JOURNAL OF ANALYTICAL ATOMIC SPECTROMETRY, APRIL 1986, VOL. 1 29R ATOMIC SPECTROMETRY UPDATE-CLINICAL MATERIALS, FOODS AND BEVERAGES Alistair A. Brown Pye Unicam Ltd., York Street, Cambridge CB7 2PX, UK David J. Halls Trace Metals Unit, Department of Biochemistry, Royal Infirmary, Castle Street, Glasgow G4 OSF, UK 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 Summary of Contents 1 Clinical Analysis 1.1, Application of Inductively Coupled Plasmas in Clinical and Biological Analysis 1.2. New Ideas and Developments 1.3. Progress for Individual Elements 1.3.1. Aluminium 1.3.2. Arsenic 1.3.3. Cadmium 1.3.4. Calcium and Magnesium 1.3.5. Chromium 1.3.6.Cobalt 1.3.7. Copper and zinc . . 1.3.8. Iron 1.3.9. Lead 1.3.10. Manganese 1.3.1 1. Mercury 1.3.1 2. Nickel 1.3.13. Platinum 1.3.14. Rubidium 1.3.15. Selenium 1.3.16. Thallium 1.3.17. Vanadium 1.4. Conclusions Table 1. Summary of An lyses of Bod! Fluids and Tissues 2 Analysis of Foods and Beverages 2.1. Sample Preparation 2.2. Analytical Techniques 2.3. Reference Materials and Quality Control Programmes 2.4. Topical Applications 2.5. Conclusions Table 2. Summary of Analyses of Foods and Beverages This review is the second Atomic Spectrometry Update and describes developments in the analysis of clinical materials, foods and beverages. It is based upon publications and conference reports, received during the period September 1st 1984 to August 31st 1985.The references cited, prefixed by S/ or, for conference reports, S/C, may be found in the supplement distributed to subscribersjo JAAS. References prefixed 86/ will appear in Volume 1 of JAAS. The format of tables is similar to that used in A a d a I Reports on Analytical Atomic Spectroscopy. In the tables, in addition to the abbreviations listed elsewhere, Hy is used to show where hydride generation was employed and S, L and G in the "Analyte form" column signify solid, liquid or gaseous sample introduction, respectively. At a first glance (see tables) it is apparent that the analysis of clinical materials attracted the majority of research interest. This may be related to the severe difficulties encountered in the analysis of these samples especially for the determination of those elements which are present at microgram per litre levels in body fluids and tissues.In contrast, the analysis of foods and beverages received little attention and this is reflected in the contributions to this subject. The authors have tried to present a critical appraisal of the reviewed literature. Readers who wish to comment on the text are welcome to express their views through the letters section of JAAS. The authors would welcome any comments or suggestions for improvements to future reviews.30R JOURNAL OF ANALYTICAL ATOMIC SPECTROMETRY, APRIL 1986, VOL. 1 1. CLINICAL ANALYSIS For the period of this review, two particular areas are prominent. One is the large number of papers on ICP-OES produced as new and experienced users try to justify a role for it in the clinical laboratory.This is reviewed separately. The second is the large number of papers on Al, which surely qualifies it for the title “Element of the Year.” Interesting new developments have occurred in FAAS, ETA-AAS and ETA-AES for clinical applications. After these developments are reviewed, progress for individual elements is summarised. 1.1. Application of Inductively Coupled Plasmas in Clinical and Biological Analysis The large number of recent papers shows an increasing interest in the use of inductively coupled plasma optical emission spectrometry in this field (TableJ). Some of the work is exploratory [e.g., Ca in serum (WOO) and Cd and Pb in whole blood (S/176)] and appears to offer little advantage over existing AAS techniques.The true multi-element capability of ICP-OES is realised in studies such as that of Smith et al. (S/592) on the determination of 12 metallic elements and B in teeth enamel and dentine and that of Wanot et al. (S/981) who determined Ca, Mg and P in stones. One advantage of the ICP-OES technique is the ability to determine non-metals such as B, P and S. This, together with the ability to determine Ba, Bi, Ca, Mg and Sr with better sensitivity than existing techniques, is cited by Allain and Mauras (S/C417) as reasons for the spread of the technique. For the last five years, they have routinely used a simultaneous Jobin-Yvon 48 spec- trometer with 35 lines for the determination of Al, B, Ba, Ca, Cu, Fe, K, Mn, P, S, Si and Sr in whole blood, plasma, urine, dialysis fluids and tissue samples.The advantages of ICP-OES in serum analysis seem somewhat limited in that only a few of the important elements have high enough concentrations to be determined. Kohl- meier and Diehn (S/C530) introduced the technique into their laboratory for the simultaneous determination of Cu, Fe and Zn giving a between-batch RSD of less than 5% for all elements. Deijk et al. (MOO) compared different pre- treatment methods for the determination of Ca in a serum reference material (Precinorm U). Direct dilution gave results which agreed with high-pressure digestion and low- temperature ashing. Pre-concentration can be used to extend the range of elements. Trace elements in digested samples can be chelated on a poly(dithi0carbamate) resin which is then digested with H202 and HN03.This procedure was applied by Mianshi and Barnes (S/984) to the determination of Cd, Co, Cu, Fe, Mo, V and Zn in serum; As and Se were determined by hydride generation ICP-OES. With all these steps, the possibilities of contamination increase and this was reflected in unrealistically high values for Mo (20.7 yg 1-1) and V (98.5 pg 1-1) in a normal serum. As others have shown [e.g., Kollmeier and Diehn (S/C530)], it is not necessary to pre-concentrate for Cu, Fe or Zn. An alternative approach is to use electrothermal atomisation. Matusiewicz and Barnes (S/974) determined A1 and Si by injecting 5 yl of serum, urine or dialysate fluid on to a Varian-Techtron CRA90 electrothermal atomiser for atomisation into a Plasma-Therm ICP.Detection limits of 1.5 and 500 yg 1-1 were achieved for A1 and Si, respectively. To determine Cd and Pb in 5 ml of whole blood, Deijck et al. (S/176) tried three different approaches: the matrix-modification procedure of Stoeppler et al. (Analyst, 1978, 103, 714), low-temperature ashing and dry ashing. The last two techniques were applied with pre-concentration using 8-hydroxyquinoline immobilised on controlled-pore glass. Final determination was by microsampling of the analyte essentially free from the matrix. While dry ashing followed by pre-concentration gave reasonable results, it was recom- mended that to determine just these elements alone, AAS or electrochemistry was simpler. The technique of ICP-OES seems more at home with the analysis of materials such as soft tissue (S/172, S/C455, S/613, S/928), hair (S/506), nails (S/C455 , S/C679), teeth (S/592), stones (S/981) and bone (S/193, S/C455, S/613).Procedures generally involve wet digestion with acid mixtures (S/193, S/928, S/981); some workers prefer pressure digestion (S/172, S/506, S/592). Chen et al. (S/506) compared dry ashing at 500-600 “C with HN03 digestion under pressure followed by a further HF - HC104 digestion. Dry ashing at 500 “C gave comparable results to acid digestion but losses of A1 and Mg were found on ashing at 600 “C. Good analytical performance for the analysis of digested material is generally achieved for a wide range of elements. For example, Marquardt et al. (S/193) achieved RSDs of 0.5-1.6% for P; 1-3% for Ba, Ca, Cu, Fe, Mg, Mn, Sr and Zn; and 4 9 % for K and Na.Sulphur determination by ICP-OES using a VUV spec- trometer at 180.7 nm has been investigated by Morita et al. (S/921). Standard reference materials (serum, hair, liver, mussel, Chlorella and orchard leaves) were analysed with satisfactory accuracy following a high-pressure digestion with HN03. The detection limit was about 10 yg 1-1. Inductively coupled plasma mass spectrometry has yet to make its impact on clinical or biological analysis; only one p a s x (S/C294) discussed a biological analysis, oyster tissue. 1.2. New Ideas and Developments The greater speed and convenience of flame atomic absorption spectrometry compared with ETA-AAS make it the preferred technique whenever sensitivity and sample size allows. An interesting development in this area was the use of a slotted quartz tube in the flame by Brown et al.