E.
Hywel Evans
*a,
Christopher D.
Palmer
b and
Clare M. M.
Smith
c
aSchool of Geography, Earth, and Environmental Sciences, University of Plymouth, Drake Circus, Plymouth, UK PL4 8AA. E-mail: hevans@plymouth.ac.uk
bWadsworth Center, New York State Department of Health, P.O. Box 509, Empire State Plaza, Albany, NY 12201-0509, USA
cSt. Ambrose High School, Blair Road, Coatbridge, Lanarkshire, UK ML5 2EW
First published on 15th May 2012
This is the second iteration of this review covering developments in ‘Atomic Spectrometry’. It covers atomic emission, absorption, fluorescence and mass spectrometry, but excludes material on speciation and coupled techniques which is included in a separate review. It should be read in conjunction with the other related reviews in the series.1-5 A critical approach to the selection of material has been adopted, with only novel developments in instrumentation, techniques and methodology being included. The major growth areas in evidence were the use of MC-ICP-MS as the method of choice for isotope ratio analyses, and new applications of AMS. The decline in the number of fundamental studies and developments in chemometrics continued. Some novel instrumental methods, such as the portable liquid electrode plasma, were reported and there were many new applications of solid phase extractants for on-line sample pretreatment, particularly using carbon nanotubes.
A critical discussion of selected reports dealing with pretreatment methods of oily samples and the determination of their organic and inorganic constituents using flow systems and spectrometric methods has been presented,7 including the empirical preparation of certain organised assemblies like micelles, and emulsions.
Tian et al.8 used a MnO2 mini-column to preconcentrate the As species AsIII, AsV, MMA, and DMA, during which process AsIII was converted to AsVvia oxidation by the MnO2, while the other species remained unchanged. Following hydride generation, detection was made by quartz flame-AAS. The recovery of AsV (i.e., the total amount of arsenate and arsenite in the original sample), DMA and MMA from the MnO2 mini-column is facilitated by TMAH. The procedure was applied for the determination of As in snow water and Hijiki (seaweed) samples. An on-line sequential injection system has been developed9 based on SPE coupled with FI-HG-AAS with a quartz tube atomiser for the determination of AsIII in groundwater, removing the need for sample pretreatment. The method was based on the selective retention of inorganic AsV, carried out by passing the original sample through a cartridge containing a chloride-form strong anion exchanger. Since the uptake of AsV could be interfered by several anions in waters, the effect of Cl−, SO42−, NO3−, HPO42− and HCO3− on the retention was evaluated and discussed. Rahman et al.10 employed a combination of SPE columns for the selective separation of water-soluble As species; arsenite, arsenate, MMA and DMA. The SPE columns, namely AnaLig TE-01, AnaLig AN-01, and AnaLig As-01 PA, contain immobilised macrocyclic material as the sorbent, commonly known as molecular recognition technology (MRT) gel. Quantitative separation of AsIII, AsV, MMA, and DMA was achieved based on the differences in extraction and recovery behaviour of the MRT gel SPE columns for different As species. Novel mercapto- and amine-bifunctionalised silica compounds have been synthesised11 for the speciation/sorption of inorganic As prior to ICP-MS. The bifunctional (NH2 + SH) silica retained AsIII at a wide pH range, from 1.0 to at least 9.0 with the exception at pH 2.0, and also exhibited quantitative sorption to AsV at pH 3.0. A field method for the speciation of As in water samples has been reported12 that was simple, rapid, cost effective, and safe to use outside laboratory environments. The method used SPE in series for selective retention of As species, followed by elution and measurement of eluted fractions by ICP-MS. The method was suitable for on-site separation and preservation of As species from water. The authors used their system at test sites in Argentina and the UK; in Argentinian groundwater, 4–20% of speciated As was present as MMA, and 20–73% as AsIII, in UK surface waters, speciated As was measured as 7–49% MMA and 12–42% DMA.
A procedure for the separation and preconcentration of Bi in a sequential injection system employing bamboo carbon as sorbent has been developed.13 For a 1 mL sample volume, the HG-AFS detection limit was 13 ng L−1. The system was validated against a river sediment CRM (CRM 320), along with spiked recoveries of Bi in human whole blood.
The capabilities of carbon nanotubes (CNTs) and modified CNTs have been evaluated14 to serve as sorbents for preconcentrating Cd with on-line ultrasonic nebulisation (USN) and ICP-OES. Three different CNT substrates, namely CNTs, oxidised-carbon nanotubes (ox-CNTs) and L-alanine-carbon nanotubes (ala-CNTs) were studied systematically. They showed dissimilar adsorption behaviours leading to increased preconcentration factors when used in the proposed on-line SPE system as follows: CNT < ala-CNT < ox-CNT. Under optimal conditions, the adsorption capacity on ox-CNTs was found to be 130 μmol g−1, and the LOD was 1.03 μg L−1 Cd. Zeng et al.15 used hollow-fibre-supported liquid-membrane extraction coupled with thermospray-flame-furnace-AAS for the determination of Cd in water. 1-Octanol was immobilised in the pores of the polypropylene hollow fibre as liquid membrane and used as the acceptor solution. Under optimised conditions, an enrichment factor of 146 could be obtained, leading to a detection limit of 5 ng L−1 Cd. Tang and Hu16 used a cone-shaped pipette tip packed with human hair as a solid phase for the preconcentration of Cd based on its complex formation with APDC. The enhancement factor was 19.
Karve and Gholave17 preconcentrated CoII using a C18 bonded silica membrane disc impregnated with Cyanex 272. CoII was quantitatively sorbed at pH 6.0 from sample solution and eluted using 10.0 mL 1.0 M HNO3 prior to FAAS determination. The LOD was 1.4 μg L−1, and the preconcentration factor was >200. The method was applied for the determination of Co in urine specimens, and industrial sludge samples.
The determination of Cr species in water and human serum samples by FAAS after magnetic SPE was reported.18 Silica-coated Fe3O4 magnetic nanoparticles modified with N-(2-aminoethyl)-3-amino-propyltrimethoxy-silane were prepared and employed for differentiation between CrIII and CrVI; this was selective towards CrIII but not towards CrVI.
There were numerous papers reporting the use of solid-phase extraction for the preconcentration of mercury species. Chen et al.19 found that the functionalisation of cellulose fibre by grafting L-cysteine on the surface significantly improved the sorption capacity towards Hg and methyl-Hg. Trace HgII has been extracted using silica gel modified with DPC.20 At pH 4, the SG-DPC was shown to have rapid adsorption kinetics towards HgII. The same authors described the use of a second SPE21 C18 bonded silica phase loaded with 4,4′-dimethoxybenzil bisthiosemicarbazone. Oxidised and non-oxidised multi-wall carbon nanotubes (MWCNTs) of different geometrical dimensions have been used for HgII adsorption.22 The L-MWCNT-4060-NA solid phase (nitric acid-oxidised MWCNT of external diameter 40–60 nm and length 5–15 μm) had the lowest sorption energy of the MWCNTs tested, and therefore the most favourable uptake of HgII. Rofouei et al.23 employed an C18 silica membrane disc modified by the recently synthesised triazene ligand, 1,3-bis(2-ethyoxyphenyl)triazene to preconcentrate HgII, followed by CV-AAS determination. Li et al.24 studied the adsorption ability of duckweed (Lemna minor) powders for removing HgII, MeHg, and EtHg with CV-AAS detection. The significant adsorption sites were C–O–P and phosphate groups by surface electrostatic interactions with aqueous inorganic and organic Hg cations, plus selective adsorption which resulted from the strong chelating interaction between amine groups and HgII on the surface of the L Minor cells. An octadecyl silica cartridge has been employed as sorbent and 4-bpdb (1,4-bis(4-pyridyl)-2,3-diaza-1,3-butadiene) as a ligand for HgII preconcentration,25 with concentration factor and detection limit of 128 and 1.87 ng L−1, respectively. Preconcentration and speciation of HgII and MeHg on a solid-phase of poly(acrylamide) grafted onto cross-linked poly(4-vinyl pyridine) has been achieved,26 with excellent selectivity for HgII in the presence of PbII, ZnII, CuII, CdII, and FeIII ions. Chai et al.27 used a new 2-(2-oxoethyl)hydrazine carbothioamide modified silica gel for Hg determination by ICP-AES; 1-(2-thiazolylazo)-2-naphthol impregnated onto Amberlite XAD-4 resin beads for sorption of HgII yielded maximum capacity (qmax) in terms of monolayer sorption of 450 mg g−1.28 Romero et al.29 employed different Pd substrates (i.e. Pd wire, Pd-coated stainless steel wire and Pd-coated SiO2) for the microextraction of Hg prior to its release in a modified quartz cell with AAS detection. This new design allowed direct sample injection from the SPME device into the quartz cell, thus avoiding analyte dilution. Although Pd wire provided the best performance for preconcentration, Pd-coated SiO2 fibres could be more easily adapted to the sampling device. Tsoi et al.30 also used SPME but with LC-ICP-MS, to preconcentrate MeHg and EtHg from the headspace of complex urine matrices. The fibre was exposed to a sample headspace for a 45 min extraction, followed by a 5 min desorption.
