Atomic Spectrometry Update: review of advances in elemental speciation

Contents

• 1 Topical reviews

• 2 CRMs and metrology

• 3 Elemental speciation analysis
  • 3.1 Antimony
  • 3.2 Arsenic
  • 3.3 Cerium
  • 3.4 Cesium
  • 3.5 Chromium
  • 3.6 Europium
  • 3.7 Gold
  • 3.8 Halogens
  • 3.9 Iron
  • 3.10 Lead
  • 3.11 Manganese
  • 3.12 Mercury
  • 3.13 Nickel
  • 3.14 Selenium
  • 3.15 Sulfur
  • 3.16 Tellurium
  • 3.17 Thallium
  • 3.18 Tin
  • 3.19 Vanadium
  • 3.20 Zinc

• 4 Biomolecular speciation analysis
  • 4.1 Interaction of metallodrugs with biological systems

• Conflicts of interest

• 5 Abbreviations used in this update

• References

---

Abstract

This is the tenth Atomic Spectrometry Update (ASU) to focus on advances in elemental speciation and covers a period of approximately 12 months from December 2016. This ASU review deals with all aspects of the analytical atomic spectrometry speciation methods developed for: the determination of oxidation states; organometallic compounds; coordination compounds; metal and heteroatom-containing biomolecules, including metalloproteins, proteins, peptides and amino acids; and the use of metal-tagging to facilitate detection via atomic spectrometry. The review does not cover fractionation, which is sometimes termed operationally defined speciation. As with all ASU reviews the focus of the research reviewed includes those methods that incorporate atomic spectrometry as the measurement technique. However, because speciation analysis is inherently focused on the relationship between the metal(loid) atom and the organic moiety it is bound to, or incorporated within, atomic spectrometry alone cannot be the sole analytical approach of interest. For this reason molecular detection techniques are also included where they have provided a complementary approach to speciation analysis. As in previous years, As speciation continues to dominate the current literature with a significant number of publications concerning Hg and Se speciation and ‘biomolecules’, in which category a number of papers covered in the Se section could also belong. The number of elements covered in these updates also continues to rise with speciation analysis of over 20 elements reported this year.

---

1 Topical reviews

This latest update adds to that from last year¹ and complements other reviews^2–6 of analytical techniques in the series of Atomic Spectrometry Updates. Two books from the same publisher contain chapters concerned with elemental speciation. The first entitled ‘Inorganic Trace Analytics: Trace element analysis and speciation’ is divided into three sections containing a total of 10 chapters, each of which is written by a small number of expert practitioners.⁷ Part 1 covers sample preparation and analytical techniques; part 2 contains five chapters each of which deals with a particular application area, environment, food, biological, forensic and industrial; whilst part 3 is entitled “Inorganic and bioinorganic speciation analysis at trace level”. The three chapters in this section are entitled (a) quality of results in trace element and speciation analysis, (b) sample pre-treatment for trace speciation analysis, and (c) solid-phase extraction in fractionation of trace elements. The second book “Handbook of Rare Earth Elements: Analytics”, again a collection of chapters written by various experts, contains 14 chapters.⁸ However, in this case only one of them is related to speciation analysis, namely ‘analysis and speciation of lanthanoids’, which would appear to deal only with compounds of Gd (contrast agents). It appears from the publisher’s website that individual chapters of both books can be purchased separately and downloaded as files in portable document format.

As has been the pattern in recent review periods, numerous reviews of various aspects of elemental speciation have appeared in the peer reviewed literature. These reviews can be broadly classified according to the following, the atomic spectrometry technique, the separation technique, the sample material, the analytes, or various combinations thereof.

Seven review articles deal with aspects of the measurement of metal species in biological materials or systems. Professor H. Haraguchi has surveyed developments in the field since he coined the term ‘metallomics’ in 2004.⁹ The review is somewhat selective as only 67 articles are cited (titles are provided). He concludes that metallomics has been developing as both an interdisciplinary and multi-disciplinary field together with a variety of biometal sciences, as well as with genomics and proteomics. He proposes that to accelerate the development of metallomics, the organization of a scientific platform is needed so that information and technological developments can be shared among the practitioners of a number of mission-oriented sciences such as medicine, pharmacy, toxicology, food and nutrition. A prediction is made that elemental analysis of single-cells and the imaging of cells and organs will be the major research fields in the next decade, together with studies in the cell biology of metals from multi-elemental aspects, molecular and genetic features of various metals, and inorganic drug design for chemotherapy. The imaging-of-metals theme is covered in a more detailed review by Ackerman et al.¹⁰ They cite 212 articles (titles are not given) to highlight the power of complementary analytical techniques to map total metal pools and identify subsets that are (a) static and tightly bound or (b) dynamic and weakly bound. The reviewers focus on the two most abundant redox-active transition metals in living systems: Cu and Fe, both of which have been traditionally studied in bioinorganic chemistry as static cofactors tightly bound by metallochaperones and buried in protein active sites to protect cells against oxidative stress but which are now are emerging as dynamic transition metal signals that can reversibly affect the function of proteins in allosteric regions outside active sites. A significant portion of the review is devoted to the application of molecular probes (synthetic small-molecule reagents), whose use is viewed as a crucial complement to the atomic spectrometry techniques, such as LA-ICP-MS and various X-ray absorption spectrometries. The reviewers conclude that open collaboration between chemistry and biology, in which the disparate communities work together to evaluate the best chemical imaging tools for a given biological system, is key to progress. One tool that features prominently in such studies is LA-ICP-MS for which Pozebon et al.¹¹ have updated their earlier 2014 review. The key advances identified are fast wash-out ablation cells, calibration strategies (such as ink printing and dried-droplets), mapping of elemental distribution in animal, human and plant tissues, nanoparticle uptake, and protein and single cell analysis. Some 11-pages of the review (about 50% of the total) are occupied by a summary table of 70 articles with three columns: element, sample and remarks. The contents of the articles are also discussed in the text along with citations to another 70 or so, for a total of 142 references, whose titles are not given. The reviewers conclude that improvements in a number of areas are needed, including low dispersion ablation cells, optimal laser parameters, high sample transport, complete aerosol ionization with low diffusion inside the ICP, higher ion beam transmission to the detector and simultaneous detection technology. They also highlight the need for the provision of three-dimensional information. The application of ICP-MS-based techniques for the determination of trace elements and their species in cells has been reviewed.¹² In addition to LA, the review (60 references without titles) includes recent developments in multi-dimensional chromatograph and electrophoresis combined with ICP-MS and highlights the improved LODs that are obtained from the development of chip-based micro-extraction techniques. However, the reviewers conclude that LOD values are not yet good enough for the analysis of single cells and that further developments in miniature sample preparation and separation are needed to obtain speciation information. They also consider that further developments in applications of tagging (of both biomolecules and cells) for quantification by ICP-MS are needed. It is suggested that future developments would include more universal and robust elemental tagging strategies, amplification strategies to overcome current LOD values, and methods with suitable elemental tags for multi-cell lines. The use of exogenous tags for the quantitative determination of biomolecules has also been comprehensively reviewed (245 references with titles).¹³ The reviewers identify four major topics: applications of HPLC-ICP-MS in the absolute/relative quantification of proteins with exogenous tags, applications of ICP-MS in immunoassays with exogenous tags, applications of LA-ICP-MS with exogenous tags, and applications of ICP-MS in the quantification of nucleic acids with exogenous tags. Each of these topics is summarised in a table. The review features detailed tutorial introductions to the various topics, including an explanation of the benefits and advantages of ICP-MS. The reviewers are perhaps a bit optimistic in considering that responses are independent of the compound and the matrix. They conclude that further research is needed in a number of areas, pointing out that although very low LOD values can be obtained, accurate quantification of low abundance proteins is still a challenge. They also call for the development of new signal amplification techniques and increases in the speed of analysis. They indicate that further applications at both the “macroscopic and microcosmic” scales are needed, pointing out that, although mass cytometry has been developed for the single cell analysis, ICP-MS-based assays have not yet been developed for smaller parts of the organism, such as mitochondria and chloroplasts. They consider that ICP-MS based procedures will find application alongside techniques such as luminescence imaging and molecular MS in the accurate diagnosis of disease. Sample pre-treatment procedures for the elemental speciation in biological materials by methods involving ICP-MS have been reviewed.¹⁴ The review cites 219 references over the past 10 years and is somewhat selective in that methods in which LC separations are implemented are only briefly mentioned; the main focus of the review is on liquid and solid-phase micro-extractions, and chip-based manifolds that can be coupled to ICP-MS instrumentation. The former included various miniaturised LLE procedures as well as matrix solid phase dispersion. In addition to SPME on fibres, capillary microextraction and stir-bar methods are reviewed. A considerable portion of the review is devoted to on-chip procedures including CE, though as the reviewers point out, there are no applications reported for biological samples. They conclude that there is currently a shortage of calibration standards and CRMs for elemental speciation and that further development will require the use of molecular mass spectrometry. The relatively low concentrations of the elemental species means that pre-treatment methods with higher extraction efficiencies and selectivity are still urgently needed. They speculate that multiple extractions could improve method detection capability and decrease matrix interferences. Finally in this section on applications to biological systems, the application of ICP-MS in quantitative proteomics has been reviewed.¹⁵ The reviewers cite 127 references (titles given) covering approaches in which quantification is based on (a) the measurement of elements already present in the target species, such as P and S, which allows the elucidation of the degree of protein phosphorylation in addition to absolute protein quantification, and (b) on bi-functional labelling substances, containing elements detectable by ICP-MS, that undergo covalent chemical reactions of previously established stoichiometry with the protein. The reviewers point out that this latter technique is very useful for the design of precise multiplexed quantification in biomarker screening and discovery schemes. They conclude that the increased interest in ICP-MS is reflected in the large number of proteomics publications describing complementary applications of plasma-source MS and MS with soft ionization techniques, such as MALDI and ES.

A number of reviews are focused on specific aspects of the instrumental techniques with an emphasis on NP characterization. In a well-constructed review, Marcinkowska and Baralkiewicz consider elemental speciation analysis by HPLC-ICP-MS, but confine their attention to those procedures in which speciation of more than one element is obtained.¹⁶ Most of the 92 articles cited (titles are provided) have appeared since 2000. The reviewers consider that the availability of commercial instruments with collision/reaction cell technology (CCT), which they consider to be an effective tool for the elimination of spectral interferences, is responsible for the growing use of the technique. They point out that the number of papers concerning multi-elemental speciation analysis is still increasing, indicative of the continued and growing interest in this field of analytical chemistry. The review opens with a number of useful sections dealing with the need for speciation analysis, of the interaction between elements and the current status of government regulations around the world. Most of the papers cited deal with two elements, about 20% of them concern three elements with only a few dealing with the speciation of four or more elements. The determination of As and Se species (mainly As^III, As^V, MMA^V, DMA^V, Se^IV, Se^VI, SeMet, SeCys and SeCys₂) predominate. The reviewers note that Cr^VI is determined three times more often than is Cr^III, whereas Sb^III and Sb^V are determined with about the same frequency. Several procedures for Hg and Pb species were also noted, but only a few procedures include Br, Cl, Cd, Mo, Te, Tl, V, or W. The most frequently analysed liquid matrices were water, urine or serum, and for solids, the most often reported were food products, ash and soil. The review includes a section on sample preparation. Although chromatographic conditions are surveyed, there is little critical commentary of this aspect of the methodology, nor is there any discussion of compound-dependent responses. However, there is a good section on the ‘metrological approach’, in which the reviewers point out that many papers do not contain a discussion of the uncertainty budget. The reviewers conclude that the continued development in this area of chemical analysis will lead to advances in the understanding of how biologically active elements interact with and react within living organisms. In a review of recent (June 2104 to June 2016) advances in CE-MS coupling, covering instrumentation, methodology, and applications, elemental speciation by CE-ICP-MS is featured in some 18 of the 161 articles (no titles) cited.¹⁷ Although the focus is on instrumentation and methodology, examples are given of applications in the fields of proteomics, glycomics, metabolomics, biomarker research, forensics, pharmacology, food analysis, and single-cell analysis. Separations by capillary electrochromatography, and micellar electrokinetic chromatography are not covered. Several of the ICP-MS examples were concerned with the determination of metallic NPs; it was pointed out, for example, that speciation analysis of metal-containing nanoscale materials in blood serum is needed to distinguish between the intact NPs and various protein conjugates. The application of CE to this topic is the subject of a review¹⁸ covering developments in the past 5 years. In considering possibilities for detection, the reviewers conclude that because of the specificity and detection capability of ICP-MS, it has become the ‘conventional approach in nano-bio studies’, though in fact only some 15 of the 53 articles (no titles) cited describe procedures based on CE-ICP-MS. The reviewers conclude, nonetheless, that as engineered nanoparticles containing metals are designed for intravenous administration, but before arriving at the target organ or cell are transformed in the blood by interaction with proteins, a crucial step in understanding their mode of action is evaluation of the reactivity and affinity of nanoparticles with these proteins. They also point out that the systematic application of ICP-MS detection has resulted in a paradigm shift from probing simple in vitro species systems to mapping multiplex interactions in real serum samples. One of the advantages of ICP-MS is that target analyte concentrations remain above the LOD when dilution is the approach to overcoming interferences from the saline, protein-rich matrices. The interaction between engineered nanomaterials and plants has been reviewed¹⁹ with particular reference to the application of micro- and nano-XRF mapping and X-ray absorption spectroscopy. The reviewers argue that the impact of engineered nanomaterials on plants, which are central components of ecosystems and the worldwide food supply, is of primary relevance. Hence, understanding the fate and physical and chemical modifications of NPs in plants and the possible transfer into food chains is crucial. They argue that the specialised analytical techniques available at synchrotron radiation facilities are well suited for this purpose as X-ray fluorescence mapping provides multi-elemental detection with lateral resolution at the double-digit nm scale, and spatially resolved X-ray absorption spectroscopy provides speciation information. The review features the contents of 102 articles whose titles are provided. Following well-written tutorial introductory sections, considerable space is devoted to sample preparation and to data acquisition and processing. The reviewers conclude that sample preparation will continue to be critical and that appropriate sample preparation protocols need to be further developed, especially for nano-X-ray fluorescence microscopy in order to preserve ultrastructure and chemical speciation. They point out that although the sample preparation strategy depends on the facilities available at the beamline, several sites now have the cryogenic setups that are highly beneficial for speciation studies.

Several reviews cover aspects of non-chromatographic separation. In a comprehensive review²⁰ of porous monoliths for on-line sample preparation, 172 references (with titles) are cited, but only about 22 of these are reports of elemental analysis of which a subset are concerned with speciation analysis. The main areas of application are for the analyses of environmental and biological samples. The bulk of the review is concerned with developments for organic analytes, which are discussed in considerable detail. A review²¹ of another aspect of SPE, the use of advanced functional materials in the determination of trace elements and their species by ICP-MS, covered the material in 190 references (whose titles are provided). The reviewers consider advanced functional materials to include nanometre-sized materials, porous materials, ion-imprinted polymers, restricted access materials, and magnetic materials, with properties such as high adsorption capacity, good selectivity, and fast adsorption/desorption kinetics. Monolithic materials are dealt with in some detail and there are 17 citations to the use of magnetic materials, which can be readily collected after dispersion in a bulk liquid. This approach may overcome the limitations of high back-pressures when nanometre sized adsorbents are packed into small columns. On the other hand, monolithic columns, featuring low back-pressure and high extraction capacity for trace elements and their species, which are easy to connect to ICP-MS detection, have shown potential for the determination of trace elements and their species in samples with limited sample amount, such as cells. Some cells have the ability to selectively bind elemental species of interest and the progress in this area of SPE speciation analysis over the period 2005–2015 has been reviewed.²² The review contains a well-written introduction to the interaction of relevant species with both living and dead cells that covers not only the nature of the selective binding but also aspects of the toxicity of the various species. The reviewers point out the advantages of low cost, the wide range of sources, abundant binding sites and environmentally friendly features. They also claim that as cells have the capability to distinguish metal species based on the toxicity difference that this ‘endows them with incomparable superiority over synthetic adsorbents as solid phase extraction (SPE) media for heavy metal speciation’. Only 65 articles (no titles) are cited of which about half describe analytical procedures, many of which are featured in a summary table from which it can be seen that, with the exception of Hg, all the methods are concerned with the separation of inorganic species of differing oxidation state. The element for which most articles have been published is Cr (11 papers), followed by As with 9, Hg with 7, Se with 5, Sb with 3 and Sn and Tl one each. The reviewers conclude that the selectivity of cell-based heavy metal speciation is limited due to the complex nature of the surface binding groups of biological cells, and that the diversity of surface binding groups on biological cells acts as a double-edged sword: large capacity but poor selectivity towards some metal species. They propose that either chemical or genetic modification is needed to adjust the metal uptake behaviour to achieve the desired selectivity for metal speciation and speculate that this might be achieved through the expression of selected metal-binding proteins. Turker²³ has reviewed the speciation analysis of trace metals and metalloids by SPE with spectrometric detection as evidenced in 83 articles (no titles) covering the period 2012 through early 2016. Chromatographic separations and fractionations are excluded. The spectrometric detection techniques employed are almost all atomic spectrometric techniques. The review is divided into six sections, each devoted to a different element or elements, with speciation methods for As, Cr, Fe, Mn, Hg, Se and Tl discussed. Each section contains a summary table and is constructed on the one-paragraph-per-paper format. Each paragraph includes some information about reported LOD values and the validation of the method, so readers can draw their own conclusions about the suitability of a procedure for a particular analytical problem. Most of the sample matrices are environmental with an emphasis on waters.

Finally, some reviews are concerned with a particular analyte or matrix. A review of 49 articles (whose titles are provided) asks whether metal speciation analysis of petroleum is a myth or reality.²⁴ The review begins with a tutorial introduction that sets the scene, pointing out that although metals are minor constituents in crude oil, but occur in a great variety of complexed forms, they are important as they affect the refining and production operations depending on their chemical state. In particular, metals have significant detrimental impact on the conversion into transportation fuels (vanadium and nickel) and on metal constructions (mercury species), and serve as tracers in geochemical prospecting and oil maturity indicators. The article includes a table entitled ‘advanced ICP-MS-based procedures attempted to petroleum speciation analysis’ that summarises the contents of some 11 papers covering Co, Hg, Ni and V. They point out that further speciation information on the metal coordination in a variety of porphyrin compounds is available from FT ion cyclotron resonance MS and that Hg species may be quantified by GC-ICP-MS. The reviewers conclude that despite the confident progress of mainstream speciation analysis, most attempts at simultaneous determination of different metal species in petroleum samples have not been successful. They consider that the problems lie with sample preparation and separation procedures, which are unable to resolve individual species prior the detection.

2 CRMs and metrology

The development of a fruit juice CRM, certified for total As, iAs, DMA, Cd and Pb has been reported this year.²⁵ Total As and the As species were quantified by the standard additions approach whilst IDMS was used for determining the Cd and Pb content with the methodologies being performed according to ISO Guides 34 and 35. For the total elemental determinations the fruit juice was digested overnight at room temperature with concentrated HNO₃ before dilution. Surprisingly, the Cd and Pb spikes for IDMS were added after this digestion step so any analyte losses during digestion would not be accounted for by IDMS. For As speciation, approximately 1 g of fruit juice was diluted 10 fold with high purity water (HPW) before injection into a mobile phase comprising 10 mmol L⁻¹ sodium 1-butanesulfonate, 4 mmol L⁻¹ malonic acid, 4 mmol L⁻¹ TMAH, 0.05% methanol in HPW at pH 3.0 flowing at 0.5 mL min⁻¹ through a 150 mm C₁₈ column. Four As species, As^V, As^III, MMA and DMA were separated (in this order) in six minutes under these chromatographic conditions. The eluent was directly introduced to the nebuliser of an ICP-MS instrument operating in collision cell mode with He as the cell gas. The LOQ values for the As species ranged from 0.6 to 1.4 μg kg⁻¹. The certified mass fractions in the fruit juice CRM are Cd, 0.220 ± 0.011, Pb 0.245 ± 0.014, As 0.185 ± 0.015, iAs 0.124 ± 0.012 and DMA 0.0601 ± 0.0052 mg kg⁻¹.

Three federal and five state regulatory laboratories in the USA have participated in a multi-laboratory study to validate the AOAC First Action Official Method 2016.04, for the U.S. Food and Drug Administration’s method, EAM 4.10 HPLC-ICP-MS determination of four As species in fruit juice.²⁶ Briefly, the method comprises dilution of the samples followed by injection into a mobile phase of aqueous 10 mmol L⁻¹ dibasic ammonium phosphate at pH 8.25 and separation on a PRP-X100 column coupled to an ICP-MS instrument operated in collision cell mode with He as the cell gas. The full method is referenced in the paper. The method was validated for iAs measured as the sum of As^III and As^V, DMA, and MMA by analyses of 13 juice samples, including three apple juice, three apple juice concentrate, four grape juice, and three pear juice samples. In addition, two water CRMs were analysed. The method LODs and LOQs obtained among the eight laboratories were approximately 0.3 and 2 ng g⁻¹, respectively, for each of the analytes which were adequate for the intended purpose of the method. Each laboratory analysed method blanks, fortified method blanks, reference materials, triplicate portions of each juice sample, and duplicate fortified juice samples (one for each matrix type) at three fortification levels. In general, repeatability and reproducibility of the method was ≤15% RSD for each species present at a concentration >LOQ. The average recovery of fortified analytes for all laboratories ranged from 98–104% iAs, DMA, and MMA for all four juice sample matrixes. The average iAs results for CRMs 1640a and 1643e were within the range 96–98% of the certified values for total As. The authors conclude that the method has provided acceptable results at and below the level of interest (10 μg kg⁻¹) for iAs, DMA, and MMA in the matrixes studied.

3 Elemental speciation analysis

3.1 Antimony

Daus and Hansen, in their review (41 references with titles) of twenty years or so of the determination of Sb species, point out that although chemically similar to As, Sb speciation analysis present some specific challenges.²⁷ They discuss the complex chemistry of Sb^III and Sb^V, which plays a major role in chromatographic speciation of these species. They consider that, for simple matrices, such as surface or ground-waters, Sb redox speciation analysis is robust and reproducible provided that (a) a stable Sb^III complex (such as that with EDTA) is formed by incorporating the complex-forming agent in the eluent, (b) complex formation is complete (add EDTA to the samples before analysis), (c) a strong AEC is used, and (d) element-specific detection by ICP-MS or HG-AFS is employed. The reviewers point out that the ability of Sb to form rather stable complexes increases the risk of artefact formation during the extraction of solid samples.

No completely novel approaches for the determination of Sb species have been reported this year, although there have been a diverse range of applications in the literature. A study of Sb speciation in spirits stored in PET bottles has identified a novel Sb complex.²⁸ Total Sb, Sb^V and Sb^III were determined, using ICP-MS and HPLC-ICP-MS, in twelve spirit samples (Greek raki and tsipouro) stored in PET bottles. Total Sb concentrations ranged between 0.4 and 4 μg L⁻¹. Both inorganic species were observed with HPLC-ICP-MS together with an unknown Sb complex, which was found to be the predominant species in all samples analysed. The structure of this complex was investigated further using HPLC with HR ES-MS/MS and gave evidence for an acetaldehyde-bisulfite pyruvate Sb complex with the formula: C₇H₁₄O₁₂S₂Sb. In addition, the influence of high temperature storage conditions, with extended exposure times up to two weeks, on Sb migration from PET bottles into raki samples was investigated. The concentrations were in the range of 5.6 to 28 μg Sb per L spirit after a week of storage at 60 °C and Sb^V and Sb^III became the predominant species in comparison to the “novel” organic Sb complex. Elevated levels of Sb^V in drinking water samples stored in PET bottles when compared to glass were also reported by Marcinkowska et al.²⁹ The study considered the impact of storage container materials on the speciation on As, Cr and Sb using HPLC-ICP-MS. The highest levels of Sb^V recorded in the drinking water samples was 0.721 ± 0.040 μg L⁻¹ whilst Sb^III was not detected. Ion chromatography coupled with ICP-MS has been used to determine Sb in bottled waters and fruit juices.³⁰ Three Sb species, Sb^III, Sb^V and TMSb, were separated in less than 8 minutes on a Hamilton PRP-X100 column using two mobile phases: mobile phase A was 20 mmol L⁻¹ EDTA, 2 mmol L⁻¹ potassium hydrogen phthalate (KHP) in 1% v/v methanol (pH 5.5) and mobile phase B consisted of 20 mmol L⁻¹ EDTA, 2 mmol L⁻¹ KHP, 40 mmol (NH₄)₂CO₃ in 1% v/v methanol (pH 9.0). The LOD and RSD values were 0.012–0.032 ng mL⁻¹ and 2.2–2.8% respectively. In water samples, Sb^V was the major species, although two unidentified species were also recorded. In the juices, organometallic Sb species, tentatively thought to be Sb–citrate complexes were also present. The CRM material NRCC SLRS-3 riverine water was used for validation.

A method to investigate the uptake and translocation of TMSb in plant tissues has been reported by Mestrot et al.³¹ Ten mL of a solution of oxalic acid and ascorbic acid (200 and 100 mmol L⁻¹ respectively) was added to 100 mg of dried and ground plant material and shaken vigorously for 1 minute and then placed in an ultrasonic bath for 30 minutes. The extraction solution was separated and the supernatant diluted with a solution of 150 mmol L⁻¹ ammonium tartrate and stored in the fridge at 4 °C to prevent Sb^III oxidation. Using a citrus leaves CRM (NCS ZC73018 GSB-11) containing 200 ± 60 mg kg⁻¹ Sb, the extraction technique was found to remove 72.6 ± 1.3% of the Sb from the plant matrix. In a hydroponics experiment HPLC-ICP-MS was then used to assess the uptake of TMSb in ryegrass (Lolium perenne L.). The results show that TMSb and Sb^III are not converted to other species during extraction and that TMSb is taken up by ryegrass roots and translocated to the shoots, although the authors recognise that the method does not fully discriminate between Sb^III and Sb^V due to reduction during the extraction process. An unknown Sb species was found in the shoots of the TMSb-treated plants.

The surface chemistry and bulk chemical speciation of solid industrial wastes containing 8% Sb have been investigated using SR-XANES and TOF-MS.³² Leaching experiments (acid and base) were conducted over a 24 hour period in order to better understand the behaviour of Sb in waste streams. Surface analyses of the wastes before leaching showed the presence of Sb associated with S and O whilst bulk analyses revealed Sb to be present, primarily, as trivalent sulfide species. Both acid and base leaching did not change the Sb speciation on the solid. Leaching transferred about 1% of the total Sb into solution where Sb was found to be present as Sb^V. Data acquired using XANES showed similarities between leachate and FeSbO₄. During base leaching, the Sb content in solution gradually increased over time.

The speciation of Sb in water has been determined using magnetic core-modified silver nanoparticles and HR continuum source AAS.³³ The μSPE procedure for the separation and pre-concentration of Sb utilised magnetic particles coated with Ag nanoparticles functionalised with the Na salt of 2-mercaptoethane-sulfonate (about 3 mg for each experiment). After separation by means of a magnetic field, the solid phase was directly introduced into the ET-AAS for Sb determination. Alternatively, the solid was slurried and then injected into the atomiser. In both cases Pd nitrate was used as a chemical modifier. The pre-concentration factors achieved were 205 and 325, with detection limits of 0.02 and 0.03 μg L⁻¹ Sb, for the slurry and solid sampling procedures respectively. Speciation of Sb^III and Sb^V was achieved by means of two extractions carried out at different acidities (pH 2 and pH 9). The results for total Sb were verified using 5 different CRMs with recovery of the Sb ranging from 91–110%. A novel approach for the speciation of Sb by total TXRF based on the selective generation of stibine and trapping onto quartz reflectors containing immobilised Pd nanoparticles has been developed.³⁴ The Pd nanoparticles were synthesised using a water : ethanol mixture as the reducing agent and immobilizing onto quartz substrates following silanisation with 3-mercaptopropyltrimethoxysilane. For microextraction and pre-concentration of the Sb, the volatile hydrides were flushed onto the immobilised Pd nanoparticles for catalytic decomposition at ambient temperature. A factorial fractional design was used for screening the effect of the variables involved in the HG and pre-concentration steps. Trapping of Sb^III was accomplished in acidic medium (pH < 2) and the Sb^V was pre-reduced to Sb^III with a KI–ascorbic acid mixture. The LOD values were 0.50 and 0.80 μg L⁻¹ for Sb^III and Sb^V, respectively with an overall efficiency of 80%.

Table 1 shows other applications of Sb speciation presented in the literature during the time period covered by this ASU.

Table 1 Applications of speciation analysis: Sb

Analyte species	Technique	Matrix	Sample treatment	Separation	LOD	Validation	Reference
As^III, As^V, Sb^III, Sb^V	ET-AAS with tungsten chemical modifier	Standards	None	DLLME. APDC complexes of trivalent species into CHCl3. Reduction of pentavalent species with thiosulfate with	0.02–0.04 μg L^−1	None	35
Sb^III, Sb^V	HG-ICP-MS. ^118Sn internal standard, CH4 addition	Drinking water (bottled, tap and well)	Filtered, acidified	Selective HG. Addition of citrate prevented HG from Sb^V	4–6 ng L^−1	Spike recovery	36

---

3.2 Arsenic

The global interest in As speciation has continued this year with numerous publications reporting on a diverse range of applications. A number of reviews have been published, including two focusing on human exposure to dietary iAs and other As species. Researchers associated with the Collaborative on Food with Arsenic and associated Risk and Regulation (C-FARR), based at Dartmouth College (USA) have comprehensively reviewed (244 references with titles) human exposure to organic arsenic species from seafood.³⁷ The reviewers consider that in addition to inorganic As compounds, a full assessment of As in food, must account for organic As compounds in seafood, as this area is a crucial missing link in the understanding of exposure and informing regulatory practices. Much of the review is devoted to elaborating the present state of knowledge of these compounds and their toxicity and metabolism. There is a critical section devoted to chemical analysis, as the reviewers consider that shortcomings in evaluating exposure to organic As in seafood are largely due to analytical complications in reliably determining the complex distribution of As species in some of these samples. They conclude that the two most pressing needs for As speciation analysis of seafood are commercially available standards of As species and CRMs. It is of interest to note, therefore, that the synthesis and characterization of a number of methylated and thiolated arsenic species for environmental and health research has been extensively reviewed.³⁸ In the review, which cites 154 sources (titles provided), more than a dozen methylated and thiolated As compounds that are not commercially available are discussed, including the following; AB, an arsenicin-A model compound, DMA^III, DMA^V, dicysteinylmethyldithioarsenite, dimethylarsino-glutathione, dimethylmonothioarsinic acid, dimethyldithioarsinic acid, MMA^III, MMA^V, monomethylmonothioarsonic acid, monomethyldithioarsonic acid, monomethyltrithioarsonate and TMAO^V. The reviewers compared the available methods, synthesised the arsenic compounds in their laboratories, provide characterisation information, and recommend a synthesis procedure for each of the compounds. Methods for the determination of inorganic As in food, that have appeared in the period 2010 to 2015, have been examined in very considerable detail.³⁹ The review, based on 280 articles (titles given), summarises the state of the art of analytical methods and highlights the tools for the examination of quality assessment of the results, such as the production and evaluation of CRMs, and the availability of proficiency testing programmes. The reviewers point out that the need for robust and reliable analytical methods is recognised by various international safety and health agencies and by organisations in charge of establishing acceptable concentrations of inorganic As in food. They conclude that mild extraction of inorganic As species, with separation and detection by HPLC-ICP-MS detection, is the most popular approach. However, they do acknowledge that some non-chromatographic approaches are faster and more cost-effective. It is also noted that although numerous methods use CRMs to validate As measurements, few CRMs are certified for the iAs content or other As species. In particular, the differences reported between the concentration of iAs in seafood CRMs illustrate the difficulty in obtaining a consistent value and reinforce the need for further development of reliable methods. They conclude that (a) the production of seafood CRMs, and (b) more proficiency tests for the determination iAs in seafood are needed. A review of human exposure to dietary iAs and other As species also considers the role of analytical methodology.⁴⁰ The authors note that, despite classification as a human carcinogen based on data from populations exposed through contaminated drinking water, only recently has a need for regulatory limits for iAs in food been recognised. The reviewers consider the delay was due to the difficulty in assessing the risk associated with dietary iAs, which critically relies on speciation analysis. They also recognise the extensive use of HPLC-ICP-MS, though they do also mention AFS as a possible element-specific detection mode. They also state that because only quantification of iAs is needed from a food safety standpoint (which may not in fact be correct given current research) HG coupled with AAS, AFS or even ICP-MS offers a simple and cost-effective approach provided HG of methylated species is suppressed. Once again, the lack of suitable CRMs is highlighted, and, like the previous review,³⁹ a list of available CRMs that are certified for the As species is provided. The reviewers point out that if the mass balance, between the sum of As species measured and the total As concentration, is not in agreement then the speciation analysis should be viewed as questionable.

Although Fu et al. must have been writing their review⁴¹ of aromatic arsenical additives in soil at around the time (2015) that the US FDA was withdrawing approval (which they acknowledge) for the major players: roxarsone, arsanilic acid, carbarsone and nitarsone as animal feed additives, it is difficult to account for their description of these compounds as “safe and excellent feed additives. Widely used in swine, chicken, turkey, and other animal products in the United States”. The 150 articles cited (with titles) are mostly about the transformations of these compounds in the environment and fate of the resultant species, including the availability to plants. Limited space is devoted to aspects of chemical analysis: there is a brief section on extraction procedures, with more space given to the subsequent determination together with some discussion of ICP-MS isobaric overlap interferences. The limited number of SR X-ray studies are highlighted.