(S/198,86/193) to give greater sensitivity allowing the determination of Cd, Cu and Pb in urine and Cd and Pb in blood after deproteinisation with TCA. Devitrification of the quartz tube by urine samples was reduced by coating the tube with La203. The technique has also been applied to the determination of Cu and Zn in serum (S/996) where the greater sensitivity allows greater dilution of the sample (1 + 20) and a reduction of the sample volume, if necessary, down to 20 yl. For the determination of Cu and Zn in fractionated blood plasma, Bahreyni-Toosi et al. (S/999) looked at various ways to increase sensitivity and decrease sample requirement: pulse nebulisation, the slotted quartz tube with pulse nebulisation and atomisation from a Delves cup and from a Mo wire loop.They found only modest improvements with the slotted quartz tube, preferring direct pulse nebulisation for their particular determination. The range of analyses possible by FAAS can also be extended by pre-concentration. Platzer et al. (S/C514) enriched trace metals from 100 ml of sample by using dithiocarbamate groups bound to a solid carrier. The dithio- carbamate chelates were then volatilised directly into the burner of an FAAS instrument. Detection limits for Cd, Co, Cu, Ni and Pb were between 0.02 and 0.2 yg 1-1. Realistic applications of this technique are awaited. Rocks and co-workers have coupled flow injection with FAAS for the determination of Ca and Mg in serum (S/964) and have taken the technique one step further with a new development called “controlled-dispersion analysis’’ (CDA) (Analyst, 1985, 110, 493).In this, a computer-controlled peristaltic pump takes up a fixed volume of sample. The sample probe then moves to the carrier solution and the pump propels the sample slug within the carrier stream to the spectrometer. This eliminates sample wastage as in conven- tional FI sampling systems. Controlled-dispersion analysis has been applied to the determination of Ca, Cu, Li, Mg and Zn in serum. If electrothermal atomic absorption spectrometry is to cope with increasing workloads in clinical laboratories, it must runJOURNAL OF ANALYTICAL ATOMIC SPECTROMETRY, APRIL 1986, VOL. 1 31R Table 1. SUMMARY OF ANALYSES OF BODY FLUIDS AND TISSUES Element h/nm A1 A1 A1 A1 A1 A1 A1 A1 A1 A1 A1 A1 A1 A1 A1 A1 A1 As As As As Ca Ca Ca Ca Cd Cd Cd Cd Cd Cd - - - - - - - 309.3 - - 308.2 - - 309.3 - 309.3 309.3 - - - - 393.37 - - 315.9 228.8 220.35 228.8 - 228.8 - Matrix Concentration Dialysis fluids 4-67 pg I-' Serum 6-220 pg I-' Biological samples, - acid rain Serum 5-50 pg 1-1 Bone - Serum, blood, 2-77 pg 1-1 urine and tissues Serum Haemodial ysis concentrates Serum, hair Body fluids Biological Serum materials Biological samples Dialysis fluids Biological Serum materials Biological materials Body fluids, foods / Marine biological tissues Dietary supplements Biological samples Serum Tissue Serum Human stones Urine Blood Biological materials Biological materials Liver tissue Hair 1-150 pg I-' - 9-300 pg 1-1 0-80 pg 1-1 0-100 pg I-' 0-100 pg I-' 0.3-300 pg 1-1 1-120 pg 1-1 - - - - - - - - - - - 0-2 pgl-1 - - - - - Technique; a tomisa tion ; analyte form AA; ETA; L AA; ETA; L AA; ETA; L AA; ETA; L AE; ICP; Lor AA; F, AA; ETA; L NzO - CzH2; L AA; ETA; L AA; ETA; L AA; ETA; L AA; ETA; L AE; ICP with ETA; L AA; ETA; L AE; ICP; L AA; ETA; L AA; ETA; Lor AA; ETA; L AE; ETA; L NAA AA; Hy; L AA; Hy; L AA; ETA; L AA; Hy; L AE; ICP; L AA;-;L AA; F, air - C2H2; L AE; ICP; L AA; ETA; L AE; ICP; L AA; ETA; S AA; ETA; L AA; ETA; L or S AA; -; L Sample treatment Reference ETA-AAS with L'vov platform ETA-AAS with Zeeman background correction ETA-AAS with metal atomiser and Zeeman background correction Samples analysed directly or with dilution, graphitelzr-coated graphite furnace Comparison of ICP-OES versus FAAS; both rechniques gave reliable results, ICP-OES gave better detection limit Liquid samples diluted with Triton X-100, tissues wet ashed Samples diluted 1 + 1 with H20, platform atomisation, integrated absorbance, matrix modifications plus O2 ashing Concentrated NO3 added as matrix modifier Matrix destroyed by O2 ashing in a Zr-coated graphite tube at 500 "C Samples incubated with 2,4-pentanedione, extracted with 4-methyl-2-pentanedione None Comparison of a direct analysis using L'vov platform and matrix modification with a protein precipitation method internal standard HN0, as matrix modifier Matrix modification with Cs, Ga used as Matrix effects overcome by addition of 2% Comparison of ETA-AAS versus NAA Optimisation of gases, tube materials, wavelengths and matrix modifiers Samples digested in HNO, Total As determined after wet ashing with HNO, - HC104 - H2S04 Also includes determination of Hg and Se; samples digested in pressure bomb with HNO, for Hg; digested further with HC104 - H2S04 for As and Se determination Metals separated from the matrix by solvent extraction with APDC - IBMK system Wet digestion with HNO, - H2S04 - K2Cr207; both organic and inorganic As compounds determined High-pressure ashing, low-temperature ashing and direct dilution of sample investigated; simple dilution preferred Content and intracellular distribution of Ca and Mg studied Flow injection coupled with FAAS, 4 pl of sample injected into a flowing, non-segmented reagent stream Samples digested in HN03 - HC104 Standard-additions procedure; samples Comparison of digestion procedures Solid sampling on to a graphite boat; Zeeman background correction Applications of O2 ashing in ETA-AAS acidified with HN03 Comparison of solid sampling and Zeeman Samples soaked in neutral detergents, ETA-AAS versus a wet-digestion procedure washed with de-ionised H20, treated with 95% ethanol; hair finally digested with 7 ml HN03 + 3 ml HC104, evaporated, re-dissolved in 0.1% Na2S04 Sl118 9192 SIC268 s1479 Sl480 s1642 Sl858 Sl965 Sl967 s/973 s1974 Sl985 86147 86165 86lC 160 861195 861197 Sl166 S/590 Sl863 Sl976 s1100 S1647 Sl964 Sl981 s1159 Sl176 Sl182 SIC553 SIC555 Sl58932R JOURNAL OF ANALYTICAL ATOMIC SPECTROMETRY, APRIL 1986, VOL.1 Table 1.SUMMARY OF ANALYSES OF BODY FLUIDS AND TISSUES-continued Element Cd Cd Cd Cd Cd Cd Cd c o c o Cr Cr Cr Cr Cr c u c u c u c u c u c u c u Fe Fe Fe Hg Hg Hg Hi? Hg K hJnm - 228.8 228.8 - 228.8 or 326.1 228.8 - - - - - - - 357.9 324.7 - 324.7 - - 324.7 324.7 - 248.3 - 253.7 - - - - 404.4 or Matrix Concentration Dietary - supplements Urine ' - Urine 0.1-164 yg 1-' Teeth - Pancreatic tissue - Urine - Liver, renal cortex - Plasma 0.15-10 1.18 I-' Whole blood, - urine Hair - Plasma, urine 0-3 yg 1-1 Serum - Enteral nutritional - solutions Faeces Plasma protein fractions Muscle tissue Liver Liver Serum Serum, urine Plasma protein fractions Liver Liver Serum Blood, urine Biological materials Biological materials Marine biological tissues Blood, fish tissue Biological 766.5 samples - - - 0-1000 yg 1-1 - - - - 0-5 yM 0-1500 yg 1-1 - 1.7-5 pg 1-1 - - - (Moo pgl-1 1-600 yg ml-1 Technique ; atomisation; analyte form AA; ETA; L AA; ETA; L AA; ETA; L AA; ETA; L AA; ETA; S AA; ETA; L AA; ETA; L AA; ETA; L AA; ETA; L AA;-;L AA; ETA; L AA; ETA; L AA; -; L AE; ICP; L AA; ETA; L AA; F, air - AA; F, air - CzH2; L C2H2; L AA; -; L AA; ETA; L AA; F, air - CzH2; L C2H2; L AA; F, air - AA; F; L, or AA; ETA; L AA; F, air - AA; ETA; L AA; cold vap.; L AA; cold vap. ; L C2H2; L AA; ETA; S AA; cold vap. ; L AA; cold vap.; L or AE; DCP; L AA; F, air - C2H2; L Sample treatment Reference See As, ref. Sl863 Investigation of various matrix modifiers, HN03 was preferred Samples diluted 1 + 1 with HzO, standard- additions method and L'vov platform Samples wet digested, solvent extracted into CHC13 then injected into graphite furnace Samples incubated at 37 "C with physiological saline, freeze dried, solid sampling of 1-13 pg sample masses into graphite furnace Comparison of calibration procedures Investigation of Cd body burden between non- , medium- and heavy-smokers Samples digested with HN03 - HC104; solvent extraction with APDC - IBMK; direct injection into graphite tube Samples homogenised then direct injection into graphite furnace; results compared with an electrochemical procedure See Cd, ref.Sl589 Standard-additions procedure Samples dry ashed with Mg(N03)z, re-dissolved in 0.1 M HC1 Wet ashing of samples Samples digested with HN03 - HC104 Comparison of 5 different graphite furnace 0.8-12.0 mg of sample extracted with HN03 systems Samples digested with HN03 - H2S04 for 1 h at 120 "C, diluted to 500 pl Samples digested with concentrated HN03 at 60 "C; collaborative study Samples diluted 1 + 2 1 with 1% Triton X-100, M HNO, and 30 mM NH4 NO3; platform atomisation plus peak-area measurement enabled calibration with acid standards; continuum source AAS (SIMAAC) slotted quartz tube; sensitivity improvement Samples diluted 1 + 20, aspirated into a X 2-3 Comparison of FAAS techniques for the determination of Cu and Zn in plasma protein fractions; techniques included pulse nebulisation, slotted quartz tube and Delves cup Comparison of 3 analytical methods (FAAS, ETA-AAS, colorimetric analysis) for the determination of Fe in liver biopsies See Cu, ref.Sl785 See Cu, ref.Sl987 Automated - computerised procedure for Hg determination Comparison of digestion procedures, semi- closed or closed silica vessels required; digest with HN03 - HC104 at 200 "C, time required varied from 1.5 to 8 h depending on sample See Cd, ref. Sl182 See As, ref. Sl590 Various digestion procedures including HNO,, HN03 - H2S04 and H2SO4 - KMn04 1 + 99 with 1% CsCl Solid samples dry ashed, samples diluted Sl863 Sl963 Sl966 Sl970 Sl975 9998 86/60 Sl982 Sl994 Sl589 Sl615 Sl616 Sl789 Sl980 Sl184 S1645 Sl987 Sl923 Sl785 Sl996 Sl999 Sl646 Sl785 Sl987 Sl2 9157 SO82 Sl590 Sl904 Sl787JOURNAL OF ANALYTICAL ATOMIC SPECTROMETRY, APRIL 1986, VOL. 1 33R Table 1. SUMMARY OF ANALYSES OF BODY FLUIDS AND TISSUES--continued Element Mg Mg Mg Mg Mn Mn Mn Na Ni Ni Ni Pb Pb Pb Pb Pb Pb Pb Pb Pb Pb Pb Pb Pb Pb Pb Pt Pt Rb Rb S Se Se Se hlnm - - - 323.83 - 279.5 403.08 330.2 or 589.0 - - 232.0 - 283.3 220.35 - - - - - - - - - - 283.3 - - - 780.0 - 180.7 - 196.0 196.0 Matrix Concentration Whole blood, - serum, urine Muscle - Serum - Human stones - Whole blood - Urine - Urine - Biological samples 1-600 pg ml-1 Rat tissue Serum Serum, water Blood, melted Blood snow Blood Biological materials