A MnII imprinted 3-mercaptopropyltrimethoxysilane (MPTS)-silica coated capillary has been synthesised and used to preconcentrate MnII from biological samples prior to ICP-MS determination,31 with a sample throughput of 9 h−1 and an enrichment factor of 16.7. The proposed method was successfully applied for the determination of trace MnII in serum samples. Chen et al.32 used single-walled carbon nanotubes as a new adsorbent for the simultaneous preconcentration and determination of MnII and MnVII by ICP-MS with LODs for MnII and MnVII of 0.031 and 0.054 ng mL−1, respectively.
A selective solid-phase for the extraction of NiII from biological samples has been developed33 by synthesis of a non-imprinted polymer material by copolymerisation of 2-vinylpyridine as monomer, ethyleneglycoldimethacrylate as crosslinking agent and azobisisobutyronitrile as initiator in the presence of Ni-5-(4-carboxylphenylazo)-8-hydroxyquinoline complex.
Bakircioglu et al.34 physically immobilised coliform bacteria on titanium dioxide nanoparticles as a biosorbent for the SPE of PbII using FIA-FAAS. The LOD for water samples was 0.90 μg L−1. The procedure was applied to the determination of PbII in river water, wine, and baby food. PbII ions were extracted35 using an ion-imprinted functionalised sodium trititanate whisker adsorbent prepared by a surface-imprinting technique. The PbII-imprinted Na2Ti3O7 materials were capable of selectively adsorbing PbII ions and were used for the preconcentration of Pb in traditional Chinese medicine samples.
An automated low pressure flow analysis method with online columns has been developed for the HG-ICP-MS determination of SbIII and SbV in aqueous environmental samples.36 A chelating resin (1,5-bis(2-pyridyl)-3-sulfophenyl methylene) thiocarbonohydrazide immobilised on aminopropyl-controlled pore glass (PSTH-cpg), and an anion exchanger (Amberlite IRA-910) were used for the columns. After removal of SbIII by PSTH-cpg, SbV was collected on the Amberlite. Zeng et al.37 employed a new method of hollow fibre supported liquid membrane extraction coupled with thermospray flame-furnace-AAS for the speciation of SbIII and SbV in environmental and biological samples. The method was based on the complexing of SbIII with sodium DDTC. The hydrophobic complex formed was subsequently extracted into the lumen of the hollow fibre, whereas SbV was retained in solution.
The preconcentration of Se from mineral water and beer samples has been achieved38 using living bacteria Lactobacillus plantarum followed by continuous powder introduction into MIP-OES. The use of bacteria immobilised on silica permits on-column preconcentration of Se with minimum sample pretreatment by pH adjustment. A LOD of 52 ng g−1 corresponded to 0.06 ng mL−1 in the sample solution, implying a preconcentration factor of 1000. Dai et al.39 developed a new type of cross-linked chitosan with diethylene triamine (DCCTS). The results indicated that the DCCTS could concentrate and separate SeIV at pH 3.6; the maximum adsorption efficiency was 94%, and the maximum adsorption capacity was 42.7 mg g−1.
Wu et al.40 preconcentrated and speciated two analytes, inorganic As and Sb in waters by SPE coupled with HG-AFS. The speciation scheme involved the on-line formation and retention of the ammonium pyrrolidine dithiocarbamate complexes of AsIII and SbIII on a single-walled carbon nanotube packed micro-column. The total As and total Sb were determined by the same protocol after AsV and SbV were reduced by thiourea, with AsV and SbV concentrations obtained by difference. Anthemidis et al.41 formed an on-line chelate complex of Pb and Cd with ammonium diethyldithiophosphate (DDTP), and retained these complexes on the surface of reversed-phase poly(divinylbenzene-N-vinylpyrrolidone) co-polymeric beads, prior to elution with methanol and FAAS determination. The proposed method was well suited to detect the target elements at concentration levels below the maximum allowed concentrations endorsed by the European Framework Directive (2008/105/EC) in inland and coastal waters. Kalyan et al.42 preconcentrated HgII and UVI ions using an itaconic acid functionalised adsorptive membrane. The micro-gel was formed by in situ photo-polymerisation of itaconic acid (ICA) along with co-monomer acrylamide and crosslinker pentaerythritoltetraacrylate (PETA) in pores of the host membrane. Afkhami et al.43 developed a new SP for PbII and CrIII preconcentration by immobilising 2,4-dinitrophenylhydrazine on nano-alumina coated with sodium dodecyl sulfate. The ions were eluted from the sorbent with a mixture of nitric acid and methanol. The FAAS detection limit for PbII and CrIII was 0.43 and 0.55 μg L−1, respectively, and the maximum preconcentration factor was 267.
Magnetic solid phase microextraction on a microchip combined with ETV-ICP-MS has been used for the determination of Cd, Hg, and Pb in cells.44 Gamma-mercaptopropyltrimethoxysilane modified silica-coated magnetic nanoparticles were synthesised and used for the extraction of Cd, Pb, and Hg. Under an external magnetic field, these magnetic nanoparticles self-assembled in microchannels to form a solid phase packed column which enabled enrichment of Cd, Hg, and Pb by a factor of >40, and resulted in improved limits of detection for Cd (0.72 ng L−1), Hg (0.86 ng L−1), and Pb (1.12 ng L−1) using ETV-ICP-MS. Similarly, Zhang et al.45 immobilised iminodiacetic acid on mesoporous Fe3O4·SiO2 microspheres and utilised the material for the fast and selective magnetic solid phase extraction of Cd, Mn, and Pb in environmental and biological samples. The microspheres were separated from the aqueous sample solution by applying an external magnetic field. Mashhadizadeh and Karami46 preconcentrated Ag, Cd, Cu, and Zn using magnetic nanoparticles coated by 3-(trimethoxysilyl)-1-propantiol and modified with 2-amino-5-mercapto-1,3,4-thiadiazole. Under the optimal conditions, high concentration factors (194, 190, 170, and 182) were achieved for Ag, Cd, Cu, and Zn.
A poly(N-isopropylacrylamide) gel as a novel polymer coating for fibre-in-tube capillary microextraction has been used47 for trace determination of Co, Ni, and Cd followed by on-line ICP-MS detection. The PNIPA coated fibre-in-tube capillary can be used for more than 150 times without decreasing the extraction efficiency. The method was used for the determination of Co, Ni, and Cd in human urine specimens. Huang and Hu48 synthesised a titania hollow fibre membrane via a template method, coupled with a sol–gel process, for the online ICP-MS determination of 4 elements, Cd, Co, V, and Ni in human serum samples. Zhao et al.49 adsorbed the rare earths, La, Ce, Nd, and Y on di(2-ethylhexyl) phosphoric acid resin in the presence of complexing agent, EDTA, with a FI-based means of sample introduction. MIP-AES detection limits were 1.09 μg L−1 (La), 3.31 μg L−1 (Ce), 2.05 μg L−1 (Nd), and 1.25 μg L−1 (Y). Li et al.50 preconcentrated 6 noble metals (Ru, Rh, Pd, Pt, Ir and Au) using FI-online displacement SPE via magnetic immobilisation of mercapto-functionalised magnetite microspheres onto the inner walls of a knotted reactor. The sample throughput of the developed method for ICP-MS was about 20 samples h−1. The recovery of all 6 metals was maintained at 90% after 140 successive cycles.
