropriate polyethylene vials and irradiated with a Atomization 2300 0 5 0 thermal neutron flux of about 5×1012 n cm-2 s-1 for 8 h, in Clean 2600 1 3 250 the nuclear research reactor IEA-R1 of the Instituto de Pesquisas Energe�ticas e Nucleares (IPEN/CNEN/SP).After Programme time: 97 s Injection temperature: 20 °C cooling times of 15, 25 or 40 d,15 the induced radioactivity (169Yb, t1/2=32.02 d, 177.18 and 197.95 keV) was measured by using an HPGe semiconductor detector (GMX 45190P,EG&G (CaCO3, Merck), 1000 mg l-1 Mg (MgO, Riedel de Ha�en), Ortec, 50% relative efficiency and 1.9 keV resolution at the 500 mg l-1 Fe (Fe2O3, Johnson Matthey), 500 mg l-1 Al 1332 keV 60Co photopeak) coupled to a computerized multi- (AlCl3, Titrisol, Merck), 500 mg l-1 Zn (Zn, Carlo Erba), channel buffer.IAEA Reference Materials (V 10 and Soil 7) 500 mg l-1 Mn (MnSO4 H2O, Merck) and 500 mg l-1 Cu were co-irradiated with the samples for calibration purposes.(Cu, Merck), with and without 10.0 mg l-1 Yb in acidic medium Neutron monitor wires were used with all samples. Data (0.014–0.56 mol l-1 HCl, HNO3 or H2SO4). acquisition was monitored by the software Maestro II (EG&G Ortec). Spectrum deconvolution was accomplished by using Sample Preparation for Graphite Furnace the software Regulus (Sinapse Informa�tica).Digesta and faeces were dry-ashed according to the following RESULTS AND DISCUSSION procedure: approximately 500 mg of ground (60 mesh) and oven-dried (60 °C to constant mass) sample were accurately Preliminary experiments carried out with 10.0 ml of ytterbium weighed in a porcelain crucible, which was placed in a pro- reference solutions in 0.014 mol l-1 HNO3 led to the heating grammable muffle furnace, and heated to 400 °C for 1.5 h. programme given in Table 3.Under these conditions, maximum Then, the temperature was gradually increased to 650 °C and pyrolysis and atomization temperatures not exceeding 1400 and the sample was ashed for at least 12 h. The residue was 2300 °C were found to be satisfactory, and this programme was moistened with 0.5 ml of concentrated HCl and 0.5 ml of water. adopted throughout unless stated otherwise. A good and repro- The crucible was covered and placed on a hot-plate and heated ducible characteristic mass (3.0±0.1 pg Yb) based on integrated to boiling.After cooling, the sample was quantitatively trans- absorbance values was always attained, indicating that the ferred into appropriate calibrated flasks and the volume was graphite atomizer could be employed for the direct atomization made up to 25 or 50 ml with water. of ytterbium without any surface pre-treatment. Each THGA tube supported at least 500 firings of 10 ml of reference solutions. Sen Gupta12 also observed that ytterbium could be directly Figures of Merit atomized from the wall of longitudinal pyrolytically coated The detection limit for ytterbium was determined after 20 graphite tubes, suggesting that this rare earth element does not form undesirable stable carbide species.consecutive measurements of the blank solution (0.24 mol l-1 476 Journal of Analytical Atomic Spectrometry, April 1997, Vol. 12Experiments carried out with HNO3 , HCl and H2SO4 within the concentration range 0.014–0.56 mol l-1 showed that these acids did not affect the atomization of 100 pg of ytterbium.These results were important, because many possibilities could be tested for developing and defining the method of sample pre-treatment. Thereupon, it was decided to dry-ash the samples and dissolve the ash in HCl so that the final acidity was 0.24 mol l-1, cf., the procedure described previously.4 In this acidic medium, the sample solutions were stable for at least 6 months when stored in poly(propylene) bottles.Fig. 1 Ytterbium absorbance profiles: 1, 100 pg Yb with or without The effects of the main concomitants present in faeces and pyrolysis; 2, 100 pg Yb+10.0 mg Ca with pyrolysis; and 3, digesta on the atomization of 10.0 mg l-1 (100 pg) of ytterbium 100 pg Yb+10.0 mg Ca without pyrolysis. are shown in Table 4. Up to 2500 mg l-1 of Na, K, P or Si, 1000 mg l-1 of Ca or Mg, and 500 mg l-1 of Fe, Cu, Al, Zn or Mndid not cause variationshigher than 8% in the ytterbium signals, based on integrated absorbance values.Taking into account the mean chemical composition of the related materials, these results were a good indication of the robustness of the proposed method, even when the ratio of sample mass to final volume was as low as 1 g510 ml. Although no differences were found in the shape of the ytterbium transient signals when the heating programmes were performed with and without pyrolysis, this step was decisive for overcoming interferences and improving peak profiles.For instance, when 100 pg (10.0 mg l-1) of ytterbium were atomized without pyrolysis, 10.0 mg (1000 mg l-1) of Ca and Mg caused a 65 and 22% suppression of the ytterbium integrated absorbance, respectively; however, with pyrolysis no significant interferences were observed. Fig. 1 shows how the transient signal of ytterbium was affected by the presence of Ca. Probably, without pyrolysis, ytterbium has insufficient time to diffuse out of the calcium or magnesium oxide before it exits the furnace (appearance time is increased by 800 ms), whereas Fig. 2 Pyrolysis curves of 100 pg ytterbium in the absence and with pyrolysis, the analyte is first brought to the oxide surface, presence of 10.0 mg each of Al, Mg, Si or P. Atomization tempera- and subsequently diffuses out of the oxide layer, as suggested ture: 2300 °C. by Qiao and Jackson.16 In addition, a slight diminution in the background was observed when the pyrolysis step was included tested contents, the pyrolysis and atomization temperatures of in the heating programme.the heating programme shown in Table 3 were considered The presence of 10.0 mg of each of Mg, P, Si and Al extends suitable for the analysis of these samples. the maximum pyrolysis temperature of 100 pg of ytterbium as The absorbance profiles for ytterbium without and with a shown in Fig. 2: Mg by 200 °C, P and Al by 400 °C and Si by mixture of concomitants (5 mg of Na, K, Ca and Mg, 50 ng of 500 °C.The atomization temperature of ytterbium in the pres- Cu, 100 ng of Mn, Zn, Al and Co, 150 ng of Fe, and 250 ng of ence of Si and Al was increased by 100 °C. No variation in Cr) are shown in Fig. 4. Changes were observed in the presence atomization temperature was observed in the presence of the of these concomitants; the peak was sharper (the height was other concomitants. Also, 10.0 mg of each of Al or Si strongly enhanced by 22%), although the integrated absorbance was affect the absorbance peak profiles of 100 pg of ytterbium virtually unaffected. (Fig. 3). Double peaks became noticeable when the amounts of Al and Si were higher than 2.5 and 5 mg, respectively. It should be mentioned that the characteristic mass of ytterbium was not Ytterbium Determination in Digesta and Faeces affected by the presence of 10.0 mg of Si, when the atomization temperature was kept at 2400 °C; however, a 15% increase was The results of ytterbium determination in faeces and digesta by the proposed method, and by INAA, are shown in Table 5.observed at 2300 °C. As the Si and Al concentrations present in faeces and digesta samples are usually much lower than the An inter-method correlation graph for ETAAS versus INAA Table 4 Effect of concomitants on the atomization of 100 pg of Yb Relative absorbance mg l-1 Na K Ca Mg Si P Fe Al Cu Zn Mn 0 1.00 1.00 1.00 1.00 1.00 1.00 1.00 1.00 1.00 1.00 1.00 10 1.02 0.98 1.01 0.99 1.02 0.97 0.92 1.00 1.02 0.92 1.00 50 1.03 0.98 0.99 1.01 1.02 0.98 0.92 0.98 1.04 0.99 1.01 100 1.01 1.01 0.99 1.00 0.97 1.00 0.93 1.03 1.02 1.00 1.02 250 0.99 1.02 0.98 0.98 0.97 0.97 0.93 0.98 0.98 0.99 1.05 500 0.99 1.00 0.97 0.95 0.99* 0.98 0.93 1.04 0.99 0.99 1.04 1000 1.05 0.95 0.98 0.94 0.95* 0.97 — — — — ; 1500 1.02 0.96 — — 1.00* 0.98 — — — — — 2000 1.01 0.99 — — 1.03* 0.95 — — — — — 2500 1.07 0.98 — — 0.97* 0.96 — — — — — * Atomization temperature, 2400 °C.Journal of Analytical Atomic Spectrometry, April 1997, Vol. 12 477Table 5 Analysis of digesta and faeces by the proposed procedure and by INAA* Ytterbium/mg g-1 Calculated t-value at Sample ETAAS INAA 95% confidence level Horse faeces 1 0.107 (3.3)† 0.121 (6.2) † 0.628 Horse digesta 2 0.037 (4.6) 0.038 (0.7) 0.048 Horse digesta 3 0.101 (6.9) 0.105 (4.2) 0.188 Horse digesta 4 0.200 (3.9) 0.204 (4.5) 0.250 Horse digesta 5 0.130 (7.0) 0.141 (6.5) 0.391 Horse digesta 6 0.109 (5.0) 0.116 (5.5) 0.360 Cow faeces 7 0.138 (1.3) 0.150 (2.8) 1.383 Cow faeces 8 0.122 (3.0) 0.135 (1.4) 1.333 Cow faeces 9 0.078 (7.9) 0.087 (11.3) 0.225 Cow digesta 10 0.760 (0.8) 0.747 (5.8) 0.767 * ttab=3.182 for a=0.05 and nETAAS=nINAA=3.† Values in parentheses are RSDs in %. precision by using ETAAS with a THGA and Zeeman-effect background correction. A practical advantage of the method is the large decrease in the operational costs related to the ytterbium salt (acetate or chloride).One kilogram costs about US$ 3500.00 and between 2 and 5 kg of the ytterbium compound are necessary in field experiments. With the THGA method, the cost caused by ytterbium addition decreases by Fig. 3 Ytterbium absorbance profiles: (a) 100 pg ytterbium alone; (b) 100-fold in comparison with the conventional FAAS method. 100 pg ytterbium plus 10.0 mg Al; and (c) 100 pg ytterbium plus Taking into account the data obtained with high concen- 10.0 mg Si.trations of several concomitants, it can be inferred that ytterbium can be determined in other materials such as soils, sediments and rocks. We thank Adibe L. Abdalla (CENA-USP) and Ana Rita A. Nogueira (EMBRAPA-CPPSE-Sa�o Carlos) for providing samples of digesta and faeces; Daniela Candello Salvadori for sample preparation; Helder de Oliveira (CENA-USP) for his helpful assistance in INAA assays; and Elisabeth de Oliveira (IQ-USP) for loan of the ytterbium hollow cathode lamp.We also thank the Fundac�a� o de Amparo a` Pesquisa do Estado de Sa�o Paulo for financial support (Processo 1995/5782–7), and the Conselho Nacional de Desenvolvimento Cientý�fico e Tecnolo�gico (CNPq) for fellowships to E. C. L. and F. J. K. Fig. 4 Atomic absorption and background signal profiles: AA1= 100 pg Yb; AA2=100 pg Yb plus 5 mg Na, K, Ca and Mg, 250 ng Cr, REFERENCES 150 ng Fe, 100 ng Mn, Zn, Al and Co, and 50 ng Cu. BG1 and BG2= 1 Satter, L. D., Combs, D.K., Lopez-Guisa, J. M., and Nelson, background signals in the absence and presence of the concomitants, W. F., Nuclear and Related Techniques in Animal Production respectively. and Health, Proceedings of a Symposium, IAEA (Vienna), 1986, pp. 469. was constructed (r=0.9997, n=10), and application of the t- 2 Siddons, R. C., Paradine, J., Beever, D. E., and Cornell, P. R., Br. test demonstrated that there was no significant difference at J. Nutr., 1985, 54, 509. 3 Yao, J., Jiang, B., and Huang, W., Spectrosc.Spectral Anal., 1984, the 5% probability level between each pair of results, which 4, 40. was a good indication that ETAAS with a THGA and longi- 4 Miaokang, S., and Yinyu, S., J. AOAC Int., 1992, 75, 667. tudinal Zeeman-effect background correction compared well 5 Yizai, M., and Di-Jun, S., Spectrochim. Acta, Part B, 1992, 47, 459. with INAA. In addition, 100±1% recoveries of small amounts 6 Sneddon, J., and Fuavao, V. A., Anal. Chim. Acta, 1985, 167, 317.of ytterbium added to the same sample digests were obtained, 7 Bellanger, J., At. Spectrosc., 1987, 8, 140. which was also good evidence that the proposed method can 8 Sneddon, J., and Fuavao, V. A., At. Spectrosc., 1982, 3, 51. 9 Fuavao, V. A., and Sneddon, J., At. Spectrosc., 1983, 4, 179. be used for the routine analysis of dry-ashed faeces and digesta. 10 Sen Gupta, J. G., T alanta, 1981, 28, 31. The relative standard deviation of measurements (n=10) for 11 Fujino, O., Nishida, S., Togawa, H., and Hiraki, K., Anal. Sci., typical samples containing 5–10.0 mg l-1 Yb was always lower 1991, 7, 889. than 3%. Each THGA tube supported at least 350 firings for 12 Sen Gupta, J. G., J. Anal. At. Spectrom., 1993, 8, 93. a 20ml injection of dry-ashed samples. The analytical cali- 13 Jie, Z., and Sixuan, G., Analyst, 1995, 120, 1661. 14 Commission on Spectrochemical and other Optical Procedures bration graph was linear (r=0.9997 for n=7) within the range for Analysis. Nomenclature, Symbols, Units and their Usage in 0–15.0 mg l-1, the detection limit was 0.15 mg l-1 Yb or 3.0 pg Spectrochemical Analysis—II, Spectrochim. Acta, 1978, 33, 241. Yb, and the characteristic mass based on the integrated 15 Fernandez, E. A. N., Ferraz, E. S. B., and Oliveira, H., J. Radioanal. absorbance was 3.0 pg Yb. Nucl. Chem., 1994, 179, 251. 16 Qiao, H., and Jackson, K. W., Spectrochim. Acta, Part B, 1991, 46, 1841. CONCLUSION It was possible to determine the natural contents of ytterbium Paper 6/07351D Received October 28, 1996 in digesta and faeces of ruminants with suitable accuracy and 478 Journal of Analytical Atomic Spectrometry, April 1

