Atomic spectrometry update. Clinical and biological materials, foods and beverages

Andrew Taylor *a, Simon Branch b, Martin P. Day c, Marina Patriarca d and Mark White e
aSupra-regional Assay Service, Trace Element Laboratory, Faculty of Health and Medical Sciences, University of Surrey, Guildford, Surrey, UK GU2 7XH. E-mail: A.Taylor@surrey.ac.uk
bMcCormick(UK) Ltd., Haddenham, HP17 8LB, UK
cThe Australian Wine Research Institute, PO Box 197, Glen Osmond, SA 5064, Australia
dIstituto Superiore di Sanità, Viale Regina Elena 299, 00161, Roma, Italy
eHealth and Safety Laboratory, Harpur Hill, Buxton, Derbyshire SK17 9JN, UK

Received 17th January 2011 , Accepted 17th January 2011

First published on 7th February 2011


Abstract

The period covered by this review has seen a number of interesting developments. In a novel technique to preconcentrate analytes in drinking water, metal complexes with 8-hydroxyquinoline were adsorbed onto miniature C18-bonded silica gel columns on centrifugal microfluidic discs. Centrifugal action forced the fluid through the column and measurements were made directly on the columns by LA-ICP-MS. In a variation on an old theme, carbon nanotubes were used for preconcentration by some workers. An elegant solution was devised to calibrate SEC columns, used to separate metalloproteins, through measuring S, found in virtually all proteins, at the same time as the metal of interest. Separate detection of protein by UV absorbance was thereby unnecessary. In this work interference from 16O16O+ on 32S was avoided by measuring SO+ at a m/z ratio of 48. A similar approach to removing polyatomic interferences was seen when methane was included in the collision cell. The 38Ar40Ar+ and 40Ar40Ar+ polyatomic interferences on 78Se+ and 80Se+ were avoided and As was measured as 75As12CH2+ at m/z 89. The flame furnace technique continues to stimulate innovation. In a system reminiscent of the Delves micro cup technique, pellets of lyophilised blood, mixed with high purity graphite powder, were placed on a quartz holder and introduced into a glass combustion chamber which connected to the flame furnace via a quartz or PTFE tube. Ammonium nitrate solution on paper was placed in the chamber, and with a flow of O2 and focussed microwave radiation, the sample ignited and the combustion products were led into the flame furnace for measurement of Cd. The commercial development of a high resolution-continuum source spectrometer is leading to some new applications for AAS. In one piece of work P was measured at both atomic and molecular lines within an electrothermal furnace. In another, Zn in a simple slurry of yoghurt was measured by FAAS. UV photo-oxidation for HG-AFS was examined and it was seen that adding organic acids to aqueous samples promotes vaporization. In other work TiO2 nanoparticles considerably enhanced the efficiency of HG from SeVI and TeVI. Much of the applications work relates to the structure and function of the measurand(s) i.e. metallomics. Overviews of analytical techniques and methods applied to metallomics were seen in several reviews. A sophisticated example of metallomic research is given by the work with ESI-MS, to identify how Cu progresses from one intra-cellular protein to another in a way that avoids the presence of toxic free Cu ions. Understanding of both As and Se metabolism has been further compounded by the identification of yet more intermediates although the question still remains as to whether these are artefacts produced during extraction and the analytical procedure. The UK Total Diet Study noted that exposure to metals has generally declined over 30 years. Of 24 elements in 20 food groups, with the exception of Al, Ba and Mn, all were found to be similar to, or at lower concentrations than when measured six years earlier. A comprehensive study of the packaging material for bottled water sold across the European Union was reported which showed that there were significant contributions of metals from plastics and glass and that the colour of the glass also influenced concentrations. Some particularly critical comment has been included in this review. A few papers have reported results that are clearly wrong. Many others include numerical data with an excessive use of significant figures. This material, appearing in the so-called scientific literature, highlights important failures by authors, referees and editors to ensure that data are quality controlled. Finally, we have to report that this is the last ASU to which Simon Branch will be contributing. Simon has been a mainstay of the writing team for many years but a change in career means that he can no longer be involved. His efficient preparation of the foods table will be much missed.


1 Reviews

This latest Update adds to that from last year1 and complements other reviews of instrumental techniques in the series of Atomic Spectrometry Updates from the last year2–6 and the special article that provides a perspective review of analytical developments and applications covering the 25 years of JAAS and ASU.7

As is evident from our recent Updates and from even a cursory glance at articles in primary journals, interest in metallomics in general and in speciation in particular, shows no sign of abating. This is reflected in the number of reviews addressing these topics. In a comprehensive overarching review, Ge and Sun8 succinctly defined metallomics and metallomes and describe the range of analytical techniques applied to the identification of bio-metallic species and the relationships between their structure and biological functions. In a more focussed review, Lobinski and his colleagues addressed similar territory but restricted their discussion to how ICP-MS may be used not only to detect proteins and peptides following HPLC or gel electrophoresis but also to quantify post-translational modifications to molecular structure.9 Further applications, such as quantification of metal-labelled proteins and antibodies were also discussed.

Several useful articles concentrated specifically on speciation. The advantages of low LODs and wide linear calibration range, afforded by AFS coupled to vapour generation methodologies, were reviewed by Sanchez-Rodes et al.10 Preparation of a range of sample types, along with the measurement of As, Hg, Sb and Se by AFS were discussed. Ogra and Anan11 refer to the ambitious objective of identifying every seleno-metabolite using elemental and molecular MS techniques, the ultimate intention being to prepare a chart showing the pathways of Se metabolism in animals and plants. They refer to previously published attempts and point out that not all of the key intermediates, especially selenophosphorus species, have actually been identified by MS and recent advances have added to the number of known species. For some, e.g., selenosugars, the route of synthesis is unclear. In a separate review, Ogra also considered speciation of Sb and Te from the viewpoint of understanding the metabolism and toxicity of these elements.12 The relevance of speciation in the context of understanding how Pt-containing drugs are effective in the treatment of cancer, was reviewed by Michalke.13 Methodologies to determine interactions with S-containing ligands, reaction kinetics with proteins, and forms entering the cell were discussed.

A number of other topics were reviewed during this period. The medicinal use of As compounds was discussed by Nicolis et al.14 including metabolism and pharmacokinetics. Application of XRF and micro-XANES to determine the location and form of As within hair was highlighted. Fairweather-Tait et al.15 reflected on what is understood of the bioavailability of Se compounds and of other features which are relevant to the derivation of dietary recommendations. Electrothermal AAS to determine As, Cd, Cu, Hg and Pb in biological samples and studies using XRF for in vivo measurement of Pb in bone, were reviewed by Sardans et al.16 and Rosin,17 respectively. Gonzalvez et al.18 reviewed how trace element content and stable isotope ratios are used to authenticate the origin of foods. Consideration was given to analytical techniques and the chemometric analysis of results. The prolific work of Becker, using LA-ICP-MS to image the distribution of metals over the surface of brain and other tissues has been a feature of these Updates in the past few years. This experience has been captured in a series of reviews.19–21 The all important aspect of contamination control during sample collection and analysis was thoroughly considered by Rodushkin et al.22 who provided a valuable list of key points for consideration by the analytical laboratory.

2 Metrology, interlaboratory studies, reference materials and reference ranges

The uncertainty associated with the measurement of 12 elements (Be, Cd, Hg, Ir, Pb, Pd, Pt, Rh, Sb, U, Tl and W) in serum from a simple dilution method by SF-ICP-MS was determined23 according to the Eurachem/Citac Guide, using method validation data. The relative expanded uncertainties ranged from 11.6% (Hg) to 27.6% (Pt). The greatest contribution came from the intermediate precision.

Davis et al.24 at NIST quantified seven As species in a frozen human urine certified reference material (CRM) (SRM 2669, levels 1 and 2) while Kruger at al.25 developed four caprine liver RMs with demonstrable homogeneity, stability and quoted concentrations that were traceable to CRMs. These secondary RMs may be used for quality control, method validation etc. A serum CRM is available from BCR with a certified value for total Se. Jitaru et al. determined indicative concentrations for Se in glutathione peroxidase, selenoprotein P and selenoalbumin from published data and from their own work.26

Using a simple ICP-MS method, with dilution of urine in 0.5% HNO3 and 0.005% Triton X-100, Batista et al. measured 16 elements in 412 samples from healthy Brazilian adults to establish reference ranges.27 Kim et al.28 determined the serum Se concentrations of 50 male and 50 female Korean adults to be 103.29 ± 31.05 and 120.81 ± 27.37 μg L−1, respectively. Selenium was also determined in 35 paired serum and CSF samples, by Michalke and colleagues.29 The concentration ranges were 42–130 and 1.63–6.66 μg L−1, respectively and there was no correlation between the two sets of results. In a further report,30 ultrafiltration was used to separate low Mr species of Ca, Cu, Fe, Mg, Mn and Zn in serum and CSF. The proportion of low Mr species within CSF tended to be much higher than in serum. Michalke was also involved in a joint Germany-Spain project in which Ce was measured in human milk and blood at 4 weeks post partum. Human milk samples from both countries had concentrations of 5 to 65 ng L−1 but the Spanish blood samples at 21.6–70.3 ng L−1 were much higher than those from Germany which were around the LOD of 10 ng L−1. It was suggested that the difference may be attributed to greater exposure to airborne particulates in Madrid compared to Munich. Reference ranges for Cu, Se and Zn in serum of non-pregnant women and during weeks 10–14 and 16–20 of pregnancy were determined by Kilinc et al.31 The reported concentrations, for these three groups, respectively were: Cu 104.75 ± 39.14, 132 ± 38.24 and 164.86 ± 39.69; Se; 55.38 ± 8.81, 44.85 ± 9.23 and 47.18 ± 10.92; Zn: 121.41 ± 29.22, 81.3 ± 31.94 and 74.25 ± 22.47. Results for male subjects were 78.29 ± 20.9, 72.24 ± 9.28 and 134.85 ± 15.95 for Cu, Se and Zn. Concentrations were stated as μg dL−1 which are obviously incorrect for the Se data and forces the reviewer to question either the work reported or the refereeing process of the journal.

3 Sample collection and preparation

3.1 Collection and storage

Two reports were concerned with metals in liver tissue. Miranda et al.32 obtained liver biopsies from calves fed a high-copper diet. Samples were taken from six areas of the liver, digested in acid and the copper quantified using ICP-MS. Highest concentrations were found in the left lobe with lowest values obtained from caudate and quadrate lobes. Concentrations differed among different breeds but the distribution pattern remained unaltered. Johnston et al.33 used dog liver to determine whether measured concentrations were influenced by the size of the specimen, formalin fixing, storage in wax blocks and deparaffinization. A number of important observations were made: metal distribution (Cu, Fe and Zn) was heterogeneously distributed and single wedge biopsies also demonstrated this variation, deparaffinized samples had high Zn concentrations compared with fresh tissue but this difference was not seen for Cu or Fe. The authors question the value of needle biopsy specimens. The discussion on sources of contamination and strategies to reduce effects on results22 was mentioned in a previous section.

3.2 Extraction and digestion

Few publications now feature developments with digestion. Ashoka et al.34 compared six methods, involving HNO3 and other reagents, to prepare fish CRMs for the determination of 40 elements by ICP-MS. They recommended using HNO3 and H2O2. As an alternative to complete digestion, extraction using acids provides for simple and rapid methods and examples of various sample types and applications were reported.35–37 Work involving cloud point extraction38,39 and co-precipitation techniques40,41 related to new application areas rather than being particularly innovative. The interesting work in recent years has been led by the needs of metallomic research for methods that preserve molecular structure. Thus, gentle extraction into water,42 buffered saline,43 and the use of enzymes,44 with or without a measure of microwave or ultrasonic energy to promote extraction are regularly used. Fang et al.45 compared extraction with water, 0.1 M HCl and enzymatic hydrolysis to extract seleno-species from rice. Using protease XIV and amylase in an ultrasonic bath was the most effective procedure. It is noted that the popularity of alkaline extraction using TMAH has increased and was used for a wide range of sample types.46–50

3.3 Preconcentration

Lafleur and Salin51 developed a novel SPE approach to preconcentrate analytes in drinking water. Metal complexes with 8-hydroxyquinoline were adsorbed, from 1–600 μL sample, onto miniature C18-bonded silica gel columns on centrifugal microfluidic discs. Centrifugal action drove fluid flow through the column. Measurement was made directly on the columns by LA-ICP-MS. Absolute LODs for Co, Cu, Ni, Pb and V were from 0.1 to 12 ng. Carbon, in various forms, has long been used for preconcentration. Ozcan et al.52 used columns of carbon nanotubes for this purpose. Samples prepared from foods and water, at pH 9, were added to a column and ions, of Cu, Fe, Mn, Pb, were eluted using 1 M HNO3 in acetone to achieve an enrichment factor of 20. Gil et al.53 also investigated the potential for using carbon nanotubes. They measured Tl in drinking water by ETAAS with an on-line FI column for preconcentration. An LOD of 0.009 μg L−1 was reported. Analyte enrichment associated with cloud point extraction and co-precipitation was mentioned in the previous section.

4 Progress with analytical techniques

4.1 Mass spectrometry

4.1.1 ICP-MS. The success of ICP-MS in recent years has led to problems associated with increasing workloads. Tu et al.54 cleverly used an HPLC-ICP-MS sample introduction system in FI mode to increase the analysis rate. With the well-plate option from the system, the injection cycle time was reduced considerably. To increase the sample throughput for HPLC-ICP-MS when measuring oxaliplatin in urine, Koellensperger and Hann55 used a stationary phase with a particle size of 1.8 μm.

Methods to improve sensitivity were described for applications where the analytes were present at particularly low concentrations. Lanthanides were determined in human placental tissues by Kruger et al.56 using a membrane desolvation system for sample introduction which provided for lower LODs and also reduced oxide formation. Because of the very low concentrations of As, Pb and Se in wines, Grindlay et al.57 investigated ETV for introducing samples into the ICP. It was found that matrix effects required calibration by standard additions. Results for Pb and Se compared well with those obtained when samples were introduced by nebulization but As values were much lower. This difference was attributed by the authors to the absence of ArCl+ interference at a m/z ratio of 75. It would be expected that the comparison would have involved use of a collision cell and that the accuracy would be checked by analysis of suitable CRMs. Despite the use of the ETV the LODs were not low enough to determine As and Se in the wines tested.

An elegant solution to the necessity of calibrating SEC columns when used to separate metalloproteins was devised by El Balkhi et al.58 Instead of separately identifying protein peaks by UV detection they determined S, which is found in virtually all proteins, at the same time as measuring the metal of interest. To avoid the interference from 16O16O+ on 32S, they measured SO at an m/z ratio of 48 and used O2 to promote the formation of the 32S16O species in the reaction cell. The technique was used to identify Cu-containing proteins in human plasma.

After separating As and Se species in fish liver and dietary supplements by CE, Hsieh et al. used ICP-MS for quantification.59 With methane in the collision cell, the 38Ar40Ar+ and 40Ar40Ar+ polyatomic interferences on 78Se+ and 80Se+ were avoided and As was measured as 75As12CH2+ at m/z 89.

The interesting application using LA-ICP-MS to measure metals on centrifugal microfluidic discs was described in section 3.3 above. Copeland et al.60 measured 87Sr:86Sr isotope ratios in fossil teeth with a multi-collector LA-ICP-MS system. The precision obtained was sufficient to be able to assess the geographical origins of the samples while analyses were less expensive than with solution MC-ICP-MS. Imaging of tissues by LA-ICP-MS is discussed in section 5.2, below.

4.1.2 Other MS techniques: TOF-MS, AMS, SIMS. An approach to standard additions calibration for LA-ICP-MS was explored by Narukawa and Willie.35 Samples were solubilised in formic acid and then internal standard and calibration standards added. Portions of the solutions were dried under a heat lamp and the dried spots volatilised by LA for ICP-TOF-MS. Borges-Alvarez et al.61 investigated the complex structure of the enzyme, Cu-Zn- superoxide dismutase. Mutant isoforms of this enzyme have been associated with a condition of neurological degeneration amyotrophic lateral sclerosis. Using two different techniques, i.e., MALDI-TOF-MS or CE with either UV spectrometry or ESI-MS detection, it was seen that metal ions were lost from the apo-protein during preparation for CE. The results indicated the importance of selecting appropriate non-denaturing separation and detection conditions to obtain reliable structural information about non-covalent protein complexes.

Hernandez-Mendoza et al.62 developed a method for monitoring Pu in urine, and other samples, from workers in the nuclear industry. Acidified urine was evaporated and subjected to microwave heating followed by anion exchange chromatography. The concentrated solution was taken for measurement of 239Pu by AMS. The entire procedure required 10 h compared with 1 week necessary for alpha spectrometry and had an LOD of 13 fg. Accelerator MS was also employed to investigate placental transport of 26Al in rats.63

Subcellular measurements of As and Se in rice and wheat samples using nanoSIMS in combination with localization by SRXRF was accomplished by Moore et al.64 in a project to identify how processes such as milling might influence the dietary intake of these elements. Complex distribution patterns were identified.

4.2 Atomic absorption spectrometry

4.2.1 Flame AAS. In keeping with the observations of recent Updates, there are few interesting publications involving FAAS, either in technology or applications. An exception was the work of Flores and colleagues,65 who combined the flame furnace technique to a rather complicated arrangement for introduction of Cd atoms from a solid sample. With a system reminiscent of the Delves micro cup technique, pellets of lyophilised blood, mixed with high purity graphite powder, were placed on a quartz holder and introduced into a glass combustion chamber which connected to the flame furnace via a quartz or PTFE tube. Ammonium nitrate solution on paper was placed in the chamber, and with a flow of O2 and focussed microwave radiation, the sample ignited and the combustion products were led into the flame furnace. With sample masses up to 56 mg the LOD was 0.018 μg g−1. By contrast, Brandao et al.66 reported a very simple method to determine Zn in yoghurt by forming a slurry by sonicating 1 g in 25 mL 2 mol L−1 HCl for 20 min. Measurement of seven samples by FAAS gave results of 2.19 to 4.85 μg g−1 with an LOQ of 0.32 μg g−1.

Citak and Tuzen39 produced yet another cloud point extraction method to pre-concentrate metals to permit measurement in water by FAAS. These workers found that, under optimized conditions, complexes with 1-phenylthiosemicarbazide would extract into surfactant-rich phase of Triton X-114 with an enrichment of 25-fold. The LODs for Co, Cu and Pb were 1.00, 0.67 and 3.42 μg L−1, respectively. A pre-concentration column in an FI system was used to measure Pb at ng mL−1 concentrations in digests of medicinal plants.67

4.2.2 Electrothermal AAS. The review of the current state of the art with respect to ETAAS for the analysis of biological samples was mentioned above.16 Features such as atomizer design, correction for non-atomic absorption, chemical modifiers, and the use of high resolution continuum source technology were discussed in detail.

Measurement of P by ETAAS is complicated by practical difficulties associated with the P resonance lines: complex chemical behaviour (e.g., formation of volatile compounds and a tendency to interact with graphite) and overlap between molecular and atomic lines. Resano et al.68 approached this problem by using high resolution-continuum source ETAAS. Methods, based on using either atomic or molecular lines, were compared. Atomization at 2650 °C and a chemical modifier consisting of Pd, ascorbic acid and tungsten (as permanent modifier) were used. For the PO lines, tungsten coating and vaporization at 1900 °C were appropriate. The LODs were 5 and 20 μg g−1 level for atomic or molecular measurement, respectively.

The P work involved solid sampling as did two further papers, both of which involved analyses of 0.5 cm segments of hair samples. Washed and dried samples were assayed without further treatment. The LODs were reported to be 2.8 ng g−1 for Cd and 40 ng g−1 for Pb in one report69 and 0.3 ng g−1 for Pb, in the second.70

Most of the recent studies for Se speciation have exploited ICP-MS as the detector following some form of separation such as HPLC. Vale et al.,44 however, demonstrated that ETAAS could be equally as effective when analysing Se-supplemented foods. Samples of around 10 mg were sonicated together with enzymes in 1 mL solution and then added to an HPLC column. It was demonstrated that ultrasonication did not alter the following Se species: Se(IV), Se(VI), SeMet, SeMeSeCys, and SeCys2.

Further work investigating Ru modifiers for simultaneous ETAAS was undertaken by Freschi et al.71 who were analysing samples of milk that had been diluted 1 + 9 with 1% HNO3. Slightly different modifier mixtures were used for volatile, (As, Bi, Pb, Sb, Se), or for less volatile (Co, Cr, Cu, Fe, Mn).

4.3 Atomic emission spectrometry

The high excitation energy of Cl, with only around 5% of sample contributing to emission in an ICP, limits the sensitivity. Hashimoto et al.72 used an oxidation vaporization reaction to deliver gaseous Cl from liquid samples which was then introduced directly into the plasma. This provided for a 20-fold improvement in sensitivity and the procedure was applied to the analysis of mineral water. Results were consistent with those obtained by ion chromatography while the measurement was complete within 1 min.

Following a simple and rapid procedure to prepare blood and tissue samples by homogenizing in phosphate buffered saline, Kollander et al.43 measured the concentrations of 14 elements by ICP-AES. Despite the nature of the slurry formed in this way, there were no issues associated with clogging or extinguishing the plasma. Sensitivity with a glass expansion concentric Meinhard type nebuliser afforded the best sensitivity of the three types tested.

A method in which 20–50 μL of blood could be used to measure the concentrations of 6 elements was presented by Zaksa et al.73 This employed what the authors described as, two-jet plasma emission spectrometry. While accurate results were obtained for a CRM, the method required the blood to be ‘carbonized’ by evaporation on graphite powder which is impractical for more than a few samples.

Campillo et al.74 developed a procedure involving SPME and GC-MIP-AES to determine dimethylselenide and dimethyldiselenide in milk and milk products. When applied to 23 samples, none contained these compounds above the LODs of around 110 pg mL−1 and 400 pg mL−1, respectively.

The feasibility of using laser induced breakdown spectrometry (LIBS) to determine a range of elements in pharmaceutical tablets, was investigated by de Carvalho et al.75 Emission signals generated by a Nd:YAG laser were focused to an echelle spectrometer with a high resolution charge coupled device for measurement. The method was demonstrated to be satisfactory and gave results in agreement with those by ICP-AES.

Relatively new or uncommon topics to which ICP-AES was successfully applied included detection of Gd species in serum following separation by HPLC,76 measurement of Li concentrations in gallstones,77 and Hg in honey following CV generation of the atoms.38

Another example of how improbable data are published despite the refereeing and editing systems is presented in a report of the concentrations of 5 elements in blood samples of smokers, as measured by ICP-AES.78 For example, the mean concentration of Se in blood was given as 332 μg L−1 (0.332 mg L−1 in the original text) which is at least twice the probable true value. Values given for Cd, Cu, Pb and Zn were equally unlikely.

4.4 Atomic fluorescence spectrometry and vapour generation procedures

While almost all work relating to AFS is concerned with vapour generation techniques, Frentiu et al.79 have a system designed around a single ring electrode rf capacitively coupled plasma and an EDL light source. This was used to measure concentrations of Zn in biological and environmental samples, as both liquids and solids. The LODs were 8.2 μg L−1 and 0.8–2.8 mg kg−1 for liquids and solid samples, respectively. They reported that there were virtually no non-spectral interferences and that operating costs are low compared to ICP-MS.

A number of interesting publications featuring vapour generation procedures were seen in the last year, some of which were applied to speciation of Hg. Instead of the more conventional UV or microwave digestion techniques, Yin et al.80 used a flame atomizer, in conjunction with chemical CV generation, to prepare eluted species for detection by AFS. The performance was similar to that seen with other systems and the instrumentation was much simpler. Chen et al.81 simply used a short guard column to separate Hg+ and MeHg, with photocatalyst assisted vapour generation and ICP-MS. Detection limits were improved considerably compared with sample introduction by nebulization.

A clever experimental system was devised by Cizdziel et al.82to investigate sample combustion as a preparative technique and also to compare AAS with AFS. Mercury vapour produced by combustion was pre-concentrated on a gold-trap, then desorbed and led first to AA and then AF spectrometers. Absolute LODs were 0.016 ng and 0.002 ng, respectively. Careful timing control over the amalgamation was necessary to prevent unwanted combustion products reaching the gold-trap.

Investigations of the Hg vaporization included UV photo-oxidation and electrolytic methods. Liu83 determined Hg in vinegar having shown that, under UV irradiation, the acetic acid in the sample reduced Hg2+ to the elemental form. Analytical conditions were optimized (3% acid, 30 s irradiation an 20 W Hg lamp) giving and LOD of 0.08 ng mL−1 with detection by AFS.

Two groups in China reported new electrolytic systems for CV-AFS. Zhang et al.84 used a Pt/Ti cathode with formic acid catholyte instead of a typical inorganic acid. This had the effect of increasing the Hg signal, reducing erosion of the cathode and had better tolerance to interferents. The LOD was 1.4 ng L−1 with an RSD of 2.3% at 1 μg L−1. Traditional Chinese medicines and CRMs were analysed. In a method that was applied to measurement of Hg in hair, Jiang et al.85 used a polyaniline-graphite cathode which performed with improved sensitivity, excellent stability and lower memory effects than a usual graphite electrode. Performance was similar to that reported in the previous paper with an LOD of 1.3 pg mL−1 and RSD of 2.7% at 2 ng mL−1.

Fewer reports were seen that were concerned with hydride generation. In addition to their work with Hg,84 Zhang et al.86 reported on a systematic study of electrolytic HG to determine As, Bi, Ge, Sb and Se by HG-AFS. Cathode materials, shape and dimensions, current and catholyte solution were all studied and careful optimization proved to be essential. Analysis of CRMs showed good agreement with certified values and the method was used to examine traditional Chinese medicines.

The UV photo-oxidation approach to HG-AFS was examined by two groups. Zheng et al.87 studied the impact of adding organic acids to aqueous samples to promote vaporization and also reported that TiO2 nanoparticles considerably enhanced the efficiency of HG from SeVI and TeVI. The LODs for As, Bi, Sb and Te were 0.5, 0.1, 0.2 and 0.08 ng mL−1, respectively. A comparison of UV photo-oxidation, with or without post-reaction heating, against microwave heating, for determination of As in urine by HG-AFS, was reported by Ito et al.88 Efficiency of HG was similar with all three arrangements but it was necessary to use matrix matched calibration to achieve accurate results with the microwave heating procedure. A lower LOD (2.7 μg L−1) and a shorter analysis time were obtained using photo-oxidation with post-reaction heating.

4.5 XRF

The ASU that reviews advances in XRF, and is published in JAAS each October, discusses technical developments and applications of interest.5
4.5.1 In vivo XRF. There is very little to review compared with previous years. Rosin summarised how K-XRF has been used to measure Pb stored in bone with reference to studies in children and adults, following occupational and environmental exposures, and in relation to neurological function. Changes following the introduction of unleaded petrol were mentioned and recommendations to reduce the concentrations of Pb in blood of children were discussed.

Gherase and colleagues described their work to validate measurements in skin. A polyester resin skin phantom was used to demonstrate that As and Se could be measured together using an X-ray tube and silicon PiN detector. There was no overlap between the peaks of the two elements.89 In another paper, descriptions of the detection geometry of the primary fluorescence and calculations applied to the signals were given. This information was used to establish the optimal positions and orientations of the detector and collimator so as to maximize the XRF signal.90

4.5.2 Quantitative analysis. Two papers by Anderson91,92 described the evaluation of a handheld, portable X-ray tube-based spectrometer for screening foods and beverages and the glazes on ceramic containers, where toxic metal contamination is suspected. With careful use and calibration, quantitative results could be obtained within a minute, without any sample preparation. The tube analyser afforded superior sensitivity and lower LODs compared to radioisotopic XRF.