Baytak et al.51 preconcentrated 6 elements, Cr, Cu, Fe, Mn, Ni, and Zn from water samples on a minicolumn of yeast (Yamadazyma spartinae) immobilised on TiO2 nanoparticles. The column, containing 100 mg sorbent, offered the capacity to preconcentrate up to 500 mL of sample solution to achieve an enrichment factor of 250 using 2 mL of 5% v/v HNO3 as eluent.
Two papers reporting the speciation of Hg by dispersive liquid–liquid microextraction combined with HPLC-ICP-MS have been published by the same group. MeHg and HgII were speciated from water samples55 by complexing them with sodium diethyldithiocarbamate, and then extracting into carbon tetrachloride using DLLME. Under the optimized conditions, enrichment factors of 138 and 350 for MeHg and HgII were obtained from 5.00 mL sample. The detection limits of the analytes (as Hg) were 0.0076 ng mL−1 for MeHg and 0.0014 ng mL−1 for HgII. In the second paper, the authors speciated Hg compounds from liquid cosmetic samples.56 HgII, MeHg, and EtHg were complexed with APDC then the complexes were extracted into 1-hexyl-3-methylimidazolium hexafluorophosphate using DLLME. Enrichment factors of 760, 115 and 235 were obtained for HgII, MeHg, and EtHg, respectively, for a 5.0 mL sample size. Liquid-phase microextraction has been proposed57 as an attractive tool in combination with XRF spectrometry for multielement analysis (Fe, Co, Zn, Ga, Se, and Pb). Since the X-ray beam can be focused on a small spot size and simultaneously DLLME produces a very small drop ranging from 2 to 30 μL, the combination of these two techniques is a very promising tool for multielement trace and ultratrace analyses. For this work, the authors used DLLME with APDC as a chelating agent.
Zeeb and Sadeghi58 preconcentrated ZnIIvia ionic-liquid cold-induced aggregation DLLME followed by FAAS. The ionic-liquid 1-hexyl-3-methylimidazolium hexafluorophosphate is dispersed into a heated sample solution containing sodium hexafluorophosphate as a common ion source. The solution is then placed in an ice bath upon which a cloudy solution forms due to the decrease in solubility of the ionic liquid. Zn is complexed with 8-HQ and extracted and this enriched phase is dissolved in a diluting agent and determined using FAAS, with LOD of 0.18 μg L−1 for Zn. Zhang and Fan59 separated and preconcentrated SbIII and total Sb by DLLME, prior to determination by ETAAS with Rh as a permanent modifier. At pH 2, SbIII was complexed with N-benzoyl-N-phenylhydroxylamine and extracted into fine droplets. The LOD for SbIII was 0.03 μg L−1. Serafimovska et al.60 described a modified highly effective liquid phase semi-microextraction (LSME) procedure for preconcentration of inorganic Sb in waters by ETAAS. SbIII species were selectively extracted as dithiocarbamate complexes from 100 mL sample into 250 μL xylene at pH 5–8. Total Sb was determined using the same extraction system over a sample acidity range 0–1.2, without the need for pre-reduction of SbV to SbIII.
Durukan et al.61 described their solidified organic drop microextraction and FI-FAAS method for the determination of Cu in water samples. The complexing agent was 3-amino-7-dimethylamino-2-methylphenazine. Similarly this technique has been used62 to preconcentrate Cu in cereal samples and for the preconcentration of SeIV prior to determination by HG-AFS.63
A high performance concentric nebuliser (HPCN) operating at sample uptakes rates of <10 μL min−1, was described by Inagaki et al.71 The HPCN had a triple tube concentric structure, and consisted of a fused silica capillary mounted in the centre of the nebuliser body. When the capillary position was recessed by 20–50 μm with respect to the nebuliser nozzle tip, a liquid jet was produced inside the nozzle by a flow focussing effect. The sensitivity in ICP-MS with the HPCN was over two-fold higher that those with commercially available microflow nebulisers. Matrix effects in ICP-OES have been compensated for using three novel multiple sample introduction systems based on Flow Burring® technology.72 When the multi-nebulisation systems were employed, the main difference, compared to routine nebulisation, was the significant reduction in ‘resource consumption’ (i.e., samples, reagents, and time).
Paredes et al.73 studied the characteristics of aerosol generation for 26 concentric nebulisers; 17 of the A-type, 6 of the C-type, and 3 of the K-type. For a given nebuliser design, the gas exit cross-sectional area was shown to critically influence the aerosol characteristics. The aerosol generation mechanism was explored and it was concluded that, for A-type nebulisers the nebulisation is more efficient than C and K-type.
A demountable capillary microflow nebuliser (d-CMN) was described by Cheng et al.74 It consisted of a nebuliser body, a fused silica capillary with a tapered tip, and a PTFE adapter. The demountable construction of the d-CMN permitted that a blocked or broken solution capillary could be easily replaced. The low self-aspiration rate (4.8 μL min−1), and the analytical characteristics comparable to commercial microflow nebulisers made the d-CMN a good choice for coupling capillary electrophoresis and microbore HPLC to ICP-MS.
An air-segmented flow-injection method has been developed75 developed based on the use of a high temperature single pass spray chamber and the injection of a sample plug into an air carrier stream, to mitigate non-spectral interferences caused by organic solvents and petroleum products, and to reduce solvent loading in ICP-OES. The evaluated sample introduction systems were a 12 cm3 inner volume single pass spray chamber (also called a torch integrated sample introduction system) with and without heating, and a 40 cm3 cyclonic spray chamber as reference. For a 5 μL sample plug and a temperature of 200 °C, evaporation of the sample was complete before it entered the plasma, thereby reducing interferences.
The analytical potential of a coupled continuous-microflow ultrasonic nebuliser triple-mode micro-capillary system has been investigated76for the determination of iodine in biological samples by direct iodine vapour generation-ICP-OES. The reaction products were obtained by mixing the sample and reaction products at the quartz oscillator, converting liquids into aerosol at the entrance to the spray chamber. The accuracy of the method was verified using NIST 1549 and 1566b SRM's, and a simple external calibration.
Recent achievements in CHG hyphenated with ICP-OES and MIP-OES have been reviewed,78 including new universal sample introduction interfaces operating as both conventional nebulisers and as gas-liquid phase separators; vapour generation from transition and noble metals; HG from slurry samples; and hyphenation with chromatographic techniques. Another review79 highlighted the lower limits of detection achievable in HPLC speciation analysis when the technique is coupled with a CVG system. The development of a HPLC-CVG techniques for ultratrace-elemental speciation in a variety of matrices, and the application of these techniques was discussed. The use of AAS, OES, AFS, and MS techniques as detectors following chemical vapour generation in slurry samples has been evaluated80 and practical application of the slurry chemical vapour generation technique to biological, environmental and food samples was discussed.
A new design of CVG integrated atom trap atomiser has been proposed82 for FAAS. The overall efficiency of the CVG process was found to be ∼50%, allowing for a 200-fold sensitivity enhancement for Ni when a 2 min in situ pre-concentration time was employed.
An automatic mesofluidic CV-AFS method incorporating a lab-on-a-valve (LOV) manifold for Hg determination has been developed.83 The LOV integrated a readily exchangeable commercially available sorptive microcartridge, a microscale reaction chamber for in-line vapour generation, and a membrane-less gas–liquid separator. HgII was adsorbed onto the surfaces of reversed-phase co-polymeric Oasis™ HLB beads, containing a balanced ratio of hydrophilic and lipophilic monomers, namely poly-divinylbenzene-co-N-vinylpyrrolidone. The applicability and reliability of the automated miniaturised method were ascertained through the analysis of spiked environmental waters.
