JOURNAL OF ANALYTICAL ATOMIC SPECTROMETRY APRIL 1992 VOL. 7 67R ATOMIC SPECTROMETRY UPDATE- CLINICAL AND BIOLOGICAL MATERIALS FOODS AND BEVERAGES Andrew Taylor" Supra- Regional Assay Service Metals Reference Laboratory Robens Institute of Industrial and Environmental Health and Safety University of Surrey Guildford Surrey GU2 5XH UK Simon Branch The Lord Rank Research Centre R. H. M. Research and Engineering Lincoln Road High Wycombe Buckinghamshire HP12 3QR UK Helen M. Crews Ministry of Agriculture Fisheries and Food Food Safety Directorate Food Science Laboratory Colney Lane Norwich NR4 7UQ UK David J. Halls Trace Element Unit Institute of Biochemistry Royal Infirmary Castle Street Glasgow G4 OSF UK Summary of Contents 1 Analysis of Clinical and Biological Materials 1.1.General Reviews and Comments 1.2. Sampling and Sample Preparation 1.3. Developments in Multi-element Analysis 1.3.1. Inductively coupled and direct current plasma atomic emission spectrometry 1.3.2. Inductively coupled plasma mass spectrometry and other mass spectrometric techniques 1.3.3. X-ray fluorescence spectrometry 1.3.4. Other multi-element techniques and studies 1.4. Developments in Single-element Techniques 1.5. Reference Materials and Inter-laboratory Trials 1.6. Hair Analysis 1.7. Progress for Individual Elements 1.7.1. Aluminium 1.7.2. Arsenic 1.7.3. Boron 1.7.4. Cadmium 1.7.5. Calcium 1.7.6. Chromium 1.7.7. Copper 1.7.8. Fluorine 1.7.9. Germanium 1.7.10. Gold 1.7.11. Iron 1.7.1 2. Lanthanides 1.7.1 3. Lithium 1.7.14. Lead 1.7.1 5. Magnesium 1.7.1 6.Manganese 1.7.1 7. Mercury 1.7.1 8. Nickel 1.7.19. Potassium and sodium 1.7.20. Platinum 1.7.21. Selenium 1.7.22. Silicon 1.7.23. Silver 1.7.24. Strontium 1.7.25. Thallium 1.7.26. Tin 1.7.27. Uranides 1.7.28. Vanadium 1.7.29. Zinc 1.8. Conclusions Table 1. Summary of Analyses of Clinical and Biological Materials * Review Co-ordinator to whom correspondence should be addressed.68R JOURNAL OF ANALLYTICAL ATOMIC SPECTROMETRY APRIL 1992 VOL. 7 2 Analysis of Foods and Beverages 2.1 . Sample Preparation 2.1.1 Preconcentration 2.1.2 Digestion 2.1.3 Solid Sampling 2.2. Developments in Hydride Generation Techniques 2.3. Speciation Studies 2.4. Developments in Methodology for Atomic Absorption Spectrometry 2.4.1. Flame atomic absorption spectrometry 2.4.2. Electrothermal atomic absorption spectrometry 2.5.Developments in Methodology for Plasma Emission Spectrometry 2.6. Developments in Methodology for Inductively Coupled Plasma Mass Spectrometry 2.7. Multi-element Analyses of Foods 2.8. Progress on the Determination of Some Individual Elements 2.8.1. Aluminium 2.8.2. Mercury 2.8.3. Selenium 2.8.4. Vanadium 2.9. Dietary Intake Studies 2.1 0. Characterization Studies 2.1 1. Reference Materials and Collaborative Trials Table 2. Summary of Analyses of Foods and Beverages This the seventh Update to review developments in atomic spectrometry as applied to the analysis of clinical and biological materials foods and beverages and covers the references 91 /826-91/4050 and 92/1-92/338. These are listed in detail as Atomic Spectrometry Updates References in Volumes 6 and 7 of JAAS.The regular increase in the number of papers reviewed has been maintained. Application of more multi-element techniques to this particular topic area is noted with several alternatives to ICP-AES now being employed. How useful these are likely to be in the world of working analysts should become evident in future Updates. 1. ANALYSIS OF CLINICAL AND BIOLOGICAL MATERIALS David J. Halls and Andrew Taylor This review covers recent publications and conference proceedings relating to the analysis of clinical and biologi- cal samples by atomic spectrometry and reflects a year of consolidation rather than of major advances. Interest in multi-element techniques continued to increase but much valuable work was still done by single-element techniques for which developments in methodology continue to be made.Information on the methods is summarized in Table 1. 1.1. General Reviews and Comments For any work on trace elements in disease states reference needs to be made to normal values. The values need to be accurate and should be in accordance with the best values in the literature. Particularly relevant this review year is the publication of an extensive study on normal values for urine blood and serum by Minoia et al. (91/2405). Using NAA ETAAS and ICP-AES 46 elements were determined in urine 35 in blood and 26 in serum of 350 healthy Italian subjects. Other studies from Italy produced reference values for trace elements in hair for subjects under the age of 15 years (91/887) and minor and trace elements in lung tissue (92/78).Not all studies meet the high standards found in the studies mentioned above. Cornelis a champion of accuracy in analysis has discussed the possible errors in the determi- nation of trace elements in body fluids and tissues (9 1K1828). All stages need to be considered representative samples contamination- and loss-free sampling specificity and selectivity in the analytical method quality control of the analysis and evaluation of the data. The theme of quality assurance was taken further by Heydorn (9 l/C 1829) particularly for the laboratory carrying out analyses for which there are at present no quality assessment schemes or where the numbers of analyses do not justify participation. Methods described were analysis of the precision of results detection of analytical bias by comparison with another method using a different detection principle traceability of calibration standards and verification of accuracy by analysis of CRMs.How much dirt do kids eat? Barnes’ review (91/3263) has some of the answers. Analytical approaches to monitor soil irzgestion by children were described using marker elements determined by ICP-AES and ICP-MS. 1.2. Sampling and Sample Preparation An extensive survey of possible trace element contamina- tion in sampling blood was made by Paudyn et al. (9 1/3338). Inductively coupled plasma MS was used for the determination of a range of 29 elements. Plastic containers were found to be generally suitable for most elements. Blood sample tubes containing EDTA were found to be particularly unsuitable. Errors caused by sampling blood with stainless-steel needles or PTFE catheters were also examined.To reduce contamination from stainless-steel surgical tools in collecting tissues Pietra et al. (91/827) coated the tools with TIN by a reactive ion-plating technique. The TiN layer had high chemical stability and good wear resistance. The tools were applied to the determination of Co Cr and W in the skin of patients on regular dialysis and of control subjects. Progress in sample digestion by microwave heating isJOURNAL OF ANALYTICAL ATOMIC SPECTROMETRY APRIL 1992 VOL. 7 69R being made by attempting to automate the process. Kings- ton et al. (911C1811) saw this as an opportunity also to standardize procedures between laboratories.Expert sys- tems have been used to develop suitable standard methods. Grillo et al. (911C3672) saw microwave digestion as part of a complete robotic sample handling area producing samples for determination by ICP-AES and AAS. An automated on- line flow-through technique is a simpler and cheaper solution with which Barnes and Martines (9 11C369 1) have been working. In this method sample and acid were pumped into the cavity of a laboratory microwave oven for heating. More conventional applications of microwave heating continue to be reported (9111494 9113271). Matu- siewicz et al. (91/3776) however have adapted the mi- crowave pressure digestion technique to allow vapour- phase digestion. The advantages of vapour-phase attack are the absence of acid in the sample minimization of contamination and very low blanks.Within the PTFE bomb a PTFE microsampling cup containing the sample was located above the acids (HN03-HF for a marine sediment RM and HN03 for the marine biological tissue TORT-1). The residue after attack was dissolved in 0.5 mol dm-3 HN03. Analysis of the sample digests by FAAS and ETAAS gave results that were in good agreement with certified values for 15 elements. A simple procedure for conventional digestion of marine organisms was described by Hernandez et al. (91/1493). Lyophilized samples were digested with HN03 in Erlen- meyer flasks at 70 "C for 24 h. The elements Cd Cr Cu Mn and Pb were determined by ETAAS and Hg by cold vapour AAS. No losses of the volatile elements Cd and Hg were seen.Lee et al. (921255) compared a modified dry ashing technique that they had developed with conventional dry ashing wet ashing and acid extraction techniques for the determination of Cd Cr Cu Fe Mn Pb and Zn in mussels. Results were comparable but they preferred their modified dry ashing technique as it was easier faster and less vulnerable to contamination. The advantages of slurry sampling continue to be advo- cated by Miller-Ihli (91K1907 9 lK3653). Ultrasonic agitation was used to ensure adequate mixing. Results for seven elements in a wide range of biological materials were obtained with excellent accuracy. It is becoming difficult to tell whether some slurry methods are true analyses of solids or analyses of solutions containing remnants of undigested material.This applies to the work of Fagioli et al. (9 11868 9 111174) who produced carbonaceous slurries by treating biological samples with concentrated H2S04. The tempera- ture was increased to 350 "C over 1 h and then maintained at that temperature for a further 1 h. Results on a wide range of NIST SRMs including Bovine Liver using determi- nation by ICP-AES gave good results. The technique however seems to have little point in that it takes at least 2 h to prepare a sample. True slurry sampling is faster and so is total digestion using microwave heating. Campos et al. (9 1127 10) developed a simple direct solid- sampling technique for AAS. The sample was weighed onto a graphite platform which was placed in the side arm of a quartz T-tube through which air was passed.The main part of the T-tube was in the optical beam of an AA spectrometer and heated by an air-C,H flame or an electrical coil. Combustion was initiated by a flash from three IR lamps focused on the sample. Sample size was limited to about 2 mg and calibration required the use of similar solid SRMs or previously analysed samples. Good agreement was obtained for measurements of Cd and Pb in grass samples analysed by ETAAS and results for Cd Cu Pb and Zn on SRMs were generally within the certified range. The technique could also be used to measure Bi Hg and TI. Poorest results were obtained for Cu probably due to incomplete vaporization. Standardization of solid sampling ETAAS is generally a problem and formed the basis of a study by Atsuya et al. (91/1509).For Cu they evaluated standardization with NIST SRM Tomato Leaves Cu coprecipitated with magnesium 8-hydroxyquinolinolate and Cu coprecipitated with nickel dimethylglyoxime. The first two gave correct results when Cu was determined in a range of biological RMs whereas the last gave low results. Nordahl et al. (91/1512) determined A1 in tissue biopsy specimens by the cup-in-tube technique with a modifier containing Mg(N03) HN03 and Triton X- 100. Variation in A1 concentrations could be overcome by choosing one of eight wavelengths for measurement. Contamination was the main problem found. Heavy metals (Cd Cr Hg Ni and Pb) in urinary calculi were determined by Struebel et al. (921 1 5 13) using direct solid sampling ETAAS in a graphite boat and ETAAS after wet digestion.Results by the two methods generally agreed but Ni could not be determined by solid sampling and Hg only by solid sampling. Hahn et a!. (9 11909 9 11 1 5 1 1) used solid sampling ETAAS for the determination of Cd and Pb along the length of the vane of magpies' feathers. Heavy metal content increased as a function of exposure time. The concentrations were higher in the more exposed parts of the feathers confirming that exogenous deposition primarily determined the concentra- tion. The feathers were considered to be a useful indicator of heavy metals in the environment of the bird's home range. Solid sampling ETV was used by Slinkman and Sacks (9 113266) to introduce samples into a magnetron rotating DCP. Detection limits were in the pg g-* range. A number of NIST SRMs were analysed in this way with calibration by standard additions.Applications offlow injection to AAS in clinical chemistry were reviewed by Shenvood and Rocks (9 11955). 'Their review includes information on controlled dispersion analy- sis and specialized applications such as HG for the determination of Bi and coupling chromatography with AAS. Fang et al. (91/3779) used an FI system to preconcen- trate Pb by coprecipitation with Fe"-hexamethyleneam- monium hexamethylenedithiocarbamate complex. The pre- cipitate was collected in a knotted reactor (interlaced knots in 0.5 mm i.d. plastic tubing). After collection IBMK was pumped through the reactor to dissolve the precipitate and to pass the Pb into the nebulizer of an FAAS system. An enhancement factor of 66 was obtained for a 30 s collection period which gave a sampling frequency of 90 h-l.Work on the use of algae for preconcentrating trace elements has continued. Pappas et al. (91/2217) used Chlorella vulgaris to remove Cr from solution. For concen- trations in the range 5-100 mg l-l at the optimum pH of 3 the maximum percentage of Crvl that could be removed was 75%; Cr1I1 was not so effectively captured. Lowering the pH allowed the Cr to be removed from the algae. Unicellular green algae Chlorella sp. were used by Shengjun and Holcombe (9111550 911C1908) to preconcentrate Cu. The solution containing Cu was agitated with the algae for 20 min and then centrifuged to collect the algal pellet. This was suspended in 1 ml of 0.5% v/v HN03 for determination of Cu by ETAAS using slurry sampling. The efficiency of extraction 65% was relatively unaffected by the matrix composition and volume of sample.The method was applied to sea and river water RMs. Solvent extraction was seen by Parsley (9 113777) as a way of achieving a common sample preparation for the determi- nation of Co Cu Mo and Se in bovine liver a routine analysis in the veterinary field for detection of deficiencies of those elements. Previously different preparation tech- niques and methods of determination had been used which were time-consuming. Liver samples after digestion with HN03-HC104 were extracted with APDC into CHC13. The CHC13 was boiled-off from the separated extract and the residue was treated with HN03 and then dissolved in dilute70R JOURNAL OF ANALYTICAL ATOMIC SPECTROMETRY APRIL 1992 VOL.7 formic acid. Copper was determined by FAAS and Co Mo and Se by ETAAS using wall atomization. For Se a Pd-Mg modifier was used and calibration was performed with Se standards in a Se-free liver digest that were extracted in the same way as the samples. Good accuracy was shown by determination of the elements in NIST SRM Bovine Liver and by comparison with alternative techniques. 1.3. Developments in Multi-element Analysis 1.3.1. Inductively coupled and direct current plasma atomic emission spectrometry Further results from the group at Ispra Italy on reference values for about 50 elements in human lung tissue have been published (92/78). Determinations were made by ICP- AES and NAA. Another study featured in last year’s Update a comparison of NAA and ICP-AES for the analysis of human brain by Andrasi et al.has also been published (92/76). The distribution of the major elements Ca Fe K Mg Na P and S and the trace elements Al B Co Cr Cu Mn Ni Pb and Zn was investigated. Coni et al. reported their extensive study on reference concentrations for 12 elements in 59 samples of human milk (91/879 91/C2879). Samples were lyophilized for storage. The material was dry ashed and then redissolved in dilute HN03. The elements Al Ba Cd Cr Cu Fe Mg Mn Ni and Zn were determined by ICP-AES. Lithium and Pb which gave problems in ICP-AES because of interferences and unsatisfactory detection limits were determined by ETAAS. Satisfactory results were demonstrated for three milk powder RMs.Subjects were divided into urban and rural dwellers and into smokers and non-smokers but no clear differences were seen. Changes in the concentration of Ca Cu Mg P S and Zn in the milk of one mother were studied by Suzuki et al. (91/3968) over a period from 2 to 196 d postpartum. Results obtained by ICP-AES showed decreases in Cu Mg S and Zn from the highest concentra- tions in colostrum milk; Ca and P increased in the transitional milk. Speciation by gel-filtration HPLC-ICP- AES of the whey showed dramatic changes in the distribu- tion of the elements. Several groups investigated ways of handling small samples by ICP-AES. Li et al. (91/1481) designed a recycling nebulization system to measure 1 ml samples particularly digested saliva. The total nebulization time was 20 min and the emission was stable during that time. Flow injection was used by Kamat and Sneddon (91K3685) for direct determination of Cu Fe Se and Zn in blood.For discrete nebulization Isoyama et al. (9 1 / 1 487) developed a small spray chamber and a time-sharing background correc- tion system to cope with the rapid measurements. Solid biological RMs were digested with HN03-HC1-HC104-HF in a PTFE pressure vessel with microwave heating. Results were in good agreement with the certified values and with those obtained by continuous nebulization. Concentrations of eight elements in gallstones were determined by Uchiyama et al. (9 1/4006) using ICP-AES. Differences in element concentrations were seen between the three types of gallstones chlolesterol bilirubin and black stones.Inductively coupled plasma AES is a natural choice for the multi-element determination of trace metals in pharma- ceuticals. Niebargall and Wennrich (9 1 /C2876) described the determination of 1 1 elements in Penicillin G using a V- groove nebulizer for high dissolved solids. The trace element ‘fingerprint’ of drugs of addiction such as cocaine and heroin was investigated by Caroli et al. (9 K2878) as a way of identifying the country of origin. Digested samples were analysed by ICP-AES for 11 elements to provide the fingerprint. Yu et al. (9 1/2485) determined trace elements in a number of traditional Chinese medicines. Samples were first either dry ashed at 490 “C or wet digested with H[W03-HC104. In both cases the dried material was dissolved in 5% HN03 for analysis by ICP-AES.The work of Heltai et al. on the use of an electrothermal atomizer in conjunction with a toroidal Ar MIP has now been published (9 1/1457). This combination gave detection limits ranging from 0.1 to 100 pg l-l which were lower by factors of 3-10 than those obtained with a filament type Ar MIP. The technique was applied to a number of biological SRMs after wet digestion and a separation step to remove the alkali metals. Isoyama and co-workers (9 1/3 105) found it necessary to calibrate by standard additions when using ETV for the determination of trace elements in biological materials by ICP-AES. This was carried out directly in the fiirnace by injecting the sample solution together with a separate addition of a standard solution containing NH4H2P04 as a chemical modifier.A miniature cup technique for handling powdered samples for ETV into the ICP was described by Atsuya et al. (91/3804). The advantages of the Hildebrand grid nebulizer were explored by Musil (911C2758) for the direct analysis of acidified urine and slurries of biological materials. Using solvent extraction Kobayashi and Imaizumi (!>1/4030) could determine Cd Co Cu Ni and Pb in urine simultaneously by ICP-AES with detection limits of 0.4 0.4 0.4 0.2 and 11.7 pg l-l respectively. An internal standard yttrium was used. 1 .:3.2. Inductively coupled plasma mass spectrometry and other mass spectrometric techniques The trend highlighted in last year’s Update (9 1/3594) away from using ICP-MS as a multi-element technique towards applications for which it is ideally suited continues i.e.sitable isotope studies as a very sensitive detector for sjpeciation studies and for measurements of low concentra- tions of elements that are difficult to measure by other techniques. Much of the work covered here under multi- element studies is not in fact new but has been reported previously as conference papers. The work of the group in Ghent Belgium on multi- element determinations in serum has now been published (!J1/1545 91/2965 92/73). In this work serum was diluted 1 + 4 or 1 + 9 with 0.14 mol dm-3 HNO containing l151n as an internal standard for the determination of Br Co Cs Cu Fe Mo Rb Sr and Zn. Correction for polyatomic interferences was necessary for Co Cu Fe and Zn by measurement of a suitable blank solution.The methods were tested on the Versieck second-generation serum RM and the results compared with those obtained by NAA PIXE and atomic spectrometric techniques. Yoshinaga et al. (91/1485) determined a range of trace elements in human liver and kidney by ICP-MS after pressure digestion of the sample with HNO,. Results obtained for NIST Blovine Liver SRM were generally within the certified reference range. Lyon (9 1/C1659) also reported satisfactory analysis of the NIST Bovine Liver SRM as well as the IAEA Kidney and Animal Muscle CRMs. Comparison of ICP-MS data on digests of human autopsy tissue samples with results obtained by AAS highlighted a problem with Zn; this was corrected by reducing the deadtime from the manufac- turer’s recommendation of 68 ns to 3 ns.With this modification results obtained by AAS and ICP-MS com- pared well. Herbal remedies were analysed for 2 1 elements by Thompson and Ward (911933) using ICP-MS. For Ca Ch Fe Mg and Zn results were compared with those obtained by FAAS. The potential of semi-quantitative multi-element analysis by ICP-MS was explored by Amarasiriwardena et al. (Y1/857). Best results were obtained by dividing the element response table into three groups with vanadium caesiumJOURNAL OF ANALYTICAL ATOMIC SPECTROMETRY APRIL 1992 VOL. 7 71R Table 1 CLINICAL AND BIOLOGICAL MATERIALS Element Matrix Ag Plasma A1 Serum A1 Serum bone A1 Plasma urine A1 Urine A1 Biological tissues A1 Serum water A1 Pharmaceutical solutions A1 Serum A1 Serum dialysis fluid urine water A1 Plasma A1 Bone A1 Serum A1 Brain A1 Brain A1 Pharmaceuticals A1 Bone soft tissues Technique; atomization; analyte form* AA,ETA;L AA ; ETA; L AA;ETA;L AE;ICP;L AA;ETA;L AA,ETA;S AA;ETA;L AE;ICP;L AA,ETA;L AE;ICP;L AA;ETA,L AA;ETA;L AE;ICP;L AA ET A;L LMMS;-;S AA;ETA;L S S1 AA;ETA;L AA; ETA;L Sample treatmentlcomments Ag and Au were measured in studies of an anti- inflammatory compound (NH4)2HP04 was added as a chemical modifier and in situ O2 ashing used to reduce matrix effects Discriminant analysis showed that in patients with chronic renal failure serum A1 concentrations were predictive of bone levels within a 7% margin of error An increase in plasma and aluminium concentrations was noted in healthy subjects who ingested sucralfate (4 g d-') for 21 d Renal excretion of A1 was measured in subjects with normal and impaired renal function following normal and enhanced exposure to A1 during plasma exchange therapy Very careful precautions to eliminate contamination were necessary for the direct determination of Al.Chemical modifiers were 0.2% Mg(NO,) in 2% HN03 and 0.2% Triton X-100. Eight wavelengths were available and the one selected depended on concentrations to be measured. Zeeman-effect background correction was used Serum proteins were separated by HPLC and the A1 detected by AAS. Complexes of AlF2+ were identified in water by ion-exchange chromatography and ICP-AES +preparations from glass containers was investigated. Lowest A1 release was found with glass having a dealkalinized surface Instrumental parameters were carefully evaluated to establish acceptable sensitivity The influence of sample container material temperature and treatment on the storage and stability of samples was investigated. Poly(propy1ene) tubes were suitable with storage at -20 "C.Water samples were acidified to prevent loss of A1 Two methods were described. Samples from patients with chronic renal failure were diluted 1 + 3 with HNO,-Triton X-1 00. Specimens with low A1 concentrations were diluted 1 + 1 with Mg(N03)2 solution introduced to eliminate contamination. Bone was solubilized in HN03 at 95 "C and diluted for analysis. Standard additions was used for calibration and other bone consituents did not interfere The effect of gaseous C species on the formation of A1 atoms was studied LMMS studies of neurofibrillary tangles in patients with Alzheimer's disease and other dementia conditions were reviewed Interference from P04'- and signal enhancement by Mg(N03)I was investigated together with an assessment of sample introduction techniques Samples containing alkaline earth elements were successfully analysed (in German) Tissues were heated with HNO at 90 "C for 3-4 h in closed PTFE tubes and the solutions diluted with H20.A1 in bone was determined by reference to standards in a similar matrix prepared from rat bone. For other tissues the standards were in 0.2 g 1-I Ca solution. The bone specimens were analysed in uncoated graphite tubes and other solutions were placed on a L'vov platform in a pyrolytic graphite coated graphite tube ' A1 contamination of the contents of pharmaceutical Extensive cleansing and monitoring steps were Reference 9 1 I2470 9 1 I965 91/1139 91/128 1 9111348 91/1512 9 1 IC 1 693 9 1 lC17 17 9 1lC2 102 91/2171 9112525 911321 4 9113288 9 113462 9 1 lC3758 9113974 9211 372R JOURNAL OF ANALYTICAL ATOMIC SPECTROMETRY APRIL 1992 VOL.7 Table 1 CLINICAL AND BIOLOGICAL MATERIALS-continued Technique; atomization; analyte form* Element Matrix A1 Serum urine Sample treatmentlcomments Reference AA;ETA;L Specimens were diluted with 0.1% Triton X-100. The measurements were calibrated by use of matrix- matched standards and good accuracy was Idemonst rated AA;ETA;L Specimens from subjects with occupational exposure 921250 to alumina in air had pre-shift concentrations of 20-30 pg I-'.Levels in plasma and urine increased during the work period 9212 10 A1 Plasma Al Brain Urine XRF-;- MS;ICP;L Deposits of A1 could not be detected in cellular !organelles of brain tissue from patients with .4lzheimer's disease Ion chromatography was used to separate As species in urine and the eluted material was analysed by (on-line ICP-MS. Samples were diluted 20-fold to eliminate the interference from ArCI+ Techniques for sample preparation separation of As :species and analysis were reviewed Samples were heated with HNO + HClO in sealed PTFE vessels. MgO was added to remove residual 'HN03 prior to As reduction (in Chinese) 921299 911853 As Biological fluids Hair AA;-;- A F Hy;L 9 11886 9112390 As As ASH was formed and collected onto the surface of a graphite tube previously coated with reduced Pd Sep,aration and identification of As from dietary and environmental sources was achieved by HPLC with ICP-AES for detection The interference due to ArCl+ was removed by addition of N2 to the Ar flow.Results were !;uccessfully obtained for total As concentrations i3nd for As species separated by HPLC compounds were established with detection by thermospray MS Arsenic-binding compounds in homogenized liver samples were separated on a Sephadex G75 column. Arsenic in the eluted samples was measured with 0.2% Ni- 1 % HNO as chemical modifier which also removed the C1- and eliminated the Interference from this ion measurement of low levels of As in biological samples (in German) A photoreactor was constructed with 5 m of PTFE tubing (0.5 mm id.) wrapped around a Hg lamp.!$ample solution pumped at 2 ml min-I was mixed with potassium persulfate (0.6 ml min-') before entry into the reactor. The emergent dution went to a continuous ASH generation device for measurement of the released As"'. A series of As compounds were separated by chromatography prior to oxidation and the conversion efficiencies with 36 s reactor dwell lime were more than 95% Conditions for HPLC separations of several As 9 113407 9 llC3686 Six As species were separated by ion chromatography 9 1lC3738 An automated procedure was described for the 92182 921 I29 See Ag ref. 9112470 Samples were heated with HN03 and then with I-IN03-H202 in open tubes on a sand-bath at 1150 "C 50 mg specimen was solubilized with H2S04 and In a collaborative trial one method was used by all participants.Samples were heated at 150 "C with HN03 under pressure the pH adjusted to 2 and IAC~ solution added Results with routine AA and AE methods were closer to reference method values than those determined by colorimetric procedures 14202 Urine Biological fluids AA;ETA;L AE; 1CP;L 9 112930 9 113 1 7 1 As As MS;ICP;L 9 113206 As Urine Biological samples Liver MS;-;L AA;ETA;L As As As Urine As Liver As Biological samples MS;ICP;L AA;H y;L AA;Hy;L AA;ETA;L AE;ICP;L Au Plasma B Biological samples 9 112470 9 11105 1 91197 9111 108 AE;DCP;L AA;F;L B Tissues Ca Biological samples Ca Serum AA;F;L AE;F;L 9111280JOURNAL OF ANALYTICAL ATOMIC SPECTROMETRY APRIL 1992 VOL. 7 73R Table 1 CLINICAL AND BIOLOGICAL MATERIALS-continued Technique; atomization; analyte form* AE;F;L Element Matrix Ca Vaginal fluid serum Sample treatmentkomments Concentrations of Ca K and Mg in vaginal fluid were much higher than in serum but the Na was about 50Yo lower (in German) Chemical modification with 6 pg of Pd and 500 pg of (NHJ2N03 allowed determination of Cd in undiluted samples with aqueous standards.Zeeman-effect background correction and platform atomization were also used The concentration of metals in bird feathers correlated with habitat and length of exposure (in German) Samples were heated in a closed vessel microwave system. Atoms were formed and then heated for AE by an atmospheric pressure r.f. plasma established within a graphite furnace [furnace atomization plasma emission (FAPES)] Powdered samples were suspended by ultrasonic agitation and placed into a molybdenum atomizer with S as chemical modifier.Reproducibility was poor column to retain low molecular mass material. These compounds were then separated by elution onto and through a second column (DEAE anion exchange) with UV and ICP-AES detection 1,5-Bis(di-2-pyridylmethylene) thiocarbonohydrazine was used to extract Cd from a large aqueous volume into IBMK. Extraction was complete over a 3.4- 1 1.4 pH range Preparation of bovine blood based RMs was described Methadone was indirectly measured in urine after reduction on a Cd or Zn microcolumn 1 ml of 5Oh HN03 was added to 0.5 ml of blood. The supernatant was taken for analysis. Urine was diluted 1 + 1 with 5% HN03 (in Japanese) O2 gas included in the ash phase greatly reduced non- atomic absorption and signal depression A large number of specimens were collected from three cities.Higher concentrations were found in smokers (in Chinese) Homogenates of liver and kidney were centrifuged and the supernatant taken for HPLC to separate metallothioneins 1 and 2. Eluate was taken for measurement of protein Cd Cu and Zn. The concentrations of metallothioneins were calculated from these data Tissue samples were solubilized with a quaternary ammonium hydroxide Plasma and blood samples were deproteinized with HN03. Cd in the supernatants and in urine was concentrated by anion-exchange chromatography to improve the detection limit An excellent review of Cd toxicology and analytical procedures Cd Cu and Zn were measured in liver from patients with Alzheimer’s disease and from controls.Increased concentrations of Cd and Zn with reduced Zn binding to metallothionein were found in the patients’ samples coated with TiN to reduce contamination prior to analysis. The effectiveness of the coating was investigated by NAA Samples were passed through a gel permeation Tissue specimens were removed with instruments See Co ref. 911827 Reference 9 114046 Cd Urine milk blood AA;ETA;L 9 11846 Cd Cd Feathers Biological tissues AA;ETA;S 91/909,9111511 9111432 AE;r.f. p1asma;L Cd Cd Biological materials Biological tissue AA;ETA;SI 9111544 9 11C 1694 AE;ICP;L Cd Biological tissues AA;ETA;L AE;ICP;L 9 11c 1 753 Cd Cd Cd Blood Urine Blood urine AA;-;- AA;F;L AA;ETA;L 9112 170 9 112469 9112566 Cd Cd Blood Blood AA;ETA;L AA;ETA;L 9 11265 1 9112680 9 113 1 02 Cd Metallothioneins AA;F;L Cd Cd Lung tissue Blood plasma urine AA;-;L AA;ETA;L 9113378 9113509 Cd Cd Biological tissues Liver 9 113766 9212 1 5 c o Cr Skin Skin 91/827 9 1182774R JOURNAL OF ANALYTICAL ATOMIC SPECTROMETRY APRIL 1992 VOL.7 Table 1 CLINICAL AND BIOLOGICAL MATERIALS-continued Technique; atomization; analyte form* AA;F;L Element Matrix Cr Serum Reference 9111041 Sample treatmentlcomments Saniples were prepared in a solution of 0.2% Na citrate and 0.2% Na,SO,. The reported detection limit (0.0007 pg ml-I) was much lower than other workers have previously obtained with FAAS (in Chinese) Procedures for specimen collection processing and analysis were investigated to determine if tissue fixed in formalin would give reliable results Cattle faeces were dried and digested with HN03.Recovery was lower with forced-draught or microwave oven drying when compared with analysis of fresh specimens Cr was removed from water by algal cells HN03 and H2S04 were used to digest the urine sample and Cr in the residue was extracted with a secondary alkyl amine N-235 in C6H6 at a pH of 2.8. Background correction was unnecessary and the detection limit was 0.2 pg for 20 pl injected suppress interferences. The reported concentrations were higher than those found by other workers CrV[ was separated from Cr1I1 in aqueous solution. Red cells immobilized on Ca alginate beads were added to take up the Crn. The cells were separated from the beads and taken for measurement of Cr; O2 was used to facilitate the ash step Levels in normal lung tissues were determined.Large intra-organ variations were detected Saniples were analysed after dilution with H20 (in Chinese) Several chelating agents for the coprecipitation of Crul and Crw were evaluated. Manganese- diethyldithiocarbamate gave best recovery (in Japanese) Deuterium background correction removed interferences and samples from diabetic and healthy subjects were analysed Three materials were investigated as calibration agents for analysis of solid samples. NIST SRM Tomato Leaves and a synthetic material formed by coprecipitation of Cu with magnesium 8- hydroxyquinolinolate gave satisfactory results Cu in samples was concentrated by algae. The algal 'cells were collected by centrifugation and :suspended in 0.5% HN03 A chemical modifier with Ca and Mg was used to Specimens were taken without pre-treatment.See Cd ref. 911C1694 Cu was measured in proteins separated by agarose EXAFS analyses were used to examine Cu-binding An FI system was developed for indirect electrophoresis (in Chinese) :sites measurement of the fungicide diethyldithiocarbamate. The Cu-complex was extracted into IBMK See Cd ref. 9 113 102 Lyophilized samples were heated in closed vessels with H2S04-HN03 and isoamyl alcohol. The solution was filtered diluted and analysed assembly. Samples reacted with powdered CuC03 and the glycine-Cu complex eluted to an on-line ,4AS detector Glycine was measured using a continuous-flow AA;ETA;L 9 111079 9 111 369 Cr Lung tissue Cr Faeces AA;-;- Cr Algal cells Cr Urine A E; I CP; L AA;ETA;L 9112217 9112945 Cr Serum urine AA;ETA;L 9112958 AA;ETA;L 9113153 Cr Red cells water Cr Lung tissue Cr Blood Cr Urine AA;-;- AA;ETA;L AA;ETA;L 9 113234 911341 5 9113865 AA;ETA;L 9212 12 Cr Urine c u Biological tissues AA;ETA;S 9111 509 AA;ETA;Sl cu Biological samples 9111550 c u c u Biological tissues Serum proteins AE;ICP;L AA;ETA;L 9 1lCl694 9112375 Yeast cells 9112627 c u c u Diethyldithiocarbamate AA;F;L 9112715 Metallothionein Liver AA;F;L AA,F air-C2H2;L 9 113 102 91/3103 c u c u c u Amino acids AA;F;L 9113181JOURNAL OF ANALYTICAL ATOMIC SPECTROMETRY APRIL 1992 VOL.7 75R Table 1 CLINICAL AND BIOLOGICAL MATERIALS-continued Technique; atomization; analyte form* AA;F air-C,H,;L Element Matrix c u Urine Sample treatmentlcomments The pH of diluted urine samples was adjusted to 4.5.Copper 8-hydroxyquinolinolate in isoamyl alcohol was added and the samples mixed for 30 min. Copper oxalate was formed in the aqueous phase and the residual Cu in the organic layer was measured for the indirect determination of oxalate specimen. Samples were digested with H2S04 and HC104 Cu and Zn were evenly distributed within a stool See Cd ref. 9212 15 Cu" as the NTA complex was extracted at pH 8.5-10.0 into 5% Aliquat 336 in C6H,. The complex was stripped into 0.1 mol dm-3 HClO. and the Cu measured principal component analysis used successfully to eliminate these effects 57Fe and s8Fe were measured to determine non-haem Fe bioavailability in children s7Fe and 58Fe were measured to determine incorporation of Fe into red cells s7Fe and s8Fe were measured to determine absorption of Fe from baby foods Poor correlations between results were obtained when samples from patients receiving iron-Dextran were analysed by a Kodak Ektachem procedure by a centrifugal analyser method or by AAS See Cu ref.9 113 103 The specimens were diluted 1 5 with 1 mol dm- Polyatomic interferences were investigated and a H2S04. Biliary Fe excretion was similar in control subjects and patients with cirrhosis Serum specimens were deproteinized by precipitation with HCl and TCA. This procedure removed Fe which can be present if there is haemolysis of red cells into HCl. Atomization from a tantalum boat was found to be free from the memory effects associated with measurement of lanthanides interferences.Sensitivity was improved by inclusion of HNO Samples were digested and introduced into the ICP as the hydride. Concentrations of Ge in plant and animal tissues were 8-203 ppb Ni(N03)2 was used as the chemical modifier with 2 pg giving an absorption of 1% (in Chinese) Samples were oxidized in 02 the Hg concentrated by amalgam formation and then vaporized by rapid heating. The Drocedure was claimed to be faster than conventional methods (in Hungarian) In an interlaboratory comparison project a range of methods were used to determine MeHg in tissues. All methods gave similar results H2S04 was added within a closed headspace vial to release MeHg and iodoacetic acid was added to form the volatile iodide derivative.The MeHgI was measured by GC with MIP detection using standard additions for calibration Gd was extracted into IBMK and then re-extracted Pd(N03)* or Mg(N03)2 was used to reduce See Cd ref. 9 112 170 The analysis was calibrated by injection of known volumes of Hg-saturated air. Other aspects of the procedure were discussed generation was from a novel reaction flask Good recovery was reported for digested samples. H g Reference 9113216 c u Faeces 9 113 504 CU c u Liver Biological tissues AA;-;L AA,FL 9212 15 921259 Fe Biological fluids MS;ICP;L 9 11CI 662 9 1/23 14 Fe Fe Fe Fe Blood Red cells Red cells Serum MS;ICP;L MS;ICP;L MS;ICP;L AA;FL I2315 I2316 12942 13103 I32 1 5 Fe Fe Liver Bile fluid AA;F air-C2H2;L AA;ETA;L Fe Serum AA,F;L AA;ETA;L 921246 Gd Biological samples 9 I I3268 AA;ETA;L Ge Biological samples 9113228 AA;ETA,L Ge Biological tissue AE;ICP;L 9 1 I4045 Ge Herbal medicines Hg Urine 921 I54 9 111 249 AA;ETA,L AA;cold vapour;L HI3 Biological samples Hg Biological samples 346 800 AA;cold vapour;L MS;ICP;L 911 911C AE;MIP;G Hg Blood Hg Biological samples AA;-;- AA;cold vapour;L 9 112 170 9 1/25 1 3 Hg Hair AA ;cold vapo u r; L 9 11254876R JOURNAL OF ANALYTICAL ATOMIC SPECTROMETRY APRIL 1992 VOL.7 Table 1 CLINICAL AND BIOLOGICAL MATERIALS-continued Element Matrix Hg Biological tissues Hg Liver Hg Biological samples Hg Biological specimens Hg Urine K Eye lenses K Vaginal fluid serum Li Serum Li Serum urine Li Serum urine Li Brain Li Tissue homogenates Methadone Urine Mg Tubular fluid Mg Biological samples Mg Aortic plaque Mg Serum Mn Aortic plaque Mn Serum Technique; atomization; analyte form* AA;ETA;S AA;-;L AE;F;L AA;F;L AE;F;L AA; ETA; L Sample treatment/comments Reference 91/3 180 Inorganic Hg species were considered as ionizable (toxic) and stable (non-toxic) fractions.Total Hg was measured after digestion with H2S04 HN03 and V205 with SnCl reduction; total inorganic Hg was determined with an HzOz and KCN reducing agent; ionizable inorganic Hg was released from tissue homogenate by addition of H,SO and NaCl with SnCl reduction. The vaporized Hg was retained on a gold trap and released by heating the graphite furnace to 700 "C Hg vapour formed by reduction with SnCl was trapped with a bundle of gold wires. The Hg was released as a sharp pulse by heating and transferred to a flow cell positioned in a conventional spectrometer (in Japanese) closed PTFE vessels.An improved gas-liquid separator was developed A series of extraction stages was applied to samples containing MeHg. As the final step a pentane solution was evaporated and the residue redissolved in toluene for GC with an AA detector. Column eluate passed to a quartz tube at 900 "C to split the MeHg. Hg was trapped on gold-platinum mesh and the concentrated Hg liberated by heating for measurement by AAS. EtZo3HgCI was added to the original sample as an internal standard (in German) KBr03-KI and SnC1 in the presence of NH,OH-HCl formed BrCl which caused breakdown of organomercury compounds within 3 h. Total Hg was then determined with recoveries of 92-1 08% of added HgCl CH3HgCl HgKCN (C,H,),Hg (CH3),Hg and N02Fe6H4Hg (in Chinese) K and Na were extracted by soaking intact or ground lenses with H20.Further dilution with CsCl solution increased sensitivity and decreased variability of the measurement 9 1/3978 AA;cold vapour;L AF;cold vapour;L Samples were dissolved with microwave heating in AA;ETA;G AA,cold vapour;L 91/884 See Ca ref. 9 1/4046 9 1 I4046 See Ca ref. 911'1280 91/1280 AA;ETA;L AA;ETA;L AA;F;L AA; ETA; L AA;F;L AA;-;- AA;F;L LEI;FL AA;-;- 92/49 92/90 921121 92/2 1 8 921274 A previously described procedure was used to 9112656 determine Li clearance as an index of renal distal delivery of Na and H20 Serum proteins were precipitated with 10% HN03 (1 + l) urine was diluted 1 +4 with 5% HN03. The graphite furnace was coated with Ta to reduce carbide formation and background interference was removed by inclusion of a small flow of Ar during atomization.Normal concentrations were determined Li was distributed heterogeneously within eight regions of brain Results given by two techniques were compared and indicated that SIMS can be used for measurements at 0.1 ppm (in French) 91/3199 Se Gd ref. 91/2469 A detection limit of 0.02 mmol dm-3 was achieved See Ca ref. 91/1108 M,g Mn and Zn concentrations were higher in plaque See Ca ref. 91/1280 See Mg ref. 91/1144 Alkali and alkaline earth elements were removed material compared with healthy aortic tissue from the specimens with an automated chelation (Chelex 100) chromatographic system under iobotic control. Eluant was aspirated into the flame for LEI analysis 9 112469 9111059 9111 108 91/1144 91/1280 91/1144 9111425JOURNAL OF ANALYTICAL ATOMIC SPECTROMETRY APRIL 1992 VOL.7 77R Table 1 CLINICAL AND BIOLOGICAL MATERIALS-continued Technique; atomization; analyte form* AA;ETA;L Element Matrix Mn Urine Sample treatment/comments Developments with a new design for a furnace and platform were described. The platform can be reproducibly positioned and securely fixed Mn was extracted from salicylate solution by Aliquat 336 in xylene See Li ref. 921274 Reference 9 1lC3627 Mn Pharmaceuticals AA;-;L 921 103 921274 9 1l884 9 1 f 4046 9 11 1079 91/1425 91lC1662 9 1/C 1 796 9 113234 9 113464 921 106 Mn Tissue homogenate MS;-;- AE;-;- AA;-;L AE;FL AA;ETA;L LE1;F;L MS;ICP;L AE;ICP;L AA;ETA;L AA;-;- Na Na Ni Ni Ni Ni Ni Ni Eye lenses Vaginal fluid serum Lung tissue Serum Biological fluids Biological tissues Lung tissue Urine See K ref.911884 See Ca ref. 9114046 See Cr ref. 9111079 See Mn ref. 91/1425 See Fe ref. 91lC1662 See Ni ref. 921106 See Cr ref. 9113234 Samples acidified with HCl were analysed (in Chinese) The pH of digested specimens was adjusted to 2-5 for the formation of a complex with 1,5-bis(di-2- pyridylmethy1ene)thiocarbonohydrazide. The complex was extracted into IBMK and a 15:l aqueous to organic ratio could be tolerated See Cu ref. 91/3216 The distribution of Pb in sections of tissue was determined by a microPIXE system. Concentrations were in good agreement with those obtained by AAS See Cd ref. 911909 Specimens were diluted with 4% NH4HzPO4-6% ascorbic acid (in Chinese) Pb was measured in vivo in finger tibia and calcaneus bones using two different radiation sources With samples that had normal Pb concentrations there was no correlation between blood and hair levels Improved instrumental design reduced the detection limit for in vivo measurement to 10 pg g-I See Cd ref.9111432 Samples were taken into solution and the Pb methylated by addition of sodium tetraethylborate. The tetraethyllead was purged from solution and trapped into a hot (400 "C) graphite furnace. Atomization was accomplished by further heating Triton X-100 and atomized from a graphite probe under isothermal conditions. Calibration was possible with aqueous standards and accurate results were achieved Low recoveries were reported when calibration standards were prepared in bovine blood compared with human blood when NH4H2N03 was used as chemical modifier.Improved recoveries were obtained with NH4H2P04 but ash residues then accumulated within the graphite tube Samples were diluted 10-fold with 0.1% Triton X-100 and injected onto a graphite platform. An equal volume of chemical modifier (0.2% Pd-2% citric acid-0.01 rnol dm-3 HN03) was added. The pyrolysis step included the addition of O2 Pb was coprecipitated with Pd by addition of 15% ascorbic acid. The procedure was successful over a wide range of acidity Blood samples were diluted 10-fold with 0.05% See Cd ref. 91/21 70 Ni Biological samples AA;ETA;L Oxalate Urine Pb Cerebellum AA;F air-C,H,;L PIXE;-;- AA;-;L 9 1/3216 9 11883 Pb Feathers Pb Urine AA;ETA;S AA;ETA;L 91/909,9111511 9111 149 Pb Bone XRF-;S 91/1189 Pb Blood.hair 91/1259 AA; ETA;L Pb Bone XRF;-;S 9111385 9 1/1432 9111546 Pb Biological tissues Pb Biological tissues AE;r.f. p1asma;L AA;ETA;L Pb Blood AA;ETA;L 91lC1793 Pb Blood AA;ETA;L 9 1lCl830 Pb Blood AA;ETA;L 9 1/C 194 1 Pb Biological samples Pb Blood AA;ETA;L AA;-;- 9 1 /C1943 9 112 17078R JOURNAL OF ANALYTICAL ATOMIC SPECTROMETRY APRIL 1992 VOL. 7 Table 1 CLINICAL AND BIOLOGICAL MATERIALS-continued Technique; atomization; analyte form* AA;-;L Element Matrix Pb Urine Reference 9112255 Sample treatmentlcomments The sample was added to a separative column atomizer (particles of graphite glassy carbon and activated charcoal or alumina). Pb was atomized by heating the tube at 1500 "C.Non-atomic absorption peaks were temporally separated from the Pb signal Urine samples were digested with HNO and HC104 and Pb extracted as the APDC chelate into IBMK. For blood see Cd ref. 9112566 (in Japanese) Blood concentrations were compared with Pb in dentine and with relevent environmental factors Sea Cd ref. 9 11265 1 See Cd ref. 9112680 Separation of Pb-containing protein species was achieved by size-exclusion chromatography coupled with ICP-MS The use of a graphite probe was described The specimens were digested with HN03-HC104 and Pb Blood urine AA,ETA;L 9112566 Pb Blood teeth AA;ETA,L 9 112593 Pb Blood Pb Blood Pb Serum red cells AA;ETA,L AA;ETA;L MS;ICP;L 9112651 9112680 91lC2828 AA ;ETA; L AA;ETA;L Pb Milk Pb Serum 9113 176 9113536 residues were dissolved in I% HN03 for analysis tissue with HNO,.Samples were further heated with HClO. the residues dissolved in H20 and a reducing agent added. This solution was taken to an FI system which involved concentration of Pb by on-line coprecipitation with iron-(11)- hexamethylene ammonium hexamethylene- dithiocarbamate. The precipitate was dissolved by a stream of IBMK which took the Pb to a very lean flame. The detection limit was 2 pg 1-1 at a sampling frequency of 90 h-' microwave system was used to digest blood or Pb Blood liver AA;F air-C2H2;L A 9113779 AA; ETA; L 9113913 Pb Tooth enamel Surface enamel was removed by acid etching. Factors that contribute to the Pb concentration; residential area age tooth type and position were investigated Blood was diluted 1+4 with a solution containing 2.5 ml 1-1 of Triton X-100 and 5 ml 1-1 of Antifoam B.A 1.6 mol dm- HNO solution was added to give a 10-fold dilution of the original sample. Diluted samples were centrifuged and 10 pl of the supernatant placed into the furnace with 5 pl of NH4H2P04 and 10 pl0.16 mol dm-3 HNO,. Calibration was by standard additions. This procedure was developed for accurate measurement of low concentrations of Pb in blood H20. A small aliquot 10 pl was placed on the tip of a carbon probe which formed the cathode for GDMS. Pb mobilization was studied in patients receiving Pt chemotherapy. A Bi internal standard was used Urine was heated with FINO and redissolved in Plb in bone was measured with an in vivo technique A 200 pg sample was placed onto a laboratory made boat. NH,H2P04 was added to remove matrix interferences and the detection limit was 0.007 ng (in Chinese) Erposure to Pb was monitored by measurement of hair Pb by SR-induced XRF.Blood Pb was determined by micro-PIXE Serum was diluted with a chemical modifier containing 1 % Triton X- 100 HNO and Cr(N03)3. The detection limit was 0.22 pg 1-1 and the normal level was 1.3 pg 1-1 (NH4)2EDTA-NH4H2P04-NH40H-octoxynol chemical modifier A 1 + 19 dilution was made with an Pb Blood AA;ETA;L 921 14 Pb Urine MS;GD;L 92lC40 Pb Pb Bone Tissues XRF-;S AA,ETA,S 92172 9211 53 Pb Pb Hair blood Serum PIXEi-;- XRF-;- 921209 921279 AA;ETA;L Pt Plasma urine AA;ETA;L 9111 138JOURNAL OF ANALYTICAL ATOMIC SPECTROMETRY APRIL 1992 VOL. 7 79R Table 1 CLINICAL AND BIOLOGICAL MATERIALS-continued Technique; atomization; analyte form* MS;ICP;L AA ETA;L Element Matrix Pt Biological tissues Sample treatment/comments Tissues red cells and plasma were heated with HN03 at 100 "C until dry.The residue was dissolved in 1% HCI with In as internal standard and aspirated into the ICP Samples were digested with HN03 and diluted to minimize salt effects See Pb ref. 92/C40 Mechanisms of action of chemical modifiers were investigated by surface analysis of the graphite tube. Cu was superior to Ni (in Chinese) Hydride generation was used for sample introduction in a study which demonstrated that Se in bulk tank milk was proportional to the mean blood concentration in a herd of cattle The serum was diluted with Ioh Triton X-100 and Pt added (in Chinese) An interference associated with phosphate was detected when deuterium or Zeeman-effect background correction sytems were used.A Pd-Mg(N03)2 modifier was added with detection at 196.0 nm heated to remove the acid. The solution was diluted 1 + I with 0. I% Triton X-100 and analysed with Zeeman-effect background correction. Aqueous standards and standard additions were used for calibration Species of Se were separately determined. Sew was extracted with APDC into IBMK Se" had to be reduced using TiC13 and then likewise extracted. Ni and Mo were shown to be suitable chemical modifiers (in Chinese) Se concentrations were higher in thyroid than in liver After digestion samples were introduced via an FI system to the hydride generator and heated quartz cell and sperm count or motility HCL.The lamp discharge enabled AE from the sample. The instrumental parameters required for optimum performance were determined The nitrate salts of many elements were investigated as chemical modifiers for the measurement of Se. A mixture of Mg and Pd was said to be the most convenient modifier as chemical modifier. Good recovery and precision were obtained (in Chinese) Patients with liver disease and who had low concentrations of Se in blood were given Se-rich yeast. Concentrations increased but remained below the reference range techniques for certification and to select potential reference met hods A novel interface based on thermochemical hydride generation was constructed to link methanolic eluate from an HPLC column to an AA spectrometer.Seleniocholine and (CH3)3Se+ were among the Se species measured in urine Tissues were digested with HN03 H2S04 and HCIO. by microwave heating. Pd was added and the solution extracted with DDDC in CHC13. The organic phase was taken for analysis in a Ta- coated graphite furnace. The reported detection limit was 0.002 pg g-l and accurate results were achieved with RMs Tissues were freeze dried solubilized with HN03 and There was no correlation between Se concentrations Samples were dried onto an A1 cathode cup of an Digested samples were analysed with Pd-Mg(N03)2 A series of RMs were analysed by different Reference 9111428 9 1 XI66 I 92lC40 91/1018 9 I l l 256 91/1271 9 1 /C 1 78 9 9 l/C 195 1 Pt Tissue specimens MS;ICP;L MS;-;L AA;ETA;L Pt Urine Se Hair Se Blood milk AE;ICP;L Se Serum Se Serum AA,ETA;L AA;ETA;L Se Heart tissue AA;ETA,L Se Serum hair AA;ETA;L 9112520 Se Thyroid gland liver Se Serum breast milk AA;Hy;L AA;Hy ;L 9 9 I252 1 12563 /2617 12718 Se Semen Se Serum 9 9 Se Biological fluids AA;ETA,L 91lC2736 Se Tissues Se Serum AA;ETA;L AA;ETA;L 9113015 9 1/32 17 Se Biological samples Se Urine .. -- 9 9 AA;Hy;L 9113229 9113591 9 113597 Se Biological tissues AA; ETA;L80R JOURNAL OF ANALYTICAL ATOMIC SPECTROMETRY APRIL 1992 VOL. 7 Table 1 CLINICAL AND BIOLOGICAL MATERIALS-continued Element Matrix Se Biological tissues Se Liver Se Blood Si Serum urine Si Plasma urine Sn Biological samples Sn Biological samples Sn Biological samples Sn Tissues Sn Blood Sr Hair Sr Serum Te Urine Th Biological tissues Technique; atomization; analyte form* Sample treatmentlcomments AA;ETA;L Different furnace types and the addition of Ni were investigated for the elimination of the interference from Po4,- using deuterium background correction. Successful measurements were achieved with an uncoated tube L'vov platform and Ni at 50-100 pg AA;Hy;L AA;ETA,L AE;DCP;L AA;ETA;L AA;Hy;L AA;Hy;L AE;ICP;L AA; H y ; L AA;ETA;L AA;F;L MS;ICP;L AA;ETA;L MS;ICP; L A dried powdered sample was ignited in an O2 combustion bomb containing H20.The dissolved Se042- was reduced by heating with HCl and taken for hydride generation with NaBH4 (in Japanese) Samples were diluted 5-fold with 0.25% Triton X- 100. Diluted specimens were mixed with an equal volume of 10% HCl and two volumes of chemical modifier [300 mg of Pd in 1 ml of HN0,-HC1 (3:l) and 200 mg of Mg(N03)2 per 200 ml at pH 71.Deuterium background correction was used No interferences were found with samples diluted in 10 ml 1-l HNO,. Concentrations were increased in subjects with chronic renal failure and were normal in patients with renal stones Silicone tubing and furnace fittings were replaced to reduce possible contamination. A chemical modifier (K2EDTA-K.H2PO4-Ca2+-NaC1- C2H50H) at pH 6.0-6.5 was used to dilute samples. The graphite platform and tube were coated with Mo. Matrix interferences were not evident specimens sample with 2 mol dm- HCl separated by GC and measured with a coupled AAS system Digested samples were reacted with NaBH and TCA for formation of SnH,. A novel gas-liquid separator was used to remove liquid and excess of CIH,C02H was used to leach butyltin from the Organotin compounds were extracted from the H2 The gaseous hydride was trapped in a warm graphite furnace and the Sn atomized by electrothermal heating.A detection limit of 7 ng 1-I was reported with 10 ml of sample ml of H20 2.5 ml of HN03 and 0.5 ml of HCl. The dry residue was dissolved in 3 ml of 6 mol dmP3 HC1 5 ml of 20% ascorbic acid and 5 ml of HCI were added. This solution was extracted with IBMK and the Sn re-extracted into 2 ml of 0.5 rnol dm- HCl-100 ppb Pb. Ascorbic acid 0.1 ml of 20% was added and the sample analysed. Good recovery and a 2.2 ppb detection limit were reported (in Japanese) Hair was digested with HNO,-H2O2 and EDTA (0.1 rnol dm-3) added to remove interferences from Ca and Mg (in Chinese) The sample was diluted 5- or 10-fold with a solution of 0.14 mol dm-3 HNO containing In as internal standard 50 ml of urine were digested with HN03-HC10 with evaporation to dryness.Tew was reduced to Tew with HCl dissolved in 7 rnol dm- HCl and extracted twice with 5 ml of IBMK. Repeated sampling-drying cycles were employed to introduce 200 p1 of IBMK extract into the furnace. Sensitivity was increased with PdCI2. Reference levels were (0.1-1.2 pg 1 - I TI or Bi were used for internal standardization A 1 ml sample of blood was heated at 100 "C with 1 Reference 9113775 9 114019 92/21 1 9112631 9112713 9 11994 9111362 91/3150 91lC3630 91/4038 9111527 9 1122 19 9113350 91/2251JOURNAL OF ANALYTICAL ATOMIC SPECTROMETRY APRIL 1992 VOL. 7 81R Table 1 CLINICAL AND BIOLOGICAL MATERIALS-continued Technique; atomization; analyte form* MS;ICP;L AA;ETA:L AA;F;L Element Matrix Th Bone TI Blood urine serum Sample treatmentlcomments Reference 9113293 9112941 Bone samples were heated with HNO and HF Samples collected during investigation of a case of severe toxicity were analysed.The T1 was extracted as the NaDDC chelate into IBMK enhanced by inclusion of CH30H 9 llC3678 The sample was heated in a graphite furnace. AF was See Th ref. 911225 1 9112251 See Th ref. 9113293 9113293 9 llC1950 Mean concentrations of V were 14.5 and 4.7 pg I-' in samples from patients with chronic renal failure and healthy controls respectively. Specimens were diluted 2-fold in a chemical modifier (Pd 100 pg citric acid 20 g and Triton X-100 1 g per 1 of 0.01 mol dm- HNO,) Urine was diluted with 2% HNO and Triton X-100.The detection limit was 1.5 pg 1 - I . Aqueous calibration was satisfactory See Co ref. 911827 See Mg ref. 9111 144 Samples were diluted 1 + 1 with water and standard solutions were in 5% glycerol (in Chinese) A collaborative study was employed to evaluate a proposed official method. The specimen was diluted with 0.03% Brij 35 and deuterium background correction was used See Cd ref. 91/2469 Zn associated with a2 macroglobulin was separated from loosely-bound Zn and albumin-bound Zn by ultrafiltration. The loosely-bound Zn is claimed to be the dynamic physiologically active fraction See Cd ref. 9 113 102 Test compounds were dissolved in HCI-C2H50H-2- See Cu ref.91f3504 See Cd ref.9212 15 Dried samples were mixed with HN03 and heated in an autoclave which contained HN0,-HCI in a second compartment (Cr Cu Mn Zn) (in Russian) commercial instrument was investigated by analysis of RMs. Results were within 30% of the certified values Powdered samples (RMs) were heated at 350 "C with H2S04 to produce carbonaceous slurries. Good precision and accuracy were obtained with this simple preparative procedure (Ca Cu Fe K Mg Mn Na Zn) contamination and loss to prepare reference values for trace element concentrations in human milk Reference values for trace element concentrations were established with specimens from 1000 children The composition of 11 remedies was measured A review of trace element determination by AAS combined with FI analysis Samples were digested with acids and the effects of residual matrix acid and oxidation states on the measurements were determined (As Cd Cr Cu Hg Mn Se Zn) (in Chinese) heated at 120 "C.Digestion was continued with addition of 30% H202-HN03 (3+ 1) and heating until solutions were clear. Samples were further diluted 10-fold with 0.1 mol analysis (B Ca Mg Na P) butanone and aspirated into the flame The semiquantitative facility available on a The study investigated causes of possible HNO was added to samples in PTFE tubes and HC1 for T1 Liver AF;laser;S U Biological tissues U Bone V Plasma serum MS;ICP;L MS;ICP;L AA;ETA;L V Urine AA;ETA;L 9 1/3 101 W Skin Zn Aortic plaque Zn Plasma 911827 9111 144 9111336 . . -9-,- AA;-;- AA;F;L Zn Serum AA;F air-C2H2;L 9 112410 Zn Urine Zn Serum proteins AA;FL AA;ETA;L 9 112469 9 112629 Zn Metallothionein Zn Pharmaceuticals AA;F;L AA;F air-C2H2;L 9 1 I3 102 9113392 Zn Faeces Zn Liver Various (4) Medicinal plants AA;-;L AA;-;L AE;ICP;L 9113504 9212 1 5 9 1 I209 Various (23) Biological samples Various (8) Biological samples MS;ICP;L AE;ICP;SI 9 11857 9 11868 Various (1 2) Breast milk AE;ICP;L 9 11879 Various (1 8) Hair AE;ICP;L 9 11887 Various (2 1) Herbal remedies Various Clinical samples MS; 1CP;L AA;-;- 9 11933 911955 Various (8) Biological samples AE;ICP;L AA;Hy;L 9111022 Various (5) Biological materials AE;ICP;L 91/105 182R JOURNAL OF ANALYTICAL ATOMIC SPECTROMETRY APRIL 1992 VOL.7 Table 1 CLINICAL AND BIOLOGICAL MATERIALS-cont inued Technique; atomization; analyte form* XRF-iS Element Matrix Various Tissues Reference 91/1054 Sample treatmentlcomments Art EDXRF system for mapping the elemental distribution in tissue specimens with a resolution of 200 pm was described (in Japanese) In-house internal quality control procedures necessary to maintain reliability of results were discussed concentrations increased C1 decreased and Ca and P were unchanged with age (in Polish) Concentrations were determined in excavated bones from various prehistoric eras and compared with values found in present day samples (Ca Cd Cu Fe Mn Ni Pb) examining correlations between concentrations of elements in these tissues From 2 18 samples it was seen that K S and Zn PIXE analysis was shown to be suitable for See Various 91/868 (in Hungarian) The distributions of elements in 50 areas of brain tissue were determined (Ca Cu Fe K Mn Rb Se Zn) Dined powdered medicines were digested with HN03 and HC104 and the residues dissolved in 6% HCl (in Chinese) RM. A large number of elements were determined and small but acceptable variations were noted for Al Br C1 Coy Fe 1 and Na adult rib bones enamel.Ca interference in the determination of Mn and Pb was overcome by use of standard additions (Cay Cd Mn Pb) (in Japanese) elements were eliminated by removal of the elements by an automated chelation chromatography procedure XRF spectrometry with reference to instrumentation and analytical applications Analyses with a toroidal Ar MIP and a cylindrical He MIP were investigated using tissues digested with acid A recycling nebulizer system was used with 1 ml of sample analysed for 20 min (in Chinese) Hair was dissolved and the solution heated in HC104-HN03. Residues were dissolved in HCl with Ga (internal standard) and diluted with H20.A wide range of recoveries was obtained (Cd Co Cu Fe Mn Mo Ni Pb Zn) (in Chinese) Tissues were heated with HN03 in closed PTFE vessels at 120 "C for 3 h. The residual solution was diluted with HzO a portion reduced almost to dryness and further diluted with H20 Solutions prepared from digested tissues were introduced by discrete nebulization into an instrument with a small spray chamber and a time sharing background correction system. Results were comparable to those obtained by continuous aspiration (Al Cu Fe Mn Zn) Specimens were heated with HN03 in conical flasks covered by watch glasses.No loss of volatile elements was observed with heating at 70 "C for 24 h (Cd Cr Cu Hg Mn Pb) Between-bottle variations were determined for an Concentrations were determined in foetal infant and These elements were measured in dentine and Interferences from alkaline and alkaline earth A review of recent developments in inorganic MS and AA;-;- 9 1/1087 Various Blood plasma Various (6) Teeth XRF-;- 91/1136 91/1143 Various (7) Bone PIXE;-;- 91/1163 Various Hair liver kidney Various Biological samples Various (8) Brain AA;ETA;Sl PIXE;-;- 91/1174 91/1187 Various (30) Traditional medicines AE;ICP;L 91/1227 Various Hepatopancreas AA,ETA;L 9 1/1247 Various Bone Various (4) Teeth AA ; ETA L AA;-;- AE;ICP;L 9 1/1329 91/1376 Various Biological samples LEI;-;- 91/1425 Various Biological materials MS;ICP;- XRF-;- 91/1436 Various Biological tissues AEMIPL 91/1457 Various (1 0) Saliva Various (9) Hair AEICPL AE1CP;L 91/1481 9 1/ 1482 Various Liver kidney Various ( 5 ) Biological samples MS;ICP;L AE;ICP;L 9111485 9111487 Various (6) Biological tissues AA,ETA;L AA;cold vapour;L 9 1/ 1493JOURNAL OF ANALYTICAL ATOMIC SPECTROMETRY APRIL 1992 VOL.7 83R Table 1 CLINICAL AND BIOLOGICAL MATERIALS-continued Element Matrix Various (1 6) Serum Various (5) Urinary calculi Various (7) Serum Various (1 1) Blood serum Various (4) Hair Various Biological samples Various (9) Biological tissues Various (1 2) Biological tissues Various Serum CSF Various (4) Hair Various (5) Biological samples Various (5) Hair Various Biological samples Various Biological samples Various (4) Biological fluids Various (8) Oral mucosa Various (46) Urine blood serum Various (6) Hair Various (5) Traditional medicines Various ( 13) Hair Various (7) Blood serum urine AA;ETA;S MS;ICP;L AE;d.c.arc$ AE;ICP;L MS;ICP;L MS;ICP;L MS;ICP;L MS;ICP;L AA;-;- AA;Hy;L Technique atomization; analyte form* Sample treatmentlcomments XRF-;S Specimens were heated with O2 in a low temperature asher dissolved in 2 mol dm-3 HN03 and the pH adjusted to 3.8. This solution was added to a Hyphan cellulose column and the analyte elements eluted with 6 mol dm-3 HCI. The eluant was evaporated on a quartz probe and covered with hexamethyldisilazane film for TXRF (in German) Results of direct analyses compared well with those obtained after digestion (Cd Cr Hg Ni Pb) Internal standard (In) was added to samples with subsequent dilution in 0.14 mol drn-3 HN03 (Co Cs Cu Fe Mo Rb Zn) Digested samples were dried under an IR lamp restored to pH 6 and elements sorbed onto a polyarsenate sorbent.The mixture was heated and the ash mixed with graphite powder for excitation from the graphite electrodes graphite cup and volatilized into the ICP (Cd Cr Ni Pb) (in Japanese) Applications of ICP-MS in clinical biochemistry were described Tissues were digested with HN03 in a microwave system (As Cd Co Cu Fe Mg Mn Pb Zn) CRMs were analysed and good accuracy was noted Open digestion microwave digestion and sample dilution were evaluated Structure of hair was examined by SEM and changes were associated with increased levels of these metals measured by AAS (Cd Pb Sb T1) Digestion procedures were investigated.Interferences from Cu and Ni were eliminated by a mixed reductant (3% NaBH + 2% KI) and accuracy was established by measurement of RMs (As Bi Sb Se Sn) described (Cd Hg Pb Se Zn) Sample solutions were placed in a large volume Preparation and certification of a hair RM was A review of LMMS analysis in biomedical research Atomic spectroscopic procedures were described with reference to their potential for direct analysis of solid samples lithium bis(trifluoroethy1)dithiocarbamate chelate was used to allow separation and detection of low concentrations of different isotopes by GC-MS (Cr Cu Ni Zn) Biopsy specimens were freeze dried embedded and sectioned for LMMS analysis to show the composition of amalgam tattoos (Ca Cr Fe Hg Mo P S Se) reference values for a wide range of elements in biological fluids from Italian subjects Reference values for Greek subjects were determined in hair samples which were cleaned with detergent and digested with acid (Cd Cr Cu Ni Pb Zn) Dried powdered compounds were heated in a crucible with HN03 + HClO charred suspended in 5% HN03 and filtered (Al B Ba Cu I) (in Chinese) EDXRF was applied to analysis of hair samples.Large individual variations associated with age sex race hair colour and treatment and environment were observed Methods for use in industrial medicine and toxicology were given (Al Cd Co Cr Cu Pb Se) Thermostable volatile chelates were prepared.The A valuable and comprehensive determination of MS;GC;L LMMS;-;- AA; ETA,L AE;ICP;L AA;ETA;L AE;ICP;L XRF;-;- AA,ETA;L Reference 9111 505 9111513 9111545 911 911 602 605 91lC1657 91lC1658 9 11C 1659 9 1lC1679 9 1lCl7 13 91lC1790 9 11C 184 1 9112155 9 112232 9112256 9112382 9 1 I2405 9 1 I2406 9 1 12485 9 112527 911257184R JOURNAL OF ANALYTICAL ATOMIC SPECTROMETRY APRIL 1992 VOL. 7 Table 1 CLINICAL AND BIOLOGICAL MATERIALS-continued Technique; atomization; analyte form* XRF;-;- Element Matrix Various Biological tissues Sample treatment/comments Mcthods for correlation of data were applied to results of analyses from tissues and hair. It was found that concentrations of toxic elements in hair are poor predictors for tissue levels EDXRF was used for a screening analysis of cell cultures (Cd Cu Fe Ni Se Zn) The technique was applied to a range of applications (in Japanese) Samples were collected after accidental death.No differences associated with gender were found. Concentrations of non-essential elements increased with age (Cd Cr Cu Mg Mn) the different elements were noted (Ca Cd Cu Fe Mg Mn Se Zn) (in Chinese) graphite platform is sited at the focus point of three IR lamps. The lamps cause rapid combustion and the smoke which contains the volatile elements is transported to a heated T- tube for atomization and AA measurements (Bi Cd Cu Hg Pb Ti Zn) A I-Iildebrand grid nebulizer was used for the analysis of samples with high solute contents and slurries A !