Most published articles reported results from projects where XRF was used simply as the measurement technique or where some attention had been given to sample preparation. The concentrations of Cu, Fe and Zn in placental tissue were determined by WDXRF and the results assessed in relation to mothers age and neonatal birth weight.93 The same technique was used to measure the concentrations of 20 elements in blood serum from Type II diabetics and healthy controls.94 Cancerous and healthy samples of colon, breast and stomach tissue were examined using TXRF.95 Concentrations of As, determined by SRXRF, in toenail clippings from children were 0.15–2.1 μg g−1 while soil As levels were 3.3–130 μg g−1. Associated XANES spectra provided some insight as to the possible As speciation within the nails.96 To investigate a case of suspected metal poisoning, Borgese et al.97 placed a 1 cm segment of hair onto a quartz carrier, added 10 μL HNO3 and allowed this to dry before analysing it by TXRF. They suggested the technique is sufficiently fast that it could be used for emergency medical situations. Tateishi et al.42 also developed a simple procedure in order to determine Br in fresh and dried fruits. Samples were extracted with H2O and 0.5 mL added to filter paper, dried and taken for analysis by WDXRF.

5 Applications: Clinical and biological materials

5.1 Metallomics

A comprehensive discussion of elemental speciation is provided in the review by Harrington et al.2 Some of the more interesting work, that is relevant to the topic of this ASU, is presented in this section. Much of this work relates to methodologies to separate and identify or quantify inorganic or molecular species but an increasing number relate also to the structure and function of the measurand(s) i.e metallomics. Overviews of analytical techniques and methods applied to metallomics are given in reviews by Mounicou et al.,9 Ge and Sun,8 Jiminez et al.,98 and Hennion et al.99 The review by Michalke13 concentrates on studies that investigate how Pt-containing drugs act to destroy cancer cells. Two elements that have received less attention are Sb and Te, and Ogra reviewed what is understood of their metabolic toxicity.12

An example of metallomic research is given by the work of Banci et al.100 Since Cu is highly toxic, there are various binding proteins within biological systems that direct the metal to sites where it is required, or for storage. To identify likely routes of Cu trafficking, these workers used ESI-MS to determine the Cu-binding affinities of a number of intra-cellular proteins. The notion that Cu may be involved in vascular disease has been discussed for a number of years. Easter et al.101 summarized research findings on this particular topic highlighting the use of analytical techniques to understand the biological role of Cu. They suggest that Cu, acting as a dynamic signalling molecule, represents a new area of possible interest. The metallo-enzyme Cu,Zn-superoxide dismutase has a crucial role in protection of cells against reactive oxidant species. A number of mutant isoforms have been found which are associated with a neurodegenerative disorder. Understanding their structure is relevant to appreciating the biological implications of these variants. Borges-Alvarez et al.61 used a range of techniques, CE, MALDI-TOF-MS, ESI-MS to study the structure of this enzyme. Their initial results indicated that the metals were released from the enzyme and that its dimeric structure was lost, during the preparation and analysis stages. With careful control of pH and background electrolyte the integrity of the enzyme was maintained allowing further characterization of the mutant isoforms.

Chung et al.102 analysed 208 urine samples from subjects living in a region with high natural exposure to As, to determine the proportions of methylated species. The measurements were repeated 15 years later on samples from the same individuals. It was shown that subjects with high percentage MMA within the total As concentration had the earliest incidence of cancer and that this could be related to a gene polymorphism affecting the arsenic-methyltransferase enzyme activity.

Ogra and Anan11 described how analytical techniques have been employed to identify all Se metabolites (the selenometabolome) and how these inter-relate to metabolic pathways in animals and plants. In a more focussed piece of work, Ohta et al.103 used HPLC-ICP-MS and GC-ICP-MS to analyse urine and exhaled breath from rats administered 76SeMet, 77MeSeCys and 78methyl seleninic acid. From the metabolites separated and detected they concluded that MeSeCys is converted to methylselenol (thought to be protective against cancer) more efficiently than SeMet, and that trimethylselenide and dimethylselenide, in urine and breath respectively, are useful biomarkers of Se cancer chemoprotection.

Ruthenium compounds are being actively explored as anti-cancer agents and the intracellular protein binding patterns of two Ru-drugs were investigated by Heffeter et al.104 Using SEC-ICP-MS methodologies, uptake into cancer cell lines was determined and movement within the cytoplasm from a large protein complex to smaller, soluble proteins demonstrated.

Many of the reports that simply describe speciation work are variations on established techniques or present results from different groups. Details are given in Tables 1 and 2 while the following sections draw attention to the more original work.

Table 1 Clinical and biological materials
Element Matrix Technique:atomization:presentation Sample treatment/comments Ref.
Actinides Urine MS;ICP;L An improved procedure for the separation of actinides (Am, Cm, Np, Pu and U) from urine was reported, which was compatible with ICP-MS analysis as well as with α-spectrometry determinations. The simultaneous extraction of 24 samples was carried out, followed by precipitation with Ca3(PO4)2 for preconcentration 40
Al Parenteral nutrition solutions AA;-;- Al was not available from parenteral nutrition solutions with a Si:Al molar ratio >5, possibly because of the presence of stable hydroxyaluminoslicates. The presence of aminoacids showed a similar effect, independently of the Si concentration 126
Al Serum MS;ICP;HPLC A method was developed for the simultaneous direct quantification of Al complexes with both small and large molecules. To this aim, a C18 HPLC column dynamically coated with zwitterionic bile acid derivative, 3-[(3-cholamidopropyl)-dimethylammonio]-1-propanesulfonate was used for separation followed by both and UV and ICP-MS detection of the eluted compounds. The separation of Al complexes with transferrin and citrate was achieved in only 4 min. LODs were 0.74 and 0.83 ng mL−1 and RSD 2.8% and 3.0% (n = 7), respectively 105
Al Serum AA;ETA;L An investigation of Al exposure in chronic renal failure patients in East China highlighted significant Al accumulation despite low Al levels in dialysis fluids 127
Al Brain AMS;-;- The placental transfer of Al was investigated in an experimental study using 26Al as a tracer. Al incorporated in liver and kidney of newborn rats decreased rapidly after birth, but 15% of the dose was still found in the brain of adult rats, 730 days after birth 63
Al Placenta, umbilical cord AA;ETA;L Samples were digested with TMAH and EDTA at atmospheric pressure prior to analysis. The Al geometric mean concentration was 0.5 μg g−1 in placenta bodies and membranes and 0.3 μg g−1 in umbilical cords 50
As Biological samples MS;ICP;HPLC Analytical procedures, based on cation exchange and RP chromatography coupled with ESI-MS and ICP-MS, were developed for the identification and quantification of As metabolites (AsIII, MMAIII, DMAIII, AsV, MMAV, DMAV), arsino-glutathione, monomethylarsino-glutathione and dimethylarsino-glutathione) in human biological samples 106
MS;ESI;HPLC
As Hair XRF;-;- XANES, SRXRF and micro-XRF cartography were applied, respectively, to obtain information about the oxidation state and the longitudinal and cross-sectional distribution of As in hair from cases of acute promyelocytic leukaemia treated with As2O3 14
XANES;-;-
As Hair MS;ICP;LA As concentration was measured in single hair strands from 45 pre-Columbian mummies from different locations in Northern Chile. Values ranged from <0.8 to 262.2 μg g−1 113
As Kidney (cortex) SRXRF;-;- In As dosed rats, analysis by SRXRF of 2 mm thick renal cortex slices showed As and Cu co-distribution and peri-glomerular accumulation 120
As Urine AA;HG;L The effect of seafood consumption on the excretion of As metabolites (iAs, MMAV, DMAV) was investigated in urine samples from 16 volunteers, over a period of 14 days 184
MS;ICP;HPLC
As Urine AA;HG;HPLC The influence of genetic polymorphism on the risk of cancer associated with As toxicity was assessed in a 15 year follow up study of residents in Taiwan. Blood and urine samples were collected on two occasions (1988 and 2004) and analysed for both the distribution of As species and genotyping. Increased cancer risk was associated with a higher baseline proportion of MMA and lower %MMA change over the 15 year period. Data were available for a subgroup of 205 subjects, to investigate the changes in As species in urine 15 years after As exposure ceased 102,131
As SRMs, urine MS;ICP;LC The concentration of seven As species (AsIII, AsV, MMA, DMA, AB, AC and TMAO) was determined in a new SRM (NIST SRM 2669) by means of two independent analytical procedures, based on LC-ICP-MS, developed for this purpose. For both methods, LODs ranged from 0.2 to 0.8 μg L−1 as As 24
As Urine AA;ETA;L In a cross sectional study, measurements of urinary As and its species were used to assess the exposure of residents in an area with As-rich soil 199
MS;ICP;HPLC
As Skin XRF;-;- New developments in methodology for the application of XRF to the detection of As in skin were reported 90
As Skin phantoms XRF;-;- Simultaneous determinations of As and Se in polyester skin phantoms were achieved, using an XRF system equipped with a commercial miniature X-ray tube and silicon PiN detector, with LODs of 1.05 ± 0.02 μg g−1 and 0.88 ± 0.02 μg g−1, for As and Se, respectively 89
As Water AA;HG:L The As exposure of residents in Central Europe was assessed from cumulative data regarding the As concentration in drinking water available to them and the individual consumption of water, e.g. in drinks and food 190
As Urine AF;HG;L Three on-line oxidation methods (microwave-assisted heating, UV-photooxidation and UV-photooxidation with post-reaction heating), aimed to convert urinary As species (AsIII, AsV, MMA, DMA, TMAO, AB and AC) to AsV, prior to determination by HG-AFS, were evaluated 88
As Urine, saliva MS;ICP;L The concentrations of As species (AsIII, AsV, MMA and DMA) in urine and saliva of children playing in areas treated with chromated copper arsenate were similar to those of controls 130
As Herbal preparations MS;ICP;L A survey of 292 samples of traditional Asian herbal medicines, carried out in the Netherlands between 2004 and 2007 revealed the presence of As, Hg and/or Pb in 64% of the samples, leading to potentially excessive intakes in 20% of the cases 164
As Skin MS;ICP;LA The uptake of AsIII, DMAV and arsenosugar from skin was assessed and compared to that of AsV 132
As Nails SRXRF;-;- As concentration and distribution in toenail samples from children living in a historic gold mining area in Victoria, Australia, were investigated 96
As Blood, urine MS;ICP;HPLC The separation of five As species (AsIII, AsV, MMA, DMA and AB) was carried out on a C18 column using a 0.75 mM tetrabutylammoniumhydroxide–25 mM (NH4)2CO3–3% v/v MeOH mobile phase. Urine samples were diluted 1 + 4 with the mobile phase and filtered. Blood samples were diluted 1 + 4 with 3 mM HgCl2, to displace protein-bound AsIII, and ultrafiltered 200
Cd Blood AA;FF;L The performance of a metallic FF-AA spectrometer, coupled with a microwave-induced combustion system, was evaluated. LOD was 0.018 μg g−1 and RSD was <12% 65
Cd Hair AA;ETA;S The direct determination of the longitudinal distribution of Cd and Pb in single strands of hair was achieved with LODs of 2.8 ng g−1 (Cd) and 40 ng g−1 (Pb). The effect of two different washing procedures was compared 69
Ce Human milk, serum, plasma MS;ICP;L Ce levels between 5 and 65 ng L−1 were observed in human milk samples and between 10 and 70 ng L−1 in the corresponding blood plasma samples 138
Co Blood, plasma and urine MS;ICP;L Potential biomarkers of exposure to Co and W were investigated in a study involving Co and W determinations in samples of biological fluids from 55 workers in a factory producing cutting tools for carpentry and 34 controls 158
Cr Proteins SRXRF;-;- Cr and Fe speciation were used as examples of the potential of native- or SDS-PAGE blotting, coupled with rapid SRXRF and micro-XANES for the investigation of the role of metals in biology 115
XANES;-;-
Cu Tissues MS;ICP;L In an experimental study, after nasal instillation of Cu nanoparticles in mice, liver, kidneys and the olfactory bulb showed the highest concentrations of Cu 135
Cu Placenta WDXRF;-;- An investigation of the content of essential elements (Cu, Fe and Zn) in the placenta pointed out possible associations with the newborn's weight at birth and the mother's age 93
Cu Soft tissues MS;ICP;LA Calibration with newly developed polymer film standards was applied to improve the quantification of trace element concentrations in biological soft tissues by means of LA-ICP-MS. Experimental data were shown for the determination of Cu and Zn 111
Cu Proteins MS;ESI;L An analytical strategy, based on ESI-MS, elucidated the mechanisms of Cu transport between different cellular compartments, based on gradients of increasing Cu-binding affinity 100
Cu Reference materials MS;ICP;LA Reference materials for 2-D maps of elemental distributions were proposed, based on Cu-containing cyan ink distributed by means of an ink jet printer 110
Cu Kidney (cortex) SRXRF;-;- See As, ref. 120 120
Cu Proteins MS;ICP;LA Electrophoresis methods and conditions were investigated to highlight potential sources of error in studies of metal-binding proteins, taking superoxide dismutase and alcohol dehydrogenase, as examples 98
Cu Liver AA;F;L Heterogeneity of the distribution of Cu, Fe and Zn in canine liver was the most important factor influencing the reliability of information from liver biopsies 33
Cu Blood, brain MS;ICP;LA An investigation of Cu distribution in mouse brain was carried out using a radioactive Cu isotope (67Cu) and combining information from three analytical techniques—autoradiography, to establish Cu uptake; immuno-histochemistry, to assess the concentrations of Cu-containing enzymes; and LA-ICP-MS to determine the 67Cu distribution in brain tissue 201
Cu Serum AA;ETA;L Reference values for the concentrations of serum Cu, Se and Zn at various stages of pregnancy were established from a sample of Turkish women 31
AA;F;L
Cu Serum MS;ICP;L The concentrations of the trace elements Cu, Se and Zn and of the antioxidant enzyme glutathione peroxidase were higher in serum samples from young female gymnasts than in those from sedentary controls, whereas superoxide dismutase levels were lower 202
Cu Proteins MS;ICP;SEC Cu binding proteins (transcuprein, ceruloplasmin and albumin) from human plasma were separated and identified using SEC-ICP-MS. A novel approach was proposed for the identification of the protein peaks in the eluate, based on the determination of S content as SO in the presence of O2 as the reaction gas 58
Fe Placenta WDXRF;-;- See Cu, ref. 93 93
Fe Brain (Substantia nigra) PIXE;-;- A new, non-interceptive, technique was developed to improve the precision of charge measurement of a micro-PIXE apparatus. The application of this technique to the mapping of Fe concentrations in paraffin-embedded sections of brain tissue (substantia nigra) from 3 cases of Parkinson's disease and 3 controls did not show significant differences 119
Fe Proteins SRXRF;-;- See Cr, ref. 115 115
XANES;-;-
Fe Liver AA;F;L See Cu, ref. 33 33
Fe Nails XRF;-;- Determinations of the spatial distribution of Fe in nails revealed areas of higher Fe concentration. However, these were not associated with health or illness in the studied cases 116
XANES;-;-
Fe Liver SRXRF;-;- Mapping the distribution of Fe in the liver lobule revealed higher Fe concentrations in the periportal hepatocytes, gradually decreasing toward the perivenous area, and associated with a similar distribution of oxidative stress markers 117
Gd Contrast media solutions, human serum AE;ICP;HPLC Five GdIII complexes, used as contrast media in MRI, were determined by RP HPLC-ICP-AES, using a mobile phase of 10 mM CH3COONH4 at pH 7.0, and a sixth one by addition of 5% acetonitrile to this mobile phase. LODs ranged between 8 and 35 μg L−1 76
Gd Bones MS;ICP;L Higher Gd concentrations were reported in bone samples from the femoral heads of 13 patients with documented exposure to Gd-based contrast agents than in similar samples from 17 controls 137
Gd Kidney extracts AE;ICP;HPLC Combined information from HPLC-ICP-AES and LC-ESI-MS allowed characterization of the Gd species present in rat kidney extracts after administration of a Gd based MRI contrast agent 108
Gd Tissues MS;ICP;LA Information about the spatial distribution of Gd in tissues from a mouse tumour model after administration of PEGylated Gd liposomal nanoparticles was reported 112
Gd Brain phantoms XRF;-;- A method to determine Gd fluorescence in tumours of different size and shape was tested using a head phantom as a model and different concentrations of Gd added to spherical and spheroid simulated tumours 203
Hg Blood MS;ICP;FI-CV 0.5 ml of 10.0% TMAH was added to the same amount of blood sample and incubated for 3 h at room temperature, prior to dilution 1 + 4 with 2% v/v HCl. Two different procedures (on-line addition of L-cysteine, followed by reduction with SnCl2vs. on-line addition of KMnO4 and reduction with NaBH4) were used to determine Hgi and total Hg, respectively. Matrix matched calibration was necessary. LODs were 0.80 μg L−1 (Hgi) and 0.08 μg L−1 (total Hg). The method was validated by the analysis of NIST SRM 966 and by comparison with the results obtained on human blood samples by CVAAS 46
Hg Urine AA;-;- To investigate Hg cardiovascular toxicity, the concentration of urinary Hg and serum lipids, serum total antioxidant status, carotid intima-media thickness (IMT) and intraventricular septum diastolic diameter were measured in a sample of 154 workers, occupationally exposed to Hg. In the group of non smokers, a correlation was observed between urinary Hg concentrations and IMT 147
Hg Biological CRMs AF;CV;HPLC The simultaneous determination of MeHg+, EtHg+ and Hg2+ was achieved by means of a novel measuring system, based on HPLC separation, followed by HG with KBH4, further atomization and detection by AFS. For a 100 μL injection, LODs were 0.2, 0.4 and 0.4 μg L−1 (as Hg) for MeHg+, Hg2+, and EtHg+, respectively 80
Hg CRMs, urine MS;ICP;HPLC A hyphenated system for the speciation of Hg2+ and MeHg in aqueous and urine samples was developed and tested. The separation of the Hg species was achieved on a short cation-exchange guard column. The column effluent was mixed, using a peristaltic pump, with a TiO2 suspension and a mixture of HCOOH and HNO3 (flow rates: 0.25 mL min−1), prior to introduction into a UV-nano-TiO2 device for the reduction of the Hg species to Hg0 under UV irradiation 81
Hg Blood MS;ICP;L From the outcome of The Childhood Autism Risk from Genetics and the Environment (CHARGE) study there was no evidence of association between blood total Hg concentrations and the development of autism or autism spectrum disorder in children 148
Hg Brain AA;CV;L Simultaneous prenatal exposure to MeHg and Hg vapour led to high levels of both MeHg and Hgi in the brain of neonatal rats 204
Hg CRMs, hair AF;CV;L A modified polyaniline-graphite electrode was used as the cathode in a system for CV generation prior to AFS and applied to Hg determination in CRMs and human hair 85
Hg Herbal preparations MS;ICP;L See As, ref. 164 164
Hg Biological CRMs AA;CV;L Total and Hgi were determined in biological CRMs on the basis of selective sample pretreatment. The concentration of MeHg was calculated by difference. Extraction procedures based on either acid leaching (with HCl or HCOOH) or solubilisation with TMAH were reported and discussed 48
Hg Biological materials (blood, fish, hair) AA;-;- The Magos method was revisited in an attempt to provide a simple and reliable method for the determination of the MeHg fractions in biological materials. The proposed procedure involved sample digestion with 0.1% L-cysteine–5 M NaOH at 70 °C, followed by extraction with CHCl3 and hexane to remove fat, addition of HBr and CuCl2 to the defatted aqueous solutions, extraction of MeHgBr in toluene and back extraction of the MeHg-Cys complex in water, prior to determination by AAS 205
Hg Desulfobulbus propionicus culture media MS;ICP;GC Mass-specific methylation of Hg by the anaerobic bacteria Desulfobulbus propionicus under dark conditions was demonstrated using GC-MC-ICP-MS 149
Hg Hair MS;ICP;L Exposure of pregnant women to Hg through fish consumption was assessed from the content of Hg in hair 206
Hg CRMs, Chinese medicines AF;CV;L An electrolytic system for CV generation was evaluated, which used a Pt-Ti cathode in the presence of organic acids. The LOD was 1.4 ng L−1 and the RSD% was 2.3 at 1 μg L−1 84
Hg Sediments, hair, fish AA;-;L An investigation of the potential health risks for residents near the Babeni reservoir (Romania), due to Hg pollution, was assessed by measuring Hg and MeHg concentrations in sediments, fish and hair samples 166
MS;ICP;GC
Lanthanides Placenta, umbilical cord MS;ICP;L Using a membrane desolvation introduction system, LODs ranging from 0.3 ng g−1 to 4 ng g−1 were obtained. The method was applied to investigate the distribution of lanthanides in placenta tissue 56
Li Gallstones AE;ICP;L In a study of 170 Polish subjects, the influence of gender, area of residence and smoking status on Li concentrations in gallstones was investigated 77
Mn Cells SRXRF;-;- In an investigation of the mechanisms of Mn cytotoxicity, Mn was found to accumulate within the Golgi apparatus of dopaminergic cells 118
Mn Blood MS;ICP;L The possible influence of exposure to Mn in early life on neurodevelopment was investigated in a group of 448 Mexican children 146
P Biological samples AA;ETA;S High resolution-continuum source ETAAS with solid sampling was applied to the direct determination of P in biological samples. Atomic P was detected with an atomization temperature of 2650 °C and a palladiumascorbic acidtungsten mixture as permanent modifier. Alternatively, the molecular absorption of PO was monitored at 1900 °C in a tungsten-coated furnace. Both methods offered LODs at the μg g−1 level and RSDs in the 5–10% range 68
Pb Blood AA;ETA;L An investigation carried out on a sample of 1017 pregnant women in France revealed significantly higher (p = 0.02) blood Pb levels in pregnancy induced hypertensive cases (mean ± SD, 2.2 ± 1.4 μg dL−1) than in normotensive women (1.9 ± 1.2 μg dL−1) 207
Pb Bone XRF;-;- In an investigation on a sample of 50 to 70 years old adults from the US, tibia Pb concentration, as a marker of cumulative long term exposure to environmental Pb, was shown to be associated with persistent impairment of cognitive function 139
Pb Hair AA;ETA;S Sample pretreatment involved washing with acetone, followed by rinsing three times with water and again once with acetone and then drying at 75 °C. Aliquots of 0.05–1.0 mg were used for analysis, without addition of modifier. The LOD (3 sigma, N = 10) was 0.3 ng g−1. Longitudinal analysis of Pb distribution along hair strands was achieved by analysing 0.5 cm sections of the same strand 70
Pb Teeth MS;ICP;LA An investigation of the Pb distribution in two human teeth from an archaeological site in Manhattan provided evidence of possible Pb exposure 114
Pb Medicinal plants AA;F;FI Plant samples were digested with HNO3–HClO4. An FI system equipped with a pre-concentration column, was devised and coupled with FAAS for quantification. Measured Pb concentrations ranged from 3.3 ± 0.25 to 7.03 ± 0.51 μg g−1 dw 67
Pb Blood AA;ETA;L Early environmental exposure to Pb, as measured by the Pb concentration in blood samples taken at 30 months of age, was associated with poorer educational and behavioural outcomes at school at age of 7–8 years 140
Pb Blood, serum, saliva MS;ICP;L In a search for novel biomarkers of environmental Pb exposure, blood, serum and saliva samples were collected from a group of 444 Brazilian children, aged 6 to 8 years. Median blood Pb was 2.1 μg dL−1, with 10% of the children showing blood Pb levels between 4.0 and 9.4 μg dL−1. There was no significant evidence of a correlation between blood Pb and the other proposed markers. 143
Pb Blood AA;ETA;L The association of blood Pb levels with intelligence and personality indices was investigated in a sample of 302 Korean school children 141
Pb Herbal preparations MS;ICP;L See As, ref. 164 164
Pb Hair MS;ICP;L The 207Pb: 206Pb and 208Pb:206Pb isotope ratios were determined in acid (HNO3) digested hair samples from Japanese women, who collected them between 1910 and 1968 for personal reasons. Changes in both isotope ratios and total Pb concentrations were observed and discussed 145
Pb Hair AA;ETA;S See Cd, ref. 69 69
Pb Plasma, serum MS;ICP;L Both serum or plasma collected in heparin were shown to be suitable sample types for the determination of the Pb concentration 144
Pb Blood AA;ETA;L A comparison of the performance of the LeadCare II system (a point of care analyser based on ASV) with ETAAS was carried out on cow blood samples with Pb concentrations ranging from 3 to 91 μg dL−1. LeadCare II results were found to correlate well with atomic absorption over a wide range of blood Pb concentrations 208
Pt Blood fractions MS;ICP;L The relative distribution of cisplatin, carboplatin and oxaliplatin among different human blood fractions was assessed by means of SF-ICP-MS, after microwave aided digestion. An alternative simpler and faster treatment of plasma samples (precipitation of the plasma proteins with acetonitrile) provided comparable results and could be combined with speciation via hydrophilic interaction liquid chromatography (HILIC) with ICP-QMS detection 150
MS;ICP;HILIC
Pt Urine MS;ICP;HPLC A method was reported for the determination of oxaliplatin in urine in less than 1 min 55
Pt Leukocytes MS;ICP;HPLC A method for the determination of 1,2-GG cisplatin adducts in DNA was developed and tested on two different cell lines. The LOD was 0.21 fmol μg−1 DNA. In patients' samples, taken after treatment with cisplatin, the concentrations of 1,2-GG cisplatin adducts ranged from 113 to 1245 fg μg−1 DNA 152
Pu Urine AMS;-;- In an alternative to α-spectrometry, urine samples were acid digested in a microwave oven, followed by separation of Pu on BioRad AG1 × 2 anion exchange resin, prior to determination of Pu concentration and isotope composition by AMS 62
Ru Proteins MS;ICP;L The intracellular protein binding patterns of the anticancer Ru drugs KP1019 and KP1339 were investigated in in vitro studies. A coupled SEC-SEC system was used to separate high and low Mr species 104
MS;ICP;SEC
Sb Antileishmaniotic drug AE;ICP;L The SbIII and SbV species in meglumine antimoniate, an antileishmaniotic drug, were separated from the organoantimonial compounds by means of 1.5 M HCl and selective retention on a column filled with Dowex 1 × 4 resin 128
Sb Plasma, skin MS;ICP;L The levels of Sb in plasma and skin samples from 24 patients with cutaneous leishmaniasis treated with Sb-based drugs were measured and potential side effects assessed. Evidence was reported of a dose-dependent relationship between Sb concentration in plasma and skin and side effects 129
Se Serum plasma, aqueous humour MS;ICP;HPLC An association was reported between higher Se levels in plasma and aqueous humour and primary open-angle glaucoma 157
Se Serum MS;ICP;L In a study of 100 healthy Korean subjects (50 males and 50 females), the mean serum Se level was found to be significantly higher in women (120.81 ± 27.37 μg L−1) than in men (103.29 ± 31.05 μg L−1). No effect of age was observed 28
Se Skin phantoms XRF;-;- See As, ref. 89 89
Se Serum selenoproteins, CRMs MS;ICP;HPLC A method was developed for the quantification of the concentration of SeAlb in serum, based on selenoalbumin hydrolysis and subsequent quantification of the resulting SeMet. The results were compared with those of the determination of intact selenoalbumin in serum, together with selenoprotein P and glutathione peroxidase in the same chromatographic run in CRMs. Expected values for these materials were derived from a survey of existing literature and unpublished data 154
26
Se CSF, serum MS;ICP;FI The concentrations of Se in 35 paired serum and CSF samples were determined by means of a FI-ICP-MS procedure, in less than 2 min per sample. The LOD was 26 ng L−1 and RSD% about 5 at 2 μg L−1. A comparison with an ETAAS procedure was carried out. Se concentration ranges were 42–130 μg L−1 in serum and 1.63–6.66 μg L−1 in CSF samples 29
Se Breath, urine MS;ICP;GC The metabolism of different Se species present in food was investigated using SeMet, methylselenocysteine and methylseleninic acid labelled with stable Se isotopes. These compounds were orally administered to rats and their metabolites determined in breath and urine samples 103
MS;ICP;HPLC
Se Tissues, urine MS;ICP;HPLC The distribution and metabolism of selenohomolanthionine, a newly identified Se compound from selenised Japanese pungent radish, were compared with those for SeMet and showed different target organs (kidney vs. pancreas) but similar levels of the two major Se metabolites in urine 155
Se Blood, tissue MS;ICP;HPLC A novel Se compound, given the name selenoneine, with the molecular formula C18H29N6O4Se2 was identified in blood and tissue from bluefin tuna (Thunnus orientalis) and also in other species (mackerel, squid, tilapia, pig and chicken), but at lower concentrations 156
Se Serum AA;ETA;L See Cu, ref. 31 31
AA;F;L
Se CSF, plasma MS:ICP:L Se concentrations in paired serum and CSF samples from 89 paediatric patients were determined 209
Se Serum MS;ICP;L See Cu, ref. 202 202
Sr Teeth MS;ICP;LA The performance of LA-MC-ICP-MS for the determination of the Sr87:Sr86 ratio was tested on samples of fossil rodent teeth. In comparison with conventional sample introduction, LA-MC-ICP-MS showed an average offset of 0.0005 ± 0.0010 (2 sigma) 60
U Tissue MS;ICP;L The concentrations of 236U and 235U in human tissues were determined by SF-ICP-MS after SPE. Results for 235U were compared with those obtained by α-spectrometry 159
U Hair MS;ICP;L A method for the determination of depleted U in hair samples was reported, with an LOD of 7.21 μg kg−1 160
W Blood, plasma and urine MS;ICP;L See Co, ref. 158 158
Zn Placenta WDXRF;-;- See Cu, ref. 93 93
Zn Prostate tissue, cells SRXRF;-;- Samples of prostate cancer tissues, as well as cells, were analysed to investigate the local distribution and chemical form of Zn 161
XANES;-;-
EXAFS;-;-
Zn Soft tissues MS;ICP;LA See Cu, ref. 111 111
Zn Bones, tissues, biological fluids and excreta MS;ICP;L In an experimental study, Zn concentration and Zn stable isotope composition were determined in tissues, fluids and excreta of sheep, by means of Q-ICP-MS and MC-ICP-MS, respectively. The results provided further evidence of differential distribution of Zn isotopes in biological samples of different origin and chemical composition 162
Zn Biological and environmental samples, CRMs AF;-;- A new AFS system based on medium power single ring electrode rf capacitively coupled plasma torch and EDL lamp was applied to the determination of Zn, with LODs of 8.2 μg L−1 and 0.8–2.8 mg kg−1 Zn in liquid and solid samples, respectively. RSD between 1.0% and 10.0% and recovery of 103 ± 9% were reported 79
Zn Proteins MS;ICP;LA See Cu, ref. 98 98
Zn Liver AA;F;L See Cu, ref. 33 33
Zn Serum AA;ETA;L See Cu, ref. 31 31
AA;F;L
Zn Serum MS;ICP;L See Cu, ref. 202 202
Various Pharmaceutical samples MS;ICP;HPLC A novel instrumental setting developed for the fast screening of samples for their metal (Cr, Pd, Rh, etc.) concentration was described. An interface, consisting of a desolvating unit, an eluent splitter and a built-in peristaltic pump, allowed to perform automated FI- or HPLC-ICP-MS analyses, using reduced sample volume (μl) 54
MS;ICP;FI
Various Daphnia magna SRXRF;-;- The coupling of information from SR-based confocal micro-XRF analysis and laboratory absorption microtomography defined the 3-D distribution of chemical elements in specific areas of the crustacean Daphnia magna, used as a model 121
Various Biological tissues MS;ICP;LA A novel sample preparation was proposed, to produce solid deposits from aliquots of biological samples solubilised with HCOOH, by drying under a heat lamp. With this procedure, both internal standardization and calibration by the standard addition method can be easily achieved by spiking the solubilised sample solutions 35
Various Proteins MS;ICP;L The performance of a new TOF-mass spectrometer, combining ESI and ICP as the extraction sources, was reported in two papers 210,211
Various (4) Blood MS;ICP;L The effect of malnutrition on the uptake of Cd from food was investigated by determining the association between the concentrations of erythrocyte Cd and of nutritional markers (erythrocyte Ca, Mn and Zn; plasma ferritin) in a sample of pregnant Bangladeshi women 134
Various (4) Human milk MS;ICP;L The concentrations of Cu, Mn, Se and Zn were measured in samples of human milk taken over a period of 4 months from birth, from 31 breast-feeding women. Trends and correlations with total antioxidant activity were discussed 177
Various (4) Placenta AA;ETA;L The concentration of Pb (but not Cd, Cr or Ni) in placenta samples from an area of China, where recycling of electronic waste occurred, was found higher than that in controls 124
Various (4) Prostate tissue XRF;-;- The associations between the concentrations of Cu, Fe, Mn and Zn in prostate cancer tissue samples taken at operations and the clinical stage of the cancer were investigated to identify potential diagnostic markers 212
Various (12) Serum MS;ICP;L The performance of a method for the simultaneous determination of Be, Cd, Hg, Ir, Pb, Pd, Pt, Rh, Sb, Tl, U, and W in diluted serum by SF-ICP-MS was reported 23
Various (14) Hair MS;ICP;L 2 mL of 20% HNO3 were added to 25 mg of hair, the mixture sonicated for 2 min (50 W, 100% amplitude) then diluted to 10 mL with water. After centrifugation of the resulting slurry, 14 elements (Al, As, Ba, Be, Cd, Co, Cr, Cu, Mn, Pb, Tl, U, V and Zn) were determined in the supernatant, using aqueous calibrating solutions and Rh as internal standard. LODs (3 s, n = 20) ranged between of 0.04 and 0.9 ng g−1, but for Cu (2.9 ng g−1) and Zn (4.2 ng g−1). The method was validated by the analysis of CRMs and participation in Proficiency Testing 213
Various (16) Urine MS;ICP;L Urine samples from 412 Brazilian unexposed subjects were diluted 1 + 19 with 0.5% v/v HNO3–0.005% v/v Triton X-100, prior to determination of the concentrations of Al, Ba, Be, Cd, Co, Cs, Cu, Mn, Ni, Pb, Pt, Sb, Se, Sn, Tl and U. Matrix–matched calibrating solutions and three (In, Rh and Y) internal standards were used for quantification purposes. LODs (3s, n = 20) were in the ng L−1 range. The method was validated by the analysis of CRMs and participation in Proficiency Testing 27
Various (16) Blood, liver, CRMs AE;ICP;L A rapid (less than 4 min) procedure for the screening of trace elements (Al, B, Ba, Ca, Cd, Cu, Fe, K, Mg, Mn, P, Pb, S, Sr, Ti and Zn) in biological samples was reported, based on sample homogenisation with phosphate buffered saline solution prior to introduction of the resulting slurry in the ICP-AE spectrometer 43
Various (16) Breast, intestine and stomach tissue XRF;-;- The distribution of 16 elements (Br, Ca, Cl, Cr, Cu, Fe, K, Mn, Ni, P, Pb, Rb, S, Se, Sr and Zn) in cancerous and healthy tissue samples from breast, intestine and stomach was studied to investigate potential diagnostic value 95
Various (18) Colorectal biopsies MS;ICP;L Chemometric methods, including principal component analysis and linear discriminant analysis were applied to investigated the possible diagnostic value of trace element profiles including Ag, Al, Ca, Cd, Co, Cr, Cu, Fe, K, Mg, Mn, Mo, Ni, P, Pb, S, Se and Zn in colorectal cancer biopsies 125
AE;ICP;L
Various (20) Serum XRF;-;- The profile of 20 chemical elements (Al, Ca, Cd, Cl, Cr, Cu, Fe, Hg, K, Mg, Mn, Mo, Na, Ni, P, Pb, S, Se, Si and Zn) in serum samples from type II diabetic patients, with and without complication, and controls was evaluated by WDXRF 94
Various (31) Serum AE;ICP;L Partial least squares discriminant analysis, applied to the results of measurement of 31 trace elements in serum samples from patients with Parkinson's disease and controls, suggested that the profile of serum Al, Cu, Fe, Mn and Zn may have diagnostic value 214
MS;ICP;L
Various (31) Drinking water, urine MS;ICP;L An investigation of the potential exposure to As and other toxic elements from drinking water available to five Andean villages in Argentina was carried out 123
Various (5) CSF, serum MS;ICP;L To investigate metal transport across the blood-brain barrier, the advantages of ultrafiltration and SEC as fractionation methods to study the distribution of Ca, Cu, Fe, Mg, Mn and Zn in low and high Mr fractions of paired CSF and serum samples were compared 30
MS;ICP;SEC
Various (5) CRMs, Chinese medicines AF;CV;L Optimal conditions (cathode material, shape and area; sample flow rate; applied current; type of catholyte and its concentration) for the electrochemical generation of As, Bi, Ge, Sb and Se hydrides were investigated 86
Various (6) Serum, liver and kidney tissue AA;ETA;L In an experimental study using diabetic rats as a model, the effect of supplementation with chromium histidinate on the levels of Cr, Cu, Fe, Mn, Se and Zn in serum and tissues was investigated 215
Various (6) Blood AE;-;L Freeze dried blood samples, mixed with graphite powder containing 15% w/w NaCl were analysed directly by means of jet plasma AES to determine the concentrations of Ca, Cu, Fe, Mg, P and Zn. Liquid blood samples were analysed by the same method after evaporation and carbonization on graphite powder 73
Various (7) Pharmaceutical tablets LIBS;-;- The performance of laser induced breakdown spectrometry for the determination of Ca, Cu, Fe, Mg, Mn, P and Zn in pharmaceutical tablets was evaluated. Experimental conditions were: 29 J cm−2 laser fluence, 165 mm lens to sample distance (f = 200 mm), 2.0 μs delay time, 5.0 μs integration time and 5 accumulated laser pulses. Results obtained with the proposed method were compared with those obtained by means of ICP-AES, after acid digestion of the samples 75
AE;ICP;L
Various (7) Homeopathic medicines AA;ETA;L A survey of homeopathic products, marketed in Croatia, involving the determination of the concentrations of As, Cd, Cr, Hg, Ni, Pb and Zn, provided no evidence of a risk of undue exposure to toxic metals 163
Various (8) Drinking water AA;ETA;L A survey of the trace element (Ag, As, Cd, Cr, Hg, Mn, Pb and Se) concentrations in samples from 530 village drinking water sources in Kurdistan (Iran) revealed levels of As, Cd and Se exceeding national or WHO limits. A subsequent cross-sectional survey, involving 587 residents, provided evidence of clinical signs of chronic As poisoning in 30% of the participants 189
AA;F;L
Various (8) Serum AE:ICP;L Serum levels of Ca, Cu, K, Fe, Mg, Na, P and Zn were assessed in a group of 30, 2-year-old female native camels, from the Eastern Region of Oman 122
Various (9) Serum MS;ICP;LA Combined information on the distribution of selected chemical elements in serum from patients with bipolar disorder (treated or not treated with Li) and controls was obtained on 2-D-PAGE gels using LA-ICP-MS and MALDI-TOF-MS/MS 109


A C18 column coated with 3-[(3-cholamidopropyl)-dimethylammonio]-1-propanesulfonate, a zwitterion bile acid derivative was used by Chen et al. to speciate Al in serum.105 Complexes of Al with citrate and with transferrin were separated within 4 min and detected by ICP-MS. With this simple and rapid procedure, both species of Al in normal serum could be quantified.

There is some uncertainty over the detailed pathway of As metabolism, particularly whether some intermediates contain AsIII or AsV and whether they may be artefacts produced during extraction and the analytical procedure. In addition, a number of As species containing glutathione have been identified and again, the validity of these to the in vivo situation is uncertain. These issues were addressed by Yehiayan et al.106 Two separation methods were developed; cation exchange chromatography for AsIII, AsV, MMAV and DMAV, with ICP-MS and ESI-MS, while RP chromatography with a Spherisorb C8 column was used to separate trivalent and As-glutathione species. The chromatography was validated using synthesised standard samples and the authors recommended that their methods will be useful for investigating As metabolism. Four dry seafood products were analysed by Gao et al.107 using HPLC-ICP-MS. The major proportion of the total As was represented by three unknown species. The authors speculated, on the basis of their molecular weights determined by TOF-MS, that these were dimethylarsinic-glycerol ribose, dimethylarsinic-phosphate ribose and dimethylarsinic-sulfate ribose, and presented no harm to consumers.

In our last update, we mentioned the developing interest in Gd1 and this has continued with the demonstration of at least six GdIII complexes measured in human serum, following RP-HPLC-ICP-AES.76 Kahakachichi and Moore found that when the contrast agent ‘Optimark’ (gadoversetamide; Gd-DTPA-BMEA) was injected into rats, this was present unchanged as the major species in kidney.108

A new method to separately measure concentrations of inorganic and total Hg in blood was described by Rodrigues et al.46 Samples were diluted 1 + 1 with 10% TMAH, incubated at room temperature for 3 h and further diluted 1 + 4 with 2% HCl. An FI system was used to add L-cysteine to release Hg2+, then to add the reductant and finally to transfer the Hg vapour to an ICP spectrometer. Total Hg concentrations were determined by on-line addition of KMnO4 and NaBH4. A similar procedure was described by de Carvalho et al.49 to analyse lyophilised fish samples.

While chromatographic techniques are widely used for speciation studies, electrophoretic methodologies are important too. Becker and colleagues20,109 describe how LA-ICP-MS may be used in clinical disorders to detect serum metalloproteins that have been separated by gel electrophoresis and that further analyses using MALDI-TOF-MS and MALDI-ESI-MS help to characterise the proteins.

5.2 Imaging: LA-ICP-MS and XRF

While most reports of LA-ICP-MS refer to specific applications, a few addressed issues concerned with fundamental aspects of the technique. Bellis and Santamaria-Fernandez110 found a way to accurately assess resolution and sensitivity of measurement. Using an ink-jet printer to prepare patterns and lines of varying thickness and density, they measured Cu concentrations in the dried ink by LA-ICP-SFMS. Precision improved with increasing size of image from 100 μm and 60 ng Cu cm−2 up to 1 mm and 250 ng Cu cm−2, at which point the RSD was 3%. Anything less than 500 μm gave poor signal intensity. How to quantify images produced by LA-ICP-MS was addressed by Austin et al.111 who compared homogenised soft tissue spiked with Cu and Zn and thin films prepared from spiked solutions of polymethylmethacrylate. Excellent agreement was obtained with concentrations up to 400 μg g−1, demonstrating that the spiked films may be used for calibration.

In recent years, Becker has been particularly active in using LA-ICP-MS for bio-imaging especially in the context of metal distribution in the brain and in identifying metalloproteins after electrophoretic separation. Together with her co-workers, she has produced further examples of how the technique may be exploited and reports that, in thin brain tissue sections, it is possible to achieve spatial resolution in the μm range with the prospect of working in the nm scale in the future.19–21

Kamaly et al.112 developed liposomal nanoparticles containing a Gd-PEG complex, which were injected intravenously into a mouse tumour model. After 2 h MRI showed where Gd had accumulated and then the distribution was confirmed by LA-ICP-MS. The results demonstrate how LA-ICP-MS may be used to show the tissue distribution of MRI contrast agents.

Three publications report on LA-ICP-MS used to analyse pre-historic samples. Concentrations of As in hair from Chilean mummies found at various sites were from <0.8 to 262.2 μg g−1 indicating severe exposure had occurred in some locations.113 It would be expected that those individuals with the very high concentrations must have died as a consequence of the As absorbed. The distribution of Pb in human teeth from an archaeological burial site in Manhattan, New York, USA, was observed to be similar to that in present-day Pb-dosed goats.114 Concentrations in the cementum, at around 2000 μg g−1 were very much greater than in enamel and dentine at approximately 5 μg g−1. The 87Sr:86Sr ratio in the enamel of fossil rodent teeth in South Africa was found to be essentially identical to that of modern day teeth whereas in fossil dentine there appeared to have been considerable change due to diagenesis.

Complementing the work using LA-ICP-MS, metal metabolism may also be studied by electrophoresis and XRF. Combined with XANES analysis, information relating to structure of the species investigated can be obtained. Finney et al.115 illustrated how these techniques may be used showing that the oxidation state influences binding of Cr to serum proteins. Analysis of nails using XRF and XANES showed that As was present in discrete layers and in two forms; a low oxidation state species, possibly with mixed S and –CH3 coordination, and a higher oxidation state species. Depth profiling suggested that surface contamination was unlikely.96 In a similar study, the distribution and speciation of Fe in nails was examined.116 The XRF mapping identified areas where the concentration of Fe was up to six times greater than background levels. The number of these Fe-rich inclusions was increased in a sample from an ill subject compared to a healthy individual. Both Fe2+ and Fe3+ was detected in these spots.

Several reports revealed the distribution patterns of metals in soft tissues. Kinoshita et al.117 compared SRXRF mapping with histological staining, to determine the distribution of Fe in liver. Both techniques indicated deposits of Fe in the periportal area which were consistent with the location of oxidative stress markers. Two investigations of the relationship between metals in brain and neurological disorders were reported. Using SRXRF, Carmona et al.118 showed Mn, which causes a Parkinsonism-like disorder, accumulates in the Golgi apparatus of dopaminergic cells, They also observed that exposure to Mn causes a decrease in total Fe content and suggested that this may also be contributing in some way to neurotoxicity. The possibility of a role for Fe in Parkinson's disease is a subject of much current neurological research. Increased concentrations are found in the brain of these patients and perturbation of normal metabolism may be a feature of the disease. Barapatre et al.119 developed a micro-PIXE method for mapping trace elements in the brain. A positive association between As and Cu in the kidney of animals exposed to As has been seen in the past. Using SRXRF, Birri et al.120 showed co-distribution of these elements around the glomerulus in renal cortex slices. Damage to the glomerulus was seen by light microscopy and the blood urea concentrations were increased.

De Samber et al.121 produced 3-D maps of metal distribution within Daphnia magna, a crustacean often used as an ecotoxicological model. A confocal micro-XRF system was used for fast dynamic scanning at different cross sections of the organism and showed specific areas of metal accumulation.

5.3 Multielement applications

5.3.1 Biological fluids. Reports of multi-element analyses of blood serum included Ca, Cu, Fe, K, Mg, Na, P and Zn determined by ICP-AES in samples from Omani racing camels.122 The concentrations were similar to those reported for racing camels elsewhere in the Arabian Gulf region. In a more clinical setting, concentrations of 20 elements from patients with Type 2 diabetes were measured.94 Levels of Ca, Cu and Se tended to be lower, while Si was higher than in healthy controls. Some of these results are contrary to other reports and raise doubts about the sample collection and preparation or the accuracy of the WDXRF technique used. As mentioned in Section 2, Michalke and colleagues30 measured total and the low Mr species of Ca, Cu, Fe, Mg, Mn and Zn in serum and CSF. The proportion of low Mr species within CSF tended to be much higher for Cu, Fe, Mn and Zn than in serum with smaller differences for Ca and Mg.

Conch et al.123 found high concentrations of As, B, Ce and Li in the drinking water of five Andean villages in Argentina. As a consequence, they also measured the concentrations of these elements in urine samples from 198 women living in these areas. With median values (in μg L−1) of 26–266 for As, 2980–16[thin space (1/6-em)]560 for B, 34–531 for Ce, and 340–4550 for Li, further studies of possible adverse health effects and investigations in other villages were warranted.

5.3.2 Hair and nails. There were remarkably few reports of studies in which measuring concentrations of elements in hair or nails were featured. Nomura and Oliveira69 described methods for the direct determination of Cd and Pb in hair segments (5 mm long weighing approximately 80 μg) by ETAAS. With the conventional IAEA protocol (wash with acetone, rinse in water, wash again in acetone and dry) the concentrations of Cd and Pb increased from the root towards the tip of hair. However, when an additional wash with 0.1 mol L−1 HCl was included, Cd concentrations were reduced in all segments while Pb concentrations were decreased close to the root (5 to 12 mm) and then increased systematically towards the tip. The authors interpreted the results as differentiating the endogenous and exogenous contents. The imaging work to demonstrate the distribution of As and Fe in nails using XRF was discussed in an earlier section.96,116
5.3.3 Other tissues. A sudden burst of interest in the trace element content of placental tissue is noted during this review period. Ozdemir et al.93 used WDXRF to investigate relationships between maternal features and Cu, Fe and Zn in the placenta. Placental Cu and Fe concentrations increased as the neonatal birth weight increased. There was a similar positive relationship between maternal age and the Zn concentration. Concentrations of Cd, Cr, Ni and Pb in placentas from women living close to an electronic waste re-cycling plant and from a control area were compared by Guo et al.124 Worryingly large ranges of results were reported, e.g., 6.51–3465.16 ng g−1 for Pb in the ‘exposed’, and 4.59–3176.12 ng g−1 for Pb in the ‘control’ samples Nevertheless, the median values were significantly higher for Pb and lower for Ni in the ‘exposed’ and ‘control’ samples, respectively. No significant differences were found for Cd and Cr. Lanthanide elements were determined in placental tissues (placental body, membrane, cord) by Kruger et al.56 Detection limits were from 0.3 to 4 ng g−1. The heavier elements (Eu–Lu) were not detected in any samples but the lighter elements were found in the ng g−1 range, with more in the placental body than in other tissues. Concentrations of lanthanides in placentas matched the same pattern as in the Earth's crust indicating natural exposures.

As in previous years there has been some interest in comparing concentrations of elements in cancerous and normal tissue. Eighteen elements were determined by ICP-MS and ICP-AES in tissues from 38 patients with colorectal cancer, by Lavilla et al.125 Chemometric analysis of the results provided for some discrimination but it was not clear what this would achieve as far as diagnosis or prognosis are concerned. Thin sections of tissue from colon, breast and stomach cancers were analysed using TXRF.95 Sixteen elements were determined and there was no apparent consistency with the results, different concentration patterns were seen with different tissues.

5.4 Progress for individual elements

This review period has seen something of a revived interest in biological monitoring methods for occupational and environmental exposures after a few review periods focusing primarily on metal speciation studies. Again a select number of elements tend to dominate the literature but Gd has emerged as an element receiving a considerable degree of interest from its use as a medical imaging contrast agent.
5.4.1 Aluminium. There is evidence of a renewed interest in the determination of Al in biological matrices given the number of papers considered in this latest review period. Bohrer and colleagues126 investigated the potential for silicates, released from glass containers, to bind to Al in parenteral nutritional solutions, thereby reducing the bioavailability of the element. Parenteral nutrition solutions were heated in glass containers and the total concentrations of Al and Si determined using AAS. Unbound Al was determined by reaction with morin reagent. The researchers observed that there was no unbound Al in solutions with a Si:Al molar ratio >5 and that in solutions with a Si:Al molar ratio <5 the amount of ‘free’ Al increased as the ratio fell. They hypothesised that the concomitant presence of Si and Al in parenteral solutions reduced the bioavailability of Al.

A method was described for the determination of Al species in human serum using separation with HPLC and on-line determination with ICP-MS.105 Aluminium species were separated on a C18 column coated with the zwitterionic bile acid derivative 3-1-propane sulphonate (CHAPS).105 Satisfactory separation of large Al-protein and Al-small molecule species was achieved in four min under optimal conditions. The reported LODs were 0.74 ng mL−1 for Al-transferrin and 0.83 ng mL−1 for Al-citrate. The researchers used the method to examine Al species in human serum from healthy individuals and patients on haemodialysis. They considered this method had a number of advantages over other published methods, in particular the speed of analysis with no need for sample pretreatment. Hou and colleagues127 examined Al exposure in a group of chronic renal dialysis patients resident in Eastern China. Concentrations of Al in serum, tap water and dialysate solutions were determined using ETAAS. Levels of serum Al in a group of patients with renal failure who had yet to start dialysis treatment were compared with those from a group who were undergoing dialysis. Serum Al concentrations were significantly higher (mean 29.56 μg L−1) in dialysis patients than in non-dialysis patients (mean 1.77 μg L−1) and were also significantly higher than concentrations found in dialysis patients from centres in Belgium and Portugal. The authors concluded that Al accumulation was still a problem for dialysis patients in Eastern China despite the low level of Al in dialysis solutions. They considered that uptake via food and medication were significant contributory factors.

Two papers reported studies on the determination of Al in foetal and placental tissues. In the first,50 concentrations of Al in placental membranes, placental body and umbilical cord were determined using ETAAS. Samples were prepared for analysis using digestion with TMAH. An LOD of 0.25 μg L−1 was reported. Mean Al concentrations in the different tissues were 0.5 μg L−1 for placenta body and membranes, and 0.3 μg L−1 for umbilical cord. Aluminium was detected in 95% of placental body samples but only in 46% of umbilical cord samples. In the second paper, Yumoto et al.63 used AMS to investigate the incorporation of 26Al into the rat foetus via the placental barrier. Pregnant rats were injected with 26Al between 15 and 19 days of gestation and 26Al quantitatively determined in tissues of the newborn rats. Significant quantities of Al were determined in the kidney and liver immediately after birth but the concentrations declined rapidly during postnatal development. However, around 15% of Al incorporated into the foetal brain remained into adulthood.

5.4.2 Antimony. Interest in Sb during this review period has again focused on its role in the treatment of Leishmaniasis. A simple method for the speciation of SbIII and SbV in the anti-Leishmaniotic drug meglumine antimonate was described by Lukaszczyk and Zyrnicki.128 The two species were released from the drug using 1.5 M HCl and separated on a Dowex 1 × 4 ion exchange resin for quantitative determination using ICP-AES. Interference from a range of contaminating elements was investigated and only Bi was identified as a significant interferent. Another group129 monitored levels of Sb in plasma and skin samples from patients being treated with meglumine antimonite. Plasma and skin Sb concentrations were quantitatively determined using ICP-MS. A mean skin concentration of 9.24 μg g−1 was recorded at the end of the treatment period. The authors reported a significant correlation between Sb concentrations in the clinical samples and the reporting of adverse side effects (primarily myalgia and arthralgia).

Ogra12 presented a review of speciation studies on the “exotic elements” Sb and Te. The reported studies identified that the major excretory pathways for Sb were oxidation and methylation, whilst for Te it was methylation to trimethytelluronium.

5.4.3 Arsenic. A review was presented on the medicinal use of therapeutic doses of As2O3 in the treatment of acute myelocytic leukaemia.14 The paper described results of hair analysis for biomonitoring of As2O3therapy.

Lew et al.130 established that there was no potential health risk from exposure to As for children using playground structures treated with copper chrome arsenate wood preservatives.

Given the wealth of published studies on As speciation reported in this review series, the work presented by Yehiayan et al.106 merits highlighting. The researchers developed a method for separation of ten As species based on cation-exchange and reverse phase chromatography coupled with ESI-MS for molecular identification and ICP-MS for quantitative determination of As. In particular, they studied the effect of glutathione (GSH) concentration on the formation of As-GSH complexes and the effect of pH on the stability of As-glutathione and trivalent As species. They established that, in the presence of GSH, chromatographic artefacts were observed from the cation-exchange column. Their findings led them to re-emphasise the importance of sample preparation and choice of analytical conditions for As speciation studies. They argued that previous speciation studies needed to be re-investigated for verification of the occurrence of As-GSH complexes and DMAIII in biological matrices.