ISSN:0267-9477    

DOI:10.1039/a607351d

出版商:RSC    

年代:1997

数据来源: RSC

摘要:

Evaluation of Various Sample Preparation Procedures for the Determination of Chromium, Cobalt and Nickel in Vegetables ALATZNE CARLOSENA†, MERCEDES GALLEGO AND MIGUEL VALCA� RCEL Department of Analytical Chemistry, Faculty of Sciences, University of Co� rdoba, E-14004 Co� rdoba, Spain Three methods for the determination of Cr, Co and Ni in cedure for digestion of food samples has been superseded by vegetables were tested with a view to the evaluation of various solid or slurry sampling analysis in recent years.5 The slurry sample preparation procedures involving dry ashing, the direct technique combines the advantages of both liquid and solid slurry technique and continuous flow microwave-assisted sampling, reduces analysis times and minimizes contamination digestion. NIST SRM 1572 Citrus Leaves was used for this risks.Recently, a comprehensive review of the direct analysis purpose. For the determination of the analytes by AAS, the of foods as slurries by atomic spectrometry was published.6 temperature programmes of the graphite furnace were Most applications involve animal tissues and plant materials; optimized and the use of some chemical modifiers was studied.by contrast, few determine Cr, Co and Ni. Thus, Cr has Direct slurry analyses of vegetables (in samples prepared at a been determined in kale7 and spinach,8 and Co and Ni in 2% m/v concentration in 1% HNO3 containing 10% H2O2) vegetables.9,10 were generally accurate and precise for Co and Ni, but gave The introduction of a food slurry into a graphite tube leads low recoveries (about 85%) for Cr.Continuous digestion of the to a matrix excess that can be minimized by using a chemical samples in a microwave oven was necessary for the reliable modifier. Magnesium nitrate is reportedly the best modifier determination of Cr (about 100% recovery) by use of 5% for Cr determination in foods;11 on the other hand, Co and HNO3 as the digesting reagent.The best method for each Ni are usually determined with no chemical modifier.6 In the application was used for the determination of Cr, Co and Ni atomization of slurries of biological materials, carbonaceous in a variety of vegetables, using direct calibration with aqueous residues build up inside the graphite furnace as a consequence standards. of incomplete ashing of the organic matrix. There are two ways of overcoming this problem, viz., by infusing oxygen at Keywords: Electrothermal atomic absorption spectrometry; the ashing stage12 or by adding a chemical oxidant to the continuous flow microwave-assisted digestion ; chromium, slurry.Vin�as et al.10,13 studied the latter option extensively cobalt; nickel; vegetables ; slurry and succeeded in determining several elements in vegetables as slurries, using H2O2 in the suspensed medium; in this way, The contents of heavy metals in vegetables may be increased the carbonaceous residue was virtually completely avoided.by various contamination sources; it is therefore important to In order to improve the accuracy and precision of analyses test and maintain the quality of vegetables in terms of heavy of food slurries, pre-digestion can be of assistance for extracting metals in view of recent dietary guidelines which recommend the analyte into the liquid phase (usually by adding HNO3 or increasing their intake.1 Cr, Co and Ni are essential trace with preliminary ashing to remove the bulk of the organic elements but can be toxic above certain limits.CrIII is involved material).5,6 The combination of microwave oven digestion in glucose and fat metabolism and/or the mechanism of action and a flow injection (FI) interface with atomic spectrometric of the pancreatic hormone insulin; on the other hand, CrVI is detection (flame or plasma) increases sample throughput and extremely irritating and toxic to humans owing to its high minimizes sample contamination;14–19 however, few appli- oxidizing potential.Co is required in the normal human diet cations use ETAAS.20–25 The most recent developments in as vitamin B12 (cyanocobalamin), a deficit of which causes on-line digestion or reaction with a microwave oven and pernicious anaemia. The essentiality of Ni for humans remains atomic spectrometry have been reviewed.26 The growing inter- uncertain and controversial; thus, the significance of Ni to est in this tandem technique has led to the inception of human health is almost exclusively related to its toxicity.Ni commercially available instruments based on it; all process compounds are currently considered to be carcinogens.2 Soil liquid samples or solid slurries.23 However, there seems to be is the main way by which heavy metals enter plants; in no reference to the sequential determination of Cr, Co and Ni response, a European Union directive (86/278/CEE) has been in vegetables by using direct slurry analysis or the on-line issued to control the maximum allowable contents of some microwave-assisted digestion technique.metals in agricultural sewage sludge.3 The average daily intake The purpose of this work was to evaluate various reported in food for Cr, Co and Ni is well documented;4 these elements methods for the determination of trace metals in foods by occur at very low concentrations in vegetables. ETAAS. Cr, Co and Ni were selected because they occur at Three analytical techniques are sufficiently sensitive for low levels in foods and have scarcely been studied in this determining metals at trace levels, viz., NAA, ICP-MS and context.Vegetables were chosen as the samples because they ETAAS. The last is available in most laboratories, so there is are the primary sources of these elements for the human diet; a wealth of literature on the determination of Cr, Co and/or moreover, these samples can readily be dried to obtain finely Ni in biological samples and foods by ETAAS following sample powdered forms that facilitate preparation of the slurry.Three digestion by dry or wet ashing. The tedious traditional prosample preparation procedures for application before ETAAS measurements were tested, namely: dry ashing (the most widely † On leave from the Department of Analytical Chemistry, University of La Corun�a, La Corun�a, Spain. recommended mineralization procedure for non-volatile Journal of Analytical Atomic Spectrometry, April 1997, Vol. 12 (479–486) 479elements in vegetables); the direct slurry technique (the best ground in a household grinder for 10–20 s and then in an agate mortar. All these operations must be completed as procedure for avoiding mineralization of the sample); and continuous flow microwave-assisted digestion (the most expeditiously as possible, to avoid degradation or/and contamination of the sample. No sieving was needed. An SRM supplied recently developed technique for rapid sample digestion). The three methods were evaluated by application to a reference by NIST, viz., SRM 1572 (Citrus Leaves) was dried to a constant mass at 85 °C for 2 h, as per the supplier’s recommen- material and/or lettuce samples, and the method that performed best in each case was used for the real samples.dations, and used for method validation. Two sample preparation procedures were used; blanks were also assayed in parallel in all instances.EXPERIMENTAL Apparatus Dry ashing A Perkin-Elmer Model 1100-B atomic absorption spectrometer A 0.2–0.3 g portion of dried vegetable was accurately weighed (U� berlingen, Germany) equipped with deuterium arc back- in a platinum crucible and placed in a muffle furnace and ground correction, an HGA-700 graphite furnace and an AS-70 ashed at 600–650 °C overnight. The fully ashed sample was autosampler was used. Measurements were made by using dissolved with 2 ml of concentrated HNO3 and evaporated to Perkin-Elmer single-element hollow cathode lamps.Pyrolytic dryness in a sand-bath. Finally, the residue was taken up in graphite-coated tubes (Perkin-Elmer Part No. BO-121092) 10 ml of 0.2% v/v HNO3. with L’vov platforms (Perkin-Elmer Part No. BO-121091) were also employed. Argon was used as the inert gas. All Direct slurry preparation measurements were based on integrated absorbance peak areas. The nal parameters for the spectrometer A 0.1–0.15 g portion of dried vegetable was weighed in a are listed in Table 1.The FI system consisted of a Gilson PTFE vial and mixed with 2.5 ml of 2% v/v HNO3 containing Minipuls-2 peristaltic pump (Villiers-le-Bel, France) fitted with 2.5 ml of H2O2 (20% v/v) to obtain a final slurry volume of poly(vinyl chloride) tubes, PTFE transmission lines of 0.8 mm 5 ml (2–3% m/v). The slurry was then placed in an ultrasonic id, a switching valve [Rheodyne 5301 (Cotati, CA USA)], and bath for 10 min immediately prior to direct analysis (Fig. 1A) a laboratory-made three-piece injector commutator described or before mineralization in an on-line digestion system elsewhere.27 A household microwave oven (AEG Micromat- (Fig. 1B), and then analysed. For on-line digestion, a larger Duo, Madrid, Spain) equipped with a magnetron of 2450 MHz amount of sample (0.2–0.3 g in 5 ml, 4–6% m/v) was used in with a maximum power of 800W was used without further order to ensure a slurry concentration similar to that employed modification. Samples were ground in a household grinder in the direct procedure.