The vast majority of the publications on As speciation in this review period are applications of established approaches, often with modifications to suit a particular matrix. One of the few instrumental developments reported is a study of HG sample introduction combined with ICP-OES for non-chromatographic As speciation from solutions of As^III, As^V, DMA and MMA.⁴² Hydrides were generated by reaction with NaBH₄ in acidic medium using a continuous flow system directly coupled to the ICP. The different reactivities of the As species under various reaction conditions in terms of the type and concentration of acids (HCl, acetic and citric), buffers (acetate and citrate), the NaBH₄ concentration and presence of pre-reducing agents (KI–ascorbic acid, thiourea–ascorbic acid and L-cysteine) were optimised for selective generation of their individual hydrides. The relationship between As^III, As^V, DMA and MMA signals under different reduction conditions was also studied. In total, five speciation procedures for species-selective HG of As in one solution were evaluated. These were for As^III, for As^III + DMA, for As^III + As^V + MMA, for DMA + MMA and for As^III + As^V + DMA + MMA. Differentiation between the four As species was achieved.

The use of nanoparticles has been reported in a number of applications for As speciation. A novel magnetic metal–organic framework (MOF-199/dithiocarbamate modified magnetite nanoparticle composite) has been synthesised and used for the speciation of As^III and As^V by ET-AAS.⁴³ The synthesised sorbent offered selectivity toward As^III at pH 3 while As^V remained in solution. The total As in the samples was determined after reduction of the As^V species to As^III ions with a mixture of Na₂S₂O₃ and KI. Under the optimal conditions the LOD, linearity and RSD of the method for As^III were 1.2 ng L⁻¹, 4–300 ng L⁻¹ and <8.4%, respectively. The method was validated by analysing two NIST SRMs: a water, 1643e, and rice flour, 1568a. The method was then successfully employed to determine As^III and As^V in water samples and total As in rice and canned tuna samples. A method for As speciation using a nano-TiO₂ photocatalytic HG system has also been reported.⁴⁴ The nano-TiO₂ particles were used as a photocatalyst in the reduction of As species (As^III, As^V and DMA) in water, sediment and plant samples by UV irradiation in the presence of formic acid. Following after UAE with solutions of 1.0 mol L⁻¹ phosphoric acid HG-AAS was used for detection of the extracted As species. The LOD values for As^III and total As were 0.418 and 0.574 g g⁻¹, respectively, with an RSD of up to 8.6%.

The speciation of As in geological samples has been popular this year. An evaluation of techniques for sampling volatile As species (AsH₃, MeAsH₂, Me₂AsH, Me₃As) released from volcanoes has been reported.⁴⁵ Needle trap devices, cryotrapping, and Tedlar bags were each tested with respect to quantitative and species-preserving sampling. The needle trap devices did not trap AsH₃, MeAsH₂, Me₂AsH, and did not release sorbed Me₃As quantitatively. The use of these devices also led to artefact formation of dimethylchloroarsine. Cryotrapping in dry ice was insufficient to trap AsH₃ and MeAsH₂ whilst Me₂AsH and Me₃As were only partially retained. Sampling in Tedlar bags was found to be the best alternative. The stability of the four arsines was evaluated by storing in the dark at 5 °C for 19 days in a matrix of dry N₂, 11 days in 20% O₂, and 19 days in 3800 ppm CO₂. Better than 80% recovery was obtained for all species for all storage conditions. In the presence of H₂S, Me₃As recovery was only 67% and in the presence of SO₂, Me₂AsH and Me₃As recovery was 40 and 11%, respectively. The authors suggest that these results question the reliability of previous reports using SPE with the same sorption materials. Finally, removing interfering reactive gases using a NaOH trap, the authors sampled natural volcanic emissions at fumaroles of Vulcano and Solfatara (Italy). Detected total As concentrations of 0.5–77 ng m⁻³ were 1–2 orders of magnitude higher than the calculated background. The distribution and geochemical behaviour of As redox couples in hydrothermal waters from Bali and Java, Indonesia have been investigated.⁴⁶ Simultaneous speciation analysis of As and Sb was carried out using SF-ICP-MS coupled to HPLC. The concentration and distribution of As and Sb were closely related to host rock–water interaction and water types (Cl-rich, SO₄-rich and HCO₃⁻ rich), e.g. in HCO₃⁻ rich samples, As^V was the main form. Arsenic concentrations on Java were generally much higher (up to 9220 μg L⁻¹) than those from Bali (highest concentration <40 μg L⁻¹). In five of the samples an unknown As species was detected, and in two of these it was the dominant species, probably due to the influx of seawater. An experimental evaluation of sampling, storage and analytical protocols for measuring As speciation in sulfidic hot spring waters has been reported.⁴⁷ Thioarsenates are unstable and can degrade upon handling and storage. The effect of exposure to air on As and sulfur-enriched geothermal waters was therefore investigated and demonstrated a near to complete loss of thioarsenate species to As^III or As^V during short oxidation times. In contrast, thioarsenic standards were stable for the duration of analysis in spite of exposure to air. For samples containing thio-methylated As species, the extent of oxidation varied for different methylated As species. The study recommended flash freezing of samples in liquid nitrogen immediately after recovery and further storage under anaerobic conditions at −80 °C. A second experiment evaluating different HPLC columns for the separation of As thioanion species resulted in the preference for an IonPac column with NaOH as the mobile phase over the commonly used PEEK PRP-X100 anion exchange and Atlantis C₁₈ reverse phase column with ammonium phosphate mobile phases. However, distinct separation of thio-methylated As species with the IonPac column was not successful, potentially due to the matrix components. Acceptable detection, separation and quantification of thio-methylated As species were only achieved with the Atlantis C₁₈ column. The study demonstrated that preservation and analysis of these samples were matrix dependent, which has important implications for efforts to interpret As speciation in geothermal waters, especially those of low pH (2–3), low oxygen (≤49% saturation), low iron (≤5 mg L⁻¹) and high sulfur concentrations (≥91 mg L⁻¹).

Element substitution during fluid-rock interaction allows an assessment of fluid composition and interaction conditions in ancient geological systems, and provides a way to fix contaminants from aqueous solutions. Liu et al.⁴⁸ have conducted a series of hydrothermal mineral replacement experiments to determine whether a relationship can be established between As distribution in apatite and fluid chemistry. Calcite crystals were reacted with phosphate solutions spiked with As^V, As^III, and mixed As^III/As^V species at 250 °C and water-saturated pressure. Arsenic-bearing apatite rims formed in several hours, and within 48 h the calcite grains were fully replaced. Data acquired using XANES showed that As retained the trivalent oxidation state in the fully-reacted apatite grown from solutions containing only As^III. EXAFS data reveal that these As^III ions occupy tetrahedral phosphate sites. The use of μXANES imaging data revealed that apatite formed from solutions spiked with mixed As^III and As^V retained only As^V after completion of the replacement reaction; in contrast, partially reacted samples revealed a complex distribution of As^V : As^III ratios, with As^V concentrated in the centre of the grain and As^III towards the rim. Overall the study showed that the observed oxidation state of As in apatite may not reflect the original As^V : As^III ratio of the parent fluid, due to the complex nature of As^III uptake and possible in situ oxidation during recrystallisation.

The use of As^V : As^III ratios in barite as a geochemical proxy for paleo-environmental reconstruction has been reported.⁴⁹ Laboratory experiments showed that both As^III and As^V could be incorporated into barite as As^III- and As^V-co-precipitated barite, respectively, and could retain most of the information on the As^V : As^III ratio in coexistent water, if under equilibrium in terms of the redox reactions. The As^V : As^III ratios in natural barite were then collected in a hot spring and a hydrothermal vent using μXRF and μXAFS to evaluate the reliability of the As^V : As^III ratio in barite as a redox indicator in natural system and to estimate the depositional redox conditions in water for barite precipitation. The results indicated that one barite particle could provide information on the As^V : As^III ratio in water, and additionally whether the barite precipitated under oxic, suboxic, or anoxic redox conditions. The sorption of As to biogenic Fe (oxyhydr)oxides produced in circumneutral environments has been studied using sorption isotherm and kinetics experiments by Sowers et al.⁵⁰ Sorption of As^V and As^III to synthetic 2-line ferrihydrite and Fe biominerals harvested from the hyporheic zone of an uncontaminated creek was used and quantified using ICP-MS. XAS was utilised to obtain As and Fe K-edge spectra for As^V and As^III sorbed to environmentally collected and laboratory produced Fe^III minerals. All environmental Fe^III biominerals were determined to be structurally similar to 2-line ferrihydrite. Whereas the extent of sorption was similar for As^III on all minerals, As^V sorption to environmental Fe^III biominerals was approximately three times higher than what was observed for synthetic 2-line ferrihydrite. Structural modelling of EXAFS spectra revealed that the same surface complexation structure was formed by As^V and by As^III on environmental Fe^III biominerals and ferrihydrite. These results suggest that, despite similarities in binding mechanisms, Fe^III biominerals may be more reactive sorbents that synthetic surrogates often used to model environmental reactivity.

Concentrations of As in soil found in the vicinity of the former Zloty Stok gold mine in Poland are known to exceed 1000 μg g⁻¹ posing an inherent threat to neighbouring water bodies. A three month study to investigate the As mobilisation under reducing conditions and the capacity of autochthonic microflora that live on natural organic matter as the sole carbon/electron source has been reported.⁵¹ A biphasic mobilisation of As was observed. In the first two months, As mobilisation was mainly conferred by Mn dissolution despite the prevalence of Fe (0.1 wt% vs. 5.4 for Mn and Fe, respectively) as indicated by multiple regression analysis. Thereafter, the sudden increase in aqueous As^III (up to 2400 μg g⁻¹) was attributed to an almost quintupling of the autochthonic dissimilatory As-reducing community. The aqueous speciation influenced by microbial activity led to a reduction of solid phase As species (determined by XAFS) and a change in the elemental composition of As hotspots (μXRF mapping). The depletion of most natural dissolved organic matter and the fact that an extensive mobilisation of As^III occurred after two months, raises concerns about the long-term stability of historic As-contaminated sites.

Although the behaviour of As under reducing conditions in periods of high water levels in wetlands is well understood and documented, there is less information on what happens under oxidising conditions when the water level decreases. In a recent study, Guenet et al.⁵² looked at the first stage of the oxidising period, when oxidation products are still in suspension. A soil sample from the Naizin Kervidy wetland (France) was incubated in the laboratory to produce a reduced soil solution. The reduced solution was subsequently oxidised, filtered and ultrafiltered using decreasing pore size membranes (5 μm, 3 μm, 0.2 μm, 30 kDa and 5 kDa). The distribution of As and Fe was investigated in each size fraction of the oxidized solution and their speciation studied using XAS, HPLC and SEC-ICP-MS. Organic matter was characterized using thermally assisted hydrolysis and methylation GC-MS. The majority of the As was present as As^V but a small amount of As^III still remained despite the advanced oxidised conditions. In the >0.2 μm fractions, the XAS analyses showed that As was associated, in the second shell, with Fe (As–Fe – 3.35 angstrom) as bidentate binuclear complexes and C (As–C = 2.90 angstrom), suggesting the integration of As in biological objects. In the <30 kDa fraction, As was directly bound to C (As–C = 1.96 angstrom) in the first shell indicating the presence of organic As species. In the second shell, an As–Fe distance of 3.35 angstrom was found showing that part of the As was still complexed with Fe. The 0.2 μm to 30 kDa fraction was a transitional fraction in terms of the Fe species and organic matter composition. In this fraction, organic matter exhibited a more humic character (aromatic molecules) inducing an increasing cation binding capacity. As a consequence, in this fraction and in the smallest one, As, Fe and OM seemed to form ternary complexes in which the Fe or nano-oxides in the >30 kDa fraction and as monomer, or cluster in <30 kDa fraction acted as a bridge. In all of the fractions, a proportion of As^V was present as organic methylated species. Mass calculations provided evidence that 90% of the As was contained in the >5 μm particulate fraction and thus was hardly mobile. The study showed that although wetlands have been identified as a potential source of As, a number of biological and geochemical trapping mechanisms also favour As stabilisation.

The factors influencing As concentrations and species in mangrove surface sediments have been studied.⁵³ The sediments contained variable silt-clay (2–30% w/w), Fe (670–12 700 μg g⁻¹), Mn (1–115 μg g⁻¹), S (70–18 400 μg g⁻¹) and C (5–90 mg g⁻¹) concentrations. The As concentrations were in the range 1.6–8.6 μg g⁻¹ dry mass and correlated with the silt and clay content, and the Fe and Mn concentrations. This indicated that the silt-clay particles were covered with Fe and Mn (oxy) hydroxides and scavenged the As. The As extracted with 0.5 M phosphoric acid (68–95%) was present only as iAs (55–91%), indicating that other As species such as AB derived from marine animal tissues rapidly degrade in the sediments. The unextractable As was correlated with increases in OC, Fe and Mn content.

Arsenic speciation in groundwater used for drinking has been the focus of much research. The use of HR-CS-GF-AAS, combined with on-site separation and SPE has been developed for the speciation of iAs in geothermal and drinking water.⁵⁴ The HR-CS-GFAAS calibration curves were linear up to 200 μg L⁻¹ As, but using second order polynomial fitting, an accurate calibration could be performed up to 500 μg L⁻¹. It was found that the sample pH should not be higher than 8 for accurate speciation of As^V with a recovery of approximate to 95%. The geothermal water had reasonably high salt content (approximate to 2200 mg L⁻¹) due to the presence of chlorides and sulfates at mg L⁻¹ levels, and a two-fold dilution of these samples prior to SPE was recommended. Effects of natural organic matter on As adsorption has been investigated to evaluate the efficiency of modified granular natural siderite as an adsorbent for groundwater arsenic remediation.⁵⁵ Humic and fulvic acids were selected as model natural organic matter compounds. In batch tests, the humic and fulvic acids were either first adsorbed onto the modified granular natural siderite, or applied together with dissolved As. The kinetic data showed no significant effects for both adsorbed and dissolved humic and fulvic acids on As^III adsorption. However, As^V removal was increasing inhibited with increasing organic matter concentration. Fulvic acid exhibited higher inhibitory effect than humic acid at the same concentration. The use of steric exclusion chromatography and SEC-UV-ICP-MS revealed that As^V removal was mostly achieved by the oxyanion adsorption and adversely affected by dissolved fulvic acids via competitive adsorption for surface sites. In addition to oxyanion adsorption, removal of As^V was related to scavenging of ternary humic acid–As–Fe complexes. Novel cerium-loaded pumice (Ce-PU) and red scoria (Ce-RS) adsorbents have been developed to remove both As^III and As^V ions from water.⁵⁶ The Ce-PU and Ce-RS adsorbents were characterised using ICP-OES, EDX, and SEM. The experimental equilibrium sorption data fitted well with Freundlich and Dubinin–Radushkevich (D–R) isotherms. The adsorption was fast and reached an equilibrium within 2 hours. Both Ce-RS and Ce-PU showed high As^III and As^V removal efficiency over a pH range between 3 and 9, which is important for practical applications. The Ce-PU and Ce-RS adsorbents could be recycled and used up to three adsorption cycles without significant loss of their original efficiency.

A nanocomposite-coated microfluidic-based photocatalyst-assisted reduction device has been developed for As speciation when using HPLC-ICP-MS.⁵⁷ Gold nanoparticles were deposited on TiO₂ nanoparticles to enhance the conversion efficiency of the nanocomposite photocatalytic reduction. The sensitivity for As was significantly increased by employing the nanocomposite photocatalyst and using pre-reduction and signal-enhancement reagents. Under optimal operating conditions, the LOD values (3σ) of the online device for As^III and As^V were 0.23 and 0.34 μg L⁻¹, respectively. The results were validated using NIST SRM 1643e and used to determine iAs in both groundwater and fresh water samples. Amine-functionalised bimodal mesoporous silica nanoparticles immersed in ionic liquid as an extraction phase have also been used to determine iAs and total organic As in waters and human urine samples.⁵⁸ The above mixture was injected into 10 mL of sample containing As^III, As^V and organic As at pH 3.5. After UAE and centrifugation As^V was extracted with the nanoparticles. Total iAs was determined after oxidization of As^III to As^V by the addition of KMnO₄ solution in acidic medium and the As^III content was calculated by subtraction. The concentration of As^V was determined by ET-AAS after back extraction in separated phase by KOH solutions. Under the optimum conditions, the linear range, LOQ (3σ), precision (% RSD) and enrichment factor for As^V were calculated as 0.02–1.65 μg L⁻¹, 11 ng L⁻¹, 43% and 100 respectively. Recoveries were in the range of 95–103%. A novel analytical approach for the speciation of iAs by TXRF based on the selective generation of arsine by trapping onto quartz reflectors containing immobilised Pd nanoparticles has been developed.³⁴ The Pd nanoparticles were synthesised by chemical reduction under mild conditions using a water : ethanol mixture was used as the reducing agent, and then immobilised onto quartz substrates following silanisation with 3-mercaptopropyltrimethoxysilane. For microextraction and pre-concentration of iAs, volatile hydrides generated in a continuous flow system were flushed onto the immobilised Pd nanoparticles for catalytic decomposition at ambient temperature. A factorial fractional design was used for screening the effect of the variables involved in the HG and pre-concentration steps. Selective trapping of As^III was accomplished using citric acid/citrate buffer medium at pH 4.5. In addition, As^V was pre-reduced to As^III with a KI–ascorbic acid mixture. The LOD values were 0.09 and 0.10 μg L⁻¹ for As^III, As^V, respectively.

Household water sources in Pakistan have been the focus of a study of the potential health risks associated with chronic As exposure.⁵⁹ The study assessed 228 ground water sources supplying six villages for total, inorganic and organic As species using IC-ICP-MS. Samples were collected in duplicate in HDPE bottles (125 mL), and one sample was filtered and acidified on-site with concentrated HNO₃. A non-uniform distribution of As was found, with maximum values of 3430 pg L⁻¹ (median = 52) for As^V and 100 μg L⁻¹ (median = 0.37) for As^III. The levels exceeded the WHO provisional guideline value for As in drinking water (10 μg L⁻¹) in 89% of water sources and were the highest yet recorded for Pakistan. The maximum daily intake of total As for the 398 residents in the study was 237 μg per kg per day. This exposure estimate indicated that 63% of the rural residents exceeded the WHO provisional tolerable daily intake of 2.1 μg per kg per day body weight. Average daily intake of As^V was found to be 15.6 μg per kg per day for children ≤16 and 15 μg per kg per day for adults. A mean daily intake of 0.09 μg per kg per day was determined for As^III for children and 26 μg per kg per day for adults. Organic As species such as MMA, DMA and AB were found to be below the method LOD. A study to determine the total As and its speciation in saliva and urine samples collected from 70 people exposed to As in drinking water from an As-contaminated area of China has also been reported.⁶⁰ The results obtained using HPLC-ICP-MS showed that total As concentration in saliva were lower than in urine samples, but there was a strong positive correlation between total urinary As, drinking water As and different skin lesions. In the As metabolism study, As^III, As^V, MMA, and DMA were detected in all of the urine samples with the dominating species being DMA (73.2%). In contrast, most As species in saliva were not methylated. The major species in saliva were iAs (As^III + As^V, 76.2%), followed by DMA (13.1%) and MMA (9.13%). The primary methylation index (PMI), secondary methylation index (SMI) and proportion of the four different species (As^III, As^V, MMA, and DMA) in saliva showed no significant positive relationship with that of urine. The authors conclude that saliva may be used as a useful tool for biological monitoring of total As exposure rather than assessing As metabolism in the human body after being exposed to As.

The speciation of As in plants used in bioremediation studies has been the subject of several publications. The role of phosphorus in the metabolism of As^Vby a freshwater green alga, Chlorella vulgaris has been studied.⁶¹ The microalga was grown in the presence of varying phosphate concentrations (<10–500 μg L⁻¹ P) and environmentally realistic concentrations of As^V (5–50 μg L⁻¹ As). Arsenic speciation in the culture medium and total cellular As were measured using AEC-ICP-MS and ICP-DRC-MS, respectively, to determine As biotransformation and uptake in the various phosphorus scenarios. At high phosphate concentration in the culture medium, >100 μg L⁻¹ P, the uptake and biotransformation of As^V was minimal and DMA^V was the dominant metabolite excreted by C. vulgaris, albeit at relatively low concentrations. At common environmental P concentrations, 0–50 μg L⁻¹ P, the uptake and biotransformation of As^V increased. At these higher As-uptake levels, As^III was the predominant metabolite excreted from the cell. The concentrations of As^III in these low P conditions were much higher than the concentrations of methylated As observed at the various P concentrations studied. The switchover threshold between the (small) methylation and (large) reduction of As^V occurred around a cellular As concentration of 1 fg per cell. The observed near quantitative conversion of As^V to As^III under low phosphate conditions indicated the importance of As^V bio-reduction at common freshwater P concentrations. Nostoc sp. PCC 7120 (Nostoc), a filamentous cyanobacterium ubiquitous in aquatic system, is recognised as a model organism to study prokaryotic cell differentiation and nitrogen fixation. Nostoc cells were incubated with As^III for two weeks and then extracted with dichloromethane/methanol and water.⁶² The As species in the aqueous and dichloromethane layers were then determined using HPLC-ICP-MS/ES-MS. In addition to iAs, the aqueous layer also contained MMA^V, DMA^V, and two arsenosugars (a glycerol arsenosugar (Oxo-Gly) and a phosphate arsenosugar (Oxo-PO₄)). Two major arsenosugar phospholipids (AsSugPL982 and AsSugPL984) were detected in the DCM fraction. Arsenic in the growth medium was also investigated by HPLC-ICP-MS and shown to be present mainly as the inorganic forms As^III and As^V accounting for 29–38% and 29–57% of the total As respectively. The total As (methylated As, arsenosugars, and arsenosugar phospholipids) in Nostoc cells with increasing As^III exposure were not markedly different, indicating that the transformation to organoarsenic in Nostoc was not dependent on As^III concentration. A novel bioremediation system based on the symbiosis between leguminous plant and genetically engineered rhizobia has been tested and reported.⁶³ The As^IIIS-adenosylmethionine methyltransferase gene (CrarsM) from the alga Chlamydomonas reinhardtii was inserted into the chromosome of Rhizobium leguminosarum bv. trifolii strain R3. The As methylation ability of the recombinant rhizobium was tested under free living conditions and in symbiosis with red clover plants. Arsenic speciation was determined using HPLC-ICP-MS. Under free-living conditions, CrarsM-recombinant R. leguminosarum gained the ability to methylate As^III to methylated As species, including MMA^V, DMA^V and TMAO. Red clover plants were inoculated with either control (non-recombinant) or CrarsM-recombinant R. leguminosarum and exposed to 5 or 10 μM As^III. No methylated As species were detected in red clover plants inoculated with control R. leguminosarum. In contrast, all three methylated species were detected in both the nodules and the shoots when the recombinant rhizobium established symbiosis with red clover, accounting for 74.7–75.1% and 29.1–42.4% of the total As in the two plant tissues, respectively. The recombinant symbiont also volatilised small amounts of As. The study demonstrated that engineered rhizobia expressing an algal arsM gene can methylate and volatilise As, providing a proof of concept for potential future use of legume-rhizobia symbionts for As bioremediation.

Arsenic speciation in food and dietary studies has remained one of the key areas of research during this review period. The contamination of cereals with As is a global health and agronomic concern. The As uptake and As speciation in the grains and above ground tissues of 20 wheat cultivars exposed to 5 mg As per kg soil as either As^V or DMA under glasshouse conditions has been reported by Duncan et al.⁶⁴ The As species were extracted from grains using a 2% v/v HNO₃ extraction prior to determination of the As species using ICP-MS. Concentrations of both total and speciated As in the CRM ERM BC211 were in agreement with certified values. Germination rates for the majority of cultivars exceeded 80% when exposed to As^V, but fell to 20–40% when exposed to DMA. For a number of cultivars, grain yields were 20–50% lower when plants were exposed to DMA compared to As^V. Grain As concentrations were between 0.6 and 1.6 μg As per g grain across the twenty cultivars when exposed to As^V, whereas grain As concentrations were much higher (2.2–4.6 μg As per g grain) when exposed to DMA. When plants were exposed to As^V, 100% of the As present in the grain was found as iAs, while in plants exposed to DMA, 70–90% of As was present as DMA with the remainder found as iAs. The decreased germination rates and grain yields in the presence of DMA is similar to the symptoms described for straight head disease in rice, which has been linked to DMA exposure. The authors hypothesise that exposure to DMA may have inhibited Si-metabolism and translocation which resulted in both developmental impairment and possibly an increased susceptibility to soil pathogens. The determination of total iAs and As speciation in Mexican maize tortillas has been undertaken.⁶⁵ HG-MP-AES with Te^IV as the internal standard yielded total iAs results in good agreement with those provided by ICP-MS. In seven products from local markets, total iAs was found in the range of 21.8–192 μg As per kg. Analysis of As^III, As^V, MMA^V, and DMA^V using AEC-ICP-MS showed that iAs corresponded to 72.3–98.0% of total As. For total As by HG-MP-AES (with the internal standard) the LOQ was 47.6 μg As per kg, although when using SPE low recoveries (<30%) and an LOQ of 95 μg As per kg was obtained.

The determination of iAs and water-soluble As species in human milk by HPLC-ICP-MS has been reported by Stiboller et al.⁶⁶ A comparison of single and triple quadrupole mass analysers showed comparable performance, although the triple quadrupole instrument was more effective at dealing with the ArCl⁺ interference resulting from the natural chloride present in milk, without the need for gradient elution HPLC conditions. The method incorporated a protein precipitation step with trifluoroacetic acid followed by addition of dichloromethane or dibromomethane to remove the lipids. The aqueous phase was separated using anion-exchange and cation-exchange/mixed mode chromatography with aqueous ammonium bicarbonate and pyridine buffer solutions as mobile phases, respectively. For method validation, a human milk-sample was spiked with defined amounts of DMA, AB and As^V. The method showed good recoveries (99–103%) with an LOD (in milk) in the range of 10 ng As per kg. The method was further tested by analysing two Norwegian human milk samples where AB, DMA, and a currently unknown As species were found, but iAs was not detected. A slurry sampling procedure for As speciation in baby food by arsine generation, cryogenic trapping and detection with AAS has been presented.⁶⁷ Several procedures were evaluated for the preparation of the slurry, including different reagents (HNO₃, HCl and TMAH) and their concentrations, water bath heating and ultrasound-assisted agitation. The best results for iAs and DMA were obtained with 3 mol L⁻¹ HCl plus heating and ultrasound-assisted agitation. The developed method was applied for the analysis of five porridge powder and six baby meal samples. The trueness of the method was checked with the rice CRM ERM-BC211. Arsenic recoveries (mass balance) for all samples and CRM were performed by determining the total As by ICP-MS after microwave-assisted digestion and comparison against the sum of the results from the speciation analysis. The LODs were 0.44, 0.24 and 0.16 μg kg⁻¹ for iAs, MMA and DMA, respectively. The concentration of the more toxic iAs in the baby food samples ranged between 4.2 and 99 μg kg⁻¹ which were below the limits of 300, 200 and 100 μg kg⁻¹ set by Brazilian, Chinese and European legislation, respectively.

Arsenic speciation in vegetables (black radish, black salsify, lettuce, parsnip, savoy cabbage and swede turnip) grown in contaminated soils has been assessed using model pot experiments.⁶⁸ The soils used in the experiments originated from two mining and smelting sites in the Czech Republic with As levels reaching 36.0 ± 1.0 and 473 ± 10 mg As per kg, respectively. As^III, As^V, DMA, and MMA were determined by HPLC-ICP-MS. The concentration of As species determined in edible plant parts decreased in the following order: higher proportions of both DMA and MMA were found in the aboveground edible parts (leaves) compared to the underground parts (tubers). The results indicate that the distribution of As compounds differed predominantly according to individual plant species whereas almost no effect was observed due to the different soil properties. However, a higher As concentration in soils resulted in more As in the plant independently of the aboveground biomass (leaves) or the underground plant parts (tubers). An in vivo XANES measuring technique for studying the As uptake in cucumber plants has been reported.⁶⁹ The method employed a liquid nitrogen steam flow for cooling the samples and keeping them under cryogenic conditions during the measurement in order to preserve the original chemical states of As species in the hypocotyl. The aim of the study was to determine the As oxidation state in order to identify possible metabolic processes during nutrient uptake in cucumber (Cucumis sativus L. cv Joker) which was used as a model plant. The resistance level of the plants to As toxicity when different chemical forms of Fe (FeCl₃ and Fe-ascorbate) were added to the nutrient solution was also assessed. The ratio of As^III and As^V was determined in the hypocotyls of the cucumber samples. The cucumber plant was found to be able to develop biochemical resistance against As poisoning effect when it was treated with Fe-ascorbate added to the nutrient solution. A significant correlation was found between the ratio of concentration of As^III and As^V and the composition of the nutrient content, i.e. the chemical form of the Fe. The concentration of As species (AB, AC, DMA, MMA, As^Vand As^III) in edible mushrooms has been determined using HPLC-ICP-MS following microwave digestion and UEA.⁷⁰ Using an IC column (Dionex IonPac AS19) and 50 mmol L⁻¹ (NH₄)₂CO₃ (pH 9.7) as the mobile phase, the 6 species were separated within 20 minutes. The total As ranged from 0.04 to 212.3 mg kg⁻¹ in the test mushrooms. Under optimum conditions, the LOD were <2.5 μg kg⁻¹ for all six As species and the recoveries were >91.0% with an RSD of <5.5%. The uptake and transformation of As during the reproductive life stage of edible species of the Agaricus genus has been studied.⁷¹ Growth substrate and fungi were collected during the commercial growth of Agaricus bisporus and analysed for As speciation using HPLC-ICP-MS. AB was found to be the major As compound in the fungus at the earliest growth stage of fruiting (the primordium). The growth substrate mainly contained As^V. The distribution of As in an A. bisporus primordium grown on As^V treated substrate, and in a mature Agaricus campestris fruiting body collected from As contaminated mine tailings, was also mapped using two dimensional XAS imaging. The primordium and stalk of the mature fruiting body were both found to be growing around pockets of substrate material containing higher As concentrations, and AB was found exclusively in the fungal tissues. In the mature A. campestris the highest proportion of AB was found in the cap, supporting the idea that AB may be accumulated as an osmolyte (AB is structural similarity to glycine betaine).

The potential health risks associated with As speciation in the medicinal plant Panax Notoginseng have been assessed.⁷² The focus was on the presence of possible methylated As species, determined using HPLC-HG-AFS. The As^III methyltransferase (ArsM) gene abundance was determined using quantitative reverse transcription PCR (q-RTPCR). MMA and DMA accounted for 43 ± 30% of the total As in Panax Notoginseng from planting areas, while the primary species in soil was As^V (94 ± 0.12%). In the pot experiments, methylated As accounted for 37–49% of the total As in Panax Notoginseng, and As^V was the primary species in soil (>98%). The methylated As contents in the Panax Notoginseng root were positively correlated with the ArsM gene abundance in soil.

The speciation of As in livestock feed continues to attract interest. The determination of p-arsanilic acid (ASA), roxarsone (ROX) and nitarsone (NIT) in livestock feeds by HPLC-UV HG-AFS after microwave assisted extraction has been reported.⁷³ Chromatographic separation was achieved on a C₁₈ column with 2% acetic acid/methanol (96 : 4, v/v) as the mobile phase. The LODs were 0.13, 0.09 and 0.08 mg L⁻¹, and the LOQs were 0.44, 0.30 and 0.28 mg L⁻¹ with RSDs of 3.3, 5.3, and 5.4%, for ASA, ROX and NIT respectively. Microwave assisted extraction of phenylated As compounds using 1.5 M H₃PO₄ at 120 °C for 45 min allowed for 97% recoveries of total As and 95.2% for other organoarsenic species, with no degradation of these compounds. The results were validated using mass balance and the method was successfully applied to determine the presence of these compounds in feed samples. ASA was the only As species detected in chicken feed samples, with a concentration between 0.72 and 12.91 mg kg⁻¹. The oxidative transformation of roxarsone (3-nitro-4-hydroxyphenylarsonic acid) has also been investigated using electrochemistry coupled to HILIC-MS.⁷⁴ Although roxarsone is claimed to be relatively stable in organisms, relatively little is known about its metabolism. Using ESI-MS analysis potential transformation products were identified including toxic As^V. Thus, an oxidative C–As bond cleavage was observed. A HILIC separation coupled to ESI-MS enabled polarity estimation and the identification of isomeric products. Additionally, HILIC-ICP-MS facilitated speciation analysis and quantification of As containing products, revealing As^V to be the main species formed under the applied oxidative conditions. Two electrochemically generated species were also identified, one quinone and one quinone imine, indicating reactivity towards free thiol groups of the tripeptide glutathione and the proteins beta-lactoglobulin A and human serum albumin, which were used as model compounds for adduct formation. A study to evaluate As accumulation and speciation in rice grown in arsanilic acid-elevated paddy soil following the use of p-arsanilic acid as a feed additive for livestock and the subsequent use of contaminated animal manure as a fertiliser has been reported.⁷⁵ Five rice cultivars were grown in soil containing 100 mg As per kg soil. The total As concentration (measured by HG-AFS) of the hybrid rice cultivar was more than found in conventional rice cultivars for whole rice plant with the highest concentration of As found in roots. The p-arsanilic acid could be absorbed by rice, partly degraded and converted to As^III, MMA, DMA, and As^V as determined by HPLC-ICP-MS. The number of As species and their concentrations in each cultivar were related to their genotypes. The study showed that soil containing 100 mg As per kg or more is unsuitable for growing rice, and indicated that the disposal of animal manure containing arsanilic acid requires attention.

Following the trend in recent years, rice has remained a very popular matrix for As speciation studies. A study to determine iAs in husked rice using HG-AAS to meet the regulatory of the European Commission (EC) and Codex guidelines which limit iAs rather than total, organic, or individual As species such as As^III and As^V has been reported.⁷⁶ Dry rice samples (0.5 g) were digested and oxidised using 0.1 M HNO₃–3% H₂O₂ and heated in a water bath (90 ± 2 °C) for 60 min. The centrifuged extract was loaded onto a preconditioned and equilibrated strong anion-exchange SPE column (silica-based Strata SAX), followed by selective and sequential elution of As^V to enable the selective quantification of iAs. A recovery of 94% and a LOQ of 0.025 mg kg⁻¹ were achieved. The repeatability and reproducibility met the performance criteria mandated by the EC. Another ‘fit-for-purpose’ method to determine iAs has been reported by Gray et al.⁷⁷ In the method, As^III is intentionally oxidised to As^V with H₂O₂ during sample preparation, converting all iAs to the As^V form. Arsenic species were separated in less than 2 minutes using a 50 mm × 5 μm Hamilton PRP-X100 AEC column with a mobile phase was 40 mmol L⁻¹ (NH₄)₂CO₃ with 3% v/v methanol at pH 9.0. This analysis time is 10 times faster than the current FDA regulatory method. The use of O₂ reaction gas with ICP-QQQ with MS/MS capability avoided spectral interferences and dramatically increased sensitivity, allowing for low volume injections. The small injection volume (5 μL) and modified mobile phase composition mitigate non-spectral interferences such as carbon enhanced ionization. Furthermore, the shortened analysis time significantly increases sample throughput. Validation using CRMs and data from two laboratories demonstrate the method’s accuracy and reproducibility for both rice and wine matrices in a single analytical batch.