The suitability of different borane reductants and acids has been investigated84 for the generation of vapour from AsII, BiIII, and SbIII, prior to ICP-OES detection. Effects of the HCl, CH3COOH, and C6H8O7 acids as well as the NaBH4. H3B–NH(CH3)2, and H3B–NH2C(CH3)3 reductants on efficiency of the As, Sb, and Bi HG were investigated and discussed. Similarly, Duan and Wang85 investigated the use of sodium tetradecahydroundecaborate (NaB11H14) as an alternate reductant to NaBH4 for Au, Pt, Pd, Co, Zn, and Mn chemical vapour generation.
Currier et al.89 demonstrated that HG-cryotrapping-AAS could be used to quantify both methylarsonate (MAsIII) and dimethylarsinite (DMAsIII) in livers of mice exposed to inorganic As. No sample extraction was required, thus limiting MAsIII or DMAsIII oxidation prior to analysis. The limits of detection were below 6 ng g−1As of tissue, making this method suitable for studies examining low exposures to inorganic As. Roman et al.90 also studied As concentrations in tissue specimens. Specifically, they studied the As behaviour in cardiovascular tissues from an As exposed coronary heart disease patient group from Antofagasta, Chile, against a small unexposed As coronary heart patient group. Measurement for total As was made by HG-AAS, and speciated As by HPLC-ICP-MS. The As concentrations in auricle, and mammary artery tissues were significantly different between both groups of patients. Also, it was demonstrated that auricle is an ‘AsIII target tissue’.
Several papers were published in which arsenic was determined with one or more vapour-forming elements. For example, HG-ICP-OES was employed91 for the determination of As and Sb in fly ash samples. The determination of Sb was significantly interfered by HF in the digestion step, but the interference could be eliminated by the addition of boric acid. An As recovery of 96% for NIST SRM 1633b was obtained following a two-step ultrasound-assisted digestion. Likewise As and Sb have been determined92 simultaneously by HG-AFS with a dielectric barrier discharge as the hydride atomiser. The low-temperature and atmospheric-pressure micro-plasma was generated in a quartz cylindrical configuration device, which was constructed of an axial internal electrode and an outer electrode surrounding the outside of the tube. The method was validated against quartz sandstone (GBW07106) and argillaceous limestone (GBW07108) reference materials. The basic principles and application of HG multi-channel-AFS for soil analysis have been described.93 As, Bi, Te, and Se were determined simultaneously; method detection limits in the soil matrix were 0.19 μg g−1, 0.10 μg g−1, 0.11 μg g−1, and 0.08 μg g−1 for As, Bi, Te, and Se, respectively. Zhang et al.94 determined the same four elements in tea leaves by HG-AFS. The operating parameters of an in-house made HG-AFS were optimised, along with the hydride forming chemistry.
Several authors reported the use of a photochemical reactor for the generation of Hg vapour. Gao et al.101 used diethyldithiocarbamate (DDTC), not only as a chelating reagent to form a hydrophobic compound for Hg preconcentration, but also as a reductant for in situ photochemical vapour generation and desorption of Hg from a coiled UV reactor. By controlling the desorption and photochemical vapour generation conditions, the proposed method can be applied for rapid speciation of HgII and MeHg+. Angeli et al.102 developed a photochemical method for the online oxidation of p-hydroxymercurybenzonate (PHMB), an organic Hg specie widely used for mercaptan and thiolic compound labelling. An integrated online UV/microwave chemical reactor led to the quantitative conversion of PHMB and thiol-PHMB complexes to HgII, with a yield between 91% and 98%. Gil et al.103 noticed a remarkable improvement in sensitivity when stopped-flow was applied to their PVG system, suggesting that kinetics play an important role in the photo-induced reduction of HgII to Hg0. In the presence of acetic acid, a LOD of 0.3 μg L−1 could be obtained, which represented a 6-fold improvement with respect to that obtained without stopped-flow.
A novel solution cathode glow discharge (SCGD) induced vapour generation system for the HPLC-SCGD-AFS determination of HgII, MeHg+ and EtHg+ has been developed.104 The decomposition of organic species and the reduction of HgII could be completed in one step with the proposed system. The system was simple in operation, and eliminated the need for auxiliary redox reagents.
Electrolytic cold vapour generation (ECVG) with AFS detection has been reported105 for the determination of trace concentrations of Hg in waters. Several buffer solutions were used to improve the detection limit by a factor of 1–2 more than in a conventional ECVG system. A phosphate buffer solution increased the signal intensity of Hg vapour from electrolytic generation on the Pt cathode, and reduced the impact of cathode erosion on the stability of signal intensity. Under the optimised conditions, the detection limit for Hg in waters was found to be 0.27 ng L−1.
Reyes et al.113 described a non-chromatographic speciation method for the determination of SbIII, SeV (and TeIV and TeVI) in cereal samples. Ultrasound assisted extraction was employed, along with HG-AFS determination. H2SO4 was found to be the most suitable extraction media (>90% for the total content of Sb) achieved without species interconversion.
Electrochemical HG has been employed115 to determine Se (and As) in drinking waters with AAS detection. Three types of electrolytic cells were constructed and optimised. Two cells (thin-layer and tubular) were finally chosen for their low inner volume and high hydride generation efficiency (ca. 90%). HG headspace SPME with ICP-OES for the determination of Se (and As) has been investigated,116 wherein hydrides were generated from the headspace of the sample solution in a 10 mL septum-sealed vial and collected onto a polydimethylsiloxane/Carboxen SPME fibre, then the fibre was transferred into a thermal desorption unit and Se was directly introduced to either an argon ICP or helium MIP.
Sun and Feng117 speciated organic and inorganic Se in Se-enriched eggs by HG-AFS by precipitating albumen with trichloroacetic acid to effect separation, with LOD of 0.07 μg L−1.
For a number of years, high-resolution continuum source AAS has been presented as a potential revolution in the field, in particular for direct analysis of complex samples. Resano and Garcia-Ruiz have attempted to evaluate the advantages of the approach, particularly for direct solid sampling.120 Three solid sampling applications are discussed, each one highlighting one of the main advantages of this technique. The review also attempts to clarify some misconceptions on the true potential of the instrumentation that is currently commercially available, such as its performance for multielemental analysis.
For several decades and for many applications, rf-GDOES has proved to be a powerful tool for the direct solid bulk materials and in-depth profile analysis. Rf-GD is generally operated in continuous mode, but the interest in pulsed rf-GD) is steadily increasing. Alberts et al.123 have attempted to assess the analytical performance of pulsed rf-GD-OES. The technique was carefully evaluated for the analysis of three conducting materials with different matrices (copper, steel and aluminium) and for bulk homogeneous non-coated glasses of varying thicknesses (1, 1.8 and 2.8 mm). The comparison was performed in terms of crater shapes, sputtering rates and emission yields, demonstrating that increased emission yields could be obtained in pulsed mode working with small pulse widths (i.e. at high pulse frequency and relatively short duty cycle). In addition, pulsed rf-GDs was applied to depth profiling analysis of a thin gold layer (approximately 30 nm) deposited on conductive and insulating substrates and an improvement in depth resolution, compared with continuous rf-GDs was observed. Successful analysis of samples including commercial tinplates and multilayered glass was also performed. Pulsed GD with TOFMS detection has also been used124 for the characterisation of nanostructured materials. Two types of nanostructured materials were successfully studied proving that pulsed rf-PGD-TOFMS allows for fast and reliable depth profile analysis as well as for the detection of contaminants introduced during the synthesis process.