(Taurus, Madrid, Spain) and slurries were homogenized in an ultrasonic bath (Bandelin, Tk52, Berlin, Germany) or a vortex Conventional Procedure mixer (Heidolph, Kelheim, Germany). Calibration graphs for Cr, Co and Ni were obtained for variable volumes of standard solutions containing 20 mg l-1 Chemicals Cr, 50 mg l-1 Co and 50 mg l-1 Ni mixed in the graphite tube with appropriate amounts of 0.2% (v/v) HNO3 to a volume All reagents used were of analytical-reagent grade, and high- of 20 ml.The concentration ranges encompassed were 0–20 purity water (Milli-Q Water System, Millipore, Madrid, Spain) mg l-1 for Cr, and 0–50 mg l-1 for Co and Ni. The blanks was employed throughout. Metal standards were prepared contained 20 ml of 0.2% v/v HNO3. Dilute samples (either daily by diluting appropriate aliquots of a 1000 mg l-1 stock mineralized by dry ashing or as slurries) were introduced metal solution (Panreac, Barcelona, Spain) with 0.2% v/v directly into the graphite furnace via the autosampler.For HNO3 . A 10.0 g l-1 MgII solution prepared from magnesium direct slurry analysis (Fig. 1A), samples must be mixed in a nitrate hexahydrate (Merck, Darmstadt, Germany) and a 1.0 vortex shaker between measurements in order to ensure repro- g l-1 PdII solution made from palladium nitrate (Sigma, ducible results for Cr (viz. to avoid particle settling).All Madrid, Spain) were employed as chemical modifiers. Triton measurements were made in triplicate. X-100 (Serva Feinbiochemica, Heidelberg, Germany) was also used. Flow Digestion Systems Flow system connected to the autosampler cup for microwave- Sample Preparation assisted digestion of slurry samples Vegetable samples were purchased at a local supermarket. Only edible parts were taken, washed with high-purity water, The flow manifold used is shown in Fig. 1B. The slurry sample (4–6% m/v), in 1% HNO3 containing 10% H2O2, was cut and oven-dried at 85 °C for 24 h. Next, they were Table 1 Instrumental parameters and optimized furnace conditions for the determination of Cr, Co and Ni* Cr Co Ni Lamp current/mA 10 30 25 Wavelength/nm 357.9 240.7 232.0 Bandpass/nm 0.7 0.2 0.2 Temperature/ Ramp/ Hold/ Temperature/ Ramp/ Hold/ Temperature/ Ramp/ Hold/ Step °C s s °C s s °C s s Dry 1 130 5 10 130 5 10 130 5 10 Dry 2 160 10 20 160 10 20 160 10 20 Pyrolysis 1600 20 20 1500 15 20 1400 15 20 Atomize 2650 0 5 2600 0 5 2650 0 6 Clean 2650 1 3 2650 1 2 2650 1 2 * A stream of argon at 300 ml min-1 was used (the flow was stopped during the atomization step); injected volume, 20 ml. 480 Journal of Analytical Atomic Spectrometry, April 1997, Vol. 12Fig. 1. Approaches used for the determination of metals in vegetables by the slurry-sampling technique: A, direct slurry analysis (for replicate measurements of Cr in the same sample, homogenization in a vortex mixer was necessary); B, semi-on-line digestion system for slurry samples; and C, semi-on-line system for automatic mineralization of the sample; R, 1% HNO3 plus 10% H2O2; R¾, 5% HNO3; LS and LR¾, loops of sample (400 ml) and R¾ (600 ml), respectively; W, waste; IC, injector commutator.Flow rates: B, 0.5 ml min-1; C, 0.25 ml min-1. homogenized and fed into the microwave oven at 0.5 ml min-1 simultaneously unloaded into the air carrier streams at a flow rate of 0.25 ml min-1 and driven to the merging point, X.The for 1 min (sample volume, 0.5 ml). An air stream at the same flow rate was used to transfer the sample to the PTFE digestion mixed solution was then led to the PTFE digestion coil and the sequence was continued as described above, except for the coil (400 cm×0.8 mm id) located inside the microwave oven. The coil was wrapped around an Erlenmeyer flask filled with introduction of reagent R, which was omitted.In both methods, calibration graphs were constructed as water and positioned in front of the magnetron, as per previous experience.24 About 1 min after the sample had been introduced described under Conventional Procedure since no digestion of the standards was necessary. Sample blanks were introduced (when the slurry plug was inside the digestion coil), the oven was switched on at a power of 500 W for 2 min. The treated directly into the manifolds. sample was collected in an autosampler cup (the cup was fitted with a pierced anti-evaporation stopper allowing insertion of RESULTS AND DISCUSSION the flow line, in order to minimize contamination and losses).Furnace Programme and Chemical Modifier for the A 0.5 ml volume of a 1% HNO3 solution containing 10% Determination of Cr, Co and Ni H2O2 (R in Fig. 1B) was introduced, after the oven had been switched off, via the same tube as the air stream in order to Accurate, reproducible determinations of metals by ETAAS transport any residual treated sample and flush the manifold.with the slurry-sampling technique rely on appropriate heating A coil of 150 cm (0.8 mm id), inmersed in a water-filled beaker, programmes (especially as regards the pyrolysis conditions), as was inserted after the oven in order to allow digestion fumes well as on the use of chemical modifiers. The furnace conditions to expand. and the effect of two chemical modifiers were thus carefully examined for both standards and slurries.Lettuce was selected as the representative vegetable and the conventional method Flow injection system connected to the autosampler cup for illustrated in Fig. 1A as the analytical tool. automatic mineralization of the sample A 0.2–0.3 g amount of dried vegetable suspended in 5 ml of Chromium 0.2% HNO3 was homogenized by ultrasonication for 10 min and then used to fill the loop (LS) of the injector (400 ml); A 15 ml volume of a standard containing 10 mg l-1 Cr or a sample of 2% m/v lettuce slurry (both prepared in 1% simultaneously, a stream of digesting solution (R¾) containing 5% HNO3 was fed to the other loop (LR¾, 600 ml).In the HNO3+10% H2O2), and 5 ml of chemical modifier (10.0 or 1.0 g l-1 magnesium or palladium nitrate, respectively) were ‘inject’ position (Fig. 1C), the contents of both loops were Journal of Analytical Atomic Spectrometry, April 1997, Vol. 12 481injected into the pyrolytic graphite coated tube (with platform).Peak areas were used for quantification. Two drying steps were required to evaporate the solvent completely with no splattering (as observed visually); the slurry sample required higher temperatures (130 and 160 °C) than the standard (100 and 140 °C). The pyrolysis and atomization temperatures were examined over the ranges 500–2000 and 1900–2650 °C, respectively. In the absence of modifier, the optimum pyrolysis temperature was 1600 °C for the standard and sample.This temperature is significantly higher than those reported by other workers (1200–1300 °C)28,29 for Cr in the absence of modifier probably because the HNO3–H2O2 medium acts as a chemical modifier. Palladium nitrate resulted in no additional improvement in the thermal stability of Cr in the standard or slurry. In contrast, the use of magnesium nitrate permitted the pyrolysis temperature to be raised to 1700 °C; however, it caused the background signal for the slurry sample to rise dramatically (see Fig. 2A). Although this modifier has been recommended for the determination of Cr in biological matrices, its use in this case stabilized both the analyte and its concomitants. Cobalt The drying conditions were the same as for Cr. A 25 mg l-1 Co standard, a 2%m/v lettuce slurry and the above-mentioned modifiers were used. The pyrolysis temperature (in the absence of modifier) was critical for the determination of Co in the slurry sample; thus, as can be seen from Fig. 2B, raising the temperature from 1350 to 1500 °C decreased background noise and caused the high peak obtained before that of the analyte peak virtually to disappear. The behaviour of the standard solution was similar at both temperatures; no Co was lost up to 1500 °C. No practical advantages were obtained from using magnesium or palladium nitrate; on the other hand, magnesium caused an undesirable increase in the background signal (similar to Cr, Fig. 2A). At a pyrolysis temperature of 1500 °C, good results were obtained at an atomization temperature of 2600 °C. Nickel The standard solution (25 mg l-1 Ni) or slurry (2% lettuce in the HNO3–H2O2 mixture) and magnesium or palladium nitrate modifier were employed. The temperature programme was started with two drying steps (130 and 160 °C). The maximum Fig. 2. Atomization profile for the analytes in lettuce slurry samples pyrolysis temperature with both modifiers was the same as (2% m/v in 1% HNO3 containing 10% H2O2).A, Cr determination: (a) in the absence of a chemical modifier, at a pyrolysis temperature that in their absence: 1400 °C. Moreover, the addition of 50 mg of 1600°C (background signal, 0.025 A s); and (b) in the presence of of magnesium nitrate increased the background integrated 50 mg of Mg(NO3)2 ,at a pyrolysis temperature of 1700 °C (back- absorbance of the slurry sample (0.034 A s without modifier ground signal, 0.120 A s).B, Co determination without a chemical versus 0.119 A s with modifier), which was also similar to those modifier, at a pyrolysis temperature of (a) 1350 °C (background signal, reported for Cr and Co. As can be seen from Fig. 2C, a good 0.202 A s), and (b) 1500 °C (background signal, 0.076As). C, Ni atomic signal was rapidly obtained with negligible background determination without a chemical modifier (pyrolysis temperature, 1400 °C). Solid and dashed lines represent atomic and background integrated absorbance in the absence of modifier.signals, respectively. The furnace programme, unless otherwise noted, The above-described experiments revealed that the method was that shown in Table 1. performed similarly for the three elements studied. The addition of a chemical modifier did not improve the atomization signal; slurries (prepared in the HNO3–H2O2 medium), with no rather, it increased the background signal because it stabilized chemical modifier.the sample matrix. Although magnesium allows the pyrolysis temperature to be raised to 1700 °C in the determination of Cr, the problems arising from the use of this modifier offset Direct Slurry Sampling Technique the advantages. The advantages of both chemical modifiers for the determination of these elements in biological matrices are The precision and accuracy of determinations using the slurry techniquerely basically on sample homogenization, the particle well documented; however, there is increasing evidence that the presence of H2O2 plays a prominent role in the destruction size of the solid material, slurry stability, the liquid medium employed to prepare the slurry, and the slurry concentration of the total carbonaceous residue10,13 because the peroxide acts as a true oxidant modifier.In conclusion, the temperature (% m/v). A homogeneous slurry can be obtained by manual, mechanical, magnetic and ultrasonic means, as well as by programmes shown in Table 1 were used for standards and 482 Journal of Analytical Atomic Spectrometry, April 1997, Vol. 12passing a gas stream through the sample. Various stabilizing agents (mainly surfactants) have been tested in order to improve slurry homogeneity and stability (e.g., by avoiding too early settling of particles). Homogenization and stability of slurries The effectiveness of two types of mixing for slurry homogenization, viz., mechanical (vortex) and ultrasonic agitation, was examined. For this purpose, various samples of 2% m/v lettuce slurry (in 1% HNO3 containing 10% H2O2) were mixed in a vortex shaker for different times (from 30 s to 30 min) and the metals (Cr, Co and Ni) were determined.The same protocol was followed with an ultrasonic bath. Both homogenization systems required at least 10 min to ensure stabilization of Fig. 3. Influence of the slurry concentration (% m/v) for a lettuce integrated absorbance average values (n=3 at each time); sample on the atomic signal in the determination of Cr, Co and Ni by direct slurry analysis (see Fig. 1A). Solid line, working range. however, the mechanical stirrer provided lower precision (RSD=17.6, 3.6 and 6.2% for Cr, Co and Ni, respectively) than ultrasonic mixing (RSD=13.1, 3.0 and 4.0%). Since the slurry aliquot.30 A linear response was obtained for Co precision for Cr was inadequate, Triton X-100 was tested as a throughout the concentration range tested. However, a slurry stabilizing agent.Thus, several samples of 2% m/v lettuce concentration above 3% m/v caused a significant decrease in slurry (in 