Blood, serum, urine, tissues Tissue Biological Tissue Hair Whole blood Urine materials Dietary Whole blood Whole blood supplements Whole blood Animal tissue Plasma ultrafiltrate, urine Blood, plasma Erythrocytes, Dlasma Biological samples - Whole blood - Body fluids 5-100 pg I-' Marine biological - tissues Technique ; atomisation; analyte form AF; F; L AA;-;L AA; F; L AE; ICP; L AA; ETA; L AA; F, air - C2H2; L AE; ETA; L AA; F, air - AE; ICP; L AA; ETA; L C2H2; L AA; ETA; L AA; ETA; L AA; ETA; L AE; ICP; L AA; ETA; S AA; ETA; L or S AA; ETA; S AA; ETA; L AA; ETA; S AA; F; L AA; ETA; L AA; ETA; L AA; ETA; L AA; ETA; L AF; F, air - C2H2; L AA; ETA; L AA; ETA; L AA; ETA; L AE; F, air - C2H2; L AA; ETA; L AE; ICP; L AA; ETA; L AA; Hy; L AA; Hy; L Sample treatment Reference Whole blood and serum deproteinised with 1 M HNO,; urine samples acidified with 0.04 M HCI; 200 PI sample aliquots aspirated into flame; line and continuum source AFS See Ca, ref.Sl647 See Ca, ref. Sl964 See Ca, ref. Sl983 Dilute 1 + 4 with 0.5% Triton X-100, pyrolytic Coprecipitate Mn with La(OH)3, re-dissolve; Samples acidified with 10-2 M HN03, 1 + 1 graphite tube, platform atomisation direct aspiration or 1 + 5 dilution, acidified aqueous standards for calibration See K.ref. Sl787 Study of Ni uptake in the rat Matrix modification with NH4 oxalate, standard additions for calibration Minimal sample preparation Samples introduced into a pre-heated furnace on a W-wire probe Sample added to solution containing 1 M HN03 plus 0.1% Triton X-100, centrifuged; 10 or 20 pl of supernatant injected into graphite furnace See Cd, ref. Sl176 See Cd and Hg, ref. Sl182 Radiotracer study on the usefulness of various matrix modifiers; platform atomisation and NH4H2P04 recommended Slurry atomisation into graphite furnace after homogenisation of tissue samples See Cd, ref. SIC553 See Cd, ref. SIC555 See Cd, ref. S1589 Use of O2 ashing in the graphite furnace Determination of Et2Pb in urine, chelation with glyoxalbis(2-hydroxyanil), extraction into IBMK See As, ref.Sl863 Review on the use of reference samples rather than reference methods Samples diluted 1 + 9 with H20, fluorescence excitation at 283.3 nm, measured at 405.8 nm, laser-excited FAFS method Studies on the storage of blood samples for Pb determination Wet digestion of samples, interference from HN03 removed by addition of NH3 to the graphite furnace during the ash stage Samples reacted with NaDDC, extracted into CHC13, direct insertion into graphite furnace solution containing 2.5 g 1-1 Cs, standards matrix matched with inorganic constituents Blood diluted 1 + 99, plasma 1 + 9, with a - High-pressure decomposition with HNO, Study on the use of O2 ashing, Ni matrix modification and the interference of Fe on Se Samples digested in HNO, - HC104 - H2S04 with a final temperature of 310 "C See As, ref.Sl590 s17 Sl647 s1964 Sl981 SIC528 s1977 8612 1 1 9787 Sl187 Sl196 Sl810 Sl28 Sl158 Sl176 Sl182 S/191 SIC246 SIC553 SIC555 Sl589 SIC69 1 Sl811 Sl863 Sl895 s1995 Sl1200 s1979 Sl986 Sl497 Sl857 s1921 SIC27 1 s1494 s159034R JOURNAL OF ANALYTICAL ATOMIC SPECTROMETRY, APRIL 1986, VOL. 1 Table 1. SUMMARY OF ANALYSES OF BODY FLUIDS AND TISSUES--continued Element Se Se Se Si T1 Tl Zn Zn Zn Zn Zn Zn Zn Zn Zn Zn Zn Various Various (7) (6) (11) (18) (11) (7) Various Various Various Various Various ( 5 ) Various Various Various (11) (12) (4) Various (17) Various (13) Various Various (21) hlnm - - 196.0 250.7 276.8 276.8 213.9 213.9 213.9 - - 213.9 213.9 - 213.9 213.9 213.9 - - - - - - Matrix Liver Enteral nutritional solutions Serum Serum, urine, haemodialysis fluids hair Blood, urine, Urine Serum Tissue Red blood cells Plasma protein fractions Blood cells Muscle Liver Serum Serum, urine Serum, plasma Plasma Biological materials Animal tissue Hair Biological samples Biological samples Serum Human placenta Brain tissue, bone, nails Blood, plasma, urine Biological materials Hair Teeth Biological material Nails Technique; atomisation; Concentration analyte form - AA; Hy; L - AA;-; L 0.01-0.1 pg ml-l AA; ETA; L 0-100 pg 1-1 0-200 pg 1-1 0-1.65 yg 1-1 - 40-400 pg 1-1 - - - - 0-1000 pg 1-1 - - - - - - - - - - AE ;' ICP ; L AA; ETA; Lor AA; ETA; S AA; ETA; L AA; ETA; L AA; F, air - CzHz; L or AA; ETA; L AA;-;L AA; ETA; L AA; F, air - C2H2; L AA; F, air - W z ; L AA; F, air - C2H2; L AA; ETA; L AA; F, air - CzHz; L AA; F, air - CzH2; L AA; F, air - CzH2; L AE; DCP; L AE; ICP; L AA; For ETA; L AE; ICP; L AE; ICP; L AA; ETA; L AA; F; L AE; ICP; L AE; ICP; L AE; DCP, Hy; G AE; ICP; L AE; ICP; L AE; ICP; L AE; ICP; L Sample treatment Reference Samples digested in HNO, - HC104 with a See Cr , ref.Sl789 final temperature of 210 "C Addition of 1 YO Ni as matrix modifier See Al, ref. 9974 Blood and urine samples diluted with 0.1 M HN03, hair analysed directly by solid sampling Urine pH adjusted to pH 7 , NaDDC added, extracted into toluene Serum proteins separated by gel-filtration chromatography and affinity chromatography, Zn associated with az- macroglobulin Samples digested in HNO, Samples diluted 1 + 19 with HzO See Cu, ref.Sl184 Blood samples separated on a discontinuous See Cu, ref. Sl645 gradient of colloidal Percoll See Cu and Fe; ref. Sl785 See Cu and Fe, ref. Sl987 See Cu. ref. Sl996 Comparative study of FAAS methods for the determination of Zn in serum and plasma See Cu, ref. Sl999 Optimisation of parameters for d.c. arc Samples decomposed by dry ashing at 700- excitation (Ag, Li, Mn, Ni, Pb, V and Zn) 800 "C or wet digested with HN03 - HC104 (Ce, La, Nd, Pr, Sm and Y) Multi-element analyses of hair samples Samples digested with HNO, - HC104 Samples digested with HN03 - HC104, Co used as internal standards Simultaneous multi-element determination of trace elements by SIMAAC, 2-ml serum sample dry ashed with Mg(N03)2, re- dissolved with 0.5 ml Of 0.5% HNO,, platform atomisation (Al, Co, Cr, Mn, Mo, Ni and V) Optimisation of digestion procedures; H2S04, HNO,, HN03 - HZSO4 all gave good results (Ca, Cu, Fe, Mn and Zn) - Advantages of ICP-OES in the clinical laboratory Continuous-flow hydride or cold vapour generation with d.c.arc determination (As, Se, Te and Hg) Samples digested in HNO, - HF - HC104 Teeth cleaned with H202, digested in Review pressure bomb with HN03 International comparison of the concentration of 21 elements in human nails SIC680 Sl789 Sl968 s1974 Sl160 9643 s1103 s1104 Sl106 Sl184 9617 Sf645 9785 9987 Sl996 s1997 s1999 S18 Sl108 Sl156 s1172 s1193 s1194 s1195 SIC455 SIC417 SIC473 Sl506 Sl592 Sl613 SIC679JOURNAL OF ANALYTICAL ATOMIC SPECTROMETRY, APRIL 1986, VOL.1 35R Table 1. SUMMARY OF ANALYSES OF BODY FLUIDS AND TISSUES-continued Element hlnm Various - Various - (20) Various - Various - (7) (4) Various - ( 5 ) Various - (9) Various - Various - Various - (4) (4) (4) Various - (6) Various - (4) Matrix Biological materials Blood, urine Animal tissue Whole blood, serum, urine Hair, urine Serum Technique; atomisation; Concentration analyte form - AE; ICP; L - AA; For ETA; L - AA; F, air - CZHZ; L AA; ETA; L Serum, urine - Lung tissue - Platelets - Urine - Blood, serum, - urine AE; ICP; L AA; ETA; L AA;-;L AA; F; L AE; ICP; L AA; F, air - C2H2; L Sample treatment Reference Multi-element analysis in veterinary medicine, review Electrothermal vaporisation for ICP-OES, review Automated wet digestion of samples (As, Cd, Cu, Hg, Pb, Se and Zn) General review on the advantages of the slotted quartz tube and FAAS (Cd, Cu, Pb and Zn) in PTFE bomb, platform atomisation and matrix modification recommended (As, Cd, Cr , Pb and Se) Samples digested in HNO, - HCIO, in pressure bomb, pre-concentrated on ion- exchange resin, resin digested with H202 - HNO, (As, Cd, Co, Cu, Fe, Mo, Se , V and Zn) Improving analysis time of ETA-AAS analytical methods (Al, Cd, Cu and Pb) Review of digestion procedures (Al, Ga, Mg and Si) Samples digested in sealed decomposition bomb, discrete nebulisation (Ca, K, Mg and Pressure digestion of samples with HN03 zn) Sample introduced into ICP on a wire loop; sample dried near base of ICP for 30 s (Al, As, Be, Cd, Pb and Se) Applications of a slotted quartz tube to the analysis of biological materials; serum diluted 1 + 20, whole blood deproteinised with TCA, urine acidified with 0.1% HCI (Cd, Cu, Pb and Zn) SIC709 SIC719 9862 S/918 S1969 S1984 Sl993 SIC1136 s11203 86lC157 861193 faster.Again in the pursuit of a rapid, sensitive method for the determination of Cu and Zn in plasma fractions, Bahreyni- Toosi et al. (S/184) compared five separate laboratory-built ETA systems. A conventional Massman furnace was the most sensitive, but also the slowest. None of the systems were rapid enough to meet the stipulated requirement (cycle time < 30 s). Halls (S/993), using a commercial Perkin-Elmer furnace with uncoated tubes, was able to achieve cycle times of around 60 s for Cd and Pb in blood after deproteinisation, A1 in water and Cu in urine; the short programmes accentuated the slowness of the autosampler (operation time 29 s).Drying of samples could be achieved in under 10 s and in some analyses (Pb in blood and Cu in urine) the ashing stage could be omitted without affecting the performance of the method. Similar ideas were applied to the determination of A1 in dialysate fluids (86/65), which resulted in a two-step programme (combined dry - ash and atomise) of cycle time 56 s. Development of the stabilised temperature plalform furnace concept (STPF) continues. Voellkopf and Grobenski (9969) have applied it to the determination of As, Cd, Cr, Pb and Se in milk powder, human hair and urine using Zeeman-effect background correction. Different matrix modifiers were applied for each element.Stoeppler et al. (S/CSSl) have described their experience in the use of the STPF concept in the determination of Al, Cd and Pb in biological and environmental materials. They concluded that for measure- ment near detection limits, peak height should be used and direct comparison made with practically identical control materials or a matrix-matched standard curve should be used. This would seem to indicate that the STPF concept, designed to eliminate interferences and allow direct analysis, may not always be successful. Another way of ensuring that the temperature within the furnace is higher than the sample is to introduce the sample on a wire probe, as Edmonstone and Van Loon (S/28) have described for the determination of Pb in blood.Simultaneous multi-element atomic absorption spectrometry with continuum source (SIMAAC) and with electrothermal atomisation has been applied by Lewis et al. to the determina- tion of Cu, Fe and Zn in serum (S/987) and to Al, Co , Cr , Mn , Mo, Ni and V in serum (S/194). In the first paper, serum samples were diluted 1 + 20 with a diluent containing 0.1% WVTriton X-100,O.Ol M HN03 and 0.03 M NH4N03. In the second, a four-fold pre-concentration was achieved by dry ashing 2 ml of serum with Mg(NO& as an ashing aid - matrix modifier; the ash was re-dissolved in 0.5 ml of 5% V/V HN03. While normal levels of Al, Cr, Mn and Ni could be determined successfully, insufficient sensitivity was available for Co, Mo and V determination.Day-to-day precision was of the order of Clinical applications of electrothermal atomisation atomic emission spectrometry continue to show that for many analyses, it is at least as sensitive and accurate as ETA-AAS. Two recent applications are the determination of Mn in urine by Frech et al. (86/211) and A1 in blood, cortex and liver by Baxter et al. (86/197). There have been some interesting developments in solid sampling for ETA-AAS. Fietkau (S/C246) homogenised animal tissue in a commercial homogeniser to a median particle size of about 6 pm, diluted it to 1-10% m/m total solids and pipetted this directly into the graphite furnace. The problem of smoke causing a high non-specific absorbance was overcame by using air ashing as part of the furnace pro- gramme.Lead could be determined down to 60 pg kg-1 in liver tissue giving results in good agreement with those obtained after wet ashing. Rosopulo et al. (S/182) used the graphite-boat technique for sample introduction and a Zee- man system to overcome the effect of high background. 2O-3OYo RSD.36R JOURNAL OF ANALYTICAL ATOMIC SPECTROMETRY, APRIL 1986, VOL. 1 Results obtained on the analysis of SRMs (e.g., bovine liver) were within the certified range. The precision of solid sampling analysis was found to equal that of conventional ETA-AAS determination after wet digestion. The same approach was applied to the determination of Cd and Pb in liver tissue from animal carcasses for legal inspection purposes (S/C555). In situ ashing with 0 2 in the graphite tube has been used by Mohl et al.(S/C553) for the pre-treatment of difficult materials, such as cow and human milk, serum and animal muscle. Liquid samples were injected directly after addition of Triton X-100, whereas solid samples were inserted directly using the Perkin-Elmer solid sampling accessory. 1.3. Progress for Individual Elements 1.3.1. Aluminium It is a measure of the difficulty of an analysis when a large number of papers on that determination appear (Table 1) often with the adjective “improved” in the title. Amidst the claims and counter-claims of the accuracy of analysis based on simple aqueous standards, it is refreshing to find a paper that looks at some of the fundamental problems of serum A1 determination by ETA-AAS and helps to clarify the situation.Gardiner et al. (86/195) compared signals from A1 standards and A1 in serum in the presence and absence of various matrix modifiers for a standard uncoated tube, a pyrolytically-coated tube and a L‘vov platform in a pyrolytically-coated tube. The effect of different sheath gases was also examined. Whereas with uncoated tubes, serum could be analysed directly against aqueous standards using 0.1% Triton X-100 in 0.001 M HN03 as a diluent with or without the addition of Mg(N03)2 as a matrix modifier, the use of pyrolytically-coated tubes gave a matrix effect which was not eliminated by the use of Mg(N03)2. With the platform, Mg(N03)2 did eliminate the interference and allowed calibration against standards simil- arly diluted. Other factors that were considered important in the choice of tube were sensitivity (the platform is the most sensitive), sample load and number of firings possible with each tube.An interesting suggestion was the use of the alternative 396.2-nm line for analysis, which gave greater linearity with only about 20% loss in sensitivity. Brown et al. (S/985) found that the method of Leung and Henderson (Clin. Chem., 1982, 28, 2139) based on a L’vov platform and Mg(N03)2 as a matrix modifier did not allow the use of simple aqueous standards as had been claimed. They found mean slopes of 1.35, 3.44, 2.41 and 1.80 X 10-3 absorbance 1 pg-1 for aqueous standards, normal sera, uraemic sera (<300 yg 1-1) and uraemic sera (>300 pg l-l), respectively. By contrast, Bettinelli et al. (S/858), using a very similar procedure, but with an O2 ashing step, found good agreement between calibration slopes in aqueous solution and serum samples.Brown et al. (S/985) recommended the method of standard additions for calibration or alternatively the pre-treatment of the serum with HN03 to precipitate the proteins and then the analysis of the supernatant against simple aqueous standards. Good agreement was obtained between the two approaches. Whether a platform is necessary for A1 determination is debatable. D’Haese et al. (S/642) using an uncoated graphite tube, determined A1 in serum against simple aqueous stan- dards using H20 only as a diluent. Identical results were obtained with direct and standard-additions calibration and excellent results in the Trace Elements Quality Assurance Scheme run by the University of Surrey were obtained. The mean serum A1 concentration for ten healthy controls was 2.0 k 0.4 pg 1-1. Zirconium-coated tubes have been used by a number of workers (9479, S/642, S/967). Whole blood analysis by ETA-AAS using dilution of samples 1 + 3 with 2 g 1-1 Triton X-100 and direct analysis against standards also in Triton X-100 was examined by D’Haese et al.(S/642). For patients on haemodialysis, serum and whole blood concentrations were not significantly differ- ent, although whole blood concentrations for healthy subjects (12.1 k 1.5 pg 1-1) were higher than for serum (see above). Dialysate fluids cause a negative matrix interference in the determination of A1 in an uncoated graphite tube. Fagioli et al. (S/118) used a L’vov platform to overcome this and found good recoveries both with and without the addition of Mg(N03)2 as a matrix modifier.Total furnace programme time was 97 s. Halls and Fell (86/65) overcame the interference by acidifying to 2% V/V with HN03 and used a two-stage programme (combined dry - ash and atomise) to achieve a furnace programme time of only 30 s. Nitric acid was also used by Allain et al. (S/965) as a modifier for the analysis of haemodialysis concentrates by ETA-AAS . Even faster throughput of samples is undoubtedly possible with inductively coupled plasma optical emission spectrometry. Mauras and Allain (86/47) have described their automated method for A1 in plasma, dialysate fluid and water using Ga as an internal standard and Cs as a matrix modifier.Good correlation of results was obtained with ETA-AAS for plasma and water samples. As this direct approach is possible, the use of ICP-OES combined with electrothermal atomisation pro- posed by Matusiewicz and Barnes (S/974) for the determina- tion of A1 in these matrices seems unnecessarily complicated. An alternative approach to A1 determination is the use of electrothermal atomisation atomic emission spectrometry. Baxter et al. (861197) applied this technique with a constant- temperature furnace to the analysis of whole blood, cortex and liver samples after digestion with HN03. Good agreement was found with results obtained using a conventional ETA-AAS system and the L’vov platform. Other biological samples that have been examined are hair (S/967), soft tissues (S/642) and bone (S/480, S/642).Homogeneity of material can be a problem as found by Kratchovil et al. (86/C160) who examined a reference material, TORT-1 (marine biological material), and found that when sample aliquot size was decreased from 300 to 30 mg, the between-aliquot variation increased. This was blamed on inhomogeneity caused by microparticulates of Al. 1.3.2. Arsenic For the hydride generation AAS analysis of marine biological tissue, Welz et al. (S/590) applied a digestion with HN03, H2S04 and HC104 at a final temperature of 310 “C. Good agreement was found with results obtained after combustion of the sample in a stream of 02. Webb and Carter (S/976) preferred a digestion with HN03, H2S04 and K2Cr207 to give a quantitative breakdown of dimethylarsonic acid.They applied their method to the determination of As in water, urine, faeces and whole blood with recoveries in the range More important often than a knowledge of total As is to know the proportions of the various species. Stoeppler and Ape1 (S/166) differentiated between toxic inorganic As and its metabolites and the more stable and much less toxic arseno- betaines (which occur in fish). Wet digestion with HN03, HC104 and H2S04 was used followed by hydride generation AAS. 92-105%. 