The coupling of GDs as ion sources for TOFMS chemical analysis has been extensively investigated during the last two decades. In particular, the last five years have seen a number of important instrumental advances in GD-TOFMS as well as a significant increase in the description of unique analytical applications. Pereiro et al.125 provide a useful review covering instrumental developments and recent applications of GD-TOFMS, both for elemental and molecular direct solid analysis and for analytes in gaseous phase. The potential of GD-TOFMS for use in analytical research is also critically discussed.
LA in liquid (LAL) has been evaluated129 as a sampling technique for solids. LAL is widely used at present for the micro fabrication or production of nano-particles in various materials. In this study, a solid sample was placed into deionized water, and LA was carried out within the water using a Ti:S femtosecond laser operating at fundamental wavelength (780 nm). The laser-induced sample particles were trapped and collected in the ambient water. Elemental and isotopic fractionation during the LAL sampling was investigated. Five isotopic ratios were measured on the resulting sample particles obtained with various ablation conditions (fluence and repetition rate) using ICP-MS. The data obtained demonstrates the potential of the LAL technique for elemental and isotopic analyses with the careful optimization of laser ablation conditions.
A procedure enabling multi-point calibration of LA-ICP-MS for the analysis of powdered solid samples has been developed.130 CRMs available in the form of powders, e.g., soils, sediments or ashes, were used for this purpose. Stable, homogeneous and mechanically resistant targets with immobilized powder were prepared by mixing powdered solids with zinc oxide, which were then solidified by 2-methoxy-4-(2-propenyl)phenol via the formation of a zinc complex. Prepared targets were subjected to LA for multi-element ICP MS measurements. For a number of elements: Al, Ba, Co, Cr, Fe, Mg, Mn, Pb, Sb, U and V, the linear correlation coefficients between the mass of the CRM and the intensity of the signal were above 0.99. The accuracy of the proposed calibration procedure, with the use of matrix matched calibration standards, was demonstrated by successful analysis of the CRMs.
A new tungsten coil atomiser for AAS and AES has been evaluated131 which provides a more isothermal environment around the atomizer, more concentrated atomic cloud and a smaller background signal, resulting in LODs of e.g. 0.6 ppm for Cu (324.7 nm) by AAS and 4.5 ppm for Cr (425.4 nm) by AES. A portable liquid electrode plasma (LEP) atomic emission spectrometer has been described.132,133 The instrument requires no plasma gas and has been used for in situ analysis of preconcentrated extracts such as 1 M HCl soil leachate, with RSD of 20%. Plasma excited AAS has been developed for on-line analysis of industrial waters,134 in which samples are injected into, and atomised in, a nitrogen plasma jet. It is not clear from the abstract how this facilitates AA measurements with LOD of 0.25 ppm for Cu. A wavelength calibration method for flat-field grating spectrometers has been developed which provides information about the alignment of optical elements as well as higher calibration accuracy.135
Developments in detectors for AES and MS continue, with a report136 on the evaluation of a fourth-generation focal plane detector, containing 1696 Faraday-strip detectors fitted to a Mattauch-Herzog MS, detailing LODs in the single to tens of ng L−1 range for most elements, LDR of 109 and IR precision of better than 0.02%. In a related paper,137 a similar detector has been evaluated for use with SIMS and parallel detection of 29Si and 28SiH or 18O and 16OD at a sensitivity level of 230 counts s−1. A so-called ion CCD detector has been developed138 by replacing the semiconductor part of the CCD pixel with a conductor, and used in conjunction with Mattauch-Herzog MS for the detection of electrons and charged biomolecular ions. A high resolution Rutherford Backscattering method has been developed139 for the determination of trace elements in thin films of blood dried on a carbon substrate with lod of 50 ppm for Fe. A solar-blind UV detector, based on a ZrxTi1−xO2 thin film prepared by a sol–gel method, has been developed140 with dark current of less than 7 nA at 5 V bias and responsivity rate of 470 A W−1 at 270 nm.
Miniaturisation is a theme which has been prominent in recent years.141 A low power CCP micro-torch (30 W, 13.56 MHz, 0.5 L min−1 Ar) has been coupled with a commercially available microspectrometer for multi-element analysis by AES, with LODs in the range 0.003 and 1.5 μg mL−1 for Li and Mn respectively.142 Staying with the mini theme, a novel detector for GC, based on plasma excitation and AES, has been developed143 which comprises a microhollow cathode surrounded by an array of elliptically shaped collectors milled out of aluminium which couple light, via fibre-optics, into a spectrometer.
The attenuation of interferences is a topic which continues to exercise researchers. A collision reaction interface (CRI) has been utilized144 for the reduction of spectral interferences in ICP-MS, by introduction of hydrogen through the sampler or skimmer cones for correcting polyatomic interferences on 75As+ and Se isotopes in matrixes containing up to 2% v/v chloride. A different approach145 to correction of isobaric interferences in ICP-MS has been taken whereby the ArX+ signals are used to correct for variations in the signal of the interfering rather than the analyte ion; the authors claiming comparable results to those obtained with a collision–reaction interface.
Multivariate optimisation has been used to minimize interferences in Ar–N2 mixed gas ICP-MS,146 with 0.13% v/v N2 in the plasma gas and 0.11% in the central channel as a sheath gas around the nebulizer gas flow found to be optimal for interference-free determinations in the presence of 0.1 M Na.
Online electrothermal heating in the temperature range between 900 and 1200 °C has been used147 for attenuating isobaric interferences in LA-ICP-MS by up to 103 for volatile interfering elements such as Ag and Cd.
A standard addition and indicative dilution method has been developed which consists of addition of a standard solution to a sample and then successive dilution of this until the signal is equal to that produced by the undiluted sample.148 The authors claim that this approach works better than the normal standard additions approach providing the interferences are not multiplicative.
The father of GFAAS, Boris L'Vov, has written a history of the formation and development of the theory of solid-state decomposition reactions.149 The essay is based on a discussion of the mechanism of congruent dissociative vaporization of a solid with simultaneous condensation of the supersaturated vapour of the low-volatility product. The review covers a 30-year period (1981–2010), beginning with basic experimental studies in the decomposition process by ETAAS and QMS and ending with measurements of the decomposition kinetics by thermogravimetric analysis. The conclusion contains remarks and recommendations based on the L'Vov's long term participation in both the development of ETAAS and in the theory of solid-state reactions.
A variety of analytical techniques were used to characterise the condensed phase beryllium species occurring on graphite and tungsten platforms in the presence of aluminium and silicon matrices.150 The solid residue was viewed by SEM, while the chemical composition was probed by ED-XRF, FTIR spectrometry and Raman microanalysis. Beryllium oxide phases were found to be the predominant species over a wide temperature range, persisting up to about 1800 °C on the graphite platform. Beryllium metal species were also identified at high temperatures (1500 °C), but the transformation of beryllium oxide to beryllium was influenced by the amount and localised behaviour of concomitant species on the platform surface. For graphite platform atomisation, aluminium and silicon concomitants are present as metal oxides. Other silicon species, such as silicon carbide, were found mainly at temperatures above 900 °C. Little or no beryllium oxide was found on tungsten platforms up to 1800 °C, although there was evidence of some beryllium alloyed to tungsten. Tungsten from the platform supports some hydration forming different tungsten oxidation states (W6+, W5+, W4+). Also, at 900 °C, silicon was present as an oxide, but also as elemental silicon, silicon carbide, and silicon alloyed to tungsten forming tungsten disilicide at the surface interface. When tungsten platform atomisation was used for samples containing silicon, evidence of degradation of the graphite tube through formation of carbon clusters and nanostructures was more easily noticeable and evaluated by Raman spectrometry.
Ilyashenko et al.151 describe a model used for the verification of measurements made in ETAAS. The basis of the approach involves comparing the atomisation parameters obtained from samples and from calibration standards. The closer the atomisation behaviour in sample and standard, the more reliable the results are perceived to be. CRMs were analysed for heavy metal content to illustrate the feasibility of the method.
Effective simultaneous multi-element ETAAS demands rigid requirements of the design and operation of the atomiser. The system must provide a high degree of atomisation for the selected analytes, independent of the vaporisation kinetics and heating ramp residence time of atoms in the absorption volume and absence of memory effects from major sample components. Use of a low resolution spectrometer with a continuum radiation source (for simultaneous multi-element ETAAS) offers reduced sensitivity compared with traditional ETAAS. However, this reduction can be compensated, at least partially, by creating high density of atomic vapour in the absorption pulse. This has been achieved152 by using a longitudinally heated graphite atomiser, 18 mm in length and 2.5 mm in internal diameter fitted with a 2–4.5 mg ring shaped carbon fiber yarn collector. The collector is located next to the sampling port and provides a large substrate area that helps to keep the sample and its residue in the central part of the tube after drying. The collector also provides a “platform” effect that delays the vaporization and stipulates vapour release into an absorption volume that has reached a stabilised gas temperature. This furnace system allows direct simultaneous determination of a group of elements within 3–4 order concentration range. Limits of detection are close to those for sequential single element determination in FAAS.
Few fundamental studies of plasmas have been published during the review period, presumably reflecting the maturity of these atom cells. An LTE modelling approach has been used153 to determine Tion and ne in an argon ICP from MS measurements of ion ratios and their dependence on ionisation potentials, with results of 7000–8000 K and 1015 cm−3 respectively. Similarly, refinements in the method for the determination of Tgas using polyatomic ions in ICP-MS have been published,154 with several main conclusions. First, excited electronic states should be included for ArO+, neutral NO, and O2; second, a 10% error in the solvent load, sample gas flow rate, vibrational constant, rotational constant or measured ion ratio produces only a 1 to 3% error in Tgas; third, a 10% error in dissociation energy results in a 10% error in Tgas; and high temperature corrections to the partition functions can generally be neglected. The temperature of an argon-hydrogen MIP operated at 915 MHz was determined to be ca. 7000 K,155 though the intended use was for deposition of silicon coatings. PCA has been used to study intensity enhancements of 34 atomic spectral lines belonging to 33 elements and 35 ionic spectral lines belonging to 23 elements in an oscillating DCP.156 Ionisation energies, oxide bond strengths, energies of vaporisation, atomisation, and excitation were also considered in the analysis, resulting in negative correlation between enhancements and first ionisation energies for atomic lines; and positive correlation with the sum of the first ionisation energies and oxide bond energies for ionic lines, with four distinctive groups of elements.
A novel method for the determination of pressure in rock vapour plumes has been proposed157 which makes use of line broadening of Fe I at 381.58 nm and Ca I at 646.26 nm, with Stark broadening predominant for >1% ionization whereas resonance and van der Waals broadening predominates for <1% ionization.
A number of papers describing the application of chemometrics for comparisons of datasets are normally published during the review period, and this year is no exception. There is generally very little novelty in this. However, the interested reader is directed to one review of the application of ICP-MS in plant science, which includes the use of chemometrics on multi-element datasets.158
Some new chemometric algorithms have been used in spectroscopy, such as the Tikhonov regularization which is a multivariate calibration method with numerous variants such as ridge regression, which has been used for near-infrared, ultraviolet visible, and synthetic spectral data sets.159 A new method called sequential PCA has been used to identify the atmospheric sources of trace elements (determined by ICP-MS) in rainwater.160 The method of angle constrained alternating least squares has been used161 to enhance contrast in either the spectral or concentration domain by obtaining unique solutions for variables in mixture datasets such as m/z values in mass spectrometry. A neural network has been used in conjunction with AES analysis of emission from a plasma processing chamber to detect leaks, presumably because the spectral emission characteristics change.162
A generalized calibration method for analytical chemistry has been proposed163 which combines sample dilution and standard additions such that, at each stage of sample dilution, the result is estimated from six values which respond differently to errors caused by interference and non-linearity. A related paper describes its use for the determination of Cr.164
Several reviews of a general nature have been published relating to ICP-MS166 and both ICP-MS and GD-MS.167 The development of magnetic resonance mass spectrometry (MRMS) and radio frequency mass spectrometry (RFMS) has been reviewed.168 These instruments have been around since the ‘50s (and are distinct from FT-ICR-MS instruments) and have been used to determine many fundamental physical constants such as the magnetic moment of the proton in nuclear magnetons and tritium half-life.
Many reviews focus on specific fields of research, namely: isotope ratio mass spectrometry (IRMS) in the earth and environmental sciences;169 the application of LA-ICP-MS in general170 and for use in geological research;171 the use of ICP-MS for provenancing purposes;172 the study of fractionation effects for non-traditional or heavy isotopes;173,174reference materials for geological research;175,176 and the application of SIMS in geological research;126
The use of IDMS for the certification of reference materials has been reviewed,177 the relative merits of ICP-MS and TIMS compared, and sources of errors, bias and IR correction models discussed.
The accuracy and precision achievable using IDMS are exemplified by its use in an international project to determine Avogadro's constant using MC-ICP-MS,178 wherein the atoms in two 1 kg crystal spheres of isotopically enriched silicon were ‘counted’. Several papers have been published which report the steps in this process.179–181 The molar mass of Si in an artificial silicon crystal material highly enriched in 28Si was determined to be 27.97697027(23) g mol−1 with urel = 8.2 × 10−9 (k = 1), resulting in a value of 6.02214082(18) × 1023 mol−1 for Avogadro's constant.
A novel method for the determination of Ni in natural waters has been published182 which involved the extraction of Ni into MIBK as a Ni-diethyldithiocarbamate complex and direct analysis using nitrogen MIP-MS which had been doped with isotopically enriched oxygen, presumably to act as the isotopic spike for the O-containing Ni complex.
Accelerator mass spectrometry (AMS) for IDA continues to grow in popularity now that more compact systems are available. Methods have been developed for: the determination of tritium,183 an isotope which occurs at extremely low concentrations as a by-product of nuclear reactors and in landfill leachate; and for the measurement of 126Sn, via the formation of SnF2 as the target and selection of SnF3− molecular ions, yielding a sensitivity of 1.92 (±1.13) × 10−10 for the 126Sn/Sn ratio.
Several methods have been developed for absolute protein quantification by ID-ICP-MS using isotopically labelled tags. For example, Eu labelled metal coded affinity tags (MeCATs) have been used to label peptides184 with subsequent IDA for quantification. A separate group has also used 153Eu labelling by conjugating proteins with an element tag (1,4,7,10-tetraazacyclododecane-1,4,7-trisacetic acid-10-maleimidoethylacetamide loaded with europium).185 Contrastingly, LA-ICP-MS has been used for species-specific IDA of transferrin in human serum using an isotopically enriched 57Fe-transferrin complex after separation by gel electrophoresis. The authors reported ‘good agreement’ with a serum certified reference material (ERM-DA470k/IFCC) with RSD of between 0.9 and 2.7%.
Online spiking for IDA is often used for these type of analyses where a chromatographic separation is performed prior to detection using ICP-MS.3 A method has been proposed186 whereby a post-column, online double spike is introduced and the mass flow of sample and spike solution is measured with a balance, thereby eliminating errors due to changes in viscosity.
Developments in geological research include a method for the simultaneous ID-ICP-MS determination of Ca, Mg and Sr using a mixed spike of 25Mg, 43Ca and 87Sr,187 with the long-term precision of 0.4% (2s) RSD achieved for the calculation of Mg/Ca and Sr/Ca ratios in calcium carbonate.
IDMS is widely used in the nuclear industry, for which reliable isotopically enriched standards are required. The IRMM has undertaken an intercalibration exercise188 using various Pu spike reference materials for isotopes 239Pu, 240Pu, 242Pu and 244Pu and present results with uncertainties calculated using the GUM protocol.