1% HNO3 containing 10% H2O2) were spiked with the absorption signals for Cr and Ni, the limit being more different amounts of the surfactant (from 0 to 0.025% m/v) critical for Cr (see Fig. 3). and homogenized by ultrasonication for 10 min before analysis. Based on the results, the determination of Cr in lettuce In the presence of Triton X-100, the sensitivity for Cr decreased slurry is more critical than those of Co and Ni.In order to gradually as the concentration of surfactant was increased (e.g., determine the three elements in the same sample sequentially, the signal decreased by about 50% for 0.025% Triton X-100). the optimum conditions for Cr were selected. Therefore, sample On the other hand, the surfactant made no difference to the slurries were prepared at concentrations between 1 and 3% Co and Ni determinations.The precision of Cr measurements m/v in 1% HNO3–10% H2O2 medium (without surfactant), (after homogenization of the slurry) in the autosampler cup and homogenized by ultrasonication. Additional stirring of the was similar in the presence (0.005% m/v Triton X-100) and cup was required for replicatemeasurements of Cr (see Fig. 1A). absence of surfactant (RSD=10.4 and 11.9%, respectively). Under these conditions, the precision, as RSD, for the three This repeatability can be raised to 6.8% with additional elements ranged from 3.0 to 7.5%.Finally, in order to investi- mechanical mixing of the autosampler cup contents between gate the efficiency of the slurry procedure, five lettuce samples measurements (see Fig. 1A). Higher precision (RSD<5.0%, were analysed in triplicate and the results compared with those n=3) was obtained for Co and Ni with or without surfactant, obtained by ashing the sample and analysing it directly by probably because both elements are extracted more efficiently ETAAS.The results are listed in Table 2. A comparison of the into the liquid phase than is Cr. two sets of results allows one to draw the following conclusions: (1) The direct slurry technique is usable for the determination of Co and Ni as it provides results similar to those obtained Suspended medium composition by ashing the sample. (2) The direct slurry procedure is The effect of the HNO3 concentration was studied over the inadvisable for the determination of Cr because it gives only range 0–2% v/v by preparing various 2% m/v lettuce slurries about 85% of the recovery obtained with ashing mineraliz- (in 10% H2O2).A concentration of 1% was required to ensure ation; this can be ascribed to Cr being partially extracted into good sensitivity and precision (RSD<5%) for Co and Ni; the the liquid phase of the slurry and resulting in incomplete precision for Cr (RSD=7.5%) was not improved by higher release and occlusion of the solid matrix.concentrations of HNO3. The effect of H2O2 was studied over the range 0–15% v/v. The determination of Co and Ni required no addition of H2O2 to the slurry sample. The analytical signal Flow Digestion Systems for Cr was dependent on the H2O2 concentration; thus, the Microwave-assisted digestion of the slurry sample peak areas of Cr were 0.040, 0.070 and 0.095 for 0, 5 and 10% v/v H2O2, respectively. Higher concentrations increased the The direct slurry technique provided poor results for Cr in lettuce samples; hence, the use of a microwave-assisted diges- blank signals for Cr with no additional advantage (e.g., 0.015 A s for 1% HNO3 versus 0.042 A s for 1% HNO3 plus tion system was tested in order to increase the recovery and avoid the need for mechanical mixing between measurements. 10% H2O2 ). The blank signals for Co and Ni were not altered by the H2O2 concentration and remained below 0.005 A s. In The most simple system possible was assembled from a household microwave oven as shown in Fig. 1B. A slurry concen- subsequent experiments, slurry samples (for the determination of Cr, Co and Ni) were prepared in 1% HNO3 containing tration of 4% m/v was prepared in 1% HNO3–10% H2O2, taking into account that the sample was to be subsequently 10% H2O2. The presence of H2O2 favoured the decomposition of the carbonaceous residue in the graphite tube. diluted twice (with a washing solution of 1% HNO3 plus 10% H2O2).The microwave power was set at 500 W and the oven started when the sample (0.5 ml), carried by an air stream, Slurry concentration was inside the digestion coil (about 1 min). The sample was digested in the oven for different periods (1, 2 or 4 min) by Fig. 3 shows the influence of the concentration of slurry in the lettuce sample on the analytical signals for Cr, Co and Ni. actuating the pump as required. The maximum recovery of Cr was about 90% for a 2 min digestion time and did not The concentrations tested ranged from 0.5 to 5% m/v.Concentrations below 1% were unsuitable owing to the low appreciably increase at longer times. The next experiment was intended to determine the most suitable microwave power; levels of these metals in vegetables; on the other hand, concentrations above 5% may result in inefficient pipetting of the over 500W caused dispersion of the sample plug, which was Journal of Analytical Atomic Spectrometry, April 1997, Vol. 12 483Table 2 Cr, Co and Ni contents (mg g-1) in a lettuce sample and an SRM (SRM 1572, Citrus Leaves) obtained with dry ashing, direct slurry sampling and flow digestion methods Concentration mg g-1±s (% RSD, n=5) Dry ashing Sample Element (n=3) Direct slurry Flow system 1* Flow system 2* Lettuce Cr 0.18±0.02 (11.1) 0.15±0.01 (7.5) 0.16±0.02 (12.5) 0.18±0.01 (5.6) Co 0.70±0.02 (2.9) 0.68±0.02 (3.0) 0.71±0.02 (2.8) 0.69±0.02 (2.9) Ni 3.14±0.20 (6.4) 3.26±0.13 (4.0) 3.19±0.14 (4.4) 3.10±0.10 (3.2) Citrus Leaves Cr 0.81±0.06 (7.4) 0.66±0.05 (7.6) 0.70±0.07 (10.0) 0.79±0.05 (6.3) (SRM 1572) (0.8±0.2) † Co 0.025±0.005 (20) — — — (0.02)‡ Ni 0.64±0.06 (9.4) 0.56±0.04 (7.1) 0.58±0.04 (6.9) 0.61±0.03 (4.9) (0.6±0.3)† * Flow systems 1 and 2 are depicted in Fig. 1B and C, respectively. † Certified value. ‡ Non-certified value. swept out of the oven, thereby decreasing the precision of the determinations (RSD>20%). The length of the digestion coil could be adjusted to between 200 and 400 cm (0.8 mm id).Based on the above results, the stopped-flow mode was unnecessary as the sample plug remained inside the digestion coil for a sufficiently long time (the oven was operated for 2 min at 500W). A coil of 150 cm (0.8 mm id) immersed in a water-filled beaker was required at the oven outlet to allow for expansion of evolved gases. The optimized semi-on-line digestion system is shown in Fig. 1B. A comparison between this digestion system and dry ashing was made by analysing five individual samples in triplicate. As can be seen in Table 2, the recovery of Cr was only 90% because the slurry was only partially mineralized in the flow system.Moreover, if no mechanical mixing was applied between measurements, the precision was rather low (RSD about 12.5%). The results obtained for Co and Ni were similar to those provided by direct slurry analysis. The microwave digestion system (see Fig. 1B) offers no advantages over the direct slurry procedure Fig. 4. Response surfacefor the determinationof Cr in lettuce samples (see Fig. 1A); rather, it is more time consuming and involves (2% m/v) at different HNO3 and H2O2 concentrations (R¾). The manifold used is depicted in Fig. 1C. more complex operations. Automatic mineralization of samples advisable in order to prevent carry-over. The length of the digestion coil must be at least 400 cm (0.8 mm id) because the The lettuce sample was suspended at a concentration of 4% injected volume was twice that used in the manifold of Fig. 1B. m/v in 0.2% HNO3. After ultrasonication for 10 min, 400 ml The optimized FI manifold for the automatic mineralization of sample were simultaneously injected with 600 ml of dilute of samples is depicted in Fig. 1C. Because the sample was HNO3 . Both volumes were injected into air carrier streams completely digested under the conditions used, slurry sample (flow rate, 0.25 ml min-1) to offset the sample dilution taking concentrations above 6% m/v could be processed; however, into account the low levels of Cr, Co and Ni in vegetables.the manifold lines (e.g., the injector commutator) were clogged The operating conditions for the oven were the optimum found by high slurry concentrations. Five individual lettuce samples in the previous study (2 min at 500 W). The digesting reagent were prepared and analysed in triplicate. The results were (R¾ in Fig. 1C) was prepared by spiking the samples with compared with those obtained by dry ashing (Table 2).The variable amounts of HNO3 and H2O2 over the ranges 0.2–20 precision for Cr was good (about 5.6%) and its recovery was and 0–15% v/v, respectively. The influence of both chemical increased to about 100% (see Table 2); also, the need for variables on the Cr analytical signal is shown graphically as a additional mixing between measurements was avoided. This three-dimensional plot in Fig. 4. Based on the response surface, automatic mineralization system also gave good results for Co H2O2 improved the recovery of Cr at low concentrations of and Ni, which, however, were similar to those obtained by HNO3 (e.g., 3%); however, its presence was unnecessary direct slurry analysis. above 5% HNO3 (in addition, it increased blank signals for this element, similarly as in direct slurry analyses). On the other hand, a 5% HNO3 concentration was sufficient for the Validation of the Methods: Accuracy and Precision complete mineralization of the sample (as checked visually); above 15% HNO3, the signal decreased slightly because the In order to validate the accuracy of the methods tested, an SRM was used.Penninckx et al.31 reported a strategy for high pressure inside the manifold swept the sample out of the furnace before it could be mineralized. The digesting reagent choosing reference materials for the validation of atomic absorption methods for food analysis; they used two param- selected for this flow method was thus 5% HNO3 .The determinations of Co and Ni were not influenced by the HNO3 eters (inorganic materialand dietary fibre) that can significantly affect the determination of metals by ETAAS but are not (above 1%) or H2O2 concentration. Flushing the sample loop with a washing solution (0.2% HNO3) between samples was included in the AOAC’s classification. Taking into account the 484 Journal of Analytical Atomic Spectrometry, April 1997, Vol. 12recommendations of these workers and available materials five individual samples in triplicate; as can be seen in Table 3, the results support the above conclusions. with certified contents of Cr, Co and Ni, the most appropriate reference material for the intended purpose was NIST The analytical figures of merit of the methods tested were determined by using aqueous standards (0.2% HNO3) and the SRM 1572 (Citrus Leaves), where the concentration of Co is only orientative, however. Several 2 or 4% m/v slurry samples temperature programmes shown in Table 1.The average slopes (n=5) of the calibration graphs obtained were 1.21±0.04, (for direct slurry sampling and flow systems, respectively) were prepared. The average contents as determined in five individual 0.41±0.02 and 0.23±0.01 A s ng-1 for Cr, Co and Ni, respectively. In order to confirm the absence of any matrix effects, determinations of Cr and Ni by using each method are listed in Table 2.The Co content could only be determined by using several samples of SRM 1572 (for Cr and Ni) and lettuce slurries (for Cr, Co and Ni) were prepared by using the the dry ashing method owing to its extremely low concentration in the reference material (other methods would require at least different procedures studied and standard additions calibration graphs were obtained from them. No significant differences at about 25% m/v as a slurry). Direct slurry analysis was a good choice for the determination of Ni and, also probably, Co (see a confidence level of 95% were found between the slopes of the additions and calibration graphs for the three elements in the results for lettuce samples in Table 2); however, this method provided a lower accuracy for Cr; hence, it is not advisable for all instances.These results show that direct calibration with aqueous standards is valid for all the methods tested. However, this element.Of the two flow systems studied, one (Fig. 1B) offers no advantages over the direct slurry sampling method; each method requires an appropriate sample blank. Table 4 shows the detection limits (calculated as three times the the other (Fig. 1C) provides good accuracy for the elements studied, probably owing to complete mineralization of the standard deviation of the signal obtained from 15 sample blanks) and the characteristic masses (the amounts, in picog- sample, which facilitates the determination of Cr.Finally, in order to identify any significant differences between the certified rams, that provided a signal of 0.0044 A s) for each method studied; the detection limits are expresed in ng ml-1 or ng g-1 contents and those provided by the methods studied, the Student’s t-test was applied at a confidence level of 95%. The for a 2% m/v slurry sample. There are significant differences in the detection limits for Cr because the sample blanks above conclusions were statistically confirmed; thus, the determination of Cr was only sufficiently accurate when the containing H2O2 provided higher signals than those obtained with 5% HNO3 only (FI system in Fig. 1C). No significant automatic mineralization FI system was used. The precision of the methods was evaluated by using the differences between the methods for Co and Ni were observed, however. same reference material (for Cr and Ni) and the lettuce sample (for Cr, Co and Ni). First, the repeatability of the autosampler injection of 20 ml of the same sample was assessed; the best Analysis of Vegetable Samples results for the three elements (Table 3) were obtained when the sample was mineralized because it remained homogeneous The selection of vegetables was made according to the recommendations of Penninckx et al.;31 thus, the samples should (Fig. 1C). Second, the precision of the sample digestion procedure was calculated by using the same sample and both flow have a similar chemical composition to that of the reference material studied (SRM 1572).The two methods selected, direct systems (Fig. 1B and C). The results were good and similar for Co and Ni (see Table 3); however, good precision for Cr slurry sampling and automatic mineralization, were used to determine Cr, Co and Ni in a variety of vegetables. Aqueous was only achieved with the automatic digestion system. Finally, the total precision of the methods was calculated by analysing standards for calibration graphs and sample blanks were Table 3 Precision for the autosampler injection, sample digestion and procedure used by the different methods tested Precision as RSD (%) Autosampler injection* Sample digestion* Procedure† (n=3) (n=5) (n=15) Method Element SRM 1572 Lettuce SRM 1572 Lettuce SRM 1572 Lettuce Direct slurry sampling Cr 6.1 6.8 — — 7.6 7.5 (2% m/v) Co — 2.4 — — — 3.0 Ni 4.8 3.3 — — 7.1 4.0 Flow system 1 Cr 7.9 8.7 9.8 10.3 10.0 12.5 (4% m/v) Co — 1.9 — 2.5 — 2.8 Ni 4.1 2.6 5.7 3.8 6.9 4.4 Flow system 2 Cr 0.9 1.1 4.7 4.1 6.3 5.6 (4% m/v) Co — 1.3 — 2.2 — 2.9 Ni 2.0 1.6 4.2 2.3 4.9 3.2 * From a single sample.† From five individual samples. Table 4 Detection limit (DL) and characteristic mass (m0) (±standard deviation, n=5) for the different methods tested Cr Co Ni DL/ DL/ m0/ DL/ DL/ m0 / DL/ DL/ m0/ Method* ng ml-1 ng g-1† pg per 0.0044 A s ng ml-1 ng g-1† pg per 0.0044 A s ng ml-1 ng g-1† pg per 0.0044 A s Direct slurry sampling 1.69 84.5 5.2±0.3 0.56 28.0 9.9±0.3 0.84 42.0 16.6±0.5 Flow system 1 1.53 76.5 4.1±0.4 0.49 24.5 11.2±0.5 0.92 46.1 17.1±0.6 Flow system 2 0.47 23.6 3.7±0.2 0.61 30.4 10.4±0.3 0.87 43.5 15.9±0.4 * Direct slurry sampling and flow systems 1 and 2 are depicted in Fig. 1A, B and C, respectively. † For a 2% m/v slurry sample. Journal of Analytical Atomic Spectrometry, April 1997, Vol. 12 485Table 5 Cr, Co and Ni contents (mg g-1) in a variety of vegetables as determined by the direct slurry sampling and automatic mineralization methods (n=3) Cr/mg g-1 Co/mg g-1 Ni/mg g-1 Vegetable Direct slurry FI system* Direct slurry FI system* Direct slurry FI system* Lettuce 0.15±0.01 0.18±0.01 0.68±0.02 0.69±0.02 3.26±0.13 3.10±0.10 Endive 0.14±0.01 0.18±0.01 0.072±0.004 0.067±0.003 0.76±0.04 0.79±0.04 Cauliflower 0.11±0.01 0.14±0.01 0.33±0.01 0.34±0.01 0.31±0.03 0.29±0.02 Cabbage 0.13±0.01 0.17±0.01 0.096±0.005 0.103±0.004 0.88±0.04 0.92±0.04 Leek <DL† 0.095±0.008 0.041±0.005 0.045±0.004 0.38±0.03 0.39±0.03 Watercress 0.11±0.01 0.15±0.01 <DL <DL 1.62±0.07 1.68±0.06 * FI system of Fig. 1C. † <DL, lower than the detection limit. 5 Bendicho, C., and de Loos-Vollebregt, M. T. C., J. Anal. At. employed. The average contents for three individual samples Spectrom., 1991, 6, 353. analysed in triplicate are listed in Table 5. The two sets of 6 Arruda, M. A. Z., Gallego, M., and Valca�rcel, M., Quý�m. Anal., results for Co and Ni are consistent; however, direct slurry 1995, 14, 17.sampling provided only about 80% of the Cr recovery obtained 7 Minami, H., Zhang, Q., Itoh, H., and Atsuya, I., Microchem. J., with the FI digestion system. The Cr, Co and Ni contents 1994, 49, 126. 8 Miller-Ihli, N. J., J. Anal. At. Spectrom., 1988, 3, 73. found in all vegetables were lower than their tolerated limits. 9 Dobrowolski, R., and Mierzwa, J., Fresenius’ J. Anal. Chem., It should be noted that the highest values were obtained for 1993, 346, 1058.the lettuce sample because this type of vegetable has a great 10 Vin�as, P., Campillo, N., Lo�pez-Garcý�a, I., and Herna�ndez- ability to accumulate metals.1 Co�rdoba, M., At. Spectrosc., 1995, 16, 86. 11 Wagley, D., Schmiedel, G., Mainka, E., and Ache, H. J., At. Spectrosc., 1989, 10, 106. CONCLUSIONS 12 Ebdon, L., Fisher, A. S., Parry, H. G. M., and Brown, A. A., J. trom., 1990, 5, 321. A mixture of HNO3–H2O2 acts as an oxidant modifier that 13 Vin�as, P., Campillo, N., Lo�pez-Garcý�a, I., and Herna�ndez- allows one to dispense with conventional chemical modifiers Co�rdoba, M., T alanta, 1995, 42, 527.(e.g., Mg and Pd); moreover, its presence affords a high 14 Morales-Rubio, A., Salvador, A., and de la Guardia, M., Fresenius’ pyrolysis temperature for Cr and Co. Based on the results, J. Anal. Chem., 1992, 342, 452. appropriate selection of a sample preparation procedure relies 15 Carbonell, V., Morales-Rubio, A., Salvador, A., de la Guardia, M., Burguera, J.L., and Burguera, M., J. Anal. At. Spectrom., essentially on the particular elements involved; thus, the auto- 1992, 7, 1085. matic mineralization system is the best option for the sequential 16 Haswell, S. J., and Barclay, D., Analyst, 1992, 117, 117. determination of Cr, Co and Ni in vegetable samples with 17 Karanassios, V., Li, F. H., Liu, B., and Salin, E. D., J. Anal. At. good accuracy and precision.However, if only Co and Ni are Spectrom., 1991, 6, 457. to be determined, direct slurry sampling can be used instead 18 Gluodenis, T. J., Jr., and Tyson, J. F., J. Anal. At. Spectrom., 1993, for the sake of simplicity. Both proposed methods save much 8, 697. 19 Martines Stewart, L. J., and Barnes, R. M., Analyst, 1994, time relative to conventional dry ashing followed by acid 119, 1003. dissolution; also, they reduce the risk of contamination and 20 Burguera, M., and Burguera, J.L., L ab. Robot. Autom., 1993, 5, 277. losses. The concentrations of Cr, Co and Ni found in the 21 Burguera, J. L., and Burguera, M., J. Anal. At. Spectrom., 1993, vegetables analysed demonstrated that this type of food con- 8, 235. tributes only slightly to their total intake. Thus, increasing the 22 Arruda, M. A. Z., Gallego, M., and Valca�rcel, M., J. Anal. At. consumption of vegetables does not significantly affect the Spectrom., 1995, 10, 501. 23 Sturgeon, R. E., Willie, S. N., Methven, B. A., Lam, J. W. H., and intake of the metals studied. Matusiewicz, H., J. Anal. At. Spectrom., 1995, 10, 981. 24 Arruda, M. A. Z., Gallego, M., and Valca�rcel, M., J. Anal. At. The Spanish DGiCyT is acknowledged for financial support Spectrom., 1996, 11, 169. (Grant PB95–0977). A.C.Z. is also grateful to the Xunta of 25 Chakraborty, R., Das, A. K., Cervera, M. L., and de la Guardia, Galicia for the award of a research grant (DOG No.147, M., Fresenius’ J. Anal. Chem., 1996, 355, 99. 2-8-95). 26 Taylor, A., Branch, S., Crews, H. M., Halls, D. J., and White, M., J. Anal. At. Spectrom., 1996, 11, 103R. 27 Arruda, M.A.Z., Gallego, M., and Valca�rcel, M., Anal. Chem., REFERENCES 1993, 65, 3331. 28 Chakraborty, R., Das, A. K., Cervera, M. L., and de la Guardia, 1 Kabata-Pendias, A., and Pendias, H., T race Elements in Soils and M., J. Anal. At. Spectrom., 1995, 10, 353. Plants, CRC Press; Boca Raton, FL, 2nd edn., 1992. 29 Thomaidis, N. S., Piperaki, E. A., Polydorou, C. K., and 2 Seiler, H. G., Sigel, A., and Sigel, H., Handbook on Metals in Efstathiou, C. E., J. Anal. At. Spectrom., 1996, 11, 31. Clinical and Analytical Chemistry, Marcel Dekker, New York, 30 Lynch, S., and Littlejohn, D., T alanta, 1990, 37, 825. 1994. 31 Penninckx, W., Smeyers-Verbeke, J., Vankeerberghen, P., and 3 Council Directive of June 12, 1986, On the Protection of the Massart, D. L., Anal. Chem., 1996, 68, 481. Environment, and in Particular of the Soil, when Sewage Sludge is used in Agriculture, Official Journal of the European Communities, Paper 6/07939C No. L 181/6–12, Brussels, 1986. Received November 22, 1996 4 Belitz, H. D., and Crosh, W., Food Chemistry, Springer-Verlag, New York, 1987. Accepted January 16, 1997 486 Journal of Analytical Atomic Spectrometry, Apr

