Atomic spectrometry update: review of advances in atomic spectrometry and related techniques
E. Hywel
Evans
*a,
Jorge
Pisonero
b,
Clare M. M.
Smith
c and
Rex N.
Taylor
d
aThe Open University, UK. E-mail: e.h.evans@open.ac.uk
bUniversity of Oviedo, Faculty of Science, Department of Physics, c/Federico Garcia Lorca, 18, Oviedo 33006, Spain
cSchool of Education, College of Social Sciences, University of Glasgow, University Avenue, Glasgow G12 8QQ, UK
dSchool of Ocean and Earth Science, University of Southampton, NOC, Southampton SO14 3ZH, UK
First published on 5th April 2023
Abstract
This review of 200 references covers developments in ‘Atomic Spectrometry’ published in the twelve months from December 2021 to November 2022 inclusive. 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 previous review1 and the other related reviews in the series.1–6 A critical approach to the selection of material has been adopted, with only novel developments in instrumentation, techniques and methodology being included. After several years of consolidation of research in the area of elemental tagging, with very similar variations on the same themes being published, this has now evolved to include detection at the cellular level. This is largely a result of advances in single particle-ICP-MS and LA-ICP-MS mapping, coupled with multiple tagging approaches. These combined methods are very useful for diagnostics and imaging and perhaps show the way forward for this type of analysis. A more comprehensive assessment of the range of factors affecting the accuracy of isotope ratio determinations of non-radiogenic isotopes has been in evidence. Where corrections could not be made, some studies specified a range of concentration or molarity in which results remained within tolerance. In the future, this type of assessment should lead to more robust uncertainty estimates and potentially reduce the possibility of samples with erroneous deviations in their stable isotopes measurements.
1 Liquids analysis
1.1 Sample pre-treatment
Much of the focus of recent research has been on the development of ‘green’ sample preparation methods using alternative solvents or analytical approaches. Santana et al.7 reviewed this area (341 references) by investigating the role of microwave-assisted wet digestion, microwave-assisted extraction, ultrasound-assisted extraction, microwave-induced combustion and alternative LPE methods using e.g. alkaline and deep eutectic solvents. There was also a section on direct analysis methods which do away with the sample preparation step altogether (ideally). The review contained comprehensive tables of methods arranged by analyte.Sorouraddin et al.11 synthesised a ternary DE solvent from menthol, mandelic acid, and glycolic acid for the extraction of Pb from gasoline and its determination using ETV-AAS. The LOD and LOQ were 1.6 and 5.0 ng L−1 respectively, with an enrichment factor of 166× and 91.3% recovery. A similar approach, using ETV-AAS, was adopted by Shamsipur et al.,12 who used a DE solvent prepared from salicylic acid and l-menthol for the extraction of Cd from water and food samples. They achieved an LOD and LOQ of 0.37 × 10−4 and 1.24 × 10−4 μg L−1 respectively, with a preconcentration factor of 125×. The authors of the two papers claimed these as ‘green’ methods due to the absence of complexing agents.
After several years of consolidation, with very similar variations on the same themes being published, these types of tagging approaches have now evolved to detection at the cellular level. Yao et al.13 attached bacteria-specific peptides to a range of magnetic nanoparticles (MNPs). Hence, they were able to selectively tag S. aureus, S. typhimurium and V. parahaemolyticus with Au-, Ag- and Cu-nanoclusters which had been functionalised with thiol-modified aptamers for use as recognition probes. The bacteria were captured by the MNPs@peptide and aptamer modified nanoclusters to form sandwich-type MNPs@peptide-bacteria-element tag structures. After magnetic separation, the unbound nanoclusters were extracted, digested and determined using ICP-MS, thereby enabling quantitation of the original bacteria. The minimum detected concentration of bacteria was 10 CFU mL−1. The authors cited three main advantages of this method: speed (<30 min); simplicity compared with methods that involve nucleic acid extraction; and multiplex advantage. A similar approach was adopted by Huang et al.,14 who used concanavalin A-conjugated Fe3O4@SiO2 NPs and aptamer-modified AuNPs to tag S. aureus bacteria. Both ICP-MS and SERS were used for detection, potentially allowing selective determination of different strains of bacteria. The LOD for S. aureus was 11 CFU mL−1. If this method could be commercialised on a large scale so as to be affordable in most labs, and proven to be accurate and reproducible, this would be a game changer for bacterial food safety, as most tests take 3–5 days.
Sun et al.15 reported an interesting and novel adaption of the tagging method to count cancer cells, using amino modified NaYF4:Yb, Er nanoparticles. 100 μL of a suspension of MCF-7 cells was labelled with 10 μL of the probe solution, diluted 20× with PBS buffer, then separated into a single cell flow using a microfluidic chip separator. The eluting solution was nebulised directly into ICP-MS using mini nebulisation and the 89Y-immunolabeled MCF-7 cells detected in a time-resolved manner. In order to test the approach for a real sample, blood samples from three breast cancer patients and healthy volunteers were analysed. Cancer cells were observed in blood samples from patients with breast carcinoma, which were found to be 1–3 cancer cells (with a mean of 2.00 ± 0.87) per 20 μL of blood, whereas no cancer cells were detected in the healthy volunteers. An alternative approach to cancer diagnosis was reported by Wen et al.16 They developed a tagging approach using multiple exosomal miRNAs conjugated with magnetic beads and NaLnF4, and catalytic hairpin assembly amplification. They used ICP-MS to determine ten types of miRNAs simultaneously, with an LOD of 0.01 fM. Linear discriminant analysis was also applied to achieve a discrimination accuracy, between the RNA from ten cell exosomes, of 98.6%. This analysis achieved 100% cancer diagnosis accuracy for a sample of 42 patients, including five cancer types and healthy controls.
Elemental tagging to facilitate cellular imaging is another example of tagging at the cellular level. Reifschneider et al.17 used a double labelling strategy to quantitatively determine Ag NPs in lung tissue. Lung tissue samples were incubated with an anti-rat CD68 antibody. Double labelling was achieved by first incubating with Alexa Fluor®488 conjugated anti-mouse IgG, then Au-coupled anti-mouse IgG antibody. This allowed identification of regions of interest with fluorescence microscopy, followed by detection of identified cells using LA-ICP-MS. This ‘fluorescence plus gold’ double immunostaining method allowed high resolution optical detection with quantitative elemental mass spectrometry for the detection of Ag NPs, calibrated with matrix-matched standards.
Elemental tagging in combination with sp-ICP-MS is an approach which offers considerable promise. Xu et al.18 used rolling circle amplification (RCA) and sp-ICP-MS to determine DNA from the hepatitis B virus. Tagging was undertaken using DNA probes conjugated with Au NPs. Agglomeration of Au NPs into larger particles, by the addition of RCA products and spermidine, was necessary to achieve an LOD of 5.1 fmol L−1 for the target DNA. Tanaka et al.19 developed a method for the quantitation of a histidine-tagged recombinant protein produced in E. coli. A plasmid vector containing green or red fluorescent protein was used to tag E. coli which was then incubated in a solution of Ni or Co to label the His-tag. Single cell -ICP-MS was used to determine the amount of labelled protein on the basis of the signal of Ni or Co bound to the His-tag. In order to do this quantitatively it was necessary to determine both the number of E. coli cells and the amount of intracellular recombinant protein. This was achieved by measuring the number of transient signals for P as 31P16O to estimate the number of cells in a suspension. Transport efficiency (f) was corrected for using the f value for SiO2 NPs.
Nosti et al.20 examined the status of ICP-MS methods used for protein quantitation. They concluded that prior separation using LC is necessary for accurate quantitation and that parallel analysis using ESI-MS is probably necessary. Furthermore, certification of standard protein concentration and mass purity is one area where ICP-MS shows promise, because of its applicability to both simple and relatively complex protein mixtures, particularly when using IDMS and label-free methods.
1.2 Sample introduction
Aerosol dilution (AD) is a sample introduction tool that allows the direct analysis of complex matrix samples by plasma-based analytical techniques. In AD, a controlled gas stream (dilution gas) is added at the tertiary aerosol zone, between the spray chamber and the torch. As a result of the dilution of the aerosol, there is an improvement in plasma matrix tolerance and plasma robustness, when compared with pneumatic nebulisation without AD. Therefore, samples with higher total dissolved solids can be introduced with minimal sample treatment. The use of AD can also help to minimise dilution errors, liquid waste, and sample contamination. The use of AD has been reviewed, with 109 references.22,23 The effects of AD on plasma, analytical signal, signal: background ratio and interferences were described. Strategies for AD ICP-MS optimisation resulting in maximum performance monitored other parameters, such as plasma applied power and sampling depth, in conjunction with the dilution gas flow rate, while maintaining combined nebuliser gas and dilution gas flow rates of around 1 L min−1. Applications of AD for the analysis of food, environmental, geological, pharmaceutical, biological and clinical samples were also discussed.
As the use of microfluidic chips for handling biological samples increases, so does the need to combine them with sensitive techniques for analyte determination such as ICP-MS. Coupling a microfluidic chip to an ICP-MS has been demonstrated using conventional pneumatic micro-flow nebulisers. However, these nebulisers can cause dead volume issues and liquid suction exerted on the outlet channel of the chip. Mavrakis et al.24 described the use of a microfluidic chip with a pneumatic nozzle for liquid nebulisation ICP-MS. They demonstrated that coupling a chip-based supersonic microfluidic nebuliser (chip-μf-Neb) to ICP-MS can be achieved using a spray chamber with a laminar flow makeup gas. Operation of the combined system was evaluated at low liquid flow rates (0.5–20 μL min−1), while nebulisation and makeup Ar flow rates were optimised with respect to maximising in sensitivity and minimising oxide formation. A maximum sensitivity of 40![[thin space (1/6-em)]](https://www.rsc.org/images/entities/char_2009.gif)
000 cps (μg L−1)−1 was achieved at 10 μL min−1. The system was evaluated for its performance in single-particle analysis, featuring a TE of 46% for Ag NPs. Single-cell analysis was also demonstrated by detection of 80Se and 75As in individual Chlamydomas reinhardtii cells, after incubation in 20 μM of selenate and 300 μM of arsenate, respectively. Efficient operation at low liquid flow rates along with the absence of self-aspiration were aspects that showed the nebuliser to be a promising tool for combining microfluidics with ICP-MS.
In sp-ICP-MS, TE is fundamental for the correct determination of both PNC and size. In an interlaboratory comparison study, TE was systematically determined on three different days with six characterised AuNP suspensions in seven European and US laboratories using different ICP-MS instruments and sp-ICP-MS software.26 Both particle size-(TES)-and particle frequency-(TEF)-methods were applied. The reported TE did not deviate by much under ideal conditions. However, the TEF results provided systematically lower TEs. The extent of this difference (0–300% relative difference) depended largely on the choice and storage conditions of the NP suspensions used for the determination. The TES method was, therefore, recommended when particle size measurements are the primary aim. If PNC determinations are required, the TEF approach was preferred as it appears to compensate for particle losses in the sample introduction system. Analysis of NPs in solution using sp-ICP-MS requires measurement of the sample TE when using solutions for calibration. Zhou et al.27 described a total consumption IR heated sample introduction system to eliminate this requirement while preserving plasma robustness. A 50 mL cyclonic spray chamber was modified so that a pen IR heater could be introduced within the baffle in its centre. Two different gaps between the top of the baffle and the top service of the spray chamber were tested (2 and 6 mm). Sample uptake rate (25–75 μL min−1) and IR heating temperature (20–300 °C) were optimised for the detection of Au. Using the spray chamber with a 2 mm gap at a 50 μL min−1 sample uptake rate and 110 °C IR heating temperature, 99.2% transport efficiency was achieved versus 17% using the standard pneumatic nebulisation system with a double-pass spray chamber. As a result, the IR-heated system decreased the solution detection limit by a factor of 5 and improved the method size detection limit from 26 to 16 nm, versus the standard system with a Scott double-pass spray chamber, without degrading the accuracy for the measurement of 60 nm Au NPs.
Although sp-ICP-MS is currently widely applied to different fields, challenges remain in obtaining accurate and consistent PNC measurements. A systematic assessment of sp-ICP-MS capabilities for measuring the PNC of AuNP suspensions of different sizes and coatings was carried out.28,29 sp-ICP-MS direct and derived determination of PNC and reference PNC derived based on the mean particle size or the particle size distribution obtained by different reference sizing techniques were compared using NIST AuNP RM 8012, nominal diameter 30 nm. Regardless of NP size or coating, good agreement (90–110%) between sp-ICP-MS direct determination of PNC and reported PNCs was obtained for all suspensions studied when reliable in-house Au mass fractions and thorough mean particle size determinations were included in the calculation of the derived PNCs. The use of the particle size distribution over the mean size to derive PNCs resulted in larger differences for materials with a low concentration (<2%) of smaller NPs (30 nm), materials with a higher polydispersity, or materials with two distinct subpopulations of particles, regardless of NP coating.
Double-viewing-position (DVP) sp-ICP-OES was used to investigate the effects of temperature gradient in the ICP central channel on sp-ICP intensity.30 The correlation plot of the intensities of 203Yb+ from individual Yb2O3 particles at observation positions of 8.5 and 19.5 mm above the load coil showed extensive scattering. The lack of correlation was attributed to the relatively large gradient of ICP gas temperature in the radial direction at the lower observation position. The hypothesis was supported by results of sp-ICP-OES measurements obtained using a sheath gas device to confine the particles in either the centre or the outer region of the ICP central channel. Particles in the two regions show distinct patterns in the correlation plot. Computer simulation shows that the heat required to vaporise Yb2O3 particles is substantial and that the duration of the heat transfer process increased with particle mass. As a result, large particles had a relatively small degree of vaporisation (DOV) at low observation positions where sp-ICP-OES intensity was not proportional to particle mass. Calibration graphs become concave and errors in particle size determination at low observation positions were noted. A simulation of particle vaporisation with a wide range of particle sizes and boiling points was carried out and a lookup graph of the observation position for DOV = 70% versus particle diameter was provided to aid selection of initial observation positions for sp-ICP-OES measurement.
Dual-mass measurement of individual particles using quadrupole-based ICP-MS, with the assistance of O2 collision gas, was described by Chun et al.31 Simultaneous measurement of the intensity of 107Ag and 109Ag in AgNPs showed particle recovery of 100% and a Pearson correlation coefficient of 0.97, indicating effective sampling of all particles in the ICP. The technique was applied to determine the elemental content and isotope ratios of NPs and to study cell viability after cisplatin staining. The results were comparable to that of existing TOF and multi-collector ICP-MS, indicating that quadrupole-based ICP-MS could be a cost-effective alternative for simultaneous measurement of two isotopes.
Matrix interferences remain a significant challenge to accurate NP sizing and number concentration determination by sp-ICP-TOF-MS. Online microdroplet calibration, involving multi-element droplet standards mixed with nebulised aerosols from a NP-containing sample in a dual-sample introduction system before entering the ICP, was used to determine absolute sensitivities and calibrate the mass amount(s) of analyte element(s) in particle events.32 A plasma-uptake standard was added to NP-containing samples to account for variations in the nebulisation and TE of aerosols into the plasma, which were used in the determination of PNCs. Since the microdroplets experienced the same plasma conditions as analyte NPs, matrix-matched calibration was achieved. Online microdroplet calibration was evaluated for the quantification of NPs in three matrices: ethanol, acetone and acetonitrile. The approach was found to correct for the commonly observed increase in sample uptake and plasma-related signal attenuation in the determination of NPs in organic matrices. The results demonstrated accurate NP sizing and PNC determinations in organic matrices up to 98% (v/v), even when these matrices caused signals to be attenuated up to 35× and nebuliser TE to be up to 4× higher than that of a pure water matrix.
NPs with high mono-dispersity, size, shape and surface chemistry control are frequently synthesised using hot-injection methods, using hydrophobic organic ligands which are only soluble in non-polar organic solvents. Due to several instrumental limitations, suspensions in organic solvents are not commonly analysed by sp-ICP-MS. The direct introduction of toluene and mesitylene into an ICP-MS using a microdroplet generator was investigated by Kocic et al.33 in order to overcome these limitations. With this configuration the solvent load in the ICP was substantially reduced and soot formation (which causes instrumental drift) was minimised while maintaining a high TE. The effect of different vacuum interface configurations and the addition of O2 or N2 on the detection efficiency and instrumental background signals was investigated for Al, Ag, Cd, Cu, Fe, Pb, Si, and Ti. The highest DE was obtained for a “jet” interface with the addition of N2 at a flow rate of 10 mL min−1, resulting in an increase of a factor of between 2× and 8× depending on the element. The lowest detectable mass, based on counting statistics, was 1.4 ag for Pb, which corresponded to a diameter of 6.1 nm of a spherical, metallic NP. The approach was demonstrated for NP characterisation and showed promise for the sensitive determination of trace elements in organic solvents.
Using FI for sample introduction in sp-ICP-MS can simplify the determination of NPs by eliminating the requirements to measure the sample uptake rate and the TE. However, FI results in sample dispersion that degrades both the sensitivity and the LODs obtained. To avoid this degradation, monosegmented FI was used.34 This involved the injection of a discrete volume of sample solution into a pocket of air within the carrier. This not only minimised sample dispersion on the way to the nebuliser, but also increased sample TE from 6.3–8.4% compared with the unsegmented approach. This, in combination with the square-wave shape of the peak, improved sensitivity for the determination of dissolved analyte without degrading the measurement of NPs. With a 10500 NPs mL−1 suspension to avoid NPs coincidence, the size of Au NPs was measured with 0.6% RSD. However, the RSD for Pt NPs of similar size was found to be 14% because of the higher LOD compared to that for monoisotopic Au.
Zhen et al.37 presented a review (86 references) of PVG which focused on methods to improve efficiency. They also investigated the PVG of Re, observing a 19-fold enhancement in the presence of 20% (v/v) formic acid, 15% (v/v) acetic acid, 30 mg L−1 of Cd2+ and 5 mg L−1 of Co2+. EPR spectroscopy was used to investigate the mechanism, giving evidence that the radical adducts DMPO–˙CO2− and DMPO–˙CH3 increased in the presence of Cd2+ and Co2+. The authors speculated that these ions accelerated the ligand-to-metal charge transfer process, and the increase of strongly reducing ˙CO2− may have been the main reason for the improvement of vapour generation efficiency of Re, though this was unproven. An LOD of 1 ng L−1 was obtained for Re using ICP-MS. Musil et al.38 also observed a 3200-fold enhancement in PVG using Co2+ and Cd2+ catalysts in 8 M formic acid, with a thin-film flow-through photoreactor coupled to ICP-MS. An LOD of 20 pg L−1 was obtained. Zeng et al.39 reported up to a 133-fold enhancement for PVG of Sb in the presence Co2+ and formic-acetic acid, with an LOD of 0.001 μg L−1 using ICP-MS detection. Dong et al.40 observed a 55-fold enhancement in the PVG of Te in the presence of formic-acetic acid and a VO3− catalyst, with an LOD of 2.9 ng L−1 using ICP-MS. Cai et al.41 used a metal organic framework (MOF) supported catalyst to both preconcentrate and enhance PVG for Ni. An LOD of 0.03 μg L−1 was obtained using DBD-OES detection. Separation of matrix ions from solution was an advantage compared with direct PVG, resulting in a 43-fold improvement in the LOD.
A miniaturised UV-LED on-a-chip was developed by Li et al.42 for PVG of Se. The PV generator was comprised of a quartz coil, internally coated with nano-TiO2 as catalyst, placed over a UV-LED chip which contained 9 LED beads (370 nm, 50 W). The reaction solution was pumped through the PVG reactor and then into a gas–liquid separator before detection using a hollow electrode point discharge microplasma OES instrument. LODs of 3.9 μg L−1 and 4.1 μg L−1 for SeIV and SeVI, respectively, were obtained in a reaction medium of 10% (v/v) acetic acid.
Plasma induced vapour generation (PIVG) shares some similarities with PVG because some of the same mechanisms may be involved. The DBD has proved a popular choice of plasma for vapour generation. He et al.43 developed a variation of this approach by nebulising the sample solution directly into a DBD chamber. The DBD was constructed from an inner glass sleeve containing a copper electrode, surrounded by an outer tube wrapped with copper foil to serve as the counter electrode. The maximum power output was 65 W at 28 kHz and 25 kV. The plasma was sustained by the nebuliser gas flow of Ar directed into an ICP-MS instrument. Volatile chelates of REEs were generated using 2,2,6,6-tetramethyl-3,5-heptanedione as chelating reagent mixed with the sample solution, resulting in up to 9-fold increased sensitivity compared with solution nebulisation. LODs for 15 REEs were between 0.0009 and 0.11 ng L−1. The same group of researchers44 also used this system for the PIVG of CrVI using sodium diethyldithiocarbamate as a chelating reagent, achieving an LOD of 0.023 μg L−1 using ICP-MS measurement.
Pan et al.45 investigated PIVG mechanism for Cd, Hg, and Zn in a LEP discharge, using OES, GC, GC-MS and AFS. The role of e−, H˙, H3C˙and other intermediates in the argon plasma was thought to be influenced by the addition of methanol to cause enhancement in PIVG of 2.7-, 4.8-, and 7.9-fold for Hg, Cd, and Zn respectively, compared to those obtained in the absence of methanol, with respective LODS of 0.007, 0.05, and 0.5 μg L−1 using AFS detection. A solution anode GD was developed by Cai et al.,46 formed from a tungsten rod cathode and a protruding glass capillary carrying the sample solution, which overflowed from the tip, and formed a discharge with a graphite rod anode. LODs for Cd (0.3 μg L−1) and Hg (0.2 μg L−1) were improved by 12- and 90-fold respectively, in comparison with conventional PN using ICP-OES detection.
Research into the mechanisms of chemical vapour generation (CVG) is still an active area despite this method being very well established. Zhen et al.47 studied the effect of arsenazo III (ARS) on HG of REEs using NaBH4. Both REE NPs and REE-ARS chelates were identified as products of the HG reaction, suggesting a dual-route mechanism. LODs of between 0.004 and 0.15 μg L−1 were obtained using ICP-MS detection. Kratzer et al.48 studied the atomisation of SeH2 AsH3, CH3AsH2 and (CH3)2AsH in a DBD. They used AAS and selected ion flow tube mass spectrometry (SIFT-MS) to simultaneously study atomised and non-atomised species formed by HG using NaBH4. Their observations suggested that the As- species were completely atomised but SeH2 was less than 80% atomised in a ∼17 W DBD. In order to confirm the role of H2, which is formed during HG and is necessary for atomisation in the DBD, they also studied AsH3 in the absence of H2 by introducing AsH3 gas directly from a cylinder. Under these conditions As atomisation was not observed, rather the hydride was decomposed by an unresolved reaction pathway. The DBD is very sensitive to the quenching effects of water vapour, so SIFT-MS was used to determine the amount of water vapour either transported to, or produced in the DBD, and thus evaluate the effect of various drying methods. Unfortunately, the optimal drying method was found to be analyte dependent so more research in this area to find a universal dryer would be useful.
A novel extraction/preconcentration method was developed by Lourido-Grovas et al.49 They immobilised Ag NPs on a cellulose substrate by the simple expedient of immersing a filter paper, or cellulose acetate membrane filter, in AgNO3 solution then reduction with NaBH4. AsH3, SbH3 and BiH3 were generated using continuous flow HG, and retained on the AgNP-cellulose substrate upon catalytic decomposition. The substrates were subsequently digested, or the analytes desorbed, and analysed using ICP-MS. A 10 min preconcentration yielded enrichment factors of 16×, 9× and 26× for As, Sb and Bi respectively, resulting in respective LODs of 15, 2 and 1 ng L−1.
1.3 Direct methods
Solution cathode (SC-) GD is typically sustained between a metallic pin-anode and a liquid cathode. Sample solutions are introduced to the GD as a flowing stream from a cylindrical capillary. Hazel et al.50 described a system which sustained the SC-GD plasma in a horizontal arrangement between a flat anode and a liquid cathode constructed from a thin, rectangular capillary from where the liquid emerged. This arrangement created a sheet-like plasma where the negative glow of the SC-GD approximated the shape of the entrance slit of the spectrometer, improving the efficiency of optical sampling. In the determination of 24 elements, LODs up to 33× lower that those obtained using conventional SC-GD were obtained with the new arrangement. Improvement in LOD was particularly significant for elements with atomic emission concentrated near the liquid cathode surface. Zheng et al.51 suggested a refrigerated anode to improve the analytical performance of SC-GD-OES. The spectral characteristics of metal anodes at different temperatures were studied in the range from 10 to 30 °C and the spectral emission signals of metals were found to be enhanced at lower anode temperatures during the plasma discharge. In addition, the signal stability was improved. Compared with the unrefrigerated system under the same experimental conditions, the LODs determined for Ag, Cd, Cu, Mn and Zn were 0.6, 9.1, 5.3, 8.7 and 16 μg L−1, respectively, which represented improvements by factors of 6.5, 3.5, 3.2, 3.9 and 3.1, respectively at the lowest temperature compared with the highest. Analysis of CRM BWB2446-2016 and a water sample from the Yangtze river agreed with the reference values measured by ICP-OES. Results suggested that refrigerated anode SC-GD-AES provided an effective analytical method for in situ, real-time and on-line determination of metal elements in water samples. Improvements in the detection of metal ions in aqueous solution by SC-GD using a pulsed power supply to the plasma have also been reported by Zheng et al.52 The operating parameters, including pulse excitation and maintenance discharge voltage, pulse width, pulse modulation duty ratio, solution flow rate, and detector delay time, were described. The LODs obtained for Na, K, Cu, Co, Cd, Mn, and Ag were 0.080, 0.050, 3.4, 64, 69, 9.1, and 2.0 μg L−1, respectively. These values represented improvements by factors of 2.9, 1.5, 18, 5.5, 25, 1.6 and 2.1 times, respectively, compared with DC excited SC-GD.A liquid anode (LA)-GD was used as a miniaturised excitation source for OES with FI and a fibre spectrometer to form a LA-GD-OES system for the detection of Hg in water and ore samples.53 Addition of 1.5% methanol and 5% formic acid was used to enhance the signal intensity of Hg and reduce interference from Ag+, Al3+, Ga3+, Mn2+, F− and I−. The LOD for Hg was found to be 8.0 μg L−1 for 1.5% methanol and 4.5 μg L−1 for 5% formic acid. The results of the analysis of spiked water and ore samples were in good agreement with the spiked value and the verification values measured by ICP-OES, with recoveries from water and ore samples in the range from 88 to 16% and from 101.6 to 106.3%, respectively. LA-GD-AES was proposed as a useful technique for the determination of Hg due to its compact and portable design, low energy consumption, high discharge stability and no inert gas requirement.
The direct determination of ultratrace Pb in a single human hair by DC-APGD-OES coupled electromagnetic heating vaporisation (EMV) was described by Cai et al.54 A sample mass of 0.15 mg (often a single human hair) was required for the analysis. An LOD of 30.8 μg kg−1 (4.8 pg) for Pb was obtained under optimised conditions. The method was verified by analysis of CRM GBW09101b (human hair) and human hair samples from three volunteers. This new method was reported as a simple, efficient and low-cost method for detecting Pb in human hair.
The use of GD-OES for matrix independent determination of oxygen was investigated using the spectral lines of O at 130 nm and 777 nm and standard conditions for DC discharge with a 4 mm anode (700 V, 20 mA).55 Analysis of calibration samples of Cu-, Al- and Mg-powder mixed with their oxides, at 130 nm, confirmed the matrix dependence of the emission yield. At 777 nm, however, O had the same emission yield in these matrices. In order to compare the emission yield of O with the emission yield of Fe, a thick 43 μm FeO-layer was prepared and characterised by Rutherford backscattering spectrometry, XRD and GD-OES. At 130 nm, the emission yield of O in FeO was similar to that in an Al-matrix. At 777 nm, the calibration revealed a higher emission yield for O in FeO in comparison to the common emission yield of O in Cu-, Al- and Mg-matrices.
Swiderski at al.56 described the use of a hanging drop cathode (HDC) APGD coupled with a Dove prism as an excitation microsource for OES. The effect of the addition of low molecular weight organic compounds on the intensity of the analytical lines of Ag, Ca, Cd, Hg, Mg, Mn, Pb and Tl were evaluated. The greatest enhancement was obtained for 8% (m/m) formic acid. The 2-D spatial distribution profiles of the emission lines of the selected elements) and the background species (emission of H, O I, O II, N2, OH, and NO) were obtained. Enhancement of the Ag, Pb and Tl atomic emission lines resulted in LODs of 0.54, 14 and 2.2 μg L−1, respectively.
2 Solids analysis
2.1 Direct methods
2.2 Indirect methods
Advances in LA-ICP-MS in recent years have led to reduction in analysis time, higher spatial resolution and improved sensitivity. Quantification and accurate analysis are still problematic, however, and matrix matched calibration solutions and internal standards are required. Additional uncertainties associated with laser fluence and beam size using various ablation cells and interfaces make quantification even more challenging. These additional factors are considered by Jerse et al.63 using approaches based on pulse intensity, LA spot volumes and noise characteristics for As, Gd, La, Ni, Te and Zn. A range of analyte concentrations (between 10 and 1000 μg g−1), and matrices (gelatine standards and NIST SRM 612) were studied. The findings indicated that selection of the appropriate laser fluence, just above the ablation threshold, and beam size, depending on the interface of LA and ICP-MS, were critical for reliable quantification and should be properly adjusted to avoid excessive Poisson and flicker noise, achieve maximum sensitivity, and prevent the formation of double peaks in single pulses. Recent advances in the application of LA-ICP-MS for the analysis of metal-binding proteins were discussed by Chen et al.64 The focus was on the identification of metal-binding proteins via the determination of endogenous metals. The main strategies, including in situ analysis of biospecimens and ex situ analysis with gel electrophoresis were reviewed. These approaches allowed researchers to describe how biologically essential and toxic metals are incorporated into biological tissues.
In situ Mg isotope analysis of geological materials by LA-MC-ICP-MS is a powerful tracer technique for tracking geological processes. For this application, medium or high mass resolution is commonly applied to avoid spectral interferences. Improved mass resolution can result in reduced intensity and an aggravated matrix effect, however. Lin et al.65 demonstrated that the main interferences (12C, 14N+ and 48Ca2+) in Mg isotopic analysis can be suppressed, and the matrix effects can be reduced overall by adding water after the LA cell (to create wet plasma conditions) using fs-LA-MC-ICP-SFMS. Using optimised conditions, the authors reported accurate and precise Mg isotopic determination in silicate RMs with various matrices. Over the last 20 years, LA-ICP-MS has become the method of choice to quantify element concentrations in fluid inclusions within geological samples. There is a drawback, however, as ablation of fluid inclusions typically produces short, transient signals that are difficult to representatively sample with sequential analysis inherent to single collector ICP-MS instruments, especially when small inclusions (<20 μm) and/or low-concentration elements (<10 μg g−1) are studied. This issue can be overcome by significantly reducing quadrupole settling times, allowing faster cycling through a given element list at constant duty cycle and therefore, better temporal resolution of the signal. Laurent et al.66 presented results from previously characterised fluid inclusion assemblages performed with a “fast-scanning” quadrupole ICP-MS, i.e. characterised by extremely short settling times (default 0.2 ms and down to 0.065 ms). This capability, which does not impede instrument performance, improved the results of LA-ICP-MS fluid inclusion analyses in several ways. It made possible accurate and reproducible quantification of element concentrations with fluid inclusions of <10 μm and LODs in the same range obtained using a conventional quadrupole. Secondly, it removed the need to compromise between full chemical characterisation of the fluids and accurate determination of key elements, because short settling times offered the possibility to rapidly scan through large element numbers such as the scan of 52 elements in 52 ms at 80% duty cycle reported in their work.
The acquisition speed in LA-ICP-MS element imaging is highly dependent on the laser aerosol transport system. The faster the aerosol washout, the faster the acquisition can be carried out. A modified ablation cell based on a tube cell design was proposed for LA-ICP-TOFMS element imaging to provide shorter signal durations.67,68 The modified parallel flow ablation cell included a recess in the cover to improve the gas flow pattern at the ablation site. This achieved signal durations of 0.25 ms (FW0.1M) for 44Ca and 0.29 ms (FW0.1M) for 238U on NIST SRM 610 using an LA repetition rate of 100 Hz. The shortest signal duration was achieved using an inner cell-to-sample surface distance of 700 μm, which is several times larger than with previous fast washout ablation cells. A shift in arrival time from the ablation to TOFMS extraction was observed when comparing light to heavy ions. An H2 collision and reaction cell was not able to correct for this shift and so slightly extended signal durations were observed. The robustness of the washout makes this a promising ablation cell for fast aerosol transport and quasi-simultaneous detection of ions using an ICP-TOFMS and element imaging using a ≥ 1000 Hz laser ablation rate while maintaining the possibility for pulse-to-pulse signal separation.
An infrared laser (808 nm) coupled with a DBD was used for the OES determination of S and Cl in organic compounds.69 By using a continuous wave IR laser with an output power of 1–2 W, volatilisation of analytes from condensed surfaces was achieved. The analytes were then excited and atomised in the DBD plasma at atmospheric pressure. Direct analysis of S- and Cl-containing organic compounds, in tablets, resulted in a dynamic range of 0.5–20% with linearities (R2) above 0.93 and LODs in the μg g−1 range. The precision was determined by measuring inter-day and intra-day reproducibilities, leading to RSDs ranging from 4.6% to 15.0%. The feasibility of this LA-DBD-OES system was demonstrated through analysis of commercial pharmaceutical tablets of sulfadiazine and chloramphenicol with results consistent with the indicated concentrations and verified by HPLC. The system shows promise for the online analysis of solids and pharmaceutical tablets.
The analytical figures of merit of a low-dispersion aerosol transport system for high-throughput bulk and spatially resolved analysis via LA-ICP-MS were reported by Van Acker et al.70 The system was designed to maximise the collection of aerosol particles generated during the ablation process. This ensured a high transport efficiency and minimised aerosol dispersion during transport of entrained particles through the transport tubing – which included mixing of the He carrier gas flow with Ar make-up gas flow in a co-axial mixing bulb, then on-axis introduction into the ICP torch injector. The compression of the aerosol particles in an optimised space and time window resulted in an increase of the mass flux into the ICP, accompanied by an enhancement in sensitivity and throughput. A low-dispersion ablation cell and a 1 kHz ns laser were used. A linear response of the integrated 238U+ signal intensity was observed upon ablation of a NIST SRM 610 glass as a function of the laser repetition rate up to 1 kHz. Single pulse responses with a full peak width at 50%, 10% and 1% of the maximum peak height of only 0.3 ± 0.1 (FWHM), 0.5 ± 0.1 (FW0.1M) and 1.2 ± 0.1 ms (FW0.01M) can be achieved. These peak profiles were similar to profiles generated by sp-ICP-MS for individual metallic NPs and microplastics.
Significant improvements in the speed, sensitivity and spatial resolution of elemental mapping by LA-ICP-MS have been reported in recent years. However, certain elements show a significantly longer single pulse response duration due to their specific physicochemical properties. Van Helden et al.71 investigated Hg and Se, as examples of elements with long responses, in biological tissues. The effects of the type of ablation cell, mixing bulb and transfer line, on the shape and duration of the single pulse response profiles were evaluated. By optimizing the instrumental setup, a full peak width at 10% of the maximum peak height (FW0.1M) of 50 ± 2 and 61 ± 4 ms were obtained for 202Hg+ and 77Se+, respectively, when using a laser beam of 20 μm diameter. This constitutes a >5-fold improvement in peak duration compared to a standard setup. These peak durations were still considerably longer than those obtained for, e.g., 65Cu+ (7 ± 1 ms), however. When selecting the instrument settings and data acquisition conditions for elemental mapping of Hg and/or Se based on the peak profiles obtained, for e.g.65Cu+, the pixel acquisition rate attainable was overestimated for 202Hg+ and 77Se+ and significant aerosol carryover or pixel crosstalk occurred, resulting in smearing effects in the elemental maps. With the optimum setup and adequate selection of all settings, multi-elemental mapping of Hg and Se was performed for a mushroom tissue section, demonstrating a significant improvement in pixel acquisition rate (up to 20 pixels per s).
Nanoparticle (NP) analysis performed by LA-sp-ICP-MS allows information to be obtained regarding NPs in biological tissues. This information includes NP localising, sizing and counting, and differentiating between the elemental NP signal and its ionic form. In these analyses, a low laser fluence (<1 J cm−2) has been used because it was observed that higher laser fluences can induce NP degradation. Recent work by Metarapi et al.72 focussed on the degradation of AuNPs during LA. The authors provided guidelines for the selection of the optimal laser fluence associated with NP analysis in gelatine as a matrix which mimics biological tissues. AuNPs of known size and with a narrow size distribution (CV < 5%) were used to monitor the measured NP size with changing laser fluence. Optical profilometry was also used to measure the amount of material ablated as a function of the laser fluence and provided additional information on NP degradation.
The analysis of complex nano-fraction particle samples has traditionally been carried out using sp-ICP-MS with nebulised samples. Holbrook et al.73 described a method for the direct analysis of multi-element particles in sediment samples using LA-sp-ICP-TOFMS. A comparison was made between nebuliser and LA sample introduction using standard Ag–Au core–shell particles and environmental samples obtained via cloud point extraction procedures from a road runoff sedimentation basin. Using the results of the method comparison, metrics were used to test the applicability of LA-sp-ICP-TOFMS for analysis of an unextracted sediment. Two main groups of signals were identified: overly abundant signals which cannot be used for sp analysis at the chosen measurement parameters; and highly abundant signals that, when compared with the two previous methods, produced comparable results for elemental ratios and sp fingerprinting. Lastly, low abundant well-defined elements such as the PGEs were found to be ideally suited for measurement from unextracted sediment. The LA-sp-ICP-TOFMS method was proposed for the analysis of multi-element particles for groups 2 and 3 in sediment samples with minimal sample preparation and capable of source allocation for the reported element groups.
The effect of adding H2 to the carrier gas for the determination of F in solid samples in ETV-ICP-OES was investigated by Maung and Beauchemin.75 By incorporating H2 as the reaction gas through the ETV graphite furnace, H2 was introduced into the Ar plasma through the central channel of the ICP. The optimal flow rate for H2 was determined to be 3 mL min−1, with a pyrolysis temperature of 200 °C and a vaporisation temperature of 2200 °C for the ETV program. Two SRMs, NIST 8432 Corn Starch and 8437 Hard Spring Wheat, were used to observe the changes in sensitivity and LOD for F, following the addition of H2. These materials were also used to verify the accuracy of the method using external calibration, with internal standardisation using the Ar 404.442 nm line to compensate for sample loading effects on the plasma. Compared to ETV-ICP-OES analysis without the addition of H2, LODs were improved between 17- and 140-fold, depending on the F emission line used. Similar or improved LODs were obtained for Cd and Hg in NIST 8437 under the conditions optimised for the determination of F compared to those previously obtained with methods optimised for multi-elemental detection (excluding F).
3 Instrumentation, fundamentals and chemometrics
3.1 Instrumentation
Miniaturisation of analytical instrumentation has attracted much research interest over the years, to reduce cost, increase portability or for in situ field analysis. Advances in microelectronics, battery technology and nano-engineering have meant that some of the practical problems have now been overcome. However, one of the most intractable problems still remains the interface between sample and atomisation/excitation cell such that a representative analytical signal can be obtained, either from an inhomogenous sample, or a homogeneous sample that has a matrix which will cause an interference. Zhang et al.76 reviewed (131 references) advances in miniaturised OES systems using microplasmas as excitation sources, with this problem in mind. They discussed current examples of micro-plasmas, including the DBD, point discharge (PD), LEP, various types of atmospheric pressure GD, microstrip plasmas and a microfabricated ICP. Sampling methods discussed were direct solution sampling, VG, ETV and LA. Sections on microfluidic technology and speciation analysis were also included. Their conclusions succinctly sum up the challenges still to be addressed, with each type of microplasma having advantages and disadvantages. For example, direct solution sampling using a flowing electrode AP-GD is a good choice for a field based instrument because it requires no special gases or sample introduction system, but suffers from matrix effects. Conversely, microplasmas (e.g., DBD, PD) which require some form of sample introduction (e.g., ETV, CVG) benefit from the ability to perform matrix separation and preconcentration, but at the cost of added complexity. They conclude by mentioning that the way forward may be achieved by integrating several aspects of the process such as in situ trapping and on-line sample digestion using a DBD, which is also used as the atomisation/excitation source.Yang et al.80 developed a miniature, 3D-printed CVG-PD-OES system for field analysis. The paper included photographs and a video of the system in operation and it is notable that, while the 3D-printed analytical unit itself is quite compact, it is accompanied by a battery and reagent containers which form most of the physical bulk. Nevertheless, this prototype system, which also had a bespoke HV generator to power the PD, was used in the field for Hg speciation by using CVG and PVG to selectively vaporise and determine MeHg, inorganic Hg, SeIV, and SeVI with LODs of 0.1, 0.1, 5.2, and 3.5 μg L−1, respectively. Deng et al.81 used a tungsten coil for sample introduction into a PD microplasma for the determination of Ag, As, Bi, Cd, Cu, In, Pb, Sb and Zn, with LODs of 0.6, 45, 40, 0.08, 15, 8, 8, 41 and 5 mg L−1, respectively. A cross-point, microplasma discharge was developed by He et al.82 using four tungsten electrodes. The four-electrode design resulted in a larger discharge compared to a single-point discharge, thus increasing the chance that most of the HG analyte vapour entered the discharge region and became excited. LODs for As, Hg and Pb were 2.4, 0.15 and 1.9 μg L−1, respectively, with a 3× to 4× improvement in sensitivity compared to the single point discharge.
Sample introduction methods such as VG or ETV are useful because they separate the analyte from the matrix, improve sensitivity and allow sample introduction into miniature plasmas – but are limited to volatile analytes or those which can be generated as volatile species. In contrast, Yang et al.83 developed a nebulised discharge plasma operated at atmospheric pressure. The AC plasma was generated in the 2 mm gap between two tungsten needle electrodes and the tip of a pneumatic nebuliser aspirated with Ar carrier gas. The effect of applied voltage, gas flow rate, liquid pH, and organic agents on the emission intensity of Cu I was investigated. Optimal conditions were an HNO3 solution with pH = 3.59, 5.6 kV applied voltage, 1.68 L min−1 Ar flow rate, and 2% (v/v) ethanol addition, to achieve an LOD for Cu of 0.083 mg L−1. Trot and Tvib, calculated by using N2 spectra, were ∼2190 K and ∼3165 K, respectively.
Direct analysis does away with the problem of sample introduction but means that the matrix is present. A LEP-OES system with axial viewing was reported by Huang et al.84 which resulted in higher sensitivity for Zn, Cd, Pb, Ca, and K compared with radial viewing. Pulsed voltage operation was superior to continuous discharge with LODs of 0.24, 0.051 and 0.85 μg L−1 for Zn, Cd, and Pb, respectively.
Soft ionisation methods have been developed to obtain structural information or circumvent specific problems. White et al.85 used nanospray induced chemical ionisation in the afterglow of an ICP to determine F in liquid samples. The high ionisation energy of F means that it has poor sensitivity, with respect to F+ detection, in conventional ICP-MS. One solution is to promote formation of BaF+, but this is also inefficient. The authors used a new method whereby the sample solution, containing a organofluorine, was introduced by nanospray into the ICP afterglow, resulting in formation of HF. Subsequent reactions with acetates of Na and Ba in the nanospray electrolyte solution resulted in Na2F+ and BaF+ ions. These ions could be determined with sensitivity 100× higher than conventional ICP-MS, resulting in an LOD of ∼11 ng mL−1 for F using the BaF+ ion for detection. In addition, because detection was at a much high m/z, there were fewer isobaric interferences. A hectowatt MPT was used as a soft ionisation source by Jiang et al.86 to determine metal ions in lake water. Samples were conventionally nebulised and then desolvated using a membrane separator and H2SO4 trap. Ionisation of Cu and Zn in solution resulted in the formation of [M(NO3)x(H2O)y(OH)z]+ complex ions in the MPT. These ions were then subject to MS4 multi-stage fragmentation using CID to obtain the ions 63Cu+ and 66Zn+ which yielded LODs of 0.23 and 1.1 μg L−1, respectively.
Ikeda87 used a semiconductor microwave generator to deliver sub-ms bursts of microwave energy to atmospheric LIBS and SIBS plasmas. This allowed better control of the plasmas which could be tuned by varying the microwave power and pulse duration. The system was developed primarily for industrial monitoring applications, e.g. steel, foods, soil, minerals, drugs and radioisotopes and was suitable for both molecular analysis of gases, and atomic spectroscopy of metal powders without pellet compression.
3.2 Fundamentals
:Sr ratios in radioactive samples, with a mean error of 3.86%. Thus, calibration-free analysis could be achieved. Note, also, that they published a subsequent correction to one of the equations used.89 Naoi et al.90 determined the transition probability of the previously unreported Er II 393.863 nm line, by applying its intensity to the linear function obtained in the Boltzmann plot, to be 1.2(2) × 107 s−1. Irvine et al.91 evaluated previously unreported transition probabilities for five Eu I and three Eu II lines, ranging from 0.172 to 7.38 × 107 s−1 and from 1.56 to 6.75 × 107 s−1 respectively.
The photoionisation (PI) cross section of Yb was determined by Shafique et al.92 They used two lasers to excite Yb atoms formed in a vapour. A 722.6 nm laser was used to excite atoms via the 6s2 1S0 → 6s5d 1D2 two-photon transition from the ground state. This excited state population was then promoted to the ionisation threshold using a 439.2 nm ionising laser and above the ionisation threshold at 435 nm, 355 nm, and 266 nm laser wavelengths. The PI cross-sections at each of these wavelengths were then determined using the saturation technique. At the first ionisation threshold, the PI cross-section from the 6s5d 1D2 excited state was determined to be 40.5 ± 6.2 Mb. At higher energies in the continuum, the PI cross-section decreased up to 6.4 Mb at 266 nm, corresponding to 1.84 eV excess energy. Saakyan93 also the determined PI cross-section of 7Li, by using three commercially available low-power UV-C LEDs to measure PI-induced losses in a magneto-optical trap. The PI cross-sections from the 2P3/2 state at 259, 280, 307 nm were determined to be 5.4 ± 0.4, 8.1 ± 0.8 and 9.5 ± 1.1 Mb respectively. The authors state that the results were in agreement with earlier reports and consistent with theoretical data.
3.2.2.1 Plasmas. The excitation mechanisms for Ni I and Ni II in an Ar GD were studied by Weiss.94 It was not possible to model the charge transfer behaviour of Ni II lines using an experimental Boltzmann plot because of the lack of transition probability data. However, some insight was gained from a transition rate (TR) diagram (used to identify selective excitation processes by a peak at the resonance energy in the depopulation rate diagram) because it could be constructed based solely on experimental spectral data.
A collisional radiative (CR) model for a neutral and singly ionised Ar plasma was developed by Han et al.95 The plasma in question was a 15 kW applied-field magnetoplasmadynamic thruster so it is debatable how relevant it is to analytical atomic spectrometry. They used 31 Ar I and 95 Ar II levels to build the model, taking into account radiative decay, electron impact ionisation, excitation and de-excitation, fast ion and atom impact excitation, and three body recombination. Measurements of Te, ne, nAr were made using OES and validated using Langmuir probe measurements. An optical immersion probe was used to determine that the central region of the plasma plume had high ne but low Te, and nAr conformed to the thermal diffusion distribution. The authors concluded that electrons were extracted from the cathode with low Te and heated as they are transferred to the anode in the plume region.
Xie et al.96numerically modelled the gas dynamics in a an ICP generated with a Fassel-type torch. A high-speed camera was used to image the gas pulsation, which was found to be insensitive to: sample type; sample gas flow rate; and auxiliary flow rate, but increased with the r.f. power and coolant gas flow rate. The authors concluded that this supported the theory that audio frequency noise was due to the turbulent entrainment of ambient air into the outer region of the plasma. This could be accurately predicted based on the simulated gas profile at the torch outlet, in the absence of an MS interface. When an interface was present pulsation frequency decreased with sampling depth, attributed to a coupling effect of vortex dissipation in lateral and longitudinal directions. The use of a torch bonnet or extension was observed to move air entrainment away from the analytical zone, reducing gas pulsation and thus noise.
Vonderach and Gunther97 studied the droplet throughput in a downward-pointing ICP-TOFMS instrument. Monodisperse droplets of a multielement solution were generated using a piezo-actuated autodrop pipette. A triple-pulse mode was used to produce 80 μm diameter droplets from a 70 μm capillary. The droplets then passed through a downward-pointing, heated, glass sample introduction tube which had three gas inlets to allow the addition of varying flows of He and Ar make-up gas. The droplets were used to transport individual mouse lung cells, with detection aided by addition of a Cs tracer, Ir nucleus marker, surface markers and P content. Time-resolved TOFMS spectra were collected at a spectral averaging rate of 5.56 kHz (180 μs per spectrum) for 184 ms (1024 spectra), thus individual cells could be detected and counted using the ICP-TOFMS. A random selection of droplets were inspected optically and found to contain intact cells, providing confidence that the system could be used for cell cytometry.
Wiltsche et al.98 investigated the effect of introducing Ar to a 1.5 kW, microwave-sustained, nitrogen ICP (MICAP). Replacing N2 with Ar in the nebuliser gas flow resulted in signal suppression of up to 70% for 65 emission lines of 29 elements, whereas negligible effects were seen when the intermediate gas flow was replaced with Ar. Interestingly, when Ar was used in both the intermediate and nebuliser gas flows, with N2 in the outer gas flow, and enhancement of up to 60% was observed for high excitation energy atom and ion lines. Various combinations of Ar and N2 in the nebuliser, intermediate and outer gas flows were observed to change the density and shape of the plasma, which probably partially accounted for the change in emission intensities – e.g. a more diffuse analyte channel resulting in lower emission intensity. High-speed imaging revealed changes in the plasma structure as the gas composition was varied. When Ar in the outer gas flow exceeded ∼71%, plasma filamentation to form a thin-walled ring discharge occurred, under which conditions the MICAP could not be operated for long.
Shi et al.99 studied the spatial emission profile from a μDBD flowing directly onto a sample surface. Radially resolved maps of N2+, N2, He2, He I and O I emission, and Trot, Te and ne, were used to gain insights into excitation mechanisms. The authors proposed that, in the upstream region of the He μDBD, N2+ was produced by Penning ionisation with He metastables, but downstream by charge transfer with He2+, followed by electron recombination to yield excited N2. An increase in Te and ne close to the surface of conducting samples was observed, which promoted electron impact excitation. This was due to a faster ionisation wave during the start of the positive half-cycle followed by a diffuse glow-like discharge.
Temperature measurements are useful because they provide insights into the thermal and excitation conditions in the plasma. It should be noted that most of the theoretical underpinning for such measurements assumes the condition of thermal equilibrium, though this condition is rarely met. Volker and Gornushkin100 addressed the correct use of units in the Boltzmann plot method. It seems obvious that consistent use of units is necessary in order for correct application of the theory, but anyone familiar with the literature will know that consistency is often not encountered. In particular, they note that the logarithmic nature of the plot requires that dimensionless quantities are used as an argument, achieved by normalising the equation for the integral line intensity by the unit intensity in the chosen system of units. This requires calibration of the spectrometer and detector so that the emission signal can be measured in physical units. In addition, the abscissa on a Boltzmann plot has units of energy, but the logarithmic function gives a dimensionless unit so the ordinate is also dimensionless.
Tibere-Inglesse et al.101measured Trot, using OES and Raman spectroscopy, in a non-equilibrium, non-analytical, 4 MHz, premixed Ar–N2 ICP. They measured emission spectra of the first (B3Πg → A3Σu+) and second (C3Πu → B3Πg) positive systems of N2, In order to correct the Boltzmann method for the non-equilibrium conditions, vibrational overpopulation factors were determined based on the measured absolute density of vibrational levels of N2(B), the spectrum was normalised at the (2, 0) bandhead and the residual between the experimental and SPECAIR (a spectral analysis programme) spectra was calculated. This procedure was repeated for various Trot measurement until the smallest residual was obtained. A similar method was applied to the first negative system of N2(C), however, the rotational levels of N2+ (B2Σu+ → X2Σg+) were found to follow a Boltzmann distribution. The Tgas was measured by Raman spectroscopy to be ∼3200 K and was used for comparison. For the N2(B) and N2(C) systems Trot was in good agreement, but found to be significantly higher for the N2+(B) system. The authors postulated that this was a consequence of N2+(B) being formed by a charge exchange reaction between N2(A3Σu+) and N+, and calculated a rate constant of k3 ≃ 4.8–7.5 × 10−10 cm3 s−1 for this reaction.
Garcia et al.102 used OES to determine Tgas of a He MIP used for medical applications. They used a method based on collisional broadening of the He I 667.82 nm and He I 728.13 nm emission lines. They indicated that this was particularly suitable for low gas temperature plasmas (<1000 K) and He plasmas with ne < 1014 cm−3, because Stark broadening did not significantly contribute to the total collisional broadening. They also used this method to estimate the amount of air entrainment for three different designs, by comparing Tgas with Trot determined using the N2(C–B) rovibrational band (2nd positive system).
Gajo et al.103determined Te in DC and pulsed, wall-stabilised arc plasmas, using the ratio of ne values obtained from hydrogen Hα and Hβ lines, based on Griem's theory. The spectral lines were corrected for the spectral response of the optical system and detectors. A Voigt function was fitted to the Hα line profiles because the Gaussian part of the line profile was 0.44 Å, which corresponded to the convolution of the instrumental width and the Doppler width for 10000 K. Values of ne were between (1.24 and 1.37) × 1016 cm−3 for DC arc plasmas and (1.69·and 1.80) × 1017 cm−3 for pulsed arc plasmas. The Te values ranged from 9100 to 13600 K. The ne and Te of a low pressure, Ar CCP was determined by Wu et al.104 using OES and a CR model. They found that in triple-frequency (2, 13.56, and 27.12 MHz) plasmas ne was more independent of Te, compared with dual-frequency (2 and 27.12 MHz) plasmas, as the intermediate frequency power increased. 2-D imaging with a CCD camera revealed that the axial and radial distributions of ne were more uniform at lower power, but Te was uniform under all conditions. A transition from α to γ mode was observed at 13.56 MHz with increasing RF power, accompanied by an increase in ne and a decrease in Te.
Xiao et al.105 developed a simple method to determine ne in an air MIP. Rather than use a spectral method, they utilised the power reflection ratios to build a 3-D Gaussian simulation of ne, then adjusted this ne value until the calculated power reflection ratio equalled the measured results. The results were compared with values calculated from Stark broadening of the Hβ spectral line at 486.1 nm, and agreement within 8% was obtained for experiments using different flow rates and different N2:O2 gas ratios.
Nedić et al.106estimated the electric field strength in a Grimm-type GD. They made spectral measurements of six Ne I and four Ar I lines in a broad range of discharge conditions (pressure, voltage, current) and cathode materials. A linear correlation (1.083 in Ne and 1.129 in Ar) was observed between shifts in line broadening features recorded end-on (Δλe) and side-on (Δλs) at the position of the maximum electric field in the cathode sheath. The correlation coefficients could be used to estimate the maximum electric field strength and thickness of the cathode sheath.
3.2.2.2 Graphite furnaces. XPS was used to characterise the solid-state surface phases formed on pyrolytic graphite platforms during the atomisation of V in ETAAS by Ruiz et al.107 V8C7 was confirmed as the carbide phase formed at the ashing temperature (1400 °C) and above. The thermodynamically stable phase under these conditions (V2C) was not observed, suggesting a kinetic control of the carbiding process. Interestingly, several oxides detected by XPS were suspected to be artifacts of the XPS technique itself. In particular, VO, a V2+ species produced at the low temperature of 120 °C and probably originating from carbo-reduction due to irradiation with X-rays. V2O5 was also observed after ashing at 1400 °C, which could be due to reoxidation of the lower V oxides during transfer from the ETV-AAS equipment to the UHV system for XPS.
3.3 Chemometrics
Chemometric approaches for interference correction are useful because they only require data treatment rather than additional hardware. Popova et al.108 presented a non-linear theory to account for the effect of interfering elements on analytical determinations using OES. The authors start from the proposition that most correction methods are linear in nature, which in fact is often not the case. Development of their theory led them to conclude that third-element effects are additive and are caused by changes in plasma conditions and spectral interferences. Hence it was possible to represent the relationship between the concentration of the analyte and the intensity of its analytical line as a polynomial with some degree (N0) that coincides with the degree of the corresponding polynomial in the absence of third-element effects. The coefficients of this polynomial depended on the intensities of the analytical lines, Ik, of the influencing elements and, at Ik = 0, were equal to the corresponding coefficients in the absence of third-element effects.The same group of researchers109 also proposed a mathematical correction method for blooming of high intensity spectral lines with CCD detectors. The method seemed to involve using intensity measurements at different points in the data acquisition time frame, when there was no blooming, to correct the intensity of the spectral limes during the time frame when blooming occurred. The algorithm was tested for the determination of Al in iron based alloys using spark source-OES at concentrations of several thousandths of a percent. At lower concentrations blooming was not observed.
Correction of M2+ interferences on As and Se in ICP-MS was investigated by Smith et al.110 They performed a multivariate analysis of interferences caused by a mixture of 19 REEs over a two month period and with two different tuning conditions. PCA indicated that M2+ clustered better than M+ over the factor space resulting in better 95th percentile confidence bounds (n = 704) when used for dual M2+ internal standards for determination of As (0.3 ppb) and Se (5.4 ppb), compared with M+ internal standards (As = 0.6 ppb, Se = 12.0 ppb, n = 1056). Hierarchical modelling indicated that using M2+ internal standards also reduced variability caused by cone changes and shifts in the M2+:M+ ratio, in a 250 ppm Na, Ca and Mg matrix.
Nakadi et al.111 used a time-absorbance profile method to correct spectral interferences in GF-CSAAS, by subtracting the normalised spectrum of the interfering species from that of the atomic line. They demonstrated the efficacy of the approach by determination of Pb and Cu in biological CRMS. In order for the method to work effectively it was necessary to determine the maximum interference level and ensure that its time-absorption profile did not completely overlap the analytical signal. Hence, some adjustment of the GF temperature program may be necessary. Also, the model only worked for a single interfering species. They also noted that the method was developed for interferences caused by diatomic molecular species, but the correction can still be used for atomic line interferences provided that atomic lines of the interfering element fall within the spectral window.
AFS methods can suffer from scattering interference. Wang et al.112 addressed this problem by using a real-time suppression method. They analysed AFS line spectra for As using a dispersive UV digital micromirror spectrometer to identify the source of scattering interference and the effect of external factors. The ratios of non-fluorescence to fluorescence lines for As were determined and their stability under different conditions studied. These ratios were used to correct for scattering effects resulting in >99% accuracy for the analysis of water CRMs.
Finally, Grebneva-Balyuk et al.113 proposed a simple method for determination of LOQ using an equation which included only the analytical signal for the blank, and analyte signal and concentration of a working standard. They determined LOQs for a variety of elements using ICP-MS and ICP-OES and tabulated them in detail. However, LOQ is often a subjective measurement which is specific to a particular standard operating procedure, so this method may not be universally applicable.
4 Laser-based atomic spectrometry
Atomic spectrometry techniques where the laser is used as either an intense energy source or a source of precise wavelength (e.g. LIBS or LIF) are considered here. However, studies related to LA-ICP-MS/OES, and to the use of lasers for fundamental studies of the properties of atoms or for thin film deposition are not reviewed.4.1 Laser induced breakdown spectroscopy (LIBS)
This section particularly describes the latest instrumental developments and fundamental studies related to LIBS. Reviews that cover detailed applications can be found elsewhere; for instance: Zhang et al.114 reviewed the recent advances in the use of LIBS for in situ and online detection of pollutants under atmospheric environments; Liu et al.115 reviewed novel approaches of LIBS for coal analysis; Khan et al.116 discussed the pros and cons of different LIBS methods to improve the detection of trace elements; and Marina-Montes et al.117 described a new protocol for particulate matter characterisation in filters using LIBS. Hu et al.,118 Wang et al.119 and Zhang et al.120 described advances in LIBS quantification methods, in particular related to calibration-free LIBS methods (CF-LIBS), which allows quantitative analysis without the use of standards. In these studies, the authors included a description of the basic theory, novel modifications/variations of the quantification methods, applications in a variety of fields and a discussion about the existing drawbacks.The effect of laser polarisation on the molecular emission bands from fs-LIBS was investigated by Chen et al.,123 who observed that higher AlO and CN molecular emission was obtained using circular polarisation and increasing laser energy. These studies were supported with additional Tvib calculations at different waveplate angles. LIBS molecular emissions are also considered to play an important role in the differentiation of organic compounds of interest in astrobiology. In this context, Delgado et al.124 investigated the influence of a Martian atmosphere on the recombination mechanism in laser induced plasmas of organic compounds. This study highlighted how the presence of minor amounts of nitrogen in the low-pressure Mars environment affected LIBS plasma chemistry and the formation pathways leading to molecular species. Additionally, statistical analysis suggested that molecular emission signals were the major contributors to achieve high LIBS sorting performance and feasibility of discriminating closely related organic compounds in the Martian atmosphere. Vogt et al.125 performed a spatiotemporal characterisation of a LIP plume in simulated Martian conditions. From the investigation of a calcium sulphate sample, they found unique spatial distributions, which suggested a strong influence of the outgoing shock wave at selected atmospheric conditions. After plasma formation, the plasma centre was found to rapidly become cooler and more rarefied than the outer plasma regions. The authors claimed that molecular emissions originated in the cold plasma centre.
LIBS molecular emission was also investigated to determine N in water by Ma et al.,126 employing Ar or He gas environments to improve sensitivity and LODs. In particular, N was indirectly quantified by making use of the emission from the violet system B2Σ+ − X2Σ+ (Sequence Δv = 0) of CN spectra. The authors achieved an LOQ of 1.98 μg mL−1, in an Ar atmosphere, that was good enough to meet the surface water environmental quality standards of China for N.
The detection of REEs by LIBS molecular emission is also an interesting hot topic. Gaft et al.127 observed that oxides of REEs Ce, Pr, Nd, Sm, Eu, Gd, Tb, Dy, Ho, Er, Tm, Y, and Lu were appropriate for LIBS molecular analysis, whereas previously only La, Y and Sc were considered to form analytically useful molecules. Moreover, the authors claimed that these molecular emissions opened an opportunity to apply LAMIS and molecular LIF-LIBS in combination for isotopic analysis of REEs.
The process of deposition of Ti oxides from a Ti LIP to the Ti target surface was investigated by Gornushkin et al.,128 by making use of a novel chemical-hydrodynamic model. The biocompatibility and durability of Ti-based implants are improved by texturing and coating them by LA across the implant surfaces. The developed model included plasma chemical reactions, formation of condensed species inside the plasma plume, and deposition and accumulation of these species on the ablation surface. Moreover, the model was used to visualise how condensed titanium oxides, mostly TiO2, were formed in a peripheral plasma zone and gradually adhered to the surface during the plasma plume evolution. The authors claimed that the model might be of great interest to predict and optimise these surface modification treatments.
The effects of oceanic pressure and temperature on in situ LIBS analysis at different depths during sea trials were investigated by Li et al.129 Strong variations of the spectral features were observed at increasing depth, including a linear increase of line broadening. Results indicated that emission signals from depth below the thermocline were mainly influenced by pressure. Moreover, a novel model that exponentially fitted the relationship between the LIBS signals and the oceanic depth and temperature was developed. The same research group investigated the temporal emission characteristics of underwater LIBS at different pulse duration conditions.130 Results indicated that a long laser pulse duration (∼189 ns) produced a compact plasma with stronger emission and longer lifetime compared to short laser pulses (∼ns); in particular, for spectral lines with a high-energy transition level.
T e and ne of LIPs are typically derived from plasma emission spectra, however, these spectra might be affected by self-absorption and plasma inhomogeneity effects. A complementary approach would be to use Thomson scattering (TS) spectra, but this also suffers from spectral interferences and low SNR in low-temperature LIBS plasmas. However, Zakuskin et al.131 developed a reliable and robust method that combined pre-processing TS spectra, using a wavelet denoising method, and a nomograms approach, which consisted of two curves calculated from theoretical TS spectra, for plasma parameter calculations.
Arora et al.132 investigated the detection ns-LIBS emission signals from Al I lines at long delay times (≥200 μs) in an Ar atmosphere, under different conditions of laser power density, distance from the target and pressure. The authors considered that the excitation and subsequent emission from neutral Al species might be the result of energy transfer from metastable species of Ar.
The influence of LA angles on LIBS spectra was investigated by Wu et al.133 This study was related to the application of LIBS under complex working conditions, such as those required for the evaluation of the plasma–wall interaction processes in a Tokamak. The spatiotemporal dynamic evolution of a laser-induced WC/Cu plasma at variable angles was investigated under simulated experimental advanced superconducting Tokamak vacuum conditions.
A temporal analysis of self-reversed Ag I resonant lines, which are thought to originate from the combined effect of self-absorption and plasma inhomogeneity, was performed by Urbina et al.134 using different laser pulse energy conditions and in different atmospheres (e.g., He, Ar, Xe and air). The laser-induced shock wave generated was demonstrated to play a key role in the production of plasma inhomogeneities. In particular, it was observed that plasma expansion in a molecular gas induced strong self-reversal of Ag lines, whereas plasma expansion in an atomic noble gas resulted in negligible self-reversal effects. In another work, self-absorption temporal evolution of alkali and alkaline earth metal elements in soils was investigated by Tang et al.135 In this case, weaker self-absorption effects in air were observed at the early stage of the plasma. The authors suggested that the phenomenon might be caused by plasma expansion and temperature reduction. Analytical error growth caused by self-absorption in continuous flow LIBS was investigated by Taleb et al.136 The error was found to be significantly larger for line shapes dominated by Doppler broadening and for line-centre intensity measurements, while line-integrated intensity measurements of Stark broadened lines allowed accurate results if line width and plasma size were precisely known. The authors considered that these findings could enable straightforward selection of the most appropriate analytical lines.
Hu et al.138 developed a multidimensional-plasma-grating-induced breakdown spectroscopy (MIBS) system to further improve the analytical performance of single plasma-grating-induced breakdown spectroscopy (GIBS). The superior performance of MIBS in detection of trace elements, compared to filament-induced breakdown spectroscopy (FIBS), was due to its higher intensity and electron density inside the plasma waveguide. In particular, the authors generated a 2D plasma grating using the interference of three non-collinear and non-coplanar fs-filaments. MIBS was shown to be greatly affected by the laser polarisation and inter-pulse delay. However, compared to GIBS, signal intensity was 2× greater and the induced plasma lifetime 20% higher.
The coupling of atmospheric pressure GD and cylindrical confinement with LIBS was investigated by Li et al.,139 and found to enhance LIBS signals by up to ∼20× using optimised experimental conditions. The authors developed the device as a miniaturised, low-cost and low-power consumption instrument. Xu et al.140 developed a hybrid method that combined discharge-assisted LIBS with wavelet transform de-noising, for trace elemental analysis in liquid samples (e.g., oil pollutants).
Sun et al.141 developed a system where two laser beams were generated by splitting a single laser beam, then focused and crossed orthogonally in air. This resulted in signal enhancement and stabilisation by confining the plasma volume and controlling its position. Using this experimental set-up improved the spatial resolution of LIBS for gas analysis and enhanced the OE signals from H, N and O by 2-fold with respect to conventional LIBS. Furthermore, it was demonstrated that this technique was moderately tolerant of dust particles present in the environment, thus avoiding the spark occurring outside the detection volume.
A LIBS system was attached by He at al.142 to a unilateral-shift-subtracting confocal microscope, to achieve fast, high spatial resolution, multi-element maps of sample surfaces. This system, called unilateral-shift-subtracting confocal controlled LIBS microscopy (UCCLIBS), had a surface axial focusing resolution of ∼18 nm under a 20× objective lens with a numerical aperture value of 0.4. This was more than two orders of magnitude higher than that of traditional LIBS. The authors claimed that UCCLIBS might provide a new approach for 3D elemental characterisation of minerals and chemical materials.
A LIBS system based on Bessel beam, which is a kind of “non-diffracting beam” with extended linear focus, was developed by Lv et al.143 for the analysis of uneven samples. The authors applied this technique to the analysis of steel tubes, demonstrating that RSDs were reduced, and that higher accuracy and stability was achieved, compared to LIBS based on Gaussian beam.
Buday et al.146 collected simultaneous LIBS spectra and data from several sources, including direct imaging of the plasma, shadowgraphy and sound signals. This data was found to be correlated with the optical emission signals and the correlation coefficients were significantly improved by correcting the data using the energy of laser pulse. In a similar context, Chen et al.147 developed a method, based on the correlation between plasma image information and screening methods (e.g., SNR), to identify effective LIBS spectra from coal particle flow, thus allowing accurate online analysis of these particles. The proposed image auxiliary method application was shown to be feasible and a prospect for improving the quantitative analysis of coal particle flow.
Mapping data, obtained using single-shot LIBS measurements, have intrinsically low SNRs and high dimensionality (high number of spectra acquired). Finotello et al.148 developed a new analysis tool (HyperPCA) to efficiently extract physico-chemical information (e.g., elemental maps) from noisy and large datasets. This tool for hyperspectral images was based on a sparse representation of the data using discrete wavelet transform (DWT) and kernel-based sparse PCA to reduce the impact of noise on the data and to consistently extract the spectroscopic signal. A novel method to achieve elemental mapping of F by means of molecular LIBS was developed by Weiss et al.149 Fluorine containing molecules (CuF or CaF) were generated by recombination of F atoms from the original polymer sample, with the molecule-forming partner stemming from a surface coating that was deposited using spray coating or sputter coating techniques. The authors claimed that sputter-coating of Cu yielded better sensitivity, while spray coating of Ca provided higher spatial resolution.
A novel method for quantitative analysis of alloy steel samples with changing temperature was proposed by Chang et al.150 A conversion model for finding the non-linear relationship of LIBS spectra at varying sample temperatures was constructed using functional data analysis. Spectra measured at any temperature could then be easily converted into their equivalents at room temperature. A calibration model, with spectra at room temperature as the input and elemental concentrations as the output, was established based on least squares support vector machines. The authors claimed that the proposed method provided a feasible and effective way for LIBS analysis of samples at varying temperatures in iron and steel smelting production processes. Lithological recognition and quantitative determination of multiple chemical elements in rock samples was investigated by Chen et al.,151 by combining LIBS and deep convolutional neural network. A 2D convolutional neural network algorithm was employed to provide two different outputs, which could complete classification and regression tasks at the same time. This method was considered to provide reliable LIBS data processing to complete multi-task analysis with high efficiency and good accuracy.
A new method, based on wavelength artificial shift subtraction, was developed to remove the spectral background for underwater detection of LIBS.152 After conducting a subtraction between wavelength shifted- and unshifted spectra, the high baseline of the LIBS signal was properly corrected, especially for weak signals.
4.2 Laser induced fluorescence (LIF)
Ultraviolet laser-induced fluorescence of dense plumes for ultra-trace material analysis was thoroughly reviewed covering all related publications from 2005 to 2021.153 It was highlighted that this technique might be able to induce simultaneous fluorescence from multiple atomic and molecular analytes imbedded in a dense plume (e.g., in a pulse LIP), using a single VUV excitation wavelength.4.3 Laser atomic absorption spectroscopy (LAAS)
A transient Al plasma generated by excimer LA in vacuum was investigated through LAAS to probe the space-time evolution of neutral species.154 Results indicated that the column density of Al atoms decreased temporally, over a plasma lifetime of about 10 μs, and spatially, over distances in the range 0.5 to 8.5 mm from the target surface. Calculations showed the kinetic temperature of Al ground states dropped significantly a few millimeters from the target.5 Isotope analysis
5.1 Reviews
Developments in the field of IRA using MC-ICP-MS have enormously developed our ability to measure the mass dependent isotopic compositions of small amounts of sample material. Thus expectations of achievable precision and accuracy have increased over the last 25 years. Gerber and van Zuilen155 reviewed these developments and summarised most of the basic elements of non-traditional stable isotope determinations. This extended from instrumentation and measurement parameters, through laboratory processing and blanks, to data correction and reduction. Additionally, a section was included that itemised a series of tests to assess accuracy, robustness and precision of the isotopic measurements. A tutorial-style review of isotopic measurement by ICP-MS was completed by Penanes et al.156 This covered two aspects of isotopic measurement: natural variations in the isotopic composition of the elements, and the use and measurement of enriched stable isotopes used as tracers. Guidance and equations for isotope abundance calculations, delta values, mass bias, isobaric/polyatomic interferences, blanks and memory effects were covered in the first part of the review. Further sections then detailed the treatment of uncertainty, analytical optimisation and error propagation. Clearly the different aspects covered in their review were generated by different co-authors which is noticeable through the diversity of the examples used. These ranged from archaeological provenance of metals using Pb isotopes of ores, enriched isotopes for metabolism studies and the environmental mobility of Hg. Overall, the review provided a good coverage of ICP-MS basics alongside interesting case studies.5.2 Radiogenic isotope ratio analysis
Radiogenic Nd isotope measurement using MC-ICP-MS was reconsidered by Feres et al.157 Their study assessed the affects of wet and dry plasma conditions, cone geometry, oxide formation and normalisation protocol in the accuracy and precision of Nd isotopic determinations. It was concluded that if sample Nd concentrations were high then wet plasma and standard cones provided the most stable oxide and fractionation conditions. If concentrations were lower, then desolvation nebulisation combined with standard dry plasma cones produced the best results. At the lowest concentrations, it was recommended that a small N2 addition to the sample line, in conjunction with high sensitivity skimmer cones, produced the lowest oxide formation while maintaining sensitivity.The effects of laser focus on U–Pb isotope measurement by LA-ICP-MS were investigated by Huang et al.158 This revealed that variation in the laser focus by 30 μm shifted 206Pb/238U by between <1% and over 10% depending on the composition of the target mineral. However, little variation in 206Pb/207Pb was identified, which meant that a combination of these effects could result in discordant data on concordia plots. It was estimated that if laser focus was adjusted to within 5 μm of the ablation surface then the offset was reduced to less than the analytical uncertainty.
Luu et al.159 demonstrated the use of a TIMS instrument equipped with a dynamic zoom lens capable of switching the ion beam between the 16 detectors. Sr and Nd isotopes were measured in this mass spectrometer using multidynamic routines. A key feature of the analyses was that peak shape and peak alignment could be adjusted for each line in the multidynamic routine by varying the zoom lens parameters. Results show that SRM NIST987 gave accuracy and long-term reproducibility of 87Sr/86Sr = 0.7102467 ± 0.0000043 (2 sd; n = 38), while the Rennes-Ames Nd standard solution gave 143Nd/144Nd = 0.5119537 ± 0.0000022 (2 sd; n = 31). These values were a factor of two or three times more precise than typical TIMS analyses.
Rösel and Zack160 investigated Rb–Sr dating of minerals using LA-ICP-MS-MS. This mass spectrometric technique utilised reaction cell technology to suppress interferences, but also to generate 86Sr16O+ and 87Sr16O+ by reaction with N2O. Hence 86Sr and 87Sr were measured at m/z 102 and 103, respectively. 85Rb was measured at m/z 85 and 87Rb calculated using the natural Rb isotope abundance ratio. 85Rb16O+ was also measured to check and correct for any Rb reaction products. The method was found to be highly effective for Rb-rich Sr-poor minerals such a micas. A key benefit of the technique was that dating could be achieved on a single spot analysis. A caveat was identified in the need to assume an initial 87Sr/86Sr. However, for high Rb-minerals with 87Sr/86Sr ≫1 and a realistic assumption of the primary geological setting, this value was found to have a minor contribution to the final age.
5.3 New developments
Martin et al.161 developed a new method for direct U/Th dating of small carbonate samples. This utilised a femtosecond LA-ICP-SFMS system with a single collector. To target areas of samples that were unaffected by contamination or U mobility, an isotopic map covering an area of 5–8 mm2 was acquired by surface ablation, with each isotope integrated for 1 s in a 30 μm spot. From these maps, regions of interest were then identified for longer integration ablation analysis which acquired U and Th ratios from between 1 and 2 mg sample. Data for speleothems of known age produced similar dates within the measurement uncertainty. In comparison to solution-derived data from MC-ICP-MS the laser technique was found to have significantly worse uncertainty (±6–22%; 2 sd). However, the data acquisition was extremely rapid without the need for purification and the ability to target the analytical regions of interest, i.e., those areas of the sample where detrital 232Th had not accumulated, was seen as being of great interest. The U–Th–Pb system was also investigated by Rush et al.162 who utilised a new extreme UV LA system in conjunction with a TOFMS spectrometer. The aim of their study was to see if matrix effects from geological specimens were reduced when determining 206Pb/238U and 232Th/238U. Results indicated that Pb–U isotope ratios were determined with a systematically low bias that could be corrected with external calibration using SRM NIST610. However, Th–U ratios did not show any systematic bias. Overall, the system was found to be a good candidate to isotopically analyse geological materials on a scale of <10 μm.Masuda et al.163 developed a technique which effectively used LA as a microsampling tool. This method directed the output from the LA system to pass through a PTFE membrane filter that trapped the sample particulates. Experiments with NIST610 glass demonstrated that sample material collected were sufficient to put through a separation procedure that isolated Nd, Sr and Pb for isotopic analysis with minimal blank contributions. The key advantage of this technique was the high spatial resolution of sampling, such that material could be extracted from crystal zonation, archaeological artifacts or accumulated biological growth.
A new system that separated and measured Mo isotopes was developed by Yobregat et al.164 This used a AG50W-X8 cation resin that eluted using HCl and separated Mo from matrix elements, followed by an anion AG1-X8 column that purified Mo from elements such as Ti, Nb, Ta, W and Ru. Isotopic determination was then made by ionising Mo as oxide species from Re filaments on negative ion TIMS. Corrections were made for oxygen isotopic isobaric interferences during the three-sequence multidynamic routine. For most Mo isotope ratios this method resulted in a long-term reproducibility of ∼±6 ppm, e.g. for SRM NIST3134 94Mo/96Mo = 0.552569 ± 0.000003.
5.4 Geological studies
Advances in measurement capability of K isotopes has meant that they are increasingly used to investigate geological processes. Parendo et al.165 measured K isotopes on samples from the Izu-Bonin volcanic arc using MC-ICP-MS equipped with a collision and reaction cell. Some CaH+ formation was generated in the reaction cell, but the overlap with 41K was estimated by monitoring Ca-doped solutions. Results indicated that fluids released from the subducting ocean crust preferentially mobilise heavy K, resulting in higher δ41K (−0.22‰) in arc-front volcanoes relative to documented mantle source values (−0.44‰). Volcanoes furthest from the subduction zone were noted to have intermediate δ41K (−0.36‰), thus may represent the progressive removal of heavy K as subduction progresses.Sr isotopes were measured using a novel “tribrid” mass spectrometric system by Bevan et al.166 This combined QMS and SFMS with a CRC and MCs, ICP-CRC-QMS-MC-SFMS. NBS SRM987 doped with Rb was used to demonstrate the resolution of Sr from Rb by the collision cell, which is an important facet of in situ analysis where chemical isolation of these elements is not possible. The high ion transmission and multi-collection capabilities were utilised in geochronological applications. This was demonstrated by producing an accurate age for Dartmoor granite using multiple Rb–Sr analyses of a single feldspar crystal within this lithology. Sr isotopes are still preferentially measured by TIMS which is less affected by the Rb and Kr interference present with MC-ICP-MS instruments. Di et al.167 utilised TIMS to measure high-precision μ84Sr/86Sr to understand the volatile depletion history in the early Solar System. Based on comparison of multidynamic and multistatic acquisition routines, TIMS measurements in this study were identified as being susceptible to bias generated by drift in isotopic fractionation during the run. The μ84Sr/86Sr was found to be more effected than the radiogenic 87Sr/86Sr ratio. A linear interpolation was applied to correct the fractionation drift and produced equivalent dynamic and static results. When applied to the NIST SRM 987 it was found that this reference standard has an enriched μ84Sr/86Sr relative to terrestrial samples which have values of −31 ± 8 (2 sd). Liu et al.168 also investigated Sr isotopes and aimed for high-precision determinations by MC-ICP-MS. The study combined Zr external normalisation with an 84Sr–87Sr double spike, which was found to be beneficial in overcoming the mass fractionation effects observed in the isobaric Kr and Rb interferences. Ultimately, the results were found to approach those of double spike TIMS methodology. However, it was noted that the highest internal precision and best reproducibility would still be provided by TIMS analysis, albeit with longer acquisition times.
5.5 Stable isotope ratio studies
Vogl et al.169 presented a re-evaluation of uncertainty for stable IR studies centred on the calculation of the delta parameter. Their rationale is that delta values are defined by nonlinear relationships which means that linear approximations are not sufficiently exact. They provide exact equations for converting delta values of the same isotope ratio, but measured on different scales. Measurement uncertainty between studies and laboratories was also considered as a contributor to these calculations and a new method for its calculation, which involved covariation between isotope ratios, was presented.An et al.170 examined K IR measurements by HR-MC-ICP-MS. Their methodology resolved 41K as a “shoulder” on the low mass side of the combined 41K+ + 40Ar1H+ peak. This position enabled a static measurement of 41K+ together with the 39K+ peak, following adjustment for the K+ + ArH+ tail. Further rigorous assessment of this system was made by investigating the effects of concentration mismatch between sample and standard, acid molarity differences and matrix effects. It was concluded that the concentration mismatch had the most significant effect on accuracy, but δ41K could be reproduced to within ±0.1‰ in consistent conditions. A similar instrument, but fitted with a collision cell, was used by Li et al.171 to evaluate K isotopes. The study measured international seawater standard IAPSO, alongside Mn-nodule (NOD-P-1) and iron formation standards to test the accuracy of their system. Long-term reproducibility was estimated to be ±0.04‰ for δ41K. Wang et al.172 developed a K isotope measurement strategy using MC-ICP-MS specifically designed for measuring the low K content of marine carbonate minerals. They used a dual-column purification the first AG50W-X8, to separate Na, Al, Mg and Ca from K; the second to minimise Na and Rb. The method achieved precise results in samples of 100 to150 mg of carbonate, which was highlighted as a great benefit for future palaeoceanographic research. Zheng et al.173 investigated the bias effects of the Sapphire collision cell on concentration mismatch between sample and standards. These authors proposed that the term concentration mismatch should perhaps more correctly be ion intensity mismatch as this includes the effects of matrix elements and instrument drift in the signal bias. A correction method to eliminate the mismatch bias was presented in the study, resulting in an intermediate precision of ±0.05‰ (2 sd) for δ41K. Gu and Sun174 also investigated K isotopes using MC-ICP-MS without using collision cell technology in low resolution mode, with cool plasma conditions to suppress the formation of Ar hydrides. Residual isobaric ArH+ was accounted for by bracketing samples with blank acid measurements. Final precision was better than ±0.08‰ for δ41K. As in the study of An et al.,170 the effect of acid molarity and concentration mismatch was considered. However in the Gu and Sun study, it was concluded that the acid molarity had the greatest effect on K isotope systematics.
Nd isotope ratios are generally measured to obtain information on the radiogenic 143Nd/144Nd ratio. However, a study by Bai et al.175 combined this measurement with a determination of the stable Nd isotope composition using MC-ICP-MS. Isolation of Nd was completed using the Eichrom TODGA resin and was found to give near perfect separation of Nd from Ce, Pr and Sm (Ce/Nd < 0.003). This resulted in δ142Nd/144Nd of ±0.027‰ (2 sd) with similar precision for the other stable Nd ratios, in conjunction with 143Nd/144Nd better than 0.00001‰ (2 sd). The study also reported combined radiogenic and stable Nd isotope ratios for a number of geological materials, and highlighted a range in δ142Nd/144Nd from −0.24‰ to +0.23‰.
B isotopes were again a subject of interest, due to their use as a proxy for ocean pH in palaeoceanographic reconstructions. Buisson et al.176 strove to achieve high precision δ11B on nanogram levels of boron. Their technique isolated B from a carbonate sample using a microdistillation system and then transferred the analyte to a MC-ICP-MS by coupling it to a demountable direct injection high efficiency nebuliser (d-DIHEN). An X skimmer, Jet cones and two 1013 Ω amplifiers enhanced sensitivity and SNR resulting in δ11B external repeatability of ±0.24‰ using <1.2 ng B. Chanakya and Misra.177 used a QQQ-ICP-MS instrument to determine δ11B in solutions containing HF and HNO3 used to provide high sensitivity and rapid washout of the notoriously sticky boron. These workers also used a microdistillation system to concentrate boron, and produced δ11B with 2 sd reproducibility of ±0.4‰ from samples containing between 2 and 100 ng B.
The potential for Fe isotopes to be measured without chromatographic purification of iron-rich samples was investigated by Chen et al.178 Their study measured 40 Fe-based RMs using MC-ICP-MS. Following dissolution using HNO3 and aqua regia, aliquots of these samples were purified using AG1-X8 cation exchange. The results indicated that there was little bias in δ56Fe between the purified and unpurified samples. This allowed for a simplified and efficient procedure for Fe isotope studies centred on Fe-dominated minerals. Lei et al.179 used a large-geometry, high-resolution MC-ICP-MS for high-precision Fe isotope analysis, following an anion exchange purification step. δ56Fe determinations were found to be strongly correlated with acid molarity and to a lesser extent with sample-standard concentration mismatch. The study also investigated the effect of residual HCl in the analyte following chromatographic separation, and this was observed to make δ56Fe heavier with greater HCl concentrations. When comparable conditions were used, external precision of the measurements were found to be better than ±0.03‰ (2 sd) for δ56Fe. Wang et al.180 used a collision-cell equipped Nu Sapphire MC-ICP-MS to make Fe isotope measurements in a low-resolution mode. Helium and H2 were added to the hexapole collision cell, where H2 was a reaction gas for charge exchange with interfering argide species. Results were observed to be comparable in precision and value to δ56Fe measured in high-resolution mode. The key advantage of this low resolution system was the significantly greater sensitivity, which enabled Fe measurement at 30 ng mL−1. A nice feature of this paper was a figure that summarised the δ56Fe values determined for a spread of geological RMs in a number of different studies across a range of different instruments and techniques. Xu et al.181 analysed Fe isotopes in solids using a fs-LA-MC-ICP-MS system. Their study used wet plasma conditions generated by adding nebulised water, doped with Ni, into the ablation stream before the plasma. This was found to minimise matrix-dependent fractionation of Fe isotopes without reducing signal intensity and improved instrumental bias based on a Ni correction. As such, this method would be suitable to investigate Fe isotopes in a range of solids and potentially within individual minerals or mineral growth bands.
Poole et al.182 devised new methods to determine Pt isotope compositions. The study was targeted at both mass-independent and mass-dependent isotope variations in iron meteorites. Measurements were completed using HR-MC-ICP-MS and utilised a198Pt–196Pt double spike. An improved separation procedure generated high Pt yield and purity enabling essentially interference-free isotopic analysis. Of particular note was the determination of the mass-dependent Pt isotope compositions in the meteorites which demonstrated the importance of correcting for cosmogenic Pt isotope effects caused by exposure to galactic cosmic rays.
Chen et al.183 devised a method of creating S isotope RMs for S isotope determination by LA-MC-ICP-MS. This involved mixing powdered sulphide and sulphate minerals with epoxy resin to generate resin-preserved powders (RPPs). δ34S ratios measured in these RPPs were better than ±0.47‰ (2 sd; n = 20), which indicated that this is a suitable method to develop matrix-matched standards for laser S isotope analysis via bracketing.
Ca isotopes were measured, on samples collected by spacecraft from the Ryugu asteroid, by Moynier et al.184 (following the method detailed in Dai et al.185) who used MC-ICP-MS equipped with a collision cell to minimise Ar interference. They focussed on how Ca concentration, acid molarity and matrix elements effected IRs, and found that matching Ca intensities to within 2% of each other produced the most consistent δ40Ca/44Ca data. The radiogenic ratio, expressed as ε40Ca, is determined with internal normalisation to 42Ca/44Ca and was found to be reproducible to ±0.48‰. From this, the Moynier et al. study was able to determine an average value of δ44Ca/40Ca for the Solar System (relative to SRM915a) of 0.57 ± 0.04‰ (2 sd). Razionale et al.186 proposed an automated chromatographic separation and membrane desolvator to allow measurement of Ca isotopes in much larger quantities. In particular, the concept of their study was to achieve a high throughput of geological carbonate samples. Automated ion chromatography was used to isolate Ca from carbonate prior to analysis using MC-ICP-MS. This method was found to be capable of measuring ∼140 δ44Ca/40Ca values per week with precision of ±0.14‰ (2 sd).
High-precision Rb isotope measurements were made by Hu et al.187 using MC-ICP-MS. Although not strictly a stable isotope system as 87Rb decays to 87Sr, the long half-life means it can be treated as such. A key issue with Rb is that it only has two main natural isotopes and hence is not suitable for double spike fractionation correction. This means that chromatographic purification needs to have a Rb recovery close to 100% to avoid bias. At the same time Rb needs to be effectively free from K and other major elements during analysis. A two-stage chromatographic purification was developed, with a mixed HCl–HF cation exchange stage followed by a Sr-spec Resin to remove residual K. Tests were completed to assess the influence of acid molarity on the isotopes by diluting the SRM984 standard with various molarities of HNO3. Interestingly, this demonstrated that no Rb isotope offset was present only when the acid molarity difference was ≤5%. Similar effects were also found with increasing concentrations of K and Mn as impurities, notably with an offset in δ87Rb of 0.5 with Mn/Rb = 5. The technique was calculated to have an external precision better than ±0.05‰ (2 sd). As a range of igneous rock standards were found to have a range of 0.2‰ this was taken to indicate that Rb isotopes can be fractionated by high temperature processes.
Huang et al.188 investigated the determination of Ti isotopes using fs-LA-MC-ICP-MS. Following signal stability optimisation, external reproducibility of δ49Ti/47Ti was estimated as ±0.07‰ for spot-to-spot analysis. However, it was found that mineral matrix effects had a significant effect on accuracy and robust data could only be achieved using matrix-matched RMs as the bracketing standard. Ti isotopes were measured by Li et al.189 They used a47Ti–49Ti double spike to correct for mass fractionation in MC-ICP-MS following a 2 or 3 column Ti separation procedure. Results were evaluated using SRM3162a alongside the SRM979 Cr standard using its certified 53Cr/52Cr of 0.11339. A range of igneous RMs were measured for δ49Ti/47Ti with an estimated reproducibility of ±0.047‰ (2 sd).
A separation of Cu for isotopic determination was designed by Li et al.,190 using MC-ICP-MS. A two-stage tandem column was set up with a Cu-selective resin passing onto an AG50W-X12 cation exchange resin. HNO3 supplemented by HF and H2O2 were used as the eluants. Long-term reproducibility was estimated as ±0.07‰ for δ65Cu.
Lin et al.191 determined the isotopic composition of Yb in natural materials to act as RMs for future Yb IR measurements in environmental and geoscience applications. A regression model was used in determining the absolute IRs following MC-ICP-MS analysis.
Cd isotopes were determined by plasma electrochemical VG combined with MC-ICP-MS by Liu et al.192 This was specifically aimed at low-Cd complex materials by greatly simplifying the purification process and consuming minimal material. Compared to conventional wet and dry plasma modes, the VG technique produced sensitivity around 6× higher than wet plasma. An additional benefit was found to be the significantly greater tolerance to organic solvents and matrix elements such as Sn, Mo and Zn. Typical internal precision for determinations was found to be ±0.06‰ (2 sd).
Stable Sr isotope measurement of barite was investigated by Sun et al.193 Barite is difficult to dissolve, and strategies have to be deployed to achieve complete solution, yet not fractionate or contaminate Sr. The study used a Na2CO3 exchange method with excessive Na2CO3 which avoided Sr isotopic fractionation during this reaction. Synthetic barites and a synthetic solution of SRM987 mixed with matrix elements were used to validate that the δ88Sr/86Sr was free from bias. Results from natural barite standards revealed significant stable Sr fractionation by up to 0.37‰.
Mo isotopes were measured on low-concentration samples (∼5 ng g−1) by Zhu et al.194 They used a97Mo–100Mo double spike added to sample solutions prior to Mo isolation and measurement using MC-ICP-MS. Results for δ98Mo/95Mo were reproducible to ±0.03‰.
5.6 Nuclear forensics
Bosco et al.195 combined SIMS with RIMS to examine the actinide isotopic composition of individual microparticles. These particles were analysed at a sub-micrometre spatial resolution. An important aspect of this study was that the isobaric interferences, common in techniques involving chemical separation, were suppressed by five orders of magnitude. The study examined 238Pu and 242Am on Chernobyl hot particles to demonstrate the effectiveness of this technique.U isotopes were determined by Jin et al.196 using a static array of Faraday detectors on an MC-ICP-MS instrument. The study deployed a 1011 Ω amplifier for 238U and 1013 Ω for 234U. The relative gain of the 1013 Ω amplifier was calculated by switching 238U/235U between a 1011 Ω pair and a 1011–1013 Ω combination. Results indicated that precision of 238U/235U was in the range 0.3–0.7ε and 234U/238U between 0.3 and 1.0‰, which was considered appropriate for environmental and geochronological applications. U isotopic ratios were measured using triple Q ICP-MS by Lindahl et al.197 The study concluded that drift in the mass calibration of two quadrupoles significantly affected the precision and accuracy of U IRs. Following optimisation of the mass resolution for each quadrupole, results agreed with U reference values and long-term measurement of 233U/235U was reproducible to ±0.07% (2 sd).
Ni et al.198 devised a method to concurrently measure Th and U isotopes from water samples. An automated platform was developed to pass analyte through a UTEVE resin and deliver an ICP-MS-compatible final eluent. ICP-MS/MS was used with a He collision gas to obtain data for 230Th, 232Th, 234U, 235U and 238U. Based on the results for seawater, isotopes were quantified to within ±1% except for 230Th, which with concentrations at ∼0.2 fg mL−1 was ±10–30%.
Quemet and Baghdadi199 provided a methodology for the optimisation of the double IDMS technique. This technique involves adding an isotopic spike of one element with an isotope spike of another element to a sample. From this the ratio between two isotopes of two different elements can be determined, for example 230Th/238U or 238Pu/238U, and hence is a useful technique in geochronological or nuclear applications. A key benefit of this double IDMS method was it did not involve a gravimetric determination in its deconvolution, hence there was a greatly reduced level of uncertainty compared to normal IDMS. The Quemet and Baghdadi study defined the minimum uncertainty envelopes for the spike–spike-sample mixture and included all calculation methodology as well as a piece of code to calculate the optimal spiking parameters.
90Sr measurement is of importance in understanding the effect of anthropogenic radionuclide inputs to humans. Wakaki et al.200 developed a technique which accurately measured small amounts of 90Sr in environmental samples. Using TIMS fitted with a retarding potential quadrupole energy filter to reduce tailing from 88Sr, they were able to quantify 90Sr with a detection limit of 0.23 ag (1.2 μBq).
6 Abbreviations
Whenever suitable, elements may be referred to by their chemical symbols and compounds by their formulae. The following abbreviations may be used without definition. Abbreviations also cover the plural form.| AAS | atomic absorption spectrometry |
| ac | alternating current |
| AES | atomic emission spectrometry |
| AFS | atomic fluorescence spectrometry |
| AP | atmospheric pressure |
| CCD | charge coupled detector |
| CR | collisional radiative |
| CRC | collision/reaction cell |
| CRI | collision/reaction interface |
| CRM | certified reference material |
| CS | continuum source |
| CVG | chemical vapour generation |
| DBD | dielectric barrier discharge |
| DE | deep eutectic |
| DIHEN | direct injection high efficiency nebuliser |
| DSN | desolvation nebulisation |
| EIE | easily ionisable element |
| EPR | electron paramagnetic resonance |
| ESI | electrospray ionisation |
| ETV | electrothermal vaporisation |
| FAES | flame atomic emission spectrometry |
| FIB | fast ion bombardment |
| FI | flow injection |
| FLA | flowing liquid afterglow |
| FLC | flowing liquid cathode |
| FWHM | full width at half maximum |
| GC | gas chromatography |
| GD | glow discharge |
| GF | graphite furnace |
| GIS | gas injection system |
| GO | graphene oxide |
| HG | hydride generation |
| HR | high resolution |
| HV | high voltage |
| ICP | inductively coupled plasma |
| IDMS | isotope dilution mass spectrometry |
| IP | ionisation potential |
| IRA | isotope ratio analysis |
| KED | kinetic energy discrimination |
| LA | laser ablation |
| LAMIS | laser ablation molecular isotopic spectrometry |
| LC | liquid chromatography |
| LDPE | low density polyethylene |
| LEP | liquid electrode plasma |
| LIBS | laser induced breakdown spectroscopy |
| LIF | laser induced fluorescence |
| LIP | laser induced plasma |
| LLE | liquid–liquid extraction |
| LOD | limit of detection |
| LOQ | limit of quantitation |
| LPE | liquid plasma electrode |
| MC | multicollector |
| MDG | microdroplet generation |
| MDL | method detection limit |
| MIP | microwave induced plasma |
| MNP | magnetic nanoparticle |
| MS | mass spectrometry |
| n e | electron number density |
| n ion | ion number density |
| n Ar | number density of Ar atoms |
| NIST | National Institute of Standards and Technology |
| NP | nanoparticle |
| OES | optical emission spectroscopy |
| PAH | polyaromatic hydrocarbon |
| PD | point discharge |
| PGE | platinum group element |
| PIVG | plasma induced vapour generation |
| PN | pneumatic nebulisation |
| PNC | particle number concentration |
| PTFE | poly(tetrafluoroethylene) |
| PVG | photochemical vapour generation |
| QMS | quadrupole mass spectrometry |
| REE | rare earth element |
| RF | radio frequency |
| RM | reference material |
| SC | solution cathode |
| SEM | scanning electron microscopy |
| SERS | surface enhanced Raman spectroscopy |
| SFMS | sector field mass spectrometry |
| SIBS | spark induced breakdown spectroscopy |
| SIMS | secondary ion mass spectrometry |
| SNR | signal-to-noise ratio |
| SPE | solid phase extraction |
| sp | single particle |
| SRM | standard reference material |
| TE | transport efficiency |
| T e | electron temperature |
| T exc | excitation temperature |
| TIMS | thermal ionisation mass spectrometry |
| T ion | ionisation temperature |
| TISIS | total sample consumption system |
| TOF | time-of-flight |
| TOFMS | time-of-flight mass spectrometry |
| T rot | rotational temperature |
| T vib | vibrational temperature |
| UHV | ultra high vacuum |
| UV | ultra violet |
| VG | vapour generation |
| XPS | X-ray photoelectron spectroscopy |
| αion | degree of ionisation |
Conflicts of interest
There are no conflicts to declare.References
- M. Patriarca, N. Barlow, A. Cross, S. Hill, A. Robson, A. Taylor and J. Tyson, J. Anal. At. Spectrom., 2022, 37(3), 410–473 RSC.
- E. H. Evans, J. Pisonero, C. M. M. Smith and R. N. Taylor, J. Anal. At. Spectrom., 2022, 37(5), 942–965 RSC.
- R. Clough, C. F. Harrington, S. J. Hill, Y. Madrid and J. F. Tyson, J. Anal. At. Spectrom., 2022, 37(7), 1387–1430 RSC.
- C. Vanhoof, J. R. Bacon, U. E. A. Fittschen and L. Vincze, J. Anal. At. Spectrom., 2022, 37(9), 1761–1775 RSC.
- J. R. Bacon, O. T. Butler, W. R. L. Cairns, O. Cavoura, J. M. Cook, C. M. Davidson and R. Mertz-Kraus, J. Anal. At. Spectrom., 2022, 37(1), 9–49 RSC.
- S. Carter, R. Clough, A. Fisher, B. Gibson and B. Russell, J. Anal. At. Spectrom., 2022, 37(11), 2207–2281 RSC.
- A. P. R. Santana, P. D. Nascimento, T. G. S. Guimaraes, I. Menezes, D. F. Andrade, A. Oliveira and M. H. Gonzalez, TrAC, Trends Anal. Chem., 2022, 152, 116596 CrossRef CAS.
- V. Jalili, R. Zendehdel and A. Barkhordari, Rev. Anal. Chem., 2021, 40(1), 93–107 CrossRef CAS.
- V. Andruch, R. Halko, J. Tucek and J. Plotka-Wasylka, TrAC, Trends Anal. Chem., 2022, 147, 116510 CrossRef CAS.
- B. Herce-Sesa, A. J. Lopez-Lopez and C. Moreno, TrAC, Trends Anal. Chem., 2021, 143, 116398 CrossRef CAS.
- S. M. Sorouraddin, M. A. Farajzadeh, H. Dastoori and T. Okhravi, J. Iran. Chem. Soc., 2022, 19(6), 2591–2599 CrossRef CAS.
- M. Shamsipur, N. Mafakheri and N. Babajani, Food Anal. Methods, 2022, 15(5), 1203–1213 CrossRef.
- S. Yao, B. Pang, Y. L. Fu, X. L. Song, K. Xu, J. Li, J. Wang and C. Zhao, Sens. Actuators, B, 2022, 359, 131581 CrossRef CAS.
- X. Q. Huang, Z. B. Zhang, L. Z. Chen, Y. J. Lin, R. M. Zeng, J. Xu, S. Z. Chen, J. L. Zhang, H. H. Cai, H. B. Zhou and P. H. Sun, Biosens. Bioelectron., 2022, 212, 114414 CrossRef CAS PubMed.
- X. Y. Sun, X. Wei, X. Liu, X. Zhang, N. Wu, J. H. Liu, Y. Wang, M. L. Chen and J. H. Wang, Anal. Chim. Acta, 2022, 1201, 339639 CrossRef CAS PubMed.
- Y. Wen, X. W. Zhang, Y. Y. Li, S. Chen, Y. L. Yu and J. H. Wang, Anal. Chem., 2022, 94(46), 16196–16203 CrossRef CAS PubMed.
- O. Reifschneider, F. Blaske, A. Vennemann, C. Köppen, M. Sperling, U. Karst and M. Wiemann, J. Anal. At. Spectrom., 2022, 37(5), 1084–1089 RSC.
- Y. Xu, G. Y. Xiao, B. B. Chen, M. He and B. Hu, Anal. Chem., 2022, 94(28), 10011–10018 CrossRef CAS PubMed.
- Y. K. Tanaka, S. Shimazaki, Y. Fukumoto and Y. Ogra, Anal. Chem., 2022, 94(22), 7952–7959 CrossRef CAS PubMed.
- A. J. Nosti, L. C. Barrio, F. C. Celis, A. