摘要:

Determination of arsenic in organic solvents and wines using microscale flow injection inductively coupled plasma mass spectrometry Sunanta Wangkarn and Spiros A. Pergantis* Department of Chemistry, Birkbeck College, University of London, Gordon House, 29 Gordon Square, London, UK WC1H 0PP. E-mail: s.pergantis@chemistry.bbk.ac.uk Received 4th January 1999, Accepted 24th February 1999 The presence of carbon-containing substances can enhance As signals and elevate background levels observed in inductively coupled plasma mass spectrometry (ICP-MS), thus causing potential errors in the quantification of As.Most approaches for eliminating these interferences are tedious and time consuming and increase the risk of sample contamination and analyte loss. A microscale flow injection (mFI) system employing a microconcentric nebulizer (MCN) for eYcient sample introduction into the ICP-MS system was used to reduce signal enhancement eVects. Also investigated was the suitability of several elements (Se, Y, In and Sb) to be used as internal standards for As in samples containing organic solvents.The mFI-ICP-MS system developed in this study was shown to reduce the signal enhancement caused by organic solvents by a factor of 2–3 compared with a conventional FI-ICP-MS system. Sample volumes of 0.2, 0.5 or 1.0 ml were injected into the mFI-ICP-MS system at carrier flow rates ranging from 50 to 200 ml min-1. Response profiles obtained following the injection of 1.0 ml solutions containing 20 pg ml-1 As at carrier flows of 50, 100 or 200 ml min-1, allowed for throughputs of approximately 80, 140 or 180 samples per hour, respectively. The relative standard deviation (%RSD) of the transient signals, determined over a 20 min period at the previously mentioned flow rates, ranged from 2 to 5%.The calculated absolute limit of detection of the mFI-ICP-MS system, ranging from 25 to 59 fg As, demonstrates the method’s potential for determining As at ultratrace levels.The mFI-ICP-MS method was subsequently used to determine As in red and white wines. Diluted wine samples (1+1 dilution in de-ionised water) were analysed without any further sample preparation. When using In as the internal standard the average recovery of As was found to be 100±2%. The concentration of As was determined to be between 7 and 13 pg ml-1 for all wines examined. These values are significantly lower than the reported maximum permissible concentration limits for As in wine.matrix-matched standards, however, is only possible if the Introduction exact composition of the sample matrix is known and can be Arsenic, a well-known poison and suspected carcinogen, is prepared easily. Most often this is not the case, especially found in its various forms in food products, and biological when analysing waste materials containing unknown amounts and environmental materials, at trace and ultratrace levels.1–4 of organic solvents.On the other hand, employing an internal Despite the fact that As is ubiquitous in the environment, little standard to correct for As signal variation is not always a is known about its chronic sub-lethal eVects. Since considerable straightforward procedure. This is mainly because As and the interest is currently directed towards understanding these element used as the internal standard do not always respond eVects it is of particular importance to develop robust analyt- identically towards matrix interferences. Selecting an approical techniques suitable for providing high sensitivity, accuracy, priate internal standard becomes even more diYcult when precision, and sample throughput.Inductively coupled plasma samples containing high levels of dissolved solids cause mass spectrometry (ICP-MS), one of the most sensitive tech- additional interferences. Other approaches for reducing or niques for elemental determinations, has been used extensively even eliminating non-spectral interferences may include the for determining As in a wide range of materials.1,2,5–9 partial removal of organic components from the sample.This Unfortunately, however, both spectral and non-spectral inter- approach is time consuming and tedious, and also adds to the ferences have been reported for this element.6,8,10–12 One overall complexity of the analysis by increasing the risk of particular non-spectral interference originates from carbon- sample contamination and analyte loss.To emphasise further containing substances, which have been reported to cause the eVects of introducing large amounts of volatile solvents pronounced eVects on the intensity of As signals in into an ICP, it should be mentioned that this can cause plasma ICP-MS.8,11–13 Although the mechanism responsible for the instability because of energy withdrawal, formation of carbonsignal variation is not fully understood, it is believed to involve containing molecular species and carbon deposition on the changes in plasma conditions which cause modification of the torch and sample cones.Prevention of such eVects requires elemental ionisation equilibrium over a limited range of ionis- specialised equipment suitable for spray desolvation, addition ation energies.11 In practice, this can cause errors in the of oxygen gas, and spray chamber chilling.14–16 In many cases quantification of As in samples containing elevated levels of the cost and complexity of additional equipment has restricted organic substances.their usefulness and general applicability. Measures taken to eliminate quantification errors may The purpose of this study was to investigate the eVects of include using matrix-matched standards or internal standards carbon-containing solvents, i.e., methanol and isopropyl alcohol, on As signals obtained using ICP-MS. In addition, a capable of compensating for As signal variations.The use of J. Anal. At. Spectrom., 1999, 14, 657–662 657microscale flow injection (mFI)-ICP-MS technique was devel- 0.1% v/v nitric acid. Aqueous calibration standards were also prepared to contain the same amount of acid. All wines used oped and evaluated for overcoming the aforementioned eVects. The technique is based on the eYcient introduction of 0.2, 0.5 in this study were commercially available. All wine samples were filtered, through a 0.45 mm pore size membrane, prior or 1.0 ml samples into the ICP-MS system via a commercially available microconcentric nebuliser (MCN).The mFI-ICP-MS to injection. technique was characterised initially and subsequently evaluated for its suitability for determining As in organic solvents Results and discussion and wine samples. Performance characteristics of the mFI-ICP-MS system Experimental Responses obtained for As by using mFI-ICP-MS were evaluated at carrier flow rates of 50, 100 and 200 ml min-1.A Instrumentation typical response profile obtained following consecutive 0.5 ml injections of solutions containing 20 pg ml-1 As is presented A VG PlasmaQuad II ICP-MS instrument (VG Elemental, Winsford, Cheshire, UK) was used throughout this study. The in Fig. 1. It is evident from this profile that the resulting narrow peaks allow for high sample throughput. More specifi- quadrupole mass filter was operated in the single-ion monitoring mode (m/z 75) when monitoring As+ and in the peak cally, when running the carrier at a flow of 50, 100 or 200 ml min-1 and injecting 1.0 ml samples it was possible to jump mode for multi-element detection. Multi-element detection was applied when Se+ (m/z 77), Y+ (m/z 89), In+ make 80, 140 or 180 injections per hour, respectively.The relative standard deviations (%RSD) of the transient signals (m/z 115) and Sb+ (m/z 121) were used as internal standards. The instrument operating conditions were optimised for maxi- determined over a 20 min period, at the previously mentioned flow rates, ranged from 2 to 5%.The system dispersion was mum As+ signal (m/z 75). Typical operating conditions and data acquisition parameters are given in Table 1. determined to be 10. Dispersion was calculated as the ratio of the net As signal obtained following continuous sample intro- Throughout this study the MCN-100 (CETAC Technologies, Omaha, NE, USA) was used. It has been demonstrated that duction to the peak height of the As transient signal obtained following mFI using a 1.0 ml internal chamber injector.this nebulizer allows for the eYcient introduction of analytes into the ICP at low ml min-1 flow rates.17,18 The MCN was The limits of detection (LOD) obtained for As using the MCN-based mFI-ICP-MS system are summarised in Table 2. mounted directly onto a standard double-pass water-cooled spray chamber without the need for any modifications. A All reported LOD were determined at a carrier flow rate of 200 ml min-1.The concentration LOD (cLOD) obtained for reciprocating pump (Shimadzu, LC-5A, Kyoto, Japan) was used to deliver carrier liquid into the ICP-MS system at flow 0.2 and 0.5 ml samples were, on average, a factor of 15 higher than those typically obtained for 20 ml samples. However, the rates between 50 and 200 ml min-1. Samples were introduced into the carrier stream via an internal chamber micro-injector absolute LOD (aLOD) obtained after injecting 0.2 and 0.5 ml samples were, on average, 6 and 2.5 times, respectively, better (Rheodyne 7520, Cotati, CA, USA).Internal chambers of 0.2, 0.5 and 1.0 ml were used in this study. The mFI-ICP-MS than those obtained for 20 ml samples. The aLOD, ranging from 25 to 59 fg As, demonstrate the potential of mFI-ICP-MS response was optimised for maximum signal at m/z 75 using a 190 ml loop fitted onto a Rheodyne 7125 injector. This for determining As at ultratrace levels.The performance characteristics, i.e., LOD, %RSD and sample volume was suYcient to provide a quasi-continuous signal that lasted long enough for signal optimisation. Reagents and samples Inorganic As, Se, Y, In and Sb standard solutions were prepared by diluting individual standards (VHG Lab, Manchester, UK) containing 10 000 mg ml-1 of each element in de-ionised water acidified to 0.05% v/v nitric acid (69% m/m, AnalaR, BDH, Poole, Dorset, UK). Methanol (99.8%, BDH) and isopropyl alcohol (99.7%, BDH) were used to provide a convenient carbon source that was miscible with water.Individual solutions containing 20 pg ml-1 of As and 5–100% v/v methanol or 3.5–100% v/v isopropyl alcohol were prepared and used for investigating the eVect of carboncontaining substances on As signals in ICP-MS. These solutions were also used to test for As recoveries. Wine samples Fig. 1 Peak profiles for 0.5 ml injections (n=10) of 20 pg ml-1 As; were diluted 1+1 with de-ionised water and acidified to carrier flow rate was 200 ml min-1.Table 1 Typical ICP-MS operating conditions Table 2 Limits of detectiona obtained for As by using MCN-based FI-ICP-MS Argon coolant flow 12.5–14.5 l min-1 Argon auxiliary flow 1.2–2.0 l min-1 Injected volume/ Concentration LODb/ Absolute LODb/ Argon nebulizer flow 0.76–0.80 l min-1 ml fgml-1 fg Rf forward power 1350 W Reflected power <5 W 0.2 125c 25c Nebulizer Microconcentric nebulizer (MCN)-100 0.5 118d 59d Spray chamber Water-cooled (5 °C) Scott-type (double-pass) 20 8e 160e Sample cone Nickel; aperture diameter 1.0 mm Skimmer cone Nickel; aperture diameter 0.75 mm aCalculated as three times the background standard deviation. bAll LOD were determined at 200 ml min-1 carrier flow.cCalculation based mFI carrier flow rate 50–200 ml min-1 Data acquisition Peak jump or single-ion monitoring (SIM) on the peak area corresponding to 50 pg ml-1 As. dCalculation based on the peak area corresponding to 10 pg ml-1 As.eCalculation mode Dwell time 10.24 ms based on the peak area corresponding to 2 pg ml-1 As. 658 J. Anal. At. Spectrom., 1999, 14, 657–662sample throughput, of the MCN-based mFI-ICP-MS system observed for the 20 ml sample we propose that the presence of carbon in the plasma causes accumulative eVects, which most compare well with those previously reported for a highe Yciency nebulizer (HEN)-based mFI-ICP-MS system.5 likely occur because of the long washout times required to remove the carbon from the plasma.In studies reporting on Judging from the systems’ analytical figures of merit both the MCN and the HEN perform equally well when used for the development of liquid chromatographic methods for separating As species it was mentioned that even when the introduc- determining As at ultratrace levels. However, the obvious advantage of the MCN, over the HEN, is that it does not tion of organic solvent into the plasma was discontinued, a relatively long period of time was required before the ICP-MS require a high-pressure nebulizer gas line and a mass flow controller capable of withstanding approximately 175 psi back- sensitivity returned to its pre-solvent level.9,20 When the absolute amount of carbon injected into the pressure.The MCN can operate eYciently with a mass flow controller that can deliver nebulizer gas at back-pressures of ICP-MS system is plotted against the recorded As signal intensity, a continuous increase in As signal intensity with about 50 psi.This type of mass flow controller is a standard component of almost all commercial ICP-MS instruments. It increasing amount of carbon injected is observed [Fig. 2(c) and (d)]. A maximum occurs when approximately 1 mg of carbon should, however, be mentioned that the additional hardware required for the eYcient operation of the HEN is no longer is injected; signals start to decrease when higher amounts of carbon are aspirated.The fact that the signal intensity for In required for a modified version of the HEN; the direct injection HEN (DI-HEN).19 The DI-HEN sits directly inside the ICP and Y also starts to decrease at these levels of organics is a strong indication that plasma overloading is occurring. torch and maintains its eYciency even at low nebulizer gas flows, #0.25 l min-1 compared with 1.0 l min-1 required typi- Together these findings oppose the hypothesis that signal enhancements originate solely from increased nebulization cally by most other types of nebulizers.The DI-HEN has been reported to provide excellent cLOD for As (17 fg ml-1) in the eYciencies resulting from organic solvents present in the aspirated solution. If signal enhancements were caused exclus- continuous flow mode; however, aLOD for As in the FI mode have not yet been published.19 ively because of increased nebulization eYciencies then signal enhancements would be expected to be equal for samples EVects of carbon on arsenic signal containing the same concentration of carbon.Our present findings, however, do not support this theory. The report by It has been shown that carbon-containing substances can cause Allain et al.,11 in which signal enhancements for partially signal enhancement for partially ionised elements when analysed ionised elements were still observed when a post-nebulizer by ICP-MS.11 Although the mechanism responsible for this methane gas was added to the plasma, also supports the eVect is not fully understood, it is currently believed that the hypothesis that signal enhancement is mainly caused because presence of carbon in the plasma induces changes in the thermoof the presence of carbon in the plasma and not because of ionisation of some elements over a limited range of ionisation increases in nebulization eYciencies.energies. Allain et al.11 first reported signal enhancements for After comparing signal intensities obtained on injecting 20 partially ionised elements (Au, As, Hg, Se, Te) in the presence and 0.5 ml solutions, both containing the same absolute amount of organic substances and briefly commented on the diYculty of carbon, it was observed that signal enhancement was associated with finding suitable internal standards to compenapproximately the same for both sample sizes [Fig. 2(e) and sate for this. In view of these and later findings it is apparent (f )].However, on further inspection a slightly greater enhance- that signal variations originating from the carbon present in ment was observed for 0.5 ml samples. This minor diVerence samples can cause errors when quantifying As by ICP-MS. To is believed to occur as a result of diVerences in the rates correct for signal variations, samples must either be pre-treated at which carbon is nebulized into the ICP-MS system for the to reduce their carbon content or analysed along with matrixtwo diVerent sample sizes.For example, on injecting samples matched standards or with appropriate internal standards. All containing identical absolute amounts of carbon, a 20 ml three approaches, however, exhibit limitations. As already dissample results in a 40-fold lower rate of carbon introduction cussed, sample pre-treatment is not only time consuming but compared with a 0.5 ml sample. In practice, this suggest that also involves additional sample preparation steps that can lead similar reductions in signal enhancement for As in samples to analyte loss or sample contamination.Employing matrixwith elevated levels of carbon can be achieved by using either matched standards is extremely diYcult, especially for samples mFI (0.5 ml internal volume injector) or by diluting the sample containing unspecified or variable amounts of carbon, e.g., 40-fold and using conventional FI (20 ml injection loop). Of waste mixtures or alcoholic beverages.Also, samples containing course some disadvantages associated with sample dilution large amounts of dissolved solids can cause additional noninclude the risk of contamination and additional time required spectral interferences. per analysis. Also, larger injection loops give broader peaks, As part of this study a mFI-ICP-MS system was used to which result in reduced sample throughput. examine the eVects of organic solvents on the intensity of As signals.The investigation was carried out by analysing solu- Selection of internal standard tions containing constant amounts of As along with various amounts of carbon. In these experiments methanol and isopro- As part of this study, the use of mFI-ICP-MS for overcoming interferences caused by organic compounds was also investi- pyl alcohol were used as carbon sources. Initially, the eVects of increasing concentrations of methanol and isopropyl alcohol gated. The fact that the mFI system only requires small sample volumes (1 ml ), and, as a consequence, only introduces a on As signal intensity, for 0.2 and 20 ml samples, were investigated.The results obtained from these experiments are sum- small absolute amount of sample matrix into the ICP-MS system, prompted us to evaluate mFI-ICP-MS for its suitability marised in Fig. 2. As expected, non-spectral interferences, manifested as signal enhancements, were observed when to eliminate or reduce non-spectral interferences caused by organic substances.However, as already mentioned, in the samples containing organic solvent were introduced into the ICP-MS system via FI. The signal enhancement was signifi- presence of elevated levels of organic substances, accurate quantification of As can only be achieved if a suitable internal cantly greater for 20 ml samples compared with 0.2 ml samples, even though the organic solvent concentration was the same standard is used to compensate for all non-spectral interferences.In general, an internal standard should be selected in both cases [Fig. 2(a) and (b)]. Since the carrier flow rate was also the same in both cases, i.e., 200 ml min-1, the rate of with consideration to the type of sample to be analysed and the ionisation eYciency of the element. The elements evaluated carbon aspiration into the ICP-MS system was identical in both cases. In order to explain the greater signal enhancement in this study as internal standards for As were Se, Y, In and Sb.J. Anal. At. Spectrom., 1999, 14, 657–662 659Fig. 2 Comparison of the eVect of carbon-containing compounds on As signal intensity for solutions containing 20 pg ml-1 As: (a), (c) and (e) with isopropyl alcohol (i-PrOH) as carbon source; (b), (d) and (f ) with methanol (MeOH) as carbon source. The eVects of methanol and isopropyl alcohol on the lysed using FI-ICP-MS the response ratio As+/Se+ deviated from the reference value of 100%, reaching a maximum of response ratios of As-to-internal standard were investigated in detail (Table 3).When Se was used as the internal standard it 129%. This finding indicates that Se can only compensate partially for the enhancement of As signals caused by elevated compensated successfully for the enhancement of As signals caused by elevated levels of organics. The measured As+/Se+ levels of organics. Also, unacceptably high standard deviations were observed for triplicate 20 ml injections of these samples.ratio remained relatively constant (99±3%) for both 0.2 and 0.5 ml samples containing 0–100% v/v isopropyl alcohol or This was most likely caused because organic solvents remain in the plasma for a prolonged time, thus aVecting samples methanol. When larger samples, i.e., 20 ml samples, containing 69.3% v/v isopropyl alcohol or 30% v/v methanol, were ana- injected subsequently. Table 3 EVect of organic solvents on the As-to-internal standard ratio for diVerent sample volumes (mean±s; n=3)a % of original ratioc As+/Se+ As+/Y+ As+/In+ As+/Sb+ Volume injected/ml Volume injected/ml Volume injected/ml Volume injected/ml Solventb 0.2 0.5 20 0.2 0.5 20 0.2 0.5 20 0.2 0.5 20 i-PrOH (%): 3.5 96±1 101±1 90± 5 98± 1 108±1 135±7 103±1 108±1 132±4 89± 3 106±1 134±6 6.9 101±3 100±1 95± 5 102±2 110±2 178±6 103±5 109±3 166±3 112±2 107±2 166±4 10.4 97±7 101±3 98± 9 102±1 108±2 234±8 104±6 108±3 202±5 114±6 105±2 185±8 20.8 93±4 98± 4 106±8 106±5 106±5 241±12 102±13 111±6 222±9 124±9 102±5 209±18 69.3 98±6 99± 4 110±15 119±5 124±5 474±23 101±20 124±6 403±17 142±11 112±15 208±25 100 94±6 107±6 n.d.d 85±12 127±12 n.d. 107±24 134±19 n.d. 174±15 168±21 n.d. MeOH (%): 5 101±1 101±1 96± 4 102±3 106±3 135±5 102±1 108±2 134±5 102±2 106±2 124±4 10 98±1 99± 1 100±5 101±1 107±2 171±5 102±1 109±1 167±3 102±1 107±2 151±5 15 101±3 98± 2 107±4 105±2 103±3 216±7 110±3 108±3 199±6 105±2 104±2 178±6 20 99±2 98± 2 104±7 103±2 100±6 265±15 104±4 106±3 242±11 104±1 102±5 201±14 30 102±3 98± 2 129±13 108±3 106±6 364±18 106±7 111±6 305±16 107±6 106±7 249±19 100 98±4 103±4 n.d. 86±3 116±8 n.d. 121±9 127±5 n.d. 99±7 137±7 n.d. aCarrier flow at 200 ml min-1. bi-PrOH=Isopropyl alcohol; MeOH=methanol. cOriginal ratio is defined as the ratio of As signal intensity to that of Se, Y, In or Se in aqueous solutions. dn.d.: Not determined. 660 J. Anal. At. Spectrom., 1999, 14, 657–662When the As+/Y+ ratio was examined it was observed that were analysed.High recoveries, however, were observed when solutions containing elevated levels of isopropyl alcohol were Y did not compensate adequately for As signal fluctuations in the presence of elevated levels of organic solvents (Table 3). analysed. Although this finding is not fully understood, it may relate to diVerences in the washout characteristics of methanol When injecting 0.2 or 0.5 ml samples the As+/Y+ ratio only remained close to the reference value for solutions containing and isopropyl alcohol, which may relate to diVerences in their viscosity and/or vapour pressure.up to 20.8% isopropyl alcohol or 20% methanol. Yttrium did not compensate for As signal variations for solutions containing higher concentrations of organic solvents. This was further Determination of arsenic in diluted wine samples observed when 20 ml samples were injected. These samples According to Baluja-Santos and Gonzalez-Portal,21 the concen- gave 135–474% enhancements for As signals when Y was used tration of As in wines depends on a variety of factors such as as the internal standard.soil composition, grape variety, climatic conditions, use of Indium, one of the most commonly used internal standards pesticides, vinification process and storage conditions. The in ICP-MS, compensated adequately for As signal fluctuations OYcer Internationale de la Vigne et du Vin has set the maximum when 0.2 or 0.5 ml solutions containing up to approximately limit of As in wines at 200 pg ml-1.Although very little data 20.8% v/v isopropyl alcohol or 20% v/v methanol were injected. regarding As concentrations in wine has been published, it is However, for solutions containing higher amounts of organics accepted that uncontaminated wines contain only a few pg ml-1 severe interferences were observed. When 20 ml samples were of As.21–24 In the past, the classical Gutzeit method was used injected, enhancements ranging from 132 to 403% were for determining As in wine.The method is based on the observed, indicating that as little as 5% v/v methanol causes generation of volatile arsines, followed by their reaction with considerable As signal enhancement when using a conventional silver diethyldithiocarbamate and subsequent measurement of FI system. Similar observations were made for Sb (Table 3). the As-containing complex using molecular absorption.25 The LOD for this method, estimated at 10 pg ml-1, is probably Determination of arsenic in solutions containing organic solvents insuYcient for the analysis of most wines. Hydride generation atomic absorption spectrometry has been used for the determi- Solutions containing Se (300 pg ml-1), Y (10 pg ml-1) and In (10 pg ml-1), and 0.05% HNO3, along with various amounts nation of several hydride-forming elements, including As, in wine and beverages.21 The technique normally requires sample of methanol and isopropyl alcohol, were spiked with As and subsequently analysed for their As content.External calibrations decomposition, a time consuming procedure which may result in sample contamination or analyte loss. Also, when electrother- were applied in order to quantify As and determine As recoveries. No organics were added to the calibration standards used mal atomic absorption spectrometry is used it is preferred that the wine sample is decomposed prior to the analysis, especially for the external calibration.The results obtained from this experiment are summarised in Table 4. The average As recovery when As concentrations are close to the method’s LOD, which was reported to be 1.8 pg ml-1 for untreated wine.22 Following determined using mFI-ICP-MS with the Se internal standard was 105±4%, whereas the As recoveries obtained using Y or wine decomposition, the cLOD was determined to be 0.5 pg ml-1. The main diYculties encountered when analysing wines In as internal standard were 114±3 and 109±2%, respectively.Arsenic recoveries obtained using Se as the internal standard are caused by their complex matrix. Wines contain ethanol in addition to a variety of inorganic and other organic substances reached 119% when 20 ml samples were injected. Samples were subsequently diluted 40-fold so that a 20 ml sample contained sometimes at concentration levels of up to 1%. The main inorganic ions, present at about 1 g l-1, are K+, Na+, Mg2+ the same absolute amount of carbon as a 0.5 ml non-diluted sample.Arsenic recoveries obtained for the diluted samples and Ca2+. The organic substances are mainly citric acid, glycerol, polyphenols, various proteins, amino acids and were determined to be 98±2%, with Se as the internal standard. These findings suggest that sample dilution and subsequent polysaccharides. White wines are less abundant in dissolved substances than red wines, particularly with respect to injection via a conventional 20 ml injection loop give results similar to those obtained when analysing non-diluted samples polyphenols.26 The mFI-ICP-MS system developed in this work was tested via a 0.5 ml injector.This means that both approaches can be used with similar results. However, the mFI procedure oVers for determining As recoveries in wine samples spiked with As. The average As recoveries were determined to be 76±7 and superior aLOD (Table 2), higher sample throughput, and, most importantly, can handle samples with a high organic content 100±2% when using Se and In as internal standards, respectively (Table 5).These findings suggest that Se does not com- without the need for sample dilution. In almost all cases, the In internal standard also provided pensate adequately for As signal variations in wine samples. This is probably due to the eVects caused by the elevated acceptable recoveries for As when non-diluted 0.5 ml samples Table 4 Arsenic recoveries from de-ionised water solutions containing various amounts of organic solvents (mean±s; n=3).MeOH=Methanol; i-PrOH=isopropyl alcohol % Recovery of As from de-ionised water spiked with organic solvents Internal 5% MeOH 10% MeOH 20% MeOH 30% MeOH 5% MeOH 5% MeOH 5% MeOH standard +5% i-PrOH +5% i-PrOH +5% i-PrOH +5% i-PrOH +10% i-PrOH +20% i-PrOH +30% i-PrOH Sea 102±2 104±2 106±3 107±5 107±6 109±2 102±6 Seb 99±2 102±1 111±1 119±3 105±1 117±1 112±2 Sec 98±2 100±1 96± 5 100±5 96± 2 100±3 100±6 Ya 105±4 104±2 110±1 116±3 114±5 121±4 132±3 Yb 183±5 199±5 242±5 308±9 232±1 315±4 364±18 Yc 78±1 88± 3 92± 4 90± 8 86± 2 94± 3 102±5 Ina 102±2 101±2 106±2 110±5 107±3 114±3 122±7 Inb 184±2 206±3 255±4 315±12 233±2 320±5 300±15 aSample chamber 0.5 ml, carrier flow 200 ml min-1.bConventional loop 20 ml, carrier flow 200 ml min-1. cConventional loop 20 ml, carrier flow 200 ml min-1 and diluted sample. J. Anal. At. Spectrom., 1999, 14, 657–662 661Table 5 Arsenic recoveries from spiked wine samples (mean±s; n=3) As recovery (%) Spiked As Se internal In internal Type of wine (ppb) standard standard Vin de Pays Des Cotes de Gascogne, France (white) 10 67±11 95±3 20 75±10 103±1 Vin de Pays de l’Herault, France (red) 10 81±7 98± 1 20 84±7 103±1 Vino de la Tierra, Spanish (white) 10 76±4 92± 1 20 77±1 103±3 Maureillera Cotes de Provence, France (rose�) 10 74± 9 103±4 20 78±6 102±1 Table 6 Concentration of As in winesa (mean±s; n=3) Concentration of As/pg ml-1 Type of wine External calibration Standard additions Vin de Pays Des Cotes de Gascogne, France (white) 10.4±1.3 9.7±1.1 Vin de Pays de l’Herault, France (red) 13.2±1.0 12.6±0.7 Vino de la Tierra, Spanish (white) 7.7±0.2 7.2±0.2 Maureillera Cotes de Provence, France (rose�) 8.3±0.6 8.3±0.2 aIn internal standard used in all determinations. 2 E. H. Larsen, G. Pritzl and S. H. Hansen, J. Anal. At. Spectrom., concentrations of other matrix elements (Ca, Na, K or 1993, 8, 1075.Mg).12,26 Indium, on the other hand, seems to be a suitable 3 W. R. Cullen and K. J. Reimer, Chem. Rev., 1989, 89, 713. internal standard for determining As in wine. 4 K. Ogoshi, I. Mori, K. Gotoh and K. Ogawa, Appl. Organomet. Four diVerent wines, diluted 1+1 with de-ionised water, Chem., 1996, 10, 757. were analysed for their As content. Indium was used as the 5 S. A. Pergantis, E. M. Heithmar and T. A. Hinners, Anal. Chem., internal standard during the analysis.External calibrations 1995, 67, 4530. were constructed by using aqueous standards containing As 6 B. S. Sheppard, J. A. Caruso, D. T. Heitkemper and K. A.Wolnik, Analyst, 1992, 117, 971. at a range of concentrations and 5 pg ml-1 In in 0.1% nitric 7 M.nd X. C. Le, Clin. Chem., 1998, 44, 539. acid. The concentration of As in these wines, as determined 8 J. Goossens, F. Vanhaeche, L. Moens and R. Dams, Anal. Chim. using external calibration and standard additions methods, is Acta, 1993, 280, 137.reported in Table 6. It was observed that both methods found 9 E. H. Larsen and S. Stu�rup, J. Anal. At. Spectrom., 1994, 9, 1099. As to be present in the wines at approximately 10 pg ml-1. It 10 S. H. Tan and G. Horlick, Appl. Spectrosc., 1986, 40, 445. should also be mentioned that the slopes of the calibration 11 P. Allain, L. Jaunault, Y. Mauras, J.-M. Mermet and plots obtained using external calibration (0.0191±0.0003) T.Delaporte, Anal. Chem., 1991, 63, 1497. and standard additions (0.0195±0.0004) were almost identical, 12 S. A. Pergantis, G.-M. Momplaisir, E. M. Heithmar and indicating that the In internal standard compensated T. A. Hinners, in Proceedings of the 44th ASMS Conference on Mass Spectrometry and Allied Topics, Portland, Oregon, 1996, adequately for the matrix eVects caused by the wine sample. American Society for Mass Spectrometry, 1996, p. 21. These findings also indicate that the external calibration pro- 13 S.Saverwyns, X. Zhang, F. Vanhaecke, R. Cornelis, L. Moens cedure is equally suited to determining As in wines by mFIand R. Dams, J. Anal. At. Spectrom., 1997, 12, 1047. ICP-MS, and thus the time consuming and tedious stan- 14 B. Magyar, P. Lienemann and H. Vonmont, Spectrochim. Acta, dard additions methodology is not necessary. Although the Part B, 1986, 41, 27. As concentrations found in the wines analysed are in the same 15 D. W. Hausler and L. T. Taylor, Anal. Chem., 1981, 53, 1223. range as those determined by other workers21–24 for other 16 T. J. Brotherton, P. E. Pfannerstill, J. T. Creed, D. T. Heitkemper, types of wines, it was extremely diYcult to establish the J. A. Caruso and S. E. Pratsinis, J. Anal. At. Spectrom., 1989, 4, 341. accuracy of the present mFI-ICP-MS method for quantifying 17 F. Vanhaecke, M. van Holderbeke, L. Moens and R. Dams, As in wines, especially since no wine reference material with J. Anal. At. Spectrom., 1996, 11, 543. certified concentrations of As is currently available. However, 18 S. D. Lofthouse, G. M. Greenway and S. C. Stephen, J. Anal. At. the fact that the external calibration and standard additions Spectrom., 1997, 12, 1373. methodologies gave almost identical results supports the val- 19 J. A. McLean, H. Zhang and A. Montaser, Anal. Chem., 1998, idity of the method. It should also be mentioned that when 70, 1012. monitoring m/z 77 we did not observe any evidence for the 20 S. A. Pergantis, E. M. Heithmar and T. A. Hinners, Analyst, 1997, presence of 40Ar37Cl+, which can cause interference at m/z 75 122, 1063. (40Ar35Cl+ on As+). Because wines contain chloride at about 21 C. Baluja-Santos and A. Gonzalez-Portal, Talanta, 1992, 39, 329. 22 B. T. Kildahl and W. Lund, Fresenius’ J. Anal. Chem., 1996, 100 mg ml-1, it is essential to check for any chloride inter- 354, 93. ferences when using mFI-ICP-MS for their analysis. 23 S. N. F. Bruno, R. C. Campos and A. J. Curtius, J. Anal. At. Spectrom., 1994, 9, 341. Acknowledgements 24 G. A. Pedersen, G. K. Mortesen and E. H Larsen, Food Addit. Contam., 1994, 11, 351. S. W. acknowledges the Royal Thai Government for finan- 25 P. D. Handson, J. Sci. Food Agric., 1984, 35, 215. cial support. 26 S. Augagneur and B. Medina, J. Anal. At. Spectrom., 1996, 11, 713. References 1 A. Lasztity, A. Krushevska, M. Kotrebai, R. M. Barnes and D. Amarasiriwardena, J. Anal. At. Spectrom., 1995, 10, 505. Paper 9/00067D 662 J. Anal. At. Spectrom., 1999, 14, 657–6

ISSN:0267-9477    

DOI:10.1039/a900067d

出版商:RSC    

年代:1999

数据来源: RSC

摘要:

Multi-element analysis of airborne particulate matter collected on PTFE-membrane filters by laser ablation inductively coupled plasma mass spectrometry Ching-Jer Chin, Chu-Fang Wang* and Su-Ling Jeng Department of Nuclear Science, National Tsing Hua University, Hsinchu, Taiwan. E-mail: cfwang@ins.nthu.edu.tw Received 17th September 1998, Accepted 4th January 1999 Direct analysis of airborne particulate matter collected on PTFE-membrane filters using laser ablation inductively coupled plasma mass spectrometry (LA-ICP-MS) was performed.Laboratory prepared standard filter samples were analyzed. Several parameters, including the laser energy, pulse type and beam focus, thermal properties and homogeneity of the sample and the carrier gas flow rate, may aVect the LA-ICP-MS measurement. The influences of these parameters were thoroughly examined. Empirical results obtained here demonstrated that applying LA-ICP-MS for multi-element analysis of airborne particulate matter collected on PTFE-membrane filter is feasible.More than 20 major, minor and trace elements in airborne particles on PTFE membrane filters can be determined. It was found that the optimum ablation eYciency can be achieved using a 160 mJ single shot laser operated in the free-running mode with a 6.5 mm defocus distance from the filter surface. The optimum transport eYciency for most of the elements can be obtained if the argon gas flow rate is kept at 0.8 L min-1.employed LA-ICP-MS to determine simultaneously 21 Introduction elements in atmospheric particulate matter. However, the For the elemental analysis of airborne particulate matter, calibration was performed using a standard filter prepared by particle samples collected on various filter media are often dropping an aqueous standard solution on to the surface of a used. Since trace levels of many elements need to be determined membrane filter, the matrix of which was totally diVerent from within a limited amount of particles, high-purity polytetra- that of real samples.A prerequisite for the quantitative analysis fluoroethylene (PTFE)-membrane filters have been considered of air particle samples collected on membrane filters is that as one of the most appropriate choices for collecting air the matrices of both the standard and samples be similar. In samples.1 However, owing to its low loading capacity, finding fact, calibration is the most important aspect and a limiting a suitable analytical technique to determine the various elemen- factor for accurate and precise LA-ICP-MS analysis.tal concentrations of air particles on the filter medium is During our previous work, closely matched standard filters essential. were prepared for calibration and quantitative analysis of air Inductively coupled plasma mass spectrometry (ICP-MS) particles.16 To examine further the feasibility of applying this incorporated with an acid mixture digestion method has been technique to real samples, it was necessary to explore the optimum experimental parameters, such as laser energy, pulse employed in our laboratory to analyze airborne particulate type and beam focus, thermal properties and homogeneity of samples.2 Compared with other analytical techniques, ICP-MS the sample and carrier gas flow rate.has several attractive features, e.g., very low detection limits In this work, an investigation into the direct analysis of for many elements, multi-element analysis capability, rapid airborne particulate matter collected on PTFE-membrane fil- scan and simple, easily interpreted spectra.However, many ters using LA-ICP-MS was performed. This study not only analytical problems have been encountered in airborne particudemonstrates the advantages of the laser ablation technique late analysis. Samples are diYcult to digest, trace elements employed in the introduction system of ICP-MS, but also may be lost or contaminated during the digestion process and oVers practical observations and suggestions for optimizing spectrometric interference may be serious owing to the acidand applying the technique. During the course of this work, derived background ions formed during the ion extraction LA-ICP-MS was employed to analyze quantitatively standard process of ICP-MS.3–7 filters prepared in the laboratory.Conditions for such analyses Applying a laser ablation technique, which involves directly and methods of calibration for obtaining accurate results were removing particulate matter from the filter media by ablation investigated in detail.with laser pulses, incorporated with ICP-MS (LA-ICP-MS) may provide a solution to reducing the problems associated with the conventional acid-digestion/ICP-MS method.8–11 Experimental With the advantages of no requirement for sample pre- Sample preparation treatment, reduction of sample contamination, loss and spectral interference, laser ablation has been considered to be by PTFE-membrane filters [37 mm diameter, 2 mm pore; Gelman far the most versatile and promising of the solid introduction (Ann Arbor, MI, USA) R2PJO37] with polyester supports methods for ICP-MS.However, there is still a lack of experi- were used. Standard filters with diVerent particulate concenence of LA-ICP-MS to analyze quantitatively air particles trations were prepared in the laboratory by depositing NIST standard reference material SRM 1648 Urban Particulate collected on membrane filters.12–14 Tanaka et al.15 recently J.Anal. At. Spectrom., 1999, 14, 663–668 663Matter on the PTFE-membrane filter in a designed sampling chamber.16 The certified elemental concentrations for each filter were verified using the acid-digestion/ICP-MS method after the LA-ICP-MS analysis. Apparatus The ICP-MS instrument used was a Perkin-Elmer SCIEX ( Thornhill, Ontario, Canada), Elan Model 5000 quadrupole mass spectrometer and the laser ablation instrument was a Perkin-Elmer SCIEX Model 320 sampler equipped with Quanta-Ray GCR-11 pulsed Nd5YAG laser (wavelength 1064 nm).The laser was operated in either the free-running (pulse width about 230 ms) or Q-switch (pulse width 8–9 ns) mode. The laser power used in this work varied from 100 mJ to 180 mJ for both the free-running and Q-switch modes. The laser beam was focused with a lens of focal length 60 mm and the estimated diameter of the spot size was about 0.6 mm.An argon flow was used as the carrier gas from the laser sample cell to the torch. A 3 m×5 mm id PVC hose was employed to transfer the sample vapor to the torch. Table 1 gives the operating conditions of both the laser ablation and the ICP-MS instruments. Experimental approach A direct laser ablation technique coupled with ICP-MS was tested first to examine the capability of this technique for qualitative and quantitative analyses of airborne particulates on PTFE-membrane filter samples.During the course of this work, several parameters, including the energy, pulse type and focus condition of the laser and the transport eYciency of the sample, were examined to determine the optimum ablation conditions. The laser was operated in such a way that seven successive shots were randomly fired on diVerent positions of the same filter to normalize the imhomogeneity of the filter Fig. 1 Responses of (a) Na, (b) K, (c) Fe and (d) Pb ablated by a medium or particles on the filter and instability of the ablation single shot laser in the free-running mode with 160 mJ laser beam instrument.To obtain an accurate and precise analytical result, energy from blank, standard and real sample filters. The mass loading a standard calibration experiment was performed using the of the standard filter sample is 0.67 mg per filter. prepared standard filters. of the laser was also observed. The abscissa represents the Results and discussion transient time in the peak hopping scanning mode of ICP-MS.After the laser shot, a delay of the signal peak was generally Qualitative LA-ICP-MS analysis observed, which might be attributed primarily to the transport Fig. 1 displays the responses of the four elements Na, K, Fe time from the laser ablation equipment to the ICP-MS. Several and Pb (other elements exhibited similar responses) when a seconds later, the signal reached a maximum and then gradusingle laser pulse in the free-running mode with 160 mJ laser ally decreased.A total of 30 s replicate time was included to beam energy was fired on to blank and standard sample filters. ensure a complete measurement. During the course of this A similar but less intense response for the Q-switched mode work, 23 elements of interest were determined sequentially. A 20 ms dwell time and 50 repeated scans in ICP-MS were Table 1 LA-ICP-MS operating conditions selected for each element, thereby extending the total acquisition time for a single element to at least 1000 ms.A significant ICP-MS high background intensity for elements such as Na was Instrument PE SCIEX Elan 5000 observed, which could have been contributed by the contami- Power 1100 W nation. On the other hand, the high background intensity of Coolant argon flow rate 14 L min-1 Auxiliary argon flow rate 0.8 L min-1 K could be attributed to the polyatomic interference of Carrier argon flow rate 0.8 L min-1 38Ar1H+ during the ICP-MS measurement.Except for Na, K Dwell time 20 ms and Ca, the background contributions for most of the elements Resolution Normala of interest were negligible and quantitative analysis could be Readings per replicate 50 performed by integrating the signal peaks as shown in Fig. 1. Sweeps per reading 1 Fig. 2 exhibits the mass spectra of (a) the standard and (b) Scanning mode Peak hopping transient Laser ablation loaded filter samples that were obtained using a mass scan from Instrument PE Laser Sampler Model 320 20 to 215.The interference from a blank filter [Fig. 2(c)] and Operating mode Free running argon gas [Fig. 2(d)] were also presented as background contri- Ablation type Single pulse butions. It was observed that the only significant interference Laser wavelength 1.06 mm was in the region of m/z <40, which was primarily due to the Flash lamp energy 40 J ( laser beam energy ~160 mJ) formation of polyatomic ions from clustering reactions during Defocus distance 6.5 mm the ion-extraction process, e.g., ArH+, ArN+ and ArO+. As aThe peak width at 10% of peak maximum is about 0.832 mass units. can be seen in Fig. 2, the interference due to the acid-derived 664 J. Anal. At. Spectrom., 1999, 14, 663–668Fig. 2 Mass spectra of no filter, blank, standard and loaded filter samples obtained with a mass scan from 21 to 215. background ions was greatly reduced.For instance, 75As (100% abundance) could be determined without the interference of 40Ar35Cl+, which usually forms from a combination of chloride introduced via reagents with argon. Fig. 2 reveals that LA-ICP-MS is most appropriate for determining transition and rare earth elements. On the other hand, for elements such as Na, K and Ca, associated spectral interference may cause larger errors during the LA-ICP-MS measurement. Consequently, determining these elements using LA-ICP-MS is not recommended.The elements determined using this technique are Fig. 3 Enlarged photographs (×30) of loaded PTFE-membrane filter shown at the masses employed in this study (24Mg, 27Al, 29Si, samples ablated at diVerent laser energies. (A) Free-running, 120 mJ; 35Cl, 49Ti, 51V, 52Cr, 55Mn, 57Fe, 59Co, 60Ni, 63Cu, 66Zn, 75As, (B) free-running, 140 mJ; (C) free-running, 160 mJ; (D) free-running, 180 mJ; (E) Q-switch, 120 mJ; (F) Q-switch, 140 mJ; (G) Q-switch, 77Se, 88Sr, 114Cd, 121Sb, 138Ba and 208Pb). 160 mJ; and (H) Q-switch, 180 mJ. Selecting the laser operation mode sample filter in a very short period of time (<10 ns), which Two types of operation mode, free-running and Q-switch may provide suYcient energy (109–1012 W cm-2) to melt the modes, were examined at diVerent laser energies to explore filter media even before the particles are vaporized. This the optimum energy and operation mode for the laser. Fig. 3 phenomenon can be avoided by altering the operation mode, displays the enlarged photographs (×30) of loaded PTFE- lowering the laser energy or defocusing the beam.membrane filter samples ablated at diVerent laser energies. Table 2 compares the relative integral signal intensities gener- Lower laser energy tends to ablate a smaller mass and avoids ated per unit energy at 100, 120, 140, 160, 180 and 200 mJ. As physically disturbing the sample surface. However, the laser can be seen in Fig. 3, a wider area of particles should be not only needs to remove the particles from the filter medium vaporized with increasing energy applied on the filter.The data but also to free them for transport so that they can be delivered in Table 2 also show a threshold eVect for the ablation from 100 to the ICP for further analysis. To achieve a more representa- to 120 mJ energy. However, a saturated ablation phenomenon tive ablation, particles of a given spot size should be completely becomes significant if the energy is higher than 160 mJ.vaporized with suYcient energy.17 It was observed from Fig. 3 As a result, diVerent laser energy and operation modes that no ablation phenomena occurred until the flash lamp significantly aVect the coupling eYciency of energy to the energy was raised to 30 J ( laser beam energy about 120 mJ) sample and also the precision and sensitivity of the measureand a doubtful TEM01 cylindrical symmetry mode profile ment. The recommended low energy and normal operating appeared on the burn patterns.The ablated area is significantly mode as optimum conditions may suggest that particle cleaning larger than the cross-section of the laser beam. It appears that is the dominant mechanism for removing particles from the the area is wider only because the energy increase brings to filter surface and not the ablation process.19 A single shot the outer edge of the spatial profile a high enough energy level laser in the free-running mode with 160 mJ laser beam energy to cause ablation.Fig. 3 also shows that the highly focused was therefore employed in the investigation as the optimum Q-switch ablation mode has a tendency to burn a large hollow laser operating condition for LA-ICP-MS. hole through the membrane. A pronounced burn pattern occurred in the 180 mJ Q-switch mode. On the other hand, Optimizing laser focus the PTFE substrates still remained after ablation in the low energy density free-running mode.The eVect of laser focusing on precision and sensitivity was evaluated from the integrated signal data. Fig. 4 indicates that, During the ablation process, the focused laser beam initially produces a laser induced plasma of ionized argon, sample and with the proposed optimized laser operating conditions, the integral intensities of several elements vary with the defocus filter media vapor and free electrons above the filter surface. 18,19 The low b.p.PTFE substrates have been considered distance, which is defined as the distance where the laser becomes focused below the surface of the sample.20 An asymp- as a material which conducts heat very well. In the high power density Q-switch mode, focused energy is transferred to the totic increase in intensity was generally observed for most J. Anal. At. Spectrom., 1999, 14, 663–668 665Table 2 Relative integral signal intensities (ions J-1) generated per unit energy in free-running and Q-switched modes at diVerent laser beam energies (n=7) Element Laser mode 100 mJ 120 mJ 140 mJ 160 mJ 180 mJ 24Mg Free-running 600 (6)a 11000 (13) 21000 (11) 24000 (4) 24000 (3) Q-switch 610 (5) 5100 (10) 13000 (5) 16000 (3) —b 48Ti Free-running 170 (7) 3000 (15) 6500 (17) 8500 (10) 8300 (2) Q-switch 220 (6) 2000 (16) 5200 (7) 9300 (10) — 51V Free-running 380 (8) 5200 (8) 9400 (8) 11000 (3) 11000 (3) Q-switch 400 (6) 4000 (6) 7800 (3) 8300 (2) — 55Mn Free-running 300 (3) 16000 (16) 29000 (9) 34000 (4) 32000 (3) Q-switch 300 (3) 9800 (7) 21000 (1) 23000 (3) — 57Fe Free-running 270 (3) 4400 (12) 7300 (5) 8400 (3) 7900 (3) Q-switch 260 (9) 38000 (10) 7500 (4) 7500 (2) — 60Ni Free-running 220 (10) 1500 (19) 2400 (9) 2800 (4) 2700 (6) Q-switch 230 (12) 950 (15) 1800 (2) 1900 (5) — 63Cu Free-running 940 (8) 17000 (14) 29000 (8) 34000 (2) 32000 (4) Q-switch 860 (4) 14000 (10) 23000 (3) 25000 (2) — 66Zn Free-running 200 (11) 31000 (7) 51200 (8) 60000 (2) 59000 (6) Q-switch 220 (11) 15000 (9) 30000 (7) 37000 (5) — 75As Free-running 200 (5) 370 (10) 540 (5) 500 (4) 450 (5) Q-switch 190 (9) 320 (9) 420 (7) 370 (2) — 88Sr Free-running 150 (14) 760 (9) 1200 (10) 1500 (6) 1400 (8) Q-switch 160(14) 440 (15) 790 (11) 980 (7) — 114Cd Free-running 130 (12) 290 (12) 310 (11) 330 (15) 400 (4) Q-switch 150 (14) 440 (10) 620 (6) 670 (8) — 121Sb Free-running 130 (3) 1000 (16) 1800 (5) 2100 (2) 19000 (4) Q-switch 140 (7) 800 (6) 1500 (3) 1600 (6) — 138Ba Free-running 120 (8) 12000 (17) 23000 (10) 27000 (5) 26000 (3) Q-switch 110 (8) 6200 (10) 13000 (7) 15000 (4) — 208Pb Free-running 103 (8) 7100 (7) 14000 (7) 16000 (2) 15000 (6) Q-switch 90 (13) 4200 (10) 8000 (8) 9800 (7) — aData in parentheses are the relative standard deviations (%).bNot determined because of burning through the membrane filter. Fig. 4 Integral intensities of 52Cr, 57Fe, 60Ni, 121Sb, 138Ba and 208Pb vary with the defocus distance using a 160 mJ single shot laser operated in the free-running mode.elements, which then reached a plateau with a defocus distance of almost 6 mm. A wider area could be ablated by defocusing the laser beam and rastering it over the sample surface. Fig. 5 illustrates the linear relationship between the defocus distance and the size of the ablated area on the filter at a defocus distance of 5.5 mm, and then the asymptotic plateau at a greater defocus distance. Obviously, the ablated area was dependent on the energy dissipated to the sample, which was alternatively controlled by the laser energy and by the defocus distance of the sample.Airborne particles usually contain a variety of matrix constituents including organic compounds, oxides and silicates. A larger plasma plume can be formed with a deeper defocus position because more particles on the filter can be directly evaporated by the laser beam. However, since only limited energy was transferred with each laser setting, the observed Fig. 5 Relationship between the defocus distance and diameter of intensity did not rise with a defocus distance to much higher ablated crater on the filter with a 160 mJ single shot laser operated in the free-running mode. values. 666 J. Anal. At. Spectrom., 1999, 14, 663–668Fig. 6 Relationship between the signal intensities of 27Al+, 39K+, 57Fe+, 88Sr+, 121Sb+ and 208Pb+ in the filter samples and the carrier gas flow rate at 160 mJ in the free-running mode at a 6.5 mm defocus distance.Fig. 8 Calibration curves for Fe obtained from NIST SRM 1648 Urban Particulate Matter; the total suspended particulate (TSP) mass loading on the filter ranged from 0 to 5.5 mg per filter. The experiment was performed with a 160 mJ single shot laser operated in the freerunning mode with a 6.5 mm defocus distance and 0.8 L min-1 carrier gas flow rate. Table 4 Calibration data r2 and LODs for airborne particulate loaded filters determined by LA-ICP-MS Slope Intercept LOD Element (×104) (×104) r2 (ng per filter) 23Na 28 665.0 0.782 820.0 24Mg 70 2.2 0.722 3.2 29Si 1.9 8.9 0.905 250.0 39K 32 92.3 0.791 74.6 49Ti 11 0.8 0.924 15.6 Fig. 7 Transient spectra of 121Sb with various carrier gas flow rates at 51V 500 1.3 0.925 0.7 160 mJ in the free-running mode at a 6.5 mm defocus distance. 52Cr 200 9.1 0.768 4.3 55Mn 300 1.0 0.933 0.6 Table 3 Relative standard deviations (%) of elements on the loaded 57Fe 5.7 0.8 0.945 41.8 PTFE-membrane filter determined using LA-ICP-MS in Q-switch and 60Ni 100 0.7 0.965 1.8 free-running laser ablation modes with 160 mJ laser energy (n=7) 63Cu 300 0.8 0.905 0.4 66Zn 38 0.7 0.942 5.2 Element Q-switch mode Free-running mode 75As 84 0.7 0.978 2.2 77Se 9.9 0.5 0.981 18.2 24Mg 8 10 114Cd 68 0.4 0.829 1.8 27Al 12 11 121Sb 300 0.3 0.947 0.3 29Si 9 6 138Ba 200 0.3 0.824 0.8 35Cl 2 3 208Pb 74 0.3 0.840 1.6 44Ca 9 10 49Ti 9 10 51V 6 10 Influence of the carrier gas flow rate 52Cr 5 11 55Mn 3 9 The ablated material was entrained in an argon gas stream 57Fe 6 10 and transported by tube to the ICP for subsequent excitation. 60Ni 4 10 63Cu 5 9 Many researchers have evaluated gas flow entrainment.21,22 66Zn 6 10 Generally, the flow rate of carrier gas will influence the 75As 7 7 transport eYciency and the signal sensitivity. Fig. 6 illustrates 88Sr 8 9 the relationship between the signal intensities of 27Al+, 39K+, 114Cd 6 7 57Fe+, 88Sr+, 121Sb+ and 208Pb+ (other elements also showed 121Sb 7 9 similar trends) in the filter samples and the carrier gas flow 138Ba 8 10 208Pb 12 10 rate.It was observed for most elements that maximum sensitivity could be reached with a carrier gas flow rate ranging between 0.8 and 0.9 L min-1 and that the peak position shifted slightly towards a higher flow rate with increasing atomic As Fig. 4 indicates that, in the free-running mode with 160 mJ laser beam energy, the optimum sensitivity of most number.We believe that a lower carrier gas flow rate might degrade the transport eYciency and cause overheating of the elements determined using LA-ICP-MS was found at a defocus position of 6.5 mm. It was reasonable to expect that optimum plasma in the ICP, which may also increase the production of divalent ions. On the other hand, incomplete excitation could sensitivity also leads to an optimum precision because in that position the highest eYciency of energy coupled to the sample occur in the ICP process with a much higher flow rate, which is attributed to the short dwell time of the analytes and cooler is achieved.J. Anal. At. Spectrom., 1999, 14, 663–668 667temperature in the plasma. Significant spectroscopic inter- energy, pulse type and beam focus, thermal properties and homogeneity of the sample and the carrier gas flow rate, on ference from the formation of molecular ions was therefore expected. Fig. 7 indicates that with the proposed laser the LA-ICP-MS measurement were thoroughly examined.The empirical results obtained demonstrated that applying operating conditions, steady transient spectra of 121Sb can only be obtained at a 0.8–1.0 L min-1 carrier gas flow rate. LA-ICP-MS for the multi-element analysis of airborne particulate matter collected on PTFE-membrane filters is feasible. For elements such as Na, K and Ca, the decrease in intensity with increase in flow rate is illustrated by the curve for K in This suggests that particulate matter loaded on the filter will be eYciently ablated if a moderate power input is utilized.If Fig. 6. The K signal is possibly related to the fact that an increase in 38Ar1H in the background with decreasing nebulizer the power input is too high, a burning pattern occurs on the filter sample, leading to irreproducible results. As a result, the gas flow rate as the temperature of the plasma increases is not expected. Possibly it is due to tailing of increased 40Ar.In optimized ablation eYciency can be achieved using a 160 mJ single shot laser operated in the free-running mode addition, it was found from a subsidiary experiment that the smallest fluctuation could also be obtained at a 0.8 L min-1 with a 6.5 mm defocus distance from the filter surface. The optimum transport eYciency for most of the elements can be flow rate for most elements, which was therefore selected as the optimum carrier gas flow rate for this investigation.obtained if the argon gas flow rate is kept at 0.8 L min-1. More than 20 major, minor and trace elements in airborne particles on PTFE membrane filters can be determined. Homogeneity study of filter membrane samples Since only a small area on the filter could be ablated using Acknowledgments single pulse operation, poor reproducibility due to the heterogeneity in both the sample and the filter matrix was expected. The authors thank the National Science Council, Republic of To minimize this source of variability, multiple ablations for China, for financially supporting this research under Contract each sample analysis were therefore suggested.Table 3 lists No. NSC87–2113-M007–039. Professor M. H. Yang, the results of seven successive repeated analyses on a PTFE- Department of Nuclear Science, National Tsing Hua membrane standard filter sample. University and Professor P. C. Chiang, Institute of The use of internal standards to compensate for temporal Environmental Engineering, National Taiwan University are signal variations has been recommended by many workers in thanked for their long-term assistance and valuable comments.diVerent studies.11,23 However, selecting appropriate elements for trace element determinations in airborne particulate matter References collected on a membrane filter is not simple because their concentrations vary from sample to sample. Since Si possessed 1 C. F. Wang, E.E. Chang, P. C. Chiang and N. K. Aras, Analyst, 1995, 120, 2521. the highest concentration in most of the real samples, it can 2 C. F. Wang, M. F. Huang, E. E. Chang and P. C. Chiang, Anal. be used as the reference element for internal standardization. Sci., 1996, 12, 201. Based on the analytical methods developed previously in our 3 H. Kawaguchi, T. Tanaka, T. Nakamuram, M. Morishita and A. laboratory, Si may be rapidly determined using XRF spec- Mizuike, Anal. Sci., 1987, 3, 205.trometry without destroying filter samples. A reproducible 4 J. A. Olivares and R. S. Houk, Anal. Chem., 1986, 58, 20. signal for Si can be obtained with normalization of the XRF 5 B. S. Sheppard, D. T. Heitkemper and C. M. Gaston, Analyst, 1994, 119, 1683. measurement, and thus be utilized as a naturally occurring 6 L. Ebdon, A. S. Fisher and P. J. Worsfold, J. Anal. At. Spectrom., internal standard. 1994, 9, 611. 7 D. E. Nixon and T. P. Moyer, Spectrochim. Acta, Part B, 1996, Calibration of LA-ICP-MS system 51, 13. 8 N. Imai, Anal. Chim. Acta, 1989, 235, 381. Calibration curves were generated by integrating the transient 9 S. F. Durrant and N. I. Ward, Food Chem., 1994, 49, 317. signals of elements in various prepared standard filters. For 10 C. F. Wang, S. L. Jeng and F. J. Shieh, J. Anal. At. Spectrom., example, Fig. 8 shows the calibration curve of Fe established 1997, 12, 61. from the prepared standard filters with the proposed optimum 11 C.Luedke, E. HoVmann and J. Skole, Fresenius’ J. Anal. Chem., 1994, 350, 272. operating conditions. The calibration curve depicted here indi- 12 F. N. Abercrombie, M. D. Silvester, A. D. Murray and A. R. cates that measured intensities first increase with increasing Fe Barringer, in Application of ICP to Emission Spectrometry, ed. concentration and reach a plateau with much larger fluctuations Barnes, R. M., Franklin Institute Press, Philadelphia, 1987, p. 121. at higher concentrations.We believe that incomplete ablation 13 E. R. Denoyer, K. J. Freeden and J.W. Hager, Anal. Chem., 1991, might be the cause of such ‘saturation’ in the excess mass loading 63, 455A. of particulates on the filter media. An increase in laser energy 14 S. A. Darke and J. F. Tyson, J. Anal. At. Spectrom., 1993, 8, 145. 15 S. Tanaka, N. Yasushi, N. Sato, T. Fukasawa, S. J. Santosa, K. revealed a shift of saturation to higher concentrations and could Yamanaka and T. Ootoshi, J. Anal. At. Spectrom., 1998, 13, 135 possibly result in burning through the very thin filter. Fig. 8 16 C. F. Wang, S. L. Jeng, C. C. Lin and P. C. Chiang, Anal. Chim. reveals that an excellent linear calibration curve was obtained Acta, 1998, 368, 11. with a mass loading of less than 2 mg. This has also been 17 L. Moenke-Blankenburg, M. Ga�ckle, D. Gu�nther and J. Kammel, considered the limit of the loading capacity of a PTFE-membrane in Plasma Source Mass Spectrometry, ed. K. E. Jarvis, A. L. Gray, filter for collecting airborne particles.1 I. Jarvis and J. G. Williams, Royal Society of Chemistry, Cambridge, 1990, p. 1. The limits of detection (LODs) for determining various 18 D. Pottor and I. Abell, Anal. Sci., 1991, 7, 1239. elements using LA-ICP-MS were calculated using the 3s cri- 19 R. L. Armstrong and A. Zardecki, Appl. Opt., 1990, 29, 1786. terion on 10 ablations of blank filters.24 Table 4 gives the slopes, 20 I. D. Abell, in Applications of Plasma Source Mass Spectrometry, intercepts, r2 and LODs obtained from the calibration curves. ed. Holland, G., and Eaton, A. N., Royal Society of Chemistry, Since a white blank filter membrane can reflect laser energy, a Cambridge, 1991, p. 209. higher LOD should be expected in real sample analysis. 21 P. Arrowsmith and S. K. Hughes, Appl. Spectrosc., 1988, 42, 1231. 22 T. Mochizuki, A. Sakashita, H. Iwata, T. Kagaya, T. Shimamura and P. Blair, Anal. Sci., 1988, 4, 403. Conclusions 23 S. T. Durrant and N. I. Ward, Food Chem., 1994, 49, 317. 24 L. A. Currie, Anal. Chem., 1968, 40, 586. Direct analysis of airborne particulate matter collected on PTFE-membrane filters using LA-ICP-MS has been tested. Paper 8/07275B The influences of various parameters, including the laser 668 J. Anal. At. Spectrom., 1999, 14, 663&ndash

ISSN:0267-9477    

DOI:10.1039/a807275b

出版商:RSC    

年代:1999

数据来源: RSC

摘要:

Characterization of a hydrogen flame as an ion source for mass spectrometry Lee L. Yu,* Gregory C. Turk and S. Roy Koirtyohann† Analytical Chemistry Division, National Institute of Standards and Technology, Gaithersburg, MD 20899, USA Received 16th December 1998, Accepted 16th December 1998 A commercial inductively coupled plasma (ICP) mass spectrometer was modified to employ an air–hydrogen flame in place of the ICP as an ion source. A liquid nitrogen trap was placed in the vacuum line to remove water.A very simple intrinsic mass spectral background was obtained with the hydrogen flame ionization mass spectrometry (FIMS). Molecular ions such as K(H2O)+, Na(H2O)+, Ca(H2O)+ and CaOH(H2O)x+ (x=0–2) were observed when solutions containing Na, K or Ca were aspirated. Although the presence of the molecular ions complicated the mass spectra, it also provided a wider choice of analytical masses for an analyte. Isotope ratio measurements of Ca were made with both Ca+ and CaOH+ species at masses 40, 44, 57 and 61.Better isotope ratio precision was obtained at CaOH+ masses relative to those for Ca+ because the sensitivity was about 10 times higher. Isotope ratio measurement of K was made at masses 39 and 41. A ratio precision of about 0.2 and 0.5% was obtained for K and Ca, respectively. The results suggest that the FIMS is suitable for the isotope ratio measurement of K and Ca in simple matrices, and that the air–hydrogen flame is a more desirable ion source than an air–acetylene flame for FIMS.populations cannot be dramatically reduced without Introduction extinguishing the plasma. The spectral background of inductively coupled plasma mass Replacing the plasma with a diVerent ion source is another spectrometry (ICP-MS) has been well documented.1,2 While possible solution to the background problem. Zhang et al.10,11 many interferences from the plasma background can be mini- developed a helium inductively coupled plasma mass specmized by desolvation3,4 or by using an Ar–N2 mixed gas trometer.Elements such as K and Fe that suVer from spectral plasma,5,6 solutions to the interferences from Ar+ are surpris- interferences in Ar ICP-MS were detected at 0.1 and ingly few. As a result, isotope ratio measurements of K and 4 pg mL-1, respectively. Mass spectrometers were extensively Ca are severely impaired. One approach that has gained used to study the molecular species in the combustion flame limited success is to minimize the amount of Ar+ being in order to elucidate the combustion mechanism.12–14 extracted by the mass spectrometer.In this eVort, Smith et al.7 Conversely, flames can serve as alternative ion sources to the used an Ar–Xe supported plasma and a high resolution mode plasma for elemental mass spectrometry. Taylor et al.15 used of the mass spectrometer. The background at mass 40 was an air–acetylene flame to this eVect. A lower detection limit greatly reduced and its eVect on K masses was minimized.of K was reported in comparison with that obtained by The interference of 38ArH+ at mass 39 was negligible; neverthe- ICP-MS. Isotope ratio measurements of K were made, and less, the 40ArH+ interference at mass 41 was still the equivalent the feasibility of flame ionization mass spectrometry (FIMS) of about 0.9 ppm K. Despite the 40ArH+ background, the for ground water studies by using 41K as a stable tracer was measured K isotope ratio was reasonably close to the demonstrated. Unfortunately, the organic combustion prodaccepted value.7 ucts of an acetylene flame resulted in instrument contami- Jiang et al.8 virtually eliminated the Ar+ and ArH+ species nation.A significant amount of a wax-like residue was in the spectral background by operating an ICP-MS under observed inside the interface of the instrument after several cool plasma conditions. A detailed study of the ‘cold plasma’ hours of operation.It is likely that this accumulation has a conditions by Tanner9 showed that interferences from Ar+ deleterious eVect on instrument performance, making an air– were minimized whereas those from ArH+ persisted. A non- acetylene flame less suitable as an ion source. Alternatively, linear relationship was observed between the count rate and an air–hydrogen flame can be used for mass spectrometry, as the concentration of K, but the non-linearity was not expected reported by Turk et al.16 Unlike the residue from an acetylene to aVect the isotope ratio measurements.8 A 9% mass bias was flame, the water resulting from combustion was easily observed for the 39K/41K ratio.8 Despite the fact that the removed from the vacuum system with a liquid nitrogen count rate for 39K+ was lowered by about two orders of trap.16 In that work,16 the flame was used as an atom source magnitude, the detection limit was similar to that obtainable and ionization of ground state analyte atoms was with conventional instrument settings.8 Argon ions are an accomplished by laser-enhanced ionization (LEI) with subintegral part of energy transfer to the plasma; therefore, their sequent mass spectrometric detection.At a temperature of about 2300K,17 an air–hydrogen flame is capable of ionizing most of the alkali and alkaline earth elements to a significant degree.18 Therefore, it can also be an eVective ion source for †Present address: Department of Chemistry, University of Missouri at Columbia, Columbia, MO 65211, USA.these elements. J. Anal. At. Spectrom., 1999, 14, 669–674 669the National Institute of Standards and Technology (Gaithersburg, MD, USA) in 0.2 M HNO3 prepared locally by sub-boiling distillation. Sample solutions contained a similar acid concentration. Procedure An air–hydrogen flame was used for all the elements studied except Ca, for which an oxygen-enriched air–hydrogen flame was used. The gas flow rates for the air–hydrogen flame were 4.3 L min-1 for air and 3.7 L min-1 for hydrogen.For the spectral background studies, a mass scan consisting of six sweeps over the entire mass range (1–250 u) was performed with a blank solution. The total dwell time at each mass was 1.8 s. For the molecular ion studies, four single element solutions containing 100 ppm Na, 10 ppm K, 40 ppm Ca and Fig. 1 Schematic diagram of a hydrogen burner and MS interface. 10 ppm Fe were used.Mass scans from 39 to 97 u were performed for each solution. A calibration curve for K was constructed using solutions Table 1 Instrument parameters containing 0, 0.078, 0.38, 1.9, 9.3, 31 and 100 ppm K. For K Mass spectrometer Perkin-Elmer SCIEX Elan 5000 isotope ratio measurements, five replicate measurements were Spectral resolution 0.8 u (normal ) made on solutions containing 32 and 100 ppm K. The total Sample uptake rate 0.94 mL min-1 dwell time for each measurement was 36 and 144 s at masses Ni sampler and Ni skimmer 0.5 mm 39 and 41, respectively.orifice diameter An oxygen-enriched air–hydrogen flame was used for the Spacing between the burner 20 mm studies on Ca. The flow of the oxygen was regulated using the and the sampler Interface pressure 230 Pa (1.7 Torr) oxygen channel of the mass flow controller of the ELAN 5000 Quadrupole pressure 2.9 mPa (2.2×10-5 Torr) ICP-MS. The oxygen entered the burner through the auxiliary oxidant port of the spray chamber.Typical flow rates were 0.20 L min-1 of O2, 3.6 L min-1 of air and 4.6 L min-1 of H2. A calibration curve for Ca was constructed with solutions Experimental containing 0, 1.0, 2.0, 2.7, 3.9, 4.9, 6.5, 8.4, 9.3, 11, 23, 32 and Reagents and equipment‡ 40 ppm Ca. Isotope ratio measurements were made using 10 replicates of a 40 ppm Ca solution. The total dwell time for An Elan 5000 ICP-MS (Perkin-Elmer SCIEX, Thornhill, each measurement was 5, 35, 5 and 35 s at masses 40, 44, 57 Ontario, Canada) and a pre-mix burner and nebulization and 61, respectively. For studies of ionization interferences, a system designed for flame atomic absorption spectrometry blank and five samples were prepared.The five samples from Perkin-Elmer (Norwalk, CT, USA) were adapted for contained 10 ppm Ca, 10 ppm K, 10 ppm K plus 10 ppm Ca, this work. A schematic diagram of the FIMS is shown in 100 ppm Na and 100 ppm Na plus 10 ppm Ca. The count Fig. 1.Model 604 and 603 tapered rotameter flow meters from rates at masses 40, 44, 57 and 61 were measured. Matheson Gas Products (Secaucus, NJ, USA) were used to The detection limit was determined as the concentration control the flows of the air and the hydrogen, respectively. A giving a signal equivalent to three times the noise, calculated capillary burner head was used rather than the normal slotfrom the standard deviation of 11 repetitive measurements type burner head. It was constructed with a bundle of 43, with 3 s integration of the background intensity. 4 cm×1 mm id stainless steel capillaries clustered into 1 cm diameter and mounted in a water-cooled holder in the position where the ICP torch normally is located.The burner head was Results and discussion connected to the spray chamber with a 3 cm id flexible hose. General properties of the hydrogen FIMS A Gilson (Middleton, WI, USA) peristaltic pump was used to regulate the sample uptake.The vacuum pressure of the When a commercial ICP-MS instrument is retrofitted with a instrument was maintained by decreasing the orifice of both sampler of a smaller orifice, the geometry of the supersonic the sampler and the skimmer to 0.5 mm, and by increasing jet changes. It is important to ensure that the skimmer is the pumping capacity by using a RUVAC WAU-250 Roots within the boundary of the Mach disk to minimize interactions blower from Leybold Vacuum Products (Export, PA, USA) that could alter the chemical composition of the sample in series with the instrument roughing pump.Water was extracted from the ion source.14 Some researchers have sugremoved from the vacuum line by a Reliance Glass Works gested that the optimum sampler–skimmer separation for an (Bensenville, IL, USA) Model R-7210 liquid nitrogen trap Ar plasma ion source is two thirds of the Mach disk location installed upstream of the Roots blower. A Convectron Model relative to the sampler.19 The position of the Mach disk was 275 vacuum gauge from Granville-Phillips (Boulder, CO, calculated19 using the measured interface pressure (see Table 1) USA) was also installed to monitor the pressure in the interface and the diameter of the sampler orifice. This calculation places region.Control over the interlocks of the mass spectrometer the location of the Mach disk approximately 7 mm behind the was achieved by using the service mode of the Elantools orifice.Therefore, the skimmer, which was about 6.1 mm from software. Table 1 gives the instrument parameters used. the sampler orifice, was inside the Mach disk. All standard solutions were prepared by appropriate The mass spectrum of a 0.2 M nitric acid solution is shown dilutions of 3100 Series Standard Reference Materials from in Fig. 2. The H3O+, CH2O+, NO+ and NO2H+ peaks found in the background spectrum of an acetylene FIMS were not ‡Certain commercial instruments are identified in this paper to specify observed in the hydrogen FIMS.15 The temperature of an air– adequately the experimental procedure.Such identification does not hydrogen flame is about 150 K lower than that of an imply recommendation or endorsement by the National Institute of air–acetylene flame.17 The lower temperature and a lack of Standards and Technology, nor does it imply that the equipment identified is necessarily the best for the purpose. flame carbon relative to the air–acetylene flame are probably 670 J.Anal. At. Spectrom., 1999, 14, 669–674sensitivity of 4 counts s-1 mM-1 [the equivalent of 100 counts s-1 (ppm)-1 at mass 39] is needed for an analyte. Substitute Ia, Vai and ba in eqn. (6) with the sensitivity of 350 counts s-1 mM-1, the ionization potential of 4.3 eV and atomization factor of 0.4 for K.17 Substituting Ib in eqn. (6) with the limit sensitivity of 4 counts s-1 (ppm)-1 and using a flame temperature of 2300 K, eqn. (6) is reduced to Vbi=6.3+0.46 log bb (7) The ionization potential Vbi has a maximum of 6.3 eV since bb cannot be greater than 1.Hence, the usefulness of the hydrogen FIMS is limited to elements with ionization poten- Fig. 2 Mass spectral background of a blank containing 0.2 M nitric tials below 6.3 eV, which include most of the alkali and acid. alkaline earth elements. If the atomization factor of an analyte is less than 1, the corresponding Vbi for the element is lowered. An example is Ca, which has an atomization factor of 0.15 in directly or indirectly20 responsible for the absence of these an air–hydrogen flame.17 Vbi for Ca is calculated to be 5.9 eV, background species.The background intensity of the hydrogen which is less than the first ionization potential of 6.1 eV for FIMS at any mass was below 14 counts s-1. Data from the the element. This suggests that Ca+ is not sensitive enough to six replicates of a scan showed that the average of the be useful for isotope ratio measurement with the hydrogen intensities at any given mass was smaller than the standard FIMS, and therefore the measurement of Ca has to be made deviation of the intensities at the mass, suggesting that the with other alternatives, as will be discussed later.The limited features observed in the background scan were due primary ionization capability of the hydrogen FIMS is an advantage to statistical noise. The hydrogen flame is, therefore, a clean rather than a disadvantage for the determination of the alkali source for mass spectrometry.and alkaline earth elements, because it promises a simpler The signal intensity of a 10 ppm K solution was about mass spectrum relative to that of ICP-MS for samples of 9×104 counts s-1 at mass 39, yielding a sensitivity of 350 complex matrices. counts s-1 mM-1. With some reasonable assumptions, the sensitivity and hence the eYcacy of the FIMS for the other Molecular ions elements can be estimated by using the ionization potential of the analyte.The ion number density in the hydrogen flame is The primary application of the hydrogen FIMS research is related to the ionization potential of the element by the Saha the isotope ratio measurements of K and Ca for which the equation:21 conventional ICP-MS technique is inadequate because of the isobaric interferences. Na and Fe are present in most sample log [M+]2 [M] =-5040 Vi T +2.5 logT-6.5 (1) matrices; therefore, they were included in the study.Spectra of the four elements using single element solutions are shown where [M+] and [M] are the ion and atom number densities, in Fig. 3. respectively, of the analyte, Vi is the ionization potential of The mass spectrum of a 100 ppm Na solution showed two the analyte in electronvolts and T is the flame temperature peaks at K masses, one at mass 39 with 700 counts s-1 and (2300 K). The mass spectrometric signal intensity is directly the other at mass 41 with 3000 counts s-1 [Fig. 3(a)]. The proportional to the ion number density of the isotope: peak at mass 39 was probably K from the contamination of the burner head after the prolonged use with high concen- [M+]=kI (2) trations of K solutions. The more intense peak at mass 41 where I is the signal intensity and k is the FIMS response cannot be explained by the K contamination because of the constant for the element. The atom number density is directly low natural abundance of 41K.The probable designation for proportional to the free atom fraction of the analyte and the this peak was Na(H2O)+. The background from the molecular molarity of the analyte solution aspirated: ions of the 100 ppm Na solution was the equivalent of about 2 ppm natural K at mass 41. A separation may be necessary [M]=fbC (3) for isotope ratio measurement of K in samples containing where f is a constant of the flame, b is the atomization factor high concentrations of Na.Alternatively, the interference on of the analyte and C is the analyte molar concentration. Let 41K may be avoided by preparing the samples in D2O. us assume that the mass response curve of the spectrometer is The mass scan of a 10 ppm K solution [Fig. 3(b)] showed flat and therefore the response constant k is approximately a peak at mass 57 with 1000 counts s-1 and another at mass the same for all elements. Then for isotopes of element a and 59 with 100 counts s-1 in addition to the peaks at K masses.element b of the same molarity (Ca=Cb), the following applies: The probable species responsible for the peaks were the hydrated analyte ions 39K(H2O)+ and 41K(H2O)+. The intenlog [Ma+]2 [Ma] =log (kIa)2 fCaba =-5040 Vai T +2.5 log T-6.5 sities of K+ at masses 39 and 41 were 1×105 and 1×104 counts s-1, respectively. Assuming a flat mass response curve (4) of the FIMS in the mass range of this study the K+ species were estimated to account for over 99% of the total K log [Mb+]2 [Mb] =log (kIb)2 fCbbb =-5040 Vbi T +2.5 log T-6.5 containing ions.The mass scan of a 40 ppm Ca solution [Fig. 3(c)] showed (5) four groups of peaks in the mass range studied. The first group from masses 39 to 44 included peaks of 39K+ as a Subtracting eq. (4) from eq. (5) we obtain contaminant at 900 counts s-1, 40Ca+ at 3000 counts s-1 and 44Ca+ at 80 counts s-1. The second group included peaks log Ib2 Ia2 =-5040 Vbi-Vai T +log bb ba (6) from masses 57 to 65.Based on the intensities of these peaks relative to one another, the probable identifications of the peaks were CaOH+ species at masses 57 (30 000 counts s-1), For isotope ratio measurements to be practical, a minimum J. Anal. At. Spectrom., 1999, 14, 669–674 671ure relative to the local dew point.22 The gas may become supersaturated at some stage of the expansion, resulting in homogeneous nucleation and the formation of the hydrate ions.22 These hydrate ion species are rarely observed in ICP-MS, probably owing to the high temperature of the plasma and the low partial pressure of water relative to that of Ar.The presence of the hydrates of Na+, K+, Ca+ and CaOH+ complicates the mass spectra of the hydrogen FIMS. Analyte ion–solvent clusters are frequently observed23,24 in the spectra obtained using electrospray mass spectrometry (ES-MS). The extent of the analyte ion–solvent cluster depends on the electric field strength between the sampler and the skimmer.Increasing the voltage promotes collisions in the free jet,23 resulting in diminished hydrate ion intensities relative to singly charged analyte ions.23,24 This de-clustering principle of ES-MS should also be applicable to the hydrogen FIMS, and the hydrate ions in the hydrogen FIMS probably could be minimized by appropriate biasing of the sampler and the skimmer. No attempt was made to do so in the present work. Analytical performance with potassium Taylor et al.15 reported that with an air–acetylene flame, a 100 ppm Cs solution aspirating between each sample analysis was essential to keep the K signal stable.Without the Cs solution the K signal would drift down to the background level in 10–15 min.15 This phenomenon was not observed with the air–hydrogen flame. Calibration standards of K in 0.2 M nitric acid were measured and the intensities were plotted against the concentrations of K. A linear curve (correlation 0.9997) spanning over three orders of magnitude was indicated.The detection limit for K was about 0.2 ppb. This is comparable to the 0.2–0.3 ppb reported by Taylor et al.15 using an acetylene FIMS and it is slightly better than the 1 ppb obtained with ICP-MS;25 however, a much better detection limit of 0.1 pg mL-1 can be obtained with helium ICP-MS.10 Two solutions containing natural K were used for isotope ratio measurements. The mass bias calculated from the measured ratio of 13.51 was about 2.5% lower relative to the accepted ratio of 13.86, compared with the 9% bias observed in a cool Fig. 3 Mass spectra of (a) 100 ppm Na, (b) 10 ppm K, (c) 40 ppm Ca plasma.8 The results for five replicate measurements gave 0.19 and (d) 10 ppm Fe. and 0.14% relative standard deviation (RSD) for 32 and 100 ppm K, respectively. A mass scan of a 1000 ppm K solution is shown in Fig. 4. 59 (300 counts s-1), 60 (80 counts s-1), 61 (1000 counts s-1) The magnitudes of the three peaks at masses 39, 40 and 41 and 65 (60 counts s-1), and 40Ca(H2O)+ at mass 58 (500 were 1.63×106, 238 and 1.19×105 counts s-1, respectively, counts s-1).The third group included peaks from masses 75 yielding abundances of 93, 0.014 and 6.8% (not corrected for to 83. The probable identifications for these peaks were mass bias), respectively. They were in good proportion relative CaOH(H2O)+ species at masses 75 (20 000 counts s-1), 77 to the accepted abundances of 93, 0.012 and 6.7% for the (200 counts s-1), 78 (30 counts s-1) and 79 (300 counts s-1) corresponding K isotopes.The fact that the peak at mass 40 and Ca(H2O)2+ at mass 76 (30 counts s-1). The fourth group is well resolved from those at 39 and 41 suggests that the included peaks from masses 93 to 97. The probable identifi- measurement of 40K is possible. cations for these peaks were CaOH(H2O)2+ at masses 93 Potassium-40 is sometimes used for mineral dating.26 The (1000 counts s-1), 95 (20 counts s-1) and 97 (40 counts s-1).Assuming a flat mass response curve of the FIMS, the relative abundance of the Ca species was about 55, 37, 5, 2 and 1% for CaOH+, CaOH(H2O)+, Ca+, CaOH(H2O)2+ and Ca(H2O)+, respectively. Therefore, CaOH+ was the dominant Ca species in the FIMS. The spectrum of a 10 ppm Fe solution [Fig. 3(d)] showed two major peaks at masses 39 and 56, respectively. The peak at mass 39 was probably K contaminant from the burner head discussed earlier.The peak at mass 56 with an intensity of about 120 counts s-1 was from Fe. Apparently, the high ionization potential of Fe (7.87 eV), relative to the limit of 6.3 eV discussed earlier, was responsible for the low sensitivity. The hydrate ions observed in Na, K and Ca spectra were probably formed during the supersonic jet expansion.22 As the gas expanded from the flame into the vacuum through the Fig. 4 Mass spectrum of a 1000 ppm K solution. sampler orifice, its temperature dropped faster than the press- 672 J.Anal. At. Spectrom., 1999, 14, 669–674previously. A measurement of a 2.7 ppm Ca solution with the air–hydrogen FIMS gave a count rate of about 160 counts s-1, or a sensitivity of about 60 counts s-1 (ppm)-1. Since the rate of ionization is exponentially proportional to the temperature of the flame, the higher the flame temperature, the higher is the sensitivity. An oxygen–hydrogen flame would be more desirable for Ca because of its higher temperature relative to an air–hydrogen flame; however, the oxygen–hydrogen flame is extremely susceptible to flash back.In this work, an oxygen enriched air–hydrogen flame was used to enhance the Ca sensitivity. As the flow of the air was decreased, the flow of the oxygen was increased, and the flow of hydrogen was also increased in order to maintain the linear flow velocity of the gas mixture. This procedure reduced the chance of a flash back. The sensitivity of Ca was increased by a factor of about Fig. 5 Calibration curve for Ca. 8 in the oxygen-enriched flame relative to that in the air– hydrogen flame. Fig. 5 shows the calibration curve of Ca using capability of the FIMS for the measurement of 40K can be signal intensities at mass 40. The curve is linear up to about compared with radiochemical scintillation counting. At the 10 ppm (correlation 0.9993), beyond which it bends toward signal intensity of the experiment, it would take about 70 min the concentration axis.The curvature at concentrations higher to collect 106 counts at mass 40. A 70 mL volume of 1000 ppm than 10 ppm is probably a result of auto-ionization suppres- K solution would be needed with a sample uptake of 1 sion, since the element is highly susceptible to ionization mL min-1 of the current nebulizer. The sample uptake can be interferences to be discussed later. The detection limit was reduced easily without sacrificing the sensitivity or the signal about 2 ppb at mass 40 compared with the 2 ppb at mass 44 stability by using a direct injection high-eYciency nebulizer by ICP-MS.25 As discussed earlier, the sensitivity of CaOH+ (DIHEN).27 An estimated 7 mL of the solution are needed was about 10 times higher than that of Ca+.Consequently, a with a sample uptake of 0.1 mL min-1 of the DIHEN;27 better detection limit of 0.2 ppb was obtained at mass 57. therefore, the calculated volume of 70 mL is the upper limit.A 40 ppm Ca solution was used to demonstrate an isotope The 40K in 70 mL of the solution is about 1.30×1017 atoms. ratio measurement. Signal intensities at masses 40, 44, 57 and Assuming a 100% detector eYciency, the time it would take 61 were recorded. Results of 10 replicate measurements are to collect N counts is summarized in Table 2. The ratios of 0.0241 and 0.0249 obtained for Ca+ and CaOH+ species are about 12–16% t= Nt1/2 N0 ln 2 (8) higher than the accepted ratio of 0.0215 for 44Ca/40Ca.On the per u basis the bias is about 3–4%. Although this mass bias is where t1/2 is the half-life of 40K (1.27×109 years) and N0 the larger than the per u bias of about 1% for K discussed number of 40K atoms in the K solution at the beginning. A previously, it is within the range of the mass bias reported in simple calculation shows that with 70 mL of 1000 ppm K, the literature.8 The discrepancy between the mass bias obtained scintillation counting would take over 5 d to collect the same with K and Ca in this work is probably attributable to the number of counts.Therefore, the mass spectrometric method diVerent flames used for the two elements. has the potential to be more practical than scintillation count- The feasibility of using CaOH+ species for isotope ratio ing for 40K determinations. analysis was evaluated by comparing the isotope ratio and the ratio precision at CaOH+ masses with those at Ca+ masses. Analytical performance with Ca It is expected that the intensity ratio at CaOH+ masses would be identical with that of the corresponding Ca+ masses.The According to the discussion on the general properties of the 3% diVerence between the two ratios in Table 2 is probably hydrogen FIMS, Ca has a sensitivity below the limit of 4 counts s-1 mM-1 [or about 100 counts s-1 (ppm)-1] set forth due to the mass bias since these ratios are not corrected for Table 2 Isotope ratio measurement of a 40 ppm Ca solution Average intensity at mass Ratio 40 44 57 61 44/40a 61/57a 6659 160.7 8.361×104 2085 2.41×10-2 2.49×10-2 RSD (%) 8 9 7 7 3 0.5 aNot corrected for mass bias.Table 3 Intensity and the standard deviation (n=5) of K, Na and Ca solutions at Ca and CaOH masses Samplea 40 (40Ca+) 44(44Ca+) 57(40CaOH+) 61(44CaOH+) Blank 1±0.3 0.8±0.2 4±0.8 1±0.4 Ca 3000±40 70±2 40000±200 900±200 Na 1±0.3 0.9±0.3 8±0.7 0.8±0.4 Na+Ca 50±4 2±0.5 700±10 20±2 K 100±3 0.8±0.3 3000±20 1±0.4 K+Ca 300±3 4±0.9 5000±80 50±2 aNa, 100 ppm Na; K, 10 ppm K; Ca, 10 ppm Ca.J. Anal. At. Spectrom., 1999, 14, 669–674 673the bias. The measured ratio precision of 0.5% RSD at CaOH+ References masses is much smaller than the 10% predicted by the law of 1 M. A. Vaughan and G. Horlick, Appl. Spectrosc., 1986, 40, 434. error propagation with 7% RSD each at masses 57 and 61 2 S. H. Tan and G. Horlick, Appl. Spectrosc., 1986, 40, 455. (Table 2), suggesting that the intensities at masses 57 and 61 3 A. Montaser, H.Tan, I. Ishii, S. Nam and S. Cai, Anal. Chem., are highly correlated. This high correlation substantiates the 1991, 63, 2660. viability of using CaOH+ species for isotope ratio analysis. 4 L. C. Alves, D. R. Wiederin and R. S. Houk, Anal. Chem., 1992, The ratio precision listed in Table 2 is apparently limited by 64, 1164. 5 A. Montaser and H. Zhang, in Inductively Coupled Plasma Mass the counting statistics since the total counts measured at each Spectrometry, ed.A. Montaser, Wiley-VCH, New York, 1998. mass are much less than 106. As a result, a better ratio 6 J. W. Lam and J. W. McLaren, J. Anal. At. Spectrom., 1990, precision was obtained at CaOH+ masses because of the 5, 419. higher sensitivity relative to that at Ca+ masses, thus making 7 F. G. Smith, D. R. Wiederin and R. S. Houk, Anal. Chem., 1991, CaOH+ the species of choice for isotope ratio measurements. 63, 1458. The eVects of easily ionized elements (EIE) Na and K on 8 S.Jiang, R. S. Houk and M. A. Stevens, Anal. Chem., 1988, 60, 1217. Ca signal intensities at masses 40, 44, 57 and 61 are summarized 9 S. D. Tanner, J. Anal. At. Spectrom., 1995, 10, 905. in Table 3. The intensities for single element solutions of Na 10 H. Zhang, S. Nam, M. Cai and A. Montaser, Appl. Spectrosc., and K are also included. The count rate of 100 and 3000 1996, 50, 427. observed at masses 40 and 57 for the 10 ppm K solution are 11 S. Nam, H.Zhang, M. Cai, J. Lim and A. Montaser, Fresenius’ probably due to 40K+ and 39K(H2O)+ species discussed J. Anal. Chem., 1996, 355, 510. earlier. The ionization suppression from Na and K is evident 12 G. C. Eltenton, J. Chem. Phys., 1947, 15, 455. 13 A. N. Hayhurst and T. M. Sugden, Proc. R. Soc. London, Ser. A, when the signal from the Ca solution is compared with those 1966, 293, 36. with alkali elements in addition. A 60-fold and a 20-fold 14 A. N. Hayhurst, D. B. Kittleson and N.R. Telford, Combust. suppression of Ca signal were observed when 100 ppm Na and Flame, 1977, 28, 123. 10 ppm K were added to the 10 ppm Ca solution, respectively. 15 H. E. Taylor, J. Garbarino and S. R. Koirtyohann, Appl. In addition to the multiplicative interferences due to ionization Spectrosc., 1991, 45, 886. suppression, the 10 ppm K in the Ca solution resulted in 16 G. C. Turk, L. Yu and S. R. Koirtyohann, Spectrochim. Acta, Part B, 1994, 49, 1537. additive interferences at masses 40 and 57 due to 40K+ and 17 Handbook of Flame Spectroscopy, ed.M. L. Parsons, B. W. Smith 39K(H2O)+ species. A separation may be necessary if Ca is to and G. E. Bentley, Plenum Press, New York, 1975. be determined in high concentrations of easily ionized elements. 18 Instrumental Methods of Analysis, ed. H. H. Willard, L. L. Merritt, Jr and J. A. Dean, 5th edn., Van Nostrand, New York, 1974. 19 D. J. Douglas and J. B. French, J. Anal. At. Spectrom., 1988, 3, 743. 20 J. A. Green and T. M. Sugden, in 9th International Symposium on Conclusion Combustion, ed. W. G. Berl, Academic Press, New York, 1963, The mass spectral background of the air–hydrogen flame was pp. 607–621. 21 M. N. Saha, Philos. Mag., 1920, 40, 472. very simple; however, the analyte spectra were complicated. 22 F. T. Greene and T. A. Milne, Adv. Mass Spectrom., 1964, 3, 841. The low flame temperature relative to an ICP resulted in a 23 D. J. Douglas, in Inductively Coupled Plasma in Analytical Atomic limited number of alkali and alkaline earth elements becoming Spectrometry, ed. A. Montaser and D. W. Golightly, VCH, New appreciably ionized. The isotope ratio measurement of K and York, 2nd edn., 1992. Ca was successfully demonstrated with the FIMS. Two major 24 G. R. Agnes and G. Horlick, Appl. Spectrosc., 1995, 49, 324. 25 Perkin-Elmer Technical Summary, TSMS-12, Perkin-Elmer, limitations relative to ICP-MS are severe EIE interferences Norwalk, CT, 1991. and the presence of analyte ion-solvent clusters in the spectra. 26 Nuclear and Radiochemistry, ed. G. Friedlander, J. W. Henedy, Although the molecular ions complicate the analytical spectra, E. S. Macias and J. M. Miller, Wiley, New York, 3rd edn., 1981. they can sometimes be used to advantage for elements such as 27 J. A. McLean, H. Zhang and A. Montaser, Anal. Chem., 1998, Ca. Compared with the acetylene flame, the air hydrogen 70, 1012. flame is equally capable while being more practical as an ion source for FIMS. Paper 8/09797F 674 J. Anal. At. Spectrom., 1999, 14, 669–674