/-groove nebulizer was used for the high dissolved solids content.Samples were analysed to check for conformity with Pharmacopoeia values Wet and dry ashing procedures were employed for the analysis of drugs in an attempt to define a ‘fingerprint’ for their geographical origins Specimens were lyophilized digested with HN03 and the solutions taken for analysis. Extensive precautions were taken to avoid contamination during the collection and preparative steps An In internal standard was added and the sample was diluted with 0.14 mol dm-3 HN03. Polyatomic interferences were corrected for by analysis of a suitable blank solution (Co Cs Cu Fe Mo Rb Zn) transferred to a graphite furnace for vaporization into the ICP. Calibration was by additions of solutions containing NH4H2P04 and the analyte standards within the furnace TXRF.Results were in agreement with certified concentrations Sarnples were analysed without special treatment apart from dilution with Triton X- 100. Chemical modifiers were included if necessary and calibration was by standard additions (Cd Cu Pb Se Zn) Reference values for children living in Italian urban areas were presented Powdered sample was vaporized from a graphite furnace into a magnetron rotating DCP A study of potential contamination of blood specimens from collection and storage devices showed that plastic containers were most suitable The composition of senile plaques and neurofibrillary tangles was studied to investigate mechanisms of formation. It was concluded that aluminosilicates are not required for plaque genesis (Al Ca Fe Si) Samples from subjects with occupational exposure to HF were analysed by EDXRF Fe was the major impurity in the pharmaceuticals examined (Cr Cu Fe Mn Ni Pb Zn) (in Russian) Various correlations between the concentrations of In this novel device a solid sample placed on a Powdered samples were digested and the solutions RM samples were digested prior to analysis by Reference 9 11’2609 12628 12672 12686 9112694 9 1/27 10 9 1lC2758 9 1/C2876 91/C2878 9 l/C2879 XRF;-;- Various (6) Leukaemic cells Various Biological samples Various (5) Liver MS;ICP;L AA,FL AA;ETA;L Various (8) Seminal fluid AA;-;L AA;F air-C,H2;S Various (7) Biological tisues Various Urine Various (1 1) Penicillin AE;ICP;L SI AE;ICP;L Various (1 1 ) Addictive drugs AE;ICP;L Various (12) Breast milk AE1CP;L Various (7) Serum MS;ICP;L 9 112965 Various Biological tissues AE;ICP;L 9113105 Various ( 1 1) Blood serum Various (5) Serum blood XRF;-;S AA;ETA;L 9013 1 5 I 9 1/31 54 Various (10) Hair Various Biological samples Various (29) Blood AE;ICP;L AE;DCP;S MS;ICP;L 91/3168 9 113266 9113338 XRF;-;S Various (4) Brain inclusions 1346 1 13467 f3473 Various (1 1) Hair Various (7) PharmaceuticalsJOURNAL OF ANALYTICAL ATOMIC SPECTROMETRY APRIL 1992 VOL.7 85R Table 1 CLINICAL AND BIOLOGICAL MATERIALS-continued Element Matrix Various (6) Gallstones Various (4) Teeth Various (5) Plasma urine Various Clinical and biological Various Urine samples foods Various Biological samples Various Biological samples Various Biological materials Various (4) Liver Various (5) Various (6) Various (6) Various (6) Various (8) Various Various (4) Various ( 5 ) Biological samples Biological samples Breast milk fractions Urinary calculi Gallstones Biological tissues Brain Liver tissue Various Teeth bone Various ( 16) Brain Various Chinese medicines Various (50) Lung Technique; atomization; analyte form* XRF-;- XRF;-;S MS;ICP;L MS;ICP;L AE;ICP;L .. -- Y AA;F;L AA; ETA; L AE;ICP;S PIXE;-;- AE;ICP;L AA;ETA;L AE;ICP;L XRF-;- PIXE;-;- XRF-;S XRF;-;S PIXE;-;- AE;ICP;L PIXE;-;- AE;ICP;L Sample treatmentlcomments The stone was powdered and made into a pellet (Ca The distribution of elements in teeth of Zn-deficient Specimens were diluted 1 + 5 in 1% vlv HN03 Cu Fe Mn Pb Zn) rats was investigated (Cu Fe Mn Zn) containing Eu internal standard. Concentrations of rare earth elements were all less than 0.3 pg 1-I (Ce Gd La Tb Yb) The 1990-1991 ASU An ion chromatography system with iminodiacetate chelating resin was coupled to the nebulizer. Samples at pH 5.5 were applied to the resin and analytes were eluted with 1.25 mol dm-3 HNO,.The system preconcentrates analytes and also separates Group I and I1 metals and some anions Plans for a robotic system to automate every stage of an analysis were presented and progress with construction was described Parameters associated with a new ultrasonic nebulizer were examined PTFE digestion vessels that provide for vapour-phase digestion with HN03-HF with low blanks and no acid accumulation in the residues were developed and evaluated. Digestion with microwave heating was complete within 45 min Liver samples were digested with HNO and HC104 complexed with APDC and extracted into CHCI,.Cu was measured by FAAS other elements by ETAAS (Co Cu Mo Se) Powdered samples were vaporized from a cup placed within a graphite furnace into the ICP (Al Cu Mn Pb Zn) Sampling technique was investigated to assess how representative subsamples from specimens could be obtained (Br Ca C1 Cu K Mn) Changes in concentrations of Ca Cu Fe P S and Zn were measured as a function of time for up to 196 days after birth of the baby Variations in the trace element content of urinary calculi were demonstrated (Cd Cr Cu Ni Pb Zn) (in German) The element contents of different types of gallstones were determined to investigate the mechanism of formation (in Japanese) EDXRF and PIXE techniques were used to examine normal and cancer tissues Ayurvedic drugs and blood samples Levels of Cu Hg Se and Zn were imaged in brain regions.The resolution was poor but there was little tissue destruction separately analysed in specimens of normal and cirrhotic liver. Some differences in concentrations of Br Cu Fe K and Zn were seen but it was concluded that for practical purposes it was not usually necessary to separate the fractions Ca P and trace elements were measured in calcified tissues 12 areas of brain were separately analysed by NAA and AES. Results compared well with those obtained by other methods to their essential trace element contents analysis was established to determine the reference values for samples from urban dwellers The cellular and connective tissue fractions were An attempt was made to relate activity of medicines A careful protocol for sampling pre-treatment and Reference 9113525 9113564 9113585 9113594 91lC3663 9 llC3672 91lC3705 9 113776 9113777 9113804 9113829 9 113968 9 1 I3975 9 1 I4006 9 1 I4008 9114031 9 1 I4040 92/66 92176 92177 9217886R JOURNAL OF ANALYTICAL ATOMIC SPECTROMETRY APRIL 1992 VOL.7 Table 1 CLINICAL AND BIOLOGICAL MATERIALS-continued Technique; atomization; analyte form* AA;-$ Element Matrix Various (4) Hair Sample treatment/comments Reference 92/95 Various (4) Sperm AA;-;- Various (24) Biological samples AE;ICPL AA;ETA;L Various Kidney Various (4) Blood XRF;-;S XRF;-;- Various (7) Mussels Various (4) Teeth AA;ETA;L 92196 921 147 921 193 921208 921255 921260 Inconsistent results were obtained with different washing methods.The influences of age pollution cosmetics and other factors on results were discussed (Cd Cu Mg Pb) (in German) The concentrations of Cd Pb and Zn in sperm were similar to those measured in seminal fluid. Levels of Se were higher in sperm (in German) Specimens were heated by microwave radiation in PTFE vessels. After investigation of spectral overlapping and acid effects a full analytical protocol was established (in Russian) The relevance of losses of C CI H and S observed in the analysis of freeze-dried tissue was discussed Specimens from patients with cardiomyopathy had higher Br and Rb lower Fe and similar Zn concentrations to the control samples A simpler faster dry-ashing procedure was developed that was reproducible and free from contamination. Accurate results were obtained with an RM for Cr Cu Fe Mn and Zn but not for Cd and Pb Upper central incisors were divided into sections corresponding to prenatal and postnatal formation.Analytes were separated by cation- exchange chromatography and then measured (in Japanese) Various Calculi AAi-i- Different concentration ratios were found in oxalate phosphate and urate stones (Cu Fe Mn Zn) (in French) samples from children living in Indian slums. Other elements were also determined (Br Ca Cu Fe K Pb Rb Se Zn) tumour bearing tissues was described medicines were measured (in Chinese) Various (9) Blood PIXEi-;- High blood lead concentrations were found in 921315 Various (6) Bone teeth XRF-$5 The distribution of As Ca Fe Pb Sr and Zn in 9213 1 8 Various (1 1) Traditional medicines AA;FL The concentrations of metals in six Chinese 921338 9213 1 1 *Hy indicates hydride generation and S L G and SI signify solid liquid gaseous or slurry sample introduction respectively.Other abbreviations are listed elsewhere. (or indium) and iridium (or bismuth) as respective internal standards. For a digest of NIST Bovine Liver SRM measured concentrations of 13 elements were within 20% of the certified concentrations except for As Cr and Hg. For NIST Oyster Tissue SRM all of the 15 elements measured except for Ag and Ni were within 30% of the certified concentration. The use of ICP-MS for stable-isotope studies has been reviewed by Janghorbani et al. (9 1/2963). Recent publica- tions from the same group focused on the use of the isotope 58Fe for determining the bioavailability of Fe in meals (9 1/32 14) the incorporation of Fe into erythrocytes (91/23 15) and Fe absorption from infant foods (91/23 16). These studies are discussed in more detail in section 1.7.1 1.Further work by Dever and Bresee on the use of 65Cu to trace the incorporation of dietary Cu into hair has been published (9 113380). Practical problems concerned with the use of ion chromatography to preconcentrate and separate trace ele- ments for determination by ICP-MS were discussed by Long and Rivielio (91K3663). By using a chelating resin containing iminodiacetate groups at pH 5.5 trace elements were retained while metals of Groups I and I1 and major anions passed straight through.The trace elements were eluted with 1.25 mol dm-3 nitric acid and pumped directly into the ICP. The system was used for the analysis of urine. Inductively coupled plasma MS is proving especially useful for the determination of elements with high relative atomic mass many of which are difficult to measure by other techniques. Relative freedom from isobaric interfer- ences in the higher mass range is an obvious advantage. Elements in this category include Pb (91lC1657 91/C1658) Pt (91/C1661) long-lived radionuclides such as Th and U (91/C1681 91/2251) and the rare earths (9 ~3585). Further details of these applications can be found in section 1.7 under the respective elements Recent applications in the biomedical field of isotope dilution using LC and TIMS were reviewed by Yergey et al.(9 112358). Isotope dilution using SIMS measurements was applied to the determination of Th and U in NIST Oyster Tissue SRM by Simon et al. (90/26 17) and the results were compared with those obtained by TIMS. According to a recent review by Kaufmann et al. (9'1/2155) about 30% of the publications using LMMS relate to the biomedical field. Their review summarized achievements and limitations of the technique for the analysis of biomedical samples and attempted to elucidate its future potential. A further paper from the same workersJOURNAL OF ANALYTICAL ATOMIC SPECTROMETRY APRIL 1992 VOL. 7 87R (9012382) described a study by LMMS of amalgam tattoos in the oral mucosal membrane. These are generally considered to be the result of high-speed penetration of the mucosa by dental amalgam particles during dental treatment. A novel application of GDMS to the determination of trace elements in body fluids was developed by Evetts et al.(92/C40). Urine samples were first diluted 1 + 10 with concentrated HN03 and evaporated to dryness to drive off organic material. The ash was dissolved in 5 ml of H20 to which Bi was added as an internal standard. An aliquot of prepared sample (10 pl) was deposited onto a carbon tip which formed the cathode and dried with a hot air drier. Concentrations of Pb and Pd in the urine of patients treated with cisplatin were measured which revealed some Pb mobilization as a result of the patients’ treatment. As a result of a study of 41Ca:40Ca ratios in bones by accelerator MS Fink et al. (92/23) concluded that radiocal- cium dating of bones was not likely to be a viable technique.The 41Ca concentrations in bone were found to be disappo- intingly low markedly variable and close to the sensitivity limit. 1.3.3. X-ray fluorescence spectrometry Kobayashi (9 1 / 1054) developed an energy-dispersive system for element mapping by XRF especially for biological samples. A point source was generated with a conventional X-ray tube and a pinhole collimator. An area about 200 mm square could be surveyed with a resolution of about 200 pm producing a two-dimensional distribution of up to three elements shown as different colours. Methods were sug- gested to improve the resolution to about 30 pm. While the small sampling area is a distinct advantage in element mapping for bulk analysis the sample area ana- lysed may not be representative of the whole.Spyrou et al. (9113829) studied this for PIXE with a 0.5 mm diameter proton beam in order to determine sampling factors which could be used to calculate representative masses for a given subsampling error. Seventeen elements were determined in a number of biological RMs for this study. Pinheiro et al. (92/74) developed an acid digestion procedure with microwave heating for preparation of biological samples for analysis by PIXE. Eight CRMs were analysed to assess precision and accuracy. Deviations from certified values were less than 5% and the over-all precision was 2-5% RSD at concentrations greater than 2 pg g-l. The applicability of EDXRF and PIXE to the measure- ment of trace elements in biological samples was discussed by La1 and Choudhury (91/4008).Examples were given from their work on trace elements in cancerous tissues and on the toxic elements Hg and Pb. Prange et al. (91/3151) showed that it was possible to determine 11 elements in whole blood and serum by TXRF. Samples were digested with HN03 using microwave heating for the determination of Ca Cu Fe K P Pb Rb S Se Sr and Zn. Extension to Mn and Ni was possible if separation was carried out to remove Fe and the salt matrix. Analysis of NIST SRMs gave results that compared well with certified values. Bethel et al. (9 1/ 1505) also using TXRF decomposed samples in a plasma asher with 02. After dissolution in dilute HN03 the major electrolytes were removed on a column of Hyphan cellulose. The trace elements were eluted with 6 mol dm-3 HCl and the eluate was evaporated for determination.The ion-exchange separ- ation procedure improved detection limits to the range 0.6-2.0 pg 1-l. By analysing both the direct digest and the trace elements separated as above it was possible to determine 16 elements with results in accordance with literature values. The less-sensitive radionuclide XRF technique was used by Bumbalova et al. (92/208) to determine Br Fe Rb and Zn in whole blood from patients with dilated cardiomyopathy. Bromine and Rb concentra- tions were higher than in controls whereas Fe was lower and Zn was the same. One advantage of XRF is the ability to measure fairly small samples making it very suitable for studying cells.Total-reflection XRF was used by Niemann et al. (9 1/2626) to study changes in Ca P and trace element concentrations during biomineralization in an in vivo tissue culture system. The thin film approach was adapted by Bray et a[. (9 1/2628) to study by EDXRF the concentrations of Cd Cu Fe Ni Se and Zn in leukaemic cells. Spectra obtained by X-ray absorption spectroscopy showed that Cu accumulated by yeast cells was univalent and exclusively coordinated to S atoms (9112627). The disposition of the atoms and the interatom distances were not in agreement with those found in purified yeast thionein. The distribution of Ca Cu Fe K Mn Rb Se and Zn in 50 different regions of 12 normal human brains was studied by Duflou et al. (9 111 187). Measurements by PIXE revealed that concentrations of Cu Fe K Rb Se and Zn (expressed per dry mass) were higher in the gray than the white matter. In the brains of patients with Alzheimer’s disease A1 accumulates in the neurofibrillary tangles associated with the disease.Per1 and Good (91/3462) reviewed their studies by X-ray techniques on this accumulation of A1 while Moretz et al. (9 1/3461) found that the levels of A1 measured by EPMA were very variable and concluded that A1 was not involved in the formation of the neurofibrillary tangles. Lead concentrations in the cerebellum of suckling rats was studied by Lindh et al. (911883) with element mapping by micro-PIXE using a 3 pm proton beam. Results obtained on bulk analysis were in agreement with those by AAS. X-ray microanalysis of cryofixed nervous tissue for elemental analysis and mapping was reviewed by Wroblewski (9 1 /935).Concentrations of Br Cu Fe K and Zn in cirrhotic human livers and normal livers were measured by Laursen et al. (9 1/4040). Cellular and connective tissue fractions were separated and analysed by XRF. In cellular fractions of cirrhotic livers concentrations of Br and Cu were higher than normal Zn lower and Fe and K within the normal range. For connective tissue cirrhotic livers had again levels of Br and Cu higher than normal with K and Zn lower and Fe within the normal range. The authors concluded that in measuring the concentration of elements in cirrho- tic livers correction for the connective tissue content made no practical difference to the result. Further developments have been made for in vivo techniques of measuring Pb in bone (91/1189 91/1385 92/72) and are described in more detail in section 1.7.14.Teraki and Uchiumi (92/318) used element mapping XRF to study the distribution of As Ca Fe Sr and Zn in the teeth and bones of control and tumour-bearing rats. Concentrations of K S and Zn in teeth increase with age while C1 concentrations decrease according to measure- ments on 2 18 permanent teeth by Struzak-Wysokinska and Kot (91/1136) using XRF. Calcium and P concentrations showed no change with age. Plaque-like black deposits on teeth in Zn deficiency were studied by Teraki and Ishiyama (91/3564) using a rat model. Examination of the teeth of Zn-deficient rats by element mapping XRF and EPMA revealed the presence of Fe on the surface of the teeth.This was related to higher than normal concentrations of Fe in saliva and plasma. In examining why some dental implants gave rise to rejection phenomena Simonoff et al. (9 1 /25 79) found that the pegs and receptacle of the implant were of different alloy compositions leading to a galvanic cell being set up with saliva. Dissolution of the alloys gave a migration of Fe and Ni from the implant into tissue as revealed by XRF analysis of the implants before and after implantation. A range of elements was determined in gallstones by88R JOURNAL OF ANALYTICAL ATOMIC SPECTROMETRY APRIL 1992 VOL. 7 Tsukada et al. (9 1/3525) by XRF. Samples were ground and then pelletized. Results obtained for Ca Cu Fe Mn Pb and Zn were compared with those obtained by AAS.These workers suggested that the results of gallstone analysis could be used as an index of environmental pollution. X-ray techniques have become particularly important in studies of trace elements in hair (91/1025 91/1163 91/2527 91/3467 9 1/4000 92/209 92/285). Further de- tails can be found in section 1.6. Concentrations of the essential elements in traditional Chinese medicines measured by PIXE did not allow a direct correlation with pharmacological effects as Xiao and Qin (92/77) had hoped. 1.3.4. Other multi-element techniques and studies The use of the Hitachi 2-9000 simultaneous AA spectro- meter with ETA to determine Al Cd Co Cr Cu Pb and Se in blood serum and urine was described by Jagdt and Schlesing (9 1 /257 1). Procedures for the sequential determi- nation of Pb in blood and Cd Cu Se and Zn in serum by ETAAS with Zeeman-effect background correction were developed by Zhao et al. (91/930 91/3514).Sample preparation was by dilution with a diluent containing Triton X-100 and a chemical modifier when appropriate. Platform atomization was used but calibration was by standard additions. Concentrations of Cd Pb and Zn in normal human sperm determined by AAS were the same as in seminal plasma but Se concentrations were higher leading De Beer et al. (92196) to conclude that Se was bound to spermato- zoa. Li et al. (9112694) used AAS to produce reference ranges for Ca Cd Cu Fe Mg Mn Se and Zn in the seminal plasma of normal fertile men. Infertile men were studied by Wang and Liu (91/1090) who used AAS to measure Ca Cu Fe K Mg Mn Na Ni and Zn in seminal plasma.They concluded that together with other para- meters of prostatic function the concentrations of Fe Mg Mn and Ni would allow diagnosis of male infertility. Copper Fe Mn and Zn in urinary calculi were deter- mined by Garban et al. (92131 1) using AAS. The ratios of these elements differed greatly depending on the type of stone. Similar findings were reported by Struebel and Rzepka-Glinder (9113975) who measured Cd Cr Cu Ni Pb and Zn in calculi by ETAAS with platform atomization and either Pd + Mg(N03)2 or Mg(N0,)2 alone as modifier. They also found differences in trace element composition according to the sex of the subject. Milk teeth with caries contained less Cu Fe Mg Mn Ni and Sr and much more Zn than healthy teeth according to measurements made by Struzak-Wysokinska and Wujec (91/3977) using AAS.Ichida et al. (91/1376) studied Cd Mn and Pb in permanent teeth separating the teeth into dentine and enamel. Concentrations of Cd and Mn in dentine measured by AAS correlated well with each other while in the enamel the Mn and Pb concentrations showed good correlation. Tange (92/260) divided deciduous teeth into prenatally and postnatally formed parts. After diges- tion of the parts Cd Cu Pb and Zn were separated by ion- exchange chromatography and determined by ETAAS. Lead and Zn were higher in the postnatally formed enamel than in that formed prenatally. Cadmium and Cu in the teeth accumulated mainly in the prenatal period. A study of Cd Mn and Pb concentrations in bones from urban and industrial areas of Germany was reported by Hedrich et al.(9111329). Measurements were made by ETAAS on digests of rib bones from 83 foetuses and children aged < 2 years and from 55 adults. A depletion of 14% of Mn per month was seen in the infant bones up to the age of 10 months. Hisanaga et al. (91/1143) measured by ETAAS seven elements in ancient bones from six historic eras and compared the results with values for modern bones. Particularly notable were differences in Cd concen- tration which were much higher in early bones. Shah and Belonje (91/2686) reported the results of an extensive study of Cd Cr Cu Mg and Mn concentrations in the livers of 1 18 accident victims aged from 0 to 89 years. High levels of Cu in liver concentrations were found at :birth declining to adult levels by the age of 2 years.Cadmium levels were low at birth and increased with age. Concentrations of Cr Mg and Mn were not affected by age. Magnesium was determined by FAAS and the other elements by ETAAS. In a study of Cd Cu and Zn in the llivers of patients with Alzheimer's disease (92/2 1 9 mea- surements by AAS revealed that concentrations of Cd and Zn were higher than normal. By chromatography on Sephadex G75 it was shown that there was a reduction in Zn bound to metallothionein and an increase in Zn bound it0 fractions with high molecular mass. Metalloproteins in lliver or kidney homogenates were separated by HPLC and Cd Cu and Zn determined by FAAS in a procedure described by Van Beek and Baars (91/3102) to study in ]particular the metallothioneins. Mendis (9 1/1144) found that Mg and Zn concentrations in the fibrous plaques of aortic tissue in atherosclerosis were significantly higher than in normal aortic tissue.Manganese concentrations were also higher but the difference did not reach statistical significance. Measurements were by AAS. Species of sharks from British and Atlantic waters which live inshore have higher concentrations of trace metals than offshore species according to a study by Vas (92/297). Tissue samples from 46 sharks were analysed for Cu Cd IFe Mn Ni Pb and Zn by AAS. Ii "4. Developments in Single-element Techniques Two recent papers exploited the sensitivity obtainable with laser-excited fluorescence techniques. Thallium was deter- mined directly in solid samples of bovine liver by Anzano et LEI.(91/C3678) by laser-excited AF in a graphite furnace. Methanolic solutions of T1 were used for calibration over the range 0-50 pg as the presence of methanol considerably increased the signal compared with aqueous solutions. Butcher et al. (9 1/3 186) described the determination of F by laser excitation of the MgF molecule in a graphite furnace. With front surface illumination the sensitivity was excel- lent giving a detection limit of 0.3 pg enabling urine to be diluted 1 00-fold thus removing interferences. Laser-enhanced ionization spectrometry suffers from in- terference from alkali and alkaline earth ions when analysis of biological materials is attempted. To overcome this handicap Turk and Kingston (9 1/1425) developed a robot-operated chelation chromatography system using a1 Chelex 100 column to separate off the trace elements. The elements Cd Co Cu Mn and Ni were determined in digests of a range of NIST SRMs including Bovine Serum amd Total Diet.The first real application of furnace atomization plasma AES has been reported. Sturgeon et al. (91/1432) deter- mined Cd and Pb in DORM- 1 Dogfish Muscle and TORT- 1 Lobster Hepatopancreas. Calibration was by standard additions as matrix effects were evident. The development of indirect methods for the determina- tion of pharmaceuticals and anions by A S has been continued by workers in Spain. A group at the University of Cordoba described their development of automatic methods based on continuous separation by FI (91/4011).For example methadone (9 112469) was determined by passing a solution of the drug through a microcolumn containing Cd or Zn metal. Reduction of the keto group released metal ions which were determined by FAAS. The methadone could also be determined in urine after extrac-89R JOURNAL OF ANALYTICAL ATOMIC SPECTROMETRY APRIL 1992 VOL. 7 tion of the drug with CH2C12 from the alkaline urine. Other drugs were said not to interfere. Martinez Calatayud and Garcia Mateo (9 1/3 18 1) determined glycine in pharmaceu- tical preparations by passing a solution through a reactor containing finely divided CuCO in an FI system to form a Cu complex determined by FAAS. High pressure from an HPLC pump was required for the system. Researchers at the University of Cadiz (9 1/32 16) developed an indirect method for the determination of oxalate in urine based on the oxalate forming a water soluble Cu complex when shaken with a solution of Cutl oxinate in isoamyl alcohol.The decrease in the concentration of Cu in the organic phase measured by FAAS or by molecular spectrophotome- try at 395 nm was proportional to the concentration of oxalate. Interferences from other organic constituents of urine were generally small but larger interferences were found from Fe Mg and PO4,-. 1.5. Reference Materials and Inter-laboratory Trials Several papers reflect the activity of the Community Bureau of Reference in producing new CRMs a human hair material (9 1 /C 184 1) certified for Cd Hg Pb Se and Zn; a mussel tissue CRM (9 11C1834) certified for As Cd Cr Cu Fe Hg Mn Pb Se and Zn and the improvement of methods for the future certification of MeHg (9 1 /C 1 83 1).The objective of the last mentioned was to improve the performance of the collaborating laboratories in the mea- surement of MeHg. Three collaborative exercises were carried out which highlighted particular problems in the analysis of fish extracts. The intention to produce two freeze-dried tuna fish RMs with different levels of total Hg was outlined. Canadian workers have assessed the homo- geneity of the marine biological CRM LUTS-1 (Lobster Hepatopancreas) by NAA ETAAS and ASV (9 1 / 1 247). Differences were found for Al Br C1 Co Fe I and Na in samples bottled on different days but not to an extent to affect its use as a CRM. Cox (91/2170) described methods for the preparation of bovine blood quality control materials for Cd Hg and Pb and evaluated results obtained by AAS.Bergstrom et al. (91/C1830) found that in an external quality assessment scheme based on bovine blood samples their results for Cd and Pb by ETAAS were too low at higher concentrations. Investigation showed that their recoveries of added Cd and Pb were lower for bovine blood than for human blood. Replacing the NH4N03 modifier with NH4H2P04 removed this difference. Quality control procedures for the determination of trace elements in whole blood and plasma were discussed by Bradley and Leung (9 1/ 1088). Control materials produced in-house need to be calibrated against CRMs. Daily performance should be plotted as means (k SD) on Levey- Jennings charts.Methods to detect analytical drift were described. Participation in an external quality-assessment scheme was recommended. A collaborative study between nine laboratories for the Association of Official Analytical Chemists to produce a standard method for the determination of Zn in serum was reported by Perry (9 1/24 10). Serum was diluted with 0.03% Brij 35 (a surfactant) and determined by FAAS. Reproduci- bility between laboratories was 6.1 and 12.9% for serum Zn concentrations of 6.36 and 0.63 mg l-l respectively. The results of an interlaboratory study to test a method for the determination of Ca and Mg in foods and biological materials including mussel tissue was described by Vaessen and Van de Kamp (91/1108). The method involved pressure digestion with HNO dilution of the digest and addition of LaCl followed by determination by FAAS.The between-laboratory RSDs ranged from 5.3 to 15.3% for Ca and 3.1 to 6.0% for Mg. Interestingly three of the 12 laboratories failed to follow the specified method. 