The measurement of As species in rice by HPLC-ICP-MS after extraction with sub-critical water and H₂O₂ has also been presented by Maher et al.⁷⁸ Samples (0.2 g dry mass) were extracted with 9 mL of deionised water and 1 mL of 30% v/v H₂O₂ at 200 °C at 25 bar using microwave heating. Converting the As^III to As^V made quantification of the iAs easier using HPLC-ICP-MS by ensuring that the As^V eluted well after the solvent front and distinct from the DMA peak. The system employed a PEEK PRP-X100 AEC (250 mm, 4.6 mm, 10 mm) using 20 mmol L⁻¹ ammonium phosphate buffer mobile phase at pH 4.5, and a flow rate 1.5 mL min⁻¹, column temperature of 40 °C and injection volume of 40 mL. Recoveries of As in five rice CRMs and eleven rice samples were 100 ± 6% (n = 50) and 103 ± 14% (n = 97) respectively. Comparison of the measurements of total As in extracts by ICP-MS and ETAAS (used for total As measurements) were also equivalent, indicating that organic carbon from the rice extracts did not cause an enhancement of the ICP-MS signals. There was no significant difference or bias in As speciation between the water–H₂O₂ and 2% v/v HNO₃ procedures. The reproducibility for the rice samples containing As concentrations at legislative limits were: total As (0.192 ± 0.004 μg g⁻¹), iAs (0.108 ± 0.002 μg g⁻¹) and DMA (0.085 ± 0.003 μg g⁻¹). A method to determine iAs in rice using online anion suppression with IC and ICP-MS has also been reported.⁷⁹ The optimal extracting agent for As^III, ASV, DMA and MMA was found to be HCl of 0.01 mol L⁻¹. A mixture of 38 mmol L⁻¹ sodium carbonate and 15 mmol L⁻¹ sodium acetate were used as the mobile phase. The results showed that there were no significant losses or transformations with the anion suppressor and an improvement in sensitivity. The LOQ were 0.1 μg L⁻¹ for DMA, As^III and MMA, and 0.2 μg L⁻¹ for As^V. As^III and DMA were the primary forms present in the rice sample with recoveries ranging from 99.76 to 110.42%. A monitoring method using HPLC-ICP-MS following heat-assisted extraction has been developed for rapid measurement of total iAs species in rice flour.⁸⁰ The As^III and As^V eluted at the same retention time and were completely separated from organoarsenic species when using an isocratic elution program on a reversed phase column. Neither ambiguous oxidation of As^III to As^V nor the integration of two peaks were necessary to directly determine directly the target analyte iAs. Rapid injection allowed measuring 3 replicates within 6 min and this combined with a quantitative extraction of all As species from the rice flour in 15 min using a HNO₃–H₂O₂ extraction makes this one of the fastest laboratory based methods for iAs in rice flour yet reported.

Total As, iAs, and DMA were determined in 37 commercial rice samples collected in France in order to evaluate the toxicological risk for different population groups as a function of age and gender, when employing different cooking methods.⁸¹ Total As was measured by ICP-MS and AEC-ICP-MS was used for iAs and DMA determinations. The two CRMs TORT 2 (lobster hepatopancreas), and BC 211 (rice powder), were used for method validation. The total As in the raw rice samples varied from 0.041 to 0.535 mg kg⁻¹, whereas iAs varied from 0.025 mg kg⁻¹ (polished basmati rice) up to 0.471 mg kg⁻¹ (organic rice duo). The intake varied between 0.002 and 0.184 μg kg⁻¹ body weight for total As and 0.002–0.153 μg kg⁻¹ body weight for iAs; levels which do not pose a chronic toxicity risk. Organic wholegrain rice may present a risk for children in the case of sole consumption at the expense of polished rice. Pre-rinsing and boiling the raw rice using an excess of water was the most efficient mode to obtain significant iAs removal, particularly for white rice varieties. In Brazil, potential differences in As speciation in organically and traditionally cultivated rice have been studied.⁸² Samples of polished and husked rice (organic and conventional) and gastronomic rice (Arborio, Carnaroli and red/black rice) were analysed using ICP-MS and HPLC-ICP-MS and the results compared to FAO/Codex maximum limits. The results showed no significant statistical differences in total As concentration in organic rice (157.7 ± 56.1 ng g⁻¹) vs. conventional rice (137.4 ± 46.6 ng g⁻¹) and also in organic husked rice (227.7 ± 95.5 ng g⁻¹) vs. conventional husked (217.7 ± 60.9 ng g⁻¹). However, iAs was 45% higher in organic polished rice than in conventional polished rice and 41% higher in organic husked rice than in conventional husked rice. Gastronomic rice presented total As ranging from 65.4 to 348 ng g⁻¹ for black and arborio rice, respectively. All results were within the maximum levels adopted by Codex for iAs (200 ng g⁻¹). Different types of rice from two areas in Korea have also been studied to estimate the potential health risk from polished rice.⁸³ Results HPLC-ICP/MS showed that brown rice (no polishing) contained the highest amount of total As followed by 5, 7 and 10% polished white rice. Among the As species, As^III was predominantly detected in brown rice with concentrations ranging from 28.51 ± 0.71–51.91 ± 1.13 μg kg⁻¹ in region A and from 62.1 to 130.4 μg kg⁻¹ in region B. When estimating the daily As exposure from consumption of polished rice, the expected daily exposure of iAs from brown and white rice was found to be below benchmark value of 3.0 μg kg⁻¹ body weight per day. A study to look at the micro-distribution and speciation of As in human hair and rice grain samples collected in one of the villages in Guizhou, China, identified to have endemic arseniasis caused by exposure to indoor combustion of high As-content coal has been reported.⁸⁴ Analyses were performed by micro-beam XRF and XAFS. The total As level in hair samples of diagnosed patients was at almost the same level as in their asymptomatic neighbours. Concentrations in the lateral cut of hair samples were high-low-high (from surface to centre). XAFS revealed the coexistence of both the As^III and As^V states in hair samples. However, the samples from diagnosed patients displayed a tendency of higher As^III/As^V ratio than the asymptomatic fellow villagers. The μXRF mapping of rice grains showed that As penetrated the endosperm, the major edible part of the grain, when rice grains were stored over the open fire burning high As-content coal. Thus rinsing the rice grains with water to remove the As was largely ineffective.

There have been fewer studies looking at As in seafood this year. However, IC-UV-HG-AFS has been used to detect five As species (AB, DMA, MMA, As^V and As^III) extracted from seafoods samples using 2% formic acid.⁸⁵ A gradient elution using 33 mmol L⁻¹ CH₃COONH₄ and 15 mmol L⁻¹ Na₂CO₃ with 10 mL CH₃CH₂OH at pH 8.4 facilitated the chromatographic separation of all species on a Hamilton PRP-X100 AEC in less than 8 min. An ultrasound extraction method was used to extract the As species from seafood recoveries from spiked samples in the range of 72.6–109%. The precision between sample replicates was better than 3.6% for all determinations. LODs were 3.543 μg L⁻¹ for AB, 0.426 μg L⁻¹ for As^III, 0.216 μg L⁻¹ for DMA, 0.211 μg L⁻¹ for MMA, and 0.709 μg L⁻¹ for As^V. The linear coefficients were greater than 0.999. The method was applied to the determination of As species in bonito, Euphausia superba, and Enteromorpha with satisfactory results. A method for the extraction and determination of total As and As species in seafood by IC axial ICP-AES has also been reported.⁸⁶ Sample preparation included a selective extraction method, and optimization of the separation and detection parameters. A LOD for As of 4.7 ng mL⁻¹ was obtained and good recovery efficiency (98.2–99.1%) for all As species (As^III, As^V, MMA, DMA, and AB). Several kinds of seafood (snow crab, red snow crab, octopus, octopus minor, and squid) and seaweed (red laver, green laver, sea tangle, sea mustard, and hizikia) were investigated by drying and grinding the samples to fine powder before use. An oyster tissue CRM was also evaluated. The main As species found in these seafoods was AB with the highest concentration found in red snow crab samples (152 ± 4 mg kg⁻¹, dry weight). Other As species were not detected in these seafood samples. The major As species found in hizikia samples was As^V with a concentration of 34.0 ± 0.4 mg kg⁻¹ (dry weight). No As was found in green laver, sea tangle, red laver, or sea mustard samples. Arsenic species have been determined by HPLC-ICP-MS in 38 species of tropical marine fish harvested from the Spratly Islands, China following a microwave digestion preparation step.⁸⁷ The average level of total As was 20.845 mg kg⁻¹. Further analysis indicated that AB (8.560–31.020 mg kg⁻¹) was the dominant As species in the fish samples and accounted for 31.48% to 47.24% of the total As. As^III and As^V were detected at low concentrations, indicating minimal arsenic toxicity.

Arsenic speciation in seafood after culinary treatments has been performed.⁸⁸ AB, As^III, DMA, MMA and As^V species were determined LC-ICP-MS/MS using O₂ as the reaction gas for the conversion of As⁷⁵ to AsO (As⁷⁵O¹⁶). The influence of the culinary treatments (boiling, frying and sautéing) with or without the addition of spices (salt, lemon juice and garlic) on the As species in blacktip shark and Asian tiger shrimp were investigated. Arsenic species were extracted by using a 30 mmol L⁻¹ HNO₃ solution. Ammonium phosphate (10 mmol L⁻¹) was used as the mobile phase and the influence of pH and the addition of 1% (v/v) methanol were also investigated. Oil, water, butter and the spices used during cooking were analysed to perform a mass balance of the total As. The method was validated using the CRM DORM-3, and the accuracy evaluated by statistical comparison between the certified value and the total As concentration determined by ICP-MS after acid digestion. It was found that the culinary treatments did not influence the stability of As species in uncooked seafood. On the other hand, significant analyte losses (15–45%) were observed for boiled seafood. The LOQ were 4, 21, 4, 9 and 18 ng g⁻¹ for AB, As^III, DMA, MMA and As^V, respectively. Arsenic speciation in seaweeds continues to be of interest since seaweed can accumulate iAs from seawater as HAsO₄²⁻ in place of the phosphate anion (HPO₄²⁻). Although it is rapidly metabolised to organoarsenic species, predominantly arsenosugars and arsenolipids, any residual iAs present in the seaweed may pose a potential health threat to consumers of seaweed products. The distribution of total As and iAs has been determined in thallus parts of the kelp Laminaria digitata and the intertidal fucoid Ascophyllum nodosum (both Phaeophyceae) using ICP-MS and HPLC-ICP-MS.⁸⁹ The total As ranged from 36 to 131 mg kg⁻¹ (dw) in L. digitata, and from 38 to 111 mg kg⁻¹ dw in A. nodosum, with no statistically significant differences between different thallus parts. Inorganic As was detected in all A. nodosum samples, comprising less than 1% of the total As content. Concentrations of iAs in L. digitata were significantly higher, ranging from 2.2 to 87 mg kg⁻¹, increasing through the thallus from the stipe to the decaying distal blades. The iAs comprised more than 50% of total As in the middle of decaying distal blades. The authors highlight the potential implications for harvesting, processing and use of Laminaria digitata in agri-, food and health applications. Inorganic As in 13 store-bought edible seaweed samples and 34 dried kelp (Laminaria digitata) samples has been determined using a new field-deployable method for As in water.⁹⁰ Results from the method compared well to results from speciation analysis using HPLC-ICP-MS (slope = 1.03, R² = 0.70). The field-deployable method consisted of a simple extraction method using diluted HNO₃ to quantitatively extract iAs without decomposing the organoarsenicals to iAs. This was followed by the selective volatilisation of iAs as AsH₃ and subsequent chemo-trapping on a filter paper soaked in HgBr₂ solution. Method optimization with a sub-set of samples showed 80–94% iAs recovery with the field-deployable method with no matrix effect from organo-As species in the form of DMA on the iAs concentration. The method provided good reproducibility with an average error of ±19%. The LOQ was around 0.05 mg kg⁻¹ (dw). The accumulation of AB and thio-arsenoribosides have been measured in common macroalgae species (8 phaeophyta, 4 rhodophyta and 2 chlorophyta) along the Australian south east coast line.⁹¹ The As species profiles were also determined in two common marine herbivores, the sea urchin Centrostephanus rodgersii and the fish Odax cyanomelas that graze on these macroalgae. Two extraction methods were employed (acetone and methanol–water) prior to As speciation by HPLC-ICP-MS. AB was found in seven of the fourteen macroalgae species investigated, but it did not contribute significantly to any of the macroalgae As content (0.01–1.2 μg g⁻¹). AB was found in only two of the brown macroalgae and all the red and green macroalgae (with the exception of Corallina officinalis). Thio-As species were found sporadically, but not in high concentrations in any of the macroalgae investigated. The authors report that the AB present in macroalgae is likely to be associated with epiphytic organisms while thio-arsenoribosides are likely to be produced by decaying parts of damaged macroalgae. A laboratory feeding experiment in which the herbivorous gastropod, Austrocochlea constricta, was fed macroalgae containing thio-arsenoribosides for a 24 h period every three days showed that these are readily accumulated over a short period. The thio-arsenoribosides in herbivores are therefore probably obtained through trophic transfer. Some AB was also obtained through trophic transfer; however, the presence of trimethylated arsonioribosides, a hypothesised precursor of AB formation in herbivores, suggested that some AB is produced within herbivores from the transformation of arsenoribosides accumulated from their diet.

The determination of arsenosugars is often difficult due to a lack of appropriate calibration standards. A paper by Yu et al.⁹² has reported on the determination of As species in a kelp extract with traceability to SI units. The hydrophilic fraction of the As species were extracted from a candidate RM (SRM 3232 Kelp Powder, Thallus Laminariae) during the study at NIST. Arsenosugars and DMA were separated into fractions using analytical LC cation and anion columns. The As in the fractions was determined using NAA. Cation exchange separation was used for the determination of arsenosugar 3-[5′-deoxy-5′-(dimethylarsinoyl)-b-ribofuranosyloxy] propylene glycol (As(328)) for the first time, while DMA and arsenosugars 3-92-2-hydroxypropyl 2,3-dihydroxypropyl hydrogen phosphate (As(482)) and 3-92-2-hydroxypropanesulfonic acid (As(392)) were determined following anion exchange separation. The As species in the kelp extract, including DMA, As(328), and iAs were determined using LC-ICP-MS. The results of DMA and As(328) were 0.024–0.485 mg kg⁻¹ and 0.03–1.14 mg kg⁻¹, respectively, which were in good agreement with those determined by NAA in fractions of the LC eluent. The most toxic species, As^III and As^V, were found to be <0.07 mg kg⁻¹ and 0.018–0.231 mg kg⁻¹, respectively.

Several dietary exposure studies focusing on As speciation have been reported. Dietary exposure of the Italian population to iAs has been assessed in the national Total Diet Study carried out in 2012–2014.⁹³ Over 3000 food samples were collected to model the Italian diet of the whole population, prepared as consumed, and pooled into 51 food groups. The iAs was determined by HPLC-ICP-MS after chemical extraction and quantified in all samples. Occurrence data were combined with national individual consumption data to estimate mean and high level dietary exposure of the general population and of population subgroups according to age and gender, both at the national level and for each of the four main geographical areas of Italy. The intakes assessed are in the lower range of iAs exposure estimates in other European countries carried out without the support of the Total Diet Study approach. However, taking the lower limit of the BMDL01 range established by the EFSA as a reference point, the margins of exposure are <2 for the mean intake in infants and toddlers and <1 for the 95th percentile intakes in all younger age groups, indicating a need to further reduce the dietary exposure to iAs. To evaluate whether differences in dietary intake of selected micronutrients are associated with the metabolism of iAs, the intake of 21 micronutrients was estimated for 1027 women living in northern Mexico using a food frequency questionnaire.⁹⁴ Concentration of urinary metabolites of iAs were determined by HPLC-ICP-MS and the proportion of iAs metabolites was calculated (% iAs, % MMA and % DMA), as well as ratios for MMA/iAs, DMA/MMA and DMA/iAs. After adjustment for covariates, it was found that methionine, choline, folate, vitamin B12, Zn, Se and vitamin C favour elimination of iAs mainly by decreasing the % MMA and/or increasing % DMA in urine, and confirming that diet contributes to the elimination of iAs. The relation of dietary iAs exposure and urinary iAs metabolites excretion has also been studied in Japanese subjects.⁹⁵ A set of 24 h duplicated diet and spot urine samples from 20 male and 19 female subjects were used. The concentrations of As species were determined by LC-HG-ICP-MS. The sum of the concentrations of urinary iAs and MMA were used as a measure of iAs exposure. Daily dietary iAs exposure was estimated to be 0.087 μg per kg per day. Analysis of covariance did not find a gender difference in regression coefficients (P > 0.05). The regression equation log¹⁰ (urinary iAs + MMA concentration) = 0.570 × log¹⁰ (dietary iAs exposure level per body weight) + 1.15 was obtained for whole data set. The equation was used in converting urinary iAs concentration to daily iAs exposure in the risk assessment.

Few publications this year have looked at occupational exposure to As. The preservation of museum objects with iAs compounds and contamination of the surroundings has previously been documented. However, in a recent study this has been extended to assess the exposure of museum staff by measuring arsenicals in urine.⁹⁶ After 1 week without exposure, urinary samples were taken before and after handling of preserved skins and analysed by HPLC-ICP-MS for iAs, As metabolites and AB. The sum of iAs and metabolites was used as an index of exposure. Information about work and seafood intake was obtained by questionnaire. One out of five subjects had a work-related rise in the exposure index of 18.1 μg As per L to a post-exposure level of 37.1 μg As per L. Four subjects had no certain exposure-related increase in the index. The study indicates that museum staff may be exposed to As from handling As-preserved objects.

Arsenic speciation as once again been reported in a range of clinical studies. Inefficient As methylation capacity has been associated with developmental delay in children. A study by Hsieh et al.⁹⁷ designed to explore whether polymorphisms and haplotypes of As methyltransferase (AS3MT), glutathione-S-transferase omegas (GSTOs), and purine nucleoside phosphorylase (PNP) affect As methylation capacity and developmental delay has been reported. In total, 179 children with developmental delay and 88 children without delay were recruited. Urinary As species, including As^III, As^V, MMA^V, and DMA^V were measured using HPLC-HG-AAS. The polymorphisms of AS3MT, GSTO, and PNP were performed using a Sequenom MassARRAY genotyping platform. Polymorphisms of AS3MT genes were found to affect susceptibility to developmental delay in children, but GSTO and PNP polymorphisms were not. The data provide evidence that AS3MT genes are related to developmental delay and may partially influence As methylation capacity. A similar paper by the same group, although with a focus on health status, has also been published.⁹⁸ Developmental delays were associated with the total urinary As concentration, iAs and MMA^V, and negatively associated with DMA^V. MMA^V was negatively associated with the health-related quality of life and functional performance, whereas DMA^V was positively associated with health-related quality of life and functional performance. A further publication by this group explored the relationship between X-ray repair cross-complementing group 1 (XRCC1) gene polymorphisms and renal cell carcinoma (RCC) and to investigate whether individuals with an XRCC1 risk genotype, a high level of 8-OHdG or a high urinary total As concentration have a modified odds ratio (OR) of RCC.⁹⁹ The study was based on 180 RCC patients and 360 age and sex matched controls from a hospital-based pool. Image guided biopsy or surgical resection of renal tumours was performed on RCC patients for pathological verification. Genomic DNA was used to examine the genotype of XRCC1 (Arg399Gln), XRCC1 (Arg194Trp), XRCC3 (Thr241Met) and XPD (Lys751Gln) by PCR-RFLP. LC-MS was used to determine urinary 8-OHdG levels and HPLC-HG-AAS was used to determine the concentrations of urinary As species. Participants with the genotype XRCC1 (Arg194Trp) Arg/Trp + Trp/Trp had a significantly higher OR of RCC than those with the Arg/Arg genotype; the OR and 95% confidence interval was 0.66 (0.45–0.97) after multivariate adjustment. The OR of RCC for the combined effect of high urinary 8-OHdG levels and high urinary total As in individuals with a XRCC1 (Arg194Trp) Arg/Trp + Trp/Trp genotype was higher than in patients with an Arg/Arg genotype, which was evident in a dose response manner. The authors state that this is the first study to show that the XRCC1 Arg194 allele is a predicting factor for RCC. The more risk factors (high urinary 8-OHdG levels, high urinary total As concentrations, and XRCC1 Arg194 allele) that were present, the higher the OR of RCC. Although chronic As exposure is known to decreases adult and children’s ability to methylate iAs, few studies have examined children’s sex differences. Torres-Sanchez et al. have measured urinary concentrations of iAs, MMA, and DMA, and calculated the primary MMA/iAs and secondary DMA/MMA methylation capacity indexes in 591 children aged 6–8 years in Mexico.¹⁰⁰ Each species was determined using HG cryotrapping AAS and lineal regression models used to estimate associations between methylation capacity and total As or iAs. Interactions with sex were tested at p < 0.10. Boys had significantly higher total As levels, (58.4 μg L⁻¹) than girls (46.2 μg L⁻¹). The authors observed negative associations between total As and primary MMA/iAs (β = −0.039; p < 0.18) and secondary DMA/MMA (β = −0.08; p = 0.002) with significant sex differences; primary MMA/iAs reduction was significant in boys (β = −0.09; p = 0.02) but not in girls (β = 0.021; p = 0.63), p for interaction = 0.06. In contrast, secondary DMA/MMA reduction was significantly more pronounced in girls. Furthermore, negative associations primary MMA/iAs (β = −0.19; p < 0.001) and secondary DMA/MMA (β = −0.35; p < 0.001) were a function of urinary iAs levels, independently of total As; however, the reduction in primary MMA/iAs was more pronounced in boys (β = −024; p < 0.001); girls (β = −0.15; p < 0.001), p for interaction = 0.04. A significant negative association was observed between secondary DMA/MMA and iAs levels without significant sex differences. Total As and iAs associations with metabolite percentages were in good agreement with those observed with methylation indexes. The results suggested that iAs plays an important role in reducing As methylation ability and that significant sex differences are present in As metabolism.

Several publications have focused on urinary As speciation in pregnant women. Exposure to iAs is of particular concern among pregnant women, infants and children, as they are specifically vulnerable to the adverse health effects of iAs, and in utero and early-life exposure. Even low to moderate levels of iAs, may have a marked effect throughout their lifespan. In a study of Spanish women by Signes-Pastor et al.,¹⁰¹ IC-ICP-MS was used to analyse urinary As speciation as a exposure biomarker. The profile of the As metabolites were determined in samples taken during pregnancy (1st trimester) and in 4 year old children from the same region. The median of the main As species found in the children was 9.71 μg L⁻¹ (AB), 3.97 μg L⁻¹ (DMA), 0.44 μg L⁻¹ (MMA) and 0.35 μg L⁻¹ (iAs). Statistically significant differences were found in urinary AB, MMA and iAs according to the region studied, and also in DMA among pregnant women and their children. Spearman’s correlation coefficient among urinary As metabolites was calculated, and, in general, a strong methylation capacity to methylate iAs to MMA was observed. Another study to investigate the urinary metabolic changes of pregnant women exposed to low-dose As, and to identify biomarkers from metabolomics analysis has been published on work conducted in China.¹⁰² Urine samples of 246 pregnant women were collected in the first trimester of pregnancy and divided into three groups based on the tertile distribution of urinary As concentrations determined ICP-MS and UPLC-Q-TOF-MS. Arsenic-related metabolic biomarkers were investigated by comparing the samples of the first and third tertiles of As exposure classifications using a partial least-squares discriminant model. Nine urine potential biomarkers were identified, including LysoPC (14 : 0), glutathione, 18-carboxydinor-LTE4, 20-COOH-LTE4, cystathionine ketimin, 1-(beta-n-ribofuranosyl)-1,4-dihydronicotinamide, thiocysteine, p-cresol glucuronide and vanillactic acid. The results indicated that environmental As exposure, even at low levels, can cause metabolite alterations in pregnant women which might be associated with adverse health outcomes. NMR-based metabolomic analysis has been used to identify metabolites in neonate cord serum associated with, prenatal iAs exposure in participants taking part in the Biomarkers of Exposure to ARsenic (BEAR) project in Gomez Palacio, Mexico.¹⁰³ Using multivariable linear regression, ten cord serum metabolites were identified as significantly associated with total urinary iAs and/or iAs metabolites, measured as % iAs, % MMAs, and % DMAs. A total of 17 metabolites were identified as significantly associated with total iAs and/or iAs metabolites in cord serum. These metabolites were indicative of changes in important biochemical pathways such as-vitamin metabolism, the citric acid (TCA) cycle, and amino acid metabolism.

Genetic variants associated with cardiometabolic traits or As metabolism biomarkers have been studied in the so called “Strong Heart Family Study”.¹⁰⁴ Using 2428 American Indian participants, urine As species (iAs, MMA, and DMA) were measured by HPLC-ICP-MS. PCA and linear regression models were used and adjusted for age, sex, total As levels, and population stratification. Single nucleotide polymorphism associations were stratified by study site and were meta-analysed. Variants at 10q24 were statistically significant for all As species and principal components of As species. The index single nucleotide polymorphism for iAs%, MMA%, and DMA were located near AS3MT, whose gene product catalyses methylation of iAs to MMA and DMA. Among the candidate As variant associations, functional single nucleotide polymorphism in AS3MT and 10q24 were most significant (p < 9.33 × 10⁻⁵). This hypothesis-driven association study supported the role of common variants in As metabolism, particularly AS3MT and 10q24. A study to elucidate the associations between As exposure, gene expression, and DNA methylation in peripheral blood has been reported.¹⁰⁵ The study used women from the Andes, Argentina, who were exposed to As via drinking water. The As exposure was assessed as the sum of As metabolites in urine using HPLC-HG-ICP-MS, and the As metabolism efficiency was assessed by the urinary fractions (%) of the individual metabolites. The genome-wide gene expression (N = 80 women) and DNA methylation (N = 93; 80 overlapping with gene expression) in peripheral blood were measured using Illumina DirectHyb HumanHT-12 v4.0 and Infinium Human-Methylation 450K BeadChip, respectively. The results showed concentrations of As metabolites in urine, ranging 10–1251 μg L⁻¹, associated with decreased gene expression: 64% of the top 1000 differentially expressed genes were down-regulated with increasing As metabolites in urine. No overlap was found between As-related gene expression and DNA methylation for individual genes.

Arsenic trioxide has been successfully used in the treatment of acute promyelocytic leukemia. To determine the As species in such patients, HPLC-HG-AFS and HG-AFS have been used to quantify the plasma concentrations of As^III, As^V and methylated metabolites (MMA and DMA), together with the total As in the blood cells and plasma.¹⁰⁶ Blood cells and plasma were digested with mixtures of HNO₃–H₂O₂ and analysed by HG-AFS. For As speciation, plasma samples were prepared with perchloric acid to precipitate protein. The supernatant was separated within 6 min using AEC and a mobile phase consisting of a mixture of 13 mmol L⁻¹ CH₃COONa, 3 mmol L⁻¹ NaH₂PO₄, 4 mmol L⁻¹ KNO₃ and 0.2 mmol L⁻¹ EDTA-2Na, at an isocratic elution flow rate of 1.0 mL min⁻¹. The methods provided linearity range of 0.2–20 ng mL⁻¹ for total As and 2.0–50 ng mL⁻¹ for the four As species. The spiked recoveries ranged from 81.2–108.6% and the coefficients of variation for intra- and inter-batch precision were less than 9.3% and 12.5%, respectively. A strategy using cisplatin as a viability dye together with conjugating lanthanide tags to marker proteins has also been developed to examine As₂O₃ cytotoxicity in single leukaemia (NB4/HL60) cells using ICP-MS.¹⁰⁷ A positive correlation among intracellular As contents, the ratios of apoptosis and dead cells can be used as a basis for the design of more effective As drugs. A method to determine toxicologically important As species in the blood of leukemia patients by selective HG with cryotrapping coupled either to AAS with a quartz multi-atomiser or ICP-MS has also been reported.¹⁰⁸ A five times sample dilution with addition of Triton X 100, Antifoam B, and L-cysteine was used to suppress excessive foaming during HG. Calibration slopes for whole blood and blood plasma spiked with As^V, MMA, and DMA at 0.25–1 μg L⁻¹ As and 0.025–0.1 μg L⁻¹ As for AAS and ICP-MS detection, respectively, did not differ from slopes in aqueous solutions. HG- cryotrapping-AAS was used to analyse samples with elevated levels of As species in blood plasma from patients treated with AsO₃ for acute promyelocytic leukemia. In the next step, plasma and whole blood from healthy donors that were spiked with the plasma from leukemia patients at levels of 0.15–0.4 μg L⁻¹ As were analysed by direct HG- cryotrapping -ICP-MS. Good recoveries (88–104%) were obtained for all species. The LODs in blood and plasma were 0.014 μg L⁻¹ for iAs and below 0.002 μg L⁻¹ As for methylated As species. However, recent clinical trials have shown that As^III is not always effective for non-APL malignancies. Another arsenical, S-dimethylarsinoglutathione (DMA^III(GS)), which is a putative metabolic intermediate in the hepatic metabolism of As^III, has shown promise for treating several types of lymphoma. However, the metabolism of DMA^III(GS) has not been fully investigated, probably because DMA^III(GS) is not stable in biological fluids where the concentration of glutathione is low. In study by Kato et al.¹⁰⁹ DMA^III(GS) was injected intravenously into mice, and the tissue distribution and metabolic dynamics of DMA^III(GS) compared with those of NaAsO₂. Each plasma sample was diluted two folds with elution buffer, filtered through a 0.44 μm-pore membrane filter, and then a 20 μL sample was applied to a Shodex Asahipak GS-220HQ or GS-520HQ HPLC column. The column was eluted with 50 mmol L⁻¹ Tris-HNO₃ buffer (pH 7.4, 25 °C, flow rate: 0.5 mL min⁻¹ or 1.0 mL min⁻¹) and the concentrations of As and S were continuously measured by ICP-MS. The authors found an organ preference for the distribution of DMA^III(GS) to the lung and brain in comparison to NaAsO₂. Furthermore, DMA^III(GS) appeared to bind to serum albumin by exchanging its glutathione moiety quickly after administration, providing insights into the longer retention of DMA^III(GS) in plasma.

Studies on As biotransformation mechanisms have mainly focused on As^III oxidation, As^V reduction, and As methylation; however less is known about the pathway for the biosynthesis of arsenosugars. The involvement of As^IIIS-adenosylmethionine methyltransferase (ArsM) in arsenosugar synthesis has now been demonstrated for the first time by Xue et al.¹¹⁰Synechocystis sp. PCC 6803 incubated with As^III or MMA^V produced DMA^V and arsenosugars, as determined by HPLC-ICP-MS. Arsenosugars were also detected in the cells when they were exposed to DMA^V. A mutant strain Synechocystis AarsM was constructed by disrupting arsM in Synechocystis sp. PCC 6803. Methylation of As species was not observed in the mutant strain after exposure to As^III or MMA^V; when Synechocystis AarsM was incubated with DMA^V, arsenosugars were detected in the cells. These results suggest that ArsM is a required enzyme for the methylation of iAs, but not required for the synthesis of arsenosugars from DMA, and that DMA is the precursor of arsenosugar biosynthesis.

Finally, a few studies have considered As speciation in mice. Arsenic methyltransferase (As3mt) catalyses the conversion of iAs to its methylated metabolites, including toxic MMA^III and DMA^III. A study to compare the oxidation states of As species in tissues of male C57BL/6 As3mt-KO and wild-type mice exposed to As^III in drinking water has been reported.¹¹¹ Wild type mice were exposed to 50 mg L⁻¹ As and As3mt-knockout mice (that cannot tolerate 50 mg L⁻¹ As) were exposed to 0, 15, 20, 25 or 30 mg L⁻¹ As. The As^III accounted for 53% to 74% of total As in liver, pancreas, adipose, lung, heart, and kidney of As3mt-KO mice; As^III and As^V methylated arsenicals did not exceed 10% of total As. Tissues of wild type mice retained iAs and methylated arsenicals: As^III, MMA^III and DMA^III represented 55–68% of the total As in the liver, pancreas, and brain. High levels of methylated species, particularly MMA^III, were found in the intestine of wild type, but not As3mt-KO mice, suggesting that intestinal bacteria are not a major source of methylated As. Blood of wild type mice contained significantly higher levels of As than blood of As3mt-KO mice. This study is reported to be the first to determine oxidation states of As species in tissues of As3mt-KO mice. Eicosanoids are a series of bioactive lipid mediators that are metabolised from polyunsaturated fatty acids and generated majorly from the precursor arachidonic acid. A study by Chen et al.¹¹² has identified the profiles of target eicosanoids after acute exposure to As. The principle objective was to determine and validate 10 eicosanoids in mouse serum using on-line SPE integrated with LC-MS. The LOD and LOQ were 0.003 to 0.288 ng mL⁻¹ and from 0.009 to 0.962 ng mL⁻¹, respectively and the intra-day and inter-day repeatability was 82–120 and 87–124%, respectively. The levels of 7 of the 10 eicosanoids in mouse serum significantly increased after As exposure compared with the levels in the vehicle control group.

Table 2 shows other applications of As speciation presented in the literature during the time period covered by this ASU.