1.3.3. Cadmium In tissue analysis for Cd, possibilities of solid sampling have been explored. Nilsson and Berggren (S/975) freeze-dried pancreatic tissue and then analysed 1-13 yg amounts directly in a Varian CRA90 furnace. For Cd in liver tissue, Rosopulo et al.(S/C553) used direct solid analysis with Zeeman AAS and found good agreement with results after wet digestion. More conventional wet digestion - AAS procedures have been used for Cd in prostate glands (S/920), teeth (S/970) and liver and renal cortex (86/60). The last study, an estimate of the Cd body burden of an occupationally non-exposed population inJOURNAL OF ANALYTICAL ATOMIC SPECTROMETRY, APRIL 1986, VOL. 1 37R Southern Bavaria, again confirmed the great increase in body burden caused by smoking. Determination of Cd in urine by ETA-AAS continues to be a problem. Feitsma et al. (S/963) concluded that of the various matrix modifiers they examined , HN03 was preferable. Nitric acid was also used by Dungs and Neidhart (S/998) in the determination of Cd in urine on the L'vov platform by a standard-additions procedure.A graphics display, giving both corrected and uncorrected signals, was considered of great importance and was shown to be of value when examining a reference material (Lanonorm Control Urine). This sample gave problems with background correction, even with a Zeeman system, which were not seen with natural urines. Further modification of the method was needed for this sample. McAughey and Smith (S/966) tried a combination of the L'vov platform and (NH4)2HP04, but found that the modifier could only be used after time-consuming purification to reduce blank levels. Instead they developed a method based on selective volatilisation of Cd at 800 "C from a L'vov platform. Calibration was by standard additions and a detection limit of 0.06 pg 1-1 was achieved.A similar method based on standard additions but using an uncoated graphite tube was used by De Groot et al. (S/159). 1.3.4. Calcium and magnesium Although Ca and Mg determinations in serum, urine and tissues are well established, there is still some development in the measurement of Ca and Mg levels in blood cells, some at very low levels. Gerlach et al. (S/C531) determined low levels of Ca in erythrocytes (<1 pg g-1) by ETA-AAS. All steps in the analysis were verified by the use of a 47Ca tracer. Dust-free conditions had to be used to reduce contamination. Mono- nuclear cells were separated on a discontinuous Ficoll - Hypaque gradient by Elin and Hosseini (S/922). After lysing and dilution with La solution, the Mg content was determined by FAAS. Makino (S/1203) separated platelets by differential centrifugation and digested them in a PTFE bomb.Discrete nebulisation FAAS was used for the measurement of Ca and Mg* 1.3.5. Chromium To determine Cr in serum by ETA-AAS, Veillon et al. (S/616) used Mg(NO& as an ashing aid and matrix modifier. Samples with Mg(N03)2 added were lyophilised and dry ashed in silanised quartz tubes. Chromium was determined after the ash had been re-dissolved in 0.1 M HC1. A mean normal Cr concentration of 0.11 k 0.07 pg 1-1 was found for 15 adults. Very strict control of contamination was necessary. The method has been applied to studies of the effect of Cr supplementation and glucose loading (S/927) , from which it was concluded that serum Cr concentration does not appear to be a meaningful indicator of Cr status.The same dry-ashing procedure was used by Lewis et al. (S/194) to give a four-fold pre-concentration for determination by ETA-SIMAAC. Their mean value for 30 normal subjects was 0.24 1-18 1-1. Morris and Kemp (S/615) determined Cr in plasma and urine by ETA-AAS with a W lamp source for background correc- tion. Samples were diluted 1 + 1 with 0.3% V/VTriton X-100 and analysed by use of standard additions. Mean normal values of 0.82 k 0.52 yg 1-1 for plasma Cr and <1.0 pg 1-1 for urine were obtained. Since the value for plasma is higher than what is now regarded as a realistic value (<0.5 pg 1-I), it can be concluded that insufficient precautions were taken in sample collection (e.g., stainless-steel needles were used).1.3.6. Cobalt The problem of determining the low levels of Co in serum by ETA-AAS was approached by Anderson and Hoegetveit (S/982) through the use of solvent extraction of digested plasma with APDC into IBMK. Recovery was checked with a ~ C O tracer. A detection limit of 0.05 pg 1-1 and a mean normal concentration of 0.15 yg 1-1 was obtained. The application of the ETA-SIMAAC system to the determination of Co (S/194) gave a detection limit of 0.57 pg l-l, which was insufficient to give reliable values for normal levels. Similar problems of insufficient sensitivity were encoun- tered by Heinrich and Angerer (S/C525) in the development of a method for Co in whole blood. To remove matrix interferences in ETA-AAS , samples were diluted eight-fold and a multi-stage ashing procedure was used. The method was suitable for occupational levels, but not for natural levels. Urine was analysed after solvent extraction.1.3.7. Copper and zinc As with Ca and Mg, developments for Cu and Zn are mainly in studies of the cellular components of blood and the distribution in plasma. Milne et al. (S/617) separated platelets, mono- nucleated cells, polymorphonucleated cells and erythrocytes by use of a discontinuous Percoll gradient and measured the Zn content of the cells by HN03 digestion and FAAS using pulse nebulisation. Because of the high Zn concentration of platelets, the values for mononuncleated cells depended greatly on the extent of contamination from platelets. Makino (S/1203) measured Zn in platelets using separation by differential centrifugation, digestion in a PTFE bomb and analysis by FAAS using pulse nebulisation.Red cells were analysed by Chen and Zhao (S/106) for Zn by FAAS after dilution 1 + 19 with distilled water. Their conclusion that erythrocyte Zn was a better index than plasma or serum for nutritional and metabolic studies is one that few are likely to agree with. Using gel-filtration chromatography combined with affinity chromatography (to separate albumin) and ETA-AAS, Foote and Delves (S/103) found that most of the Zn in the globulin fraction of human serum was bound to a*-macroglobulin; no Zn was associated with transferrin. Although tissue analysis for Cu and Zn seems straight- forward, it is apparent from a collaborative study on the determination of Cu by Osheim and Ross (S/923) that agreement between laboratories could still be improved.In this study, 11 laboratories analysed four blind duplicate pairs of bovine liver samples by digestion with HN03 at 60 "C and determination by AAS. Within-laboratory variation ranged from 5.6 to 19% RSD and between-laboratory variation was 7.1-21% RSD. Most methods for tissue analysis use acid digestion followed by FAAS determination and similar methods have been applied to Cu and Zn in liver biopsies (S/785), Zn in prostate tissue (S/920) and Zn in rat intestine, kidney and liver (S/104). For determinations of Cu and Zn in human placenta, Imaeda et al. (S/195) used acid extraction with 0.5 M H2SO4 or 1 M HN03. Results compared well with those obtained after wet digestion in a PTFE bomb.1.3.8. Iron Kreeftenberg et al. (S/646) have compared the determination of Fe in liver biopsies by the three different methods: colorimetry using thiocyanate as a complexing agent, FAAS and ETA-AAS. A good correlation was found between all three methods. Of the three, the FAAS method was quickest, but the ETA-AAS method was more suited to small sample masses. Serum Fe has been determined after deproteinisation with TCA by Zhong et al. (S/105). Triton X-100 and NH4N03 were used as matrix modifiers in the final determination by ETA-AAS with Zeeman-effect background correction. Li et al. (S/971) used FAAS with tartaric acid as a matrix modifier to increase the sensitivity and reduce interferences. 