Davis and colleagues24 developed two independent LC-ICP-MS methods for the separation and determination of As species in urine. The authors determined seven As species in a new urine SRM, NIST-2669, certified for As species, namely AsIII, AsV, MMA, DMA, AB, AC and TMAO. Reported LODs for the species ranged from 0.2 μg L−1 to 0.8 μg L−1. Ito et al.88 developed a method for the determination of five As species, AsIII, AsV, MMA, DMA and AB, in whole blood using LC-ICP-MS. Blood was diluted 1 + 4 v/v with a diluent containing HgCl2 to prevent binding of AsIII to the LC column and the diluted sample was ultra-filtered to remove macromolecules above 3000 Da. The same group examined three on-line oxidation methods for the determination of total As in urine using HG-AFS. The methods were microwave-assisted digestion and UV photo-oxidation with and without post-reaction heating. All three methods used a K2S2O8 + NaOH oxidant. Aqueous calibration was satisfactory for the photo-oxidation methods but matrix matched standards were mandatory for the microwave method. The group considered the photo-oxidation method with heating had advantages over the other methods with respect to sensitivity and speed of analysis. A Taiwanese group102,131 examined profiles for As species in urine samples collected from subjects living in an arseniasis hyperendemic region. Urine samples were collected from 208 subjects in 1988 and again 15 years later. Concentrations of As species were determined using HPLC-ICP-MS and expressed as a percentage fraction of total As. The researchers established that individuals with a higher level of baseline %MMA and a lower change in %MMA over the 15 year period had the earliest incidence of cancer. They concluded that %MMA may be a potential predictor of cancer risk. Hsiech et al.59 separated As and Se species from a range of biological matrices using CE for quantitative determination using DRC-ICP-MS. Arsenic and Se species were extracted from the matrix by treatment with protease XIV and lipase. Separation of five As species and five Se species was achieved on a 60 cm fused silica column. Interferences in the determination of Se at masses m/z 78 and 80 were reduced by using CH4 as the reaction cell gas.

A study on the dermal penetration of As species was reported.132 A Franz cell system was used to investigate penetration of As species through full thickness human skin. Accumulation of As in the skin was examined using LA-ICP-MS. The authors observed that dermal transportation was heavily species dependent and AsIII and DMA were taken up at rates more than 29 and 59 times higher than AsV respectively. They concluded that dermal uptake rates of As had previously been underestimated in risk assessments where exposure to As species occurred.

5.4.4 Bismuth. Ngu et al.133 examined the binding of Bi to cysteine-rich human renal metallothionein (hMT) using ESI-MS. The authors established a binding stoichiometry of Bi-7-S-20 for the entire hMT protein, by titrating Bi3+ with apo-hMT, and also identified that the binding of Bi to the protein occurred in a “non-cooperative manner”.
5.4.5 Cadmium. It is encouraging to see an ever-decreasing number of publications on the biological and environmental impact of this toxic element, which is indicative of the rapidly declining industrial use of this element. Only three papers merited comment in this review period. In the first,65 Cd in whole blood samples was quantitatively determined using FF-AAS. Blood samples were prepared as lyophilised pellets, by mixing with a high purity graphite powder, and introduced on a quartz holder into a graphite chamber for microwave-induced combustion. Samples were ignited using a paper impregnated with NH4NO3 solution. The glass chamber was connected to the flame furnace with a PTFE and quartz tube and the combustion products transported to the air-C2H2 flame in an O2 stream. Calibration was achieved using graphite pellets with adsorbed Cd from a reference solution. An LOQ of 0.018 μg g−1 was reported for sample masses of up to 56 mg. Satisfactory values were reported for a range of CRMs (95%–99% of certified values).

In the second,69 a method was described for the quantitative determination of Cd and Pb in 5 mm hair segments using solid-sampling ETAAS. The researchers used two washing procedures to try to distinguish exogenous and endogenous levels of Cd and Pb. Hair samples were washed using either the IAEA protocol, (acetone-water-acetone), or the IAEA protocol combined with a final wash in 0.1 M HCl. The method was used to study the longitudinal distribution of Cd and Pb in hair strands. Both Cd and Pb concentrations increased from the hair root to the root tip. The researchers considered the method to be useful for biological and environmental investigations of exposure.

The third paper134 reported the findings of a study to investigate how different nutritional factors influence Cd uptake in pregnant Bangladeshi women. The study was undertaken in a rural Bangladesh area where malnutrition was prevalent and exposure to Cd via food appeared elevated. Blood samples were taken in early pregnancy and 6 months post partum. Erythrocyte Cd was determined using ICP-MS. The erythrocyte Cd concentration showed a positive correlation with erythrocyte Mn and a negative correlation with erythrocyte Ca and plasma ferritin. The authors hypothesised that the increased uptake of Cd was due to an up-regulation of a common divalent metal-ion transporter and that the negative association with Ca may suggest inhibition of Ca transport by Cd.

5.4.6 Chromium. A combination of electrophoresis and XRF was used by Finney et al.115 for the identification and quantitative determination of Cr-protein adducts in human serum. Proteins were separated by native or SDS-PAGE and Cr in the protein blots determined using micro-XANES. The authors observed that the in vitro binding pattern of Cr to serum proteins was markedly influenced by the oxidation state of the Cr species and the treatment conditions employed.
5.4.7 Copper. A comprehensive review101 was presented on the application of atomic spectroscopic techniques for metallomic studies of Cu. The authors produced a table comparing the different analytical techniques and also summarised evidence of the role of Cu in human cardiovascular disease. Two papers investigated the optimal conditions for the separation and quantitative determination of metal-binding proteins containing Cu and Zn. In the first, an important study examining the impact of electrophoresis conditions on the determination of Cu and Zn metal-binding proteins by PAGE-LA-ICP-MS was described.98 The authors identified that satisfactory determination of these metal-binding proteins depends on the nature of the electrophoretic process, the chosen trailing ion and the applied current. They considered that non-denaturing conditions were preferred for alcohol dehydrogenase, whilst for superoxide dismutase a denaturing method with tricine as the trailing ion was recommended to avoid metal-protein dissociation during the process. It was reported that protein metal losses increased with higher currents and that post-electrophoresis staining of gels was not recommended in any situation. In the second, both MALDI-TOF-MS and CE coupled with ESI-MS were used to investigate the structural characterisation of bovine superoxide dismutase (SOD).61 The results indicated that both Cu and Zn were released from the Cu-2, Zn-2 dimer of SOD during either sample preparation or ionization. The authors emphasised the point that the correct non-denaturing conditions were necessary during CE separation and detection in order to obtain reliable structural information on non-covalent metal-protein complexes. Banci and colleagues100 presented the findings of an elegant study on the intracellular transport of Cu. The researchers used ESI-MS to study the binding affinities of intracellular copper-binding macromolecules in an artificial media that mimicked the “cellular redox milieu”. The results led the researchers to hypothesise that Cu ions were transported to the proteins that require them by passing from one protein to another down a gradient of increasing Cu affinity.

Despite increasing industrial use of nanomaterials, there has been relatively little evidence in the published literature of investigations into their potential health impact. Liu and colleagues135 investigated the toxicity of nasally instilled Cu nanoparticles (23.5 nm diameter) in mice. Copper concentrations in tissues were quantitatively determined using ICP-MS and accumulation of Cu was observed in olfactory tissue, kidney and liver. Pathological changes in the liver and kidney target organs were only observed at high exposure doses (40 mg kg−1). The authors considered that the findings contributed significantly to the risk assessment for Cu nanoparticles.

Miranda et al.32 used ICP-AES to determine the distribution of Cu in liver of different breeds of beef calf fed a copper-rich diet. They also examined whether needle biopsy sampling gave a satisfactory measure of overall hepatic Cu status. Samples were taken from six regions of the liver, obtained from different cattle breeds at time of slaughter and acid digested for quantitative determination of Cu. The authors reported that the breeds differed in the absolute levels of hepatic Cu but that intralobular Cu distribution was not breed dependent. Biopsy samples did give accurate estimates of liver Cu status.

5.4.8 Iodine. Minakata et al.136 described a rapid method for the determination of the Ianion in urine using ESI-MS-MS. Urine samples were treated with ascorbic acid to reduce urine IO3 and then with NaAuCl4 to form an AuI complex, which was extracted into IBMK. Between one and five μL of the extract were injected directly into the ESI-mass spectrometer and I quantitatively determined by reaction monitoring of the I ion at m/z 127. A quantification range from 10−7 M to 10−5 M was reported for an initial sample volume of 50 μL and the method was validated by analysing a human urine CRM.
5.4.9 Iron. Research groups have reported on the determination of the spatial distribution of Fe in biological matrices using XRF methodologies. Methods using both micro-XRF and micro-XANES were described for establishing the spatial distribution of Fe in human nails.116 Analysis with micro-XRF revealed that Fe was aggregated in ‘islands’ where concentrations were six times higher than in other parts of the nail matrix. Analysis with micro-XANES revealed that the bonding environment of the Fe ion varied between different Fe- rich spots. Both Fe2+ and Fe3+ species were identified in distorted octahedral bonding arrangements. An SRXRF method was described by Kinoshita et al.117 to examine the distribution of Fe in the human liver lobule. The authors compared the distribution of Fe, determined using SRXRF, with that obtained from histological staining with Berlin Blue. Iron was preferentially distributed in the periportal hepatocytes with a decreasing gradient from the periportal to the perivenous area. The authors noted that the distribution of Fe was consistent with that of oxidative stress markers and they hypothesised that hepatocytes in this region of the liver may be primed by Fe induced free radical damage.
5.4.10 Lanthanides. Interest in the lanthanide series of elements has increased significantly in this review period. There has been further interest in the quantitative determination of Gd in biological matrices as a consequence of its continued use in contrast agents for medical magnetic resonance imaging. A robust and sensitive method was described for the separation and determination of six GdIII complexes in both native contrast medium and spiked human serum using HPLC-ICP-AES.76 Gadolinium complexes were satisfactorily separated on a C18 RP column using either 10 mM CH3COONH4 or 10 mM CH3COONH4–5% CH3CN as mobile phase. Separated complexes were quantitatively determined over the concentration range 2.5–500 μM Gd and reported LODs for the Gd species ranged from 8 to 35 ng mL−1. Chromatographic profiles were presented for the contrast agents: Magnevist, Dotarem, ProHance, Omniscan, Optimark and MultiHance. The same researchers108 characterised GdIII complexes in rat kidney following administration of the contrast medium Optimark. Gadolinium species were determined in rat kidney supernatant using HPLC-ICP-AES. Darrah and colleagues137 found high concentrations of Gd in the heads of femoral bones of patients administered chelated Gd for medical contrast imaging. Gadolinium and other lanthanides were quantitatively determined in bone digests, from 35 patients with known histories of Gd exposure, using ICP-MS. In healthy controls, reference intervals were 95% CI 0.23, 0.41 nmol g−1 for cortical bone and 95% CI 0.054, 0.107 nmol g−1 for trabecular bone. Levels in exposed subjects were up to 4800 times greater than the control values, which led the authors to conclude that the administered Gd was incorporated into bone and retained for more than 8 years. The authors also identified different levels of uptake between patients with osteoporosis and osteoarthritis, which led them to hypothesise different mechanisms for incorporation and retention of the element in these disease states. Kamaly and colleagues112 used LA-ICP-MS to localise Gd in mouse tumour tissue following administration of Gd-containing liposomes to Balb/C mice with IGROV-1 tumours. The authors reported that Gd was most prevalent in tissue regions of high vascularity. The presence of Gd in the kidney was also confirmed.

The determination of Ce in human milk and blood samples using ICP-MS was described in a study to evaluate the relationship between Ce concentrations in the two matrices.138 Blood and breast milk samples were obtained from two healthy female populations, one from Munich (n = 42) and one from Madrid (n = 26). Concentrations of Ce in breast milk samples from both populations ranged from 5 ng L−1 to 65 ng L−1 and were eight times lower than values previously reported. Serum Ce concentrations on the other hand were higher in the Spanish mothers than the German mothers, which the authors hypothesised could be due to environmental exposure to Ce in urban particulate from the increased traffic density in Madrid. They considered that serum Ce could be used as a marker of environmental lanthanide exposure. Kruger et al.56 determined the concentrations of lanthanide elements in human placental tissues using ICP-MS. Improved sensitivity was achieved with a membrane desolvation sample introduction system, which significantly reduced oxide formation and gave reported LODs between 0.3 ng g−1 and 4 ng g−1 for the individual elements. The researchers observed that the pattern of lanthanide concentrations in the placenta was very similar to that of the earth's crust, indicating natural environmental exposure to the elements.

5.4.11 Lead. Rosin17 reviewed the literature evidence of long term health effects of elevated levels of bone Pb, determined using XRF, that had arisen from either environmental or occupational exposures. The review presented evidence of an association between elevated bone Pb and cognitive decline, renal disease and hypertension. The relationship between cumulative Pb dose and cognitive function in a group of 50–70 year old Americans was investigated by Bandeen-Roche et al.139 A mean tibia Pb level of 18.8 μg g−1 was reported. After adjusting for demographic characteristics, increased tibia Pb levels were associated with a progressive decline in eye-hand co-ordination, verbal memory and learning.

The adverse effects of Pb on early childhood development were discussed in several papers. In a very interesting study, a UK research group140 investigated whether blood Pb levels below 10 μg dL−1 had an adverse impact on educational and behavioural outcomes at school. Blood samples were collected from a group of children at 30 months of age and blood Pb determined using AAS. Developmental and educational test outcomes were collected on the same children at 7–8 years of age. After adjusting for confounding factors, blood Pb levels showed a significant relationship with educational outcomes. Blood Pb levels of between 5–10 μg dL−1 were associated with a reduction of scores in reading and writing, whilst levels >10 μg dL−1 were associated with scores for antisocial behaviour. Another group141 examined the same relationship between blood Pb level and intelligence and personality scores in Korean children. Blood Pb concentrations were quantitatively determined using ETAAS and mean Pb levels of 2.79 μg dL−1 and 2.54 μg dL−1 were reported for boys and girls respectively. A four point lowering of IQ scores was observed in the group of children with higher blood Pb levels. The authors also noted that hyperactivity scores were higher in the group with higher blood Pb levels. Given the significance of the findings of these two papers, the relevance of the study by Lin et al.142 cannot be ignored. The authors investigated the potential paediatric exposure to Pb from imported Indian spices and cultural powders. Levels of Pb in a selection of imported spices (86) and cultural powders (71) were determined using XRF and bioavailability was assessed by a simple extraction test. Twenty two of the spice samples and 46 of the cultural powders contained Pb levels >1 μg g−1 and three products contained >47% Pb. The authors concluded that exposure to these products could cause elevated blood Pb levels and supported this with four case studies of paediatric Pb poisoning. Despite repeated evidence of the limitations of measurement of salivary trace elements, groups continue to investigate this matrix for biological monitoring of environmental or occupational exposures. Barbosa et al.143 evaluated salivary Pb measurements for monitoring low level Pb exposure in children. Blood Pb, serum Pb and saliva Pb was quantitatively determined using ICP-MS. It was of no surprise to this reviewer that the researchers found no significant correlation between blood, serum and saliva Pb levels, although the researchers did report a weak relationship between serum Pb and parotid saliva Pb. The question of whether serum or plasma Pb determinations are appropriate for assessing Pb exposure was posed.144 Concentrations of Pb in serum and heparin-plasma were determined using ICP-MS. The authors concluded that either matrix was appropriate for assessing Pb exposure and that the measured concentration of Pb was not affected by the standing time prior to separation.

Two papers reported on the determination of Pb in hair. Baysal and Akman70 determined Pb in human hair using solid-sampling ETAAS. Hair was washed following the IAEA protocol with acetone and water (see 5.4.5, Cd above),and 0.5–1 mg samples inserted onto the solid-sampling autosampler for quantitative determination. The authors reported an LOD of 0.3 ng g−1. Matsumoto and Yoshinaga145 used ICP-MS to study the isotope ratios of Pb in stored hair samples originally taken from Japanese women between the 1910s and late 1960s. Hair samples were digested with HNO3, diluted with H2O and the 207Pb:208Pb and 208Pb:206Pb ratios quantitatively determined. The authors reported that the earliest hair samples from the 1910s–1920s had the highest concentrations of Pb and most distinct isotope ratios, indicating likely exposure from cosmetics used heavily during this period. Contemporary levels of Pb were reported to be quite low (<1 mg kg−1) and, interestingly, showed ratios consistent with those of samples from the 1960s where hair Pb was assumed to be heavily affected by Pb from leaded gasoline.

Finally, a French group investigated the relationship between environmental Pb exposure and pregnancy induced hypertension (PIH). Blood samples were collected from 1107 pregnant women from two French municipalities and Pb levels determined using AAS. Pregnancy induced hypertension was diagnosed in 107 of the women and blood Pb levels were statistically significantly higher in these cases. The authors concluded that environmental Pb exposure may play a role in the etiology of PIH.

5.4.12 Lithium. The accumulation of Li in human gallstones was investigated by a Polish research team.77 Gallstones were collected during surgery from 170 Polish subjects and Li concentrations quantitatively determined using ICP-AES. The Li concentrations ranged from 250 μg kg−1 to 6790 μg kg−1 in men and 250 to 17840 μg kg−1 in women. The lowest Li concentrations were observed in gallstones from subjects living in rural areas of Poland, whilst much higher levels were found in subjects from Silesian cities associated with high levels of environmental pollution. The authors reported reference intervals for Li in gallstones of 800–2180 μg kg−1 and 530–2890 μg kg−1 for women and men, respectively. However, as gallstones are a feature of morbidity and not normally present within the body, the concept of reference intervals for this sample type makes little sense!
5.4.13 Manganese. Past reviews have recorded continuing growth in the application of XRF techniques for determining the intracellular location of essential and toxic elements. A French research group118 used SRXRF imaging to study the intracellular distribution of Mn in dopaminergic cells. The researchers reported that Mn accumulated in the Golgi apparatus until a cytotoxic concentration was reached, at which point it also accumulated in the cell cytoplasm. Treatment of the dopaminergic cell with Brefeldin A, a molecule that causes collapse of the Golgi body, led to a dramatic redistribution of Mn into the cell nucleus and cytoplasm, which led the authors to hypothesise that the Golgi apparatus plays an important role in the cellular detoxification of Mn.

A relationship between early postnatal exposure to Mn and neurodevelopment was hypothesised from the research conducted by Henn and colleagues.146 The group used ICP-MS to determine blood Mn levels in blood samples from 448 children from Mexico City at 12 months and 24 months of age. Mental and psychomotor development was monitored using standard tests at six-monthly intervals. A mean blood Mn concentration of 24.3 μg L−1 was reported at 12 months, which declined to 21.1 μg L−1 at 24 months. These levels are quite high in comparison with levels determined in the UK and may reflect the high environmental levels in Mexico City. The authors noted an inverted U-shaped relationship between 12 month blood Mn levels and concurrent mental development scores. However, this relationship diminished at later ages and 24 month blood Mn levels were not associated with mental development. The authors suggested a biphasic dose-response relationship between early life low level Mn exposure and infant neurodevelopment.

5.4.14 Mercury. This review has seen a return to simpler direct methods for the determination of organic and inorganic Hg in biological materials. Several groups have presented methods for the determination of Hg in biological matrices using AFS. Cizdziel et al.82 described a method for the determination of Hg in biological samples using combustion AFS. The researchers coupled a direct mercury analyser, based on sample combustion, gold amalgamation and CV-AAS, with a mercury-specific AF spectrometer to enable dual measurements from a single combustion event. Absolute LODs of 0.002 ng Hg and 0.016 ng Hg were reported for AFS and AAS detection respectively. The researchers analysed a variety of CRMs and obtained comparable results between the two detection systems. They considered combustion AFS was feasible for the direct analysis of low levels of Hg in solid biological samples thereby avoiding contamination-prone digestion pre-treatments. Three Chinese groups described methods for the direct determination of Hg in biological matrices using AFS. The first80 coupled HPLC with CVG and flame atomization AFS for speciation of EtHg, MeHg and Hg2+ without the need for any post-column sample digestion. The chromatographically-separated organomercury species were reacted with KBH4 and atomized in the flame atomizer for quantitative determination of Hg. With optimized conditions, reported LODs were 0.2 μg L−1 for MeHg and 0.4 μg L−1 for Hg2+ and EtHg respectively. The method was validated by analysing biological CRMs. The second group84 described a novel electrolytic CV generation system for the determination of Hg by AFS. Mercury vapour was generated on a Pt/Ti cathode in the presence of CHOOH. With optimized electrolytic conditions an LOD of 1.4 ng L−1 was reported for aqueous solutions. Again the method was validated by analysing CRMs and was used to determine concentrations of Hg in Chinese medicines. In the third,85 an electrochemical CV generation system with a polyaniline-coated graphite cathode was developed. The authors reported that the polyaniline-modified cathode exhibited greater sensitivity and lower memory effects than the unmodified electrode. An LOD of 1.3 pg mL−1 was reported and the method evaluated by analysis of human hair (GBW09101) and Laminaria Japonica (GBW08517) CRMs.

Groups have used solubilisation with TMAH as a sample pre-treatment for the determination of inorganic Hg in biological matrices. Matusiewicz and Stanisz48 investigated different sample treatments for the determination of Hg in CRMs. The researchers examined microwave-assisted acid digestion, ultrasonic extraction with HCl–CHOOH and solubilisation with TMAH for treatment prior to analysis with CV-AAS. The authors determined inorganic Hg and total Hg in a range of CRMs and calculated the MeHg fraction as the difference between the total and inorganic Hg measurements. The group of de Carvalho and colleagues49 used a similar approach to determine total Hg and inorganic Hg in fish tissues using FI-CV-AFS. Lyophilised fish samples were treated with 25% w/v TMAH and inorganic Hg determined by adding 0.1% L-cysteine followed by 1% SnCl2 to the FI system. Total Hg was determined using the same FI manifold by adding 0.1% KMnO4 followed by 1% SnCl2. The organic Hg fraction was again calculated as the difference between total and inorganic Hg measurements. An LOD of 37 ng L−1 was reported and no loss of Hg was observed from sample solutions reanalysed after 7 days of refrigerated storage. Rodrigues et al.46 described a simple method for the quantitative determination of total and inorganic Hg in whole blood. Aliquots of whole blood (500 μL) were diluted 1 + 1 v/v with TMAH and then 1 + 4 v/v with 2% HCl. Inorganic Hg was determined by on-line addition of L-cysteine and SnCl2, whilst total Hg was determined by on-line addition of KMnO4 and NaBH4. Reported LODs were 0.8 μg L−1 and 0.08 μg L−1 for inorganic Hg and total Hg respectively and the method was validated by analysing SRM 996 bovine blood. Chen et al.81 combined short-column ion-chromatography with online photocatalysed VG for the quantitative determination of Hg2+ and MeHg in aqueous solutions using ICP-MS. The authors reported that the photocatalysis assisted VG overcame some of the shortcomings of more conventional chemical VG using SnCl2. With optimized conditions, LODs of 0.1 ng ml−1 and 0.3 ng ml−1 were reported for Hg2+ and MeHg, respectively. The authors validated the method by analysing a urine CRM and other urine samples.

A Polish research team147 evaluated the relationship between occupational exposure to Hg and carotid artery thickness in order to identify suitable markers for Hg induced cardiovascular toxicity. The study group consisted of 154 workers in a chlorine production plant. Urine Hg concentrations were determined using AAS and two artery thickness measurements (intima-media thickness (IMT) and intraventricular septum diastolic diameter) made using ultrasound imaging. Mean urine Hg concentrations were 1.9 μg g−1 creatinine and 5.6 μg g−1 creatinine in women and men respectively. They reported a positive linear correlation between urine Hg and IMT in the non-smoking workers (n = 102) and a negative relationship between high density lipoprotein and IMT. They concluded that defensive anti-atherosclerotic mechanisms were associated with high density lipoprotein and were disturbed in smokers.

It is also sometimes welcoming to report negative findings in relation to toxic trace element determinations. An American group148 investigated blood Hg levels in children diagnosed with autism spectrum disorders. They found that blood Hg levels were typical of those of a nationally representative population and concluded that there was no evidence of a relationship between environmental Hg exposure and autism.

Finally, Rodriguez et al.149 reported exciting results indicating species-specific Hg isotope fractionation during methylation of Hg2+. The authors used GC-ICP-MS to determine the isotopic composition of Hg species during methylation of Hg2+ SRM-NIST 3133, by a pure bacteria strain (Desulfobulbus propionicus) under dark anaerobic conditions. The methylation led to a mass-dependent fractionation of Hg isotopes during the exponential growth phase as the bacteria preferentially methylated the lower mass isotopes of Hg.

5.4.15 Phosphorus. The difficulties in obtaining reliable results for P in biological matrices using ETAAS have been well documented and this analytical method has consequently not been routinely employed for quantitative determination of the element. Resano et al.68 examined the potential of high-resolution continuum-source ET spectrometry for the quantitative determination of P in solid biological samples. The researchers compared the results obtained by monitoring the atomic and molecular P lines. For AA measurements, optimum performance was obtained using a high atomization temperature (2650 °C) and a combination of Pd, ascorbic acid and tungsten chemical modifier. For molecular absorption measurements, a much lower vaporization temperature of 1500 °C and tungsten-coated graphite tube gave optimum results. Satisfactory results were obtained using aqueous calibration standards and an LOD at the μg L−1 level was reported.
5.4.16 Platinum. Methods for the quantitative determination of Pt group elements have again focused primarily on their application as active components of chemotherapeutic drugs. Michalke13 comprehensively reviewed analytical methods for Pt speciation in relation to the use of platinum-based anti-cancer drugs. The review examined the intracellular metabolism of different drugs, the reaction kinetics with different target proteins and the urinary excretion patterns, all of which enabled an assessment of the in vivo metabolism of platinum-based drugs. Five research groups presented studies on this topic. The ex vivo distribution of platinum-containing drugs in different blood compartments was investigated by determining Pt in microwave digests of whole blood, plasma, plasma ultrafiltrate and plasma pellet fraction using ICP-MS.150 The authors noted that the distribution of Pt amongst the blood compartments was dependent on the drug type but was independent of the initial drug concentration. Plasma ultrafiltrate proportions of Pt were 16.5%, 56.8% and 10.4% of the total blood Pt concentration for cisplatin, carboplatin and oxaliplatin respectively. The localization of oxaliplatin and its metabolites in tissue sections of kidney from rats treated in a similar manner as heated interoperative chemotherapy, was described.151 Oxaliplatin and its metabolites were detected using MALDI-MS and were localized exclusively in the kidney cortex. Koellensperger and Hann55 described a method for the rapid quantitative determination of oxaliplatin in urine using HPLC coupled with ICP-MS. An LOD of 0.05 μg L−1 was reported. The authors confirmed that, if samples were not analysed immediately, accurate determination of oxaliplatin required that samples were stored below −80 °C and were analysed immediately after thawing. Finally, a method for the determination of Pt-DNA adducts using HPLC coupled with ICP-MS was described.152 Adducts were purified using a highly specific enzymatic isolation procedure and HPLC for quantitative determination of the Pt content using ICP-MS. An LOD of 0.21 fmol μg−1 DNA was reported corresponding to 0.8 pg Pt for the 1,2 GG adduct. The method was used to analyse samples from patients following administration of cisplatin. Levels of the 1,2 GG adduct ranged from 113 to 1245 fg Pt μg−1 DNA after treatment compared to levels below the LOD prior to treatment.
5.4.17 Ruthenium. Heffeter and colleagues104 used SEC coupled with ICP-MS to investigate the intracellular binding of Ru to proteins following administration of the ruthenium-based drugs KP1019 and KP1339. Distinct intracellular distribution patterns were described, which indicated that the target for the drug was cytosolic rather than nuclear. The authors compared the binding pattern of ruthenium-based drugs with that of cisplatin and found distinct differences. They concluded that the ruthenium-based drugs rapidly entered cells and initially bound to larger protein complexes and organelles before the Ru was redistributed to a soluble protein fraction below 40 kDa.
5.4.18 Selenium. A review of the current understanding of Se bioavailability was presented by Fairweather-Tait and colleagues.15 The reviewers highlighted the complications associated with the quantitative determination and identification of Se species and noted that improvements in the standardization and availability of techniques for species quantification were required. Another group153 presented a comprehensive review on recent trends in the determination and speciation of Se in biological matrices using chromatographic and ICP-mass spectrometric techniques. The review addressed important aspects such as sample preparation and interferences on quantitative determination. Given the challenges identified in these Se reviews, two important papers were published on the accurate quantitative determination of Se-containing species in a human serum CRM (BCR-637).26,154 The authors reported indicative values for the Se proteins; glutathione peroxidase, selenoprotein P and selenoalbumin, based on results obtained using 13 different analytical methods all using chromatography coupled to ICP-MS.