ISSN:0267-9477    

DOI:10.1039/a809797f

出版商:RSC    

年代:1999

数据来源: RSC

摘要:

Laser ablation of synthetic geological powders using ICP-AES detection: eVects of the matrix, chemical form of the analyte and laser wavelength M. Motelica-Heino,*a O. F. X. Donarda and J. M. Mermetb aLaboratoire de Chimie Bio-Inorganique & Environnement, EP CNRS 132, France bLaboratoire des Sciences Analytiques, UMR 5619, Universite� Claude Bernard-Lyon I, France Received 19th October 1998, Accepted 24th February 1999 A systematic study of the influence of the physical and chemical properties of analogues of geological samples and of laser wavelength on laser ablation was conducted.DiVerent chemical forms of several test elements were investigated, first as pure elements, then mixed in various matrices. Synthetic model samples were prepared from diVerent crystalline compounds of Mg, Al and Fe spiked in SiO2 or CaCO3 and manufactured in the form of pressed pellets that were analysed by laser ablation inductively coupled plasma atomic emission spectrometry (LA-ICP-AES).This study was performed at two laser wavelengths (1064 and 266 nm) with a Nd5YAG laser and using ICP-AES as an elemental detector. These various steps made it possible to demonstrate that the LA-ICP-AES response factors of the analytes are strongly dependent on their chemical form and on the bulk composition of the matrix. These eVects were also found to be laser wavelength dependent and the use of a UV laser did not lead to any improvements to minimise these eVects.In contrast, there was no influence of the grain size or binding pressure of the pressed pellets. However, normalisation to a matrix element could compensate for diVerences in the response factors of the analytes. Implications for direct quantitative elemental analysis of geological materials by laser ablation related techniques are that calibration without matching in terms of the mineralogical and chemical composition of the matrix and of the chemical forms of the analytes can lead to systematic errors in the determination of the element concentration.many other fields of application, LA has been particularly 1 Introduction considered for geochemistry.3,4 The direct trace elemental analysis of solid samples is still a Trace element analysis of powdered or particulate material challenge for the analyst. Plasma spectrochemical methods is often required in geochemistry or environmental monitoring oVer high sensitivity and low detection limits. However, they as many natural samples such as soils and sediments occur as require solid samples to be dissolved prior to analysis, which a mixture of discrete grains.Before analysis, bulk rock samples is often diYcult and time consuming. Therefore, instrumental are also commonly reduced to powders. Powders can be techniques that allow direct analysis of solid samples with introduced into the plasma after digestion or even directly little preparation are particularly interesting in the analysis of through slurry nebulization if a micrometre particle size can diYcult solid samples such as refractory or geological mate- be achieved.5 Little consideration has been given to LA of rials.So far, physical methods such as X-ray fluorescence powders although it provides rapid analysis. Moreover, the (XRF), particle induced X-ray excitation (PIXE), electron fundamental processes involved in the LA of particulate probe microanalysis (EPMA), neutron activation analysis materials are poorly understood.Previous work with LA-ICP- (NAA) and ion microprobe (SIMS) have been extensively AES/MS on powdered samples has been conducted in the used for micro-, bulk and surface analysis. More recently, field of geology and the environment. Since the pioneering lasers have been coupled to plasma-based analytical techniques work of Gray,6 bulk analysis by LA-ICP-MS of geological such as inductively coupled plasma mass spectrometry materials using pressed pellet samples has continued to receive (ICP-MS) or atomic emission spectrometry (AES).Laser attention. Several workers have performed analysis of reference ablation (LA) allows direct introduction into the ICP and silicate rocks7–13 and carbonates.14,15 Samples were prepared expands the scope of ICP-MS/AES to any solid sample. It as pressed pellets using a binder after grinding. For environavoids the fastidious preparation steps and problems that mental monitoring, the same strategy was used for the analysis occur with dissolution such as contamination, loss of volatile of sediments by LA-ICP-MS16 and for the analysis of soils by elements and dilution of the elements.LA is a powerful LA-ICP-AES.17 DiVerent calibration strategies were used for sampling tool for the ICP as it can perform in situ analysis or major, minor and trace element analysis such as calibration be a suitable alternative to dissolution when bulk analysis is with a geological reference material, or calibration against required.LA-ICP-MS/AES is a tandem analytical technique synthetic powders spiked with analytes occurring in solid or where the laser beam is used as a sampling probe and the liquid form. Globally, only semiquantitative analysis could be ICP-MS/AES system as a detector. When targeted on the achieved, i.e., accuracy was within one order of magnitude, sample surface the laser beam generates microparticulate mate- while the use of an internal standard significantly improved rial that is brought with an argon stream into the ICP where results.However, quantitative analysis using external caliit is atomised. LA has been extensively reviewed for its bration proved possible only if close matching of the matrices of the sample and the standard could be achieved in terms of fundamental aspects1 and analytical applications.2 Among J. Anal. At. Spectrom., 1999, 14, 675–682 675bulk chemistry and, moreover, in terms of mineralogical translational x–y stage displacement with a translation rate of 10 mm s-1.The sampling pattern was a 1 mm long raster line. composition. Variability of the elemental response factors between diVerent matrices was thought to be controlled by The ablated material was transported by an argon carrier gas into the injector tube of the ICP-AES system. the physical properties of the individual minerals that host the elements. A Perkin-Elmer (Norwalk, CT, USA) Optima 3000 DV ICP-AES instrument was used for elemental analysis.The LA is a complex process that depends on both laser parameters and physico-chemical parameters of the materials. The detector is a custom segmented-array charge coupled device allowing simultaneous multi-element measurements of the line influence of the laser wavelength on elemental response factors was not investigated in previous studies using infrared (IR) intensity and adjacent background.All measurements were performed with the axial plasma viewing mode. Acquisition laser ablation (with Nd5YAG lasers operating at their fundamental wavelength). As the general trend in LA is to move started after a pre-ablation of 30 s to reach a steady signal. Elemental signals were then averaged over the acquisition towards shorter wavelengths in the ultraviolet (UV) region (working with quadrupled Nd5YAG systems or with excimer period. Each measurement was performed five times on diVerent sample locations and the mean calculated.Experimental lasers), it would be interesting to know if there are any significant improvements in terms of matrix eVects when settings for LA-ICP-AES are detailed in Table 1, together with spectral lines and integration parameters. shorter wavelengths are used. Since elements can be present in diVerent chemical and physical (e.g., grain size) forms in geological samples, it is also crucial to verify whether LA 2.3 Method eYciency is similar, regardless of the form of the element.The 2.3.1 Model samples. Synthetic powders of diVerent com- aim of this work was, therefore, to investigate the ablation position as previously described were used to prepare simple behaviour of model geological samples at two laser waveor complex matrices with controlled physical and chemical lengths using a Nd5YAG laser operating at 266 and 1064 nm. properties. Samples were prepared as 12 mm diameter pellets Synthetic mol samples were prepared as pressed pellets from with a press using a pressure of 10 tonnes without any binder.diVerent crystalline compounds of Mg, Al and Fe spiked in SiO2 and CaCO3. The aim was to move from pure compounds Single phase samples. Silica was used to prepare model to complex matrices, with controlled physical and chemical matrices with controlled physical (grain size, binding pressure) matrix properties. The eVects of matrix properties and also of properties.SiO2 powders of diVerent grain sizes (5, 10, 10–40 the chemical and physical form of the analytes on the elemental and 100–200 mm) were prepared under two pressures, 5 and response factors were investigated by using ICP-AES as an 10 tonnes, respectively. Pure compounds of Mg, Al and Fe elemental detector. Finally, the analytical matrix eVects of were used to prepare model phases with diVerent chemical LA-ICP-MS/AES were highlighted and the potential of these forms.Pure chemical synthetic powders of Mg (oxide, sulfate, techniques for direct quantitative analysis of real geological metaphosphate, acetate and fluoride), Al (oxide, phosphate, materials was evaluated. sulfate and fluoride) and Fe (reduced, oxide and sulfate) were pressed under a pressure of 10 tonnes. 2 Experimental Samples with complex matrix. Pure compounds of Al, Mg 2.1 Samples and Fe were spiked in SiO2 and CaCO3 at diVerent concen- Synthetic crystalline compounds of Mg, Al and Fe occurring trations to prepare dual phase matrices with diVerent bulk in chemical forms were selected: magnesium oxide (Fluka, composition and chemical form of the minor phases.Sigma-Aldrich, St. Quentin Fallavier, France), sulfate Magnesium oxide, sulfate, metaphosphate, acetate and fluoride (Janssen, Beerse, Belgium), metaphosphate, fluoride (Prolabo, Fontenay sous Bois, France) and acetate (Aldrich); aluminium Table 1 Instrumental operating conditions oxide (Merck, Darmstadt, Germany), sulfate, phosphate (Prolabo) and fluoride (Riedel-de Hae�n, Hannover, Germany; Laser ablation device— Sigma-Aldrich); iron oxide (Merck), reduced and sulfate Laser type Surelite Nd5YAG (Prolabo).The grain size of these powders was in the Wavelength 1064 nm 266 nm (quadrupled) 10–200 mm range. Four pure SiO2 powders (Prolabo) used for Energy 300 mJ per pulse (1064 nm) chromatography with controlled grain size were chosen to 4 mJ per pulse (266 nm) represent the silicate matrix. The diVerent grain sizes were, Laser repetition rate 10 Hz respectively, 5, 10, 10–40 and 100–200 mm.Pure CaCO3 Sampling pattern 1 mm straight line (Prolabo) was used for the carbonate matrix. Translation rate 10 mm s-1 Focus position 0 2.2 Instrumentation ICP-AES— A Nd5 YAG Continuum Surelite laser (Continuum Lasers, Type Perkin-Elmer Optima DV Santa Clara, CA, USA) was used under Q-switched operation. ICP parameters: This device can generate laser light at the fundamental wave- Rf power 1100 W length (1064 nm) or at the fourth harmonic (266 nm), pro- Nebuliser gas flow 0.65 l min-1 cessing the fundamental laser beam with harmonic generators.Integration parameters: The laser had a maximum laser energy delivery per pulse of Processing mode Area 6 mJ at 266 nm and 350 mJ at 1064 nm. The laser repetition Integration time 2 ms Sampling time 1 s rate could be adjusted to between 1 and 20 Hz. A UV beam Replicates 50 expander (BXGU-10–3x-266) from CVI Laser Corporation (Albuquerque, NM, USA) was positioned after the output of Spectral lines: Si 251 nm the laser to limit divergence of the laser beam when working Mg 279 nm with the 266 nm wavelength. The laser beam was focused Al 396 nm perpendicularly onto the sample surface via a plano-convex Fe 238 nm 200 mm focal length lens.The sample was mounted in a Ca 422 nm movable ablation cell controlled by step motors, allowing 676 J. Anal. At.Spectrom., 1999, 14, 675–682powders were spiked in a SiO2 matrix at relative phase proportions of 10, 20, 30 and 40%, respectively. MgO was spiked in a CaCO3 matrix at 10, 20, 30 and 40%. Aluminium oxide, phosphate, sulfate and fluoride were spiked in a SiO2 matrix at phase proportions of 10, 20, 30 and 40%, respectively. Al2O3 was spiked in a CaCO3 matrix at 10, 20, 30 and 40%. Fe (reduced, oxide and sulfate) was spiked in a SiO2 matrix at 10, 20, 30 and 40%. Fe was spiked in a CaCO3 matrix at 10, 20, 30 and 40%.Powders were mixed and ground with a pestle before being pressed. 2.3.2 Analysis by LA-ICP-AES. LA-ICP-AES analysis of the SiO2 samples. The SiO2 samples were analysed for Si by LA-ICP-AES with the laser working at 266 nm. Each analysis was performed for laser pulse energies of 1 and 4 mJ. LA-ICP-AES analysis of the single phase samples. The samples of Mg (oxide, sulfate, metaphosphate, acetate and fluoride), Al (oxide, phosphate, sulfate and fluoride) and Fe (reduced, oxide and sulfate) were analysed by LA-ICP-AES at 1064 and 266 nm using the standard conditions previously defined.Mg, Al and Fe were measured for the three diVerent families of samples, respectively. LA-ICP-AES response factors Fig. 1 EVects of grain size (a) and binding pressure (b) on the response for Mg, Al and Fe were obtained for each sample in the of Si in the SiO2 matrix using the laser at 266 nm. following way: the response factor for element E, R(E), was calculated as the AES elemental mean signal I(E) divided by the elemental concentration in the matrix C(E), i.e., R(E)= deviation) between five diVerent analyses. Within each set of I(E)/C(E).experiments the signal appears to be proportional to the energy. Although a slight increase can be observed in the LA-ICP-AES analysis of the complex matrices. The samples 10–40 mm size range at 4 mJ, the signal is almost independent of Al, Mg and Fe spiked in SiO2 or CaCO3 were analysed by of the grain size.Results show that the ICP-AES response is LA-ICP-AES at 1064 and 266 nm, respectively, using the mainly influenced by the energy level. In contrast, the eVect operating conditions previously defined. Mg, Al and Fe were of the grain size is almost negligible. There is no significant measured together with Si or Ca. LA-ICP-AES response eVect of the binding pressure on the signal when the binding factors were calculated as described above.For the samples pressure changes from 5 to 10 tonnes regardless of the laser with a SiO2 matrix, response factors relative to Si were energy and the grain size. calculated as response factors divided by the response factor The laser spot size was of the order of 20 mm and the grain of Si: I(E)/I(Si) C(Si)/C(E). sizes were above and below the laser spot size. Therefore, results show that the laser–powder interaction corresponds to a true ablation process and not to a dislocation of the 3 Results and discussion compressed grains, regardless of the grain size.A binding 3.1 LA-ICP-AES response factors for single phase matrices pressure of 10 tonnes was suYcient to neutralise the eVect of the grain size. The response factors were indeed independent First, LA was investigated for pure chemical compounds. The of the structural characteristics of the powder such as the aim was to understand the eVects of the physical and chemical grain size and binding pressure.These results clearly suggest form of the analytes on LA-ICP-AES response factors together that the eVects of the physical properties of the matrix in with the role of laser wavelength. terms of grain size do not significantly influence the LA processes when powders are prepared as pressed pellets, at 3.1.1 EVects of the physical properties of the matrix. First, it is important to understand whether the ablation of powders least under the conditions used in this work.In fact, laser sampling of pressed pellets was found to be true ablation and consists of either grain removal or dislocation of the matrix. Therefore, the eVect of the physical properties of the pressed not dislocation of the grains, and the amount of ablated material was not dependent on the grain size of the powders. pellets on the ablation behaviour was investigated for the SiO2 model samples. Si was chosen for this study because it is a However, for bulk alysis of trace elements in powdered materials prepared as pressed pellets, the grain size would major geochemical component.The Si emission signal was used to monitor the amount of material removed from the certainly play a significant role. The distribution of trace elements within the matrix is likely to be dependent on the matrix by the laser. The relative dimensions of the laser spot and the grains are likely to be important in the sampling grain size. Therefore, the sample has to appear homogeneous within the sampling volume probed by the laser beam to be process.The laser spot size was of the order of 20 mm when using the UV wavelength and the grain sizes (respectively, 5, representative of the distribution of trace elements. This would certainly depend on the grain size of the trace elements and 10, 10–40 and 100–200 mm) were selected above and below the laser spot size. The laser sampling area was a 1 mm long on their concentration. Thus, to eliminate the influence of grain size on the distribution of trace elements, the sampling straight line.The width of this line was governed by the laser spot size, i.e., 20mm. Therefore, the area of the sampling area has to be larger than the matrix grain size. This should be the subject of another study. However, for all the experi- pattern was about 0.2 mm2. The length of the sampling area was set far above the grain sizes to be representative of the ments presented in this work, the dimensions of the sampling area were selected far above the grain sizes.bulk of the pellets. Fig. 1 displays the ICP-AES response factors of Si using the laser at 266 nm as a function of grain size, binding pressure 3.1.2 EVect of the chemical form of the element. The influence of the chemical form of an element on its ablation behaviour and laser energy. Error bars express the dispersion (standard J. Anal. At. Spectrom., 1999, 14, 675–682 677was studied for Mg, Al and Fe with the pure compounds dent on the laser wavelength, the response factor being generally higher in the UV mode.described above for both UV and IR laser ablation. Fig. 2 shows the ICP-AES response factors for Al, Mg and Fe for the single-phase matrices as a function of the chemical form 3.2 LA-ICP-AES response factors for complex matrices of these elements and the laser wavelength. It can be seen that After studying the eVect of the chemical form of an element the response factors obtained for Mg are much higher (at least on elemental response factors with pure compounds, the next five times) for the UV wavelength than for IR in all samples.step was to repeat the experiments with analytes spiked in a Response factors also vary with the chemical form of Mg. matrix. SiO2 and CaCO3 matrices were used for this study as This is most pronounced for the IR. The organic form of Mg being representative of two important geochemical poles, displays a higher signal response than the inorganic forms for silicates and carbonates.The combined influence of the laser both laser modes. These results clearly show that the response wavelength, chemical form, the nature of the matrix and the factor of Mg is significantly dependent on the laser wavelength concentration of the analyte in the matrix on the ICP-AES and also on the chemical form of Mg. The IR ablation mode response factor was evaluated. Samples with diVerent com- is less sensitive than UV to the chemical form.This trend can pounds of Mg, Al and Fe spiked in SiO2 or CaCO3 at diVerent also be observed with Al. There are also response factor concentrations, as described above, were used for this study. variations between the diVerent compounds of Al. The larger For both UV and IR laser ablation modes, ICP-AES variations are obtained with UV. The response factor is also responses of Mg, Al and Fe versus their concentration in the much higher (at least three times) in the UV mode than for complex matrices are presented in Fig. 3. The variations of IR. A slightly diVerent trend can be observed with Fe. the ICP-AES signal of Mg with its concentration in the matrix Response factors are similar for UV and IR ablation modes show diVerent patterns according to the chemical form of Mg, regardless of the chemical form of Fe but vary between the the nature of the matrix and the laser wavelength. For most diVerent forms.In this case, results obtained show that the experiments, the signal for Mg varies linearly with the concen- response factor of Fe is dependent on the chemical form of tration from 0 to 15%. The response of the organic form is Fe, regardless of the laser wavelength. In summary, for the non-linear for both UV and IR modes. In the CaCO3 matrix, three families of model samples tested, the LA-ICP-AES the response factors appear to reach a plateau after 15%. The response factor was strongly dependent on the chemical form results show that the slope of Mg signal versus concentration of the element for both UV and IR ablation.This trend was significantly more pronounced in the UV laser ablation mode but it was not a general rule. IR laser ablation appeared to be less sensitive than UV ablation to the chemical form of the compound. The LA-ICP-AES response factor was also depen- Fig. 3 LA-ICP-AES responses versus concentration for pure com- Fig. 2 LA-ICP-AES response factors calculated for Mg (a), Al (b) pounds spiked in SiO2 and CaCO3 matrices for Mg (a), Al (b) and Fe (c).and Fe (c) in pure compounds. 678 J. Anal. At. Spectrom., 1999, 14, 675–682curve is constant for all the inorganic forms of Mg spiked in ship between the ICP-AES response of Mg, Al and Fe and the elemental concentrations was linear up to at least 10%, SiO2 whereas it is constant for a lower concentration range (under 15%) except in the CaCO3 matrix.Therefore, the ICP- which is a large range. For both UV and IR laser ablation, this trend was noticed for the SiO2 and CaCO3 matrices and AES response factor of Mg is independent of the concentration for these conditions. The relationship between the signal and regardless of the chemical form of Mg, Al and Fe. Thus, this result indicates that the response factor of an analyte is the concentration of Al in the diVerent matrices is also linear up to 20% except in the IR mode for the sulfate and fluoride constant when its concentration in the matrix changes.However, this might not be true over larger concentration compounds where the linear concentration range is limited to 5%. Similar conclusions to those for Mg can be obtained. The ranges as the composition of the host matrix changes signifi- cantly. Normalisation to a matrix element can compensate for relationship between the signal of Fe and the concentration is also linear up to 40% except for the experiment with the changes in matrix composition.At this stage, the ICP-AES response factor appeared to be significantly dependent on the CaCO3 matrix where the signal behaves linearly in a smaller range (up to 30%). The response factor of Fe appears to be chemical form of the element and laser wavelength and to a lesser extent on the matrix composition. The concentration of independent of the concentration over a large concentration range. The ICP-AES responses normalised to Si versus the an analyte in the matrix occurring as a minor element does not have an eVect on the response factor.relative concentration (i.e., elemental concentration divided by the concentration of Si) for Mg, Al and Fe in SiO2 are Response factors for Mg, Al and Fe in SiO2 and CaCO3 matrices were derived from previous data using a simple linear presented in Fig. 4. First, it can be noted that the relative signal for Mg varies linearly with the relative concentration regression in the concentration range where the ICP-AES response was linear.The aim was to compare elemental within the whole concentration range. Al and Fe present a similar trend. The slopes of the diVerent curves are diVerent response factors according to the chemical form of the element, the nature of the matrix (SiO2 or CaCO3) and the LA mode. with respect to the chemical form of the analyte, the nature of the matrix and the laser wavelength. However, at least for First, response factors calculated for Mg are higher for the UV laser mode than for the IR mode.For the UV mode, the Mg and Al, the diVerences between the experiments are less important than the simple response factors. In contrast to response factor for MgO is diVerent with respect to the matrix in which it was included (2.0×107±0.72×107 and what was noticed for simple response factors, the relationship between the relative response and the relative concentration is 9.5×107±2.1×107, respectively, for SiO2 and CaCO3).In contrast, for IR the response factor is similar regardless of the always linear. In this set of experiments, the eVects of matrix composition matrix (7.2×106±1.2×106 for SiO2 and 5.3×106±0.63×106 for CaCO3). Response factors show the same dependence on and concentration of the analyte in the matrix on the response factors were evaluated. For the complex matrices, the relation- the chemical form similarly to what was previously observed for the pure compounds. Response factors calculated for Al are slightly higher or similar for UV than for IR in contrast to what was observed with pure compounds of Al.The response factors for Al2O3 in SiO2 and CaCO3 are not significantly diVerent (respectively, 2.1×105±6.3×105 and 1.9×105±0.23×105). They show the same dependence on the chemical form of Al as for the pure compounds. The response factors calculated for Fe show the same general trend for the chemical form of Fe as for the pure compounds except for Fe2O3.In the UV the response factor is lower in SiO2 than in CaCO3 (respectively, 8.4×103±2.9×103 and 2.6×104±0.72×104). The same trend occurs in the IR mode but with less variations (6.5×103±0.67×103 for SiO2 and 1.6×104±0.44×104 for CaCO3). Because the eVects of concentration were not significant for relative ICP-AES responses, relative response factors for Mg, Al and Fe in the SiO2 matrix, i.e., response factors normalised to Si, were derived from previous data using simple linear regression.Relative response factors allow one to compensate for variations in the amount of ablated material between the diVerent samples and thus compare the role of chemical form and laser wavelength on the ablation eYciency. In contrast to simple response factors, normalised response factors calculated for Mg, Al and Fe are similar in the UV and IR laser modes. There seems to be no significant variations for the normalised response factors calculated for Mg between diVerent forms of Mg for both UV and IR laser modes.For Al similar conclusions can be obtained except for the phosphate form which has a higher response. In contrast to Mg and Al, relative response factors calculated for Fe are dependent on the chemical form. They show the same trend according to the chemical form as for simple response factors but with lower fluctuations. These results suggest that response factors normalised to Si are less dependent on wavelength and also far less sensitive to the chemical form than the simple response factors.Globally, response factors presented the same dependence on the chemical form as for pure compounds. The LA-ICP- Fig. 4 Relative response versus concentration ratio for pure compounds spiked in SiO2 for Mg (a), Al (b) and Fe (c). AES response factors were also dependent on laser wavelength, J. Anal. At. Spectrom., 1999, 14, 675–682 679the response factor being generally higher in the UV, but sediments were analysed by LA-ICP-MS (Nd5YAG 1064 nm) by Williams and Jarvis.9 Variations of the response factors diVerences between UV and IR for a given compound spiked in a matrix were lower than for pure compounds.The response normalised to Ba for major and trace elements varied according to the host matrix. Major elements gave similar factors were also dependent on the nature of the matrix and this influence was found to be higher in the UV than in the responses only for matrices with similar mineralogical composition.LA was thought to be controlled by the physical IR. These results underline the occurrence of two factors: an important contribution of the chemical form of the element properties of individual minerals in which the elements occur rather than by the bulk rock chemistry. For silicate geostand- and an influence of the nature of the matrix. It appeared that the eVect of the chemical form of the element and laser ards, LA-ICP-MS response factors for major, minor and trace elements showed the same dependence on matrix wavelength are predominant eVects, even if the nature of the matrix plays a role.Clearly, the influence of the chemical form properties in the UV and IR modes using a Nd5YAG laser.19 Use of shorter wavelengths led to improvements in terms of of the element is complex and is laser wavelength dependent. However, the eVects of the chemical form and laser modes sensitivity for LA but not in terms of matrix dependence.In summary, the response factors of analytes in geological were lower for the normalised response factors than for the simple response factors. Moreover, for some elements normal- materials are influenced by matrix properties such as bulk chemical composition and mineralogy. Moreover, the elemen- isation to Si could overcome the influence of the chemical form and laser wavelength. tal response factor is also likely to be dependent on the individual mineralogical phase including the element.Indeed, diVerences in the LA eYciency of diVerent mineralogical 3.3 Influence of the matrix on the ablation eYciency of the phases, related to their specific physical properties, will analytes influence the response factors of the elements included in the phases. Response factors are, therefore, likely to exhibit a The influence of the nature of the matrix on LA was studied for diVerent compounds of Mg, Al and Fe spiked in SiO2 or complex dependence on global matrix properties and on individual properties of the host mineral. Thus, variability in CaCO3 matrices (cf.Section 3.2). It appeared that the nature of the matrix had an influence on the elemental responses elemental response factors will occur between diVerent families of materials because of the dependence on global matrix factors of the analytes spiked in the matrix. EVects of the composition of the matrix were investigated with elements properties. However, diVerent elemental distribution patterns between the mineralogical phases within the same matrix spiked in the SiO2 or CaCO3 at high concentration; changes in the composition of the matrix were able to influence the would also generate variations in response factors.These matrix eVects can be overcome by using normalisation to a response factors of the analytes, although changes were limited and normalisation to a matrix element could overcome matrix element, at least for matrices of a similar nature.For more complex matrices, from a mineralogical point of view, these eVects. Although only two diVerent matrices were used in the experiments, the nature of the matrix was thought to normalisation by means of an internal standard should be performed with an element representative of the mineral. have an eVect on the response factors of the elements spiked in the matrix, whereas eVects of the matrix composition were more limited.The same trend was also noticed for LA-ICP- 3.4 Influence of the chemical form AES determination of REE and refractory elements (W, N, Zr) in model geological samples with a CO2 laser at The influence of chemical form on the LA eYciency of the analytes was studied for diVerent compounds of Mg, Al and 10.6 mm.18 Model matrices were prepared by mixing diVerent oxides (Si, Ca, Fe, Al and Mg) and were then spiked with Fe prepared as pure samples (cf.Section 3.2.1). The LA-ICPAES response factors were strongly dependent on the the elements of interest. Variations in the composition of the matrix had no significant eVect on the response factors of the chemical form of the element. The same study was performed for complex matrices using diVerent compounds of Mg, Al analytes. Other studies from the literature suggest that elemental response factors in geological materials are influ- and Fe spiked in SiO2 or CaCO3 matrices (Section 3.1.2).In general, response factors displayed the same dependence on enced by matrix properties such as bulk chemical composition and mineralogy. Indeed, the influence of the physico- the chemical form as for pure compounds. Therefore, ablation eYciency depends significantly on the chemical form of the chemistry of the matrix on the response factors for major, minor and trace elements was confirmed for a large range of element. These results are, to a first approximation, related to the melting temperatures of the compound or to its ability geochemical materials.Several workers have investigated LA of geological or synthetic materials prepared as pressed pellets to be easily decomposed. For instance, the response factor is lowest for the MgO form, which exhibits a melting tempera- with LA-ICP-MS or LA-ICP-AES but these earlier studies were mostly based only on IR lasers. Jarvis and Williams7 ture of 2826 °C, while MgSO4 and MgF2 have melting temperatures of 1127 and 1263 °C, respectively.The highest analysed basaltic rocks and sediments with ICP-MS coupled to a Nd5YAG laser operating at 1064 nm. Significant response factor corresponds to Mg(CH3CO2)2, which is not stable. A similar conclusion can be obtained for Al, where diVerences between rock types for elemental response factors relative to Ba for REE were noted. These workers concluded the lowest response factor corresponds to the most stable compound, i.e., Al2O3, with a melting temperature of 2054 °C.that samples with diVerent chemistry and mineralogy gave diVerent responses for trace elements because of the specific The role of the chemical form of an element on its response factor has not been well studied. However, the influence of location of the REE in mineral phases that had diVerent behaviour during the LA process. Morrison et al.10 performed the chemical form of the elements on their LA eYciency has also been described for some other types of matrices.LA-ICP-MS analysis of geostandards (igneous rocks and shale) also with a Nd5YAG laser at 1064 nm. One reference Similarly to what was previously observed, response factors for metallic additives in polymers were dependent on their material was selected as the standard and the other materials were analysed as samples using Y as an internal standard. chemical form within the matrix.20 Goodall and Johnson21 also noticed a compound-specific response for uranium when Relative responses for trace elements varied with the diVerent matrices but these variations were similar for elements within determining this element in lithium chloride from fuel reconditioning materials by LA-ICP-AES.A Nd5YAG laser selected geochemical groups such as REE. DiVerences in the interaction of the laser with these matrices were thought to operating at 1064, 532 or 355 nm was used and Li was chosen as the internal standard. A compound-specific be the cause of such a dispersion.Silicates, basic rocks and 680 J. Anal. At. Spectrom., 1999, 14, 675–682response of uranium(III) chloride or uranium(IV) chloride was elements in polymers.20 It was also noted that the ablation process was dependent on the chemical environment, i.e., the observed and calibration using uranium compounds other than the specific compound was unsuccessful. Eppler et al.22 structures of the other additives. Because normalisation to a matrix element can compensate for diVerences between the investigated the eVects of matrix composition and chemical form of Pb and Ba spiked in soil and sand matrices with chemical forms and also laser modes, diVerences in the ablation eYciency are related to the amount and nature of the ablated laser-induced breakdown spectroscopy using a Nd5YAG laser at 1064 nm. Emission signals of Pb and Ba present in material.DiVerent mechanisms for LA have been proposed.1 The two major mechanisms significant for solid sampling are the samples as crystalline compounds were influenced by compound speciation (oxide, sulfate, chloride or carbonate).vaporisation and ablation itself, i.e., explosion. The LA process is usually considered to be qualitatively diVerent for UV and Results presented in this work on synthetic analogues of geological materials and also from other workers showed IR. For IR ablation this process results from laser light– plasma–material interactions where fusion and vaporisation that the variability in LA-ICP-AES response factors was related not only to diVerences in matrix properties but also are predominant.In contrast, UV laser beams couple directly onto the sample. Explosion is the predominant mechanism to the chemical form of the analytes, at least for elements occurring as minor discrete phases in the matrix, and this and removal of material from the sample is mainly caused by mechanical fragmentation. As the UV laser interacts directly influence of the chemical form of the analytes on the response to LA was very important.This suggests that even if the with the sample, it is likely to be more sensitive to matrix properties and physical properties of the analytes spiked in response to LA is controlled by matrix properties and by the location in mineralogical phases as previously suggested, the matrix than IR ablation. elements present in the matrix have a specific response to the laser, depending on their chemical form.In geological materials, location in a mineral and chemical form of the 4 Conclusion element are generally interdependent in the sense that an element is enclosed in a given mineral in a specific form. Geological materials such as rocks, sediments or soils have However, these results go further and suggest that elements complex matrices, and are a mixture of several mineralogical can have a specific response to LA depending on their phases.Trace elements occur as major elements of minor chemical form. This is very important in environmental individual minerals or are included or associated with major analytical chemistry where metallic pollutants can occur in mineralogical phases in diVerent chemical forms. This work diVerent forms in natural matrices. dealt with metals spiked in solid form in synthetic analogues of geological materials. Results showed that analytical matrix eVects in LA-ICP-AES were related to diVerences in global 3.5 Influence of the laser wavelength matrix properties and individual properties of the analytes The laser wavelength appeared to be a very important such as chemical form.With a Nd5YAG laser, the use of UV parameter in the diVerent sets of experiments. For pure instead of IR wavelengths did not overcome these matrix compounds, LA-ICP-AES response factors were strongly eVects. Quantitative analysis using external calibration thereinfluenced by the chemical form of the element but this trend fore requires matching of the unknown sample and the stanwas also laser wavelength dependent as IR ablation was less dard in terms of matrix properties such as chemical and sensitive than UV to the chemical form of the compound (cf.mineralogical composition and also in terms of the chemical Section 3.1.2). The LA-ICP-AES response factors were also form of the analytes. Thus, even under consistent laser conhigher in the UV mode than for IR.For analytes spiked in ditions, the non-matching of the standard and the sample will SiO2 or CaCO3 matrices, similar conclusions as for pure lead to systematic analytical errors. Internal standardisation compounds were obtained even if the nature of the matrix with a matrix element may compensate for matrix eVects due also plays a role (cf. Section 3.2). However, diVerences to diVerences between the host matrix and chemical forms for between UV and IR were less important for some elements some elements.in complex matrices than for pure compounds. It should also be noted that working with relative responses, i.e., response factors normalised to a matrix element, reduced the eVects of References both the laser wavelength and chemical form of the analyte for some elements. 1 R. Russo, Appl. Spectrosc., 1995, 49, 14A. EVects of the chemical form of the elements and of the 2 S. A. Darke and J. F. Tyson, J. Anal. At. Spectrom., 1993, 8, 145. 3 C.Neal, Rev. Geophys., 1995, 33, Suppl. nature of the matrix on response factors could be assigned 4 W. T. Perkins, in Microprobe Techniques in the Earth Sciences, either to the ablation process itself or to the transport of the ed. P. J. Potts, J. F. W. Bowls and S. J. B. Reed, Chapman and ablated material. It seems that the laser plays a key role, Hall, London, 1995, ch. 7, pp. 291–325. because the results were significantly dependent on the laser 5 K. E. Jarvis, Chem. Geol., 1992, 53, 335. wavelength. Clearly, the LA eYciency of an element is directly 6 A. L. Gray, Analyst, 1985, 110, 551. dependent on the laser wavelength. Because elemental response 7 K. E. Jarvis and J. G. Williams, Chem. Geol., 1993, 106, 251. 8 N. Imai, Anal. Chim. Acta, 1990, 235, 381. factors were related to the physical properties of the diVerent 9 J. G.Williams and K. E. Jarvis, J. Anal. At. Spectrom., 1993, 8, 25. compounds such as melting temperature, ablation eYciency 10 C. A. Morrison, D. D. Lambert, R. J. S. Morrison, W. W. Ahlers depends on chemical form. It was also found to be dependent and I. A. Nicholls, Chem. Geol., 1995, 119, 13. to a lesser extent on the matrix. In contrast to previously 11 T. Mochizuki, A. Sakashita, H. Iwata, T. Kagaya, T. Shimamura published work, it was found that the use of an IR laser was and P. Blair, Anal. Sci., 1988, 4, 403. better in terms of minimizing the matrix eVect and chemical 12 A. A. Heuzen and J. B. W. Morsink, Spectrochim. Acta, Part B, 1991, 46, 1819. form of the element. However, the role of the laser is probably 13 W. T. Perkins, N. J. G. Pearce and T. E. JeVries, Geochim. more complex than presented in this study as the energy Cosmochim. Acta, 1993, 57, 475. available at the fundamental wavelength of the Nd5YAG laser 14 W. T. Perkins, R. Fuge and N. J. G. Pearce, J. Anal. At. Spectrom., is far higher than that available at 266 nm. The eVect of laser 1991, 6, 445. energy should, therefore, also be considered. It should be 15 N. J. G Pearce, W. T. Perkins and R. Fuge, J. Anal. At. Spectrom., noted that the use of a UV laser did not lead to significant 1992, 7, 595. 16 E. R. Denoyer, J. Anal. At. Spectrom., 1992, 7, 1187. improvements to minimise the eVects of the chemical form of J. Anal. At. Spectrom., 1999, 14, 675–682 68117 L. Moenke-Blankenburg, T. Schumann and J. No� lte, J. Anal. At. 21 P. Goodall and S. G. Johnson, J. Anal. At. Spectrom., 1996, 11, 469. Spectrom., 1994, 9, 1059. 18 S. Lin and C. Peng, J. Anal. At. Spectrom., 1990, 5, 509. 22 A. S. Eppler, D. A. Cremers, D. D. Hickmott, M. J. Ferris and A. C. Koskelo, Appl. Spectrosc., 1988, 50, 1175. 19 M. Motelica-Heino, unpublished work. 20 M. Hemmerlin and J. M. Mermet, Spectrochim. Acta, Part B, 1997, 52, 1687. Paper 8/08088G 682 J. Anal. At. Spectrom., 1999, 14, 675–6

ISSN:0267-9477    

DOI:10.1039/a808088g

出版商:RSC    

年代:1999

数据来源: RSC

摘要:

Development of a new high temperature/high pressure flow system for the continuous digestion of biological samples Christiane Gra�ber and Harald Berndt* Institut fu�r Spektrochemie und Angewandte Spektroskopie, Bunsen-KirchhoV-Str. 11, 44139 Dortmund, Germany Accepted 22nd January 1999 The development of a new high temperature/high pressure flow system that incorporates resistively heated capillaries for the continuous digestion of various samples is presented. The system allows the continuous digestion of samples at temperatures of up to 260 °C and a pressure of up to 30 MPa (300 bar).Aqueous solutions of simple organic compounds such as glucose, as well as slurries of diVerent biological materials, e.g., Bovine Liver, milk or juice, were digested. The determination of the residual carbon content was performed oV-line by ICP-OES. Compared with other continuous flow digestion systems, unusually low residual carbon contents were obtained. For the majority of biological samples digested, the residual carbon content was 1–13%, corresponding to a digestion eYciency of 87–99%.The potential of the continuous digestion system was illustrated by the digestion of three certified reference materials (Dried Whole Animal Blood, Bovine Liver and Tomato Leaves) and subsequent determination of Mn, Cu, Zn and Fe with ICP-OES. Recoveries between 76 and 109% were found. The RSDs ranged from 2.5 (Mn in Tomato Leaves) to 15% (Fe in Dried Whole Animal Blood).injection analysis and microwave digestion techniques. In an Introduction eVort to shorten digestion times and to eliminate or reduce The determination of elements in solid samples requires the other disadvantages of batch digestion procedures, flow digesdissolution or digestion/manipulation of the solid sample,1,2 if tion systems were developed.15–17 The main advantage of such one refrains from the determination of elements via electrother- flow digestion systems is the possibility for complete automal ablation,3 laser ablation4 or the direct slurry nebulization mation inherent in almost every flow system,18 in addition to approaches5 or direct determination by X-ray spectroscopy.6 others such as high sample throughput, minimized reagent Digestion of solid samples — or aqueous solutions containing consumption, reduced analyte losses and low contamination an organic matrix that can interfere with the analytical determi- and simplified sample handling.Automation is especially nation of an element — can be performed in many diVerent desirable in the analysis of waste water, soil samples, food and ways, e.g., open-vessel hot-plate digestion, and with many biological samples, as well as in process analysis. Several diVerent commercially available digestion systems. In principle, workers have developed and presented diVerent flow digestion high temperatures are essential to achieve as complete a systems in which the determination is carried out oV-line after digestion as possible.7 ‘If the temperature is increased by 10 K the sample has passed the digestion system as well as real the reaction rate is increased by a factor of 2 to 4 on average’.8 on-line systems which are directly coupled to a spec- Convectively heated pressure bomb vessel systems have proved trometer19,20 or a hydride system.21,22 In other systems digesto be the most valuable systems to guarantee complete or tion is carried out in a stopped-flow mode.23–25 There are also almost complete dissolution of solid samples because they several commercially available continuous digestion systems provide elevated digestion temperatures of about 200– (e.g., from CEM, Questron) which can be used for diVerent 230 °C.9,10 One of the most ingenious systems is the high applications.26,27 pressure asher (HPA) developed by Knapp which oVers diges- In all these systems the sample to be digested is transported tion temperatures up to 320 °C.11 Such systems only suVer through a microwave heated capillary consisting of the same from one severe disadvantage: the digestion procedures are, fluoropolymer compounds that are commonly used for batch although complete in most cases, very time consuming.microwave digestion systems, i.e., PTFE or PFA. One excep- Remedial measures concerning time consumption and labour- tion is the system presented by Gluodenis and Tyson.18 Here, intensive manipulation were taken when microwave digestion the PTFE tubing is loosely embedded in a resistive heating systems appeared since they facilitate direct heating of the block.By using PTFE tubing the maximum digestion is sample and therefore generally oVer much shorter times for restricted to less than 250 °C. The limited mechanical strength sample preparation.12–14 Because the power supply is realized of the material merely allows maximum working pressures of by microwave energy the vessel materials have to be micro- up to 35 bar.Since the mechanical strength of the material is wave-transparent. The most commonly used vessel materials decreased with increasing temperature it is not possible to are based on fluoropolymer compounds such as PTFE (poly- apply the maximum admissible temperature of 250 °C and at tetrafluoroethylene) or PFA (perfluoroalkoxy). The general the same time the maximum admissible pressure of 35 bar. problem with these materials is that, due to their limited Therefore, the usual working temperature and pressure are mechanical strength, working temperatures and pressures are much lower than these values (about 200 °C and/or 10–20 restricted to values that are not always suYcient for the bar).The system oVered by CEM, for example, utilizes a complete digestion of certain sample materials. structurally reinforced polymer digestion tube (with a restrictor at the end) that can be operated at 120 °C and 23 bar at the Several workers have tried to combine the merits of flow J.Anal. At. Spectrom., 1999, 14, 683–691 683same time. These conditions are not suYcient for the total principle behind the high temperature/high pressure digestion system required many fundamental studies. The main problem digestion of several types of sample. This can also be concluded from the reports of several workers, who found that after was to find a suitable arrangement for such a system.passing the continuous digestion system there was still particulate matter in the digested solution which had to be filtered.24,26 Experimental A special system was presented by Knapp et al.28 Here, the PTFE tubing is positioned within an autoclave which is filled Instrumentation with nitrogen to a pressure of up to 25 bar. By embedding the A simultaneous ICP-OES instrument (IRIS/AP, Thermo tubing within an autoclave the problem of the limited mechan- Jarrell Ash, Franklin, MA, USA) with axial observation of ical strength of the material is by-passed because the same the plasma and Echelle optics and a CID detector was used pressure exists inside as well as outside the tubing.28 for the determination of elements in the digested solutions.It is obvious that, at present, there exists no material that Instrumental and analytical parameters are given in Table 1. is transparent to microwaves and that is also resistant to high Calibration functions were generated from 0.1–1.0 mg l-1 temperatures and high pressures.Concerning high temperasolutions of the elements. The residual carbon was also deter- tures and high pressures, microwave heated flow digestion mined with this instrument. The calibration function for systems have already reached limitations that can hardly be carbon was generated from 0.1–1 g l-1 solutions. surpassed. However, it is possible to deal with high tempera- A total carbon analyzer (built in-house) was used for the tures and high pressures in a flow system, namely, high temdetermination of the original carbon content of solid samples perature–hydraulic high pressure nebulization (HT-HHPN), and slurried material.where an indirect resistively heated metal capillary is used.29 A standard HPLC pump, a UV-HPLC photometer, a six- A modified arrangement should provide a suitable system port-injection valve and connectors were obtained from Dr. continuous digestion. The high temperature/high pressure flow Knauer, Wissenschaftliche Gera�tebau, Berlin, Germany. system for the continuous digestion of samples that is presented A power supply was built in-house implementing a PID in this work results from a pure nebulization system, viz., (proportional, integral, derivative) controller (Horst, Lorsch, hydraulic high pressure nebulization (HHPN),30 a further Germany). development of which is HT-HHPN.29 In HHPN, a liquid is forced with the aid of an HPLC pump under a high pressure Arrangement of the high temperature/high pressure flow (50–400 bar/5–40 MPa) through a nozzle with a very small digestion system orifice (10–30 mm).In this case, nebulization is realized only by the hydraulic component of the liquid. Apart from pure The arrangement used for the high temperature/high pressure nebulization, HHPN can be considered as an interface between flow digestion system is shown schematically in Fig. 1. An HPLC techniques and atomic spectrometry.Hence, various HPLC pump was used to transport the liquid through the on-line techniques such as matrix separation,31 preconcen- system. The pump head is made of titanium. The valve used tration and trace element speciation analysis are possible.32,33 for sample injection is a six-port injection valve made of PEEK In HT-HHPN, the liquid, which is also transported with the (polyether ether ketone). The connection tubes to the digestion aid of an HPLC pump, is heated to temperatures of up to system consist of PEEK (id=0.75 mm, od=1/16 in).The 360 °C. For nebulization, the superheated liquid is forced sample loop (PEEK, id=0.75 mm, L=5 m, volume=2.2 mL) through a nozzle at pressures of about 200 bar. The aerosol is filled with the aid of a syringe. To connect the PEEK supply eYciency is increased from about 45 to 80% in flame AAS. capillaries (od=1/16 in) to the digestion capillary (od=1/8 in), The essential part of both nebulization systems, viz., HHPN special PEEK connectors with a low dead volume (1/16–1/8 in) and HT-HHPN, is the nebulization nozzle.Its high flow were used. Sealing is guaranteed with PEEK or Teflon ferrules. resistance demands the use of an HPLC pump for the transport The temperature control device consists of a PID controller of a liquid through the flow system. With such a nozzle at the which controls two solid-state relays. Via the solid-state relays end, the system can be considered as a quasi-closed system.the PID controller regulates a variable transformer which is The nozzle can also be considered as a restrictor which connected to a power transformer. The latter is a customary provides a certain back-pressure for the whole flow system. A transformer normally used for halogen bulbs with a maximum liquid can be heated far beyond its atmospheric boiling-point power of 300 W and 12 V. To reach a high temperature within the heated system. Vapour generation is restricted as stability in a simple manner, one of the solid-state relays long as the back-pressure of the nozzle exceeds the vapour delivers a power of 100% while the other via a variable number pressure of the liquid.In HT-HHPN, only a relatively small of resistors delivers a power of about 90%. These resistors are heating zone is necessary (about 12 cm) to reach a temperature connected in series with the variable transformer. If the equilibrium for the liquid and to ensure reproducible nebulization conditions.For the development of a flow digestion Table 1 Instrumental operating parameters for ICP-OES system, it was necessary to scale up the dimensions of the high temperature/high pressure flow system used for nebulization Rf frequency/MHz 27.12 to provide a larger reaction zone and sample residence times Rf power/kW 950 long enough to guarantee complete digestion. Instead of the Argon gas flow rates/L min-1: HHP nozzle a capillary with a small inner diameter can Outer 16 Intermediate 0.5 function as a restrictor.Provided that the flow is Laminar, the Nebulizer 0.6 length of the restrictor capillary necessary to achieve a certain Spray chamber Cyclone type back-pressure at a given flow rate can be calculated using the Sample flow rate/mL min-1 1.4 Hagen–Poiseuilles law. It has to be considered that a deviation Integration time l<360 nm: 30 s l>360 nm: 10 of the nominal inner diameter of 10% leads to a deviation of s the calculate pressure of 46% (p#1/r4).To avoid vaporization Analytical wavelength/nm: C 193.091 inside the capillary, the liquid has to be cooled below its Cu 224.7 atmospheric boiling-point in front of the restrictor capillary. Mn 257.6 In this paper, the development and performance of such a Fe 248.8 high temperature/high pressure system for the continuous Zn 206.2 digestion of biological samples is described and evaluated. The 684 J. Anal. At.Spectrom., 1999, 14, 683–691Fig. 1 Schematic arrangement of the high temperature/high pressure flow digestion system. Upper part: HHP nebulization nozzle as restrictor. Lower part: 50 mm capillary as restrictor. nominal heating value is exceeded by more than 2 °C, the temperatures, high pressures and high acid concentrations as has been shown for the HT-HHPN system.33,34 device ensures that the controller is switched oV completely. This short-term switch-oV is especially important for strongly exothermic digestion reactions.This device ensures the perma- Quartz glass tubing. In the first arrangement a quartz capillary of 2.5 m in length that was specially produced for nent control of energy flux into the sample. Furthermore, it provides the possiblity to heat electric components of low as these experiments was used (Heraeus Quarzglas, Hanau, Germany). The inner diameter of the capillary was 1 mm and well as high resistance by simply changing the power transformer, thus being very flexible.The cables of the power the outer diameter was 5 mm, thus giving a wall thickness of 2 mm. The main problem to be solved was the connection to transformer are connected via strip connectors either directly to the tubing (direct heating) or to a heating wire (indirect the PEEK supply capillaries. Connectors that screw on to the capillary can destroy the material and tend to leak if the heating). The heated zones of the capillaries were insulated with mineral wool to reduce heat losses.system is exposed to high pressure. Therefore, special connectors were fabricated from PEEK that can be glued on the For cooling, either a water/Liebig cooler or an air cooler was used. For the air cooler, two electronic cooling elements quartz capillary with a two component glue that exhibits a particularly high firmness for materials such as glass, metal, (20×4x10 cm; 0.55 kW-1) were screwed together. A channel was drilled through the cooler into which the tubing was ceramics and a variety of plastics (e.g., Aremco-Bond 631; T-E-Klebetechnik, Hannover, Germany). A schematic diagram pressed.The restrictor capillaries used were made of Polysil or PEEKsil (Scientific Glass Engineering, Ringwood, Victoria, of such a connector is shown in Fig. 2. These connectors were able to withstand pressures of more than 20 MPa. Australia). Quartz capillaries with an inner diameter of 50 mm are common in gas chromatography but they are diYcult to Unfortunately, the quartz capillary was destroyed at a pressure of about 18 MPa.A second capillary of 2 m in length was connect to common HPLC capillaries. The Polysil and PEEKsil capillaries have an inner diameter of 50 mm, but they melted together from two 1 m pieces (id 1 mm, od 6 mm, wall thickness 2.5 mm). According to the manufacturer are synthetically coated. The outer diameter is 1/16 in so that the capillaries can easily be connected to normal HPLC (Vogelsberger Quarzglastechnik, Freiensteinau, Germany), the maximum inner overpressure for this capillary is about capillaries with common connectors.Implementation of a restrictor capillary is shown in the upper part of Fig. 1. 6.4 MPa.35 It should be much lower because of the unevenness produced by the melting procedure. However, several experi- The HHPnozzle plate that was used for some experiments consists of Pt–Ir (95+5) and has an orifice of 20 mm (or ments showed that these capillaries were stable beyond 10 MPa at a temperature of 250 °C.The quartz capillary has to be 15 mm). A special nozzle holder made of titanium holds the nozzle. To protect the nozzle from particles that could cause clogging, a titanium filter sieve (d=3 mm) is mounted in front of the nozzle holder. It must be emphasized that this filter is not used to filter any possible particulate residues after digestion. The arrangement implemented with an HHP nozzle as restrictor is shown in the lower part of Fig. 1. Selection of tubing material DiVerent materials were investigated for their suitability as digestion capillaries. The tubing material should withstand high temperatures and high pressures and also be resistant to high acid concentrations. Metal capillaries, e.g., stainless steel, are not suitable as they are partly dissolved under the influence of acids and thus would give high blanks for several elements. Fig. 2 High pressure connector between standard HPLC capillary (1/16 in od) and quartz capillary (6 mm od).An exception are Pt–Ir capillaries. They can be used at high J. Anal. At. Spectrom., 1999, 14, 683–691 685heated indirectly with a heating wire. The heating wire consists electrical isolator. The total length of the tubing is 2.9 m. The volume of the tube is 7.2 mL. The heated zone was coiled of Ni 99.6 (d=1 mm) with a specific electrical resistance of 0,.9 V mm2 m-1 (Isabellenhu� tte, Dillenburg, Germany).A (distance between the coils about 0.5–1 cm; diameter of coiling about 10 cm). The tubing was coiled to obtain a more compact 70 cm length of the quartz capillary was wrapped with 5 m of the wire starting at about 15 cm beyond the inlet. In front of arrangement for ease of handling. The heated part was coated with fibre glass tubing. In the middle of the heated zone a the connection to the restrictor the quartz capillary is cooled with a Liebig cooler (L=20 cm).thermocouple (Ni–CrNi) was fixed which was connected to the PID controller. Finally, the heated zone was insulated with mineral wool to reduce heat losses. The total length of the Glass lined tubing (GLT A). This material consists of a stainless-steel capillary that is coated on the inside with a heated zone is 2.2 m, corresponding to a volume of 5.5 mL. The electrical power required for heating depends on the flow borosilicate glass layer. According to the manufacturer (Scientific Glass Engineering), GLT should withstand strong rate and is shown in Fig. 3. Even at the highest flow rate of 2 mL min-1, the power required to reach a temperature of acids and bases and temperatures of up to 600 °C and briefly up to 800 °C. The material is available in several dimensions, 250 °C was lower than 100 W. The highest voltage necessary was about 5.3 V and the current was 13.1 A. This means that e.g., 1/8 in od, id=1.8 mm or 1/16 in od, id=0.7 mm, standard length up to 1.8 m.The GLT capillary is heated directly. The a relatively small power supply would be suYcient for the eYcient heating of the TL tubing to high temperatures. borosilicate glass layer functions as an electrical isolator so that no electrochemical reaction should take place. After To ensure that the temperature inside the TFE tubing is equal to the temperature measured outside the stainless-steel several digestions it became evident that the borosilicate glass layer was decomposed under high temperature/high pressure tubing, the capillary was heated slowly, without a restrictor at the end of the arrangement, at diVerent flow rates conditions in combination with nitric acid digestion (250 °C, 15 MPa, approximately 6 M nitric acid).Pictures of several (1–4 mL min-1). Additionally, the temperature of the water was measured at the end of the heated part. The diVerence segments of the GLT were taken by a scientific video camera, showing that the glass layer was totally dissolved under the between the temperatures measured outside and inside the capillary was 1–2 °C, thus proving a satisfactory heat transfer digestion conditions stated.Additionally, owing to the strong corrosion of the glass layer, fairly high concentrations of Na, from the stainless-steel capillary to the TFE tubing. Ca and Fe, the main components of the glass layer and the stainless steel, could be detected in the digested solutions.It Pt–Ir capillary. This material can withstand temperatures up to 400 °C and pressures up to 40 MPa. It was tested with was noticed that the glass layer had been destroyed only within the heated part of the capillary. The inlet and outlet of the strong acids and showed good resistance and low blank values for Pt and Ir. For the HT-HHPN system, capillaries of 15 and capillary were still coated. Additionally, the restrictor (nozzle or even the 50 mm restrictor capillary) was blocked frequently 30 cm in length were used.During the manufacturing procedure these capillaries were stretched/drawn. For the continu- by particles that were dissolved from the borosilicate glass layer. The material only seems to be inert up to temperatures ous digestion system, Pt–Ir capillaries of 1.5 m in length were ordered from another manufacturer. Unfortunately, these of about 200 °C. Therefore, GLT is not suitable for a true high temperature/high pressure digestion system.On the other capillaries had hairline cracks, so it was impossible to apply high pressure. These defective capillaries were manufactured hand, this result shows that even glass is dissolved under the stated conditions. Therefore, it is indirect proof of the eYciency by welding. Nevertheless, some orientating experiments were carried out with the 30 cm Pt–Ir capillary which showed that of a high temperature/high pressure flow digestion system. this material should be suitable for incorporation in a high temperature/high pressure flow system.Further investigations Teflon lined tubing (TLT ). Teflon lined tubing (e.g., Alltech Associates, Deerfield, IL, USA) consists of a Teflon (TFE) with this material will follow as soon as another capillary from another manufacturer is available. tube that is mechanically inserted inside a stainless-steel tube. It is available with TFE tubing of, e.g., 3.2 mm od, 1.8 mm id and about 0.19 mm wall thickness. Because most of the results Reagents were obtained with this tubing, its behaviour will be described Analytical-reagent grade water was used for all solutions and in detail.as a carrier stream. Single or multi-element calibration stan- In preliminary experiments the TL (Teflon lined) tubing dards were prepared by dilution of standard solutions contain- was used only for the digestion of glucose solutions. After a ing 1000 mg L-1 of the elements. few months, frequent clogging of the nozzle and restrictor capillary was apparent.Thereupon, the capillary was cut with a silicon carbide saw blade to investigate whether the TFE tubing had been destroyed. It was found that the tubing had snapped and been reduced in diameter and length. It became obvious that during mechanical insertion the TFE tubing must have snapped. Therefore, a second TL tubing was prepared by passing a sturdy PEEK capillary through the TFE tubing to widen it. Furthermore, short pieces of a PEEK capillary (7 cm) were inserted into each end of the TL tubing and glued to the TFE tubing with a special glue (Aremco-Bond 526, T-E-Klebetechnik) appropriate for fluoropolymer materials.This was done to ensure that the liquid is transported only through the TFE tubing and not in the space between the TFE tubing and the stainless steel capillary and to avoid contamination for several trace elements. The TL tubing was connected to the supply capillaries with special connectors (1/16–1/8 in).The TL tubing was heated directly by connecting it to the Fig. 3 Power consumption as a function of flow rate and temperature of the system. power transformer. The inner TFE tubing functions as an 686 J. Anal. At. Spectrom., 1999, 14, 683–691A carbon stock standard solutn was prepared by dissolving residual carbon. The residual carbon can be calculated from eqn. (1): 5.248 g of oxalic acid (C2H2O4·2H2O) in 100 mL of analyticalreagent grade water, thus giving a carbon content of 10 g L-1.RC (%)=RCC×100/Cor (1) All other carbon standards (10, 100, 500, 1000 mg L-1) were freshly prepared by dilution of the 10 g L-1 carbon stock Cor: original carbon content (determined, e.g., with total standard immediately before the measurements. carbon analyzer) in g (carbon) per g (sample) Digestions were carried out using nitric acid (65% m/v, pro RCC: residual carbon content in the digested solution (deteranalysi; Merck, Darmstadt, Germany). mined, e.g., by ICP-OES) refering to original carbon content Cor, in g (carbon) per g (sample) RC (%): residual carbon as percentage of the original carbon Procedure Cor Sample preparation.For the initial digestion experiments DE: digestion eYciency: 100-RC (%). glucose solutions were used since glucose is relatively easy to The original carbon content was determined with a total digest and contains no particulate matter. To prepare these carbon analyzer.Therefore, an amount of the powdered solutions an appropriate amount of glucose was weighed into sample material was weighed with a micro-balance into small a 50 mL (20 mL) flask to which were added about 10 mL Ni containers and transferred into the apparaturs. The material (4 mL) of water followed by diVerent amounts of concentrated was burned in an oxygen stream and the amount of CO2 nitric acid to give a final acid concentration of 10–40% m/v generated was determined by an infrared detector.For the (1.7–7.9 mol L-1). After cooling, the solution was diluted to determination of the original carbon content of milk, the volume with water. sample was diluted by a factor of 20 because of the high As a natural suspension, conventional mixed vegetable juice carbon content. For the determination of the original carbon (a commercial mixture of, e.g., tomato, red beet, onion, celeriac content of vegetable juice, the juice was freeze-dried to reduce and carrot juice) was chosen as an initial sample material that the water content (preconcentration of carbon).The dried contains fibrous particles. To prepare a sample of juice the material was then reacted in the carbon analyzer. bottle originally containing the juice was agitated to provide The eYciency of the digestion was determined via the homogeneity. About 5 g of juice were then weighed into a residual carbon. To obtain the accurate carbon content of the 20 mL calibrated flask and diluted to volume after the addition digested sample by an ICP-OES measurement it was necessary of concentrated nitric acid.to de-gas the sample. If the carbon content was measured To obtain slurries of powdered materials the following without de-gassing the results were found to be rather high. certified reference materials (CRMs) were used: NIST SRM Initially, de-gassing was realized by ultrasonic shaking but 1577 Bovine Liver, NIST SRM 1567 Wheat Flour, IAEA A-2 even after this procedure the carbon contents were still very Dried Whole Animal Blood, NIST SRM 1573 Tomato Leaves high.Hence, the samples were finally de-gassed by passing and Red Beet Powder (Commission of European nitrogen through the digested samples for at least 20 min. The Communities). To prepare slurries of the CRMs, about 0.2 g residual carbon for digested samples (25 g L-1 glucose) that of the sample was weighed (±0.1 mg) into a 20 mL calibrated were de-gassed with nitrogen was ‘0’% and for samples that flask to which were added 5 mL of water followed by the were not de-gassed 37% and if de-gassing was carried out by addition of 1 mL increments of concentrated nitric acid to ultrasonic shaking 29%.These values show that it is necessary obtain a suspension of 20% m/v (3.5 mol L-1) HNO3. Between to de-gas the digested solutions prior to determination of the the addition of each increment of concentrated HNO3 the carbon content by ICP-OES.flask was shaken vigorously to ensure a homogeneous distribution of the solid sample. The slurry was then diluted to Basic investigations volume with water. With some slurry materials frothing was observed; hence, it became necessary to remove the bubbles Most of the following experiments were carried out using a and foam by ultrasonic shaking. TL capillary (od 1/8 in). Because of the large length (3.05 m) and the relatively large inner diameter (1.8 mm) of the capillary, dispersion was expected to be rather large.The dispersion Digestion. About 5 mL of the slurry were drawn into a disposable syringe and injected into a six-port injection valve profile was investigated in two ways. First, by transporting an aqueous solution of Mn through the system and measuring equipped with a sample loop of volume 2.2 mL, corresponding to a sample amount of 22 mg. By turning the valve to the the Mn content in each fractionated sample volume by ICPOES (0.5 mL diluted to 10 mL).Additionally, the solutions injection position the sample was inserted into the aqueous carrier stream and transported towards the heated zone. The contained glucose. The carbon content was also measured by ICP-OES. It was decided to measure the carbon content also sample was digested while passing the heated zone. Subsequently, the sample reached the cooling zone and then because in later experiments the carbon content of the digested solutions acted as an indicator of the eYciency of the sample left the system via the restrictor.The digested sample was collected at the outlet of the system in a single fraction into a digestion procedure. The dispersion profile for an undigested glucose solution containing Mn is shown in Fig. 4. 20 mL flask, thus resulting in a dilution factor of 9.1 (2.2 mL per 20 mL). For a flow rate of, e.g., 1 mL min-1, a sample Additionally, dispersion was monitored by transporting a dye (Eosin Blue) through the system and measuring the absorption volume of 18 mL was collected.Such a comparatively large volume of sample was collected to take into account the large with a UV-HPLC photometer at 252 nm. These measurements led to the same result. For a flow rate of 1 mL min-1, the dispersion within the system. Collection was started 2 min after injection of the sample. To investigate the dispersion maximum of the profile appears at 6–7 min. Theoretically, dilution with the carrier and thus dispersion should be profile, the sample was collected in 0.5 mL fractions and each increment diluted to a final volume of 10 mL.After collection decreased by using capillaries with a smaller inner diameter,36 but the smallest inner diameter available for TL tubing is and dilution, the carbon content and several trace elements were determined oV-line by ICP-OES. 1.8 mm. In low pressure flow systems, dilution with the carrier and The carbon content of the digested solutions was determined by measuring the carbon emission by ICP-OES at 193.091 nm. thus dispersion can be avoided by deliberately embedding the sample in small volumes of air (segmented flow technique).The residual carbon content expressed as a percentage of the original carbon content of the undigested sample yields the For a high pressure system, a relatively large volume of air J. Anal. At. Spectrom., 1999, 14, 683–691 687indicating that temperatures of at least 210 °C are necessary for the eYcient digestion of a glucose solution.Further elevation of the temperature (220–240 °C) does not seem to increase the oxidation rate of carbon significantly and thus the digestion eYciency any further. For an acid concentration of 20% HNO3, the carbon content is already decreased at a temperature of 190 °C. After evaluation of the variation of the carbon profile with increasing temperature, solutions containing the same amount of glucose (25 g L-1) were digested and the residual carbon (RC%) was determined quantitatively.For this digestion, an acid concentration of 20% HNO3 was used. Again, the temperature was varied from 25 to 250 °C (25, 130, 150, 170, 190, 210, 230, 250 °C). For a temperature of 130 °C, the residual carbon was 68% (RSD: 1%); for 150 °C, 23.5% (RSD: 9%); for 170 °C, 6.5% (RSD: 6%); for 190 °C, 1.6% (RSD: 15%); Fig. 4 Dispersion profile of undigested glucose solution with Mn for 210 °C, 0.5% (RSD: 20%); and for temperatures of 230 as indicator. and 250 °C, no carbon could be detected in the degassed solutions (n=5) (detection limit for carbon: 1.5 mg L-1).has to be introduced as the air becomes compressed under In further experiments the amount of glucose to be digested high pressure. An initial volume of 2.2 mL of air was injected was elevated (25, 50, 75, 100, 150, 200, 250 g l-1). Whenever that should result in a buVer volume of 10 mL after com- digestion at 250 °C was found to be incomplete (RC >2%) pression.With an inner diameter of 1.8 mm of the capillary, the acid concentration was increased (10, 20, 30, 40% m/v). the length of the compressed volume under a pressure of Fig. 6 shows the residual carbon (RC%) as a percentage of 20 MPa should be 4 mm, thus providing a satisfactory distance the original carbon content of the digested sample for diVerent between sample and carrier. However, obviously, the air is initial glucose contents and acid concentrations. For small compressed and probably becomes completely dissolved in the amounts of glucose (25, 50 g L-1), a nitric acid concentration carrier stream.After introduction of the air, the pressure of up to 20% is suYcient to reach RC values of less than 2%. suddenly drops considerably which can be observed by follow- For higher amounts of glucose (75–250 g L-1), higher acid ing the pressure display of the HPLC pump. Pressure is rebuilt concentrations are necessary (30–40% m/v).It can be seen only very slowly as the air becomes compressed and dissolved that even a 25% (250 g L-1) glucose solution can be digested in the carrier stream. Owing to the inevitable pressure drop with 40% HNO3, resulting in a residual carbon content of after injection of a discrete volume of air, the segmented flow about 2%. The relative standard deviation is 3–16% for the technique, often used in low pressure flow systems, cannot be determination of the RC%.For original glucose contents transferred to high pressure flow systems. Furthermore, the higher than 150 g L-1, the relative standard deviation is less pressure drop within the system leads to an undesirable vapour than 10% (n=6). generation along the heated zone because boiling of the liquid occurs. Residual carbon (digestion of slurries) The first suspension to be digested was mixed vegetable juice Results (5 g juice per 20 mL digestion suspension). Digestion of the vegetable juice was carried out using a TL capillary with a Residual carbon (digestion of glucose solutions) restrictor capillary at the end of the digestion system.With an Solutions containing 25 g l-1 glucose were digested with 10% acid content of only 10% HNO3 and temperatures between HNO3. The digested sample was collected in 0.5 mL fractions 200 and 230 °C, frequent clogging of the restrictor capillary into 10 mL flasks as mentioned above. The carbon content of occurred.At a temperature of 240 °C and an acid concentration each fraction was measured by ICP-OES without any of 20%, a series of digestions were carried out without any de-gassing because only the variation of the carbon content should be observed. In a series of experiments the digestion temperature was increased from 190 to 240 °C. The resulting profiles are shown in Fig. 5. At temperatures below 210 °C, no digestion takes place. Starting at a temperature of 210 °C the carbon content of the digested solution suddenly decreases, 10 RC (%) 9 25 g L–1, 10% HNO3 25 g L–1, 20% HNO3 50 g L–1, 10% HNO3 50 g L–1, 20% HNO3 75 g L–1, 10% HNO3 75 g L–1, 20% HNO3 75 g L–1, 30% HNO3 100 g L–1, 30% HNO3 150 g L–1, 30% HNO3 200 g L–1, 30% HNO3 200 g L–1, 40% HNO3 250 g L–1, 40% HNO3 8 6 7 5 4 3 2 1 0 Fig. 6 Residual carbon as a function of the original glucose content Fig. 5 Residual carbon profile at increasing digestion temperatures. and of the concentration of the digestion acid. 688 J.Anal. At. Spectrom., 1999, 14, 683–691clogging. Even samples containing higher amounts of vegetable means at a 95% probability level. Table 3 additionally contains the calculated texper, the critical t-value, tcritical, and the result juice (e.g., 10 g juice per 20 mL) could be digested without any problem. These initial experiments with vegetable juice as of the t-test. the sample showed that, in general, the system is suitable for the digestion of samples that contain particulate matter.Discussion Thereupon, various pulverized CRMs (see under Procedure) The aim of this work was to develop a new system for the were selected for digestion. Arbitrarily, a slurry content of continuous digestion of diVerent sample materials and to prove 10 g L-1 (1% m/v) was chosen. Since digestion of vegetable the validity of the principle and the feasibility of a high juice with only 10% HNO3 led to frequent clogging of the temperature/high pressure flow digestion system.Heating of restrictor and thus incomplete mineralization of particulate the tubing was accomplished by direct or indirect resistive matter, for the digestion of the pulverized material an acid heating. DiVerent tubing materials were tested such as quartz concentration of 20% HNO3 was provided. After optimization, capillaries, glass lined tubing and Teflon lined tubing. a digestion temperature of 250 °C was found to be suitable. Under these conditions each of the listed sample materials (see Development of the arrangement/selection of tubing material under Procedure) could be digested.The digested solutions were in all cases clear and colourless. Furthermore, no eventu- The most practical material and the least expensive was Teflon ally undigested particles led to blockage of the restrictor. The lined tubing. The majority of the presented results were only material that gave rise to complications was Pine Needles obtained with this material.From a diVerent point of view the (NIST CRM 1575). The particles within the slurry were too TL tubing can be regarded as a pressure bomb stretched in large and stringy so that not even injection into the sampling length (‘long tubing bomb’) as, usually, pressure bomb vessels valve was possible. This problem can be overcome by drilling such as the ‘classical’ pressure bomb described by To� lg consist holes in the injection valve. With an injection valve containing of a stainless-steel vessel in which is embedded a PTFE liner.holes of a larger inner diameter (e.g., 1/8 in HPLC sample However, the best tubing material for a continuous digestion introduction valve) and by the use of a larger sample injection system should be Pt–Ir capillaries as they allow temperatures coil this problem can be overcome. To date, no erosion of the up to 300 or 400 °C. However, these capillaries have to be valve components was observed.In addition, slurries of Pine produced to order. Yet it seems to be important that the Needles can be injected and digested if they are left standing capillaries are manufactured by stretching/drawing. for at least 1 d after preparation of the slurry. Concerning safety it can be stated that, in general, the The carbon content measured in the digested solutions of handling of high pressure is substantially safer in metalvarious sample materials is listed in Table 2. In addition to covered tubes of a small diameter than in bombs or in the relative standard deviation, the original carbon content unprotected PTFE tubing.Temperature is controlled by a PID (Cor), the residual carbon content (RCC) and the residual controller that ensures total shut-down of energy flux in the carbon (RC%) are listed. It can be seen that the residual event of highly exothermic reactions. Pressure is controlled by carbon is rather low. For the majority of digested samples it the aid of an HPLC pump.If clogging should eventually is less than 10%. Only for Bovine Liver and Dried Whole occur, the mp immediately stops if the maximum pressure Animal Blood is the value above 10%. This may be attributed is reached. By using a TFE tube that is surrounded by a to the comparatively high original carbon content and to the stainless-steel capillary, pressure is no longer a limiting paramfact that the slurry consisted of rather large clots that may eter in digestions.not have been totally digested. Applicability/application of the system Elemental recoveries After the development of the system it had to be tested concerning its applicability to the digestion of diVerent sample For three CRMs, viz., Tomato Leaves, Dried Whole Animal Blood and Bovine Liver, the content of some selected elements materials. Initially, only glucose solutions were used as they can be digested relatively easily and contain no particles. The in the digested solutions was determined oV-line by ICP-OES.The results were compared with the certified values. In Table 3 residual carbon was evaluated as a measure of the eYciency of the digestion system. These preliminary experiments showed the analytical results and the certified values are listed. In general, the results are in the range of the certified values. that a considerable amount of carbon was present in the solutions after digestion in the form of CO2 that falsifies the Only the value for Fe in Dried Whole Animal Blood (76% recovery) is significantly lower than the certified value.To actual digestion eYciency. Complete de-gassing was realized by passing a stream of nitrogen through the digested solutions date, no explanation can be given for this behaviour. In Table 3 the relative standard deviations and the recovery are before measurement of the carbon content by ICP-OES. Highly concentrated glucose solutions up to 250 g L-1 can be also listed.A significance test was performed for the resulting Table 2 Digestion eYciency for diVerent biological materials Originala Residual Residualc Digestion carbon content, Carbonb RSD carbon content, Carbon (%) eYciency (%) Material Cor/g g-1 content/mg L-1 (%) RCC/g g-1 (RC%) (100-RC%) Tomato Leaves 0.34 25±5 20 0.023 6.7 93.2 Wheat Flour 0.38 14±3 21 0.013 3.4 96.6 Red Beet Powder 0.33 20±3 15 0.018 5.5 94.5 Bovine Liver 0.44 53±8 15 0.048 11 89 Whole Animal Blood 0.52d 71±25 35 0.065 12.5 87.5 Cocoa Powder 0.38 5±1 20 0.005 1.2 98.8 Vegetable Juice 0.02 8±1 13 0.0003 1.4 98.6 Milk 0.07 12±1 8 0.0011 1.6 98.4 aMeasured with total carbon analyzer.bCarbon content measured in digested solution by ICP-OES. cResidual carbon calculated as percentage of original carbon. dValue from the literature.37 J. Anal. At. Spectrom., 1999, 14, 683–691 689Table 3 Trace determinations in digested slurries of CRMs Concentration RSD Certified Recovery Significant measureda/mg g-1 (%) value/mg g-1 (%) texper tcritical diVerence NIST SRM 1573 Tomato Leaves— Cu 11±1 9.1 11±1 100 0 2.26 No Mn 244±6 2.5 238±7 103 2.13 2.23 No Fe 593±35 5.9 690±25 86 6.4 2.36 Yes Zn 67±3 4.5 62±6 108 2.92 2.09 Yes IAEA-A2 Dried Whole Animal Blood— Cu 45±3 6.7 45±4 100 0 2.18 No Fe 2.6±0.4 15 3.41±0.26 76 4.7 2.36 Yes Mn 114±11 9.6 123±21 93 1.46 2.10 No Zn 77±5 6.5 89±9 87 4.4 2.11 Yes NIST SRM 1577 Bovine Liver— Cu 207±11 5.3 193±10 107 2.84 2.31 Yes Zn 128±8 6.3 130±13 98 0.48 2.13 No Fe 292±10 3.4 268±8 109 5.5 2.36 Yes aMean and standard deviation (n=6).digested with a nitric acid concentration of 40%. In further hydride system to obtain on-line digestion and determination. Initial experiments showed that it is possible to connect the experiments, natural suspensions and slurries of powdered CRMs were digested. Even these suspensions containing proposed digestion system directly to an ICP-OES instrument. In these experiments the outlet of the restrictor capillary was diVerent types of particulate matter could be digested under the influence of high temperature and high pressure, resulting coupled to the Meinhard nebulizer of the spectrometer.Problems arose due to the dissolved CO2 and NOx present in in high digestion eYciencies between 87 and 99% depending on the respective material. The rather low amounts of residual the digested sample. To prevent the plasma from being extinguished, only small sample amounts can be digested and directly carbon should even make possible the subsequent determination of elements by electrochemical detection methods.In led into the spectrometer. The problem of extinguishing is not as serious when coupling the digestion system directly to a general, it was proved that temperatures of 250 °C are suYcient for the almost complete digestion of a variety of biological flame atomic absorption spectrometer. Working with a nozzle as restrictor it should also be possible to use the HHP nozzle and botanical materials.Because of the rather large inner diameter of the TL tubing, dispersion leads to the dilution of for direct nebulization of the digested samples into the plasma or even an HT-HHP nozzle. In both cases for ICP-OES, the sample in the carrier. In spite of the large dilution a sample volume of 2.2 mL was suYcient for the digestion of biological desolvation of the aerosol of the digested sample solution is necessary because of the high aerosol yield provided by this slurries and the subsequent determination of minor, major and trace elements.Carry-over from one sample to the next nebulization technique. A further possibility of direct nebulization of the sample into a spectrometer can include the use of was not observed as the elements under investigation are present in the biological materials at rather low concentration the restrictor capillary. As mentioned above, the digested solution is cooled below its atmospheric boiling-point before levels.Additionally, the system is cleaned continuously by the carrier. It is not yet known whether carry-over would occur leaving the system via the restrictor capillary. However, if the liquid were not cooled it would reach the restrictor capillary for the subsequent digestion of samples which contain the elements at considerably diVerent concentrations. at rather high temperatures. This should result in thermospray eVects that can be used for direct nebulization. If suYcient sample material is available, e.g., urine, waste water or juices, larger sample volumes (10–20 mL) can be Furthermore, the system can be automated to achieve computer-controlled digestion procedures.Another area for injected and a larger segment of the sample can be collected after digestion, rejecting the diluted front and end (tailing) of investigation is the use of alternative reagents for the digestion of diVerent materials, e.g., mixtures of HNO3 and HCl, the sample.peroxodisulfate and H2O2. Further developments In a further development of the digestion system the use of a Conclusion Pt–Ir capillary is intended. This capillary is available in a A new continuous flow digestion system using resistively much smaller inner diameter, thus reducing the dilution of the heated metal-covered tubing has been presented that works sample in the carrier stream. Since this material can be heated under real high temperature/high pressure conditions (260 °C, to temperatures up to 300 or even 400 °C it is expected that >200 bar).The feasibility of the continuous digestion of the digestion eYciency can be increased further. Higher digesvarious sample materials, liquid as well as slurrry samples, has tion temperatures should allow higher flow rates and shorter been demonstrated. The eYciency of the oxidation of the residence times and a reduction of the digestion time as well organic matrix was about 90–99% depending on the digested as the use of a shorter capillary and thus a more compact sample material.Several minor, major and trace elements were system for ease of handling. In the near future further investidetermined with recoveries of about 87–109%. This work has gations will be carried out with a system incorporating a Pt–Ir shown that the hitherto existing limitations concerning tem- capillary, especially for the digestion of waste waters.This perature and pressure can be overcome. It can be expected sample material can contain rather large amounts of silicates that in future still higher temperatures can be reached. that are hard to digest. However, with temperatures up to 400 °C, even the digestion of such a complex sample material should be possible. Acknowledgement In addition, it is intended to connect the continuous digestion system directly to an ICP optical emission spectrometer, This work was carried out with the financial support of the Bundesministerium fu� r Forschung und Technologie as well as a flame atomic absorption spectrometer, a photometer or a 690 J.Anal. At. Spectrom., 1999, 14, 683–69117 H. Matusiewicz and R. E. Sturgeon, Fresenius’ J. Anal. Chem., the Ministerium fu� r Wissenschaft und Forschung des Landes 1994, 349, 428. Nordrhein-Westfalen. 18 T. J. Gluodenis, Jr. and J. F. Tyson, J. Anal. At. Spectrom., 1992, 7, 301. 19 V. Carbonell, M. de la Guardia, A. Salvador, J. L. Burguera and References M. Burguera, Anal. Chim. Acta, 1990, 238, 417. 20 V. Carbonell, A. Morales-Rubio, A. Salvador, M. de la Guardia, 1 R. Bock, Aufschlußmethoden der Anorganischen und Organischen J. L. Burguera and M. Burguera, J. Anal. At. Spectrom., 1992, Chemie, Verlag Chemie,Weinheim, 1972. 7, 1085. 2 T. T. Gorsuch, The Destruction of Organic Matter, Pergamon 21 C. P. Hanna and S. A. McIntosh, At. Spectrosc., 1995, 16, 106. Press, Oxford, 1970. 22 D. L. Tsalev, M. Sperling and B.Welz, Analyst, 1992, 117, 1729. 3 R. L. Watters, Jr., J. R. DeVoe, F. H. Shen, J. A. Small and R. B. 23 T. J. Gluodenis, Jr. and J. F. Tyson, J. Anal. At. Spectrom., 1993, Marienko, Anal. Chem., 1989, 61, 1826. 8, 697. 4 S. A. Darke, S. E. Long, C. J. Pickford and J. F. Tyson, J. Anal. 24 V. Karanassios, F. H. Li, B. Liu and E. D. Salin, J. Anal. At. At. Spectrom., 1989, 4, 715. Spectrom., 1991, 6, 457. 5 S. A. Darke, S. E. Long, C. J. Pickford and J. F. Tyson, Fresenius’ 25 L. J. Martines Stewart and R. M. Barnes, Analyst, 1994, 119, 1003. J. Anal. Chem., 1990, 337, 284. 26 R. E. Sturgeon, S. N. Willie, B. A. Methven, W. H. Lam and H. 6 A. T. Ellis, P. J. Potts, M. Holmes, G. J. Oliver, C. Streli and P. Matusiewicz, J. Anal. At. Spectrom., 1995, 10, 981. Wobrauschek, J. Anal. At. Spectrom., 1997, 12, 461R. 27 E. S. Beary, P. J. Paulsen, L. B. Jassie and J. D. Fassett, Anal. 7 M.Wu� rfels, E. Jackwerth and M. Stoeppler, Fresenius’ Z. Anal. Chem., 1997, 69, 758. Chem., 1988, 330, 160. 28 G. Knapp, M. Michaelis and U. Pichler, paper presented at 8 A. Eucken and E. Wicke, Grundriss der Physikalischen Chemie, the Colloquium Analytische Atomspektroskopie (CANAS), Akademische Verlagsgesellschaft, Leipzig, 1959. Konstanz, Germany, April 2–7, 1995. 9 L. Kotz, G. Kaiser, P. Tscho� pel and G. To� lg, Fresenius’ Z. Anal. 29 H. Berndt and J. Yanez, J. Anal. At. Spectrom., 1996, 11, 703. Chem., 1972, 260, 207. 30 H. Berndt, Fresenius’ Z. Anal. Chem., 1988, 331, 321. 10 E. Jackwerth and S. Gomiscek, Pure Appl. Chem., 1984, 56, 479. 31 H. Berndt and A. Mu� ller, Fresenius’ J. Anal. Chem., 1993, 345, 18. 11 G. Knapp, Fresenius’ Z. Anal. Chem., 1984, 317, 231. 32 G. Schaldach and H. Berndt, Fresenius’ J. Anal. Chem., 1994, 12 H. Matusiewicz, R. E. Sturgeon and S. S. Berman, J. Anal. At. 350, 481. Spectrom., 1989, 4, 323. 33 S. K. Luo and H. Berndt, Spectrochim. Acta, Part B, 1994, 49, 485. 13 H. M. Kingston and L. B. Jassie, Introduction to Microwave 34 I. May and J. J. Rowe, Anal. Chim. Acta, 1965, 33, 648. Sample Preparation: Theory and Practice, American Chemical 35 E. Greil, GIT , Merk- und Arbeitsbla�tter, 1963, 7, Heft 1. Society, Washington, DC, 1988. 36 J. Ru° z¡ ic¡ka and E. H. Hansen, Flow Injection Analysis,Wiley, New 14 A. Abu-Samra, J. S. Morris and S. R. Koirtyohann, Anal. Chem., York, 1988. 1975, 47, 1475. 37 M. Wu� rfels and E. Jackwerth, Fresenius’ Z. Anal. Chem., 1985, 15 M. Burguera, J. L. Burguera, C. Rondon, C. Rivas, P. Carrero, 322, 354. M. Gallignani and M. R. Brunetto, J. Anal. At. Spectrom., 1995, 10, 343. 16 S. J. Haswell and D. Barclay, Analyst, 1992, 117, 117. Paper 9/00634F J. Anal. At. Spectrom., 1999, 14, 6