Table 2 Applications of speciation analysis: arsenic

Analyte species	Technique	Matrix	Sample treatment	Separation	LOD	Validation	Reference
As^III, As^V, Cr^III, Cr^VI	HPLC-ICP-MS, collision cell gas He	Iron supplements	MAE with water 90 °C for 30 min	RP ion-pair, C18 column, isocratic, 1.0 mmol L^−1 tetrabutylammonium hydroxide (TBAH), 0.7 mmol^−1 EDTA, 5% methanol at pH 7.2	2–140 μg kg^−1	Spike recovery	113
As^III, As^V	FAAS	Waters and hair	Filter, adjust to pH 3 (for water), aqua regia leach (hair)	Hollow fiber (containing Triton X-100) LPME of As^III APDC complex enhanced by ionic liquid. Total As after reduction with L-cysteine	0.08 μg L^−1	GBW08666 and GBW08667, (both waters) GBW07601 (hair) and spike recovery	114
As^III, As^V	CS-ET-AAS	Drinking water	None	SPE on strong anion-exchanger. Previous published method	Not given	Comparison with results by TXRF	115
As^III, As^V	ET-AAS	Drinking water	Adjust to pH 7	DLLE of As^III with 1,2,6-hexanetriol trithioglycolate in methanol and CCl4	0.03 μg L^−1	GSBZ 50004-88 (water), spike recovery	116
As^III, As^V	ET-AAS	Natural waters (mineral tap and river)	Adjust to pH 7	SPE of As^III on moringa oleifera seeds	6	APS 1071 (water), spike recovery	117
As^III, As^V	ET-AAS	Waters (lake, mineral, tap and river). Edible mushrooms fish sediment, green tea, black tea, rice, cigarettes and soil (total arsenic)	Filtered (waters). Nitric acid and H2O2 digestion (biologicals). Adjust to pH 5	Deep eutectic solvent DLLME of As^III with DDTC and choline chloride–phenol into tetrahydrofuran. Reduction with KI and ascorbic acid	10 μg L^−1	NIST SRM 1946 (lake superior fish tissue) and CS-M-3 (mushroom powder) and spike recovery	118
As^III, As^V	ET-AAS	Environmental water, sediment and soil samples	Adjust to pH 3.5. Soils and sediments extracted with HNO3 + HCl + H2O2	Magnetic ionic liquid, air-assisted, liquid–liquid microextraction of As^III APDC complex into [C4mim][FeCl4]. Reduction with thiosulfate and KI	0.03 μg L^−1	GBW08605 (water) GBW 080231 (sea water) GBW07402, GBW070008 (soils), GBW07309 (sediment) and spike recovery	119
As^III	Batch-HG-AAS	Waters (hot spring, tap, ground, lake, river, sea)	None	Hydrophobic to hydrophilic switchable liquid–solid DME with diethylenetriamine immobilised on multiwalled carbon nanotubes	0.003 μg L^−1	TM-28.3 (Lake Ontario water) NRCC-SLRS-5 (river water) and spike recovery	120
As^III, As^V	Continuous flow HG-ICP-OES,	Water, drinking	None	Not explained. Selective generation of As^III assumed	As^III, 0.07 μg L^−1 : As^V 0.4 μg L^−1	Spike recovery	121
As^III, As^V	FI-HG-AFS	Rice	Ground, MAE with 0.28 M HNO3, centrifuged	Methylated arsines cryotrapped with Peltier-cooled device	1 μg kg^−1	NIST 1568b (rice flour). Comparison with results from an SPE-HG-AFS method	122
As^III, As^V	FI-HG-AFS	Algae	Washed, freeze dried, ground, hot 1% HClO4, centrifuged, filtered	Reduction of As^V by thiourea. Bromination, SPE on polystyrene, eluted with H2O	3 μg kg^−1	Spike recovery	123
As^III, As^V, MMA^V, DMA^V	HPLC-ICP-MS	Wine	Diluted, filtered	FDA method EAM 4.10. AEC. Standards matrix matched with 3% ethanol	0.2 μg kg^−1	Multi-laboratory validation. NIST 1643e (trace elements in water). Spike recovery	124
As^III, As^V, MMA^V, DMA^V, DMDTA^V, DMMTA^V	HPLC-ICP-MS. Compound dependent responses	Standards	None	RP, C18 isocratic 5.0 mmol L^−1 tetrabutyl ammonium phosphate, pH 7.7, 5% methanol. 70 min	0.04–0.26 μg L^−1	TWMW (trace metals in drinking water)	125
As^III, As^V, MMA^V, DMA^V	HPLC-HG-AFS with post column oxidation with persulfate	Urine	Filtered, refrigerated, filtered again	AEC, Hamilton PRP X-100, gradient: (A) NH4H2PO4, 2 mmol^−1 pH 4.6; (B) NH4H2PO4, 30 mmol^−1 pH 7.6	LOD not given, LOQ 2 μg L^−1	NIES no. 18 (human urine)	126
As^III, As^V, AB	HPLC-ICP-MS	Water and chicken liver	Minced, lyophilised, frozen, shaking with water, centrifuged, filtered	AEC (Dionex AS9) isocratic, sodium bicarbonate 50 mmol^−1 pH 10	6–16 ng L^−1	NMIJ 7901-a (AB), and UME 1201 (spring water)	127
As^III, As^V, MMA^V, DMA^V	HPLC-ICP-MS	Urine	Diluted (1 + 9) with water	AEC, Hamilton PRP X-100, isocratic 20 mmol^−1 NH4HCO3, 8 mmol^−1 CH3COONa, 2.4 mmol^−1 NaNO3 and 1% (v/v) ethanol pH 8.9	0.1 μg L^−1	Samples from two external quality assessment schemes	128
As^III, As^V, MMA^V, DMA^V	HPLC-ICP-MS	Apple juice	Diluted, filtered	RP, ion-pair, isocratic 2.5 mmol^−1 K2HPO4 and 5 mmol^−1 TBAH, pH 6	Not given	None. Experiment is designed for an instrumental analysis class	129
As^III, As^V, MMA^V, DMA^V	HPLC-HG-AFS	Soil and sediment	Freeze dried, ground (<75 μm), extracted 100 mmol^−1 oxalic	AEC, isocratic 15 mmol^−1 (NH4)2HPO4 and 6.6 mmol^−1 oxalic acid pH 6	Not given	GBW07405, GBW07406, GBW07407, GBW07408 (soils) and GBW07301a, GBW07407 (sediments)	130
As^III + As^V + MMA^V, DMA^V, AB	HPLC-ICP-MS	Fish	Washed, blended, frozen, lyophilised, ground, extracted with water, centrifuged, decanted, residue re-extracted twice, ultrafiltered (10 kDa)	Cation-exchange, Spheris S5, isocratic 10 mmol^−1 pyridine pH 2.05 in 5.0% (v/v) methanol	0.2–0.3 μg L^−1	NIST RM-50 (albacore tuna), DOLT-4 (dogfish liver)	131
As^III, As^V, MMA^V, DMA^V, TMAO and five unknown species	HPLC-ICP-MS cooled spray chamber. ^74Ge internal standard	Spirulina	MAE with H2O + H2O2 (1 + 4), filtered	AEC. Dionex IonPac AS23 isocratic 10 mmol L^−1 (NH4)2HPO4 pH 5.5 (with HNO3)	0.1–0.2 μg kg^−1	Spike recovery	132
AsH3	GC-ICP-MS	Gas and liquefied gas	Vaporization, online dilution	J&W GS-GasPro column (60 m, 0.32 mm i.d.), He/Xe, 40 to 250 °C in 5 min, 2 min at 250 °C	Not given	None	133
Inorganic As	Micro-XRF and micro-XANES	Nano zero-valent iron	Freeze-dried deposited on strip of Kapton tape	Spectral interpretation	Not given	None	134

---

3.3 Cerium

The antioxidant capacity of CeO₂ nanoparticles (NPs) was evaluated using HeLa humans cells as an in vitro model and applying a plethora of techniques including bioassays (MTT cytotoxicity assay, Comet assay), imaging techniques (light and electron microscopy) and XANES and μXRF measurements.¹³⁵ HeLa cells were supplemented with 200 μg L⁻¹ of CeO₂ NPs dispersion and incubated for between 15 minutes and 24 h. After exposure, the cells were harvested, pelleted and processed for their evaluation by the different techniques and assays. Data from SEM, TEM, light microscopy and μXRF measurements evidenced a preferential accumulation of nanoparticles in the endolysomes. XANES revealed that the Ce^III : Ce^IV ratio was dependent on the location of the internalised nanoparticles. Cerium was mainly found as Ce^IV in endolysomes whereas Ce^III was predominantly found in surroundings areas. These findings suggest that the effectiveness CeO₂ NPs to interact with cells depends on their location with endolysomes playing key roles.

3.4 Cesium

Only one paper covers the detection of Cs species. An analytical approach based on in situ XAFS analysis was used by Shiota et al.¹³⁵ as a rapid method for investigating chemical kinetics of Cs species during reaction in an alkali-activated municipal solid waste incineration fly ash. The Cs–K edge XAFS spectra, collected under different curing temperatures (room temperature, 60 °C, 80 °C, 105 °C), were processed using chemometric tools such as PCA and Linear Combination Fitting (LCF). It was found that Cs speciation changes over time during curing, and that when curing temperature was higher, those changes were faster and of a greater magnitude. The LCF results revealed that changes in in situ XAFS spectra reflect changes in pollucite (CsAlSi₂O₆) content, the formation of which contributes to Cs stabilisation, thus preventing Cs leaching. A simple reaction model for the reaction was used to calculate the reaction rate and its constant, which was in agreement with the LCF results.

3.5 Chromium

The redox speciation of Cr in food samples, covering the work published since 2012 in 68 articles (titles not given) has been reviewed.¹³⁶ That Cr^III is essential and Cr^VI is a carcinogen is re-iterated and, therefore, speciation analysis of Cr in food is important. Following a short tutorial introduction, the review is divided into two main parts. The first deals with the analytical methodology: non-chromatographic speciation, based on selective LLE, co-precipitation, selective SPE or complexation reactions, and HPLC separations. The detection is almost exclusively by atomic spectrometry (including X-ray techniques). The second part is devoted to food type and covers water, beer and wine, tea leaves and infusions, milk and dairy products and flour and bread. Some criticisms are made: in the section on tea leaves and infusions readers are told (about three published articles) that the reports concerning the content of Cr^VI in some foodstuffs with ETAAS detection without its identification are artefacts of inappropriately applied analytical methodology, which may lead to mistaken interpretations. The reviewer concludes that HPLC with ICP-MS detection seems to be the best technique, the application of the collision/reaction cell could reduce polyatomic interferences and that the use of enriched stable isotopes of ⁵³Cr^III and ⁵⁰Cr^VI allows the interconversion of species during the whole analytical procedure can be monitored (provided V does not co-elute).

Compared to the activity covered in the previous ASU, there has been a significant decrease in the numbers of papers published describing methods for the elemental speciation of Cr. This is particularly true for methods featuring separation by HPLC, for which there is only one for the current review period as opposed to 11 such reports last year. Most of the 18 or so articles include in this update feature methods based on selective extraction (either LLE or SPE) for the Cr^III and Cr^VI species. Papers that are primarily the application of methodology to a particular sample or only describe a procedure for the determination of one species are summarised in Table 3. The only review article published, devoted to the determination of Cr^III and Cr^VI in foodstuffs,¹³⁶ was discussed earlier in Section 1 on Topical reviews. Many of the new methods are based on separation followed by quantification with ETAAS.

Table 3 Applications of speciation analysis: Cr

Analyte species	Technique	Matrix	Sample treatment	Separation	LOD	Validation	Reference
Cr^VI	FAAS, ETAAS	Soils	Extraction with alkali	LLE of complex with methyltrioctylammonium chloride (Aliquat 336) into xylene	FAAS 2 mg kg^−1, ETAAS 25 μg kg^−1	Fluka CRM 041–30 G (chromium VI soil), spike recovery from the extracts	148
Cr^III	FAAS	Waste waters	Filter and adjust to pH 9	SPE with ion-imprinted polymer: nicotinate, acrylamide and ethylene glycol dimethacrylate, elute with 4 mol L^−1 g acetic acid	80 μg L^−1	Ielab Calidad, Spain municipal waste water RES 25.2 and spike recovery	149
Cr^VI	ETAAS	Vinegar	None	LLE with DDTC into amyl alcohol	0.3 μg L^−1	Spike recovery	150
Cr^VI	ETAAS	Seawater	Adjust to pH 4	LLE of the ion-pair with methyltrioctylammonium chloride in a single drop of chloroform	0.04 μg L^−1	Spike recovery from tap water	151
Cr^III	ICP-OES	River and tap water	Filter and adjust to pH 2	SPE with azodicarbonamide-functionalised attapulgite, elute with 0.5 mol L^−1 HCl + 5% thiourea	0.25 μg L^−1	GBW 08301, (river sediment), GBW 08303, (polluted farming soil), spike recovery	155
Cr^VI	LIBS	Lake water	Adjust to pH 4	Collection on amino-functionalised chitosan-modified graphene oxide	12 μg L^−1	Spike recovery	152
Cr^III, Cr^VI	HPLC-ICP-MS H2 + He collision gas	Exhaled breath condensate	Addition of EDTA, adjust to pH 8	AEC Dionex AG7 (5 cm long). Conditions as reported previously	0.007 μg L^−1 Cr^III, and 0.002 μg L^−1 Cr^VI	Sigma Aldrich proficiency testing material (Cr^VI in drinking water 1045/PE1453, lot LRAA1427)	153
Cr^III, Cr^VI	EXAFS, XANES	Ionic liquid for soil remediation	None	Spectral deconvolution	Not given	None	154
As^III, As^V, Cr^III, Cr^VI	HPLC-ICP-MS, collision cell gas He	Iron supplements	MAE with water 90 °C for 30 min	RP ion-pair, C18 column, isocratic, 1.0 mmol L^−1, TBAH, 0.7 mmol^−1, EDTA, 5% methanol at pH 7.2	2–140 μg kg^−1	Spike recovery	113

---

The only HPLC method involved separation by AEC with gradient elution (mobile phase A was water, mobile phase B was 2 mol L⁻¹ nitric acid).¹³⁷ The Cr^III was complexed with EDTA. No citation was provided for the HPLC separation, which was not subject to any optimization, so it is not clear whether the method is new or not. The ICP-MS was operated in kinetic energy discrimination mode, with helium as the collision gas, to minimise the interferences of ³⁵Cl¹⁶O¹H⁺, ³⁵Cl¹⁷O⁺ and ⁴⁰Ar¹²C⁺ on the Cr peak at m/z 52. For the various dairy products analysed, three different sample preparation procedures were used (a) total Cr (hot concentrated nitric acid in a sealed vessel in a microwave oven), (b) Cr^VI (alkaline extraction in an ultrasonic bath), and (c) Cr^III (0.05 M EDTA at pH 8.5 in a closed quartz flask in microwave oven). This latter procedure was developed after a substantial optimization. The method was validated by spike recovery of Cr^VI, and the analyses of a fruit yoghurt and a CRM (ERM-CE278k mussel tissue), both of which had known Cr^III concentrations. Of the 17 real samples analysed, only 3 contained measurable concentrations of Cr^III, and 7 contained measurable concentrations of total Cr. The researchers identified Cr^III in everything that was injected into the system, which they ascribed to contamination from the HPLC injection system. They applied a correction, which they acknowledged affected the LOD of 13 μg kg⁻¹ (dry weight) for a 0.3 g sample. As the Cr^III values found were not significantly different from those for total Cr, it was concluded that the foodstuffs concerned did not contain Cr^VI.

A method has been developed based on selective volatilisation of a Cr^IIIcomplex from an electrothermal atomiser.¹³⁸ Thenoyltrifluoroacetone (TTA) was added to the sample solution and allowed to react for 100 min under the optimised conditions of pH and TTA concentration. Reaction temperature was not investigated, though immediately after the TTA addition, the solution was heated for 8 min at 40 °C in an ultrasonic bath. At a pyrolysis temperature of 1200 °C, the Cr^III–TTA complex evaporated, leaving Cr^VI, which was then detected on atomization at 2400 °C. The procedure was validated by the analysis of two CRM (GBW(E)080403a and GBW(E)080404b), which are certified only for total Cr, and were also analysed by an HPLC-ICP-MS method (for which a citation is provided). They applied both procedures to the analysis of three water samples (tap, bottled and well), finding both species in all them, except for Cr^III in the tap water as the concentration was below the LOD of both methods. Both species were also found in the CRM. The LOD values were 0.05 μg L⁻¹ and 0.04 μg L⁻¹ for Cr^VI and total Cr respectively. The characteristic mass was 8 pg.

A number of SPE separation procedures for Cr speciation have been developed, several of which feature magnetic separation. Wei et al. investigated¹³⁹ iminodiacetic acid (IDA) functionalised magnetic nanoparticles (Fe₃O₄@SiO₂@IDA) finding that both Cr^III and Cr^VI were retained in the pH range of 2.5–3.5, while only Cr^III was adsorbed in the pH range of 4.0–9.0. The species retained on 10 mg of sorbent, ultrasonically suspended in a 50 mL water sample, were dissolved, after collection with an Nd–Fe–B strong magnet, in 0.5 mL of 1.5 mol L⁻¹ nitric acid and determined by ETAAS. The researchers do not mention a second magnetic separation, but this seems likely as the desorption step involved ultrasonication for 30 min. Nor are any details of the calibration procedure given, so it is not clear which species was used for calibration or whether calibration standards were taken through the pre-concentration procedure. The method was validated by the analysis of CRM GSBZ50009-88 (environmental water), which is certified for total Cr (at 0.5 μg mL⁻¹). Their results indicated that the CRM contained between 30 and 50 μg L⁻¹ Cr^VI, but no comment was made. The LOD values were 9 and 13 ng L⁻¹ for Cr^III and total Cr, respectively and both species were quantifiable in two river water samples. Spike recovery at concentrations of 100 and 200 ng L⁻¹ were between 89 and 99%. A somewhat similar procedure was developed by Sarikhani and Manoochehri¹⁴⁰ in which the species were extracted by an imidazolium-functionalised magnetite graphene oxide nanocomposite material, whose synthesis and characterization are described in some detail. At pH 2, only Cr^VI was extracted, whereas both species were retained at pH 6. From a 500 mL sample the appropriate Cr species were extracted with 17 mg of extractant by stirring for 9 min. After magnetic separation, the Cr was eluted into 1.4 mL of 2.2 mol L⁻¹ HCl by stirring for 16.5 min, and the extractant was again collected magnetically. The Cr in a 20 μL subsample was determined by ETAAS with palladium nitrate as the chemical modifier. The furnace operating parameters are listed, but there is no information about the calibration procedure. Two SRMs, spinach leaves (NIST SRM 1570a) and natural water (NIST SRM 1640), were accurately analysed for the total Cr content. The LOD values (for a 500 mL sample volume) were 1 ng L⁻¹ for Cr^VI and 2 ng L⁻¹ for total Cr. The speciation method was applied to the analysis of two real waste water samples, in which both species were found at concentrations ranging from 2–30 μg L⁻¹ and from which spike recoveries of 5 and 10 μg L⁻¹ were 92–104%. The procedure was also applied to the analysis of several real vegetable samples, but no attempt was made at speciation even though the researchers state in their abstract that “the method was utilised for rapid extraction and speciation analysis in food samples”. The sample preparation involved digestion with hot concentrated nitric acid and hydrogen peroxide. Concentrations of Cr ranging between 3 and 10 mg kg⁻¹ were found, and spike recoveries of 5 mg kg⁻¹ were 86–104%. The researchers conclude that “the new nanosorbent can be used for the speciation analysis of other metal ions”, but the basis for this statement is not explained. A method involving two SPE materials has been developed¹⁴¹ based on UAE and magnetic extraction on a nanocomposite prepared from magnetite nanoparticles, manganese oxide and alumina oxide, some of which was functionalised with¹⁴¹ trimethoxysilane (AAPTMS). The Fe₃O₄@MnO₂, Al₂O₃ material oxidised Cr^III to Cr^VI and was used for the pre-concentration of total Cr, whereas the AAPTMS-modified material was selective for Cr^VI. After elution with nitric acid, Cr was quantified by ICP-OES. Although a detailed optimisation of some of the relevant parameters was carried out, no information is provided about the sample volume, but the eluent (dilute nitric acid) volume is “about 5 mL”. A pre-concentration factor of 94 was reported, so the sample volume must have been about 500 mL. The procedure was validated by the analysis of a CRM (environmental express trace metals in drinking water), diluted to a concentration of 10 mg L⁻¹ and the method was applied to the analysis of 6 river water samples in which both species were found between 1.5 and 4.5 μg L⁻¹ and for which the total Cr values were not significantly different from those obtained by analysis by ICP-MS. The synthesis and characterization are described in great detail, but much of the information about the optimization of the analytical method is relegated to the supplemental information. Calibration was based on taking Cr^VI standards (0–500 μg L⁻¹) through the procedure. The LOD for Cr^VI was 20 ng L⁻¹. Unlike the situation for the two other methods based on magnetic nanomaterials, the Fe₃O₄@MnO₂, Al₂O₃ materials were reusable for up to 10 times. Each of the three articles include a table comparing the performance of method developed with those of other similar procedures.

Luz et al. developed a SPE procedure for the sequential determination (in synthetic saliva) of total Cr and Cr^VI by ETAAS.¹⁴² Under optimal conditions, any Cr^III was adsorbed on to silica nanoparticles functionalised with (3-aminopropyl)triethoxysilane and separated by centrifugation, leaving the Cr^VI in the supernatant. Compound-dependent responses related to the presence of the saliva matrix were observed, so calibration with a mixture of Cr^III and Cr^VI was used for total Cr determination and calibration using Cr^VI was used for the determination of this species. The researchers claimed that this was the first time that the different atomisation behaviours of the atomization of Cr^III and Cr^VI had been observed for ETAAS. Adding a chemical modifier (magnesium nitrate) decreased the sensitivities of both species in the saliva matrix, but did not remove the sensitivity difference. The LOD was 0.1 μg L⁻¹. A similar method, incorporating a SPE cartridge containing 500 mg of a strong anion-exchange material (quaternary ammonium groups), was developed for the determination of Cr^VI in airborne particulate matter.¹⁴³ The Cr species were extracted with an alkali and the pH adjusted so that only Cr^VI was retained. After washing off the unretained Cr^III, the Cr^VI was eluted (volume not given) with a solution containing equimolar concentrations of ammonium sulfate and hydroxide. Total Cr was determined following digestion of the particulate material and filter. The LOD was 0.1 μg L⁻¹. The method was validated by spike recoveries from filters and by comparison with the results of a spectrophotometric method. The procedure was applied to six work place environments, in three of which (welding foundry and electroplating) Cr^VI was detected following an 8 hour sampling period. The researchers speculate that the method may be an alternative to those based on the high-cost techniques of HPLC and ICP-MS for monitoring occupational exposure to Cr^VI. In principle, a procedure based on retention on a SPE cartridge has the potential to incorporate a pre-concentration step. Tiwari et al.¹⁴⁴ devised such a method in which Cr^III was retained on 78 mg of Amberlite XAD-16 loaded with salicylic acid in a column (3 mm × 3 cm). An automated flow injection (FI) system was connected directly to a FAA spectrometer. The sample volume was 10 mL, loaded at 5 mL min⁻¹, and the eluent was 0.1 mol L⁻¹ nitric acid at 4 mL min⁻¹. Based on the absorbance of the peak maximum the enhancement factor was 79 resulting in a LOD of 0.1 μg L⁻¹. There did not appear to be a wash step, so the fate of any Cr^VI remaining in the column is not clear. Total Cr was determined following reduction of Cr^VI by hydroxylamine hydrochloride (45 min at room temperature). The Cr content of SRM NIST 1640a (trace elements in industrial water) was determined accurately and the method was applied to the analysis of three industrial water samples in all of which both Cr species were found at concentrations between 15 and 50 μg L⁻¹. Recoveries of 20 μg L⁻¹ spikes were between 96 and 101%.

An electromembrane extraction method, one of a number of methods based on LLE, has been devised.¹⁴⁵ In this procedure Cr^VI was selectively extracted as the cationic complex with diphenylcarbazone into 30 μL of a dilute acid acceptor solution in the interior of a hollow fiber membrane loaded with 2-ethylhexanol. Presumably, the species actually transported across the supported liquid membrane was an ion-pair. The complex was formed by the addition of diphenylcarbazide to an acidified sample solution, when Cr^VI was reduced to Cr^III and was complexed by the oxidation product, diphenylcarbazone. It would appear that any Cr^III also present in the sample did not react with the diphenylcarbazone. To accelerate the extraction, electrodes were placed in the sample solution (positive) and in the acceptor solution (negative) and a potential of 300 V applied. The method was optimised by a univariate search of the relevant parameters, which did not include the nature and dimensions of the hollow fibres, which were porous PPQ3/2 poly-propylene with 300 μm wall thickness, 1200 μm internal diameter, and 0.2 μm pores. The end of the 31 mm long fibre was sealed with hot pliers. The optimization clearly showed the electrokinetic effect, as the signal in the absence of the applied potential but with vigorous stirring was much lower than that obtained when the potential was applied. The optimization also showed that stirring was also needed as a higher signal was obtained when the solution was stirred. Following extraction, 20 μL were transferred to the atomiser of a continuum-source atomic absorption spectrometer. Total Cr was determined after oxidation with hot acidified permanganate, the excess of which was removed by dropwise addition of a sodium azide solution. The enrichment factor was 110, corresponding to 66% recovery, and the LOD was 0.02 μg L⁻¹. The method was applied to the analysis of three water samples (tap river and mineral), in which both species were found, apart from Cr^III in the tap water, at concentrations ranging from 0.06 to 0.44 μg L⁻¹. Spike recoveries of 0.25 μg L⁻¹ of both species were 76–108%. Porto et al. used a continuum-source flame atomic absorption spectrometer in a DLLME method¹⁴⁶ optimised by both univariate search and response surface methods. With APDC as the complexing agent, Cr^VI was extracted (at pH 2) into 50 μL of 1-undecanol with the aid of 300 μL ethanol. After collection by centrifugation, the organic solvent was removed with a micropipette, mixed with 200 μL of ethanol and introduced into the spectrometer by a discrete nebulisation device (a micropipette tip). To also extract Cr^III, the sample (pH 7) was heated at 80 °C for 15 min and only 275 μL ethanol was needed. For both analyses, sodium chloride was added to the solutions at a concentration of 5% (m V⁻¹). No information about calibration was given, other than the standards were prepared in ethanol. Enhancement factors of 19 (for the conditions for total Cr) and 26 (Cr^VI) were obtained with LOD values of 7 and 0.4 μg L⁻¹, for total Cr and Cr^VI, respectively. The procedure was applied to the analysis of a certified water reference material (NIST SRM 1643e), for which a value of 16 ± 2 (standard deviation n = 3) μg L⁻¹ was obtained. The certificate value is 20.4 ± 0.2 (expanded uncertainty with a coverage factor of 2). The researchers claim that there is no significant difference between these numbers. The method was applied to the analysis of one tap and three mineral water samples, in none of which could the analytes be determined. Recoveries for spikes between 2 and 50 μg L⁻¹ ranged from 88–115%. Given the rather large difference in LOD values, a better approach might be to oxidise any Cr^III to Cr^VI as was done by Tafti et al.¹⁴⁷ who devised procedure they called supramolecular DLLME-based solidification of floating organic drops (SM-DLLME-SFO) in which Cr^VI was extracted (from 10 mL of sample at pH 2.5) as the APDC complex into decanoic acid (70 mg), which was rapidly injected as a solution in tetrahydrofuran (700 μL) into the aqueous sample, whereupon a cloudy suspension of water-immiscible, fine droplets of coacervates of decanoic acid formed immediately. The organic solvent was collected on the top of the aqueous phase by centrifuging, solidified by cooling and transferred to a conical vial where it melted immediately at room temperature. Finally, 15.0 μL of the extract along with 10.0 μL of a palladium magnesium modifier was manually injected into the graphite furnace atomiser. Total Cr was determined after oxidation with permanganate as described above.¹⁴⁵ The enhancement factor was 127 and the LOD was 2 ng L⁻¹. The method was applied to the analysis of a CRM (NRCC river water SLRS-1) for which a total Cr concentration of 0.35 ± 0.25 μg L⁻¹ was obtained. The ± term was not specified, but might be the standard deviation for n = 3. As the 95% confidence interval would be −0.26 to 0.97 μg L⁻¹, it is perhaps not surprising that the researchers concluded that the result was not significantly different from the certificate value of 0.36 ± 0.03 μg L⁻¹. This rather large ± term was not in agreement with the RSD determined for replicate (n = 6) analysis of a 0.1 μg L⁻¹ standard of 4.2%. They also applied the procedure to the analysis of three water samples (well, tap and river) and two serum samples (prepared by adding trichloroacetic acid to precipitate proteins, centrifugation and dilution), in all of which both species were found at concentrations ranging from 15 to 515 ng L⁻¹. Recoveries of spikes (20–100 ng L⁻¹) ranged from 95 to 104%.

A number of papers are summarised in Table 3. Some additional information may be of interest. Researchers compared the efficiency of extraction of Cr^VI from soils by dilute sodium hydroxide solution with that obtained with US EPA method 3060A.¹⁴⁸ A novel polymeric solid phase extractant was synthesised selective for Cr^III, but there was no pre-concentration.¹⁴⁹ So although some potential interferences were removed, the method could not detect any analyte in the real samples. On the other hand, a method for the determination of Cr^VI in vinegar by LLE¹⁵⁰ detected the analyte at single-digit μg L⁻¹ concentrations in ten samples. The procedure, carried out in a flow-batch analyser, involved a single vinegar–amyl alcohol–ethanol phase that was broken by the addition of Britton–Robinson buffer. A single-drop LLE method, also carried out in a flow system, was able to detect Cr^VI in six seawater samples at triple-digit ng L⁻¹ concentrations.¹⁵¹ When compared with the determination of Cr^III in water, compound-dependent responses were observed. A method¹⁴⁹ based on SPE onto surface-modified attapulgite (a fibrous crystalline hydrated magnesium silicate) for Cr^III was able to detect the analyte in two water samples at concentrations around 1 μg mL⁻¹. However, a LIBS method,¹⁵² in which Cr^VI was collected by electric-field assisted absorption onto chitosan-modified graphene oxide, was unable to detect the analyte in a lake water sample. To facilitate the in situ generation of the laser-induced plasma, the liquid sample was diverted from the optical path by a gas pulse. A previously developed HPLC-ICP-MS method was able to quantify both Cr^III and Cr^VI in exhaled breath condensate of many tens of samples from individuals occupationally exposed to Cr^VI compounds by inhalation.¹⁵³ The researchers provide a useful tutorial introduction and discussion of occupational exposure to chromium. To follow the conversion of Cr^VI to Cr^III in the ionic liquid 1-butyl-3-methylimidazolium chloride being evaluated for the remediation of contaminated soil, EXAFS and XANES indicated that about 70% of the Cr^VI was extracted in 30 min of which about 48% of the Cr^VI complexed with humic acid was reduced to the corresponding Cr^III species during extraction.¹⁵⁴

3.6 Europium

A CE-ICP-MS method has been developed for the determinations of a paralytic shellfish toxin, saxitoxin (STX).¹⁵⁶ The STX was labelled with Eu^IIIvia a coupling agent, diethylenetriamine-N,N,N′,N′′, N′′-pentaacetic acid (DTPA), the CE system was interfaced with the ICP-MS instrument, fitted with a micro-concentric nebuliser of optimum flow 100–400 μL min⁻¹, via a 60 cm length of 75 μm i.d. × 375 μm o.d. Fused-silica capillary. The running buffer was 20 mmol L⁻¹ NaH₂PO₄⁻ and 5.0 mmol L⁻¹ Na₂B₄O₇ adjusted to pH 6.0 with a runtime per sample of 13 minutes. The method was used to determine trace STX in seafood samples with a LOD of 0.38 fmol for a 100 nL sample injection and a RSD of <7% (n = 5). Spike recoveries, of between 0.05 and 5 μg g⁻¹ of STX, ranged between 93 and 110%. The authors state that the method provides an alternative to conventional methods for determining STX content, such as LC/LC-MS and ELISA, by offering greater sensitivity and a lower LOD and cost.

3.7 Gold

An HPLC-ICP-MS method has been developed to distinguish between Au NPs and Au^III ions. Separation of the two analytes was achieved using a mobile phase of 1 mmol L⁻¹ NaH₂PO₄, 1 mmol L⁻¹ Na₂HPO₄ and 10 mmol L⁻¹ sodium dodecyl sulfate at pH 7.3 flowing at 0.5 mL min⁻¹ and a 250 × 4.6 mm RP C₁₈ column of 7 μm particle size and 1000 Å pore size. The separation followed a size exclusion mechanism. A human cervical adenocarcinoma cell line (HeLa) was exposed to differing amounts of Au NPs and/or Au^III ions. After centrifugation the supernatant from the exposure medium was injected directly onto the column. The detection limits for AuNPs and dissolved Au³⁺ were 0.62 and 2.3 μg L⁻¹, respectively, whereas the quantification limits for AuNPs and dissolved Au³⁺ were 1.8 and 6.9 μg L⁻¹, respectively. It was observed that AuNPs can undergo an oxidation process in the supernatants and that about 10% of AuNPs and 8–24% of dissolved Au³⁺ was associated with cells. To evaluate the biological impact of AuNPs, a classical viability assay onto HeLa cells was performed in the presence of increasing dosage of 10 nm AuNPs. The results showed that 10 nm AuNPs exhibit a slight toxic effect.

3.8 Halogens

Following on from their work reviewed last year on the measurement of halogen (Cl and Br) containing drugs in human samples from a clinical trial, the Ghent research group under Vanhaecke¹⁵⁷ have refined the method for the determination of Cl-containing drugs, so it can cope with gradient elution separations. The report describes the determination of the non-steroidal anti-inflammatory drug diclofenac, its metabolite 4′-hydroxy-diclofenac and other related Cl containing compounds, in human plasma by RP HPLC coupled to triple quadrupole ICP-MS. As with most methods for the HPLC separation of polar drugs and their metabolites, gradient elution was required to provide sufficient resolution of the peaks and the authors showed that under different methanol and acetonitrile eluent compositions the plasma gave a variable response for the measurement of Cl. For peak quantitation this effect was overcome by using a mathematical function to correct the effect of the eluent composition on the sensitivity for Cl, which was measured as ³⁵ClH₂⁺ at m/z 37. The LOQ for Cl (as diclofenac) was 50 μg L⁻¹, which could be significantly improved to 2 μg L⁻¹ using on-line sample pre-concentration on a chromatographic column and a time-programmable 10 port, 2 position micro-valve. To validate the method a synthetically degraded diclofenac sample spiked with 4′-hydroxy-diclofenac was used and showed a recovery of 95–105% for this metabolite. When applied to spiked human plasma samples an acceptable recovery of 92–98% and precision of <4% RSD was shown for both 4′-hydroxy-diclofenac and diclofenac.