ISSN:0267-9477    

DOI:10.1039/a607939c

出版商:RSC    

年代:1997

数据来源: RSC

摘要:

INTER-LABORATORY NOTE Determination of Selenium in Geochemical Samples by Flow Injection Hydride Generation Inductively Coupled Plasma Atomic Emission Spectrometry Following On-line Removal of Iron Interference by Anion Exchange L. D. MARTINEZ, E. SAIDMAN, E. MARCHEVSKY AND R. OLSINA Area de Quý�mica Analý�tica, Facultad de Quý�mica, Bioquý�mica y Farmacia; Universidad Nacional de San L uis, Chacabuco y Pedernera, 5700-San L uis, Argentina Selenium was determined in geological samples with a high lanthanum hydroxide,10 or isolating it by means of ion exchange in a batch system have also been employed,11–13 but content of iron by HG–ICP-AES in combination with flow they have the disadvantage of being tedious, labour intensive injection (FI ) on-line iron separation.Anion-exchange and to some extent dependent on operator skill. Moreover, separation with hydrochloric acid was applied for the Thompson et al.14 found that the iron interference still persisted separation of selenium from iron.The iron–chloro complexes when the method based on coprecipitation with lanthanum were adsorbed on an anion-exchange (Dowex 1X-8) hydroxide was used. Good results were obtained by Martinez microcolumn (3.0 mm id×50.0 mm bed length), while the et al.15 for the determination of selenium by HG–ICP-AES in analyte was introduced into the hydride generation system. samples with high iron content, after separation of the iron by After the removal of the iron, a 700 ml sample was injected liquid–liquid extraction with APDC–chloroform.However, we into a water carrier stream. This was merged with continued investigating the possibility of developing a method hydrochloric acid and sodium tetrahydroborate in order to that allows the separation of interferents and determination of generate selenium hydride. The system was found to have a selenium hydride in a single on-line step in a continuous limit of detection of 0.3 ng ml-1 and an RSD of 2% for a flow system. 20 ng ml-1 selenium concentration. The application of the Flow injection (FI) HG–AAS has been reported to reduce method to geochemical standard reference samples (GSD-4, interference effects in the determination of selenium.16–18 In GSS-7, GSR-6, JB-2 and JGb-1) demonstrated that the the last decade, on-line FI separation techniques have become results were statistically indistinguishable from the certified increasingly popular for the determination of trace elements values.in different types of samples. Good results were obtained when Keywords: Selenium determination ; geochemical samples; using cation-exchange resins (on-line) in continuous flow syshydride generation inductively coupled plasma atomic emission tems for the elimination of interferents such as copper and spectrometry ; iron interference removal; anion exchange; flow nickel.19,20 Likewise, we obtained good results in the determiinjection nation of selenium in geological samples with high iron contents using a strongly acidic cation-exchange resin in a continuous flow system for the separation of iron.21 However, The HG technique in combination with ICP-AES is a commonly used method for the determination of selenium.1,2 The some difficulties arise in the application of this type of method- HG–ICP-AES technique minimizes spectral interference prob- ology to samples in which the selenium concentration is of the lems caused by other matrix components.However, there order of a few ng l-1. This is fundamentally due to the large remain other types of interferences, such as those affecting the difference between the hydrochloric acid concentration neces- HG process. Interference caused by iron is a serious problem sary for the pre-reduction (4–5 mol l-1) and that for the for the determination of selenium by HG–ICP-AES, since this retention of interferents on the resin (0.5 mol l-1) since the necessary dilutions would be incompatible with the detection interference results in the inhibition of selenium hydride formalimits required.tion.3–5 Removal of this interference has been attempted by This paper reports an FI system for the determination of addition of masking agents,6–8 although the results were not selenium in geological samples by means of HG–ICP-AES. satisfactory for high iron concentrations. This is partly due to This technique is also applicable to any type of sample.difficulties encountered in finding masking agents that are Elimination of the interference from iron is achieved by a effective under the highly acidic conditions required for HG. continuous flow matrix isolation procedure based on a micro- Indeed, even if such agents were found, the high iron concen- column of an ion-exchange resin, Dowex 1-X8 (chloride form). tration would prevent them from being effective. Wickstrøm et al.9 made an important contribution in this respect for the elimination of interferences of some transition metals, using EXPERIMENTAL masking agents in alkaline media.However, the interference Apparatus caused by iron could not be totally eliminated when its concentration was over 1000 mg l-1. Procedures for separating The measurements were made with a sequential ICP spectrometer [Baird (Bedford, MA, USA) ICP 2070]. The 1 m selenium from the matrix by means of coprecipitation with Journal of Analytical Atomic Spectrometry, April 1997, Vol. 12 (487–490) 487Czerny–Turner monochromator had a holographic grating which was pumped continuously through the microcolumn at a flow rate of 2 ml min-1 for 60 s. The microcolumn was then with 1800 grooves mm-1. The 196.026 nm spectral line was used and the measurements of the FI system were expressed regenerated by washing with 2 mol l-1 hydrochloric acid at a flow rate of 2 ml min-1 for 60 s. Following the period of as peak-height emission, which was corrected against the reagent blank.The ICP operating conditions were rf generator regeneration, the microcolumn was switched back in-line and the sampling procedure repeated, after washing the sample line power 1 kW, frequency of rf generator 40.68 MHz, plasma gas flow rate 8.5 l min-1 and observation height 15 mm (above the with distilled water. Standard solutions, requiring no iron isolation, were pumped through the injection valve, bypassing load coil). The FI system used is shown in Fig. 1. The hydride unit used was a PS Analytical (Sevenoaks, Kent, UK) hydride the microcolumn. The HG manifold was employed continuously throughout the whole procedure, independently of generator and the flow rates of the reagents were controlled by a Watson-Marlow (Falmouth, UK) 303X peristaltic pump. the matrix isolation unit. The operating conditions were established and the determination was carried out. The volatile hydrides were passed through a U-type gas–liquid phase separator and then to the plasma by means of a stream of argon.Sample injection was achieved using a Rheodyne (Cotati, CA, USA) Model 50 four-way rotatory valve. Sample RESULTS AND DISCUSSION loops of 700 ml were prepared using PTFE tubing. A microbore glass column (50 mm×3mm id) fitted with porous 25 mm Fig. 2 shows the results for the determination of selenium in glass frits was used as the resin holder; the microcolumn length synthetic samples of high iron content by means of HG–ICPcan be increased up to 100 mmwithout changes being observed AES under different operating conditions.It can be concluded in the magnitude of the signal. that although the FI system is attractive because it avoids the use of a separating system or masking agents for the elimination of interferences, it fails when the iron concentration exceeds Reagents 1000 mg l-1. As suggested by A° stro�m,23 the former improvement is probably due to the fact that the residence time of the A 0.6% m/v sodium borohydride solution(Aldrich, Milwaukee, WI, USA;>98%) was prepared in 0.5% m/v sodium hydroxide sample in the FI system is very short and well controlled.This aspect is particularly favourable for the formation of hydrogen solution and was filtered through Whatman (Maidstone, Kent, Ufilter-paper in order to remove undissolved solids. selenide which is fast, while the slower interfering reactions are suppressed. This solution was prepared freshly each day.Standard solutions of selenium were prepared by dissolution of suitable Hence for many samples, mainly in those of geological origin where the selenium content is at the trace level and iron is amounts of the oxide (99.999% pure; Johnson Matthey, Royston, Herts., UK) in small volumes of nitric acid, followed present as a major component, necessarily, a separation technique must be used. In this work, we used a microcolumn by dilution with Milli-Q purified water.The anion-exchange resin was Dowex 1-X8 (100–200 dry mesh, hydrogen form, filled with an anion-exchange resin (Dowex 1-X8) which has the capacity to retain iron (as anionic chloro complexes) and 8% cross-linkage) (Aldrich). All other solvents and reagents were analytical-reagent grade or better. not selenium when the concentration of hydrochloric acid is between 3.5 and 7 mol l-1; when this concentration is less than 3.5 mol l-1 iron is hardly retained owing to incomplete Procedure chloro complex formation. On the other hand, although the method for the separation of iron and selenium is optimum The decomposition of the reference geological materials used up to a hydrochloric acid concentration of 9 mol l-1, at was carried out by conventional acid digestion.22 After the concentrations above 7 mol l-1 some problems related to the acid digestion of 0.5 g of sample, the resulting solution was plasma stability arise which cause a decrease in precision.This placed in a closed vessel, together with 12.5 ml of 12 mol l-1 type of instability is independent of the iron concentration in hydrochloric acid and distilled water up to a volume of the solution. Consequently, a concentration of 6 mol l-1 was approximately 20 ml. The solution was then subjected to a adopted, since with this concentration, the optimum iron– reduction treatment of SeVI to SeIV (90 °C, for 1 h). The chloro complex formation is ensured24,25 and also a suitable resulting solution was transferred into a 25 ml calibrated flask acid medium is generated for the reduction of SeVI to SeIV.and diluted to volume with water. The solution was then ready One of the most important characteristics of this work is for analysis. the wide interval of acid concentration that can be used, since The microcolumn was filled with the anion-exchange resin. it permits the separation to be carried out immediately before Before use, it was conditioned by running 2 mol l-1 hydrochloric acid through the column at a flow rate of 2 ml min-1 for 2 min.The injection volume was 700 ml, the loop charge flow rate was 2 ml min-1 and the carrier flow rate was 6 ml min-1. After triplicate injections, the two-way valve was changed to line B and the iron was eluted using 2 mol l-1 nitric acid, Fig. 2 Effect of iron concentration on the response of selenium(IV) (20 ng ml-1). A, anion-exchange microcolumn and FI steps prior to Fig. 1 Schematic diagram of the instrumental set-up. Line A, sample; determination; B, only FI prior to determination; C, neither anionexchange microcolumn nor FI prior to determination. line B, HNO3 (2 mol l-1) for 60 s, then HCl (2 mol l-1) for 60 s. 488 Journal of Analytical Atomic Spectrometry, April 1997, Vol. 12Table 1 Concentrations (mg g-1±95% confidence interval, n=6) of selenium in geochemical standard reference samples Iron content Sample* Ref. Certified Found (%, m/m)† GSD-4 26 0.28±0.13 0.27±0.02 4.12 GSS-7 27 0.32±0.09 0.34±0.03 13.12 GSR-6 27 0.099±0.028 0.095±0.007 1.76 JB-2 28 0.18‡ 0.16±0.02 10.03 JGb-1 28 0.17‡ 0.18±0.02 10.61 * GSD-4, stream sediment; GSS-7 soil; GSR-6, rock; JB-2, basalt; JGb-1, gabbro.† Literature values. ‡ Recommended values. Fig. 3 Effect of the sample injection volume on the response of 20 ng ml-1 of selenium(IV). moderate amounts of Cu, Ni, Co and Zn (which are common interferents in this type of methodology), since under the conditions used they are retained as chloro complexes25,29 by performing the pre-reduction without additional conditioning the resin.of the solution. This is also advantageous compared with the use of cation-exchange resins where rigorous control of pH is essential; further, it is very difficult to make these values of acidity compatible with the detection limit of the method, since CONCLUSIONS before the pre-reduction (4–5 mol l-1 hydrochloric acid) From the results obtained it can be concluded that it is possible it is necessary to dilute the solution at least 10-fold and this to determine selenium in samples with a high iron concen- is unacceptable when trace levels of selenium are to be tration by using HG–ICP-AES with an FI method for the determined.removal of iron on-line by means of a rapid and simple ion- The range of sodium tetrahydroborate concentration where exchange procedure. Selenium can be determined precisely and maximum sensitivity for selenium determination was estab- accurately in geochemical samples using the proposed method, lished was between 0.5 and 0.75% m/v.At concentrations as demonstrated by the analysis of five fundamentally different greater than 0.75% the sensitivity decreased slightly and at geochemical standard reference materials. concentrations greater than 1.2% the plasma became unstable and easily extinguished. Variations in sodium tetrahydroborate This work was supported by the Universidad Nacional de San concentration within the above range did not have any effect Luis and the Consejo Nacional de Investigaciones Cientý�ficas on the interference from iron in the three cases shown in Fig. 2.y Te� cnicas (CONICET). A concentration of 0.6% m/v was subsequently adopted. The optimum flow rate for sample introduction into the hydride generation system was found to be 6 ml min-1. However, if the sample was run through the column at this REFERENCES rate, resin compaction occurred, with subsequent clogging of 1 Nakahara, T., Prog.Analt. At. Spectrosc., 1983, 6, 163. the column. This problem was solved by using an FI system 2 Nakahara, T., Spectrochim. Acta Rev., 1991, 14, 95. for introduction of the sample. This system allowed the separa- 3 Campbell, A. D., Pure Appl. Chem., 1992, 64, 227. tion of iron in the ion-exchange column at a flow rate of 4 Bax, D., Agterdenbos, J., Worrell, E., and Kolmer Beneken, J., 2 ml min-1; subsequently the sample-containing loop was Spectrochim.Acta, Part B, 1988, 43, 1349. injected into the carrier (water) at 6 ml min-1. 5 Qiu, D.-R., T rAC, T rends Anal. Chem. (Pers. Ed), 1995, 14, 76. 6 Belcher, R., Bogdanski, S. L., Henden, E., and Townshend, A., Each measurement represents the mean of results from three Analyst, 1975, 100, 522. 700 ml injections, after which the resin was regenerated by 7 Kirkbright, G., and Taddia, M., Anal. Chim. Acta, 1978, 100, 145.running 2 mol l-1 nitric acid for 60 s and then 2 mol l-1 8 Uggerud, H., and Lund, W., J. Anal. At. Spectrom., 1995, 10, 405. hydrochloric acid for 60 s. A 700 ml injection volume was 9 Wickstrøm, T., Lund, W., and Bye, R., J. Anal. At. Spectrom., chosen because it is the minimum volume with which maximum 1995, 10, 803. sensitivity is obtained (Fig. 3). 10 Bedard, M., and Kerbyson, J., Can. J. Spectrosc., 1976, 21, 64. 11 Jones, J. W., Capar, S. G., and O’Haver, T.C., Analyst, 1982, For the purpose of checking the proposed method for 107, 353. determining selenium, a recovery study was carried out on 12 Narasaky, H., Anal. Sci., 1986, 2, 141. synthetic samples; several progressive additions of selenium 13 Hershey, J., and Keliher, P., Spectrochim. Acta, Part B, 1989, (for a concentration level of selenium of 20 ng ml-1) were 44, 329. performed in order to establish a maximum concentration of 14 Thompson, M., Pahlavanpour, B., Walton, S.J., and Kirkbright, 80 ng ml-1. The recoveries were found to be in the range G. F., Analyst, 1978, 103, 705. 15 Martinez, L., Baucells, M., Pelfort, E., and Roura, M., Fresenius’ 98.5–99.7%. The preon of the system was 2% RSD, based J. Anal. Chem., 1996, 354, 126. on 10 injections of a 20 ng ml-1 aqueous selenium standard 16 Yamamoto, M., Yasuda, M., and Yamamoto, Y., Anal. Chem.,1985, solution, and the detection limit was 0.3 ng ml-1. The system 57, 1382. was calibrated using aqueous standards (2–30 ng ml-1); the 17 Chan, C.C. Y., Anal. Chem., 1985, 57, 1482. calibration graph was linear up to at least 30 ng ml-1 of 18 Fang, Z., Xu, S., Wang, X., and Zhang, S., Anal. Chim. Acta, 1986, selenium(IV) and a correlation coefficient of 0.9995 was 179, 325. 19 Offley, S. G., Seare, N. J., Tyson, J. F., and Kibble, H. A. B., obtained. J. Anal. At. Spectrom., 1991, 6, 133. The results obtained when the method was applied to the 20 Riby, P. G., Haswell, S. J., and Grzeskowiak, R., J. Anal. At. determination of selenium in geochemical standard reference Spectrom., 1989, 4, 181. samples of known composition are shown in Table 1. 21 Martinez, L., Baucells, M., Pelfort, E., Roura, M., and Olsina, R., The proposed methodology can be applied to routine analy- Fresenius’ J. Anal. Chem., in the press. ses without taking special precautions. Further, it can be 22 Subramanian, K. S., Fresenius’ Z. Anal. Chem., 1981, 305, 382. 23 A° stro�m, O., Anal. Chem., 1982, 54, 190. satisfactorily applied to the analysis of samples containing Journal of Analytical Atomic Spectrometry, April 1997, Vol. 12 48924 Kolthoff, Y., and Elving, P., T reatise on Analytical Chemistry, 28 Ando, A., Mita, N., and Terashima, S., Geostand. Newsl., 1987, 11, 159. Part II, Wiley, New York, 1962, vol. 2, p. 267. 25 Vijan, P., and Leung, D., Anal. Chim. Acta, 1980, 120, 141. 29 Fukuda, M., Hayashibe, Y., and Sayama, Y., Anal. Sci., 1995, 11, 13. 26 Xie, X., Yan, M., Li, L., and Shen, H., Geostand. Newsl., 1985, 9, 83. 27 Xie, X., Yan, M., Wang, C., and Shen, H., Geostand. Newsl., 1989, 13, 83. Paper 6/06884G Received October 8, 1996 Accepted January 2, 1997 490 Journal of Analytical Atomic Spectrometry, April 1997, Vol.