ISSN:0267-9477    

DOI:10.1039/a900634f

出版商:RSC    

年代:1999

数据来源: RSC

摘要:

A pulsed Grimm glow discharge as an atomic emission source Chenglong Yang, Kristofor Ingeneri and W. W. Harrison* Department of Chemistry, University of Florida, Gainesville, FL 32611-7200, USA Received 15th September 1998, Accepted 25th January 1999 A high voltage, microsecond regime pulse dc glow discharge was applied to a standard Grimm source on a modified LECO SA-2000 direct reader spectrometer and to a duplicate Grimm source coupled to a scanning monochromator. The eVects of experimental conditions, including pulse voltage, pulse frequency, pulse width and Ar pressure, on Cu atomic and ionic emission intensities were examined and compared with continuous discharge results.The pulsed Grimm source has a higher sputtering rate, greater signal intensity and lower detection limits. The pulsed signal for atomic lines shows a diVerent temporal response from ionic lines. Pulsed Grimm GD-AES also oVers control features advantageous in the measurement of thin surface layers.Introduction The glow discharge (GD) has proven to be a very useful atomic spectroscopy source for the direct analysis of solid samples owing to its numerous advantages, which include the simplicity of the source itself, excellent detection limits (ppb), minimal matrix eVects and a broad elemental coverage.1–3 The Grimm FD4 is the most widely applied source for atomic emission spectrometry, a technique represented by several commercial instruments. GD-AES using a Grimm source has the advantages of easy sample interchange, good precision, wide dynamic range and relatively small matrix eVects.It has the ability to achieve both bulk and depth-resolved elemental analysis. Therefore, it is now a routine tool for the direct elemental analysis of solids.5 The Grimm GD has been powered by direct current (dc) Fig. 1 Schematic diagram of pulsed Grimm GD optical system and and radiofrequency (rf ) sources.6,7 Both modes are routinely gated detection.operated at power levels of tens of watts. The instantaneous power of the microsecond pulsed GD can routinely reach several hundred watts, and even up to more than 1000 W in shaped anode of the source extended to approximately 0.2 mm some conditions.8 Pulsed operation of conventional glow from the sample. The inner diameter of the anode was 4 mm. discharges and hollow cathode lamps has been shown to The Grimm glow discharge was operated in both the enhance sputtering rates and signal intensities by 1–4 orders continuous and pulsed modes.In the dc mode a high voltage of magnitude in emission, fluorescence and mass spec- dc power supply from the LECO SA-2000 system was used. trometry.9,10 The purpose of this study was to extend the In the pulsed mode, a high voltage pulsed power supply investigations of the microsecond pulsed discharge to include (Model 350, Velonex, Santa Clara, CA, USA) was employed. its application to the important Grimm source, including The pulse width and frequency of the pulsed generator were certain initial evaluations of critical parameters.varied from 0.1 to 300 ms and from 1 to 100 000 Hz, respectively. Voltages up to 3000 V could be applied to the glow discharge source. The pulsed power supply is capable of internally controlled pulsed operation, but a separate wave- Experimental form generator [Hewlett-Packard (Palo Alto, CA, USA) Model 8003A] was used to trigger both the pulse power supply Apparatus and the oscilloscope in order to synchronize the discharge and A schematic diagram of the microsecond pulsed dc Grimm resulting emission signal.GD-AES system is shown in Fig. 1. Two spectrometers were An alternative signal readout and data handling system was used in the experiment. One was a modified (pulsed source used to monitor the microsecond pulsed signal, as the standard and gated detection) multi-channel direct-reading spec- SA-2000 system is designed for continuous signals.Output trometer, LECO (St. Joseph, MI, USA) SA-2000, which uses current pulses taken directly from the photomultiplier tubes a 0.4 m direct-reading spectrometer with a 2400 groove holo- were converted to pulsed voltage signals by measurement graphic grating. Spectra collection was done with a sequential across a 66.3 kV resistor. The signals were processed in two monochromator [Chromex (Albuquerque, NM, USA) ways. The first system consisted of a 2 Gsa s-1 (gigasamples 500IS/SM] equipped with a 1200 grooves mm-1 grating.The per s), 400 MHz, four channel, digital storage oscilloscope standard LECO Grimm GD sources have been described in (Hewlett-Packard Model 54542C). The output signal, discharge voltage and discharge current were displayed simul- detail elsewhere.7 In the model used for this study, the tubular- J. Anal. At. Spectrom., 1999, 14, 693–698 693taneously on the oscilloscope. The voltage was attenuated through a 100051 voltage probe, while the discharge current was determined by calculation from the peak voltage dropped across a 5.1 V carbon resistor.The second readout system consisted of a gated boxcar and a computer. The signals from the PMTs were temporally gated and processed with a boxcar integrator (SR250, Stanford Research Systems, Sunnyvale, CA, USA), then the signals from the boxcar were sent to a computer via an analog-todigital (A/D) converter (SR245, Stanford Research Systems).The timing diagram of the system involves a waveform generator to control both the pulsed power supply and the boxcar integrator. A variable delay time (td) of 1 ns–10 ms with a variable pulse width (tw) of 1 ns–15 ms was available in the boxcar integrator. The delay window selected the measurement Fig. 2 Typical time-resolved profiles for a pulsed Grimm GD at interval before emission data were taken after initiation of the 100 Hz. pulse discharge. A custom 32-channel gated data system has now been charging of the argon gas and after breakdown continues to developed and added to the SA-2000, but was not available rise slowly even to discharge termination.at the time of this work. Applying a high voltage does not immediately yield a copper emission signal, which appears only after the current builds Sample preparation and suYcient sputtering has taken place (~1 ms after initiation). At that point, the sputter injection of copper atoms A pure copper sample (99.9%) and iron standard samples causes Cu I emission to increase sharply with current, peaking (spectrometric reference materials cast iron C18.8, CKD232– around the 3 ms point, then decreasing slowly over the remain- 239, Research Institute CKD, Prague, Czech Republic) were der of the pulse.The emission remains high even after the used as flat-faced blocks to couple to the Grimm source. current has dropped to its plateau value owing to the slow Copper films on polished steel substrates were prepared in the diVusion rate of the sputtered atoms and the presence of Department of Materials Science and Engineering, University metastable Ar atoms to sustain excitation at high levels.of Florida. Because of the emission intensity available from a microsecond pulsed discharge, the response of the elemental ions is of interest, as these species are enhanced greatly compared Results and discussion with the case of dc discharges.8 We are also interested in ions Compared with other GD configurations, which have a signifi- for their direct role in GD mass spectrometry.11 In the case of cant, but variable cathode and anode distance, the Grimm copper (but not with all elements12), ion emission lags behind model depends on an anode that is positioned very close to that from atoms.As shown in Fig. 2, the Cu II emission the cathode, within the dark space distance. In this way, the exhibits a brief delay before its appearance and then increases discharge is confined and the sampled surface area is well gradually to a maximum around 15 ms, which is 10–12 ms later defined by the diameter of the anode.This makes the Grimm than the maximum for Cu I. The mechanism for this response source particularly advantageous for the study of thin layer is not clear, although it could be related to the timing of samples. Some disadvantages are also encountered with a maximum argon metastable formation and the well known Grimm source.The sample must be very flat to permit a related ionization step.13 In any case, the timing delay set for vacuum seal against an O-ring fitting and to allow a continuous optimum signal acquisition for elemental emission will be a cathode–anode distance around the sampled area. Because this function of the species measured. In some cases, there could distance is so small, sputtering can produce non-uniform be an advantage in selecting an intense ion line and setting a deposition around the crater, eVects that may short out the delay time that provides temporal discrimination against electrodes or cause discharge instabilities, although this was potentially interfering atomic lines.not encountered in these experiments. In the case of the microsecond pulse discharge, high sputter Electrical characteristics of pulsed Grimm GD rates are developed during the discharge on-time, so we were The analytical performance of a GD source is closely related interested in detecting any deleterious eVects that might be to its electrical properties.While the standard operating produced with a Grimm source. In a broader sense, the GD characteristics of a Grimm source are well known, it was plasma in the Grimm source is confined within a narrow important to determine how the Grimm source responded to tubular anode, so another aspect of this study was to determine the high voltage, high current pulses over a range of conditions.whether high power discharge pulses would exhibit diVerent The relationship between discharge current and voltage at a responses to the more open GD design. given source pressure provides the most fundamental display of GD characteristics. Normally, the discharge of a Grimm Typical pulse profiles lamp is self-sustaining and operates in an abnormal region. Therefore, increases in discharge current must be accompanied Fig. 2 shows atomic and ionic emission profiles for two copper lines, relative to the pulse voltage and current waveform as by increases in operating voltage at a constant pressure.Fig. 3(a) indicates that this is true for the pulsed operation, displayed on the oscilloscope. Typically, a 10 ms 2500 V pulse was applied to the Grimm source with 100 Hz operating showing a plot of the peak pulse voltage of the GD source as a function of peak current for two diVerent pressure regimes, frequency. At the beginning of the pulse cycle, the current rises sharply within the first microsecond after the voltage 3 and 4.5 Torr argon.A linear relationship is observed at each pressure, which implies a similar response of power with application, reaches about 3 A at 2 ms, then with a rapid drop in discharge resistance decreases to a plateau region of 0.5 A increased current, and likewise of sputtering rate. It can also be seen in Fig. 3(b) that the two pressures create diVerent at 3 ms and maintains this value for the duration of the pulse.The discharge voltage rises abruptly owing to capacitative plasma impedances as the power levels increase. Peak powers 694 J. Anal. At. Spectrom., 1999, 14, 693–698Fig. 4 EVect of discharge pressure on Cu II-to-Cu I ratio at 2000 V, 10 ms and 50 Hz pulse. Cu I, 327.39 nm; Cu II, 219.23 nm. current response takes the form of a sharp initial pulse, followed by rapid decay to a lower power (resistance) level. The response is probably also aVected by the power supply being unable to sustain the high peak currents.In any case, Fig. 3 Electrical characteristics for pulsed Grimm GD lamp at 3 and wider pulses have limited advantage in that the low current 4.5 Torr Ar using a Cu cathode. (a) Voltage versus current; (b) power plateau is extended, but with a decrease in the initial peak versus voltage. current amplitude. The eVect of pulse width is shown for copper ion emission in Fig. 5. Longer pulses decrease the peak of 2–3 kW are attained for the 10 ms pulse, compared with emission while adding only marginally to the temporal typical continuous power levels of 60 W for the dc mode.It is response, so that no analytical advantage is gained. A 10 ms significant that the usual pulse operating conditions produce pulse provides the optimum response in balancing emission, a discharge in which the average power is lower than that of power and stability. No significant eVect from frequency the dc mode because of the reduced duty cycle.Pulse frequency variation was observed. and pulse width determine the average power delivered to the discharge. Normally, with a Cu cathode, the peak power is Comparison of emission and sputtering rates 750 W with a pulse of 2000 V, 100 Hz frequency, 10 ms pulse The peak power of the pulsed Grimm GD is considerably width and 3 Torr pressure. However, the average power is higher than the steady-state level from the dc Grimm GD, only 0.56 W, so the cooling system required for normal Grimm resulting in higher short term intensity and sample sputtering.GD operation is not necessary in the pulse mode. A direct comparison of the emission signals, with each source operating at its normal conditions, is shown in Fig. 6 for Cu Optimization of operating parameters I and Cu II. A data gate set to sample each discharge at its Studies on the eVect of pulse operating voltage, pulse width, maximum intensity shows that the Cu I and Cu II intensities pulse frequency and support gas pressure on copper atomic increased by factors of 15 and 16, respectively.The net and ionic emission intensity were carried out using a pure Cu advantage includes the ability to operate at low average power sample. The emission response from Cu I (327.396 nm) and while still obtaining higher peak emission. Cu II (219.226 nm) was monitored as a function of pulse Operating a GD at high peak power implies enhanced voltage and current.Compared with the standard dc discharge, sputter rates during the pulse period, both from an increased the pulse discharge would be expected to produce enhanced number of sputtering ions striking the cathode and also from emission intensities, which is indeed the case, as will be shown the fact that the average energy of the impacting ions is higher, in a later section. The pulse discharge exhibits a disproportion- which increases the sputter yield.The pulsed nature of the ate enhancement of ion emission over that from neutral atoms, particularly as the pressure and related power of the pulse increase. Fig. 4 shows the eVect of increasing pressure on Cu II and Cu I intensities, both of which increase with pressure, but the ion signal rises more rapidly. At low pressures (and thus low powers) the ion line shows no special prominence, but as power is increased the ion signal intensity approaches that of the atom signal.Bogaerts and Gijbels14 suggested that under high power conditions, asymmetric charge transfer may become a dominant ionization process in comparison with Penning ionization, producing more eYcient ionization. The practical significance of this eVect is to make certain ion lines suYciently intense as to be analytically viable alternatives in cases where spectral interferences may preclude the use of neutral atom lines. The emission spectrum of a pulsed GD shows many more prominent ion lines.In principle, the use of extended pulse widths beyond the standard 10 ms should increase the discharge duty cycle accordingly and produce a linear increase in power and net sputtering. Fig. 5 EVect of pulse width on Cu II emission at 2000 V pulse with operating frequency of 50 Hz and 3 Torr Ar. Cu II, 219.23 nm. In practice, this is not observed. As shown in Fig. 2, the peak J. Anal. At. Spectrom., 1999, 14, 693–698 695Fig. 7 Depth profile of a 10 nm Cu film deposited on steel.Conditions: 1000 V, 10 ms, 200 Hz pulse and 4 Torr Ar. Cu I, 327.39 nm; Fe I, 327.99 nm. Fig. 6 Comparison of Cu I and Cu II emission profiles from dc and pulsed Grimm GD-AES. Pulse: 2500 V, 972 W (peak), 0.49 W (average) and 3.5 Torr Ar. Dc: 1200 V, 38 W and 5 Torr Ar. discharge permits controlled erosion of a surface by variation of the pulse frequency. We carried out direct mass loss measurement for sputtering rates with both types of discharge.The net sputter rate of a continuous discharge is greater than with this pulsed dc, but during the measurement time of the microsecond discharge, much higher sputtering rates occur. A Fig. 8 Depth profile of a computer hard-disk. Conditions: 1200 V, 10 ms, 600 Hz pulse and 4 Torr Ar. C I, 165.70 nm; Cr I, 425.43 nm, 10 ms 2000 V pulse at 3 Torr pressure at 422 W peak power, Co I, 345.35 nm; Ni I, 349.30 nm; P I, 178.29 nm. but only 0.42W average power, removes about 2000 A° of a Cu disk (4 mm eVective surface diameter) per minute with a pulse frequency of 100 Hz.From this, the net sputtering rate tive stripping of atoms from a thin film. Layers of only a few atomic distances can be removed during the period of several is about 3.5×1013 atoms per pulse. In the dc mode at 1200 V, 5 Torr pressure and 37 W continuous power, a hundred pulses extending over as much time as the pulse frequency determines. There is therefore suYcient time to 132 000 A° min-1 removal rate is measured.If we calculate then on comparable time-scales, the net sputtering rate of the stabilize the discharge and collect depth profile data, whereas with a continuous discharge a very thin film may be removed dc discharge in a 10 ms interval is roughly 2.5×1012 atoms. On this basis, the sputtering rate of a pulsed Grimm GD is in seconds. The pulsed nature of the discharge may also reduce somewhat redeposition from collision with other sputtered enhanced about 14 times compared with that of dc Grimm GD during the pulse cycle.However, since the on-time is very atoms due to the removal of these atoms between pulses. Fig. 7 shows the removal of a 10 nm deposited copper layer short and each pulse is reproducible within about 5%, careful control of the surface removal process is possible by adjustment from a steel substrate. The uniformity of both the deposited film and the base material was not optimum, but the results of pulse frequency.It is also important to note that we have not experienced showed that layers of this magnitude could be removed with about 5% reproducibility (three trials), as determined from any problems with sputter deposition of cathode material relative to shorting out the cathode–anode gap. Craters formed the total time to the crossing point of the plots. Thin film standard samples were not available to us, but we did examine have been symmetrical with smooth edges.Even after several hours of discharge time, no instability has been observed. a well known sample type that contains multiple thin layers, namely magnetic disc hard drives taken from spent computers. Fig. 8 shows results taken from an IBM computer hard drive, Thin-layer sample analysis indicating that the pulsed GD method can resolve and identify The Grimm GD has proven to be an excellent tool for depth the several disk layers. Manufacturers of these disks do not profile analysis,7 using both dc and rf discharges. The appli- make readily available details of construction and composition cations have tended to be comparatively thick layers in the for comparison with our results.The advantage of the applimicrometer range, such as metal or alloy protective coatings. cation of Grimm GD to such samples is the complete lack of Limitations for GD thin layer analysis include possible uneven preparation time required. The disks are simply placed against crater formation and also atomic mixing due to redeposition the source O-ring mount and analyzed directly.of sputtered atoms back on to the surface, caused by collisions with discharge gases. High vacuum techniques such as SIMS Cu emission spectra avoid some of the problems found with a GD, although it is not without its own limitations. The minimal matrix eVects in Previous work in our laboratories has achieved a beneficial separation of the signal from background in pulsed GD time- GD analysis promote eVorts to extend the GD to thin film analysis.15 of-flight mass spectrometry.16 Similarly, work with the pulsed GD hollow cathode lamp has shown some interesting temporal The microsecond pulsed discharge allows short term, repeti- 696 J.Anal. At. Spectrom., 1999, 14, 693–698Fig. 9 Time-resolved spectra from a pulsed Grimm GD lamp (Cu sample) at 2000 V, 10 ms, 600 Hz pulse and 3 Torr Ar. separation among various plasma species.11 Fig. 9 shows the only a few microseconds after pulse termination.Both types of emission depend on sputter ablation of the cathode atoms, results of a comparable study with the pulsed Grimm source by delaying the gate time after discharge initiation. Even at with subsequent collisional eVects in the GD plasma to cause excitation. The time-scale we see here seems much too slow to the initial delay time of 2 ms, a signal for Cu I is observed, but no Cu II emission is present. The Cu I emission reaches its be attributed to any fundamental stepwise ionization process.More likely is some discharge termination eVect that creates maximum at a delay time of 5 ms. However, Cu II emission forms later and maximizes at about 15 ms. Temporal responses electrons or other species more optimal for ionization (e.g., charge transfer possibilities). Recall, of course, that we are not from the pulsed signal for several other elements were also measured and are presented later.In principle, by optimizing observing actual atom or ion populations as such; we see instead that fraction excited. We can only suggest at this point the time delays appropriately for specific elemental lines, better analytical signals can be obtained, particularly for ion lines that there are diVerent temporal responses of atoms vs. ions for these particular elements. vs. atom lines. Analytical characteristics of the pulsed Grimm source DiVerent temporal response for atomic and ionic lines For potential utility of a pulsed Grimm GD, we should be Examples of diVerent temporal responses for atomic and ionic interested in the source stability, sensitivity and analytical lines of four elements (all at 2000 V, 10 ms, 50 Hz pulse and 3 response. The stability characteristics of pulsed Grimm GD- Torr pressure) are shown in Fig. 10. In the case of each AES were evaluated in terms of a relative standard deviation element, the sputtered atom emission appears earlier than the (RSD) of about 5% for 10 consecutive pulsed signals, as ionic emission.The atomic lines reach a maximum during the shown in Fig. 11. Obviously, signal averaging would be used duration of the pulse, but the ionic lines contrast by maximizing in real analytical situations (data taken over a longer period of 1 min, as evidenced in Fig. 8 for nickel and phosphorus, yield RSDs near 1%), but we wished to determine the reproducibility of the raw pulses. In these measurements, the pulse width was reduced to 5 ms to achieve more intense pulse conditions. A standard material (CKD238) containing Cu at 0.92% served as the sample, with pulse conditions of 2500 V, 50 Hz, 5 ms and 2.5 Torr pressure.The RSDs for Cu I and Cu II for peak areas of single pulses are 4.5 and 9.2%, respectively. With signal averaging, better RSDs result. By averaging 64 pulses (slightly over 1 s in time of 50 Hz), RSDs of 1.0 and 2.5% were obtained for Cu I and Cu II, respectively.Calibration curves were obtained for copper in a range of CKD standard samples, as shown in Fig. 12. Standard conditions employed were 2000 V pulses at 50 Hz, 5 ms pulse width and 3 Torr pressure. Data were obtained for both an atom line and an ion line, with good linearity for each. The regression coeYcients for Cu I and Cu II are both 0.996. One hundred Fig. 10 Temporal elemental emission profiles of pulsed Grimm GD at 2000 V, 10 ms, 50 Hz pulse and 3 Torr Ar.pulses at 100 Hz were sampled for each measurement using a J. Anal. At. Spectrom., 1999, 14, 693–698 697Conclusions The pulsed operation of a Grimm GD shows several interesting analytical advantages, including enhanced sputtering rate and emission intensity compared with the dc mode. The pulsed method also permits more control over sample removal rate, which may be advantageous in the analysis of thin layer samples. The electrical characteristics of a pulsed Grimm GD and the diVerent temporal responses for atomic and ionic lines oVer opportunities to elicit a deeper understanding of the discharge mechanism.In order to extract the full potential from the pulsed Grimm GD, a suitable signal accumulation system is required to handle the fast microsecond pulse emission. Based on the results obtained in these optical emission studies, we now plan to modify the source suitably for interfacing with our time-of-flight mass spectrometer to gain complementary information.Fig. 11 Single pulse stability comparison of copper atom and ion emission using a CKD238 standard sample (0.92% Cu) at 2500 V, 50 Hz, 5 ms pulse and 2.5 Torr Ar. Cu I, 327.29 nm; Cu II, 219.23 nm. Acknowledgements The authors express their gratitude for the support of this research by the Department of Energy, Basic Energy Sciences and by the LECO Corporation. Also acknowledged are valuable discussions with Drs. Ben Smith and Jim Winefordner. References 1 Glow Discharge Spectroscopies, ed.R. K. Marcus, Plenum Press, New York, 1993. 2 W. W. Harrison, C. M. Barshick, J. A. Klingler, P. H. RatliV and Y. Mei, Anal. Chem., 1990, 62, 943. 3 J. A. C. Broekaert, J. Anal. At. Spectrom., 1987, 2, 537. 4 B. N. Chapman, Glow Discharge Processes, Wiley, New York, 1980. 5 A. Bengtson, J. Anal. At. Spectrom., 1996, 11, 829. 6 K. Wagatsuma and S. Suzuki, Fresenius’ J. Anal. Chem., 1997, 358, 581. 7 A. Bengtson, Spectrochim. Acta, Part B, 1994, 49, 411. Fig. 12 Calibration curves for Cu in CKD standard samples (atom 8 X. M. Yan, W. Hang, B. W. Smith, J. D. Winefordner and W. W. and ion lines at 2000 V, 5 ms, 100 Hz pulse and 3 Torr Ar. Harrison, J. Anal. At. Spectrom., 1998, 13, 1033. 9 W. Hang, W. O. Walden and W. W. Harrison, Anal. Chem., 1996, 68, 1148. 2.5 ms sampling window. Error bars show the RSD of three 10 W. O. Walden, W. Hang, B. W. Smith, J. D. Winefordner and sample runs, each at diVerent sample locations, requiring W. W. Harrison, Fresenius’ J. Anal. Chem., 1996, 355, 442. breaking of the sample vacuum between each measurement. 11 W. W. Harrison, W. Hang, X. Yan, J. Ingeneri and C. Schilling, The composition of detection limits for the pulsed discharge J. Anal. At. Spectrom., 1997, 12, 891. 12 C. Yang and W. W. Harrison, in preparation. vs. dc operation is of interest, but we are constrained by our 13 E. B. M. Steers and F. Leis, J. Anal. At. Spectrom., 1989, 4, 199. present data system, so only relative indications are justified. 14 A. Bogaerts and R. Gijbels, Anal. Chem., 1997, 69, 721A. Although the signal intensities of most elements in pulsed 15 K. Wetzig, S. Baunack, V. HoVmann, S. Oswald and F. Praessler, Grimm GD-AES are enhanced more than 10-fold compared Fresenius’ J. Anal. Chem., 1997, 358, 25. with the dc Grimm GD-AES, the background emission in the 16 W. W. Harrison and W. Hang, J. Anal. At. Spectrom., 1996, pulsed Grimm GD also increases. Therefore, DLs (3× RSD) 11, 835. for the pulsed mode improve by a factor of ~2–8, dependent upon the line and background response. Paper 8/07204C 698 J. Anal. At. Spectrom., 1999, 14, 693–698

ISSN:0267-9477    

DOI:10.1039/a807204c

出版商:RSC    

年代:1999

数据来源: RSC

摘要:

Low-level determination of non-metals (Cl, Br, I, S, P) in waste oils by inductively coupled plasma optical emission spectrometry using prominent spectral lines in the 130–190 nm range K. Krengel-Rothensee, U. Richter and P. Heitland* Spectro Analytical Instruments GmbH, Boschstr. 10, D-47533 Kleve, Germany Received 8th September 1998, Accepted 12th February 1999 The non-metals Cl, Br, I, S and P were determined in waste oils by inductively coupled plasma optical emission spectrometry (ICP-OES).Prominent spectral lines in the vacuum ultraviolet spectral region between 130 and 190 nm were applied simultaneously with a nitrogen-filled Spectro polychromator. For example, the intense spectral lines for chlorine (Cl 134.72 nm), bromine (Br 163.34 nm), iodine (I 161.76 nm), sulfur (S 180.70 nm) and phosphorus (P 177.50 nm) were used. The detection limits of Cl, Br, I, P and S in undiluted waste oils are 0.9, 1.6, 0.47, 0.04 and 0.07 mg kg-1, respectively. To achieve these detection limits the standard oils and samples were diluted 1+4 (m/m) with kerosene.For the analysis of waste oils, small amounts of oxygen (80 ml min-1) were added to the outer plasma gas. In order to establish optimized plasma operating conditions for non-metals, a simplex optimization was performed. Compared with metals such as Ba, V, Cd, Al or Ni, a lower nebulizer gas flow and a higher generator power is advantageous to improve the detection limits for the non-metals.Because the detection limit for Cl is more than a factor of 1000 below tolerated limits of waste oil regulations, highly viscous oils could be analyzed after a suitable dilution with kerosene. The determination of the total Cl concentration by ICPOES can be used as a means of screening for the highest possible concentration of polychlorinated biphenyls (PCBs) in waste oils. or metallic surfaces in the refining process by impurities. Oil Introduction additives (b) often have the task to enhance the lifetime of an The determination of non-metals in organic samples with ICP oil and must be controlled precisely.Therefore, the use of an optical emission spectrometry (ICP-OES) has increasingly internal standard is often necessary.Wear metal concentrations become a point of interest since intense spectral lines for non- in (c) lubricating oils are determined as an indicator for metals in the vacuum ultraviolet wavelength region below bearing wear in engines.Monitoring wear metals ensures the 190 nm can be used with ICP spectrometers. In this wavelength reliability of engines and optimizes servicing periods. When region are prominent spectral lines for many environmentally analyzing (d) waste oils the determination of heavy metals, or industrially important elements for ICP-OES.1–3 Examples halogens and sulfur has become an environmental analytical are aluminum (167.08 nm), boron (182.64 nm), bromine task for ICP spectrometry.In particular, the determination of (154.07), chlorine (134.72 nm), lead (168.21), mercury Cl is of great interest in environmental element analysis, (184.84 nm), nitrogen (149.25 nm), sulfur (180.73 nm) and because of the hazardous potential of chlorinated organic phosphorus (177.50 nm). compounds, such as polychlorinated biphenyls (PCBs), penta- Because wavelengths below 200 nm are absorbed by oxygen chlorophenol (PCP), polychlorodibenzodioxins (PCDDs) or or water vapour, spectrometers were evacuated or purged.polychlorodibenzofurans (PCDFs). Vacuum optical systems for ICP spectrometry have the prob- Apart from halogen determination in waste oils, the lem that they are often not able to attain a perfect vacuum determination of PCBs is often required. In the past, PCBs and that back-streaming vapours from the oil-filled vacuum have been used in various technical applications, i.e., in pumps cause coatings on the optical surfaces of the spec- electrical transformers or capacitors, because of their chemical trometer.1 A successful approach to avoid these disadvantages and physical properties, such as thermal stability, low flammis a nitrogen-filled optical system.This technique allows ability and low conductivity. Since it was shown that PCBs element determinations down to a wavelength of 120 nm. are carcinogenic and persistant, their use was limited. Today, A wide range of organic samples are analyzed by ICP the production and import of PCBs is prohibited in many spectrometry, such as solvent extracts, chromatography countries, but waste oils will still be the main source for PCBs eZuents or the organic solutions themselves.Therefore, the in the future. The recycling and disposal of waste oils is characteristics of an ICP operating with organic aerosols were regulated by the metal, the halogen and the PCB content in studied in several papers.4–9 However, the majority of organic many countries. In the US, waste oil containing more than samples are various oils such as crude oils, lubricating oils, 0.1% (m/m) of Cl is considered to be hazardous waste. In refined oils or waste oils.Typical examples of ICP applications Germany,22 the tolerance limit for the total Cl content in are the determination of (a) impurities in crude oils or refined waste oils is 0.2% (m/m) and the tolerance limit for PCBs is oils, (b) additives in oils or in gasoline, (c) wear metals in 20 mg kg-1.PCB analysis by gas chromatography with mass lubricating oils, or (d) elements in waste oils.10–21 Monitoring spectrometric or electron-capture detection is often time conof (a) crude oils or refined oils, for instance, could be required suming and therefore rapid screening tests for PCBs are useful.23 to characterize the oil itself or to avoid destruction of catalysts J. Anal. At. Spectrom., 1999, 14, 699–702 699This paper describes the determination of non-metals in oils PCBs, diVerent weighed portions of Arochlor 1260 were diluted to 2 g with the base oil and made up to 5 g with using prominent spectral lines in the vacuum ultraviolet spectral region down to Cl 134.72 nm.The optimization of the kerosene at room temperature. ICP operating conditions for the non-metals compared with some metals was performed and the detection limits in oils for Results and discussion new prominent spectral lines in routine analysis were calculated.Further, ICP-OES was investigated as a screening test For non-metals such as Cl, Br, I, N, P and S the most intense spectral lines are in the vacuum ultraviolet spectral region for the highest possible PCB concentration in waste oils calculated from the total Cl concentration because total Cl below 190 nm. Typical examples of these spectral lines are given in Table 2. The measured intensity of the most intense can be determined at low ppm levels in oils.spectral line for each element in Table 2 is standardized to 100. Several of the spectral lines in Table 2 have not been Experimental discussed in the literature in combination with ICP-OES for routine analysis. For Cl, which is environmentally one of the Apparatus most important elements to be determined in organic The investigations described below were performed using a compounds, the intense spectral line at 134.72 nm was used. Spectro ICP (Spectroflame Compact E, Spectro Analytical When analyzing organic samples with ICP-OES sometimes Instruments, Kleve, Germany) with a free-running 27.12 MHz the addition of oxygen is required.Reasons for oxygen addition generator. For sample introduction, a cross-flow nebulizer are (a) the prevention of carbon deposits on the injector quartz (Spectro) was used in combination with a Scott-type double tube or (b) the minimization of spectral interferences by pass spray chamber. Because of the excellent plasma stability carbon or carbon molecules (C, CH, C2, etc.).In Fig. 1 the for all samples a cooled spray chamber for oil analysis was eVect of adding small amounts of oxygen to the outer gas on not required. Small amounts of oxygen (80–240 ml min-1) the background equivalent concentration (BEC) for diVerent were added to the outer plasma gas to avoid carbon deposits metals and non-metals is shown. The BEC was calculated on top of the quartz injector tube. Further ICP operating from BEC=c/SBR, where c is the concentration of the conditions are described in Table 1.elements in the oil standard and SBR the signal-to-background The polychromator optics of the ICP spectrometer are ratio. The BECs for the elements are displayed in Fig. 1 as enclosed in a chamber filled with nitrogen to slightly above relative values. For the non-metals Cl, Br, I and S, oxygen atmospheric pressure. By the use of a membrane pump the addition was found to cause an overall increase in the BEC nitrogen filling gas is circulated through a filter which removes [Fig. 1(a)]. Similarly, the BEC of the metals is significantly oxygen, water vapour and other species that might absorb increased on adding oxygen to the outer gas, except for Na electromagnetic radiation. This ensures the best possible light [Fig. 1(b)]. Hence, oxygen addition is disadvantageous for the transmission in the wavelength range down to 120 nm and in limits of detection for all the determined elements with the addition the optical system is protected from contamination.exception of Na. The main spectral line of Na (Na I 589.59 nm) The use of a nitrogen-filled spectrometer saves costs compared in oil analysis is aVected by a carbon emission. The improved with nitrogen-flushed systems. values of the BEC for Na are due to a reduced carbon spectral interference on adding oxygen. Line selection The eVect of the nebulizer gas flow on the relative BEC values of the elements is shown in Fig. 2. The optimum BEC The non-metals determined in oils were Cl, Br, I, P and S at (the lowest BEC) for the non-metals Cl, Br, I and S is achieved 134.72, 163.34, 161.76, 177.50 and 180.73 nm, respectively. at 0.5–0.6 l min-1 [Fig. 2(a)]. For the elements Al, Ba, Cd, Ni Metals determined were Al, Ba, Cd, Ni and V at 308.21, and V, a higher nebulizer gas flow of 0.9–1 l min-1 is advanta- 455.40, 226.50, 231.60 and 292.46 nm, respectively. Sample preparation Table 2 Prominent non-metal spectral lines below 190 nm.a Spectral line intensities were measured with the described instrument A 0.1–1 g amount of the standard oils (Alpha Resources, Stevensville, MI, USA) or 1 g of the sample was made up to Element Line/nm Relative intensity 2 g with a base oil (Conostan, Ponca City, OK, USA) and Cl 134.72 100 then diluted to 5 g with kerosene (Sigma-Aldrich, Steinheim, 135.16 35 Germany) at room temperature.The base oil and kerosene 136.34 30 were applied to compensate for diVerences in the viscosity of Br 148.84 90 sample and standard solutions. 153.17 26 The PCB samples were prepared from Arochlor 1260 154.07 100 157.48 10 (Promochem, Wesel, Germany), and were diluted 1+9 (m/m) 163.34 12 with kerosene at 40 °C. Arochlors were the most widely used I 142.56 75 industrial PCBs. The Arochlor 1260 sample is a reference 158.27 50 material for the analysis of PCBs with a Cl concentration of 161.76 95 60% (m/m). For the measurements of the Cl recoveries in 179.91 85 183.04 100 N 149.26 100 Table 1 ICP operating conditions 174.27 40 174.52 20 Generator Free-running at 27.12 MHz P 177.50 100 Power/W 1200 178.29 75 Nebulizer ‘Cross-flow’ (Spectro) S 166.67 10 Spray chamber Double pass, Scott-type 180.73 100 Outer gas/l min-1 12 182.04 75 Intermediate gas/l min-1 2 182.62 25 Nebulizer gas/l min-1 0.9 Nebulizer pressure/bar 3.1 aAll intensities are standardized to the most intense line of that Sample uptake rate/ml min-1 0.8 particular element. 700 J. Anal. At. Spectrom., 1999, 14, 699–702Fig. 1 EVect of added oxygen to the outer gas on the BEC for Fig. 3 EVect of the generator power on the BEC for diVerent nondi Verent non-metals (a) compared with some metals (b): A, S metals (a) compared with some metals (b): A, S; B, Br; C, I; D, Cl; 180.7 nm; B, I 161.7 nm; C, Br 163.3 nm; D, Cl 134.7 nm; E, Cd E, V 292.5 nm; F, Ba; G, Al; H, Cd; I, Ni. Wavelengths as in Fig. 1. 226.5 nm; F, V 292.5 nm; G, Al 308.2 nm; H, Ba 455.4 nm; I, Ni 231.6 nm; J, Na 589.59 nm.higher generator power seems to be advantageous for the determination of halogens and sulfur with higher excitation energy for their spectral lines compared with the determined metals Al, Ba, Cd, Ni and V. Thus, for the simultaneous determination of non-metals and metals, compromise ICP operating conditions should be selected. For the determination of Cl, Br, I and S, a higher generator power and a lower nebulizer gas flow is required to improve the limits of detection for these elements.Followed by univariate searches for the optimum values of power, nebulizer gas flow and additional oxygen gas flow, a multivariate simplex optimization was used to establish the optimum plasma parameters for a low BEC of Cl. The parameter ranges for the power, nebulizer gas flow and additional oxygen gas flow are 1000–1600 W, 0.4–1.2 l min-1 and 80–240 ml min-1, respectively. For the simplex procedure the power, nebulizer gas and added oxygen used were varied in steps of 100 W, 0.1 l min-1 and 80 ml min-1, respectively.The lowest value for the BEC of Cl was achieved at 1500 W generator power, 0.6 l min-1 nebulizer gas flow and 80 ml min-1 additional oxygen flow to the outer plasma gas. In comparison with Cl, the lowest BEC of the metals V, Ni and Cd was found at 1200 W, 0.9 l min-1 and 80 ml min-1 after simplex optimization. Table 3 lists the detection limits for non-metals and metals in oil.The detection limits were calculated from the standard deviation (3s) of ten measurements of the blank solution (base oil/kerosene). The detection limits for the non-metals Cl, Br, Fig. 2 EVect of the nebulizer gas flow on the BEC for diVerent nonmetals (a) compared with some metals (b): A, S; B, Br; C, I; D, Cl; I, S and P in the undiluted oil are 0.9, 1.6, 0.47, 0.07 and E, Al; F, V; G, Ba; H, Cd; I, Ni. Wavelengths as in Fig. 1. 0.04 mg kg-1, respectively.The detection limit for Cl (0.9 mg kg-1) is more than a factor of 2000 below the tolerated limit stated in the German waste oil regulations geous [Fig. 2(b)]. This eVect could be caused by the higher excitation energy of the non-metals compared with the metals (2000 mg kg-1). Therefore, it is possible to analyze highly viscous oils using suYcient dilution with kerosene to adjust for the applied spectral lines. As shown in Fig. 3 there is also a diVerence in the eVect of the viscosities of samples and standards. The following example discusses the ability of ICP-OES to the generator power on the relative BECs of the non-metals and metals.For Cl, Br, I and S, the lowest BECs are achieved monitor the maximum PCB content in waste oils by the determination of Cl. It is assumed that the Cl concentration at 1400–1600 W [Fig. 3(a)], while the lowest BECs for Al, Ba, Cd, Ni and V are in the range 1200–1300 W [Fig. 3(b)]. A in a PCB mixture is 50% (m/m).This assumption may be J. Anal. At. Spectrom., 1999, 14, 699–702 701Table 5 Comparison of certified and experimentally determined Cl Table 3 Detection limits (LOD) of the elements in the undiluted oil after simplex optimization of the ICP operating conditions for Cl contents in Arochlor 1260 samples diluted with a base oil and kerosene 134.724 nm (power: 1500 W, nebulizer gas: 0.6 l min-1, O2 additional gas: 80 ml min-1). The oil samples were diluted 1+4 (m/m) with kerosene for analysis Sample 1 2 3 PCB concentration/mg kg-1 1.77 16.13 32.68 Metals Non-metals State of ionization Line/nm LOD/mg kg-1 Certified Cl concentration/mg kg-1 1.06 9.68 19.61 Determined Cl concentration/mg kg-1 1.08 9.63 19.28 Al I 308.21 0.15 Ba II 455.40 0.003 Relative standard deviation (%) 4.8 1.6 0.9 Cl recovery (%) 101.8 99.5 98.3 Br I 163.34 1.6 Cd II 226.06 0.009 Cl I 134.70 0.9 I I 161.76 0.47 result in this concentration range.Consequently, ICP-OES is Na I 589.59 0.08 suitable as a fast screening test of the highest possible PCB Ni II 231.60 0.012 concentration calculated from the total Cl concentration to P I 177.50 0.04 monitor limiting values of PCBs in waste oils.S I 180.73 0.07 V II 292.46 0.018 Conclusions Using a nitrogen-filled spectrometer, ICP-OES provides the regarded as practical because the Cl concentration in PCBs ability to determine the non-metals Cl, Br, I, S and P at their for technical applications is between 30 and 60% (m/m) with intense spectral lines in the range 130–190 nm.These non- the higher concentrations being preferred. In Table 4 the PCB metals and a few metals in oils were determined simultaneously isomers are listed according to their Cl concentration. The with a nitrogen-filled polychromator. In particular, the deter- mean value for the Cl concentration in a PCB isomer is 52.8%. mination of Cl in organic waste oils is of great environmental If the Cl concentration in the PCB mixture is 50% (m/m), interest because of the hazardous potential of chlorinated then 20 mg kg-1 PCB of this mixture in an oil corresponds to organic compounds such as PCBs, PCDDs and PCDFs.The 10 mg kg-1 Cl. This is more than a factor of 10 higher than problem of determining these hazardous organochlorine com- the above-determined detection limit for Cl in oil. The pounds will be one of the important analytical tasks of the requested limiting value according to the German waste oil future.Because of the low detection limits for the halogens at regulations22 is 20 mg kg-1 PCB. Thus, a fast screening test spectral lines down to 134.7 nm for Cl, the determination of for the highest possible PCB content in waste oils calculated halogens in waste oils can be performed with ICP-OES. from the Cl concentration can be accomplished with ICPFurther fast screening tests for an upper limit of the PCB OES, even if the determination of the total Cl content in waste concentration calculated from the Cl concentration are possible oils does not necessarily determine the PCB content.A reliable with ICP-OES. statement of the PCB concentration calculated from the total Cl concentration could not be given, but it is possible to state whether a limit value for PCBs in waste oil regulations (a) References might be exceeded or (b) is in no case exceeded. In the first 1 V. B. E. Thomsen, G. J. Roberts and D. A. Tsourides, Int.Lab., case (a) further PCB analyses should be conducted, but in the 1997, 27, 9A. second case (b) no further eVorts are required to show that 2 J. Alvarado and J. W. Carnahan, Appl. Spectrosc., 1993, 47, 2036. the PCB concentration is below a limiting value. Therefore, 3 D. D. Nygaard and D. A. Leighty, Appl. Spectrosc., 1985, 39, 968. quantitative PCB determination by GC-MS which often 4 C. Pan, G. Zhu and R. F. Browner, J. Anal. At. Spectrom., 1990, requires time consuming sample preparation may be avoided, 5, 537.resulting in considerable savings of time, personnel and 5 A. W. Boorn and R. F. Browner, Anal. Chem., 1982, 54, 1402. 6 A. W. Boorn, M. S. Cresser and R. F. Browner, Spectrochim. expenditure. Acta, Part B, 1980, 35, 823. To verify the applicability of the Cl determination in PCBs, 7 M. W. Blades and B. L. Caughlin, Spectrochim. Acta, Part B, the technically applied PCB mixture Arochlor 1260 containing 1985, 40, 579. 60% (m/m) Cl was analyzed at three diVerent dilution levels 8 D.G. Weir and M. W. Blades, J. Anal. At. Spectrom., 1994, with base oil and kerosene as described above. Table 5 lists 9, 1323. the experimentally determined Cl concentration compared 9 D. G.Weir and M. W. Blades, J. Anal. At. Spectrom., 1996, 11, 43. 10 R. I. Botto and J. J. Zhu, J. Anal. At. Spectrom., 1996, 11, 675. with the certified Cl concentration in Arochlor 1260. Although 11 P. N. Bangroo, C. R. Jagga, H. C. Arora and G.N. Rao, At. sample 1 was analyzed close to the detection limit, the recover- Spectrosc., 1995, 16, 118. ies and the relative standard deviation were satisfactory. The 12 X. R. Liu and G. Horlick, J. Anal. At. Spectrom., 1994, 9, 833. relative standard deviation for the Cl concentration in the 13 M. Bettinelli and P. Tittarelli, J. Anal. At. Spectrom., 1994, 9, 805. diluted PCBs is in the range 0.9–4.8% and represents a good 14 R. I. Botto and J. J. Zhu, J. Anal. At. Spectrom., 1994, 9, 905. 15 M. Murillo and J. Chirinos, J. Anal. At. Spectrom., 1994, 9, 237. 16 M. Murillo, A. Gonzales, A. Ramirez and N. Guillen, At. Table 4 Chlorine concentration in PCB isomer groups Spectrosc., 1994, 15, 90. 17 M. Murillo, N. Carrion and J. Chirinos, J. Anal. At. Spectrom., Number Cl concentration 1993, 8, 493. PCB Formula of isomers (%) 18 R. I. Botto, J. Anal. At. Spectrom., 1993, 8, 51. 19 R. I. Botto, Spectrochim. Acta. Rev., 1991, 14, 141. Monochlorobiphenyl C12H9Cl 3 19 20 J. D. Algeo, D. R. Heine, H. A. Phillips, F. B. G. Hoek, Dichlorobiphenyl C12H8Cl2 12 32 M. R. Schneider, J. M. Freelin and M. B. Denton, Spectrochim. Trichlorobiphenyl C12H7Cl3 24 41 Acta, Part B, 1985, 40, 1447. Tetrachlorobiphenyl C12H6Cl4 42 49 21 J. L. Fabec and M. L. Ruschak, Anal. Chem., 1985, 57, 1853. Pentachlorobiphenyl C12H5Cl5 46 54 22 Alto�lverordnung, (BGBl 1), 1987, 2335. Hexachlorobiphenyl C12H4Cl6 42 59 23 P. Richner and S. Wunderli, J. Anal. At. Spectrom., 1993, 8, 45. Heptachlorobiphenyl C12H3Cl7 24 63 Octachlorobiphenyl C12H2Cl8 12 66 Nonachlorobiphenyl C12H1Cl9 3 69 Paper 8/07024E Decachlorobiphenyl C12Cl10 1 71 702 J. Anal. At. Spectrom., 1999, 14, 699–7

ISSN:0267-9477    

DOI:10.1039/a807024e

出版商:RSC    

年代:1999

数据来源: RSC

摘要:

Development of an analytical scheme for the direct determination of antimony in geological materials by automated ultrasonic slurry sampling-ETAAS M. Jesu�s Cal-Prieto, Alatzne Carlosena,* Jose� M. Andrade, Soledad Muniategui, Purificacio�n Lo� pez-Mahý�a, Esther Ferna�ndez and Darý�o Prada Department of Analytical Chemistry, University of La Corun�a, Campus da Zapateira s/n, E-15071 La Corun�a, Spain. E-mail: alatzne@udc.es Received 21st September 1998, Accepted 7th January 1999 The ongoing application of ultrasonic probes for slurry homogenization gives new additional advantages with respect to the traditional slurry analyses by ETAAS, namely speed, automation, improved reproducibility, etc. Despite all these advantages, there are still diYculties in the optimization stages of the analytical procedures because of the large number of variables to take into account.Accordingly, several chemometric techniques were applied to help in the development of a systematic analytical scheme to determine Sb directly in soils and sediments by means of ultrasonic slurry sampling (USS)-ETAAS. They are intended not only to optimize the slurry preparation but also to diminish the amount of benchwork to be done.The chemometric techniques and their objectives were as follows: Plackett–Burman designs, to assess the influence of six analytical variables; optimization of the important analytical variables; a two-way ANOVA, to investigate the influence of sample cups and replicates; and control charts, to monitor the graphite tube performance.Additionally, two pipetting options and three quantification methods were evaluated. Analytical results for Sb determined in five certified reference materials confirmed the usefulness of the ultrasonic slurry sampler (USS-100) combined with ETAAS for a direct and almost fully automatic analysis of complex matrices. determination of this element in several environmental samples Introduction such as soils, sediments and air.Additionally, this technique Traditionally, very little attention has been paid to antimony in is increasingly being used in the direct analysis of solid samples soils, sediments and biomaterials, mainly because it is not either with solid sampling or slurry sampling. The main recognized as having nutritional significance and its contents in benefits of direct solid analysis by ETAAS are reduced time most materials are very low.Nevertheless, compounds of anti- and sample preparation requirements, decreased likelihood of mony are moderately toxic to most organisms, Sb3+ compounds analyte loss prior to analysis, minor use of acids and their being more toxic than Sb5+ compounds (similarly to the corresponding reduction in the production of wastes and a corresponding arsenic compounds).1 Exposure to antimony lower probability of sample contamination. During the 1990s, compounds causes cellular damage in the lungs, heart and the use of slurry sampling-based analytical procedures has kidneys, but the toxicity mechanisms are still not well known.2 been gaining acceptance to cope with diVerent types of samples, Many industrial processes such as the manufacture of alloys, such as soils and sediments,7–14 thanks to several practical paints, compact discs and bactericides involve the use of anti- advantages.Some of them are as follows: (i) conventional mony compounds which appear in most industrial eZuents.liquid autosamplers can be used; (ii) several replicates can be Furthermore, antimony may reach soils by wet and dry depos- made in just one aliquot; (iii) handling of higher masses of ition following emissions from incineration and fossil fuels sample; (iv) ability to change the slurry concentration; (v) combustion, vehicle traYc and by the addition of soil amend- better analytical performance can be achieved since the slurry ments such as chemical fertilizers, sewage sludge and fly-ash.3,4 technique combines the benefits of solid and liquid sampling; Antimony in air is considered to be associated with the smaller and (vi) improved repeatability and reproducibility.Moreover, particles, inhalation being the major cause of antimony intoxi- the development and application of an ultrasonic probe cation. The occurrence of this element in airbone dust and its (Perkin-Elmer USS-100)15–21 for slurry homogenization feacorresponding deposition and accumulation in vegetation were tured new additional advantages such as higher speed, autostudied by Dietl et al.,4 who confirmed antimony enrichment mation of slurry sample preparation and its introduction into near roadways by monitoring it in standardized grass cultures the furnace, improved reproducibility and analyte extraction and total depositions.Moreover, the antimony deposition might and improved representativeness of the aliquot analysed.also aVect the antimony cycling in the oceans.In that sense, Despite all these advantages, there are diYculties in the Takayanagi et al.5 investigated vertical profiles of dissolved optimization stages of the analytical procedures because of inorganic antimony in sea-water, reflecting higher concen- the large number of variables to take into account: mass of trations on the surface which decreased to a near constant level sample, volume of slurry, mixing time, power output of the with depth. Despite this, the natural low levels of antimony in ultrasonic probe, use of thixotropic agents, chemical characterthe earth’s crust (0.2–0.3 mg g-1)6 make it a potential marker istics of the diluent, number of replicates per sample, number of environmental metallic contamination.of particles present in the injected sample volume, etc. The good sensitivity of electrothermal atomic absorption Although the classical schemes of work (many of them ‘trial and error’ type) have given good results in previous method spectrometry (ETAAS) makes it a suitable technique for the J.Anal. At. Spectrom., 1999, 14, 703–710 703developments, few attempts have been made to incorporate a polyethylene autosampler cup and suspending it in a small chemometric techniques not only with optimization purposes volume of diluent (typically 0.5 or 1 ml ). Then, an aliquot is but also as a way to diminish the amount of work to be done. injected into the graphite furnace.The mass of the powder In this way, Koch et al.22 have shown an excellent approach will depend on the total amount of Sb that the sample is to deal with these kind of developments focused on the final expected to present. The higher the content, the lower is the measurement step. Here, the following topics are applied to sample mass. There are practical limits for both the minimum achieve a reliable and optimized procedure for the direct and maximum amounts of sample (see Results and discussion) determination of Sb in soils and sediments by ultrasonic slurry that can be considered (the latter situation is mostly caused sampling (USS)-ETAAS: by the relatively poor eYciency of conventional autosamplers (i) a Plackett–Burman saturated factorial design23 to deter- when used for the aspiration of a concentrated slurry aliquot). mine the influence of six analytical variables on the preparation The experimental conditions were carefully optimized for of the slurries; slurry preparation whereby sample aliquots (2–150 mg) were (ii) optimization of the critical variables; directly weighed in a polyethylene cup and suspended in 1 ml (iii) a two-way ANOVA to investigate the influence of both of 0.5% v/v HNO3 and the ultrasonic power setting and the the number of sample cups and number of replicates per agitation time were set at 40W and 10 s, respectively to obtain sample cup; levels for cups are 1, 2, 3 and 4 and levels for homogeneous slurries. The major steps are depicted in Fig. 1, replicates are 2, 3, 4 and 5; which illustrates the straightforward operations required by (iv) comparison of three quantification methods using the this analytical approach. Hotelling confidence intervals associated with the calibration Three roadside soil samples and three marine sediment line;24 samples were collected. Soil samples were from a metrol charts to monitor the behaviour and ageing of city, La Corun� a, NW Spain, at a 0–5 cm depth; two of them the pyrolytic graphite tubes (with a L’vov platform).were from city gardens and the other was from a crop soil. At every sampling site, several sub-samples were collected from a Experimental 0.5 m2 grid area and mixed to obtain one bulk sample. Regarding the sediments, they were collected from the La Apparatus Corun� a estuary in coastal areas. Sampling was made with a The determination of Sb was performed on a Perkin-Elmer ‘box-corer’ probe.The superficial sediment layer (0–10 cm) was (U� berlingen, Germany) Model 4100 atomic absorption spec- taken. Samples were lyophilized (-40 °C) for 36–48 h until trometer equipped with deuterium background correction, an constant mass, then sieved using a 2 mm mesh sieve and ground. HGA-700 graphite furnace, an AS-70 autosampler and a Accuracy and precision were assessed by using five certified USS-100 ultrasonic slurry sampler. Argon was used to provide reference materials, namely two NCR CRM (National the inert atmosphere within the furnace.Pyrolytic graphite- Research Council of Certified Reference Materials, China) coated graphite tubes (Z-tek, Amsterdam, The Netherlands) soils (GBW07401 and GBW07409), one NCR CNRC with preinserted pyrolytic L’vov platforms were employed. (National Research Council of Canada) marine sediment Atomization was accomplished by using the maximum power (BCSS-1) and two BCR CRM (Community Bureau of heating and gas stop flow.Peak heights were recorded and Reference, Brussels, Belgium) materials (a calcareous loam used to quantify the Sb analytical signal. Measurements were soil, CRM 141, and an estuarine sediment, CRM 277). made by using a hollow cathode lamp (Perkin-Elmer) at the 217.6 nm Sb wavelength, using a 0.7 nm spectral bandwidth. Analytical procedure For the statistical treatment the Statgraphics software package Sb was determined directly by using automated USS-ETAAS.was used. Slurry aliquots (5–20 ml ) were injected after ultrasonic agitation, then dried using two steps: 100 °C (5 s ramp, 10 s hold) Reagents and 130 °C (30 s ramp, 10 s hold). The optimum pyrolysis and All reagents were of analytical-reagent grade. High-purity atomization temperatures were found to be 900 °C (10 s ramp, water (Milli-Q Water System, Millipore, Madrid, Spain) was 30 s hold) and 1700 °C (0 s ramp, 2 s hold), respectively. Two employed throughout. The acids (HNO3 and HCl) were of slurries were prepared for each sample and they were analysed Suprapur grade (Merck, Darmstadt, Germany).An Sb stock in triplicate. Quantification was performed using peak height standard solution (1.000 g l-1) for atomic absorption was measurements and the method of addition calibration obtained from Panreac (Barcelona, Spain) and working Sb implemented in the spectrometer management software. The standard solutions were prepared daily by diluting appropriate slurry, Sb spike and diluent were introduced into the furnace aliquots of the commercial stock standard solution.Standard by the autosampler using the ‘pipette separate’ mode. The aqueous solutions of the chemical modifiers containing total volume injected was 20 ml. 10 000 g l-1 of both magnesium and ammonium dihydrogenphosphate were prepared from Mg(NO3)2 (99.995%) Results and discussion (Aldrich, Milwaukee, WI, USA) and NH4H2PO4 (99.999%) (Aldrich), respectively.A palladium solution was made by Results are presented for studies carried out to develop and dissolving Pd powder (99.999%) (Aldrich) in the minimum validate the analytical method intended to determine Sb in volume of aqua regia, followed by evaporation to near dryness geological materials by USS-ETAAS. The flow chart depicted and dilution to a final concentration of 1.000 g l-1 with in Fig. 2 shows each major analytical step and the associated 0.5% v/v HNO3. The same procedure was followed to prepare studies.This flow chart is not ‘strict’; this means that the order nickel solution. of each test or particular study might be changed to other All glassware, plasticware, pipette tips and storage bottles sequences. Such diYculty arises as a direct consequence of the were soaked in 10% v/v HNO3 for 24 h and rinsed with high- intrinsic complexity of the slurry-based analytical techniques. purity water at least three times prior to use. Sometimes, several tests have to be performed to establish some particular conditions, but which is first? Hence, this Slurry and sample preparation scheme represents an approach to the development of analytical methods to determine trace metals directly by slurry Grossly speaking, the preparation of the sample for the sampling-ETAAS and it resembles a general-purpose pro- ultrasonic slurry sampling using the USS-100 probe consists in weighing a certain amount of sample powder directly into cedure already published.17 704 J.Anal. At. Spectrom., 1999, 14, 703–710Fig. 1 System used for the direct and almost fully automatic analysis of soils and sediments by ultrasonic slurry sampling-ETAAS for the determination of Sb. media, mixed using the USS-100 probe and the Sb content determined in the supernatants (separated by centrifugation). In all cases, the Sb extraction into the liquid phase was negligible. This reveals that the use of high concentrations of nitric acid is not essential (as usually recommended for other metals12,16,17).Also, this will increase the lifetime of the graphite tube. Therefore, the representativeness of the analysed slurry aliquot becomes more critical. Selection of the critical variables. Regarding the slurry preparation itself, there are a large number of variables to deal with9,18 and, also, diVerent ways to study them simultaneously to evaluate their influence on the analytical response, e.g., simplex22 and evolutionary operation procedures (EVOP), Latin squares and experimental designs.22 Of these, the Plackett–Burman partial factorial designs are very appealing since they allow the maximum extraction of information with a minimum level of laboratory work.They have also been proposed by ASTM26 as a simple way to obtain ‘rugged’ analytical methods. Accordingly, one of the Plackett–Burman Fig. 2 General flow chart of the main steps needed to carry out designs was chosen to elucidate which variables are important analyses by slurry sampling-ETAAS showing the main statistical studies associated with each. (and to what extent) when preparing slurry solutions.Table 1 presents the variables considered in this study and also the experimental conditions for each trial, the correspond- The inherent diYculties with this analytical technique call ing experimental absorbance and the calculated eVect for each for an iterative process in order to optimize the method but variable (experimental t-value). All the calculations were made this is highly expensive and time- and labour-consuming, so it as usual.was only applied when strictly necessary. Taking the tabulated t-value (95% confidence) as 1.94, it can be observed that only the F factor (time of agitation) was Slurry preparation statistically significant. Nevertheless, the relatively high t- Stability of the slurry. Precision is highly dependent on the values for factors A (sample mass) and B (slurry volume) homogeneity of the analyte distribution in the sample and make it interesting to consider both of them in some more analyte partitioning between the solid and liquid phases.16 detail (although they are not statistically significant). When the analyte was extracted easily into the liquid phase Moreover, the confounding eVect most likely to happen is the (which is improved by ultrasonic agitation and acid content), interaction between variables A–B and variable F.This means slurry stability became less critical. that varying both factors, the mass of sample and the volume Accordingly, to study the slurry stability, various slurries of of slurry, may have the same eVect as varying the time of the same sample were prepared (50 mg ml-1 in 0.5% v/v agitationover, the negative sign means that the HNO3) using the USS-100 probe at a power setting of 40 W absorbance diminishes when the time of agitation increases for 10 s.Several aliquots were taken periodically and measured or, alternatively, when the mass of sample and volume of (homogenization was carried out before the first sampling). slurry increase. Of course, both variables point towards the The average absorbances showed that after 185 s the so-called mass/volume ratio so important in slurry studies.19 absorbance dropped from 0.158 to 0.003 owing to the high The explanation for the negative sign is that when the agitation sedimentation rate after stopping agitation and the small to time is too long (alternatively, the mass/volume ratio), a trend negligible amount of analyte extracted into solution.to form emulsions in the autosampler cup was observed. Unfortunately, this is the minimum time to make two consecutive injections using the AS-70 autosampler and the furnace Optimization of the selected variables. Following the experimental design, it seems that there is only one variable to be programme.Hence the need for ultrasonic agitation prior to each measurement was clearly established (see also Miller- optimized (time of agitation) and, therefore, a classical univariate scheme is adequate. All the variables not statistically Ilhi25 for analogous results for materials with high densities). To evaluate the analyte extraction eYciency, diVerent HNO3 significant were fixed at values considered either in the experimental design or within the experimental range, as follows: concentrations (from 0.5 to 5% v/v) were assayed as suspension J.Anal. At. Spectrom., 1999, 14, 703–710 705Table 1 Experimental conditions for each trial of the experimental design and statistical eVect calculated for each variable (experimental t-value, 95% confidence interval ) Variable Mass/ Volume/ [HNO3] [Triton X-100] Power/ Time/ Experimental Experiment mg ml (%) (%) W s ‘Dummy’ absorbance 1 100 1 5 0 30 10 — 0.00517 2 50 1 5 0.04 50 10 — 0.00428 3 50 0.5 5 0.04 30 30 — 0.00656 4 100 0.5 0.5 0.04 50 10 — 0.00447 5 50 1 0.5 0 50 30 — 0.00532 6 100 0.5 5 0 50 30 — 0.00523 7 100 1 0.5 0.04 30 30 — 0.05080 8 50 0.5 0.5 0 30 10 — 0.00642 Statistical eVect— Variable Mass Volume [HNO3] [Triton X-100] Power Time ‘Dummy’ EVect -0.00065 -0.00071 -0.000014 -0.00044 0.000126 -0.00097 0.000463 texp 1.38 1.52 0.03 0.94 0.27 2.07 1 mass=50 mg; volume=1 ml; Triton X-100=not used; L’vov platform.Therefore, the upper limit for the sample mass is well established (150 mg), and it seems reasonable to assume HNO3=0.5% v/v; and power=40 W.Fig. 3 shows that after a minimum agitation time of 5 s, no influence was observed that the lower limit will depend on the Sb content of the samples (ensuring a representative number of particles in the up to 30 s. This surprising result strongly suggests that the mass/volume ratio is the variable that needs more attention, furnace). The upper limit achieved here is much larger than those usually reported when slurries are prepared directly in so its optimization is required (agitation time fixed at 10 s).Hence the mass/volume ratio was studied as usual in the the sample cup, which allows one to improve the sensitivity and detection limits. Small masses of another certified reference related literature.27 Seven levels of mass from the NCR CRM GBW07401 soil material (BCR CRM 277 estuarine sediment; 3.9±1 mg g-1 Sb content, not certified) were assayed and the results obtained sample were suspended in 1 ml of the diluent and then analysed.For each level two samples were prepared, three were in good concordance with the certified value. There is another important factor to deal with, which is replicates per sample. No diVerences were observed in the Sb response employing masses from 20 to 150 mg (Table 2). In that the mass/volume ratio may also limit the volume of slurry to be introduced into the furnace. Table 3 summarizes the contrast, high and low responses were obtained when 10 and 200 mg were used, respectively.When 200 mg of sample were results obtained for several masses of the certified reference soil GBW07401, injecting diVerent volumes of each one. An weighed, the high concentration of the slurry led to ineYcient pipetting of the aliquot, whereas 10 mg of sample gave poor important diminution in absorbance can be observed when 200 mg are used. Both the mean value and associated RSD results owing to the small amount of Sb introduced onto the became negatively aVected to a large extent, as stated previously.Such a diminution becomes much greater as the volume of slurry injected in the furnace increases. From the foregoing, the following operational guidelines can be established: (i) samples with moderate Sb contents (2–<20 mg g-1) would be analysed by diluting the sample, either by using low masses (2–10 mg) or by injecting small volumes of slurry into the furnace (5–10 ml ); for higher concentrations, less sensitive non-resonance lines may be useful;9,17,18 Fig. 3 Peak height absorbances recorded after applying diVerent ultrasonic homogenization times for a soil slurry (0.5% v/v HNO3; Table 3 Relationship between the pipetted volume of slurry and the Tpyrolysis=700 °C; Tatomization=1700 °C; power=40 W). sample mass used to prepare 1 ml of soil slurry (total volume injected fixed at 20 ml using 0.5% v/v HNO3) Table 2 Analytical response and limits of detection (LOD) and quanti- Sample mass/mg fication (LOQ) as a function of the soil sample mass employed to prepare 1 ml of slurry (volume injected=20 ml ) 20 50 200 Sample mass/ Corrected LOD/ LOQ/ Slurry volume/ml Corrected absorbancea (n=3) mg absorbancea (n=3) mg g-1 mg g-1 5 7.1 6.9 5.3 2 — 0.620 2.06 10 6.2 7.0 4.4 10 7.0 0.124 0.41 15 6.8 6.6 3.6 20 6.3 0.062 0.21 20 6.3 6.7 2.6 40 6.4 0.031 0.10 x: 6.6 6.8 4.0 50 6.7 0.025 0.08 RSD (%) 6.4 2.7 28.9 92 5.7 0.013 0.04 150 5.8 0.008 0.03 aCorrection made for each volume injected and sample mass 200 2.6 — — Aabsorbance×103 mg × total volume/ml volume injected/mlB. aCorrection made for each sample mass (103 absorbance mg-1). 706 J. Anal. At. Spectrom., 1999, 14, 703–710(ii) samples with low Sb contents (<0.5 mg g-1) would be Therefore, it was decided to check the influence of changing processed either by weighing high masses of sample (<150 mg) the graphite tube (i.e., ‘tube 1’ and ‘tube 2’) on the absorbance.or injecting large volumes of slurry (20 ml ). The results obtained for the 2-way ANOVA are summarized in Table 4 considering absorbance values (they were also made Representativeness of the aliquot injected into the graphite considering concentrations and all the results agreed fairly furnace well ). The eVect of changing the number of replicates per sample cup is statistically significant, which can be verified An important concern when direct analyses of solid samples from Fig. 4, where all the confidence intervals (95%) for the are performed by ETAAS is the representativeness of the average values for each level overlap substantially except for aliquot introduced into the graphite furnace.8 The representathat corresponding to the two replicates level. Accordingly, tiveness of the aliquot will determine both the accuracy and three replicates should be performed to improve the precision precision of the analytical results, and it depends on the of the within-sample measurements and to avoid the slight homogeneity of the analyte distribution in the sample and the negative trend associated with performing only two replicates.analyte partitioning between the solid and liquid phases of the No special benefits are obtained with more than three repli- slurry.16,21 Taking into account that extraction of Sb into cates. These results reflected that the automatic ultrasonic the liquid phase was negligible, it seems mandatory to evaluate mixing provided a uniform distribution of particles in the the representativeness of the analysed slurry aliquot.autosampler cup and that the autosampler exhibited a low performance variability. Hence, the amount of sample ali- Number of particles. It is accepted that at least 50 individual quoted by the autosampler from the slurry is representative of particles of the sample should be present in the aliquot injected the slurry itself. into the graphite tube to avoid statistical limitations of the Regarding the number of sample cups, this factor was not sampling techniques.18 statistically significant (see Table 4).Therefore, it can be The sample powder density and the particle size, assuming deduced that Sb exhibits a homogeneous distribution in spherical particles, allow the determination of the number of the studied sample (certified reference soil GBW07401). particles present in an aliquot. The method previously proposed Nevertheless, it was decided to use two sample cups in all for the determination of density18 was found not to be well measurements combined with three replicates.One sample cup suited for the materials covered in this work because the could be adequate although the variance of the results might reproducibility was poor. Consequently, a standardized pycbe too high. These results agree with previous considerations nometer was used. The densities of the samples and soil and for slurry-sampling analyses of CRMs, although when real sediment standards were found to be around 2.6 g cm-3, which samples are slurry analysed this should be confirmed owing to agrees with densities found for other certified soils and their more questionable homogeneity.16,17 Attention should be sediments.8,18 drawn to the slightly negative trend observed in Fig. 4 when The particle size was determined by means of an optical four sample cups were considered. This could be attributed to microscope, placing 20 ml of a slurry (10 mg ml-1) on a slide the tube ageing, which decreases the analytical signal at the and drying it in an oven at 60 °C.A small area of the drop end of its lifetime. was examined to determine the number of particles and their Now, the eVects caused by the ageing of the graphite tube maximum and minimum diameters. The total area occupied can be studied by means of an additional one-way ANOVA. It by the drop was also determined to estimate the total number was considered that this factor should not be combined with of particles.For the GBW07401 CRM soil, the diameters were the previous two variables to develop a three-way ANOVA between 93.3 and 0.15 mm and the estimated number of because it is not current practice to analyse one sample using particles was 2912. This procedure was repeated for all the diVerent tubes. Nevertheless, the tube seemed an important certified materials and samples and similar results were factor when batches of samples are to be analysed.As expected, obtained. In all cases, the number of particles was much larger the factor ‘tube’ was significant and it caused large changes in than 50, indicating that after slurry homogenization using the the absorbances (Fig. 4). This confirms the need to maintain ultrasonic probe, the aliquot can be said to be representative strict control over the graphite tube performance and to recali- of the weighed sample.brate and validate carefully the results obtained after each change of tube. Moreover, it is important to take into account Number of replicates and sample cups. Despite the importance that faulty packs of tubes could be delivered and hence pur- of this topic, there is little agreement when dealing with slurry chased. Unfortunately, the influence of such tubes on the suspensions. Miller-Ihli17 recommended five replicates in each measurements is not always equal and their eVects depend upon of five sample cups for each sample. Here, it was thought that which element is being measured.One of the more straightfor- some kind of ‘optimization’ could be made to ascertain the ward and eYcient ways of checking tube performance is to adequacy of such a rule. Hence a two-way ANOVA was develop and maintain control charts (e.g., by using either an designed where ‘number sample cups’ ( levels=1, 2, 3, 4) and aqueous standard or a slurry sample).In this work, the control ‘replicates per sample cup’ ( levels=2, 3, 4, 5) were the factors. chart for an aqueous Sb standard (20 ng ml-1) revealed that The eVect of changing the graphite tube used for the analyses up to 350 atomization cycles can be made without a significant was also studied as a factor because one of the typical problems loss of sensitivity. Several control charts for soil slurries (10 ml associated with the slurry-based determination of metals by ETAAS is the accelerated degradation of the L’vov platform.of 50 mg ml-1 solution) allowed 90–100 atomization cycles to Table 4 Two-way ANOVA considering number of sample cups and number of replicates Source of Sum of Degrees of Mean Statistical variation squares freedom square F-ratio diVerence (95%) Sample cups 0.0006796 3 0.0002265 1.0 NO Replicates 0.0023607 3 0.0007869 3.6 YES Interaction 0.0038752 9 0.0004306 1.9 NO Residual 0.0250987 116 0.0002164 — — Total 0.0317421 131 — — — J.Anal. At. Spectrom., 1999, 14, 703–710 707Fig. 4 Average values and confidence intervals associated with each factor of the two diVerent ANOVA studies: (a) number of replicates; (b) number of sample cups; and (c) tubes. be established as the upper limit for the use of L’vov platform the reduced palladium provided the best results, as has been observed elsewhere.31 tubes when analysing this kind of slurry. In conclusion, 0.5% v/v HNO3 was found to be the most appropriate chemical modifier for the determination of Sb in Measurement procedure slurry soils by ETAAS, providing a good peak shape and a Chemical modifiers and temperature programme.The high satisfactory separation between the atomic and background volatility of Sb and the presence of several concomitants in peaks. The use of HNO3 as chemical modifier is also advantathe sample matrices can cause diYculties in its determination geous because its optimum concentration can be obtained by ETAAS.28 The use of the L’vov platform and chemical directly by preparing the slurries in this suspension medium.modification has been reported for the determination of this element in several matrices. Especially for geological samples, palladium, magnesium, lanthanum, nickel, nitric acid, copper Quantification and molybdenum have been reported.28,29 Also some suitable Pd mixtures such as Pd–Zr, Pd–Zr–citrate and Pd–W–citrate Calibration. There are two general calibration modes when working in ETAAS, namely direct calibration and the method have been used in the determination of volatile elements.30 Recently, a 0.1% m/v nickel nitrate hexahydrate solution has of the standard additions. The former assumes that the analytical signal is only caused by the analyte itself and that all been recommended as a chemical modifier for Sb determination in soil and sediment slurries.14 samples and standards are treated considering the same conditions.This method has the important advantage that one Systematic experiments were carried out to optimize the temperatures and times for the drying, pyrolysis and atomi- calibration line is useful for analysing many samples.The standard additions method is time consuming since a cali- zation steps considering nitric acid, palladium nitrate, a mixture of palladium and magnesium nitrate, palladium nitrate– bration line has to be obtained for each sample. This method accounts for most of the possible matrix interferences and it ascorbic acid, magnesium nitrate, ammonium dihydrogenphosphate and nickel as chemical modifiers.These conditions were is considered as the ‘reference’ method. More details can be found elsewhere.32 optimized for a slurry soil sample (90 mg of certified reference soil GBW07401 in 1 ml of 0.5% v/v HNO3) and aqueous Sb A third calibration method was implemented in the spectrometer software, called the method of addition calibration, standard solutions (20 ng ml-1).When a slurry sample is atomized, the high background and it takes characteristics from the two previous modes.33 It implies the development of one standard addition line for the causes the deuterium background correction system to overcorrect the atomic signal, sometimes giving negative integrated first sample and then the absorbance value of the sample is subtracted from the absorbance of each of the absorbance values. Therefore, only peak height absorbance was taken into account in this work.sample+standard solution. Subsequently, the absorbance of the second, third, etc., sample is added to each of the above In all cases, two drying steps (100 and 130 °C) were needed for eYcient, uniform drying of the soil slurries, being deter- calibration points and the sample is quantified. This method can be applied only if the matrix is more or less constant in mined through careful visual observation of the inner part of the tube.The use of 0.5% v/v HNO3 gave the most symmetric all the samples, which is the case here. To elucidate the most suitable calibration mode and to atomic peak and the best separation between the atomic and background signals, with pyrolysis and atomization tempera- assess if the calculated concentrations showed a dependence on sample mass, three diVerent masses of one reference soil tures of 700 and 1700 °C, respectively. NH4H2PO4, Mg(NO3)2 and Ni provided a slightly lower pyrolysis temperature of (GBW07401) were taken (number of cups and replicates as stated above) and analysed.Direct calibration was made with 800 °C (atomization temperature 1750 °C). These modifiers decreased the atomic peak height and altered the background aqueous standards in 0.5% v/v HNO3. Table 5 presents the quantification results and the recoveries. The slope of the signal appearance. Regarding the Pd-based modifiers, palladium nitrate and direct calibration regression did not diVer statistically (Student’s t-test, 95% confidence) from that of the standard reduced palladium (palladium nitrate–ascorbic acid) increased the pyrolysis temperature to 1000 °C without Sb losses (atomi- additions method, the main disadvantages of the direct calibration being the large confidence interval associated with zation temperature 1750 °C).In contrast, the atomic absorbance decreased considerably and the peaks obtained each prediction and the poor recoveries.The method of addition calibration and the method of standard additions were worse than those with HNO3. Pd(NO3 )2–Mg(NO3)2 (1.5+1) led to the highest pyrolysis temperature, 1200 °C gave the most satisfactory recoveries (83–99%) and quantifi- cations (accuracy and precision). The calibration confidence (atomization temperature 1800 °C), but the amount of modifier employed became critical. Thus, on increasing its amount, the intervals associated with each prediction were calculated considering the Hotelling confidence intervals of the regression atomic signal diminished and the overlap between the atomic and background signals increased.From all these Pd modifiers, lines.31 Accordingly, the method of addition calibration was 708 J. Anal. At. Spectrom., 1999, 14, 703–710Table 5 Comparison of three calibration methods for the quantification of Sb in one certified reference soil (GBW07401) using three diVerent levels of sample mass.Certified value: 0.75–0.87–0.99 mg g-1 Concentration/mg g-1 (average value±Hotelling confidence intervals) Direct calibration Standard additions method Method of addition calibration Sample mass/ mg Mean value Recovery (%) Mean value Recovery (%) Mean value Recovery (%) 40 0.09–0.66–1.23 76 0.52–0.86–1.19 99 0.39–0.72–1.05 83 92 0.46–0.69–0.94 76 0.60–0.86–1.11 99 0.77–0.84–1.19 97 150 0.94–1.04–1.15 119 0.94–1.09–1.26 125 0.77–0.82–1.04 94 selected because of its advantages (narrow confidence intervals, both the LOD and limits of quantitation (LOQ) (10 times the standard deviation of the blank).Note that for 200 mg ml-1, good recoveries, lower costs, high throughput, etc.). no limits were calculated owing to operational problems (the upper limit was set at 150 mg ml-1). The instrumental LOQ Pipetting options. Simultaneously with the calibration tests, we assessed how to introduce the slurries and metal standards was 4.13 ng ml-1. The characteristic mass (mo) was calculated for several into the furnace.Previously, the pipetting sequence was optimized in order to achieve good repeatability. The AS-70 sampler aqueous standards and slurry samples containing diVerent Sb concentrations. The mean values obtained were 17.3±1.7 and allows two pipetting options: (i) ‘pipette separate’: each of the slurry, standard aliquots 12.0±1.6 pg, respectively. and diluent are injected separately, the slurry being the first to be introduced, and a washing step is performed before Accuracy. From an operative point of view, accuracy can aspirating the standard aliquots; be studied in two ways: first, check the accuracy of the (ii) ‘pipette together’: slurry, standards and diluent are instrumental measurements; and second, test the overall pipetted and injected in one step.method accuracy. They allow one to investigate eventual bias A simple way to compare their performance is to develop in diVerent stages of the methodology.The use of the USS-100 either direct or standard addition regressions employing each probe did not cause any bias in the results (as shown below). pipetting mode and compare their slopes statistically. The lines The ‘instrumental accuracy’ has to be assessed using recovery for the pipette separate mode were as follows: calibration, assays. Accordingly, diVerent amounts of an aqueous Sb absorbance=0.0170+0.0045[Sb]; addition, absorbance= standard were added to several slurries obtained from the 0.0827+0.0053[Sb].The regression lines for the ‘pipette certified soil GBW07401, homogenized and then measured. together’ mode were as follows: calibration, absorbance= All recoveries were satisfactory, ranging from 93 to 110%. The 0.0155+0.0042[Sb]; addition, absorbance=0.1042+ method accuracy was evaluated using the reference materials 0.0027[Sb]. mentioned above. The ‘pipette separate’ mode gave much better results (the Despite all the method development carried out using certitwo lines are parallel at the 95% confidence level, texp=2.07 fied soils, several studies were replicated for sediments as a and ttab=2.23, n-1=10) than the ‘pipette together’ option.tentative extension of the developed methodology. Therefore, To our knowledge, this has not been reported previously and, if the two CRM sediments also performed well, this is strong therefore, the ‘pipette separate’ mode is recommended. It is evidence to consider that the methodology can be applied for thought that the diVerences between these two pipetting modes both soils and marine sediments.Table 6 collects the main increased when working with slurry samples because after results for each CRM. As for the instrumental accuracy, the aspirating the slurry the sampler tip remains dirty. This seems recoveries ranged from 82 to 109% and all the results were justified because whenever aqueous calibrations were carried very satisfactory.Not only were the experimental values out they showed the same slope using both pipetting modes. included in the confidence intervals of the certified or indicative In contrast, when the ‘pipette together’ mode was employed values, but they also exhibited excellent agreement with the to construct the standard additions line for slurry solutions, reference values. This confirms that the developed method can the slope diVered statistically from that of the aqueous be used successfully for the direct analysis of soils and sedicalibration. ments for Sb determination.Moreover, it was observed that the atomic peaks presented better characteristics when the ‘pipette separate’ mode was Precision. First, the precision of the instrumental measure- used and the quantification of one reference material ments was studied using two aqueous Sb standards (20 and (GBW07401) was much improved using such a pipetting mode 60 ng ml-1), either applying ultrasonic mixing prior to the (average value and confidence intervals for the ‘pipette separinjection into the furnace and without such homogenization.ate’ mode 0.60–0.86–1.11 mg g-1 and for the ‘pipette together’ Five replicates were performed for each situation and their mode 1.63–2.07–2.50 mg g-1; certified value, 0.75–0.87– relative standard deviations (RSD) were calculated. The values 0.99 mg g-1). were satisfactory and no statistical diVerences (Student’s ttest, 95% confidence) were found considering this variable: Analytical figures of merit with agitation, 20.14±0.59 ng ml-1 (RSD=2.9%); 61.60± 1.60 ng ml-1 (RSD=2.6%); without agitation, 20.55± Sensitivity. The instrumental limit of detection (LOD) is the signal which is statistically diVerent (99.9% confidence) from 0.47 ng ml-1 (RSD=2.3%); 60.63±1.80 ng ml-1 (RSD= 3.0%).Consequently, the ultrasonic homogenization achieved the blank.24 It was calculated from 10 independent blank measurements and using the average slope from six indepen- using the USS-100 probe did not increase the analytical response uncertainty.dent calibrations (which was 0.0048 ml absorbance ng-1). The instrumental LOD for the Sb determination was 1.24 ng ml-1. Second, two CRMs (BCR CRM 141 and BCR CRM 277) with diVerent contents of Sb were selected to evaluate the The method LOD, i.e., the LOD calculated following all the analytical methodology, has to be calculated for each mass to precision.One slurry of each of the two CRMs was prepared and then analysed six times. The RSDs obtained were 3.6 and volume ratio as is usual in slurry techniques.11 Table 2 gives J. Anal. At. Spectrom., 1999, 14, 703–710 709Table 6 Sb contents of five certified reference materials and six ‘real’ samples (average values±standard deviation; n=6) Concentration/mg g-1 RSD (%) Material Obtained value Certified value Recovery (%) (n=6) Soil GBW07401 0.86±0.06 0.87±0.12 99 7.0 Soil GBW07409 0.23±0.01 0.21±0.03 109 6.1 Calcareous loam soil BCR CRM 141 0.69±0.05 (0.7±0.3)a 99 7.2 Estuarine sediment BCR CRM 277 3.21±0.15 (3.9±1)a 82 4.7 Marine sediment BCSS-1 0.58±0.03 0.59±0.06 98 5.5 Soil 1 0.29±0.02 — — 7.0 Soil 2 0.46±0.02 — — 4.3 Soil 3 0.68±0.04 — — 5.9 Sediment 1 0.54±0.05 — — 9.2 Sediment 2 1.06±0.05 — — 4.7 Sediment 3 1.35±0.07 — — 5.2 aNot certified. 7 M. W. Hinds, K. W. Jackson and A. P. Newman, Analyst, 1985, 2.8%, respectively. Although the use of slurries increased the 110, 947.within-sample RSD (as expected), the precision values were 8 M. Hoening and P. Regnier, J. Anal. At. Spectrom., 1989, 4, 631. very satisfactory (<4%). It is thought that these good values 9 C. Bendicho and M. T. C. de Loos-Vollebregt, J. Anal. At. are derived from the excellent homogenization provided by Spectrom. 1991, 6, 353. the USS-100 probe and also the optimized preparation of the 10 P. Bermejo-Barrera, C. Barciela-Alonso, M. Aboal-Somoza and A.Bermejo-Barrera, J. Anal. At. Spectrom., 1994, 9, 469. slurry samples. 11 P. Bermejo-Barrera, J. Moreda-Pin� eiro, A. Moreda-Pin� eiro and Next, six diVerent portions of each CRM were weighed, A. Bermejo-Barrera, Anal. Chim. Acta, 1994, 296, 181. slurry prepared and analysed to calculate the overall method- 12 I. Karadjova, P. Mandujukov, S. Tsakovsky, V. Simeonov, J. A. ology precision. The precision values were good (7.2–4.7%). Stratis and G. A. Zachariadis, J. Anal.At. Spectrom., 1995, 10, Similar figures were achieved when six ‘real’ samples (three 1065. soils and three sediments) were analysed applying the above 13 P. Bermejo-Barrera, M. C. Barciela-Alonso, J. Moreda-Pin� eiro, C. Gonza�lez-Sixto and A. Bermejo-Barrera, Spectrochim. Acta, developed methodology (Table 6). This strongly suggests that Part B, 1996, 51, 1235. the numbers of cups and replicates optimized using CRMs are 14 I. Lo�pez-Garcý�a, M. Sa�nchez-Merlos and H. Herna�ndez-Co�rdoba, also valid when dealing with real samples.16 Therefore, these Spectrochim.Acta, Part B, 1997, 52, 437. results can be used to confirm that Sb is also homogeneously 15 N. J. Miller-Ilhi, J. Anal. At. Spectrom., 1989, 4, 295. distributed in the soil and sediment samples studied in this 16 M. S. Epstein, G. R. Carnrick and W. Slavin, Anal. Chem., 1989, work. The Sb contents for soils were comparable to natural 61, 1414. 17 N. J. Miller-Ihli, At. Spectrosc., 1992, 13, 1.reported values (0.2–1 mg g-1).34 Regarding sediments, they 18 N. J. Miller-Ihli, Fresenius’ J. Anal. Chem., 1993, 345, 482. exhibited slightly higher concentrations than soils. Never- 19 N. J. Miller-Ihli, J. Anal. At. Spectrom., 1994, 9, 1129. theless, these levels were similar to others already reported for 20 W. Klemm and G. Bombach, Fresenius’ J. Anal. Chem., 1995, estuarine sediments which are considered as normal values 353, 12. (0.04–1 mg g-1).35 For both types of samples, the Sb contents 21 R. Dobrowolski, Spectrochim. Acta, Part B, 1996, 51, 221. corresponded to unpolluted sites. 22 I. Koch, C. F. Harrington, K. J. Reimer and W. R. Cullen, Talanta, 1997, 44, 771. 23 R. L. Plackett and J. P. Burman, Biometrika, 1946, 33, 305. 24 J. C. Miller and J. N. Miller, Statistics for Analytical Chemistry, Acknowledgements Ellis Horwood, Chichester, 2nd edn., 1992. 25 N. J. Miller-Ilhi, Spectrochim. Acta, Part B, 1997, 52, 431. This research was supported by the Vicerrectorado de 26 Standard Guide for Conducting Ruggedness Tests, ASTM E 1169, Investigacio�n of the University of La Corun� a. The authors American Society for Testing and Materials, Philadelphia, PA, thank B. Ferna�ndez Souto (Servicios Xerais de Apoio a� 1997. Investigacio� n) for his collaboration in obtaining the optical 27 A. Carlosena, M. Gallego and M. Valca�rcel, J. Anal. At. microscopy photographs. Professor X. Toma�s is also acknowl- Spectrom., 1997, 12, 479. edged for his kind advice on the chemometric concepts. 28 H. Niskavaara, J. Virtasalo and L. H. J. Lajunen, Spectrochim. Acta, Part B, 1985, 40, 1219. 29 Y. Morishigue, K. Hirokawa and K. Yasuda, Fresenius’ J. Anal. Chem., 1994, 350, 410. References 30 I. Havezov, A. Detcheva and J. Rendl, Mikrochim. Acta, 1995, 1 S. Allen, Chemical Analysis of Ecological Materials, Blackwell, 119, 147. 31 E. Bulska and W. Jedral, J. Anal. At. Spectrom., 1995, 10, 49. Oxford, 2nd edn., 1989. 32 M. J. Gardner and A. M. Gunn, Fresenius’ J. Anal. Chem., 1988, 2 G. B. van der Voet and F. A. de WolV, in Toxicology of Metals, 330, 103. ed. L. W. Chang, CRC Press, Boca Raton, FL, 1996. 33 Atomic Absorption Laboratory Benchtop. User’s Guide, Perkin- 3 B. J. Alloway, Heavy Metals in Soils, Wiley, Chichester, 1st edn., Elmer, U� berlingen, 1991. 1990. 34 A. Kabata-Pendias and H. Pendias, Trace Elements in Soils and 4 C. Dietl, W. Reifenha�user and L. Peichl, Sci. Total Environ., 1997, Plants, CRC Press, Boca Raton, FL, 1992. 205, 235. 35 U. Fo� rstner and G. T. W. Wittman, Metal Pollution in the Aquatic 5 K. Takayanagi, D. Cossa and J.-M. Martin, Mar. Chem., 1996, Environment, Springer, Berlin, 1983. 54, 303. 6 N. N. Greenwood and A. Earnshaw, Chemistry of the Elements, Pergamon Press, Oxford, 1997. Paper 8/07359G 710 J. Anal.

ISSN:0267-9477    

DOI:10.1039/a807359g

出版商:RSC    

年代:1999

数据来源: RSC

摘要:

EVectiveness of fullerene as a sorbent for the determination of trace amounts of cobalt in wheat flour by electrothermal atomic absorption spectrometry Ma Mar Gonza�lez, Mercedes Gallego and Miguel Valca�rcel* Department of Analytical Chemistry, Faculty of Sciences, University of Co�rdoba, E-14004 Co�rdoba, Spain Received 19th October 1998, Accepted 12th February 1999 A flow injection system for the on-line preconcentration of ultratrace levels of cobalt by sorption with ammonium pyrrolidinedithiocarbamate on a C60 fullerene column was developed. The eluate volume is reduced and dispersion minimized by using a nitrogen (or air) stream to carry the eluent (IBMK) for elution of the adsorbed chelate, which is collected in a 500 ml PTFE autosampler cup.The analytical figures of merit for the determination of cobalt are as follows: limit of detection (3s), 8 ng l-1; precision (RSD), 4% for 0.2 ng ml-1; preconcentration factor, 40 (using 25 ml of sample and 500 ml of IBMK).Similar experiments involving C18-bonded silica as sorbent revealed ethanol to be the most suitable eluent under these conditions. The results obtained in the determination of cobalt in wheat flour testify to the usefulness of the proposed method. The high hydrodynamic impedance of the column to using Introduction high sample loading flow rates, and its short lifetime, can both The significance of cobalt to human and ruminant nutrition be overcome by adsorbing metal complexes on the inner walls has aroused great interest in its determination in a variety of of a knotted PTFE reactor.These novel preconcentration sample types. Cobalt, as vitamin B12, is essential to living systems have been extended to ETAAS for the automatic species; however, it is toxic in large amounts. Plants are determination of metals in water samples.14–16 By using this generally low in cobalt; however, they are the main source of preconcentration system, the adsorbed Co–PDC complex is this metal for the human diet.1 quantitatively eluted with 45 ml of ethanol and transferred The very low concentration of cobalt in foods and the directly to an ETAAS instrument.16 An enhancement factor presence of matrix interferences call for the use of a separation of 47 (6.7 ml of sample) was thus obtained at a sampling technique for its determination.2 Recently developed automatic frequency of 30 h-1.techniques provide a variety of eVective approaches to the Ever since its existence was confirmed,17 fullerene has problem, particularly those including a continuous preconcen- received substantial attention in various scientific fields.18 The tration unit.3–5 The advantages and pitfalls of liquid–liquid main problem with fullerenes continues to be the separation extraction in relation to other alternatives such as column of isomers and homologues; Jinno et al.have published about extraction procedures based on chelating resins and sorbents 25 papers on the separation and isolation of fullerenes from have been discussed.6 The use of on-line flow injection (FI ) carbon soot on common and specialized stationary phases by column preconcentration for the atomic spectrometric determi- HPLC.19,20 Numerous reviews and a recent monograph on nation of cobalt was reviewed by Fang et al.7 Based on this topic have also been published.21 The purity of fullerene solid-phase extraction with C18 silica gel, Sperling and precursors and products has also been examined; thus, data co-workers8–11 developed FI systems for determining this metal on trace element impurities in these products have been by electrothermal atomic absorption spectrometry (ETAAS).reported.22 The analytical potential of C60 fullerene as a For example, ultratrace amounts of cobalt in sea-water were sorbent material for the preconcentration of metals was first determined by using NaDDC as chelating reagent and demonstrated by Gallego et al.;23 subsequent experiments with reversed-phase C18-bonded silica as sorbent;10 an enhancement C60 and C70 fullerenes in continuous systems (and IBMK as factor of 42–210 was obtained with a 2–10 min sample loading eluent) were successfully applied to FAAS for the determitime (5–30 ml of sample).The principle of eluate zone sam- nation of copper24 and cadmium25 in biological samples. Both pling, whereby a 40 ml portion of the eluate containing the fullerenes were found to exhibit better sorbent properties in most concentrated fraction of eluted analyte (in ethanol ) is metal preconcentration than did conventional solid materials introduced directly into the graphite furnace through the (e.g.RP-C18, activated carbon and resins). Better sensitivity capillary in the autosampler arm, enables processing of high and selectivity were obtained with neutral chelates than by sample volumes with minimal contamination and analyte loss.ion-pair formation.24,25 Ma and Adams12 have investigated the applicability of alkyldi- Cobalt concentrations in cereals can be as low as 10 ng g-1 thiophosphates as chelating agents for cobalt and other metals; dry matter; the determination of such small amounts obviously C18-bonded silica gel as sorbent, methanol or ethanol as eluent requires a very sensitive analytical method. The purpose of and FAAS/ ETAAS detection were used. More recently, a this work was to extend the high analytical usefulness of preconcentration column packed with Muromac A-1 chelating fullerenes for metal preconcentration to ETAAS in order to resin was inserted at the tip of an autosampler arm to determine ultratrace amounts of cobalt in wheat flour.Following decomposition of organic matter by a simple, rapid preconcentrate several metals (cobalt excluded) in sea-water.13 J. Anal. At. Spectrom., 1999, 14, 711–716 711wet ashing procedure, cobalt, in 0.1 mol l-1 HNO3, is intro- columns packed with 80 mg of C60 fullerene or silica RP-C18.The mini-columns were made from PTFE capillaries of 3 mm duced into an FI system. The adsorbed Co–PDC complex is quantitatively eluted with IBMK and the whole eluate is id and sealed at both ends with small cotton-wool plugs to prevent material losses. They were initially flushed with directly transferred to a stoppered autosampler cup. 0.1 mol l-1 HNO3 (subsequent use of IBMK as eluent in each operating cycle was suYcient to make them ready for re-use). Experimental C60 fullerene columns were usable for at least 9 months.Reagents and standard solutions Sample preparation A 1000 mg l-1 cobalt stock solution was prepared by dissolving 1.000 g of the metal in a small volume of concentrated nitric Flour samples were prepared at a pilot plant; the raw material was wheat produce from diVerent Spanish locations. An acid and diluting to 1 l with 1% v/v nitric acid. A 0.1% m/v aqueous solution of APDC (Aldrich, Madrid, Spain) was amount of 0.5–1 kg of wheat containing ca. 15% m/m water was ground in a metal rolling mill to obtain 60% of white prepared; the solution remained stable for at least 3 d. A 1% m/v solution of neocuproine (Merck, Darmstadt, flour and 40% of by-products. The flour was screened through a 132 mm sieve. Flour samples were dried to constant mass in Germany) in ethanol was also prepared. Hydroxylammonium chloride and IBMK (Merck) were also used.C60 fullerene an oven at 103 °C. The wheat flours were prepared as follows: an accurately (>99.4%, Hoechst, Frankfurt-am-Main, Germany) and polygosyl- bonded silica reversed-phase sorbent with octadecyl weighed amount of 0.25 g was digested with 2 ml of 65% HNO3 and a few drops of 30% H2O2 in a glass beaker. The functional groups (RP-C18), 60–100 mm particle size (Millipore, Madrid, Spain), were employed as sorbent mate- mixture was heated on a hot-plate at about 200 °C to neardryness (about 8 min).Once cool, the residue was diluted with rials. Standard solutions (25 ml ) containing 0.02–1 ng ml-1 cobalt were all freshly prepared by appropriate dilution of a 0.1 mol l-1 HNO3 and transferred quantitatively into a 25 ml calibrated flask to which 1 g of hydroxylammonium chloride stock standard solution (1000 mg l-1) in 0.1 mol l-1 HNO3. An organic stock solution ofh;PDC was made by placing and 0.5 ml of 1% neocuproine (to avoid the interference of iron and copper, respectively) were added before making up 1 ml of the 1000 mg l-1 cobalt solution, 10 ml of 1 mol l-1 acetic acid–acetate buVer (pH 4.7) and 10 ml of the 0.1% to volume with the same nitric acid solution. A reagent blank was prepared in parallel.The diluted sample (25 ml ) was APDC solution in a separating funnel and adding 50 ml of IBMK. The mixture was then shaken gently for 5 min; after analysed immediately after preparation by inserting it into the manifold of Fig. 1. the two phases had been decanted, the IBMK phase was transferred to a PTFE stoppered bottle containing anhydrous sodium sulfate. Working-strength solutions were prepared Preconcentration procedure daily by appropriate dilution of the above 20 mg ml-1 stock The continuous preconcentration and elution system is shown solution with IBMK. in Fig. 1. In the preconcentration step, a volume of 25 ml of Solutions of potentially interfering ions were prepared by standard or treated sample, containing 0.02–1 ng ml-1 Co in dissolving the amount of each metal/salt needed to obtain a 0.1 mol l-1 HNO3, was continuously introduced into the 100 mg ml-1 concentration of each ion.system at 3.0 ml min-1 and mixed thoroughly with the chelating reagent solution (0.1% APDC) at 0.3 ml min-1. The cobalt Apparatus chelate was adsorbed on the C60 mini-column, located inside the loop of the injection valve (IV1), the sample matrix being A Perkin-Elmer Model 1100-B atomic absorption spectrometer (U� berlingen, Germany) equipped with deuterium arc back- sent to waste.Residual aqueous solution inside the column and FI connectors was flushed by passing a nitrogen stream ground correction, an HGA-700 graphite furnace atomizer and an AS-70 furnace autosampler were used throughout. at 2.0 ml min-1 through the carrier line of the second valve (IV2) for 2 min; simultaneously, the loop of IV2 was filled Measurements were made at 240.7 nm, using a single-element hollow cathode lamp for cobalt that was operated at 30 mA with eluent (IBMK).As IV2 was switched, 500 ml of IBMK were injected into the nitrogen stream and passed through the and a bandwidth of 0.2 nm. Argon at a flow rate of 300 ml min-1 was used as the inert gas except during the column to elute the chelate (position in bold lines in Fig. 1). The extract was collected in the stoppered PTFE cup (500 ml atomization step, where the flow was stopped; the injected volume was 20 ml.Pyrolytic graphite L’vov platforms inserted into pyrolytic graphite coated tubes were obtained from Perkin-Elmer. Background-corrected integrated absorbance was used as the analytical signal. The furnace programme is shown in Table 1. The flow manifold consisted of a Gilson Minipuls-2 peristaltic pump ( Villiers-le-Bel, France) furnished with poly(vinyl chloride) tubes, two Rheodyne 5041 injection valves (Cotati, CA, USA) and laboratory-made sorption mini- Table 1 Graphite furnace temperature–time programme for the determination of cobalt following sorbent extraction in the FI system Time/s Step Temperature/ °C Ramp Hold 1 90 5 15 2 300 10 20 Fig. 1 FI manifold for the on-line preconcentration of cobalt and its 3 1500 6 20 oV-line determination by ETAAS. Bold lines denote lines relevant to 4 2600 0 6 the elution step. IV, Injection valve; W, waste; IBMK, isobutyl methyl 5 2650 1 3 ketone (eluent); GT, graphite tube. 712 J. Anal. At. Spectrom., 1999, 14, 711–716capacity) of the instrument’s autosampler. A blank consisting solution extracted with IBMK prepared in parallel with the organic standards but in the absence of cobalt; the other was of 20 ml of IBMK (0.005 A s) was used. an IBMK solution. Identical results were obtained with both blanks (ca. 0.005 A s), so the IBMK blank (20 ml ) was used Results and discussion in subsequent experiments. There are two methods for the determination of trace amounts Sorption preconcentration system of cobalt in feed by ETAAS following complexation with 2-nitroso-1-naphthol and solvent extraction with xylene26 or The eVect of chemical and flow variables was studied by heptan-2-one.27 In this work, APDC was selected to form a introducing an aqueous standard solution of 0.2 ng ml-1 neutral chelate with cobalt because this is the most widely cobalt into the flow system for ca. 8 min (sample flow rate used chelating reagent for metal enrichment in AAS techniques 3 ml min-1, sample volume 25 ml ) and merging it with a 0.1% and in combination with C60 fullerene it exhibits improved APDC stream.The mini-column (1.0 cm×3 mm id) was consensitivity and selectivity.23–25 Isobutyl methyl ketone is the structed from a PTFE capillary and packed with 80 mg of C60 preferred eluent because the PDC chelate is the easiest to fullerene. For comparison, a similar mini-column packed with dissolve and hence to desorb in this medium.A previously 80 mg of RP-C18 (1.5 cm×3 mm id) was also tested. The reported on-line column preconcentration system based on eluent used was IBMK in all instances. A volume of 500 ml FAAS, with some changes (e.g., the eluent was carried by a (the capacity of the PTFE autosampler cup) was initially stream of nitrogen instead of water), was used.25 Using air as selected. The cup was fitted with a pierced anti-evaporation eluent carrier was initially discarded as cobalt(II) complexes stopper allowing insertion of the flow line in order to minimize are known to be easily oxidized by oxygen to trivalent contamination and losses.complexes that are too inert for elution.28 The eVect of pH on chelate adsorption was studied over the range 0–7 by adjusting the cobalt sample with dilute HNO3 Graphite furnace heating programme or NH3 as required. The optimum pH range (1–5) was wider for C60 fullerene than for RP-C18 (1–1.5), probably because The low surface tension of IBMK makes sample delivery the adsorption constant of the former is greater, consistent diYcult as the ketone tends to creep along the length of the with experimental results obtained for Pb–PDC (adsorption furnace tube, thus severely limiting sample volumes; in constants were 575 and 155 for C60 and RP-C18, respectaddition, the analytical response from organometallic com- ively),23 and Cd–PDC chelates.25 Although the maximum pounds is often diVerent from that of inorganic salts.29 chelate adsorption for RP-C18 was achieved at a low sample Additional problems often posed by IBMK include the fact pH, a further plateau was observed between pH 3 and 5, in that the analytical response varies with the injected solvent addition to a signal decrease by about 25% relative to the first volume and potential pre-atomization losses of volatile metal plateau.It is interesting that the acid zone (lower than 2) is chelates or their decomposition products.In this work, pyro- unusual in liquid–liquid extraction procedures based on this lytic graphite coated graphite tubes with platforms were used ligand; this suggests that the acid medium favours retention in order to restrict penetration of the solvent into the graphite of the chelate on both sorbents. In order to simplify the and spreading of the sample.30 A simultaneous study was operating procedure and increase the selectivity of the method, performed with the Co–PDC chelate in IBMK and aqueous 0.1 mol l-1 HNO3 was used to prepare the samples for both standards of cobalt in 0.2% HNO3.No problems were encoun- mini-columns. The eVect of the APDC concentration was tered in pipetting 20 ml volumes of IBMK into the graphite studied over the range 0.001–0.5%; both mini-columns profurnace. Aqueous and organic standards of 25 ng ml-1 cobalt vided similar results (the preconcentration eYciency gradually were employed in all instances.No significant influence on increased as the APDC concentration was raised to 0.01% m/v sensitivity or precision was observed at drying temperatures and then levelled oV ). For subsequent work, a 0.1% m/v of ca. 110 °C for aqueous solutions; on the other hand, two APDC solution was chosen. Replacing the APDC streawith drying steps at 90 and 300 °C were necessary to avoid splatter- a water stream (the sample was also circulated) resulted in an ing of the organic solutions. The integrated absorbance of area in the elution step similar to that obtained by replacing cobalt remained constant on varying the pyrolysis temperature the sample stream with 0.1 mol l-1 HNO3 (blank), ca.from 500 to 1500 °C in both types of medium. This indicates 0.005 A s, which corresponds to the eluent signal. IBMK was that the thermal stability of Co–PDC in IBMK is similar to thus used as the blank. that of cobalt in the aqueous solution. The high pyrolysis The influence of the sample flow rate on the signal was temperature used (1500 °C) favoured separation of the cobalt examined over the range 0.5–4.0 ml min-1 by using an overall from more volatile concomitants present in the matrix sample, sample volume of 25 ml; the signal remained constant through- thus reducing the background signal.At a pyrolysis temperaout the range studied, so 3.0 ml min-1 was selected. A reagent ture of 1500 °C, peak areas for cobalt increased with increasing flow rate of 0.3 ml min-1 was chosen because higher flow rates atomization temperature up to 2600 °C (with both aqueous resulted in concomitant/sample dilution and hence in decreased and organic standards).On the other hand, the use of mag- atomic signals. The optimum length of the preconcentration nesium nitrate as modifier (15 ml volume of standard and 5 ml coil ( located before the sorbent column) ranged between 100 of 10 g l-1 chemical modifier) provided no advantages with and 300 cm for both mini-columns; a length of 200 cm was aqueous or organic standards.At a pyrolysis temperature of thus used throughout. 1500 °C and an atomization temperature of 2600 °C, the integrated absorbance was 10% higher with Co–PDC in IBMK Elution process standards than with identical amounts of cobalt in 0.2% HNO3. Background absorption was found not to depend on As shown elsewhere,10,12,16 ethanol is the most widely used eluent by virtue of its eVective elution of adsorbed chelates the nature of the cobalt solution medium; thus, the signals were similar for aqueous and IBMK standards and ranged from classical sorbents; it has thus been frequently used with RP-C18 sorbents.As in this paper a parallel study was per- from 0.004 to 0.012 A s. The blank proved the most important single sample to be formed with C60 and RP-C18, we compared ethanol and IBMK as eluents with both sorbents. A volume of 25 ml of a standard carried through the analytical procedure; running two blanks simultaneously with the samples for the organic medium solution containing 0.2 ng ml-1 Co in 0.1 mol l-1 HNO3 was introduced into the FI system of Fig. 1 for this purpose. proved advantageous. One blank consisted of a PDC buVer J. Anal. At. Spectrom., 1999, 14, 711–716 713Following retention, the column was dried with nitrogen and chelate were completely eluted by ethanol, even if the volume needed exceeded that of IBMK as eluent, the analytical signal the retained chelate was eluted by using a nitrogen stream at 2.0 ml min-1 as eluent carrier.The eZuent from the sorbent of cobalt should be similar under the optimum conditions for ethanol with equal volumes of both eluents. Because the column was collected in a PTFE autosampler cup of 500 ml capacity. favourable eVect might lie in boosted sorption (i.e. sorption of the Co–PDC chelate might be increased by conditioning The eVect of the eluent volume was studied between 100 and 500 ml by using loops of variable length in the injection with IBMK), a series of quintuplicate experiments was conducted that provided the following results: (i) when the fullerene valve (IV2 in Fig. 1).In order to study only the desorption process and to correct the opposing eVect of dilution, the column was conditioned with 500 ml of IBMK before the eluate was always diluted to 500 ml with IBMK or ethanol. preconcentration step and further elution with 500 ml of etha- With C60, desorption eYciency increased with increasing nol, the signal was 90% of that obtained when eluting with injected IBMK volume up to 200 ml, above which the analytical 500 ml of IBMK; (ii) when, following conditioning with 500 ml signal remained constant.A second injection of the same of IBMK, the column was flushed with 1 ml of 0.1 mol l-1 eluent volume (200 ml ) revealed the absence of carry-over. The HNO3, elution with 500 ml of ethanol or IBMK provided elution experiments with ethanol revealed that at least 400 ml signals which bore the same relation as in (i), but were both were necessary for complete desorption of the Co–PDC che- 30% lower; and (iii) conditioning with 500 ml of ethanol and late; in addition, as can be seen in Fig. 2, the peak area eluting with IBMK provided results similar to those obtained decreased by ca. 40% relative to IBMK eluent. Fig. 2 also by eluting with ethanol. One can therefore conclude that shows the atomization signal for cobalt following preconcen- adsorption is favoured by conditioning the C60 fullerene tration on C60 and elution with 500 ml of 2mol l-1 HNO3. column with IBMK, probably because its keto group (an Nitric acid resulted in no signal diVerence between the sample electron donor) boosts the sorption capacity of fullerenes and blank.IBMK provided the better results (diVerence (electron acceptors), which seemingly act via p-electron interbetween sample and blank); also, the background signal was actions.21 Alternatively, the water immiscibility of IBMK negligible and similar for the three eluents tested.A second solutions may give rise to the formation of a film within the study was performed by using RP-C18 as sorbent under the column that might subsequently facilitate the adsorption of above-described conditions. The better eluent in this case was the Co–PDC chelate; however, this eVect is not observed with ethanol, which resulted in an atomic signal 15% higher than the RP-C18 sorbent, so the previous explanation seems the with IBMK.In addition, the signal for 0.2 ng ml-1 cobalt more plausible. On the other hand, if the potential bonding obtained with RP-C18 (and ethanol as eluent) was similar to mechanism for the adsorption of Co–PDC on RP-C18 is based that obtained with C60 (with IBMK as eluent). Therefore, on a hydrophobic eVect, then the increased eluting capacity calibration graphs with RP-C18 were constructed by using of ethanol can be ascribed to the chelate desorption being ethanol and IBMK as eluents (500 ml ).The sensitivity (slope governed by a reduced solvent polarity. of the calibration graphs) for ethanol and IBMK was 0.40 Finally, in order to optimize the performance of the and 0.35 A s per ng ml-1, respectively. proposed method, an injected volume of 500 ml of IBMK was The above experiments with C60 fullerene provided better used in order to ensure complete elution of the chelate and results with IBMK than with ethanol as eluent (see Fig. 2). conditioning of the column before the aqueous sample was However, this is diYcult to explain because, if the Co–PDC preconcentrated. Flow rates of the nitrogen stream (the carrier of the eluent volume) between 1 and 3 ml min-1 had no eVect on the cobalt signal. No blank was required; the sample signal was calculated by subtracting the eluent signal (20 ml of IBMK) obtained before the sample extract was injected into the graphite furnace tube.Air, introduced via the peristaltic pump, can be used to dry the sorbent column and FI connections, and also as the eluent carrier as it provides results similar to those of nitrogen, probably because the Co–PDC chelate is not oxidized. Analytical performance Several calibration graphs were run by using mini-columns packed with 80 mg of C60 or RP-C18 and IBMK as eluent. Aqueous standard solutions were processed along the preconcentration flow system depicted in Fig. 1. The sensitivities (expressed as the slopes of the calibration graphs) were 0.40±0.02 and 0.35±0.03 A s per ng ml-1 and the linear ranges for cobalt were 0.02–1 and 0.05–1 ng ml-1 for C60 and RP-C18, respectively, equivalent to a sample volume of 25 ml at a sampling time of ca. 8 min. The detection limit was calculated as three times the standard deviation of the peak area for 15 injections of 20 ml of IBMK (blank) and was found to be 8 and 20 ng l-1 for C60 and RP-C18, respectively.The precision (as relative standard deviation) was checked on 11 standard solutions containing 0.2 ng ml-1 cobalt with both Fig. 2 Atomization signals for 0.2 ng ml-1 Co after on-line mini-columns. It was higher for C60 (4%) than for RP-C18 preconcentration of Co–PDC chelate on C60 fullerene sorbent, using (7%). The preconcentration factor (viz. the ratio between the 500 ml of diVerent eluents: (a) IBMK, peak area (AA-BG)= slope of the calibration graphs provided by the proposed 0.078 A s, peak area (BG)=0.011 A s; (b) ethanol, peak area method and by direct insertion of Co–PDC standards in (AA-BG)=0.048 A s, peak area (BG)=0.010 A s; and (c) 2 mol l-1 HNO3, peak area (AA-BG)=0.002 A s, peak area (BG)=0.009 A s.IBMK) was 40 and 35 for C60 and RP-C18, respectively. 714 J. Anal. At. Spectrom., 1999, 14, 711–716Table 3 Cobalt content (ng g-1±s) in various wheat flour samples as Table 2 Tolerated concentrations of foreign cations in the determination of 0.2 ng ml-1 cobalt with APDC and C60 sorbent determined by the proposed FI method (n=9) and directly in mineralized samples (n=6) Tolerated Ion [metal ]5[Co] ratio Sample FI-ETAAS Conventional ETAAS 1 11.0±0.9 9.5±0.7 Al3+ 10 000 Mn2+ 10 000 2 19.9±0.7 20.1±0.5 3 18.3±0.8 19.1±0.6 Sn2+ 10 000 Hg2+ 8000 4 16.6±0.8 15.5±0.5 5 16.3±0.7 15.5±0.5 Zn2+ 7000 Bi3+ 2000 6 18.2±0.7 19.0±0.6 7 13.3±0.6 12.8±0.4 Ni2+ 700 Cu2+ 400–1000a 8 13.8±0.6 13.5±0.4 Cd2+ 300 Fe3+ 100–8000b Pb2+ 100 a,bTolerated ratio with 0.02% m/v neocuproine and 4% m/v replicate are given in Table 3.For comparison, identical hydroxylammonium chloride, respectively. samples (No. 1, 2 and 5 in Table 3) were pre-treated by ashing and analysed by direct ETAAS using aqueous standards of cobalt. For this purpose, about 5 g of flour were charred in a Interference testing burner for ca. 2 h and then ashed at 650 °C for 3 h. The results were consistent with those provided by the proposed FI The influence of metals that might react with APDC and method; however, the procedure was more time-consuming replace cobalt in the original chelate was investigated in order and used a greater amount of sample.The proposed method to identify potential interferences. Table 2 lists the cations is, therefore, suitable for the determination of cobalt in this examined and their tolerated ratios in the determination of type of sample; the sample preparation time is minimal (ca. 0.2 ng ml-1 cobalt with C60 at pH 1; the maximum concen- 10 min) as the preconcentration method aVords the processing tration tested was 10 000 times that of the analyte. Interferents of small amounts of sample (ca. 0.25 g) that can be readily decreased the cobalt signal in all instances by competing for mineralized within a short time. and consuming the reagent; as a result, uncomplexed cobalt was not retained on the column. As can be seen in Table 2, no interference was caused by Al3+, Mn2 + or Sn2+ at the Conclusions highest concentration tested (2 mg ml-1).Only Pb2+ and Fe3+ The ETAAS determination of trace amounts of cobalt in interfered at concentration ratios up to 100. Unless the samples organic solvents warrants several interesting comments. Thus, are highly contaminated, only Cu2+ and Fe3+ are encountered the complexing agent (APDC) and the solvent (IBMK) have at concentrations of ca. 6 and 70 mg g-1, respectively, in flour negligible eVects relative to aqueous standard solutions (peak samples; such contents interfere with the proposed method.areas are 10% higher in the organic medium and background Masking agents may thus be required to avoid the interference signals are similar and negligible). The analytical potential of of high concentrations of Fe3+ and Cu2+ in analysing wheat fullerene as a sorbent in ETAAS was studied here for the first flour. We examined the eVectiveness of several commonly used time.Fullerenes perform better in metal preconcentration than masking agents. Sodium citrate (8% m/v), tartaric acid (4% do conventional C18-bonded silica sorbents when using FAAS m/v) and Tiron (10% m/v), which forms a highly stable and IBMK as eluent;23–25 however, water-miscible solvents complex with Fe3+ in an acidic medium, were tested as (e.g., ethanol ) should be avoided in order to minimize dilution masking agents for iron, with unfavourable results at high while water is being driven to the nebulizer.In this work, iron concentrations. Since Fe2+ has a lower aYnity for APDC eluents were carried by a nitrogen (or air) stream, and IBMK than Fe3+, reductants such as hydroxylammonium chloride and ethanol were tested in parallel with C60 and RP-C18 and ascorbic acid were tested. Hydroxylammonium chloride sorbents. The results aVord some interesting conclusions. First, and ascorbic acid at levels up to 4% m/v increased the amount the sensitivity is higher with C60 than with RP-C18 when using of tolerated iron to 1.6 mg ml-1 (Fe5Co ratio 800051).IBMK as eluent; however, RP-C18 is similar to C60 when using Bathocuproine and neocuproine at variable concentrations ethanol instead of IBMK. Second, adsorption is increased by were tested to overcome the interference of copper. Only conditioning C60 fullerene with IBMK before the preconcen- neocuproine was found to complex Cu2+ eVectively (in the tration step, probably as a result of the keto group in IBMK presence of a reductant such as hydroxylammonium chloride) favouring the adsorption mechanism via p-electron inter- with no disturbing eVect on cobalt. As can be seen in Table 2, actions.Future experiments will be conducted to identify, in 0.02% m/v neocuproine suppressed the interference of Cu2+ more rigorous terms, the best conditions (e.g., chemical vari- at levels at least up to 1000 times higher than that of cobalt, ables, selectivity) for RP-C18 and Co–PDC, using ethanol as which suYced since copper normally occurs at concentrations eluent.If similar figures of merit are obtained under the only ca. 600 times higher than that of cobalt in flour. optimum conditions for each sorbent (C60 and RP-C18), then the sorbent of choice will be the more aVordable and readily Analysis of wheat flour available. Because no certified values for the cobalt content in wheat flour reference materials were available, the applicability of Acknowledgements the proposed method was checked by analysing wheat flour samples from diVerent Spanish locations.These materials were This work was supported by grant PB95–0977 from Spain’s dissolved as described under Experimental and the cobalt DIGICyT. content in the resulting solutions was determined by using the recommended procedure. The signals for the blanks, which References contained the reagents in the digested sample, corresponded to a cobalt concentration below the detection limit.The results 1 World Health Organization, Trace Elements in Human Nutrition and Health, WHO, Geneva, 1996. obtained for three sample replicates and three injections per J. Anal. At. Spectrom., 1999, 14, 711–716 7152 A. Mizuike, Enrichment Techniques for Inorganic Trace Analysis, 17 H. W. Kroto, J. R. Heath, S. C. O’Brien, R. F. Curl and Springer-Verlag, Berlin, 1983. R. E. Smalley, Nature, (London), 1985, 318, 162. 3 Flow Injection Atomic Spectroscopy, ed. J. L. Burguera, Marcel 18 J. F. Anacleto and M. A. Quilliam, Anal. Chem., 1993, 65, 2236. Dekker, New York, 1989. 19 K. Jinno and H. Nakamura, Chromatographia, 1994, 39, 285. 4 Z. Fang, Flow Injection Atomic Absorption Spectrometry, Wiley, 20 K. Jinno, C. Okumura, M. Harada, Y. Saito and M. Okamoto, New York, 1995. J. Liq. Chromatogr., 1996, 19, 2883. 5 Z. L. Fang and G. H. Tao, Fresenius’ J. Anal. Chem., 1996, 355, 21 M. S. Dresselhaus, G. Dresselhaus and P. C. Eklund, Science of 576. Fullerenes and Carbon Nanotubes, Academic Press, San Diego, 6 B. Welz, Microchem. J., 1992, 45, 163. CA, 1996. 7 Z. L. Fang, S. Xu and G. H. Tao, J. Anal. At. Spectrom., 1996, 22 H. Rausch and T. Braun, Anal. Chem., 1997, 69, 2312. 11, 1. 23 M. Gallego, Y. Petit de Pen�a and M. Valca�rcel, Anal. Chem., 1994, 8 Z. L. Fang, M. Sperling and B.Welz, J. Anal. At. Spectrom., 1990, 66, 4074. 5, 639. 24 Y. Petit de Pen� a, M. Gallego and M. Valca�rcel, Anal. Chem., 1995, 9 M. Sperling, X. Yin and B. Welz, J. Anal. At. Spectrom., 1991, 67, 2524. 6, 295. 25 Y. Petit de Pen� a, M. Gallego and M. Valca�rcel, J. Anal. At. 10 M. Sperling, X. Yin and B. Welz, J. Anal. At. Spectrom., 1991, Spectrom., 1997, 12, 453. 6, 615. 26 O. K. Borggaard, H. E. M. Christensen and S. P. Lund, Analyst, 11 M. Sperling, X. Yin and B. Welz, Spectrochim. Acta, Part B, 1991, 1984, 109, 1179. 46, 1789. 27 W. J. Blanchflower, A. Cannavan and D. G. Kennedy, Analyst, 12 R. Ma and F. Adams, Anal. Chim. Acta, 1995, 317, 215. 1990, 115, 1323. 13 Y. H. Sung, Z. S. Liu and S. D. Huang, J. Anal. At. Spectrom., 28 K. Isshiki and E. Nakayama, Anal. Chem., 1987, 59, 291. 1997, 12, 841. 29 R. E. Sturgeon, S. S. Berman, A. Desaulniers and D. S. Russel, 14 M. Sperling, X. P. Yan and B. Welz, Spectrochim. Acta, Part B, Talanta, 1980, 27, 85. 1996, 51, 1891. 30 E. Tserovsky and S. Arpadjan, J. Anal. At. Spectrom., 1991, 6, 15 E. Ivanova, K. Benkhedda and F. Adams, J. Anal. At. Spectrom., 487. 1998, 13, 527. 16 X. P. Yan, W. Van Mol and F. Adams, Lab. Rob. Autom., 1997, 9, 191. Paper 8/08095J 716 J. Anal. At. Spectrom., 1999, 14, 7

ISSN:0267-9477    

DOI:10.1039/a808095j

出版商:RSC    

年代:1999

数据来源: RSC

摘要:

ATOMIC SPECTROMETRY UPDATE Clinical and biological materials, foods and beverages Andrew Taylor,*a Simon Branch,b David J. Halls,c Linda M. W. Owend and Mark Whitee aSupra-Regional Assay Service Trace Element Laboratory, Centre for Clinical Science and Measurement, School of Biological Sciences, University of Surrey, Guildford, Surrey, UK GU2 5XH. E-mail: A.Taylor@surrey.ac.uk bThe Lord Rank Centre, R. H. M. Technology, Lincoln Road, High Wycombe, Buckinghamshire, UK HP12 3QR cTrace Element Unit, Department of Clinical Biochemistry, Glasgow Royal Infirmary University NHS Trust, Castle Street, Glasgow, UK G4 0SF dMinistry of Agriculture, Fisheries and Food, CSL Food Science Laboratory, Norwich Research Park, Colney Lane, Norwich, UK NR4 7UQ eHealth and Safety Laboratory, Health and Safety Executive, Broad Lane, SheYeld, UK S3 7HQ Received 28th January, 1999 Analysis of clinical and biological materials 11.1 General reviews and articles 1.2 Sampling and sample preparation 1.2.1 Sample collection and pretreatment 1.2.2 Solid and slurry sampling 1.2.3 Sample digestion 1.2.4 Preconcentration 1.3 Developments in and applications of multielement techniques 1.3.1 Atomic emission spectrometry with the inductively coupled plasma and the microwave induced plasma 1.3.2 Inductively coupled plasma mass spectrometry and other mass spectrometric techniques 1.3.2.1 Reviews 1.3.2.2 Overcoming interferences in quadrupole ICP-MS (Q-ICP-MS) 1.3.2.3 Multielement determination by Q-ICP-MS 1.3.2.4 Laser ablation ICP-MS 1.3.2.5 Electrothermal vaporization ICP-MS 1.3.2.6 Double focusing magnetic sector ICP-MS 1.3.2.7 Determination of stable isotopes by mass spectrometry 1.3.2.8 Accelerator mass spectrometry 1.3.3 X-ray fluorescence spectrometry 1.3.4 Other multielement techniques and studies 1.4 Developments in single element techniques 1.5 Reference materials and quality assurance 1.6 Hair and nail analysis 1.7 Drugs and pharmaceuticals 1.8 Marine and freshwater biology 1.9 Progress for individual elements 1.9.1 Aluminium 1.9.2 Antimony 1.9.3 Arsenic 1.9.4 Bismuth 1.9.5 Boron 1.9.6 Bromine 1.9.7 Cadmium 1.9.8 Calcium 1.9.9 Chromium 1.9.10 Cobalt *Review co-ordinator, to whom correspondence should be addressed and from whom reprints may be obtained.Analysis of foods and beverages Sampling and sample preparation 1.9.11 Copper and zinc 1.9.12 Gallium 1.9.13 Germanium 1.9.14 Indium 1.9.15 Iodine 1.9.16 Iron 1.9.17 Lanthanides 1.9.18 Lead 1.9.19 Magnesium 1.9.20 Manganese 1.9.21 Mercury 1.9.22 Molybdenum 1.9.23 Nickel 1.9.24 Platinum group metals 1.9.25 Selenium 1.9.26 Silicon 1.9.27 Silver 1.9.28 Sodium and potassium 1.9.29 Strontium 1.9.30 Titanium 1.9.31 Thallium 1.9.32 Tin 1.9.33 Tungsten 1.9.34 Uranides 1.9.35 Vanadium 1.10 Conclusions 22.1 2.1.1 Extraction 2.1.2 Digestion 2.1.3 Preconcentration 2.2 Speciation 2.3 2.4 2.5 Developments in methodology for flame atomic absorption spectrometry Developments in methodology for electrothermal atomic absorption spectrometry Developments in methodology for inductively coupled plasma mass spectrometry 2.6 Progress in individual elements 2.6.1 Arsenic 2.6.2 Iodine 2.6.3 Mercury 2.6.4 Selenium 2.7 Single and multi-element analysis of food 2.7.1 Wine and alcoholic beverages J.Anal. At. Spectrom., 1999, 14, 717-781 7172.7.2 Metal contamination in food 2.8 Dietary intake studies 2.9 Characterization studies 2.10 Reference materials and collaborative trials Table 1 Analysis of clinical and biological materials, foods and beverages References 3T his update presents work published through to October, 1998. Of the developments mentioned in our 1998 review we observed continuing interest in in vivo measurement of lead in bone by XRF, new (or renewed) methods for tissue solubilization and increasing use of high resolution ICP-MS.Application of tungsten coil atomizers were again reported. T his included further use of portable atomic absorption spectrometers but a new area was electrothermal vaporization of analytes into an ICP mass spectrometer. Of new equipment, electrospray mass spectrometry was used by several workers.A resurgence of interest in the metabolism of aluminium was seen and there are reports of lead in plasma being associated with caeruloplasmin, the copper containing protein. Finally, various authors made use of Monte Carlo systems. None, however, have suggested that the chances of winning at roulette are increased. 1 Analysis of clinical and biological materials Advances in the determination of major, trace and ultratrace elements in clinical and biological materials by atomic spectrometric techniques are reviewed for the year ending approximately October, 1998.Table 1 summarizes the publications and relevant conference presentations while the text describes the more important publications and covers themes of current interest. 1.1 General reviews and articles 18 Speciation of clinical and biological materials is discussed in three recent general articles. Cornelis et al.1 give background information about sample preparation and handling, which is critical to maintain the integrity of the species.The need for speciation is discussed by Sanz-Medel2 and analytical strategies to tackle speciation are reviewed. If speciation is to become established in routine laboratories for quality control, environmental protection and regulation, then it must be faster and more reliable, as £obinski et al.3 rightly argue. They described processes developed for microwave leaching and derivatization for alkyl and aryl compounds of Hg, Pb and Sn, reducing preparation time to a few minutes.Separation by GC was possible in under 2 min on an isothermal multicapillary column only 22 cm long; this has the potential of producing a compact accessory for an atomic spectrometer. They presented a solution to the problem of an ICP mass spectrometer not tolerating the high concentrations of methanol used in reversed-phase HPLC. On-line dilution was used and the diluted eluate was passed to a low-consumption DIN.Faster LC is possible by decreasing the particle size and increasing the homogeneity of the packing. They demonstrated a complete separation of cobalamin species in under 2 min by using a 50 mm long column of non-porous C packing of particles of size 1.5±0.17 mm. Schramel4 reviewed modern techniques for trace element analysis and speciation. Recent developments in analysis of trace elements in clinical samples, which have been covered in recent Updates, feature in a review by Taylor.5 Last year's Update6 reviews publications appearing in the year up to about October, 1997.718 J. Anal. At. Spectrom., 1999, 14, 717-781 1.2 Sampling and sample preparation 1.2.1 Sample collection and pretreatment. The risk of infection from viruses can be reduced by lowering the pH of the sample, according to Veillon et al.7 Samples were diluted 1+4 with 0.1 M HNO3 which lowered the pH to about 1.5 without precipitating proteins.Samples could then be transferred safely from the sample handling area to the analytical laboratory for determination of Se by ETAAS. Microwave irradiation can facilitate deproteinization steps, as Bohrer et al.8 demonstrated. More than 99% of the protein in serum was removed after irradiating a 1+1 dilution with 1% m/v TCA. This concentration is 10-fold lower than is conventionally used for deproteinization. They applied the technique in the determination of Al in serum by ETAAS. Precipitates often form in urine samples on standing.Burden et al.9 showed by X-ray microanalysis that this contained mainly Ca and P but also could include some trace metals. To ensure complete dissolution in the determination of Al in urine by ETAAS, they diluted the urine 1+1 with 0.22 M HCl, warmed it to 40 °C and then allowed it to cool to room temperature.Krachler et al.10 compared simple dilution, UV irradiation and microwave digestion with HNO3-H2O2 for determining a range of trace elements in urine by ICP-MS. Simple dilution required a higher dilution ratio to obtain a stable reading than the digested samples but the LODs were comparable and all techniques gave accurate results. Comparison of results on digestion of urine with and without the sediment showed significant diVerences for Pb but not for Cd, Co, Cr, Ni, Sb, Sr and V.1.2.2 Solid and slurry sampling. In two recent applications of direct solid sampling, the high organic content found in most biological samples was removed in a pretreatment step. By using prior dry ashing externally, the resulting ash contains the element at a concentration 10-40-fold higher than in the original material. Applying this to the determination of Se in biological materials, Minami et al.11 used a Pd modifier to prevent volatilization.Selenium in the ash was then determined 3.3 ng g-1 compared to 0.13 mg g-1 for determination in the by solid-sampling ETAAS. The LOD in this case was original material directly. Results on CRMs using simple aqueous solutions for calibration were in good agreement with the certified values. To determine Cd and Pb in biological materials, Okamoto et al.12 used ETV-ICP-AES. The sample, mixed with (NH4)2HPO4, was ground to a fine powder. A portion (10 mg) of the powder, placed in a tungsten boat furnace, was digested in-situ with TMAH at 130-150 °C.After pyrolysis, a vaporization step produced Cd and Pb atoms which were transported into the ICP. Calibration was with CRMs. Although 2% nitric acid was shown by Bermejo-Barrera et al.13 to eYciently extract Ni from slurries of hair in determination by ETAAS, no improvement in results was seen when compared with simple suspension of powdered hair in deionized H2O.Glycerol was added as a stabilizing agent and Mg(NO3)2 was the chemical modifier. However, Meeravali and Kumar14 found that slurries of biological CRMs were not stable for most elements in glycerol but were stable in 5% HNO3, which extracted a high percentage of the elements Cd, Cu, Mn and Pb that they studied. Determination was by transverse heated ETAAS using a programme with no charring stage but a high-temperature drying stage. Although no modi- fier was used, direct aqueous calibration could be used without chemical interference for Cd, Mn and Pb, but for Cu, standard additions calibration was necessary.This was also the element with the lowest extraction (57%) into acid. Calcium-containing pharmaceuticals needed to be ground down to an average particle size of 3 mm to produce satisfactory results in thedetermination of Pb by slurry-sampling ETAAS using a Mo-tube atomizer.15 Thiourea was added to remove interferences and H2, added to the Ar atmosphere, gave increased sensitivity.Matrix-matched standards were used for calibration. Good agreement was found with results obtained after acid digestion. 4 1.2.3 Sample digestion. More applications of alkylamines as solubilizing agents were described. Tao et al.16 found that, when they used TMAH to digest biological tissues for the determination of Hg by FI-CVAAS, the sensitivity for various species of Hg diVered. To convert all forms to inorganic Hg, 0.2% m/v KMnO was added on-line.Dissolution in TMAH was complete in 30 min, allowing a high throughput of samples. Animal tissue CRMs were dissolved in TMAH by Pozebon et al.17 for the determination of a range of trace elements by ETV-ICP-MS. A small mass of sample, 20-100 mg, was mixed with 10-200 ml of 25% m/v TMAH to dissolve. In determination, Pd was added separately as a modifier. Tertiary amines are very suitable for retaining volatile elements such as I for determination as Krushevska et al.18,19 demonstrated.Simultaneous determination of I with other trace elements was then possible by ICP-MS. On-line digestion systems have been developed further. Gra�ber and Berndt20 described a high pressure system using a HPLC pump to supply pressure in a capillary which was resistively heated to 250-350 °C. Residual C was less than 1% after digestion. Previously, systems have been low pressure with microwave heating.Such a system was designed by Huang et al.21 for automated preparation of blood and serum for determination of trace elements by ICP-MS. Samples were pumped with an acid mixture into a Teflon tubing coil in a focused microwave oven. After passing through a cooler, the sample was transferred to an iminodiacetate-based resin column for retention of trace elements, separating them from the main matrix components. The analytes were subsequently eluted for determination. The whole system was controlled by an expert system.However technologically interesting this system is, the application to the determination of Cu, Fe, Ni, Pb and Zn in blood and serum at a rate of 6 samples h-1 does not seem to be progress when more direct and faster approaches are possible. For on-line determination of Hg in slurried samples, Lamble and Hill22 used a carrier stream of 3- and Br- and HCl, mixed the sample with a solution of BrO heated it in a coil in a microwave digestor for complete breakdown of methylmercury to Hg.Determination was by CVAFS. To speciate As in biological materials,23 they separated the species by HPLC, broke down the species to inorganic As by on-line microwave digestion, reduced AsV to AsIII with L-cysteine and then determined As by HGAAS. In the system developed by Burguera et al.24 for urine, the column eluate was directly coupled to HGAAS for determination of AsIII, AsV, monomethylarsonic acid (MMA) and dimethylarsinic acid (DMA).Total As was determined in a parallel flow system in which the sample was digested in a microwaveheated coil before determination. The advantage of the Lamble and Hill approach is that species, such as arsenobetaine, that are not reduced to a hydride can be determined, allowing a complete speciation. An alternative to microwave digestion for breakdown of As species to inorganic As is UV photooxidation.Tsalev et al.25 used this on-line after separation by HPLC on an anion-exchange column and prior to determination of As by HGAAS. The sample was mixed with alkaline peroxydisulfate and pumped to a PTFE knotted reactor, where it was irradiated for 93 s. Their system used an 8 W lamp but Ritsema and van Heerde,26 in their method, used a 500 W Hg lamp, adding H2SO4 and H2O2 for oxidation oV-line for 8 h before determining total As by HGAAS.The conditions seem somewhat excessive in view of the results obtained by Tsalev et al.25 A recirculating system for digestion using microwave heating was developed by Perez-Jordan et al.27 A powdered tooth sample mixed with HNO3 in a flask was pumped through PTFE digestion coils in a microwave oven and back to the flask. The flask was cooled in ice and stirred. After 10 min, the digest was analysed for Mg and Sr by FAAS and for Na and K by FAES.Modifications to microwave digestion systems to handle small sample masses have been described. Deaker and Maher28 used two 7 ml screw-topped vials placed within a normal 120 ml vessel. Samples of less than 0.1 g digested with 1 ml HNO3 in a multistage heating programme lasting 49 min gave quantitative recovery of Se from three marine biological CRMs when measured by ETAAS. Vapour-phase digestion with HNO3 in a commercial pressurized microwave digestion system was applied to samples of biological CRMs of weights 0.05-0.1 g in a study by Amarasiriwardena et al.29 Addition of 250 ml of H2O allowed coupling of microwave energy to the sample. Ultrasound was used with good eVect to assist the extraction of Se from biological materials with 4% v/v HNO3 by Mierzwa et al.30 Comparison with results from full digestion and with the certified values of the CRMs showed the eYcacy of the extraction.1.2.4 Preconcentration.The changing pattern of sample enrichment is illustrated by recent publications. Conventional liquid-liquid extraction was used by Bergerow et al.31 in the determination of Au, Pd and Pt in urine by ETAAS. An enrichment of 25-fold was achieved with extraction into MIBK with APDC. Liquid-liquid extractions can be carried out on-line in an FI system but it has been found more convenient to use sorption of the metal chelates on a knotted reactor; these are subsequently eluted. Ivanova et al.32 applied this to the determination of Cu, Mn and Ni in biological materials comparing APDC and 8-hydroxyquinoline as chelating agents. For Cu and Ni, APDC was better, giving enhancement factors o44 and 21, respectively, but could not be used for Mn.This was enriched up to 8-fold with 8-hydroxyquinoline. Developments in supported liquid membrane technology could rekindle interest in on-line liquid-liquid extraction. In their method for Pb in urine, Djane et al.33 used a fixed layer of 40% m/m di- 2-ethylhexylphosphoric acid in kerosene, forming a membrane above which was pumped sample with buVer at pH 3 and below 1 M HNO3.An extraction eYciency of 95%into the HNO3 phase with enrichment factors up to 200 could be attained in 45 min.Work by He et al.34 showed the possibilities with solid-phase microextraction. In determining methylmercury in biological materials, they introduced KBH4 into a GC vial to convert methylmercury to its hydride.A capillary fused-silica fibre, etched with HF, was inserted into the headspace above the sample for 90 min. The fibre was then transferred to a GC column where the methylmercury was desorbed and the Hg eluted from the column detected by AAS. In a study of the use of living organisms for metal speciation, Pe�rez-Corona et al.35 showed that baker's yeast immobilized on silica gel and packed in a PTFE microcolumn could retain both methylmercury and inorganic Hg, the former on the yeast and the latter on the silica.ganic Hg with 0.8 M CN-. Methylmercury could be eluted with 0.02 M HCl and the inor- 1.3 Developments in and applications of multi-element techniques 1.3.1 Atomic emission spectrometry with the inductively coupled plasma and the microwave induced plasma. An evaluation by De Wit and Blust36 of the microconcentric nebulizer 719 J. Anal. At. Spectrom., 1999, 14, 717-781(MCN) in the determination of metals by axial ICP-AES showed that the sample volume could be reduced to a few hundred ml for multielement analysis but the performance was poorer than with a standard concentric glass nebulizer.The eVects were element-dependent but for the MCN the LOD was on average 3.4-fold worse and the eVects of salt on signal stability were more pronounced. Nevertheless, the MCN could tolerate continuous nebulization of solutions containing up to 3.5% salt for several hours without blocking and, in the analysis of digests of biological CRMs, showed results with good accuracy and precision.Speciation studies by Bra�tter et al.37,38 using HPLC coupled on-line to ICP-AES (or ICP-MS) showed that the proteinbinding of trace elements in human milk was significantly diVerent from that in infant milk formulae. In the formulae, the binding pattern depended on the main component (cows' milk or soya milk), the processing and the chemical form of the trace element additive.To compensate for poor absorption, some elements were added at higher concentrations than are found in human milk. For example, Fe was up to 20-fold higher with a very diVerent binding pattern to that of human milk. In a second study from the same Institute,39,40 speciation of Cu, Fe, Sn and Zn in red-cell lysates by SEC-ICP-AES allowed identification of superoxide dismutase (containing Cu and Zn), catalase (Fe), haemoglobin (Fe, S), glutathione peroxidase (Se, S) and carbonic anhydrase (Zn).In healthy infants from birth to 4 months, the superoxide dismutase and catalase increased with age. The mineral composition of bones and similar materials is conveniently determined by ICP-AES. Changes with age in men's and women's calcanei were studied by Tohno et al.41 with both ICP-AES and dual-energy X-ray absorptiometry. Both showed age-dependent decreases in the mineral content for men over the age range 40-98 but not for women.In a second study on the same subjects,42 they looked at the mineralization of femoral arteries. The relative contents of Ca and P started to increase before the age of 60 but Mg only increased after the age of 70. Sulfur showed no significant change.Wollastonite (CaSiO3), a potential bone implant material which readily forms a layer of hydroxyapatite when in contact with body fluids, was analysed by De Aza et al.43 using ICP-AES for Ca and Si and for the impurities introduced in processing.The precision obtained in the determination of Ca and Si was said to be comparable to that achievable by gravimetric procedures. In a study on teeth, Capota et al.44 determined Al, Cd, Cr, Cu, Fe, Pb and Zn after ashing the samples in O prior to digestion with HNO 2 3-H2O2. The suitability of ICP-AES to determine refractory elements is shown in two recent studies.For determination of Ti in blood plasma of patients treated with a Ti-containing anticancer drug, Einha�user et al.45 investigated measurement by ICP-AES with an USN. Even with the preferred dilution of 1+99 with water, an LOD of 1-2 mg l-1 was achieved. Methods were also developed using ETAAS with longitudinal and transverse heating, the latter giving less tailing. The authors preferred ICP-AES as ETAAS suVered more from matrix and memory eVects.Tungsten was determined by ICPAES by Marquet et al.46 in the investigation of acute toxicity in a young soldier in the French artillery. He had drunk a mixture of beer and wine used to rinse a hot gun barrel. Higher than normal concentrations of W were found in his blood, urine, hair and nails. Iohexol, an iodinated radiographic contrast medium, can be used to estimate the glomerular filtration rate by following the elimination of I by XRF from blood plasma after injection.Braselton et al.47 showed that, by using a more sensitive ICPAES method to measure the I, a much lower dose of iohexol could be given. They determined I at the 178.3 nm line on the supernatant after precipitation of proteins from the plasma 720 J. Anal. At. Spectrom., 1999, 14, 717-781 sample with a TCA-based medium. Inter-element correction was necessary for interference by P which was measured at 214.9 nm. 1.3.2 Inductively coupled plasma mass spectrometry and other mass spectrometric techniques. 1.3.2.1 Reviews. Moens48 has reviewed applications of mass spectrometry in the determination of trace elements in biological materials, discussing also stable isotope tracer studies. 1.3.2.2 Overcoming interferences in quadrupole ICP-MS (Q-ICP-MS). Interferences in the determination of trace and ultratrace elements in biological fluids by Q-ICP-MS were discussed by Hsiung et al.49 They advocated the use of internal standards of close mass to correct for non-spectral interferences.Dilution also helped. For multielement analysis, fivefold dilution and four internal standards (Be, Ga, In and Ir) across the entire mass range gave adequate accuracy for most elements with simple external standards, but for Cs and Zn calibration with internal standards was necessary. Accurate results for Cu and Zn in serum and urine could be obtained by measuring the 65Cu and 68Zn isotopes, respectively.Cool plasma conditions can eliminate some polyatomic interferences, as Hedrick50 demonstrated. Addition of O2 as a C scavenger also helped, enabling the determination of 52Cr and 24Mg. A diVerent approach was used by Krushevska et al.51 for 52Cr. To reduce the C and Cl, they optimized acid digestion and then corrected for remaining ArC+ interference at mass 52 mathematically. The same approach was used for V, but Minnich et al.52 preferred cryogenic desolvation to condense Cl as HCl.This allowed the determination of V in urine for occupational monitoring. 1.3.2.3 Multielement determination by Q-ICP-MS. Twentysix major and trace elements in tissues (brain, lung, spleen, kidney, heart and liver) of 21 human fetuses of gestational age 16-22 weeks were determined in a study by Gelinas et al.53 Samples were prepared by microwave digestion. The range of concentrations found was generally much lower and much narrower than in adult tissues.In an assessment of variation within diVerent regions of five brains, no significant diVerence was found for any of the elements. Amyloid plaques in the brains of patients who had Alzheimer's disease were analysed by FI-ICP-MS by Beauchemin and Kisilevsky.54 The plaque cores, isolated by homogenization and gradient ultracentrifugation, were suspended in salt buVer and dissolved in HNO3. Concentrations of Al, Fe and Zn were relatively high in the extra the range 2-20 mg l-1, whereas Cr, Cu, Mn, Ni and Pb concentrations were between 0.2 and 0.8 mg l-1.A study, reported by Lutter et al.,55 used ICP-MS, ICPAES and NAA to determine toxic and essential metals and radionuclides in breast milk from mothers in Kazakhstan. Concentrations found were not abnormal compared with data from other countries and breast-feeding was considered safe. Bra�tter et al.38 used ICP-MS and ICP-AES to detect metals in the speciation of breast and formula milk by HPLC.The protein-binding pattern was very diVerent between breast and formula milk and even within the formulas there were diVerences depending on the main component, its treatment and the form of the inorganic supplement added. Copper and Zn were determined in plasma and urine after 20- and 10-fold dilution, respectively, by Szpunar et al.56 An internal standard, Y, was added and standard additions were used to eliminate matrix eVects.Copper was measured as 65Cu. Flow injection coupled to a DIN allowed determination in samples as small as 1-2 ml. 1.3.2.4 Laser ablation ICP-MS. Laser ablation ICP-MS in combination with gel electrophoresis was investigated jointlyby workers in SheYeld, UK, and Copenhagen, Denmark, for the speciation of trace elements in biological samples.57-59 Nielsen et al.58 separated the proteins in a serum spiked with Co by using two-dimensional electrophoresis.Proteins were identified by a staining technique. Laser ablation ICP-MS produced a map of the distribution of Co from which it was possible to conclude that the Co was associated principally with six major proteins. Further work59 assessed metal binding patterns in therapy with Au and Pt. To allow laser ablation of soft tissues, Ek60 froze samples been more publications and conference presentations on double focusing magnetic sector ICP-MS [or high-resolution to produce a hard surface for ablation.A special cell allowed ICP-MS (HR-ICP-MS)], enabling a better appreciation of its periodic introduction of liquid N2 to keep the sample frozen. potential. In this respect, the most informative are those which Calibration was with frozen standard solutions. Imaging of trace element distribution was possible. 'Fingerprints' of cannabis samples from Western Australia were obtained by Watling61 using laser ablation. Ground samples were pressed into cardboard mounts under high pressure. Ternary plots of ratio per cent.of particular elements measured by laser ablation ICP-MS allowed identification of the source of the cannabis. 1.3.2.5 Electrothermal vaporization ICP-MS. Electrothermal vaporization allows the use of slurries and more concentrated solutions, requires smaller volumes of solution and allows removal of water and matrix components, thus reducing interferences in the determination.Lanthanides are not an obvious candidate for ETV as they are refractory but Buseth et al.62,63 showed that the addition of Freon-23 (CHF3) to the Ar purge gas lowered the vaporization temperature and reduced memory eVects. Absolute LODs were 1-20 fg for all the lanthanides, allowing determination of lanthanides in serum down to LODs of 0.05-1.2 mg l-1. Blood samples from 90-350 ng l-1 for Pr, Nd and Gd but Ce was around 13 mg l-1, 30 healthy volunteers gave mean concentrations in the range possibly because of contamination from sample tubes.Comparison of results on tissue samples analysed by slurry sampling and after digestion showed similar accuracy and precision. Chemical modifiers are as important in ETV-ICP-MS as they are in ETAAS and much of the knowledge has been transferred from that technique. Pozebon et al.17 determined a range of elements in biological SRMs by ETV after treatment with TMAH to form solutions or slurries. Palladium was the chemical modifier for Ag, As, Cd, Co, Cr, Cu, Mn, Ni, Se, Te and V, but for Bi, Sb, Sn and Pb, an iridium-coated tube was used. The use of iridium-coating was examined by these authors in a separate publication.64 For the determination of As, Pb and Se, this allowed the pyrolysis temperature to be increased to 1300 °C.Although more than 100 firings could To monitor workers involved in nuclear fuel reprocessing, be obtained with each coating, gradual loss of sensitivity of Nanni et al.72 used HR-ICP-MS to measure Th and U in about 50% occurred over this lifetime, requiring frequent human urine.Samples were diluted ten-fold with addition of re-sloping of the calibration curve. Reduction of sensitivity matched standards. The LODs (0.26 and 0.01 mg l-1 for Th for Pb was much less significant. Water and urine CRMs were Bi as an internal standard and were measured against matrixanalysed with good accuracy.A novel method for speciation and U, respectively) can be compared to the LODs of 0.5 mg l-1 of Hg in biological tissues using ETV-ICP-MS was described for both elements reported by Schramel et al.73 for determiby Willie et al.65 Samples were solubilized with TMAH. Total above this (mean 8 and 10 mg l-1 for Th and U, respectively), Hg was determined without further treatment. Through the nation by Q-ICP-MS.As concentrations found72 were much addition of iodoacetic acid, acetic acid and sodium thiosulfate, either instrument would seem to have adequate sensitivity. To methylmercury was converted to methylmercury iodide, which measure very low levels of Pu and lanthanides in blood and was driven oV in the drying stage allowing only inorganic Hg urine, Pickford et al.74 used chemical separation methods to to be measured. The method was tested on marine biological preconcentrate samples and to reduce the matrix concen- CRMs.For determination of Hg in urine, Lee et al.66 compared tration. To eliminate interference from C, Pb and Th in the modifiers and found a mixed modifier of Pd, Mg(NO3)2 and determination by HR-ICP-MS, high grade reagents and clean 0.2% v/v HCl to be suitable. Spiking with the isotope 201Hg allowed quantitation by ID. Mercury was determined in NIST SRM 2670 Toxic Metals in Urine and in several samples from normal subjects.The LOD of the method was 0.02 mg l-1. In In a conference presentation, Garbe-Scho�nberg and Sievers75 further work,67 these workers described the determination of Cd and Pb in urine by ID using ETV-ICP-MS. A relatively low volatilization temperature (1000 °C) with 1% HNO3 as chemical modifier allowed separation of the analyte from the 0.005 mg l-1, respectively. major matrix components. The LODs were 0.02 and 1.3.2.6 Double focusing magnetic sector ICP-MS.There have have carried out a direct comparison with Q-ICP-MS. Kishiki et al.68 compared multielement determination of digests of animal tissue CRMs by both techniques. Results on NIST SRM 1577b Bovine Liver by HR-ICP-MS agreed well with certified values, except for P, Pb and V. The poor result for P could be explained by detector saturation. The reasons for low Pb and V results were to be investigated further. Results by Q-ICP-MS were good at high mass, but gave low results for Cd, Cu, Pb and Zn.In a comparison of the determination of the Pt group metals in urine, Krachler et al.69 concluded that Q-ICP-MS with USN was totally inadequate to quantify baseline concentrations but HR-ICP-MS with USN operated in low-resolution mode had adequate sensitivity, although it was still subject to potential interferences from 106Cd and 40Ar66Zn on 106Pd and 206Pb2+, 87Sr16O and 40Ar63Cu on 103Rh. Samples were digested by UV photolysis and diluted 1+19 for analysis.The LODs were 5, 0.6 and 0.5 ng l-1 for Pd, Pt and Rh, respectively, enabling studies to be made on the eVect of release of Pt group metals from catalytic converters in cars on urban children. On 30 young people, median concentrations found were 9.5, 1.0 and 11.7 ng l-1 for Pd, Pt and Rh, respectively. Using a similar method, Begerow et al.70 achieved better LODs in urine, 0.17 and 0.24 ng l-1 for Pd and Pt, respectively.Lower dilution, 4.2-fold, was used. They found higher levels in 21 adults (mean 140 and 1.8 ng l-1 for Pd and Pt, respectively). In a comparative study of ETAAS, Q-ICP-MS and HR-ICP-MS for the determination of Al in serum, Sariego Mun~ iz et al.71 found that lower LODs were possible with the IS techniques (0.35 and 0.85 mg l-1 for HR- and Q-ICP-MS, respectively) than with ETAAS (2 mg l-1). Although there were negligible matrix eVects for serum diluted 1+4 on the Q-ICP-mass spectrometer, matrix eVects were evident with the HR-ICP mass spectrometer even at 1+9 dilution, requiring the use of internal standards (Be and Sc).Even with HR-ICP-MS, most of the serum samples from healthy individuals were below the LOD (i.e., less than 0.35 mg l-1). room conditions were necessary to keep the concentrations of these elements low. In this way, it was possible to determine 244Pu down to 0.1 fg l-1 in urine.claimed that, using HR-ICP-MS, they could determine 721 J. Anal. At. Spectrom., 1999, 14, 717-781accurately major, trace and ultratrace elements in serum and urine using only a small volume of serum diluted by a factor of 20- to 100-fold. An internal standard was added to correct for matrix interference and for drift. Publication is awaited with interest. In order to monitor workers exposed to Cd and other metals, Townsend et al.76 developed a method of determining Cd, Cu, Pb and Zn in urine by HR-ICP-MS.A spectral resolution of 3000 was required to separate Cu and Zn isotopes from interference, but the resolution was dropped to 300 to achieve greater sensitivity in the determination of Cd and Pb. Samples were diluted 1+9, incorporating In as an internal standard. Simple external calibration gave results on urine RMs which were in the accepted range. Comparison of Cd results with those obtained by ETAAS showed good agreement.1.3.2.7 Determination of stable isotopes by mass spectrometry. In a comparison of the measurement of isotope ratios of Mg by electron-impact mass spectrometry (EI-MS) of volatile Mg derivatives with determination by ICP-MS after acid digestion of samples, Batel et al.77 showed that ICP-MS gave better precision (0.19% and 0.38% RSD within- and between-batch, respectively) than EI-MS (1.28% and 2.6%). The method was used to evaluate the bioavailability of a Mg pharmaceutical form in humans.A method developed by Woolard et al.78,79 used double focusing magnetic sector ICP-MS for the determination of all four stable Pb isotopes. Each sample was scanned 300 times and compared with a certified standard that had been isotopically characterized by TIMS. Isotope ratios could be measured to a precision of 0.07-0.22% RSD. Total analysis time was about 5 min. The method was applied to Pb isotope tracer studies in animals and humans.In a further variant of this method,80,81 improved precision (<0.1% RSD) was obtained by using 1200 scans per sample. A comparison of results on blood and dust samples with those obtained by TIMS showed agreement within the propagated s of both methods, but there was a small but significant bias which could not be explained. The method provided an equally precise but less costly and more rapid alternative to determination by TIMS. To interpret strontium isotope ratios in bones as a tool for studying dietary input from discrete geochemical environments, a precision in measurement of at least 0.1% RSD is required. Latkoczy et al.82 showed that, using HR-ICP-MS, they could achieve a precision of about 0.05% RSD.Compared to other more accurate and precise alternatives such as TIMS, the ease of operation and sample preparation together with the short analysis time (10 min) allowed HR-ICP-MS to handle numerous samples at reasonable cost.The technique was used in a study of skeletal remains from a Neolithic settlement in lower Austria. Becker and Dietze83 showed that HR-ICP-MS allowed interferencefree determination of the 39K540K ratio with a precision of 0.7% RSD and showed promising results for the determination of Ca and Mg isotope ratios. 1.3.2.8 Accelerator mass spectrometry. As in previous years, the main clinical application of this technique is to study the metabolism of Al using the isotope 26Al.On the basis of their measurements on human subjects, Kislinger et al.84 developed a model to describe the kinetics of elimination of Al. This model comprised a central compartment (blood and interstitial fluid) feeding three peripheral compartments. Excretion of Al was mainly from the kidneys into urine and, to a lesser extent, via the liver with faeces. In two patients with renal failure, a diVerence in the pattern of excretion was seen in comparison with that for healthy subjects.In rats, Zafar et al.85 found that 26Al accumulated to the greatest extent in bone, followed in order by spleen>kidney=liver>brain. Only 0.97% of the 722 J. Anal. At. Spectrom., 1999, 14, 717-781 dose was absorbed. The absorption of Al from Al-containing vaccine adjuvants was studied by Flarend et al.86 using rabbits. The area under the blood 26Al concentration curve for 28 d showed that three times more Al was absorbed from aluminium phosphate than from aluminium hydroxide.Yumoto et al.87 found that 26Al could cross the blood-brain barrier in rats and was not eliminated over 270 d whereas the 26Al in blood had mainly gone after 75 d. Studies by Barker et al.88 on mice showed that the role of transferrin in transporting Al may be less than originally believed. In rats with low serum transferrin concentrations, uptake of 26Al was similar to that in controls, except in bone. Mice treated with antibodies against the transferrin receptor also showed similar results to controls, except in spleen and muscle.They concluded that transport by citrate and other low-molecular mass complexes was important. The analytical methods used by the Manchester group were described in more detail.89 After pressure digestion of blood, plasma and tissue samples with HNO3, Al carrier was added and the Al isolated by selective precipitation with quinolin-8-ol.Ignition at 800 °C produced Al2O3 which was by AMS. The LOD for 26Al was 10-18 g. mixed with Ag powder for measurement of the 26Al527Al ratio Bone resorption measured with a 41Ca tracer by AMS gave results which were comparable to those obtained with conventional stable isotope measurements.90 Freeman et al. indicated that, in contrast with other Ca isotope tracers, 41Ca could be administered in such a way that a quasi steady state was achieved between the various body pools.Measurement of 36Cl and 129I in teeth and bones can provide information on exposure of individuals to fallout from nuclear weapons tests or emission of radionuclides from nuclear power stations. Cornett et al.91 developed a pyrolysis technique to liberate these elements from teeth and bone for measurement of the isotopes by AMS. Samples from human subjects who had been exposed to weapons fallout had elevated levels of these isotopes.1.3.3 X-ray fluorescence spectrometry. Development of invivo XRF techniques for estimation of heavy metal burden continues, although some evidence from clinical studies suggests that more traditional and simpler measurements of metals in blood and urine are more eVective. The first phase in the development of a faster, more accurate XRF system for invivo measurement of Pb in bone was reported by Niemela and they were able to reliably measure 17 mg Pbg-1 in a bone Grodzins.92 Using a monochromatic, polarized X-ray beam, below 5 mg Pb g-1.Ao et al.93,94 developed a Monte Carlo phantom and expected with further optimization to measure technique for the simulation of in-vivo XRF and applied it to reduce the eVect of Compton scattering by optimizing the source-sample-detector geometry in K-shell XRF and by using a polarized source for L-shell XRF. A Monte Carlo approach was used by Hugtenburg et al.95 to model the measurement of cis-platin uptake by in-vivo XRF.Under optimized conditions, an LOD of 50 mg g-1 was found. Clinical studies of the estimation of Pb burden were reviewed by Rosen.96 Kovala et al.97 studied the eVects of low-level exposure to Pb on neurophysiological functions amongst Pb battery workers. Exposure to Pb was assessed by both in-vivo XRF of Pb in tibial and calcaneal bones and from regular blood Pb measurement. Assessment of peripheral and central nervous system functions showed a correlation with long-term exposure to Pb.Blood Pb history showed a better correlation with the eVects of Pb than the bone measurements. They concluded that a more accurate estimate of health risk from Pb could be obtained from a history of blood Pb measurement. Cadmium in superficial liver tissue and in kidney cortex measured by in-vivo XRF was compared with blood and urine Cd measurements and records of Cd in workplace air levelsfor 30 Ni-Cd battery workers in a study by Boerjesson et al.98 Kidney Cd concentrations were above the LOD (30 mg g-1) for 63% of the workers while 48% had liver Cd concentrations above the LOD for that substrate (3 mg g-1).Kidney Cd correlated significantly with current blood and urine Cd concentrations; liver Cd correlated significantly only with urine Cd concentrations. No correlation was found with the integrated levels of Cd in the workplace air.They concluded that current urine Cd measurements can be used to predict the kidney and liver Cd concentrations. Comparison of data with more traditional measurements featured also in a study by Gerhardsson et al.99 of in-vivo XRF measurement of Cd in kidney and Pb in bone for 22 smelter workers. O'Meara et al.100 examined the possibility of in-vivo XRF measurement of U in bone. A 57Co source (1 mCi) was used The LOD found was 20 mg g-1, which they concluded was with 180° backscatter geometry and a hyperpure Ge detector.not suYcient for use in occupational monitoring. Apart from in-vivo techniques, measurement of previous Pb exposure can be obtained by measurement of Pb in shed teeth or in bones post-mortem. Lead in shed teeth measured by k-shell XRF was compared in children from Beijing, industrial sites in the Urals and urban USA in a study by Bloch et al.101 Calibration was by aqueous solutions of Pb and the signal response of Pb in teeth was divided by two to account for the diVerence in density between teeth and the standard solutions.Bone samples from a late 19th century graveyard in Colorado, USA, analysed for several elements by XRF and ICP-AES, showed Pb concentrations comparable to those found in modern Pb smelter workers.102 In order to overcome the high dead time associated with a high matrix interference in the determination of trace elements in blood plasma by TXRF, Savage and Haswell103 examined the use of chemical modifiers in drying and ashing the sample in order to prevent losses of volatile elements and to modify the matrix.Despite the addition of Ni and Mg nitrates, losses of As and Se were found when using microwave ashing but dead time was reduced to about 20%. However, since microwave ashing took about 20 min, it was more practical to accept a higher dead time (and consequently longer measurement) with simple air drying.The use of Mg(NO3)2 together with APDC was found to be beneficial in improving precision by ensuring even dispersion of the sample on drying. Impressive results were obtained on 16 elements in the Versieck second generation CRM and in a range of samples from the Trace Elements Quality Assurance Scheme, Guildford, UK. Sample volume was only 10 ml and precision was generally better than 8% RSD. Normal concentrations of bromide in human blood were determined by Olszowy et al.104 on 183 random blood samples by WDXRF. Two samples from each donor were mixed to give 16 ml (!) which were put into an Al sample cup to give a depth of 20 mm for irradiation.A range of 2.5-11.7 mg l-1 was found (mean±s: 5.3±1.4 mg l-1). Zaichick105 compared three methods for estimating extracellular water by determination of the corrected bromine space using stable Br and EDXRF measurement of Br in blood, urine and saliva.All approaches gave results which were in agreement with those obtained by use of a conventional 82Br measurement. Fukumoto et al.106 described applications of a newly developed XRF system for element mapping with a spatial resolution of better than 20 mm. Examination of a sliced tissue sample taken at autopsy from a patient with brain calcification revealed high concentrations of Ca, Fe and P on the internal surfaces on capillary vessels.Improvement in data handling enabled Brands et al.107 to produce an element map of 1024 pixels in less than 1 min and thus allowed continuous updating of the map. Imaging of Hg in hair108 and kidney109 using SRXRF enabled workers at the University of Tsukuba, Japan, to study the accumulation of Hg in rats. After endogenous exposure to methylmercury, Hg accumulated in the cortex of the hair while the predominant site after exogenous exposure was the cuticle.1.3.4 Other multielement techniques and studies. Simultaneous ETAAS and FAAS was used by Gottelt et al.110 to determine Cd, Cu, Fe, Mn, Ni, Pb and Zn in liver and kidney samples from cattle taken from 5 areas of Brandenburg, Germany. The data were analysed by multivariant techniques to define the origin of the sample. Data on reference values are always useful, particularly for children. Rukgauer et al.111 reported serum concentrations of Cu, Mn, Se and Zn, measured by ETAAS, for healthy children 20.4 mmol l-1, 25.5 nmol l-1, 0.77 mmol l-1 and 12.6 mmol l-1 in eight age groups. Overall mean concentrations were for Cu, Mn, Se and Zn, respectively.Concentrations of Cu, Fe and Zn in 157 broncho-alveolar lavages before and after centrifugation were measured by Harlyk et al.112 Zinc was measured by FAAS and Fe and Cu by ETAAS using standard additions calibration. Most of the trace elements were in the supernatant.An inverse relationship between Zn and Cu concentrations was apparent. Centeno et al.113 measured essential and toxic trace element levels in human placental tissues by FAAS and ETAAS after microwave digestion. Reference values were established. Some data on Cu concentrations from pregnancies at risk for Menkes disease showed higher values in parenchyma and membrane than those for controls. Values for trace elements in the lung lobes of Japanese urban dwellers were established by Tsuchiyama et al.114 using FAAS and ETAAS.Copper and Zn concentrations showed a smaller inter-individual variation and were not apparently aVected by smoking or occupational exposure. The elements Al, Cd, Cr, Mn, Ni and Pb showed much larger inter-individual variation and were aVected by environmental or occupational exposure. Concentrations of Cd, Cr, Ni and Pb were associated with smoking and exposure to metal-containing dusts at work gave higher concentrations of Al, Cr, Mn and Ni.In a study of Blackfoot disease (a vascular disorder resulting in gangrene of the foot), Horng and Lin115 found concentrations of As, Hg and Pb in urine that were significantly higher than in controls and concentrations of Se and Zn that were lower. Mercury and As were determined by continuous flow CVAAS and HGAAS, Pb and Se by ETAAS and Zn by FAAS. 1.4 Developments in single element techniques Tungsten-coil atomizers are small, relatively low in cost and use low power.They have been used to their best advantage in a portable analyser for Pb in blood developed by workers at Wake Forest University, USA.116 This, powered by a 12 V car battery, used a conventional Pb HCL as the source and a CCD array detector mounted in a laptop computer. Background correction was by near-line correction at the nonabsorbing 280 nm line and the atomizer was purged with 10% H2 in Ar to prevent oxidation.The application of this device Cd in aqueous solution was 3 mg l-1. Precision was about 10% to determine Cd in urine has been published.117 The LOD for RSD. The same group investigated the potential of the W-coil atomizer for ETV-ICP-AES, using the same inexpensive CCD detector for measuring the emission from the plasma.118 The LODs were comparable to those obtained with conventional ETAAS but the advantage was the ability to measure several elements simultaneously.A parallel development119 has been a system for multi-element ETAAS using four EDLs or HCLs and three 45° coarse grating beam combiners. The instrument had been used to measure Cd, Cr, Cu and Pb. Further details of these systems are awaited with interest. In their application 723 J. Anal. At. Spectrom., 1999, 14, 717-7813. Precision was improved by using an alignment by ICP-MS, were statistically equivalent to the best-estimate of the W-coil atomizer, Parsons et al.120 diluted blood samples reliable based on the agreement of two or more laboratories.1+4 with a modifier containing NH4H2PO4, Triton X-100 The atomic abundances of 208Pb, 207Pb and 206Pb, determined and HNO device to guide the micropipette into the atomizer for injection. It was possible to achieve an LOD of 10-20 mg l-1 and results TIMS values. The preparation of serum materials for interlaboratory comparison of selenium measurement was described on blood RMs were encouraging.In an investigation of by Morisi et al.133 Bovine serum was spiked with H SeO3 to chemical modifiers for the determination of Pb in digests of give five pools of serum with Se in the range 26-1202 mg l-1. blood and hair with the W-coil atomizer, Bruhn et al.121 found Unspiked materials suitable for speciation studies were prethat, of NH4H2PO4, (NH4)2HPO4 and Pd, the best accuracy pared by mixing bovine and horse serum with Se concenand precision was obtained with Pd.Sensitivity was improved trations of 25 and 150 mg l-1, respectively. by including a cooldown step prior to atomization. Results on hair and blood CRMs gave results in agreement with the 1.6 Hair and nail analysis certified values. The mean precision was 5.0% RSD and the LOD was 1 mg l-1. For Cd122 and Cr,123 the best modifier was again Pd, which eliminated chemical interferences, reduced the background and increased the lifetime of the tube by 20% in the case of Cd determination.An LOD of 0.05 mg l-1 was found for Cd. Kitigawa et al.124 have modified their vertical column atomizer to allow temperatures above 2400 °C to be maintained. The modifications were: the use of an open column of glassy C rather than a packed column; suspension of the sample cup with a glassy C rod rather than a W wire; a lower graphite electrode assembly to protect the glassy C column when expanding; and the use of a Si controlled rectifier to give higher power outputs.As in previous designs, determination could be made free of background and this was illustrated by the determination of Mn in a bovine liver CRM. The simplicity and reliability of the flame atomizer have encouraged many to explore ways of increasing its sensitivity. The intriguing combination of ETV with FAAS was investigated by Yoo et al.125 A Ta-filament atomizer was used to introduce 10 ml of blood or urine into a flame for the determination of Al, Ca, Cd, K, Na, Pb and Zn.Publication of results by this technique will be of particular interest. Liang and Lin126 used a slotted quartz tube to increase the sensitivity for the determination of Pb in herbal medicines by FAAS. Presputtering the tube with a solution of vanadate produced an orange-red coating and prevented the deposition of sodium salts and C on the surface of the tube, thus prolonging its lifetime.The high sensitivity achievable by laser-excited atomic fluorescence spectrometry with electrothermal atomization has been exploited by Winefordner's group to measure Ge127 and Pt128 in biological samples. A dye laser pumped by a Cu vapour laser gave excitation at 269.13 nm for Ge and 270.24 nm for Pt. The LODs were 1.1 pg for Ge and 50 fg for Pt. For Pt,127 wall atomization was preferred to platform atomization as black body radiation from the platform aVected the reproducibility of the results.Despite a high dilution of blood (1+24) with H2O, interferences were still present and standard additions calibration had to be used, whereas for urine, a dilution of 1+4 removed interferences, allowing calibration with aqueous standards. 1.5 Reference materials and quality assurance An overview by Byrialsen et al.129 of trends in quality assurance in the determination of metals in clinical samples highlighted the introduction of the concepts of traceability and uncertainty.Haraguchi et al.130 reported on an interlaboratory comparison of results in the determination of 11 elements in human plasma. Of these elements, most variation was seen in the results obtained for Se by various techniques. The trace element concentrations and Pb isotope composition in NIST SRM 1400 Bone Ash were measured by quadrupole and magnetic sector ICP-MS (4 laboratories) and TIMS (2 laboratories) in a study reported by Hinners et al.131,132 Results for 42 analytes were listed, 26 of which were considered to be 724 J.Anal. At. Spectrom., 1999, 14, 717-781 Kruse-Jarres134 has stressed the limitations of hair analysis, indicating that hair acts as a kind of ion-exchanger with small superficial holes and fissures operating as ports of entry. Thus hair contains elements both of exogenous and endogenous origin, which cannot be diVerentiated by prior sample pretreatment.Imprecision in determination of hair samples of nearly identical origin is 30% or more. He concludes that hair analysis is unsuitable for the detection of clinically relevant deficiencies of essential trace elements and only has a limited role in assessment of exposure to heavy metals. Evidence for this can be seen in the study by Ali et al.135 on hair levels of Ca, Cr, Cu, Fe, Mn, Ni, Pb and Zn in malnourished and healthy children in Bangladesh.Despite the lack of nourishment, concentrations in the hair of those children showed no significant diVerence from the results for the healthy children. Measurements were made by PIXE after grinding the hair and forming a pellet. However, there are always reports of positive findings from studies involving hair analysis for the essential elements. Hussein et al.136 reported that hair Zn concentrations in Egyptian pre-school children corresponded to the children's growth rate, nutritional and socio-economic status.The average dietary intake of Zn was 3.8 and 3.6 mg for boys and below 70 mg g-1, their normal cut-oV value. For men with girls, respectively, and 13% of the children had a hair Zn alopecia, Jin et al.137 found hair Mn and Zn concentrations lower than those for healthy men but Cu concentrations were higher. Both of the latter two studies used FAAS for measurement. In an analysis of data on hair, sweat and serum Cr concentrations measured by ETAAS in their laboratory from 40 872 patients over the years, Davies et al.138 found that these measurements were significantly inversely related to age and that hair, sweat and serum measurements correlated well with each other.This apparent decrease in Cr status with age was discussed in terms of increased incidence of diabetes mellitus and heart disease with age. In a study of 13 members of the Italian expedition to the Antarctica, Caroli et al.139 measured, by ICP-AES, concentrations of Ca, Cu, Fe, Mg and Zn in hair samples taken before departure and on return.While Ca 21.7 mg g-1 before to 11.9 mg g-1 after. Increases were seen and Fe showed no changes, Mg dropped from a mean of for Cu (13.6-16.2 mg g-1) and Zn (174-217 mg g-1). The authors concluded that the changes were a result of stress, irritability and anxiety and could be corrected for by a change of diet. In view of the comments by Kruse-Jarres (see above), the wisdom of recommending diet changes based only on changes in hair concentration seems questionable. The relatively high Sb concentrations in the hair of infants, which appeared to support a hypothesis that sudden infant death syndrome (or 'cot death') was a result of SbH3 generated by bacteria from Sb in fire-retardants in mattresses, is discussed with other investigations in section 1.9.2.Yun and Li140 described their work in establishing values for the rare earth elements in the Chinese CRM, GBW09101 Human Hair, by ICP-MS.Dry ashing and microwave digestion were shown to be suitable methods for sample preparation.The results of analysis of vegetable CRMs and spiked samples showed accuracy which was confirmed by comparison with data from other laboratories. The concentrations found were generally more than 10 times the LOQ of the method. Further results have appeared from the ongoing Seychelles study of prenatal methylmercury exposure from a high fish hair mercury concentrations of the mothers was 5.8 mg g-1 diet on developmental milestones in children.141 The median measured by AAS.No association was found between the ages at which the children walked and talked and the mother's hair Hg concentration. High hair Hg concentrations were also found in a study of mothers and infants from the Amazon basin, Brazil.142 Total mercury was determined by CVAAS after alkaline digestion.The median hair Hg concentration was 14.1 mg g-1 in non-Indian mothers and 8.30 mg g-1 in Indian mothers. A statistically significant correlation between mother's and her infant's hair was only found for non-Indians, whereas a correlation between length of breast feeding and the infant's hair Hg concentration was only significant for Indians. 1.7 Drugs and pharmaceuticals Identification of the source of drugs by their trace element fingerprint is not a new idea, but the use of laser-ablation ICP-MS to provide this information is novel.Watling61 showed that this oVered a quick, convenient and relatively contamination-free method.Ground cannabis was pressed into a pellet and irradiated with an Nd5YAG laser. Ratios between the signals from the elements obviated the variability between diVerent ablation events and were conveniently presented as ternary ratio per cent discriminant diagrams expressing the relationship between three particular elements.The technique had been used to identify the source of cannabis in Australia from the relationship of the element patterns to the geological environment. Indirect methods of determination of pharmaceuticals through complexing with metal ions continue to be developed. Issa and Mohamed143 determined antazoline, hydralazine, amiloride and thiamine hydrochlorides and quinine sulfate after formation of insoluble ion-association complexes with [Cd(SCN)4]2- or [Zn(SCN)4]2-.The excess Cd or Zn was determined by FAAS or DCP-AES. El-Brashy and Al-Ghannam144 used the reaction with FeII or HgII to determine histidine by FAAS. A method for the determination of cinnarazine in pharmaceutical preparations by FAAS after com- 4]2- was shown by Hassan et al.145 plexation with [Co(SCN) to give results in good agreement with results obtained by molecular spectrophotometry and potentiometry.Interferences in the determination of Bi in prescription medicines by HGAAS were successfully overcome by masking with a solution of thiourea (0.2% m/v) and KI (10% m/v) in a method developed by Cadore et al.146 The same method could be used for determination of Bi in urine to follow the kinetics of excretion of Bi. The mineral composition (Ca, Cr, Mg, Mn and Zn) of 3 multi-vitamin and mineral tablets was determined by ICPAES after microwave digestion by Krampitz and Barnes.147 When open-vessel digestion with HNO -HCl was used, low results were obtained.New regulations in California, USA, require Pb exposure to be limited to 0.5 mg per day. High intake of Ca supplements taken, for example, for osteoporosis can result in this being exceeded and, as a consequence, there has been interest in measuring Pb concentrations in Ca supplements. As an LOD of below 0.5 mg g-1 has been stipulated, ICP-MS has been the method of choice.148-150 Sharma et al.150 used high pressure digestion with HNO3, while Bakowska149 found it suYcient to dissolve the pills (calcium carbonate or calcium chelates) in 3 concentrated HNO and then dilute with de-ionized water.Highest Pb concentrations (8-28 mg g-1) were found in supplements from natural sources (dolomite or bone meal ).150 Isotope ratios were also explored as a way of identifying the source of the product.1.8 Marine and freshwater biology 4 Moreton and Delves151 found that in the presence of NH4OH and (NH4)2H2EDTA, wash-out times in the determination of mercury by ICP-MS with conventional nebulization were reduced to less than 3 min. Applying this to the determination of Hg in fish, they digested fish tissue with HNO3 at room temperature for 24 h. The digests were made alkaline by careful addition of 6 M NH4OH and then diluted further with NH OH-(NH4)2H2EDTA solution.Levels of Hg in fish from the Tapagos valley, Brazil, were in the range 0.009-2.58 mg kg-1 (wet mass), and from this and from results on blood and urine Hg concentrations of the inhabitants, it was concluded that the levels of Hg in fish were a more general health problem than just for the gold miners who were directly involved in the use of Hg. Total Hg could be determined by FI-CVAAS at a rate of 100 measurements h-1 with a method developed by Tao et al.16 Samples were digested with TMAH for 30 min. Organic Hg was converted to inorganic Hg by 0.1 mg l-1.The method was validated on marine biological addition of 0.2% KMnO4 in the FI system. The LOD was CRMs. Organic Hg in fish was determined by Scerbo and Barghigiani152 after extraction with toluene. The Hg in the extract was back-extracted with cysteine solution and the Hg determined on an automatic Hg analyser, the AMA 254, which uses controlled pyrolysis and amalgamation to collect Hg for subsequent determination by AAS.More complete information is obtained by a GC separation technique, as demonstrated by Gerbersmann et al.153 They used a rapid microwaveassisted dissolution technique with TMAH. Two approaches were then used for analysis by GC-AES. In the first, derivatives formed by ethylation with sodium tetraethylborate were extracted into hexane and injected into a cooled injection system.In the second, hydrides were generated with NaBH4 and introduced with a purge-and-trap injector. Both CRM but the former gave the lower LOD (3.0 pg g-1 versus approaches gave satisfactory results on a freeze-dried tuna 12.5 pg g-1). Haraguchi's review in Japanese154 covers methods for the 4 determination of organotins in environmental and biological samples and deals with the eVects on the human endocrine system of consumption of food contaminated with these compounds.Rapid one-step extraction and derivatization of organotins from fish tissue was developed by Schmitt et al.155 In this, 0.1-0.2 g of tissue was microwaved at 40W power with acetic acid, nonane and NaBH solution for 3 min. The organic phase was analysed by GC-AES after cleanup on an alumina column. The method, validated on a fish tissue CRM, was applied to the analysis of mussels and sea-urchin eggs. Abnormal arsenic concentrations were found in fish from 4 3 Lake Usonko, a naturally acidified lake in Japan, by Takatsu and Uchiumi.156 The fish tissues were digested with HNO -HClO and the As determined by HG-ICP-AES. The accumulation of As, concentrated mainly in the eye tissues of the fish, was thought to be related to the high As and low P content of the lake water.Trace amounts of vanadyl porphyrin in marine mussels were determined by Rivaro and Frache157 using HPLC with ICP-AE and UV detection.It was possible to detect this compound in mussels accumulating high concentrations of V from their environment. A method for the determination of Cd in mussels was developed by Enriquez-Dominguez et al.158 After microwave digestion with HNO3, the Cd was collected at pH 4.5 725 J. Anal. At. Spectrom., 1999, 14, 717-781on a column of poly(aminophosphonic acid) resin in an FI system. Elution with 0.5 M HCl into a flame AA spectrometer allowed the determination of Cd down to an LOD of 0.56 mg l-1.Forty samples h-1 could be analysed in this way. The preparation and certification of Plaice and Shrimp RMs by the National Food Agency, Denmark, was described by Larsen et al.159 Total trace elements were determined by ETAAS, CVAAS, ICP-MS and ID-ICP-MS. Certified values were established for As, Cd and Hg in the Shrimp RM and As, Hg and Se in the Plaice RM. Indicative values for arsenobetaine amd tetramethylarsonium ion were found by cation-exchange HPLC coupled to ICP-MS as detector. The eVects of pollution from a disused Ni mine and smelter on fish in the river Otra, South Norway, were investigated by Brotheridge et al.160 The liver and muscle of brown trout, sampled from several sites along the river, were digested with HNO3 and Co, Cu, Ni and Zn determined by AAS and/or AES. Concentrations were highest nearest the mine and gradually decreased down river.1.9 Progress for individual elements 1.9.1 Aluminium.Several diVerent approaches have been The use of AMS for measurement of 26Al in biological samples continues to be well represented. King et al.89 described sample preparation methods for the determination of 26Al in a variety of clinical samples by AMS. Following addition of 27Al carrier, blood and soft tissues were acid digested and Al selectively extracted with quinolin-8-ol. The complex was ignited in air to form Al2O3.Urine Al was precipitated with phosphate, extracted with acetylacetone and back extracted into HNO for ignition with air to form Al The method gave an absolute 3 LOD of 10-18 g. Kislinger 2O3. et al.84 studied Al biokinetics in renal failure patients by AMS, following administration of a 26Al tracer, and developed a pharmacokinetic model to describe the experimental data. They observed that Al was mainly excreted from plasma citrate into urine and from plasma transferrin into faeces via the liver.They also noted that the size of two peripheral Al compartments were diVerent in renal failure patients compared with controls. Barker and colleagues88 used AMS to study the 4, followed by solid phase extrac- role of transferrin in Al uptake and distribution in hypo- 3 taken to overcome matrix interference on the determination of Al in serum by ETAAS. Chen et al.161 used microwave digestion with HNO -HClO tion with 1% quinolin-8-ol on an XAD chromatographic transferrinaemic mice and mice treated with antibodies to the column.The bound AlIII was eluted with 1 M HCl and determined by ETAAS. An LOD of 0.3 mg l-1 could be achieved if transferrin receptor. They observed that non-transferrin mediated Al transport occurred in many tissues and concluded careful attention was paid to determination of the analytical that transport by citrate and low molecular mass complexes blank.Bu and Sun162 used a molybdenum coated tube and a may be important. Uptake of Al was greatest in bone and K2Cr2O7-Triton X-100 chemical modifier for the direct deter- least in spleen and brain. Yumoto et al.87 used AMS to mination of Al in serum of cancer patients. An absolute LOD of 21 pg was reported. The graphite tube was pre-coated by injection and atomization of 50 ml of a 10% ammonium molybdate solution. A sample deproteinization procedure by microwave digestion with 1% TCA was described by Bohrer et al.8 Low acid concentrations produced complete protein reduction with minimal dilution (1+1 v/v).The deproteinization method gave results in good agreement with a direct Kisters et al.166 used AAS to determine Al levels in lymphodilution method, but sample deproteinization allowed a simpler and faster temperature programme to be used without matrix interferences. Burden et al.9 investigated optimum collection and sample preparation procedures for the determination of Al and other trace elements in urine samples by ICP-AES.They noted that precipitated salts could contain Al and, therefore, these were redissolved for accurate quantitation of Al. Satisfactory results were obtained by standard additions or matrix matched calibration but aqueous calibration was inappropriate. Muniz and colleagues71 described a comprehensive evaluation of ICP-MS and ETAAS methods for the determination of Al in serum.They obtained better LODs with HR-ICP-MS (0.35 mg l-1) and Q-ICP-MS (0.85 mg l-1) compared with ETAAS (2 mg l-1). Matrix eVects were overcome by simple dilution (1+4 v/v) with H2O for Q-ICP-MS, but were still evident at a 1+9 dilution with the high resolution instrument. These were corrected for with Be and Sc internal standards. Analysis of serum samples from healthy subjects showed that HR-ICP-MS was necessary for the determination of Al levels as they were at or below the LOD of 0.35 mg l-1.The same research group163 also coupled FPLC with HR-ICP-MS to study the speciation of Al in human serum from unexposed individuals and uremic patients. Serum proteins were separated by anion exchange chromatography with a CH3COONH4 gradient at pH 7.4. The results indicated that transferrin was the only significant Al binding protein in uremic serum. The combination of HPLC with oV-line ETAAS was used by Roellin et al.164,165 to study Al binding to serum proteins.726 J. Anal. At. Spectrom., 1999, 14, 717-781 Serum was ultrafiltered and chromatographed by diVerent size exclusion techniques, using a phosphate-NaCl mobile phase. Both high and low molecular mass complexes were identified. The authors applied the method to the analysis of serum from occupationally exposed workers and observed that increased Al exposure was associated with an increase in the high molecular mass fraction.investigate Al uptake and accumulation in rat brains, five days following injection of 26Al. A considerable amount was measured in the brain and the levels remained constant for over 270 d. Blood Al levels, however, fell significantly after 75 d. The authors considered the data supported the hypothesis of Al accumulation in the brain being a causative agent for Alzheimer's disease. cytes from dialysis patients. T lymphocytes were separated by rosetting with sheep red blood cells.Intracellular Al concentrations were increased in B lymphocytes of dialysis patients (2.92 versus 1.44 mg g-1 protein) but not in T lymphocytes (2.24 versus 2.18 mg g-1 protein). Golub et al.167 measured Al in tissues of guinea pig fetuses by ETAAS in a study examining the distribution of Al in the central nervous system during development. They noted that Al distribution did not follow that of essential cations.1.9.2 Antimony. A sensitive ETAAS method, using a transversely heated furnace with STPF conditions and longitudinal Zeeman eVect background correction, was presented by of mg l-1 levels of Sb in tap water, serum and whole blood. Subramanian et al.168 for the quantitative determination The use of a Pd-Mg(NO3)2 chemical modifier eliminated the need for standard additions calibration and an LOD of 2.6 mg l-1 was reported. The method was evaluated by analysis of leachate tap water from plumbing systems containing Sn-Sb soldering and determination of Sb in blood samples taken from rats given Sb supplemented drinking water.A coupled HPLC-ICP-MS method was described by Koch et al.169 for the determination of Sb species in biological samples. Both ion exchange and ion pairing techniques were used to separate SbIV, SbV, trimethylantimony dichloride and phenylstibonic acid. The hypothesis that sudden infant death syndrome ('cot death') was a result of SbH3 generated by bacteria from Sb in fire retardants in mattresses was supposedly supported by high hair Sb levels.This hypothesis, publicized3-H2O2. The low LODs for connected to a quartz cell in an air-C2H2 flame. Limits of by a British TV programme 'The Cook Report' led to an extensive investigation by an expert committee appointed by the UK Government. In a large scale study, commissioned by the committee and reported by Keenan et al.,170 the concentrations of Sb in umbilical cord and matched infant hair samples were determined by HG-AFS, following closed vessel microwave digestion with HNO the methods (1 ng g-1 for cord and 3 ng g-1 for hair) enabled Sb to be quantified in 90% of both sample types.Mean and median Sb values of 27.7 and 9 ng g-1 were determined in cord blood whilst for hair the mean and median values were 826 and 300 ng g-1, respectively. Concentration distributions were heavily skewed in both sample types.The investigators found no correlation between cord Sb and hair Sb in matched samples. They considered the likely origin of the Sb to be environmental via food, airborne dust and water. 1.9.3 Arsenic. There has been a wealth of publications on 2 to the nebulizer gas flow or the addition of ethanol to were determined by HPLC-ICP-MS, using an anion exchange the determination of As in biological materials in this review period.Although it is recognized that As speciation is very important for monitoring occupational and environmental exposure, methods for the determination of total As continue to be reported. Pozeban et al.64 found Ir to be a suitable chemical modifier for the determination of As, Se and Pb in urine by ETV-ICP-MS, allowing pyrolysis temperatures up to 1300 °C. The modifier was injected and atomized before the sample to form an Ir coating, as this removed contamination from the modifier and lowered blank signals.Amarasiriwardena et al.171 compared diVerent interference correction methods for the determination of As in urine by solution nebulization ICP-MS. The authors claimed that accurate determination of As could be achieved by the addition of 3% N the samples, using Te as an internal standard. The methods were evaluated by analysis of NIST SRM 2670 (toxic elements in urine). The majority of clinical studies, however, have described methods for the speciation of As in order to distinguish metabolites of inorganic As from less toxic dietary forms.Benramdane et al.172 selectively extracted toxic As species from dietary As prior to analysis by ETAAS. Optimum extraction was achieved by reduction with KI-Na hypophosphite to form 'iodide arsines', which were extracted into toluene and back extracted into NaOH. Recoveries of 100% were reported with an LOQ of 7 mg l-1.Guo et al.173 observed low recoveries of toxicologically relevant As species in urine when KI-ascorbic acid was used as a pre-reductant for FI-HGAAS. They obtained optimum recoveries for AsIII, AsV, dimethylarsinic acid (DMA) and monomethyarsonic acid 4 (MMA) following reaction of diluted urine with 1% cysteine in 0.03 M-1 HCl. A high NaBH concentration (0.5%) and longer residence times in the HG system eliminated residual kinetic eVects in the formation of hydrides from the diVerent As species.Sperling et al.25,174 examined in depth diVerent chromatographic and HG conditions for the determination of As species 1.9.4 Bismuth. A HGAAS method for the determination of in urine by HPLC-HGAAS. The influence of stationary and Bi in plasma was developed by Shen et al.183 to monitor mobile phases, conditions for complexation and design of gas- plasma Bi concentrations in a patient administered a Bi liquid separator were evaluated.The authors found that containing anti-ulcer compound. Heparinized plasma was optimum separation was achieved on a strong anion exchange digested with HNO -HClO4, evaporated to near dryness and column with a phosphate buVer. On line photo-oxidation with redissolved in HCl for hydride generation. Standard additions K2S2O8-NaOH oxidized all As species to AsV for HG. calibration was used for quantitation and the LOD was Burguera et al.24 described an on-line FI-HGAAS method for 1 mg l-1.The authors compared the HG method with an the separation and determination of AsIII, AsV, DMA and extraction method and obtained comparable results but with MMA in urine from occupationally exposed workers. improved analytical precision. Cadore and colleagues146 also Separation was achieved on a combined anion-cation described a HGAAS method for the determination of Bi in exchange column. Arsenite was not retained on the column both urine and prescription medicines.In their approach and MMA, AsV and DMA were eluted with 0.006 M TCA, samples were acid digested and potential interferences from 0.2 M TCA and 5 M NH4OH, respectively. The eluted fractions metal ions were overcome by the addition of a thiourea-KI were reacted with 15% KI-5% ascorbic acid and 1% NaBH4. Ng and colleagues175 developed an HGAAS method with cold trapping for the determination of As metabolites in urine from exposed workers.The hydride trapping system consisted of two water traps and a chromatographic column of 10% polyphenyl ether on 100-200 mesh Chromosorb, and was detection for the species were 1, 1.3 and 3 ng for AsIII+AsV, DMA and MMA, respectively. The authors observed an increased excretion of AsIII+AsV in exposed workers and hypothesized that methylation of inorganic As was compromised. Ma et al.176,177 described a fast HPLC method for separating As species in urine with detection by HGAFS.Fast separation was achieved by ion-pair chromatography on a short column, giving baseline separation of four As species in minutes. The method was validated by analysis of CRMs and was used to study the eVect of dietary As intake on urinary As excretion. The authors found that arsenosugars in seaweed were metabolized to DMA. Speciation methods have been applied to studies of As metabolism in various study populations. Zhang et al.178 measured As species in serum from groups of uraemic patients.Low molecular weight species were quantified by ion exchange chromatography-HGAAS, whilst high molecular weight species were separated by SEC or aYnity chromatography, digested and quantified oV-line by HGAAS. All groups of uraemic patients had increased serum As levels compared with controls and the major As species were identified as DMA and arsenobetaine. Only inorganic As species were bound to proteins.Kavanagh and colleagues179 reported findings of a pilot study of urine As species in residents from a heavily As contaminated area of the United Kingdom. Arsenic species column with a phosphate mobile phase coupled to a high pressure nebulizer. They found an increased excretion of total As and an increase in the percentage of inorganic As, indicating chronic exposure to inorganic As. Del Razo et al.180 also found a significant increase in inorganic As species, with a corresponding decrease in methylated species, in the urine of adults chronically exposed to As in drinking water.The group used HPLC-HGAAS for determination of As species. They noted that individuals with cutaneous signs of arsenicism had significantly higher levels of MMA and inorganic As and hypothesized that chronic exposure to As led to a decrease in methylating capacity. Hsueh et al.181 determined As species in urine from exposed populations drinking As contaminated water, using HPLC-HGAAS.They observed that individuals a cumulative As exposure>20 mg l-1 year-1 had an increased with elevated proportions of MMA in their urine (>27%) and multivariate odds ratio of 20.9 for developing skin cancer. In a study of potential biomarkers for cardiovascular disease, Jensen et al.182 found increased levels of glycosylated Hb in As exposed workers which was associated with an increase in systolic blood pressure.3 727 J. Anal. At. Spectrom., 1999, 14, 717-781masking solution. Improved sensitivity and precision were achieved with a low residence volume gas-liquid separator, which was treated with HF and chlorodimethylsilane to eliminate reactive sites on the glass surface, giving an LOD of 0.3 mg l-1 and analytical precision of 4.7% RSD. 1.9.5 Boron. In a comprehensive review containing 175 references, Sah and Brown184 examined the advantages and disadvantages of diVerent analytical methods for the quantitative determination of B in biological samples.Whilst spectrophotometric methods, based on colorimetry or fluorimetry, were most commonly used, they were noted to suVer from numerous interferences and to have low analytical sensitivity and poor precision. Amongst nuclear methods, only alpha spectrometry was considered to be of practical use. The development of ICP-MS methods overcame many of the drawbacks of spectrophotometric methods and oVered the advantage of B isotope measurements for both biological tracer studies and isotope fingerprinting.The authors considered ICP-MS to be the method of choice for accurate B determination in biological matrices. A simple automated method, which the authors Anderson and Tschirgi185 described as benign-by-design, was developed for the determination of B in microsamples of biological tissue. The acid digestion of 100-400 mg tissue samples minimized waste and eliminated the need for toxic reagents previously used for B determinations. The method was exhaustively evaluated by more than 250 determinations of B in a variety of CRMs and recoveries between 82% and 104% were reported. Price et al.186 described a high temperature alkali ashing sample preparation procedure for the determination of B in whole blood by ICP-AES.The method was used to study blood B levels in pregnant rats administered B throughout gestation.The researchers observed that blood B was positively correlated with both maternal dietary B and with embryo/fetal toxicity. They reported a no observed adverse eVect level of 10 mg kg-1 d-1 for developmental toxicity. 1.9.6 Bromine. Using WD-XRF, Olszowy et al.187 determined a reference interval for bromide in human blood. Blood samples were dispensed to a depth of 20 mm into aluminium sample cups, with a Mylar film base, which were placed in the spectrometer for quantitation of Br.Correction for matrix diVerences between samples and aqueous calibration standards centration of 5.3 mg l-1 and a reference interval of was made using the Compton scatter line. A mean Br con- 2.5-11.7 mg l-1 was determined. The authors noted that blood Br levels tended to be higher in older people and women. Zaicheck105 used EDXRF to determine Br in diVerent body fluids in order to estimate the extracellular water space.Volunteers were given an oral dose of 14 mg kg-1 stable Br as an NaBr solution. Blood, urine and saliva samples were collected 6 h and 243 h after administration and the Br concentration determined by XRF, using a 109Cd source and 25 mm2 Si(Li) detector. The method gave results that were in good agreement with a conventional 82Br measurement method. 1.9.7 Cadmium. The eVectiveness of diVerent chemical modifiers for the determination of Cd in acid digested hair and blood samples by ETAAS, using a tungsten coil atomizer, was examined by Bruhn et al.122 Optimum performance was obtained with a Pd modifier, which both reduced background interferences and improved atomizer lifetime.Several groups have described diVerent sample introduction procedures for the quantitative determination of Cd in biological matrices by ICP-MS. Goenaga Infante and colleagues188 compared an FI-ICP-MS method with in-situ trapping ETAAS for the determination of Cd in urine at baseline levels.Both methods were based on the vesicle assisted HG procedure 728 J. Anal. At. Spectrom., 1999, 14, 717-781 3 previously described by this group.189 Simple dilution 1+1 v/v with H2O minimized contamination risks. Aqueous calibration gave satisfactory results and the use of an antifoam agent for HG enabled a throughput of 20 h-1 for the FI-ICP-MS method. Absolute limits of detection were 7 pg and 14 pg for ICP-MS and ETAAS, respectively, and both methods were validated by analysis of a urine CRM. In the study described by de Boer et al.190 for the direct determination of Cd in urine by ICP-MS, various approaches to overcome or correct for matrix interferences were investigated.The eVect of nonspectral interferences from NaCl were corrected for by internal standardization with Rh and the use of a correction equation, calculated from a series of recovery experiments. Sample dilution 1+9 v/v with 1% HNO and measurement at the 114Cd mass were found to give satisfactory results for CRMs.A particularly sensitive FAAS method for the determination3 of urine Cd species separated by HPLC was described by Chang et al.191 A novel thermospray nebulizer design produced a 104 fold improvement in sensitivity over conventional flame nebulization. An LOD of 3-7 ng l-1 was reported for a 100 ml urine sample chromatographed on a 5 cm C8 column, using 2O as the mobile phase at a flow rate of 2.5 ml min-1. Lee Het al.67 developed an ETV-ID-ICP-MS method for the direct determination of both Cd and Pb in urine.A 1% HNO solution was used as a chemical modifier and a low vaporization temperature of 1000 °C employed to separate the analytes from matrix components. Isotope ratios were calculated from peak area measurements of each isotope. The ID method was evaluated by analysis of NIST SRM 2670, freeze dried urine, and comparison with standard additions calibration.Detection limits of 0.02 mg l-1 and 0.005 mg l-1 were reported for Cd and Pb, respectively. In an eVort to better understand the bioavailability of dietary Cd, Langford et al.192,193 developed a coupled SEC-ICP-MS method to characterize Cd species in human ileostomy fluid. Ileostomy patients were given a porridge meal spiked with 106Cd and ileostomy fluid collections were made at various times over a 24 h period.The samples were ultracentrifuged and separated by SEC with ICP-MS detection of the 106Cd and 111Cd isotopes together with Cu, Pb and Zn. Both Cd isotopes eluted in a single peak corresponding to a molecular weight of 15 kDa. The Cd peak was also associated with Zn but not Cu or Pb. The portable battery powered tungsten coil AA spectrometer described in the previous Clinical update6 was used by However, the reported LOD of 3 mg l-1 is considered by this Batchelor et al.194 to determine Cd in urine, soil and tap water.reviewer to be insuYciently sensitive to monitor low-level currently reported to be well below 1 mg l-1. A detailed environmental exposure as 'normal' levels of Cd in urine are investigation of Cd exposures in battery workers was reported by Boerjesson et al.98 in which in-vivo measurements of liver and kidney Cd by XRF were compared with Cd in blood and urine and cumulative air Cd measurements. Several correlations were observed, most notably, kidney Cd with urine Cd (r=0.70), liver Cd with urine Cd (r=0.58) and kidney Cd with blood Cd (r=0.49). No correlation between cumulative air Cd exposure and kidney or liver Cd were observed.The authors considered that uncertainties in both the estimates of individual cumulative exposures and in the in-vivo measurements may have contributed to the lack of correlation. Finally, an ETAAS method was developed by Bermejo- Barrera et al.195 for the determination of ultra-trace levels of Cd in cocaine and heroin samples.A 0.2% (NH4)2HPO4 was found to be the optimum chemical modifier and standard additions calibration was necessary to overcome matrix interferences. Cadmium levels in cocaine ranged from 5-45.6 mg kg-1 and in heroin ranged from 10-192 mg kg-1. The authors considered the measurement of Cd in both drugsto be one of several parameters which could help elucidate the geographical provenance of the illicit drugs. 1.9.8 Calcium.The high sensitivity of AMS to 41Ca was used by Freeman et al.90 to study the kinetics of Ca in the human. The authors considered that a 41Ca tracer could be administered to achieve a quasi-steady-state situation, and thus bone resorption could be quantitatively measured directly. They tested the hypothesis by comparison with a conventional stable isotope study and obtained good agreement between the methods.To study the role of Ca in the pathogenesis of pre-eclampsia, Kisters et al.196 determined plasma and erythrocyte membrane Ca levels in healthy pregnant and pre-eclamptic women by AAS. A decrease in plasma Ca levels (1.98 versus 2.43 mM) and corresponding increase in membrane Ca content (1.21 versus 0.73 mM) was found in pre-eclamptic women, from which the authors concluded that the condition was associated with a disturbed Ca homeostasis.1.9.9 Chromium. The quantitative determination of Cr in biological matrices by ICP-MS is aVected by interferences on the most abundant isotope 52Cr by the polyatomic species 40Ar12C and 35Cl16OH and interference from 54Fe on 53Cr and 54Cr. The severity of these interferences normally precludes the determination of Cr in blood and urine by ICP-MS, using conventional pneumatic nebulization. In the method described by Gupta and Barnes,197 sample introduction by ETV was used to minimize interferences from the organic matrix.Replacement of the graphite furnace with a tungsten filament removed the interference from 40Ar12C. In a second paper, the same research group51 described a high pressure vapour phase acid digestion procedure, using specially designed microvessels, for the determination of Cr and V in biological matrices by ICP-MS with pneumatic nebulization. The digestion procedure markedly reduced C and Cl polyatomic interferences and the remaining 40Ar12C interference on 52Cr was corrected mathematically. The method was verified by analysis of biological SRMs.Chromium in biological samples is normally determined by ETAAS. With great improvements in ETAAS instrumentation and the recognition and elimination of contamination sources over past decades, the levels of Cr determined in healthy populations have declined markedly. In a recent survey of the Italian population, Apostoli et al.198 reported a reference value of Cr in urine of 0.08 mg l-1 (non-detected-0.24 mg l-1).Measurements were made at three laboratories using similar methodologies and a between laboratory standard deviation of 0.049 mg l-1 was reported. Using multiple regression analysis to study confounding factors, only geographic location and gender were found to have any influence on urine Cr levels. Interestingly, populations residing at coastal locations had the lowest levels.Davies et al.138 reported age related decreases in hair, sweat and serum Cr levels in a study of over 40 000 patients, using ETAAS to determine Cr in all sample matrices. Correlations were found between hair Cr, sweat Cr and serum Cr, leading the authors to consider both hair and sweat to be suitable additional matrices for assessing Cr status. The authors discussed the implication of the age-related decline in Cr on the development of several age-related diseases.In a review of TOF-LMMS, Hachimi et al.199 critically evaluated the potential of the technique for the speciation of Cr and Ni in biological specimens, such as macrophages. 1.9.10 Cobalt. A novel crossed immunoelectrophoresis-LAICP-MS method was developed by Neilsen et al.58 for the identification and quantitation of Co binding proteins in human serum. Serum, enriched with Co, was electrophoresed in two dimensions on an agarose gel.Cobalt associated with proteins was quantified in the gel by LA-ICP-MS and the main binding proteins identified by comparison with Coomassie Blue stained gels. The main Co binding proteins were identified as albumin, a-2 macroglobulin, a-1 and b-1 lipoproteins and haptoglobin. Owing to present limitations in the sensitivity, the authors considered the technique to be most appropriate for clinical studies of metallodrugs containing Au and Pt.1.9.11 Copper and zinc. Two groups evaluated methods to minimize matrix interferences on the determination of Cu and Zn in biological fluids by ICP-MS. Szpunar et al.56 compared cross-flow and FI-direct injection nebulization for introduction of serum and urine samples diluted with 0.05% HNO3. Results obtained for CRMs agreed well with the certified values. In the study by Hsiung et al.49 simple dilution and calibration with an internal standard was found to be satisfactory to minimize polyatomic interferences on the determination of Cu and Zn in urine and serum. Accurate results for CRMs were obtained with measurements of 65Cu and 68Zn isotopes.Kondo et al.200 described a TXRF method for the quantitative determination of Cu and Zn in small (mg) biopsies of bladder cancer tissue. Measurements were made using 11 keV X-rays and calibration was by standard additions. Levels of Cu were significantly lower in carcinoma tissue compared with healthy tissue, whereas Zn levels remained relatively unchanged.The authors concluded the method was suYciently sensitive to monitor time-dependent changes in trace element levels in developing carcinomas. A computer controlled electrodialysis unit was designed by Jacobus et al.201 and coupled to an FI-AAS system for the determination of Cu in multi-vitamin and mineral tablets. The dialysis unit separated and preconcentrated soluble CuII ions from tablet suspensions with an eYciency of 94%, which was a marked improvement on earlier passive dialysis designs.An LOD of 1.4 mg l-1 was reported with a sampling frequency of 14 h-1. HuVer et al.202 compared a microwave digestion procedure with conventional hotplate wet digestion for the determination of Zn in isolated erythrocytes by AAS. The group also described a sample preparation method for the quantitative determination of 70Zn in erythrocytes by ICP-MS which involved solvent extraction with diisopropyl ether and ion exchange chromatography.In a study of trace elements in broncheolar lavage, Harlyk et al.112 used ETAAS to determine Cu and Fe, and FAAS to determine Zn concentrations. External calibration was satisfactory for Zn but quantitative determination of Cu and Fe required standard additions calibration. For all three elements, highest concentrations were determined in the supernatant after ultracentrifugation.High levels of Cu in the lavage fluid were associated with low Zn levels and vice versa. Bibi et al.203 determined Cu levels in electrophoresed protein fractions of serum from healthy and rheumatoid arthritis patients, using AAS. They observed that Cu binding to a-2-globulin was higher in arthritis suVerers but binding to albumin was similar in both healthy and diseased groups. They proposed that the method may be used in diagnosis of the condition.Hallfrisch et al.204 described the results of a study of plasma Cu and Zn levels in healthy, well nourished, American adults which demonstrated the 'antagonism' between the elements. At high levels of Cu intake, a 'quadratic' decline in Zn was observed in both men and women, which was overcome by increasing vitamin C intake. A closer correlation between caeruloplasmin and plasma Cu was observed in younger people than the elderly. A study of Cu levels in human placental tissue, using ETAAS with Zeeman eVect background correction, led Centeno et al.113 to propose that measurement of placental Cu may be a useful marker for the diagnosis of Menkes disease.Artacho et al.205 determined Zn in serum from insti- Zn concentration of 10.49 mmol l-1. The authors found no tutionalized elderly subjects, using ETAAS, and found a mean 729 J. Anal. At. Spectrom., 1999, 14, 717-781significant influence of gender on serum Zn levels and no statistical correlation between serum Zn and dietary Zn intake, indicating that serum Zn may not be a suitable index of Zn status in the elderly.Krivorucho et al.206 determined plasma and erythrocyte levels of Zn, and plasma Cu levels, in iron deficient pregnant women by AAS, to examine the influence of these elements on the progress and outcome of parturition. They found that by giving a Zn supplement as well as standard antianaemic treatment in the third trimester, problems at parturition were reduced and proposed that blood Zn and Cu be monitored to prevent possible obstetric complications.An interesting application of TXRF for the quantitative determination of Zn in alleged bee and snake venom samples was described by Blumelhuber et al.207 Combining the spectroscopic data obtained by FT-IR with the quantitative measurement of Zn in the samples enabled the researchers to identify the sources of the diVerent venoms.1.9.12 Gallium. With continuing advances in ICP-MS instrumentation, methods are being developed for a wider range of ultra-trace elements using the technique. Okazaki et al.208 developed an ICP-MS method for the determination of Ga in serum. They observed that a polyatomic interference from 37Cl16O2 on 69Ga was particularly significant at 1+9 v/v dilutions of serum. To overcome this and the observed matrix suppression from NaCl, measurements were made at the 71Ga mass using Co as an internal standard. 1.9.13 Germanium.A very sensitive LEAFS method for the direct determination of Ge in water and blood samples was developed by Aucelio et al.,127 in which a graphite furnace and L'vov platform were used to improve the eYciency of sample atomization. A number of chemical modifiers were evaluated. Optimum results were achieved with a Cu vapour laser for excitation and fluorescence detection at either 275.45 nm or 326.95 nm.The authors noted that the major source of background noise arose from black body radiation from the graphite furnace. Standard additions calibration was necessary for complex sample matrices containing high concentrations of Cl and an absolute LOD of 1 pg for Ge in whole blood was reported. 1.9.14 Indium. Imai et al.209 used Ni(NO3)2 as a chemical modifier for the determination of In in human milk and infant formula by ETAAS with Zeeman eVect background correction.Both sample types were diluted 1+1 v/v with 2% Triton X-100 and 10 ml volumes injected into the furnace with an equal volume of 5 g l-1 Ni in 1 M HNO the two matrices were 2.5 mg l-1 for milk and 2 mg l-1 for 3. Detection limits for In in infant formula. 1.9.15 Iodine. The eYcacy of a water soluble tertiary amine solution for the quantitative determination of total I in biological samples by ICP-MS was investigated by Gelinas et al.18,210,211 Two diVerent sample pretreatments were examined: simple dispersion into a 10% solution of the tertiary amine matrix or combustion with O2 in a closed or flow through system and collection of the residue into a 5% solution of the tertiary amines.The presence of low levels of I in the amine matrix solution restricted the limit of determination to 10 ng g-1, but use of the solution enabled the simultaneous determination of several other elements of toxicological significance.Good agreement with the certified value for I in six biological CRMs was obtained with the combustion method. Wardley and colleagues212 developed a rapid ICP-MS method, using standard additions calibration, to determine I in urine. An LOD of 0.38 nmol l-1 and precision of 15% at the 3 mmol l-1 level were reported. No age or gender diVerences were observed and the authors identified the potential of the 730 J. Anal. At. Spectrom., 1999, 14, 717-781 method for the diVerential diagnosis of thyroid dysfunction. Nixon et al.213 described a simple method for the determination of I in serum, whole blood and urine by ICP-MS, which was fully compliant with Good Laboratory Practice (GLP).Samples were diluted with 1% TMAH and matrix suppression eVects compensated for with Rh as internal standard. The method was used to establish reference values for a healthy population. The reference values determined were 33-81 mg l-1 for serum, 35-91 mg l-1 for blood and 100-450 mg l-1 for urine.Zaichick and Zaichick214 used both XRF and INAA to study the age dynamics of intrathyroid I. Both concentration and total I content of the thyroid were determined in necropsy samples from 90 subjects aged between 2 and 87. A mean thyroid I concentration of 345 mg g-1 was determined for healthy individuals aged 26-65, whilst significantly higher concentrations were found in both the young and elderly populations.The authors considered that the range of I values observed was suYciently wide to explain the sensitivity of diVerent populations to excesses or deficiencies in I intake. The determination of I by ICP-AES was used by Braselton et al.47 to monitor the plasma clearance of the radiographic contrast agent Iohexol. Samples were deproteinized with TCA- hydroxyamine and centrifuged. The supernatant was directly nebulized without decantation.To correct for interference from P at the 178.276 nm I measurement line, measurement of P was also made at the 214.914 nm line. The researchers found that for domestic pets an iodine dose of 300 mg kg-1, with blood sampling at 3-7 h, was satisfactory for accurate determination of the glomerular filtration rate. Determination of 129I and I isotope ratios in teeth and bones by AMS was made by Cornett et al.91 to assess their suitability as a dosimetric record of exposure to high neutron fluxes and as a method to determine the age of bone samples covering the period 15 million-75 million years before present.Sample treatment involved removal of contaminating soft tissue residue by a beetle colony, after which the samples were dried, ground and pyrolysed to extract 129I. Elevated levels of 129I were determined in samples exposed to fall out from nuclear weapons testing. 1.9.16 Iron. In a very interesting investigation of the chemical interactions between asbestos fibres and cultured lung cells, Seal et al.215 used electron spectroscopy, AAS and ESR to study the behaviour of chrysotile silicates in selected cultures of lung cells.The study findings led the authors to hypothesise that Fe species associated with the asbestos silicates are drawn into the cells where they may generate toxic oxygen species through Fenton-type reactions. Menendez-Fraga and colleagues216 used ETAAS to determine total and ultrafilterable Al and Fe in serum from dialysis patients administered desferrioxamine (DFO).Samples were simply diluted with ultra-pure water for both analytes and aqueous calibration was satisfactory for accurate quantitation. The researchers found that DFO administration 1 h before dialysis increased the ultrafilterable fraction of Al and Fe to 75% of the total, but this value declined to 38% 1 h after infusion. Determination of 54Fe and 57Fe isotopes in serum by ICP-MS with conventional nebulization are severely aVected both by polyatomic interferences from 40Ar14N and 40Ar16OH and by reduced nebulization eYciency caused by high serum protein concentrations.To reduce these interferences, Varnes217 developed a method in which serum samples were deproteinized and introduced into the ICP by ultrasonic nebulization Ar flow rates, an LOD of 1 mg l-1 was reported for both with membrane desolvation.With optimized RF power and Fe isotopes.3 3 1.9.17 Lanthanides. Buseth et al.63,62 developed an ETVICP-MS method in order to achieve the necessary detection limits for quantitative determination of lanthanide elements in various biological matrices. Comparable results were obtained for microwave digested samples and undigested slurried samples. The injection of H2O2 into the graphite furnace prevented the build-up of carbonaceous residue from undigested samples and trifluoromethane was found to be an eVective chemical modifier both for lowering the vaporization temperature and reducing memory eVects.Absolute LODs were in the range 1-20 fg for all lanthanides. Using the method the researchers found detectable levels of all lanthanides in human plasma from 30 healthy volunteers and in liver samples from unexposed rats. To determine La in urine quantitatively, Vicente et al.218 developed a complicated FI-ICP-AES method with oV-line chelation and preconcentration of the analyte.Lanthanum was extracted from 1 l of urine by chelation with quinolin-8-ol into CHCl and back extracted into H2O. The solution was reacted again with ethanolic quinolin-8-ol, preconcentrated onto an Amberlite ion exchange column and LOD of 0.09 ng ml-1 and recoveries of 98-101% were eluted with HNO directly into the nebulizer of the ICP. An reported. 1.9.18 Lead. There continues to be considerable interest in the development of low cost portable instruments for blood Pb analysis.A review of past, present and future developments of low cost tungsten filament atomization spectrometers for the determination of trace elements in biological matrices was presented by Parsons and Slavin.219 The authors focused on the problems in determining Pb in blood using such instruments. The same authors and colleagues120 evaluated a new tungsten filament atomizer for the direct determination of Pb in blood.Blood samples were diluted 1+4 v/v with an NH4H2PO4-HNO3-Triton X-100 chemical modifier and 15 ml volumes were deposited on the filament with a micropipette guided by an alignment device to ensure reproducible deposition. With power controlled heating, detection limits of 1-2 mg l-1 were reported with a precision that satisfied US Centre for Disease Control requirements. Bruhn et al.121 critically evaluated diVerent chemical modifiers for the determination of Pb in acid digests of blood and hair by ETAAS with a non-enclosed tungsten coil atomizer.More accurate and precise results for blood and hair CRMs were obtained with a Pd chemical modifier compared with phosphate modifiers. Djane et al.33 described an interesting approach to sample clean up and pre-concentration for the determination of Pb in urine by ETAAS or FAAS. Lead ions were extracted by proton gradient mass transfer across a supported liquid membrane of 40% di-2-ethylhexylphosphoric acid in kerosene into 1 M HNO and 6 mg l-1 were reported for ETAAS and FAAS, respectively.3. Extraction eYciency was 95% and LODs of 0.1 The method was validated by comparison with results obtained by direct ICP-MS. The method, however, is unlikely to be considered for routine analysis owing to analysis times of up to 45 min per sample. This review period has seen an enormous increase in studies to determine isotope ratios of trace elements in biological samples.A large proportion of the work has focused on Pb isotope measurements. Gwizada et al.80 described a double focusing magnetic sector ICP-MS method for the accurate determination of Pb isotope ratios in biological samples. Samples were digested with acid and spiked with Bi as internal standard. The method gave a measurement precision of<0.1% for 206Pb5204Pb, 207Pb5206Pb and 208Pb5206Pb ratios. Accuracy of the method was validated by comparison with a reference TIMS method.The authors used the method to identify and distinguish household Pb sources leading to elevated blood Pb levels in children. Amarasiriwardena et al.220 determined blood Pb levels and isotope ratios in Ecuadorian children from an area of high Pb use. Blood Pb levels in children exposed to emissions from lead pottery glazing were markedly elevated (43 mg dl-1) and the isotope ratio closely matched that found in soil samples collected close to the kilns.Hall and Zhu221,505 used HPLC coupled to ultrasonic nebulization ICP-MS to study isotope ratios of Pb associated with plasma proteins. Plasma samples were collected from patients who ingested soil samples with a modified Pb isotope ratio in order to study the kinetic parameters of plasma Pb binding. Ingested Pb was initially associated with albumin but transferred to caeruloplasmin, competing with Cu for the binding sites.Hinners et al.131,132 reported results of an inter-laboratory comparison of MS methods for the measurement of Pb isotope ratios in NIST SRM 1400 (bone ash). Lead isotope measurement by TIMS agreed to within 0.09% of previously certified values. Measurement of Pb isotope abundances by ICP-MS diVered by less than 0.17% with the values determined by TIMS. Wooland and colleagues79 used magnetic sector ICP-MS for rapid and precise measurement of Pb isotope ratios in biological materials.Their method measured Pb at six isotope masses, using 209Bi as internal standard, with a total measurement time of 5 min. The method was evaluated by analysis of NIST of 0.013 mg kg-1 and measurement precision for isotope ratios SRMs. Recoveries of 96-101% were reported, with an LOD of 0.07-0.22% RSD. Finally, Franklin et al.222 sequentially administered stable Pb isotopes to monkeys in order to investigate changes in blood Pb during pregnancy.Blood and bone Pb isotope ratios were determined by TIMS. The authors observed that bone Pb mobilization increased in late pregnancy and that there was substantial transplacental transfer of Pb. They estimated that between 7 and 39% of fetal skeletal Pb originated from the maternal skeleton. A 1997 American legal proposition relating to Pb contami- 3 3 nation of calcium supplements included a requirement for monitoring of Pb contamination by ICP-MS.This has resulted in several reports of methods developed for such monitoring. Bakowska149 described the determination of Pb in more than 20 types of calcium supplement by ICP-MS. Samples (0.1 g) were simply dissolved in HNO and diluted 1+49 v/v with H2O. Results were reported to reflect the amount of Pb contamination corresponding to a Ca dose at the recommended daily amount. Ahson et al.15 described an ETAAS method for the direct determination of Pb in calcium supplements using a molybdenum atomizer and thiourea as a chemical modifier.Sample slurries were prepared by ultrasonic agitation. Satisfactory results were obtained with calibration standards prepared in a CaCO solution and the analytical precision improved with smaller particle size. Results obtained from slurry samples agreed well with results from digested samples. Sharma and Barnes150 measured Pb in calcium supplements by ICP-MS following acid digestion under high temperature and pressure.Calibration standards were prepared in CaCO3 and the method was validated by analysis of animal bone SRM. Lead levels in natural supplements ranged from 8-28 mg g-1 Ca, whilst refined supplements had much lower levels (1-2 mg g-1 Ca). Zerwekh and Pak223 used ETAAS to measure bone Pb in osteoporosis patients given calcium supplements. The results suggested to the authors that Ca citrate supplementation did not pose a Pb intoxication risk. Rosen96 presented a review of L-XRF methods used to estimate bone Pb levels in adults and children from lead contaminated suburbs.Tantari et al.224 used XRF to measure finger bone Pb levels in occupationally exposed workers. Elevated bone Pb levels were observed even though blood Pb levels indicated very low Pb exposure. Ao et al.94 described a Monte Carlo statistical approach for correction of interference 731 J.Anal. At. Spectrom., 1999, 14, 717-781from Compton scatter in the in-vivo determination of Pb by L-XRF. 1.9.19 Magnesium. A review of current methodologies used 3 in clinical chemistry laboratories for the determination of Mg was presented by Ryan and Barbour.225 The article described the analytical precision that could be achieved by the diVerent techniques and highlighted the generally poor performance of laboratories in external quality assessment programmes.The article also discussed the relevance of methods for the determination of intracellular Mg. Pruvost et al.77 compared electronic impact mass spectrometry (EI-MS) and ICP-MS for the determination of Mg isotope ratios, using various statistical approaches to evaluate the correlation between methods. Improved precision<0.4% was obtained for Mg isotope ratios by ICP-MS compared with precisions of 1.3-2.6% for EI-MS.The ICP-MS method was used to diVerentiate exogenous and endogenous Mg, enabling the bioavailability of exogenous Mg from pharmaceutical preparations to be determined. Godlewska-Zylkiewicz et al.226 combined LC with ETAAS to determine Mg-protein species in human serum. Proteins were separated on an ion-exchange column and detected with a diode-array system. The authors observed that a CH COONa mobile phase stabilized Mg-protein complexes. The Mg concentration in the collected fractions was determined by ETAAS and the main Mg binding proteins were identified as albumin and globulin fractions.Several groups have studied Mg status in diVerent pathological conditions. Starobrat-Hermelin et al.227 investigated the value of Mg supplementation for children with attention deficient hyperactivity disorder (ADHD). Children with recognised ADHD and Mg deficiency, confirmed through determination of serum and hair Mg by AAS, responded positively to Mg supplementation over 6 months.Magnesium supplementation led to reduced hyperactivity and increased hair Mg levels. Rob et al.228 used AAS to monitor serum Mg levels in patients undergoing plasma exchange with albumin based substitutes. Plasma exchange led to a negative Mg balance and hypomagnesemia. Yoshida et al.229 determined Mg levels in bone and ligament samples from rats given an unbalanced mineral diet mimicking that found in amyotrophic lateral sclerosis (ALS) foci, and also in ligament and bone samples from cases of calcification of spinal ligaments (CSL).In both cases the Mg levels were significantly lower than controls indicating that low dietary intake of Mg may be a contributory factor for CSL and ALS. 1.9.20 Manganese. The determination of Mn by FI-ETAAS was used by Lang et al.230 as an indirect method for the quantitative determination of vitamin B6 in pharmaceutical preparations.Solubilized samples were reacted on-line with MnO2 at pH 4.5 and the reaction solution transferred to the atomizer of the AAS for determination of Mn. Recoveries of 98-102% were achieved for both tablets and injection solutions, but interferences from thiamine and riboflavin were identified. An ETAAS method for the determination of Mn in hair was described by Zhou and Ge,231 in which the acid digested hair was introduced into the furnace using a laboratory made probe system.An IR source was used to dry the 5 ml sample before introduction into the furnace followed by ashing at 300 °C and atomization at 2400 °C. The authors observed no interferences from other elements nor any memory eVect from the probe. An LOD of 11.7 pg was found. 1.9.21 Mercury. A variety of sample pretreatment procedures have been described for the determination of total Hg in biological matrices by atomic spectroscopic methods. Lamble and Hill22 developed an on-line microwave digestion procedure for the determination of total Hg in solid biological 732 J.Anal. At. Spectrom., 1999, 14, 717-781 3 3 samples by CVAFS. Slurried samples were mixed with HCl and KBr-KBrO and reacted in a PTFE coil within the microwave digester. Reduced HgII was subsequently reacted on-line with SnCl2 and the released Hg determined by AFS. The method was validated by analysis of several CRMs. Fernando et al.232 also developed a CVAFS method to determine Hg in blood and tissues of rats exposed to Hg.Their sample preparation also involved acid digestion with HNO3-H2SO4 followed by reduction with KBr- KBrO -hydroxylamine hydrochloride and reaction with SnCl2. Adeloju et al.233 observed that addition of H2SO4 prior to or following wet digestion of biological samples produced a significant improvement in sensitivity for the determination of Hg by CVAAS. A simple direct method for the determination of Hg in 4 urine, blood and digested tissues by ICP-MS was described by Moreton and Delves.151 Memory eVects from Hg were which complexed the Hg.Detection limits were 0.13 mg l-1 minimized by diluting samples with (NH4)2H2 EDTA in NH3, and 0.27 mg l-1 for blood and urine, respectively. The method was validated by analysis of CRMs and participation in external quality assessment schemes. The method was used to determine levels of Hg in blood and urine samples collected from Brazilian gold mine workers.All but one of the workers had blood and urine Hg levels above the 95% value for for UK industrial workers (19 mg l-1) and 17% of workers unexposed subjects, 55% had blood levels above the 95% value ance value (20 mmol mol-1 of creatinine). More importantly, had urine Hg levels above the UK occupational health guidsubjects living in a fishing village 100 km from the mine had significantly higher blood Hg levels than the mine workers, which the authors presumed to be related to fish consumption.As hepatitis and HIV were prevalent among the study subjects, a novel disinfection procedure was developed in which samples were pre-diluted with a virucide. Blood proteins were precipitated by the virucide and were redissolved with TMAH. In the ICP-MS method described by Mickley and Amato,234 samples were introduced into the plasma by electrothermal vaporization.Samples were combusted in a minitube furnace at 950 °C using an O2-Ar stream and cobalt oxide-alumina catalyst. The method was evaluated by analysing hair CRM IAEA-086 and an LOD of 19 pg was reported for this matrix. Lee et al.76 described an ETV-ID-ICP-MS method for the determination of Hg in urine. A mixed Pd-Mg(NO3)2-HCl chemical modifier gave optimum sensitivity for Hg, with an LOD of 0.02 mg l-1 being found. Analysis of CRM NIST 2670 toxic elements in freeze dried urine gave good agreement with the certified value.Rudner et al.235 described a novel separation and preconcentration treatment for the determination of Hg in biological samples by ICP-AES. Mercury was concentrated on a column containing a methylsalicylate chelating resin coupled to an FI manifold. Synthesis of the resin was reported to be faster and simpler than previously prepared chelating resins and the authors comprehensively evaluated factors aVecting the preconcentration step.Complexed Hg was eluted with thiourea and reacted on line with NaBH4 to generate Hg vapour. The method gave acceptable results for a variety of CRMs. Tao et al.16 solubilized samples of marine biological SRMs with TMAH for the determination of total Hg by FI-CVAAS. On-line addition of KMnO was necessary to obtain full recovery of both inorganic and organic Hg species. Various approaches have also been taken for the speciation of Hg.Perez-Corona and colleagues236 evaluated immobilized yeast cells on silica gel micro-columns for preconcentration and speciation of Hg. Inorganic Hg was retained by the silica support whilst methylmercury bound to the immobilized cells. for methyl Hg and 0.8 M CN- for HgII. Concentration factors Speciation was achieved by sequential elution with 0.02 M HCl of 15 and 100 were reported for organic and inorganic Hg,respectively. Gerbersmann et al.153 evaluated two derivatization-injection procedures for speciation of Hg in fish samples by GC-MIP-AES.In the first, microwave digested samples were derivatized by ethylation, extracted with hexane and injected, using a cooled injection system, for GC separation. In the second, the sample was reacted with NaBH4, followed by purge and trap injection onto the GC column. Wasik et al.76 described a method for speciation of Hg in biological samples by ICP-MS in which methyl and ethyl species were separated by purge and trap cryofocusing and chromatography at ambient temperatures.Optimization of Ar gas flows minimized peak broadening on the column. The authors noted that sensitivity and speed of analysis were limited by the acquisition rate of the quadrupole mass spectrometer. Willie et al.65 described a rapid method for the determination of total and inorganic Hg in biological samples by ETV-ICP-MS, following simple dissolution with TMAH.For determination of total Hg, injected sample was dried and atomized into the plasma. For inorganic Hg, an iodoacetic acid-acetic acid- thiosulfate chemical modifier was injected with the sample to form volatile methylmercury iodide. This was removed in the sample drying step, leaving only inorganic Hg to be quantified. The method was validated by analysis of marine tissue CRMs. In the clinical setting, Haswell et al.237 described CVAAS and ICP-MS methods for the determination of Hg in hair which were developed to support studies on transplacental transfer of Hg.The authors developed a small volume microwave digestion procedure to cope with the small quantities of hair from new-born babies. Barregard et al.238 used ELISA to determine levels of antinuclear auto-antibodies and immune complexes in Hg-exposed battery workers. Blood, plasma and urine Hg levels were determined by CVAAS. The authors found no associations between antibodies or complexes and any of the Hg exposure indices and no increased prevalence of abnormal antibody titres was observed.Barbosa et al.142 measured hair Hg levels as an indicator of Hg body burden in the indigenous population of the Amazon Basin. Samples were alkali digested and Hg determined by CVAAS. Hair Hg levels in non-Indian women ranged from 0.8-94.7 mg g-1, and in Indian women ranged from 0.8-13.3 mg g-1. A significant correlation between maternal hair Hg and hair Hg of breastfeeding infants was only observed in the non-Indian population.Segmented hair analysis indicated a 20% decrease in Hg body burden during pregnancy, which suggested significant placental transfer. 1.9.22 Molybdenum. As part of a European-wide project, Iversen and colleagues239 used ICP-MS to determine urine Mo levels in a healthy Danish population. Measurements were made at both 95Mo and 98Mo masses and no significant diVerences were observed between the levels determined at the two masses.The LOD for the method was 0.2 mg l-1 and a precision of 8.6% RSD was reported. Both uncorrected and creatinine corrected data were log normally distributed and a 95% parametric reference interval of 10-124 mg l-1 was calculated. There was no influence of gender or age on urine Mo concentrations. Of the many dietary factors examined, only butter consumption showed any significant correlation with urine Mo concentrations.An ETAAS method was developed by Haywood et al.240 to determine Mo accumulation in organs of sheep administered ammonium tetrathiomolybdate (TTM), a treatment for chronic Cu poisoning. They observed that Mo was widely distributed and accumulated in many organs including the brain and pituitary, and hypothesised that TTM treatment may redistribute excess liver Cu to the brain. 1.9.23 Nickel. An ETAAS method with slurry sampling for the determination of Ni in human scalp hair was described by Bermejo-Barrera et al.13 Washed hair was pulverized in a ball mill and the powder vigorously shaken with H2O to produce a slurry. The aqueous slurry was mixed with Mg(NO3)2, as a chemical modifier, and glycerol, and a 20 ml aliquot injected into the furnace.The method gave an LOD of 273 mg kg-1 and recoveries of NiII were 93-108%. The authors reported that matrix interferences caused less than a 5% suppression of the Ni absorbance signal.The method was evaluated by analysis of several hair CRMs and good agreement with certified values was found. In a study of the pulmonary damage in mice caused by combustion products from Ni coated polycarbonate, Larsen et al.241 used AAS to measure the Ni content of the ash following decomposition of the plastic. The findings led the authors to hypothesize that Ni lost through pyrolysis could account for the increased toxic eVects on pulmonary tissue through the formation of toxic Ni species.1.9.24 Platinum group metals. There is growing interest in the measurement of noble metals in relation to environmental contamination from automobile emissions. Very sensitive methods are required, however, to determine physiological levels of noble metals in biological fluids in the monitoring of environmental exposure. In the method developed by Begerow et al.70 for the determination of Pd and Pt in urine, measurements were made by double focusing magnetic sector ICP-MS.By using UV photolysis for decomposition of the organic matrix, very low reagent blank values were obtained and lowest detection limits were achieved operating the instrument in the low resolution mode. Practical LODs of 0.17 ng l-1 for Pd and 0.24 ng l-1 for Pt were found, but they were not for Pd and Pt in unexposed populations were 140 ng l-1 and improved with HR nickel cones.The reported reference values 1.8 ng l-1, respectively. A similar high resolution ICP-MS method was also developed by Krachler et al.69 for the determination of Pt group metals in urine. Low reagent blank values were again obtained using UV photolysis for sample digestion and sensitivity was improved further by using ultrasonic nebulization for sample introduction. The authors noted that interference on 106Pd from 106Cd and 40Ar66Zn, and on 103Rh by 206Pb2+ and 40Ar63Cu could not be disregarded at reported LODs were 0.25 ng l-1 for Pd and 0.03 ng l-1 for Pt the levels of interest. With a dilution factor of 1+20 v/v, and Rh.The authors found a mean urine Pt level of 1 ng l-1 in youngsters from urban and suburban areas of Rome. They also considered that Q-ICP-MS with ultrasonic nebulization had inadequate sensitivity for the quantitative determination of baseline levels of noble metals in urine.Sample introduction by ETV was used by Cairns and White242 for the determination of Pt in urine in order to monitor workers exposed to the element through the handling of cytotoxic drugs. In a comprehensive study of Pt levels in urban road dust, soils and biological fluids from environmentally and occupationally exposed populations, Farago et al.243 determined concentrations of Pt in blood and urine by ICP-MS. Measurable levels of Pt were reported in both occupationally and environmentally exposed groups. The authors noted that Pt levels in road dust were well correlated with traYc intensity but there was no correlation between Pt and Pb, suggesting that traYc emissions cannot be identified with certainty as the primary source of Pt.An ETA-LEAFS method was described by Aucelio and colleagues128 for the determination of Pt in biological samples. The authors thoroughly investigated furnace conditions to obtain optimum signal to noise ratios and obtained an absolute LOD of 50 fg.Analytical recoveries of 100-108% were reported for a range of biological matrices. The quantitative measurement of Pt in biological samples from patients undergoing chemotherapy with Pt drugs can usually be achieved with ETAAS. A comprehensive review, 733 J. Anal. At. Spectrom., 1999, 14, 717-7813. An LOD of leagues254 used N2 rather than Ar as the plasma gas for the with 177 references, of the clinical pharmacokinetics of carboplatin was presented by DuVull et al.244 The authors discussed methods of analysis and the development of models to describe the relationship between drug dose, toxicity and response.Milacic et al.245 described a simple ETAAS method for the determination of Pt in tissue biopsies in which 0.1 g samples were digested at 37 °C with HNO 3 mg l-1 was reported and aqueous calibration gave satisfactory results. Several papers in this review period again describe spectroscopic methods to study the pharmacokinetics of Pt based drugs for establishing appropriate administration regimes. Peng et al.246 measured free plasma Pt by ETAAS and Pt-DNA adducts by ELISA in a study of the relationship between adduct formation and pharmacokinetics of carboplatin and cisplatin.A weak correlation was found between adduct formation and the area under the unbound drug curve (AUC) several hours after infusion, but the relationship was lost beyond 24 h after infusion.Carboplatin formed lower levels of reactive adducts despite a much higher AUC for this drug. The authors concluded that interaction of the drug with the target was determined both by pharmacokinetic and cellular factors. Tokuhashi et al.246 used AAS to measure total and free Pt in plasma and urine in a study of the pharmacokinetics of cisplatin in children. Urinary excretion of Pt was only 27% after 48 h, leading the authors to hypothesise that other clearance pathways were important.They also concluded that dosing based on body surface area was inappropriate for children. 1.9.25 Selenium. In this review period, Se has replaced Pb as the element generating most scientific interest. Much of the work has arisen through interest and some concern over the Se status of many populations. A comprehensive evaluation of atomization techniques and chemical modifiers for the determination of total Se in human serum by ETAAS was undertaken by Gayon et al.247 Optimum performance was achieved with platform atomization.Satisfactory results were obtained with aqueous calibration standards if a 2CO3 Pd-Mg(NO3)2 chemical modifier was used, whereas the use as electrolyte. Selenium was measured at m/z 77 and 78. In a of Ni or Ni-Mg(NO3)2 necessitated matrix matched calibration standards. Cabon and Le Bihan248 examined the atomization behaviour of Se in the presence of high salt matrices.They noted that with a high SO42- content, Sr with rapidly taken up by erythrocytes and reduced to selenide. The Pd proved to be a better chemical modifier than Pd alone for selenide reappeared in the plasma bound to albumin and stabilizing Se species. In the method developed by Chen gradually disappeared again to be incorporated into et al.,249 serum samples were simply mixed with an selenoprotein-P and GSHPx.A robust method, using micro- HNO3-Triton X-100 chemical modifier by ultrasonic slurry bore ion exchange chromatography coupled to ETAAS was sampling for ETAAS analysis using STPF conditions. described by Emteborg et al.262 to determine Se species in Ultrasonication of the sample led to improved homogeneity biological materials. Four inorganic and organic species were and stabilization of Se in the diluted sample. Tyson et al.250 separated within six minutes.A Pd-Mg(NO3)2 chemical modidetermined total Se in human urine by FI-HG-ETAAS follow- fier was found to thermally stabilize all Se species up to ing digestion and reduction of organoselenium species by acid 1000 °C and LODs of 2.8-4.1 mg l-1 were reported. The refluxing with HBr-KBrO3. Following reduction, excess Br method was evaluated by analysing biological CRMs spiked was removed by addition of hydroxyammonium chloride. The with diVerent Se species. digest was diluted with 10% HCl and injected into a stream Various groups have determined Se in biological matrices of 0.2% NaBH4-10% HCl.Hydrides were trapped on an iridium coated graphite tube for determination by ETAAS. Recoveries >95% were reported from samples spiked with typical organoselenium compounds. Several groups have described ICP-MS methods for the determination of Se in body fluids. Sieniawska et al.251 described a simple 1+15 v/v dilution procedure for the determination of Se in blood and plasma.The diluent contained 1% butanol to overcome interferences from Ar adducts on 78Se. The method gave an LOD of 0.02 mmol l-1 and was validated by demonstration of excellent performance in two international quality assessment programmes. This same group252 and also the group of Nichol et al.253 combined aYnity chromatography with ICP-MS to quantify selenopro- 734 J. Anal. At. Spectrom., 1999, 14, 717-781 teins in human serum.Two independent aYnity columns were used to separate selenoprotein-P, glutathione peroxidase (GSHPx) and albumin. Analysis of two reference sera found the distribution of Se to be 53 and 54% with selenoprotein-P, 21 and 29% with GSHPx, and 27% and 13% with albumin. To overcome polyatomic interferences, Furuta and coldetermination of Se by ID-MIP-MS. The authors measured Se in NIES 4, human serum SRM, using an enriched 78Se spike solution and determining the 78Se580Se ratio with a precision better than 1% RSD.Jimenez et al.255 used ethylation derivatization with NaB(C2H5)4 to volatilize selectively selenoaminoacids for on-line determination by ICP-MS. Selenium alkyl derivatives were preconcentrated by cryogenic trapping or solid phase adsorption for determination of Se at ng l-1 levels. Finally, in this section, Turner et al.256 described a method for the determination of Se in serum by ETV-ICP-MS.By selecting appropriate furnace programmes, the authors eliminated interferences from 81BrH on 82Se and gave an LOD of <0.1 ng g-1. 40Ar37Cl on 77Se. The improved transport eYciency of ETV An overview of methods developed for speciation of Se in biological samples was presented by Behne et al.257 Cooney and colleagues258 presented the results of their studies on the separation and identification of Se metabolites using HPLCICP-MS. Using diVerent chromatographic techniques, they speciated seven inorganic and organic Se species in human urine, of which only one was tentatively identified as selenomethionine.Uden et al.259 developed methods employing derivatization and headspace GC with MIP-AES or ion pair chromatography with ICP-MS to study Se speciation in biological samples. They applied the methods to the determination of Se species in nutritional supplements. Michalke and Schramel260 used capillary zone electrophoresis and isoelectric focusing to separate Se species in human serum and milk for determination by ICP-MS.Six organic and inorganic species were separated on an uncoated capillary for electrophoresis and a coated capillary for isoelectric focusing using Na very elegant study, Suzuki et al.261 used HPLC-ICP-MS to study the metabolism of Se in rats administered enriched Se isotopes. The authors observed that administered selenite was to establish reference values for Se in the general populations of diVerent countries.However, there appears to be no standardization or consensus on the most appropriate biological indicators of Se status. Veillon et al.7 reported that infection risks from viruses could be overcome by simple pre-dilution with 0.1 M HNO3 for the determination of Se in serum by ETAAS. Diluted samples were mixed with an Ni-Mg(NO3)2 chemical modifier for quantitation of Se. The acid concentration was suYciently low to prevent protein precipitation whilst still inactivating viruses.Meissner and Hohne263 used ETAAS to establish Se reference values of 0.68-1.52 mmol l-1 for blood and 0.85-1.85 mmol l-1 for serum in the population of the Dresden area of Germany, which they considered to indicate marginal Se deficiency for this population. A serumSe reference interval of 60-106 mg l-1 (0.75-1.34 mmol l-1) was reported by Torra et al.264 for a healthy Northern Spanish (<45 mg l-1).Dhindsa et al.265 found a mean plasma Se level population. No subjects showed excessive Se deficiency of 91.8 mg l-1 (53-131 mg l-1) in the Sikh population of Sydney, Australia. They considered that the Se status was adequate even though a vegetarian diet was common. The establishment of reliable reference values depends on the adequacy of the analytical method and satisfactory performance, demonstrated by analysis of control materials.Morisi et al.133 presented results of an inter-laboratory study of the determination of Se in serum control materials spiked with inorganic Se. Russo266 examined the relationship between Se status and the incidence of colorectal cancer, using ETAAS to determine plasma Se levels. Mean plasma Se levels in cancer patients and controls were 107 and 120 mg l-1, respectively, and statistical analysis indicated that individuals with the highest quartile of plasma Se levels had 0.24 times the risk for colorectal adenoma compared with individuals in the lowest quartile. The authors hypothesised that Se may be a useful chemopreventative agent for colorectal neoplasia.1.9.26 Silicon. The use of silicone implants in cosmetic 2O. A 10 ml aliquot of the diluted both Mg and Sr. In the study of Tanaka et al.277 Sr levels in surgery continues to stimulate studies on the determination of Si in biological matrices. A review examining the toxicity of organic and inorganic Si compounds in humans was produced by Lugowski et al.267 Topics included the measurement of total Si in blood and other body fluids and methods for speciation of Si.A need for reliable CRMs was highlighted. Several methods for the determination of total Si in biological fluids and tissues by ETAAS have been described in this review period. A method for the direct determination of Si in urine was described by Kobayashi et al.268,269 in which urine was diluted 1+499 v/v with H sample was injected into the furnace with an equal volume of a 5% Ni solution as chemical modifier, ashed at 1600 °C and atomized at 3000 °C.The method was evaluated by comparison with an ICP-AES method. The method was also used for the determination of Si in blood.268 Blood was diluted 1+19 v/v with 1% Triton X-100 and injected with the Ni chemical modifier. An LOD of 0.7 mg l-1 was reported for Si in blood but recoveries were only around 70%.For the determination of Si in serum and acid digested breast tissue, Leung and Edmond270 diluted samples 1+3 v/v with a complex chemical modifier solution containing LaO, CaCl2, (NH4 )2HPO4, EDTA, EtOH and Triton X-100. An ashing temperature of 1400 °C and atomization temperature of 2700 °C with gas stop were found to give optimum absorbance signals. The method was used to establish reference intervals for Si samples obtained from women with silicone breast implants.Lykissa et al.271 described the results of a series of studies investigating the release of Pt and low molecular weight silicones from breast implants. The release of silicones from intact implants into lipid rich oil media, aqueous tissue culture media and emulsions of the two was investigated. Silicon released into the media was quantified by GC with detection by either AES or MS. Platinum release was quantified by ICP-MS.Leakage of silicones was greatest when the surrounding medium was lipidrich and leakage levels of up to 10 mg d-1 were observed at 37 °C. Platinum leakage levels of up to 20 mg d-1 were also recorded. The observations led the authors to conclude that leakage of both elements from breast implants could lead to significant accumulation in lipid-rich tissues. 1.9.27 Silver. A sample treatment involving digestion with H2SO4, reaction with 30% KI-2% ascorbic acid and extraction into IBMK was developed by Luo et al.272 for the determination of Ag in whole blood by ETAAS.Optimized ashing and atomization temperatures were found to be 500 °C and 2300 °C, respectively, and calibration was linear up to 100 mg l-1. Analytical recoveries ranged from 89 to 110% with a method precision of 4.8-13% RSD. The method was also used to determine Ag in serum and urine. Lech273 described an FAAS method for the quantitative determination of Ag in post-mortem samples from a fatal case of Ag poisoning.Tissues were digested with HNO3-H2SO4 and analysed by FAAS using an air-C2H2 flame. The highest concentrations of Ag were found in the spleen, heart and lungs. The authors hypothesised that diVerential tissue aYnities for Ag may be associated with the levels of Ag binding proteins in the basement membrane of each tissue. 1.9.28 Sodium and potassium. Both Zoppi and colleagues274 and Ziebig et al.275 discussed factors responsible for the observed diVerences in the determination of Na and K in sera by diVerent analytical techniques.In the first, serum osmolality was identified as the main factor for diVerent values of Na and K determined by ion specific electrodes and flame photometry. In the second, higher values for both Na and K were determined by direct potentiometry compared to flame photometry for samples with high protein concentrations. In a study on the eVect of dietary Na on biochemical markers of bone metabolism in young women, Ginty et al.276 determined urine Na and K by flame photometry and urine Ca and Mg by AAS.The researchers identified an Na induced calciuria in Na sensitive women, but observed that the calciuria was not associated with increased bone resorption or turnover. 1.9.29 Strontium. Two studies described the determination3 of Sr in teeth by AAS. In the method used by Perez-Jordan et al.,27 teeth samples were powdered and digested with HNO in a closed-flow microwave system for 10 min for analysis of enamel and dentin were compared in both healthy and diseased teeth.The authors found that in sound teeth, Sr levels were not aVected by gender or degree of dental disease. On the other hand, in carious teeth, Sr levels in enamel increased with the degree of decay. The authors hypothesised that Sr may act synergistically with other trace elements in cariostatic activity. The levels of Sr in human hair were determined by Zhang et al.278 using FAAS.Hair was digested with HNO3-H2O2 and the digest mixed with a 20% surfactant solution for nebulization into an air-C2H2 flame. Interference from Al required the use of standard additions calibration for accurate quantitation. The addition of surfactant to the sample solution gave a 55% enhancement of Sr absorbance. A HR-ICP-MS method was described by Latkoczy et al.82 for the determination of Sr isotope ratios in prehistoric human bone samples, in which optimized operating parameters including correction for mass bias and dead time were investigated.When the instrument was fitted with a microconcentric nebulizer and installed in a pressurized ultraclean room with temperature control, 87Sr586Sr isotope ratios could be determined with a precision better than 0.035% RSD. Such high precision was needed for accurate interpretation of paleoanthropological specimens.The authors used the method to analyse several 7000 yr old human skeleton samples from lower Austria in order to confirm the provenance of the individuals. 1.9.30 Titanium. A comparative study of ICP-AES and ETAAS methods for the determination of Ti in human plasma was made by Einhauser et al.,45 in order to determine Ti concentrations in plasma of patients administered the Ti containing anticancer drug Budotitane.For measurement by ICP-AES, dilution of serum 1+99 v/v with H2O and sample introduction by ultrasonic nebulization gave satisfactory 735 J. Anal. At. Spectrom., 1999, 14, 717-781results with no evidence of matrix interferences. The determination of Ti by ETAAS, however, was more problematic. A complex furnace programme was required to minimize matrix interference eVects and to reduce carbide formation. Memory eVects were observed with both longitudinally and transversely heated furnaces, due to the refractory nature of the element.The authors concluded that ICP-AES was better suited than ETAAS for the determination of plasma Ti in cases of therapeutic drug monitoring. 1.9.31 Thallium. Fleischer279 described an ETAAS method with Zeeman eVect background correction for the direct determination of Tl in urine from occupationally exposed workers. Urine was injected directly onto the platform of a pyrolytically coated graphite tube together with a tetraamine-Pd(NO fier.Standard additions calibration was necessary for accurate quantitation of Tl. 1.9.32 Tin. Saeki et al.280 described a microwave digestion 3 of ICP-MS for the quantitative measurement of uranides in biological matrices. Two groups have described ICP-MS methods for the determination of U in urine in order to monitor exposed workers. The study by Schramel et al.73,282 described a rapid, sensitive method in which urine samples were sequentially acidified with HNO and HCl for determination by ICP-MS with pneumatic nebulization.To compensate for matrix suppression and nebulization eVects due to variations in urine density, viscosity and mineral content, internal standardization with Ir was necessary. Satisfactory results were obtained with calibration standards prepared in 3.5% HNO3-1.5% HCl but more accurate quantitation at very 3)2-Mg(NO3)2-NH4NO3 chemical modilow levels was achieved with standard additions calibration.The method was equally suitable for the determination of U and Th in urine with LODs of 0.5 ng l-1 reported for each element. The method was evaluated by comparison with results obtained by alpha-spectrometry. In the ICP-MS method developed by Caddia and Iversen,283 for U in urine, sample introduction was also by pneumatic nebulization following dilution of urine samples 1+19 v/v with 1% HNO3.Standard additions calibration with Ir as internal standard was again necessary to correct for adverse matrix eVects. An LOD of 0.32 ng l-1 and precision of 2-4% RSD were reported. Analysis of samples from non-occupationally exposed individuals gave median urine U levels of 13.5 ng l-1 for males and 17.6 ng l-1 for females. The authors emphasised that careful control of blank contamination was essential to achieve suYciently low detection limits.A HR-ICP-MS method was used by Nanni and colleagues72 to determine U and Th in urine of exposed workers from a U processing plant. Organic matrix was partially removed by drying the urine samples at 70-80 °C and dissolving the residue in 1% HNO3. The authors investigated the eVect of matrix suppression and observed that signal suppression became significant at dilution factors less than 10. Optimum results were obtained with matrix matched Detection limits of 0.01 mg l-1 for U and 0.26 mg l-1 for Th calibration standards and internal standardization with Bi.were reported. procedure, using HNO3, for the determination of total Sn in biological materials by ICP-MS. With a 100 mg sample, digested and diluted to 10 ml, an LOD of 10 ng g-1 could be achieved. Good agreement with certified values was obtained for three biological CRMs. Interferences on the quantitation of Sn were noted when HCl was used in the digestion mixture.Microwave digestion procedures were also described by Schmitt and colleagues155 for the speciation analysis of organotin compounds in marine biological materials by GC-AES. One- and two-stage sample preparation procedures were examined. In the one-stage process, lyophilized tissues were digested with CH COOH-nonane-Na ethylborate. In the two stage process, tissue samples were first digested with TMAH, acidified to pH 5, and extracted into nonane-ethylborate.Organic supernatants from both procedures were chromatographed on a DB-210 column with diethyl ether as eluent and Sn quantified by AES. Both procedures were validated by analysis of NIES Fish tissue CRM, which has certified values for tributyl- and triphenyltin species. Detection limits for tributyltin and triphenyltin were 2-10 and 20-40 ng g-1, respectively. Martin et al.281 described a method combining HG, cryogenic trapping and GC separation with AAS detection of Sn for the determination of low molecular weight organotin compounds in brain tissue of rats exposed to trimethyltin.The method was suYciently sensitive to determine levels of Sn below 0.2 ng in tissue from exposed animals. The method was subsequently used to measure Sn in human brain tissue showing evidence of Alzheimers disease. No low molecular weight organotin species were detected in these samples. A review of AAS, GC and LC methods for the determination of organotins in biological and environmental samples was presented by Horiguchi.154 The author also reviewed the eVects of organotin compounds on the human body, paying particular attention to their role in endocrine disruption.1.9.33 Tungsten. A simple method for the quantitative deter- 3 at 60 °C for 12 h. Tungsten was measured mination of W in biological fluids, hair and nails by ICP-AES was described by Marquet et al.46 in order to monitor a bizarre case of severe acute W intoxication, in which a quantity of wine and beer that had been rinsed in a hot gun barrel was consumed by a military recruit. Hair and nail samples were digested in HNO at 207.91 nm with the 193.03 C line as reference and using Ar as nebulizer, sheath and coolant gas.Initial blood and urine W levels, following hospital admission, were determined to be 5 mg l-1 and 101 mg l-1, respectively, and were confirmed by ICP-MS. Elevated W levels were also found in hair and nails.736 J. Anal. At. Spectrom., 1999, 14, 717-781 Blood levels remained elevated (>5 mg l-1) after 6 courses of haemodialysis over 13 days. 1.9.34 Uranides. There has been growing interest in the use 3 An ICP-MS method with sample introduction using a microconcentric nebulizer was described by Pietrzak and Kaplan284 for the quantitative determination of Pu in an artificial urine reference material. Plutonium isotopes were extracted from the urine matrix by co-precipitation and acid digestion, followed by anion exchange chromatography.All procedures were performed in a clean room. The absolute LOD for the method was reported to be 1-2 fg for 239Pu, which was comparable with the fission track analysis method currently used for accurate quantitation of Pu. 1.9.35 Vanadium. To overcome polyatomic interferences on the determination of V in urine by ICP-MS, Minnich et al.52 designed a cryogenic desolvation system, which was coupled to an ultrasonic nebulizer.Cryogenic desolvation markedly reduced the 35Cl16O interference on V by reducing the amount of Cl and O reaching the plasma through condensation of HCl and removal of water vapour. Matrix suppression was also corrected for by internal standardization with Sc. The method was found to be satisfactory for rapid screening of urine samples from occupationally exposed workers in order to assess the degree of exposure. 1.10 Conclusions Speciation features prominently in this review.Elements featured in previous Updates, such as As, Hg and Sn, are still important but other elements have been actively studied,notably Se (section 1.9.25). Three articles dealt with general aspects of speciation, giving advice on sample preparation,1 analytical strategies2 and improving reliability and speed of analysis.3 In this last, £obinski et al.reason that, for speciation to be part of the normal output of a routine laboratory and therefore capable of being included in environmental and occupational legislation, methods need to be faster than those developed previously to make them commercially viable. Moreover, they need to be more reliable, thus reducing the standard of competence required of the analyst carrying out the determination. They envisaged that the sample preparation and chromatographic separation steps should form a compact accessory for an atomic spectrometer.Their example of replacing a commercial gas chromatograph with an isothermal multicapillary column 22 cm long illustrates this point clearly. Work on microwave-assisted sample preparation by their group3,155 and others153 has reduced preparation times to a few minutes. Many of the on-line speciation systems using FI technology seem to be approaching the goals outlined above. The introduction of UV photolysis, for example by Tsalev et al.,25 seems to have simplified on-line systems for the speciation of As.It is obvious from the publications that more laboratories now have double focusing magnetic sector ICP-MS (also called HR-ICP-MS). In multielement analysis, this technique seems capable of accurately measuring the clinically important elements, Cu and Zn, in biological samples68 which suVer from interference when determined by Q-ICP-MS.In low resolution mode, the technique oVers impressive sensitivity, which has been exploited well in the determination of physiological concentrations of the Pt group metals in urine.69,70 This is becoming a topical subject as evidence grows for the release of these elements into the environment from catalytic converters in cars (see section 1.9.24). An important advantage of HR-ICP-MS is the high precision with which isotope ratios can be measured, oVering an easier and more rapid alternative to TIMS.79,81-83 This was confirmed in an inter-laboratory comparison of the determination of Pb isotope abundance in a Bone Ash SRM in which results by ICP-MS were statistically equivalent to those obtained by TIMS.132 There have been interesting developments in the use of tungsten-coil atomizers (see section 1.4).Conference presentations on the application to ETV-ICP-MS118 and in multielement ETAAS119 are intriguing. As manual injection is generally used with these atomizers precision is not good, and this reminds this writer (DJH) of the early days of ETAAS before autosamplers were available.Parsons et al.120 used an alignment guide to improve precision and Bruhn et al.121-123 have demonstrated the importance of the modifier in achieving accuracy and precision. It is apparent that the best modifier is not necessarily that which is best for a graphite furnace. 2 Analysis of foods and beverages This review follows on from last year's,6 and covers work published during the year ended October 1998, describing the analysis of foods and beverages by atomic spectrometric techniques.It includes papers and some conference abstracts which present novel work and significant developments in instrumental and analytical techniques, and their applications. Table 1 complements the text with summaries of these publications. 2.1 Sampling and sample preparation 2.1.1 Extraction.Comparisons of diVerent extraction methods were described by three groups. Ali et al.285 compared the eYciencies of three methods (wet digestion, CHCl3 extraction and TCA extraction) in extracting Ca, Cr, Cu, Mg, Mn, 3 Pb and Zn from buValo milk for measurement by AAS. Elements such as Cu and Mn were not detected in samples prepared with TCA, which gave optimum extraction for Ca, Cr, Ni, Pb and Zn. The best results for Cu and Mn required initial digestion in HNO followed by removal of organic residues in CHCl3.A conference presentation by Ponce de Leon et al.286 compared preparation methods for trace element analysis of herbal medicinal teas by ICP-MS or ICP-AES. They used a slurry technique with samples prepared by infusion or microwave digestion, discussing the advantages and limitations of each technique and noting the trace element content of herbal medicines from diVerent regions which are claimed to have the same curative properties.It is to be hoped that this work will be published in full. Two procedures for leaching trace elements from food 3, packaging paper boards are currently proposed by the European Commission: (i ) immersion in distilled water for 24 h at 23 °C (extraction test); and (ii ) immersion in 3% v/v acetic acid for 24 h at 40 °C (migration test). Conti287 used both methods and AAS to measure the levels of Cd, Cr, Hg and Pb in 15 samples of paper boards used for packaging pasta and cereal products.It was found that most of the samples contained trace element levels higher than the limits specified by the European Commission and that the diVerence in results for the two extraction methods were considerable. The author recommended use of the more drastic migration test when a higher degree of food packaging safety is required. A simple procedure was developed by El Azouzi and co-workers288 for the room temperature leaching of trace elements from mussel samples with a solution of 1.6 M HNO 1.2 M HCl and 0.1 M H2O2.The method involved the use of a sonication time of 120 min with determination by FAAS and ETAAS. The leaching of Cs and Sc were poor, Cd, Co, Cr, Rb and Se were partially leached and quantitative recovery was achieved for Ca, Cu, Fe, K, Mg, Mn, Na, V and Zn. The results obtained by the leaching procedure were evaluated by comparison with those obtained by ICP-MS and AAS following microwave-assisted digestion and also by solid sample analysis by NAA.The method gave LODs of 0.081, 0.012, 0.059, 0.002 and 0.007 mg ml-1 for Ca, Cu, Fe, Mg and Zn, respectively, and 0.342 mg ml-1 for Mn. Recovery from spiked samples ranged from 92 to 109%, with relative errors <9% for analysis of a certified mussel tissue. Ozdemir and Gucer289 described a photodecomposition method for preparing tea infusions for trace element analysis by FAAS.The dissolved organic substances such as polyphenols and alkaloids present analytical problems, particularly tannic acid, which can cause an apparent increase in Fe and Mn concentrations. It was reported that the total dissolved organic substances were completely destroyed by a UV-H2O2 decomposition technique. Two decomposition methods were compared for the determination of Al, Ca, Cu, Fe, K, Mg, Mn, Na and Zn and a lower standard deviation was reported for the UV-H2O2 decomposition. Sample preparation techniques by dry ashing for As, Cu, Fe and Pb determination in sugar were reported by Leblebici and Volkan.290 White sugar samples were dry ashed with either H2SO4 or Mg(NO3)2 prior to measurement of As by HGAAS and Cu, Fe and Pb by FAAS.Very little diVerence in recoveries (ranging from 87.6-97.4%) was noted between the two methods for any of the elements measured.This review year saw increasing interest in the development of methods to solubilize tissues with TMAH and tertiary amines. Two methods are described here for multi-element determination, and techniques specific to the analysis of I are considered in section 2.6.2. Pozebon et al.17 proposed a method for preparing solid biological samples for analysis by ETV-ICP-MS. A solution or slurry was formed by mixing 20-100 mg sample with 10-200 ml of a 25%m/v TMAH solution.Complete dissolution 737 J. Anal. At. Spectrom., 1999, 14, 717-781was obtained for animal tissues, and slurries for plant and whole egg powder. The slurries were stirred manually during measurement after every three readings, and good results were obtained for four CRMs using external calibration and, in some cases, analyte additions method, for Ag, As, Co, Cu, Mn, Ni, Se, Te and V.However, Cd and Cr could not be measured in the bovine muscle sample owing to matrix eVects and spectral interferences, respectively.Krushevska et al. used a conference presentation to discuss the advantages of the addition of tertiary amine mixtures in the analysis of food and biological samples by ICP-MS and ICP-AES.19 They summarized the techniques developed and described applications for the accurate ICP-AES determination of low levels of Si in food, the sensitive and reliablemeasurement by ICP-MS of As and Se in milk and food, the determination of major and trace elements in powdered, skimmed and whole milk, and the ICP-MS analysis of I in food and biological samples.A commercial tertiary amine reagent, CFA-C, neutralized the eVect of HF used for sample digestion, and helped to prevent it attacking Si-containing parts of the ICP system during Si analysis. This eVect was applied to ICP-MS analysis of biological samples which had been microwave-digested in HNO3-H2O2-HF mixtures.Signal enhancement was noted for trating Pb2+ from tap and sea-water for analysis by ICP-AES As and Se. They also reported that CFA-C dissociated casein micelles in milk and stabilized the liquid phase cations, an eVect which had not been observed in the presence of other alkaline solutions such as ammonia or NaOH. This had enabled the establishment of a direct method for measurement in milk of Ca, K, Mg, Na, P and Zn by ICP-AES and Al, Ba, Cu, I, Mn, Mo, Pb, Rb, Sb, Se and Zn by ICP-MS.2.1.2 Digestion. A new high temperature/pressure flow system for sample digestion was described in a conference abstract20 for on-line digestion of biological samples with measurement with ICP-AES. Digestion temperature and pressure were significantly increased by the use of special HPLC and GC materials such as Teflon-lined tubing, glass-lined tubing, quartz capillaries with thick walls and Pt/Ir capillaries. A HPLC pump permitted a working pressure up to 40MPa and the capillaries could be heated to give digestion temperatures in the range 250-350 °C.Insertion of a restrictor with high flow resistance at the end of the capillary created a quasi-closed system within which a liquid could be heated to much higher temperatures than its atmospheric boiling-point without generating any vapour. The degree of sample destruction was characterized by residual C content after digestion, which was measured by ICP-AES as less than 1%.First results for sample digestion were presented, the applicability of diVerent tube materials discussed and the feasibility of on-line determination of trace elements by ICP-AES after digestion evaluated. The authors believe that the system has great potential for this and other detection methods, and the publication of this preliminary work and follow-up studies will be welcome, particularly with details of sample sizes used, LODs achieved and system hygiene.3-H2O2-HF for the determination of total for measuring Pb in water by FAAS. Samples (30 ml ) were 3 2O2. Excess HF 3BO3. Matrix eVects from Ca, K, Mg, segmented with IBMK. The tetrabutylammonium iodoplumb- 3 4 3 Digestion methods for the measurement of Al in foods have been described by two groups. Sun et al.291 used microwave digestion with HNO Al in seafood and meat by ICP-AES. Lyophilized samples were first digested with HNO and HF in closed vessels with an additional digestion in open vessels with H was removed with H Na and HNO were investigated, together with possible spectral interferences. Variations in microwave power and the amount of HF used were reported.Analysis of CRMs and recoveries of spikes (95.2-97.6%) were used to demonstrate the reliability of the technique. Twelve samples of meat and seafood were analysed and the results compared with those obtained following hot plate digestion with HNO -HClO and with HNO3-H2O2 microwave digestions.It was reported that tap water, rain water and sea-water were 94-100%. the digestion without addition of HF was incomplete for the 738 J. Anal. At. Spectrom., 1999, 14, 717-781 determination of total Al. A group in Poland used hot concentrated HNO3-H2SO4-HClO4 (1+1+1, details given) in a two-stage reaction to decompose portions (0.1-100 g) of various homogenized foods.292 After evaporation of the acids, the residues were dissolved in 1 ml of 50% HCl and diluted with H2O.This solution was filtered and treated with 1 ml 6 M acetic acid, neutralized with 700 ml 6 M ammonia and adjusted to pH 6.1-6.2 with 5 ml acetic acid-ammonia buVer. The solution was extracted with 2 ml of 0.3% NaDDC in CHCl3 and 5 ml CHCl3. The aqueous phase was extracted with 5 ml of 2% quinolin-8-ol in CHCl3 and the Al-oxine complex was measured by spectrophotometry at 385 nm.The recovered 309.3 nm. Detection limits of 2.5 mg per sample and 6 mg ml-1 complex in a 4% KCl matrix was measured by FAAS at were reported for spectrophotometry and FAAS, respectively, and the Al contents of a range of foods were tabulated. 2.1.3 Preconcentration. Techniques for the preconcentration of trace elements in water were described by five groups. A novel Pb2+ ion exchange resin, based on a Pb2+ ion templated polymer, was used by Bae et al.293 for removing and preconcenmolecularly imprinted for Pb2+ by creating polymers with or a simple colorimetric assay.The ion exchange resin was cavities lined with complexing ligands arranged to match the size of Pb2+. The Pb2+ resin consisted of Pb-vinylbenzoate charge, co-ordination number, co-ordination geometry and complex and divinylbenzene in styrene using chemical initiation by azobisisobutyronitrile. Details of the synthesis and characterization of the templated resin have been presented the recovery of Pb2+ measured by ICP-AES when using the previously by this group, and the current publication compared templated resin with three other ion exchange resins (Chelex-100, thiol-based Duolite GT-73 and a proprietary NASA polyacrylic acid resin).Samples preconcentrated on cations, and this resin was the most selective for Pb2+ of all the templated ion exchange resin contained almost no other those tested, giving an LOD of 1.0 ppb.The authors proposed that the templated resin is suitable for a field-ready test kit for the determination of Pb2+ in environmental samples when used with colorimetric analysis. Anezaki et al.294 adjusted 150 ml of H2O to pH 2 with 6 M HCl, then mixed it successively with 4 ml of anion exchange resin suspension, 1.5 ml 20 nM APDC and 1.5 ml 3 M NaClO4. The mixture was adjusted to pH 6 with 2 ml of 1 M ammonium acetate-1 M NaOH (1+1), stirred for 10 min then filtered under vacuum.The filter was placed in a beaker with 1 ml 0.1 M HCl containing 100 mg Pd and the beaker was sealed with film and irradiated ultrasonically for 1 min. An aliquot of the suspension was analysed Pb, respectively. Detection limits were 0.17 ng l-1 for Cd with by ETAAS with detection at 228.8 and 283.3 nm for Cd and recoveries of 98.3-100%, and 5.7 ng l-1 for Pb with recoveries of 99.3-99.6%, and interference eVects were tabulated. Tao and Fang295 described a dual stage preconcentration system using FI on-line ion exchange followed by solvent extraction pumped through a 1 ml column of Amberlite IRA 728 resin for 1 min.The pump was stopped and a second pump eluted the Pb with 1 M tetrabutylammonium bromide, which was ate ion pair was extracted into this solvent in a 30 cm reaction coil. A 50 ml portion of the extract was retained in a holding coil while excess solvent and aqueous solution were run to waste.Each extract was neutralized with NH3, mixed with 25 ml 1 M ammonium acetate buVer and diluted to 250 ml for analysis by FAAS at 283.3 nm. The enrichment factor was 550, and the LOD was 0.3 mg l-1. Recoveries of Pb added to A conference report296 described in considerable detail a, method for the preconcentration of impurities in high purity and drinking water for measurement by ICP-AES and ETAAS. The impurities (Al, Ag, As, Ba, Be, Bi, Ca, Cd, Co, Cr, Cu, Fe, Ga, Na, Hg, K, Mg, Mn, Mo, Ni, Pb, Si, Sn, Te, Ti, V, W, Zn) were co-precipitated with 8,8¾-diquinolyldisulfide (the oxidation product of 8-mercaptoquinoline) the precipitate separated on nuclear filters, followed by calcination and analysis of the concentrate. A method of electroconcentration was developed to measure trace elements in the suspended particle fraction of the water.The method consisted of capturing particles on an electrode in a constant electrical field.It was found that Fe, K, Mg and Mn were the main elements associated with the particle fraction although their contribution to the total concentration in water was <0.5%. The group reported that Si was the major impurity in the water samples, with high concentrations of Ba, Ca, Fe, Mg, Na and Zn in drinking water and concentrations of heavy metal impurities<1×10-8 mass%. Trace enrichment on an activated carbon column of Cd, Cu, Ni and Pb in drinking water was reported by Soylak et al.297 Drinking water (500 ml ) was adjusted to pH 6, mixed with 0.5 M 1-(2-pyridylazo)- 2-naphthol and, after 10 min, was applied to a column packed with 0.5 g activated carbon.The analytes were eluted with 10 ml 2 M HCl in acetone, the eluate evaporated to near dryness then diluted to 5 ml with 2 M HCl. This solution was limits were 12, 25, 59 and 43 ng l-1 for Cd, Cu, Ni and Pb, analysed by FAAS using an air-C2H2 flame.The detection respectively, with recoveries in the range 96-102%. Yaman and co-workers continued their work on trace 3 element preconcentration on activated carbon reported in previous years' Atomic Spectrometry Updates6 (refs. 244 and 245). Yaman and Gucer298 modified a previous method for the measurement of Al in animal milk and fruit juices by FAAS. The samples were wet ashed and quinolin-8-ol (oxine) and cupferron were used as complexing reagents for adsorption of the Al complexes on activated carbon.It was found that simultaneous enrichment of Al and Pb was achieved and both elements were measured by FAAS. Yaman also described the determination of Ni in vegetables by AAS following preconcentration by mixing with oxine or cupferron and activated carbon.299 A Chinese group used FAAS to measure Pb in soft drinks after preconcentration by coprecipitation.300 The sample (100 ml ) was neutralized, shaken with 2 ml 15% Mg(NO3)2 and 2 ml 20% NaOH for 1 min and stored for 1 h.The resulting precipitate was centrifuged, the supernatant discarded and the residue dissolved in HNO and diluted to 10 ml with H2O. Using the method of standard additions, plexation with ETDA to form negatively charged metal-EDTA recoveries of Pb were 90.5-112% with an RSD of 1.2-7.6%. 3 Two groups described flow injection preconcentration systems. Ivanova et al.301 developed a rapid and selective FI on-line sorption separation and preconcentration procedure for the determination of Bi in cod muscle, lake and river sediment by ETAAS.The diethyldithiophosphate complex of Bi was formed in 0.5-4% (v/v) HNO and adsorbed onto the inner walls of a PTFE knotted reactor. The complex was eluted with 30% (v/v) HCl and the eluate directly introduced into a pyrolytically-coated graphite tube without a pre-heating step. The ETAAS measurement of the concentrated analyte was carried out at the same time as the preconcentration cycle for the next sample.The enrichment factor was 74 and a detection limit of 3 ng l-1 was achieved with a sampling frequency of 28 h-1. The RSD was 3.4% for 0.1 mg l-1 (n=9) lake sediment and river sediment CRMs. A value of 14 ng g-1 and results were in good agreement with certified values for was found for BCR 422 (cod muscle). Enriquez-Dominguez et al.158 described their work on a FI preconcentration system with a chelating resin for the determination of trace and ultratrace amounts of Cd in mussels by FAAS.The metal was preconcentrated on a minicolumn packed with poly(aminophosphonic acid) resin and eluted with dilute HCl directly into the nebulizer-burner system of an FAAS. A preconcentration factor of 16-47, equivalent to a 3.4-10 ml sample, was in the sample was 0.56 mg l-1 for a sample volume of 3.4 ml achieved by using a time-based technique. The detection limit of Cd in the range 1-20 mg l-1.Chemical and flow variables and an RSD of 1.4-6.6% was obtained for diVerent amounts were studied and the system was found to be essentially unaVected by interferences. Good results were obtained for a CRM and the method was applied to the analysis of mussels from estuaries in part of Spain. 2.2 Speciation As in previous years, trace element speciation in a wide range of matrices has continued to attract considerable eVort.In a review with 41 references,302 Crews emphasised that studies on the speciation of trace elements in foods are required to better understand how the absorption and bioavailability of elements can be reduced for toxic elements, and improved for nutrients, and to validate existing risk assessment procedures for toxic elements. It was also pointed out that the methodology for speciation studies is still in the research and development stage, with few CRMs or standard procedures.Csikkel- Szolnok and Kiss303 investigated the Ca and Mg content of wheat, oats and barley and the uptake of these elements. Total Ca and Mg were measured by AAS following microwave digestion in HNO3. Uptake was apparently defined as the solubility of Ca and Mg following incubation at 37 °C for 1 h in both distilled H2O and in synthetic gastric juice (0.1 M HCl and pepsin). However, this reviewer considers that, since inorganic element absorption occurs mainly in the small intestine and not the stomach, it is unlikely that solubility of these elements under gastric conditions can be taken as an indicator of uptake, particularly if the authors intended uptake to indicate bioavailability of Ca and Mg from these foods.Total Ca in diVerent cereal products was approximately 10% of total Mg content, and it was reported that Ca solubility is lower than Mg solubility since Ca complexes are more stable.Not surprisingly, it was found that the solubility of Mg was 3-8 fold higher in HCl-pepsin than in distilled H2O. A conference presentation by Barnett and Horlick304 described the ability of ES-MS to diVerentiate between common valence states of Fe and V, and to measure metals traditionally regarded as 'problem elements' in ICP-MS, including Ca, Fe and Hg. Metal cations were determined following comcomplexes. This method had the added advantage of eliminating the oxide interferences between REEs commonly observed in ICP-MS.The authors used ES-MS to measure Ca levels in drinking water, a commercial Ca supplement and milk, and reported that results agreed well with those obtained by FAES. 3 Three groups have been working on the determination of Cr species in water and dairy products. Sahayam et al.305 measured CrVI in potable waters after selective separation of CrIII using ZnO.Water (250 ml ) was passed through a column containing 2 g of ZnO prepared by calcination of zinc nitrate; 2 ml HNO and 1 ml H2O2 were added to the eluate and the solution was evaporated to approximately 5 ml on a sand- ETAAS. The LOD for CrVI was 0.01 ng ml-1. The amount of bath. The residue was diluted to 10 ml and analysed by CrIII was calculated from the total Cr and CrVI values. A Slovakian group measured Cr in dairy products after preconcentration of CrIII on activated alumina at pH 7 in a flow system coupled online to an AAS instrument.306 Total Cr was determined after electrochemical reduction of chromate ions to CrIII, and both reduction and sorption were carried out in a combined cell containing a porous electrode made of glassy carbon particles coated with gold, a layer of the sorbent and 739 J.Anal. At. Spectrom., 1999, 14, 717-7812 the counter electrode in series. Lameiras et al.307 measured nedithiocarbamate in a medium buVered at pH 4.5 and total Cr and CrVI in UHT milk. Hexavalent Cr was separated extracted into hexane, then derivatized into volatile triin a procedure in which the proteins were precipitated and the phenylstibine, by Grignard reaction with phenylmagnesium supernatant was passed through a Chromabond NH column, bromide.A cool injection programmed temperature vaporizafter which CrVI was eluted with HNO3. Total Cr was deter- ation injector was used, and factors aVecting the extraction mined directly in the milk following addition of a surfactant and a Pd-Mg mixture as the chemical modifier, and both total Cr and CrVI were measured by ETAAS.Detection limits were 0.2 mg l-1 for total Cr and 0.15 mg l-1 for CrVI. Both procedures were validated by the standard additions method and recoveries were higher than 93%. Interference studies were undertaken using a simulated milk matrix to which total Cr and hexavalent Cr were added.The simulated milk was then separated and analysed by the procedures described and absorbance readings were interpolated on the calibration graph. It was concluded that there were no noticeable interferences from the principal constituents of total milk. Methods for the speciation of organotin were reported by 18 three groups. A Japanese team proposed a simple method for determining tributyltin (TBT) and triphenyltin (TPT) in hatchery fish.308 Extracts from the fish were concentrated and purified on an SPE column cartridge containing divinylbenzene-hydrophilic methacrylate copolymer.The sample was hydrolysed with a KOH-ethanol solution and the TBT and TPT compounds were extracted with petroleum ether. The petroleum ether was evaporated and the residue extracted in 50% ethanol solution and loaded onto the SPE cartridge. The column was washed with 10% methanol and the TBT and TPT eluted with HCl-methanol (1+9), then extracted with a hexane-cyclohexane mixture (1+1).After hydrogenation, derivatives of the extract were analysed by GC-quadrupole-MS. The LOD in the samples was estimated as 0.05 mg g-1 for TBT and 0.1 mg g-1 for TPT. Schmitt et al.155 described one- and two-stage procedures for the preparation of biological materials for speciation analysis of organotin compounds by GC-AES and GC-FPD. For the onestage procedure, 0.1-0.2 g of lyophilized material was microwave-digested at 40W power for 3 min at 130 °C with 5 ml acetic acid, 1 ml nonane and 3 ml 2% sodium tetraethylborate.The organic supernatant was analysed by GC-AES or GC-FPD after clean-up on alumina with diethyl ether as the eluent. For the two-stage procedure, the sample was microwave-digested at 50-60 W power for 3 min with 5 ml 25% TMAH. The solution was diluted to 15 ml with H2O, adjusted to pH 5 with acetic acid and extracted with 1 ml 2% sodium tetraethylborate and 1 ml nonane for 5 min.The organic phase was separated and analysed for Sn species. A DB-210 column was used with AES detection and an HP-1 column with FPD. Both methods were validated by analysing a reference fish tissue. Detection limits were 2-10 and 20-40 ng g-1 for AES and FPD, respectively. The one-stage method was applied to the analysis of mussels and sea urchin eggs. A conference report309 described the development of an LC method for the separation of dibutyltin (DBT), TBT, diphenyltin (DPT) and TPT, which is compatible with both ICP-MS and atmospheric pressure ionization (API )-MS.The chromatographic system comprised a Kromasil-100 5 mm C column and a mobile phase of 0.05% triethylamine in acetonitrile-acetic acid-H2O (65+10+25) at a flow rate of 0.2 ml min-1. The system was reported to be compatible with both ICP-MS, which oVers superior sensitivity, and API-MS, which provides intact molecular ion information on the organotin compounds.The development work was carried out on fish tissue and sediment CRMs. Two papers reported advances in the speciation of Sb. A method was described for the selective determination of SbIII in the presence of SbV in spiked tap water by GC-quartz furnace-AAS following derivatization with triphenylmagnesium bromide.310 Trivalent Sb was complexed with pyrrolidi- 740 J. Anal. At. Spectrom., 1999, 14, 717-781 and derivatization conditions, the operating conditions for the quartz furnace and the injector were all optimized.Zhang et al.311 used a miniaturized HPLC column coupled to HGAAS for the speciation of inorganic SbIII and SbV species in spiked water samples. The two Sb species were separated by using 50 mM tartrate solution at pH 5.5 as eluent, and the retention times were 0.5 and 2.8 min for SbV and SbIII, respectively. The hydrides were generated with 3% NaBH4 and 1.8 M HCl solutions.The detection limits achieved were 1.0 and 2.0 mg l-1 for SbV and SbIII, respectively. The author emphasises that to the EU drinking and surface water limits (10 mg l-1), but these LODs are adequate for screening samples with respect are not suYciently sensitive to measure real levels in natural waters. It was also noted that there is as yet no suitable CRM available. Bra�tter and co-workers have used speciation as an analytical aid in trace element research in infant nutrition.38 They studied the binding pattern of trace elements in formula milk as compared with breast milk and the relationship between trace elements in breast milk and maternal dietary intake. Protein separation by HPLC was coupled on line with ICP-MS or ICP-AES for simultaneous speciation to investigate the binding forms of the nutritional elements Ca, Co, Cu, Fe, I, K, Mg, Mn, Mo, P, S, Se and Zn, as well as the heavy metals Cd and Pb.Size exclusion chromatography minimized interactions between the labile metal protein complexes and the column material. It was reported that the binding pattern of trace elements in infant formulae was significantly diVerent from that in breast milk, and was dependent on whether its major component was cows milk or soy, its processing and the chemical form of the added compounds. An investigation of breast milk samples from diVerent regions of the world showed comparable elution profiles and, for Mo and Se, a dependence on the regional maternal dietary intake.Interestingly, the group found significant changes in the Zn-binding pattern as a function of the Se content of breast milk, and citrate was found to decrease with increasing maternal dietary Se intake. It was also noted that formula-fed infants received much higher levels of Fe than breast-fed babies, and the binding forms of Fe were very diVerent in the two milks, which poses an interesting question as tohe relative bioavailability of Fe in these foods which hopefully will be addressed in the near future.Several applications of HPLC separation with detection by ICP-MS and ICP-AES have been reported in conference presentations. It is to be hoped that these contributions to a developing area of research will be published in full. Bantan et al.312 used anion exchange FPLC-ICP-AES for speciation of negatively charged low molecular weight organic complexes of Al.Separations were carried out on a Mono-Q strong anion exchange FPLC column over a wide pH range (3-11) using linear gradient elution (0-100% 4 M NaNO3). Characteristic retention times were 5.5 min for Al-citrate, 6.0 min for Al-oxalate and 2 min for Al-EDTA. The separated Al species were determined in 0.5 ml eluate fractions by ICP-AES. Kos and Hudnik313 described a procedure for the isolation and determination of Cd species in endive grown in Cd-enriched nutrient solution.The Cd species were extracted from the endive with Tris-HCl-phenylmethylsulfonyl fluoride. The clear supernatant was passed down a Sephadex G-50 column which was coupled to ETAAS for on-line Cd measurement. Two fractions were obtained, the first one of which was identified as Cd bound to a high molecular weight protein. The authors suggested that the second Cd fraction consisted of ionic Cdand Cd bound to polypeptides, and discussed some approaches for the examination of this fraction.It is to be hoped that these approaches will be published. Yasui Akemi described a laboratory-built interface for ICP-AES to increase the transfer eYciency of sample solutions.314 The interface, consisting of a thermospray nebulizer, spray chamber and condenser, was used in conjunction with HPLC columns, Shodex IC I-524A and Tosoh G3000SWx1 for the separation of P compounds and Fe compounds, respectively.The P compounds were a mixture of polyphosphonic acids and phytic acid (IP6) hydrolysate with HCl, and ferritin, hemoglobin and iron(II) sulfate were used for the Fe separation. The sensitivity of the ICP-AES using the interface was increased by up to 30 times compared with a conventional concentric nebulizer. Szpunar and colleagues315 used a rapid SEC-ICP-MS method requiring minimal sample preparation to study Pb speciation in 20 wine samples.They reported that Pb was not present in the mineral form, used for assessment of toxicity, but was mostly associated with a biomolecular species (ca. 10 kDa), identified as the dimer of a pectic polysaccharide, rhamnogalacturonan II. Other minor Pb complexes in the range 500-3000 Da were not identified. In a conference presentation Szpunar316 described two separation techniques, SEC and reversed phase chromatography (RPC), for studying protein-bound metals based on the coupling of HPLC with ICP-MS.The tolerance of neuroblastoma cells to Al was studied with SEC-ICP-MS. Two Al-containing low molecular weight (6-8 kDa) compounds were identified in the cytoplasm of neuroblastoma cell cultures which accounted for all the Al in sample. Another application of SEC-ICP-MS investigated the binding of metals by wine proteins, polysaccharides and smaller molecules. Examples of Cd, Cu and Zn associated with metallothioneins were shown, and the bi-dimensional SEC-RPC-HPLCICP-MS speciation analysis of metallothioneins was discussed.Finally, Magnuson et al.317 coupled capillary electrophoresis (CE) to HG-ICP-MS with on-line reduction (OLR) for the determination of four arsenicals and two Se species. Selenate (SeVI) was reduced on-line to SeIV by mixing the CE eZuent with concentrated HCl. A microporous PTFE tube was used as a gas-liquid separator to eliminate the 40Ar37Cl and 40Ar35Cl interferences from 77Se and 75As, respectively.Detection limits for SeIV and SeVI for conventional pressure injection were 10 and 24 pg, respectively. The direction of the electroosmotic flow during CE could be reversed with hydrodynamic pressure, thus allowing increased freedom of buVer choice. The authors reported that dual detection of CE-speciated As and Se compounds had been demonstrated, but emphasised that further investigation was necessary to verify the use of the CE-OLR-HG-ICP-MS system for simultaneous determination of As, Se and other hydride forming species.It is anticipated that the coming year will produce more applications of CE coupled to atomic spectrometry detectors for elemental speciation. 2.3 Developments in methodology for flame atomic absorption spectrometry A method was described for the direct determination of Fe and Zn in cows' milk by FAAS with a high performance nebulizer.318 Samples of whole milk, non-fat milk and whey were analysed without chemical treatment, using a nebulizer comprising a tantalum capillary and ceramic impact bead positioned to optimize nebulization eYciency.Increased sensitivities were reported for both Fe and Zn, with LODs of 0.024 and 0.007 mg ml-1, respectively. The accuracy of the procedure was studied by analysis of non-fat milk CRMs and the method was applied to ten cows' milk samples.Zhang et al.319 reported a method for the determination of Cd by derivative FAAS with a modified water-cooled stainless-steel atom trapping tube in an air-C2H2 flame. The laboratory-made derivative measurement system and atom-trapping equipment, previously described, gave a detection limit and sensitivity improved by FAAS. The LOD (3s) for Cd was 0.02 mg l-1, with a recovery two and three orders of magnitude over those of conventional range of 85.9-114%.The method was applied to measurement of Cd in vegetable samples. 3 Three groups reported FI analysis systems coupled to FAAS. Shuai et al.320 described on-line determination of Pb in rice flour and bee chrysalis (honeycomb?). The sample was digested to dryness with HCl-HNO3, the residue taken up in 1 ml of HNO and diluted to 25 ml with H2O. An aliquot was mixed with 2 ml HCl (1+1), H2O added to 10 ml, plus 5 ml 200 g l-1 KI and made up to 25 ml with H2O.A 7 ml portion was injected into the FI equipment for on-line extraction in a coil with 250 ml of IBMK, followed by separation in a phase separator (diagram shown). The extract was transferred in an sampling frequency was 45 runs h-1 and the LOD was H2O carrier stream for Pb measurement by FAAS. The 2.8 ng ml-1, with RSD (n=11) of 2.5%. Xu and Chen321 dissolved 10 g of cooking salt in 100 ml of 1% HCl. Following further dilution, 150 ml was injected into an FI system (diagram given) in a water carrier before nebulization into the slotted tube atom trap for FAAS determination of Ag, Au, Cd, Co, Cu, Fe, Mn, Ni, Pb, Pd, Sb and Zn.The system was optimized with PTFE tubing (17 cm×0.5 mm id) for linking the sampling valve and nebulizer, and the distance between the slotted tube and 5 cm burner was 3 mm. Sensitivity was enhanced 1.2-3.3-fold compared with conventional FAAS, and the usable lifetime of the slotted tube also increased. The method was applied to water samples, with a sampling frequency of 240 runs h-1, and recoveries for Cd, Cu, Mn and Pb were 90-110% by the standard additions method.A sequential injection (SI ) system was coupled with FAAS for the determination of Ca and Mg in mineral water.322 The SI system comprised a 6-position valve, 3-way solenoid valve and a peristaltic pump fitted with 1.6 mm id PVC tubing. The holding coil was constructed with PTFE tubes (0.8 mm id×200 cm length) coiled on a rod with 1.5 cm od.The FAAS detector was connected to one of the ports of the 6-position valve through a polyacryl confluence device. Small volumes of sample (ml ) were injected into the holding coil between two slugs of 100 mM La. The flow was then reversed and the coil contents were introduced into the air-C2H2 flame. It was reported that interferences were eliminated when La solution frequency was >110 h-1 and reproducibility was better than was injected before and after each sample.The sampling 3%. The method was applied to 15 samples of mineral water and results were compared with conventional AAS analysis. 2.4 Developments in methodology for electrothermal atomic absorption spectrometry Applications of ETAAS with a transversely heated graphite atomiser (THGA) were described in two papers. Subramanian et al.168 developed a procedure for determining low levels of Sb (mg l-1) in water and blood.The method was based on the concept of STPF atomization by using a THGA furnace, longitudinal Zeeman-eVect background correction and matrix modification with Pd(NO3)2-Mg(NO3)2-HNO3. The method did not require the use of standard additions and the LOD (3s of the blank) was 2.6 mg l-1. The THGA-AAS method was reported to be simple, fast and contamination-free because the entire sampling operation was carried out in the same tube.The direct determination of Pb by ETAAS with endcapped THGA was described by a Brazilian group.323 Sweet fruit-flavoured powder drinks, honeys and syrups were analysed without sample digestion and without air-ashing during the pyrolysis step. Samples (4 g) were completely dissolved in 741 J. Anal. At. Spectrom., 1999, 14, 717-7812% lactalbumin. The LODs were 400, 26, 20 and 280 pg, respectively, and results were presented for nine samples of diVerent milk types.2 H2O, acidified and the volume made up to 50 ml to give a of 3%, 6% and 15% H2O2, 1% HNO3, 0.1% phosphate and sugar content of 8%. Aliquots of these solutions (20 ml ) with 10 ml of chemical modifier solution were delivered directly into the end-capped THGA atomizer. For sugar content higher than 8.0%, a build-up of carbonaceous residue inside the Finally, Bettencourt da Silva et al.327 validated the unceratomizer was observed and tube lifetime was reduced (100 tainty evaluation for the determination of metals in solid firings at 16% sugar, compared to 240-260 firings at 8% samples by ETAAS.Lettuce leaves were dried at 60 °C and sugar). Several matrix modifiers were tested and their eVect 200 mg of the dried material was microwave digested with on sample throughput and recovery investigated. The modifier HNO3. Concentrations of Mn in the digests were determined selected [0.05% (m/v) Pd+0.03% (m/v) Mg(NO3)2] increased by ETAAS using a Pd-Mg chemical modifier and D backsample throughput by a factor of 2 and gave recoveries of 90.2-106% with an RSD <5.9%.The LOD of 7.0 ng g-1 Pb ground correction. Assuming 100% digestion eYciency and following the recommendations given in the Eurachem guide attained the Food Chemical Codex recommendation for the ('Quantifying uncertainty in analytical measurement', Version maximum allowed Pb concentration in sugar samples. Imai 6), total analytical uncertainty and uncertainties associated et al.324 examined the eVect of coating a pyrolytically coated with each experimental operation were evaluated.A repeatgraphite furnace with hafnium, niobium, tungsten and zir- 0.82 mg kg-1 lettuce (9 degrees of freedom) against an esti- conium on the injectable sample volume in order to enhance ability test revealed a method uncertainty for Mn of the sensitivity of Pb determinations. Treatment with W was mated uncertainty of 0.73 mg kg-1 lettuce (57 500 degrees of found to give high precision (RSD <3%) for a 100 ml injection freedom) with 95% confidence.Analytical accuracy was of sample solution and good linearity was observed between assessed using spinach leaves treated by a similar method. integrated absorbance and sample volume in the range 0-100 ml. Using a Pd modifier enhanced precision (RSD<2%) 2.5 Developments in methodology for inductively coupled and stability of the integrated absorbance (RSD <2.5%) for plasma mass spectrometry up to 250 firings.The calculated detection limit and characteristic mass (sensitivity) were 0.02 mg l-1 and 12 pg, respectively, for a 100 ml injection with Pd modifier at the optimum ashing temperature of 1400 °C. An interference study was performed and the method was applied to water samples. Chemical modifiers were studied by two groups. A Japanese The popularity of ICP-MS for a wide range of applications has continued during the past year, but the majority of contributions to the literature on this topic has unfortunately appeared as conference presentations, and it is to be hoped that detailed papers will be forthcoming.A conference report by Robb328 described ICP-MS as an invaluable tool for the measurement of trace elements in food. An overview of a wellestablished approach to determination of total element concentrations in food was presented together with the measures required to ensure production of quality multi-element data. The key issues of data handling and results presentation were group investigated the eVect of Ni(NO3)2 as a matrix modifier on the determination by ETAAS of low levels of In in infant formulae and human milk.209 Twice-concentrated samples (400 ml ) were diluted with 200 ml of 4% (w/v) Triton X-100 and 200 ml of H2O.A 10 ml portion of the diluted sample was 5000 mg l-1 Ni in 1 M HNO was injected and the sample was injected into the furnace and dried. After cooling, 10 ml of 3 3 Excellent linearity, stability, spike recovery and detection limits were reported.Information on Pb intake from Ca supplements atomized. Detection limits were 2 and 2.5 mg l-1, with mean emphasised and examples of complex data sets given, with results from several types of investigation. Baxter et al.329 In recoveries of 97.2% and 95.4% in infant formulae and breast milk, respectively.The RSD was 2% at the 50 mg l-1 investigated long term accuracy and precision of multi-element measurements by ICP-MS. Their conference presentation dealt level. Karadjova et al.325 proposed a simple and fast procedure with the behaviour of CRMs measured over two years during for the direct ETAAS determination of Al, Cd, Cr, Cu, Fe, large multi-element surveys (29-36 elements determined in test Mn, Ni and Pb in olive oil.A universal modifer, N,N- samples) of total diet samples, health foods, fish, and snack hexamethylenedithiocarbamic acid hexamethyleneammonium and convenience foods. It was reported that achievable accusalt (HMDC-HMA), was used which had two functions: racy and precision during multi-element measurements varied isoformation of diVerent chemical species of metals present in with element and with concentration. Data were presented olive oil and their thermal stabilization during the pretreatment from analyses by a VG PlasmaQuad 2+ Turbo and a Perkinstep.Various organic solvents (diethyl ether, IBMK, xylene, Elmer Elan 6000. heptane and 1,4-dioxane) were studied as solvents for olive oil Two conference presentations described the determination dilution. It was reported that 1,4-dioxane was the most suitable of Pb by ICP-MS in calcium dietary supplement pills in since it improved the decomposition of triglycerides during to Pb in Ca supplements to 0.5 mg d-1.Sharma et al.150 the pre-treatment step so that aqueous standard solutions response to California Proposition 65 which limits exposure could be used for calibration. Uncoated graphite tubes with described the diYculty in measuring trace metals in Ca-rich platforms were selected as atomizers for the determination of matrices such as over-the-counter Ca supplements, which are Cd, Cu, Fe, Mn and Pb, and pyrolytic graphite coated tubes either derived from natural inorganic CaCO sources such as with grooves were most suitable for Al, Cr and Ni. The paper dolomite, oyster shell, bonemeal, or Ca chelated to organic presents instrumental parameters optimized according to pre- matrices such as gluconate or lactate.Total Pb and Pb isotope treatment and atomization curves, obtained in the presence of ratios were measured in nine supplements by ICP-MS followthe olive oil matrix and with HMDC-HMA as modifier.The method permitted the determination of 0.1 mg g-1 Fe, ing high pressure/temperature acid digestion. Matrix matched 0.05 mg g-1 Ni, 0.02 mg g-1 Al, Cu, Cr and Pb, and 0.01 mg g-1 Pb calibration standards were prepared in high purity CaCO3, and satisfactory recoveries were reported for Ca in an animal Cd and Mn with RSDs of 8-10% for all analytes in this bone SRM. Supplements containing natural Ca sources had concentration range. Vinas et al.326 established optimum con- 8-28 mg Pb per gramme of calcium whilst chelated or refined ditions for the determination of Al, Cr, Mn and Mo in cows' Ca supplements contained 1-2 mg Pb per gramme of calcium.milk, human milk and infant formulae by ETAAS following Source identification by Pb isotope ratios was investigated. A dilution in suspension media specific to each element. Dried similar procedure was proposed by Bakowska149 involving formula milk was dispersed in 25 ml of an aqueous-20% acid digestion followed by ICP-MS analysis against standards C2H5OH suspension medium whilst liquid samples were and a standard blank prepared in CaCO3 of known Pb value.diluted as necessary. For the determination of Al, Cr, Mn and Mo, the suspension media also contained various combinations 742 J. Anal. At. Spectrom., 1999, 14, 717-781was reported. Almeida and Gine�330 described the determination of Pb in vegetables by ID-ETV-ICP-MS. Sample introduction by vaporization from metallic, rather than graphite, surfaces was proposed to replace pneumatic nebulization.Direct nebulization was used to optimize Pb isotope ratio signals for aqueous standards by ICP-MS prior to installation of a tungsten coil (TC)-ETV furnace. The TC was mounted inside a glass chamber which was connected directly to the plasma torch using a tube (7 mm id×750 mm long). A gas mixture of 95% Ar-5% H2 transported the vaporized sample at 1.5 l min-1 to the plasma.It was reported that digested biological samples in a strongly acid medium damaged the tungsten coil, and that best precision was obtained by measuring isotope ratios. The most accurate quantification of Pb by TC-ETV was therefore obtained by ID. It was unclear whether the disadvantages and complexity of using TC-ETV, as opposed to direct nebulization, were really outweighed by the possibility of using minimal volumes (20 ml ) of a very concentrated digest solution, particularly for an RSD of 1-2.5% 100 mg l-1 Pb solution.obtained for 208Pb5207Pb isotope ratios measured in a 2O. The resulting solution was adjusted to pH 0-0.5 with Mestek.334 The sample (25 ml ) was acidified, heated to expel 2, cooled and treated with 1 ml of 0.0005% Y as internal 3, 500 ng Re added as internal standard, known additions were 100-106%, with an RSD of 3.2%. Using A method was reported for measuring ultra-trace elements by coupling IC to ICP-MS.Cao et al.331 described an involved procedure for measuring very low levels of REEs in tea. Samples (0.2-0.3 g) were microwave digested in capped vessels with a mixture of HNO3, H2O2, HClO4 and HF using a threestage programme, repeated three times. After cooling, the digests were washed into Teflon beakers and the solutions from two samples were combined, evaporated to near-dryness and the residue taken up in 1 ml of HNO3 and a few drops of HNaOH for chromatographic separation.The sample solution was then loaded onto a glass column packed with cation exchange resin and the alkaline metal elements were washed out with 1.75 M HCl, followed by the alkaline earth elements in 2 M HNO3. After rinsing the column with H2O and 5 M HNO3, the REEs were eluted with 5 M HNO3 (30 ml ), the fraction collected and evaporated to near dryness. The residue was dissolved in HNO and diluted to 25 ml with H2O prior to analysis by ICP-MS.as polyatomic BaO+ and BaOH+ interference with Eu. It was reported that the method eliminated matrix eVects, such Detection limits were between 0.0039 and 0.0003 ng ml-1 for REEs in solution, with spike recoveries from 90-105% and precision <9% RSD. Two conference reports were used to present work on trace element determinations by quadrupole (Q)- and high resolution (HR)-ICP-MS. Kishiki et al.68 measured 25 elements in an animal tissue SRM by a combination of conventional and HR-ICP-MS.Samples were prepared by microwavepowered digestion in high purity HNO3. It was reported that following reduction to the trivalent state with KI-ascorbic results obtained by HR-ICP-MS agreed well with certified values for all elements except P, Pb and V. Results for Al were lower than the indicative value, probably due to incomplete solubilization. A detector saturation problem was reported for P, and a high background was noted for Ni due to the Ni sampling cone.More detailed investigation for Pb and V was considered necessary. There was some discussion of preliminary work on interferences, both for Q-ICP-MS and HR-ICP-MS. Stu�rup and Larsen described methods for the determination of As, Cd, Cr, Cu, Hg, Ni, Pb, Se and Sn in mussels using either Q-ICP-MS or HR-ICP-MS.332 It was reported that throughput was faster with the Q-ICP-MS since the quadrupole could be scanned faster than the magnet of the HR-ICP-MS instrument.However, HR-ICP-MS showed better signal to noise ratios and therefore had better LODs. The report described the diYculties of measuring As, Cr, Cu, Ni and Se by Q-ICP-MS due to interferences by polyatomic ions. Using HR-ICP-MS, it was possible to resolve the analytical peaks with a resolution setting of 4000 for Cr, Cu, Ni and Se, and 10 000 for As. The results obtained for these elements were in good agreement with certified values of a mussel tissue SRM.It was also possible to identify the polyatomic interferences originating from the sample matrix, and this information could be used to set up correction equations for the Q-ICP-MS. The authors concluded that the most accurate and precise results were obtained by HR-ICP-MS but that the data were obtained at a considerably higher cost, since the instrumentation is more expensive and the sample throughput is lower.Publication of this work in full will make a valuable contribution to the literature on the relatively new technique of HR-ICP-MS. 2.6 Progress in individual elements 2 4 4 2.6.1 Arsenic. Methods for the determination of total As were described by two groups. Madrid et al.333 characterized AFS in conjunction with FI-HG for measurement of As in untreated samples of beer and wine. Several parameters were evaluated for each sample type, including HCl and NaBH4 concentrations and Ar and H flow rates.Optimum conditions were selected as 6 M HCl and 0.5% NaBH for beer and 6 M HCl and 1% NaBH for wine. Three mineralization techniques were tested, and comparison of results for untreated and mineralized samples showed no significant diVerence at the 95% confidence level. It was reported that the method was levels of 2 mg l-1 in wine. A method for determination of As fast, accurate and sensitive and could be used to quantify As in high-saline mineral waters by ICP-MS was described by CO standard and 2 ml CH3OH, then diluted to 50 ml. It was from 40Ar35Cl+ by up to 70%, and the remaining interference reported that the inclusion of CH3OH reduced the interference was corrected for on the basis of the signal ratio of 35Cl+ to 40Ar35Cl+, which was constant over the range 0.4-1 g l-1 Cl.An LOD of 30 ng As l-1 was achieved and recoveries of a calibration graph and the method of standard additions, concentrations of As in a mineral water sample were 1.50 and 1.44 mg l-1, respectively, compared with 4.2 mg l-1 when no correction was applied.Techniques for the separation of arsenite (AsIII) and arsenate (AsV) were reported by three groups. Pozebon et al.335 used an FI procedure for the separation and preconcentration of inorganic As based on complexation with ammonium diethyl dithiophosphate (DDTP) and sorption on a C18 bonded silica gel mini-column. Total As in water samples, a CRM and synthetic mixtures of AsIII and AsV was determined by ETAAS the As3+-DDTP complex was stripped from the solution and acid.Sample aliquots were loaded onto the mini-column and retained on the column, whilst other ions and AsV which did not form complexes were discarded. When 6 ml sample had been loaded onto the column, As3+-DDTP was eluted directly into an autosampler cup (120 ml ), and 0.1% Pd(NO3)2 added as a chemical modifier for measurement by ETAAS.The concentration of AsV was determined by diVerence. Good recoveries of spike were reported (97-108%), and total As concentration was in agreement with the certified value for the SRM. Preconcentration time was 120 s, and 20 samples h-1 could be processed. The LOD for AsIII was 0.15 mg l-1. An analytical method for the determination of AsIII and AsV in natural drinking ws was reported in a conference presentation.336 Solid phase extraction cartridges were used as a low pressure column for separation, with HGAAS and HGAFS for As detection.The best results were obtained using an anion exchange cartridge and 20 mM phosphate buVer at pH 7, 743 J. Anal. At. Spectrom., 1999, 14, 717-781and cation-exchange columns, with mobile phases of phosphate buVer of pH 6 and 10 mM HCl, respectively. The anionexchange column separated AsIII, DMA MMAand AsV, whilst AB and tetramethylarsonium (TETRA) ion were separated on the cation-exchange column, with trimethylarsine oxide (TMAO) and arsenocholine (AC) co-eluting. The As species were oxidized with K reduced to hydrides with NaBH4. The hydrides were carried in Ar to a drying unit and detected at 193.7 nm by AFS.The 2S2O8 with 254 nm radiation, then 2O LOD was 0.5 ng ml-1. For simultaneous GC separation, samples (100 ml ) were reduced with NaBH4 at pH 6, the AsH3 from AsIII trapped on a short column of Chromosorb W AW-DMCS cooled with liquid N2, and thermally desorbed 2 with a step gradient of 5 mM Na2B4O7 (pH 9) for elution of Samples for HPLC analysis were applied separately to anion AsIII and 0.1 M NaHCO AsIII and AsV as low as 2 mg ml-1 were separated within 3 (pH 9) for AsV.Concentrations of 1.2 min, and no interferences were observed from Cd, Cu, Ni or Pb at these concentrations. Higher levels of FeIII aVected AsV detection.Helgesen and Larsen337 investigated the bioavailability and speciation of As in carrots grown in uncontaminated plots and soil contaminated with As and Cu. The As species were extracted from the soil for 1 h by gently shaking dried samples in a 1 mM Ca(NO3)2 solution, centrifugation and separation of the supernatant. A CH3OH-H (1+9) solution was used to extract As from dried carrots with gentle microwave heating. Using a referenced method, extracts were injected onto an anion exchange column and eluted isocratically in 45 mM ammonium carbonate+H O-CH3OH for separate AFS analysis.Calibration graphs were linear up (97+3) solution at pH 10.3 directly into an ICP-mass spectrometer. Although the system was characterized by injecting aqueous standards of MMA, DMA, AsIII and AsV, only AsIII and AsV were detected in the samples. The soil-to-carrot uptake rate (bioavailability) was reported and dietary intakes from the carrots assessed.An interesting conference presentation by Heitkemper et al.338 reported that most of the As in carrots exposed to an organic arsenical defoliant (dimethylarsinic acid) was not inorganic AsIII or AsV. Both IC-ICP-MS and HPLC-ES-MS were used for the identification and quantitation of the major As species found. Results were obtained for total and speciated As, which were compared with those obtained for control samples. Full publication of this work will be a welcome addition to the literature.Lamble and Hill23 described the development of a method capable of the separation of As species by on-line HPLC prior to their on-line decomposition by microwave digestion, pre-reduction with Lcysteine and analysis by HGAAS. Full speciation of AB, MMA, DMA and total As (AsIII+AsV) was achieved in biological samples. to 10 ng in a 100 ml sample, with an LOD of 0.25 ng. The same team used a conference presentation to describe As speciation in wine and urine using HPLC-AFS.343 A novel interface between HPLC and AFS was proposed which comprised photooxidation in the presence of persulfate of the HPLC eZuent followed by HG of the decomposed As compounds.Hydride-forming compounds only (AsIII, AsV, MMAA, DMAA) were detected without on-line decomposition, but UV-decomposition permitted measurement of nonhydride-forming compounds (AB, AC, TETRA, TMAO). Determination of As compounds by ion-pair chromatography was described by van Elteren and Slejkovec.344 Samples were injected onto a cation-exchange column, with 5 mM 3-carboxy- 4-hydroxybenzenesulfonic acid at pH 1.9 as mobile phase, photochemical oxidation with 4% potassium persulfate in 3% NaOH and an 8 W 254 nm UV lamp.Hydride generation was performed in a PTFE mixing coil with 4 mM HCl, 1.5% NaBH4 in 0.1% NaOH, and As detection was by AFS. The method was used to investigate the eVect of food treatment processes (boiling, microwave heating, c-radiation and dry heating) on the stability of As compounds.Partial decomposition was observed for c radiation and dry heating, but it was reported that no health hazards were expected. Two groups reported work on inorganic and organic As compounds in mushrooms. Larsen et al.339 investigated As species in an edible mushroom, Laccaria amethystina, collected from contaminated and uncontaminated locations.The As Finally, Pergantis et al. reported the characterization of species were extracted from the samples using focused micro- arsenosugars by FAB-tandem-MS.345 Arsenosugars are of wave digestion and separated and detected by anion and cation particular interest because they are present in relatively high exchange HPLC with As detection by ICP-MS. Total As concentrations were 23 and 77 mg g-1 (dry weight) in mushconcentrations in seaweed food products, and their biochemisrooms from two uncontaminated sites and 1420 mg g-1 in the try requires further investigation.FAB was used for the eYcient generation of gas-phase ions of the various arsenocontaminated sample. The total As comprised 68-74% DMA, sugar compounds, and these were used to provide structural 0.3-2.9% methylarsonic acid, 0.6-2.0% trimethylarsine oxide, information for characterization of an arsenosugar present in 0.1-6.1% arsenic acid, and the unextractable fraction of As partially purified algal extract.ranged between 15 and 32%. It was reported that mushrooms or their bacteria were able to biosynthesise DMA from arsinic 2.6.2 Iodine. There has been an upsurge in activity this year acid in the soil, and AB, trimethylarsine oxide and other in developing methods for the determination of I. Fecher unidentified As species were detected for the first time in et al.346 described the measurement of I in food samples by Laccaria amethystina.This agrees with work originally ICP-MS after alkaline extraction. A 1 ml volume of TMAH reported in last year's ASU340 (ref. 275), and further described was added to the sample (200-500 mg dry+5 ml H2O, or this year in a conference presentation,341 which reported the 2-3 g wet sample). The sample vessel was closed gas-tight and finding of AB, previously only found in marine biota, in three heated in a dry oven at 90 °C for 3 h. The cooled sample was fungus species.This group extracted Laccaria amethystea and diluted to 25 ml with H2O and any undissolved particles three other wild-growing mushrooms into boiling bi-distilled removed by centrifugation and/or filtration (0.45 mm) to avoid H2O and determined the As compounds by HPLC-ETAAS blocking the nebulizer. After addition of Te as the internal and HPLC-ICP-MS. The mobile phase for HPLC-ETAAS standard, the solution was measured directly or following was 0.003 M potassium hydrogenphthalate solution saturated further dilution.The LOD (9s of the blank) was 30 ng g-1 with Ni(OH)2, and 0.03 M sodium dihydrogenphosphate at for a 100 mg dry sample mass. The method was applied to a pH 6 was used for ICP-MS detection. It was reported that 97% of the As in Laccaria amethystea was present as DMA with 3% as AsIII, with various distributions of diVerent As species in other mushrooms. Two papers and a conference presentation described 4 3 round-robin test on two dietetic child nutrition food samples artificially enriched with I.It was reported that all the I present was extracted completely by the TMAH digestion, and the results compared favourably with those obtained by TMAH-ID, GC, HNO -HClO decomposition and photometric techniques. The TMAH procedure was adopted as a German reference method for determination of I in dietetic foods. Radlinger and Heumann347 compared results for I methods for As speciation with detection by AFS.Slejkovec and van Elteren342 developed a dual As speciation system combining LC and purge and trap GC separation with AFS. 744 J. Anal. At. Spectrom., 1999, 14, 717-781measured by ID-ICP-MS in SRMs (three milk powders and bovine liver) following TMAH digestion (3 h at 90 °C) with those obtained following microwave digestion in HNO3-H2SO4. All digests were diluted with 25 ml H2O for acetic acid-sodium acetate, H2O and 10 g l-1 sodium tetrameasurement of the 127I5129I ratio.The detection limit was reported as 8 ng g-1 for a 0.8 g sample with RSD of 0.6-2.8% ethylborate in 20 g l-1 KOH, and the generated ethylmethylfor 0.1-5 mg g-1 I. Ge�linas et al.210 measured total I in mercury was adsorbed onto a 1.5 cm Tenax trap using N The Tenax tube was connected to a GC system, and the nutritional and biological samples by ICP-MS. The samples ethylmethylmercury was desorbed by heating at 220-250 °C were combusted in a stream of O2, and the combustion for determination on a 15% OV-3 column linked to a 3% products collected in a 5% water-soluble tertiary amine solu- OV-3 guard column with He as the carrier gas and Hg tion.Good agreement with certified I values was obtained for detection by AFS. The LOD was 0.5 pg Hg. six CRMs, although low-level I contamination was present in An interesting paper by Moreton and Delves described a the blank.It was reported that the method was also suitable for the analysis of other trace elements, including Cd, Co, Cu, Mn, Ni, Rb, Pb and Zn. In a conference presentation, Petersen and Larsen348 pro- 3 posed an acid-digestion method for preparing samples for I analysis by ICP-MS, which would overcome the problems of memory and losses during wet ashing caused by the volatility of I species. The sample was placed in a Teflon liner with during which time ashing occurred and I- was oxidized to HNO and HClO4, then pressure-digested for 4 h at 160 °C, in control blood and control urine, and blank samples were prepared by the same method.Samples of fish were cut into 3-. After dilution with H2O, the sample was analysed by small pieces, to which internal standard (Tl ) and concentrated ICP-MS and I as I+ was measured at m/z 127. It was reported IO that blanks were contaminated with I from the Teflon liners, which required rigorous cleaning with TMAH and monitoring prior to analysis to ensure low LODs (5 ng g-1 fresh weight).The LOD was improved by 50% by adding 3% CH3OH to the analyte solutions and increasing the plasma power to 1200 W, due to signal enhancement by introduction of C into the ICP. Calibration was by standard additions to overcome matrix eVects, and the method was applied to measuring I in milk, milk products and fish. 2.6.3 Mercury. As noted in previous years, most of the activity on Hg analysis in this review year has been concentrated on developing methods for the measurement of total Hg.Manzoori et al.349 described a technique for the determination of Hg by CVAAS after preconcentration with dithizone immobilized on surfactant-coated alumina. Both inorganic and methymercury cations were adsorbed from aqueous solution onto a column packed with dithizone immobilized on sodium docecylsulfate coated alumina beads. The trapped Hg was back-extracted into the aqueous phase with 1 M HBr and determined by CVAAS following reduction with NaBH4.Methylmercury and HgII cations were completely recovered from water, with a preconcentration factor of 100. Venthe and Boewe350 reported improved detection limits for Hg determination by conventional HG-ETAAS. Platinum group metals, particularly Ir, can bind metal hydrides and an Ir-coated graphite tube was used to concentrate hydrides of Hg prior to atomization and measurement by ETAAS.The method was also applied to As, Bi, Sb and Se, and LODs of 0.03-0.08 mg l-1 were reported as about one order of magnitude lower than by conventional HGAAS. An automatic mercury analyser (AMA-254) was used for the determination of organic Hg in fish samples.152 Samples were extracted with toluene and then treated with cysteine solution (1% L-cysteinium chloride in 12.5% anhydrous sodium sulfate, 0.775% sodium acetate).After shaking and centrifugation, the solution was analysed by the AMA-254: the solutions were heated and the decomposition products were passed through an amalgamator which collected Hg0. The amalgamator was then heated and the Hg released for measurement by AAS. The instrumental LOD was 0.01 ng Hg, and it was reported that good reproducibility and recovery for CRMs were achieved, especially if a re-extraction step was performed. Methylmercury was measured in fish, shellfish and environmental samples by GC-AFS after aqueous phase ethylation.351 The sample was decomposed in ethanolic KOH at 70 °C, cooled and mixed with concentrated HCl and H2O.Methylmercury was converted to ethylmethylmercury by treating the solution with 2 M2. simple direct technique for determination of total Hg in blood, urine and fish tissue by ICP-MS.151 The method included a novel disinfection procedure for urine and blood by the addition of a virucidal agent, Virkon, to each sample.Since Virkon precipitated blood proteins, subsequent dissolution with TMAH was necessary for these samples. To each solution was then added a sequence of diluents, with mixing between each one. Matrix matched calibration standards were prepared 3 HNO3 were added. After 24 h, appropriate diluents were added to the clear solutions, neutralized with aqueous NH3 and diluted with H2O for measurement by ICP-MS.Wash solutions were optimized and wash-out times were <130 s for Hg concentrations of30 mg l-1. The wash solution for blood and urine was dilute NH -NH4H2PO4 containing EDTA and Triton X-100. The best wash for the fish tissue digests was an aqueous solution of 0.1% (v/v) Triton X-100. Extensive IQC data were reported, and the LODs were 0.13 mg l-1 for Hg in blood, 0.27 mg l-1 for urine, and 0.004 mg g-1 for Hg in fish. The method did not discriminate between inorganic and organic Hg.Two groups used microwave-assisted digestion and GC-AES to determine methylmercury and HgII species in fish. Pereiro et al.352 described a simple and rapid procedure for simultaneous determination of methylmercury and HgII in fish SRMs. Samples were subjected to a rapid (2.5 min) microwaveassisted solubilization in TMAH, followed by simultaneous quantitative ethylation-extraction of the Hg species into hexane and flash isothermal separation using a multi-capillary regular column (100 cm) or minicolumn (22 cm).Using the minicolumn avoided the need for a chromatographic oven, which was replaced by a small thermally insulated compartment, maintained at a constant temperature, and coupled directly to the MIP-AES detector. No dilution of the GC sensitivity. Detection limits for dry mass were 20 ng g-1 (Hg), eluent with make-up gas was necessary to achieve optimum and 80 ng g-1 (methylmercury and HgII).Gerbersmann et al.153 used a similar approach using GC-MIP-AES with focused microwave-assisted sample digestion, but compared two diVerent derivatization/injection procedures. In the first method ethylation was carried out with sodium tetraethylborate, with extraction into hexane and injection via a cooled system. The second method involved hydride generation with NaBH4 and purge-and-trap injection, which was reported to be a new and little investigated technique for the speciation analysis of Hg.It was found that both methods were suitable for measuring Hg in fish samples, and further work was proposed to develop a combined microwave-assisted digestion/ purge-and-trap instrument which could be used for the simple determination of volatile species in solid samples. 2.6.4 Selenium. Most publications in this review year concentrated on techniques for Se speciation. Two papers, however, investigated methods for measuring total Se.Deaker and Maher28 described a digestion procedure for determination of 745 J. Anal. At. Spectrom., 1999, 14, 717-7812 was added 1 ml CuSO4 (15 g l-1 Cu), 40 mg Te and 15 ml SnCl -hydroxylamine hydrochloride (reducing agent). When a precipitate formed, it was filtered oV and the filter and solid were air dried for 1 h, placed on the instrument support and covered with a PTFE disc for XRF measurement of the Se Ka line (2h=31.87°).Background correction was by measuring 2h=31°. The LOD was 0.2 mg Se, withean recovery of 96%, and results were comparable with those obtained by AAS. Two conference presentations by Crews discussed the use total Se in small samples of fish tissue. Freeze-dried samples (<0.1 g) and HNO3 (1 ml ) were placed in 7 ml screw-top Teflon vessels which were then sealed into larger pressure vessels and microwave heated at 600W for 2 min, 0 W for 2 min and 450 W for 45 min.Digests were diluted and measured with a matrix modifier by ETAAS. Complete recovery of total Se from six SRMs was reported, and the procedure resulted in complete recovery of five Se species which were added to marine tissues (selenite, selenate, selenomethionine, selenocystine, selenocystamine). Added trimethylselenonium was not fully recovered (90-101%). Benefits of this method included increased sample throughput, reduced loss of volatile material, and low dilutions. A Chinese group measured total In a very detailed paper, Emteborg et al.359 presented a Se in pork spare ribs and vegetables by XRF.353 Sample (5 g) speciation method based on microbore ion-exchange chromawas digested overnight with 10 ml HNO3-H2SO4, then heated tography coupled to ICP-AES via DIN or coupled to ETAAS.for 4 h at 60 °C. The cooled digest was diluted to 50 ml with Selenomethionine, selenite, selenate, selenocystine and trime- H2O, boiled for 1 h and filtered if necessary.To the filtrate thylselenonium were separated using two diVerent metal-free microbore anion exchange columns. The chromatographic eluate was nebulized into the plasma at a flow rate of 80 ml min-1, and selenomethionine, selenite, selenate and selenocystine were separated within 15 min. Selenocystine was completely retained on the column and was eluted with a plug of dilute HNO3. It was reported that on-line enrichment of selenocystine was quantitative.Selenomethionine and trimethylselenonium co-eluted on the solvent front, and a second mobile phase. Detection limits were between 20 and 38 ng ml-1 column was used to resolve these species without changing the for the five Se species. Four diVerent extraction methods and clean-up procedures were investigated for preparing the sample matrix (CRM 402 white clover grown on Se-enriched soil ), and all stages of the work were checked by ETAAS analysis.of stable isotopes in speciation and bioavailability studies. The measurement of Se isotope ratios in enriched cod was described with a comparison between results from ICP-MS and GC-MS.354 Codfish had been intrinsically labelled with 74Se from two slightly diVerent sources in terms of isotopic abundance. This complicated the mathematical calculation of 80Se following ICP-MS measurements since this isotope cannot be determined because of the isobaric overlap from 40Ar40Ar.Following derivatization using phenylenediamine, 80Se was measured by GC-MS and found to have a fractional abundance of 0.4408, compared to the value calculated from ICP-MS data of 0.4417. Total Se in the sample was 135.6 mg kg-1, with recovery of approximately 100% by ICP-MS, compared with 62-64% from the GC-MS method. The second conference report355 discussed methods for intrinsically labelling foods such as fish, wheat, yeast and garlic, and Se speciation studies on these foods using enzyme extracts and chromatography coupled to ICP-MS.The measurement of blood plasma and urine by HG-ICP-MS was described. Human volunteers were fed Se as supplements to their diet and exchangeable body pool sizes were investigated. It was reported that the observed kinetic diVerences in behaviour of ingested forms of Se indicated that the forms were metabolized diVerently. It is hoped that this work will be published in detail.Speciation of Se compounds by HPLC coupled to ES-MS or ICP-MS was discussed in three conference reports. Larsen356 acknowledged the eYciency, sensitivity and selectivity of coupling HPLC to ICP-MS for Se speciation, but pointed out that identification of separated species depends on relating HPLC retention times with those of available standard substances, and unknown organometallic molecules could not be identified. It was proposed that ES-MS could be used for the detection of molecular ions after their separation by HPLC, and that the combination of ES-MS and ICP-MS holds great potential for speciation analysis.Momplaisir et al.357 described a method based on the separation of organoselenium compounds by HPLC with on-line detection by ES-MS. The organoselenium compounds (including selenocysteine, methyl selenocysteine, allyl selenocysteine, propyl selenocysteine, selenomethionine, selenoethionine, selenocystamine, selenoniumcholine and trimethylselenonium ion) were separated on a C18 or cyanopropyl whilst almost all the laboratory-extracted oils had very low bonded phase column, and under optimum conditions could be detected in the sub-nanogram range.The method was 746 J. Anal. At. Spectrom., 1999, 14, 717-781 applied to a Se-enriched food supplement and a urine SRM spiked with three organoselenium compounds. Plants and nutritional supplements enriched with Se were investigated by HPLC-ICP-MS.358 An ion pair chromatographic method with ICP-MS detection for the separation of selenoamino acid standards potentially present in real samples was presented. Ten selenoamino acid standards could be separated, including cis-trans isomers of Se-1-propenyl-diselenocysteine and selenoxides.Qualitative and quantitative results from the analysis of extracts of Se-enriched yeast-based nutritional supplements and Se-enriched allium vegetables were presented and discussed.2.7 Single and multi-element analysis of food 2 Trace elements were measured in edible oils and fats by four groups. A group at the US Food and Drug Administration reported a method for measuring Cu, Ni and Pb in edible oils by ICP-AES and ETAAS following atmospheric pressure microwave digestion.360 Method detection limits for all elements were equal to or less than 50 ng g-1 with ICP-AES and 30 ng g-1 with ETAAS.Spike and recovery experiments 100 ng g-1 level in soybean oil for Cu, Ni and Pb were between were carried out with corn and soybean oil. Recoveries at the 90 and 106% by ICP-AES, and between 89 and 106% by ETAAS. For corn oil they were in the range 93-103% for ICP-AES and 90-117% for ETAAS. Day-to-day reproducibility was demonstrated by repeat analyses of two spiked canola oils. A method was developed by van Dalen to measure P and S in edible oils and fats by WDXRF.361 The samples were measured in disposable liquid cells after solidification with 15% stearic acid to obtain a stable measurement, and results by WDXRF were compared with data obtained by ETAAS for P, and with an O combustion method for S.The LOD was 2 mg kg-1 for both P and S and the long-term precisions were 3% and 5%, respectively. It was reported that the method was also suitable for the analysis of lecithins. De Leonardis et al.362 measured Cu and Fe in 47 samples of virgin olive oil by AAS in order to evaluate their eVect on oxidative processes.Oil samples were obtained from local mills or extracted in the laboratory from various oil cultivars, and one part of each sample was analysed immediately whilst the other was stored in a sealed glass container in a cool, dark room for one year. Levels of Fe in the mill-produced oils were higher than those produced in the laboratory, although there was less diVerence in Cu levels between the two groups.After one year's storage, the mill-produced oils were highly oxidized, levels of peroxides. Giaccio et al.363 used AAS to investigate a range of elements which may act as catalysts to olive oiloxidation. The content of Co, Cr, Cu, Fe, Mn, Ni and Zn was measured in five varieties of olives harvested at various stages of ripening. The metal levels were found to be variable and there was no clear correlation to literature values for metals in virgin olive oil. 3, Multielement analysis of fruit and vegetables was reported by four groups. Thi Huynh et al.364 measured REEs in rice by both ICP-MS and INAA.Samples for analysis by ICP-MS 2.7.1 Wine and alcoholic beverages. Interest continued in were digested by microwave heating successively with HNO the elemental analysis of wine and alcoholic beverages. H2O2 and HF, and the residue was taken up in H2O. Internal Cabrera-Vique373 measured Pt in wine by ETAAS.Samples of standardization was with Be, In and Re. Two reports by Penuela et al.365,366 investigated the mineral content of frozen 3-H2O2 or dry mineralization (details given). and boiled spinach. Concentrations of Ca, Cu, Fe, K, Mg, Mn, Na and Zn were measured by AAS in three fresh samples of spinach. Each sample was then frozen using a laboratory procedure similar to an industrial process and the samples were stored frozen and studied over four consecutive months.Slight changes were noted in the mineral content throughout the storage period, and some metals increased in the last month of storage. It was reported that there was little diVerence between the mineral content of raw and boiled spinach since the trace elements were not leached into the cooking liquid. Bruggeman and Kumpulainen367 carried out a five-year study on mineral elements in various bread grains, flours and baked breads.They measured Ca, Cd, Cu, Fe, Mg, Mn, Mo, Ni, Pb, Se and Zn by AAS, and reported that individual element concentrations in grains showed only minimal year to year variations. The eVects of freezing and long-term storage on the trace element content of papaya were investigated by a Spanish group.368 Fruit samples from female and hermaphrodite flowers were analysed fresh, frozen and stored frozen for periods up to 1 year, and the elements measured by FAAS were Ca, Cu, Fe, K, Mg, Mn, Na and Zn.It was reported that K was the major element in both types of fruit, and the Na content was found to be diVerent from literature values. The freezing and storage process mainly aVected Zn concentrations in both fruit types, K in female fruits and Na, Mg and Mn in hermaphrodite fruits. Few methods for single element analysis were reported during the review year. Sun et al.369 described a method for determining B in plants and plant-derived foods by ultrasonic nebulization-ICP-AES.Samples were ashed at 550 °C for 4 h, the ash dissolved in 5% mannitol-1 M HNO3, boiled gently and filtered when cool. The resulting solution was introduced into the ICP via an ultrasonic nebulizer for determination of B at 249.773 nm. The method was validated with two SRMs, and results agreed with the certified values. The LOD was 0.5 mg l-1 and recoveries ranged from 93-106%. Plessi et al.370 measured Al by ETAAS and Zn by FAAS in powdered infant formulae and infant food following acid digestion with microwave heating.The LOD for Al was 0.44 ng ml-1. Beverages based on tea gave the highest Al levels, whilst soy-based infant formulae had higher concentrations (7 and 7.8 mg g-1) than milk-based formulae (3-3.5 mg g-1). Formulae based on hydrolysed vegetable protein contained Al levels of 13 and 17 mg g-1. Concentrations of Zn were more variable, but only one sample had a level below the recommended minimum of 13 mg g-1.The results were discussed in the paper. Two groups reported their studies on Ti. A German group used ICP-AES to measure the Ti levels in a range of foods and beverages.371 They reported that on average plants had lower levels than animal-based foods, although some plants showed signs of Ti accumulation. However, since the biological function and environmental distribution of Ti were unclear, the authors admitted that the nutritional and health signifi- cance of the data presented was uncertain. The feasibility of using Ti as a food contact material was the subject of an investigation by Feliciani et al.372 Acetic acid and acetic brine were used as aggressive food simulants to study the behaviour of stainless steel and Ti during exposure times and temperatures to mimic both long- and short-term exposure.Migration of Cr, Mo, Ni and Ti was monitored by measuring concentrations in simulants by ETAAS, and Fe was determined by FAAS.It was concluded that Ti could be regarded as a candidate food-grade material. red, white and rose� wines were prepared by either microwave digestion in HNO Analysis of Pt was at 265.9 nm with deuterium arc background correction. Matrix interferences were due to combinations of elements rather than specific ions, and the two sample preparation methods gave similar results. The LOD was 100 pg Pt, and recoveries ranged from 92.5-102% with RSDs of 7.5-10%.The method was recommended for routine analysis and quality controls of wine and similar beverages. Levels of Pt in French wines were found to be generally the same as those reported for Italian wines. The same group described the determination of Cr in French wine and grapes by ETAAS.374 Chromium was measured directly in wine and in acid-digested wine and grape samples, and no diVerence was reported between the direct and mineralization procedures.The detection limit was 1.8 pg of Cr with mean recovery of 99.3%. The Cr content of 79 wines and 12 varieties of grape was reported and the contribution to dietary intake was discussed. McKinnon and Scollary375 investigated the suitability of ultrafiltration for examining the distribution of metals in diVerent particle size fractions of wine. Bottled wine samples were shaken, filtered through a 0.45 mm cellulose acetate membrane filter and ultrafiltered through Amicon disc membranes with molecular weight cut oVs in the range 100 000-1000.Ultrafiltrate portions were analysed for Al by ETAAS, for Cu and Fe by FAAS, for Pb by stripping potentiometry, and K and Na by flame photometry. Results were interpreted in terms of interactions with potential binding agent. Schwedt et al.376 compared the results of seven diVerent methods for measuring Cu in wine.Detailed analytical procedures were described for Cu trace analysis by FAAS, ETAAS, ICP-AES, a catalytic procedure with 3-hydroxybenzaldehydrazine, urease inhibition with ammonium determination, direct potentiometry with Cu selective electrodes, and inverse diVerential-pulse polarography. The methods were also applied to water and soil leachate samples. Chmilenko and Baklanova377 described a method for de-gassing sparkling wines by ultrasound for AAS determination of metal impurities.Samples of champagne (25 ml ) were agitated at 18 kHz and 1 Wcm-2 for 2 min, then 3 W cm-2 for 2 min, for measurement of Cu (>0.1 ppm), Fe and Zn by FAAS. Treated samples were mixed with Pb(NO3)2-NH4NO3 matrix modifier for determination of Cd, Cu and Pb by ETAAS. An alternative method involved ultrasonic decomposition of 100 ml samples with 20 ml of H2O2. Cations were then extracted with DDC into IBMK for analysis by FAAS or ETAAS.Mercury was determined by CVAAS. The method was reported to be quicker than existing standard methods and gave comparable results. Considerable activity was directed to the measurement of Pb in wine. Kaufmann378 used Pb measurements from 7000 wine samples to identify possible sources of Pb contamination. Using a previously described method which avoids sample preparation, Pb was measured by ETAAS with a Pd modifier. It was reported that atmospheric pollution-related contamination was not responsible for elevated Pb concentrations.The presence or absence of an Sn-Pb capsule (cap covering cork) and its state of corrosion had a very minor influence on Pb levels. A reported positive correlation was confirmed between wine age and Pb concentration, although further 747 J. Anal. At. Spectrom., 1999, 14, 717-781statistical analysis indicated that the most significant contribu- determination of Hg in food, including conditions for GC ting factors were vintage and wine colour.It was concluded measurement of organomercury compounds. that the main contamination source was brass, and brass pipes Metal contaminant levels in peanuts were investigated by and taps were always found in wineries known for elevated two groups. Yankov et al.387 used AES to study the uptake Pb levels. The authors indicated that there has been a continu- and localization of Cd, Cu, Pb and Zn in peanuts produced ing and significant reduction in Pb levels in wine during the in an industrially polluted region of Bulgaria.They reported past seven years. Liu and Jiang379 discussed the eVect of that Cu, Pb andn accumulated selectively in the underground matrix, hydride generation and ion interference during determi- parts of the plant, whilst Cd was concentrated in the seed in nation of Pb in wine by HG-AFS. High sensitivity and selectivity were reported, with a detection limit of 0.2 ng ml-1.concentrations exceeding the legal limits, and concluded that heavy metals enter the peanut plants either through the root Chinese white wine was directly sampled for measurement of system or as an aerosol through leaves and stems. Cadmium Pb by ETAAS.380 Ashing was carried out at 600 °C for 10 s, levels in Australian peanut products were investigated by and atomization at 2000 °C for 4 s, with background correction Tinggi.388 Samples of peanut butter, raw, roasted and crushed using a deuterium lamp.The matrix was eliminated by washing peanuts from various commercial producers and local markets the sampler nozzle with C2H5OH. Using the method of were pre-digested in a microwave system prior to dry ashing standard additions, recovery of Pb was 90-108%, with RSD of 1-3.7%, and the detection limit was 31 pg. 2O and a reagent plug of the disinfection of drinking water. However, a disinfection Two publications described the elemental determination in beer by AAS.Fernandes et al.381 developed a method for measuring high levels of Ca, K, Mg and Na by FES and AAS. Two manifolds were designed to carry out substantial dilutions required to bring samples within linear concentration ranges. Samples were diluted on-line with H LaIII solution (1% 0.6 M HCl, 0.12 ml ). The methods were reported to give good agreement with results obtained by manual dilution, with a sample throughput of 120-200 samples h-1, and a significant reduction in consumption of expensive LaIII reagent.Fernandes and Rangel382 described an FI system for determination of Cu in beer by AAS using the method of standard additions. A manifold, based on the 'merging zone technique', was used to prevent the burner from clogging. Results obtained using five standard additions were comparable with those from a reference method, the LOD was 5 mg l-1, and the sampling rate was about 30 samples h-1.2H5OH. A bromide ions were separated and identified in an aqueous The sampling rate was increased to 75 h-1 by using only two standard additions, but results were less precise and accurate. An FI manifold was coupled with ICP-MS for the measurement of 27Al, 75As, 138Ba, 114Cd, 53Cr, 63Cu, 55Mn, 208Pb, 88Sr, 238U and 64Zn in vodka and brandy.383 Small volumes of C2H5OH-H2O solutions (10-40%) were injected into the plasma via the FI manifold.It was observed that this caused a change in ionization conditions, and the incident power was increased to 1.6 kW. Transient signal behaviour was studied for the 11 elements by increasing the C2H5OH content in the samples, and it was noted that the most aVected elements were As, Pb, U and Zn. There was no correlation between ionization potential and signal change in the presence of C rapid method was described by the US Bureau of Alcohol, Tobacco and Firearms for the determination of Cu in malt beverages without preconcentration.384 Samples containing up to 7% alcohol by volume were diluted five-fold with deionized water and analysed directly by ETAAS, with NH4H2PO4 as the matrix modifier.The LOD was 5 mg l-1 in the malt beverages and analytical results were verified by standard additions and spike recoveries (95-108%, RSD <10%). Copper levels between 22 and 86 mg l-1 were measured in 20 US drinks, with levels of 26-79 mg l-1 determined in 12 imported malt beverages.2.7.2 Metal contamination in food. Increasing awareness of food quality issues and proposed legislation setting limits on levels of toxic elements in food have resulted in a flurry of work to determine metal contaminants in a range of foodstuVs. Two reviews appeared during the review year on methods for the determination of Cd, Hg and Pb in foods. Janin and Schnitzer385 discussed ashing, AES, AAS, ICP-MS and other procedures, with 22 references.Boisset386 presented 29 references and discussed spectrometry and other methods for the 748 J. Anal. At. Spectrom., 1999, 14, 717-781 to 0.031 mg kg-1, were lower than the maximum permitted for Cd analysis by ETAAS. Mean values, ranging from 0.013 0.05 mg kg-1. concentration for Cd in peanuts in Australia, currently set at Two conference presentations and three papers described the determination of bromate in drinking water. Ozonation has been successfully used as an alternative to chlorination for by-product (DBP) of the ozonation of bromide-containing source water is bromate, a potential carcinogen for which the WHO has established guidelines levels in ozonated water.Creed and BrockhoV389 reported that IC-ICP-MS had excellent sensitivity and selectivity as a separation and detection system, whilst allowing the use of ID analysis. The method LOD for direct analysis of bromate was 0.3 mg l-1.Argonbased spectral interferences on mass 81 were discussed and matrix-based interferences which were chromatographically resolved from bromate were tentatively associated with phosphate and sulfate. The chromatographic resolution of bromate from other brominated DBPs was described, and results obtained by ID and direct analysis in water samples were compared. Cooney and Browner390 also used IC-ICP-MS to measure trace levels of bromate in drinking water, but reported that, since the positive mode ionization eYciency of Br was fairly poor, larger than normal chromatographic injections were necessary to obtain a measurable signal when using conventional nebulizers and lower liquid flow rates.The development of an oscillating capillary nebulizer in their laboratory resulted in an increase in transport eYciency when using lower flow rates, thereby eliminating column concentration techniques and large injection volumes.Bromate and solution at a concentration of 10 mg l-1 each using IC-ICP-MS in the positive ion mode, and the LOD was 2 mg l-1. However, further investigation was required to overcome co-elution of bromate with bromoacetates, and to reduce the LOD to submg l-1 levels. Nowak and Seubert described a method in both English391 and German392 for the ultra-trace determination of bromate in drinking waters by microbore column IC coupled on-line with ICP-MS.Water samples (0.885 ml ) were injected onto a column packed with a self-made high capacity, high performance anion exchanger [5 mm poly(styrenedivinylbenzene)-DMEA]. Adsorbed ions were eluted with 4NO3 at pH 6 containing 20 mg l-1 Sr as internal 60 mM NH standard. The eluate was nebulized directly into an ICP-MS instrument for detection of 79Br. A clear separation was achieved between bromate and other bromine species, such as bromide and bromoacetic acid.The LOD was 50-65 ng l-1 in samples with an RSD of 5% at a bromate concentration of 500 ng l-1, and no sample pretreatment such as matrix elimination or trace enrichment was required. Each complete analysis took between 10 and 15 min, depending on the bromide content of the sample. Elwaer et al.393 developed a method for measuring bromate in drinking waters using FI-ICP-MS.They identified a need to improve method sensitivity substantially in view of EC, US EPA and WHO recommendations for a maximum contaminant level of 10 mg l-1 of bromate in drinking water.An FI system with a microcolumn packed with a strong anion exchanger was interfaced to an ICP mass spectrometer to carry out on-line separation of bromate and bromide in drinking water samples. The eluent and carrier was 50 mM malonic acid and good separation of bromate and bromide was reported. The presentation included aspects of method development, including operating parameters and the eVect of other co-existing ions on analyte recovery and discussed a survey of bromate in process and mineral waters. There was continued interest in monitoring the levels of toxic elements in infant foods and breast milk.Lutter et al. carried out a survey of toxic metals and radionuclides in Kazakhstani breast milk.55 The data presented were from a larger study designed to provide a scientific basis for the development of a national infant feeding policy.Sample collection procedures were described in detail; two independent analytical techniques (ICP-MS and NAA) were used to verify the results, and good agreement was found between the two methods. The major elements Fe and Zn were measured by ICP-AES and 137Cs by a Ge detector. Median concentrations of toxic elements were within the range reported for other countries, and 137Cs was not detected in any samples. All the essential trace elements, major minerals and electrolytes were within reference values, and as a result of the study the Ministry of Health for Kazakhstan is promoting breast feeding.Baum and Shannon394 used AAS to measure the concentration of Pb in home-prepared reconstituted infant formula. Forty concentrations above the current action level of 15 mg l-1 infants were evaluated and two of the 40 samples had Pb required by the US Environmental Protection Agency for safe water.These samples contained 17 and 70 mg l-1 and were prepared with cold tap water from houses >20 years old. It was concluded that the use of infant formulae requiring reconstitution may present an inadvertent Pb hazard to young infants. Salvato395 investigated Cd levels in samples of liver, meat, fish, rice and cocoa powder and in infant foods containing these ingredients. Samples were either acid-digested or ashed and Cd was determined by ETAAS with tube and L'vov platform. The highest Cd levels, found in liver, exceeded the limit of 0.09 mg kg-1, whilst Cd was not detected in meat and fish.Concentrations in rice and cocoa powder were in line with the safety limits of 0.1 and 0.3 mg kg-1, respectively, and Cd levels in baby foods were largely within acceptable tolerances. Two papers by Danev et al. investigated residues of environmental contaminants in meat and other foods. Cadmium levels were measured by ETAAS in beef from animals grazed close to lead and zinc smelter plants in Macedonia.396 Samples of muscle, liver and kidney from 14 animals were analysed, and results showed that 29% of muscle samples, 86% of liver samples and 100% of kidney samples were contaminated above the limits set by the national regulations. In a separate paper,397 the same team reported that levels of both Cd and Pb were above the current legal standards in a range of animal tissues (cattle, sheep, goats, wild birds) and foods (sheep cheese, eggs, honey) due to contamination from smelters.A Polish group398 studied the pollution contamination of flour by Al, Cd, Four papers discussed dietary intakes of Cd and Pb. Oral Cu, Mn, Mo and Zn. Zinc was measured by FAAS and the Cd exposure of adults in Germany was discussed by Mu� ller other elements by ETAAS, following mineralization in et al.406 using market basket calculations. Representative foods HNO3+HClO4 (4+1).A wheat flour CRM was used to from 11 categories were purchased in 1988 (864 individual validate precision and accuracy of the data, and the method was applied to samples of potato, rye, corn and two wheat flours. The Pb content of edible wild mushrooms was used as an indicator of environmental contamination in north-west Spain.399 Levels of Pb were measured by ETAAS in 95 samples of 13 species, and the average Pb concentration of the samples was 1 ppm dry weight.Contamination due to traYc pollution was particularly obvious in the species Coprinus comatus, which had the highest Pb concentration (10.43 ppm) in samples collected in a city centre, and it was reported that this species could be considered as an indicator of Pb contamination. Angelova et al.400,401 used AAS to measure Cd, Cu, Pb and Zn in grapes (skins, pulp, seeds and stalks) harvested from two regions with diVering degrees of heavy metal pollution.Most of the heavy metals were concentrated in the stalks, and least in the pulp. Significant diVerences in amounts of the heavy metals were noted for skins and stalks from the two regions, but not the pulp and seeds. Cyhexatin (tricyclohexyltin hydroxide) is a non-systemic acaricide used to control the motile stages of a wide range of phytophagous mites, particularly in fruit. Anderson et al.402 used ICP-MS as an initial screen for total Sn levels to indicate the possible presence of pesticide residues in apples, pears and below 0.06 mg kg-1, equivalent to the study's target reporting kiwi fruit.Background levels of Sn in these fruit were normally limit for cyhexatin of 0.2 mg kg-1. A total of 72 samples were screened, of which three apple samples contained Sn at >0.06 mg kg-1, and were re-analysed by GC-MS as a con- firmatory method for cyhexatin. It was reported that the ICP-MS preliminary screen reduced the number of samples requiring analysis by GC-MS, with a cost saving in staV hours of approximately 30%.This screening method could potentially be applied to other organometallic pesticide residues. 2.8 Dietary intake studies There has been much interest during the past year in studying dietary intakes of minerals and particularly the toxic elements. Tripathi et al.403 investigated As intake by the adult population of Bombay City, using HGAAS to measure As in air, duplicate diets and body fluids.The LOD was 0.02 ppb and SRMs were used to assess data reliability. It was reported that dietary intake was the major contributor to total As intake, and that a mean blood As concentration of 1 mg dl-1. The daily intake the turnover time in blood was approximately 33 d, based on of As by the adult population of Bombay was much lower than the WHO recommended value of 140 mg. Daily dietary Rb intake in Belgium was evaluated by duplicate portion sampling.404 Samples were digested in a microwave oven and Rb was measured by AAS.It was reported that the mean daily intake (2.2±0.3 mg) was similar to levels in most other countries. However, it is unclear whether Rb is an essential element for humans, so the intake level could not be assessed with respect to a recommended range for safe and adequate dietary intake. Llobet et al.405 used ICP-MS to estimate dietary intake as part of a survey of environmental pollution and human exposure to metals in Tarragona, Spain.Food items from 11 food groups were analysed for As, Be, Cd, Cr, Hg, Mn, Ni, Pb, Sn, Tl and V, and it was found that Be, Tl and V were below the detection limit for all food groups. Dietary intakes of the remaining metals were established by a total diet study, and results agreed well with literature values from recent surveys in diVerent countries. It was reported that dietary intakes of As, Cd, Hg and Pb should not present a health hazard for the population of the Tarragona area.samples of 94 foods) and 1991 (642 samples of 105 foods), and prepared for cooking but left raw. Samples were dried, dry-ashed at 450 °C and the ash taken up in 2% HCl. Cadmium was measured by ETAAS using a boric acid modifier method (referenced ), which gave an LOD of 0.08 mg l-1 and a mean Cd concentration of 33.3 mg kg-1 dry weight. The average 749 J.Anal. At. Spectrom., 1999, 14, 717-7813 daily intake of Cd was reported to be 10-14 mg, which represents 16-19% of the provisional tolerable weekly intake (PTWI), and there was no diVerence between Cd intake in 1988 and 1991. There was no toxicological risk from orally ingested Cd. Kim et al.407 investigated the Cd exposure level from traditional food in the North-West Territories of Canada. A risk assessment was based on 24 h dietary recalls, traditional food use frequency and Cd levels measured in acid digests of about 30 species of wildlife and plants by FAAS and ETAAS.The range of Cd concentrations was between 0 and 1869 mg per 100 g, with lowest levels in fruits and the highest in caribou and moose oVal. Average Cd intakes were estimated to be no greater than 9% of the WHO PTWI (400-500 mg), although further clarification was necessary regarding health implications for frequent consumers of caribou and moose.Tahvonen408 investigated dietary intakes of Cd and Pb from beverages in Finland. The Cd and Pb contents of carbonated beverages, juices, beers and wines were quantified by ETAAS using Zeeman background correction, peak height mode and the method of standard additions. Wines we diluted with 10% HNO and other beverages were digested in concentrated 3. The highest level of Cd (2 mg l-1) was found in a HNO Pb concentration was 34 mg l-1 in imported red wines.It was domestic red wine made from raspberry, and the highest mean concluded that beverages contributed a negligible amount of Cd and Pb to the average Finnish diet, although beverages accounted for about a fifth of the total Pb intake. Mustafa et al.409 measured 11 elements by AAS in 20 diVerent species of vegetables commonly consumed in Bangladesh. They reported that concentrations of Cd and Pb were below the tolerable limit recommended by FAO/WHO and did not pose any health risk for consumers.Results of the other elements were also discussed. Dietary intakes of essential elements and micronutrients have also been considered. The eVect of energy restriction on the mineral content of diabetic diets was investigated by a Japanese group.410 Diabetic diets (1200 kcal and 1840 kcal ) oVered in a hospital were analysed by AAS to quantify Ca, Cu, Fe, K, Mg, Mn, Na and Zn. It was reported that the Fe level in 1840 kcal diets was 6.60±1.44 mg d-1, which is much lower than the recommended value.The Cu, Mg and Zn levels were also low. All elements measured were even lower in the 1200 kcal diet, and it was reported that these results were considered to be characteristic features of low-energy diets. Biego et al.411 measured the mineral contents of diVerent kinds of milk and estimated dietary intake in infants from birth to three months. Levels of Al, Ba, Cd, Cr, Cu, Mg, Mo, Ni, Pb, Sn and Zn were measured by ICP-MS following microwave digestion in acid of six milks (cow's milk-based formula, breast milk, soya milk, bottled milk, dried milk and evaporated milk). Soya milk contained the highest levels of Al, Ba, Cu, Mg, Mo and Ni, and except for Ni in soya milk, the dietary intakes of minerals were below or close to those recommended by the FAO/WHO.Daily dietary Fe intake in Belgium was evaluated by Van Cauwenbergh et al.412 using duplicate portion sampling. Samples were heated in a microwave oven and Fe was measured by AAS.The adult Fe intake (11.3 mg d-1) was above the recommended daily allowance of 10 mg d-1 for adult males, but below the value of 15 mg d-1 for adult women. Two papers reported dietary intake of Mo, an essential micronutrient for which the physiological requirement in adults is approximately 25 mg d-1. Holzinger et al.413 investigated the Mo intake of adults in Germany and Mexico by means of duplicate portion analysis using ICP-AES.Results showed that the Mo intake had increased from 1988 to 1996, and diVered for German adults depending on location and the kind of diet, with vegetarian women consuming 179 mg d-1, compared with 89 mg d-1 by women on mixed diets. There was a significant diVerence in Mo consumption between 750 J. Anal. At. Spectrom., 1999, 14, 717-781 German and Mexican adults, with Mexican women consuming on average 162 mg d-1, whilst Mexican men consumed 208 mg d-1 compared with 100 mg d-1 for German men.It intake, with no risk of exceeding the toxic dose of 150 mg kg-1 was concluded that Mo requirements were met by normal body weight. Daily dietary Mo intake in Belgium was studied by Van Cauwenbergh et al.414 Duplicate portion samples of meals, drinks and between meal snacks were collected during seven consecutive 24 h periods from four locations. The meals were cut up, inedible portions discarded and the remainder homogenized. Portions were freeze-dried and acid digested in a microwave oven prior to analysis by AAS with a Pd-Mg(NO3)2 matrix modifier.The dry weight LOD was 0.3 ppb and accurate results were obtained for two CRMs. The results were compared with other published data. 2.9 Characterization studies Characterization of wines according to their trace element composition was investigated by four groups. Baxter et al.415 reported a method for determining the authenticity of wines from England and Spain.Wine samples were diluted 1+1 with dilute HNO3 containing In as internal standard. Portions were injected into a carrier stream of 0.5% HNO3 containing In and 5% C2H5OH for ICP-MS analysis. Detection limits were 10-100 ppt, recoveries were 100±25%, and accuracy was tested by measuring CRMs. The data set (48 trace element concentrations in 112 wines) was subjected to discriminant analysis.Spanish wines were classified into their regions of origin, and English and Spanish white wines were diVerentiated from each other by their trace element profiles. When red and rose� wines were included in the Spanish set, English and Spanish wines were diVerentiated with 95% accuracy. Greenough et al.416 investigated the relationship between wine composition, vineyard and wine colour by elemental fingerprinting using ICP-MS. Thirty-three elements were measured in 17 white and 10 red wines from Canadian wineries.It was reported that wines from grapes from the same vineyard tend to be most similar, regardless of vintage, grape variety or processing winery, implying that the element fingerprints were derived soil signatures. Discriminant analysis indicated that combinations of approximately six elements from several geochemical groups could accurately classify the wines according to vineyard.Twenty-six elements (Al, As, B, Ba, Ca, Ce, Co, Cr, Cu, Fe, Ga, La, Mg, Mn, Mo, Ni, P, Pb, S, Sb, Sn, Sr, Ti, U, V, and Zn) correlated strongly with vineyard of origin, although the limited number of wines available for this study required that further work be done to confirm these conclusions. A Brazilian group attempted to characterize wines from Argentina, Brazil and Uruguay according to their mineral composition.417 A total of 31 wines were analysed by AAS, FES and colorimetry, and the results were interpreted by PCA.It was reported that the elements which demonstrated the highest discriminant eVect were Fe, K, Li, Mg, Mn, Na, P and Rb, and it possible to identify wines from each country. Sun et al.418 used ICP-AES to measure Al, B, Ba, Ca, Cu, Fe, K, Mg, Mn, Na, P, Rb, Sr, V, and Zn in samples of 17 wines from six regions. Results were analysed by a three-layer artificial neural network (ANN) model with background propagation of error, and compared with those obtained by cluster analysis, PCA, the Bayes discrimination method and the Fischer discrimination method.An ANN method gave 100% accurate classification of wine samples on a regional basis, and was considered to be more accurate than the other procedures. Multivariate characterization of wine vinegars from southern Spain was described by Guerrero et al.419 Forty wine vinegars were analysed for mineral content using FAAS for Ca, Cu, Fe, Mg, Mn and Zn, FES for K and Na, and HGAAS for As.Several pattern recognition techniques wereapplied to the data, and the best discriminant elements were Fe, Mg, Mn and Na. The samples were successfully divided into two groups (slow and quick vinegars) according to the rate of their fermentation. 25H2 ratios at the methyl and methylene sites of OYcial Methods to include soy-based, whey-based and enteral formulae. Four groups studied the characterization of other beverages according to their trace element content.Martý�n et al.420 measured 11 metals (Ba, Ca, Cu, Fe, K, Mg, Mn, Na, P, Sr and Zn) by ICP-AES (conditions given) to study 41 samples of arabica and robusta green coVee varieties. Cluster analysis and PCA were applied to the results, and it was reported that significant diVerences were found in the metallic content of the two varieties. The largest diVerences were found in levels of Cu, Mn and P and these were used to characterize the coVee varieties, with Cu and P higher in robusta, and Mn higher in arabica coVee.A preliminary study to classify tea by geographical origin was undertaken by Marcos et al.421 Fifteen types of tea from ten diVerent countries were digested in HNO3, and the trace element content measured by ICP-AES (Al, Ba, Ca, Cu, Fe, Li, Mg, Mn, Sr, Ti and Zn) and ICP-MS (Cd, Co, Cr, Cs, Hg, La, Li, Nd, Ni, Pb, Pr, Rb, Se, Sn, Ti, V and Zr).The data obtained were analysed by UNSCRAMBLE, a chemometrics package incorporatPCA, and the results indicated that a distinction could be made between African and Asian and between Chinese and Asian teas. The authenticity of Brazilian orange juice was determined by isotopic analysis.422 Samples of authentic orange juice were analysed by site-specific natural isotopic fractionation nuclear magnetic resonance (SNIF-NMR) to determine D C2H5OH produced by fermentation of the orange juice.Stable isotope ratio mass spectrometry (SIRMS) was used to determine the ratio of 13C512C in the C2H5OH and the ratio of 18O516O in the citrus juice water. The mean ratios for the authentic samples were used to determine whether sugar had been added or retail samples had been diluted with tap water. 2.10 Reference materials and collaborative trials An interlaboratory study of an ETAAS method for the determination of Pb in wine was conducted by Brereton et al.423 Seventeen laboratories from France, the United States and the United Kingdom took part in the study, using a variety of ETAAS instruments.The method incorporated a novel matrixmatching procedure to minimize matrix eVects between Table 1 Analysis of clinical and biological materials, foods and beverages Matrix Element Tissues Ag Ag Ag Biological materials Blood, serum, urine Wheat Al Serum Al Serum Al Technique; atomization; analyte form* AA;F;L AA;F;L AA;ETA;L -;-;- AA;ETA;L AA;ETA;L Sample treatment/comments Very high concentrations were found in specimens from an unusual case of fatal Ag poisoning.Samples were digested with HNO and H2SO4 Ag was extracted from sample solutions with 3,6-dithiaoctane (8-2S). A single drop nebulization technique was employed for atomization Samples were heated with H2SO4, diluted with H2O and then KI-20% ascorbic acid was added.This mixture was extracted with IBMK and the organic layer used for measurement of Ag (in Chinese) 26Al was used to study cellular transport of Al. The authors concluded that diVerent transport mechanisms were present in those wheat varieties that exhibited Al toxicity compared with those that did not Distribution of Al in serum of occupationally exposed subjects was investigated by three techniques: SEC using a Sephadex G-100 SF column; HPLC with an SEC TSK G 4000 SW column; ultrafiltration with a 10 000 cut oV Centricon membrane 0.5 ml serum, 0.5 ml HNO and 0.2 ml HClO were sealed in PTFE vials and subjected to microwave heating.After evaporation, the residue was dissolved in 0.1 M HCl, and the analyte preconcentrated by addition to an ion exchange column. The eluted Al was transferred to the graphite furnace 2 samples and standards. Good agreement was reported between (24-279 mg l-1) obtained by ICP-MS, and the method was results obtained from participants and target values recommended for use for oYcial purposes.Venkatesh424 described a collaborative study intended to extend the variety of reference materials certified for I. Three diVerent mineralization methods were developed to measure total I by ICP-MS in biological and nutritional materials. A mixture of water soluble tertiary amines was used as the matrix solution for two O combustion methods and a simple extraction at room temperature.Results for the two combustion methods agreed well for concentrations >0.1 mg kg-1. The amine extraction method gave the most precise and accurate results for the mixed diet, milk powder and infant formula samples, but recoveries were low for most of the biological materials owing to incomplete extraction and solubilization of I. The ICP-MS data were compared with NAA results. A collaborative study organised by the US Food and Drink Administration was reported by Cook425 to test modifications to AOAC OYcial Methods.Eight laboratories participated in a study of OYcial Method 985.35 for the analysis of Ca, Cu, Fe, K, Mg, Mn, Na, P and Zn by AAS in ready-to-feed milk-based infant formula and pet foods, and seven laboratories tested OYcial Method 986.24 for the measurement of P in milk-based infant formula by a spectrophotometric method. The samples analysed included soy-based and whey-based infant formulae and a casein-based enteral formula.On the basis of the results, which were tabulated in detail, it was decided to extend the Shrimp and plaice reference materials produced by the Danish National Food Agency were characterized for trace elements and two As species.235 The materials were prepared by dissection, drying, milling and sieving to collect particles less than 150 mm in size. Total trace element concentrations were determined by ETAAS, CVAAS, ICP-MS and ID-ICP-MS.Arsenobetaine (AB) and tetramethylarsonium (TETRA) ion were determined by cation exchange HPLC coupled with ICP-MS or with ion spray MS-MS for verifi- cation. Following rigorous statistical analysis of the analytical data, it was decided to assign certified values for As, Cd and Hg in the shrimp, and As, Hg and Se in the plaice. Indicative values were given for Cr, Pb and Se in the shrimp material, and AB and TETRA ion in both RMs.Ref. 273 3 426 272 427 164 428 4 3 751 J. Anal. At. Spectrom., 1999, 14, 717-781Table 1 (Continued) Serum Al Serum Al Serum Al Al Serum, amniotic fluid, tissues Serum transferrin Al Canned soft drinks Al Fruit juice Al Lymphocytes Al Tea, coVee Al Al Wild mushrooms, cultivated mushrooms Fruit juice, milk Al Al Infant formula, infant food Seafood, meat Al Serum Al Al Blood, plasma, tissues Serum Al Blood, tissues Al Al Blood, urine, tissues Urine Al Urine Al Blood, urine Al Tissues Al Brain Al Hair Al Hair Al Al Dialysis concentrates Tea Al 752 J.Anal. At. Spectrom., 1999, 14, 717-781 AA;ETA;L AA;ETA;L AA;ETA;L AA;ETA;L AA;ETA; HPLC AA;ETA;L AA;ETA;L AA;ETA;L AA;ETA;L AA;ETA;L AA;F;L AA;F;L AA; ETA;L AE;ICP;L MS;ICP;L MS;-;S MS;ICP;LC MS;-;S MS;-;S AA;ETA;L AE;ICP;L MS;-;S MS;-;S SIMS; -;S AA;F, N2O-C2H2; L AE;F, 2O-C2H2;L N AA;ETA;L AE;ICP;SEC 8 Serum proteins were precipitated by the combined eVects of 1% TCA and microwave radiation.The deproteinized, low acidity solution was analysed with a simple fast heating programme. There were no matrix 162 0.5 ml serum and 0.5 ml 0.5% Triton X-100 was mixed with 1 ml 0.1 mg ml-1 interferences K2Cr2O7. The sample was atomized from tube previously treated with 50 ml 216 10% ammonium molybdate (in Chinese) Al, as aluminoxamine, was separated from other species in the serum ultrafiltrate by HPLC The results revealed transplacental passage of Al after cutaneous administration 429 to pregnant mice Binding of Al to transferrin in the presence of diVerent amounts of Fe was studied 430 431 432 The eVect of can lining defects on Al concentrations in carbonated drinks was investigated using ETAAS with a L'vov platform and Mg(NO3)2 as modifier.Levels were some 20-fold higher in the defective cans Various acids were discussed with respect to the analysis of trace (nanogram) levels in juices and fruit syrups. A closed PTFE bomb in conjunction with 166 433 HNO3-H2O2 was proposed (in Polish) Al was measured in B and T-lymphocytes from subjects on haemodialysis to investigate the immunosuppression which features in Al toxicity Sample was dried at 105 °C, ashed at 450 °C and then dissolved in HCl prior to 3)2 as chemical modifier.The LOD 434 measurement using ETAAS with Mg(NO was 26 pg A large number of mushroom species were assayed for their Al content. The authors concluded mushrooms do not contribute to human Al intake 298 370 Al and Pb were simultaneously preconcentrated on activated C after complexing with cupferron Samples were microwave digested with 3 ml of 70% HNO3 as the digestion acid. With closed vessels and a 3-step programme digestion was achieved in 9 min. Al was determined by ETAAS, Zn by FAAS.The highest Al levels were found 291 in products based on hydrolysed vegetable protein Lyophilized sample was digested in closed vessels with HNO3-HF, then in open 2O2. Excess HF was removed using H3BO3. Matrix eVects and 71 vessels with H spectral interferences were assessed. HF was shown to be essential for complete recovery of Al Samples were diluted with H2O for Q-ICP-MS and double-focusing sector field ICP-MS.Interferences were removed by using Be and Sc as internal 89 standards Al from a digestion solution was precipitated with 8-hydroxyquinoline. This complex was heated to 800 °C to form Al2O3. Al in urine was precipitated with Na3PO4, complexed with acetyl acetone and Al2O3 similarly formed. The Al 163 2O3, mixed with Ag powder, was taken for AMS Proteins were separated by Fast Protein LC on a MonoQ (HR5/5) anion exchange column and the Al measured in the eluent by Q-ICP-MS and double-focusing ICP-MS.In unspiked samples, Al was bound only to transferrin Al metabolism was studied in rats. 26Al administered by oral gavage was 85 86 determined by AMS Al adjuvents, included in vaccines, were labelled with 26Al and the absorption was measured. Al was measured by AMS Rabbits loaded with Al were used to assess the eYcacy of oral chelating agents. 4359 Possible alternatives to i.v. desferrioxamine were identified It was shown that trace elements may be lost from specimens by precipitation.Methods to ensure re-dissolution were given. Standard additions calibration was necessary to achieve accurate results 26Al was measured by AMS in specimens from volunteers who had received 84 88 experimental doses in biokinetic investigations 26Al was measured by AMS in specimens from mice to investigate tissue distribution and uptake mechanisms 26Al was measured by AMS in specimens from rats to investigate uptake and 87 accumulation in the brain as a mechanism of Alzheimer's disease The hair was ashed and discs formed for irradiation 436 437 Washed hair was digested with HNO and H2O2 and diluted with H2O.The flame conditions were optimized; the 3 LODs were 0.25 and 0.07 mg ml-1 for AAS and AES, respectively (in Chinese) 438 Al was separated from the high salt content of the samples by 5-fold repetitive passage through a 50 mm3 column of 100-200 mesh H+ form of Chelex-100.The Al was then eluted for measurement and the LOD was 0.2 mg l-1 Al species in tea infusions were separated on a Sephadex 12HR10/30 column 440 with spectrometric detection. Al in the infusion was determined by ICP-AES. The same species were found in teas of diVerent originsTable 1 (Continued) As As Biological specimens Human milk, cow's milk Urine As Calcium carbonate As Urine As Urine As Urine As Urine As Urine As Urine As Urine As Waters As Urine As As Grain, cereal products Wine, urine As Drinking water As Drinking water As Mushrooms As As Biological and Food CRMs Foods As Beer, wine As Potable water As Water, urine As Seaweed As -;-;- AA;-;L AA;Hy;FI AA;Hy;L AA;Hy;L AA;Hy;L AA;ETA;L AA;Hy;L A test was devised to evaluate body load of As by administering the chelating agent 2,3-dimercaptopropane-1-sulfonic acid and measuring AsIII, MMA and DMA in urine collected for 2 h AA;Hy;L As species were separated by ion-exchange chromatography.Specimens from subjects living in areas with naturally high drinking water As concentrations were examined AA;Hy;HPLC Specimens were collected from subjects with skin cancer and who lived in an area with high concentrations of As in drinking water. AsIII, AsV, MMA and DMA were determined AsIII, MMA and DMA were separated on a 10% polyphenyl ether GC column which was part of a system which included a series of low temperature traps to collect H2O vapour released from the reaction vessel.Atomization occurred in a heated quartz tube. Samples were from subjects with occupational exposure and from rats administered As compounds AA;Hy;HPLC An FI system was elaborated which included HPLC separation of As species, UV photooxidation with 4% K2S2O3 in 1 M NaOH in a knotted reactor and continuous HG AA;Hy;FIA An FI manifold which included UV photo-oxidation to breakdown organoarsenic compounds, including arsenobetaine, was used. The LOD was 0.1 mg l-1 Inorganic As species were determined by complexation between AsIII and diethyl dithiophosphate, preconcentration onto a C autosampler cup and quantitation by ETAAS with a Pd modifier.Total As was determined directly by ETAAS and AsV obtained by diVerence AA;ETA;L AE;ICP;L AF;Hy;HPLC AA;Hy;L AA;Hy;L AF; Hy;HPLC AF;Hy;HPLC AA;Hy;L AA;Hy;HPLC On-line HPLC-microwave digestion-HGAAS was proposed for speciating As without the need to invest in more expensive instruments The eVect of various food treatments-boiling, microwaving, irradiation, etc.,- on As compounds was investigated.In some cases degradation was observed, but no health risks were thought to be present AF;Hy;L AF; Hy;L Hy;-; HPLC MS;-;- A review of methods for analysis and speciation 441 442 Total As was determined in the breast milk of 35 lactating women and were 4.219±0.079 and 4.932±0.38 mg l-1 in breast and cow's milk, respectively milk from 36 cows.All samples originated in Izmir, Turkey. The levels 24 443 As species were separated using Dowex 50W-8X-AG1 X8 ion-exchange column. The eluted species were reduced downstream by 15% KI-5% ascorbic acid and then by 1% NaBH4. AsIII, AsV, MMA and DMA were measured. A second stream of the FI manifold carried unseparated sample through a microwave oven for digestion, reduction and measurement of total As Arsenic in CaCO3 used in food products was analysed by dissolving the sample in a minimum volume of H2O then digesting with concentrated HCl.At 193.7 nm the LOD was 0.6 ng ml-1 (in Chinese) 173 Samples were diluted 1+9 and incubated for 1 h with 1% cysteine in 0.03 M HCl. A commercial FI system was then used to measure AsIII, AsV, MMA rates, 4 etc., were optimized and the LOD was 1 mg l-1 and DMA.Conditions of NaBH concentration, acid concentrations, flow 444 180 181 175 25 26 335 18 172 445 3minicolumn, elution into an HCl, KI and sodium hypophosphite were added to urine to form 'iodide arsines' of AsIII, MMA and DMA, which were extracted into toluene and back extracted with 1 mM NaOH. As from dietary sources was not measured Samples were microwave digested using HNO -HCl-HClO4. Hg was also 343 446 336 342 determined (in Chinese) A novel HPLC-HGAFS system was described in which photo-oxidation in the presence of persulfate was used to decompose As species in the column eluent, allowing both reducible and non-reducible species to be determined 3 multivariate calibration functions were employed-classical least squares, inverse least squares and Kalman filtering-but no significant diVerences were observed for recoveries of AsIII, AsV, MMA or DMA Anion exchange and reverse phase solid phase extraction cartridges were used at concentrations above 2 mg l-1 interfered in the AsV analysis as low pressure chromatography columns for separating AsIII and AsV.FeIII A dual system composed of (a) HPLC followed by photo-oxidation at 254 nm, then HGAFS in tandem with (b) HG-cryogenic trapping/thermal desorption- AFS gave results similar to HPLC-ICP-MS for 8 As species. LODs were of the order of 0.5 ng ml-1 23 217 333 447 Operating parameters for FI-HGAFS on untreated samples were described in a conference presentation. Results compared favourably with mineralization procedures and allowed 2 mg l-1 to be determined in wine LODs of 10 and 20 ng g-1 were obtained for As and Se, respectively 174 UV photolysis coupled to HGAAS allowed the measurement of a range of species at LODs of approximately 5 mg l-1.Operating parameters for HPLCHGAAS were also reported FAB-MS was used to characterize arsenosugars 345 753 J.Anal. At. Spectrom., 1999, 14, 717-781Table 1 (Continued) Drinking water As As Drinking water, urine and citrus leaf RMs Mushrooms As Urine As Urine As As Animal feed additives Carrots As Carrots As Fish, shellfish As Urine As Urine As Urine As Urine As Urine As Urine As Serum As As Blood, urine, tissues Chinese medicines As As Chinese herbal medicines B Biological tissues B SRMs B Tissues B Blood Beverages, foods Ba Tissues Ba Cod muscle Bi 754 J. Anal.At. Spectrom., 1999, 14, 717-781 MS;ICP;CE MS;ICP;HPLC Using microbore reverse phase ion-pairing HPLC-ICP-MS absolute LODs in MS;ICP;HPLC As species in 4 types of wild mushroom were determined using HPLC coupled AA;ETA; to the above named techniques. Inorganic arsenic and DMA were the main HPLC forms found MS;ICP;HPLC Increased concentrations of As species were found in urine specimens from subjects living in an area with old mines MS;ICP;ETV Urine was pipetted into an Ir-coated graphite tube.The analytes were vaporized into the ICP and the concentrations calculated by standard additions MS;ICP;HPLC An unusual As speciation study in that the species studied were the phenylarsonic acids added to poultry feed as growth promoters. It was necessary to use microbore HPLC to separate the additives from naturally occurring species present.The technique was proposed as a tool for investigating metabolites and degradation products MS;ICP;HPLC In contrast to the report in ref. 337, HPLC-ICP-MS studies of carrots found to MS;ICP;L MS;ICP;L AE;ICP;L MS;ICP;L MS;- ;HPLC AF;-;HPLC MS;ICP;HPLC of seaweed products AF;Hy;HPLC AF;Hy;HPLC AF;Hy;HPLC AA;Hy;L AF;Hy;L AF;Hy;L AE;ICP;HG -;-;- AE;ICP;L AE;ICP;L AE;ICP;L AE;ICP;L XRF;-;S AA;ETA;L 317 Coupled CE-HG-ICP-MS was used to speciate AsIII, AsV, MMA, DMA, SeIV and SeVI.A microporous PTFE tube was used as a gas-liquid separator to eliminate interferences from 40Ar37Cl+ and 40Ar35Cl+. A compromise was necessary between complete reduction of SeIV and achieving optimum CE peak shape 448 the fg range were achieved for As and Sn species. Using the same separation approach coupled to ES-MS-MS up to 10 As species could be separated. The selectivity of this method allowed co-eluting As species to be measured 341 179 64 449 338 contain 0.6-0.8 mg g-1 As revealed the species present were not inorganic in nature.The source of the As was believed to have been DMA used to treat the crop growing in the field prior to the planting of carrots 337 An interesting study of As uptake from experimental plots containing diVerent soils artificially contaminated with As was reported. Carrots did not grow in soil with As concentrations above 400 mg g-1, but did grow in soils containing up to 338 mg g-1, although with reduced growth as As concentration increased.Only inorganic As species were detected in the carrots 450 171 A conference report described the application of open focused microwave preparation for extracting As species. Great care in selection of power and exposure time was required to prevent change of species 2 to the Ar plasma, or the addition of C2H5OH and Te internal 451 452 176 177 Addition of N standard to samples, gave accurate, precise results Structural information concerning As species was obtained using HPLC-MS with electrospray ionization Eleven As and four Se species were separated and determined after consumption Metabolism of As found in drinking water and seafoods was investigated.An on-line microwave heating arrangement, prior to HG and AFS, was used New, short HPLC columns made possible the rapid separation of AsIII, AsV, MMA and DMA with LODs of 1 ng ml-1 Various columns and conditions were investigated to develop a procedure for 453 178 the separation of As species present in urine following ingestion of arsenosugars in seaweed As binding proteins and As species in uraemic sera were separated by SEC, ionexchange or aYnity chromatography.Eluted fractions were digested prior to measurement of As. DMA concentrations were increased in specimens from patients receiving dialysis The hydride was formed after acid digestion (in Chinese) 454 455 3 456 Samples were digested with HNO and HClO4.Thiourea and ascorbate were added followed by 5% HCl. The solutions were sampled into an FI manifold for HG. The LOD was 0.1 ng ml-1 (in Chinese) Leaves, etc., were boiled in H2O, cooled, acidified with HCl and mixed with KI solution. An FI manifold was used to cause reduction with NaBH4 (in Chinese) 184 369 A number of analytical methods were reviewed.It was concluded that ICP-MS is the method of choice Samples were ashed, the ash dissolved in 5% mannitol-1 M HNO3, boiled gently and allowed to cool. The resulting solution was directly aspirated, yielding an LOD of 0.5 mg l-1 185 186 457 100-400 mg samples were prepared with an automated microdigestion procedure. An LOD of 0.01 mg ml-1 was reported After high temperature alkaline ashing, samples from animals fed boric acid throughout pregnancy were analysed A detailed survey, conducted in 1988, 1992 and 1996, of Ba intake in 16 test populations in Germany and Mexico was reported.One of the German groups was vegetarian and this group consumed twice as much Ba as the others. However, no adverse health eVects were predicted (in German) 458 301 Gunshot residues were analysed in frozen tissue specimens to help demonstrate entry wounds and shooting distances Diethyldithiophosphate complexes of Bi were formed in 0.5-4% HNO3 and adsorbed onto the walls of a knotted PTFE reactor.The complex was then eluted using 30% HCl directly into the cuvette of an ETAA spectrometer. During the ETAAS determination the next preconcentration step occurred. The LOD was 3 ng g-1 for 60 s loadingTable 1 (Continued) Plasma Bi Urine, medicines Bi Br Drinking and mineral water Water Br Water Br Drinking water Br Br Blood, plasma, urine, saliva Blood Br Plasma, urine CCa Plasma, cell membranes Cereals Ca Mineral water Ca Maize flour Ca Biological samples Ca Bone Ca Beverages Cd Blood, foods Cd Foods Cd Drinking water Cd Infant foods Cd Oranges Cd Peanut products Cd Tap water Cd Tea Cd Cd Traditional Canadian foods AA;Hy;L AA;Hy;FI the residue for HG and the LOD was 13mg l-1 (in Chinese) After digestion with acid an FI manifold was used to generate BiH3.Interferences were masked by addition of thiourea-KI.The LOD was 320 pg ml-1 MS;ICP;HPLC 2 very similar papers from the same co-workers reporting the use of microbore MS;ICP;HPLC Bromate was determined in ozonated drinking water using ID and direct HPLCICP-MS, yielding an LOD of 0.3 mg l-1. The ID procedure was compromised MS;ICP;HPLC A further paper on using HPLC-ICP-MS to speciate BrO3- in water that had by interference from the 40Ar dimer on mass 81 MS;ICP;L XRF;-;S XRF;-;L MS;-;S AA;-;- AA;-;L AA;F;L AA;F, air-C2H2;Sl MS;ICP;L MS;-;S AA;ETA;L AA;ETA;L MS;ICP;L AA;ETA;L AA;ETA;L AA;ETA;L AA;ETA;L AA;ETA;L AA;ETA;L AA;ETA;L AA;ETA;L 183 Plasma samples were digested with HNO and HClO4. 10% HCl was added to 146 391, 392 species was reported. The LOD at m/z 79 was 50-65 ng l-1 (in German) column HPLC-ICP-MS to determine bromate. No interference from other Br 389 390 been disinfected with ozone.A novel device, the oscillating capillary, was used to overcome the need for injecting large sample volumes. The LOD was 393 2 mg l-1 and separation of BrO3- and Br- was achieved within 3 min 3- and Br- were determined by FI-ICP-MS using a microcolumn packed 105 BrOwith an anion exchanger. Malonic acid was used as carrier and eluent NaBr was given to human volunteers in experiments to estimate extracellular water 16 ml blood was placed into an aluminium cup to a depth of 20 mm.The 104 459 concentration of Br in specimens of normal blood was 2.5-11.7 mg l-1 AMS was used to provide very sensitive measurements of 14C, to follow the metabolism of labelled drugs Ca concentration changes during pregnancy were studied. Perturbed Ca homeost- 196 asis was observed in pre-eclampsia Ca and Mg were determined in wheat, barley and oats. Bioavailability of the 303 322 Ca and Mg were determined by FI-FAAS at LODs of 0.03 mg l-1 for both elements was assessed using a synthetic gastric digest elements.The sampling rate was 110 h-1. Lanthanum solution was injected before and after each sample to remove interferences 460 Powdered maize (0.1 g) was suspended in 0.15% agar and a 5 ml aliquot then mixed with 0.8 ml 0.1% agar-2 ml of 5% La3+ solution. An appropriate Ca standard was added, the solution diluted to 50 ml and Ca determined using standard additions. At 422.7 nm the LOD was 0.12 mg ml-1 (in Chinese) 83 90 Samples were dissolved in acid and analytes separated from the matrix by liquid- liquid extraction.Isotope ratio measurements were achieved with doublefocusing sector field ICP-MS Bone biopsy specimens were examined by AMS, using 41Ca to investigate the kinetics of bone resorption The results of a comprehensive study of Cd and Pb in Finnish beverages-beer, 408 461 carbonated drinks, juices, wines-was reported. The samples studied were found to make a negligible contribution to the diet To compare the performance of the named techniques, 418 samples of diet homogenate and blood were collected from Chinese and Japanese women and analysed.The ICP-MS results were 10-20% lower for Pb in blood and diet and Cd in blood than the ETAAS results, although equivalent results could be obtained following an unexplained mathematical correction 406 462 German Cd intake was calculated from market basket studies in the period 1988 to 1991 20 ml of sample was introduced into the autosampler without use of a modifier solution.No ash step was used to increase the rapidity of analysis. Using Zeeman-eVect ETAAS, Cd and Pb were measured at 228.8 and 283.3 nm yielding LODs of 0.05 and 0.8 mg l-1, respectively (in Chinese) 395 463 3 388 294 4. Following Samples were either mineralized with oxidizing acids or ashed at 500 °C prior to analysis using ETAAS with a L'vov platform.Neither baby foods nor selected raw materials gave cause for concern Samples were dried, ground, digested using HNO -HClO4, diluted to 25 ml with H2O and 10 ml taken for analysis by ETAAS at 228.8 nm. The LOD was 0.02 ng ml-1 (in Chinese) Levels of Cd were found to be between 0.013 and 0.031 mg kg -1 in a range of products purchased in Brisbane, Australia Water (150 ml ) was adjusted to pH 2 then mixed successively with 4 ml anion exchange suspension, 1.5 ml 20 nM APDC and 1.5 ml 3 M NaClO adjustment to pH 6 the solution was filtered.The filter was placed in a beaker, 1 ml of 0.1 M HCl containing 100 mg Pd applied to the filter and the beaker ultrasonically irradiated for 1 min; 20 ml of the suspension were taken for ETAAS. The LODs were 0.17 and 5.7 ng l-1 for Cd and Pb, respectively 464 Sample was charred, ashed at 550 °C, decomposed using HNO -HClO cooled, diluted in 0.5% HNO3, mixed with 0.2 mg l-1 PdCl 3 2-0.2 mg l-1 4, 4+1, 407 Mg(NO3)2 and Cd measured using platform ETAAS (in Chinese) Cd exposure from traditional foods, e.g., moose, caribou, was evaluated.Highest levels were found in liver and kidney of the named animals. The relatively high consumption of these organs could pose health eVects for a sub-group of the population studied 755 J. Anal. At. Spectrom., 1999, 14, 717-781Table 1 (Continued) Endive Cd Milk powder Cd Mussels Cd Vegetables Cd Urine Cd Hair, blood Cd Urine Urine, blood, hair Cd Cd Plasma, urine Cd Urine Cd Biological samples Cd Food Cd Urine Cd Urine Cd Cd Urine, faeces, intestinal fluid Liver, kidney Cd Kidney Cocaine, heroin Cd Cd Pharmaceuticals Cd Teeth, bones Cl Serum proteins Co Cinnarizine Co Preserved foods Co Milk products Cr 756 J.Anal. At. Spectrom., 1999, 14, 717-781 AA;ETA;LC AA;ETA;Sl AA;F, air-C2H2;L AA;F, air-C2H2;L AA;F;HPLC AA;ETA;L AA;ETA;L AA;CV;L AA;-;L AA;ETA;HG MS;ICP; FI-HG AE;ICP;ETV AF;Hy;L MS;ICP;L MS;ICP;ETV MS;ICP;SEC XRF;-;S XRF;-;S AA;ETA;L AA;F;L AE;DCP;L MS;-;S MS;ICP;LA AA;-;- AA;ETA;L AA;-;L 313 Endive was grown in Cd-enriched nutrient solutions.Following harvesting and extraction Cd species were determined using gel filtration LC coupled to ETAAS via a flow through cell. The results were used in an evaluation of the biochemical pathways that confer heavy metal tolerance on certain 465 4H2PO4, which also acted 158 Sample was suspended in a 1 mg ml-1 solution of NH plants as chemical modifier.Good recoveries were obtained for both Cd and Pb (in Chinese) An FI-FAAS system utilizing preconcentration on a minicolumn packed with poly(aminophosphonic acid) was described. Elution with concentrations of HCl above 0.5 M reduced recovery. Using smallest possible elution volumes resulted in an LOD of 0.56 mg l-1 319 191 122 A water-cooled stainless steel atom trap gave an LOD of 0.02 mg l-1 for a 1 min collection time.A laboratory made derivative measurement system was connected to the AA spectrometer A thermospray nebulizer to connect the HPLC column to the flame was designed and evaluated. It was reported that the sensitivity, at 3.7 pg ml-1, was 104 better than with conventional FAAS A tungsten coil atomizer was used to measure Cd after HNO3 digestion in a microwave oven.Because of vapour-phase interferences phosphate-based modifiers were not successful and 15 mg l-1 Pd solution was used Another description of a portable instrument with a tungsten coil atomizer 117 466 467 188 Samples were in thiourea, Ni, or thiourea-Ni-based media. An FI arrangement mixed the sample with 0.23 M HCl and then with NaBH4. The resultant vapour was fed to a heated quartz tube for atomization Metallothionein proteins were separated and quantified using covalent aYnity chromatography with thiol-disulfide interchange gels and AAS 2O: the hydride was trapped in the 12 Vesicular HG was utilized with urine and H graphite furnace for ETA giving an LOD of 10 ng l-1; the FI system had an LOD of 7 pg with a 50 ml sample 4)2HPO4 and 10 mg added to 2 468 0.2 M H2SO4.A portion was then treated with 30g l-1 KBH4 containing 5 Dried samples were powdered together with (NH a tungsten cuvette superimposed on a tungsten boat furnace. 80 ml TMAH was added and the cuvette heated to 130-150 °C. The temperature was further increased in stages to pyrolyse, ash and vaporize. The atomic vapour was transferred to the ICP in a stream of Ar and H The sample was low temperature ashed, the residue evaporated using H2SO4 and then ashed again at 600 °C. The residue was then shaken with 25 ml 0.2 M 2SO4, 5 ml of 0.5 g l-1 dithizone in CCl4 and diluted to 50 ml with further Hg l-1 KOH and analysed by AFS.The LOD was 0.12 mg l-1 190 3 67 3 Samples were analysed following 1+9 dilution with 1% HNO containing Rh as internal standard. Correction factors and instrumental conditions were established to minimize NaCl eVects 1% HNO was used as the chemical modifier in the ID measurement of Cd and Pb in SRMs and fresh urine specimens Porridge prepared from wheat intrinsically labelled with 106Cd was fed to infants 193 98 99 195 3 and adults in Cd balance studies In vivo measurements of liver and kidney provided LODs of 30 and 3 mg g-1, respectively.Results correlated poorly with individual Cd exposure estimates Accumulation in the kidney of smelter workers was monitored Solutions were prepared by dissolution in HNO and dilution with H2O. in cocaine and heroin at 5-45.6 and 10.2-192 mg kg-1, respectively (NH 143 4)2HPO4 was added to a concentration of 0.2% m/v.Cd was measured Solutions of drug compounds or extracts from pharmaceutical preparations were mixed with either [Cd(SCN)4]2- or [Zn(SCN)4]2- at optimized pH and then filtered. The Cd or Zn was measured in the filtrate to calculate the amount of metal taken for the formation of the insoluble ion associate. This provided an indirect determination of the drug concentration 91 58 145 A pyrolytic technique was developed and shown not to introduce contamination.Isotopes of Cl and I were extracted for measurement of 36Cl and 129I by AMS. Results were used to show exposure to radionuclides Proteins were separated by immunoelectrophoresis on agarose gels. The distribution of Co among the proteins was shown by LA-ICP-MS Powdered pharmaceutical preparations were dissolved in C2H5OH, diluted with HCl, filtered and treated with Co tetrathiocyanite solution. The complex was extracted with nitrobenzene and Co measured in the organic phase for the indirect determination of cinnarizine 469 306 Co, Cr and Ni were determined at 242.5, 357.9 and 232 nm, respectively, using ETAAS with pyrolytically coated tubes and a L'vov platform.A fast temperature programme obviated the requirement for an ashing stage Cr was determined by AAS after on-line preconcentration of CrIII on an activated column of alumina. The Cr was determined after electrochemical reduction of CrVI to CrIII.The original CrIII in the sample was also determined (in Slovakian)Table 1 (Continued) Blood Cr Foods Cr Grapes, wine Potable water Cr Cr Preserved foods UHT milk Cr Cr Urine Cr Urine Cr Hair, sweat, serum Cr Semen Cr Cr Biological specimens White asparagus Cr White asparagus Cr Cr Cr Biological specimens Body fluids and tissues Cr Biological materials Tissues Cr Green vegetables Foods Cu Cu Virgin olive oil Cu Cu Cu Serum proteins Plasma, urine Malt beverages Cu Water, wine Cu Plasma, urine Cu Urine Semen Amniotic fluid Cu Cu Cu Cu Biological specimens Tissues Tissues Cu Cu Cancer tissues Cu AA;ETA;L AA;ETA;L AA;ETA;L AA;ETA;L AA;ETA;L AA;ETA;L AA;ETA;L AA;ETA;L AA;ETA;L AA;ETA;L AA;ETA;L AA;F, air-C2H2;L AA;F, air-C2H2;L MS;ICP;L MS;ICP;L MS;ICP;ETV LMMS;-;S AA;-;- AA;-;- AA;-;L AA;-;- AA;-;L AA;ETA;L AA;F;L AA;ETA;L AE;ICP;L MS;ICP;L Samples were diluted 20- and 10-fold in 0.05% HNO3 for plasma and urine, respectively. 89Y was added as internal standard. 1-2 ml samples were taken using an FI-DIN AA;ETA;L See Cr, ref. 472 AA;ETA;L See Cr, ref. 473 MS;ICP;HPLC Metal binding proteins were separated by SEC. Caeruloplasmin was shown to AA;ETA;FI AA;F;L AA;F;L XRF;-;S 470 471 3-H2SO4 (6+1) and digested overnight. HNO3 Contamination from three types of venepuncture needles was assessed and found to be absent or negligible.The analytical method used had an LOD of 0.3 mg l-1 Samples were mixed with HNO was added dropwise and refluxed until the appearance of white fumes. The digests were diluted to final volume with H2O. The LODs were 1.1 and 1 mg kg-1 for Mn and Cr, respectively, based on 2 g of sample in a final 374 305 volume of 20 ml The Cr content of 79 wines and 12 grape varieties was determined Water, 250 ml, was passed through a 1.5×0.1 cm column containing 2 g ZnO which retained the CrIII. HNO3-H2O2, 2+1, was added to the eluate and the solution evaporated to 5 ml on a sand-bath.Following dilution to 10 ml Cr and CrVI were determined by ETAAS. The LOD was 0.01 ng ml-1 for CrVI. 469 307 2 CrIII was determined by diVerence See Co, ref. 469 CrVI was determined by protein precipitation and then elution through an NH 3. Without the precipitation step recoveries were less than 25%.The highest level found was 1.2 mg l-1 column using HNO 472 198 Specimens from subjects with occupational exposure were acid digested with microwave heating Reference concentrations in Italian subjects were determined. The LOD was 0.05 mg l-1 and results were from none detected to 0.24 mg l-1 Age-related concentration decreases in specimens from more than 40 000 subjects 138 473 3 at 85 °C, diluted with H2O and analysed.were reported The semen was heated with HNO Concentrations were not related to sperm quality 123 474 Various chemical modifiers were investigated for the measurement of Cr in acid digested materials, using a tungsten coil atomizer. Pd was the most eVective Dried, homogenized sample was sequentially ashed, dissolved in H2O-65% HNO3-30% HCl, 2+1+1, evaporated to dryness, the residue ashed at 460 °C and the resulting ash redissolved in the acid mixture. At 357.9 and 232 nm 475 the LODs were 0.04 and 0.6 mg g-1, respectively The same research group as in ref. 474 reported the eVect of freezing and storage on Cr and Ni distribution. Statistically significant diVerences were found for Ni levels Carbon-based species which interfere with the measurement of 52Cr and 24Mg 50 51 were removed when a cool Ar-ICP was used Ways to reduce polyatomic interferences were developed. Vapour-phase acid The ArC+ eVect on 52Cr was corrected mathematically.Accurate results were digestion limited contamination and interferences associated with C and Cl. 197 obtained with CRMs A Mo or W metal filament was used for ETV to avoid the introduction of C and the formation of C-containing polyatomic interferences Various Cr species were identified in tissue sections with mm resolution. Samples 199 476 477 were irradiated at 266 nm and the species were detected by TOF-MS Cu and Zn in vegetables commonly consumed in Ghana were determined Human diets from 12 locations in Austria were measured for their Cu and Ni content (in German) The influence of Cu and Fe on oxidation of 47 oil samples was evaluated 362 (in Italian) The distribution of Cu among serum proteins was determined See Cd, ref. 467 203 467 384 Samples were diluted 5-fold and analysed directly using ETAAS with NH as the chemical modifier. The LOD was 5 mg l-1. The maximum level found 4H2PO4 376 was 86 mg l-1 The above techniques and potentiometry, diVerential-pulse polarography and catalytic procedures were compared for determining Cu (in German) 56 472 473 478 32 be a major carrier for Pb An FI manifold was constructed which included a 'knotted reactor' for the adsorption of an APDC chelate. With a 30 s sampling time the preconcentration step produced a 44-fold enhancement factor and an LOD of 6 ng l-1 The samples were from a case of fatal Ag poisoning.See Ag, ref. 273 Samples from sheep with perturbed Cu status were digested with HNO 273 240 3 4 and HClO Biopsy materials were analysed by TRXRF (in Japanese) 200 757 J. Anal. At. Spectrom., 1999, 14, 717-781Table 1 (Continued) Vitamin tablets Cu Serum Cows milk FF ee Serum Fe Lung cells Fe Fe Histidine preparations Virgin olive oil Biological samples FF ee Serum Ga Blood Ge Hair Hg Hg Fruit, vegetables, cellulose Fish Hg Hg Deer liver and heart Fish, hair Hg Tap water Hg Hair Hg Hg Grain, cereal products Hair Hg Hg Biological specimens Biological samples Hg Hg Blood, plasma, urine Tissues Hg Hg Histidine preparations Fish Hg Yeast cells Hg Hg Urine, plasma, serum 758 J.Anal. At. Spectrom., 1999, 14, 717-781 AA;F;L AA;ETA;L AA;F;L MS;ICP;L AA;-;-; AA;-;L AA;-;L AA;ETA;L MS;ICP;L AF;ETA;L AA;-;- AA;-;L AA;-;L AA;CV;L AA;CV;L AE;ICP;L AA;CV;L AA;CV;L AE;ICP;L MS;ICP;ETV AA;CV;L AA;CV;L AA;CV;L AA;CV;FI AA;-;L AA;CV;L AA;CV;L AA;ETA;GC 201 A suspension of tablets in H2O was sampled with an FI system which included an electrodialyser to remove suspended solids.The system was linked to an AA spectrometer for measurement of the Cu See Al, ref. 216 Fe and Zn determinations using a high performance nebulizer were compared 216 318 217 to those of a standard nebulizer with glass impact bead. LODs were 0.024 and 0.002 mg ml-1 for Fe and Zn, respectively, for the former nebulizer and 0.044 and 0.011 mg ml-1 for the latter.The Cu and Zn distribution in cows milk was also reported Results for measurement of 54Fe and 57Fe were improved when protein was removed, membrane desolvation was employed and sample was introduced 215 144 by an ultrasonic nebulizer Asbestos silicates were included with cultured cells in an investigation of the mechanisms of cell damage.Changes of Fe and Mg concentrations were observed TLC was used to remove interferents. The extract was then reacted with FeIII. Filtrate and precipitate were separated and the Fe measured in both fractions 362 124 to calculate the histidine concentration in the original specimen See Cu, ref. 362 Further work with the glassy C column atomizer, giving no background absorbance, was reported Specimens were diluted 1000- to 10-fold. 37Cl16O caused interference on 67Ga. 208 2 127 141 Measurement on 71Ga was preferred, therefore (in Japanese) A LEAF system with a copper vapour laser was used. An LOD of 1.0 pg was reported Prenatal exposure to methylmercury from fish diets was assessed by analysis of maternal hair. The exposure to methylmercury did not influence the age at 479 480 The eVect of Ca, Cu, Mg and Zn ions and pH on the adsorption of Hg2+ by which infants began to walk or talk cellulose and other materials was investigated. The presence of the ions increased adsorption.The results were compared with those for various food products and cellulose dietary fibre. In all cases the adsorption was higher than that for cellulose alone (in Polish) The level of Hg in various fish species and monthly variations in the levels were reported The holding time for total Hg in the above tissues was calculated using EPA 481 482 method 245.6 A multi-vessel system for determination of Hg by CVAAS and ICP-AES 349 142 Hg2+ and methylmercury were adsorbed onto a column packed with dithizone was developed immobilized onto sodium dodecylsulphate coated alumina.The trapped Hg was then back extracted into aqueous solution using 1 M HBr Samples from mothers and infants living in the Amazon region were digested using alkaline conditions See As, ref. 445 445 234 The sample was placed into a minitube furnace and heated to 950 °C in a stream of O LOD was 19 pg 233 4 483 2-Ar.The gas flow transferred the volatilized Hg to the ICP-MS. The Samples were heated at 90 °C with acids and then KMnO4 added. Oxalic acid was added to neutralize KMnO for CVAAS. Sensitivity was increased by the presence of H2SO4 A microwave heating system was used to digest specimens. Hg vapour was formed in a commercial accessory and preconcentrated by gold amalgamation.The final measurement was made after thermal desorption Markers of the immune response to Hg, seen in some individuals, did not 238 16 144 152 correlate with Hg in body fluids Following solubilization with TMAH an FI arrangement was set up in which reduction 4 and vaporization of the Hg. An LOD of 0.1 4mg l-1 was reported KMnO was used to cleave the C-Hg bond. NaBH was then added for TLC was used to remove interferents.The extract was then reacted with FeIII. Filtrate and precipitate were separated and the Hg measured in both fractions to calculate the histidine concentration in the original specimen Organomercury species were extracted using toluene and the extract treated with a 1% cysteinium chloride solution. After shaking and centrifugation the cysteine solution was analysed using a commercial Hg analyser. The LOD was 0.01 ng Hg 35 Yeast cells were immobilized onto silica gel and the powder assembled into a column which was incorporated into an FI manifold for separation of Hg species and generation of atomic vapour 484 Methylmercury and dimethylmercury were extracted into benzene and xylene, respectively.The extracts were injected onto a 3% Carbowax 20M on Chromosorb W/AW DMSC (60-80 mesh) column with a temperature rise from 60 to 160 °C. Ar carrier gas carried the eluent to the atomizer for atomization and measurement at 700 °C.The LOD was 0.2 ng (in Chinese)Table 1 (Continued) Biological samples Hg Hg Chinese herbal medicines Biological samples Hg Fish Hg Fish RMs Hg Hg Fish, shellfish, human hair Hg Biological specimens Hg Biological specimens Urine Hg Blood, urine, fish Hg Biological samples Hg Hg Biological fluids, tissues Chinese medicines Hg Hair Hg Kidney cells Hg Serum I Foods II Foods I Food CRMs II Nutritional and biological CRMs Nutritional and biological CRMs I Plasma, serum, blood, urine AA;Hy;GC AE;ICP;CV AE;ICP; FI-CV AE;MIP;GC AE;MIP;L AF;-;GC MS;ICP;GC MS;ICP;ETV MS;ICP;ETV MS;ICP;L AF;CV;L AF;CV;L AF;CV;L XRF;-;S XRF;-;S AE;ICP;L MS;ICP;L MS;ICP;L MS;ICP;L MS;ICP;L MS;ICP;L NAA;-; - MS;ICP;L 485 3 4 KBH was added to the sample to give the hydride form of CH HgCl.The hydride was collected in a fused silica fibre, part of a laboratory made solid phase extraction unit.After collection the fibre was transferred to the GC column and the hydride thermally desorbed. The LOD was 26 ng See As, ref. 456 456 235 4 153 A column containing silica gel functionalized with methylthiosalicylate was included in an FI manifold for separation and preconcentration of the Hg, which was eluted with thiourea and reduced with NaBH Following microwave assisted digestion samples were either ethylated using Na tetraethylborate, extracted into hexane and injected into a GC-MIP-AES or treated with NaBH and analysed using purge and trap GC-MIP-AES, yielding LODs of 3 and 12.54 pg g-1, respectively 349 351 2O. Methylmercury was ethylated and the 486 Rapid microwave digestion using TMAH was followed by ethylation, extraction of Hg species into hexane and flash isothermal separation Sample was decomposed using methanolic KOH at 70 °C, cooled and mixed with concentrated HCl and H resulting species, ethylmethylmercury(II) adsorbed onto a Tenax tube which was then connected to a GC and desorption eVected by heating to 220-250 °C.The LOD was 0.5 pg. Interfering metal ions were removed by EDTA (in Chinese) Organomercury species in the sample were collected at -80 °C and heated for rapid injection onto a multi-capillary GC column. The separated species/ isotopes were measured by ICP-MS with an LOD of 15 pg 65 66 TMAH was used to solubilize tissues. A portion, with iodoacetate, sodium thiosulfate and acetic acid, was dried and heated.At this point methylmercury iodide was volatilized, leaving only the inorganic Hg. Total Hg concentrations were given when this treatment was omitted For this ID analysis a modifier with Pd, Mg(NO3)2 and 2% HCl was found to give best results. Accurate results with an SRM were obtained and the LOD was 0.02 ng ml-1 151 By complexing Hg with (NH4)2H2EDTA in the presence of NH3 it was possible to achieve wash out times of <130 s for concentrations above 30 mg l-1.A novel disinfection procedure for blood and urine using Virkon, a virucide, was also reported. Excellent QC results were detailed. The application of the method to various Hg studies were described 22 232 3 455 3 An on-line microwave digestion system was described for heating the sample with potassium bromide-bromate to oxidize methylmercury.The HgII was reduced by SnCl2 50 mg samples were digested with HNO and H2SO4 in heated sealed containers, reduced with SnCl2 and the Hg determined Samples were digested with HNO and HClO4. Thiourea and ascorbate were added followed by 5% HCl. The solutions were sampled into an FI manifold for vapour generation. The LOD was 0.02 ng ml-1 (in Chinese) 108 109 Cross-sectional analysis using SR-XRF showed that Hg accumulated in the inner cortex following absorption, and on the cuticle if there was exogenous exposure, to methylmercury. The profile along the hair length could also be determined Structural changes were observed in apoptopic proximal tubular cells of HgCl2- treated animals.Accumulation of Hg was evident in the altered cells, suggesting a relationship between apoptosis and Hg 47 346, 487 Serum proteins were precipitated and removed by centrifugation. The supernatant was aspirated to measure iohexol indirectly as I A conference report and then full paper on extraction of I using TMAH at 90 °C.For complete extraction from all the matrices studied it was necessary to have a sample particle size of <300 mm. The LOD was 30 ng g-1 348 3 and good agreement was obtained in comparison with conventional digestion procedures 0.5 g of sample was weighed into a PTFE liner, HNO -HClO4, 3.5+1.5 added and the insert placed inside a steel bomb at 160 °C for 4 h.Rigorous cleaning with TMAH was necessary to overcome blank problems 347 Sample, 129IO3- spiked solution and H2O2 were heated with TMAH in a closed PTFE vessel at 90 °C for 3 h, or the sample plus spike were digested using Q-ICP-MS, yielding an LOD of 8 ng g-1 H2SO4-HNO3 in a microwave. The 127I5129I ratio was measured using a 210 2 424 The samples were combusted in a stream of O and the products collected in a 5% H2O-tertiary amine solution. The same system was proposed for other elements A conference report related to the work described in ref. 210. The results obtained by the method described were compared with those obtained by NAA. Low recoveries were reported for biological CRMs due to incomplete extraction of I 213 1% TMAH was used as diluent and Rh was included for internal standardization 759 J. Anal. At. Spectrom., 1999, 14, 717-781Table 1 (Continued) Urine Urine III Tissues, biological fluids Tissues I Teeth, bones Thyroid tissue IIIn Breast milk, infant formula Serum K Tissues Biological samples Urine KKLa Mg Leucocytes, erythrocytes Cereals Clinical samples Mg Mg Mineral water Plasma, red cells Mg Mg Mg Serum, red cells, hair Lung cells Serum Mg Mg Bone Mg Mg Mg Platelets, plasma, red cells Biological specimens Plasma, urine Mg Mg Mn Biological samples Biological specimens Foods Hair Mn Mn Pharmaceuticals Mn Tea leaves, tea Mn Liver Mn Wine Mn Tea leaves, tea Mn Foods, beverages Mo Tissues Mo Foods, beverages Mo 760 J.Anal. At. Spectrom., 1999, 14, 717-781 MS;ICP;L MS;ICP;L MS;ICP;L MS;ICP;L MS;-;S XRF;-;S AA;ETA;L AE;F;L AA;F;L MS;ICP;L AE;ICP;FI AA;-;- AA;-;L AA;F;L AA;F;L AA;F;L AA;F;L AA;-;- AA;ETA;L AE;ICP;L AE;DCP;L MS;ICP;L MS;ICP;L MS;ICP;L AA;ETA;FI AA;ETA;L AA;ETA;L AA;F;L AA;F;L AA;ETA;L AE;ICP;L AA;F;L AA;-;L AA;ETA;L AE;ICP;L 212 488 solution containing 129I.The LOD was 3 Standard additions calibration was employed and recoveries of 101-113% were obtained. The LOD was given as 0.000 38 mmol l-1 Samples were diluted with an NH 0.02 mmol l-1 A mixture of water-soluble tertiary amines at pH 8 was used for sample 18 211 2 dissolution. This mixture, sold as CFA-SC, allowed simultaneous measurement of I and other elements To avoid loss of volatile I, samples were oxidized in a stream of O and the products collected in a 5% tertiary amine solution (CFA-C).This solution was used for ICP-MS and measurement of I. Other elements could be 91 214 209 determined at the same time See Cl, ref. 91 Age-related changes in concentration and total amount of I in the thyroid were determined Ni(NO3)2 was applied as chemical modifier in the Zeeman-eVect ETAAS determination of In in breast milk and infant formula; at 325.6 nm the LODs were 2 and 2.5 mg l-1, respectively.The samples, 400 ml, were diluted with 200 ml 4% Triton X-100-200 ml H2O, injected into the furnace, dried and 10 ml 274 of 5000 mg l-1 Ni solution added Measurements using ion selective electrodes were compared with results given by flame photometry The samples were from a case of fatal Ag poisoning. See Ag, ref. 273 273 83 218 3 See Ca, ref. 83 La in 1.0 l of urine was extracted at pH 9 into 100 ml quinolin-8-ol in CHCl 2O at pH 4.5.This solution was readjusted to pH 9, 3 and back-extracted into H ethanolic quinolin-8-ol was added and pumped through an ion exchange column. The retained La was eluted with HNO into the ICP with an LOD of 0.09 ng ml-1 Cells were obtained by centrifugation and then lysed for measurement of Mg. 489 303 225 Concentrations were higher with heparinized compared with defibrinated blood See Ca, ref. 303 Methods for measurement of Mg in clinical specimens were comprehensively reviewed See Ca, ref. 322 Samples were from healthy subjects, patients with steatosis and those with 322 490 227 alcoholic liver cirrhosis Mg supplements were given to hyperactive, Mg-deficient children. Hair Mg concentrations increased and the hyperactivity decreased See Fe, ref. 215 Samples diluted with pH 7.4 buVer were added to a TSK-gel DEAE-5PW ion 215 226 229 exchange column.Mg species were eluted and the fractions analysed oV-line by ETAAS. Mg was associated with albumin and globulin proteins Results from an experimental study of variations of dietary Al, Ca and Mg were reported (in Japanese) Platelet Mg concentrations were low in diabetic compared with control subjects 491 See Cr, ref. 50 50 77 83 32 Mineralized samples were analysed in a study using stable isotopes to evaluate absorption and bioavailability See Ca, ref. 83 An FI manifold was constructed which included a 'knotted reactor' for the adsorption of a quinolin-8-ol chelate. With a 30 s sampling time the preconcentration step produced an 8-fold enhancement factor and an LOD of 29 ng l-1 471 231 3 See Cr, ref. 471 After digestion with HNO and H2O2 and evaporation to dryness the residue 3. 5ml were placed onto a graphite probe for 230 was taken into 1% HNO atomization within the furnace.The LOD was 11.69 pg (in Chinese) An FI system was prepared for mixing specimen solutions with MnO2 at pH 4-5. The excess Mn was directed to the AA spectrometer for the indirect determi- 492 493 nation of vitamin B6 (in Chinese) Mn was speciated using ion exchange chromatography and FAAS. Full experimental details were given and Mn bioavailability discussed Modifications to the vertical glassy carbon atomizer were described: an open tubular column was used; the sample cup was held by a glassy C rod; a lower graphite electrode was included; an Si-controlled rectifier supplied an increased 494 Mn and Pb were determined at LODs of 0.4 and 5.5 mg l-1, respectively output rating (in Japanese) 492 414 (in Chinese) Mn was speciated using ion exchange chromatography and FAAS.Full experimental details were given and Mn bioavailability discussed Daily dietary intake of Mo in Belgium was estimated by duplicate portion sampling.Portions were freeze-dried and sub-samples digested using 240 4 3 413 HNO3-H2O-H2O2 in a microwave oven Tissues from sheep given ammonium tetrathiomolybdate to treat Cu toxicity were prepared by digestion with HNO and HClO Daily dietary intake in Mexico and Germany was estimated using duplicate portion sampling. Levels were estimated in 14 groups of adults in 1988, 1992 and 1996. Intake increased considerably over the duration of the studyTable 1 (Continued) Mo Mo Na Ni Ni Ni Ni Ni Ni Ni Ni Ni Ni Ni Ni PPPb Pb Pb Pb Pb Pb Pb Pb Pb Pb Pb Pb Pb Pb Pb Pb Pb Pb Foods Urine Serum Chicken organs and tissues Foods, faeces, urine Green vegetables Vegetables Foods Preserved foods Sterilized milk Hair Biological specimens White asparagus White asparagus Tissues Edible oils and fats Edible fats and oils Blood Beverages Blood, foods Drinking water Infant formula Powdered drinks, honey, syrups Tap water Water White wine Wild mushrooms Wine Wine Milk powder Drinks Fruit juice, milk Rice flour Water Water AE;ICP;L MS;ICP;L AE;F;L AA;-;- AA;-;- AA;-;- AA;-;L AA;ETA;L AA;ETA;L AA;ETA;L AA;ETA;Sl AA;ETA;FI AA;F, air-C2H2;L AA;F, air-C2H2;L LMMS;-;S XRF;-;- AA;ETA;L AA;-;- AA;ETA;L AA;ETA;L MS;ICP;L AA;ETA;L AA;ETA;L AA;ETA;L AA;ETA;L AA;ETA;L AA;ETA;L AA;ETA;L AA;ETA;L AA;ETA;L AA;ETA;Sl AA;F, air-C2H2;L AA;F;L AA;F;L AA;F;L AE;ICP;L 495 239 Mo in tar-contaminated soil and fruit and vegetables grown on the soil was determined.Diet evaluation indicated no risk was posed to human health (in German) 95Mo and 98Mo were measured in urine samples diluted 1+9 with 2% HNO3. A positive correlation with butter consumption was noted See K, ref. 274 The eVect of supplementing chicken diet with NiSO ·7H 274 496 4 2O, up to toxic levels, on Ni and Zn concentrations in a wide range of organs was investigated.Ni increased in all organs, whilst Zn showed more variable behaviour (in German) 497 476 299 Ni content in 7-day diets, faeces and urine was measured for 14 volunteers from Saxony. Most dietary Ni was eliminated in the faeces, but there was a positive Ni balance. The daily intakes were several times higher than daily allowances See Cu, ref. 477 Ni was preconcentrated using activated C following ashing and decomposition 3-H2O2 (2+1). Various experimental parameters were evaluated with HNO Ni was determined in 8 types of food. The LOD was 1.6 mg l-1 (in Chinese) 498 469 499 13 See Co, ref. 469 44 brands from 32 Spanish provinces were surveyed. The average Ni content was 60.7 ± 35.3 mg kg-1 (in Spanish) 0.1 g hair powdered in a zirconia ball mill was suspended in 25 ml H2O and 32 stabilized by addition of glycerol.Mg(NO3)2 was used as chemical modifier An FI manifold was constructed which included a 'knotted reactor' for the adsorption of an APDC chelate. With a 30 s sampling time the preconcentration step produced a 21-fold enhancement factor and an LOD of 7.6 ng l-1 See Cr, ref. 474 474 See Cr, ref. 475 475 199 361 500 See Cr, ref. 199 Samples were solidified with 15% stearic acid and WDXRF measurement performed in disposable liquid cells.The LOD was 2 mg kg-1 for both P and S 3-30% H2O2, 4+2, and digested using Sample, 0.5 g, was mixed with 65% HNO microwave heating. The wavelength of measurement was 213.6 nm (in Czech) Specimens from infants living in urban and rural areas of Chile were analysed See Cd, ref. 408 See Cd, ref. 461 501 408 461 462 394 323 See Cd, ref. 462 A study of the practice of infant formula reconstitution revealed that 2 of the 40 samples analysed contained Pb concentrations above approved safe levels.The samples were found to have been reconstituted using tap water drawn from houses with plumbing over 20 yr old Samples were dissolved in H2O, acidified using 0.2% HNO3 and injected into an 3)2 were 294 324 end-capped transversely heated cuvette. Carbon build-up was minimized by using two pyrolysis steps, 600 and 1000 °C. 5 mg Pd+3 mg Mg(NO used as modifier See Cd, ref. 294 Refractory elements-Hf, Nb, W, Zr-were used to treat the surface of a pyrolytically coated graphite furnace.Treatment with W allowed injection 380 399 volumes up to 100 ml, yielding an LOD of 0.02 mg l-1 Pb was measured in Chinese white wine at an LOD of 31 pg (in Chinese) 95 samples of 13 wild species in Northern Spain were analysed. Average levels were 1 mg kg-1 dry weight, with a maximum of 10.43 mg kg-1 17 laboratories participated in an international collaborative study of an ETAAS 423 378 determination for Pb in wine.The method, incorporating a novel matrixmatching procedure, gave good agreement between participants An exhaustive study of the source of elevated Pb levels in wine was reported. Analysis of 7000 wines showed that neither atmospheric pollution nor the tin-lead covering of the neck were responsible. The vintage and wine colour most strongly correlated with the elevated levels and inspection of the wineries 465 300 giving these levels always revealed the presence of brass tubes and faucets See Cd, ref. 465 100 ml samples were neutralized, shaken with 15% Mg(NO3)2-20% NaOH, 1+1, left for 1 h, centrifuged and the precipitate dissolved in HNO3 and diluted to 10 ml with H2O.Recoveries were in the range 90.5-112% (in Chinese) 298 320 See Al, ref. 298 Flour was digested using HCl-HNO3, diluted with a solution containing 200 g l-1 KI and injected into an FI system incorporating on-line extraction using IBMK.The LOD was 2.8 ng ml-1 (in Chinese) Samples were acidified, neutralized and pH adjusted to 5.5 using 1 M ammonium 295 293 acetate buVer. They were then pumped through a column containing Amberlite resin. The Pb was then eluted using 1 M tetrabutylammonium bromide and the resulting complex extracted in IBMK in a 30 cm reaction coil. The preconcentration factor was 550 and the LOD 0.3 mg l-1 A novel ion-exchange resin based on 'cavities' designed to exactly fit the charge, co-ordination number, geometry and size of Pb2+ was described.It was claimed to be almost interference free 761 J. Anal. At. Spectrom., 1999, 14, 717-781Table 1 (Continued) Blood Pb Calcium pills Pb Pb Calcium pills and supplements Maple syrup Pb Vegetables, SRMs Pb Wine Pb Hair, blood Pb Blood Pb Blood Pb Blood Blood Blood Pb Pb Pb Blood Pb Urine Pb Urine Pb Urine Urine Amniotic fluid Plasma Pb Pb Pb Pb Plasma proteins Pb Blood, tissues Pb Bone Pb Pb Blood, bone, tissues Biological samples Herbal medicines Pb Pb Ca drugs Pb Pb Antacids, dietary supplements Bone Pb Biological RMs Pb Tissues Bone Pb Pb Tissues Bone Pb Pb 762 J.Anal. At. Spectrom., 1999, 14, 717-781 MS;ICP;L MS;ICP;L MS;ICP;L MS;ICP;L MS;ICP;L MS;ICP;L AA;ETA;L AA;ETA;L AA;ETA;L AA;ETA;L MS;ICP;L MS;ICP;L MS;ICP;L AA;F;L AA;ETA;L MS;ICP;L AA;Hy;FI MS;ICP;ETV See Cd, ref. 67 MS;ICP;ETV See As, ref. 64 MS;ICP;HPLC See Cu, ref. 478 MS;ICP;HPLC Distribution of Pb among caeruloplasmin and other plasma proteins was determined by SEC. See also Pb, ref. 478 MS;ICP;HPLC Samples were collected from subjects who ingested contaminated soil. Plasma MS;-;- AA;ETA;L MS;ICP;L AE;ICP;S AA;F;L AA;ETA;Sl MS;ICP;L MS;-;S MS;ICP;L LMMS;-;S XRF;-;S XRF;-;S XRF;-;S 220 149 Samples were from Ecuadorian children living in areas with intensive occupational use of Pb.The isotope ratios in blood matched those of contaminated soils A method was described for determining Pb in calcium pills and antacids. The method was required in response to legal proceedings in January 1997 when 10 companies were reportedly sued for not warning consumers about allegedly high levels of Pb in their supplements 150 3 at 230 °C and 1770 psi. Matrix matched 3 502 Pb concentration and isotope ratios were measured in 9 brands of Ca supplement following digestion in 2 ml of HNO standards using high purity CaCO were employed.Moderate levels were found in samples originating from oyster shell Samples were digested in a microwave using HNO3 as acid. 8 brands of commercial syrup contained levels in the range 18-367 ng g-1 A tungsten coil was placed within a glass chamber which was connected directly 330 and transported at 1.5 l min-1 by a 95% Ar-5% H flow.Quantification was to the plasma torch via a 750 mm tube. Sample was vaporized from the coil 2 315 achieved using isotope dilution procedures. For 208Pb5207Pb the precision was 1-2.5% SEC-ICP-MS was used to speciate Pb in wine. A study carried out on 20 wines suggested that no free Pb was present in wine. 40-95% of Pb was complexed with rhamnogalacturon II and the remainder with other unidentified species A comparison of three chemical modifiers used with a tungsten coil atomizer 121 was reported.Pd(NO3)2 gave results that were more precise and accurate than did (NH4)2HPO4 or NH4H2PO4. The LOD was 1 mg l-1 Recent work to improve the sensitivity and accuracy of measurements using 219 116 tungsten filament atomizer were reported Another report of a small portable AA spectrometer based on a tungsten coil atomizer A further presentation of a portable tungsten coil atomiser AA spectrometer Blood specimens were digested with microwave heating. 208Pb was measured 120 503 80 81 33 Samples were digested in acid and analysed by ICP magnetic sector-MS Pb isotope ratios were measured with good accuracy and precision using double focusing magnetic sector ICP-MS Sample enrichment was achieved with a supported liquid membrane. The membrane solution was 40% m/m di-2-ethylhexylphosphoric acid in kerosene 504 from which the Pb was back extracted into 1 M HNO3.Enrichment factors of up to 200 were obtained with extraction times of 0.5-4 h. This allowed LODs of 0.1 and 6 mg l-1 for ETAAS and FAAS, respectively. Results compared well with those obtained by ICP-MS An FI-HG manifold, which used a ferricyanide oxidizing agent, was reported. The sample was led to a heated quartz tube atomizer and the LOD was 80 ppt 67 64 478 221 505 222 223 3 volume of H proteins were separated by ion exchange chromatography and SEC.Most of the lead was associated with caeruloplasmin TIMS was used to measure 204Pb, 206Pb and 207Pb in a study to monitor movements of Pb between tissues during pregnancy. A monkey model was used for this investigation 5-20 mg ashed bone was dissolved in 5% HNO and diluted with an equal 2O 78 12 126 3 Magnetic sector ICP-MS was used for precise measurement of Pb isotopes in specimens collected in studies of the eVectiveness of chelation treatment See Cd, ref. 12 1 g sample was ashed at 450 °C and further digested with HNO and HClO4. The residue was taken into 2% HNO and aspirated into a vanadium-coated slotted quartz tube. The LOD was 8.37 3 mg l-1 Ca drugs were ground to produce particle sizes of around 3 mm. Suspensions of 15 the powders were atomized from a molybdenum tube with thiourea chemical modifier. H2 added to the Ar increased the analytical sensitivity Acid dissolution was the only preparation required 148 Pb isotopes in SRM 1400 Bone ash were measured by TIMS 131, 132 79 Pb isotopes, at concentrations of 1-10 ppb, were determined by magnetic sector MS for isotope tracer studies. 209Bi was the internal standard and the LOD was reported as 0.013 ppb total Pb See Cr, ref. 199 In vivo determination of Pb in the finger bones of persons with occupational 199 224 exposure was reported. The LOD was 25-30 mg kg-1 458 97 See Ba, ref. 458 Neurophysiological studies were undertaken in subjects with occupational Pb exposure. Results were evaluated in association with long-term exposure as given by in vivo determination of Pb concentrations in tibial and calcaneal boneTable 1 (Continued) Bone Pb Bone Pb Bone Pb Teeth Pb Finger bone Urine Pb Pd Urine Pd Blood Pt Pt Plasma ultrafiltrate Plasma Pt Tissues Pt Plasma, tissues Pt Wine Pt Blood, urine Pt Urine Urine Tissues Pt Pt Pt Urine, blood Pt Anti-cancer drugs Pt Lipids, emulsions Pt Pt Tissue, in vivo Blood, urine Pu Urine Pu Foods Rb Serum REE Blood, urine Plasma, tissues REE REE Tissues Rh SRh Sb Sb Urine Edible oils and fats Red cells, serum Water, red blood cells, serum Water Sb Tap water Sb XRF;-;S XRF;-;S XRF;-;S XRF;-;S XRF;-;S MS;ICP;L MS;ICP;L AA;ETA;L AA;ETA;L AA;ETA;L AA;ETA;L AA;ETA;L AA;ETA;L MS;ICP;L MS;ICP;L MS;ICP;L AA;ETA;L AF;ETA;L laser and a detection limit of 50 fg was achieved MS;ICP;HPLC Inactive species in Pt anti-cancer drugs were separated and identified.A sensitive analysis was established with an LOD of 40-50 ng l-1 Leakage of Pt and silicones (measured by GC-MS) from breast implants into MS;ICP;L XRF;-;S MS;ICP;L MS;ICP;L AA;-;L MS;ICP;L MS;ICP;L MS;ICP;ETV AE;ICP;L MS;ICP;L XRF;-;- AA;ETA;L AA;ETA;L AA;Hy;HPLC SbIII and SbV were determined at LODs of 2 and 1 mg l-1, respectively, using was reported for all 3 sample types AA;quartz furnace;GC 96 92 Clinical application of in vivo bone Pb measurement in children and other nonoccupationally exposed groups was discussed Preliminary evaluation of a new XRF system, using bone phantoms, were reported.The results predict that a lower LOD, using less excitation energy, should be possible Tibial Pb was measured in vivo.Instrumental features were optimized using the 94 101 Monte Carlo code Children's teeth were irradiated and the X-ray spectra recorded. An aqueous calibration material was used and results were expressed as mg Pbg-1 per accumulated year. Data from China, Russia and the USA were compared Accumulation in bone was determined in smelter workers A UV photolysis procedure was the only preparation required. Concentrations 99 70 69 in untreated subjects were 32.7-219.7 ng l-1 and the LOD was 0.17 ng l-1 UV irradiated specimens were analysed.Q-ICP-MS and magnetic sector field ICP-MS were compared and superior results were found using ultrasonic nebulization high resolution ICP-MS The pharmacokinetics of total and unbound Pt species were investigated in 244 506 patients treated with carboplatin, to relate treatment response to dose A study of the maximum tolerated dose, toxicity, pharmacokinetics and pharmacodynamics of an orally administered drug, JM216, was reported The relationship between Pt pharmacokinetics and the formation of Pt-DNA 246 245 adducts was investigated in children treated with Pt Tumour tissue, 0.1 g, and 1 ml HNO were heated at 37 °C for 2 d.Pt was measured using aqueous standards and 3 the LOD was 3 mg l-1 A new cisplatin formulation with microcrystals suspended in oil was administered 507 373 243 to animals.The pharmacokinetics, therapeutic eVects and toxicity compared favourably with the aqueous drug (in Japanese) Various wines were microwave digested with HNO3-H2O2 or dry ashed then analysed using ETAAS. At 265.9 nm the LOD was 100 pg Blood was digested with HNO and the residue dissolved in H2O. HCl and a 3 solution with Triton X-100 and Tl internal standard were added. The solution was equivalent to a ten-fold dilution of the original specimen. Urine samples were similarly digested, redissolved in HNO3 and diluted to the original sample See Pd, ref 70.Normal levels were 0.48-7.7 ng l-1 and the LOD was 0.24 ng l-1 volume with H2O. Au was added as internal standard 69 70 508 See Pd, ref. 69 Uptake of Pt into cervical and endometrial cancer tissue was determined. Specimens were removed at surgery, one hour after i.v. administration of cisplatin Pt atoms were excited by radiation from a high repetition rate copper vapour 128 509 271 diVerent media was investigated.Pt leaked into lipid-rich tissues at a rate of 20-25 mg d-1 per 250 g implant A programme for the EGS4 Monte-Carlo system was prepared for in vivo 95 74 estimation of Pt uptake. The minimum detectable concentration was 50 ppm Analytes were preconcentrated by a chemical separation step. High purity reagents were used to limit potential interferences. The LOD was reported to be 0.0001 fg ml-1 The Pu isotopes were precipitated and the organic matrix destroyed by wet 284 404 ashing. After ion exchange chromatography a microconcentric nebulizer was used for sample introduction The daily Rb intake in Belgium was evaluated by duplicate portion sampling.The mean intake was 2.2±0.3 mg d-1 Specimens were digested in acid and the REEs preconcentrated using chelating 510 74 62, 63 resin See Pu, ref. 74 Lanthanides were measured. Samples were prepared by microwave digestion, or injected as slurries or solutions.H2O2 was included in the diluent to prevent build-up of ash while trifluoromethane lowered the vaporization temperature Distribution of Rh, following i.p. administration of rhodium propionate to mice, 511 was investigated See Pd, ref. 69 See P, ref. 361 3)2-Mg(NO3)2 was used as chemical modifier. The LOD was 2.3 mg l-1 69 361 168 168 Pd(NO Sb was determined using an STPF in a Zeeman-eVect ETAA spectrometer.The chemical modifier was Pd(NO3)2-Mg(NO3)2-HNO3. An LOD of 2.6 mg l-1 311 HPLC-HGAAS. A miniaturized column, 2 cm×0.4 mm id, was used in conjunction with 50 mM tartrate mobile phase. The LODs were deemed not sensitive enough for measuring real samples 310 SbIII was extracted into hexane after complexing with pyrrolidinedithiocarbamate, then derivatized to triphenylstibine using phenylmagnesium bromide and measured using capillary GC-AAS 763 J. Anal.At. Spectrom., 1999, 14, 717-781Table 1 (Continued) Body fluids, tissues Sb Sb Biological specimens Blood, hair Sb Tissues Body fluids Tissues, fluids Serum Sb Se Se Se Se Blood, plasma, urine, hair Beverages Se Se Marine tissue SRMs Serum Se Serum Se Serum Se Serum Se Serum Se Plasma Se Blood, tissues Se Nutrition liquids Se Edible mushrooms Se Foods Se Human milk Se Human milk Se Biological CRMs Se Se Se Potable water Food supplements, urine Amino acids Se Cod Se Se Foods, plasma, urine 764 J.Anal. At. Spectrom., 1999, 14, 717-781 MS;ICP;L Body fluids were diluted 1+14 with H2O. Tissues were digested with HNO3, either with microwave heating or in open quartz vessels. In was added as internal standard. Accurate results were achieved with various CRMs. LODs were 0.01 mg l-1 and 0.7-0.8 ng g-1 MS;ICP;HPLC Methods to separate at least four Sb species were developed AF;Hy;L XRF;-;S -;-;- -;-;- ;-;- AA;-;- AA;ETA;L AA;ETA;L AA;ETA;L AA;ETA;L AA;ETA;L AA;ETA;L AA;ETA;L AA;ETA;L AA;ETA;L AA;ETA;L AA;Hy;L NAA; -;- AA;Hy;L AA;Hy;L AA;Hy;L 1.15 and 10.4 ng ml-1, respectively Se content of human milk from 20 volunteers and Se intake of breast fed infants was determined at 1, 2 and 3 months postpartum.There was a slight decline as lactation advanced AE;ICP;HPLC Anion exchange HPLC-ICP-AES was used to speciate selenomethionine, SeIV, AF;Hy;L MS;-;L MS;ICP;L MS;ICP;L MS;ICP;HG 512 169 170 Samples for analysis were prepared by microwave digestion.The LODs were 3 ng g-1 (hair) and 1 ng g-1 (blood) 458 257 513 133 See Ba, ref. 458 A review of the clinical importance of Se species was presented A review of procedures for the determination of Se species RMs containing added inorganic Se were prepared for use as control specimens in speciation studies Specimens from children with phenylketonuria, aged 4.5-15 y, were analysed 514 515 (in German) Sample-Triton X-100, 1+1, was loaded into the autosampler and Pd(NO3)2 added. 20 ml aliquots were injected into the furnace. At 196 nm the LOD was 8 ngml-1 (in Chinese) Sample, <0.1 g, and HNO 28 3, were placed into 7 ml digestion vessels placed inside larger vessels and microwave digested. Total and Se species gave quantitative 516 recoveries, particularly if a multistage microwave programme was used Normal serum Se concentrations of subjects living in Barcelona were established.The range of results, 60-106 mg l-1, indicated adequate Se intake Wall, platform and probe atomization techniques were compared together with 247 an investigation of various chemical modifiers. Platform atomization was recommended. Cu or Pd, alone or with Mg, gave good results and allowed aqueous calibration. The LOD was 10 mg l-1 0.1 M HNO was added to serum, to a pH of 1.5, and left for 24 h to inactivate 7 3 517 2 249 potentially hazardous viruses A standard additions calibration procedure was used with addition of Cu or Pd chemical modifier.Accurate results were demonstrated using D background correction 200 ml sample was added to 800 ml diluent, 0.2% HNO3 in 0.5% Triton X-100. 266 The solutions were mixed ultrasonically using a slurry sampling system Patients with colorectal adenoma had lower plasma Se concentrations than controls.A protective eVect of Se was proposed A new Se agent, selol, was administered to rats. The tissue distribution and 518 519 pharmacokinetics were determined SeIV was measured using HG interfaced to an ETAA spectrometer. The hydrides were trapped in a Pd-coated graphite cuvette at 700 °C. SeVI was determined following reduction. Total Se was measured following sample dissolution with 520 A survey of 12 species revealed very high levels, up to 367 mg kg-1 dry weight, K2S2O8 in a boiling water bath.The LOD was 36 pg Se in one species from Italy and a related species from the Pacific Northwest 521 522 USA. Toxicological consequences were discussed Se was determined in >100 samples of convenience and vegetarian foods. Mushrooms, fish, oVal and some chicken dishes contained the highest levels 3-H2O2, Human milk, 2 ml, or blood, 0.5 ml, was digested with HNO 1.5+0.23 ml, in a microwave oven.The resulting solution was heated at 140 °C for 2-3 h to reduce the volume to 1 ml. HCl, 2 ml, was then added, the volume adjusted to 10 ml and the solution heated at 100 °C for 10 min. Se was measured using FI-HGAAS at 196 nm. The LODs for milk and blood were 523 359 447 357 SeVI, selenocystine and trimethylselonium. 3 diVerent microcolumns were evaluated and a DIN utilized with the ICP-AE spectrometer. LODs were in the range 20-38 ng ml-1 See As, ref. 447 9 organic Se species were separated by HPLC-ES-MS, using either a C18 or a cyanopropyl bonded column. The LODs were sub ng g-1 255 An on-line procedure using sodium tetraethylborate was used to obtain the ethyl derivatives of selenocysteine and selenomethionine. These were then trapped to preconcentrate the species before being eluted into an ICP-mass spectrometer. It was possible to determine the species at ppt levels in selenoamino acids from enzymatic systems 80Se was measured by GC-MS following derivatization using phenylenediamine 354 355 to overcome problems encountered with isobaric interferences in ICP-MS.However when the 2 methods were compared for total Se recovery it was found GC-MS only gave 62-64% whilst ICP-MS gave 100% Volunteers were fed diVerent levels of Se and exchangeable body pool sizes investigated. Plasma and urine were analysed using HG-ICP-MS. Volunteers were also fed cod or yeast enriched with 82Se.The results indicated diVerent metabolic pathways for diVerent speciesTable 1 (Continued) Se Se Se Se Se Se Se Se Se Se Se Se Se Se Se Se Se Se Se Se Se Se Se Se Se Si Si Si Si Nutritional supplements, plants Drinking water Foods Infant formula Plasma Blood, plasma Plasma, blood Plasma, serum Serum Serum, breast milk Serum proteins Serum Grain, human hair Pork spare ribs, vegetables Urine Urine Urine Urine Biological specimens Biological specimens Medicines Biological specimens Blood, urine, tissues Urine Biological specimens Serum, tissue Urine, blood Urine Hair MS;ICP;HPLC 10 selenoamino acids were separated using ion pair HPLC-ICP-MS, including the cis-trans isomers of Se-propen-1-yl-DL-selenocysteine. Quantitative results were obtained for enriched yeast and vegetables See As, ref. 317 Cation and anion exchange HPLC-ICP-MS were used to speciate Se in enriched MS;ICP;L MS;ICP;L MS;-;L MS;ICP;L MS;ICP;L of 49±11.55 mg l-1 0.5 ml plasma was heated in 2 ml HNO standard was added and the Se determined. Se concentrations of Sikhs living in Australia were measured MS;ICP;HPLC Selenite enriched with a stable Se isotope was injected into rats and the MS;ICP;L MS;ICP;L MS;ICP;ETV MS;ICP;L MS;ICP;LC MS;MIP;HG XRF;-;S AF;Hy;L NAA;-;S XRF;-;S AA;ETA;HG AA;ETA; FI-HG solution was taken for FI-HG with trapping of H2Se in a graphite furnace coated with In MS;ICP;HPLC A number of columns and conditions were explored and at least seven Se species MS;ICP;ETV AA;ETA;S AA;ETA;IC AA;ETA;L Samples contained 150-250 ppb Se MS;ICP;HPLC Microwave heating with HNO was used for measurement of total Se.Standard additions calibration was necessary to allow for variable final acid concentration.Various extraction procedures were used prior to separation of Se AF;Hy;L AF;-; HPLC MS;ICP;HPLC AE;MIP;GC AA;ETA;L AA;ETA;L AA;ETA;L SIMS;-;S 358 317 356 yeast. Acid or enzymic hydrolysis was required to recover the Se, with solvent based extraction giving recoveries of less than 20%. ES-MS was used to elucidate information about the structure of the species Analysis of 24 commercial brands using HG-ICP-MS revealed mean Se levels 365 265 3 at 100 °C for 1 h.An In internal 261 251 253 subsequent metabolism and incorporation into proteins was monitored A 1+15 dilution with 1% butanol eVectively eliminated polyatomic interferences. Accurate results were obtained with RMs A 1+10 dilution with 0.5% butanol allowed interference-free measurement of 82Se with an LOD of 0.05 mg l-1. Se concentrations in at least four separated 256 2 species were measured Se was separated from sources of interference by ETV and by addition of N 3 to the ICP 260 and CHF Capillary zone electrophoresis and isoelectric focusing were applied to the separation of selenium species Selenoprotein P, glutathione peroxidase and albumin were separated by aYnity 252 254 524 chromatography and the distribution of Se among these proteins determined in two RMs Following addition of 78Se internal standard and digestion with acids, the hydride was formed for introduction into an N2-MIP Water, soil, grain and human hair were analysed as part of a study of Se balance in China and Sri Lanka.The hair and grain samples demonstrated a systematic bias for the HGAFS results when compared to the other 2 techniques Sample was digested in a closed vessel overnight using HNO 353 2O, boiled for 3-H2SO4 and then heated to 60 °C for 4 h. The digest was diluted to 50 ml with H 1 h, filtered and 1 ml CuSO4, 40mg Te and 15 ml SnCl2-hydroxylamine hydrochloride added.The precipitate was filtered oV, dried and taken for analysis by XRF. The LOD was 0.2 mg (in Chinese) Focused microwave digestion was included in an on-line FI system. Not all 504 250 1 ml urine was heated with HBr and KBrO3 at 150 °C for 2 h. Excess Br- was species were converted to SeIV removed by addition of hydroxylammonium chloride and 10% HCl. The 258 64 11 were separated See As, ref. 64 0.1-1.0 mg powdered sample and 10 ml Pd modifier solution were placed into a graphite cup.Alternatively, to increase sensitivity, 0.2-0.4 g and 5 ml modifier were dried and heated at 600 oC for 30 min; the ash was then taken into the graphite cup. The ashing achieved a 10-40 fold preconcentration with an 262 LOD of 10 ng g-1 Se speciation was accomplished by IC using a microbore anion exchange column. The flow rate was 80 ml min-1 and 20 ml fractions were collected automatically into cups on an autosampler.These solutions were injected, together with Pd-Mg(NO 525 3)2 which stabilized all Se species Ni(NO3)2, 100 mg Ni, modifier was used in a comparison of ETAAS and ASV. 526 3 species. Extracts were further treated with a tertiary amine, CFA-C See As, ref. 454 454 See As, ref. 452 452 259 270 Volatile organoselenium species were separated by capillary GC with MIPAES detection Serum or digested breast tissue were diluted with a modifier containing lanthanum 2, NH4H2PO4, EDTA, C2H5COOH, and Triton X-100 268 oxide, CaCl Samples were diluted; urine 1+350 in H 10 ml were injected with 10 ml Ni, 0.5 mg ml-1 in 2% HCl. Recoveries were 2O, blood 1+19 in 1% Triton X-100.less than 90% and LODs were 0.07 mg l-1 and 0.7 mg l-1 for urine and blood, 269 2O were injected into the atomizer with Ni chemical respectively Samples diluted 1+500 in H modifier. The LOD was 0.7 mg l-1 See Al, ref. 436 436 765 J. Anal. At. Spectrom., 1999, 14, 717-781Table 1 (Continued) Sn Biological specimens Brain Sn Sn Sn Mussels, sea urchin eggs Biological materials Hatchery fish Sn Sn Drinking water, urine and citrus leaf RMs Fish tissue RMs Sn Fruit Sn Sr Tooth enamel, dentine Foods Sr Hair Sr Bone Sr Urine Th Urine Th Beverages, foods Ti Plasma Ti Urine Tl Urine Urine UU Urine Bone UU Urine VVW Biological fluids and tissues Blood, urine, nails, hair Clinical specimens Chicken organs Zn Zn and tissues Foods Zn Zn Zn Green vegetables Serum, spleen, thymus Plasma, urine Foods Zn Zn Serum Zn 766 J.Anal. At. Spectrom., 1999, 14, 717-781 -;-;- AA;Hy;GC AE;-;L MS;ICP;L MS;-;L MS;ICP;L MS;ICP;L MS;ICP;L AA;-;- AA;F;L AA;F;L MS;ICP;L MS;ICP;L MS;ICP;L AE;ICP;L AE;ICP;L AA;ETA;L AA;ETA;L AA;ETA;L MS;ICP;L MS;ICP;L XRF;-;S MS;ICP;L MS;ICP;L AE;ICP;L -;-;- AA;-;- AA;-;- AA;-;- AA;-;- AA;-;L AA;ETA;L AA;F;L AA;ETA;L A review of the eVects of organotins and their measurement (in Japanese) 154 281 Trimethyltin was determined using a system involving HG, cryogenic trapping, GC separation and detection by AAS.Organotins could be measured when 155 TBT and triphenyltin were determined at LODs of 2-10 and 20-40 ng g-1 using the total Sn taken was as little as 0.2 ng GC-AES and GC-FPD, respectively 100 mg samples were digested with HNO 280 3 (in Japanese) 308 Samples were hydrolysed using KOH-C2H5OH, then TBT and triphenyltin extracted using petroleum ether.The ether was evaporated and the residual extract mixed with 50% ethanol and placed in a solid phase extraction cartridge. The column was washed with 10% methanol and the species eluted using HCl-CH3OH, 1+9. The species were then extracted from the eluate using hexane-cyclohexane, 1+1, hydrogenated and analysed using GC-MS (in Japanese) See As, ref. 448 448 309 402 An HPLC separation of TBT, DBT, di- and triphenyltin, compatible with both ICP-MS and atmospheric pressure ionization-MS, was developed. The latter technique conferred the benefit of providing molecular ion information ICP-MS was used as an initial screen of fruit samples for Sn levels in excess of 0.06 mg kg-1. Samples containing above this level were re-analysed using GC-MS to identify the presence of the pesticide cyhexatin.Using ICP-MS saved a considerable amount of staV time and it was proposed other organometallic pesticides could be screened in this way Sr concentrations were related to tooth type and to stage of caries 277 527 3 Samples were dry ashed at 600 °C, dissolved in 20 ml 5% HNO containing 0.2% KCl and 1% La(NO 278 3 2O2 was mixed with an emulsifier for FAAS. 3)3 and analysed using FAAS at 460.7 nm (in Chinese) Hair digested with HNO and H Sensitivity was improved 55% by addition of the emulsifier (in Chinese) 82 87Sr586Sr ratios were measured by high resolution ICP-MS.Operating parameters were experimentally optimized and corrections for dead time and mass bias were made to achieve a precision of less than 0.03%. Remains from 7000 year old skeletons were investigated HNO and HCl were added with Ir internal standard. Samples from workers 73 3 72 3 371 45 For AES the sample was diluted 1+99 with H with occupational exposure and from controls were analysed Urine was evaporated to dryness at 70-80 °C and the residue dissolved in 1% HNO for determination of Th and U.The LODs were 0.26 and 0.01 mg l-1, respectively Ti was determined in a wide range of foods from Germany in 1988 and 1992. There was evidence of Ti accumulation in some vegetables (in German) 2O. Complex heating programmes 279 3)2 were added as chemical modifiers to were required and memory eVects were noted for the ETAAS methods Tetraamine, Pd nitrate and Mg(NO untreated urine in a procedure to analyse specimens from subjects with 73 283 72 100 occupational exposure to Tl See Th, ref. 73 Dilution of specimens 1+19 in 1% HNO with an Ir internal standard allowed an LOD of 0.32 ng l 3 -1. The mean concentration in non-exposed subjects was 16.1 ng l-1 See Th, ref. 72 Plaster of Paris phantoms with U concentrations of 0-100 mg g-1 were used for calibration.The LOD of 20 mg g-1 was inadequate for occupational monitor- 52 Interference from ClO+ was overcome by cryogenic desolvation to remove Cl- ing by in vivo measurement as condensed HCl. Sc was added as the internal standard See Cr, ref. 51 51 46 Hair and nails were heated at 60 °C for 12 h with HNO3. High concentrations were found in specimens from a subject with suspected poisoning Techniques to detect Zn deficiency were discussed See Ni, ref. 496 528 496 529 11 German test populations totalling 80 men and 80 women were evaluated with respect to Zn intake. InsuYcient dietary intake was observed for 12% of women and 8% of men (in German) See Cu, ref. 476 Concentration changes, associated with aging in rats, were demonstrated 476 530 See Cd, ref. 467 Samples, 100-200 mg, were digested with 30% HNO 467 531 3, 2 ml, for 10 min. The solution was then diluted to 10 ml. Alternatively 20 g of sample was subjected 205 to a 2-stage enzymatic digestion There was no correlation between serum concentrations and dietary intakes of Zn in elderly Spanish institutionalized subjectsTable 1 (Continued) Cows milk Foods Zn Zn Zn Infant formula, infant food Urine Semen Plasma, urine Zn Zn Zn Zn Amniotic fluid Red cells Zn Pharmaceuticals Zn Zn Zn Various Snake and bee venom Cancer tissue Biological and clinical samples, beverages, foods Various Various Various Clinical specimens Clinical samples Clinical fluids, tissues Potatoes Serum, liver Various (16) Various Plasma Herbs, spices Wheat Various (11) Various (8) Various (13) Brown Trout Various (4) Chinese teas Diabetic diets Various (12) Various Hazelnuts Various (9) Nuts, edible seeds Various (4) Olives Various (6) Orange juice Various Spinach Various (8) Various (5) Various Biological samples, foods, waters Milk, infant formula (4) Olive oil Various (8) Serum Various (4) AA;F;L AA;F, air-C2H2;L AA;F;L AA;ETA;L AA;F;L MS;-;L See Cr, ref. 472 AA;F;L See Cr, ref. 473 MS;ICP;L See Cu, ref. 56 MS;ICP;HPLC See Cu, ref. 478 Equal volumes of sample and HNO were taken to 185 °C and 185 psi. The AA;F;L AE;DCP;L XRF;-;S XRF;-;S -;-;- -;-;- -;-;- -;-;- -;-;- -;-;- -;-;- AA;-:- AA;-;- AA;-;L AA;-;L AA;-;L AA;-;L AA;-;L AA;-;L AA;-;L AA;-;L AA;ETA;L AA;ETA;L AA;ETA;L AA;ETA;L 318 532 See Fe, ref. 318 Dietary intake in Egyptian urban populations was determined. Phytate was also determined in order to assess impact on Zn bioavailability See Al, ref. 370 370 472 473 56 478 202 3 solution was evaporated to dryness and dissolved in 6 M HCl. Further purification by extraction with diisopropyl ether and ion exchange chromatography was undertaken See Cd, ref. 143 143 207 Determination of the Zn concentration was one of various investigations used to identify suspicious solutions See Cu, ref. 200 The 1997 ASU 200 5335 1291 A review of recent developments for trace element analysis This review examines protocols to achieve high quality results An introduction to procedures for sample preparation, speciation and measurement of trace elements in clinical specimens A neural network, incorporating a database containing 1000 samples was used 5344 to determine geographic origin Methods for sample preparation and analysis were described.Interferences in ICP-MS were listed and applications of coupled techniques were given (in German) A summary of interlaboratory surveys with comparisons of analytical techniques 130 535 536 (Ca, Cu, Fe, K, Mg, Na, P, Pb, Se, Sr, Zn) (in Japanese) Levels of Cd, Cr, Cu, Fe, Mn, Ni, Pb and Zn in samples imported into Poland were measured. Concentrations were found to be permissible (in Polish) Elemental profiles were used to assess species, origin and variety of wheat following analysis by both ion chromatography and AAS Ni pollution from a disused Norwegian smelter and mine was investigated by 160 determining Co, Cu, Ni and Zn in fish samples.Although the levels found were concluded not to pose a threat to human health, severe physiological aVects were reported in the fish collected from one of the sampling sites The eVect of diVerent production processes on elemental levels was investigated 537 410 538 The trace element profile of diabetic meals served in a Japanese hospital were evaluated.Low energy meals, 1200 kcal, were found to contain Cu, Fe, Mg, and Zn levels below recommended daily intakes (in Japanese) A range of nutritional parameters in 6 varieties of Tarragona hazelnuts were measured.Of the elements determined, only K showed significant variation 539 between varieties (Ca, Cr, Cu, Fe, K, Mg, Mn, Na, Zn) A survey of a wide range of samples-peanuts, pistachios, almonds, hazelnuts, etc.-found Cd, Cu and Zn in blue poppy and sunflower seeds and Zn in pumpkin seeds to exceed permissible levels. Pb was acceptable in all samples (in Polish) The content of Co, Cr, Cu, Fe, Mn, Ni and Zn were determined in 5 olive 540 541 varieties at diVerent stages of ripening. The metals were selected as they are believed to be catalysts in the oxidation of olive oil The combination of Ba, Rb, K and isotopic data allowed the geographical origin of samples from Brazil, Florida and Israel to be correctly assigned A Spanish group reported on the levels of Ca, Cu, Fe, K, Mg, Mn, Na and Zn in spinach.The 2 papers reported on commercial and then experimental 365, 366 samples. Over a 4-month period of storage the authors reported that both sample sets showed some evidence of increases towards the end of this time for some elements (in Spanish) An Ir-coated tube was used to trap and enrich hydrides of As, Bi, Hg, Sb and 350 Se, yielding LODs in the range 0.03-0.08 mg l-1 (in German) Powdered samples were suspended in a medium containing 1% HNO 326 3-20% ethanol and varying levels of H2O2.Liquid samples were diluted with the suspension medium. The resulting solution was introduced in the ETAA 325 spectrometer and measurement conducted against aqueous standards, containing 2% lactoalbumin in the case of Mo (Al, Cr, Mn, Mo) Samples were dissolved in 1,4-dioxane with the addition of N,N-hexamethylenedithiocarbamic acid hexamethyleammonium salt as a universal modifier for Al, Cd, Cr, Cu, Fe, Mn, Ni and Pb.Aqueous standards could be used for 111 calibration Reference ranges for healthy children at diVerent ages were established (Cu, Mn, Se, Zn) (in German) 767 J.Anal. At. Spectrom., 1999, 14, 717-781Table 1 (Continued) Sparkling wine Various (6) Waters Various (28) Beer Various (4) Clams, mussels Various (10) Confectionery Various (8) Dairy products Various (4) Mango juice Various (6) Mineral water Mussel CRM Various (4) Various (16) Papaya Various (8) Sugars Various (4) Teas Various (9) Vegetables Waters, salt Various (11) Various (12) Wine Various Wine vinegar's Various (9) Various (7) Human hair CRM, Dogfish liver and muscle CRMs Foods Various Beverages, foods Various Cactus juice CoVee Various (27) Various (11) Various (4) Dog fish muscle and lobster SRMs Herbal medicines Various (16) 768 J.Anal. At. Spectrom., 1999, 14, 717-781 AA;ETA;L AA;F, air-C2H2;L AA;CV;L AA;ETA;L AE;ICP;L AA;F;L AE;F;L AA;F, air-C2H2;L AA;CV;L AA;ETA;L AA;F, air-C2H2;L AA;F;L AA;F, air-C2H2;L AA;F, air-C2H2;L AA;F, air-C2H2;L AA;ETA;L MS;ICP;L NAA;-;- AA;F;L AA;F, air-C2H2;L AA;Hy;L AA;F;L AA;F;L AA;F, air-C2H2;L AA;F, air-C2H2;L AE;F;L AA;F, air-C2H2;L AE;F;L AA;Hy;L AA;F, air-C2H2;Sl AE;glow discharge;S,L AE;ICP;L AE;ICP;L AE;ICP;L AE;ICP;L AE;ICP;L 377 296 To avoid losses associated with eVervescence when de-gassing sparkling wines a method based on ultrasonic agitation was developed.The method gave comparable results to and oVered practical advantages over reference methods (Cd, Cu, Fe, Hg, Pb, Zn) A wide range of elements were determined in high purity and drinking waters following preconcentration by distillation, using a vacuum apparatus constructed from high purity silica and co-precipitation with 8,8¾ diquinyldisulfide. A round robin procedure was used to validate the method 381 A dual manifold FI system for simultaneous determination of Ca, Mg (by performed on-line dilution and addition of La3+, greatly reducing the con- FAAS), K and Na (by flame photometry) was developed.The system 542 sumption of this reagent Pollution in the Po river, Italy, was assessed by measuring Al, Cd, Co, Cr, Cu, Fe, Hg, Mn, Pb and Zn in local shellfish samples 543 544 Cu, Fe, Mg, Mn and Zn were determined directly on aqueous solutions of the sample, Cd, Pb and Ni following chelation and extraction using APDC-IBMK (in Polish) A closed vessel microwave method for determining Ca, K, Mg and Na in cheeses, caseinates and skimmed milk was described.Standard additions 545 calibration was required to achieve optimum accuracy 10 ml juice+5 ml HNO was heated to 105 °C for 45 min. Following filtration 3 the samples were diluted to 25 ml with H2O and Ca, Cr, Fe, K, Na and Zn determined at 422.7, 357.9, 248.3, 766.5, 589 and 213.8 nm, respectively.The 546 procedure was also proposed for the analysis of urine and milk Measurements were made at 589, 766.5, 422.7 and 285.2 nm for Na, K, Ca and Mg, respectively Samples were sonicated for 120 min in a solution of 1.6 M HNO 288 3-12 M HCl-0.1 M H2O2. FAAS or ETAAS was used to determine elements leached into the solution. Results were confirmed using ICP-MS and NAA. Quantitative leaching was obtained for 9 elements 368 Levels of K, Mg, Mn and Zn were found to be aVected by variations in freezing process and storage conditions.Na levels were not in agreement with those reported by other authors (Ca, Cu, Fe, K, Mg, Mn, Na, Zn) (in Spanish) Sample preparation procedures for determining As, Cu, Fe and Pb were described 290 289 Infusions were prepared and treated using UV decomposition in the presence of H2O2. Decomposition was necessary to remove tannic acid which gave elevated 409 321 levels for Fe and Mn (Al, Ca, Cu, Fe, K, Mg, Mn, Na, Zn) Elemental concentrations in 20 species of vegetable from Bangladesh were determined (Ca, Cd, Cu, Fe, K, Mg, Mn, Na, Ni, Pb, Zn) Samples were diluted in 1% HCl, then injected into an FI-FAAS system incorporating a STAT. The signal enhancement over conventional FAAS was 417 1.2-3.3-fold (in Chinese) The geographical origin of 31 wines from Argentina, Brazil and Uruguay was correctly assigned using elemental data and PCA 419 Forty wine vinegars from Southern Spain were analysed for their As, Ca, Cu, Fe, K, Mn, Na and Zn content.Pattern recognition allowed quick and slow processed vinegar's to be distinguished 547 Hair was cleaned and powdered in a zirconia mill. Portions, 0.1 g, were dispersed in H2O and diluted to 25 ml. Samples, 0.25 or 0.5 ml, were treated with HNO3 to give a concentration of 1% and the slurry diluted to 5 or 10 ml. The solution was stirred ultrasonically prior to aspiration into the FAA spectrometer (Ca, Cu, Fe, K, Mg, Na, Zn) Four sample introduction procedures were evaluated: glow discharge itself, direct 548 insertion, electrothermal vaporization, and electrospray.Direct insertion was good for some elements, but precision was poor. The other techniques oVered better precision, but were more time consuming as sample preparation was required A method was described for confirming compliance with the US Nutrition Labelling Act (1990).Sample dissolution was achieved using microwave 549, 550 551 420 digestion Variation in juice trace element and pigment content at diVerent stages of ripening was measured (in Chinese) 41 samples of green coVee belonging to the varieties arabica and robusta were analysed by ICP-AES and the samples characterized using PCA and cluster analysis (Ba, Ca, Cu, Fe, K, Mg, Mn, Na, P, Sr, Zn) As, Cd, Pb and Se were determined following on-line preconcentration 552 350 Sixteen elements were determined at LODs in the range 0.1-5 ng ml-1 following wet acid digestion using HNO and H2O2 (in Chinese) 3Table 1 (Continued) Various Human breast milk, infant formula Milk powder SRMs Various (11) Various Milk powder SRMs Multivitamin Various tablets Saline solutions, Food CRMs Various (18) Serum, red cells Wine Various (5) Various (15) Wine Various (25) Wheat Various (8) Biological CRMs Various (11) Various Biological materials, foods Various Drinking water, calcium supplements, milk Food SRMs Various (13) Various Foods, wine, neuroblastoma cells Various Various Herbal medicinal teas Human hair and wheat flour RMs Milks Various (11) Mussel SRM Various (9) Various Plant derived foods Rice Various (REE) Rice flour RM Various (4) AE;ICP;L MS;ICP;L AE;ICP;L AE;ICP;L AE;ICP;L AE;ICP;L AE;ICP;L AE;ICP;L AE;ICP;L AE;laser;S MS;ICP;L,Sl MS;ICP;L AE;ICP;L MS;ICP;L MS;-;L AE;F;L MS;ICP;L MS;ICP;L MS;ICP;L,Sl AE;ICP;L,Sl MS;ICP;L MS;ICP;L MS;ICP;L MS;ICP;L MS;ICP;L NAA; -;- MS;ICP;L 38 The paper reported studies to investigate the binding pattern of trace elements in formulas compared with breast milk and the relationship between trace elements in breast milk and in maternal dietary intake.SEC-ICP-AES or SEC-ICP-Ms was used to speciate elements of nutritional significance.Considerable diVerences in binding forms of elements, e.g., Fe, were found in the 2 fluids 553 Aqueous solutions were prepared containing Sc as internal standard and analysed directly using an axially viewed ICP-AE spectrometer. Addition of Cs as ionization buVer was necessary (Ba, Ca, Fe, K, Mg, Mn, Na, P, S, Sr, Zn) An almost identical presentation to ref. 553 was given 554 147 3 Samples were digested using HNO -HCl, 1+1.Subsequent analysis by an open vessel procedure gave levels 40% below label claims.Use of microwave digestion gave results in compliance with the label 36 An axial spectrometer equipped with a microconcentric nebulizer was compared with a concentric glass nebulizer. The analytical performance of the latter was generally better, although the scale was element dependent. Increased salt concentration aVected the stability and sensitivity of the MCN to a greater extent 40 418 Trace elements and enzymes were measured in blood of healthy infants at intervals from birth to four months of age (Ca, Cu, Mg, P, S) (in German) 17 wines from 6 regions were analysed by ICP-AES and the data interrogated using an artificial neural network. Perfect classification was achieved, better than other multivariate procedures such as PCA, Fischer discrimination and Bayes discrimination 555 Wine, must and lees were digested using HCl-HNO3, 4+1, in a microwave oven.The role of the elements determined in vine growth and wine character was discussed (in French) 556 A novel spectrometer was described. A 5 ns pulse from a Nd5YAG laser was focused onto the sample and the resulting radiation carried via an optical fibre to an e�chelle spectrometer fitted with a CCD camera. It was claimed that 60 elements could be determined simultaneously with LODs of a few mg kg-1 over the spectral range 180-750 nm (Al, Ca, Fe, K, Mg, Mn, P, Si) (in German) 17 Sample, 20-100 mg, was mixed with 25% m/v TMAH, 10-200 ml.Complete dissolution was achieved for animal tissues, slurries obtained from other samples. Electrothermal vaporization was used for sample introduction. Cr and Cd could not be determined in bovine muscle due to spectral interference and matrix eVects, respectively 19 This conference paper described numerous benefits of employing tertiary amine mixtures in dissolution media, e.g.: 1, neutralizing HF, thus preventing attack on the Si containing parts of the ICP and allowing low level determination of Si; 2, signal enhancement for As and Se, applied in the determination of these elements in diets and breast milk; 3, avoiding acid dissolution in the determination of I 304 68 316, 557 Negative-ion ESMS was used to determine metal cations as EDTA complexes.It was possible to differentiate between common valence states of Fe and V. Oxide interference in determination of REEs was overcome HR-ICP-MS was used following HNO3 dissolution in a microwave oven. Generally good results were obtained, although problems of incomplete dissolution, detector saturation and high background were reported The application of ICP-MS coupled to reversed phase or SEC-HPLC was discussed in 2 conference presentations. One application was the study of tolerance of neuroblastoma cells to Al, with 2 Al-containing low MW species identified, one of which increased in concentration when cultures were exposed to Al. Other applications included the binding of metals by wine proteins and polysaccharides and the speciation of glycoprotein-bound Se in immunologically active plants 286 140Three preparation techniques were compared-slurry nebulization, microwave digestion and simple infusion The merits of dry ashing, microwave and open vessel acid digestion for the 0.0003-0.0039 ng cm-3and RSD <10% for most elements determination of REE was discussed. The LODs were in the range 411 332 The concentration of Al, Ba, Cd, Cr, Cu, Mg, Mo, Ni, Pb, Sn and Zn were determined in cows' milk based formula, breast, soya, dried, bottled and evaporated milk. Except for Ni in the soya milk, levels were close to recommended values The performance of Q-ICP-MS and HR-ICP-MS was compared for the determination of As, Cd, Cr, Cu, Hg, Ni, Pb, Se and Sn following dissolution of the 3-H2O2 in a microwave oven.Perhaps unsurprisingly, those 558 sample using HNO elements traditionally suVering from polyatomic interferences gave more accurate and precise data using the high-resolution instrument Samples were microwave digested using H2O-HNO3, 1+3, and a stepped 364 heating programme. The detection limits were <2.2 pg g-1 REE were determined following sequential microwave digestion using HNO3, H2O2 and then HF 559 Samples were digested in a microwave oven and analysed using ID-ICP-MS (Cd, Cu, Pb, Zn) 769 J. Anal. At. Spectrom., 1999, 14, 717-781Table 1 (Continued) Serum, bone Various Serum, urine Various Tea Various (28) Teas Various (REE) Water Various Wine Various (33) Wine Various Brain, vegetables Various Various Vegetables, Chinese tea Blood Urine, blood Various (5) Various Blood Various (5) Blood cells Blood, serum Various (4) Various (5) Urine, serum Various Urine Various (5) Urine, blood Various (6) Urine Various (9) Urine Various Urine Serum Various (4) Various (4) Serum proteins Various (9) Serum Various (8) Various Serum proteins, DNA Metallothionein Various (4) Various (5) Various (6)Cerebrospinal fluid Cerebrospinal fluid 770 J. Anal. At. Spectrom., 1999, 14, 717-781 MS;ICP;L MS;ICP;L MS;ICP;L AE;ICP;L MS;ICP;L MS;ICP;L MS;ICP;L MS;ICP;L XRF;-;S XRF;-;S, LMS;ICP;L AA;F;L AA;F;ETV AA;F;L AA;ETA;L MS;ICP;L MS;ICP;L AA;F;L AA;ETA;L AA;Hy;L AA;CV;L AA;ETA;L MS;ICP;L MS;ICP;ETV MS;ICP;L -;-;L MS;ICP;LC MS;ICP;LA MS;ICP;LC AA;F;HPLC Isoforms of metallothionein were separated by SEC or ion exchange HPLC and a quartz T-tube interface linked the eluent to the AA detector (Ag, Cd, Cu, Zn) MS;ICP;HPLCA combined ion exchange chromatography and SEC procedure was developed to investigate binding of elements to proteins (Ca, Cu, Fe, Pb, Zn) Specimens from premature infants were analysed (Ca, Cu, Pb, Rb, Sr, Zn) MS;ICP;L 560 75 Concentrations of trace elements in normal and uraemic sera were measured by HR-ICP-MS The high resolution of a sector field mass spectrometer was used to eliminate spectral interferences.Samples were acidified and diluted 20-100-fold and an internal standard added Elemental analysis in conjunction with PCA was able to discriminate between 421, 561, 562 3 331 African and Asian and between Chinese and other Asian teas. A total of 15 teas from 10 countries were analysed following digestion using HNO3, 0.4 ml Sample, 0.2-0.3 g, was weighed into a PTFE insert and 3 ml HNO H2O2, 0.2 ml HClO4 and 0.2 ml HF added.The vessel was capped and the sample digested via a microwave-heating programme. After cooling the inserts were removed, the contents washed into PTFE beakers using 1 ml HNO3, evaporated to near dryness and adjusted to pH 0-0.5 using NaOH. A cation exchange column was then used to separate REE from the matrix, the former being determined using ICP-MS with Re as internal standard 563 416 The molar response curves for both a quadrupole and a sector ICP-mass spectrometer were calculated using a series of internal standard elements. The results were used in conjunction with correction factors for ionization eYciency to improve the accuracy of semi-quantitative analyses 17 white and 13 red wines from 13 wineries in the Okanagan valley, Canada, were analysed. A total of 25 elements correlated strongly with the vineyard of origin.The wine element fingerprints were deemed soil signatures that survived metabolic and winery influences 415 106 FI-ICP-MS was used to analyse 112 wines from Spain and England The region of origin of the Spanish wines was given unequivocally. It was also possible to diVerentiate Spanish and English white wines with 100% accuracy A new XRF spectrometer, capable of spatial mapping at a resolution better than 20 mm, was described. Autopsy samples showing brain calcification and leaves from plants exposed to acid rain and X-ray radiation were among the examples cited (in Japanese) 439 564 Samples were separated into a cytosol and pellet fraction by ultracentrifugation and the binding of metals to high and low molecular weight proteins investigated (in German) 10 ml capillary blood was diluted with 100 ml 0.05%Triton-100 and the full amount aspirated via the nebulizer (Ca, Cu, Fe, Mg, Zn) (in Chinese) 10 ml sample was placed onto a tantalum filament from which ETV into the 125 565 40 ml blood and 1.5 ml 5 g l-1 Triton X-100-5 g l-1 Na flame occurred 2EDTA for measurement of Cu and Zn. The solution was diluted a further five-fold for determination 566 Mg(NO of Ca, Fe and Mg (in Chinese) Cell types were isolated by density centrifugation, digested and analysed with 3)2-Pd-H2O2 as the chemical modifier (Cr, Cu, Se, Zn) 21 An automated preparative system was developed with a microwave digester and an iminodiacetate-based resin column for separation of analytes from the matrix elements. Analysis rate was 6 samples h-1 (Cu, Fe, Ni, Pb, Zn) ICP-MS for measurement of trace elements in clinical specimens was reviewed 49 115 with emphasis on ways to eliminate interferences Specimens from patients with Blackfoot disease were digested for analysis.As, Hg and Pb concentrations were increased, Se and Zn were reduced, compared with normal controls (As, Hg, Pb, Se, Zn) 567 10 Application of a controlled voltage caused electrolytic movement of analytes from the sample into the Pd-coated furnace. Residual sample, which contains the components responsible for interferences, was removed and the furnace heated for atomization (Cd, Co, Cr, Mn, Ni, Pb) A comparison of H2O dilution, UV irradiation and HNO3-H2O2 digestion, prior to measurement by Q-ICP-MS or magnetic sector field ICP-MS, was 568 reported (Cd, Co, Cr, Ni, Pb, Rb, Sb, Se, V) Advantages of ETV, of small sample volume and reduced interferences, were discussed in the context of monitoring occupational exposures After a series of investigations a sample dilution of 1+9 with In as internal 76 569 570 standard gave acceptable results for HR-ICP-MS (Cd, Cu, Pb, Zn) Conditions were optimized for separation of metalloproteins by SEC with element specific detectors. The preferred column material was Asahipak, a vinyl alcohol copolymer (Ca, Fe, Na, Zn) Serum proteins were separated on Mono Q HR 5/5 anion exchange Fast Protein LC, and the associated metals were detected by coupled double focusing 59 ICP-MS (Al, Cr, Cu, Fe, Mn, Se, Sn, Sr, Zn) Proteins were separated by gel electrophoretic techniques and the trace elements detected by LA-ICP-MS (Au, Cd, Cu, Fe, Hg, Pb, Pt, Zn)Binding of metals to proteins and DNA was investigated 571 572 573 574Table 1 (Continued) Ileostomy fluid Various (5) Intestinal fluids Body fluids Various (8) Various (11) Various (42) Hair, blood, red cells Biological tissues Various Bone Bone substitutes Various (58) Various (24) Teeth Teeth Various (4) Various (7) Fetal tissues Various (26) Placental tissues Various Neural tissues Amyloid plaques Lung Various (6) Various (8) Various (8) Liver Various Tissues Various Tissues Various (10) Tissues Various Tissues Various (10) Various Various Various Biological materials Biological specimens Biological materials Various Biological materials Biological materials Various (4) Various Biological specimens Various Various Biological specimens Biological specimens Hair Various Hair Various (4) Hair Hair Various (7) Various (5) Various Pharmaceutical preparations MS;ICP;SEC AE;ICP;L AA;Hy;L AA;ETA;L MS;ICP;L MS;ICP;L MS;ICP;L AA;Hy;L AA;CV;L AE;ICP;L AA;F;L AE;F;L AE;ICP;L MS;ICP;L AA;F;L AA;ETA;L AA;ETA;L AE;ICP;L MS;ICP;L AA;F;L AA;ETA;L the lung were determined and the eVects of smoking, sex and occupational exposure were investigated (Al, Cd, Cr, Cu, Mn, Ni, Pb, Zn) MS;ICP;HPLCProteins from a liver extract were separated and the attached metals determined MS;ICP;L MS;ICP;L MS;ICP;LA AE;ICP;L MS;-;- -;-;- MS;-;- AA;ETA;L AE;ICP;L AA;ETA;Sl MS;ICP; capillary electrophoresis AE;ETA;HG AE;ICP;ETV -;-;- AA;F;L AE;MIP;L AE;ICP;L -;-;- 192 575 Fluid collected from ileostomy patients following ingestion of a meal spiked with 106Cd was centrifuged and the Cd species separated by SEC. ICP-MS was used to measure 106Cd, 111Cd, 63Cu, 57Fe, 206Pb and 66Zn Binding of elements to tea polyphenols was investigated in an in vitro, ultrafiltration study assessing bioavailability (Ca, Cu, Fe, Mg, Mn, K, Na, Zn) 576 577 On-line solid phase extraction and other recent developments were reviewed (Al, As, Co, Cr, Fe, Mn, Mo, Ni, Se, V, Zn) (in German) 0.2 g sample and 3 ml HNO3 were heated at 115 °C for 30 min. Good recovery of volatile elements was observed and large numbers of specimens were readily prepared Equipment and conditions for the microwave digestion of small samples were 578 132 43 discussed Laboratories collaborated in providing data to add to the usefulness of SRM 1400 Bone Ash. See also Pb, ref. 132 Wollastonite (CaSiO3) preparations were heated in acid (conditions were optimized for diVerent analytes and procedures) 27 3 44Powdered tooth material was dispersed in HNO and then pumped through a coiled PTFE tube in a microwave oven (K, Mg, Na, Sr) 2O2 and ashed at 450 oC under O2. The 3 (Al, Cd, Washed teeth were soaked in 5% H residues were powdered and 0.5 g digested in 10 ml 50% v/v HNO Cr, Cu, Fe, Pb, Zn) 53 Samples collected from 21 human fetuses aged 16-22 weeks were digested with acid using microwave heating. Particular attention was given to contamination control and to demonstration of accuracy and precision. Concentrations increased with fetal age and were lower than in adult tissues Dried specimens were microwave digested and analysed to determine normal 113 167 54 values Tissue concentration changes during uterine and infant development were monitored following maternal exposure to Al (Al, Ca, Fe, Mg, Mn, Zn) Amyloid plaques were extracted using an FI system from brain preparations taken from patients with Alzheimer's disease (Al, Cr, Cu, Fe, Mn, Ni, Pb, Zn) Tissues from urban dwellers were analysed. Concentrations in diVerent parts of 114 579 580 3 in 20 ml vessels using by ICP-MS 50-100 mg samples were digested in 700 ml HNO microwave heating. Residues were diluted with H2O to reduce the acid content 581 3 Samples were heated with HNO and then with H2O2. These solutions were diluted and analysed. Procedures to compensate for interferences were 60 29 3 described (As, Ca, Cl, Co, Cr, Hg, Ni, Se, V, Zn) Soft tissues were placed into a special cell into which liquid N2 was added. The laser beam was then focused onto the surface of the frozen, solid specimen A device for vapour phase HNO digestion of 50-100 mg specimens was used and microwave heating was successfully used to achieve decomposition (Al, As, Ca, Cd, Cu, Fe, Mn, Mo, Pb, Zn) Applications of ICP-MS, TIMS and AMS were reviewed 48 582 Methods for decomposition of specimens prior to analysis were reviewed (in Czech) A bibliography of publications involving applications of MS 583 584 Dry ashing and wet digestion methods for sample decomposition were reviewed in detail. Enzyme catalysed and UV photolysis methods were also discussed (in Czech) Dried, powdered specimens were suspended in 5% HNO 14 3.The slurry was analysed using very rapid heating programmes and results were in agreement 585 with expected values Cd, Cu, Mn, Pb Capillary electrophoresis was used to speciate nl volumes of specimen. Interfaces to link the output to the ICP were discussed FANES, with solid phase preconcentration steps and HG, was reviewed 586 A tungsten coil atomizer vaporised samples into the ICP 118 134 137 3 Trace element concentrations in hair were shown to be subject to external influences and not to correlate with blood or tissue content Hair from men with alopecia was digested with HNO -HClO4. Concentrations of Mn and Zn were lower, Cu was increased and Fe was unaltered compared with control samples (in Chinese) Data from samples from Chinese schizophrenic patients were reported (Co, Cu, 587 139 588 Fe, K, Mg, Mn, Na) Acid-assisted microwave digested samples were analysed as part of a study of influences of Antarctic conditions on human health (Ca, Cu, Fe, Mg, Zn) Various atomic spectroscopy techniques suitable for pharmaceutical research were discussed 771 J. Anal. At. Spectrom., 1999, 14, 717-781Table 1 (Continued) Dialysis fluid Various (7) Cocaine, heroin Cannabis Various (4) Various Chinese herbal medicine Various (4) Various Drugs, biological specimens Bone Plasma Tooth enamel Various (12) Various (16) Various (4) Hair Various (8) Tissue slices Various *Hy indicates hydride and S, L, G and Sl signify solid, liquid, gaseous or slurry sample introduction, respectively. Other abbreviations are listed elsewhere. 3 References

ISSN:0267-9477    

DOI:10.1039/a900785g

出版商:RSC    

年代:1999

数据来源: RSC