The isotopic analysis of Cl containing compounds in environmental samples by GC coupled to various different detectors including MC-ICP-MS, IRMS or QMS has been investigated in an effort to establish the sources, transformation pathways and sinks of some Cl-containing priority pollutants. The compound-specific Cl-isotopic analysis of tetrachloromethane (CCl₄) and trichloromethane (CHCl₃) was facilitated using GC separation coupled to either IRMS or QMS.¹⁵⁸ Different parent ions were analysed in each instrument, GC-IRMS measurements analysed CCl isotopologue ions, whereas GC-QMS looked at the ions CCl₃, CCl₂, CCl, for CCl₄ and CHCl₃, CHCl₂, CHCl, for CHCl₃. The optimised GC-QMS parameters (dwell time 70 ms, 2 most abundant ions) resulted in SDs of 0.2‰ (CHCl₃) and 0.4‰ (CCl₄), which are only about twice as large as the values for GC-IRMS. The accuracy of each method was determined using samples from CCl₄ and CHCl₃ abiotic degradation experiments, calibrated against isotopically enriched reference standards. In a related approach,¹⁵⁹ overcoming the limitation of IRMS which can only be used for a narrow range of compounds due to the fixed Faraday cup configuration, GC coupled to MC-ICP-MS was developed for the compound specific stable Cl isotopic analysis of volatile aliphatic compounds. Both the approaches investigated also overcome the high sample requirements and offline sample preparation that are associated with analysis using TIMS. The MC-ICP-MS method was evaluated by using five in-house characterised standards and eight chlorinated methanes, ethanes, and ethenes. The isobaric interference of the ³⁶ArH⁺ polyatomic on ³⁷Cl⁺ was minimised by using dry plasma conditions. Samples containing 2–3 nmol L⁻¹ Cl injected on-column were sufficient to achieve a precision of 0.1‰ or better. Long-term reproducibility and accuracy was better than 0.3‰ if organics were analysed in compound mixtures. Standardization was carried out by using a two-point calibration and drift was corrected using an internal standard. The method offered a direct, universal and compound-specific procedure to measure the delta ³⁷Cl of a wide array of organic compounds overcoming limitations of previous techniques, but offering high sensitivity and accuracy.

The measurement of F by ICP-MS is difficult due to both its very high first ionisation potential (17.42 eV), which results in insufficient F⁺ formation in the plasma, and also because of the commonly occurring interferences from water derived polyatomics such as ¹⁶O¹H₃⁺, ¹⁸O¹H⁺ and ¹⁷O¹H₂⁺ at the low m/z ratio for ¹⁹F. However, it is possible to determine F in a plasma based instrument by using the formation of polyatomics with suitable elements, such as Ba. This approach¹⁶⁰ has been used for some initial studies on the speciation of F containing compounds using HPLC coupled to triple quadrupole ICP-MS. The coupling of AEC with triple quadrupole ICP-MS allowed for the separation and detection of fluoride and fluoroacetate, which was facilitated by the addition of Ba to enable the formation of BaF. Separation of the two F containing species used an HPLC system with a PRP-X100 Hamilton AEC column (4.1 mm × 250 mm). The eluent was ammonium carbonate (30 mmol L⁻¹, pH 8.7) at an isocratic flow rate of 1 mL min⁻¹ and a 100 μL sample loop was used for sample introduction. To couple the HPLC to the ICP, a transfer capillary was used to connect the chromatographic column to the nebuliser via a T-piece, which allowed the introduction of the Ba solution prior to entry into the plasma. A large number of plasma and other instrumental parameters were optimised to achieve the maximum formation of BaF⁺, with the minimum formation of the counter regulatory ion BaO⁺. The response was compound independent and the LODs for F and fluoroacetate were 0.022 mg L⁻¹ and 0.11 mg L⁻¹, respectively. Both compounds were baseline separated and detected quantitatively, making this newly developed method a promising candidate for non-target F speciation analysis in environmental samples.

3.9 Iron

A paper has reported on iron speciation analysis this year which, as might be expected given the essential role Fe plays in living processes, involves the measurement of haemoglobin (Hb) in blood to validate methods traceable to the SI.¹⁶¹ To achieve this ss-ID-ICP-MS was used, with an Hb spike enriched with ⁵⁷Fe produced in-house from a previously published method, and double and triple spiking procedures. The HPLC conditions consisted of a mobile phase of 12.5 mmol kg⁻¹ Tris, 125 mmol kg⁻¹ NH₄CH₃COOH at pH 7.8, flowing at 0.35 mL min⁻¹ through a Gel column. The Fe signals at m/z 56 and 57, measured using a quadrupole ICP-MS operated in reaction cell mode with H₂ flowing at 6 mL min⁻¹, were used to calculate the isotope ratios required for IDMS. The performance of the ss-ID-ICP-MS method was compared with five other methods, which are also described in detail, using a whole blood CRM (JCCRM912-2M ReCCS, Japan) with a certified Hb content of 132 ± 1.4 g L⁻¹. The found values were 135 ± 2.1 g kg⁻¹ and 133 ± 2.1 g kg⁻¹ for the double and triple spiking approaches with the major uncertainty contribution arising from the precision of the isotope ratio measurements in each case.

3.10 Lead

A paper on lead speciation in algae samples by CE-ICP-MS has been reported by Jing-Xi et al.¹⁶² Experimental conditions affecting sample processing and separation (running buffer and voltage) were evaluated in detail. Microwave-assisted extraction (MAE) using 50% methanol and 10 mL 0.5 mol L⁻¹ acetic acid was employed for quantitative releasing of organolead compounds and Pb^II, respectively. Following extraction, the analytes were analysed by CE-ICP-MS by using an uncoated fused silica capillary column (75 μm i.d. × 85 cm), a buffer solution containing 70 mmol L⁻¹ H₃BO₃, 17 mmol L⁻¹ Na₂B₄O₇ at pH 8.90 and a separation voltage of −13 kV. Under these conditions, Pb^II, TML and TEL were efficiently separated in less than 20 minutes. The method reports LOD values of 0.091, 0.23 and 0.030 μg L⁻¹ for Pb^II, TML and TEL, respectively with a RSD of 4% (n = 5). The results showed that the method offered good recoveries (90–103%) for lead-spiked algae samples. Only Pb^II was detected the in seaweed samples analysed using the developed method. The speciation of Pb in rice samples has also been recently reported.¹⁶³ The Pb species in ground rice were extracted with 1% EDTA in 2% methanol at 65 °C and UAE for 30 minutes followed by shaking for 4 hours, centrifugation, supernatant collection and filtration. The extracted Pb species were separated on a C₁₈ column under a gradient elution with the mobile phase consisting of (A) methanol and (B) 0.1 mol L⁻¹ acetic acid – ammonium acetate at pH = 4.7 and flowing at 0.6 mL min⁻¹. These conditions allowed the separation of Pb^II, trimethyllead chloride (TML), triethyllead chloride (TEL), and tetraethyllead (TTEL) in 25 minutes. The HPLC eluent was fed directly to an ICP-MS instrument and an ‘optional gas’ is mentioned in the table of operating conditions. Presumably, this gas was O₂ to negate the build-up of C on the sampler and skimmer cones. The LOD values obtained for the four Pb species analysed for ranged from 0.03 μg L⁻¹ for Pb^II to 0.6 μg L⁻¹ for TTEL. Only Pb^II was detected in the two rice samples analysed. Spike recoveries for Pb^II and TML were ≥90% but worse for TEL, 85% and TTEL, 75%, which the authors attribute to instability of these species. However, as a mixed spike was used this hypothesis should have been borne out by an increase in the recovery for one or both of the other two Pb species determined.

3.11 Manganese

The speciation of Mn^II and Mn^VII in water samples by ion pair chromatography coupled with ICP-MS. After optimisation of the various chromatographic factors, sample pH, ion pair reagent, Mn^II complexing agent, and composition of mobile phase. The optimal separation conditions were found to be 1 mmol L⁻¹ TBAH and 0.36 mmol L⁻¹ EDTA at pH 7.5 flowing at 1 mL min⁻¹ through a C₈ column and directly coupled to an ICP-MS instrument and a run time of 6 minutes. The detection limit for Mn^II was 0.22 μg L⁻¹, while for Mn^VII the value was 1.55 μg L⁻¹. Two water CRMs, ES-H-2 groundwater and EU-H-3 wastewater, were analysed and the found Mn^II content was in agreement with the certified value, which is presumably total Mn, in each case. However, when the CRMs were spiked with Mn^VII the recovery of this species was only around 65% of the spiked value. The Mn^II recoveries were consistent though suggesting that the added Mn^VII was being retained or lost during analysis rather than being converted to Mn^II. The method was applied to various water sample types, tap, bottled and artesian, and the Mn^II content ranged between 0.3 and 2.6 μg L⁻¹. Spike recoveries for Mn^VII for these samples were between 70 and 90%.

3.12 Mercury

Following the pattern of recent years the number of papers concerning Hg speciation continues to rise. This is probably in part due to the increase in availability of ‘dedicated Hg analysers’ and also the use of selective extraction procedures, which is a less costly approach than the more traditional approach of the online coupling of a separation technique with an element selective detector. A second factor is of course the greater awareness of regulatory bodies on the need for the speciation of an element to be understood as opposed to just the total elemental concentration. The final factor is the technological advances that have allowed the measurement of Hg isotope ratios with sufficient accuracy and precision to allow Hg biogeochemical processes to be elucidated with a greater degree of certainty.

The speciation of Hg is increasingly being determined in samples of human origin to assess exposure levels of these important toxins. One of the drawbacks in methods used for Hg speciation is that different extraction procedures are used for different sample matrices, with possible cost and throughput implications. Workers in one of the main centres for speciation studies in Europe, in conjunction with those from NIST, have therefore developed a common methodology for the simultaneous determination of MeHg, EtHg, and iHg in human blood, hair and urine by ID-GC-ICP-MS.¹⁶⁴ The extraction procedure involves the addition of ²⁰¹Hg-enriched MeHg, ¹⁹⁸Hg-enriched EtHg, and ¹⁹⁹Hg-enriched Hg^II to the sample, either 0.15 g of blood or 0.5 g of urine or 0.1 g of hair followed by 3 mL of 25% TMAH and a magnetic stir bar. Subsequently, a focused MAE, 35 W for 4.5 minutes, of the Hg species was performed. Finally, the extracted Hg species were derivatised with sodium(tetra-n-propyl)borate (NaBPr₄), extracted into hexane and pre-concentrated by evaporation of the hexane under N₂ before injection onto a DB-5MS column. The GC was coupled to a Q-ICP-MS via an in-house fabricated heated transfer line. The method was validated using three CRMs, NIST SRM 955c (Caprine Blood) Level 3 and IAEA 085 and 086, both human hair with all found values being in agreement, within uncertainty limits, with the certified values. As there was no urine CRM available the method was validated by spiking experiments at three levels, 1, 2 and 5 ng g⁻¹, with recoveries of 97% or greater for all Hg species at each spiking level except for EtHg at 1 ng g⁻¹ where the recovery was 94%. Conversion of MeHg and EtHg into Hg(II) was observed for all sample matrices under all sample preparation conditions evaluated as part of the method optimisation. The authors report that he most important factor for controlling species interconversion is the amount of amount of NaBPr₄ added for derivatisation. Dealkylation of EtHg into Hg^II was observed for all sample matrices, up to 95% in blood, 29% in hair and 11% in urine. Dealkylation of MeHg ranged from 1 to 15% in blood and hair analyses, but was not observed with urine samples. Hg methylation was not observed in any matrix. The use of spikes enriched with different Hg isotopes for each Hg species allowed the observed dealkylation effects to be corrected for. Another matrix to which Hg speciation is being increasingly being applied to is human breast milk. It would be interesting to apply the above methodology to this sample type. A paper describing the validation of a method for the determination of MeHg species in breast milk has been published, by workers in another of the main centres for speciation studies in Europe in conjunction with colleagues from elsewhere.¹⁶⁵ In this procedure a ²⁰¹Hg enriched MeHg spike was added to lyophilised breast milk. For analysis by GC-ICP-MS the MeHg was extracted from the spiked sample with 6 mol L⁻¹ HNO₃ and heating at 85 °C for two hours followed by derivatisation with NaBEt₄ and extraction into isooctane. The Hg species were separated on a 95% methyl-5% phenyl-polysiloxane column. For analysis by HPLC-ICP-MS the MeHg was extracted from the spiked sample with 0.5% 2-mercaptoethanol and heating at 85 °C for two hours. After centrifugation the supernatant was then freeze-dried, the dried material dissolved in HPW. The Hg species were separated on a C₈ column with a mobile phase of 60 mmol L⁻¹ ammonium acetate, 0.01% 2-mercaptoethanol in 5% methanol. In each case Hg isotope ratios were measured using a Q-ICP-MS. The authors report that the GC method was less robust than the HPLC method, mainly due to the effects of the sample matrix on ethylation efficiency and the need for pH measurements on each sample. However, the HPLC column required a clean-up stage after approximately 50 samples due to increasing back pressure due to the retention of compounds such as proteins. No mention of a guard column is made and the use of one could have simplified the clean-up procedure required. For the GC method a ‘progressive seizure’ of the injection syringe was observed after analysis of 10–20 samples. The LOQ of the HPLC-ICP-MS method was evaluated using a ‘trueness bias’ approach, using an in-house RM of infant formula spiked with MeHg, was evaluated to be 0.16 μg Hg per kg wet weight. The method was applied to 180 samples of breast milk with most values being <the LOQ and the highest found to be 0.43 μg Hg per kg wet weight.

The measurement of Hg species in waters can be challenging due to the low concentrations usually found and a therefore pre-concentration step is usually required. The validation of method for the determination of MeHg in seawater, one of the more challenging matrices for analysts, has been reported.¹⁶⁶ The MeHg was extracted from 70 mL of seawater, after acidification with concentrated HCl, into CH₂Cl₂ with overnight shaking, collection and then evaporation of the organic phase and derivatisation with NaBEt₄ in a citrate buffer at a pH of 4.6, sparging of the ethylated MeHg onto Tenax traps and thermal desorption at 180 °C onto a GC column. An isothermal separation (36 °C) was used followed by pyrolysis and detection by AFS. The reported LOD ranged between 7 and 25 fmol g⁻¹, a precision of 3%, a spike recovery of 79% and an expanded uncertainty (k = 2) of between 11 and 21% relative depending on the MeHg concentration of >400 fmol and <50 fmol, respectively. An alternative approach to pre-concentration is the use of ion-imprinted materials. To this end a MeHg ion-imprinted magnetic nanoparticle (MeHg IIMNP) has been synthesised.¹⁶⁷ The MeHg IIMNP employed core–shell Fe₃O₄@SiO₂ NPs, the MeHg-1-pyrrolidinecarbodithioic acid complex as a template, methacrylic acid as a functional monomer and trimethylolpropane trimethacrylate as a cross linker. The MeHg IIMNPs (50 mg) were added to 500 mL of water sample, filtered (0.45 μm) and pH adjusted to 5, which was stirred for 20 minutes and the MeHg IIMNPs collected with a magnet. The adsorbed MeHg was then eluted from the NPs with 1 mL of 0.1 mol L⁻¹ and 2 mol L⁻¹ HCl for five minutes. The desorption step was repeated, the fractions combined and the extract analysed for MeHg using CE-ICP-MS. Thus, the enrichment factor was 250 fold. The MeHg IIMNPs could be used for 50 water samples. The method was validated by spiking tap water with MeHg to give final concentrations of between 1 and 20 pg mL⁻¹ with recoveries ranging from 92–98% and a reported LOD value of 0.084 pg mL⁻¹. The use of a more conventional automated DLLME device, directly coupled to an HPLC-CV-AFS instrument, for extraction and analysis of Hg species in water samples has also been reported this year.¹⁶⁸ The water samples were filtered (0.22 μm) and the pH adjusted to 5 before subjection to the DLLME procedure. Details of the HPLC-CV-AFS system are given in a cited reference. Under optimal conditions the linear range was 10–1200 ng L⁻¹ for EtHg and 5–450 ng L⁻¹ for MeHg and Hg²⁺. The LOD values were 3.0 ng L⁻¹ for EtHg and 1.5 ng L⁻¹ for MeHg and Hg²⁺. Method validation was by spiking with recoveries ranging between 88 and 96.1% and the RSD was <5.1%. The apparatus was used to determine Hg species in six water samples, no MeHg or EtHg was detected and iHg ranged between 14 and 200 ng L⁻¹. A method has been developed using thiolated graphene (G-SH) for the CPE iHg and MeHg. The Hg species extracted were separated and quantified using HPLC-ICP-MS. The chromatographic conditions were a C₁₈ column, with a mobile phase of 1.0 g L⁻¹L-cysteine, 60 mmol L⁻¹ ammonium acetate flowing at 1 mL min⁻¹. Under optimal conditions, the enrichment factors, for a 40 mL sample were 78 and 77 for iHg and MeHg, respectively. The LOD values were 3.8 and 1.3 ng L⁻¹ for iHg and MeHg, respectively with spike recoveries of >94% for both species. The method was applied for speciation of mercury in surface water, ground water, domestic sewage and sea water with only iHg being detected the seawater and domestic sewage samples. Finally, in this section on Hg speciation in waters the development and use of flow injection (FI) photochemical vapour generation (PVG) coupled to a miniaturised solution cathode glow discharge-atomic emission spectroscopy (SCGD-AES) device has been reported.¹⁶⁹ Mercury speciation was achieved by using different irradiation powers and wavelengths from a UV lamp and irradiation times. Total Hg is converted to Hg⁰ at 8 W at 254 nm and 60 s of irradiation. Only Hg^II was reduced to Hg⁰ at 4 W and 365 nm and a 20 s irradiation time. The difference in concentration between the total and iHg values is assumed to be the organoHg fraction. The LOD value was found to be 0.2 μg L⁻¹. The device was validated using GBW09101b (human hair) and GBW10029 (fish tissue) CRMs (no sample preparation procedure is given) and spiking of tap water. All results for the CRMs were in agreement, within uncertainty limits, with the certified values and spike recoveries were 96% or greater.

Two papers report on Hg speciation in rice this year. In the first of these two different extraction procedures, MAE with either 6 mol L⁻¹ HNO₃ or alkaline TMAH, were compared.¹⁷⁰ The extracted species were ethylated and the extracts analysed using GC-pyrolysis-AFS. The method were validated using BCR-60 (aquatic plant), BCR-482 (lichen) and NCS ZC73027 (rice plant) CRMs for total Hg. As these CRMs are not certified for MeHg they were also used for MeHg spike recovery validation. No statistical difference (F and Students t-test) was found for the results obtained using either of the extraction procedures. For the CRMs the measured total Hg content was in agreement, within uncertainty limits, with the certified values and MeHg spike recoveries were 98% or greater for all extractions except for the alkaline extraction with the rice plant CRM, NCS ZC73027, which was 81% suggesting that the acidic extraction procedure is more suitable for this type or sample. The TMAH extraction also suffered from higher blank values than the acidic extraction leading the authors to conclude that the acidic extraction was more suitable for this type of work. The second paper covered here reports the MeHg and total Hg content in rice grains harvested from plants grown in the vicinity of a coal fired power station.¹⁷¹ In this work the rice seeds were dehusked, ground and the 177 μm fraction collected. The MeHg in the ground rice was extracted using 25% KOH in methanol at 75–80 °C for 3 h, leached into CH₂Cl₂ and back-extracted into HPW for a determination based on US EPA method 1630. A similar procedure, but replacing the methanolic KOH with HNO₃, was sued to extract and quantify the MeHg content of the soils in which the rice plants were growing. The method was validated using TORT-2 (lobster hepatopancreas) CRM and spiking of rice samples with MeHg. For TORT-2 the found and certified values agreed within uncertainty limits and the spike recoveries ranged from 72–99%. For the soil samples MeHg spike recoveries varied between 92 and 107%. The MeHg content of the rice samples varied from 1.7 to 3.8 μg kg⁻¹ and from 0.3 to 3.5 μg kg⁻¹ in the soil samples. There was no correlation between the MeHg content in the rice samples and distance from the power plant, which, the authors suggest, may be due to the presence of a petroleum refinery and cement plant in the near vicinity of the collection point of the rice samples having the greatest MeHg content. Positive correlations of rice MeHg content with soil Hg species, soil S, and soil HG⁰ content were however observed.

Four papers are concerned with Hg speciation in sediments and soils this year. The first of these articles reports the results of an interlaboratory comparison (n = 4) of the analysis of 17 sediment and soil samples collected from Hg contaminated areas.¹⁷² Two laboratories used GC-pyrolysis-AFS with the other two methods be GC-ECD and HPLC-CL. The four extraction methods used were as follows. Shaking with 5 mol L⁻¹ HCl, 1 5 mol L⁻¹ CuSO₄, extraction into toluene and back extraction into an EDTA solution and complexation with emetine-CS₂. The Hg complexes were separated on an RP column with CL detection due to the reaction with Ru(bpy)₃³⁺. The method LOD was 0.07 ng g⁻¹. The second method involved alkaline leaching, extraction into toluene with dithizone, back extraction into alkaline Na₂S solution, S₂⁻ removal with HCl, re-extraction of the formed dithizone–MeHg complexes into toluene, filtration and analysis by GC-ECD. The method LOQ was 0.1 ng g⁻¹. The third and fourth methods were leaching with 5 mL of 18% KBr in 5% H₂SO₄ solution and 1 mL of 1 mol L⁻¹ CuSO₄, extraction into CH₂Cl₂ and either ethylation of propylation, purging and trapping onto Tenax followed by thermal desorption and analysis by GC-pyrolysis-AFS. The method LOD was 0.05 ng g⁻¹. Each laboratory also analysed two estuarine sediment CRMs, ERM-CC580 and IAEA 405, and all results obtained, except for the HPLC-CL method for IAEA 405, were in agreement with the certified values. The results obtained for most of the samples by all methods were in agreement except for those samples which had a high S content and the procedure involved the use of a halide based reagent. The authors suggest that, for samples rich in S, the use of a stronger extractant than HBr or HCl, such as dithizone should be considered. A detailed discussion on the possible causes of poorer extraction from S rich samples is included in the article and suggestions for method improvement given. The paper highlights that information about sample chemical composition is needed before analysis is undertaken. The second paper covered here details the evaluation of Hg methylation and MeHg demethylation rates in vegetated and non-vegetated saltmarsh sediments.¹⁷³ To achieve this sediment cores were spiked, into various sediment horizons, with tracers of ¹⁹⁹Hg²⁺ and CH₃²⁰¹HgCl. Total Hg measurements were made, after an Aqua Regia extraction of sediment samples, by ID-CV-ICP-MS. A water vapour distillation method, in the presence of H₂SO₄ (9 molar) and KCl, was used to extract MeHg from the sediment samples followed by ssID-ICP-MS. The paper contains a detailed discussion on the Hg methylation and MeHg demethylation rates observed and factors which may affect these rates. The results suggest that the formation of MeHg from iHg linked to the presence of vegetation but that this observation might be site specific and further studies are needed. The third paper reviewed here covers the quantification of MeHg and Hg geochemistry at a former Au ‘exploitation’ site in Brasil.¹⁷⁴ A water vapour distillation method was employed to extract MeHg from sediment samples as described above followed by ethylation, trapping, thermal desorption and analysis by GC-pyrolysis-AFS. Method validation was by the use of ERM-CC580 estuarine sediment CRM and a recovery of 95% was obtained. The LOD and LOQ values, for a 0.5 g sample, were 0.11 mg kg⁻¹ and 0.33 mg kg⁻¹, respectively. The site is heavily contaminated in places, with a total Hg content ranging from 0.06 to 43 mg kg⁻¹ whilst the MeHg content ranged between <LOD to 8 μg kg⁻¹. The final paper in this section comprises a ‘critical evaluation’ of a water vapour distillation procedure for the determination of methylmercury in soil samples.¹⁷⁵ A Plackett–Burman design was used to evaluate the effect of the following parameters on the extraction efficiency of the procedure; distillation time, KCl%, acid concentration, gas flow, type of acid, (HCl or H₂SO₄) and water volume. The optimal conditions were found to be, 0.5 g of dried soil or sediment, 200 mL of H₂SO₄ (9 molar), 200 mL of KCl (20%) and 10 mL water and a distillation time of 70 minutes. After extraction MeHg ethylated, trapped, thermal desorbed with analysis by GC-AFS and ID-GC-ICP-MS. The results obtained by detection systems were broadly in agreement. Finally, the optimised method was applied to soil samples collected from sites impacted by coal fired power plants and the found MeHg content ranged from 0.09–3.9 ng Hg per g.

Table 4 shows examples of other applications of Hg speciation presented in the literature during the time period covered by this ASU.

Table 4 Applications of speciation analysis: Hg

Analyte species	Technique	Matrix	Sample treatment	Separation	LOD	Validation	Reference
OrganoHg, total Hg	μCCP-OES. OrganoHg by photochemical vapour generation. Total Hg by CV generation	Salt and fresh water fish muscle	Lyophilisation. Total Hg : HNO3–H2O2, MAE. OrganoHg: HBr hydrolysis, extraction into toluene, back extraction into L-cysteine	None, selective extraction procedure. Hg^2+ by difference	OrganoHg: 2 μg kg^−1. Total Hg: 3 μg kg^−1	NRCC DOLT-4 (dogfish), BCR-463 (tuna), ERM CE-464 (tuna), TORT-2 (9lobster hepatopancreas). Recoveries: Total Hg 101 ± 10%, organoHg, 100 ± 8%, Hg^2+ 102 ± 13%	176
OrganoHg	μCCP-OES. OrganoHg by photochemical vapour generation. Thermal desorption AAS	Salt and fresh water fish muscle	Lyophilisation. OrganoHg: HBr hydrolysis, extraction into toluene, back extraction into L-cysteine	None, selective extraction procedure	OrganoHg: 2 μg kg^−1	NRCC DOLT-4 (dogfish), BCR-463 (tuna), ERM CE-464 (tuna), TORT-2 (lobster hepatopancreas). Recoveries: organoHg, 100 ± 10%	177
OrganoHg, total Hg	Thermal desorption AAS	Canned tuna	Total Hg: none, direct thermal desorption. OrganoHg: MAE with toluene and HCl. Back extraction into L-cysteine	None, selective extraction procedure	OrganoHg: 1.4 μg kg^−1. Total Hg: 0.5 μg kg^−1	Total Hg: NIST SRM 1566b (oyster), 97 ± 1%. NRCC DORM-4 (dogfish muscle), 111 ± 2%	178
MeHg	GC-pyrolysis-AFS	Human breast milk	Lyophilisation. Suspended in methanolic KOH, 70 °C, 6 h. Ethylation with tetra ethyl sodium borate	None, selective extraction procedure	0.1 μg L^−1	IAEA 085 (human hair). Recovery 85–105%	179
MeHg, total Hg	ss-ID-GC-ICP-MS	Rice plants and soil	Total Hg: hot HNO3. MeHg: distillation in HPW, H2SO4, KCl, CuSO4, ethylation with tetra ethyl sodium borate	No details given, method cited	Total Hg: 0.42 ng g^−1. MeHg: 0.05 ng g^−1	Total Hg: Mess-3 (sediment), 101 ± 3% recovery. MeHg: ERM CC580 (sediment), 100 ± 11% recovery	180
Total Hg, MeHg, PhHg	Total Hg: thermal desorption AAS. MeHg, PhHg, GC-pyrolysis-AFS	Soil	Total Hg: either direct analysis or after solid phase dilution with sand. MeHg, PhHg: UAE and shaking with 6 mol L^−1 HCl, ethylation with tetra ethyl sodium borate	HP-5 silica capillary analytical column. 50 °C for 1 min, ramp at 15 °C min^−1 to 110 °C, then ramp at 30 °C min^−1 to 230 °C, held for 1 min	Not given	Total Hg: ERM-CC580 (sediment), 95 ± 5% recovery. MeHg: ERM CC580 (sediment), 99 ± 3% recovery. PhHg by the standard addition method, no data given	181
iHg, MeHg	GC-pyrolysis-AFS	Fish	Extraction with TMAH, ethylation, trapping onto n-decane	Cited reference	LOD: 2 μg kg^−1, LOQ: 6 μg kg^−1	NRCC DORM-2 (dogfish), BCR 463 (tuna). Recoveries >93% ISO17025 accredited laboratory	182
iHg, MeHg	Au trap or GC-ICP-MS, IDMS	Algae	Total Hg: US EPA method 1631. MeHg: US EPA method 1630, ethylation, trapping	Silanised glass column, packed with 15% OV-3 (phenyl methyl dimethyl silicone 10% phenyl) on Chromasorb	Total Hg: 9 pg MeHg: 0.6 pg	NRCC DORM-4 (dogfish). 95% recovery for both Hg species	183
iHg, MeHg, PhHg	HPLC-ICP-MS. O2 addition to the plasma	Fish and waters	Extraction onto Fe3O4@SiO2 NPs functionalised with MPTS	C18 column. Mobile phase was 50 : 50 methanol 0.1% mercaptoethanol, 50 mmol L^−1 NH4Ac, pH 5, 1 ml min^−1	All species <1 ng L^−1	NRCC DORM-4 (dogfish). >95% recovery for both Hg species	184
iHg, MeHg	ssID-GC-ICP-MS	Bird feathers	TMAH or HNO3 extraction with MAE or hot block, ethylation, extraction into isooctane	Cited reference	MeHg 3.2 ng g^–1 1 iHg 11 ng g^−1	NIES-13 (hair). Recoveries of 94% for MeHg and 98% for iHg	185
iHg, MeHg, EtHg	HPLC-DBD-AFS	Fish	Methanolic KOH, CH2CL2 and HCl. Back extraction into Na2S2O3	C18 column. Mobile phase: 0.06 mol L^−1 ammonium acetate, 10% methanol, 0.01% 2-mercaptoethanol, pH 6.8	iHg, 1.6 μg L^−1. MeHg, 0.42 μg L^−1. EtHg 0.75 μg L^−1	GBW10029 (tuna fish). iHg certified value <LOD. MeHg recovery 100%	186

---

3.13 Nickel

An HPLC-post column-ID-ICP-MS has been described for Ni speciation in cocoa infusions.¹⁸⁷ Separation of Ni compounds was performed by using a weak anion-exchange diethylamine (CIM DEAE, 12.0 mm i.d., length 3.0 mm) monolithic column. Moreover, structural assignment was performed by QTOF-MS. The methacrylate-based column support and the chromatographic conditions applied enabled the efficient separation of Ni species, and column recoveries ranged from 93 to 98%. Three Ni-containing peaks were first detected by ICP-MS and manually collected for subsequent QTOF-MS measurements. Ni was found to be present in the cocoa infusions as Ni^II and Ni–gluconate and Ni–citrate complexes. The HPLC post-column IDA-ICP-MS measurements revealed Ni–gluconate as the predominant Ni-specie in the cocoa infusions.

3.14 Selenium

During the past year, two reviews have appeared. A comprehensive review on Se bioavailability in food by Moreda-Piñero et al.¹⁸⁸ summarizes published data concerning to the application of in vivo and in vitro approaches as well as the effect of food processing and food composition (protein, fat and mineral content) on Se bioaccessibility. There are also extensive tables including Se bioavailable percentages in different types of food samples (cereals, vegetables, seafood, algae and seaweeds, yeast and yeast-based supplements, eggs and milk). The review cites more than 100 references. The second review reports the last achievements on Se speciation by using microextraction techniques.¹⁸⁹ The first section of the review describes the theoretical principles behind microextraction techniques focusing on solid phase microextraction (SPME) and liquid phase microextraction (LPME). The second part is dedicated to pre-concentration and speciation procedures by using microextraction techniques connected to several instruments such as AAS, GC-MS, ICP-MS, HPLC-ICP-MS and XFS. The last section deals on the future trends of Se microextraction techniques highlighting the use of nanostructured fibers for SPME. The review includes 52 references covering the period from 2008 to 2016.

Most of the papers published this year on Se are based on the elucidation of its metabolism by performing both in vivo and in vitro studies with the aim of gaining a deeper insight into Se metabolic pathway, Se role in health and diseases and Se protective effect against the toxicity of elements such as Hg and Pb. The performance of these studies usually implies the combined use of LC-ICP-MS and LC-ES-MS/MS. Related to the binomial Se and health, Bryan et al.¹⁹⁰ carried out a complete study on the role of Se in the progress of idiopathic cardiomyopathy disease by using whales as model, in particular pygmy sperm whales (Kogia breviceps). Selenoproteins were extracted from heart and liver tissues and separated and detected by a 2D-chromatography approach (2D-SEC-AE-ICP-MS). The SEC measurements were useful for observing differences in Se-containing peak patterns between stages of disease progression. Se containing fractions were further collected and subsequently analysed by RPLC-ES-MS/MS for peptide sequencing and protein identification. A database search allowed authors to identify selenoproteins such as GPX, Se-containing proteins such as SeAlb and Se-binding proteins such as metallothioneins either in liver or heart tissues. Analysis of the data by ANOVA showed significant variation in metallothionein levels between animals groups with and without pathological findings. A notable increase in the level of Se-containing metallothioneins was detected in those animals suffering cardiopathy. One of the most novel findings of this year under review is the protective effect of Se containing phycocianine (Se-PC) against paraquat-induced renal injury.¹⁹¹ Phycocianine (PC) is a type of phycobiliprotein from cyanobacteria with beneficial effect on lungs and renal diseases. The combination of this valuable product with the antioxidant properties of Se may improve the beneficial effects of PC. Based on that, Se-PC was isolated from Se-enriched N-fixing cyanobacterium (Nostoc spc) and subsequently purified by using a step–step approach that includes, (1) stepwise precipitation of contaminant proteins and the Se-PC fraction by adding 30% and 50% NH₄SO₄, respectively, (2) separation of Se-PC fraction from concomitant proteins by DEAE-sepharose chromatography using increasing NaCl concentration as mobile phase and (3) collection of Se-PC fraction and final purification by Sephacryl S-300 SEC using 2 mmol L⁻¹ Na₃PO₄ buffer (pH 7.0) as a mobile phase at 1 mL min⁻¹. The effect of the purified PC and Se-PC extracts on KH₂ cell viability after a 24 hour exposure to paraquat was further evaluated. Data from caspase-3-activity and ROS generation assays revealed that Se-PC blocked PQ-induced cell apoptosis through restraining the overproduction of superoxide and preventing therefore oxidative damage and nephrotoxicity. The work opens new insights on the use of Se-PC for treating those renal diseases mediated by oxidative stress processes. In another paper, Se deficiency has been associated with the appearance of age-related cognitive illness.¹⁹² Based on that, Se status and Se bound to proteins in different biofluids (serum, red blood cells and cerebrospinal fluid) of Alzheimer’s disease (AD) patients was determined by ICP-MS and SEC-ICP-MS measurements. No differences in Se levels in serum were observed between AD and control patient groups. However, a significant decrease in Se level was reported in erythrocytes from AD patients. Separation of serum by SEC-ICP-MS evidenced the presence of a Se-containing fraction which was collected, purified and analysed by ES-MS/MS. The application of a bottom-up proteomics approach allowed authors to confirm the presence of the selenoprotein SelP in the Se-containing fraction.