Two groups reported the identification and determination of new Se species. In the first,155 the researchers compared the metabolism and tissue distribution of selenohomolanthionine, a novel selenoaminoacid obtained from Japanese pungent radish, with that of SeMet in rats. Selenomethionine was noted to accumulate in the pancreas, whilst selenohomolanthionine accumulated in the kidney. The researchers found no difference in urinary excretion patterns of the two species, suggesting no differences in the efficiency of metabolism and elimination of the Se species. In the second paper,156 Yamashita and Yamashita described a new Se species, selenoneine, purified from blood and tissues of bluefin tuna. The species was purified using several sequential chromatographic steps and its mass and structure determined using HR-MS and NMR spectroscopy. The new species was quantitatively determined, with other Se species, in different tuna tissues using LC coupled with ICP-MS. Selenoneine was the predominant organoselenium species in tuna tissue and the highest concentrations were determined in tuna blood. The quantitative determination of Se species in urine and breath using both HPLC and GC coupled to ICP-MS was reported by Ohta et al.103 A mixture containing 76SeMet, 77Se-MeSeCys and 82Se-methylseleninic acid (MSA) was orally administered to rats. Selenium species eliminated in urine were determined using HPLC-ICP-MS and those in exhaled breath by GC-ICP-MS. The authors reported that SeMet was less efficiently metabolized to TMSe+ and dimethylselenide than either of the other methylselenium sources. They concluded that urinary trimethylselenonium and exhaled dimethylselenide may serve as useful biomarkers for the generation of cancer chemopreventive Se species.

Methods have also been described for the quantitative determination of total Se in different biological matrices. A method was developed for the determination of Se in small volumes of CSF and serum.29 Diluted samples of serum or CSF (25 μL) were injected into an ICP-DRC-mass spectrometer using FI with a liquid chromatograph. Rhodium as internal standard was added to correct for changes in sample electrolyte concentrations. The reported method characteristic parameters were; LOD 26 ng L−1, LOQ 86 ng L−1 and 4.5% day to day precision. The method was validated by analysing serum and urine CRMs. Quantitative determination of Se in 35 paired samples of serum and CSF gave median values of 3.28 μg L−1 and 86 μg L−1 for CSF and serum respectively. The concentration of Se in CSF was independent of the corresponding serum Se level. Bruhn et al.157 investigated concentrations of Se in serum and aqueous humour from patients with glaucoma using HPLC-ICP-MS. The authors concluded that although a causal relationship could not be inferred from the findings, the data suggested that glaucoma pathology was Se-related. Reference values for serum Se in a healthy Korean population were reported.28 The reported mean serum Se levels were 120.8 μg L−1 and 103.3 μg L−1 for females and males, respectively. The authors reported no effect of age on serum Se levels.

5.4.19 Strontium. Copeland and colleagues60 examined the precision and accuracy of LA-ICP-MS for the quantitative determination of Sr isotope ratios in fossil rodent teeth from different regions of South Africa. They reported that 87Sr:86Sr ratios determined by LA-ICP-MS were offset from ratio values determined from solution measurements, using MC-ICP-MS, by a mean of 0.0005. They concluded that measurements using LA-ICP-MS were perfectly satisfactory for assessing the geographical origin of fossil rodents based on the measured Sr ratio. The authors also noted that the concentrations of Sr and Sr ratios in contemporary and fossil teeth were statistically indistinguishable.
5.4.20 Tungsten. Studies from the research group of Apostoli and colleagues have been regularly reported in this review series. In this review period, the group reported the findings of a biological monitoring study of 55 workers in the hard metal industry.158 The researchers determined Co and W in both workplace air and biological samples (blood, plasma, urine) using ICP-MS. Mean urine W levels were significantly higher in exposed workers (4.12 μg L−1) compared with unexposed controls (0.06 μg L−1). The element was excreted at higher levels than Co and had lower corresponding circulating levels in blood and plasma. The authors considered that urinary W determinations could complement the more traditional determination of Co in urine or blood in the exposure assessment of workers employed in hardmetal industries.
5.4.21 Uranides. The papers described in this review period have compared or combined mass spectrometric methods with more traditional α-spectrometry for determination of these elements in biological matrices. A method for the quantitative determination of Pu in urine and other biological fluids using AMS was described62 as an alternative to α-spectrometry for biological monitoring of exposed workers in the nuclear and research industries. Urine samples were prepared for analysis using acid evaporation and acid digestion with microwave heating followed by purification of Pu using an ion-exchange column and Pu determined using a 1 mV compact AMS system. A total sample preparation time of 10 h was reported compared to the α-spectrometric method which typically required a week. The method was sufficiently sensitive to determine 239Pu activity concentrations of 30 pBq (13 fg 239Pu). Li et al.159 determined U in human tissue samples using SF-ICP-MS following separation of U from the biological matrix by automated solid phase extraction. Results obtained with the method were comparable with those obtained using α-spectrometry. Maxwell and Jones40 used ICP-MS and α-spectrometry to quantitatively determine long-lived and short-lived actinides respectively in human urine samples. Sample preparation employed stacked extraction chromatography cartridges for rapid separation of the actinides from the urine matrix followed by pre-concentration using Ca3(PO4)2 precipitation. The purified solution was split for quantitative determination of long-lived actinides using ICP-MS and short-lived actinides using α-spectrometry. The author considered that the rapidity and relative simplicity of the method made it attractive for radiological emergency responses as well as routine occupational exposure monitoring.

Finally, a group of Italian researchers160 described a method for the determination of U in human hair using Q-ICP-MS, in order to monitor for exposure to depleted U in regions of military conflict. An LOD of 7.2 μg kg−1 was reported and the method was used to study levels of depleted U contamination in hair samples with different background levels of natural U.

5.4.22 Zinc. The potential role of Zn as a biomarker of human prostate cancer has been reported in earlier reviews in this series. A paper was presented at the 9th International Symposium on Synchrotron Radiation in Natural Science,161 which described the application of SRXRF, XANES and EXAFS for the bio-imaging and localized structure of Zn in a range of prostate cancer tissues. Austin and colleagues111 described a method for the quantitative determination of Cu and Zn in biological soft tissues using LA-ICP-MS. Calibration standards of thin polymer films were prepared by spin-coating spiked polymethylmethacrylate solutions onto a quartz substrate. The method was used to determine Cu and Zn in spiked soft tissue samples and a good agreement was found with results obtained from analysis of the same samples using solution nebulization ICP-MS.

An interesting study was presented on the variability of natural Zn isotope abundances in organs of sheep raised on a controlled diet.162 Zinc was purified from different organs using anion exchange chromatography for quantitative determination of total Zn and Zn isotope ratios using Q-ICP-MS and MC-ICP-MS respectively. The researchers observed that bone, muscle, serum and urine were enriched in the lower Zn isotope masses whilst kidney, liver erythrocytes and faeces were enriched in higher Zn isotope masses. They hypothesised that the 66Zn enrichment of the serum pool took place in the digestive tract through preferential binding of the lighter isotopes to phytic acid, which thereby controlled uptake of the metal ions. A Romanian group79 described a novel system for the quantitative determination of Zn in both solid and liquid biological matrices using AFS. The system combined a medium-power radiofrequency capacitively-coupled plasma torch with an EDL excitation source. Reported LODs were 8.2 μg L−1 for liquid samples and 0.8–2.8 mg kg−1 for solid samples. The authors considered the system easy to operate with much lower running costs than ICP-MS.

6 Applications: Drugs and pharmaceuticals, traditional medicines and supplements

It is expected that regulated pharmaceuticals will be prepared to very high standards with respect to composition of the medicines provided to patients. If there is a known or suspected contamination with an inappropriate material, remedial action must be taken immediately to maintain the manufacturing specification. This can involve the analysis of a large number of samples. Tu et al.54 demonstrated that where contamination involves metals, a high analytical throughput is possible using FI-ICP-MS. If speciation is required a preliminary HPLC stage may be included. For some pharmaceuticals the active ingredient is an element. Compounds containing Sb are effective in treating leishmaniasis. Lukaszczyk and Zyrnicki128 established a method to measure SbIII and SbV in meglamine antimonite, a widely used anti-leishmaniotic drug. The two species were separated on Dowex 1 × 4 resin with measurement by ICP-AES. de Carvalho et al.75 assessed the actual composition of multi-element tablets, using LIBS. Tablets were cryogenically ground and pelletized for analysis. The instrumental conditions were optimized and results were in agreement with those obtained by ICP-MS after acid digestion. The authors recommended LIBS for the analysis of multi-element tablets.

Food supplements enriched to contain enhanced concentrations of Se were analysed by Vale et al.44 to determine whether the actual and stated contents agreed. They used ultrasonic assisted enzyme digestion and HPLC to extract and separate SeIV, SeVI, SeMet, SeMeCys and SeCys2 with measurement by ETAAS. When the total Se concentrations in 10 samples were investigated measured in this way the results obtained were similar to those found following microwave digestion. It was found that SeMet was the major species in three of the samples while two others had mainly SeIV. In the other five there were large amounts of Se species that were not identified. It was shown that the procedure used did not alter the five key species monitored and, therefore, these were not artefacts and must have been in the original sample. The authors suggested that controls on these products should become mandatory.

Campos et al. pointed out that medicinal plants have been widely used for many centuries throughout the world but that there are few studies of heavy metal contaminants.67 As strict controls over the sale of medicinal plant preparations within the European Union come into effect from 1 May 2011 (The Traditional Herbal Medicinal Products Directive, 2004/24/EC), their report of the Pb content in medicinal plants is pertinent. These workers developed an analytical procedure which included a sample enrichment step so that even low concentrations could be reported, rather than stating “below the LOD”, and used in calculations of weekly Pb intakes. The Pb concentrations in the samples analysed were 3.37 ± 0.25–7.03 ± 0.51 μg g−1 dw.

Given that homeopathic medicines are prepared by multiple dilutions of the initial active ingredient it would be anticipated that metal concentrations should be no greater than whatever is present in the water used for dilution. Nevertheless, Tumir et al.163 elected to test 30 homeopathic products for possible contamination. Although nothing was detected in most they found: As 0.22 (1), Cd 2.78 (1), Cr 0.40–10.27 (10), Hg 0.02–0.12 (24), Ni 0.45–55.00 (19), Pb 0.33–1.29 (16) and Zn 2.2–27.8 (11). The concentrations are in μg g−1 and the numbers of positive samples are given in parentheses.

Two publications report on the analysis of Asian herbal164and Ayurvedic97medicines. Results from a survey of 292 herbal preparations used in Ayurvedic, Chinese and Tibetan traditional medicines available in the Netherlands, were presented. From the concentrations of As, Hg and Pb measured, and the recommended doses, the estimated weekly intakes were calculated and 59 would exceed safety limits: 50 for Hg, 26 for As and 8 for Pb.164 In the second report Borgese et al.97 used TXRF to analyse hair from, and medicine used by, a patient with suspected poisoning.

7 Applications: Foods and beverages

7.1 Progress on individual elements

7.1.1 Arsenic. Three unknown arsenosugars were identified in various Chinese dry seaweed products (kelp, Sargassum fusiform, laver, Enteromorpha prolifera) by Gao et al.107 who undertook a detailed As speciation by HPLC-ICP-MS. The arsenosugars accounted for approximately 63.5% to 92.7% of all species detected and were identified as DMA-glycerol ribose, DMA-phosphate ribose and DMA-sulfate ribose (Mr 329.0599, 483.0738 and 409.0162 confirmed by HPLC-TOF-MS). The main species was DMA-sulfate ribose (51.1%–80.3%), while that in kelp was DMA-phosphate ribose (48.9%). Curiously the authors state that while total As exceeds the Chinese National Standard value, most of the marine plants dry seaweed products are safe to eat! Arroyo-Abad et al.165 reported the characterization of three arsenolipids extracted and purified from canned cod liver by SPE (silica gel column and ethyl acetatemethanol as eluents) using GC-ICP-MS, MIP-AES and ESI-QMS. The structures of the three lipids were confirmed by TOF-MS.
7.1.2 Mercury. Investigation of stable isotope fractionation of metals have revealed interesting information in other fields and have recently been used to obtain information concerning speciated Hg. Rodriguez-Gonzalez et al.149 examined stable isotope fractionation of Hg during methylation by anaerobic bacteria (Desulfobulbus propionicus MUD10 (DSM 6523)) of HgII from SRM NIST 3133 under dark conditions. The isotopic composition of HgII and MeHg in these cultures was measured by GC-MC-ICP-MS after two different incubation strategies. The results obtained showed that methylation of HgII causes mass-dependent fractionation of the Hg isotopes for both HgII substrate and produced MeHg preferentially on the lighter isotopes of Hg. After 96 h (continuous culture) and 140 h (single sampling cultures), a change in the fractionation trend suggested an increasing contribution of a simultaneous process balancing methylation extent such as demethylation. The authors calculated the fractionation factor, α202/198 = 1.0026 ± 0.0004 (assuming Rayleigh conditions) for the single sampling cultures. Bravo et al.166 studied the impact of Hg from chlor-alkali plants as a point Hg source for aquatic environment contamination in the Babeni reservoir (Romania). Sediments, fish and human scalp hair were collected for total Hg determination by AAS. MeHg was determined in the fish muscle and liver tissues by ID-capillary-ICP-MS. High Hg concentrations were found in the sediments and in fish from the reservoir with a median Hg concentration of 2.1 mg kg−1 in sediments and a mean value of 1.8 ± 0.8 mg kg−1 ww in fish muscle. MeHg concentrations in fish were well above the WHO guidelines for fish consumption. The scalp hair of reservoir-fish consumers was significantly higher (median value of 2.5 mg kg−1) than those consuming fish from upstream reservoirs. The authors highlighted a considerable health risk for the local fish consumers in bodies of water located near such plants. Modification of the graphite electrode with polyaniline as the cathode material in an electrochemical CV-AF spectrometer for determination of HgII was detailed by Jiang et al.85 The CV generation system exhibited better sensitivity, stability and lower memory effects than a standard graphite electrode. The precision was 2.7% (n = 11, HgII spike 2 ng mL−1) and the LOD was 1.3 pg mL−1.
7.1.3 Selenium studies. With the mechanisms of Se-enrichment in egg products still poorly understood and difficult to control, Lipiec et al.167 optimized a method for the simultaneous determination of SeMet, SeCys, and SeIV in chicken eggs. In order to achieve complete protein digestion and to avoid SeCys losses, all stages of sample preparation (defatting, protein denaturation and carbamidomethylation) were optimized. After quantitative isolation of the Se-containing fraction by SEC, RP-HPLC-ICP- MS was used to quantify the species. The LODs were 0.06, 0.003, and 0.01 μg g−1 dw for SeCys, SeIV and SeMet, respectively, with a precision of 5–10%. The yolk was shown to contain more SeCys and the white more SeMet. Digestion with simulated gastric and gastrointestinal juices followed by SEC-HPLC-ICP-MS gave some insight into Se bioaccessibility in eggs. A comprehensive Se profile of soybean (Glycine max) plant system (bean, pod, leaf and root), grown in soil supplemented with Na2SeO3, was undertaken by Chan et al.168 Size exclusion chromatography revealed that high Mr Se species characterized a large amount of the Se present in the bean compartment (82% of total Se) while low Mr (<5 kDa) Se species accumulate in the pod, leaf and root compartments. The predominant low Mr Se compounds found in the bean were SeMet and SeCys2, whereas Sei was the major species detected in other plant compartments. An investigation into the suitability of Se-enriched wheat grain and chicken meat allowed Govasmark et al.169 to point out some pitfalls in assessing bioaccessibility. The use of commercial enzymes, compared with human gastric juices, tended to overestimate Se bioavailability, and NaCl or Tris-HCl were not suitable for the extraction of Se proteins from enriched products. Cubadda et al.170 showed the highest Se concentrations ever recorded in cereal grains (29–185 μg g−1) from wheat collected in the Punjab (India). A strong positive relationship between Se in shoots and kernels showed that Se in grain could be predicted from that in plant tissues. Ultrasound-assisted enzymatic extraction followed by HPLC-DRC-ICP-MS was used to identity and quantify Se compounds in the grain samples. The sum of Se species was accounted for by SeMet (72–85%) and SeVI (2–6%). It was noted that, with increasing Se content, the proportion of Se as SeMet showed a relative reduction whereas the proportion of other organo-Se compounds increased between 18 and 22% and SeMetSeCys was detected as a minor compound (0.2–0.5%) in high-Se wheat. Although regular consumption of locally produced wheat-based food items may lead to excessive Se intake, the large predominance of SeMet shows that local wheat could be a promising raw material for naturally-enriched products. The effect of Se-enrichment of rice with foliar fertilization was studied by Fang et al.45 using ion-pair RP and strong anion exchange chromatography with ICP-MS detection, following ultrasound-assisted amylase hydrolysis. In Se-enriched rice, SeMet, confirmed by nano-ESI-IonTrap MS, was found to make up approximately 86.9% of the total Se, while this was only 26.7% in non-supplemented rice. Additionally, SeIV accounted for nearly 60% of Se present in the non-enriched rice, and only 6.8% of Se (as SeIV and SeVI) was found in the Se-enriched rice; small amounts of SeCys2 and selenomethionine selenoxide were also present. Elevated SeMet levels in foliar supplementation confirmed a high inorganic to organic conversion and, given that SeMet is readily bioavailable for animals, Se-enriched rice could be considered for supplementation in Se-impoverished agricultural regions.
7.1.4 Other elements. Variations in the in vivo distribution of Zn stable isotopes species have been studied in sheep fed a controlled diet and monitored using MC-ICP-MS. 162 Enrichment of 66Zn occurs in bone, muscle, serum and urine whereas it is depleted in faeces, red blood cells, kidney and liver because of preferential binding of lighter isotopes (64Zn) to phytic acid, the controlling factor in metallic element uptake. A simple time-dependent two compartment model, involving the gastrointestinal tract, and muscle and bone, predicts that the maximum 66Zn enrichment is reached after about 10 years. This study, which brings a better understanding of the variations of natural isotopic abundance of Zn in animals and humans, will probably bring new perspectives for the assessment of their Zn status. This dynamic role for metals in biology, through redistribution of metal ions among ligands, was further investigated by Finney et al.115 They identified and quantified metal-protein adducts in complex samples using native- or SDS-PAGE, blotting, and rapid SRXRF mapping with micro-XANES of entire blots. The authors demonstrated their approach though examination of two cases: the first tackled speciation of the in vitro binding of exogenous Cr to blood serum proteins which was influenced by both oxidation state and the treatment conditions; and the second, where in vivo changes in endogenous metal speciation probed the influence of O2 depletion on Fe speciation in the proteobacterium, Shewanella oneidensis.

Method development for trace analysis in drinking water, in particular the use of carbon nanotubes, continues to capture the worthy interest of many researchers. Trace speciation of Tl in drinking water using an FI-ETAAS method was described by Gil et al.53 Multiwalled carbon nanotubes (MWNT) were employed as the SPE sorbent in a continuous flow system. Selectivity for TlI was possible, with determination of TlIII calculated from total Tl after reduction with hydroxylamine. The LOD and precision were 0.009 μg L−1 and 3.9% respectively and recovery was 96 to 100%. Another MWNT preconcentration method was described by Ozcan et al.52 for the determination of FeIII, CuII, MnII and PbII. Adsorption was achieved quantitatively at pH 9.0 with elution by 1 M HNO3 in acetone to give a preconcentration factor of 20. LODs ranged from 3.5 μg L−1 (Mn) to 8.0 μg L−1 (Pb). Instead of SPE, Lafleur and Salin51 suggested a novel centrifugal microfluidic device for field sampling and on-site preconcentration of water samples. Metal-8-hydroxyquinoline complexes were adsorbed onto miniature C18-bonded silica gel columns on centrifugal microfluidic discs. Analytes were directly vaporized from the disc with LA-ICP-MS. Absolute LODs ranging between 0.1 to 12 ng were obtained using drinking water CRMs. The centrifugal discs required <600 μL and could be used for multiple extractions simultaneously as well as being easily transported to the laboratory for analysis.