ISSN:0267-9477    

DOI:10.1039/a606884g

出版商:RSC    

年代:1997

数据来源: RSC

摘要:

INTER-LABORATORY NOTE Determination of Cadmium in Very Low Concentration Urine Samples by Electrothermal Atomic Absorption Spectrometry JAMES P. SNELLa , SUSANNE SANDBERGb AND WOLFGANG FRECH*a aDepartment of Analytical Chemistry, Umea° University, S-901 87 Umea° , Sweden bDepartment of Occupational and Environmental Medicine, Umea° University, S-901 87 Umea° , Sweden Urine samples were collected that contained concentrations of A number of previously published papers outline method Cd between 0.05 and 0.5 mg l-1.When using ETAAS for the development for this analyte–matrix combination. For analysis of these samples, spectral interference is present that example, Ediger and Coleman,5 Yin et al.6 and Smeyerscan obscure low masses of Cd and deteriorate the precision Verbeke et al.7 have demonstrated the benefits of using and accuracy of the analysis. Instrumental advances such as NH4NO3 modifier and for the last two, in combination with electrodeless discharge lamps, the use of echelle Pd.Fraile et al.8 used a controlled atomization temperature to monochromators, solid-state detectors and end-capped separate temporally the vaporization of Cd and sodium chlorspatially isothermal atomizers have been combined to reduce ide. Our study included specimens from children and diabetics. detection limits. Spatially isothermal atomizers provided with As the physiological dilution of urine may be greater for these end-caps and modified contact cones were found to double the two groups, a significant number of samples contained much sensitivity for Cd, compared with a standard atomizer lower concentrations of Cd than those previously reported2 or configuration.A combined chemical modifier of Pd and used for method optimization.6,8–10 These papers give detection NH4NO3 can reduce interference from molecular alkali limits from 0.05 mg l-1,10 with typical concentrations between halides, on atomization of Cd. Samples were decomposed with 1 and 10 mg l-1 Cd.In our case, without careful optimization boiling HNO3 to destroy carbon compounds and ensure of procedures the very low Cd signals of the order of 0.01 homogeneity. Calibration was by standard additions to ensure absorbance could be disturbed by background absorption, accuracy in the determination; however, standard calibration which is not sufficiently reduced by chemical modification or may be used with a deterioration in accuracy of about 6%.For the use of a spatially isothermal transversely heated graphite a sample containing 0.1 mg l-1 Cd, the method including atomizer (THGA). sample pre-treatment gives a precision of about 15%. The Instrumental noise levels for the technique have been con- accuracy of the method was established by standard additions tinuously lowered by advances in instrumental AAS design. to samples and the analysis of a urine reference material; the Electrodeless discharge lamps are known to increase light instrumental detection limit is 0.008 mg l-1 Cd in urine.intensity. Echelle spectrometers and solid-state detectors were Keywords: T ransversely heated graphite atomizer; end-capped recently introduced into commercial instruments11–13 (for a tube; modified contact cone; matrix interference; cadmium; review see ref. 14). For the spectral resolution required, the urine echelle spectrometer has a more intense light flux than a conventional Czerny–Turner monochromator.At the Cd wave- Cadmium is a toxic element at relatively low concentrations length of 228.8 nm, the solid-state detector used has improved for which urine provides a reliable guide to the body’s load.1–4 quantum efficiency over photomultiplier tubes.13 In addition, Urine is also a convenient sample to collect. The accurate the solid state detector gives less noise than photomultiplier determination of Cd in urine is now possible, using modern tubes at the measured light intensities typical of absorption instrumentation, at such low levels that most samples are spectrometry.13 measurable. This allows the monitoring of the Cd load on For these state-of-the-art instruments, the modification of people who excrete low levels of the metal. The subjects of our furnace geometry can further improve sensitivities.To increase study are residents of Sweden, where natural exposure to Cd analyte peak area for a given mass, THGAs with end-capped is low, and some are children who excrete more diluted urine furnaces are used to increase analyte residence time in the than adults and have had less time to accumulate Cd.In furnace14–18 and in-house modified contact cones19 can further addition, the subjects are diabetics who may also excrete more increase residence time at lower atomization temperatures. As diluted urine than people with normal pancreatic function. It well as limiting the rate of diffusion, a reduced atomization is known that diabetes may lead to nephropathy and that this temperature is suitable for Cd determination as it achieves a can reduce the kidneys’ tolerance to the toxic effects of Cd.It partial temporal separation of Cd atoms from sodium chloride is therefore particularly important to determine Cd at low diatomic vapour. concentrations for correlation to the health of diabetics. Our study was performed using these improved conditions For the determination of Cd in urine, ETAAS is the chosen to lower detection limits and adopted methods of validation. method of many analysts because of its relative simplicity, the Standard additions were made to samples to give the recovery possibility of automation and the low instrumental detection of the preparation procedure.Sample digestion was repeated limits given. With this technique, however, matrix interferences, to check in-batch precision. Reference materials were used to particularly from various concentrations of sodium chloride, remain that deteriorate accuracy and precision.check method accuracy. Journal of Analytical Atomic Spectrometry, April 1997, Vol. 12 (491–494) 491EXPERIMENTAL injection port diameter of 1.8 mm. Two sets of graphite contact cones were used: the standard Perkin-Elmer models and modi- Reagents and Sample Preparation fied cones with a groove cut around the injection port17 (Fig. 1). Recovery tests were performed by adding 500 pg of Cd to Sample aliquots of 10 ml were used; these were introduced with 10 ml of chemical modifier and 10 ml of either diluent or approximately 2.5 g of urine, before acid decomposition.Calibration graphs were constructed using a digested urine standard containing 0.5% m/m HNO3. The chemical modifier used was 6 mg of Pd, and 0.5 mg of NH4NO3 dissolved in sample with a Cd concentration below the instrumental detection limit. The autosampler of the spectrometer was pro- 0.9% m/m HNO3.Cadmium standards were prepared daily from a 10 mg l-1 stock solution and contained 0.5% m/m grammed to pipette various volumes of Cd standard and diluent with 10 ml of the urine sample. Sample measurements HNO3 . All sample containers were new and acid-cleaned prior to were made consecutively. The instrumental detection limits for Cd in urine are given use. Samples were collected in 100 ml poly(ethene) bottles. Care was taken to avoid contamination at all stages3,4,20 and in Table 2 and are defined as 3s of ten replicates of a low concentration sample, the Cd content of which is then deter- volunteers were requested not to smoke before collection or touch the inner surfaces of the container.Midstream, morning mined by standard additions. urine was voided directly into the container, the amount recorded and the sample frozen. On receipt the samples were RESULTS AND DISCUSSION thawed, transferred into 20 ml poly(propene) scintillation tubes (Beckman Instruments, Fullerton, CA, USA) and acidified to Sample Preparation below pH 2 with HNO3 to prevent analyte loss.9 The samples Samples were decomposed by heating with HNO3 to dissolve were then frozen at -20 °C until determination to minimize solid particles and oxidize organic compounds, which was not deterioration.achieved by adding acid to a cold sample. Once a sample had Solutions were prepared from water purified with a Milli-Q been evaporated to near dryness, it was reconstituted with system (Millipore, Bedford, MA, USA) and HNO3 purified water to about 0.8 times its original mass.Further concen- was with a sub-boiling still (Acidest, Heraeus Quartzschmelze, tration is not desirable as a 10 ml injection into the graphite Hanau, Germany). Palladium modifier (pro analysi quality) furnace gives an optimum signal-to-background absorbance. and NH4NO3 (ACS reagent quality) were supplied by Merck While larger injection volumes will increase the analyte signal, (Darmstadt, Germany).Seronorm reference material used was with the matrix present the gain in signal may be less than the low values urine, batch No. 101021 (Nycomed Pharma, that predicted for the greater analyte mass. The accompanying Oslo, Norway), with a Cd concentration of 0.35 mg l-1. This was reconstituted according to the manufacturer’s instructions and prepared as a normal sample. Samples were prepared in laminar flow clean benches (Ultramare, Stockholm, Sweden), which provide a Class 100 working environment. Aliquots of 2–5 g of thawed urine samples were weighed into quartz conical flasks and about three times the sample mass of HNO3 was added.The mixture was boiled until the flasks were almost dry; they were left to cool and about 0.8 times the sample mass of water was added and weighed. The mass ratio of the original and digested samples was taken to calculate the sample concentration. This procedure avoids the use of volumetric equipment, which could increase the risk of sample contamination.The densities of the digests were measured independently for some samples and were typically 0–2% higher than those of the original samples. Therefore, the error introduced by measuring sample mass rather than volume is negligible. Prepared samples were trans- Fig. 1 Diagram of the modified contact cones after Hadgu and ferred into 20 ml poly(propene) scintillation tubes and stored Frech.19 in the dark at 4 °C until use.Table 2 Performance comparison of furnace adaptations Instrumentation Atomizer Characteristic Detection limit in A Perkin-Elmer SIMAA 6000 spectrometer (Bodenseewerk configuration Light energy* mass/pg† urine/mg l-1‡ Perkin-Elmer, U� berlingen, Germany) was run in single element Standard cone 455 2.01 0.015 mode with a Perkin-Elmer Series 2 electrodeless discharge Open furnace lamp for Cd. The furnace temperature programme is given in Standard cone 431 1.20 0.013 End-capped Table 1.Baseline offset correction was applied for 2 s and furnace taken 3 s before atomization. Graphite tubes were open or Modified cone 452 1.28 0.018 fitted with end-caps (3 mm aperture, 2 mm length) and had an Open furnace Modified cone 425 0.89 0.008 End-capped Table 1 Furnace temperature programme furnace Temperature/°C Ramp time/s Hold time/s * Transmitted light energy in arbitrary units, recorded by the spectrometer. 90 1 10 † Characteristic mass was calculated by standard additions to a 120 90 30 urine sample, with a natural Cd content below the detection limit. 