1.3.9. Lead Direct determination of Pb in whole blood by ETA-AAS has been frequently described in previous years.Oxygen ashing38R JOURNAL OF ANALYTICAL ATOMIC SPECTROMETRY, APRIL 1986, VOL. 1 has been used by Shiowatana and Matousek (S/C691) to help resolve Pb peaks from the interfering background absorption. Direct calibration with aqueous standards was possible. Schmid and Krivan (S/191) investigated the behaviour of Pb in the graphite furnace during ashing with 203Pb as a radiotracer for a number of matrices including blood, serum and urine. Optimum elimination of matrix effects was achieved by simultaneous use of a L'vov platform and NH4H2P04 as a matrix modifier. Stoeppler et al. (S/158) have described a modified version of the original Stoeppler , Brandt and Rains method for Pb in blood (Analyst, 1978, 103, 714).Samples were diluted with 1 M HN03 containing 0.1% Triton X-100. After centrifuging off precipitated cells and proteins, the supernatant was analysed by ETA-AAS. Comparison of results was made with isotope dilution MS and differential pulse ASV. Blood samples for Pb determination were shown by Wang and Frank (S/1200) to be stable for at least ten weeks whether stored at 22, 4 or -20 "C using either EDTA or heparin as an anticoagulant. Omenetto et al. (S/995) showed that the determination of Pb in blood was possible by laser-excited atomic fluorescence spectrometry. Samples diluted ten-fold were nebulised into an air - C2H2 flame. Excitation at 283.3 nm by a frequency- doubled pulsed tunable dye laser stimulated fluorescence which was measured at 405.8 nm.A detection limit of 4 pg 1-1 was achieved and results were in good agreement with those obtained by ASV and Delves cup AAS. Extracted tetramethyllead from whole blood was deter- mined by combining high-resolution GC with ETA-AAS (S/978). A detection limit of 10 pg 1-1 as tetramethyllead was obtained. Diethyllead in urine can be measured by chelation with glyoxal-bis-(2-hydroxyanil) and extraction into IBMK, as Turlakiewicz et al. have shown (SBll). Determination of Pb in the extract was made by ETA-AAS. The method success- fully discriminated against both triethyllead and inorganic lead. 1.3.10. Manganese The technique of ETA-AES has been applied to the determi- nation of Mn in urine by Frech et al. (86/211). Urines diluted 1 + 5 with 0.01 M HN03 were analysed at the 403.1-nm line directly against standards in 0.01 M HN03.The results correlated well with those obtained using an ETA-AAS method. Pre-concentration of Mn in urine by co-precipitation with La(OH)3 formed the basis of an FAAS method elabor- ated by Holler et al. (S/977). Bayer (S/C528) proposed that, since over 90% of Mn in blood is present in red cells, whole blood Mn determination may be more appropriate than serum analysis. Blood (200 pl) was diluted with 500 pl of 0.5% V/VTriton X-100 and the Mn determined by standard additions on a L'vov platform in an ETA-AAS system. The ETA-SIMAAC system of O'Haver and co-workers (S/194) gave a mean normal Mn concentration of 0.48 pg 1-1 for serum; measurements were close to the detection limit (0.11 pg 1-1).As a result of hair Mn measurements on newborns and their mothers, Saner et al. (86/3) concluded that determination of Mn in prenatal maternal hair may prove to be a reliable indicator of the risk of intrauterine malformations. Hair samples were dry ashed, dissolved in HCl and analysed by ETA-AAS. 1.3.11. Mercury Since Hg is so volatile, losses can easily occur during the sample pre-treatment stage. May and Stoeppler (S/157) avoided losses by digestion with HN03 and HClO4 in partly closed or completely closed silica vessels heating cautiously from 20 to 200 "C over 1&15 min. Welz and Melcher (S/590) recommended pressure digestion with HN03 in a PTFE bomb for marine biological tissues. Cold-vapour AAS is still the most popular technique for measurement of Hg.Welz and Melcher (S/590) combined it with an amalgamation technique to achieve greater sensitivity. Einarsson et al. (S/2) automated the determination so that the digested samples (blood, urine or other biological material) were introduced by an automatic sampler into a computer- controlled cold-vapour AAS system. Comparison of cold- vapour AAS with DCP-OES for the determination of Hg in blood and fish samples has been made by Lajunen et al. (S/904). Mercury -containing drugs have been analysed by Holak (S/1088) using HPLC with cold-vapour AAS. 1.3.12. Nickel Drazniowsky et al. (S/810) determined Ni in serum by ETA-AAS after dilution 1 + 1 with 0.2% V/VTriton X-100 using matrix-matched standards. For 71 normal subjects a median normal value of 1.0 pg 1-1 was obtained; patients on haemodialysis had much higher levels (predialysis median value 7.4 pg 1-1).Wei and Qi (9196) in their ETA-AAS method used ammonium oxalate to eliminate interference from chlorides. Their mean normal value (21.6 pg 1-1) is very high and suggests that inadequate care was taken to minimise contamination. 1.3.13. Platinum Most determinations are concerned with the use of cis- diamminodichloroplatinum as an anti-tumour agent. In their study of rats, Wagner and Engelmann (S/C522) diluted serum samples 1 + 1 with 10-4 M HN03 for analysis by ETA-AAS using matrix-matched standards. Drummer et al. (S/986) used an HPLC assay for human serum ultrafiltrate and compared it with an ETA-AAS method. Nitric acid causes an interference in the determination of Pt in tissues by ETA-AAS.Matsumato et al. (S/979) removed it by adding NH3 to vaporise it in the ashing stage as NH4N03. Their method was applied to rat organs. 1.3.14. Rubidium Flame atomic emission spectrometry was used by Allain et al. (S/497) to determine Rb in plasma and whole blood after dilution 10- and 100-fold, respectively, with a 2.5 g 1-1 Cs solution as an ionisation suppressor. A standard AA spec- trometer was used in the emission mode with an air - C2H2 flame and a red-sensitive photomultiplier. Mean normal values for plasma of 2.29 k 0.29 pmol l-1 for 27 males and 1.96 k 0.46 pmol 1-1 for 17 females were obtained. Hallis et al. (S/857) used ETA-AAS for the determination of Rb in plasma and erythrocytes using a 20- and 50-fold dilution, respectively.Calibration was by standard additions. Their normal values were 3.1 pmol l-1 for plasma and 61 pmol l-1 for erythrocytes. 1.3.15. Selenium Successful Se determination in body fluids and tissues by hydride generation depends greatly on the prior digestion. Welz et al. (S/494, S/590) have shown that the use of HN03 alone gave low recoveries. They recommended a digestion with HN03, H2S04 and HC104 at a final temperature of 310 "C to give complete breakdown of organoselenium com- pounds. This has been applied to serum, whole blood and urine (S/494), and to marine biological tissue (S/590). In their method for Se in liver, Kompiang and Coates (S/C680) used an HN03 - HC104 digestion. This gave good results with NBS bovine liver. The now well-established ETA-AAS procedure for serum analysis using Ni as a matrix modifier has been applied to sheep serum by Prosbova et al.(S/789). This approach has been found to fail when tried with whole blood because of spectral interference from Fe. According to Droessler and Holcombe (S/C271), this interference is due to a molecularJOURNAL OF ANALYTICAL ATOMIC SPECTROMETRY, APRIL 1986, VOL. 1 39R iron oxide band formed by the combination of Fe with entrained air. 1.3.16. Thallium Chandler and Scott (S/643) determined T1 in urine by extraction with NaDDC at pH 7 into toluene. The toluene extracts were analysed by ETA-AAS for concentrations of 0-1.65 pg 1-1. The method is suitable for determining T1 levels in persons not normally exposed to T1. Direct determination by ETA-AAS for both blood and urine after dilution with 0.1 M HN03 was shown to be possible by Matsuno et al.(S/160). Good agreement was found with solvent extraction methods. Wakid and Cortas (S/992) have compared a spectrophoto- metric procedure and an FAAS method after solvent extrac- tion for the determination of T1 in urine. Hair analysis has been proposed as a suitable marker for following TI poisoning (S/C529). Matsuno et al. (S/160) analysed hair directly by insertion into a graphite cup of an ETA-AAS system. 1.3.1 7. Vanadium Mousty et al. (S/988) reported that up to 20% of V is retained by the graphite tube after atomisation in ETA-AAS. Despite this, the determination of V in ten urine samples showed good agreement with NAA results. Chen and Angerer (S/989) preferred a solvent extraction step with APDC into diiso- propyl ketone and xylol in their ETA-AAS method.1.4. Conclusions With time, some of the less realistic applications of ICP-OES will disappear and it will find its true place in the clinical laboratory. The technique can have an important role, as the pioneering work of Allain and Mauras (S/C417) has shown. For Al, although ETA-AAS methods continue to be developed that work in the authors' laboratories, the scene must seem very confusing to a newcomer. The work of Gardiner et al. (86/195) has helped to clarify the situation, but more needs to be done. All developments that produce easier and quicker methods are welcome and some were featured in section 1.2. The amazing accuracy produced by direct solid sampling ETA- AAS and the use of O2 ashing for in situ pre-treatment of samples in the graphite tube are of particular interest.For elements at low concentrations in clinical samples, it is apparent that the ideas expressed by Versieck and Cornelis (Anal. Chim. Acfa, 1980, 116, 217) in the comparison of normal levels in establishing accuracy are becoming recog- nised. Unfortunately, there are still papers appearing in which previous literature on sample contamination and handling is ignored and which result in normal values that are far too high. 2. ANALYSIS OF FOODS AND BEVERAGES The use of atomic spectrometry for trace element analysis of foods and beverages is now well established and few papers describing new or novel approaches have been published recently.Owing to this rather static situation more attention has been given to analytical accuracy, which is reflected in the increasing number of papers reporting results from inter- laboratory quality control programmes. An important trend, however, is the use of direct introduction of samples into flames, plasmas and graphite furnaces as slurries, suspensions or emulsions. As in the majority of application areas, speciation is another topic of interest. A major area of concern in the food industry is the possible contamination of foods from wrapping or container materials. 