Table 2 Foods and beverages
Element Matrix Technique:atomization:presentation Sample treatment/comments Ref.
Al Parenteral solutions AA;-;L The presence of Al and Si, as concomitant impurities in parenteral solutions, reduced the Al available, due to the formation of hydroxyaluminosilicates 126
As Drinking water AA;F;L Ag, As, Cd, Cr, Hg, Mn, Pb and Se were measured in 530 drinking water sources from the Kurdistan province of Iran. Of these, 28 exceeded WHO or National guidelines. 587 participants from 211 households were examined for evidence of chronic arsenical poisoning, which was manifested in 30.7% of the participants 189
AA;ETA;L
As Water AA;HG;L Lifetime exposure of 1392 subjects from Hungary, Romania and Slovakia was calculated from new and historic As water concentrations, as part of an epidemiological case-control study of As and cancer 190
As Cow's milk, water AA;HG,FI As was determined in cow's milk in Argentina and in the drinking water of the cattle. The results showed a low biological transference to the milk from consumption of the water 178
As Wine MS,ICP;ETV AA;ETA;L MS;ICP;L The use of ETV-ICP-MS for As, Pb and Se determination was compared with more conventional techniques. The most noticeable difference was the reduced impact of polyatomic interferences on As using the ETV approach 57
As Rice MS;ICP;HPLC Total As concentration range was 65.3–274.2 ng g−1, with an average value of 114.4 ng g−1 in a study of samples from different production areas of China. AsIII, AsV, MMA and DMA were found in most samples, with the inorganic species accounting for 72% of the total. The authors concluded that the risk of As exposure, through rice, to the Chinese population was underestimated 182
As Rice MS;ICP;L A detailed and useful study of translocation was described. Total As in flag leaf, grain and husk was quantified by ICP-MS and As speciated in fresh grain by XANES. DMA was translocated to the grain with over an order magnitude greater efficiency than iAs and was more mobile than AsIII in both phloem and xylem. The study also demonstrated altered DMA speciation in planta, potentially through complexation with thiols 179
XANES
As Rice MS;ICP;L A partial digestion method, using HNO3, was used to speciate As in several rice varieties. Elevating temperatures to 100 °C increased efficiency to 90% 181
As Cod liver MS;ICP;GC 3 novel arsenosugars were identified following extraction, purification and characterisation using solid phase extraction with ethyl acetateMeOH and then MS 165
TOF-MS;-:L
As Seafood MS;ICP;L 3, initially unknown, arsenosugars were identified as DMA-glycerol ribose, DMA-phosphate ribose and DMA-sulfate ribose 107
TOF-MS;-;L
As Seaweed AF;HG;L Samples were microwave digested using HNO3–H2O2, yielding an LOD of 2.3 μg g−1 on a dw basis 185
As Seafood MS;ICP;L MS;ICP;HPLC The standard additions method in conjunction with 77Se as internal standard gave the best results. A number of extraction procedures were also investigated. MeOH was found to enhance the As signal 186
As CRMs, dietary supplements MS;ICP;CE A DRC, utilising CH4 as reaction gas, reduced Ar2+ interference in the determination of Se, as part of a wider study of As and Se speciation 59
As Cereals SIMS;-;S In an interesting study of the distribution of As and Se in grains, As was found to be concentrated in the subaleurone endosperm cells, in association with the protein matrix. Se predominated in protein surrounding the starch granules in the endosperm cells and was more homogeneously distributed in the aleurone cells but with Se-rich hotspots 64
SRXRF;-;S
As Rice and other cereal based products -;-;- Japanese products, such as misos, syrups and a fermented drink produced from rice, barley and millet were analysed for total As and a subset for As species. Rice based products displayed the highest total As content, with the predominant species being inorganic in nature 180
As Rice CRMs MS;ICP;HPLC A method based on heat-assisted aqueous extraction was shown to yield accurate results 216
As Diets AF;HG;L A dietary intake survey undertaken in Jinhu, China found that weekly As intake in 3.3% of children and 6.7% of adults exceeded the PTWI 173
Br Fruit, grain products WDXRF;-;S Sample, 5 g, (fresh, dried and frozen fruits, grain products) was twice extracted with distilled H2O and further diluted to 25 mL with distilled H2O. The resulting solution, 0.5 mL, was dripped onto a filter paper which was dried and analysed by WDXRF 42
Ca Bottled water AA;-;L Ca and F concentrations were on average higher than the label declaration on 21 branded samples from Riyadh, Saudi Arabia 217
Ce Human milk MS;ICP;L The relationship between Ce content in human milk and blood plasma or serum was studied by collecting blood and human milk samples at various stages of lactation from volunteers from Germany and Spain 138
Cl Mineral water AE;ICP;L A new oxidation-vaporisation method, based on gaseous introduction of Cl, was claimed to overcome the traditional low sensitivity of Cl analyses caused by the element's high excitation energy 72
Co Wine AA;ETA;L A method with an LOD of 0.8 μg L−1 was developed 218
Co Water, food AA;F;L Co, Cu and Pb were complexed with 1-phenylthiosemicarbazide, extracted into the surfactant-rich phase of Triton X-114, separated and taken for FAAS measurement, yielding LODs of 1.0, 0.67 and 3.42 μg L−1, respectively 39
Cr Water MS;ICP;LC CrIII and CrVI were separated on a guard column using gradient elution, ultimately achieving LODs of 0.04 and 0.02 μg L−1, respectively 192
Cr Bread AA;ETA;L CrTotal and CrVI were determined following selective alkaline extraction of the latter. In 152 bread samples CrTotal contents were 47.3 ± 20.0 and 50.9 ± 22.2 μg kg−1 dw for white and wholemeal bread samples, respectively; those for CrVI were 5.65 ± 5.44 and 6.82 ± 4.88 μg kg−1 dw 176
Cu Cane spirits AA;F;L Ag, Bi, Co and Ni were evaluated as internal standard to minimize transport interferences, with Ag at a concentration of 2 mg L−1 most effective 219
Cu Beer, Wine AA;F;L Using 2 solid phase columns, Cu was speciated into 3 groups – cationic, hydrophobic and “residual”, the last of which constituted 72–82% of the total Cu in beer with the cationic species being 27–72% of the total in wine 194
Cu Water, food AA;F;L See Co, ref. 39 39
F Bottled water AA;-;L See Ca, ref. 217 217
Fe Fortified flour, cereal based products AA;F;L In 2002, Brazilian authorities introduced requirements for wheat and corn flour to be fortified with Fe (and folic acid). This report found Fe and folic acid to be heterogeneously distributed in the target cereal products and the authors noted the average product Fe content observed could affect Zn absorption 183
Hg Vinegar AF;Vaporization;L An ingenious paper described analysis of vinegars via the application of the inherent ability of acetic acid to reduce Hg in the presence of UV irradiation. Using the approach an LOD of 0.08 μg mL−1 was achieved 83
Hg Cattle and fowl liver and muscle -;-;- Concentrations in samples taken from animals grazing in the vicinity of Tanzanian gold mines were significantly higher than reference samples. Levels up to 436 and 820 μg kg−1 ww were found in cattle and domestic fowl samples, respectively 188
Hg Biological CRMs AF;CV;HPLC MeHg, EtHg and iHg were determined at LODs of 0.2, 0.4 and 0.4 μg L−1 (for a 100 μL injection) using HPLC-CV-Flame atomisation-AF 80
Hg Fish, biological materials AA;CV;L Samples were solubilised with NaOH at 70 °C and defatted using CH3Cl–hexane. HBr was used to acidify the solubilised solution to form MeHgBr. Heating was used to promote Hg vaporization 205
Hg Fish muscle AA;CV;L Ultrasound-assisted extraction yielded an LOD and LOQ of 0.133 and 0.445 μg kg−1, respectively 37
Hg Fish AF;CV;FI Lyophilised samples were extracted using 25% w/v TMAH. By adjusting the reagents, speciation was possible 49
Hg Fish tissues, hair MS;ICP;GC AA;F;L Hg contamination originating from a chlor-alkali plant was investigated. IDGCICP-MS was used to determine MeHg in fish tissues. The authors concluded pollution in the reservoir, and probably many other lakes and reservoirs receiving Hg polluted wastewater, represented a considerable health risk for local fish consumers 166
Hg Rice AF;CV;L Samples were microwave digested using HNO3–H2O2, diluted in KBr–KBrO3 and hydroxylamine then reduced with SnCl2 in HCl. The LOD was 0.9 ng g−1 220
Hg Food RMs AA;CV;L Following treatment with TMAH, iHg was determined by CVAAS. Total Hg was determined following microwave acid dissolution and orgHg estimated by difference 47
Hg Honey AE;ICP;CV Hg was complexed with ammonium diethyldithiophosphate and Triton X-114 was used as surfactant, permitting an LOD of 2.2 ng g−1. Sonication with L-cysteine as a means of pre-treatment was not effective for total Hg but satisfactory for iHg 38
I Foods MS;ICP;L Microwave digestion using HNO3–HClO4 yielded an LOD of 29 ng g−1 and RSDs of 10 and 1.3% for 10 replicate measurements of 100 and 1000 ng g−1 standards, respectively 221
Mn Water, food CRMs AA;F;L MnII was co-precipitated using Zr(OH)4. Total Mn was determined by reducing MnVII to MnII using ascorbic acid 41
P Biological samples AA;ETA;S An HR-continuum source ETAA spectrometer was used to determine P, with monitoring of both atomic and molecular lines. Modifiers and atomization temperatures were investigated 68
Pb Water, food AA;F;L See Co, ref. 39 39
Pb Wine MS,ICP;ETV AA;ETA;L MS;ICP;L See As, ref. 57 57
Pb Spices XRF;-;S The risk of paediatric exposure in American subjects, arising from imported Indian spices and “cultural” materials (e.g. the skin dye sindoor) was determined. 22 out of 86 spices/foodstuffs and 46 of 71 cultural products contained >1 μg g−1 Pb 142
Se Milk AE;MIP;GC Using SPME, LODs ranged from 70–110 pg mL−1 for DMSe and from 80–400 pg mL−1 for DMDSe, neither of which species was found above LODs in any sample studied 74
Se Wine MS,ICP;ETV AA;ETA;L MS;ICP;L See As, ref. 57 57
Se Foods -;-;- A study of Swiss foods found that the major sources of Se in the diet were pasta and meat 174
Se Se-enriched mushrooms AA;ETA, L Se-enriched and control mushrooms were fed to rats and blood plasma analysed following sacrifice. SEC with UV detection was used to identify protein-containing aliquots which were taken for measurement by ETAAS 222
Se Food supplements AA;ETA;HPLC Ultrasound assisted enzymatic digestion combined with HPLC-ETAAS was proposed as a method for routine Se control 44
Se Yeast supplements MS;ICP;HPLC Enzymatic hydrolysis was used for speciation. The data supported a hypothesis that S-(methylseleno)cysteine is formed by reaction between DMDSe and cysteine or cystine 223
AE;-GC
Se Soybean MS;ICP;HPLC MS;ESI;- The distribution of various species was reported. SeMet and SeCys2 were predominant in the bean, while Sei was predominant in other plant compartments 168
Se Wheat MS;ICP;HPLC It was claimed that wheat collected in the Nawanshahr-Hoshiarpur Region of India contain the highest ever recorded levels of Se, being 29–185 μg g−1. Speciation was conducted using ultrasound-assisted enzymatic extraction with an ultrasonic probe, followed by HPLC-DRC-ICP-MS 170
Se Cereals SIMS;-;S See As, ref. 64 64
SRXRF;-;S
Se Bluefin tuna, tissues MS;ICP;HPLC A novel compound, named ”selenoneine” was extracted and characterised in various tissues of the Bluefin tuna. It was then found in a limited number of other tissues from squid, tilapia, pig, and chicken 156
Se CRMs, dietary supplements MS;ICP;CE See As, ref. 224 59
Se Chicken meat, wheat MS;ICP;FI A bioaccessibility study using commercial enzymes and human gastric juice suggested the former overestimated Se bioavailability 169
Se Eggs MS;ICP;HPLC,SEC Following defatting, protein denaturation and carbamidomethylation, SeCys, SeIV, and SeMet, were measured using SEC-HPLC-ICP-MS with LODs of 0.06, 0.003, and 0.01 μg g−1 dw respectively. The carbamidomethylation step was necessary to prevent losses of SeCys 167
Si Parenteral solutions AA;-;L See Al, ref. 126 126
Sr Foods MS;ICP;L As part of a critique of another paper, the author expounded his view of the geochemical and instrumental fundamentals for accurate and precise Sr isotope data 198
Tl Drinking water AA;ETA;FI Speciation was effected by virtue of the selective adsorption of TlI on carbon nanotubes. TlI was reduced to TlIII using hydroxylamine to measure total Tl 53
Zn Fortified flour, cereal based products AA;F;L See Fe, ref. 183 183
Zn Yoghurt AA;F;Sl Sample, 1 g, was mixed with 2.0 mol L−1 HCl and sonicated for 20 min before final dilution to 25 mL and measurement using a continuum source HR-FAA spectrometer. The LOQ was 0.32 μg g−1 66
Various (8) Drinking water AA;F;L See As, ref. 189 189
AA;ETA;L
Various (31) Water MS;ICP;L Samples from a number of villages in Northern Argentina were found to contain some very high elemental concentrations, including As (highest 210 μg L−1) Li (1000 μg L−1), Cs (320 μg L−1), Rb (47 μg L−1) and B (5950 μg L−1). These elements were also found elevated in urine samples from local inhabitants 123
Various (10) Milk AA;ETA;L Ru-coating was used as modifier for the simultaneous determination of 2 groups of volatile (As, Bi, Pb, Sb, Se) and non-volatile (Co, Cr, Cu, Fe, Mn) elements. Sample preparation involved a simple 1 + 6 dilution in HNO3 71
Various (4) Human milk MS;ICP;L Changes in total antioxidant status of human milk were evaluated during the first 4 months of lactation, and correlated with the content of the antioxidant elements Cu, Mn, Se and Zn. A significant difference was found only for Cu and Zn concentrations at 7 days of lactation 177
Various (38) Blood orange juice AE;ICP;L AA;ETA;L The bioavailability of nutritionally important elements in Sicilian samples was assessed. It was low for Cu, Fe and Zn 225
AA;HG;-
AA;CV;-
Various (4) Beverages XRF;-;L The feasibility of using a portable XRF spectrometer was evaluated. Accurate results were obtained for concentrations >20 mg kg−1, although for industrial use concentrations of interest would lie below these levels (As, Cd, Hg, Pb) 92
Various (4) Beverages, cocoa XRF;-; The same author as ref. 92 applied the portable XRF to other elements and food groups (As, Cr, Cu, Pb) 91
Various (57) Bottled waters MS;ICP;L Elemental concentrations in 294 samples of the same water, but stored in bottles made from glass, hard and soft PET, showed significant differences for 26 of 57 analytes. The effect of pH and bottle colour was also investigated 193
Various (5) Drinking water RM MS;ICP;LA A novel approach for in situ sample preconcentration began with analyte complexation with 8-hydroxyquinoline. The complexes were adsorbed onto C18 minicolumns on centrifugal microfluidic discs followed by vaporization in the laboratory using LA. Absolute LODs were in the range 1–12 ng (Co, Cu, Ni, Pb, V) 51
Various (4) Food, water -;-;- Cu, Fe, Mn and Pb were adsorbed on the walls of carbon nanotubes as a means of preconcentration. Elution was achieved using 1 M HNO3 in acetone yielding LODs in the range 3.5 μg L−1 (Mn)–8.0 μg L−1 (Pb) 52
Various Vines, wine must TRXRF;-;- In an investigation of heavy metal content in Portuguese vineyards, Cu in vine-leaves from an older vineyard was extremely high, with the authors speculating this was due to excessive use of Cu-based fungicides to control vine downy mildew 195
EDXRF;-;-
Various Wine -;-;- Over many years this review has reported the application of atomic spectrometry to identifying wine vintages. This paper was the first in a planned series describing a large European project to establish a databank of analytical parameters in wine. 1800 samples from 3 vintages were analysed 226
Various (14) Olive oil -;-;- By combining the isotope ratios for C, H and O with the elemental concentrations for Ca, Ce, Cs, Eu, K, La, Mg, Mn, Rb, Sr, U, V and Zn, then applying a multivariate discriminant analysis, a good discrimination between olive oils from 8 European sites was achieved 197
Various (26) Olive oil -;-;- Following ultrasound assisted extraction, it was possible to quantify rarely measured elements such Ce, Cs, La, Rb and Yb in a study aimed at developing an Italian national databank for olive oils 196
Various (16) Bovine liver, blood AE;ICP;Sl Sample was homogenised for 1 min in phosphate buffer saline solution and analysed directly, giving a total analysis time of <4 min 43
Various (4) Seafood -;-;- A survey of numerous samples from Campania, Italy was used to calculate the population's weekly intake of As, Cd, Hg and Pb 175
Various (4) Wild boar -;-;- The relationship between age and bioaccumulation of As, Cd, Hg and Pb was investigated 187
Various (40) Fish tissues MS;ICP;L Comparison of 6 different digestion methods identified microwave digestion with HNO3–H2O2 as the preferred method 34
Various (8) Foods MS;ICP;L The Cd, Co, Cu, Fe, Mn, Ni, Pb and Zn content of 1215 Lebanese foods was determined. Intakes of Mn in men and Fe in women were below nutritional recommendations 172
Various (4) Foods AF;-;- A review, with 157 references, of the application of AFS in speciation studies 10
Various (9) Foods AE;ICP;L A simple and rapid method for use in food poisoning crises was proposed. Analytes were extracted using H2SO4 in a disposable tube using a homogenizer. The extract was diluted with 1% HNO3 and analysed by ICP-AES 36
Various Biological and clinical materials, food and beverages -;-;- A review, with many references, of the development and application of atomic spectrometry to the named sample types 1


7.2 Single and multi-element applications in food and beverages

7.2.1 Dietary intake studies. The UK Total Diet Study (TDS) has noted that the UK population's metals exposure has generally declined over 30 years.171 A range of elements were measured (Al, As, Ba, Bi, Cd, Cr, Cu, Ge, Hg, In, Mn, Mo, Ni, Pb, Pd, Pt, Rh, Ru, Sb, Se, Sr, Th, Sn and Zn) in 20 food groups collected from 24 UK towns along with estimated dietary exposures. With the exception of Al, Ba and Mn, they were found to be similar to, or lower than, the previous study six years earlier. A similar TDS approach examined the intake of micronutrient elements (Co, Cu, Fe, Mn, Ni, Zn) and heavy metals (Cd, Pb) in the Lebanese urban adult population.172 Average daily intakes of micronutrient elements (11.4 μg d−1 (Co), 1104.19 μg d−1 (Cu), 13.00 mg d−1 (Fe), 2.04 mg d−1 (Mn), 126.27 μg d−1 (Ni) and 10.97 mg d−1 (Zn)) were below toxicological reference values, with the exception of Mn in men, and Fe in women, 63% of whom had inadequate Fe intake. Average dietary exposure to Cd and Pb represented 21.7% and 3.2% of the respective provisional tolerable weekly intake (PTWI). The duplicate diet protocol was used to assess the As intake of adults and children in the Jinhu province of China using HG-AFS.173 The median As intakes for the two groups were 3.8 μg kg−1 bw wk−1 (adults) and 6.7 μg kg−1 bw wk−1 (children) neither of which exceeded the World Health Organization (WHO) PTWI of 15 μg kg−1 bw wk−1. Maximum values for adults and children were 118.7% and 125.3% of the PTWI highlighting only a minority of people in this area being over-exposed to As. The sources of Se intake in Switzerland were examined by Jenny-Burri et al.174 The daily intake was estimated at 66 μg kg−1 d−1 and has remained constant for the last 25 years; the main source of Se in Switzerland is pasta made from North American durum wheat and also meat. Heavy metals (As, Cd, Hg and Pb) were surveyed in three types of seafood (162 samples of fish/cephalopods and 30 pools of mussels) available in Campania (Italy).175 All Cd and Hg concentrations were below the EU limits. Two samples of farmed European seabass and two frozen samples exceeded EU limits for Pb which could make a consumer exceed the PTWI for Pb by 16.5% but, more importantly, the Hg intake could exceed 41% of the Hg PTWI. Reported values were (fresh catch, farmed, frozen): As: <10–231 ng g−1; 12–310 ng g−1; 12–272 ng g−1; Cd: <0.8–19.8 ng g−1; <0.8–42.3 ng g−1; <0.8–93.1 ng g−1; Hg: 8–339 ng g−1; <5–226 ng g−1; <5–313 ng g−1; Pb: <2–689 ng g−1; <20–438 ng g−1; <20–541 ng g−1.

The difference between toxicity and essential element status for Cr depends on the species present and knowledge of this speciation is important in diet staples such as bread. Following an ETAAS method optimization based on wet digestion and selective alkaline extraction, total Cr and CrVI were determined in 152 bread samples from Oporto (Portugal) by Soares et al.176 Total Cr concentration was 47.3 ± 20.0 and 50.9 ± 22.2 μg kg−1 dw for white and whole bread, respectively while CrVI concentration was 5.65 ± 5.44 and 6.82 ± 4.88 μg kg−1 dw. The calculated daily intake based on three bread units, was 12.7 and 1.98 μg d−1 respectively, the total Cr contributing up to 10% of the Reference Daily Intake of 120 μg d−1.

7.2.2 Human milk and infant formula. The correlation between the change in anti-oxidant status of human milk during the first 4 months of lactation and certain trace elements (Cu, Mn, Se and Zn) was measured by Matos et al.177 Milk samples were collected from 31 lactating women in Oporto (Portugal) at 1, 4, 8, 12 and 16 weeks post partum. A decrease in total anti-oxidant status and trace element concentrations was observed at 1 week and correlations were seen between total anti-oxidant status and concentrations of Cu, Se and Zn; no correlation between measured parameters and age was evident. A contribution to setting a base level for Ce in human milk was presented by Hollriegl et al.138 who determined Ce in blood plasma and human milk using ICP-MS in 42 subjects from Munich (Germany) and 26 from Madrid (Spain) over 51 weeks post partum. Mean values ranged from 5 ng L−1 to 65 ng L−1 with a median of 13 ng L−1; a value eight times higher was measured in an earlier study conducted in an eastern German province. Interestingly plasma Ce levels were below the LOD of 10 ng L−1 in all but two German subjects but the levels in the Spanish subjects ranged from 21.6 to 70.3 ng L−1 showing environmental conditions such as traffic volume may influence plasma Ce levels.
7.2.3 Milk and dairy products. The influence of natural groundwater As levels on livestock drinking water and the subsequent effect on cow's milk was studied by Sigrist et al.178 Dry-ashed milk and well water samples were measured using optimized FI-HG-AAS. Detection limits were 0.7 μg L−1 and 0.6 μg L−1 for milk and well water, respectively. A low biological As transfer rate was observed from drinking water to raw milk. Determination of dimethylselenide (DMSe) and dimethyldiselenide (DMDSe) in milk and milk by-products was described by Campillo et al.74 using an optimized SPME method followed by GC-MIP-AES. The LODs ranged from 70 to 110 pg mL−1 for DMSe and from 80 to 400 pg mL−1 for DMDSe, depending on the sample. Standard additions were required for quantification. None of the twenty-three samples analysed contained the studied compounds at concentrations above the corresponding detection limits. An ETAAS method for the simultaneous determination of a range of elements in milk samples was proposed by Freschi et al.71 who employed different kinds of Ru coatings as chemical modifiers. A furnace pre-treatment with W and Ru was used for volatile analytes (As, Bi, Pb, Sb and Se) whereas a platform treated with W and RuCl3 was found appropriate for the transition group metals (Co, Cr, Cu, Fe and Mn). Milk samples were diluted 1 + 9 with 1% HNO3; calibration matrix matching in the range 5.0–20.0 μg L−1 yielded a linear response with LODs typically 25–36 μg L−1 for the volatile elements group and 13–53 μg L−1 for the transition metals. Brandao et al.66 developed a sensitive slurry sampling method for determining Zn in yoghurt, using high resolution continuum source FAAS, with an LOD of 0.32 μg g−1. The optimized method employed a 20 min sonication of 1.0 g of sample with 25 mL of 2.0 M HCl. The method was validated against a NIST non-fat milk powder SRM and was also found to give identical results when hot digested with concentrated HNO3 and H2O2. The Zn concentration of seven yoghurts ranged from 2.19 to 4.85 μg g−1.
7.2.4 Cereals, flour and rice. Elemental imaging is now furthering the understanding of spatial distribution in foodstuffs and is proving an exciting tool in the Food Industry. Moore et al.64 determined the subcellular localization of As and Se within cereal grains in order to understand deposition patterns and the impact of processes such as milling. The localization of Se in wheat (Triticum aestivum) and As in rice (Oryza sativa) was determined with HR-SIMS and SRXRF. They found Se to be concentrated in the protein surrounding the starch granules in the starchy endosperm cells and more homogeneously distributed in the aleurone cells but with Se-rich hotspots. Conversely, As was concentrated in the subaleurone endosperm cells in association with protein. HR-SIMS identified the high intensity of As in the SRXRF image to be localized in micron-sized hotspots near the ovular vascular trace and nucellar projection. SRXRF mapping was also used along with fluorescence micro-tomography by Cary et al.179 who revealed marked differences in the pattern of As distribution in the rice grain during a study: AsIII was retained in the ovular vascular trace and DMA was dispersed throughout the external grain parts and into the endosperm. The study, which also investigated the roles of phloem and xylem transport, found that DMA was translocated with a 10-fold efficiency compared with iAs and that the phloem transported 90% of AsIII and 55% of the DMA to the grain.

Although much is known of the major contribution of rice to the human intake of i As, the influence of other rice products oniAs intake is less well known. Signes-Pastor et al.180 analysed Japanese misos, syrups and amazake (a fermented sweet rice drink) produced from rice, barley and millet for total As and, for a subset of samples, for As speciation. Products prepared from barley and millet had lower As compared with those derived from rice, which was mainly iAs (63–83%) with the remainder being DMA. Consumption of these products alone could contribute 23% of the WHO PTWI of iAs. Nishimura et al.181 determined total As and speciation of iAs in several varieties of rice. Total As was determined with ICP-MS using CH3COOH to increase sensitivity and Ge added as internal standard to increase the precision. Partial digestion with HNO3 was carried out at 100 °C for better extraction efficiency from glutinous and coloured rice (>90%). Total As and iAs for polished and unpolished forms of non-glutinous, glutinous and coloured rice were 0.04–0.54 mg kg−1 and 0.02–0.41 mg kg−1, respectively. Rice colour did not influence total or iAs content. These values are much higher than those found by Liang et al.182 in polished rice from China. They analysed samples from various production regions for total and As species using HPLC-ICP-MS. Total As concentration ranged from 65.3 to 274.2 ng g−1 (mean 114.4 ng g−1). Four species, (AsIII, AsV, DMA and MMA) were detected in most rice samples. The predominant species was iAs (mean 82.0 ng g−1) which accounted for approximately 72% of the total. The authors compared the exposure to As through rice (37.6% of WHO maximum tolerable daily amount) to that from drinking water and found it to be 25 times higher, prompting them to comment on the underestimated risks of As exposure to the Chinese population.

The fortification status of Brazilian corn and wheat flour was assessed by Boen and Pallone183 through FAAS analysis of Fe and Zn in macaroni, pizza and bread. Levels of Fe and Zn in macaroni were 16–54 mg kg−1 and 7–10 mg kg−1, respectively, whereas concentrations were 26–73 mg kg−1 and 4–9 mg kg−1, respectively, for pizza, and 30–124 mg kg−1 and 4–8 mg kg−1 for bread. The authors indicated that the high average Fe concentration observed could result in problems with Zn absorption.

7.2.6 Fish and seafood. A novel Se-containing compound found in the blood and other tissues of bluefin tuna (Thunnus orientalis) was reported by Yamashita and Yamashita.156 The compound was identified as 2-selenyl-Nα,Nα,Nα-trimethyl-L-histidine, 3-(2-hydroseleno-1H-imidazol-5-yl)-2-(trimethylammonio)propanoate with a molecular formula of C18H29N6O4Se2 and [M + H]+ of 533.0562. The authors assigned the gross structure as the oxidized dimeric form of a Se analogue of ergothioneine therefore naming this novel Se-containing compound “selenoneine.” Using HPLC-ICP-MS, selenoneine was found to be the predominant form of organic Se and was widely distributed in various tissues of the tuna, with the highest concentration in blood, similar to that found in mackerel. The effects of a copious consumption of seafood containing high levels of As species on urinary As metabolites (iAs MMAV, DMAV), using HG and HPLC-ICP-MS, was comprehensively studied by Choi et al.184 The daily mean intake of total As given to the 16 Korean subjects was 6.98 mg which was distributed as seaweed (67%), flat fish (25%) and conch (8%). Study subjects refrained from eating seafood for 3 days prior to the first urine collection and then ingested seafood daily for 6 consecutive days and first voided urine was collected on days 1, 2, 3, 4, 5, 6, 7, 10 and 14 post ingestion. An increase in total urinary As metabolites was observed from day 1 which was attributed to increased DMA but recovered to control level by day 10. However, no significant changes in biological indexes (aspartate transaminase, glutathione, glutathione peroxidase, lipid peroxidation and uric acid) were observed. The authors recommended evaluating As metabolism when assessing iAs exposure and highlighted potential chronic health effects of seafood consumption in Korea.

With a view to recommending an appropriate methodology for use in the discrimination of fish stocks, six digestion procedures were evaluated by Ashoka et al.34 prior to multi-element (40 analytes) ICP-MS analysis. Although no one method gave optimal conditions in terms of accuracy and precision, microwave-assisted digestion in a standard polypropylene screw-cap tube with HNO3 and H2O2 gave the best compromise. Optimization of an HG-AFS method for total As in microwave-digested seaweed was discussed by Garcia-Sartal et al.185 Seaweed samples (red seaweed (nori and dulse), brown seaweed (wakame, sea spaghetti and kombu), and green seaweed) were digested by microwave heating using a mixture of HNO3 and H2O2. Parameters relating to the HG were 1.25% w/v NaBH4, and 0.75 M and 0.1 M for HCl used as carrier and sample dilution solutions, respectively, giving an LOD of 2.3 μg g−1 dw and precision <10% RSD. Optimization of sample digestion, involving a Plackett-Burman experimental design, for subsequent determination of total Hg using CV-AAS was detailed by Shah et al.37 This validated method was sensitive and sported an LOD of 0.133 μg kg−1. Another, albeit less sensitive (LOD 3.7 ng g−1), method based on FI-CV-AFS49 was validated for total Hg and Hgi. An ICP-MS method for quantification of total As in seafood was characterized.186 Using this method, they found total As in bluefin tuna, yellowfin tuna, bigeye tuna and swordfish to range from 0.74 to 6.87 mg kg−1. Speciation showed that the major species was AB.

7.2.7 Meat and poultry. The bioaccumulation of heavy metals into the human food chain can occur during the raising of livestock. Rudy187 determined the accumulation of four heavy metals (As, Cd, Hg and Pb) in tissues (longissimus muscle and hepatic tail lobe) from wild boar. Each sample type was divided into three age groups. A statistical difference was only found between the youngest and oldest animal and the maximum permissible level was not exceeded in muscle for Cd, Hg and Pb; As was not detected above 0.001 mg kg−1. In the >3 year group, the maximum permissible level of Cd (0.5 mg kg−1) was exceeded in two liver samples. Conversely, domestic cattle and fowl which grazed on contaminated soil around artisanal gold mining villages in Tanzania showed significantly higher (p = 0.05) Hg levels than those in the reference area.188 Maximum total hepatic Hg levels of 436 and 820 μg kg−1 ww were found in cattle and domestic fowl, respectively. The interlobular distribution of Cu in beef liver was assessed by biopsy and post mortem sampling by Miranda et al.32 on 29 ten-month-old beef calves fed growing and finishing diets supplemented with 35 mg kg−1 CuSO4. Needle biopsies were shown to give accurate estimates of overall hepatic Cu status. The highest Cu concentrations were found in the left lobe, followed by the processus papillaris while the lowest Cu concentrations were found in the caudate and quadrate lobes. The different breeds studied (Galician Blonds (n = 10), Holstein Friesians (n = 9) and Galician Blond–Holstein Friesian crosses n = 10)) showed different absolute Cu levels but not interlobular distribution.
7.2.8 Drinking water. Concerns for elevated levels of heavy metals in drinking water in the developing world continue to necessitate careful analysis and interpretation. Focusing on Kurdistan (Iran), the concentrations of As, Cd and Se (among a total of eight elements measured by ETAAS) in 28 village drinking water sources exceeded WHO or National Standard limits. The of concentration of As in drinking water ranged from 42 to 1500 μg L−1.189 They then surveyed a cross-section of 587 participants from 211 households for the occurrence of multi-symptomatic As poisoning and found 30.7% were affected with Mee's line, keratosis and pigment disorders accounting for 86.1%, 77.2% and 67.8% respectively of sufferers. The results show a strong linear relationship between As exposure and occurrence of multi-symptomatic arsenical poisoning (R2 = 0.76). Part of the European “Arsenic Health Risk Assessment and Molecular Epidemiology” programme assessed the lifetime exposure to As in several regions of Central Europe using data on As concentrations in water, measured by HG-AAS.190 The acquired data accounted for 80% of the recorded lifetime residential As exposure. The mean (and median) lifetime concentrations were 14.7 (13.3) μg L−1 for Hungary, 3.8 (0.7) μg L−1 for Romania and 1.9 (0.8) μg L−1 for Slovakia. Overall, 25% of the study population had average concentrations >10 μg L−1 and 8% with exposure >50 μg L−1. A multi-element (31 elements) ICP-MS screening of drinking water collected in six villages of the Argentinean Andes, and in urine of the female inhabitants, was undertaken by Concha et al.123 High concentrations of As, B, Cs and Li were found in both sample types. Maximum drinking water values were: 210 μg L−1 (As), 5950 μg L−1 (B), 320 μg L−1 (Cs) and 1000 μg L−1 (Li); and the range of village median urine concentrations: 26–266 μg L−1 (As), 2980–16[thin space (1/6-em)]560 μg L−1 (B), 34–531 μg L−1 (Cs) and 340–4550 μg L−1 (Li). Another multi-element survey of drinking water around the Narmada River of Madhya Pradesh (central India) measured concentrations of As, Be, Ca, Cd, Co, Cr, Cu, Fe, Li, Mg, Mn, Mo, Ni, Pb, Sb, Se, Sr, Ta, Ti, V and Zn using AAS in addition to measurements of temperature, pH, electrical conductivity and total faecal coliformes.191 Their overall results confirmed that this stretch of the Narmada River is contaminated with heavy metals and other contaminants that might affect human health as well as the health of the ecosystem.