Selenium metabolism in plants remains of research interest. The metabolism of organic (SeMet) and inorganic Se (⁷⁷Se^VI) in two Se accumulator plants (garlic and Indian mustard) was studied by HPLC-ICP-MS and ES-MS/MS.¹⁹³ Plants were hydroponically grown and exposed simultaneously to SeMet and ⁷⁷Se^VI over seven days. Selenium species in the water extracts of the different part of the plants were separated by means of a SEC (Shodex Asahipak GS-320HQ, 7.5 i.d. × 300 mm) column, with 50 mmol L⁻¹ NH₄AC at pH 6.5 as mobile phase flowing at 0.6 mL min⁻¹. The data obtained revealed different Se accumulation mechanisms in the two accumulator plants tested. The authors proposed several mechanisms for Se metabolisms. It was found that γ-GluMeSeCys and MeSeCys were the major metabolites of Se^IV and SeMet in garlic and Indian mustard whereas an additional Se-compound was found in Indian mustard, identified by ES-MS/MS as selenohomolanthionine (SeHLan).

In an outstanding study, the role of glutathione (GSH) and Se^IVagainst Hg toxicity was evaluated by analysing the reaction products resulting from a GSH–Se^IV binary system, a GSH–Hg^II binary system and a GSH–Se^IV–Hg^II ternary system.¹⁹⁴ The GSH–Se^IV binary system was prepared by mixing 1 mmol L⁻¹ Se^IV with varying concentrations of GSH providing GSH/Se^IV molar ratios ranged from 0.1 : 1 to 10 : 1; GSH–Hg^II binary system was obtained by incubating GSH and Hg^II at the molar ratio of 2 : 1 and finally a GSH–Se^IV–Hg^II system was prepared by mixing the binary system GSH/Se^IV at a molar ratio of 2 : 1 with Hg^II/Se^IV at a molar ratio varying from 0.125 : 1 to 2 : 1. After 24 hours of incubation, reaction products were analysed using RPLC (Kromasil C18 (5 μm, 4.6 mm × 200 mm)) coupled to ICP-MS and RPLC (CAPCELL PAK C18 MG II (5 μm, 4.6 × 250 mm) coupled to ES-MS/MS. The authors suggested both the competition between Se^IV and Hg^II for the GSH binding sizes and the formation of stable and low toxic (Hg_xSe_y)_n(GS)_m precipitates as responsible for the protective effect of Se against mercury toxicity. Similarly, Wang et al.¹⁹⁴ evaluated the role of Se against mercury toxicity by using HepG2 cells as model. For this purpose, HepG2 cells were co-treated with SeCyst₂ and MeHg at concentration ratios of MeHg : SeCys₂ of 1 : 0, 3 : 1, 1 : 1 and 1 : 3. The concentration and speciation distribution of MeHg and Se was monitored by LC-ICP-MS and a chip-based treatment method. The cytosolic fractions of HepG2 cells were first fractionated in a SEC-ICP-MS and the low molecular weight containing Hg and Se molecules were subsequently analysed by RP-LC-ICP-MS. Based on MS/MS data, the complexes MeHg–GSH and MeHg–Cys were identified. On-line chip-based MSPME-micro HPLC-ICP-MS was applied to evaluate the accumulation kinetics of MeHg during incubation studies. Data evidenced an increase of MeHg and Hg^(II) concentration with increasing incubation time and MeHg/SeCys₂ ratios. The authors suggest a Se detoxification mechanism based on aggregation of MeHg and SeCys₂ to promote the uptake of MeHg followed by transformation of MeHg into complex species (MeHg–GSH and MeHg–Cys).

The protective effect of Se against another metal such as Pb has been also evaluated as reported in the study carried out by Fang et al.¹⁹⁵ For this purpose Se-containing protein hydrolysates (SPHs) from Se-enriched rice were co-administered with lead to a PC12 (phechromocitoma derived) cell line and RAW264-7 (mouse macrophages from RAW267) cell lines. Data from MMP assay, caspase activity and Western blotting evidences that SPHs significantly protect PC12 and RAW264-7 cells from Pb^II-induced apoptosis. Additionally, HPLC (Hamilton PRPX-100, 150 mm × 4.6 mm and 5 μm)-ICP-MS and SEC (Superdex peptide 100–300 GL, 300 mm × 10 mm, 13 mm) ICP-MS measurements showed that SPHs were mainly composed of low molecular weight peptides (2220–500 Da) and SeMet. The SPHs are postulated by the authors as an active agent to preventing Pb poisoning. However, and as the authors also mention, in vivo studies need to be performed to clarify the role of SPHs on lead toxicity.

Two papers report on selenoproteins (SelP, SeAlb and GPx) quantification in plasma and serum samples by using HPLC post-column ID-ICP-MS and by employing two affinity columns: heparin sepharose (HEP) and Blue sepharose (BLUE). In the first study,¹⁹⁶ the method was applied for determining selenoproteins in 849 individuals belonging to the Arctic indigenous population of Inuit of Nunavik (Canada) who have a Se uptake which is amongst the highest in the world. Total plasma Se concentrations (median = 139 μg L⁻¹; interquartile range (IQR) = 22.7 μg L⁻¹) were markedly lower and less variable than whole blood Se concentration (median = 261 μg L⁻¹, IQR = 166 μg L⁻¹). With regard to Se distribution in Se-containing proteins, 52% was found in SelP, while GPx and SeAlb represented only 25% and 23% of total Se in plasma, respectively. Interestingly, comparisons were made between populations with high levels of Se intakes. A non-linear relationship between whole blood and plasma Se was found in the Inuit population which is not in agreement with data already published from communities consuming Brazil nuts where a linear relationship is reported. The different behaviour suggests accumulation of different Se compounds, SeMet in a Brazilian Amazon population and possibly selenonine in blood of an Inuit population. Additionally, Hg was monitored and quantified in plasma simultaneously while performing Se speciation. Mercury appeared associated to selenoproteins especially SelP that bounds 50% of total mercury. In the second paper, modifications in the HPLC post-column IDA-ICP-MS methodology were performed with the aim of improving the accuracy of selenoprotein quantification.¹⁹⁷ One of the problems encountered by the authors are the changes observed in the signal provided by ⁷⁷Se and ⁷⁸Se spikes when are continuously added after the column. Variations in the Se background during separation hampered the proper quantification of selenoproteins, especially SelAb. The reason for the unstable Se signal was attributed to the mobile phase composed mainly of NH₄Ac with concentrations ranging from 0.1 to 1.5 mol L⁻¹. The introduction of high levels of NH₄Ac into the ICP-MS increases the carbon load into the plasma affecting therefore the Se sensitivity by enhancing its signal. The alternative use of NH₄NO₃ as eluting solvent provided stable background levels and notably improved the quantification of selenoproteins in blood serum. Finally, ⁷⁷Selenite and ⁷⁶SeMet were employed as tracers to evaluate losses of low-molecular-weight Se metabolites during deproteinisation of blood and plasma samples. Two Se species, ⁷⁷Se^IV and ⁷⁶SeMet, were chosen as model compounds for iSe and seleno amino acids, respectively. The performance of four protein precipitation agents such as acetonitrile, methanol–acetonitrile (1 : 1 v/v), methanol–ethanol (1 : 1 v/v) and methanol–acetonitrile–acetone (1 : 1 : 1 v/v/v) were tested. For this purpose ⁷⁷Se^IV and ⁷⁶SeMet were added to plasma and hemolysed blood samples. The spiked samples were then submitted to protein precipitation and the resulting supernatant and pellets were analysed by SEC-ICP-MS and reverse isotope dilution. Different behaviours were observed depending on the Se compound, the solvent used for protein removal and the sample type. The ⁷⁶SeMet remained in solution after precipitation in both plasma and blood samples whereas ⁷⁷Se^IV completely disappeared in solutions of blood samples. The best results were obtained by using the methanol–acetonitrile–acetone mixture. The SEC-ICP-MS measurements evidenced that both co-precipitation phenomena and association with proteins are the main causes of selenite losses during protein precipitation in blood samples. The study highlights the importance of sample treatment in selenometabolomics and metabolomics in general.

One paper has appeared this year dealing on urine metabolites of Se but instead of detecting them in urine, SeSug1 and TMSe were toxicologically characterised for first time.¹⁹⁸ For this purpose three cellular model systems: human hepatome (HepG2), human urothelial (UROTsa) and human astrocytoma (CCF-STTG-1) cells were employed. The results obtained were compared with the cytotoxicity behaviour exerted by Se^IV, SeMet and MeSeCys. Total Se and Se species determination was achieved by ICP-QQQ-MS and HPLC-ICP-QQQ-MS, respectively. Most of the paper is devoted to the optimization of the instrumental parameters affecting Se determination by ICP-QQQ-MS. A LOD value of 20 ng L⁻¹ was obtained by using an H₂ and O₂ gas cell mixture and by adding 3% isopropanol to samples for Se signal enhancement. Selenium speciation in water soluble fractions of the incubated cells was achieved by using a polar modified RP column (phenyl and pentafluorophenyl phase, YMC-TRiart-PFC column, 3 μm, 3 × 250 mm) and 20 mM ammonium formate (pH 3, 35% methanol) as the mobile phase flowing at 250 μL min⁻¹. The results showed the lack of toxicity from SeSug1 and TMSe up to 1000 μmol L⁻¹ whereas Se^IV, SeMet and MeSeCys exerted notably toxicity, especially in human urothelial cells (UROTsa). Interestingly, speciation measurements evidenced the presence of additional unknown Se species when UROTsa cells were treated with SeSug1 which opens new insights into the metabolisms of this selenosugar.

Selenium speciation in supplements has been described in several papers. Two papers from Kubachka et al. report on Se characterization in 13 dietary supplements labelled to contain Se in various forms and under different dosage types (liquids, tablet, capsules, and softgels). In the first paper, multiple extraction methods and LC-ICP-MS measurements were applied.¹⁹⁹ Special attention was paid to selection of the best Se extraction procedure. In this vein, samples were treated with different extracting solutions (water, 0.1 mol L⁻¹ NaOH, 1 mol L⁻¹ HCl, and 4 mol L⁻¹ MSA) either under sonication or heating (room temperature, 50 °C and 95 °C), and by applying different treatment times. Heated water extraction was selected for releasing Se compounds in the supplements tested. For the separation and detection of the species, a Zorbax StableBond (SB-Aq) (4.6 mm × 250 mm, 5 mm) column coupled to the ICP-MS was used. For those extracts containing MeSeCys or SeMet, additional Se-containing peaks were detected and attributed to the presence of seleno-amino acid oxidation products. To confirm the peak identities, the extracts with and without H₂O treatment were analysed using LC with simultaneous ICP-MS and ES-MS detection. Se methylselenocysteine oxide (MeSeOCys) and methylseleninic acid (MSeA) were identified in MeSeCys-containing extracts. In the second paper,²⁰⁰ a rapid screening method based on the combination of XFR and real time resolution accurate-mass spectrometry (DART-HRAMS) was applied to directly determine iSe and selenoamino acids respectively. Samples were directly analysed without the need for dissolution of the sample. The use of DART-HRAMS in Se speciation has been limited due to its low performance when analysing Se incorporated to large molecules such as selenoproteins. Despite its limited use, the authors highlight its applicability for rapidly detecting adulteration or false label claims in selenised supplements by monitoring the presence or absence of SeMet or MeSeCys. By applying DART-HRAMS, analysis time (including sample preparation and analysis) was considerably reduced from 2 hours (LC-ICP-MS) to less than 10 minutes. The LOD was estimated to be 100 μg of Se which is suitable for detecting Se in dietary supplements with levels of total Se ranging from 600 to 1500 μg g⁻¹.

Among Se supplements, selenised-yeast characterization has been a topic of much research in past years and nowadays the characterization of seleno-compounds continues to attract the attention of several researchers. About 100 seleno metabolites, most of them never reported, were identified in selenised yeast.²⁰¹ Selenised metabolites were extracted in aqueous solution by means of probe sonication followed by SEC-ICP-MS analysis to collect a low-molecular weight fraction. The metabolites structural identity was subsequently assigned using LC-TOF-MS by means of accurate mass measurements and manual Se isotope pattern identification. Besides Se species, sulfur and S–Se species were also identified in the aqueous extract of selenised yeast. Following Se metabolism in selenised-yeast, the presence of selenohomolanthionine (SeHLan) was detected and identified for first time in water soluble extracts of Se-enriched torula yeast (Candida utilis).²⁰² The identity of SeHLan was confirmed by HILIC (TSKgel Amide-80 3 × 300 mm × 2.6 μm) -ICP-MS and HILIC-orbitrap-MS and by using a standard of SeHLan synthesised in the authors laboratory. Additionally, a methodology based on the use of HILIC-ICP-MS was developed for quantifying Se species and a LOD of 146 ng g⁻¹ was reported. The major Se compound in the trula yeast was SeHLan, accounting for a 60–80% of the total Se whereas the amount of SeMet was notably low (7–11% of the total Se). Selenium was mostly found in the water soluble extracts suggesting the non-proteinogenic origin of Se in this type of yeast. Finally, losses of Se during manufacturing of peanut flour containing Se-enriched proteins has been investigated.²⁰³ First, the authors performed a comparative study of different extracting solutions for determining selenospecies in Se-enriched peanut by RPLC-ICP-MS using a Zorbax RX-C8 (4.6 mm × 150 mm/5 μm) column and 0.1% PFPA, 2% MeOH at pH 1.98 and flowing at 0.9 mL min⁻¹ as mobile phase. Best results in terms of extraction yield were obtained when using a mixture of protease, lipase and sodium thiosulfate (2 : 1; 5 mmol L⁻¹) and after one hour of ultrasonic processing. Under these conditions, recovery rates ranged from 85 to 102% were obtained. The predominant specie found in the peanuts was SeMet, accounting for 65% of the total Se. Losses of Se, of about 37% of the total Se content, were detected during manufacturing peanut protein flour. The detected Se losses were attributed to Se volatilisation, dissolution, degradation or physical transfer that may occur during the complex process needed to manufacturing the peanut-based Se flour.

A method based on the use of Zorbax SB-C18 (4.6 × 250 mm) column on line coupled to HG-AFS was employed for performing Se speciation in food and beverages.²⁰⁴ The main novelty of the paper is the use of ionic liquids (ILs) as mobile phase modifiers for the separation of Se^IV, Se^VI, SeMet and MeSeCys. Most of the paper is devoted to performing a multivariate optimization of the parameters affecting Se species separation and hydride generation in presence of imidazolium type-ILs with different alkyl chain lengths. The Se species in the selected samples were efficiently separated in 12 minutes using 0.1% (v/v) of 1-hexyl-3-methylimidazolium chloride ([C₆mim]Cl) as a modifier. The LOD values of 0.92, 0.86, 1.41 and 1.19 μg L⁻¹ were obtained for Se^IV, Se^VI, SeMet and MeSeCys, respectively. These values were comparable or even better than those obtained when applying HPLC-ICP-MS. According to the authors, the hypothesised separation mechanisms in presence of ILs involves different processes such as: interaction between anionic Se species and IL molecules retained on the stationary phase as well as hydrophobic partitioning of the ion pairs formed between ILs and Se species with the RP stationary phase. The method was successfully applied to garlic, yeast, beer and wine samples. Recovery values of between 95 to 105% were obtained in the different food and beverage samples.

Most studies related to Se are focused on its essentiality however it is considered one of the elements with the narrowest ranges between deficiency and toxicity. Thus, several papers on the impact of Se on environment have appeared this year. In this context, ⁷⁹Se is one of the long-lived, and high priority classified radionuclides. The sorption and speciation of Se^IV and Se^VI in a boreal forest soil collected from Olkiluoto Island was evaluated.²⁰⁵ The location under study was selected as it has been used as a site for the disposal of spent fuel from nuclear power plants from Finland. Solutions of Se^IV and Se^VI were exposed to the soils under different experimental conditions with the aim of evaluating the effect of pH, time of incubation, soil depth, temperature, mineral composition and presence of microbes, all under both aerobic and anaerobic conditions. Once the experiments were completed, Se speciation was performed in the liquid-phase by using a Dionex anion exchange column AS11 (i.d. 4 mm × 250 mm) coupled to ICP-MS. Additionally, the mass distribution coefficient, Kd, for Se^IV and Se^VI was calculated. A large amount of K_D values were derived from the different experiments. Speciation studies showed the low transformation of Se^IV and Se^VI, even with changes between aerobic and anaerobic conditions. Aluminium and iron(II) oxides played a key role in Se^IV sorption in soils while Se^IV was less well retained. Selenocyanate Se(CN⁻) was detected as a major Se specie in petroleum refinery wastewaters by applying an instrumental set-up consisting on the use of ion chromatography, UV-induced volatile Se compound generation and ICP-MS detection.²⁰⁶ In brief, Se species were first separated on a Dionex anion exchange column AS11 (i.d. 4 mm × 250 mm) using a step gradient with NaOH as the eluent, and subsequently converted to their volatile forms by UV irradiation in presence of a mixture of HCOOH–HNO₃. This approach offered a LOD for Se^IV and Se(CN⁻) of 0.06 and 0.01 μg mL⁻¹. In the absence of suitable CRM, the developed method was validated independently by IC-ICP-MS and IC-HG-ICP-MS. The method was also applied to Se speciation in spiked waste water samples with recovery values in the range of 95–102%. Liu et al. describe a novel procedure based on the use of electrolytic hydride generation and AFS for the determination of total Se and Se^IV in waters and rice samples.²⁰⁷ For first time the transformation of Se^IV to SeH₂ on an L-cysteine modified carbon paste electrode (CMCE) is reported. The paper contains detailed discussion on the effect of different parameters (type of electrode surface modification, electrolytic current, temperature, interferences) on the fluorescence intensity. The LOD value was determined to be 0.065 μg L⁻¹ for Se^IV. The precision (n = 10) was less than 5% relative. The method was validated by spike recoveries and accurate analysis of the standard reference materials (rice flour GBW08502 and tea leaves GBW08505). The method was applied to the speciation analysis of Se^IV and Se^VI in spring and river water and rice.

Table 5 shows examples of other applications of Se speciation presented in the literature during the time period covered by this ASU.

Table 5 Applications of speciation analysis: Se

Analyte species	Technique	Matrix	Sample treatment	Separation	LOD	Validation	Reference
Se^IV, total Se	Magnetic ionic liquid-based DLLME and GFAAS detection	Rice	Total Se: MAE, HNO3 : H2O2 then HCl added to reduce Se^VI to Se^VI. Se^IV: selective extraction using 0.2 mol L^−1 H2SO4	None, selective extraction procedure	0.018 ng L^−1	GBW10010, GBW10043, GBW10045, (all rice). Speciation by spike recovery, 95–108%	208
Se^IV, total Se	Solidified floating organic drop microextraction GFAAS detection	Water	Se^IV extracted in the pH range of 1–3	None, selective extraction procedure	0.19 μg L^−1	Se: LGC 6010, (water). Speciation by spiking, recovery 96–98%	209
Se^IV, Se^VI SeMet, SeCys2, MeSeCys	HPLC-ICP-MS	Wine and freeze-dried samples of leaves, grapes obtained after Se^VI foliar fortification of vineyards	2 g of wine and 0.25 g of leaves and grapes were treated protease type XIV, UPS 3 minutes, 3 hours, water bath at 37 °C	Hamilton, PRP-X100, gradient elution with increasing concentrations of NH4Ac (pH = 5.2, 2% methanol)	0.3–0.7 μg L^−1 for the five species tested	Spike recovery of 84–104% for iSe, SeMet and MeSeCys, 34% for SeCys2. Sum of Se species equalled the total Se	210
Se^IV, Se^VI SeMet, SeCys2, MeSeCys	2D-HPLC-ICP-MS	Se^IV enriched lettuces	Physiologically based extraction	AE, gradient elution. Fractions collected, then CEC, gradient elution	Not given	Mass balance approach. Sum of Se species accounting for 90% of the total Se	211

---

3.15 Sulfur

An analytical procedure based on RP-HPLC-ICP-QQQ-MS was developed for determining ergothioneine in biological matrices.²¹² Chromatographic separation was performed using a RP column (Gemini C6-Phenyl 4.6 × 150 mm) and a methanol gradient as mobile phase flowing at 0.5 mL mL⁻¹. With the aim of decreasing separation time due to the presence of strongly retained sulfur species, a 6-port valve was introduced to direct the column effluent either to the ICP-QQQ-MS or to the waste. The developed separation reduced the analysis time from 120 to 30 minutes. After separation, ergothioneine and other S containing species were detected using an ICP-QQQ-MS, with an O₂ reaction gas flow rate of 0.3 mL min⁻¹ so that a mass shift of 16 was used to monitor the S signal. The developed method provided LOD and LOQ values for ergothioneine of 0.23 and 0.80 μg S per L, respectively which are slightly lower than those provided by LC-MS/MS. The method was successfully applied to detect ergothioneine in ergothioneine-treated HPG 2 cell pellets. The presence of ergothioneine in this complex biological matrix was fully confirmed by LC-MS/MS. The applicability of the method for simultaneous monitoring of S (including ergothioneine) and Se species in food samples was also checked. Measurements by both RP-HPLC-ICP-QQQ-MS and LC-MS/MS measurements evidenced the presence of ergothioneine in water extracts of oyster mushrooms. Unfortunately, no Se signal was detected which was attributed to the use of water as extracting agent. In a second paper,²¹³ a study has been conducted to detect sulfur-containing metabolites in roots and shoots of Alium sativum by employing HPLC-ICP-QQQ-MS and high resolution LC-MS/MS in parallel. Plants were hydroponically grown and exposed to increasing levels of sulfate. S-compounds were extracted by using 1% (v/v) formic acid with an extraction efficiency of 84%. For the separation of the S-species, an Agilent Eclipse C18 (4.6 × 150 mm) column with a linear water: methanol gradient was employed. The effluent was split 1 : 3 after the column with 1 part being introduced into the ICP-QQQ-MS and the rest being introduced in the ES-MS. Chromatography revealed up to 54 separated S containing compounds, which constituted about 80% of the total S present in A. sativum. Three S-containing compounds were also identified and quantified for the first time.

3.16 Tellurium

Tellurium has attracted increasing analytical attention due to its wide industrial application such as the development of new materials (quantum dots for solar panels) and the production of fluorescent probes. However, very little is known on the environmental and human health impact of this rare element. Consequently, Te speciation plays a key role for better understanding its distribution, toxicity and bioavailability. Three papers have been publishing covering the topic of Te speciation. The first one reviews the existing knowledge on Te metabolism in plants and microorganisms based on LC-ICP-MS and LC-ESI-MS/MS data.²¹⁴ Similarly to Se, several Te metabolites such as MeTeCys, TeMet and TeCys has been detected and identified. Based on their own experimental data, the authors postulate a metabolic pathway of Te in plants. However, there are still many unknown metabolites that need to be identified. In a second paper Te speciation was performed in a contaminated soil from an abandoned mining area by using μ-XRF, μ-XRD and μ-XAFs techniques.²¹⁵ Results showed that Te is present in the soil tested as a mixture of Te^IV and Te^VI and structurally incorporated into Fe^III-hydroxides. According to the experimental K_D values obtained, Te^IV is less retained in soils than Te^V1, resulting in an easier release and migration of Te^IV than Te^VI. The results are of importance for understanding the solubility and mobility of Te in contaminated soils. Finally, other researchers developed a method based on HG-AFS for determining Te^IV and Te^IV in bioleaching solutions from low-grade Te ores treated with bacteria Acidithiobacillus ferrooxidans.²¹⁶ Factors affecting HG were investigated in detail. The best Te^IV HG efficiency was achieved by using 10% HCl and 2% NaBH₄. The method was applied to total Te determination prior to reduction of Te^VI to Te^IV with a mixture of 0.01% (m/v) KI and 10% (m/v) ascorbic acid as the reducing agents and 0.05% (m/v) 1,10-phenanthroline and 20 mg L⁻¹ Fe^III as masking agents. The Te^VI concentration was calculated by difference. A LOD value 0.209 μg L⁻¹ was reported. The speciation of Te in biolechates showed that Te^II in ores samples is oxidised to Te^IV and then to Te^VI by the action of Acidithiobacillus ferrooxidans.

3.17 Thallium

Thallium speciation in different matrices has been described in three papers. Two of them are based on the application of microextraction methodologies. In a first paper,²¹⁷ a DLLME method followed by FAAS measurements has been developed for Tl^I and Tl^III speciation in hair samples. Most of the paper is devoted to the optimization of the parameters affecting microextraction efficiency, such as pH of the sample solution, concentration of the chelating agent (1-(2-pyridylazo)-2-naphthol (PAN)), type and volume of dispersant and solvent and ionic strength. Speciation of Tl^III was carried out by extracting the Tl^III–PAN complex with a mixture containing 2.5 mL of ethanol and 100 μL of 1-dodecanol. After centrifuging, the organic solvent droplet was solidified over ice, separated from the solution and melted at room temperature. Nitric acid in methanol was added prior to FAAS analysis. The method was applied to total Tl determination prior to oxidation of Tl^I to Tl^III by means of a Ce^IV solution. The Tl^I concentration was calculated by difference. The enrichment factor was reported to be 42.7 and the reported LOD value for Tl^III was 21 μg L⁻¹. The method was validated for total Tl concentration by using BCR 288 (alloy sample). Extraction efficiencies of Tl^I and Tl^III from spiked hair samples were always close to 100%. In the second paper,²¹⁸ a DSPME for Tl speciation, using magnetic Fe₃O₄@SiO₂ NPs stabilised with palmitic acid, and FAAS determination has been presented. The adsorption was based on the interaction between the nanoparticles and a Tl^I dibenzo-18-crown-6 (DB18C6) complex. The resulting Tl loaded nanoparticles were separated from the solution with an external magnetic field followed by a treatment with HCl. Total Tl was determined after the reduction of Tl^III to Tl^I by hydroxylamine hydrochloride. The LOD and enrichment factor values were determined to be 0.85 μg L⁻¹ and 298, respectively. The proposed method was successfully validated by using a BCR 288 (alloy sample certified in total Tl). The method was applied to the determination of Tl species in water and waste water samples with recoveries between 98 and 102%. Finally, a method based on IC-ICP-MS was developed to determine Tl^I and Tl^III in surface waters and springs from a mining area.²¹⁹ The unstable Tl^III was complexed with DTPA to form the stable [Tl(DTPA)]²⁻ prior separation. Tl species were separated by using a cation exchange Dionex CG-2 (4.0 mm i.d. × 50 mm length) guard column. The mobile phase consisted of 5 mmol L⁻¹ HNO₃ with 3 mmol L⁻¹ NH₄NO₃ and 0.75 mmol L⁻¹ DTPA flowing at a 1.5 mL min⁻¹. The use of a CE guard column allowed authors to drastically decrease separation time while keeping optimum peak resolution and species separation in less than 2 minutes. Method validation was undertaken by analysing a CRM (NIST1640A). The concentration of Tl found (1.61 ± 0.08 μg L⁻¹) was in good agreement with the certified value (1.62 ± 0.016 μg L⁻¹). Besides of being thermodynamically unstable, a high level of Tl^III was found in tap water intended for human consumption.

3.18 Tin

Chung and Wu have comprehensively reviewed (107 references with titles) the determination of butyltins, phenyltins and octyltins in foods.²²⁰ The review covers species stability (in both standards and samples), extraction and clean methods, separation techniques with particular emphasis on GC and associated derivatisations, and various detection methods for GC. Separation by LC is also included. The reviewers offer several critical comments about which methods do not preserve the speciation and point out that many researchers report spike recovery results that are outside the acceptable limits of 70–110%. Of the various derivatisation methods that have been proposed, HG is not recommended. Alkylation by tetraethylborate would appear to be the most satisfactory. The reviewers conclude that the sensitivity and separation power of LC methods are inferior to those of GC for which MS and ICP-MS are the best detectors. They point out that MS–MS offers better sensitivity, a much lower probability of false positives than other selective GC detectors, and the possibility of identifying unknown OTC. The review concludes with a summary table highlighting some 36 published methods.

Although the use of TBT in antifouling formulations for use on small craft has now been banned for nearly 30 years, organotin speciation in antifouling paint is still attracting some attention. In one such study,²²¹ 23 paint samples were scraped from recreational boats from three countries around the Baltic Sea and analysed for total Sn and OTCs. Two antifouling paint products were also subjected to the same analyses. Methods for the detection of both Sn and OTCs in paint flake samples were developed and found to yield more accurate results compared to four different acid digestion methods. The hull paint samples had Sn concentrations ranging from 25 to 18 000 mg kg⁻¹ paint and TBT was detected in all samples with concentrations as high as 4.7 g (as Sn) per kg paint. However, TBT was not always the major OTC. Triphenyltin was abundant in many samples, especially in those originating from Finland. Several other compounds such as MBT, DBT, MPhT, DPhT and tetrabutyltin were also detected. These could have resulted from the degradation occurring on the hull or from impurities in the paint as they were also identified in the two analysed paint products. A correlation of 0.934 was found between the total tin content and the sum of all detected OTCs. The authors suggest that existing methods for OTC analysis of sediment may not offer full recoveries of OTCs if paint flakes are present in the sample. A review of OTCs in seafood, found as a result of environmental exposure to butyltins and phenyltins has also been published²²⁰ (further details in Section 1 of this review).

The determination of TBT in seafood using magnetic molecularly imprinted polymers coupled with HPLC-ICP-MS has been reported.²²² In the study, Fe₃O₄ was adopted as a carrier for surface molecular imprinting with two-stage polymerization. First, the functional monomer methacrylic acid was modified on the surface of Fe₃O₄, which was then polymerised with the template molecule TBT, cross linking agent ethylene glycol dimethacrylate and porogen (acetonitrile). Compared with conventional molecular imprinting polymers, this polymer showed significant selective adsorption and can be rapidly separated by an external magnetic field, with a shortened pre-treatment time. The LOD, recovery rate, and linear range were 1.0 ng g⁻¹, 79.74–95.72%, and 5 ng g⁻¹ respectively. The method was used to analyse extracts of Tegillarca granosa mussels, large yellow croaker, and other seafood specimens for OTC content.

3.19 Vanadium

Two papers report on the detection of V species in the period covered by this update. The first of these details the development of an SPE method for the quantification of V^IV and V^V in water samples.²²³ In brief, a SAX column was pre-conditioned by with 2 mL of 99.9% methanol, 2 mL of HPW, 10 mL of 0.2 mol L⁻¹ EDTA and finally 20 mL HPW, followed by 10 mL of a water sample which was filtered through a 0.45 μm filter onto a SAX column. For field sampling the SPE column was then flushed with air to remove the water before storage and transportation to the laboratory. Subsequently, the retained V^IV was eluted with 10–15 mL of a mixture of methanol (4%), EDTA (5 mmol L⁻¹) and TBAH (2 mmol L⁻¹) at pH 4 whilst V^V was eluted with 15 mL of NH₄H₂PO₄ (80 mmol L⁻¹) at pH 8. Under these conditions a recovery of 100% was obtained for each species. The V content in each fraction was measured using an ICP-MS instrument operated in collision cell mode with He as the cell gas. The method was compared with an HPLC-UV method, with a mobile phase of NH₄H₂PO₄ (possibly 80 mmol L⁻¹) and EDTA (0.2 mol L⁻¹) at pH 8 with both V species eluting in 20 minutes. Statistical analysis, Students t-test, showed that the results obtained by both the SPE-ICP-MS and HPLC-UV methods were in agreement. A comparison was made between the V speciation of water samples processed in situ with SAX adsorption in the field and those transported to the laboratory, which took 15 days. In all cases significant conversion of V^IV to V^V during storage and transportation was observed for the unprocessed water samples which, as the authors point out, has implications for the validity of previously published results for V speciation in water samples where no sample preservation was incorporated into the analytical procedure. The content of V^IV ranged from 0.4–230 and V^V 0.9–1900 μg L⁻¹ in river and groundwater samples collected from various locations in Argentina. The second paper covered here reports on the identification of individual polyoxometalates (POM, {PMo_12−xV_xO₄₀}^−3−x, where x = 1–3) by HPLC-ICP-OES. Model systems of POMs were synthesised and separated on a C₁₈ column with a mobile phase containing acetonitrile, ethanoic acid, NaOH and TBAH. The proportions of each mobile phase component and elution programme are not given but the work is worth reporting as a rare example of the coupling of HPLC to ICP-OES. The HPLC eluent, 0.2 mL min⁻¹, was mixed, via a t-piece, with a deionised water diluent, 2 mL min⁻¹, before nebulisation and at least two different Mo and V emission wavelengths were monitored. The measured composition of the POMs were found to be in agreement, within uncertainty limits, with theoretical compositions and also the Mo : V stoichiometry obtained from ⁵¹V NMR measurements.