1.6. Hair Analysis Inductively coupled plasma AES is a widely-used technique for the analysis of hair allowing determination of the major elements in hair digests without particular problems (911880 911887 9111482 91/2633 91/3168). Yang and Tan (9111482) used a Ga internal standard to measure Cd Co Cu Fe Mn Mo Ni Pb and Zn. Matrix interferences were eliminated by using matrix-matched standards. A direct technique involving inserting hair into the plasma in a graphite cup was applied by Umemoto and Hayashi (91/1605) for the determination of Cd Cr Ni and Pb.Results on an NIES hair RM agreed well with the certified value. Detection limits were in the range 0.01 2-0.75 pg g-l. Flame A S is also a suitable technique where elements have to be determined at higher concentrations and is particularly useful when just a single element is being studied e.g. Sr (9111527) and Zn (91/C2103) although it has been used for the sequential determination of a series of elements e.g. Ca Cu Fe K Mg Mn and Zn (91/3838). Electrothermal AAS has been used for elements present at lower concentrations such as Pb (9111259 92/94) and Se (9 1 / 10 1 8). He et al. (9 1 / 10 1 8) compared different modifiers for the determination of Se and concluded that copper was better than nickel.Cold vapour AAS enabled the determination of Hg in hair (91/2548). Samples were digested in HN03 at 65 "C and then oxidized with KMn04. The excess of KMnO was removed with NH,OH-HCl and the Hg vapour was released by reduction with SnCl in a closed vessel. After the reaction was complete the Hg vapour was carried through a 290 mm long absorption cell where the integrated absorp- tion signal at 253.7 nm was measured. A mean normal hair concentration for Hg of 1.06 pg g-l and a range of 0.01- 10.59 was found for 351 persons. X-ray techniques offer a number of advantages for the analysis of hair the material can be analysed directly and is not destroyed in the measurement elements difficult to measure by other techniques can be determined e.g. S (91/2527 92/285) and a wide range of elements can be determined simultaneously (9 1 /3020 9 1 / 1025 9 1 /4000 9 1/2609 9 1/3467). Particle-induced X-ray emission offers greater sensitivity (9 1/1163).Synchrotron radiation XRF and micro-PIXE were used by Van Langevelde et al. (92/209) for the measurement of Pb in hair. An extensive study ( 1000 samples) to determine reference concentrations for 18 elements in the hair of healthy young persons under the age of 15 years was reported by Senofonte et al. (91/887). Measurements were by ICP-AES and each stage of the study was carefully planned to ensure reliable data. Further details of the method and quality control were given in a second publication (91/880). A Greek study also attempted to define values for people living in unpolluted areas (9 1/2406).Concentrations of six elements Cd Cr Cu Ni Pb and Zn were measured by ETAAS in acid digests of the hair of 144 agricultural workers aged less than 50 years. Strong positive correlations were found between the concentrations of Cr Cu and Pb and a negative correlation between Cd and Zn. A Chinese study (91/1025) examined concentrations of Ca Cu Fe and Zn in the hair of healthy children and children with rickets. Statistical evaluation of the data indicated significant differences which were sufficient to allow a prediction of rickets from the hair results that were in agreement with the clinical diagnosis. In a further Chinese study (9 1/4000) concentra- tions of trace elements in the hair from renal stone formers and patients with hyperthyroidism or liver disease were measured by XRF and an attempt was made to correlate the90R JOURNAL OF ANALYTICAL ATOMIC SPECTROMETRY APRIL 1992 VOL.7 data with the disease state. Unfortunately the abstract of this paper gave no details of the conclusions. Another intriguing study from China (91/1163) without any conclu- sion in the abstract looked at the concentrations of trace elements in hair liver and kidney by PIXE analysis and attempted to correlate the data between the different tissues. Concentrations of trace elements in the hair of Shanghai citizens over the age of 80 years were measured by Qin et al. (9 1/3020) using XRF. Concentrations of Fe Pb and Zn were found to increase with age. Sky-Peck's study on American subjects (91/2527) showed however a de- crease of Zn with age along with decreases of C S Se and Sr and an increase in Ni.His results obtained by EDXRF also showed differences due to sex (probably as a result of hair treatments in females) hair colour (blondes contained less Fe and Mn than brunettes) race (blacks had higher levels of As and Pb than Caucasians) and position along the strand (Ca Mn Ni Pb and Sr increased in concentration with distance from the root). In a study on Turkish children up to the age of 15 years (91/C2103) concentrations of Zn in serum and hair were measured for controls and for patients with various categories of disease. Mean values for the concentrations of Zn in hair and serum in the different groups correlated well with each other. Exposure to toxic elements causes changes in the structure of the hair as Bencze (9 l/C17 13) observed. Examination by SEM showed disintegration of the cuticle after heavy Pb exposure whereas Cd and T1 caused death to the root of the hair.Antimony exposure resulted in deformation of the outer layers of the follicle. Ahmed et al. (91/1259) failed to find any significant correlation between the concentrations of Pb in blood and hair in schoolboys whereas Van Langewelde et al. (92/209) were convinced that Pb in hair was a good guide to body burden of Pb and found a relationship between concentrations of Pb in hair and the attention capacity of children. Specchiarello et al. (9 1/2633) used concentrations of Pb in hair as an indicator of Pb accumulation from air pollution by traffic. Prucha (92/9 1) examined changes in the concentrations of trace elements in hair of 10 year old children over a six year period (1982-1988); concentrations of Cd and Pb had decreased whereas Fe and Zn had increased.Measurements by Korn et al. (92194) using ETAAS showed higher concentrations of Pb in hair in occupationally exposed than in unexposed control subjects. An extensive study by WDXRF of hair concentrations of workers exposed to HF was reported by Kono et al. (9 113467). Concentrations of F were higher than in controls and they correlated with increases in urinary concentrations of F. Concentrations of Al Ca I Mg Na Ni P and Zn were also higher in the exposed group. On the basis of a study of hair as a possible indicator of A1 exposure in patients on dialysis Wilhelm et al.(Hum. Toxicol. 1989 8 5 ) concluded that hair analysis was of very little value in this case. Concentrations of A1 in hair did not correlate with daily A1 intake cumulative A1 intake or bone and plasma A1 concentrations. Bone analysis was considered to give the most reliable assessment of A1 body burden. 'What contribution can be made to biological monitoring by hair analysis?' was the title of a two-part article by Bencze (Fresenius J. Anal. Chem. 1990,337,867 and 1990 338 58). Although Bencze does not provide a definitive answer to the question posed this article is well worth reading for the insight it gives into the complexity of the structure of hair the way in which trace elements are incorporated the different types of bonding the way in which hair should be sampled and cleaned and the effect this has on the result of analysis (see also Bencze's article 92/95).The severe limitations of interpreting results on individuals rather than groups as in epidemiological stud- ies was stressed. This is apparent also from the work of Henrion et al. (91/2609) a particularly relevant study in that they attempted to correlate results of hair analysis with data on the concentrations of 20 trace elements in autopsy samples of brain liver and kidney obtained by XRF on 41 accident victims aged between 40 and 60 years. Statistical evaluation of the correlation was carried out by three-way principal component analysis partial least squares linear characteristics and the Kendall non-parametric rank corre- lation coefficient. Strong correlation was seen between concentrations in brain liver and kidney especially for As Hg Ni Pb and Se.Correlation between the inner organs and hair was very much weaker which raised serious questions about the usefulness of hair analysis. 1.7. Progress for Individual Elements 1.7.1. Aluminium In a study on the stability of samples during storuge Wilhelm and Ohnesorge (9112171) found that serum samples were stable in poly(propy1ene) or polystyrene tubes for at least 18 months at -20 "C. Acid-washing of the tubes was found to be unnecessary. Water samples required acidification for stability but acidification of the dialysis fluid and urine samples used did not result in any noticeable !difference in stability. Nevertheless acidification was considered advisable to ensure a low pH in the sample.LJnder these conditions urine dialysis fluid and water were stable for 18 months when stored at - 20 "C. The requirements of a method for the determination of the low levels of A1 in the serum of patients with normal renal function are different from those for the higher levels fbund in patients on dialysis as Hewitt et al. (91/2525) rightly recognized. For determination of the higher levels they used ETAAS with a 1 + 3 dilution with 0.1 mol dm-3 FINO 0.5% v/v Triton X-100 and atomization off the wall of a pyrolytic graphite coated graphite tube. For low levels ;a more rigorous sample collection procedure was proposed using a PTFE intravenous catheter and syringes and sample tubes that had been acid-washed. Serum was diluted 1 + 1 with 5.5 mmol dm-3 Mg(N03)2 0.2% v/v Triton X-100 and atomized off a L'vov platform in a pyrolytic graphite coated graphite tube.The rate of analysis by this method was less than half that of the first method and the cost was estimated to be twice as high. By the more rigorous sampling procedure the mean ( -I- SD) normal serum concentration of A1 was 0.044 +_ 0.030 pmol I-' (1.2 2 0.08 pg 1-I) compared with 0.42 -I- 0.15 pmol 1-l (1 1.3 t- 4.1 pg 1-I) when samples were taken with the normal procedure. On the same samples the two methods gave results that correlated well. Delves et al. (9 1/965) preferred NH4H2P04 i l s chemical modifier in their ETAAS method for the determination of A1 in serum adding an O2 ashing step to remove carbonaceous material. The importance of the itshing temperature on sensitivity was studied by JSilroe- Smith and Rollin (91/3288).Removal of carbon species before atomization by introducing a cool stage after ashing produced an increase in sensitivity and ionization interference was reduced by lowering the ashing femperature from 1600 to 1000 "C so as not to lose the casily ionizable salts before atomization. Speciation of A1 in serum in the absence and presence of desferrioxamine was studied by Garcia Alonso et al. (91lC1693) using ion- exchange HPLC with detection on-line by fluorimetry or off-line by ETAAS. Concentrations of A1 in the blood plasma of workers in the aluminium industry were found by Schlatter et al. (92/250) to rise up to 60 pg 1-1 during the shift. Urinary excretion of A1 also rose reaching a maximum of around 5 ,ug h-l about 4 h after ceasing work.Prior to the beginning of the shift plasma concentrations of 111 as measured by ETAAS were in the range 20-30 pg 1-1 compared with 3 pg 1-l in unexposed individuals.JOURNAL OF ANALYTICAL ATOMIC SPECTROMETRY APRIL 1992 VOL. 7 91R Urinary A1 excretion was studied by Wilhelm et al. (9111438) in subjects with normal and impaired renal function. Measurements by ETAAS on samples from healthy volunteers gave a mean A1 excretion of 12.2 pg per 24 h. Six patients with renal failure but some residual renal function excreted a mean of 51.4 pg per 24 h but in their treatment by continuous peritoneal dialysis only 27.2 pg per 24 h were transferred to the dialysate fluid. Further studies with rats confirmed the limited capacity of the kidney to excrete Al.Urine and plasma A1 concentrations in patients given the intestinal agent sucralfate (basic sucrose aluminium sulfate) were measured by Allain et al. (9 1 / 128 1) using ICP-AES. Plasma A1 increased from about 2 to more than 5 pg 1-1 while urine A1 excretion increased from less than 5 pg to more than 30 pg per 24 h. Urinary concentrations of A1 remained higher than normal 5 and 10 d after cessation of sucralfate treatment. Inductively coupled plasma AES was applied by Leflon et al. (91/3214) to the determination of A1 in bone. Samples were digested in HNO and diluted with 1% v/v Triton X- 100 for nebulization into the ICP for measurement at the 396 nm line. Calibration was by standard additions.The main constituents of bone did not interfere. The method gave results about 12% higher than those obtained by ETAAS; these workers considered their method to be more accurate in view of the limited chemical interferences in the ICP. Mean normal concentrations found were 2.4 & 1.2 pg g-l (n=24) compared with a mean bone concentration of A1 in patients on dialysis of 20.3 pg g-l. In the method of Liang et al. (92/13) bone samples were digested with HNO and then diluted before analysis by ETAAS using uncoated tubes and calibration with matrix-matched standards. Problems with matrix interference were less with uncoated graphite tubes than with pyrolytic graphite coated tubes using either wall or platform atomization. Direct solid sampling of tissue for the determination of A1 by ETAAS was described by Nordahl et al.(9 1 / 1 5 1 2). Using the cup-in-tube technique with addition of 0.2% m/v Mg(N0,)2 in 2% v/v HNO and 0.2% v/v Triton X-100 as modifiers A1 was measured using one of eight wavelengths according to the sensitivity required. Minimization of contamination was found to be of paramount importance. Xu et al. (91/C3758) found that in the determination of A1 in brain tissue by ETAAS phosphate caused interference which could be overcome by modifiers such as Mg(N03)2 and Pd. However Liang et al. (9211 3) produced a simple approach applicable to a range of soft tissues. Samples were digested in HNO,. Both standards and digests were diluted 1 + 1 with 0.2 g 1-l of calcium and the A1 was atomized off a L‘vov platform in a pyrolytic graphite coated graphite tube.Recoveries ranged from 94.5 to 103%. Whether A1 is implicated in the development of Alzheimer’s disease is still a very topical issue. Per1 and Good proponents of the theory that A1 is involved have reviewed their studies on the accumulation of A1 in the brain (91/3462). Using SEM with EPMA and LMMS they found A1 in the neurofibrillary tangles (NFTs) seen in Alzheimer’s disease. Moretz et al. (9 11346 1) using EPMA also found A1 and Si but only in about half of the NFTs that they examined. They concluded that the inconsistency of detection and the low levels of A1 present indicated that A1 was not required for the formation of NFTs. French workers (92/299) using EPMA and ion microprobe microanalysis failed to find any measurable A1 in brain tissue from seven patients with Alzheimer’s disease. It should be stated that the evidence of A1 and Si in the NFTs was the initial (and as yet the most convincing) evidence for a role of A1 in Alzheimer’s disease. Lack of verification of the findings by other workers could raise serious doubts about the implication of A1 in this disease.A method for the determination of A1 in alkaline earth containing pharmaceuticals by ETAAS was developed by Reust and Seltner (9 113974). The importance of preventing contamination in all steps was stressed. Injectable solutions are distributed in vials often made of soft glass which can release significant amounts of Al. Guadagnino et al. (9 l/C 1 7 15) measured by ETAAS the release of A1 from three different types of glass.Soda lime-silica glass with a dealkalized surface released the least Al. 1.7.2. Arsenic A review of the determination of total As and As species in biological fluids by atomic spectrometric methods was produced by Violante et al. (91/886). In an attempt to obtain higher sensitivity Zheng et al. (9112390) used non-dispersive AFS for the determination of As in hair. Samples were digested in HN03-HC104 with microwave heating. The conventional hydride-generation technique was used to generate the arsine. The detection limit was 0.6 pg 1-I. Collection of hydrides onto a graphite tube using Pd as both a collector and modifier for subsequent analysis by ETAAS is not new but Chaudry et a/. (91/2930) investigated novel ways of carrying this out practically.A commercially-available probe attachment was adapted to move the hydride injection capillary horizontally into the tube while the Pd was added by the autosampler through the normal injection hole. A second approach was a two-hole tube system in which the Pd was injected by the autosampler through the normal hole and the hydride capillary was inserted manually through a second vertical hole. These two procedures were compared with a single-hole tube in which the Pd and hydride capillary were separately inserted manually. The first two approaches unfortunately gave much lower signals than the manual method probably as a result of increased diffusional losses through the extra openings. The single-hole tube was adopted and applied to various measurements including the determination of As in urine which was verified by a satisfactory result for As in the Lanonorm urine RM.Chloride interferes in the determination ofAs by ICP-MS. Branch et al. (91/3206) showed that it was possible to reduce this interference to negligible levels by introducing a small flow of N2 into the Ar flow to the plasma. Concentrations of NaCl up to 1000 mg 1-1 gave no interference. Sheppard et al. (91/583) chose to remove the C1- interference by ion chromatography. However samples had to be diluted 1 + 19 to prevent overloading the column with chloride and although the chromatography allowed separation of the species monomethylarsonic acid (MMA) and dimethylarsinic acid (DMA) were not resolved. By contrast Branch et al. (9 1/3206) used a lesser dilution (1 + 9) and were able to determine total As directly and also after separation by HPLC which completely resolved the species.Total concentrations of As agreed with the sum of the concentrations of the individual species in four samples from two volunteers who had eaten fish meals. Total As in Dogfish Muscle RM (DORM-1) was also determined after microwave digestion giving a result that was in good agreement with the certified value. Most studies have focused on the speciation of As. Using HPLC-HGAAS Kreppel et al. (92/80) found that mice injected with As111 excreted about 15% of the As in urine mainly as AsIU with small amounts of AsV and DMA and traces of MMA. When mice were injected with AsV about 40% was excreted mainly unchanged as AsV with smaller amounts of AslI* and DMA.Thermospray MS was applied by Cullen and Dodd (91/3407) to the identification of As species separated by HPLC either on an anion-exchange or a reversed-phase column. Roehl et al. (911C3738) demon- strated the suitability of a Dionex AS4A anion-exchange column with gradient elution for the complete separation of92R JOURNAL OF ANALYTICAL ATOMIC SPECTROMETRY APRIL 1992 VOL. 7 As species. Detection was by ICP-MS. Chloride in urine samples eluted at a time which did not interfere in the detection of any of the As species. The work of Atallah and Kalman discussed in last year's review (91/3594) has now been published (92/129). They used a novel on-line phatooxidation system to convert organoarsenic compounds into arsenate for detection by HGAAS.The sample was mixed with potassium persulfate in the flow system and passed through a coil of PTFE tubing wrapped round an Hg lamp. The system could be interfaced to a chromatographic column to act as a detector for As speciation. A study on As-bindingproteins in rat livers by Huang et al. (91/C3686) revealed that there were three main compo- nents two proteins of high molecular mass and one of mass around 6000. The supernatant produced by homogenizing the liver in Tris-HCl buffer was separated on a Sephadex G75 column. Fractions eluted with Tris-HC1 buffer at pH 7.4 were analysed for As by ETAAS with a Ni modifier. 1.7.3. Boron The suitability of DCP-AES for the determination of B in tissues and cells was demonstrated by Barth et al. (92/97). Samples were digested with H2S04 and H202.The method was shown to be independent of the species of B and to have a sensitivity of 0.1 mg 1-l. 1.7.4. Cadmium The toxicology of Cd and its determination in a range of environmental and clinical samples was reviewed by Ro- bards and Worsfold (91/3766). For the determination of Cd in biological fluids by ETAAS Smeyers-Verbeke et al. (9 1/846) investigated the performance of the mixed Pd-NH4N03 modifier proposed by Yin et al. (Anal. Chem. 1987 59 1462). Palladium alsc stabilizes NaCl giving rise to background problems but reduction of the amount of Pd injected from 50 pg (as used by Yin et al.) to 6 pg gave good recovery of Cd with lower background absorbance due to NaCl. The suitability of the low amount of Pd modifier for the accurate determination of Cd in blood milk and urine was demonstrated.Peterson et al. (91/3509) extended the sensitivity of ETAAS for the determination of Cd in blood plasma and urine of persons with low level environmental exposure by preconcentrating and isolating the Cd on an anion-exchange column. Whole blood and plasma were deproteinized with HN03 resulting in transfer of Cd to the supernatant of 99 and loo% respectively. The Cd was retained on the column as the anionic chlorocadmium complex. The Cd eluted was determined directly against simple aqueous standards. At the 0.1-0.3 pg 1-l concentration level the between-batch RSD was 12%. A study of concentrations of Cd in blood showed no statistically significant differences between the general populations of three Chinese cities (9 1/2680).As is to be expected the blood Cd levels of smokers were significantly higher than those of non-smokers. Ohta et al. (91/1544) applied their ETAAS method to the determination of Cd with a molybdenum tube atomizer and thiourea chemical modifier to its direct determination in slurried tissue samples. Powdered samples were dispersed ultrasonically in H 2 0 before injection into the atomizer. Comparison with determination after wet digestion showed that the slurry method gave better accuracy but the RSD was greater. Sample masses (dry) used were between 1 and 10 mg. According to Blackstone et al. (91/3378) sample masses of 200-300 mg of wet tissue were required to obtain precise values. They solubilized rat tissue in a quaternary ammonium hydroxide for determination by AAS.Despite the many satisfactory direct determinations that have been demonstrated for Cd some workers continue to develop slolvent extraction procedures. Almendro et al. (9 1/C 1753) dleveloped a procedure for ICP-AES and ETAAS based on the reagent 1,5-bis(di-2-pyridylmethylene)thiocarbonohy- dirazide which was claimed to extract Cd from digested biological materials into IBMK with preconcentration ratios of up to 30. 1.7.5. Calcium According to a study by Kuelpmann et al. (91/1280) on the influence of methods on accuracy in the determination of (:a in serum FAAS and FAES gave only a small negative bias in results for quality control materials compared with reference values (mean deviation - 1.2 and - 0.1 Yo respec- tively). Colorimetric methods however showed a significant positive bias ( f 2.3% for Methylthymol Blue and + 2.2% for Cresolphthalein methods).1.7.6. Chromium Recent papers show a variety of approaches to the determi- nation of Cr in biological fluids by ETAAS. Benling and Yongming (91/2958) found that a mixture of Ca and Mg enhanced the signal far more than the separate effects of Ca and Mg. With a Ca-Mg modifier ashing temperatures up to 1.5 50 "C were possible. Unrealistically high urine and serum concentrations of Cr were reported. A modifier containing 0.2% m/v sodium citrate and 0.2% m/v Na2S0 was found by Tang et al. (91/1041) to confer higher sensitivity in the determination of Cr in serum and a detection limit of 0.7 pg 1-l was reported. Shan et al. (9112945) used a solvent extraction step with a high molecular mass secondary alkylamine to increase the sensitivity of the determination of Cr in urine after digestion giving a detection limit of 0.01 pg 1-l.Surprisingly the step was also said to be necessary to overcome background correction problems; others (92/2 12) using direct methods confirmed that the deuterium-arc background correction system satisfactorily eliminates background interference. Since the two principal oxidation states of Cr have different toxicities speciation is important. Qi et al. (9 113865) examined various metal-chelating agent combi- nations for selectively coprecipitating CrW. Manganese with IVaDDC was the most suitable and when applied to human urine samples allowed recoveries of 87.5- 108.2% for C P and 94.4-105.3% for CrlI1 to be achieved when measured by ETAAS.The ability of human blood erythrocytes selec- tively to take up Crw as chromates in the presence of CrlI1 was exploited by Neidhart et al. (91/3153) for speciation studies in other materials. The cells were immobilized in calcium alginate beads to give increased mechanical stabil- ity. After sampling the erythrocytes were separated and subjected to a multiple-step clean up procedure before determination of the Cr by ETAAS with an O2 ashing step in the furnace programme. The potential of this novel procedure was illustrated by the measurement of Crm in airborne particulates. Measurement of faecal Cr when Cr is used as a marker is ;in analytical problem of a different nature. Concentrations :ire high allowing determination by FAAS but accuracy is important.Mir et al. (91/1369) measured the recovery of ridded Cr from cattle faeces when different pre-treatment rind digestion procedures were used. Better recovery was obtained when samples were frozen when fresh and then l.hawed for analysis than if they were dried in either a conventional or microwave oven for storage. Problems of incomplete recovery remained with the three digestion procedures tried. Concentrations of Cr in lung tissue were studied by IRaithel and Schaller (91/3234). In normal lungs Cr wasJOURNAL OF ANALYTICAL ATOMIC SPECTROMETRY APRIL 1992 VOL. 7 93R found at higher levels in the upper lung areas and showed substantial variation throughout the lung. For 30 autopsied lungs from persons without a history of occupational exposure to Cr median concentrations ranged from 742 to 1375 ng g-l (dry mass).Seeman and Wittig (91/1079) found that it was possible to use lung tissue that had been stored in formalin without invalidating the analysis provided defined precautions were taken in collecting and handling of specimens. 1.7.7. Copper Fang et al. (91/2375) separated proteins in seven human serum samples by agarose gel electrophoresis and deter- mined Cu in fractions by ETAAS. They found that 80.2% of the Cu was associated with the at,-globulin fraction and 19.8% with the albumin fraction. This is in contradiction to the majority of previous studies which show that over 90% of the Cu is associated with a single protein caeruloplas- min. Copper in animal liver was determined by FAAS in a method developed by Vidal et al.(9 1 /3 103). Freeze-dried liver was digested with HN03-H2S04 in a pressure vessel and then diluted for direct analysis. Copper was shown by Dastych (9113504) to be fairly homogeneously distributed throughout human stools; less than 13% variation was seen. The mean Cu concentration for healthy subjects was 45.9 pg g-l. Determination was by FAAS after wet digestion. 1.7.8. Fluorine Molecular fluorescence from MgF was used as the basis of measurement of F in urine by Butcher et al. (91/3 186). The MgF vapour was produced in an electrothermal atomizer and the fluorescence excited by a laser using front-surface illumination. The technique had a linear range of five orders of magnitude and an extremely low detection limit of 0.3 pg but suffered from interferences from other ions.Because of the high sensitivity urines could be diluted 100- fold to remove interferences. The value obtained for a certified RM was then within the certified range. 1.7.9. Germanium Levels of Ge in plants and animals are in the range 8-203 ng g-l according to results obtained by Hara et al. (9 1/4045) using a procedure based on wet ashing followed by determination by ICP-AES with FI hydride generation. Studies by Schleich and Henze (9113228) showed that the determination of Ge in biological samples is possible by ETAAS. The presence of nitric acid or nitrates improved sensitivity and interferences could be minimized by the use of a Pd(N03)2-Mg(N03)2 modifier. The detection limit was 20 pg 1-l.A nickel modifier was used by Lin et al. (921 154) for the determination of Ge in herbal medicines by ETAAS. Precision was better than 4% RSD and recoveries ranged from 97.8 to 99.1%. 1.7.10. Gold For a study on the pharmacological properties of a mixture of copper silver and gold salts Thunus and Dauphin (91/2470) developed a method for the determination of Au in rat plasma by ETAAS with Zeeman-effect background correction. Samples were diluted with a solution containing Triton X-100 and an anti-foaming agent. 1.7.1 1. Iron Methods for the determination of Fe and total iron-binding capacity (TIBC) in serum were compared by Vercammen et al. (9 1/2942). For patients who were given Fe-dextran for anaemia the serum Fe and TIBC concentrations measured by FAAS were much higher than results by spectrophoto- metry with ferrozine as reagent and with the Kodak Ektachem slide technique.The last two procedures measure uncomplexed Fe while FAAS measures total Fe. The complexing effect of the dextran was confirmed in vitro. For patients not on this treatment results by the three tech- niques correlated well. A confusing paper by de Benzo et al. (92/246) surprisingly accepted by a leading clinical chemis- try journal described the determination of Fe in serum by FAAS and ETAAS and attempted to compare deproteiniza- tion with direct analysis. As is to be expected deproteiniza- tion was shown to avoid the problems of haemolysis of samples. In an attempt to understand the reason for higher than normal concentrations of Fe in liver tissue in patients with alcoholic liver cirrhosis Raedsch et al.(9 1/32 15) measured Fe in bile by ETAAS. Samples were diluted 1 4 4 with 1 mol dm-3 H2S04 which was sufficient to give good accuracy as shown by correct recoveries. Biliary excretion in healthy controls was 0.32 & 0.09 pmol h-l (mean -t SD) whereas in patients with cirrhosis it was 0.45 2 0.14 pmol h-l. Thus the accumulation of Fe in the liver of patients with cirrhosis cannot be explained by reduced biliary excretion of Fe. Iron in animal liver was determined by Vidal et al. (9 1/3 103) using FAAS. Samples were digested with HN03-H2S04 in a pressure bomb for 2.5 h diluted with water and directly measured. There is nothing new in this but it does demonstrate that this determination is straight- forward as was confirmed also by Zhang and Huang who determined Fe in liver-extract tablets in a similar way (9 1/3579). The absorption of non-haem Fe from meals by children was studied by Fomon's group (9112314) using the stable isotope 58Fe and measurements by ICP-MS.The mass isotope ratio 58Fe:57Fe was measured in blood before and 14 d after giving the 58Fe spiked lunch. The geometric mean percentages of Fe incorporated into erythrocytes was 2.02 and 1.05 for lunches with beef patty and beef-soy patty respectively. A variety of iron-fortified infant foods were also studied by the same technique (9 1/23 16). Further work showed a negative correlation between the incorporation of Fe into erythrocytes and the serum ferritin concentration (9 1/23 15).Results showed a large between-subject variation for incorporation of Fe indicating that study designs based on group comparisons were likely to be of little value. For the same subject comparison of availability of Fe from different foods was a satisfactory proposition. 1.7.12. Lanthanides The concentrations of Ce Gd La Tb and Yb in the plasma and urine of healthy individuals were measured by Allain et al. (9 1/3585) using ICP-MS. Concentrations for all ele- ments in both plasma and urine were < 0.3 pg l-l except for one urine containing 1.5 pg 1-l of Ce. Samples were diluted 1 + 5 with 1% v/v HN03 containing Eu as an internal standard. They concluded that higher values obtained previously by other workers using NAA were erroneous owing to contamination. For a study of the pharmacokinetics and biodistribution of Gd in rats Liang et al.