The low levels of analytes in drinking water often pose an analytical problem, especially when only simple equipment is available, and creative solutions are by necessity still being developed. Due the high excitation energy of Cl, sensitivity is limited in ICP-AES and, because of low nebulization efficiency, only 5% of the analyte contributes to the emission intensity. Hashimoto et al.72 proposed oxidation vaporization, based on co-nebulization with KMnO4, to give a 20-times improvement in sensitivity compared with a standard nebulizer. Results were almost identical to those obtained by IC, with analysis time less than 1 min. A simple sub μg L−1 detection of Cr species in drinking water was achieved directly, without any pre-treatment, to preserve the original Cr speciation, by Xing and Beauchemin192 using IEC-ICP-MS. Separation of CrIII and CrVI was possible on an IonPac AG-7 guard column with gradient elution using 0.1 M NH4NO3 and 0.8 M HNO3. Hydrogen collision/reaction gas eliminated Cl-based and C-based polyatomic interferences. Obtained LODs were 0.04 μg L−1 and 0.02 μg L−1 for CrIII and CrVI, respectively. The same workers,41 using FAAS as detector, employed coprecipitation with Zr(OH)4 to determine Mn species. Total Mn was determined after the reduction of MnVII to MnII by ascorbic acid. Figures of merit for MnII were: precision <7%, 95–98% recovery, 50-fold preconcentration factor and LOD 0.75 μg L−1.

Studies of the leachate levels from packaging have not been reviewed here for some years but a comprehensive study of the packaging material for bottled water sold across the European Union was reported by Reimann et al.193 They used ICP-MS to determine 57 elements in 294 pairs of samples packaged both in glass and PET bottles. The results demonstrated significant differences in median concentrations of 26 elements. The most common catalyst for PET production contributes to a Sb concentration that is 21 times higher than the median value in water sold in glass bottles (0.33 vs. 0.016 μg L−1). Several other elements (listed in order of enrichment of 19-fold down to 1.4-fold): Ce, Pb, Al, Zr, Ti, Th, La, Pr, Fe, Zn, Nd, Sn, Cr, Tb, Er, Gd, Bi, Sm, Y, Lu, Yb, Tm, Nb and Cu, were at higher concentration when packaged in glass compared to PET. Results from an additional 136 bottles of the same water sold in green and clear glass bottles demonstrated the influence of green glass pigment contributing to significantly higher concentrations of Cr (7.3), Th (1.9), Ce, La, Nd, Zr (1.6), Co, Fe, Nb, Pr, Ti (1.3), and Er (1.1); the numbers in brackets indicate the enhancement factors. A further demonstration of the effect of bottle pigmentation over time (1–150 days) was carried out on 126 bottles of three different materials (glass, hard and soft PET) with colours (clear, light and dark green and blue, brown) at pH 6.5 and pH 3.5. For most elements, leaching is enhanced at pH 3.5 and, independent of bottle material, dark coloured bottles leach more than clear bottles. It is perhaps worth pointing out that this study was published in Applied Geochemistry.

7.2.9 Alcoholic beverages. As a crucial step in understanding oxidative spoilage in lager beers and red wines, three Cu species were isolated using a two-column solid phase extraction.194 Species were separated using Amberlite XAD-16 and Dowex 50W × 8–200 resins which yielded hydrophobic, cationic and residual-species fractions. Total Cu of each fraction was determined by FAAS. The polar and non-cationic fractions were most abundant in beer (72–82% of total Cu) and, in red wine, the hydrophobic species were most prevalent (27–77%); the authors presumed these contained relatively strong Cu complexes with the polyphenols. Old vines are considered by many to enhance the complexity of wine and this led Pessanha et al.195 to assess the influence of the vineyard age on the elemental content at several stages of vinification. The elemental content of vine leaves and grapes was determined by EDXRF, while TXRF was used for analysis of must and wine, sourced from two vineyards (6 and 14 years) in the Douro region (Portugal). The wine Br content from the 6 year old vineyard was an order of magnitude higher than normally expected while the vine leaves of the older had much higher Cu largely due to the use of Cu-containing fungicides. Given the low analyte concentration and the risk of interferences from wine matrix components, ETV-ICP-MS was explored for the determination of several heavy metals (As, Pb and Se) in comparison with standard nebulizer-ICP-MS or ETAAS.57 The obvious influence of matrix components on the analyte transport efficiency required the use of standard additions. Concentrations for Pb and Se were similar across techniques but reported concentrations for As were 2–4 times lower, presumably because of over-correction of the polyatomic interferences (i.e. 40Ar35Cl+vs. 75As+) obtained with ETV sample introduction. Another approach for these analyses would be to use a handheld X-ray tube analyser, and this was evaluated by Anderson91,92 for analysis of beverages using an Innov-X alpha-6000s XRF tube-type analyser. Average correction factors applied to analyser output yielded accurate (in terms of z-scores) quantitative results for As above 20 mg kg−1 and weighted quadratic fit calibrations provided accurate quantitative or semi-quantitative results for the other elements. Additionally, spiked beverages analysed through the wall of the original polyethyleneterephthalate container gave accurate results, within measurement uncertainties, after application of “container wall” correction factors. Regrettably the LOD was higher than one would want for measuring heavy metals in drinks!
7.2.10 Authenticity. Olive oil authenticity continues to attract the application of stable isotope mass spectrometry and multi-elemental analysis despite its inherent cost. To establish a national databank, Camin et al.196 used the data from C, H, O isotopic ratios (13C:12C, H2:H1, 18O:16O) in oil and extracted glycerol, and the mineral composition of authentic Italian extra-virgin olive oils (Protected Designation of Origin (PDO) and Protected Geographical Indication quality levels), collected from 2000 to 2005 (n = 539). The concentrations of 26 elements, including Ce, Cs, La, Li, Rb and Yb, which are rarely reported in the literature, were determined in well settled oils after ultrasound acid extraction. The ratios 13C:12C and 18O:16O increased from northern to southern Italy (Sicily), each year allowing their differentiation. Significant differences were found among the years and in some cases also between PDOs from the same region. The same group197 was involved in a further study which related the same parameters in 267 olive oils and 314 surface waters collected from eight European sites with climatic (mainly temperature) and geographical (mainly latitude and distance from the coast) data, as had been done for wine several decades ago. It was possible to characterize the geological origin of the olive oils by using the content of 14 elements (Ca, Ce, Cs, Eu, K, La, Mg, Mn, Rb, Sm, Sr, U V, Zn) and, by combining the 3 isotopic ratios with the 14 elements and applying a multivariate discriminant analysis. A good discrimination between olive oils from eight European sites was achieved, with 95% of the samples correctly classified into the production site. Gonzalvez et al.18 reviewed the literature on authentication of foods with PDOs from their trace element composition or stable isotope ratios, in order to evaluate the state of the art of this field. They identified the main parameters of analytical techniques and chemometric data treatments that permit accurate discrimination of samples of different geographical origins and with different characteristics. One of the most interesting applications of analytical atomic spectrometry for food authenticity is the use of stable Sr isotope ratios and Rosner198 highlighted the pitfalls some researchers can make if adequately accurate 87Sr:86Sr ratios are not achieved. The author stated that appropriate wet chemical separation techniques and mass spectrometric instrumentation enables the generation of 87Sr:86Sr data with very low uncertainty and permits the identification of even small differing Sr contributions from source reservoirs or physical and anthropogenic processes.

8 Abbreviations used in this Update

2-DTwo-dimentional
3-DThree-dimentional
AAAtomic absorption
AASAtomic absorption spectrometry
ABArsenobetaine
ACArsenocholine
AESAtomic emission spectrometry
AFAtomic fluorescence
AFSAtomic fluorescence spectrometry
AMSAccelerator mass spectrometry
ASUAtomic Spectrometry Update
ASVAnodic stripping voltametry
BCRCommunity Bureau of Reference
bwbody weight
CECapillary electrophoresis
CIConfidence interval
CRMCertified reference material
CSFCerebrospinal fluid
CVCold vapour
CV-AASCold vapour AAS
CV-AFSCold vapour AFS
CVGChemical vapour generation
DaDalton
DMADimethylarsinic acid
DMSeDimethylselenide
DMDSedimethyldiselenide
DNAdeoxyribose nucleic acid
DRCDynamic reaction cell
dwdry weight
EDLElectrodeless discharge lamp
EDTAEthylenediaminetetraacetic acid
EDXRFEnergy dispersive XRF
ESIElectrospray ionisation
ESI-MSElectrospray MS
ETElectrothermal
ETAElectrothermal atomization
EtHgEthylmercury
ETVElectrothermal vaporization
ETAASElectrothermal atomization AAS
EXAFSExtended X-ray absorption fine structure
FAASFlame AAS
FFFlame furnace
FIFlow injection
GCGas chromatography
GC-MSGas chromatography-MS
GSHGlutathione
iAsInorganic arsenic
iHgInorganic mercury
HGHydride generation
HG-AASHydride generation AAS
HG-AFSHydride generation AFS
hMThuman metallothionein
HPLCHigh performance liquid chromatography
HRHigh resolution
IAEAInternational Atomic Energy Authority
ICPInductively coupled plasma
ICP-MSInductively coupled plasma-mass spectrometry
IDIsotope dilution
IECIon-exchange chromatography
JAASJournal of Analytical Atomic Spectrometry
LALaser ablation
LCLiquid chromatography
LIBSLaser induced breakdown spectroscopy
LODLimit of detection
LOQLimit of quantitation
MALDIMatrix assisted laser desorption ionization
MCMulti-collector
MeHgMethylmercury
MeOHMethanol
MeSeCysMethylselenocysteine
MIPMicrowave induced plasma
MMAMonomethylarsonic acid
M r Relative molecular mass
MRIMagnetic resonance imaging
MSMass spectrometry
Nd:YAGNeodinium:ytrrium aluminium garnet
NISTNational Institute of Standards and Technology
PAGEPolyacrylamine gel electrophoresis
PDOProtected Designation of Origin
PEGPolyethyleneglycerol
PETPolyethyleneterephthalate
PIXEParticle-induced X-ray emission
PTFEPoly(tetrafluoroethylene)
PTWIProvisional tolerable weekly intake
Q-ICP-MSQuadrupole-ICP-MS
RSDRelative standard deviation
RPReversed phase
SDSSodium dodecylsulfate
SECSize exclusion chromatography
SeCysSelenocysteine
SeCys2Selenocystine
SeMetSelenomethionine
SeMetSeCysSe-methylselenocysteine
SF-MSSector field MS
SIMSSecondary ion MS
SlSlurry
SODSuperoxide dismutase
SPESolid phase extraction
SPMESolid phase micro extraction
SRMStandard reference material
SRXRFSynchrotron radiation XRF
TMAHTetramethylammonium hydroxide
TMAOTrimethylarsine oxide
TMSe+Trimethylselonium ion
TOFTime-of-flight
TXRFTotal reflection XRF
UVUltra violet
VGVapour generation
vs. versus
XANESX-ray absorption near edge structure
XRFX-ray fluorescence
WDXRFWavelength dispersive XRF
WHOWorld Health Organization
wwwet weight

References

  1. A. Taylor, S. Branch, M. P. Day, M. Patriarca and M. White, Journal of Analytical Atomic Spectrometry, 25(4), pp. 453–492 Search PubMed.
  2. C. F. Harrington, R. Clough, H. R. Hansen, S. J. Hill and J. F. Tyson, Journal of Analytical Atomic Spectrometry, 25(8), pp. 1185–1216 Search PubMed.
  3. E. H. Evans, J. A. Day, C. Palmer and C. M. M. Smith, Journal of Analytical Atomic Spectrometry, 25(6), pp. 760–784 Search PubMed.
  4. S. Carter, A. S. Fisher, P. S. Goodall, M. W. Hinds, S. Lancaster and S. Shore, J. Anal. At. Spectrom., 2009, 24(12), 1599–1656 RSC.
  5. M. West, A. T. Ellis, P. J. Potts, C. Streli, C. Vanhoof, D. Wegrzynek and P. Wobrauschek, J. Anal. At. Spectrom., 2009, 24(10), 1289–1326 RSC.
  6. O. T. Butler, W. R. L. Cairns, J. M. Cook and C. M. Davidson, J. Anal. At. Spectrom., 2010, 25(2), 103–141 RSC.
  7. O. Butler, H. Evans, A. Fisher, S. Hill, C. Harrington, A. Taylor, M. West and A. Ellis, J. Anal. At. Spectrom., 2010, 25(10), 1546–1566 RSC.
  8. R. G. Ge and H. Z. Sun, Sci. China, Ser. B: Chem., 2009, 52(12), 2055–2070 CrossRef CAS.
  9. S. Mounicou, J. Szpunar and R. Lobinski, European Journal of Mass Spectrometry, 16(3), pp. 243–253 Search PubMed.
  10. D. Sanchez-Rodas, W. T. Corns, B. Chen and P. B. Stockwell, Journal of Analytical Atomic Spectrometry, 25(7), pp. 933–946 Search PubMed.
  11. Y. Ogra and Y. Anan, J. Anal. At. Spectrom., 2009, 24(11), 1477–1488 RSC.
  12. Y. Ogra, Anal. Sci., 2009, 25(10), 1189–1195 CrossRef CAS.
  13. B. Michalke, Journal of Trace Elements in Medicine and Biology, 24(2), pp. 69–77 Search PubMed.
  14. I. Nicolis, E. Curis, P. Deschamps and S. Benazeth, presented at 4th International Conference on Metals and Genetics, Paris, FRANCE, Jul, 2008 Search PubMed.
  15. S. J. Fairweather-Tait, R. Collings and R. Hurst, American Journal of Clinical Nutrition, 91(5), pp. 1484S–1491S Search PubMed.
  16. J. Sardans, F. Montes and J. Penuelas, Spectrochimica Acta Part B-Atomic Spectroscopy, 65(2), pp. 97–112 Search PubMed.
  17. A. Rosin, Israel Medical Association Journal, 2009, 11(11), 689–694 Search PubMed.
  18. A. Gonzalvez, S. Armenta and M. de la Guardia, TrAC, Trends Anal. Chem., 2009, 28(11), 1295–1311 CrossRef.
  19. J. S. Becker, International Journal of Mass Spectrometry, 289(2-3), pp. 65–75 Search PubMed.
  20. J. S. Becker, A. Matusch, C. Palm, D. Salber, K. A. Morton and S. Becker, Metallomics, 2(2), pp. 104–111 Search PubMed.
  21. J. S. Becker, M. Zoriy, A. Matusch, B. Wu, D. Salber and C. Palm, Mass Spectrometry Reviews, 29(1), pp. 156–175 Search PubMed.
  22. I. Rodushkin, E. Engstrom and D. C. Baxter, Analytical and Bioanalytical Chemistry, 396(1), pp. 365–377 Search PubMed.
  23. B. Bocca, D. Mattei, A. Pino and A. Alimonti, Rapid Communications in Mass Spectrometry, 24(16), pp. 2363–2369 Search PubMed.
  24. W. C. Davis, R. Zeisler, J. R. Sieber and L. L. Yu, Analytical and Bioanalytical Chemistry, 396(8), pp. 3041–3050 Search PubMed.
  25. P. C. Kruger, C. M. Geraghty and P. J. Parsons, Accreditation and Quality Assurance, 15(8), pp. 451–458 Search PubMed.
  26. P. Jitaru, M. Roman, C. Barbante, S. Vaslin-Reimann and P. Fisicaro, Accreditation and Quality Assurance, 15(6), pp. 343–350 Search PubMed.
  27. B. L. Batista, J. L. Rodrigues, L. Tormen, A. J. Curtius and F. Barbosa, J. Braz. Chem. Soc., 2009, 20(8), 1406–1413 CAS.
  28. Y. J. Kim, O. Galindev, J. H. Sei, S. M. Bae, H. Im, L. Y. Wen, Y. R. Seo and W. S. Ahn, Biol. Trace Elem. Res., 2009, 131(2), 103–109 CrossRef CAS.
  29. B. Michalke, P. Grill and A. Berthele, J. Trace Elem. Med. Biol., 2009, 23(4), 243–250 CrossRef CAS.
  30. V. Nischwitz, A. Berthele and B. Michalke, Journal of Analytical Atomic Spectrometry, 25(7), pp. 1130–1137 Search PubMed.
  31. M. Kilinc, A. Coskun, F. Bilge, S. S. Imrek and Y. Atli, Journal of Trace Elements in Medicine and Biology, 24(3), pp. 152–156 Search PubMed.
  32. M. Miranda, J. L. Benedito, B. Gutierrez, M. Garcia-Vaquero, I. Blanco-Penedo and M. Lopez-Alonso, Journal of Veterinary Diagnostic Investigation, 22(2), pp. 277–281 Search PubMed.
  33. A. N. Johnston, S. A. Center, S. P. McDonough and K. L. Warner, Am. J. Vet. Res., 2009, 70(12), 1502–1511 CrossRef CAS.
  34. S. Ashoka, B. M. Peake, G. Bremner, K. J. Hageman and M. R. Reid, Anal. Chim. Acta, 2009, 653(2), 191–199 CrossRef CAS.
  35. T. Narukawa and S. Willie, Journal of Analytical Atomic Spectrometry, 25(7), pp. 1145–1148 Search PubMed.
  36. C. Nomura, H. Obana and H. Oda, J. Food Hyg. Soc. Jpn., 2009, 50(5), 253–255 Search PubMed.
  37. A. Q. Shah, T. G. Kazi, M. B. Arain, J. A. Baig, H. I. Afridi, M. K. Jamali, N. Jalbani and G. A. Kandhro, Journal of Aoac International, 2009, 92(5), 1580–1586 CAS.
  38. F. D. Depoi, F. R. S. Bentlin and D. Pozebon, Analytical Methods, 2(2), pp. 180–185 Search PubMed.
  39. D. Citak and M. Tuzen, Food and Chemical Toxicology, 45(5), pp. 1399–1404 Search PubMed.
  40. S. L. Maxwell and V. D. Jones, Talanta, 2009, 80(1), 143–150 CrossRef CAS.
  41. D. Citak, M. Tuzen and M. Soylak, Journal of Hazardous Materials, 173(1–3), pp. 773–777 Search PubMed.
  42. Y. Tateishi, T. Hashimoto, K. Ushiyama, N. Sakai, I. Baba and T. Nagayama, Journal of the Food Hygienic Society of Japan, 51(2), pp. 53–57 Search PubMed.
  43. B. Kollander, M. Andersson, J. Pettersson, Talanta, 80, 5, pp. 2068–2075 Search PubMed.
  44. G. Vale, A. Rodrigues, A. Rocha, R. Rial, A. M. Mota, M. L. Goncalves, L. P. Fonseca and J. L. Capelo, Food Chemistry, 121(1), pp. 268–274 Search PubMed.
  45. Y. Fang, Y. F. Zhang, B. Catron, Q. L. Chan, Q. H. Hu and J. A. Caruso, J. Anal. At. Spectrom., 2009, 24(12), 1657–1664 RSC.
  46. J. L. Rodrigues, D. P. Torres, V. C. D. Souza, B. L. Batista, S. S. de Souza, A. J. Curtius and F. Barbosa, J. Anal. At. Spectrom., 2009, 24(10), 1414–1420 RSC.
  47. D. P. Torres, V. L. A. Frescura and A. J. Curtius, Microchem. J., 2009, 93(2), 206–210 CrossRef.
  48. H. Matusiewicz and E. Stanisz, Central European Journal of Chemistry, 8(3), pp. 594–601 Search PubMed.
  49. G. G. A. de Carvalho, M. A. Feres, J. R. Ferreira and V. H. Kennedy, International Journal of Environmental Analytical Chemistry, 90(9), pp. 686–696 Search PubMed.
  50. P. C. Kruger, L. M. Schell, A. D. Stark and P. J. Parsons, Metallomics, 2(9), pp. 621–627 Search PubMed.
  51. J. P. Lafleur and E. D. Salin, J. Anal. At. Spectrom., 2009, 24(11), 1511–1516 RSC.
  52. S. G. Ozcan, N. Satiroglu and M. Soylak, Food and Chemical Toxicology, 48(8-9), pp. 2401–2406 Search PubMed.
  53. R. A. Gil, P. H. Pacheco, P. Smichowski, R. A. Olsina and L. D. Martinez, Microchim. Acta, 2009, 167(3-#x2013;4), 187–193 CrossRef CAS.
  54. Q. Tu, T. B. Wang and C. J. Welch, Journal of Pharmaceutical and Biomedical Analysis, 51(1), pp. 90–95 Search PubMed.
  55. G. Koellensperger and S. Hann, Analytical and Bioanalytical Chemistry, 397(1), pp. 401–406 Search PubMed.
  56. P. C. Kruger, L. M. Schell, A. D. Stark and P. J. Parsons, Journal of Analytical Atomic Spectrometry, 25(8), pp. 1298–1307 Search PubMed.
  57. G. Grindlay, J. Mora, L. Gras and M. T. C. de Loos-Vollebregt, Anal. Chim. Acta, 2009, 652(1-#x2013;2), 154–160 CrossRef CAS.
  58. S. El Balkhi, J. Poupon, J. M. Trocello, F. Massicot, F. Woimant and O. Laprevote, Analytical Chemistry, 82(16), pp. 6904–6910 Search PubMed.
  59. M. W. Hsieh, C. L. Liu, J. H. Chen and S. J. Jiang, Electrophoresis, 31(13), pp. 2272–2278 Search PubMed.
  60. S. R. Copeland, M. Sponheimer, J. A. Lee-Thorp, P. J. le Roux, D. J. de Ruiter and M. P. Richards, Journal of Archaeological Science, 37(7), pp. 1437–1446 Search PubMed.
  61. M. Borges-Alvarez, F. Benavente, J. Barbosa and V. Sanz-Nebot, Rapid Communications in Mass Spectrometry, 24(10), pp. 1411–1418 Search PubMed.
  62. H. Hernandez-Mendoza, E. Chamizo, A. Yllera, M. Garcia-Leon and A. Delgado, Nuclear Instruments & Methods in Physics Research Section B-Beam Interactions with Materials and Atoms, 268(7-8), pp. 1331–1333 Search PubMed.
  63. S. Yumoto, H. Nagai, S. Kakimi and H. Matsuzaki, Nuclear Instruments & Methods in Physics Research Section B-Beam Interactions with Materials and Atoms, 268(7-8), pp. 1328–1330 Search PubMed.
  64. K. L. Moore, M. Schroder, E. Lombi, F. J. Zhao, S. P. McGrath, M. J. Hawkesford, P. R. Shewry and C. R. M. Grovenor, New Phytologist, 185(2), pp. 434–445 Search PubMed.
  65. L. Hoehne, F. R. Bartz, C. A. Bizzi, J. N. G. Paniz, V. L. Dressler and E. M. D. Flores, Journal of the Brazilian Chemical Society, 21(6), pp. 978–984 Search PubMed.
  66. G. C. Brandao, R. M. de Jesus, E. G. P. da Silva and S. L. C. Ferreira, Talanta, 81(4-5), pp. 1357–1359 Search PubMed.
  67. M. M. A. Campos, H. Tonuci, S. M. Silva, B. D. Altoe, D. de Carvalho, E. A. M. Kronka, A. M. S. Pereira, B. W. Bertoni, S. D. Franca and C. E. S. Miranda, Phytochem. Anal., 2009, 20(6), 445–449 CAS.
  68. M. Resano, J. Briceno and M. A. Belarra, J. Anal. At. Spectrom., 2009, 24(10), 1343–1354 RSC.
  69. C. S. Nomura and P. V. Oliveira, Analytical Methods, 2(1), pp. 49–53 Search PubMed.
  70. A. Baysal and S. Akman, Spectrochimica Acta Part B-Atomic Spectroscopy, 65(4), pp. 340–344 Search PubMed.
  71. G. P. G. Freschi, C. D. Freschi and J. A. G. Neto, Atomic Spectroscopy, 2009, 30(5), 147–155 CAS.
  72. B. Hashimoto, H. Daidoji and H. Uchihara, Bunseki Kagaku, 59(7), pp. 627–631 Search PubMed.
  73. N. P. Zaksas, V. A. Gerasimov, G. A. Nevinsky, Talanta, 80, 5, pp. 2187–2190 Search PubMed.
  74. N. Campillo, R. Penalver, M. Hernandez-Cordoba, Talanta, 80, 5, pp. 1856–1861 Search PubMed.
  75. G. G. de Carvalho, L. C. Nunes, P. F. de Souza, F. J. Krug, T. C. Alegre and D. Santos, Journal of Analytical Atomic Spectrometry, 25(6), pp. 803–809 Search PubMed.
  76. C. L. Kahakachchi and D. A. Moore, J. Anal. At. Spectrom., 2009, 24(10), 1389–1396 RSC.
  77. M. Bogunia, J. Kwapulinskii, E. Bogunia, B. Brodziak-Dopierala, A. Paukszto, J. Kowol, E. Nogaj, B. Ahnert and R. Rochel, Fresenius Environmental Bulletin, 2009, 18(10A), 1937–1941 CAS.
  78. A. Massadeh, A. Gharibeh, K. Omari, I. Al-Momani, A. Alomari, H. Tumah and W. Hayajneh, Biological Trace Element Research, 133(1), pp. 1–11 Search PubMed.
  79. T. Frentiu, M. Ponta, M. Senila, A. I. Mihaltan, E. Darvasi, M. Frentiu and E. Cordos, Journal of Analytical Atomic Spectrometry, 25(5), pp. 739–742 Search PubMed.
  80. Y. G. Yin, Z. H. Wang, J. F. Peng, J. F. Liu, B. He and G. B. Jiang, J. Anal. At. Spectrom., 2009, 24(11), 1575–1578 RSC.
  81. K. J. Chen, I. H. Hsu and Y. C. Sun, J. Chromatogr., A, 2009, 1216(51), 8933–8938 CrossRef CAS.
  82. J. V. Cizdziel, C. Tolbert and G. Brown, Spectrochimica Acta Part B-Atomic Spectroscopy, 65(2), pp. 176–180 Search PubMed.
  83. Q. Y. Liu, Spectrochimica Acta Part B-Atomic Spectroscopy, 65(7), pp. 587–590 Search PubMed.
  84. W. B. Zhang, Z. F. Su, X. F. Chu, X. A. Yang, Talanta, 80, 5, pp. 2106–2112 Search PubMed.
  85. X. J. Jiang, W. E. Gan, L. Z. Wan, H. C. Zhang and Y. Z. He, Spectrochimica Acta Part B-Atomic Spectroscopy, 65(2), pp. 171–175 Search PubMed.
  86. W. B. Zhang, X. A. Yang and X. F. Chu, Microchem. J., 2009, 93(2), 180–187 CrossRef CAS.
  87. C. B. Zheng, Q. Ma, L. Wu, X. D. Hou and R. E. Sturgeon, Microchemical Journal, 95(1), pp. 32–37 Search PubMed.
  88. K. Ito, C. D. Palmer, W. T. Corns and P. J. Parsons, Journal of Analytical Atomic Spectrometry, 25(6), pp. 822–830 Search PubMed.
  89. M. R. Gherase, M. E. Vallee and D. E. B. Fleming, Applied Radiation and Isotopes, 68(4-5), pp. 743–745 Search PubMed.
  90. M. R. Gherase and D. Fleming, X-Ray Spectrom., 2009, 38(6), 513–518 CrossRef CAS.
  91. D. L. Anderson, J. Radioanal. Nucl. Chem., 2009, 282(2), 415–418 CrossRef CAS.
  92. D. L. Anderson, Journal of Aoac International, 93(2), pp. 683–693 Search PubMed.
  93. Y. Ozdemir, B. Borekci, A. Levet and M. Kurudirek, Appl. Radiat. Isot., 2009, 67(10), 1790–1795 CrossRef.
  94. R. Durak, Y. Gulen, M. Kurudirek, M. Kacal and I. Capoglu, Journal of X-Ray Science and Technology, 18(2), pp. 111–120 Search PubMed.
  95. T. Magalhaes, M. L. Carvalho, A. Von Bohlen and M. Becker, Spectrochimica Acta Part B-Atomic Spectroscopy, 65(6), pp. 493–498 Search PubMed.
  96. D. C. Pearce, K. Dowling, A. R. Gerson, M. R. Sim, S. R. Sutton, M. Newville, R. Russell and G. McOrist, Science of the Total Environment, 408(12), pp. 2590–2599 Search PubMed.
  97. L. Borgese, A. Zacco, E. Bontempi, M. Pellegatta, L. Vigna, L. Patrini, L. Riboldi, F. M. Rubino and L. E. Depero, Journal of Pharmaceutical and Biomedical Analysis, 52(5), pp. 787–790 Search PubMed.
  98. M. S. Jimenez, L. Rodriguez, M. T. Gomez and J. R. Castillo, Talanta, 81(1-2), pp. 241–247 Search PubMed.
  99. M. C. Hennion, P. Gareil, A. Hagege, A. Kuhn and V. Pichon, Actualite Chimique, (338–39), pp. 55–63 Search PubMed.
  100. L. Banci, I. Bertini, S. Ciofi-Baffoni, T. Kozyreva, K. Zovo and P. Palumaa, Nature, 465(7298), pp. 645–U145 Search PubMed.
  101. R. N. Easter, Q. L. Chan, B. Lai, E. L. Ritman, J. A. Caruso and Z. Y. Qin, Vascular Medicine, 15(1), pp. 61–69 Search PubMed.
  102. C. J. Chung, Y. M. Hsueh, C. H. Bai, Y. K. Huang, Y. L. Huang, M. H. Yang and C. J. Chen, Cancer, Causes Control, 2009, 20(9), 1653–1661 CrossRef.
  103. Y. Ohta, Y. Kobayashi, S. Konishi and S. Hirano, Chem. Res. Toxicol., 2009, 22(11), 1795–1801 CrossRef CAS.
  104. P. Heffeter, K. Bock, B. Atil, M. A. R. Hoda, W. Korner, C. Bartel, U. Jungwirth, B. K. Keppler, M. Micksche, W. Berger and G. Koellensperger, Journal of Biological Inorganic Chemistry, 15(5), pp. 737–748 Search PubMed.
  105. B. B. Chen, Y. Zeng and B. Hu, Talanta, 81(1-2), pp. 180–186 Search PubMed.
  106. L. Yehiayan, M. Pattabiraman, K. Kavallieratos, X. T. Wang, L. H. Boise and Y. Cai, J. Anal. At. Spectrom., 2009, 24(10), 1397–1405 RSC.
  107. Y. Gao, X. Cao, J. J. Yu, F. S. C. Lee and X. R. Wang, Chin. J. Anal. Chem., 2009, 37(12), 1738–1742 CAS.
  108. C. L. Kahakachchi and D. A. Moore, Metallomics, 2(7), pp. 490–497 Search PubMed.
  109. A. Susulini, H. Kratzin, O. Jahn, C. E. M. Banzato, M. A. Z. Arruda and J. S. Becker, Analytical Chemistry, 82(13), pp. 5859–5864 Search PubMed.
  110. D. J. Bellis and R. Santamaria-Fernandez, Journal of Analytical Atomic Spectrometry, 25(7), pp. 957–963 Search PubMed.
  111. C. Austin, D. Hare, T. Rawling, A. M. McDonagh and P. Doble, Journal of Analytical Atomic Spectrometry, 25(5), pp. 722–725 Search PubMed.
  112. N. Kamaly, J. A. Pugh, T. L. Kalber, J. Bunch, A. D. Miller, C. W. McLeod and J. D. Bell, Molecular Imaging and Biology, 12(4), pp. 361–366 Search PubMed.
  113. B. Arriaza, D. Amarasiriwardena, L. Cornejo, V. Standen, S. Byrne, L. Bartkus and B. Bandak, Journal of Archaeological Science, 37(6), pp. 1274–1278 Search PubMed.
  114. D. J. Bellis, P. J. Parsons, J. Jones and D. Amarasiriwardena, Spectrosc. Lett., 2009, 42(8), 491–496 CrossRef CAS.
  115. L. Finney, Y. Chishti, T. Khare, C. Giometti, A. Levina, P. A. Lay and S. Vogt, Acs Chemical Biology, 5(6), pp. 577–587 Search PubMed.
  116. M. Katsikini, F. Pinakidou, E. Mavromati, E. C. Paloura, D. Gioulekas and D. Grolimund, Nuclear Instruments & Methods in Physics Research Section B-Beam Interactions with Materials and Atoms, 268(3-4), pp. 420–424 Search PubMed.
  117. H. Kinoshita, Y. Hori, T. Fukumoto, T. Ohigashi, K. Shinohara, Y. Hayashi and Y. Ku, Medical Molecular Morphology, 43(1), pp. 19–25 Search PubMed.
  118. A. Carmona, G. Deves, S. Roudeau, P. Cloetens, S. Bohic and R. Ortega, Acs Chemical Neuroscience, 1(3), pp. 194–203 Search PubMed.
  119. N. Barapatre, M. Morawski, T. Butz and T. Reinert, Nuclear Instruments & Methods in Physics Research Section B-Beam Interactions with Materials and Atoms, 268(11-12), pp. 2156–2159 Search PubMed.
  120. P. N. R. Birri, R. D. Perez, D. Cremonezzi, C. A. Perez, M. Rubio and G. A. Bongiovanni, Environmental Research, 110(5), pp. 417–423 Search PubMed.
  121. B. De Samber, G. Silversmit, K. De Schamphelaere, R. Evens, T. Schoonjans, B. Vekemans, C. Janssen, B. Masschaele, L. Van Hoorebeke, I. Szaloki, F. Vanhaecke, K. Rickers, G. Falkenberg and L. Vincze, Journal of Analytical Atomic Spectrometry, 25(4), pp. 544–553 Search PubMed.
  122. Y. E. Eltahir, H. M. Ali, M. H. Mansour and O. Mahgoub, Journal of Animal and Veterinary Advances, 9(4), pp. 764–770 Search PubMed.
  123. G. Concha, K. Broberg, M. Grander, A. Cardozo, B. Palm and M. Vahter, Environmental Science & Technology, 44(17), pp. 6875–6880 Search PubMed.
  124. Y. Y. Guo, X. Huo, Y. Li, K. S. Wu, J. X. Liu, J. R. Huang, G. N. Zheng, Q. N. Xiao, H. Yang, Y. P. Wang, A. M. Chen and X. J. Xu, Science of the Total Environment, 408(16), pp. 3113–3117 Search PubMed.
  125. I. Lavilla, M. Costas, P. San Miguel, J. Millos and C. Bendicho, BioMetals, 2009, 22(6), 863–875 Search PubMed.
  126. D. Bohrer, F. Bortoluzzi, P. C. do Nascimento, L. M. de Carvalho and S. R. de Oliveira, J. Parenter. Enteral Nutr., 2009, 33(6), 694–701 Search PubMed.
  127. X. H. Hou, L. V. Lamberts, G. J. Guan and P. C. D'Haese, Trace Elements and Electrolytes, 27(1), pp. 10–19 Search PubMed.
  128. L. Lukaszczyk and W. Zyrnicki, Journal of Pharmaceutical and Biomedical Analysis, 52(5), pp. 747–751 Search PubMed.
  129. D. B. D. Neves, E. D. Caldas and R. N. R. Sampaio, Trop. Med. Int. Health, 2009, 14(12), 1515–1522 CrossRef.
  130. K. Lew, J. P. Acker, S. Gabos and X. C. Le, Environmental Science & Technology, 44(10), pp. 3986–3991 Search PubMed.
  131. Y. K. Huang, Y. L. Huang, Y. M. Hsueh, J. T. J. Wang, M. H. Yang and C. J. Chen, Environmental Health Perspectives, 2009, 117(12), 1860–1866 CAS.
  132. S. Ouypornkochagorn and J. Feldmann, Environmental Science & Technology, 44(10), pp. 3972–3978 Search PubMed.
  133. T. T. Ngu, S. Krecisz and M. J. Stillman, Biochemical and Biophysical Research Communications, 396(2), pp. 206–212 Search PubMed.
  134. M. Kippler, W. Goessler, B. Nermell, E. C. Ekstrom, B. Lonnerdal, S. El Arifeen and M. Vahter, Environ. Res., 2009, 109(7), 914–921 CrossRef CAS.
  135. Y. Liu, Y. X. Gao, L. L. Zhang, T. C. Wang, J. X. Wang, F. Jiao, W. Li, Y. F. Li, B. Li, Z. F. Chai, G. Wu and C. Y. Chen, J. Nanosci. Nanotechnol., 2009, 9(11), 6335–6343 CrossRef CAS.
  136. K. Minakata, I. Yamagishi, S. Kanno, H. Nozawa, M. Suzuki and O. Suzuki, Journal of Chromatography B-Analytical Technologies in the Biomedical and Life Sciences, 878(20), 1683–1686 Search PubMed.
  137. T. H. Darrah, J. J. Prutsman-Pfeiffer, R. J. Poreda, M. E. Campbell, P. V. Hauschka and R. E. Hannigan, Metallomics, 2009, 1(6), 479–488 RSC.
  138. V. Hollriegl, M. Gonzalez-Estecha, E. M. Trasobares, A. Giussani, U. Oeh, M. A. Herraiz and B. Michalke, Journal of Trace Elements in Medicine and Biology, 24(3), pp. 193–199 Search PubMed.
  139. K. Bandeen-Roche, T. A. Glass, K. I. Bolla, A. C. Todd and B. S. Schwartz, Epidemiology, 2009, 20(6), 831–839 CrossRef.
  140. K. Chandramouli, C. D. Steer, M. Ellis and A. M. Emond, Arch. Dis. Child., 2009, 94(11), 844–848 CrossRef CAS.
  141. D. S. Kim, S. D. Yu and E. H. Lee, Molecular & Cellular Toxicology, 6(1), pp. 19–23 Search PubMed.
  142. C. G. Lin, L. A. Schaider, D. J. Brabander and A. D. Woolf, Pediatrics, 125(4), pp. E828–E835 Search PubMed.
  143. G. R. C. de Almeida, C. F. D. Tavares, A. M. de Souza, T. S. de Sousa, C. A. R. Funayama, F. Barbosa, J. E. Tanus-Santos and R. F. Gerlach, Science of the Total Environment, 408(7), pp. 1551–1556 Search PubMed.
  144. V. B. Rezende, J. H. Amaral, R. F. Gerlach, F. Barbosa and J. E. Tanus-Santos, Journal of Trace Elements in Medicine and Biology, 24(3), pp. 147–151 Search PubMed.
  145. M. Matsumoto and J. Yoshinaga, Environmental Science and Pollution Research, 17(3), pp. 643–649 Search PubMed.
  146. B. C. Henn, A. S. Ettinger, J. Schwartz, M. M. Tellez-Rojo, H. Lamadrid-Figueroa, M. Hernandez-Avila, L. Schnaas, C. Amarasiriwardena, D. C. Bellinger, H. Hu and R. O. Wright, Epidemiology, 21(4), pp. 433–439 Search PubMed.
  147. A. Skoczynska, R. Poreba, A. Steinmentz-Beck, H. Martynowicz, A. Affelska-Jercha, B. Turczyn, A. Wojakowska and I. Jedrychowska, Int. J. Occup. Med. Environ. Health, 2009, 22(2), 135–142 Search PubMed.
  148. I. Hertz-Picciotto, P. G. Green, L. Delwiche, R. Hansen, C. Walker and I. N. Pessah, Environmental Health Perspectives, 118(1), pp. 161–166 Search PubMed.
  149. P. Rodriguez-Gonzalez, V. N. Epov, R. Bridou, E. Tessier, R. Guyoneaud, M. Monperrus and D. Amouroux, Environ. Sci. Technol., 2009, 43(24), 9183–9188 CrossRef CAS.
  150. T. Falta, G. Koellensperger, A. Standler, W. Buchberger, R. M. Mader and S. Hann, J. Anal. At. Spectrom., 2009, 24(10), 1336–1342 RSC.
  151. A. Bouslimani, N. Bec, M. Glueckmann, C. Hirtz and C. Larroque, Rapid Communications in Mass Spectrometry, 24(4), pp. 415–421 Search PubMed.
  152. C. F. Harrington, R. C. Le Pla, G. D. D. Jones, A. L. Thomas and P. B. Farmer, Chemical Research in Toxicology, 23(8), pp. 1313–1321 Search PubMed.
  153. W. H. Wang, Z. L. Chen, D. E. Davey and R. Naidu, Journal of Liquid Chromatography & Related Technologies, 33(9–12), pp. 1151–1173 Search PubMed.
  154. P. Jitaru, H. Goenaga-Infante, S. Vaslin-Reimann and P. Fisicaro, Analytica Chimica Acta, 657(2), pp. 100–107 Search PubMed.
  155. Y. Tsuji, T. Mikami, Y. Anan and Y. Ogra, Metallomics, 2(6), pp. 412–418 Search PubMed.
  156. Y. Yamashita and M. Yamashita, Journal of Biological Chemistry, 285(24), pp. 18134–18138 Search PubMed.
  157. R. L. Bruhn, W. D. Stamer, L. A. Herrygers, J. M. Levine and R. J. Noecker, Br. J. Ophthalmol., 2009, 93(9), 1155–1158 CrossRef CAS.
  158. G. De Palma, P. Manini, M. Sarnico, S. Molinari and P. Apostoli, International Archives of Occupational and Environmental Health, 83(2), pp. 173–181 Search PubMed.
  159. C. S. Li, K. Benkhedda, S. Tolmachev, L. Carty, R. Ko, D. Moir, J. Cornett and G. Kramer, Journal of Analytical Atomic Spectrometry, 25(5), pp. 730–734 Search PubMed.
  160. S. D'Ilio, N. Violante, O. Senofonte, C. Majorani and F. Petrucci, Analytical Methods, 2(8), pp. 1184–1190 Search PubMed.
  161. M. Podgorczyk, W. M. Kwiatek, D. Grolimund and C. Borca, presented at 9th International School and Symposium on Synchrotron Radiation in Natural Science (ISSRNS-9), Ameliowka, POLAND, Jun 15–20, 2008 Search PubMed.
  162. V. Balter, A. Zazzo, A. P. Moloney, F. Moynier, O. Schmidt, F. J. Monahan and F. Albarede, Rapid Communications in Mass Spectrometry, 24(5), pp. 605–612 Search PubMed.
  163. H. Tumir, J. Bosnir, I. Vedrina-Dragojevic, Z. Dragun, S. Tomic and D. Puntaric, Homeopathy, 99(3), pp. 183–188 Search PubMed.
  164. M. J. Martena, J. C. A. Van Der Wielen, I. Rietjens, W. N. M. Klerx, H. N. De Groot and E. J. M. Konings, Food Additives and Contaminants Part a-Chemistry Analysis Control Exposure & Risk Assessment, 27(2), pp. 190–205 Search PubMed.
  165. U. Arroyo-Abad, J. Mattusch, S. Mothes, M. Moeder, R. Wennrich, M. P. Elizalde-Gonzalez and F. M. Matysik, Talanta, 82(1), pp. 38–43 Search PubMed.
  166. A. G. Bravo, J. L. Loizeau, S. Bouchet, A. Richard, J. F. Rubin, V. G. Ungureanu, D. Amouroux and J. Dominik, Environmental Science and Pollution Research, 17(8), pp. 1422–1432 Search PubMed.
  167. E. Lipiec, G. Siara, K. Bierla, L. Ouerdane and J. Szpunar, Analytical and Bioanalytical Chemistry, 397(2), pp. 731–741 Search PubMed.
  168. Q. L. Chan, S. E. Afton and J. A. Caruso, Metallomics, 2(2), pp. 147–153 Search PubMed.
  169. E. Govasmark, A. Brandt-Kjelsen, J. Szpunar, K. Bierla, G. Vegarud and B. Salbu, Pure and Applied Chemistry, 82(2), pp. 461–471 Search PubMed.
  170. F. Cubadda, F. Aureli, S. Ciardullo, M. D'Amato, A. Raggi, R. Acharya, R. A. V. Reddy and N. T. Prakash, Journal of Agricultural and Food Chemistry, 58(4), pp. 2295–2301 Search PubMed.
  171. M. Rose, M. Baxter, N. Brereton and C. Baskaran, Food Additives and Contaminants Part a-Chemistry Analysis Control Exposure & Risk Assessment, 27(10), pp. 1380–1404 Search PubMed.
  172. L. Nasreddine, O. Nashalian, F. Naja, L. Itani, D. Parent-Massin, M. Nabhani-Zeidan and N. Hwalla, Food and Chemical Toxicology, 45(5), pp. 1262–1269 Search PubMed.
  173. P. Liu, C. N. Wang, X. Y. Song, Y. F. Yu and Y. N. Wu, Food Additives and Contaminants Part a-Chemistry Analysis Control Exposure & Risk Assessment, 27(8), pp. 1128–1135 Search PubMed.
  174. J. Jenny-Burri, M. Haldimann and V. Dudler, Food Additives and Contaminants Part a-Chemistry Analysis Control Exposure & Risk Assessment, 27(11), pp. 1516–1531 Search PubMed.
  175. T. Cirillo, E. Fasano, V. Viscardi, A. Arnese and R. Amodio-Cocchieri, Food Additives & Contaminants Part B-Surveillance, 3(1), pp. 30–38 Search PubMed.
  176. M. E. Soares, E. Vieira and M. D. Bastos, Journal of Agricultural and Food Chemistry, 58(2), pp. 1366–1370 Search PubMed.
  177. C. Matos, C. Moutinho, V. Balcao, C. Almeida, M. Ribeiro, A. F. Marques and A. Guerra, Eur. Food Res. Technol., 2009, 230(2), 201–208 CrossRef CAS.
  178. M. Sigrist, H. Beldomenico and M. R. Repetti, Food Chemistry, 121(2), pp. 487–491 Search PubMed.
  179. A. M. Carey, K. G. Scheckel, E. Lombi, M. Newville, Y. Choi, G. J. Norton, J. M. Charnock, J. Feldmann, A. H. Price and A. A. Meharg, Plant Physiology, 152(1), pp. 309–319 Search PubMed.
  180. A. J. Signes-Pastor, C. Deacon, R. O. Jenkins, P. I. Haris, A. A. Carbonell-Barrachina and A. A. Meharg, J. Environ. Monit., 2009, 11(11), 1930–1934 RSC.
  181. T. Nishimura, M. Hamano-Nagaoka, N. Sakakibara, T. Abe, Y. Maekawa and T. Maitani, Food Hygiene and Safety Science, 51(4), pp. 178–181 Search PubMed.
  182. F. Liang, Y. L. Li, G. L. Zhang, M. G. Tan, J. Lin, W. Liu, Y. Li and W. W. Lu, Food Additives and Contaminants Part a-Chemistry Analysis Control Exposure & Risk Assessment, 27(6), pp. 810–816 Search PubMed.
  183. T. R. Boen and J. A. L. Pallone, Cereal Chem., 2009, 86(6), 695–700 CrossRef CAS.
  184. B. S. Choi, S. J. Choi, D. W. Kim, M. Huang, N. Y. Kim, K. S. Park, C. Y. Kim, H. M. Lee, Y. N. Yum, E. S. Han, T. S. Kang, I. J. Yu and J. D. Park, Archives of Environmental Contamination and Toxicology, 58(1), pp. 222–229 Search PubMed.
  185. C. Garcia-Sartal, M. C. B. Alonso, P. Herbello-Hermelo and P. Bermejo-Barrera, Atomic Spectroscopy, 31(1), pp. 27–33 Search PubMed.
  186. S. H. Nam, H. J. Oh, H. S. Min and J. H. Lee, Microchemical Journal, 95(1), pp. 20–24 Search PubMed.
  187. M. Rudy, Food Additives and Contaminants Part a-Chemistry Analysis Control Exposure & Risk Assessment, 27(4), pp. 464–472 Search PubMed.
  188. R. T. Chibunda and C. R. Janssen, Food Additives and Contaminants Part a-Chemistry Analysis Control Exposure & Risk Assessment, 2009, 26(11), 1482–1487 Search PubMed.
  189. A. H. Barati, A. Maleki and M. Alasvand, Science of the Total Environment, 408(7), pp. 1523–1529 Search PubMed.
  190. R. L. Hough, T. Fletcher, G. S. Leonardi, W. Goessler, P. Gnagnarella, F. Clemens, E. Gurzau, K. Koppova, P. Rudnai, R. Kumar and M. Vahter, International Archives of Occupational and Environmental Health, 83(5), pp. 471–481 Search PubMed.
  191. A. Bhavana, S. Vibha, T. C. Rekha and J. Praveen, Research Journal of Chemistry and Environment, 2009, 13(4), 23–27 Search PubMed.
  192. L. Y. Xing and D. Beauchemin, Journal of Analytical Atomic Spectrometry, 25(7), pp. 1046–1055 Search PubMed.
  193. C. Reimann, M. Birke and P. Filzmoser, Applied Geochemistry, 25(7), pp. 1030–1046 Search PubMed.
  194. P. Pohl and I. Sergiel, J. Agric. Food Chem., 2009, 57(20), 9378–9384 CrossRef CAS.
  195. S. Pessanha, M. L. Carvalho, M. Becker and A. von Bohlen, Spectrochimica Acta Part B-Atomic Spectroscopy, 65(6), pp. 504–507 Search PubMed.
  196. F. Camin, R. Larcher, M. Perini, L. Bontempo, D. Bertoldi, G. Gagliano, G. Nicolini and G. Versini, Food Chemistry, 118(4), pp. 901–909 Search PubMed.
  197. F. Camin, R. Larcher, G. Nicolini, L. Bontempo, D. Bertoldi, M. Perini, C. Schlicht, A. Schellenberg, F. Thomas, K. Heinrich, S. Verkelius, M. Horacek, H. Ueckermann, H. Froeschl, B. Wimmer, G. Heiss, M. Baxter, A. Rossmann and J. Hoogewerff, Journal of Agricultural and Food Chemistry, 58(1), pp. 570–577 Search PubMed.
  198. M. Rosner, Food Chemistry, 121(3), pp. 918–921 Search PubMed.
  199. C. Fillol, F. Dor, L. Labat, P. Boltz, J. Le Bouard, K. Mantey, C. Mannschott, E. Puskarczyk, F. Viller, I. Momas and N. Seta, Science of the Total Environment, 408(5), pp. 1190–1194 Search PubMed.
  200. K. Ito, C. D. Palmer, A. J. Steuerwald and P. J. Parsons, Journal of Analytical Atomic Spectrometry, 25(8), pp. 1334–1342 Search PubMed.
  201. L. M. Wang, J. S. Becker, Q. Wu, M. F. Oliveira, F. A. Bozza, A. L. Schwager, J. M. Hoffman and K. A. Morton, Metallomics, 2(5), pp. 348–353 Search PubMed.
  202. E. Alshammari, S. Shafi, J. Nurmi-Lawton, A. Taylor, S. Lanham-New and G. Ferns, International Journal of Sport Nutrition and Exercise Metabolism, 20(4), pp. 291–298 Search PubMed.
  203. M. Almalki, S. A. Majid, P. H. Butler and L. Reinisch, Australasian Physical & Engineering Sciences in Medicine, 33(2), pp. 185–191 Search PubMed.
  204. H. Ishitobi, S. Stern, S. W. Thurston, G. Zareba, M. Langdon, R. Gelein and B. Weiss, Environmental Health Perspectives, 118(2), pp. 242–248 Search PubMed.
  205. K. Miyamoto, T. Kuwana, T. Ando, M. Yamamoto and A. Nakano, Journal of Toxicological Sciences, 35(2), pp. 217–224 Search PubMed.
  206. K. Schoeman, T. Tanaka, J. R. Bend and G. Koren, Journal of Pediatrics, 157(1), pp. 127–131 Search PubMed.
  207. C. Yazbeck, O. Thiebaugeorges, T. Moreau, V. Goua, G. Debotte, J. Sahuquillo, A. Forhan, B. Foliguet, G. Magnin, R. Slama, M. A. Charles and G. Huel, Environmental Health Perspectives, 2009, 117(10), 1526–1530 CAS.
  208. K. Bischoff, C. Gaskill, H. N. Erb, J. G. Ebel and J. Hillebrandt, Journal of Veterinary Diagnostic Investigation, 22(5), pp. 729–733 Search PubMed.
  209. M. Tondo, J. Moreno, M. Casado, N. Brandi, C. Sierra, M. A. Vilaseca, A. Ormazabal and R. Artuch, Neurochemical Research, 35(8), pp. 1290–1293 Search PubMed.
  210. D. A. Rogers, S. J. Ray and G. M. Hieftje, Metallomics, 2(4), pp. 271–279 Search PubMed.
  211. D. A. Rogers, S. J. Ray and G. M. Hieftje, Metallomics, 2(4), pp. 280–288 Search PubMed.
  212. A. Banas, W. M. Kwiatek, K. Banas, M. Gajda, B. Pawlicki and T. Cichocki, Journal of Biological Inorganic Chemistry, 15(7), pp. 1147–1155 Search PubMed.
  213. B. L. Batista, J. L. Rodrigues, V. C. D. Souza and F. Barbosa, Forensic Sci. Int., 2009, 192(1–3), 88–93 CrossRef CAS.
  214. S. Ahmed and W. Santosh, Plos One, 5(6), p. 9 Search PubMed.
  215. A. Dogukan, N. Sahin, M. Tuzcu, V. Juturu, C. Orhan, M. Onderci, J. Komorowski and K. Sahin, Biol. Trace Elem. Res., 2009, 131(2), 124–132 CrossRef CAS.
  216. T. Narukawa and K. Chiba, Journal of Agricultural and Food Chemistry, 58(14), pp. 8183–8188 Search PubMed.
  217. N. B. Khan and A. N. Chohan, Environmental Monitoring and Assessment, 166(1–4), pp. 169–176 Search PubMed.
  218. S. Catarino, A. Neves and R. B. de Sousa, Ciencia E Tecnica Vitivinicola, 2009, 24(2), 65–72 Search PubMed.
  219. K. Miranda, A. G. G. Dionisio and E. R. Pereira, Microchemical Journal, 96(1), pp. 99–101 Search PubMed.
  220. M. J. da Silva, A. P. S. Palm, M. F. Pimentel, M. L. Cervera and M. de la Guardia, Analytica Chimica Acta, 667(1-2), pp. 43–48 Search PubMed.
  221. K. Benkhedda, A. Robichaud, S. Turcotte, F. J. Beraldin and K. A. Cockell, Journal of Aoac International, 2009, 92(6), 1720–1727 CAS.
  222. M. C. S. da Silva, J. Naozuka, P. V. Oliveira, M. C. D. Vanetti, D. M. S. Bazzolli, N. M. B. Costa and M. C. M. Kasuya, Metallomics, 2(2), pp. 162–166 Search PubMed.
  223. P. O. Amoako, P. C. Uden and J. F. Tyson, Anal. Chim. Acta, 2009, 652(1-#x2013;2), 315–323 CrossRef CAS.
  224. S. Azadmard-Damirchi and P. C. Dutta, J. Chromatogr., A, 2009, 1216(1), 36–42 CrossRef CAS.
  225. D. Cautela, F. Santelli, F. Boscaino, B. Laratta, L. Servillo and D. Castaldo, J. Sci. Food Agric., 2009, 89(13), 2283–2291 CrossRef CAS.
  226. K. Schlesier, C. Fauhl-Hassek, M. Forina, V. Cotea, E. Kocsi, R. Schoula, F. van Jaarsveld and R. Wittkowski, Eur. Food Res. Technol., 2009, 230(1), 1–13 CrossRef CAS.

This journal is © The Royal Society of Chemistry 2011