3.20 Zinc

The mechanisms of Zn sequestration and Zn transfer and storage in Typha Latifolia was evaluated by combining isotopic analysis and EXAFS and μ-XRF measurements.²²⁴T. Latifolia was selected as plant model as being currently present in wetlands. After collecting plants in an urban industrialised watershed, plant organs (stems, leaves, rhizomes, roots) were separated and analysed by the different techniques selected. For EXAFS analysis, plants organs were frozen in liquid nitrogen at −80 °C while they were air-dried for elemental and isotopic analysis. Spectra from EXAFS indicated the presence of tetrahedral and octahedral Zn complexes and Zn-thiol species. μ-XRF and HCl extraction showed the occurrence of iron plaques in rhizome and roots. From isotopic measurements, an enrichment in light isotopes for Zn sorbed on the plaque relative to the soil (Δ⁶⁶Zn_plaque-soil = −0.3 to −0.1‰) was detected suggesting the dissolution of ZnS (enriched in light isotopes) in the rhizosphere with subsequent Zn²⁺ sorption on the root plaque. Similarly, enrichment in light isotopes of stems relative to leaves (Δ⁶⁶Zn_stem-leaves = −0.2‰) evidenced the remobilisation of Zn via the phloem, from leaves back to the stems. The use of this multi-technique platform allowed authors to establish the mechanisms of interaction between Zn and plant and the reported results highlight the role of thiols in controlling Zn metabolisms. In another paper²²⁵ CE-ICP-MS was employed for determining two Zn-containing dithiocarbamate fungicides (Zn-dimethyldithiocarbamate, Ziram, and Zn-ethylenebisdithiocarbamate, Zineb) in cabbage leaves. Samples were spiked by spraying different amounts of Ziram and Zineb on the leaves. Afterwards, Zn compounds were ultrasonic assisted-extracted from leaves by using 0.1 mol L⁻¹ NaOH as extracting solution. CE experimental conditions are detailed. Due to the fact that Ziram and Zineb are neutral species, μ-cyclodextrine was used as modifier. Electrophoretic resolution was further improved by using DCTA as chelating agent. Under optimal conditions LOD values of 1.90 and 3.00 ng mL⁻¹ were obtained for Ziram and Zineb, respectively. Method validation was performed in spiked cabbage leaves with recovery values ranged within 95–107%.

4 Biomolecular speciation analysis

Over the current review period some important analytical hardware innovations relating to methods specifically applicable to the measurement and detection of small biomolecules, metalloproteins and metal–protein interactions have been described in the literature. This includes: the modification of specialist sample introduction systems to use with atomic spectrometry; the application of non-specific detection methods in combination with elemental MS; and the combination of different molecular MS sources to give orthogonal data.

A review²²⁶ detailing the current state-of-the-art of analytical techniques for characterizing purified metalloproteins and metal–protein complexes has covered 129 references in the period from 2000 to 2014. Interestingly, it details some of the more overlooked mainstream direct spectrometric methods used in biochemistry and inorganic chemistry to characterise purified proteins. Techniques such as NMR and XRD, can be used for assessing the ternary protein structure, whereas other direct approaches including circular dichroism are useful for secondary protein structural elucidation. To identify and characterise the metal–protein environment, techniques such as XAS and XANES are options, if available, and for assessing metal-binding domains and metal–protein coordination EPR, FTIR and fluorescence spectrometry have been employed. Details of work involving 1D- and 2D-PAGE, CE and liquid chromatography via SEC, AEC or RPC are also covered, including the on-line and off-line coupling to MS detectors. This includes the use of ICP, ES, MALDI and LA-ICP-MS. The authors make the important point that the integrity of the metal–protein binding must not be compromised throughout the whole analytical process and that in some cases, metals bind to proteins with high-affinity (metalloproteins) and the bond is not broken during the separation process. However, metal–binding protein interactions can be of low affinity and the metal–protein bound can easily be lost during the separation process. Examples of applications in the food, environmental and clinical areas that have employed these techniques are discussed. The review also covers more recent work on nanoparticle–protein interactions and the formation of the protein corona. Unfortunately the review does not cover methods developed specifically for the characterisation of metal–binding proteins, such as immobilised metal affinity chromatography which can be used to purify native proteins or selectively isolate proteins with metal binding sites. The inclusion of papers using detection methods developed to specifically determine metal binding proteins, such as substrate-assisted laser desorption (SALD) MS, would also have been useful.

The use of non-specific detectors in biomolecular speciation studies have become less commonly used now that specific detectors such as ICP-MS are more widely available to use as HPLC detectors. However, with the advent of spICP-MS for the measurement of nanoparticles, the advantages of techniques such as multi-angle light scattering (MALS) as non-specific protein detectors are being realised. Light scattering represents a powerful technique for determining the molar mass of macromolecules in solution as well as monitoring the presence and formation of aggregates. When used as a detector for SEC separations, the light scattering and concentration are measured for each eluting fraction, the absolute molar mass and size can therefore be determined independently of the elution position. This is particularly important for species with non-globular shapes or that interact with a SEC column; such species typically do not elute in a manner that might be described by a set of column calibration standards. It is also possible to completely characterise the mass and size from separation techniques such as IE or RPC, where there are no means to calibrate these techniques by traditional column calibration methods. This approach has been illustrated by the analysis of heparin,²²⁷ a highly sulfated glycosaminoglycan, and low molecular weight heparins (LMWHs) for QC purposes by SEC coupled to RI, MALS and ICP-MS. Used in sequence, these detectors provide qualitative molecular weight and quantitative mass content analysis, without the need for authentic standards. The ICP-MS detector was used to determine the cations, which ion-pair with the anionic polysaccharide chains of heparin and LMWHs, whereas the non-specific detectors provide information on the molecular weight and size of the anionic polysaccharides. Full molecular weight (MW) profiles and mass recoveries for three commercial heparin/LMWH products, heparin sodium, enoxaparin sodium and nadroparin calcium, were obtained and all showed higher MWs than previously reported.

Two papers that describe specific new analytical innovations in the area of mass spectrometry hardware for metallomic and metalloproteomic studies, make a welcome addition to the literature. The group of Preisler²²⁸ have followed on from an earlier study on multiply – hyphenated detection series, to produce another iteration, this time coupling an LC separation to ES with a split flow to a commercial MALDI target made of conductive plastic. The target plate is then analysed off-line in a MALDI-MS followed by SALD into an ICP-MS detector. The addition of ES to the series generates multiply charged protein ions and so offers an orthogonal detection method compared to the singly charged molecular ions generated in the MALDI and the elemental ions from the ICP source. Each of the soft ionization techniques, ES and MALDI, offers specific advantages: ES can be easily coupled on-line to an LC column and can be used for analysis of non-covalent protein–metal complexes; MALDI is a simple, quick and sensitive technique, which is generally useful for the identification of substances after direct analysis of their mixtures with a suitable absorbing matrix. The ES source also overcomes to some extent the loss of the metal ion from the protein which often occurs during ionisation in the MALDI source. Separations were performed on a conventionally sized RP HPLC column (Vydac C8 208MS 5415, 150 × 4.6 mm × 5 μm) using a gradient elution over 25 min. Buffer A was composed of 5 mM ammonium acetate (pH 7.0) in 3% (v/v) MeOH and buffer B contained 5 mM ammonium acetate (pH 7.0) in 50% (v/v) MeOH. The flow from the column was split such that 884 μL min⁻¹ was directed to the ES source and 16 μL min⁻¹ was deposited as spots on a plastic MALDI target (Prespotted AnchorChip 96, Bruker, Daltonics, Bremen, Germany). The MALDI matrix applied by the manufacturer was washed off and the LC effluent was deposited using a laboratory-built spotter with x, y, z stages. The effluent was spotted in the form of 1.6 μL droplets and dry fractions on the target were overlaid with a 0.5 μL of saturated solution of α-cyano-4-hydroxycinnamic acid in 20% ACN and 1% TFA (v/v) with 1 mg L⁻¹ In as the IS for the SALD-ICP-MS. The plate was initially desorbed in the MALDI instrument and then subsequently the same plate was analysed by SALD-ICP-MS to give the elemental composition of the spots. Off-line detection is less straightforward, but it has some advantages compared to on-line coupling including: decoupling of the separation from detection allows for their independent optimization; it is useful to make modifications to the deposited fractions prior to further analysis; storage of the spots is possible; and repeated analyses under various conditions or with other detection techniques can be utilised. The system was tested using a commercially available metallothionein-1 (MT-1) preparation isolated from rabbit liver (purity 98.2%) and chromatograms showed separation of complexes of three major isoforms, MT-1a, MT-2d and MT-2e which eluted at 10.6, 11.2 and 14.0 min, respectively. No apo-protein peaks were observed in the ES mass spectra. The dominant charge states were +4 to +7, the number of detected metal ions in the MT-1 complexes were 2 to 6 for Cd and 1 to 5 for Zn with the total number of metal ions in the complexes 7. Three-point calibration using standards of 0.1, 1.0 and 10.0 mg L⁻¹ in five replicates for each yielded typically R² of 0.999. Although the analyte signals were normalised to an In internal standard the repeatability for the Zn and Cd concentrations was quite poor at between 26 and 62% RSD, which can only be described as qualitative. However, it would be interesting to see results from this approach in much simpler scenarios using single protein standards to determine what the system capabilities are for quantitative analysis. No data was given to show how the MALDI step might have affected the amount of material available in the subsequent SALD-ICP-MS analysis, or how this might affect the analytical parameters, particularly the accuracy.

The second paper, from the group led by Doble,²²⁹ describes the use of a microfluidic chip designed for nanospray organic MS in proteomics studies, but instead coupled to an ICP-MS instrument. The authors claim that by using lower flow rates and consequently lower volumes of organic solvents, they can overcome a number of issues related to plasma instability and carbon blocking the sampler cones. However, lowering eluent flow rates to nanoflow causes different issues to be problematic, such as: frequent blocking of the capillary column; requirement for sample pre-concentration with concentrator columns prior to analysis; and the need to minimise dead volumes to avoid band broadening, which can lead to poor separation efficiency. Hyphenation of nanoLC to ICP-MS involves transport of eluents with flow rates as low as 100 nL min⁻¹ into the plasma, requiring dedicated interfaces with sheath flows to boost flow rates to levels suitable for nebulisation; or employment of sheathless total consumption direct-injection nebulisers, which can also be difficult to use consistently. Some of these practical problems can be overcome by using commercially fabricated microfluidic chip devices, which can dramatically improve separation performance, efficiency and, most importantly, versatility, compared to laboratory built systems. In this study a HPLC-Chip Cube interface (Agilent Victoria, Australia), with a reusable chip was coupled to a HPLC equipped with a nano-pump system. The microfluidic nanospray device was modified by removal of the electrospray and the flow was diverted to an unused stator port on the chip cube interface prior to introduction into the nebuliser. The microfluidic chip consisted of a Polaris C18, 3 μm packing material, a 160 nL enrichment column and a 150 mm analytical column. A total consumption nebuliser (Teledyne CETAC, Omaha, NE USA) was modified to reduce post-column band broadening and the low volume spray chamber was adapted to introduce a makeup flow of argon to aid transport of the eluent aerosol into the plasma at the low flow rates used and stop losses on the spray chamber walls. The instrumental setup was evaluated using a simple experimental procedure for the measurement of cyanocobalamin in horse plasma. It displayed good linearity of 0.9999 with a seven point calibration range from 10 to 1000 μg L⁻¹ and peak area and retention time RSDs of 1.9% and 0.2% respectively. The LOD was 14 μg L⁻¹ (0.57 μg L⁻¹ Co), which was consistent with other methods using larger i.d. columns of 150 mm × 1 mm with direct injection nebulisation. No carbon deposition was observed after 6 weeks of method development and no drift was observed in the baseline. It will be interesting to see how the system handles more complex samples containing a biological matrix and how this affects the resolution between analytes, which was not tested in the current work and the long term routine reliability, particularly in relation to system blockages.

The measurement of intact metalloproteins in multi-cellular samples such as bacteria, fungi or tissue requires the sample to be aggregated into cells, which are then lysed to facilitate release of the metalloprotein analytes contained within. It is good to see this most difficult part of the analytical process being given some consideration,²³⁰ with an investigation of the optimal conditions for cell lysis to study the Zn metalloproteome of the fungus Histoplasma capsulatum, which can cause lung disease in exposed individuals. The deprivation of Zn in relation to this fungus forms part of the host defence mechanism utilised by macrophages, hence Zn and its associated binding proteins are important in understanding infection and treatment. Various cell lysis techniques were investigated in an effort to both maintain the metalloproteins during lysis and subsequent analysis while, at the same time, serving to be strong enough to break the cell wall, allowing access to cytosolic metalloproteins. Seven lysis approaches were assessed, including: glass homogeniser; bead beater; sonication probe; vortex with and without Triton X-100 (1% v/v); and sonication bath followed by vortex, with and without Triton X-100 (1% v/v). A Qubit fluorometric assay was used to determine the concentration of nucleic acids and proteins in the lysate and the concentration of Zn over a wide molecular range was measured using SEC coupled to ICP-MS. As demonstrated by the highest Zn to protein ratio, 1.030 ng Zn per μg protein and Zn distribution among high, mid, and low molecular weights, the lysis method involving vortexing the yeast cells with 500 μm glass beads in Triton X-100 (1% v/v), was found to extract the Zn metalloproteome with the least denaturation.

The absolute quantitation of peptides and intact proteins without the need for authentic molecular standards, by “direct” heteroatom analysis using the measurement of P or S by chromatography coupled to ICP-MS, has been the focus of a number of studies, which highlight the range of different application areas where it can be used. The measurement of phosphorylated proteins is a good example of how important ICP-MS has become as an HPLC detector in macromolecular speciation studies, mainly because of its ability as a generic detector for a specific class of analyte, in this case any compound containing P. Its characteristics for quantitation with only an elemental standard and provision of low LODs, even in difficult biological matrices is unsurpassed in the lexicon of MS instrumentation. Whilst the equivalent LC-MS/MS based approach would provide the necessary structural characterisation of the analyte, it would require authentic standards for full quantitation and ion generation would be suppressed by the matrix. Work has recently been reported²³¹ on the development of a hollow fibre (HF) supported TiO₂ monolithic microextraction device, in combination with capillary HPLC coupled to ICP-MS, for the measurement of phosphopeptides via³¹P¹⁶O⁺. Due to the low abundance of phosphoproteins relative to their unphosphorylated counterparts and interference from the sample matrix, sample pre-concentration is required. The membrane pores (approx. 200 nm) of the HF effectively carry more TiO₂ on the surface and under optimal conditions, a 100-fold enrichment factor was possible and the monoliths could be re-used. By monitoring ³¹P¹⁶O with O₂ as a reaction gas in the CRC, polyatomic mass interference at m/z 31 was avoided and an LOD of 2.9 nmol L⁻¹ as P was achieved. The stationary phase was a capillary C₁₈ column (Acclaim PepMap 100C18, 300 μm i.d. 3 × 150 mm), and the mobile phase consisted of 0.1% formic acid (solvent A) and 0.1% formic acid in acetonitrile (solvent B) and a gradient program was employed for the separation. The capHPLC was connected to either a CRC ICP-MS or a ESI-Q-TOF-MS via a short length of 50 μm i.d. fused silica capillary. The system was evaluated using a model phosphorylated protein β-casein, which was digested with trypsin prior to measurement of 11 P-containing peptides by ICP-MS, of which only 9 were present at sufficiently high concentrations to allow identification by ESI. In another example of the use of ICP-MS as a “direct” heteroatom detector, this time for the accurate quantification of modified cyclic peptides, S was used to determine the peptides.²³² In this case the method was less exacting, only requiring quantification for calculation of the reaction yield and for the downstream biological testing of the reaction products. Two methods were used, the first involved the use of HPLC coupled in parallel to ES and ICP-MS to quantify the S-containing peptides obtained after extraction and purification of the cyclic peptides from chemoenzymatic reaction mixtures, the second used NMR to quantify the solution concentration of the non-sulfur containing cyclic peptides. The overall methodology made the quantification of new compounds much easier, as authentic standards were unavailable. Following on from work reviewed last year on “absolute venometrics”, a refinement of the method has been published.²³³ Using triple quadrupole ICP-MS and high resolution molecular MS, it has extended the analysis of snake venom from 25 compounds measured in a single species, to other snake species. Once again the chromatographic resolution of the individual S-containing species restricted total identification of all the compounds. Generally in the biomolecular speciation literature this is the limiting factor and more work needs to be undertaken to develop high resolution separation methods for coupling to the power of ICP-MS detection.

Another recently developed approach to the analysis of peptides²³⁴ by ICP-MS, but one which will have much wider application when fully validated, involves using elemental labelling for the specific detection of a biologically transient species, in this case cysteine sulfenic acid groups (SA, Cys-SOH) in post-translationally modified peptides. These groups are important indicators of reactive oxidation species, which are implicated as causative agents in age related disease states such as Alzheimer’s disease. In a proof of concept report the SA group was generated in model peptides by oxidation with H₂O₂ and then tagged by the linear alkyne beta-ketoester (KE) linked to a lanthanide (Ln)-containing chelator (Ln-DOTA, 1,4,7,10-tetraazacyclododecane-1,4,7,10-tetraacetic acid). The linking of the KE to DOTA was performed by click chemistry, resulting in a new reagent (Ln-DOTA-KE) that permits highly sensitive elemental (ICP-MS) and molecular (ESI-MS/MS) detection of SA groups present in proteins. The reagent reacts specifically with SA, offering improved reactivity at physiological pH, an apparently simple derivatisation step and a cell-membrane-permeable compound that has promising future applications. It will be interesting to see whether this generic approach to the measurement of biomolecular species that show a high degree of reactive oxidation, correlates with the diseased state in illnesses associated with this as a causative agent.

4.1 Interaction of metallodrugs with biological systems

The investigation and measurement of the interaction of metallodrugs containing Pt or Ru with biological systems has been the focus of research interest for many years. Recent studies have looked at quantitation, fundamentals of drug–DNA interaction and hyperspectral image analysis of tumour models. A high accuracy species-specific HPLC-ICP-MS method²³⁵ for the quantitation of carboplatin DNA adducts in a lung cell line has been developed. The method involved the use of micro-flow RPC coupled to a SF-ICP-MS instrument (R = 300), achieving an LOD of 0.2 ng Pt per mg DNA. A low LOD was essential for this work because of the known poor reactivity of the drug towards DNA. To perform IDA, carboplatin–GG calibrants and spikes (¹⁹⁴Pt-enriched GG adducts) were synthesised and then characterised for Pt mass fraction, Pt distribution and structural composition. To assess the accuracy of the developed procedure in the absence of a CRM, a “reference” sample prepared by incubation of calf thymus DNA with carboplatin was characterised in house for its total P concentration, Pt content and Pt–GG concentration. This was analysed in parallel, as a form of IQC during the investigation. It will be interesting to see the results from clinical trials involving human samples and how useful such a high accuracy approach is in the clinic. The group of Timerbaev has continued to develop new approaches²³⁶ to the assessment of the properties of a proposed anticancer Ru-based drug, indazolium trans-[tetrachloridobis(1H-indazole)ruthenate(III)], in this case how it interacts with a DNA oligonucleotide, in these model experiments 5′-GTCGTACTGATACATGAGCC-3′; 6117 Da, was used. The novelty in the work was the use of affinity CE, to study the Ru DNA-adduct formation in situ using a capillary filled with an oligonucleotide-containing electrolyte. To further characterise the binding kinetics, a drug–oligonucleotide mixture was incubated for different periods of time, and then ultrafiltration used to separate the sample into two different molecular weight size fractions (>3 and <3 kDa), which were analysed by conventional CE. The time-dependent distribution profiles of the Ru drug were then assessed by CE-ICP-MS, revealing that at least two DNA adducts were present. For coupling both CE systems, a torch with a smaller inlet (1.5 mm) was utilised to minimise the influence of plasma backpressure on the electrophoretic flow and a micro-concentric nebuliser CEI-100 (CETAC, Omaha, NE, USA), was used as the ICP-MS interface. The conventional CE (HP3D CE system, Agilent, Waldbronn, Germany) and the affinity CE (CAPEL 105, Lumex, St. Petersburg, Russia) was both performed using fused-silica capillaries of inner diameter 75 μm × 70 cm and samples were introduced into the capillary hydrodynamically. The background electrolyte used was 10 mMNaH₂PO₄ to 10 mM Na₂HPO₄, 4 mM NaCl, pH 6.0 and the affinity CE separation contained 0.2 μm of the oligonucleotide. A high voltage (10 kV) with a positive polarity was placed at the inlet end of the capillary and a sheath flow of 1 mM phosphate buffer (pH 6.0), 0.4 mM NaCl, 20 μg L⁻¹ Ge as IS, was used to generate the separations. This approach to monitoring metallodrugs has potential for studying their interaction with cellular constituents and for assessing their intracellular fate. An important demonstration²³⁷ of spatial analysis using a combination of elemental and molecular methods for the investigation of metallodrug interactions to support histopathological studies, has been realised by the incorporation of multivariate statistical methods. Using thin cuts of different tumours and FTIR spectroscopy combined with LA-ICP-MS it was possible to show a correlation between alterations in the average protein secondary structure and Pt distribution, as well as between changes in the cell nuclear morphology and a reduction of physiologically relevant trace elements. FTIR spectroscopy is a non-destructive and label free technique which provides molecular species information. The resonant interaction of molecules with mid-IR radiation results in absorbance spectra and allows for the detection of molecular vibrations characteristic of chemical bonds. The IR spectrum carries information regarding molecular conformation and aggregation of biomolecules, such as proteins and nucleic acids. It does not provide any information on the distribution of elements, this is provided by the complementary technique of LA-ICP-MS. The results of this combined analysis suggested different degrees of tumour viability. Further studies are warranted to combine molecular and elemental results and bioinformatics to provide the fullest spatially resolved speciation data.

Conflicts of interest

There are no conflicts of interest to declare.

5 Abbreviations used in this update

AAS	atomic absorption spectrometry
AB	arsenobetaine
AC	arsenocholine
ACN	acetonitrile
AE	anion exchange
AEC	anion exchange chromatography
AF	atomic fluorescence
AFS	atomic fluorescence spectrometry
APDC	ammonium pyrrolidine dithiocarbamate
ASA	p-arsanilic acid
ASU	atomic Spectrometry Update
CCT	collision cell technology
CE	capillary electrophoresis
cGPx	cellular glutathione peroxidase
CPE	cloud point extraction
CRM	certified reference material
CS	continuum source
CV	cold vapour
DART-HRAMS	direct analysis in real time high-resolution accurate mass spectrometry
DBD	dielectric barrier discharge
DBT	dibutyltin
DCM	dichloromethane
DDTC	diethyldithiocarbamate
DLLME	dispersive liquid–liquid microextraction
DMA	dimethylarsenic
DMDTAV	dimethyldithioarsinic acid
DNA	deoxyribonucleic acid
DPhT	diphenyltin
DRC	dynamic reaction cell
DSPME	dispersive solid phase microextraction
DTPA	diethylenetriaminepentaacetic acid
EDTA	ethylenediaminetetraacetic acid
EDX	energy dispersive X-ray
EPR	electron paramagnetic resonance
ES-MS	electrospray mass spectrometry
ET-AAS	electrothermal atomic absorption spectrometry
EtHg	ethylmercury
EU	European union
EXAFS	extended X-ray absorption fine structure
FT	Fourier transform
GC	gas chromatography
GPX	glutathione peroxidase
GSH	glutathione
Hb	haemoglobin
HDPE	high density polyethylene
HEP	heparin
HG	hydride generation
HILIC	hydride generation
HILIC	hydrophilic interaction liquid
HPLC	high performance liquid chromatography
HR	high resolution
IAEA	International atomic energy agency
iAs	inorganic arsenic
ICP-MS	inductively coupled plasma mass spectrometry
ICP-OES	inductively coupled plasma optical emission spectrometry
id	internal diameter
ID	isotope dilution
IDA	iminodiacetic acid
IE	ion exchange
iHg	inorganic mercury
IL	ionic liquids
IR	infra-red
IRMS	isotope ratio mass spectrometry
KHP	potassium hydrogen phthalate
LA	laser ablation
LC	liquid chromatography
LLE	liquid–liquid extraction
LOD	limit of detection
LOQ	limit of quantification
MAE	microwave assisted extraction
MALDI	matrix-assisted laser desorption ionization
MBT	monobutyltin
MeHg	methylmercury
MeOH	methanol
MeSeCys	methylselenocysteine
MeSeOCys	semethylselenocysteine oxide
MeTeCys	methyltelurocysteine
MMA	monomethylarsenic
MP	microwave induced plasma
MPhT	monphenyltin
MS	mass spectrometry
MT	metallothionein
NaBPr4	sodium(tetra-n-propyl)borate
NIST	National Institute of Standards and Technology
NMR	nuclear magnetic resonance
NP	nanoparticle
NRCC	National Research Council of Canada
OTC	organotin compounds
PAGE	polyacrylamide gel electrophoresis
PC	phycocianine
PEEK	polyetheretherketone
PET	polyethyleneterephthalate
PhHg	phenylmercury
Q	quadrupole
ReCCS	reference material institute for clinical chemistry standards
ROX	roxarsone
RP	reverse phase
RSD	relative standard deviation
SAX	strong anion exchange
SeAlb	selenoalbumin
SEC	size exclusion chromatography
SeCys	selenocysteine
SeCys2	selenocystine
SeHLan	selenohomolanthionine
SEM	scanning electron microscopy
SeMet	selenomethionein
sp	single particle
SPE	solid phase extraction
SPME	solid phase microextraction
SRM	standard reference material
SR-XANES	synchrotron radiation X-ray absorption near edge structure
ss	species specific
TBAH	tetrabutyl ammonium hydroxide
TBT	tributyltin
TeCys	telurocysteine
TeMet	telluromethionein
TFA	trifluoroacetic acid
TIMS	thermal ionization mass spectrometry
TMAH	tetramethylammonium hydroxide
TMAO	trimethylarsine oxide
TMSb	trimethylantimony
TOF	time of flight
TTA	thenoyltrifluoracetone
TXRF	total reflection X-ray fluorescence
US EPA	United States Environmental Protection Agency
US FDA	United States Food and Drug Administration
UV	ultraviolet
XAFS	X-ray absorption fine structure spectrometry
XANES	X-ray absorption near-edge structure
XAS	X-ray absorption spectroscopy
XRD	X-ray diffraction
XRF	X-ray fluorescence

6 References

1. R. Clough, C. F. Harrington, S. J. Hill, Y. Madrid and J. F. Tyson, J. Anal. At. Spectrom., 2017, 32(7), 1239–1282.
2. O. T. Butler, W. R. L. Cairns, J. M. Cook and C. M. Davidsond, J. Anal. At. Spectrom., 2017, 32(1), 11–57.
3. A. Taylor, N. Barlow, M. P. Day, S. Hill, M. Patriarca and M. White, J. Anal. At. Spectrom., 2017, 32(3), 432–476.
4. E. H. Evans, J. Pisonero, C. M. M. Smith and R. N. Taylor, J. Anal. At. Spectrom., 2017, 32(5), 869–889.
5. M. West, A. T. Ellis, C. Streli, C. Vanhoof and P. Wobrauschek, J. Anal. At. Spectrom., 2017, 32(9), 1629–1649.
6. S. Carter, A. Fisher, R. Garcia, B. Gibson, J. Marshall and I. Whiteside, J. Anal. At. Spectrom., 2016, 31(11), 2114–2164.
7. Inorganic Trace Analytics: Trace Element Analysis and Speciation, ed. H. Matusiewicz, de Gruyter, Berlin, 2018.
8. Handbook of Rare Earth Elements: Analytics, ed. A. Golloch, de Gruyter, Berlin, 2017.
9. H. Haraguchi, Metallomics, 2017, 9(8), 1001–1013.
10. C. M. Ackerman, S. Lee and C. J. Chang, Anal. Chem., 2017, 89(1), 22–41.
11. D. Pozebon, G. L. Scheffler and V. L. Dressler, J. Anal. At. Spectrom., 2017, 32(5), 890–919.
12. H. Wang, M. He, B. B. Chen and B. Hu, J. Anal. At. Spectrom., 2017, 32(9), 1650–1659.
13. Z. R. Liu, X. T. Li, G. Y. Xiao, B. B. Chen, M. He and B. Hu, TrAC, Trends Anal. Chem., 2017, 93, 78–101.
14. H. Wang, X. L. Liu, K. Nan, B. B. Chen, M. He and B. Hu, J. Anal. At. Spectrom., 2017, 32(1), 58–77.
15. O. Chahrour and J. Malone, Protein Pept. Lett., 2017, 24(3), 253–266.
16. M. Marcinkowska and D. Baralkiewicz, Talanta, 2016, 161, 177–204.
17. A. Tycova, V. Ledvina and K. Kleparnik, Electrophoresis, 2017, 38(1), 115–134.
18. S. S. Aleksenko, M. Matczuk and A. R. Timerbaev, Electrophoresis, 2017, 38(13–14), 1661–1668.
19. H. A. Castillo-Michel, C. Larue, A. E. P. del Real, M. Cotte and G. Sarret, Plant Physiol. Biochem., 2017, 110, 13–32.
20. J. C. Masini and F. Svec, Anal. Chim. Acta, 2017, 964, 24–44.
21. M. He, L. J. Huang, B. S. Zhao, B. B. Chen and B. Hu, Anal. Chim. Acta, 2017, 973, 1–24.
22. T. Yang, X. Y. Wang, L. Y. Wang, M. L. Chen and J. H. Wang, Anal. Methods, 2016, 8(47), 8251–8261.
23. A. R. Turker, Turk. J. Chem., 2016, 40(6), 847–867.
24. T. A. Maryutina and A. R. Timerbaev, Anal. Chim. Acta, 2017, 991, 1–8.
25. F. Dewi, L. P. Sim, J. Wang, B. M. K. Tong, R. Y. C. Shin and T. K. Lee, Int. J. Mass Spectrom., 2016, 411, 14–20.
26. K. Kubachka, D. T. Heitkemper and S. Conklin, J. AOAC Int., 2017, 100(4), 1077–1085.
27. B. Daus and H. R. Hansen, Environ. Chem., 2016, 13(6), 913–918.
28. S. Carneado, J. F. Lopez-Sanchez, A. Sahuquillo, E. Klontzas, G. E. Froudakis and S. A. Pergantis, J. Anal. At. Spectrom., 2017, 32(6), 1109–1118.
29. M. Marcinkowska, W. Lorenc and D. Baralkiewicz, Microchem. J., 2017, 132, 1–7.
30. Y. A. Lin, S. J. Jiang and A. C. Sahayam, Food Chem., 2017, 230, 76–81.
31. A. Mestrot, Y. Ji, S. Tandy and W. Wilcke, Environ. Chem., 2016, 13(6), 919–926.
32. P. Kappen, G. Ferrando-Miguel, S. M. Reichman, L. Innes, E. Welter and P. J. Pigram, J. Hazard. Mater., 2017, 329, 131–140.
33. I. Lopez-Garcia, S. Rengevicova, M. J. Munoz-Sandoval and M. Hernandez-Cordoba, Talanta, 2017, 162, 309–315.
34. V. Romero, L. Vilas, I. Lavilla and C. Bendicho, J. Anal. At. Spectrom., 2017, 32(9), 1705–1712.
35. A. Gholami and H. Noorizade, Bulg. Chem. Commun., 2016, 48, 36–42.
36. Y. T. Li, W. Guo, A. K. Souders, L. L. Jin, Y. Q. Ke, Q. H. Guo and S. H. Hu, RSC Adv., 2016, 6(110), 108247–108254.
37. V. Taylor, B. Goodale, A. Raab, T. Schwerdtle, K. Reimer, S. Conklin, M. R. Karagas and K. A. Francesconi, Sci. Total Environ., 2017, 580, 266–282.
38. W. R. Cullen, Q. Q. Liu, X. F. Lu, A. McKnight-Whitford, H. Y. Peng, A. Popowich, X. W. Yan, Q. Zhang, M. Fricke, H. S. Sun and X. C. Le, J. Environ. Sci., 2016, 49, 7–27.
39. T. Llorente-Mirandes, R. Rubio and J. F. Lopez-Sanchez, Appl. Spectrosc., 2017, 71(1), 25–69.
40. F. Cubadda, B. P. Jackson, K. L. Cottingham, Y. O. Van Horne and M. Kurzius-Spencer, Sci. Total Environ., 2017, 579, 1228–1239.
41. Q. L. Fu, C. Liu, V. Achal, Y. J. Wang and D. M. Zhou, in Advances in Agronomy, vol. 140, ed. D. L. Sparks, Elsevier Academic Press Inc, San Diego, 2016, pp. 1–41.
42. M. Welna and P. Pohl, J. Anal. At. Spectrom., 2017, 32(9), 1766–1779.
43. A. Abbaszadeh and A. Tadjarodi, RSC Adv., 2016, 6(114), 113727–113736.
44. B. S. Pinheiro, L. L. Gimenes, A. J. Moreira, A. F. de Araujo, C. D. Freschi and G. P. G. Freschi, J. Environ. Sci. Health, Part A: Toxic/Hazard. Subst. Environ. Eng., 2017, 52(11), 1089–1098.
45. J. Arndt, G. Ilgen and B. Planer-Friedrich, J. Volcanol. Geotherm. Res., 2017, 331, 16–25.
46. D. B. Wu, B. J. Purnomo and S. P. Sun, J. Geochem. Explor., 2017, 173, 85–91.
47. K. Hug, W. A. Maher, S. Foster, F. Krikowa and J. W. Moreau, Microchem. J., 2017, 130, 162–167.
48. W. H. Liu, Y. Mei, B. Etschmann, J. Brugger, M. Pearce, C. G. Ryan, S. Borg, J. Wykes, P. Kappen, D. Paterson, U. Boesenberg, J. Garrevoet, G. Moorhead and G. Falkenberg, Geochim. Cosmochim. Acta, 2017, 196, 144–159.
49. K. Tokunaga, T. Uruga, K. Nitta, Y. Terada, O. Sekizawa, S. Kawagucci and Y. Takahashi, Chem. Geol., 2016, 447, 59–69.
50. T. D. Sowers, J. M. Harrington, M. L. Polizzotto and O. W. Duckworth, Geochim. Cosmochim. Acta, 2017, 198, 194–207.
51. L. Rajpert, B. A. Kolvenbach, E. M. Ammann, K. Hockmann, M. Nachtegaal, E. Eiche, A. Schaffer, P. F. X. Corvini, A. Sklodowska and M. Lenz, Environ. Sci. Technol., 2016, 50(17), 9124–9132.
52. H. Guenet, M. Davranche, D. Vantelon, M. Bouhnik-Le Coz, E. Jarde, V. Dorcet, E. Demangeat and J. Jestin, Geochim. Cosmochim. Acta, 2017, 203, 284–302.
53. S. R. Hettiarachchi, W. A. Maher, F. Krikowa and R. Ubrihien, Environ. Geochem. Health, 2017, 39(1), 209–219.
54. V. G. Mihucz, L. Bencs, K. Koncz, E. Tatar, T. Weiszburg and G. Zaray, Spectrochim. Acta, Part B, 2017, 127, 30–35.
55. F. L. Li, H. M. Guo, X. Q. Zhou, K. Zhao, J. X. Shen, F. Liu and C. Wei, Chemosphere, 2017, 168, 777–785.
56. T. G. Asere, K. Verbeken, D. A. Tessema, F. Fufa, C. V. Stevens and G. Du Laing, Environ. Sci. Pollut. Res., 2017, 24(25), 20446–20458.
57. C. H. Lin, Y. Chen, Y. A. Su, Y. T. Luo, T. T. Shih and Y. C. Sun, Anal. Chem., 2017, 89(11), 5892–5900.
58. H. Shirkhanloo, M. Ghazaghi, A. Rashidi and A. Vahid, Microchem. J., 2017, 130, 137–146.
59. H. Rasheed, P. Kay, R. Slack, Y. Y. Gong and A. Carter, Sci. Total Environ., 2017, 584, 631–641.
60. D. P. Wang, Y. S. Shimoda, S. X. Wang, Z. Wang, J. Liu, X. Liu, H. Y. Jin, F. Gao, J. Tong, K. Yamanaka, J. Zhang and Y. An, Environ. Health Prev. Med., 2017, 22(1), 9.
61. J. Baker and D. Wallschlager, J. Environ. Sci., 2016, 49, 169–178.
62. X. M. Xue, Y. Yan, C. Xiong, G. Raber, K. Francesconi, T. Pan, J. Ye and Y. G. Zhu, Environ. Pollut., 2017, 228, 111–117.
63. J. Zhang, Y. Xu, T. T. Cao, J. Chen, B. P. Rosen and F. J. Zhao, Plant Soil, 2017, 416(1–2), 259–269.
64. E. G. Duncan, W. A. Maher, S. D. Foster, F. Krikowa, C. A. O’Sullivan and M. M. Roper, Chemosphere, 2017, 181, 44–54.
65. M. G. Esperanza, E. Y. Barrientos, K. Wrobel, F. J. A. Aguilar and A. R. Escobosa, Anal. Methods, 2017, 9(13), 2059–2068.
66. M. Stiboller, G. Raber, E. L. F. Gjengedal, M. Eggesbo and K. A. Francesconi, Anal. Chem., 2017, 89(11), 6266–6272.
67. C. S. Huber, M. G. R. Vale, M. B. Dessuy, M. Svoboda, S. Musil and J. Dedina, Talanta, 2017, 175, 406–412.
68. J. Tremlova, M. Sehnal, J. Szakova, W. Goessler, O. Steiner, J. Najmanova, T. Horakova and P. Tlustos, Arch. Agron. Soil Sci., 2017, 63(7), 918–927.
69. A. Gerenyi, V. Czech, F. Fodor, L. Vincze and I. Szaloki, X-Ray Spectrom., 2017, 46(3), 143–150.
70. S. Z. Chen, Q. Z. Guo and L. P. Liu, Food Analytical Methods, 2017, 10(3), 740–748.
71. M. M. Nearing, I. Koch and K. J. Reimer, J. Environ. Sci., 2016, 49, 140–149.
72. M. L. Zhu, X. C. Zeng, Y. X. Jiang, X. T. Fan, S. H. Chao, H. B. Cao and W. S. Zhang, Chemosphere, 2017, 168, 1677–1683.
73. A. A. Saucedo-Velez, L. Hinojosa-Reyes, M. Villanueva-Rodriguez, A. Caballero-Quintero, A. Hernandez-Ramirez and J. L. Guzman-Mar, Food Chem., 2017, 232, 493–500.
74. L. M. Frensemeier, L. Buter, M. Vogel and U. Karst, J. Anal. At. Spectrom., 2017, 32(1), 153–161.
75. A. J. Geng, X. Wang, L. S. Wu, F. H. Wang, Y. Chen, H. Yang, Z. Zhang and X. L. Zhao, Ecotoxicol. Environ. Saf., 2017, 137, 172–178.
76. S. K. Saxena, A. R. Karipalli, A. A. Krishnan, R. Rangasamy, P. Malekadi, D. P. Singh, V. Vasu and V. K. Singh, J. AOAC Int., 2017, 100(3), 598–602.
77. P. J. Gray, C. K. Tanabe, S. E. Ebeler and J. Nelson, J. Anal. At. Spectrom., 2017, 32(5), 1031–1034.
78. W. A. Maher, S. Eggins, F. Krikowa, R. Jagtap and S. Foster, J. Anal. At. Spectrom., 2017, 32(6), 1129–1134.
79. H. Cui, Q. L. Zhu, L. Z. Dong, B. T. Shan, T. T. Han, H. X. Li and F. Cai, Anal. Lett., 2017, 50(6), 1040–1048.
80. T. Narukawa, K. Chiba, S. Sinaviwat and J. Feldmann, J. Chromatogr., 2017, 1479, 129–136.
81. P. Jitaru, S. Millour, M. Roman, K. El Koulali, L. Noel and T. Guerin, J. Food Compos. Anal., 2016, 54, 37–47.
82. F. R. Segura, J. M. D. Souza, E. S. De Paula, A. D. Martins, A. C. C. Paulelli, F. Barbosa and B. L. Batista, Food Res. Int., 2016, 89(Pt 1), 169–176.
83. S. R. Yim, G. Y. Park, K. W. Lee, M. S. Chung and S. M. Shim, Food Sci. Biotechnol., 2017, 26(1), 293–298.
84. J. Lin, G. F. Lin, Y. L. Li, X. Y. Gao, H. Du, C. G. Jia, H. C. Lu, K. Golka and J. H. Shen, EXCLI J., 2017, 16, 25–34.
85. T. T. Han, H. W. Ji, H. X. Li, H. Cui, T. Song, X. J. Duan, Q. L. Zhu, F. Cai and L. Zhang, J. Ocean Univ. China, 2017, 16(3), 455–460.
86. S. H. Nam, S. Cui and M. Y. Park, Bull. Korean Chem. Soc., 2016, 37(12), 1920–1926.
87. J. X. Li, C. J. Sun, L. Zheng, F. H. Jiang, S. Wang, Z. X. Zhuang and X. R. Wang, Mar. Pollut. Bull., 2017, 122(1–2), 464–469.
88. L. Schmidt, J. A. Landero, R. F. Santos, M. F. Mesko, P. A. Mello, E. M. M. Flores and J. A. Caruso, J. Anal. At. Spectrom., 2017, 32(8), 1490–1499.
89. J. M. Ronan, D. B. Stengel, A. Raab, J. Feldmann, L. O’Hea, E. Bralatei and E. McGovern, Chemosphere, 2017, 186, 17–23.
90. E. Bralatei, K. Nekrosiute, J. Ronan, A. Raab, E. McGovern, D. B. Stengel, E. M. Krupp and J. Feldmann, Microchim. Acta, 2017, 184(6), 1701–1709.
91. S. Foster and W. Maher, J. Environ. Sci., 2016, 49, 131–139.
92. L. L. Yu, R. C. Stanoyevitch and R. Zeisler, Anal. Methods, 2017, 9(29), 4267–4274.
93. F. Cubadda, M. D’Amato, F. Aureli, A. Raggi and A. Mantovani, Food Chem. Toxicol., 2016, 98, 148–158.
94. L. Lopez-Carrillo, B. Gamboa-Loira, W. Becerra, C. Hernandez-Alcaraz, R. U. Hernandez-Ramirez, A. J. Gandolfi, F. Franco-Marina and M. E. Cebrian, Environ. Res., 2016, 151, 445–450.
95. T. Oguri, J. Yoshinaga, Y. Suzuki, H. Tao and T. Nakazato, J. Environ. Sci. Health, Part B, 2017, 52(6), 425–429.
96. A. Mithander, T. Goen, G. Felding and P. Jacobsen, J. Occup. Med. Toxicol., 2017, 12, 5.
97. R. L. Hsieh, C. T. Su, H. S. Shiue, W. J. Chen, S. R. Huang, Y. C. Lin, M. I. Lin, S. C. Mu, R. J. Chen and Y. M. Hsueh, Toxicol. Appl. Pharmacol., 2017, 321, 37–47.
98. Y. M. Hsueh, W. J. Chen, C. Y. Lee, S. N. Chien, H. S. Shiue, S. R. Huang, M. I. Lin, S. C. Mu and R. L. Hsieh, Sci. Rep., 2016, 6, 9.
99. Y. M. Hsueh, Y. C. Lin, W. J. Chen, C. Y. Huang, H. S. Shiue, Y. S. Pu, C. H. Chen and C. T. Su, Toxicol. Appl. Pharmacol., 2017, 332, 1–7.
100. L. Torres-Sanchez, L. Lopez-Carrillo, J. L. Rosado, V. M. Rodriguez, E. Vera-Aguilar, K. Kordas, G. G. Garcia-Vargas and M. E. Cebrian, Environ. Res., 2016, 151, 38–43.
101. A. Signes-Pastor, M. Carey, J. Vioque, E. M. Navarrete-Munoz, C. Rodriguez-Dehli, A. Tardon, M. Begona-Zubero, L. Santa-Marina, M. Vrijheid, M. Casas, S. Llop, S. Gonzalez-Palacios and A. Meharg, Exposure Health, 2017, 9(2), 105–111.
102. H. Li, M. Wang, Q. D. Liang, S. N. Jin, X. J. Sun, Y. Q. Jiang, X. Y. Pan, Y. Q. Zhou, Y. Peng, B. Zhang, A. F. Zhou, Y. M. Zhang, Z. Chen, J. X. Cao, H. L. Zhang, W. Xia, T. Z. Zheng, Z. W. Cai, Y. Y. Li and S. Q. Xu, J. Chromatogr., 2017, 1479, 145–152.
103. J. E. Laine, K. A. Bailey, A. F. Olshan, L. Smeester, Z. Drobna, M. Styblo, C. Douillet, G. Garcia-Vargas, M. Rubio-Andrade, W. Pathmasiri, S. McRitchie, S. J. Sumner and R. C. Fry, Environ. Sci. Technol., 2017, 51(1), 625–633.
104. P. Balakrishna, D. Vaidya, N. Franceschini, V. S. Voruganti, M. O. Gribble, K. Haack, S. Laston, J. G. Umans, K. A. Francesconi, W. Goessler, K. E. North, E. Lee, J. Yracheta, L. G. Best, J. W. MacCluer, J. Kent, S. A. Cole and A. Navas-Acien, Environ. Health Perspect., 2017, 125(1), 15–22.
105. S. S. Ameer, K. Engstrom, M. B. Hossain, G. Concha, M. Vahter and K. Broberg, Toxicol. Appl. Pharmacol., 2017, 321, 57–66.
106. M. H. Guo, W. J. Wang, X. Hai and J. Zhou, J. Pharm. Biomed. Anal., 2017, 145, 356–363.
107. Y. Zhou, H. Y. Li and H. Z. Sun, Chem. Commun., 2017, 53(20), 2970–2973.
108. T. Matousek, Z. F. Wang, C. Douillet, S. Musil and M. Styblo, Anal. Chem., 2017, 89(18), 9633–9637.
109. A. Kato, Y. Kobayashi, O. Udagawa and S. Hirano, Biochem. Pharmacol., 2017, 126, 79–86.
110. X. M. Xue, J. Ye, G. Raber, K. A. Francesconi, G. Li, H. Gao, Y. Yan, C. Rensing and Y. G. Zhu, Environ. Sci. Technol., 2017, 51(3), 1224–1230.
111. J. M. Currier, C. Douillet, Z. Drobna and M. Styblo, J. Environ. Sci., 2016, 49, 104–112.
112. C. Y. Chen, P. P. Lin, M. H. Tsai and H. L. Lee, Arch. Toxicol., 2017, 91(9), 3079–3091.
113. U. Araujo-Barbosa, E. Pena-Vazquez, M. C. Barciela-Alonso, S. L. C. Ferreira, A. M. P. dos Santos and P. Bermejo-Barrera, Talanta, 2017, 170, 523–529.
114. C. J. Zeng, Y. Y. Yan, J. Tang, Y. H. Wu and S. S. Zhong, Spectrosc. Lett., 2017, 50(4), 220–226.
115. V. G. Mihucz, D. Enesei, A. Veszely, L. Bencs, T. Pap-Balazs, M. Ovari, C. Streli and G. Zaray, Microchem. J., 2017, 135, 105–113.
116. Y. M. Zhang, L. Y. Fang and L. Wang, Anal. Lett., 2017, 50(5), 853–865.
117. V. N. Alves, T. S. Neri, S. S. Borges, D. C. Caryalho and N. M. M. Coelho, Ecol. Eng., 2017, 106, 431–435.
118. R. A. Zounr, M. Tuzen and M. Y. Khuhawar, J. Mol. Liq., 2017, 242, 441–446.
119. X. J. Wang, G. L. Xu, P. Chen, X. Y. Liu, Y. Fang, S. Y. Yang and G. Z. Wang, RSC Adv., 2016, 6(111), 110247–110254.
120. J. Ali, M. Tuzen and T. G. Kazi, Water, Air, Soil Pollut., 2017, 228(1), 13.
121. P. Chooto, D. Muakthong, C. Innuphat and P. Wararattananurak, ScienceAsia, 2016, 42(4), 275–282.
122. G. Y. Chen, B. H. Lai, X. F. Mao, T. W. Chen and M. M. Chen, Anal. Chem., 2017, 89(17), 8678–8682.
123. W. H. Zhang, Y. H. Qi, D. Y. Qin, J. X. Liu, X. F. Mao, G. Y. Chen, C. Wei and Y. Z. Qian, Talanta, 2017, 170, 152–157.
124. C. K. Tanabe, H. Hopfer, S. E. Ebeler, J. Nelson, S. D. Conklin, K. M. Kubachka and R. A. Wilson, J. Agric. Food Chem., 2017, 65(20), 4193–4199.
125. S. Jeong, H. Lee, Y. T. Kim and H. O. Yoon, Microchem. J., 2017, 134, 295–300.
126. N. Roldan, N. Salinas-Parra, A. A. Gonzalez, F. Cifuentes-Araneda, H. Arias, M. Bravo and W. Quiroz, J. Chil. Chem. Soc., 2016, 61(4), 3214–3218.
127. M. Firat, S. Bakirdere, S. Sel, D. S. Chormey, O. Elkiran, F. Erulas and F. Turak, Int. J. Environ. Anal. Chem., 2017, 97(7), 684–693.
128. P. Heitland, M. Blohm, C. Breuer, F. Brinkert, E. G. Achilles, I. Pukite and H. D. Koster, J. Trace Elem. Med. Biol., 2017, 41, 36–40.
129. P. He, L. A. Colon and D. S. Aga, J. Chem. Educ., 2016, 93(11), 1939–1944.
130. H. L. Yang and M. C. He, Clean: Soil, Air, Water, 2016, 44(11), 9.
131. S. Ozcan, S. Bakirdere and O. Y. Ataman, Anal. Lett., 2016, 49(15), 2501–2512.
132. X. M. Li, Y. X. Chen, J. Ye, F. F. Fu, G. R. Pokhrel, H. Zhang, Y. G. Zhu and G. D. Yang, J. Sep. Sci., 2017, 40(18), 3655–3661.
133. L. Freije-Carrelo, M. Moldovan, J. I. G. Alonso, T. D. T. Vo and J. R. Encinar, Anal. Chem., 2017, 89(11), 5720–5725.
134. L. Ye, W. J. Liu, Q. T. Shi and C. Y. Jing, Environ. Pollut., 2017, 230, 1081–1089.
135. K. Shiota, T. Nakamura, M. Takaoka, K. Nitta, K. Oshita, T. Fujimori and T. Ina, Spectrochim. Acta, Part B, 2017, 131, 32–39.
136. K. Pyrzynska, Turk. J. Chem., 2016, 40(6), 894–905.
137. F. Hernandez, P. Jitaru, F. Cormant, L. Noel and T. Guerin, Food Chem., 2018, 240, 183–188.
138. H. Cui, W. Guo, L. L. Jin, Q. H. Guo and S. H. Hu, Anal. Methods, 2017, 9(8), 1307–1312.
139. W. Y. Wei, B. S. Zhao, M. He, B. B. Chen and B. Hu, RSC Adv., 2017, 7(14), 8504–8511.
140. Z. Sarikhani and M. Manoochehri, Bull. Chem. Soc. Jpn., 2017, 90(6), 746–753.
141. T. S. Munonde, N. W. Maxakato and P. N. Nomngongo, Microchim. Acta, 2017, 184(4), 1223–1232.
142. M. S. Luz, A. L. Souza, A. Haddad, A. Tartomano and P. V. Oliveira, Quim. Nova, 2016, 39(8), 951–955.
143. M. Nafti, R. Chakroun, C. Hannachi, B. Hamrouni and H. Nouaigui, Anal. Lett., 2017, 50(12), 2012–2022.
144. S. Tiwari, N. Sharma and R. Saxena, Anal. Sci., 2016, 32(12), 1321–1325.
145. Z. Tahmasebi and S. S. H. Davarani, Talanta, 2016, 161, 640–646.
146. D. S. Porto, N. Akiba, M. A. L. de Oliveira and I. Gaubeur, J. Braz. Chem. Soc., 2017, 28(2), 266–276.
147. E. N. Tafti, S. Dadfarnia and A. M. H. Shabani, Int. J. Environ. Anal. Chem., 2017, 97(5), 444–455.
148. T. Voyslavov, S. Tsakovski and S. Ganeva, Turk. J. Chem., 2016, 40(6), 944–952.
149. B. Lesniewska, I. Jakubowska, E. Zambrzycka and B. Godlewska-Zylkiewicz, Turk. J. Chem., 2016, 40(6), 933–943.
150. F. A. S. Cunha, A. S. G. Pereira, J. P. A. Fernandes, W. S. Lyra, M. C. U. Araujo and L. F. Almeida, Food Analytical Methods, 2017, 10(4), 921–930.
151. S. Kapitany, E. Soki, J. Posta and A. Beni, Acta Chim. Slov., 2017, 64(1), 248–255.
152. T. J. Jiang, M. Yang, S. S. Li, M. J. Ma, N. J. Zhao, Z. Guo, J. H. Liu and X. J. Huang, Anal. Chem., 2017, 89(10), 5557–5564.
153. E. Leese, J. Morton, P. H. E. Gardiner and V. A. Carolan, Int. J. Hyg. Environ. Health, 2017, 220(2), 415–423.
154. H. L. Huang, H. H. Huang and Y. J. Wei, Spectrochim. Acta, Part B, 2017, 133, 9–13.
155. Z. H. Li, W. Kou, S. Wu and L. Wu, Anal. Methods, 2017, 9(21), 3221–3229.
156. Y. He, F. Mo, D. L. Chen, L. J. Xu, Y. G. Wu and F. F. Fu, Electrophoresis, 2017, 38(3–4), 469–476.
157. B. Klencsar, L. Balcaen, F. Cuyckens, F. Lynen and F. Vanhaecke, Anal. Chim. Acta, 2017, 974, 43–53.
158. B. Heckel, D. Rodriguez-Fernandez, C. Torrento, A. Meyer, J. Palau, C. Domenech, M. Rosen, A. Soler, D. Hunkeler and M. Elsner, Anal. Chem., 2017, 89(6), 3411–3420.
159. A. Horst, J. Renpenning, H. H. Richnow and M. Gehre, Anal. Chem., 2017, 89(17), 9131–9138.
160. N. L. A. Jamari, J. F. Dohmann, A. Raab, E. M. Krupp and J. Feldmann, J. Anal. At. Spectrom., 2017, 32(5), 942–950.
161. C. Frank, C. Brauckmann, M. Palos, C. G. Arsene, J. Neukammer, M. Busto, S. Zakel, C. Swart, B. Guttler and R. Stosch, Anal. Bioanal. Chem., 2017, 409(9), 2341–2351.
162. J. X. Li, C. J. Sun, L. Zheng, F. H. Jiang, X. F. Yin, J. H. Chen and X. R. Wang, Chin. J. Anal. Chem., 2016, 44(11), 1659–1664.
163. J. Xia, Y. Fang, Y. Chen, Y. S. Pan, P. Li, M. Xue and Q. H. Hu, J. Food Compos. Anal., 2017, 60, 74–80.
164. S. Q. Abad, P. Rodriguez-Gonzalez, W. C. Davis and J. I. G. Alonso, Anal. Chem., 2017, 89(12), 6731–6739.
165. V. Vacchina, F. Seby, R. Chekri, J. Verdeil, J. Dumont, M. Hulin, V. Sirot, J. L. Volatier, R. Serreau, A. Rousseau, T. Simon and T. Guerin, Talanta, 2017, 167, 404–410.
166. I. Zivkovic, V. Fajon, D. Tulasi, K. O. Vazner and M. Horvat, Mar. Chem., 2017, 193, 3–7.
167. W. J. Jiang, X. Jin, X. H. Yu, W. H. Wu, L. J. Xu and F. F. Fu, J. Chromatogr., 2017, 1496, 167–173.
168. Y. M. Liu, F. P. Zhang, B. Y. Jiao, J. Y. Rao and G. Leng, J. Chromatogr., 2017, 1493, 1–9.
169. J. M. Mo, Q. Li, X. H. Guo, G. X. Zhang and Z. Wang, Anal. Chem., 2017, 89(19), 10353–10360.
170. M. Jimenez-Moreno, M. A. Lominchar, M. J. Sierra, R. Millan and R. C. R. Martin-Doimeadios, Talanta, 2018, 176, 102–107.
171. X. H. Xu, B. Meng, C. Zhang, X. B. Feng, C. H. Gu, J. Y. Guo, K. Bishop, Z. D. Xu, S. S. Zhang and G. L. Qiu, Environ. Pollut., 2017, 223, 11–18.
172. H. Kodamatani, S. J. Balogh, Y. H. Nollet, A. Matsuyama, V. Fajon, M. Horvat and T. Tomiyasu, Chemosphere, 2017, 169, 32–39.
173. R. Cesario, H. Hintelmann, R. Mendes, K. Eckey, B. Dimock, B. Araujo, A. M. Mota and J. Canario, Environ. Pollut., 2017, 226, 297–307.
174. L. A. Mendes, J. C. de Lena, C. M. do Valle, P. M. Fleming and C. C. Windmoller, Appl. Geochem., 2016, 75, 32–43.
175. P. A. Perez, H. Hintelman, W. Quiroz and M. A. Bravo, Chemosphere, 2017, 186, 570–575.
176. E. Covaci, M. Senila, M. Ponta, E. Darvasi, M. Frentiu and T. Frentiu, Food Control, 2017, 82, 266–273.
177. E. Covaci, M. Senila, M. Ponta, E. Darvasi, D. Petreus, M. Frentiu and T. Frentiu, Talanta, 2017, 170, 464–472.
178. E. L. de Paiva, M. A. Morgano and R. F. Milani, Food Addit. Contam., Part B, 2017, 10(3), 185–191.
179. F. M. Rebelo, L. R. da Cunha, P. D. Andrade, W. A. da Costa, W. R. Bastos and E. D. Caldas, J. Trace Elem. Med. Biol., 2017, 44, 99–103.
180. R. J. Strickman and C. R. J. Mitchell, Sci. Total Environ., 2017, 574, 1415–1423.
181. J. Sysalova, J. Kucera, B. Drtinova, R. Cervenka, O. Zverina, J. Komarek and J. Kamenik, Sci. Total Environ., 2017, 584, 1032–1039.
182. T. Lang, R. Kruse, M. Haarich and W. Wosniok, Mar. Environ. Res., 2017, 124, 32–40.
183. B. B. Song, H. J. He, L. F. Chen, S. F. Yang, Y. G. Yin and Y. B. Li, Anal. Lett., 2017, 50(13), 2161–2176.
184. S. Q. Zhu, B. B. Chen, M. He, T. Huang and B. Hu, Talanta, 2017, 171, 213–219.
185. M. Renedo, P. Bustamante, E. Tessier, Z. Pedrero, Y. Cherel and D. Amouroux, Talanta, 2017, 174, 100–110.
186. Z. F. Liu, Z. Xing, Z. Y. Li, Z. L. Zhu, Y. Q. Ke, L. L. Jin and S. H. Hu, J. Anal. At. Spectrom., 2017, 32(3), 678–685.
187. K. Peeters, T. Zuliani, D. Zigon, R. Milacic and J. Scancar, Food Chem., 2017, 230, 327–335.
188. J. Moreda-Pineiro, A. Moreda-Piñero and P. Bermejo-Barrera, Crit. Rev. Food Sci. Nutr., 2017, 57(4), 805–833.
189. A. S. A. Ibrahim, R. Al-Farawati, U. Hawas and Y. Shaban, Open Chem., 2017, 15(1), 103–122.
190. C. E. Bryan, G. D. Bossart, S. J. Christopher, W. C. Davis, L. E. Kilpatrick, W. E. McFee and T. X. O’Brien, J. Trace Elem. Med. Biol., 2017, 44, 40–49.
191. H. Z. He, S. J. Liu, H. S. Li and T. F. Chen, RSC Adv., 2017, 7(68), 43266–43272.
192. B. R. Cardoso, D. J. Hare, A. I. Bush, Q. X. Li, C. J. Fowler, C. L. Masters, R. N. Martins, K. Ganio, A. Lothian, S. Mukherjee, E. A. Kapp, B. R. Roberts and A. R. Grp, J Alzheimers Dis., 2017, 57(1), 183–193.
193. Y. Ogra, Y. Ogihara and Y. Anan, Metallomics, 2017, 9(1), 61–68.
194. H. Wang, B. B. Chen, M. He, X. X. Yu and B. Hu, Sci. Rep., 2017, 7, 8.
195. Y. Fang, Z. Xu, Y. Shi, F. Pei, W. J. Yang, N. Ma, B. M. Kimatu, K. L. Liu, W. F. Qiu and Q. H. Hu, Food Chem., 2017, 219, 391–398.
196. A. Achouba, P. Dumas, N. Ouellet, M. Lemire and P. Ayotte, Environ. Int., 2016, 96, 8–15.
197. S. Godin, D. Bouzas-Ramos, S. Fontagne-Dicharry, B. Bouyssiere and M. Bueno, Anal. Biochem., 2017, 530, 9–16.
198. T. A. Marschall, J. Bornhorst, D. Kuehnelt and T. Schwerdtle, Mol. Nutr. Food Res., 2016, 60(12), 2622–2632.
199. K. M. Kubachka, T. Hanley, M. Mantha, R. A. Wilson, T. M. Falconer, Z. Kassa, A. Oliveira, J. Landero and J. Caruso, Food Chem., 2017, 218, 313–320.
200. T. A. Hanley, K. Kubachka, A. M. Taylor and S. E. Kern, J. Anal. At. Spectrom., 2017, 32(6), 1196–1202.
201. B. Gilbert-Lopez, M. Dernovics, D. Moreno-Gonzalez, A. Molina-Diaz and J. F. Garcia-Reyes, J. Chromatogr. B: Anal. Technol. Biomed. Life Sci., 2017, 1060, 84–90.
202. K. Bierla, N. Suzuki, Y. Ogra, J. Szpunar and R. Lobinski, Food Chem., 2017, 237, 1196–1201.
203. L. Gao, D. S. Luo, X. S. Hu and J. H. Wu, Int. J. Food Sci. Technol., 2017, 52(3), 843–850.
204. A. C. Grijalba, E. F. Fiorentini and R. G. Wuilloud, J. Chromatogr., 2017, 1491, 117–125.
205. M. Soderlund, J. Virkanen, S. Holgersson and J. Lehto, J. Environ. Radioact., 2016, 164, 220–231.
206. K. Chandrasekaran, N. Garg and D. Karunasagar, J. Anal. At. Spectrom., 2017, 32(4), 796–806.
207. L. Liu, X. A. Yang, X. P. Lu and W. B. Zhang, Talanta, 2017, 171, 90–100.
208. X. J. Wang, P. Chen, L. Cao, G. L. Xu, S. Y. Yang, Y. Fang, G. Z. Wang and X. C. Hong, Food Analytical Methods, 2017, 10(6), 1653–1660.
209. D. Citak and M. Tuzen, Turk. J. Chem., 2016, 40(6), 1012–1018.
210. M. C. Fontanella, R. D’Amato, L. Regni, P. Proietti, G. M. Beone and D. Businelli, J. Trace Elem. Med. Biol., 2017, 43, 87–92.
211. E. D. da Silva, F. Aureli, M. D’Amato, A. Raggi, S. Cadore and F. Cubadda, J. Agric. Food Chem., 2017, 65(14), 3031–3038.
212. N. Kroepfl, T. A. Marschall, K. A. Francesconi, T. Schwerdtle and D. Kuehnelt, J. Anal. At. Spectrom., 2017, 32(8), 1571–1581.
213. A. Raab, M. Ronzan and J. Feldmann, Metallomics, 2017, 9(10), 1429–1438.
214. Y. Ogra, Metallomics, 2017, 9(5), 435–441.
215. H. B. Qin, Y. Takeichi, H. Nitani, Y. Terada and Y. Takahashi, Environ. Sci. Technol., 2017, 51(11), 6027–6035.
216. M. Chen, L. C. Wu, X. Yi, K. Z. Yang and H. G. Xie, Anal. Methods, 2017, 9(20), 3061–3066.
217. S. Z. Mohammadi, A. Sheibani, F. Abdollahi and E. Shahsavani, Arabian J. Chem., 2016, 9, S1510–S1515.
218. Z. D. Firouzabadi, A. M. H. Shabani, S. Dadfarnia and M. H. Ehrampoush, Microchem. J., 2017, 130, 428–435.
219. B. Campanella, C. Casiot, M. Onor, M. Perotti, R. Petrini and E. Bramanti, Talanta, 2017, 171, 255–261.
220. S. W. C. Chung and A. H. T. Wu, J. Chromatogr., 2017, 1505, 18–34.
221. M. Lagerstrom, J. Strand, B. Eklund and E. Ytreberg, Environ. Pollut., 2017, 220, 1333–1341.
222. H. Yang, H. E. Zhang, X. Y. Zhu, S. D. Chen, L. J. Liu and D. D. Pan, J. Food Qual., 2017, 11.
223. W. A. Al Rawahi and N. I. Ward, Talanta, 2017, 165, 391–397.
224. A. M. Aucour, J. P. Bedell, M. Queyron, R. Thole, A. Lamboux and G. Sarret, Environ. Sci. Technol., 2017, 51(15), 8350–8358.
225. J. F. Chen, F. F. Fu, S. Y. Wu, J. Wang and Z. W. Wang, J. Sep. Sci., 2017, 40(19), 3898–3904.
226. M. P. Chantada-Vazquez, A. Moreda-Pineiro, M. C. Barciela-Alonso and P. Bermejo-Barrera, Appl. Spectrosc. Rev., 2017, 52(2), 145–174.
227. Y. L. Ouyang, Y. Y. Zeng, L. Yi, H. Tang, D. X. Li, R. J. Linhardt and Z. Q. Zhang, J. Chromatogr., 2017, 1522, 56–61.
228. K. Coufalikova, I. Benesova, T. Vaculovic, V. Kanicky and J. Preisler, Anal. Chim. Acta, 2017, 968, 58–65.
229. D. P. Bishop, L. Blanes, A. B. Wilson, T. Wilbanks, K. Killeen, R. Grimm, R. Wenzel, D. Major, M. Macka, D. Clarke, R. Schmid, N. Cole and P. A. Doble, J. Chromatogr., 2017, 1497, 64–69.
230. A. M. Donnell, S. Lewis, S. Abraham, K. Subramanian, J. L. Figueroa, G. S. Deepe and A. P. Vonderheide, Anal. Bioanal. Chem., 2017, 409(26), 6163–6172.
231. S. Li, B. B. Chen, M. He and B. Hu, J. Anal. At. Spectrom., 2017, 32(6), 1186–1195.
232. R. I. Adaba, G. Mann, A. Raab, W. E. Houssen, A. R. McEwan, L. Thomas, J. Tabudravu, J. H. Naismith and M. Jaspars, Tetrahedron, 2016, 72(52), 8603–8609.
233. F. Calderon-Celis, L. Cid-Barrio, J. R. Encinar, A. Sanz-Medel and J. J. Calvete, J. Proteomics, 2017, 164, 33–42.
234. M. Sharar, E. M. Saied, M. C. Rodriguez, C. Arenz, M. Montes-Bayon and M. W. Linscheid, Anal. Bioanal. Chem., 2017, 409(8), 2015–2027.
235. S. Cuello-Nunez, R. Larios, C. Deitrich, T. Lekishvili, V. Nischwitz, B. L. Sharp and H. Goenaga-Infante, J. Anal. At. Spectrom., 2017, 32(7), 1320–1330.
236. L. S. Foteeva, M. Matczuk, K. Pawlak, S. S. Aleksenko, S. V. Nosenko, V. K. Karandashev, M. Jarosz and A. R. Timerbaev, Anal. Bioanal. Chem., 2017, 409(9), 2421–2427.
237. A. Balbekova, M. Bonta, S. Torok, J. Ofner, B. Dome, A. Limbeck and B. Lendl, Anal. Methods, 2017, 9(37), 5464–5471.

---