Research into methods for improving the accuracy and precision of IR measurements continue. The correction of instrumental mass fractionation effects is a perennial problem in IRMS. One study192 investigated calculation methods for direct internal correction by TIMS and MC-ICP-MS for elements having at least two internal reference isotopic ratios. In the former case, methods based on both power and exponential laws were derived, whereas in the latter only an exponential law was considered due to the larger mass fractionation effects. A model and equations have been developed to determine the maximum level of contamination that can be tolerated for IRMS using SF-ICP-MS.193 The authors concluded that the often used 1000:1 sample-to-contaminant concentration limit can sometimes underestimate the error compared to their model calculations. The effect of sulfate on δ30Si and δ29Si measurements in natural waters by MC-ICP-MS has been studied,194 with the recommendation that samples are matrix matched by the addition of an excess of sulfate to prevent mass bias. Odd-even isotope, mass independent W isotopes has been observed in MC-ICP-MS,195 possibly related to the high first ionization potential of W, implying that other elements with I.P. > = 8 eV may also exhibit mass independent instrumental bias. Fractionation during IRA using LA-MC-ICP-MS of 147Sm and 144Nd has been corrected196 using an independently defined isochron for an Archean monazite standard, for 16–37 μm single spots, with comparable precision and accuracy to TIMS.
Compensation for detector dead-time is another important factor which has been found to be voltage dependent for an electron multiplier,197 with only a 2 ns disparity between actual and used dead-time giving rise to significant non-linearity. The method of averaging a number of ratios collected over the course of a measurement in SIMS has been shown to cause a positive bias,198 and the authors propose the use of Beale's estimator when the bias from the summed counts is unacceptably large.
A metrological triangle has been investigated199 for the measurement of isotope amount ratios of Ag, In and Sb using MC-ICP-MS, wherein Sb and In are calibrated against Ag by construction of three pairs of interelement isotope amount ratios, the so-called ‘Harvard Method’.
Broadening the bandwidth of the first laser in three-colour, three-photon ionisation RIMS, from a bandwidth of 1.8 GHz to about 10 GHz, has been found to decrease laser-induced isotopic fractionation during 235U:238U IR measurement from 10% to 0.5%.200
Somewhat worryingly, the first international inter-laboratory comparison of AMS for measurement of 36Cl:Cl ratios in AgCl found that 36Cl:Cl data from two individual AMS laboratories can differ by up to 17%!201
High spatial resolution analysis is an area which attracts much attention. Laser assisted atom probe microscopy has been used to achieve depth and lateral resolution of 5–15 nm thick 28Si and 30Si enriched layers of single crystalline silicon,202 thereby enabling the construction of a three-dimensional image. IRA of sub-micrometer Pu particles of NBL SRM-947 has been performed203 for nuclear forensics using ICP-MS after chemical separation of Pu and Am. Similarly, IRA of sub-micron uranium oxide particles using TIMS,204,205 and the development of software that can aid in the initial location and IRA of sub-micron U particles using SIMS206 have been reported.
One of the methods used most widely to achieve high spatial resolution is laser ablation (LA). One novel LA-IRMS method207 for the measurement of δ13C, with 50 μm resolution along a piece of hair, involved quantitative combustion of the ablated particles to CO2, with subsequent cryo-focusing, before detection using IRMS. Often, transient signals result from this mode of sample introduction so particular protocols are required to achieve accurate and precise IR measurements. One such208 utilised individual synthetic fluid inclusions of known Pb and Tl isotopic compositions, prepared from dissolved NIST SRM 981 with or without SRM 997, enclosed in quartz by a hydrothermal crack annealing technique, and analysed using 193 nm LA-MC-ICP-MS to yield external precision of ±0.011% (2s) for 208Pb:206Pb.
Ion microprobe MS is another technique used to achieve high spatial resolution. Extremely small samples present particular challenges to the analyst; this has been addressed209 by the development of a sample holder disk with multiple holes, in which epoxy disks containing a single unknown sample and a standard grain can be cast and polished. Chondrule particles were located within the 500 μm and 1 mm radius of the centres of the respective sized holes for δ17O measurement.
The development of new IRA methods for particular elements is an area of considerable research interest. In particular, IRA of 11B:10B isotope amount ratios has been reported for TIMS,210 using the Cs2BO2+ ion for measurement and the addition of mannitol to suppress B volatilisation during pretreatment. Similarly, SF-ICP-MS has also been used for IR measurement of δ11B.211 A demountable DIHEN nebuliser has been used with MC-ICP-MS for IR measurement of B isotopes,212 the obvious advantage being that the solution does not come into contact with glassware so washout time was reduced ten-fold, and sensitivity improved by a factor of 2–5 compared with the traditional Scott/cyclonic spray chamber and nebuliser arrangement. Precision of 0.25 ppt (2s) was achieved for measurement of 11B:10B isotope amount ratios.
The use of SIMS for IR of O, S and Fe isotopes has been reported,213 with routine precision of 0.3 ppt for measurement of 18O:16O isotope amount ratios using MC Faraday cup detectors and a 10 μm Cs+ beam.
AMS has been used in the search for long-lived isotopes of the superheavy element roentgenium (Rg, Z = 111) in gold.214 The authors observed two and nine events for 291Rg and 294Rg respectively resulting in an abundance in the 10−15 range, however, high background signal hampered positive identification as Rg isotopes.
High resolution ICP-AES has been used for IRA of U isotopes,215 by analysis of 235U at 424.412 nm and 238U at 424.437 nm in depleted and enriched samples. The authors cite the lack of requirements for matrix separation and mass bias correction as advantages compared to more sophisticated MS methods, and propose that the method can be used for rapid, fit for purpose IR measurement with reasonable accuracy (<1.5%) and precision (1%). Extractive electrospray ionisation (EESI) MS has been used216 for rapid IR measurement of 235U:238U in uranyl nitrate samples prepared from natural waters, ores and soils by nitric acid extraction, with 0.21–0.25% accuracy and 1.54–1.81% RSD.
The addition of 250 ppm Ge and 500 ppm Re with colloidal silica was used to enhance the ion beam intensity for IRA of Pb using TIMS,217 while methods have been developed for IRA of K (TIMS)218 and Mg.219
There have been some new developments in sample introduction methods. A novel sample cone, the jet cone, has been tested with MC-ICP-MS for IRA of Hf and Pb at ‘sub-ng’ levels, with RSD for Pb isotope amount ratios ranging between 0.08% and 0.36% depending on isotopic abundance and concentration. A torch integrated sample introduction system has been used220 with liquid flow rates as low as 10–15 μL min−1 for IR-MC-ICP-MS. The authors found that flow rate had a large effect on mass discrimination and recommend the use of a syringe pump to avoid errors arising from free aspiration, which were typically 0.05 ppth for 1–2% fluctuations at 10 μL min−1. The sample introduction system for IRA of Os has been modified221 by using a small glass connector to transfer OsO4 vapour directly after dissolution and in situ distillation from a Carius tube into a nebuliser without any prior separation procedure.
Several radiochemical methods have been adapted for MS. One such is the determination of 228Ra:226Ra isotope amount ratios in seawater samples by MC-ICP-MS after purification by ion exchange.222 Better precision and low LOD was reported compared to the radiochemical method.
Fractionation of isotopes during geological processes can occur for heavy elements so there is considerable research on the development of new IR methods for a wide range of isotope systems. The use of MC-ICP-MS for fractionation studies of heavy elements has been overviewed,223 with particular emphasis on Hg. The development of methods for the precise IR measurement of Mg isotopes has attracted much recent attention and has been reviewed.224 The use of the double-spike technique has been used in conjunction with MC-ICP-MS to investigate the fractionation of Zn isotopes,225 with ±0.05 ppth (2s) precision for δ66Zn (JMC Lyon Zn), in soil samples to determine plant availability; and δ53Cr (NBS 979),226 in carbonates in continental crust and the terrestrial mantle with ±0.031 ppth precision (2s).