(9113268) developed a method using solvent extraction and ETAAS for the determination of Gd in tissues. Considerable improvement in sensitivity and absence of memory effects were achieved by atomizing from a tantalum boat rather than from the wall of a pyrolytic graphite coated graphite tube. It was possible to measure Gd concentration in the range 0.92-72.0 pg g-' with a precision of better than 10% RSD.94R JOURNAL OF ANALYTICAL ATOMIC SPECTROMETRY APRIL 1992 VOL. 7 1.7.13. Lithium For the determination of endogenous Li concentrations in serum and urine by ETAAS Sampson (91/3199) obtained much higher sensitivity by using tubes coated with tanta- lum in order to prevent the formation of lithium carbide.The tubes were coated in situ by injecting a solution of ammonium heptafluorotantalate. Background correction at the high wavelength of Li (670.8 nm) is difficult. In order to overcome this the background was reduced to a negligible level by maintaining a flow of argon (50 ml min-l) during atomization. Serum samples were deproteinized with 10% HN03 and the Li analysed against aqueous standards. Interferences were more pronounced for urine samples but reduced by dilution. Samples were diluted 5-fold with 5% HN03 and analysed against aqueous standards. For each urine a recovery measurement was made. Where recovery was low the determination was repeated using standard additions. Mean normal concentrations found were 0.17 pmol 1-l (n= 19; range 0.05-0.39 pmol l-l) for serum and 1.50 pmol 1-l (n=13; range 0.26-4.9 pmol 1-l) for urine.The method was also applied to the determination of Li in nanolitre volumes of micropuncture fluid from rat kidney tubules. The application of an ETAAS method for the assessment of Li clearances as an indicator of renal distal delivery of sodium and water was described by Durr et al. (91/2656). The method for determination was described in last year's Update (91/3594). Their approach is original in that the clearance is assessed on the measurements of endogenous Li levels in serum and urine and not on results after giving a Li load. Determination of Li in rat brain regions and sub-cellular fractions by ETAAS was described by Rios and Guzman- Mendez (921218). Of the eight brain regions studied highest Li concentrations were found in the hypothalamus corpus striatum and mid-brain.1.7.14. Lead Much has been published over the years on the determina- tion of Pb in blood by ETAAS. As a result in recent literature there is little that is new just new people rediscovering old solutions. Deproteinization of blood with nitric acid is one such solution principally introduced by Stoeppler et al. in 1978 (Analyst 1978 103 714) a method which this reviewer can endorse as capable of excellent accuracy. This approach has been used by Honda et al. (91/2566) and Jacobsen et al. (92/14). The latter workers diluted blood 1 + 4 with a diluent containing 0.25% v/v Triton X-100 and 0.5% v/v Antifoam B and then added an equal volume of 1.6 moll-' HN03.After centrifugation the supernatant was analysed with addition of NH4H2P04 as modifier. Direct aqueous calibration was used. Excellent reproducibility was obtained for blood Pb concentrations of 0.25 1.98 and 3.76 pmol l-l within-batch precision was 3.2 1.8 and 1.4% RSD respectively whereas between batch it was 7.3 2.9 and 2.2% RSD. The workers were justifiably proud of their performance in the Quebec interlaboratory comparison programme coming second out of 66 labora- tories. Vasconcelos et al. (9 1/C 1793) developed a method using graphite probe ETAAS for which no modifier was necessary and aqueous calibration could be used. Samples were simply diluted 10-fold with 0.05% Triton X-100. Accuracy was demonstrated by a satisfactory result on a BCR Blood CRM and the reproducibility at 129 pg 1-1 was 3.8%.Romero and Granadillo (91/C1941) found that 2 mg 1-l of Pd was a suitable modifier. Samples were also diluted 10-fold with Triton X-l 00 solution and sample and modifier added separately onto a L'vov platform. An O2 ashing step was incorporated into the furnace programme. In a study of blood lead levels in 9-10 year old Danish schoolchildren by ETAAS Lyngbye et al. (91/2593) at- tempted to elucidate factors that contributed to Pb burden. P,arents' tobacco smoking the child's number in the sibship gender and consumption of canned food at home were identified from interview data as contributing signifi- cantly to the blood Pb concentration. Concentrations found were in the range 0.08-0.70 pmol 1-I. Ahmed et al. (91/1259) reported a mean blood Pb concentration of 0.33 pmol 1-l for 200 schoolboys aged 6-8 years from Saudi Arabia. Hair Pb concentrations were also measured but results failed to show any significant correlation to blood Pb d,ata.High blood concentrations of Pb were found in children living in a slum area in Bombay by La1 et al. (92/3 1 5 ) using PIXE. Environmental pollution from heavy vehicular traffic and from industry was identified as the source of Pb. Blood Pb levels in adults living in three Chinese cities were studied by Qu et al. (91/2680) using ETAAS. Geometric mean concentrations of 0.56,0.62 and 0.45 pmol 1-1 were found for Hefei Shenyang and Jinxi respectively; the differences reached statistical significance. Hiigher blood Pb concentrations were found in smokers.To determine Pb in serum by ETAAS Imai et al. (921279) diluted samples with a modifier containing CT(NO~)~ and Triton X-100. In 20 sera from normal subjects the mean concentration of Pb was found to be 1.3 pg 1-l. The detection limit was 0.22 pg 1-l in serum and the precision was 4% RSD. Li et al. (91/3536) compared results of determination after wet digestion and direct determination after dilution with an emulsifier. Both approaches showed satisfactory recovery and results by the two methods were in good agreement. Speciation of Pb in serum and red cell haemolysate was studied by Gercken and Barnes (911C2838) using size- exclusion chromatography. The chromatograph was di- rectly coupled to an ICP mass spectrometer for detection. Iin serum most of the Pb appeared together with caerulo- plasmin.In red cell haemolysate the main fraction was located at 250 kDa with minor fractions at 140 and 30 kDa. The last two appeared at the same time as haemoglobin and carbonic anhydrase. Direct determination of Pb in urine was developed by k'anagisawa et al. (91/2255) using a separative column atomizer in ETAAS. Of the various column packings tried glassy carbon activated charcoal graphite and alumina graphite gave the sharpest peak. The Pb peak was satisfacto- rily resolved from the background peak. A more conven- tiional approach was described by Huang et al. (91/1149) who diluted samples with an NH4H,P04-ascorbic acid modifier for determination by ETAAS. Recoveries of 92-1 05.3% were obtained. Direct measurement of Pb exposure through in vivo measurement of Pb by XRF continues to progress.Chettle et al. (91/1385) improved their method for bone by changing the design of the source collimator and by including data from two less intense peaks. In this way the d.etection limit was reduced from 18 to about 10 pg g-l of F'b. The same group (91/1189) evaluated the use of three different bone sites finger (measured with a 57C0 source) tibia and calcaneus (both measured with a lo9Cd source). FLesults correlated strongly with each other but measure- ment precision was best for the tibia. The technique has also been used by Morgan et al. (92/72) to study Pb body burdens in a population from Swansea UK not occupation- ally exposed to Pb. Todd et al. (92/307) wrote a Monte Carlo programme to model the in vivo XRF of Pb in the kidney in order to help choose the most appropriate source and measurement geometry.The system chosen used a 99Tc source with a backscatter geometry for measurement. An in vivo acid-etching technique for sampling surface enamel of teeth was used by Cleymaet et al. (91/3913) for t!he determination of Pb by ETAAS. Factors contributing to t'he contents of Pb were found to be etch depth tooth typeJOURNAL OF ANALYTICAL ATOMIC SPECTROMETRY APRIL 1992 VOL. 7 95R and age of subject. Subjects living close to a non-ferrous metals plant had significantly higher enamel Pb concentra- tions than other urban dwellers had. The distribution of Pb in the cerebellum of suckling rats was examined by Lindh et al. (911883) using micro-PIXE for determination.Results were in good agreement with results obtained by AAS. Lead was present in higher concentrations in the cerebellar white matter (1 1-1 8 pg g-l dry mass) than in the cortical grey matter (2.0-5.5 pg g-l). The first application of furnace atomization plasma emission spectrometry the determination of Cd and Pb in sediment and biota was described by Sturgeon et al. (9 1 / 1 432). The elements were determined in DORM- 1 Dogfish Muscle and TORT- 1 Lobster Hepatopancreas. Because of spectral interference from iron at the 283.3 nm line the less sensitive 217.0 nm line was used. Matrix effects were evident (no modifier was used) but satisfactory results were obtained by standard additions. There have been developments in preconcentration tech- niques for the determination of Pb.To allow determination of Pb in biological materials by FAAS Fang et al. (9 1/3779) developed an on-line FI system using coprecipitation of Pb with the Fell-hexamethyleneamrnonium hexamethylenedi- thiocarbamate complex with a collection efficiency of greater than 95%. This was collected in a knotted reactor made of 0.5 mm i d . plastic fine-bore tubing. After collection turning an injection valve allowed IBMK to be pumped through the reactor (dissolving the precipitate) into the nebulizer of the AA spectrometer. A sampling frequency of 90 h-l could be achieved with a precision of 2.7% RSD at 200 pg I-'. The detection limit was 2 pg 1-l. The method was applied with success to the determination of Pb in bovine liver and blood RMs.Prior digestion with HN03-HC104 was necessary. A batch coprecipitation technique was used by Dabeka (91/C1943) to enhance the sensitivity of ETAAS for Pb in biological materials. To the analyte solution containing a Pd carrier a reducing agent of 15% m/v ascorbic acid was added. Precipitation was successful over a wide range of acidities and was relatively selective for Pb as elements such as Cd Co Fe and Ni were not coprecipitated. Sturgeon et al. (9 111 546) produced volatile tetraethyllead by reaction with sodium tetraethyl- boron to preconcentrate Pb. The tetraethyllead was col- lected on a pyrolytic graphite coated graphite tube heated at 400 "C and then atomized at 1600 "C at the end of the collection period. The method was evaluated by determina- tion of Pb in marine water and marine biological tissue RMs with good results.Crews (9 l/C 1658) has described the analytical conse- quences of an incident in 1989 when animal feed contami- nated with Pb was imported into the UK and distributed to farms in the West and South West of England. It was necessary to monitor animal feeds cattle tissues blood and milk for Pb and other elements. Inductively coupled plasma MS proved very useful in coping with the number of samples generated in this emergency. (N.b. Clinical labora- tories in the UK experienced in the determination of Pb in blood collaborated in this period.) 1.7.15. Magnesium Kuelpmann et al. (9 1/1280) found that FAAS methods gave accuracy superior to spectrophotometric methods in the determination of Mg in serum For a range of quality control sera the mean deviation from the reference values was - 0.4% for FAAS procedures.For spectrophotometric methods using Magon and Camalgite however mean deviations were + 8.2 and +9.3% respectively and on several occasions results failed to meet the quality control requirements. After humans are exposed to heat serum Mg concentrations fall according to Nishimuta and Suzuki (91/1183). Of a range ofelectrolytes and trace elements Mg was the only element to show such a decrease which was not due to increased loss in sweat. They suggested that as a result of the increased circulation Mg was transferred from serum to intracellular spaces. Electrothermal AAS procedures are not often used for the determination of Mg in biological samples.For the small volumes of kidney tubular fluid that are available Seow et al. (9111059) found ETAAS to be a suitable technique. Kollmier et al. (91/1178) also used ETAAS for the determi- nation of Mg in autopsy tissue samples. Mean Mg concen- trations found were 729 589 536 490 and 546 pg g-l of dry mass for heart kidney liver lung and spleen respec- tively. 1.7.16. Manganese For the determination of Mn in whole blood by ETAAS Lin et al. (9111576) diluted samples 1 + 10 with 0.1% v/v Triton X- 100 for injection into a pyrolytic graphite coated graphite tube. Knowles et al. (9 1/C3627) described a new platform-tube combination for ETAAS which gave more secure position- ing of the platform and illustrated its use with the determination of Mn in normal human urines.Multiple injection procedures were used to build up sufficient Mn on the platform to give adequate sensitivity. Using calibration by standard additions concentrations in the range 0-1.6 pg 1-l were measured for 60 urines. 1.7.17. Mercury Papers on Hg reviewed this year focus principally on three problems releasing Hg from its bound form in samples increasing the sensitivity of the cold-vapour technique and reliable determination of the species MeHg. Sun et al. (92/121) showed that complete release of Hg from all its forms could be achieved with BrCl produced from the reaction between KBr03 and IU and SnCl in the presence of NH,OH-HCI. The method was used for the determination of Hg in urine. Danha and Baloun (9 1/ 1249) oxidized biological materials in O capturing the Hg on an amalgamator which was subsequently heated for determi- nation by AAS.The method was claimed to be faster and less expensive than conventional CVAAS procedures. The greater sensitivity aflorded by AFS was exploited by Vermeir et al. (92/49) for the determination of Hg in a range of biological materials. Mercury vapour was generated in a continuous-flow system coupled to a gas-liquid separator of their own design. A detection limit of 0.9 ng 1-l was obtained. Determination by AFS was also reported by Nie and Zhou (9 112389) but in a batch process. Their detection limit was 0.15 ng. Matsumoto et al. (9 1/3978) increased the sensitivity of CVAAS by using a collector of gold wires. After collection the bundle of gold wires was heated to release a pulse of Hg vapour into the detector.They adapted a double-beam spectrophotometer for this purpose using an Hg lamp and a gas flow cell in the cell compartment. Other forms of gold collectors used have been gold-platinum mesh (92/90) or gold-coated Celite (91/3180) which was formed by making a slurry from Celite with gold chloride and heating it in a muffle furnace at 600 "C. Results of an international comparison of results for the determination of MeHg in Jish muscle and mussel soft tissue were reported by Thibaud and Cossa (9 1/1346). Although the techniques varied (including CVAAS GLC with ECD detection NAA and isotope-dilution ICP-MS) similar results were found. Problems were evident in a Yugos- lav-German collaboration outlined by Horvat et al. (9 1 / 1 343).Various isolation techniques for MeHg were compared for final determination by CVAAS or GC.96R JOURNAL OF ANALYTICAL ATOMIC SPECTROMETRY APRIL 1992 VOL. 7 Although results were comparable for almost all biological and environmental samples differences were found for soils and some sediments. They then developed a distilla- tion technique that gave good recovery (mean +- SD= 95 a 2%) and reproducibility for all types of sample and was specific for MeHg. Harms (92/90) extracted MeHg from biological materials by adding 2 mol dm-3 Me4NOH and extracting with toluene. The MeHg was back-extracted into a cysteine-phosphate buffer where the Hg was complexed with diethylammonium N N-diethyldithiocarbamate and extracted into pentane. The solution was evaporated to dryness and redissolved in toluene for determination by GC-AAS. The reason for presenting all these details is to illustrate how complicated some analysts like their methods to be.A simpler approach was described by Baeyens et al. (9 1/C 1800). Methylmercury was released from tissue by H2S04 and converted into the iodide by iodoacetic acid in a closed headspace vial. The MeHgI in the headspace was determined by GC-MIP. Calibration was by standard additions. Mercury taken up in tissues may end up as the stable (and hence non-toxic) sulfide or selenide. Suetomi et al. (9 1/3 180) developed a method to determine only ionizable Hg in animal tissues by extracting with strong NaCl (2.5 mol dm-3) and 0.15 mol dm-3 H2S04. Subsequent determi- nation was by CVAAS with preconcentration on gold- coated Celite.The method was shown to give good recovery of Hg from liver spiked with HgCl but negligible recovery from liver spiked with HgS HgSe or MeHgC1. In combina- tion with methods for the determination of total Hg and total inorganic Hg it allowed calculation of the concentra- tions of the stable form of Hg and the organic form. The work of Friese et al. on calibration for Hg determina- tion with controlled introduction of Hg vapour has now been published (9 1/25 13). Defined values of a saturated Hg vapour were introduced into the sample gas flow with a special valve. Day-to-day precision of better than 2% and within-run precision of better than 1% was claimed. The method was validated with the analysis of a number of biological SRMs.1.7.18. Nickel The variability of Ni concentrations in lung tissue was highlighted in a study by Raithel and Schaller (91/3234). A total of 495 samples were taken at autopsy from different regions of lungs from 30 persons who had no history of occupational exposure to Ni. Median Ni concentrations measured by ETAAS ranged from 107 to 1 95 ng g- of dry mass with higher concentrations in the upper lung areas. Seemann et al. (91/1079) concluded on the basis of their experiments that it was possible for practical expediency to use lung tissue which had been preserved and stored in formalin. However care was still needed in specimen collection and processing. An increase in the sensitivity of ICP-AES and ETAAS determination of Ni in tissue digests was obtained by solvent extraction with 1,5-bis(di-2- pyridylmethy1ene)thiocarbonohydrazide into IBMK in a procedure outlined by Vereda Alonso et al.(91/C1794 92/ 106). The extractant allowed preconcentration factors of up to 15. Vaughan and Templeton (9 1/C 1 662) overcame problems of isobaric interferences in the determination of Ni in urine by ICP-MS through the use of principal components analysis. A value of 69 pg 1-1 was obtained for NIST SRM Urine (recommended value 70 pg 1-I). 1.7.19. Potassium and sodium Soederberg et al. (911884) studied the determination of K and Na in rat lens by FAAS. The K and Na were released from the lens by soaking in water and this was diluted with CsCl solution for measurement. Neither grinding the lens nor dissolution in HN03 gave higher concentrations indi- cating that the soaking procedure was sufficient.1.7.20. Platinum For the determination of Pt in the plasma and urine of patients on cisplatin chemotherapy Hopfer et al. (91/1138) developed a method based on ETAAS with Zeeman-effect background correction. Samples were diluted 1 + 199 with a solution containing the diammonium salt of EDTA blH4H2P04 NH40H and a detergent octoxynol and were analysed using simple aqueous standards for calibration. Cheater sensitivity is possible by using ICP-MS as Tothill et al. (9 I / 1428 9 1/C 166 1) demonstrated. Samples of blood and tissue required prior digestion with HN03 to remove organic material whereas urine samples were analysed directly after dilution. With In as an internal standard matrix effects were minimal.Comparison was made with determination by ETAAS which appeared to be more prone to interferences particularly from residual nitric acid. Fksults obtained by ICP-MS on samples of rat liver were shown to correlate well with results obtained by ETAAS. 1.7.2 1. Selenium FLobberecht et al. (Biol. Trace Elem. Res. 1990 25 149) r,eviewed procedures for the determination of Se in serum pdasma and whole blood by XRF and PIXE. Concentra- t ions found from various countries were tabulated. According to Heydorn and Griepink (91/3229) evidence from BCR certification analyses on biological materials by experienced laboratories within the EEC indicated that the most unbiased methods at concentrations less than 0.3 nng kg-' were obtained by fluorimetry whereas for higher concentrations INAA was recommended.Obviously atomic spectrometric techniques have not yet progressed to produce sufficiently reliable results for this element. For the determination of Se in biological materials by ETAAS a modifier of Pd with Mg(N03)2 has featured most in recent papers (91/C1789 91/C2736 91/3015 92/21 1) although others have used Pd alone (91/1299) Ni (91/2520 91/3775) or Pt (91/1271). Despite the claimed improve- rnent in accuracy for the determination in serum through the use of Zeeman-effect background correction Schnipper amd Jans (9 1/C1789) found that phosphate at physiological concentrations did cause a decrease in the Se signal of about 30%. Structured background interference by iron in whole blood normally means that the determination of Se by E3TAAS requires Zeeman-effect background correction.However Van Cauwenbergh et al. (92/211) described a rnethod suitable for use with instruments with deuterium- a m background correction. It was necessary to dilute samples 25-fold to remove interference from iron even when using platform atomization and a Pd-Mg(N03) modifier. The diluent included HCl to prevent coagulation of samples by the modifier. The detection limit in the original sample was 10 pg 1-1 while reproducibility at rrormal levels was better than 3% RSD. Calibration was by standard additions. The method was used to determine Se levels in the blood of donors submitting blood to the Belgian Blood Bank (Van Cauwenbergh et al. J. Trace Elem. Electrolytes Health Dis. 1990 4 2 15). Regional differences were seen with highest values for the coastal region centred on Bruges and lowest for the southern region around Namur which was presumed to be related to nutritional differences.The high concentrations of phosphorus in marine biologi- cal tissue can similarly cause problems in the determination of Se by ETAAS with continuum-source background correc-JOURNAL OF ANALYTICAL ATOMIC SPECTROMETRY APRIL 1992 VOL. 7 97R tion. Maage et al. (9113775) found that the best atomizer for overcoming the interference was an uncoated graphite tube with a pyrolytic graphite platform. Amounts of Ni of 100 pg per injection would remove interference from up to 5 pg of P. When samples of lower P content were analysed 50 pg of nickel sufficed. Samples were digested with HN03-HC104 and calibration was by standard additions.Accuracy was demonstrated by results on NIST SRM Oyster Tissue and on eight marine biological samples that had been previously analysed in intercalibration exercises. Some of the problems of Se determination by ETAAS can be avoided by using solvent extraction. Chi et al. (91/2520) determined Se in hair serum and water by extracting with APDC into IBMK. Selenium(vI) was reduced to SeIv with TiCl before extraction. A similar reagent diethylammon- ium diethyldithiocarbamate in CHC1 was used by Hoc- quellet and Candillier (9 113597) for the determination of Se in digests of animal tissues. The modifier Pt was coex- tracted with the Se. An unusual method for the determination of Se in serum based on AES was described by Szilvassy-Vamos et al.(9 1/27 18). Samples were dried onto the wall of aluminium cups which later formed the cathode in a demountable hollow cathode lamp. The intensities of the two Se ion lines 444.62 and 444.95 nm were measured using photographic detection giving a detection limit of 1 pg 1-I. Results by standard additions and direct calibration showed no signifi- cant difference indicating lack of interference. An automated microtechnique for the determination of Se in body fluids by FI hydride generation AAS was developed by Negretti de Braetter et al. (9112563). Aliquots (350 pl) of the acid-digested samples were injected into the FI system. An on-line gas-liquid separator passed the hydride to a heated quartz tube in an AA spectrometer. Accuracy was evaluated by determination of Se in RMs and by comparison of results with INAA.Arikawa and Iwasaki (9 1 /40 1 9) combusted biological samples directly in high pressure O2 in a Parr bomb containing 1 ml of water. The Se dissolved in the water and was reduced to Sew by boiling with 6 mol dm-3 HCl. Subsequent determination was by hydride generation AAS. Values obtained on RMs showed good accuracy. Selenium in pig tissue was determined by Chen and Chen (9 1/3015) by ETAAS with a Pd-Mg modifier after digestion of the sample with HN03-H202. Recoveries ranged from 92 to 112% and the RSD was 1.4%. Blais et al. (9 1/359 1) have developed a prototype system for speciation of organoselenium compounds in urine. Extracts of organoselenium compounds were separated by HPLC on a cyanopropyl stationary phase with a methanolic mobile phase. The eluant was nebulized by an on-line thermospray device O2 and H2 were added for combustion and the Se detected by AAS using an unheated quartz T- tube in which the H2-02 mixture burned.On the basis of their trapping experiments they concluded that hydrogen radicals in the initial combustion initiated the production of hydrogen selenide which was then broken down in the H2-02 flame. The method could detect selenoniocholine and trimethylselenonium cations in urine spiked with these compounds but further development of the extraction process was considered necessary to measure natural levels. As an alternative to blood testing of cattle for Se status Lean et al. (9 1 ~ 2 5 6 ) investigated whether measurement of bulk tank milk Se was suitable for assessing the herd as a whole.The concentrations found were directly proportional to the mean herd blood Se levels and the mean herd milk levels. Within a herd milk Se concentrations were directly related to the cow's blood Se concentration. Measurements were by hydride generation ICP-AES and concentrations found were in the range 13-42 pg 1-l. Further studies have been made to understand the role of Se in human metabolism. Normal levels of Se in the thyroid gland are surprisingly high according to data obtained by Aaseth et al. (9112521). The mean normal concentration was 0.72 ~f 0.44 pg g-l compared with 0.45 k 0.1 1 pg g-l in the liver. Measurements were by HGAAS. The high concentration suggested that Se has important functions in the thyroid gland.In an attempt to understand the reduced levels of Se found in primary biliary cirrhosis Valimaki et al. (911321 7) supplemented patients and controls with Se- rich yeast. Serum Se increased in both groups but the difference between them remained. They concluded that impaired hepatic production of Se compounds present in serum was responsible for reduced serum Se concentra- tions. Previous papers have shown the relevance of Se to male fertility. However Roy et al. (9 1 /26 1 7) in an extensive study using HGAAS found no significant correlation of any kind between the Se level in seminal plasma and sperm count or motility. 1.7.22. Silicon Interest in Si in biological fluids is increasing. Roberts and Williams (91/263 1) found increased concentrations of Si in the serum of patients with renal failure and even higher levels when such patients were on haemodialysis.Concentrations were measured by DCP-AES after dilution of serum or urine with 1% v/v HNO,. The method was demonstrated to be free of interference to be accurate (recoveries of 95-105% in serum and 96-99% in urine) to be linear up to 1000 pmol 1-l and to be adequately sensitive (detection limit = 2.0 pmol 1-l). Gitelman and Alderman (9 1/27 13) found that disposable syringes and Vacutainer tubes gave contamination and described a cleaning procedure for syringes to remove silicone coating. Silicone rubber components in the graphite furnace used for their measurements were replaced with alternative materials. 1.7.23. Silver Thunus and Dauphin (91/2470) determined Ag in rat plasma by ETAAS using a diluent containing Triton X- 100 and an antifoaming agent.Concentrations of greater than 1 pg 1-l could be measured. 1.7.24. Strontium Vandecasteele and co-workers (9 1/22 19) compared mea- surements of Sr in the Versieck human serum reference material by ICP-MS and NAA after radiochemical separa- tion. With ICP-MS In was used as an internal standard to compensate for matrix effects. Results obtained after 5- and 10-fold dilution using both S4Sr and 86Sr isotopes were in good agreement. Results showed better precision (3%) than those by NAA ( looh) and were significantly higher. 1.7.25. Thallium Using a method based on solvent extraction followed by ETAAS Chandler et al. (91/2941) reported T1 levels in whole blood serum and urine of a patient found to be suffering from TI poisoning.The initial concentrations were so high ( 1 pmol 1-l ) that they could be determined by a solvent extraction-FAAS procedure. Thallium levels in blood and serum decreased with a half-life of about 4 d. However the data on urine excretion could not be fitted to a descriptive pharmacokinetics curve.98R JOURNAL OF ANALYTICAL ATOMIC SPECTROMETRY APRIL 1992 VOL. 7 1.7.26. Tin Tin concentrations in blood were determined by Nakazaki and Shiroishi (9 1 /4038) using ETAAS. Samples were digested taken to dryness redissolved in 6 mol dm-3 HCl and extracted as a chloride complex into IBMK. The Sn back-extracted into 0.5 mol dm-3 HCl was injected into the furnace. A detection limit of 2.2 pg 1-l was reported. Another Japanese group (9 1 /3 1 50) preferred the more sensitive combination of HG and ICP-AES for which they developed an improved gas-liquid separator. Biological samples were first digested with HNO3-HC1O4 and calibra- tion was by standard additions.Their detection limit was 30 ng 1-I. Li et al. (91/C3630) developed an even more sensitive system (detection limit = 7 ng 1-') using FI for HG and trapping of the hydride in a graphite furnace. They demonstrated the accuracy of their method by determining Sn in a number of SRMs including Bovine Liver. Studies of organotins in marine life continue using combinations of GC and various atomic spectrometric techniques. Harrison and Rapsomanikis (9 I/ 1362) ex- tracted butyltin from oyster tissues with 2 mol dm-3 HC1 generated the hydrides which were collected then thermally desorbed into a gas chromatograph and the Sn detected by quartz furnace AAS.Tributyltin was the principal compo- nent in oysters from a UK coastal site. For a similar technique the efficiency of leaching procedures for extract- ing the organotins was studied by Desauziers et al. (9 1/994). The most efficient extractant was found to be cold pure acetic acid for a period of 4 h. Krull and co-workers (911995) preferred to extract with an organic solvent to allow the speciation of the original organotin species with a combination of GC with either a flame photometric detector or a DCP-AE spectrometer. 1.7.27. Uranides Applications of ICP-MS to the measurement of long-lived radionuclides in environmental and biological samples were reviewed by Igarashi et al.(9 1/225 1). Simultaneous deter- mination of 232Th and 238U in a range of biological RMs was described using T1 or Bi as an internal standard. Further papers from the same group described the use of isotope dilution with 230Th for the determination of Th in a range of biological CRMs (91/C1681) and the determination of Th and U in bone ash (91/3293). Samples were digested in HN03-HF-H202 evaporated to dryness and dissolved in dilute HN03. Accuracy was verified by a satisfactory determination of Th and U in NIST SRM Human Lung. Workers at NIST (90/26 17) described isotope dilution determination of U and Th in NIST SRM Oyster Tissue by SIMS. Comparison was made with determination by TIMS.1.7.28. Vanadium A direct method for the determination of V in urine by ETAAS was developed by Paschal and Bailey (91/3101). Urine was diluted with 2% v/v HN03 and 0.001% v/v Triton X-1 00. Matrix effects were absent allowing the use of simple aqueous standards for calibration. The detection limit was 1.5 pg 1-I. For determination of V in serum by ETAAS using platform atomization Navarro et af. (9 1K1950) diluted samples with a modifier containing Pd citric acid Triton X-100 and HN03. Oxygen ashing was incorporated into the furnace programme. Elevated concen- trations of V were found in the serum of patients with renal failure. 1.7.29. Zinc The homeostatic regulation of Zn in the human body was studied by Taylor et al. (92136) using 70Zn as a marker and measurements by TIMS.Reduction of the dietary intake of four male subjects from 85 pmol to 12 pmol per day led to a mean reduction of 48% in urinary Zn excretion and 46% in faecal Zn excretion. The efficiency of Zn absorption had risen from a mean of 38 to 93%. Zinc concentrations in stools were found to be fairly hlomogeneous with a variability of (13% in a study by Dastych (9 1/3504). This method involved determination by FAAS after digestion with H2S04-HC104. The mean concentration for healthy subjects was 408 pg g- I . The physiologically active part of serum Zn was deter- mined by ultrafiltration and ETAAS in a method described by Faure et al. (91/2629). In 20 controls the mean ultrafiltrable Zn was 0.31 pmol dm-3 and the albumin- bound Zn 12.1 pmol dm-3. The procedure was claimed to be simple and rapid requiring only a small volume of serum.A mixed solvent system H20-HCl-EtOH-butan-2-0ne was used by Anwar el al. (9 113392) to dissolve pharmaceuti- cal preparations for the determination of Zn by FAAS. 1.8. Conclusions One of the most noticeable features of this review year has been the extensive work carried out in Italy on the establishment of reference ranges. The comprehensive work by Minoia et al. (91/2405) on normal values for urine bllood and serum is most impressive and will undoubtedly ble widely cited in many future publications on clinical analysis. Coni et al. (9 1/879,91/C2879) produced reference ranges for human milk Sabbioni et al. (92/78) for human lung tissue and Senofonte et al. (91/887) for children's hair.All have provided valuable data. Biotechnology seems to be making its mark on analytical chemistry. In recent years the use of algae to preconcen- trate trace elements has featured. Work covered in this review shows applications to Cu (9 1 / 1 5 50 9 1 /C 1 908) and Cr (9 1/22 17). A novel approach has been used by Neidhart et al. (91/3153) for speciation of Cr. They elcploited the property of human erythrocytes selectively to take up Crw in the presence of CrlI1. Some events in clinical chemistry are so rare that when they do occur detailed investigation and publication is a nnust. Such a case is the T1 poisoning reported by Chandler et al. (9112961) which allowed them to exploit their previously developed expertise with this element to the full. At the time they reported the reason why this happened in the first place unfortunately had not been found.Such is the awareness now of validation of methods with CRMs that it is almost obligatory for analytical publica- tions to include some data on the analysis of CRMs. This is an important step forward. What is becoming disturbing is the number of papers where only CRMs are analysed and a claim is made that the method is suitable for analysis of those materials. Real samples are often very different from CRMs. In some CRMs the trace element concentrations are much higher than those found naturally but more importantly the physical nature of real samples often differs greatly from the neat bottled materials available as CRMs. For example blood RMs are generally lysed to rnake them homogeneous and they have the consistency amd appearance of red ink. Real blood samples need careful rnixing and are more difficult to handle in pipetting operations.The biggest differences are in the solid RMs which are supplied dried and pulverized to a small particle size to make them homogeneous. How different it is to weigh and dissolve a powder from a bottle labelled Bovine Liver CRM than to tackle a piece of soft tissue oozing blood which is the way real samples of bovine liver appear in laboratories.JOURNAL OF ANALYTICAL ATOMIC SPECTROMETRY APRIL 1992 VOL. 7 99R 2. ANALYSIS OF FOODS AND BEVERAGES Simon Branch and Helen M. Crews A summary of the published and conference papers covered by this review is given in Table 2. More papers are reviewed than in the previous Atomic Spectrometry Update (9 1/3594) for foods and beverages reflecting the increased awareness of and interest in food composition.The format of the review of foods and beverages has been changed. The sample preparation and sample introduction sections have been combined as in the clinical and biological materials part of this review. There is no section for specific developments in FAAS this year. Much of the work in this area has been associated with analyte enrichment and preconcentration techniques and is therefore discussed under sampling and sample preparation. 2.1. Sampling and Sample Preparation 2.1.1. Preconcentration The use of preconcentration techniques for the analysis of waters has been reported for several elements.Cadmium was determined in drinking waters (9 1 / 1 302) after enrich- ment and separation of the Cd on a column of an adsorbent resin impregnated with diphenylthiocarbazone-IBMK. An- dreeva and Drogobuzhskaya (9 111 593) determined As Cr Mo V and W in waters after percolating 0.5 1 samples through a column filled with polyacrylonitrile modified with polyethylene polyamine. They found that although AES was overall the least precise method of analysis when compared with XRF and ETAAS Mo could not be measured by XRF and ETAAS could not be used for Cr and W. The over-all limit of detection was cited as approximately 10 pg 1-l. Ultra-trace levels of Ni were determined in tap water by atom-trapping FAAS after elution of samples (2 1) from a strong cation-exchange resin by 3 mol dm-3 HC1(91/3341). There were no interferences from Ca2+ K+ Mg2+ or Na+ and the use of ion exchange gave a 1 2800-fold increase in sensitivity.Determinations were possible at ng kg-l levels. Other papers reported the use of FAAS for measuring Cd Co Fe Mn Pb and Zn after coprecipitation with Zr(OH) (9 1/3524) and for the deter- mination of Ag Pb and Zn after concentration on Silo- chrome SG-80 modified with 2-aminothiazole (9 1 /3574). Two conference presentations described on-line precon- centration for FAAS. Stresko et al. (9 l/C2768) investigated the parameters influencing the use of the chelating resin Chelex 100 for the determination of Cd Cu and Pb in surface and mineral waters. For a 5 ml sample the use of the resin increased sensitivity between 45- and 80-fold.Schlemmer et al. (9 l/C366 1) described on-line preconcen- tration combined with FI to improve the sensitivity and detection limits of FAAS. They found that the formation of metal-DDC complexes of the analytes of interest on bonded silica gels with C,* functional groups was preferable to ion exchange because monovalent cations were not chelated and absorbed. This enabled the analytes (Cd Cr Cu Ni and Pb) to be separated from the major matrix components such as NaCl. Lead was determined (9 1/3779) in biological samples by FAAS using an FI system with on-line coprecipitation. Samples (2 ml of whole blood or 1 g of dried biological material) were digested with HN03 in a microwave diges- tion system. The clear digests were transferred into diges- tion tubes 100 pl of HClO added and the mixture heated and then evaporated to near dryness.The pH was adjusted to 2-3 and the residues made up to 10 ml. Aliquots (2.5 ml) of the sample were used to coprecipitate Pb with the iron- hexahydroazepinium hexahydroazepin- 1 -ylformate (hex- amethyleneammonium hexamethylenedithiocarbamate) complex and the precipitate dissolved on-line with IBMK. An enrichment factor of 20 was obtained with a detection limit in the digest of 2 pg 1-l. The sample throughput was 90 h-? Three different enrichment techniques were tested for the determination of Pb in water by FAAS (911989). Acidified water samples (1 1) were either evaporated to 50 ml prior to analysis or 250 ml of the sample were extracted with NaDDC and IBMK and the organic phase analysed.The third alternative was a combination of evaporation and extraction and this was the preferred method. It gave a concentration factor of 100. Solvent extraction has also been used to determine Mn (911321 1) and Cd Cu Pb and Zn in water (91/3412) Co in feed grains and forages (91/833) and Co Cu Mo and Se in NIST SRM 1577a Bovine Liver (91/3777). This last paper reports that the same digestion and extraction procedure can be used prior to measuring Cu by FAAS and Co Mo and Se by ETAAS enabling ten samples per day to be analysed by one operator. Minimum sample handling and the prevention of contam- ination were given priority in a method for the direct determination of Cr in Danish milk products and cheese (92/335) by ETAAS. Homogenous samples of milk and milk products were injected into the graphite furnace ashed in O2 at 650 "C and then further ashed at 1 100 "C under Ar.Atomization was at 2300 "C with Zeeman-effect back- ground correction. The detection limit was 0.7 ng g-l. Heterogenous and solid samples such as cheeses were ashed with HN03 under pressure before analysis. This method was also used for the determination of Cd and Pb. Minimum sample handling was achieved by Norheim (9 1/1130) by using the same modified Tecator tube during the digestion dilution and analysis of food samples. 2.1.2. Digestion A closed vessel digestion procedure was evaluated as a method for plant analysis (9 1/33 19). The samples were heated in an oven (up to 1 10 "C) in a closed Teflon container with HN03 overnight. When compared with results obtained after a standard HN03-HC104 digestion there were significant differences in the values obtained for Al Cu Fe and Mg.For NIST SRMs the closed vessel method gave lower results for A1 and Fe only. Other workers (9 1/ 1282) have used a closed vessel microwave digestion technique prior to the measure- ment of ten elements by ICP-AES. Depending upon the sample type 0.1-0.2 g was digested with either HN03-H202 or HN03-H202-HF in a PTFE digestion vessel using microwave heating for 2-3 min at 500 W ofmicrowave power. Undiluted or diluted samples were measured. Good agree- ment was found with certified values. Microwave-assisted acid digestion was also used as part of the sample preparation procedure for the analysis of eight CRMs by PIXE (92/74).For elements heavier than Kand for~oncentrationsof2pgg-~ upwards the total random error for a single analysis was in the range 2-5% whilst the accuracy was better than 5%. The lowest detection limit was 0.3 pg g-l. Three conference presentations described various aspects of microwave sample preparation. Hasty et al. (9 1 /C3692) discussed the optimization of closed vessel microwave digestion by the use of parameter feedback control. The use of flow-through microwave techniques was described by Barnes and Martines (9 1/C369 1). Continuous or stopped- flow on-line microwave digestion should increase sample throughput and extend the automation of sample handling. A robotics system for use with both microwave and traditional digestion methods was introduced by Grillo et al.(91/C3672). The system included an articulated arm1 OOR JOURNAL OF ANAL,YTICAL ATOMIC SPECTROMETRY APRIL 1992 VOL. 7 robot to process the sample by dispensing weighing reagent addition and digestion. 2.1.3. Solid sampling The analysis of several foods for Cd and Pb by solid sampling was reported by Fecher and Malcherek (92/84). Solid-phase analysis was achieved using direct heating over dual graphite electrodes with deuterium-arc background correction and Pd(N03)2-Mg(N03)2 as chemical modifier. The method worked well for grain meals powdered milk cocoa chocolate and dough with RSDs of ~ 1 0 % . The method was not useful for meat and heterogeneous foods such as oilseeds. In contrast Luecker et al. (92183) used Zeeman-effect background corrected ETAAS with solid sampling for the determination of Cd and Pb in liver and kidney samples.They found that for liver there was no difference in Pb concentrations at the higher concentrations (5- 1 2 ng g-I) when compared with results where sample decomposition was used. At lower concentrations (GO. 1 ng mg-l) solid sampling gave generally higher Pb values. For Cd in liver and Cd and Pb in kidney both solid sampling and decomposition methods gave equivalent results. Signal summation was applied to the determination of Cd and Pb in malt and yeast by direct solid analysis (91/2506). The resulting improvement in the S/N made it possible to determine concentrations at levels where it would previously have been necessary to resort to a preliminary chemical concentration technique. Lucker et al.(91/C1840) discussed the use of solid sampling with ETAAS as a method to improve analytical quality control in the production of animal tissue CRMs. By using solid sampling some of the analytical errors which can occur during sample preparation should be eliminated. Sample preparation is not entirely eliminated when suspensions or slurries of samples are analysed. However preparation is minimized and thus the risk of contamina- tion at this stage should also be minimal. The production of samples of suitable particle size to form a slurry may involve charring and grinding with the associated risk of contamination. However chemically prepared slurries (9 1/868) have the advantage of simplicity and effectiveness in reaching a particle size of below 10 pm. The analysis of carbonaceous slurries (9 1/868) produced by warming NIST SRMs with concentrated H2S04 was reported.Eight ele- ments were determined by ICP-AES and RSDs were often less than 5%. De Andrade et al. (91/1584) used calibration graphs obtained with white bean homogenates as standards for the determination of Fe and Zn in foods by slurry nebulization with F'AAS. The foods included grains vegetables fruits and sausage. Homogenization of semi-prepared samples to form slurries took 4 min. The slurries were introduced into the spray chamber by a single-line FI system which permitted 120 injections per hour; carry-over and memory effects were negligible. Detection limits were reported as 0.6 for Fe and 0.3 pg ml-I for Zn. Aqueous standards were used by Lynch and Littlejohn (91/2220) for the determination of Cd in slurries of milk powder liver and olive leaves by ETAAS.Chemical modifi- ers were compared and whilst both Pd and NH4H2P04 stabilized the Cd to a similar extent the latter increased the background signal. An analytical procedure based on Pd as chemical modifier and platform atomization with a pre- atomization cooling step was developed. This method allowed aqueous calibration up to concentrations of 50 mg ml-l of Cd in the slurries. The accuracy was within 15% and the LOD for Cd in the analysis of a 50 mg ml-1 slurry was 10 ng g-l. Manganese Pb and T1 were measured by either ETA- LEAFS or ETAAS in slurried and dissolved samples of three NIST SRMs (91/3275). The LOD for T1 by the former technique was 1-2 orders of magnitude below that for the latter.Precision was similar for all the analytes by both techniques with ETA-LEAFS giving relatively small back- ground signals when compared with ETAAS. Data from both dj ssolved and slurried samples agreed with certified values. 2.2. Developments in Hydride Generation Techniques Solid food samples were analysed as slurries for Pb using H'G in a lactic acid-K,CrO medium (92/41). Lactic acid proved to be the most suitable of the acids tested for the generation of lead hydride stabilizing the Pbw compounds farmed as intermediates and producing a rapid reaction and thus high and sharp Pb peaks. Interference by Cu could be masked with the use of oxalic acid or by using the method of standard additions. Detection limits were 0.04 ,MI; g-l for fish and 0.1 ,ug g-l for vegetables.The proposed method gave satisfactory results for tap water (LOD=5 ng ml-I) but ethanol interfered with wine analyses. There- fore wine samples were treated with 0.5% HNO before being analysed by the method of standard additions (LOD = 20 ng ml-I). Vujicic and Steffan (9 1 / 103) described a new continuous hydride generator for ICP-AES. The generator allows sample uptake of 1 ml min-l without loss of sensitivity. The small volume of the reactor and neutralization of the reaction mixture with NaOH reduce the memory effect. The use of NaOH also prevents the generation of excess of H2 thus eliminating its suppressive effect on the plasma. Detection limits were 0.1-0.3 ng ml-1 far As Bi Hg Sb Se and Sn.Arsenic was determined in grain and soils by a hydride non-dispersive A F method (9112553). The AF method was found to be rapid sensitive and low in interferences. Pireconcentration of the generated hydride was used to determine the level of As in Dutch milk powder (91/2670). After combustion of milk powder samples with O2 in a mall quartz ashing chamber volatile As was condensed on a cold finger cooled with liquid N2 and subsequently stripped by refluxing with a small volume of ultrapure HCl. The AsV was reduced by borohydride and the arsine trapped at liquid N2 temperature released by heating and then swept into the measuring cell. Arenas et al. (92/82) reported automation of an HGAAS system with preconcentration for the determination of As in biological samples.The auto- mated routine was found to be faster and more sensitive than a standard procedure. 2.3. Speciation Studies The speciation of trace metals using HPLC-ICP-MS was reviewed at the Euroanalysis VII conference by Ebdon et al. (9 1 /C 1692). Examples were given of the determination of tributyltin and As species in waters and some foods. For Sn species a strong cation-exchange (Partisil 10) column was used whilst As species were separated with an anion- exchange resin (Benson 7- 10 pm). In an interesting paper Blais et al. (9 111 57) determined arsenobetaine arsenocho- line and tetramethylarsonium cations using LC with ther- mochemical HGAAS. Arsonium species were separated by HPLC on a cyanopropyl-bonded silica column with a C1H30H-diethyl ether mobile phase.The column eluate was nebulized and pyrolysed in a CH30H-02 flame; the As compounds were then converted thermochemically in a calol-flame atomizer into ASH for detection. Blais et al. recommend the use of a full-face shield until the operator is familiar with the system cautioning the reader '. . . at lower he:ating element temperature (or insufficient heating of the thermospray tube) an accumulation of CH30H in the interface due to an unsuccessful thermospray ignition caused noisy explosions in the optical tube; however theJOURNAL OF ANALYTICAL ATOMIC SPECTROMETRY APRIL 1992 VOL. 7 lOlR Table 2 SUMMARY OF ANALYSES OF FOODS AND BEVERAGES Technique; atomization; analyte form* AA;-;L Element Matrix Ag Water Sample treatmentkomments Reference 91/3574 Ag Pb and Zn were determined by AAS following preconcentration on Silochrome SG-80 modified with 2-aminothiazole.LODs for Ag Pb and Zn were 0.05 0.03 and 0.1 mg 1-I respectively A1 Milk AA,ETA;L Milk samples were diluted 1 + 1 with a solution containing 2Oh Triton X-100 and 0.025% Mg(NO,) prior to A1 determination by Zeeman corrected ETAAS. LOD was 2-5 pg 1-I (in Italian) A1 was determined in a wide range of foodstuffs. High fat samples were extracted with petroleum ether and A1 back extracted with 0.1 mol dm-3 HNO,. The concentrations and additives were discussed Three wet digestion methods were assessed high pressure microwave digestion using HNO the commonly used HNO - HClO digestion and the latter using an HF- HNO pre-treatment. Highest recoveries were found for the method involving the pre-treatment step 9 11922 A1 Foods AA; ETA;L 91/965 A1 Food AE;ICP;L 9 l/C1822 A1 Infant formulae A A; E TA ;- A1 was determined in 307 milk replacement formulae from 14 countries.The mean A1 concentration was 1.4 mg kg-I.In soybean based milk substitutes the global mean was 18.4 mg kg-l 9 112430 A1 Infant formulae AAiETAi- A1 was determined in 282 cans of Canadian infant 9 1 /2449 formulae. Concenlrations varied according to brand dairy or soybean product and whether the product was ready-to-use or a powder A1 Potable water A1 Milk AA ; ETA; L AA;-;L 91/3497 91/3520 Sonication was found to increase recoveries of A1 from potable water due to the release of colloidally bound A1 (in French) AA data in combination with formation constants were used to construct computer generated speciation models for A1 in milk. The major A1 species present were found to be charged citrate complexes A1 Foods AA;ETA;L Samples were digested in 5 + 1 HNO - HClO and the resulting residue dissolved in 0.5% Mg(NO,),-0.3% HNO,.A1 was determined by STPF ETAAS at 237.5 nm. LOD was 0.24 ng A digestion procedure for determining As and Pb in vegetable matter was described (in Japanese) Various approaches to using coupled HPLC- ICP-MS for speciating As and Sn in foods were described The optimum pH of the medium reducing agent concentration and flow of KBH carrier gas flow rate and atomization temperature were determined for the hydride non-dispersive AFS measurement of As 9212 1 3 9 111033 9 1 /C 1692 91/2553 As Vegetables As Foods As Grains AA;ETA;L MS;ICP;L AFHy;L As Milk powder milk AA; H y ; G Milk powder or freeze-dried milk were combusted with 0 in a quartz ashing chamber. Gaseous As was condensed on a cold finger stripped by refluxing with HCl reduced by tetrahydroborate further trapped by liquid N and finally swept by He and H carrier gases into a measurement cell.Milk contained approximately 0.3 pg kg-I of As 9 1/2670 As Tinned mussels AE;ICP;L As was determined at 193.7 nm by ICP-AES 9 K 2 9 2 1 following sample mineralization at 450 "C with Mg(NO,) and MgO. Detection limit was 0.1 mg kg-102R JOURNAL OF ANALYTICAL ATOMIC SPECTROMETRY APRIL 1992 VOL. 7 Table 2 SUMMARY OF ANALYSES OF FOODS AND BEVERAGES--continued Element Matrix As Foods B Biological materials B Plants Bi Water Ca Milk Ca Food stuffs Ca Milk powder Ca Cheese Cd Milk Cd Liver wheatflour Technique; atomization; analyte form* AA;Hy;- AE;ICP;L AE;ICP;L AA ; ETA,L AA;F N,O- C,H;L AA;-;L XRF;-;S AA;-;L AA;ETA;L AA;-;L Cd Biological reference materials AE;ETA;L Cd Ginseng Cd Vegetables berries Cd Bovine liver AA; ET A; L AA;ETA;L AA; ETA$ Sample treatmentlcomments Reference 9113529 A 7 day duplicate diet study was undertaken and As speciated in foods urine and faeces from Japanese adults An open vessel wet ash low temperature (< 140 "C) PTFE tube digestion procedure followed by ICP- AES was described for the measurement of B and other mineral elements 9111051 ICE'-AES based on a linear self-scanning photodiode 9 113323 array was utilized for the detection of B in plants.Detection limits of 10- 15 p g 1-' were possible A method for the coprecipitation of Bill1 with hafnium hydroxide was outlined. Bi was measured at 223.1 nm by ETAAS using the standard additions technique Free Ca total Ca and total C1- were determined using an automated three component flow injection procedure. Samples are directed by 2 dialysers to 3 channels; free Ca is detected by spectrophotometry at 580 nm total Ca by AAS at 422.7 nm and Cl- by a coated tubular C1- selective electrode 9111119 9111514 AAS and KMnO titration methodology were Ca and K were quantified in milk powder by compared for the determination of Ca in foods EDXRF. LODs for Ca and K were 0.007 and 0.0 1 1 % respectively A collaborative study of methods for Ca Mg and P in cheese was described.Ca and Mg were determined by AAS and P by colorimetry. The method was adopted official first action by the AOAC A combination of 6 pg of Pd with 500 pg of NH,NO was proposed as a suitable chemical modifier. The Cd detection limit in milk was calculated to be 0.5 pg 1-1 Results of collaborative studies indicated that the pressurized degradation of samples with HNO for Cd and Pb determination is equivalent to non- pressurized digestion with H202 - H,SO - HNO (in German) Cd and Pb were determined in biological CRMs at 228.8 and 217 nm respectively by furnace atomization plasma emission spectrometry using the additions method. LODs for Cd and Pb of 68 ng g-l and 3.1 pg g-I were computed Samples were heated at 75 OC for 8 - 10 h pulverized digested with HNO and HClO dried redissolved with HNO solution then measured for Cd and Pb by platform ETAAS.Mixed Pd-La was the chosen chemical modifier (in Chinese) 9 113430 9113818 9 113843 9 11846 91/1300 9111432 9111573 Berries and vegetables from Finland and elsewhere 9 1lC1709 were surveyed for their Cd and Pb content. The results confirmed previous findings that these foods contribute little to Cd and Pb intake in Finland strategy for use in the certification of liver RMs was related The philosophy behind adopting a solid sampling 9 1 x 1 840JOURNAL OF ANALYTICAL ATOMIC SPECTROMETRY APRIL 1992 VOL. 7 103R Table 2 SUMMARY OF ANALYSES OF FOODS AND BEVERAGES-continued Technique; atomization; analyte form* Element Matrix Cd Foods Sample treatmentlcomments Reference 9 112220 AA;ETA;Sl Treated samples (0.1-0.5 g) were diluted to 10 or 20 ml with H,O modified with NH,H,PO or Pd(NO,) and Antifoam B added.Concentrated NH solution was added to give a 5Oh vlv slurry. After 15 min shaking 25 pl aliquots were analysed using Zeeman-corrected platform ETAAS Cd Foods AA;ETA;L Two methods for Cd and Pb measurement using Zeeman ETAAS were described. The difference between methods was the nature of the digestion procedures bomb digestion or dry ashing. Detection limits for Cd and Pb were 0.5- 1 and 5 - 20 pg kg-I respectively The concentration and distribution of Cd and Pb in fresh Spanish asparagus was reported The merits of time-resolved signal processing for solid samples were illustrated with application to food samples ashing stage of Cd and Pb determinations (in Japanese) Preserved eggs were digested with H,SO,-HNO and then mixed with a solution containing KI and 20% H,PO,.IBMK was used to extract the oxide complexes of Cd Cu and Pb. Good recoveries were reported (in Chinese) A survey of 2 17 samples demonstrated that under similar growing conditions berries accumulated more Cd and Pb than other seed containing fruits (in Polish) Intakes of Cd and Pb from pulses and cereals were calculated and found to be below the recommended tolerance levels of the WHO. Maximum Pb concentrations (mean 452 pg kg-l) were in pulses and maximum concentrations of Cd in cereals (mean 75 pg kg-l) Cd and Pb were assayed in foods by solid-phase AAS using direct heating over dual graphite electrodes background corrected with a high-intensity deuterium lamp and Pd - Mg(NO,) chemical modifier (in German) Experimental methodology for determining Cd Cr and Pb in Danish milk products was detailed.The merits of various in situ ashing strategies e.g. 0 pressure were assessed furnace and complexed with 2-nitroso- 1 -naphthol (20 g 1-' in glacial acetic acid). Following solvent extraction Co was determined by ETAAS pericarps of cereal and grains. Concentration varied with area of origin and variety. Generally greater than 50% was bioavailable 0 gas (0.5 1 min-' for 20 s) was added during the Samples (5 - 10 g) were ashed in a 600 "C mume c Cr was found to be mostly concentrated in the See Cd ref. 921335 Cu levels in edible oils were evaluated by diluting samples with DMF and IBMK prior to ETAAS.The LOD was 5 pg kg-i (in Italian) Vegetable oil or margarine samples (5 g) were homogenized at 50 "C with 2 ml of HNO (65%); 8 ml of H,O were then added with further homogenization. Cu and Ni in the aqueous phase were determined by ETAAS with LODs of 4 and 30 pg kg-I respectively (in Italian) 9 112448 Cd Cd Asparagus Malt yeast AA;-;- AA;-;S 9 112450 9112506 Cd Milk orange juice AA;ETA;L 9112651 Cd Preserved eggs AA;F;L 9112995 Cd Berries fruits AA;-;- 9113437 Cd Cereals pulses AA;ETA;- 9 114022 Cd Foods AA;€TA,S 92184 Cd Milk products AA;ETA;L 921335 c o Forages grains AA;ET A,L 9 11833 Cr Grain and cereal products 9 11241 2 921335 9 1 I960 Cr c u Milk products Edible oils AA;ETA,L AA; ETA;L AA ; ETA; L c u Margarine oils 9111318104R JOURNAL OF ANALYTICAL ATOMIC SPECTROMETRY APRIL 1992 VOL.7 Table 2 SUMMARY OF ANALYSES OF FOODS AND BEVERAGES-continued Technique; atomization; analyte form* AE;MIP;G Element Matrix c u CRMs Sample treatmentlcomments The capabilities of a toroidal Ar and a diffuse He :MIP in combination with ETV have been studied for AES Reference 9111457 AA;ETA;L A method was established through a collaborative itrial for determining Cu Fe and Ni by ETAAS (in .Japanese) See Cd ref. 9112995 Citric acid in soda water was measured indirectly as Cu following precipitation with Cu,(PO,),. Recoveries were 90- 103% (in Chinese) A method involving liquid anion-exchange separation ,and FAAS was described for measuring Cu in lbiological samples A sludy of Eu Tm and Yb demonstrated that the first two elements were suitable markers for butterfat ,and mixtures containing butterfat.The LODs were 5 pg kg-1 or less (in German) lelectrothermal furnace and molecular absorption was proposed. Approaches for differentiating lbetween ionic and labile F- and total F- were presented. LODs were sub-mg kg-L An indirect determination of F- in water by ICP-AES combined with FI solvent extraction was reported. F- was measured as an La complex at the La I1 333.75 nm line chromatography with fluorimetric and ICP-AES (detection molecular fluorescence spectrometry in an unmodified graphite tube furnace. Front-surface illumination was used yielding an LOD of 0.3 pg of F at 358.82 nm A method for determining F as AlF using an F- 'was measured indirectly as A1F2+ by liquid F was determined in tap water by laser-excited See Cu ref.9111457 FI-AAS with slurry nebulization was used to determine Fe and Zn in foods. LODs were 0.6 ;and 0.3 mg kg-1 respectively S*Fe was used to study non-haem Fe absorption by school children By using isotope ratio ICP-MS analysis of blood samples the bioavailability of Fe from iron- fortified infant foods was evaluated A comparative study of the determination of Fe in foods was carried out using AAS and phenanthroline colorimetric methods. Both methods gave similar results See Cu ref. 9112696 Total Ge and inorganic or organic Ge in fruit juices were determined by AAS. Commercial health juices contained 15-463 mg kg-I organic Ge (in Japanese) A method involving acid digestion (HN0,- H,SO,) j'allowed by organic solvent extraction (naphthenic iicid in CHCI,) and AAS detection gave an LOD of 5 pg 1-' (in Russian) Samples were dry mineralized in 0 absorbed in an amalgamator and following heating Hg measured using a commerical Hg analyser.Results were compared with those obtained by cold vapour AAS (in Czechoslovakian) c u Fats oils 9 112696 AA;F;L AA;-;L 9112995 9211 18 c u Preserved eggs c u Soda water AA;F;L c u Crabs oysters prawns 921259 AA;ETA;L Eu F Butterfat fats Human diets 9 113463 Molecular absorption; ETA;- 9 1/C1759 F Water AE;ICP;L 9112459 Water Tap water AE;ICP;L 9 1lC29 17 9 113 1 86 Molecular fluorescence; LE;L AE;MIP;G AA;F air- C,H,;Sl Fe Fe CRMs Foods 9111457 9111584 Fe Fe Foods Infant foods MS;ICP;L MS;ICP;L 9112314 9 1/23 16 Fe Foods 9112676 Fe Ge Fats oils Health beverages AA;ETA;L AA;F N,O- C,H,;L 9 112696 9 113410 Plants Foods AA;-;L 9 1/1062 91/1249 AA;cold vapour;LJOURNAL OF ANALYTICAL ATOMIC SPECTROMETRY APRIL 1992 VOL.7 105R Table 2 SUMMARY OF ANALYSES OF FOODS AND BEVERAGES-continued Technique; atomization; analyte form* AA;cold vapour; G L Element Matrix Hi3 Biological samples Reference 9111343 Sample treatment/comments Parameters affecting the accuracy of various approaches (cold vapour AAS GC) to methylmercury determination were studied Hg Fish muscle mussel soft tissue AA;cold vapour,- MS;ICP;- An intercalibration exercise for comparing methylmercury measurement techniques was described 9 111 346 Hg Fish MS;ICP;- 9 1lC 1663 Approaches using either peak jumping or scanning Samples ( 100 - 200 mg) were analysed by AAS using were outlined a dedicated Trace Mercury Analyser 254.Detection limit was 1 pg kg-l (in Czechoslovakian) proteins using NaCl in a solution of high ionic strength. Vaporization amalgamation and AAS determination followed. LOD was 0.6 ng 9112663 9 1/3 180 Toxic (ionizable) Hg was released from tissue Hi2 Sugar confectionary AA;-;G Hg Animal tissue AA;ETA;L Hg Water AA; ETA L Hg was solvent extracted and 400 pl of the extract injected directly into the cuvette for Zeeman- corrected ETAAS. A characteristic concentration of 0.8 ng 1-' was reached Hg was preconcentrated from water by passage through a column of powdered Cu or by electrodeposition on a Cu cathode. After concentration Hg was desorbed thermally and determined by AAS See Ca ref.9 1/38 18 The sample was concentrated by passing through a cation-exchange resin eluted evaporated to dryness and redissolved with dilute HCI. Interferences reproducibility and recovery were investigated (in Spanish) See Ca ref. 9113843 Solvent extraction was combined with one-step LEI spectrometry for quantifying trace amounts of Mn in water. The LEI signal was measured at 279.5 nm giving an LOD of 90 ng I-' ETAAS and ETA LEAFS were compared for the determination of Mn Pb and T1 in RMs 9 1/3309 Hg Water AA;cold vapour;L 91/3561 K Milk powder Li Drinking water XFR;-;S AA;F;L 9113818 92/232 Mg Cheese Mn Water AA;-;L LE1;F;L 9 113843 9 1/32 1 1 Mn Foods LEAFETA; L s1 AA;ETA; L SI AA;ETA;L AA;F air- C,H,;Sl 91/3275 Ni Margarine oils Ni Water See Cu ref.9111318 Ni was determined at 232 nm following preconcentration on a strong cation-exchange resin column. Detection at levels of 3.5 ng I-' was possible (in Chinese) Phytic acid was determined indirectly by ICP-AES measurement of P. The method was simpler faster and more accurate than other procedures Three methods were studied for preconcentration of samples prior to Pb determination by FAAS at 283.3 nm See As ref. 9111033 See Cd ref. 9 11 1 300 See Cd ref. 9111432 See Cd ref. 9111573 See Cd ref. 91/C1709 See Cd ref. 91/C1840 See Cd ref. 91/2448 See Cd ref. 9112450 See Cd ref. 9112506 See Cd ref. 91/2651 See Cd ref. 91/2995 91/1318 9113341 P Cereals AE;ICP;L 9 113844 Pb Spring water tap water AA;FL 9 1/989 Pb Pb Pb Pb Pb Pb Pb Pb Pb Pb Pb Vegetables Liver wheatflour Biological reference materials Ginseng Vegetables berries Bovine liver Foods Asparagus Malt yeast Milk orange juice Preserved eggs AA;ETA;L AE;ETA;L AA;ETA;L AA;ETA;L AA; ETA$ AA; ETA;L AA;-;L AAi-i- AA;-;S AA ; ETA; L AA;FL 91/1033 9 1/ 1300 9111432 9111573 91/C1709 91/C1840 9 112448 91/2450 9 1 12506 9 1/265 1 9 1 I2995106R JOURNAL OF ANA.LYTICAL ATOMIC SPECTROMETRY APRIL 1992 VOL.7 Table 2 SUMMARY OF ANALYSES OF FOODS AND BEVERAGES--continued Technique; atomization; analyte form* Reference 9113275 Sample treatmentlcomments Element Matrix Pb Foods LEAF;ETA See Mg ref. 9113275 L s1 AA;ETA; L S1 9 113437 9113574 9 113779 Pb Berries fruit Pb Water Pb Bovine liver AA;-;- AA;-;L AA;FL See Cd ref. 9113437 See Ag ref. 9113574 A procedure for the coprecipitation of Pb in the presence of high Fe concentrations was adapted to on-line preconcentration using FI for FAAS.A detection limit of 2 pg 1-I in the sample solution was obtained See Cd ref. 9114022 Optimum conditions for PbH generation were found to be 2% (vlv) lactic acid 0.3% (rnlv) K,Cr,O and 4% (rnlv) NaBH,. Oxalic acid was used to mask Cu which interfered seriously See Cd ref. 92/84 See Cd ref. 921335 Sul fate-S was extracted with HOAc trapped on a weak anion-exchange resin eluted with 1 mol dm- HCl and detected by ICP-AES at 180.731 nm Sample preparation and experimental conditions for HG-ICP-AES determination of Se as Seiv in foods were described The influence of oxidation state chemical modifer and thermal pretreatment were critically assessed (in French) The merit of Se testing in bulk milk tank samples rather than Se testing in blood for evaluating herd Se status was evaluated using HG-ICP-AES Soils in Finland are fertilized with Na,SeO,.The study showed that this led to elevated levels in meat and organ meats but not fish or wild animals Sei\' was determined by FI on-line anion-exchange preconcentration HGAAS at 196.0 nm. The LOD was 2 ng 1-I HNO - HC10,- H,SO diluted mixed with 10% K3Fe(CN) diluted to a known final volume and a sub-portion analysed by HGAAS. The protocol yielded an LOD of 2 ng (in Chinese) Samples were wet oxidized Se and added Pb extracted by DDDC in CHCl and Se in the organic phase determined by ETAAS. The merit of microwave digestion was discussed Ato'mizer design and choice of chemical modifer were assessed for overcoming phosphate background signal in the ETAAS determination of Se Se was determined in cereals and bakery products by Zeeman-effect ETAAS using Pd - Mg(NO,) modifier (in German) Flow was slurried with H,O and triethanolamine heated on a water bath for 3 min and then Se determined by platform ETAAS using Pd as chemical modifier at 196.0 nm (in Chinese) Powdered foods (0.5 -4.0 g) were digested with AAiETAi- AA; H y ; L 9 1 I4022 9214 1 Pb Pb Cereals pulses Beverages fish vegetables 92/84 921335 9113320 AA;ETA;S AA;ETA;L AE;ICP;- Pb Pb S Foods Milk products Plants Se Foods AE;Hy;L 911997 Se Cereals AA;ETA;- 9111066 Se Milk AE;H y;L 9 111256 AA; ETA;- Se Fish meat 9 111279 Se Waters Se Food AA; H y ; L AA;Hy;L 9112653 9 112695 AA ; ETA; L Se Feeds plant and animal tissues 9113597 Oyster tissue AA;ETA;L AA;ETA;L AA;ETA;Sl Se Se Se 9113775 92192 921127 Grain grain products Flour Si Si Pineapple juice Beverages AA;F N,O- C,H,;L The results of a collaborative study of polydimethylsiloxane in pineapple juice were reported.The method was approved an interim official first action by the AOAC The effects of various experimental conditions acids salts and tube design were investigated (in Chinese) AA;ETA;L 9112410 9113345JOURNAL OF ANALYTICAL ATOMIC SPECTROMETRY APRIL 1992 VOL. 7 107R Table 2 SUMMARY OF ANALYSES OF FOODS AND BEVERAGES-continued Technique; atomization; analyte form* AA;ETA;L Element Matrix Si Jam Sample treatmentfcomments ETAAS and IR were used to quantify the antifoaming agent dimethylpolysiloxane in jam.The detection limit was 2 mg kg-' (in German) compounds by acid leaching was evaluated using HG-GC - quartz furnace AAS The coupling of GC with DCP-AES and an FPD was reported. Experimental conditions sample preparation and extraction procedures were described Thermal desorption GC-quartz furnace AAS were interfaced for the determination of butyltins after HG. Tributyltin in oysters from UK coastal waters was between 27 and 1667 jfg kg-I of Sn dry mass The efficiency of the extraction of butyltin See As ref. 91fC1692 Sulfhydryl cotton was used to adsorb and concentrate Sn4+ from acid digested food samples. Recoveries were 99.8 - 10 1.3% (in Chinese) digested with HN0,- HCIO dissolved in HNO (7%) and Th and U were determined by ICP-MS using Bi as internal standard Whole diet samples were dry ashed homogenized See Mn ref. 91/3275 Reference 91/3552 Sn Biological samples AA;Hy;G 91/994 Sn Fish shellfish AE;DCP;G 91/995 Sn Oysters AA;Hy;G 91/ 1362 MS;ICP;L AA;Hy;L 91x1 692 91/3565 Sn Sn Foods Foods Th Total diets MS;ICP;L 9 1/3785 TI Foods LEAFETA; L s AA;ETA; L S1 LEAFETA; S1 91/3275 An LEAF method for determining TI in bovine liver was reported.Methanol was found to increase the signal See Eu ref. 91/3463 See Th ref. 91/3785 Mineral waters were pH adjusted to pH 1.6 - I .8 boiled with Br2 preconcentrated on a cation- exchange column eluted and V determined by ETAAS at 381.4 nm. The LOD was 0.03 pg 1 - I (in Czechoslovakian) V was determined by ETAAS using hot injection and preconcentration on the graphite tube. The method gave an LOD of 0.58 jfg I-' following extraction with 0.5 mmol dm- 33- dibromosalicylaldehyde - 2- benzothiazolylhydrazone - CHCI solution in the presence of zephiramine (in Japanese) Trace amounts of V were quantified by ETAAS See Cu ref.91/1457 A slurry atomization method for determining Zn in flour was presented Sample preconcentration and hybrid methods of analysis were discussed A study was undertaken to assess element levels in human milk of rural and urban populations which in turn were sub-divided into smokers and non-smokers between metals were especially observed in liver paste Samples were digested in concentrated HNO,; then 3Ooh H202- 16.1 mol dm-3 HNO and finally diluted 10-fold in 0.1 mol dm- HCI prior to detection by Ar ICP-AES (B Ca Mg Na P) Principal component analysis was used to assign the origin of 35 grape samples Heavy metals in meats were studied; correlations TI Bovine liver 91fC3678 /3463 /3785 / 1007 Tm U V Butterfat fats Total diets Mineral waters AA;ETA;L MS;ICP;L AA;ETA;L V Water V Salt AA;ETA;L AA;ETA;L 91/2474 91/3479 Zn RMs z u Flour AE;MIP;G AA;ETA;SI 91/ 1457 9 1/C 1802 AF-iG 91/848 Various Plants milk powder water Various (1 2) Human milk AE;ICP;L 91/879 Various (7) Meats AA;e- AE;ICP;L 91/891 91/1051 Various ( 5 ) Biological materials foodstuffs 9111129 Various (1 0) Grapes AA;ETA;L108R JOURNAL OF ANALYTICAL ATOMIC SPECTROMETRY APRIL 1992 VOL.7 Table 2 SUMMARY OF ANALYSES OF FOODS AND BEVERAGES-continued Element Matrix Various Foodstuffs Technique; atomization; analyte form* AA;-;L Various (6) Dried fruit nuts spices A&-.- Various (22) Garlic Various (10) Animal organs Various (6) RMs Various (5) Waters Various ( T l ) Beverages Various (9) Animal feed cattle tissue milk Various (22) Diets Various (6) Asparagus Various (4) Water LE1;F;- AE;arc;- XRF;-;- AA;ETA;- MS;ICP;L MS;ICP;L MS;ICP;L AE;ICP;L AA;-;- AF;tungsten spiral; L Various (10) Human milk AE;ICP;L Various Bacterial cultures eggs honey AEICP;L S1 Various (1 3) Royal jelly AE;ICP;- Various (6) Beer AE;Hy;L Various Ham Various (9) Cheese Various (4) Water AE;-;- AA;ETA;L AA;FL 9 111 425 9111593 Sample treatmentlcomments Reference 9111 139 Fi!jh were found to be the major source of heavy metals in hospital impatient diets (As Cd Hg Pb) (in German) Concentrations of selected heavy metals in fruit nuts and spices were determined (Cd Cu Fe Mn Pb 9111 190 Zn) DCP-AES was used to analyse garlic (in Chinese) The concentration of essential trace metals in various animal organs were found to decrease in the order live> kidney> muscle and concentration of non- essential metals found in the order kidney >liver>muscle Automated chelation chromatography was used to remove interfering elements namely alkali and alkaline earths prior to LEI spectrometric analysis of reference materials Waters were filtered percolated through a column filled with polyacrylonitrile modified with polyethylene polyamine the sorbent dried and metals determined by AES XRF and ETAAS.All three methods gave similar results Semiquantitative mode ICP-MS was used to monitor 71 elements in a variety of drinks.The results were compared with EC drinking water directives This presentation described methodology and quality assurance procedures used to monitor various elements following the UK import of contaminated animal feed The design and results of a study of diet samples from Japan and the USA were reported Six essential elements were monitored in fresh and canned asparagus at each step of the canning process to assess the contribution to the daily intake of these elements in Spain Using a tungsten spiral atomizer heated to 2500 OC the following LODs in water samples were obtained 3 and 1 pg kg-1 for Cu and Pb and 3 and 5 ng kg-* for Cd and Zn respectively (Cd Cu Pb Zn) (in Russian) Weight stature sex and age were found to be factors related to inter-individual variation of certain elements in human transitional milk Research towards the direct analysis of homogenized tissues was described Samples (10 g) were evaporated with H,SO to near dryness ashed at 500 "C and the residue dissolved in HCl and HNO prior to final dilution with H,O (in Chinese) generation multi-element pre-reduction and loss of analytes in the HNQ,-HClO,-HF evaporation process in ICP-AES were described in detail (As Bi Sb Se Sn Te) (in Chinese) Several detection systems were applied to characterize the flavour components of ham.More than 60 components were tentatively identified Principal component analysis was used to ascribe the origin of cheeses Cd Cu Pb and Zn were extracted from water by APDC and NaDDC in IBMK butyl acetate and cyclohexane mixed solvents.Detection limits were below 5 pg 1-' (Cd Cu Pb Zn) (in Chinese) 9111337 9111384 91lC1680 9 1lC 1685 9112259 9112430 9 1 I2480 9112558 9 1lC2758 9 1 I3005 9113358 The effect of HNQ and HClO on hydride 9113383 911341 1 9 113412JOURNAL OF ANALYTICAL ATOMIC SPECTROMETRY APRIL 1992 VOL. 7 109R Table 2 SUMMARY OF ANALYSES OF FOODS AND BEVERAGES-continued Element Matrix Various (6) Fruits jam Various (1 3) Vegetables Various (7) Water Various Foods and beverages Various (1 4) Drinking water Various - Various Health foods Various (4) Bovine liver Various (4) Drinking water Various (1 I ) Fast foods Various (7) Sharks Various (8) Vegetables Technique; atomization; analyte form* AA;-;- XRF;-;L AA;F;L .. -- 1 MS;ICP;L AA;-;S L AE;-;S L AA;-;L AA;ETA;L AA;-;L XRF;-;- AA;-;- AA;-;L Sample treatmentlcomments Heavy metals in Polish bottled fruits and jams were quantified (As Cu Fe Pb Sn Zn) A method for simultaneous multi-element determinations using TXRF was described A micro-scale monitoring procedure in water samples utilizing coprecipitation with zirconium hydroxide gave recoveries of better than 95% (Cd Co Cu Fe Mn Pb Zn) analytical topics relationship between trace elements in mothers drinking water and infants born with neural tube defects The feasibility of a totally automated sample preparation robotics system was discussed The benefits of microwave digestions in food laboratories were presented Acid digestion and APDC solvent extraction were optimized to allow a single preparation and detection method for four elements of vital importance to livestock farming (Co Cu Mo Se) dithiocarbamate in IBMK allowed Cr Fe Mn and Ni to be determined down to levels of 0.4 0.3 0.22 and 0.36 pg 1-l respectively (Cr Fe Mn Ni) Fast foods (pizzas sandwiches breakfasts etc.) were analysed by XRF.Variations among franchise chain and outlet locations were significant in half of the determinations concentrations were found in inshore demersal species; lowest levels were found in offshore pelagic species (Cd Cu Fe Mn Ni Pb Zn) Results of a survey of element concentrations in vegetables grown around a nickel works led to the recommendation that the foods should not be for direct consumption due to the increased levels of heavy metals (Cd Co Cr Cu Mn Ni Pb Zn) (in Czechoslovakian) A review of recent publications concerned with A Canadian clinical study strongly suggested a Solvent extraction by pentamethylene In a survey of shark species highest metal Reference 9 113446 9113501 9 1 J3524 9 1 J3594 9113605 91lC3672 9 1 JC3702 9113777 9 113898 9212 14 921297 9213 1 2 *Hy indicates hydride generation and S L G and S1 signify solid liquid gaseous or slurry sample introduction respectively. Other abbreviations are listed elsewhere.interface remained intact.’ Detection limits for this method designed for use with commercial fisheries pro- ducts ranged from 15 to 27 ng. Organotins were determined in fish and shel&sh using GC with FPD and DCP (91/995). A commercially available GC-FPD system was used with a fused silica megabore column with a thin immobilized stationary phase (DB-17 1 pm thickness).No prior alkylation or hybridization was performed; the organotins were separated as the native species. Isothermal GC-FPD-DCP conditions permitted baseline resolution of all four Sn species of interest monobutyl- dibutyl- tributyl- and tetrabutyltin. Simulta- neous detection by FPD and DCP on a single injection was suggested for routine qualitative and quantitative determi- nations of the organotin species in complex food matrices. Improved sample extraction procedures were also des- cribed for organotins from fish. Butyltin compounds in oyster tissues from UK coastal sites were determined using an interfaced thermal desorption-GC-quartz fur- nace AAS system (91/1362).Quartz furnace AAS with HG was used to test the efficiency of the extraction of butyltin compounds from biological materials and sedi- ments by acid leaching procedures (91/994). The most efficient method used cold pure CH3C02H over a period of 4 h. Analytical figures of merit were given in a conference presentation describing the use of SFC with ICP-MS detection for the measurement of organotin com- pounds (9 lIC3666). Tetraalkyllead compounds were determined in seafood after enzymic hydrolysis and extraction with hexane (9 11993). Extracts were butylated prior to measurement by GC-AAS. Instrumental LODs were 1.6-2.3 pg of Pb. Total and bioavailable (CH3CH20H extractable) Cr in cereals legumes oil seeds and pastes were measured by AAS (9 1/24 12).The bioavailable fraction was in most cases >5O0/o of the total.11OR JOURNAL OF ANALYTICAL ATOMIC SPECTROMETRY APRIL 1992 VOL. 7 Free Ca total Ca and total CI- in milk were determined by FI and on-line dialysis (9 1/ 1 5 14). The three species were separated on-line from a single injection by directing them to three different channels using two dialysers in series. Samples (30 pl of milk) were processed at a rate of 60 h-l. Free Ca was determined by spectrophotometry at 580 nm and total Ca by FAAS at 422.7 nm. Dialysed Cl- was measured with a coated tubular Cl-selective electrode. In two conference reports the determination of F- was described. Labile-ionic and total F- in human diets were measured by graphite furnace molecular absorption spec- trometry (91/C1759).Total F- was determined after O2 flask mineralization whilst ionic and acid-labile F- were isolated by acidic diffusion. Fluoride was determined indirectly in waters by adding an excess of A13+ to form the AlF2+ complex (9 1 /C29 1 7). The complex is separated from A13+ by ion chromatography and detected by ICP and fluorescence techniques. Total F- was determined in tap water by laser-excited molecular fluorescence spectrometry with an unmodified AA graphite tube furnace (91/3186). After optimization of the system the LOD was 0.3 pg of F as F-. 2.4. Developments in Methodology for Electrothermal Atomic Absorption Spectrometry The effectiveness of chemical modijers for use with ETAAS continues to be investigated. The work of Lynch and Littlejohn (9 1/2220) was reported earlier in this review.They found that Pd was an effective modifier for the determination of Cd in food slurries. Smeyers-Verbeke et al. (91/846) used Pd to reduce the background absorbance due to NaCl. They stated that the use of large amounts of Pd should be avoided and that a combination of 6 pg of Pd and 500 pg of NH4N03 allowed the direct determination of Cd in undiluted urine against aqueous standards. The method was also applicable to milk and blood analyses. Detection limits were 0.1 pg 1-I for urine and 0.5 pg 1-1 for the milk and blood matrices. Cadmium was determined directly in aqueous suspensions of liver oyster tissue and plant leaves using S as modifier and an Mo tube atomizer (91/1544). Accuracy was better than with wet digestion but the RSD was inferior.Oxygen gas was investigated as a modifier for the direct determination of Cd and Pb in blood and food samples by ETAAS (9 11265 1). The background absorption for matrices such as milk orange juice and blood was greatly reduced and signal depression of Cd and Pb was prevented. The analysis of milk and dairy produce by ETAAS with (NH4)2HP04 as chemical modifier was reported (91/C2770). Sample preparation was very simple a 1 +3 dilution with water-xylene in the graphite tube followed by addition of the modifier to the tube. A programmable sample dispenser was used and the method allowed rapid determination of Cd Cu Pb and Zn. The severe effects of phosphorus on the determination of Se by ET'S were ameliorated by using Ni as a chemical modifier (9 1/3775).An uncoated graphite tube with a L'vov platform inserted was found to be the best atomizer. The amount of Ni added varied with the level of P present in the sample solution. The precision of the method with NIST SRM 1566 Oyster Tissue was 3.5% RSD; the measured concentration was within the certified range. Copper and Fe were determined in edible salad oil using a low temperature ashing pre-treatment (9 11920) with ETAAS and ICP-AES. The same elements were measured in butter (90/1391) after the fat had been removed by extraction with light petroleum and the sample decomposed in an autoclave at 150 "C for 1 h. Standard additions were used for quantification by ETAAS. An accurate method was developed for measuring trace amounts of Si in drinks using ETAAS (9 1/3345).Silicon from the antifoaming agent dimethylpolysiloxane was determined in jam by ETAAS (91/3552). 2.5. Developments in Methodology for Plasma Emission Spectrometry Heltai et al. (91/1457) studied the potential of both a toroidal Ar MIP and a cylindrical He MIP in combination with graphite furnace vaporization for the analysis of biological samples. The toroidal MIP improved the power of detection with respect to a filament Ar MIP by a factor of 3- 10. The LODs using 50 p1 aliquots of solution for a large series of elements were between 0.1 and 100 ng ml-l. Matrix effects associated with 10 pg ml-1 of Na occurred with the toroidal Ar MIP and could not be eliminated by the use of NH4N03. A diffuse He MIP in combination with graphite furnace vaporization gave LODs of 10 ng ml-l for P and 1 ng ml-I for Pb.However in the He MIP volatilization interferences were found for Ca and P. An ICP spectrometer equipped with a photodiode array was used to determine B levels in plant digests (91/3323). The emission spectrometer had three integrated optical subsystems a pre-selection polychromator in which an interchangeable mask at the Rowland circle allowed simul- taneous selection of four emission wavelengths for B a recombination system and an Cchelle-based high resolution spectrometer. In soil solutions B was measured down to 10-15 ng ml-l with varying sensitivities on the different emission lines. A method for B and other elements in foodstuffs was developed using open-vessel wet-ash and low-temperature digestion with ICP-AES (9 1/105 1).The efficiency of the method was tested using NIST SRM 1572 Orchard Leaves as this material is relative to other biological RMs difficult to digest. It is certified for several elements but not for B. The LOD for B was 15 ng g-l and for other elements ranged from 0.5 to 500 ng g-l. An indirect method for the determination of F- in waters by ICP-AES and FI with solvent extraction was reported (91/2459). A manifold with a single coil was used to first form and then extract the La-alizarin complexone-F- complex into hexanol containing N,N-diethylaniline. The organic layer was introduced into the ICP and the La content measured. The RSD was 2.16% for 200 pl of water containing 1 pg ml-l of F-. 2.6. Developments in Methodology for Inductively Coupled Plasma Mass Spectrometry Dale (921237) discussed the use of ICP-MS for trace element determination in food as well as other applications in this worker's laboratory.In a conference presentation also by Dale (91/C1663) the use of ICP-MS for determin- ing Hg in marine biota was discussed. The often low levels of Hg encountered are made more difficult to measure by ICP-MS by the relatively low abundance of the major isotope. Two procedures were tested one that used peak jumping to measure Hg and 11 other elements and the other that used scanning to measure Hg with thallium as an internal standard. Both proved adequate for measuring Hg but the accuracy and precision were much improved with the scanning procedure developed specifically for Hg. The application of HPLC-ICP-MS to speciation studies (91/C1692) is discussed in section 2.3.McLaren et al. (9 1/379 I ) have published some of their applications of ICP- MS in relation to marine samples in particular the certification of RMs. Multi-element analysis of Lobster Hepatopancreas Tissue (LUTS- 1) included microwave di- gestion with HN03-H202 and isotope dilution. Two conference reports described the analysis of animal feeds. Using a Perkin-Elmer Sciex Elan 500 Averitt andJOURNAL OF ANALYTICAL ATOMIC SPECTROMETRY APRIL 1992 VOL. 7 l l l R Wallace (9 1/C3667) compared data for feeds and phospha- tic fertilizers from ICP-MS measurements with those obtained by AAS and ICP methods. A VG Elemental PlasmaQuad was used for multi-element analyses of cattle feed animal tissues and milk (9 1 /C 1658) following sample dissolution in a microwave oven. Iodine was determined in fresh and powdered milk (9 1/3239).Simple sample preparation combined with the selectivity and sensitivity of ICP-MS allowed a rapid sample throughput with the detection of I in the lower ng ml-l range. Shiraishi et al. (9113785) measured Th and U in Japanese total diet samples. Internal standardization using bismuth was adopted to correct for non-spectral interferences. The diet samples were dry ashed and then subsamples (0.25 g) of the ash were digested with HN03- HC104 until only a white residue remained. This residue was dissolved in 7% HNO,. The final volume was made up to 25 ml and the acid strength adjusted to 10% HN03. The mean concentrations and standard deviations of Th and U in the total diet samples were found to be 25k 12 and 44+20 ng g-l of ash respectively.Although the ash contents of the diets were not given it is apparent that these two elements were determined at pg ml-* levels in these samples. The results for the NIST SRM 1571 Orchard Leaves were in agreement with certified values. The LODs for this study were 1.3 pg ml-l for Th and 2.9 pg ml-l for U. There were three reports on the use of the semi- quantitative mode of analysis with ICP-MS. Several NIST SRMs were analysed (9 11857) by dividing the elements into three groups according to their atomic number each group having its own internal standard. The internal standards were vanadium caesium or indium and iridium or bismuth. Concentrations within 30% of the certified value could be obtained for 23 elements over a certain concentra- tion range.Larger errors occurred for elements with certified values of ~ 0 . 1 pg g-l because of dilution during sample preparation or because of interferences on some elements. Drinking water (9 1 /C 1665) mineral waters and soft drinks (9 1 /C 1 680) were also investigated. Drinking water was analysed for 14 trace elements by Longerich et al. (91/3605) in order to investigate any possible relationship between levels of trace elements in water and neural tube defects in infants. No sample pre- treatment or preconcentration was used. The LODs ranged from 0.03 ( Ce U and Y) to 67 (I) pg kg-l and were mostly less than 10 pg kg-l. The ability to measure isotope ratios by ICP-MS was used in a series of three publications by Fomon and co-workers (9 1/23 14 9 1/23 15 9 1/23 16).They were investigating the availability and absorption of Fe from infant foods and childrens' meals. Labelling with 58Fe avoided the adminis- tration of radioisotopes. The ratio 58Fe:57Fe was measured in blood by ICP-MS. 2.7. Multi-element Analyses of Foods The number of papers presenting data for more than one element has increased this review year. Some examples are highlighted here. Not only has the range of analytes expanded but also the types of foods analysed. The techniques used to generate this data are not always simultaneous multi-element methods. Thus a rapid method for the determination of Cr Fe Mn and Ni in drinking waters was developed using chelation and extraction to achieve a 35-fold preconcentration prior to measurement by AAS (9 113898).In contrast 52 elements were measured in drinking waters using the multi-element techniques of instrumental NAA and PIXE with no chemical treatment of the samples before irradiation (92171). Since PIXE is sensitive enough using proton beam scattering to detect many trace elements simultaneously and without any separation processes Yukawa et al. (9111 158) used this technique to investigate the spatial distributions of various elements in human tissues and foods. Four papers reported multi-element data for infant foods. Human milk was analysed by ICP-AES (91/879). Twelve elements were determined and the paper details the strategy used to ensure that the data were reliable and representa- tive.Possible causes of analyte loss or contamination were also investigated and minimized. Human transitory milk collected on days 6-9 postpartum was also analysed by ICP- AES (9112558). The nutritional elements Ca K Mg and Na were determined by FAAS in infant foods (91/1263). Eight popular brands of infant formula and two cornflour infant foods commonly consumed in Nigeria were com- pared. The cornflour foods were found to have considerably lower amounts of the four elements. Industrially produced baby food was analysed for the toxic contaminants Cd Cu Hg Pb and Zn by Yugoslavian workers (9 1/2541) by FAAS and cold vapor AAS. In most cases the levels were less than those in the literature and did not exceed Yugoslavian permitted levels.Simultaneous multi-element analysis of vegetables was reported using TXRF (9 1/350 1). Samples were freeze dried homogenized and digested with HNO,. Gallium was added as an internal standard and 20 pl of the solution applied to a vitreous silica target and evaporated to dryness. Measure- ments could be made with the same internal standard over a concentration range of four orders of magnitude. Seventeen elements were determined in 134 vegetable samples 67 fruit samples 10 samples of roots and tubers and 15 samples of mushrooms by ICP-AES (92/3 19) and garlic was analysed by DCP-AES for 26 elements (9 111 337). Uhnak and Rippel (92/3 12) measured the levels of Cd Co Cr Cu Mn Ni Pb and Zn in ten types of vegetables grown around a Czechoslovakian nickel smelter.Cadmium and Pb levels were higher than those in similar vegetable samples taken in Slovakia and in some cases the Cd levels exceeded the permitted values. The levels for the other elements were elevated when compared with those from other regions of Slovakia and from other countries. However they did not exceed admissible values. Royal jelly (91/3005) was one of the more unusual food items analysed (for 13 elements) by ICP-AES. Fresh and canned asparagus samples were analysed (9 11243 1 91/2450) by AAS. Cadmium and Pb were found to concentrate at the tip of the asparagus. Fast-food samples (239 representing sandwiches Mexican foods pizzas deep- fried food salads desserts breakfast foods and beverages) were analysed by XRF (92/2 14) for 1 1 elements. Results for each element were validated by using between 7 and 13 NIST SRMs.Results obtained with these SRMs were on average within 7% of the certified values with a mean bias of - 2.8%. Selected heavy metals were determined in spices dryfruits and nuts using a combination of potentiometric stripping analysis and AAS (9 111 190). Mixed spices gener- ally had high levels of Cd (0.65-1.34 pg g-l) Fe (142.3-285.0 pg g-l) Pb (6.9-9.2 pg g-') and Zn (64.2-65.8 pg g-l). Almonds contained higher levels of Cd (0.24 pg g-') and Pb (1.02 pg g-l) than other nuts and dry fruits. The lanthanides Eu Tm and Yb were tested for their suitability as markers for butterfat and mixtures containing butterfat (91/3463). Palmitate salts of these REEs at ppb levels were blended with the fats and determined by ETAAS.For fat solutions an organic extraction was carried out and for foods such as chocolate ashing and standard additions were used. Ytterbium was found to be unsuitable as a marker for foods containing plant derived products because of its variable occurrence in such materials. Europium and Tm did not have this restriction. Several papers report values for Cd and Pb in foods. The mean levels for Cd and Pb in Polish berries were 12-77 and112R JOURNAL OF ANALYTICAL ATOMIC SPECTROMETRY APRIL 1992 VOL. 7 40-150 pg kg-l respectively (91/3437). The mean Cd and Pb contents of berries on the Finnish market in 1987- 1989 were 2-24 and 11-27 pg kg-l (91/C1709) respectively. Rahman et al. (9 1/4022) measured Cd and Pb in pulses and cereals by ETAAS. Milling the cereals increased the metal contents of the flour when compared with the whole grains.Preserved eggs (91/2995) and feed yeasts (91/1147) were analysed by FAAS. Some samples of eggs contained 2.14-3.1 1 ppm of Cd and 0.92-2.25 ppm of Pb. Sheu et al. (9 1/1023) developed a method for determining trace levels of Cd and Pb in cheese cream and butter using standard additions and ETAAS. 2.8. Progress on the Determination of Some Individual Elements 2.8.1. Aluminium Delves et al. (91/965) have determined A1 in foods by ETAAS. Liquid foods were diluted with 1% m/v (NH4)2HP04 and analysed directly. Fruit and vegetable samples were digested in HN03 and diluted with 1% (NH4)2HP04. High fat samples were extracted with petro- leum ether and A1 was back-extracted with 0.1 mol dm-3 HN03.The residue that was insoluble in petroleum ether was oxidized with HNOJ and both phases were analysed. Recoveries of added A1 from all types of foods were 97.8-102.9% for concentrations of 0.02-3.2 pg g-l of Al. Aluminium in foods was determined by STPF-ETAS following digestion with HN03-HClO (92/2 13). The di- gested residue was dissolved in OS0h m/v Mg(N03)2 with 0.3% v/v HN03. Average recoveries were 86-1 11% with an LOD of 0.24 ng. Three different digestion techniques for the determination of total A1 in foods were compared (9 l/C1822). The methods were a high pressure microwave digestion a conventional HN03-HC104 digestion and the latter in combination with an HF-HN03 pre-treatment step. There was a slightly higher recovery for A1 from the high pressure microwave digests compared with the HN03-HC104 digests.For some samples the pre-treatment step resulted in higher A1 concentrations being measured and the increase varied with the amount of HF added. Addition of too much HF however can start to decrease the recovery of Al. The HF was thought to release A1 tightly bound possibly to Si in some foods particularly those which may have been exposed to dust and soil such as flour and spinach. Samples such as meat and milk gave the same results for all digestion methods. In an important paper (91/2430) Woollard et al. ana- lysed 307 samples of infant milk formulae from 14 countries by ETAAS. Infants in particular neonates are at risk from A1 overload because of their renal immaturity. The mean A1 concentration was 1.40 mg kg-l with a 95% confidence interval of 0.17-3.84 mg kg-l.In the products from New Zealand and Australia the mean and range of the A1 levels in infant formulae were statistically the same as those in standard whole milk powders despite their more sophisticated processing and storage in A1 cans. The concentration of A1 in 55 soybean milk substitutes from seven countries was much higher with a global mean of 18.4 mg kg-l with a 95% confidence interval of 10.4-37.6 mg kg'l. The contribution of A1 from vegetable oil and vitamin and mineral additives was shown to be insignifi- cant in these products. Dabeka and McKenzie (9 1/2449) have also published some important data for Canadian infant formulae. Alumi- nium was determined by ETAAS with a L'vov platform and pyrolytic plateau-type graphite tubes; deuterium-arc back- ground correction was used. Milk based formulae contained average (range) concentrations of 0.129 (0.010-0.36) 0.2 17 (0.17-0.56) and 0.7 17 (0.19-2.49) pg g-l for (as sold) ready- to-use concentrated liquid and powder formulae respec- tively.The corresponding concentrations for soybean-based formulae were 1.98 (0.40-6.4) 1.4 1 (0.59-2.29) and 9.44 (3.15-1 8.0) pg g-l. It was estimated that consumption by 1-3 month old infants of only the formula brand having the highest mean A1 level could result in an intake of 2088 pg of A1 per day. The estimate does not include A1 from any other source such as water or other foods. Two other papers described methods for determining A1 in milk by ETAAS (91/922) and by ICP-AES (9113289). The chemical speciation of A1 in milk (9113520) was studied using the ECCLES computer program.Formation constants were reported for the A13+ citrate succinate piccolinate and malate systems under specified conditions. It was proposed that the major A1 species were charged citrate complexes which it was claimed would present little or no threat to a healthy human being as such complexes would not be absorbed through the intestinal walls. The influence of ultrasonics on the determination of A1 in potable waters was investigated (9 113497). In some in- stances a higher value was obtained by ETAAS after sonication of samples in poly(propy1ene) bottles. Subse- quent membrane filtration studies indicated a destabilizing and solubilizing effect of ultrasonic vibrations on colloi- dally bound Al.2.8.2. Mercury Sonication was used in a decomposition technique for the determination of Hg in drinking and surface water (92187). Total Hg is measured after sonication-aided oxidation of organically bound Hg followed by treatment with a reducing agent and cold vapour AAS of elemental Hg. Solvent extraction was used to determine Hg down to 1 ng 1-1 in water by ETAAS (9 113309). Mercury concentrations of a 6 0 pg ml-l in waters after preconcentration on a column of powdered Cu or 2 16 pg ml-l after electrodeposition were determined by Luca et al. (91/3561). The use of ICP-MS for measuring Hg in marine biota (911C1663) was discussed in section 2.6. Horvat et al. (9 1 / 1343) made comparative studies of methylmercury determination in j s h mussels shrimps and algae.Different isolation methods were tested and the detection methods cold vapour AAS or GC compared. Results from the various isolation techniques (ion exchange extraction volatiliza- tion and distillation) were comparable for most biological and environmental samples except for soils and sediments where there was disagreement between results from AAS and GC. In an international collaborative exercise (9 1 / 1 346) various analytical methods were used by 13 laboratories from 7 countries to measure methylmercury in fish muscle and mussel tissue. All of the methods gave similar results. IJrich et al. (9 1/88 1) described the optimization of a method to determine Hg in fish muscle and two other publications (91/1357,91/2496) reported the use ofcold vapour AAS for the determination of Hg in fish and water.Danha and Baloun (9 111 249) compared the capabilities of a commercially available instrument the TMA (trace mercury analyser) 254 with those of cold vapour AAS. Mercury was absorbed on an amalgamator and then released by heating the amalgamator. Total Hg was deter- mined in food and urine. Analyte Concentrations and LODs were the same for both methods but the TMA 254 method was faster and less expensive. The TMA 254 was also used (91/2663) to determine Hg in products of the Czechoslo- vakian sugar and confectionary industries. The analysis time was 5-7 min and the LOD was 1 ,ug kg-I. 2.0.3. Selenium The influence of difSrerent chemical modifiers on the determi- nation of Se in foods has been investigated in severalJOURNAL OF ANALYTICAL ATOMIC SPECTROMETRY APRIL 1992 VOL.7 113R studies. Erard and Zimmerli (9 1/ 1066) analysed wheat flour and cereals by ETAAS. Combinations of chemical modifi- ers (Ni Cu-Mg Pd-Mg-ascorbic acid Pd-Mg-glycerol and Pd-Mg) were investigated as well as the effect of thermal pre-treatment and the oxidation state of Se. The Pd-Mg modifier was the most suitable for wheat samples containing approximately 1.0 pg g-l of Se. For those containing about 0.1 pg g-' of Se this modifier was unsuitable as double atomization peaks were formed. These peaks were due to a non-uniform atomization cloud and increasing the pre-treatment temperature in an attempt to overcome this led to large losses of Se.The same group (92/92) determined Se in cereals and bakery products and found that sample concentrations of 2.5-5 ng ml-I of Se could be analysed by Zeeman-effect ETAAS with a Pd-Mg modifier. The detection limit for the determination of Se in pig tissues (9 1/30 1 5) was reported as 1.4 ppb using Zeeman- effect ETAAS with a L'vov platform and a Pd-Mg chemical modifier. Maage et al. (9 1/3 1 75) used Ni as a modifier with an uncoated graphite tube and L'vov platform to overcome the severe effects of phosphate on the Se signal found with deuterium-arc background correction. The amount of Ni required varied with the amount of P present. Selenium was measured directly in slurries of flour using ETAAS with Pd as chemical modifier (92/ 127). Hocquellet and Candillier (9 1/3597) evaluated mi- crowave digestion combined with solvent extraction to determine Se in feed and animal and plant tissues.Samples were digested in a microwave oven with HN03-H2S04- HClO followed by co-extraction of Se and added Pd with DDDC in CH3Cl. An LOD of 0.002 pg g-' of Se was reported. Kardos et al. (91/997) achieved an LOD of 0.7 ng ml-1 of Se for the determination of Se in selected Hungarian and Italian foods by ICP-AES with HG. Values ranged from 0.03 pg g-l in fruit cocktail to 0.90 pg g-' in canned fish. Ultra-trace amounts of Selv in water and soil extracts were measured after FI on-line preconcentration using an anion-exchange resin (9 ~2653). After elution with 1 mol dm-3 HCl the eluate was mixed with NaBH in 0.1% NaOH and Se determined by HGAAS.The LOD was 2 ng 1-l with a sample throughput of 50 h-'. Potable water was treated (91/177) with NaDDC extracted with tributyl- phosphate and reacted with NaBH in DMF and anhydrous CH3C02H. The hydride was measured by AFS and the LOD was 18.1 pg of Se with an RSD of 2.7%. In the San Joaquin Valley of California USA measure- ment of Se in bulk tank milk has been found to be an effective way of evaluating the Se levels of the herd in order to detect Se deficiency (9 1/1256). Mean levels of Se in blood and milk from the herd were directly proportional to bulk tank milk levels which ranged from 0.0125 to 0.0418 mg 1-l. Ekholm et al. (91/1279) reported that Finnish multi- nutrient fertilizers have been supplemented with Na2Se04 since 1984 in order to raise the Se content of Finnish foods.Since the start of the intervention Se contents have been regularly monitored in key foods. The Se content of meat and offal but not that of fish or wild animals has been increased. The present Se intake in Finland is 3-4-fold that in the mid-1970s with meat and fish contributing 52% of the Se intake in 1988 compared with about 63% in 1975. 2.8.4. Vanadium Four papers described the determination of V in waters after some form of preconcentration. Bermejo-Barrera et al. (90/3506) used 8-hydroxyquinoline in IBMK and measured the organic phase by ETAAS. The method was free from interferences and had an LOD of 0.16 pg 1-I. In a second publication Bermejo-Barrera et al. (9 112474) described the determination of V in water by ETAAS using hot injection and preconcentration on the graphite tube.One 50 pl portion was injected onto the furnace which was heated to 110 "C for 200 s during which three further 50 pl portions were introduced at 50 s intervals. The sample was then charred and atomized for measurement; Mg(NO,) was added to the samples as modifier. The limit of determina- tion was 0.58 pg 1-I. Nevoral (91/1007 91/C2759) used Ostion KS 0807 a strong acidic cation-exchange resin. The V was selectively eluted with 1 Yo v/v H202 and measured by ETAAS. The LOD was 0.03 pg 1-I. Trace levels of V were measured in common salt after extraction into an organic layer followed by back extraction into HCl for ETAAS (9113479). 2.9. Dietary Intake Studies Capar (9 113956) reviewed aspects of assessing dietary intake using analytical methods.The US Food and Drug Administration's current method development programme for monitoring elements in foods was designed to reduce the number of mineralization procedures expand the number of elements determined and to use ICP-AES for determining all elements of interest. These methods were reviewed and the value of currently used methods for assessing dietary intake discussed. A 7 d duplicate diet study was conducted to study the As intake and excretion by Japanese adults (9113529). The study was small with four volunteers. However their conclusion was that trimethylarsenic compounds in urine should be the preferred indicator of As arising from the ingestion of seafood. The mean daily intake of inorganic As from the diet (0.18 pg kg-I) did not exceed the Food and Agricultural Organization World Health Organization (FAO-WHO) Tolerable Daily Intake of 2 pg kg-l of inorganic As.The daily dietary intake of potential contaminants (As Cd Hg and Pb) in hospital food was determined by Stelz et al. (91/1139) using AAS. The highest concentrations of As and Hg were found in diets containing fish. However As was found in only 6 out of 43 samples at levels greater than the LOD of 0.04 mg kg-l whilst only 1 sample contained Hg above the LOD of 0.025 mg kg-'. Average Cd and Pb contents of daily rations were 12 and 65 mg kg-' respectively giving average weekly intakes of 100 and 544 pg per person. In contrast the weekly intake of Cd was 133-139 mg and that of Pb 504-952 mg from the daily food rations of Polish families of medium-income workers (91/3562).Mercury intake was 61.6-164.0 mg per week. None of these values exceeded the 50% maximum permis- sible levels but they were higher than reported in previous years and these workers stated that this is indicative of increasing contamination of food products. They also found that in areas of industrialization the dietary intake of elements was higher. 2.10. Characterization Studies Favretto et al. have published two papers (91/1129 91/3411) reporting the use of ETAAS to investigate the trace element content of grapes and cheese. Principle component analysis was used to test for associations among the mineral constituents. For both food matrices the clusters of elements appeared to be determined by their origin.In a conference presentation McKay and Baxter (9 UC1678) discussed the suitability of Pb isotope data obtained by ICP-MS for characterizing wines. They had found that the difference in the range of 2oaPb:207Pb ratios measured in wines from throughout the WOTM was about 1 1090 which was at least a factor of 10 greater than the worst114R JOURNAL OF ANiALYTICAL ATOMIC SPECTROMETRY APRIL 1992 VOL. 7 measurement error. Atomic emission was used as part of the characterization procedure for ham flavour (9 1/3383). Volatile flavour compounds were extracted using a Likens-Nickerson apparatus this involved steam distilla- tion of the sample followed by organic extraction of the distillate. The flavour isolates were concentrated for GC analysis. Atomic emission was used as a selective detector for N- 0- and S-containing compounds.2.1 1. Reference Materials and Collaborative Trials Inhat and Stoeppler (9 1 /3232) reported the preliminary assessment of ten new agriculturalflood reference materials. Twenty-five elements were determined. Solid sampling ETAAS was used for the determination of Cu and Pb in small (sub-milligram) samples. Other elements were deter- mined by a variety of techniques on sub-samples ( 100-2000 mg) as part of an interlaboratory cooperative scheme. Good homogeneity was observed for all of the analytes except for Cr and Pb in a limited number of instances. Further measurements will be made to bring the materials to CRM status. The production of a Community Bureau of Reference (BCR) reference material was described by Pauwels et al.(9113233). The material to be certified for Cd Fe Hg Pb and Zn is to be prepared from codfish and the paper described the production processes by which the fresh material becomes an homogenous powder. The base ma- terial was subjected to cryo-grinding in Teflon equipment and then freeze-dried to give a dry powder. Solid sampling Zeeman-effect ETAAS was used to measure the elements of interest and to assess homogeneity. Analyses at different stages of the production process showed that the heterogen- eities existing in the starting material were gradually eliminated during processing and that no external contamination occurred. Whilst it is encouraging to read of new reference materials it is to be hoped that the range of certified elements will be expanded for future CRMs.With the increasing use of multi-element techniques it should be possible to achieve this. Collaborative trials were carried out to test FAAS determinations of Ca and Mg in milk powder and four CRMs (9111 108) to determine Ca Mg and P in cheese (9 ~ 3 8 4 3 ) and to measure polydimethylsiloxane (as ex- tracted Si) in pineapple juice by FAAS (91/2411). Two papers reported tests of digestion methods for the determination of Cd and Pb in foods. Collet et al. (9 1/3 100) presented data from a collaborative study which showed that when wheat flour and freeze-dried beef liver were analysed by AAS there was no difference between wet ashing and pressurized digestion. Ellen et al. (9 1 /2448) found that the LODs using ETAAS for Cd and Pb were 1 and 20 pg kg-’ respectively for pressure bomb digestion (1 g samples) and 0.5 and 5.0 pg kg-I for dry ashing (5 g samples) when 16 RMs were analysed.There were two reports of collaborative trials involving fats and oils (91/1318 91/2696). Copper and Ni were determined in vegetable oils and margarine by ETAAS following homogenization with HNO and H20 (9 1 / 1 3 1 8). The aqueous phase was analysed and LODs for this method were 0.004 and 0.030 mg kg-I for Cu and Ni respectively. In the second paper (9112696) samples (1 g) were dissolved in IBMK-HN03 (approximately 10 ml) and aliquots (20 pl) measured directly for Cu Fe and Ni by ETAAS. The study was conducted at seven laboratories using different equip- ment and conditions. Data were obtained for soybean oil with the analytes added (0.6 ppm) and the RSDs obtained were 5% for Cu 18% for Fe and 10% for Ni.It was concluded that the method was simple and rapid and thus useful for the analysis of metals in oils. LOCATION OF REFERENCES The full list of references cited in this Update have been published as follows 91/826-91/C1687 J. Anal. At. Spectrom. 1991 6(3) 109R-136R. 91/C1688-91/2702 J. Anal. At. Specrrom. 1991 6(4) 153R-185R. 91/2703-91/C2928 J. Anal. At. Speci’rom. 1991 6(5) 221R-227R. 91/2929-9113584 J. Anal. At. Spectrom. 1991 6(7) 257R-280R. 9 113585-9 114050 J. Anal. At. Spectrom. 199 1 6(8) 323R-340R. 92/1-921338 J. Anal. At. Spectrom. 1992 7( l) 53R-66R. Abbreviated forms of the literature references quoted (excluding those to Conference Proceedings) are given on the following pages for the convenience of the readers.The full references names and addresses of ithe authors and details of the Conference presentations can be found in the appropriate issues of JAAS cited above. Abbreviated List of References Cited in Update 91/97. Mikrochim. Acta 1989 3(3-6) 299. 911103. Mikrochim. Acta 1989 3 267. 911105. Mikrochim. Acta 1989 3 291. 911157. Anal. Chem. 1990,62 1161.911177. Fenxi Huaxue 1989 17 909. 911209. Zh. Anal. Khim. 1990 45 29. 911827. Analyst 1990 115 1025. 911833. Analyst 1990 115 1323. 911846. J. Anal. At. Spectrom. 1990 5 393. 911848. J. Anal. At. Spectrom. 1990 5 407. 911853. J. Anal. At. Spectrom. 1990,5,431.91/857. J. Anal. At. Spectrom. 1990 5 457. 911868. J. Anal. At. Spectrom. 1990 5 519. 911879.J. Anal. At. Spectrom. 1990 5 581. 911880. Acta Chim. Hung. 1989 126 311. 911881. Acta Cient. Venez. 1988 39 380. 911883. Acta Neuropathol. 1989 79 149. 911884. Acta Ophthalmol. 1989 67 582. 911886. Ann. Ist Super. Sanita 1989,25 499. 911887. Ann. Ist. Super. Sanita 1989 25 385. 911909. Verh. - Ges. Oekol. 1988 18 317. 911920. HuaHsueh 1989,47(1) 30. 911922. Ind. Aliment. (Pinerolo Italy) 1989 28 589. 911930. J. Environ. Sci. (China) 1989 1(2) 104. 911933. Micronutr. Anal. 1989 6 85. 911935. J. Microsc. (Oxford) 1989 155 81. 911955. Pract. Spectrosc. 1989 7 (Flow Injection At. Spectrosc.) 259. 911960. Riv. Soc. Ital. Sci. Aliment. 1989 18 163. 911965. Spec. Pub1.-R. SOC. Chem. 1989 73 (Alum. Food Environ.) 52. 911989. Ann. Chim. (Rome) 1989 79 529. 911993. Appl.Organomet. Chem.,JOURNAL OF ANALYTICAL ATOMIC SPECTROMETRY APRIL 1992 VOL. 7 115R 1989 3 21 1. 911994. Appl. Organomet. Chem. 1988 3 469.911995. Appl. Organornet. Chem. 1989,3,295.911997. Ann. Ist. Super. Sanita 1989 25 505. 9111007. Chem. Listy 1990 84 544. 9111018. Gaodeng Xuexiao Huaxue Xuebao 1989 10 1 139.9111022. Hua Hsueh 1987,45(3) 97. 9111023. Hua Hsueh 1989 47( l) 2 1. 9111025. Huanj- ing Huaxue 1990 9(1) 73. 9111033. Iwate-ken Eisei Kenkyusho Nenpo 1988 31 37. 9111041. Jinan Daxue Xuebao 1989 (3) 63.9111051. J. Micronutr. Anal. 1989,6 16 1. 9111054. Kagaku Gijutsu Kenkyusho Hokoku 1989 84 643. 9111059. Kidney Int. 1990 37 818. 9111062. Khim. Sel’sk. Khoz. 1990 (4) 70. 9111066. Mitt. Geb. Lebensmittelunters. Hyg. 1989 80 452. 9111079. Pathol.Res. Pract. 1990 186 197. 9111088. Sci. Total Environ. 1989,89 353.9111090. Shengzhi Yu Biyun 1990,10( l) 56. 9111108. Z. Lebensm.-Unters. Forsch. 1990 190 199. 9111119. Bull. Chern. SOC. Jpn. 1990 63 544. 9111129. Chem. Express 1989 4 495. 9111130. Chem. Mag. (Rijswijk Neth.) 1989 (Feb) 7 1.8111136. Czas. Stomatol. 1989 42 95. 9111138. Doga Turk Kim. Derg. 1988 12 290. 9111139. Dtsch. Lebensm-Rundsch. 1990 86 10. 9111143. Gaodeng Xuexiao Huaxue Xuebao 1989,10,4 1 8. 9111144. Gaodeng Xuexiao Huaxue Xuebao 1989,10,429. 9111147. Gidroliz. Lesokhim. Prom-st. 1990 (2) 13. 9111149. Gongye Weisheng Yu Zhiyebing 1989 15 38. 9111153. Huaxue Shijie 1989 30 163. 9111158. Hyomen Kagaku 1989 10 488. 9111163. J. Environ. Sci. (China) 1989 1 91. 9111174. Kem. Kozl. 1989 69 206.9111178. Laboratoriumsmedizin 1990 14 12 1. 9111183. Magu- neshumu (Kyoto) 1989 8 9. 91/1187. Neurochem. Res. 1989 14 1099. 9111189. Phys. Med. Biol. 1989 34 1833. 9111190. Plant Foods Hum. Nutr. (Dordrecht Neth.) 1989 39 279. 9111227. Zhongguo Yaoke Daxue Xuebao 1989 20(1) 43. 9111247. Can. J. Chem. 1990 68 735.9111249. Cesk. Hyg. 1989 34 529. 9111256. Cornell Vet. 1990,80 41. 9111259. Environ. Pollut. 1989,62 103.9111263. Food Chem. 1990 37 269. 9111271. Huanjing Huaxue 1989 8(4) 41. 9111279. J. Agric. Food Chem. 1990 38 695. 9111280. J. Clin. Chem. Clin. Biochem. 1989 27 631. 9111281. Br. J. Clin. Pharmacol. 1990,29 39 1.9111282.9. Environ. Sci. (China) 1989 1(2) 69.9111299. LaborPraxis 1990 14 244. 9111300. Lebensrnittelchemie 1990 44 3. 9111302. Lizi Jiaohuan Yu Xifu 1988,4 358.9111318.Riv. Ital. Sostanze Grasse 1989 66 33. 9111329. Trace Elem. Anal. Chem. Med. Bio. Proc. Int. Workshop 5th 1988,236. 9111336. Yaoxue Xuebao 1990,25 157.9111337. Yingyang Xuebao 1989 11(2) 159. 9111343. Appl. Organomet. Chem. 1988 2 515. 9111346. Appl. Organomet. Chem. 1989 3 257. 9111348. Biol. Trace Elem. Res. 1988 21 241. 9111357. Cienc. Cult. (Sao Paulo) 1989 41 88. 9111362. Heavy Met. Hydrol. Cycle 1988 41 9. 9111369. J. Dairy Sci. 1989 72 2549. 9111376. Koku Eisei Gakkai Zasshi 1989,39 598.9111384. Pak. J. Sci. Ind. Rex 1989 32 175. 9111385. Phys. Med. Biol. 1989,34 1295.9111391. Radioisotopes 1989 38 179. 9111425. J. Anal. At. Spec- trom. 1990,5 595.9111428. J. Anal. At. Spectrom. 1990,5 619.9111432. J. Anal. At. Spectrom. 1990 5 635.9111436.J. Anal. At. Spectrom. 1990 5 243R. 9111438. Fenxi Huaxue 1989 17 1068. 9111457. Spectrochim. Acta Part B 1990 45 857. 9111481. Fenxi Shiyanshi 1990,9(2) 15. 9111482. Fenxi Shiyanshi 1990 9(2) 28. 9111485. Anal. Sci. 1989 5 355. 9111487. Anal. Sci. 1990 6 385. 9111493. Analusis 1990 18(5) 327. 9111494. Analusis 1990 18(5) 33 1.9111505. Fresenius Z. Anal. Chem. 1989 335 855. 9111509. Fresenius J. Anal. Chem. 1990 337 294. 9111511. Fresenius J. Anal. Chem. 1990 337 306. 9111512. Fresenius J. Anal. Chem. 1990,337,3 10.9111513. Fresenius J. Anal. Chem. 1990,337 3 16.9111514. Fresen- ius J. Anal. Chem. 1990,337,393.9111527. Lihua Jianyan Hauxue Fence 1990 26(2) 94. 9111544. ,Wikrochim. Acta 1990,1 8 1.9111545. Anal. Chem. 1989,61 185 1.9111546. Anal.Chem. 1989 61 1867. 9111550. Anal. Chem. 1990 62 1994. 9111573. Guangpuxue Yu Guangpu Fenxi 1990 10(1) 27. 9111576. Guangpuxue Yu Guangpu Fenxi 1990 10(1) 46. 9111584. Talanta 1990 37 71 1. 9111593. Zh. Anal. Khim. 1990 45 904. 9111602. Zh. Anal. Khim. 1990 45 1234. 9111605. Bunseki Kagaku 1990 39 317. 9112155. Microbeam Anal. 1989,35 (Proc. 24th Ann. ConJ Microbeam Analysis Soc. Asheville NC USA 16th-2 1 st July 1989). 9112170. J. Anal. Toxicol. 1989 13 367. 9112171. J. Anal. Toxicol. 1990,14 206.9112217. Talanta 1990 37 707. 9112219. Talanta 1990 37 8 19. 9112220. Talanta 1990,37 825.9112232. Anal. Proc. 1990,27,336. 9112251. Anal. Sci. 1990,6 157. 9112255. Anal. Sci. 1990 6 471. 9112256. Anal. Chim. Acta 1989 224 83.9112259. Anal. Chim. Acta 1990 229 157. 9112314.Pediatr. Res. 1988,23,495.91/2315. Pediatr. Res. 1988,24,20.91/2316. Pediatr. Res. 1989 26 250. 9112358. Pract. Spectrosc. 1990 8 413. 9112375. Fenxi Huaxue 1990 18 578. 9112382. Fenxi Shiyanshi 1990 9(2) 66. 9112389. Yanku- ang Ceshi 1990 9 48. 9112390. Yankuang Ceshi 1990,9 65.9112405. Sci. Total Environ. 1990 95 89.9112406. Sci. Total Environ. 1990 95 149. 9112410. J. Assoc. 08 Anal. Chem. 1990 73 619. 9112411. J. Assoc. 08 Anal. Chem. 1990,73,72 1.9112412. J. Assoc. O g Anal. Chem. 1990,73 798. 9112430. Food Chem. 1990 37 81. 9112431. Food Chem. 1990,38 1 13.9112448. Food Addit. Contam. 1 990 7 265. 9112449. Food Addit. Contam. 1990 7 275. 9112450. Food Addit. Contam. 1990,7,38 1.9112459. Anal. Chem. 1990 62 2457. 9112469. Anal. Chim. Acta 1990 234 433.9112470. Anal. Chim. Acta 1990 235 393. 9112474. Anal. Chim. Acta 1990,236,475.9112480. Zavod. Lab. 1990 56 38. 9112485. Guangpuxue Yu Guangpu Fenxi 1990 10(1) 50. 9112496. Lihua Jianyan Huaxue Fence 1990 26 33. 9112506. Fresenius J. Anal. Chem. 1990 337 275. 9112513. Fresenius J. Anal. Chem. 1990 337 860. 9112520. Beijing Shifan Daxue Xuebao Ziran Kexueban 1989 3 61. 9112521. Biol. Trace Elem. Res. 1990 24 147. 9112525. Clin. Chem. ( Winston-Salem N. C.) 1990 36 1466. 9112527. Clin. Physiol. Biochem. 1990 8 70. 9112541. Hrana Ishrana 1988 29 22. 9112548. J. Anal. Toxicol. 1990 14 235. 9112553. J. Environ. Sci. (China) 1990 2 109. 9112558. J. Nutr. Sci. Vitaminol. 1990 36 65. 9112563. J. Trace Elem. Electrolytes Health Dis. 1990 4 41. 9112566. Kanazawa Ika Daigaku Zasshi 1989,14 337.9112571.LaborPraxis 1989,13 1036,1038. 9112579. NATO ASISer. Ser. A 1988 158.9112593. Scand. J. Clin. Lab. Invest. 1990 50 441. 9112609. Z. Chem. 1990 30 204. 9112617. Arch. Androl. 1990 25 59. 9112626. Biochem. Biophys. Res. Commun. 1990 170 1216. 9112627. Biol. Met. 1990,3 45. 9112628. Biol. Trace Elem. Res. 1989 23 65. 9112629. Biol. Trace Elem. Res. 1990 24 25. 9112631. Clin. Chem ( Winston-Salem N. C.) 1990 36 1460. 9112633. ECO Not. Ecol. 1989 7(4) 30. 9112651. Bunseki Kagaku 1989 38 734. 9112653. Kexue Tongbao 1990,35 526.9112656. Kidney Int. Suppl. 1990 37(28) Sf 58. 9112663. LC Listy Cukrov. 1989,105(6) 12 1. 9112670. Neth. Milk Dairy J. 1989 43 477. 9112672. Nippon Iyo Masu Supekutoru Gakkai Koenshu 1989 14 10 1. 9112676. Pertanika 1989 12 3 1 3.9112680.Shandong Yike Daxue Xuebao 1990 28 30. 9112686. Trace Elem. Med. 1990 7 11. 9112694. Xi’an Yike Daxue Xuebao 1989 10 350. 9112695. Yingyang Xuebao 1990 12 63. 9112696. Yukagaku 1990 39 634. 9112710. J. Anal. At. Spectrom. 1990 5 669. 9112713. J. Anal. At. Spectrom. 1990 5 687. 9112715. J. Anal. At. Spectrom. 1990 5 693. 9112718. J. Anal. At. Spectrom. 1990,5 705.9112930. Anal. Proc. 1991 28 44. 9112941. Clin. Chem. 1990 36 1506. 9112942. Clin. Chem. 1990 36 1812. 9112945. At. Spec- trosc. 1990 11 11. 9112958. At. Spectrosc. 1990 11 229. 9112961. Appl. Spectrosc. 1990 44 1044.9112963. Mikro- chim. Acta 1989 3 3 15. 9112965. Mikrochim. Acta 1989,116R JOURNAL OF ANALYTICAL ATOMIC SPECTROMETRY APRIL 1992 VOL. 7 3 373.9112995. Lihua Jianyan Huaxue Fence 1990,26(4) 241.9113005.Guangpuxue Yu Guangpu Fenxi 1990,10(2) 72. 9113015. Guangpuxue Yu Guangpu Fenxi 1990,10(3) 69. 9113020. Hejishu 1990 13(6) 377. 9113100. At. Spectrosc. 1990 11 19-23. 9113101. At. Spectrosc. 1990 11 65. 9113102. At. Spectrosc. 1990 11 70. 9113103. At. Spectrosc. 1990 11(2) 75. 9113105. Anal. Sci. 1990 6 555. 9113150. Anal. Biochem. 1990 190 71. 9113151. Fresenius Z. Anal. Chem. 1989,335 9 14. 9113153. Fresen- ius J. Anal. Chem. 1990 337 853. 9113154. Fresenius J. Anal. Chem. 1990 337 877. 91f3158. Fresenius J. Anal. Chem. 1990 338 255. 9113168. Ann. Ist. Super. Sanita 1989 25 385. 9113171. Ann. Ist. Super. Sanita 1989 25 499. 9113175. Lab. Pract. 1989 38(3) 70. 91f3176. Lab. Pract. 1990 39(5) 56. 9113180. Analyst 1991 116 261.9113181. Analyst 1991 116 327. 9113186. J. Anal. At. Spectrom. 199 1,6 9.9113199. J. Anal. At. Spectrom. 199 1 6 115. 9113206. J. Anal. At. Spectrom. 1991 6 151. 9113211. J. Anal. At. Spectrom. 1991,6 173.9113214. Cline Chim. Acta. 1990 191 31. 9113215. Clin. Chim. Acta 1990 193 49. 9113216. Clin. Chem. Acta 1990 195 47. 9113217. Clin. Chim. Acta 199 1 196 7.9113228. Fresenius J. Anal. Chem. 1990,338 140.9113229. Fresenius J. Anal. Chem. 1990 338 287. 9113232. Fresenius J. Anal. Chem. 1990 338 455. 9113233 Fresenius J Anal. Chem. 1990 338 515. 9113234. Fresenius J. Anal. Chem. 1990 338 534. 9113239. Fresenius J. Anal. Chem. 1990 338 809. 9113263. Anal. Chem. 1990 62 1023A. 9113266. Anal. Chem. 1991,63 343. 9113268. Anal. Chem. 1991,63,423. 9113271. Spectroscopy (Eugene Oreg.) 1990 5(4) 49.9113275. Appl. Spectrosc. 1990 44 152 1. 9113288. Micro- chem. J. 1990 42 349. 9113289. Analusis 1990 18 562. 9113293. J. Radioanal. Nucl. Chem. 1990 138 103. 9113309. Talanta 1990 37 1 1 19. 9113319. Commun. Soil Sci. Plant Anal. 1990 21 1437. 9113320. Commun. Soil Sci. Plant Anal. 1990 21 1577. 9113323 Commun. Soil Sci. Plant Anal. 1990 21 1645. 9113338. Sci. Total Environ. 1989 89 343. 9113341. Fenxi Huaxue 1990 18 58 1. 9113345. Fenxi Huaxue 1990 18 741.9113350. Anal. Sci. 1990 6(7) 83. 9113358. Guangpuxue Yu Guangpu Fenxi 1990 10(4) 30. 9113378. Report 1989; Order No. PB90-129438 15 pp. Available from NTIS. Gov. Rep. Announce. Index (U.S.) 1990 90(5) Abstr. No. 009,884. 9113380. J. Invest. Dermatol. 1990 94 322. 9113383. J. Agric. Food Chem. 1990,38,2021.9113392. J. Chem. Soc. Pak. 1990 12 157. 9113407. Appl. Organomet. Chem. 1989 3 401. 9113410. Eisei Kagaku 1990 36 219. 9113411. Food Addit. Contam. 1990 7 425. 9113412. Fujian Shifan Daxue Xuebao Ziran Kexueban 1989,5(4) 61. 9113415. Gongye Weisheng Yu Zhiyebing 1989 15 229. 9113430. Pertanika 1989 12 303. 9113437. Rocz. Panstw. Zakl. Hig. 1989 40 274. 9113446. Z. Gesamte Hyg. Ihre Grenzgeb. 1990 36 489. 91f3461. Environ. Geochem. Health 1990 12 15. 9113462. Environ. Geo- chem. Health 1990 12 91. 9113463. Fett Wiss. Technol. 1990 92 322. 9113464. Gongye Weisheng Yu Zhiyebing 1990 16 106. 9113467. Int. Arch. Occup. Environ. Health 1990 62 85. 9113473. Khim.-Farm. Zh. 1990 24(9) 75. 9113479. Nippon Kaisui Gakkaishi 1990,44,262.91/3497. Tech. Sci. Methodes Genie Urbain Genie Rural 1990 (2) 81. 9113501. Z. Lebensm-Unters. -Forsch. 1990 190 331. 9113504. Aerztl. Lab. 1990 36 330. 9113509. Anal. Biochem. 199 1 192 434.9113514. Ber. Kernforschungsan- lage Jiilich 1989 (2301) 28 pp. 9113520. Chem. Speciation Bioavailability 1990 2 3.9113524. Eisei Kagaku 1990,36 21. 9113525. Eisei Kagaku 1990 36 20. 9113529. Food Chem. Toxicol. 1990 28 521. 9113536. Hebei Shifan Daxue Xuebao Ziran Kexueban 1990 (2) 62. 9113552. Mitt. Geb. Lebensmittelunters. Hyg. 1990 81 3 19. 9113561. Rev. Roum. Chim. 1989,34 1849.9113562. Rocz. Panstw. Zakl. Hig. 1990 41 35. 91/3564. Shigaku 1990 78 251. 9113565. Shipin Yu Fajiao Gongye 1990 (3) 58. 9113574. Vestn. Mosk. Univ. Ser. 2 Khim. 1990 31 513. 9113575. Wear 1990 140 49. 9113579. Yaowu Fenxi Zazhi 1990 10 173.9113585. Clin. Chem. 1990,36,2011. 9113591. J. Anal. At. Spectrom. 1991 6 225. 9113592. J. Anal. At. Spectrom. 1991 6 233. 9113594. J. Anal At. Spectrom. 1991 6 69R. 9113597. Analyst 1991 116 505. 9113605. Can. J. Appl. Spectrosc. 1991 36 15. 9113766. Analyst 1991 116 549. 9113775. J. Anal. At. Spectrom. 1991 6 277. 9113776. J. Anal. At. Spectrom. 1991,6 283. 9113777. J. Anal. At. Spectrom. 1991 6 289. 9113779. J. Anal. At. Spectrom. 1991 6 301. 9113785. J. Anal. At. Spectrom. 199 1,6 335. 9113791. Fresenius J. Anal. Chem. 1990 337 721. 9113804. Spectrochim. Acta Part B 1991 46 103 9113818. X-Ray Spectrom. 1990 19,285.9113829. Nucl. Instrum. Methods Phys. Res. Sect. A 1990 299 589. 9113838. Guangpuxue Yu Guangpu Fenxi 1990 10( 5) 76. 9113843. J. Assoc. 08 Anal. Chem. 1991,74 27.9113844. J. Assoc. 08 Anal. Chem. 1991 74 32. 9113865. Bunseki Kagaku 1991 40 39. 9113898. Talanta 1991 38 175. 9113913. J. Toxicol. Environ. Health 1991 32 11 1. 9113956. ACS Symp. Ser. 1991 445 181. 9113968. Biol. Trace Elem. Res. 1991 28 109. 9113974. Colloq. Atomspektrom. Spurenanal. 5th 1989 657. 9113975. Col- loq. Atomspektrom. Spurenanal. 5th 1989 747. 9113977. Czas. Stomatol. 1990 43 3 16. 9113978. Eisei Kagaku 1990,36 430. 9114000. Hejishu 1990 13(9) 540. 9114006. lgaku no Ayumi 1990 154 135. 9114008. Indian J. Phys. B 1991,65 30.9114011. J. Pharm Biomed. Anal. 1990,8 655. 9114019. Nippon Kagaku Kaishi 1991 (2) 120. 9114022. Pak. J. Sci. Ind. Res. 1990 33 85. 9114030. S‘angyo Igaku 1990,32,2J2.9114031. Sangyo Igaku 1990 32 276. 9114038. Toyama-ken Eisei Kenkyusho Nenpo 1989 13 230. 9114040. Trace Elem. Med. 1990 7 118. i9114045. 2. Naturforsch. C Biosci. 1990 45 1250. ‘9114046. Zentralbl. Gynaekol. 1990,112 1175.92113. Clin. Chem. 1991 37(3) 461. 92114. Clin. Chem. 1991 37(4) 515. 92118. Appl. Opt. 1991 30(13) 1723. 92123. Nucl. lnstrum. Meth. Phys. Res. 1990,47 79. 92136. Am. J. Clin. Nutr. 1991 53 755. 92141. Anal. Chim. Acta 1990 237 181. 92149. Anal. Chim. Acta 1991 242 203 92166. Biol. Trace Elem. Res. 1990 26 149. 92171. Biol. Trace Elem. Res. 1990 26 363.92172. Biol. Trace Elem. Res. 1990,26 407. 92/73. Biol. Trace Elem. Res. 1990 26 553 92174. .Biol. Trace Elem. Res. 1990 26 589. 92176. Biol. Trace .Elem. Res. 1990 26 691. 92177. Biol. Trace Elem. Res. 1990,26 75 1. 92178. Biol. Trace Elem. Res. 1990,26 757. 192180. Colloq. Atomspektrom. Spurenanal. 1 989 5 439. ‘92182. Colloq. Atomspektrom. Spurenanal. 1989 5 483. !92183. Colloq. Atomspektrom. Spurenanal. 1989 5 49 1. !?2184. Colloy. Atomspektrom. Spurenanal. 1989 5 501. !92187. Colloq. Atomspektrom. Spurenanal. 1989 5 7 1 1. !92190. Colloq. Atomspektrom. Spurenanal. 1989 5 737. 92/91. Colloq. Atomspektrom. Spurenanal. 1989 5 755. !92192. Colloq. Atomspektrom. Spurenanal. 1989 5 789. !92194. Colloq. Atomspektrom. Spurenanal. 1989 5 855. !92195. Colloq. Atomspektrom. Spurenanal. 1989 5 863. !92196. Colloq. Atomspektrom. Spurenanal. 1989 5 887. 92197. Anal. Chem. 1991 63 890. 921103. Analyst 1991 1116 541. 921106. Anal. Lett. 1991 24 153. 921121. Fenxi Huaxue 1990 18( lo) 982. 921127. Fenxi Huaxue 1990 lM(1 I) 1064. 921129. Talanta 1991 38 167. 921147. Zh. .4nal. Khim. 199 1,46 292.921153. Fenxi Shiyanshi 1990 !#> 71. 921154. Fenxi Shiyanshi 1991 10 30.921193. Nucl. instrum. Methods Phys. Rex Sect. B 1991 54 98. 921208. J. Radioanal. Nucl. Chem. 199 1 153,257.921209. J. Trace iWicroprobe Tech. 1989 7 247. 921210. J. Anal. Toxicol. 11991 15 66. 921211. J. Trace Elem. Electrolytes Health Dis. 1990 4 127. 921212. J. Pharm. Biomed. Anal. 1991 9 19 1. 921213. J. Agric. Food Chem. 199 1,39,724.921214. J. Agric. Food Chem. 199 1,39 887.921215. J. Am. Geriatr.JOURNAL OF ANALYTICAL ATOMIC SPECTROMETRY APRIL 1992 VOL. 7 117R SOC. 1990 38 633. 921218. J. Pharmacol. Methods 1990 24 327. 921232. Bol. Acad. Galega Cienc. 1989 8 19. 921237. Chem. Aust. 1990 57 94. 921246. Clin. Chim. Acta 199 1 197 14 1. 921250. Environ. Geochem. Health 1990 12 59. 92/255. Haeyang Yongu 1989 11 1.921259. Indian J. Environ. Prot. 1990 10 7. 921260. Kanagawa Shigaku 1990,24 653.921274. C. R. Seances SOC. Biol. Ses Fil. 1989 183 522. 921279. Eisei Kagaku 1991 37 22. 921285. Hebei Shijian Dame Xuebao Ziran Kexueban 1991 1 63. 921297. Mar. Pollut. Bull. 1991,22,67.921299. Neurosci. Lett. 1991 123 61. 92/307. Phys. Med. Biol. 1991,36 439. 921311. Rev. Roum. Biochim. 1990,27,209. 921312. Rostl. Vyroba 1990 36 387. 921315. Sci. Total Environ. 1991 103 209. 921318. Shigaku 1990 78 259. 921319. Shokuhin Eiseigaku Zasshi 1990 31 382. 92/335. Z. Lebensm.-Unters. Forsch. 1991 192 136. 921338. Zhongguo Yaoke Daxue Xuebao 199 1 22( 1 > 33.