2.1. Sample Preparation In most reports, sample preparation was achieved by acid digestion but in a comparative study this technique was found to give inconsistent results for As, Hg and Se (S/590).Pressure decomposition in a PTFE bomb led to low results for As and Se (3590) although other workers obtained satisfactory results for similar samples with this procedure (S/936). The comparative study obtained best results following combustion in a stream of O2 (S/590). Extraction of trace elements from foods merely by leaching with concentrated acid was found to be as effective as acid digestion or dry ashing but was simpler and faster (YC362). With suitable nebulisers tolerant to high suspended or dissolved solids, slurries and homogenates have been assayed (S/C298, S/C321). Some interest in speciation was evident with the publication of recommendations from the Trace Element Speciation in Foodstuffs Sub-committee of the Analytical Methods Com- mittee of the Royal Society of Chemistry (S/926).This Sub-Committee proposed that enzyme digestion at body temperature should be used for fractionation studies. They considered that the work was of sufficient importance to suggest the establishment of a full-time team of scientists. Separation of HgC12, MeHgCl and EtHgCl by HPLC with on-line formation of Hg vapour and detection by ICP-OES was described. This method resulted in detection limits of 50-75 p.p.b. and was applied to the analysis of fish samples (S/C334). Extraction of inorganic As from acid solution allowed the selective determination of inorganic and organic As compounds; further solvent extraction separated the dimethylarsenic acid from phenylarsonic acid (S/180). Sample collection and storage was considered in a study of the determination of Pb in drinking water (S/832).Acidification was shown to be necessary but that this could be delayed for up to 14 d after collection without irreversible loss of Pb. 2.2. Analytical Techniques The most commonly employed analytical technique was AAS although a wide range of other procedures were also used. Inductively coupled plasma optical emission spectrometry was adopted for a number of studies (Table 2) including inter- laboratory comparisons (S/482, S/938) and the determination of trace elements in reference materials (S/lO, S/C366, S/482, 9897, S/937). Unusual or novel techniques included the complexometric determination of Ca in cheese (86/1) , dithi- zone colorimetry (S/9), the determination of Pb in beer using the slotted quartz tube and FAAS (S/893) and hydride generation AAS for the determination of As, Sb and Se in a single-cell protein.This last report referred to the use of a liquid N2 trap to collect and concentrate the hydrides and an air - H2 flame within the quartz atomiser tube to increase atomisation efficiency (86/C146). Detection limits of 0.1-0.5 ng were obtained with this technique. A procedure to measure the proportion of residual bone within beef products was described that involved the preparation of homogenised sample, to contain 5% solids, which was immediately aspir- ated into a slurry nebuliser for Ca emission analysis with an air - C2H2 flame (S/C298). Studies for the certification of reference materials involved many other techniques (Table 2).40R JOURNAL OF ANALYTICAL ATOMIC SPECTROMETRY, APRIL 1986, VOL.1 ~~ Table 2. SUMMARY OF ANALYSES OF FOODS AND BEVERAGES Technique ; atomisation; analyte form AA; Hy; L Matrix Concentration Sample treatment Reference Element Unm S/180 As Foods 5-10 g sample homogenised with 20 ml30% HC1 and 2 ml2O0/0 ascorbic acid, extracted into CH2C12 for As(II1) determination; organo-As compounds determined after decomposition of HCl phase with HN03 Comparison of 3 digestion procedures: pressure decomposition with HN03, decomposition with HN03 - H2S04 - HClO,, decomposition in O2 stream. Pressure decomposition with HN03 for Hg, followed by H2S04 - HClO, treatment for As and Se recommended Wet digestion of samples; hydride collected in liquid N2 trap, released by rapid heating; T-shaped quartz tube with internal air - H2 flame as atomiser Digest in HN03 - NaOAc As 193.7 Marine tissue - AA; Hy; L St590 As Single-cell protein 10-500 pg 1-1 AA; Hy; L 86lC146 AA; F, air - AA; F, air - ,C2H2; L C2H2; L Ca Ca Foods s/110 St242 - 422.7 622.5 Animal feedstuffs 1-5 g sample ashed at 550 "C, re-dissolved in concentrated HC1, evaporated to dryness, residue dissolved in 6 M HC1 and filtered homogenised, diluted with 2% HN03 to 5% solids range, shaken for 10 s, aspirated via slurry nebuliser into flame Sample treated for 2 min in Polytron, Ca Beef 50-200 mg 1-1 AE; F, air - C2H2; slurry SIC298 AA; F, air - AA; ETA; L AA; ETA; L C2H2; L Sl1211 Ca Fruit juice Cd Cd Baby foods Food packaging papers Breast milk IBMK solvent extraction procedure 100-200 mg sample digested with HN03 - Breast milk from 10 women analysed for Cd H2S04 at 140 "C for 30 min and Pb; samples examined at regular intervals over a 3-month period Wet digestion of sample, Fe removed by liquid - liquid extraction Samples digested with HN03 Samples wet ashed St863 Sl1207 AA; -; L 86/59 Cd Foods AA; F; L 86/52 c o AA; ETA; L AA;-; L Cr Cr Beer Enteral formulations Meats, frozen vejetables Milk 0.246 pg 1-1 - St243 St789 AA; -; L Nutritional survey of various foods 9111 c u c u AA; ETA; L Lipids and proteins removed by shaking with H2S04 - CCl,; no interferences found; residual C removed from furnace by O2 introduction during tube clean phase Accuracy comparison between dithizone method of Hg determination and cold- vapour AAS Speciation of HgC12, MeHgCl and EtHgCl by an HPLC - cold vapour-ICP technique; separation using C18 reversed-phase column with mobile phase containing 2-mercaptoethanol See As, ref. Sl590 Solid sample (300 mg) mixed with mixture of A1203 - MgO, heated in O2 stream; Hg collected in acidified permanganate - dichromate solution Comparison of two different cold-vapour AAS instruments; comparison of 8 different digestion methods Comparison of different digestion procedures using HN03 - H2S04 and H2SO4 - KMnO, mixtures; comparison of cold- vapour AAS and DCP-OES 5-50 g sample digested in HN03 - H2S04 Sl593 Foods AA; cold vap.; L s/9 AE; ICP, cold vap.; L Shark or swordfish 50-75 pg 1-1 SIC334 Marine tissue - Vegetation - AA; coldvap.; L -; cold vap .; gold film analyser St590 SIC692 StC693 Fish tissue 0.79-2.1Opgg-' AA;coldvap.; L Fish tissue 0-600 pg 1-1 AA; cold vap.; L or AE; d.c. arc; L St904 Foodstuffs Foods Wine - AA; F; Lor AE; F; L - AA; F, air - C2H2; L - AA;-;L 861201 s/110 St925 See Ca, ref. S/110 Studies on Ni content in wine pre- and post storage in stainless-steel tanks and glass bottlesJOURNAL OF ANALYTICAL ATOMIC SPECTROMETRY, APRIL 1986, VOL. 1 41R Table 2. SUMMARY OF ANALYSES OF FOODS AND BEVERAGES-continued Technique ; atomisation; analyte form AA; F, air - C2H2; L Matrix Foods Milk Baby food Beer Element P Pb Pb Pb Pb Pb Pb Sb Se Se Se Se Se Si Sn Zn Zn Concentration - Sample treatment Reference s/110 SIC94 S1863 S/893 S1940 s/1206 86/59 86lC146 S1243 S1590 S1786 9789 86lC146 9197 S1861 Sllll Sl648 s/10 S116 s/49 Sl50 S/109 s/112 SIC321 SIC362 SIC366 SIC367 S/482 SIC548 SIC549 SIC727 Phospho - vanado - molybdate amplification procedure for P determination (see Ca, ref.S/110) - See Cd, ref. S1863 Samples degassed by rapid shaking, acidified with 1% HC1, aspirated directly into a slotted quartz tube mounted above a conventional burner of Pb in 3 infant formulas HCl, HNO, added to dissolve precipitated Cu, solution analysed for Pb by FAAS Round-robin study for the determination Samples placed in solution of CuS04 - See Cd, ref. 86/59 See As, ref. 86lC146 Digest with HN03 - HC104 (see Cr, ref. See As, ref. Sl590 Digest samples with HNO, - HClO, - 9243) HC104; results compared with alternative procedures See Cr, ref. 9789 AA; ETA; L AA; ETA; L AA; F, air - C2H2; L Infant formulas AA; ETA; L Tin coatings AA; F; L Milk Single-cell protein Beer AA; -; L AA; Hy; L AA; ETA; L AA; Hy; L AA; Hy; L - 10-500 pg 1-1 0.2-15 pg 1-1 - - Marine tissues Foods Enteral formulations Single-cell protein Food AA;-;L 10-500 pg 1-1 - AA; Hy; L AA; -; L AA; F, N20 - C2H2; L AA; F, air - C2H2; L AA; -; L See As, ref.86lC146 Samples dry ashed, fused with Na2B407, re-dissolved in dilute HNO, Samples digested in HNO, - HC1 Canned foods 10-450 pg g-' Meat, frozen vegetables Fruit juice See Cu, ref. S/111 Samples digested in concentrated HNO,, evaporated to dryness, re-dissolved in 2 M HCl, concentrated on anion-exchange resin eluted with 0.01 M HCl, evaporated to dryness, re-dissolved in 1% HN03 Certification of a skim milk powder reference material (Ca, K, Mg, Na and P) Samples mineralised with HN03 - HC104 (Co, Cu, Fe, Mn, Ni, Pb and Zn) Acid - peroxide wet ashing of samples Study of the variation of element concentrations in the milk of lactating women (Ca, Cu, Fe, Mg and Zn) Concentration of trace-heavy elements in foods with chelating resins (Ca, Cd, Co, Cu, Fe, Mg, Mn, Ni, Pb and Zn) Wet digestion of samples with HN03 - HC104 - H2S04 or HN03 - HC104 (Ca, Cu, Fe, K, Mg , Mn, Na, P and Zn) Dilution of sample (Ca, Cu, Fe, K, Mg, Mn, Na, P and Zn) Acid leach of sample (2-10 g) with 20 ml of 1 + 9 concentrated HC1- HN03, heated at 82-93 "C for 30 min, filtered through Whatman No.541 paper, aspirated directly Certification of trace element content by a variety of analytical techniques Study involving the leaching of trace elements from various containers Certification of trace elements in 3 samples of skim milk powder by a variety of analytical techniques (Cd, Cu, Fe, Hg and Zeeman background correction (Cd, Hg, Pb Study of toxic and essential trace elements Pb) and Zn) pre- and post-canning (Cd, Cr, Cu, Ni, Pb and Zn) General review on the use of ZCP-OES to solve problems related to the food- processing industry Various - Various - Various - Various - ( 5 ) (7) ( 5 ) Various Skim milk powder Rhubarb AA; F, air - C2H2; L AA;-; L AA; F; L Canned food Breast milk Various - (10) Foods Trace metals AA;-;L Various - (9) Maternal diet solutions <25 pg I-' to >lo0 mg 1-1 AE; ICP; L Various - Various - (9) (11) Nutritional Foods formulations AE; d.c. arc; L AA; F; L Trace elements Various - Various - Various - (17) (22) Diet reference material Orange and grapefruit juice Skim milk powder Trace elements AA; For ETA; L AE; ICP; L Various Various - Various - (4) (6) Fish tissue Tuna fish Foods and beverages AA; ETA; L AA; For ETA; L Various - AE; ICP; L42R JOURNAL OF ANALYTICAL ATOMIC SPECTROMETRY, APRIL 1986, VOL.1 Table 2. SUMMARY OF ANALYSES OF FOODS AND BEVERAGES-continued Technique ; atomisation; Element Unm Matrix Concentration anal@ form Various - Milk and infant - AA; ETA; L (6) formulas Various - Peanuts and - AA;-;L (7) peanut oil Various - Skim milk powder - AE; ICP; L Various - Food - AA;-;- (4) Various - Various - Various - Various - Food Food Food Foods Various - Food and biological Various - Foods - materials Various Various .. --- 3 , Various AE; ICP; Lor AA; F; L . . --- 7 , Various - Spinach - AE; ICP; L Various - Foods - AE; ICP; L (14) Sample treatment Reference Study of the levels of trace elements in human milk, cow milk and infant formulas (Co, Cr, Fe, Mn, Mo and Ni) Study of the differences in the concentration of trace elements in two different varieties of peanut (Ca, Co, Cu, Fe, Mg, Pb and Zn) Pressure digestion with HN03 (Ca, Mg, Na and P) Report on the work and recommendations of the Trace Element Speciation in Foodstuffs Sub-committee of the Analytical Methods Committee of the RSC and trace elements in foods A review on the determination of mineral A review with 90 references A review with 29 references on sampling methods and data evaluation and handling in the food industry A review with 91 references on principles and instrumentation of atomic spectrometry with particular reference to the food industry described programme for the determination of 12 elements in raw foodstuffs and crops a graphite cup direct insertion technique recommended A new low-contamination digestion bomb is Development of a quality assurance Sample introduction into the ICP by Wet digestion with HN03 - HC104 Sf788 Sf814 Sf897 Sf926 St932 st933 Sf934 Sf935 Sf936 Sf937 Sf939 Sf1208 2.3.Reference Materials and Quality Control Programmes Preparation of materials for reference purposes were des- cribed for skim milk (S/lO, S/482), vegetables (S/937) and a freeze-dried “diet” (S/C366). Suggested concentrations were set out for Ca, C1, K, Mg, Na, Ni and P (S/lO) and Cd, Cu, Fe, Hg and Pb (S/482) in milk, for 12 elements in vegetables (S/937) and for 17 elements in the lyophilised diet (S/C366). Analytical quality assurance was comprehensively reviewed (S/934) and an inter-laboratory quality assessment programme for trace elements in infant formulas (S/938) was described.Between-laboratory coefficients of variation for the nine elements in the infant formula study were less than 9% with two-thirds of the samples. This programme had six partici- pants and all used the same analytical procedure (S/938). 2.4. Topical Applications Two areas of particular concern are evident from recent work: firstly, the effect of food packaging upon the trace element concentrations of their contents; and secondly, the assessment of likely intakes of trace elements by neonates from milk and infant food formulas.The concentrations of Cr, Cu, Ni and Pb in tuna fish, analysed before and one year after canning, showed no significant change, but the mean Cd and Zn concentrations were significantly increased from 0.05 and 16.49 to 0.07 and 18.45 p.p.m., respectively (S/C549). In a second study, many metals were measured in a wide range of canned foods, fruit juices and syrups and in the alloy of the container (S/49). The pH of the contents was an important factor in the dissolution of metal from the can and the authors made various recommendations concerning packaging and storage (S/49). Nikdel and Carter (S/C367) found that orange and grapefruit juices showed minimal changes in trace element concen- trations even after prolonged periods of storage in various containers.Defective packaging, however, resulted in major contamination from the container materials (S/C367). Glass was shown to adsorb Ni from wines (S/925). Papers used for wrapping foods, for bags and for kitchen towels contained Cd at concentrations too low to be of toxicological importance (S/1207). Improved methods to determine Pb in tin coatings (S/1206) and Sn in foodstuffs (S/861) were described while, as an example of the versatility of ICP-OES, the determination of trace elements in foods and containers following customer complaints was cited (SK727). The concentrations of Co, Cr, Fe, Mn, Mo and Ni were measured in human milk, cow milk and infant formulas. The levels in formulas were at least those found in breast milk with much greater concentrations of Fe and Mn noted (S/788).Samples of human milk and colostrum from Nigerian women were analysed for Ca, Cu, Fe, Mg and Zn to determine changes with lactation age. The concentrations of all of the elements studied, but especially Zn, steadily decreased during at least nine months of lactation and intakes were inadequate when compared with recommended dietary intakes (S/50). In other studies Cd (S/863, 86/59), Cu (S/593) and Pb (S/C94, S/863, 86/59) were also determined in human and cow milk and in infant foods. Intakes of Cd and Pb by breast-fed infants were higher in urban compared with rural areas (86/59). Four review articles with particular emphasis on the methods of analysis for the determination of inorganic elements in foods were published (S/932, S/933, S/934, S/935).JOURNAL OF ANALYTICAL ATOMIC SPECTROMETRY, APRIL 1986, VOL. 1 43R 2.5. Conclusions In addition to methods applied to a general range of food types or to reference materials (Table 2) and the measurements in milk and formulas, the more specific applications during the year have included analyses of enteral products, animal feeds (laboratory and farm), oils and fats, fruit juices, wine and beer and a novel single-cell protein source, pruteen. The determination of Hg in fish continued to receive much attention. The proportion of papers reporting multi-element analysis has increased, reflecting the growth of ICP-OES in this area. Results from inter-laboratory quality control programmes indicated that with a selected group of centres, acceptable analytical performance was possible, but this scheme was too limited for more useful conclusions to be reached.44R JOURNAL OF ANALYTICAL ATOMIC SPECTROMETRY, APRIL 1986, VOL. 1 Glossary of Abbreviations Whenever suitable, elements may be referred to by their chemical symbols and compounds by their formulae. The following abbreviations are used extensively in the Atomic Spectrometry Updates. a.c. AA AAS AE AES AF AFS APDC ASV CMP CRM cw d.c. DCP DMF DNA EDL EDTA ETA FAAS FAES FAFS FI GC GDL HCL h.f. HPLC IBMK alternating current atomic absorption atomic absorption spectrometry atomic emission atomic emission spectrometry atomic fluorescence atomic fluorescence spectrometry ammonium p yrrolidinedithiocarbamate (ammonium tetramethylenedithio- carbamate) anodic-stripping voltammetry capacitively coupled microwave plasma certified reference material continuous wave direct current d.c. plasma N, N-dimethylformamide deoxyribonucleic acid electrodeless discharge lamp e t h y lenediamine t et raace tic acid electrothermal atomisation flame AAS flame AES flame AFS flow injection gas chromatography glow discharge lamp hollow-cathode lamp high-frequency high-performance liquid chromatography is0 but yl methyl ketone (4-me th y lpen t an- 2-one) ICP IR LC LTE MECA MIP MS NAA NaDDC NTA OES PMT p.p.b. p.p.m. PTFE r.f. REE RM RSD SBR SEM SNR SSMS TCA TLC TOP0 u.h.f. uv VDU vuv XRF inductively coupled plasma infrared liquid chromatography local thermal equilibrium molecular emission cavity analysis microwave-induced plasma mass spectrometry neutron-activation analysis sodium diethyldithiocarbamate nitrilotriacetic acid optical emission spectrometry photomultiplier tube parts per billion parts per million polytetrafluoroethylene radiofrequency rare earth element reference material relative standard deviation signal to background ratio scanning electron microscopy signal to noise ratio spark-source mass spectrometry trichloroacetic acid thin-layer chromatography t rioct y lp hosp hine oxide ultra-high-frequency ultraviolet visual display unit vacuum ultraviolet X-ray fluorescence