600 15 30 ‡ Detection limits are 3s of ten replicates of a urine sample containing 1300 0 6 0.059 mg l-1 Cd. 2300 1 2 492 Journal of Analytical Atomic Spectrometry, April 1997, Vol. 12rise in background absorbance will increase measurement error light attenuation by background species is reduced as these are now vaporized over a longer period of time and, more and thus deteriorate the detection limit.21 significantly, an increase in sensitivity is observed at lower temperatures as the atomic residence time is increased.The Drying latter effect is significantly enhanced by using an end-capped tube with grooved contact cones (Table 2). For a 30 ml injection volume, a slow temperature ramp as part With the standard design of the THGA, the base of the of a 2 min drying step is necessary to preserve the graphite injection port of the cone is in close proximity (0.5 mm) to the tube, which is prone to corrosion around the injection port tube.When a stream of inert sheath gas is passed through this under rapid drying conditions. space, it creates a volume of reduced pressure around the injection port of the tube. The difference in pressure forces Pyrolysis gases through the injection port of the tube and thus permits efficient removal of vapour during drying. On the other hand, Because Cd is a fairly volatile element, the possibility of with this type of cone, analyte atoms are removed more quickly complete removal of matrix constituents during pyrolysis is due to this convective flow, particularly at low atomization limited.In complex sample matrices such as urine, careful temperatures.19 For the grooved cone, the space around the optimization and temperature control of pre-treatment protoinjection port of the tube eliminates convective flows,19 thus cols, as well as the use of a suitable modifier, are therefore increasing sensitivity.required. Here, a mixture of Pd and NH4NO3 was added to The improvement in sensitivity gained by the removal of the sample as NH4NO3 facilitates removal of chlorides6,7 and convective flow is greatest when using end-capped tubes. While Pd is known to ensure the stability of Cd during pre-treatment end-capped tubes increase sensitivity by reducing the rate of at 600 °C. It was also important to apply a fairly slow ramp analyte diffusion, this increase is further enhanced by virtually during the pyrolysis step to prevent sputtering of the sample removing convection.constituents. After lower pyrolysis temperatures, the shape of the background absorbance during the atomization of Cd typically shows signals in two time domains (Fig. 2). By 1 teff = 1 tD +1 tC decomposing the urine in HNO3, the background obtained during the first 2 s is decreased significantly, indicating that Here teff, tD and tC are the effective, diffusional and convective this absorbance originates from organic compounds.mean atomic residence times, respectively.18 As tC tends to The spectrum for the later background signal correlated in high values, the residence time of atoms in the atomizer part to that of diatomic sodium chloride.6,22 This shows that becomes entirely dependent on diffusion. the reaction of the NH4NO3 modifier with the sodium chloride However, with the SIMAA 6000 instrument, end-capped in urine in the presence of Pd is not complete at the temperature furnaces reduce the light throughput, which can increase noise. used here.This agrees with the results of Yin et al.6 For The size of the end-caps, therefore, has to balance the increase optimum S/N, it was essential to keep the background as low in sensitivity with the increase in noise. Table 2 shows the as possible and to maintain the set pre-treatment temperature difference in light throughput and sensitivity for standard open accurately.There is always some uncertainty and variation and end-capped furnaces. While light throughput is not signifi- between the set and real pre-treatment temperatures. To avoid cantly affected, the detection limits from the two furnace types losses of Cd, the temperature chosen here was about 100 °C demonstrate the benefits of end-caps for this analysis. lower than the maximum permissible. While the detection limit can be reduced by improving the sensitivity of measurements, it has been shown14,21 that longer integrationtimes, a consequence of lower atomization tempera- Atomisation tures, will increase noise and thus deteriorate precision.The atomization temperature of 1300 °C is 100 °C lower than However, at higher atomization temperatures, the matrix the recommended value and gives two benefits. The maximum vapour causes light attenuation that coincides with the analyte absorbance peak.23 The lowered light transmission through the furnace will deteriorate the stability of the analyte baseline, increasing noise.L’vov et al.21 demonstrated that, by using a longer time for baseline offset correction (BOC) as well as a shorter time for absorbance peak integration, the S/N of measurements is improved and detection limitscan be considerably reduced. While increasing the BOC time may be adntageous, this setting cannot be changed on the SIMAA 6000 instrument. In summary, the possible deterioration by noise of the instrumental detection limit for Cd in urine, given by reduced light throughput due to end-caps and a longer atomization time, is more than compensated for by the increase in sensitivity brought about by increasing the atomic residence time.Method Validation To check the accuracy of the sample preparation method, recovery tests were performed by adding Cd to samples before acid digestion. Three samples with different concentrations were chosen for the test and three digestions were made of one sample (Table 3).The recovery values given are reasonably close to 100%, suggesting that Cd is completely recovered. Fig. 2 Zeeman corrected atomic absorbance and background signals The deviation seen is determined by instrumental precision at for Cd in Seronorm urine. Ashing temperature, 300 °C; atomization temperature, 1300 °C. low concentration, which will be discussed later. Journal of Analytical Atomic Spectrometry, April 1997, Vol. 12 493Table 3 Recovery tests Sample+addition/ Sample/mg l-1 mg l-1 Recovery (%) n 0.101 0.193 95.7±14.6 3 0.144 0.227 84.0 1 0.161 0.253 96.8 1 * Concentration values were determined by standard additions calibration; the number of repetitions is given in the final column and the uncertainty for the first sample is the standard deviation of the three sample digestions. Recovery tests were made by adding approximately 250 pg of Cd to approximately 2.5 g of each urine sample before digestion in acid.The precision of the digestion method was checked by repeating the digestion of five different samples (Table 4). The standard deviation of the two observations was calculated for each sample and the square root of the sum of the variances is given. The value of 0.003 mg l-1 is below the instrumental Fig. 3 Atomic peak area for masses of Cd introduced with urine. detection limit, which suggests that the digestion process and Percentage relative standard deviation for three replicates is given for each point.instrumental determination is repeatable. Seronorm reference material was used to check the accuracy The authors thank Bodenseewerk Perkin-Elmer for the loan of determination. It was prepared according to the manufacof the SIMAA 6000 spectrometer, Yngvar Thomassen of the turer’s instructions, then digested as a normal sample and Norwegian Institute of Occupational Health for supplying the analysed with a batch of samples (Table 5).On six occasions, Seronorm reference material and for valuable discussions and Seronorm was supplied without the analyst’s knowledge and Karin Olsson of Umea° University for assistance with labora- labelled as a normal sample. Both mean values given are tory work. within 5% of the recommended concentration, which suggests that determination is accurate at this level. Calibration by standard additions was used for the urine REFERENCES samples in this study; however, standard calibration may be 1 Stoeppler, M., Spectrochim.Acta, Part B, 1983, 38, 1559. used without a significant deterioration in accuracy. The 2 Herber, R. F. M., Stoeppler, M., and Tonks, D. B., Fresenius’ deviation of the characteristic mass between 20 samples on J. Anal. Chem., 1990, 338, 279. one day was found to be 6%, which is only slightly higher 3 Nordberg, G. F., and Nordberg, M., in Biological Monitoring of than that expected from the instrumental deviation.One T oxic Metals, ed. Clarkson, T. W., Friberg, L., Nordberg, G. F., and Sager, P. R., Plenum Press, London, 1988, pp. 151–168. addition of 0.2 mg l-1 is normally made to the sample. The 4 Diamond, G. L., in Biological Monitoring of T oxic Metals, ed. absorbance-to-mass of analyte relationship must be linear for Clarkson, T. W., Friberg, L., Nordberg, G. F., and Sager, P. R., this method of calibration. As a test for linearity, different Plenum Press, London, 1988, pp. 515–529. masses of Cd were pipetted into the furnace with a urine 5 Ediger, R. D., and Coleman, R. L., At. Absorpt. Newsl., 1973, 12, 3. sample showing no detectable Cd. Fig. 3 is the calibration 6 Yin, X., Schlemmer, G., and Welz, B., Anal. Chem., 1987, 59, 1462. produced from three replicates of ten standards up to 7 Smeyers-Verbeke, J., Yang, Q., Penninckx, W., and Vandervoort, F., J. Anal. At. Spectrom., 1990, 5, 393. 0.5 mg l-1. Evidently, the absorbance-to-mass relationship is 8 Fraile, R., de Benzo, Z.A., and Velosa, M., Fresenius’ J. Anal. linear, even below 0.010 absorbance, and gives a correlation Chem., 1992, 343, 319. coefficient of 0.999. To demonstrate instrumental precision, the 9 Halls, D. J., Black, M. M., Fell, G. S., and Ottaway, J. M., J. Anal. relative standard deviation is included on the plot. At. Spectrom., 1987, 2, 305. 10 Dube, P., Krause, C., and Windmu�ller L., Analyst, 1989, 114, 1249. 11 Harnly, J. M., and Radziuk, B., J.Anal. At. Spectrom., 1995, 10, 197. Table 4 Repeatability of digestions 12 Radziuk, B., Ro�del, G., Stenz, H., Becker-Ross, H., and Florek, S., J. Anal. At. Spectrom., 1995, 10, 127. Mean/mg l-1 ±/mg l-1 13 Radziuk, B., Ro�del, G., Zeiher, M., Mizuno, S., and Yamamoto, K., 0.039 0.001 J. Anal. At. Spectrom., 1995, 10, 415. 0.046 0.001 14 Harnly, J. M., Fresenius’ J. Anal. Chem., 1996, 355, 501. 0.096 0.006 15 Hadgu, N., Ohlsson, K. E. A., and Frech, W., Spectrochim. Acta, 0.123 0.001 Part B, 1995, 50, 1077. 0.130 0.001 16 Hadgu, N., Ohlsson, K. E. A., and Frech, W., Spectrochim. Acta, Part B, 1996, 51, 1081. Mean±/mg l-1 0.003 17 Hadgu, N., and Frech, W., Spectrochim. Acta, Part B, 1994, 49, 445. 18 L’vov, B. V., and Frech, W., Spectrochim. Acta, Part B, 1993, * Two aliquots of five different samples were digested and their Cd 48, 425. contents were determined. The mean of two observations is shown; 19 Hadgu, N., and Frech, W., Spectrochim. Acta, Part B, in the press. mean uncertainty is the square root of the sum of the variances. 20 Cornelis, R., Heinzow, B., Herber, R. F. M., Molin Christensten, J., Paulsen, O. M., Sabbioni, E., Templeton, D. M., Thomassen, Y., Vahter, M., and Vesterberg, O., Pure Appl. Chem., 1995, 67, 1575. Table 5 Quality control with Seronorm reference material* 21 L’vov, B. V., Polzik, L. K., Borodin, A. V., Dyakov, A. O., and Novichkhin, A. V., J. Anal. At. Spectrom., 1995, 10, 703. Run type Mean/mg l-1 ±/mg l-1 n 22 Culver, B. R., and Surles, T., Anal. Chem., 1975, 47, 920. Known 0.364 0.058 16 23 L’vov, B. V., Polzik, L. K., and Fedorov, P. N., Spectrochim. Acta, Blind 0.343 0.032 6 Part B, 1992, 47, 1411. * Results for Seronorm Batch 101021; recommended Cd concen- Paper 6/06150H tration 0.35 mg l-1. Shown are the mean, standard deviation and Received September 6, 1996 number of determinations. Determinations were made on different days; blind samples were supplied without the operator’s knowledge. Accepted January 20, 1997 494 Journal of Analytical Atomic Spectrometry, April 1997, V

ISSN:0267-9477    

DOI:10.1039/a606150h

出版商:RSC    

年代:1997

数据来源: RSC