There have been a number of publications on the use of LA-ICP-MS for IRA of B isotopes in situ,227,228 rather than the traditional method of solution analysis by TIMS which requires lengthy sample preparation. Numerous publication on the use of B isotope ratios as a pH proxy have appeared, but few detail advances in instrumental or methodological developments. One paper of note229 utilizes 11B MAS NMR and EELS to determine the speciation of trigonal B(OH)3 and tetrahedral B(OH)4−. Ion microprobe SIMS is capable of extremely high spatial resolution, and has been used to study fractionation of Fe isotopes,230 with 0.2–0.3 ppth (2s) precision compared with 0.1 ppth for MC-ICP-MS for δ56Fe measurements in magnetite. High-resolution multiple sulfur isotope analysis in sandstone has also been performed using nano-SIMS to determine δ33S (VCDT) values between −1.65 and +1.43 ppth and δ34S between −12 and +6 ppth in pyrite grains over only 5–10 μm.
Isotope ratio analysis is widely applied for geochronology. There have been numerous papers on the application of LA-MC-ICP-MS,231 SIMS and SHRIMP232 to the U–Pb dating of zircons because of the extremely high in situ spatial resolution afforded by these techniques, however, these are primarily applications, and few new instrumental developments are reported. One exception is the use of SIMS for U–Th dating of fine-grained zircons at <5 μm scale233 by using Gaussian mode primary O2− and O− probes of 5.2 μm and 4.5 μm i.d. respectively, with beam intensities of ca. 100 pA, yielding precision of 1–2%. An algorithm for U–Pb IR uncertainty propagation has been developed234 which, the authors claim, incorporates all input uncertainties and correlations without limiting or simplifying covariance terms to propagate them though intermediate calculations. This should allow a more transparent method of uncertainty propagation without making assumptions about the various contributions.
A new method has been developed for IRA of Ar, the gas being extracted from the sample by successively heating the samples with a continuous infrared laser, pre-concentrated into a capillary, separated by GC, then analysed by IR-MS at the 10−12 g level.
The use of ICP-MS and IRA for provenancing studies has been reviewed.238 MC-ICP-MS has been used for high precision IRA of Mg to determine the absolute composition of the Earth's mantle, and hence that of bulk silicate earth,239 yielding values of 25Mg:24Mg = 0.126896 ± 0.000025 and 26Mg:24Mg = 0.139652 ± 0.000033. The authors propose that this provides a basis for comparison with suspected extra-terrestrial specimens. A combination of MC-ICP-MS and IRMS has been used for IRA of Sr and O respectively240 to determine the origin of green coffees after PCA, with good discrimination between South American and islands point of origin. This statistical approach is also evident in a study in which Q-ICP-MS was used241 to determine 11B:10B and 87Sr:86Sr isotope amount ratios to distinguish between South African wines. The use of B isotope ratios together with elemental concentrations and LDA was found to be the most successful approach.
In the nuclear forensics field, isobaric interference of 238U on 238Pu has been reduced by using 0.4 to 1.5 mL min−1 CO2 as a reactive gas in a collision-reaction cell coupled with MC-ICP-MS,242 thereby facilitating IRA of U and Pu isotopes for nuclear waste management, with only 0.38% and 0.11% disparity from a chemical separation method respectively. Staying with the nuclear theme, a method has been developed204 using SEM to identify uranium oxide particles <1 μm in diameter and then TIMS to determine the U and Pu isotope amount ratios at the fg level.
8-HQ | 8-Hydroxy quinoline |
AA | Atomic absorption |
AAS | Atomic absorption spectrometry |
AAS | Atomic absorption spectrometry |
AFS | Atomic fluorescence spectrometry |
AMS | Accelerator mass spectrometry |
APDC | Ammonium pyrrolidine dithiocarbamate |
CCD | Charge coupled detector |
CCP | Capacitively coupled plasma |
CHG | Chemical hydride generation |
CNT | Carbon nano-tube |
CPE | Cloud point extraction |
CRI | Collision reaction interface |
CRM | Certified reference materials |
CS | Continuum source |
CV | Cold vapour |
CV-AAS | Cold vapour atomic absorption spectrometry |
CVG | Cold vapour generation |
DCP | Direct current plasma |
DDTC | Diethyldithiocarbamate |
DIHEN | Direct injection high efficiency nebulizer |
DLLME | Dispersive liquid–liquid microextraction |
DMA | Dimethylarsinic acid |
DPC | Diphenylthiocarbazone |
ECVG | Electrolytic cold vapour generation |
EDTA | Ethylenediaminetetraacetic acid |
EDXRF | Energy dispersive X ray fluorescence |
EELS | Energy electron loss spectroscopy |
ETAAS | Electrothermal atomic absorption spectrometry |
ETV | Electrothermal vaporization |
FAAS | Flame atomic absorption spectrometry |
FI | Flow injection |
FIA | Flow injection analysis |
FTIR | Fourier transform infrared spectroscopy |
GC | Gas chromatography |
GD | Glow discharge |
GD-MS | Glow discharge mass spectrometry |
GD-OES | Glow discharge optical emission spectroscopy |
GF | Graphite furnace |
GFAAS | Graphite furnace atomic absorption spectrometry |
GUM | Guide to the expression of uncertainty in measurement |
HG | Hydride generation |
HG-AAS | Hydride generation atomic absorption spectrometry |
HPCN | High performance concentric nebuliser |
HPLC | High-performance liquid chromatography |
ICP | Inductively coupled plasma |
ICP-AES | Inductively coupled plasma atomic emission spectrometry |
ICP-MS | Inductively coupled plasma mass spectrometry |
ICP-OES | Inductively coupled plasma optical emission spectrometry |
IDA | Isotope dilution analysis |
IDMS | Isotope dilution mass spectrometry |
IR | Isotope ratio |
IRA | Isotope ratio analysis |
IRMM | Institute for Reference Materials and Measurements |
IRMS | Isotope ratio mass spectrometry |
IUPAC | International Union of Pure and Applied Chemistry |
LA | Laser ablation |
LA-ICP-MS | Laser ablation inductively coupled plasma mass spectrometry |
LEP | Liquid electrode plasma |
LIBS | Laser induced breakdown spectroscopy |
LIPS | Laser induced plasma spectroscopy |
LLME | Liquid–liquid microextraction |
LOD | Limit of detection |
LOQ | Limit of quantification |
LTE | Local thermal equilibrium |
MA | Methylarsonic acid |
MC-ICP-MS | Multicollector inductively coupled plasma mass spectrometry |
MIBK | Methyl isobutylketone |
MIP | Microwave induced plasma |
MIP-AES | Microwave induced plasma atomic emission spectrometry |
MIP-MS | Microwave induced plasma mass spectrometry |
MIP-OES | Microwave induced plasma optical emission spectroscopy |
MMA | Monomethyl arsenic |
MRMS | Magnetic resonance mass spectrometry |
MS | Mass spectrometry |
MWCNT | Multi-wall carbon nanotubes |
n e | Electron number density |
NBL | New Brunswick Laboratory |
NIST | National Institute of Standards and Technology |
NMR | Nuclear magnetic resonance |
OES | Optical emission spectroscopy |
PCA | Principal component analysis |
ppb | Parts per billion |
ppm | Parts per million |
PTFE | Poly(tetrafluoroethylene) |
QMS | Quadrupole mass spectrometry |
QT | Quartz tube |
rf | Radio frequency |
RFMS | Radio frequency mass spectrometry |
RIMS | Resonance ionization mass spectrometry |
RSD | Relative standard deviation |
SEM | Scanning electron microscopy |
SF-ICP-MS | Sector field inductively coupled plasma mass spectrometry |
SHRIMP | Sensitive high mass resolution ion microprobe |
SIMS | Secondary ion mass spectrometry |
SPE | Solid phase extraction |
SRM | Standard reference material |
T e | Electron temperature |
T gas | Gas temperature |
TIMS | Thermal ionization mass spectrometry |
T ion | Ionization temperature |
TMAH | Tetramethylammonium hydroxide |
TOFMS | Time of flight mass spectrometry |
USN | Ultrasonic nebuliser |
v/v | Volume per volume |
VCDT | Vienna Canyon Diablo Troilite |
w/v | Weight per volume |
XRF | X-ray fluorescence. |
This journal is © The Royal Society of Chemistry 2012 |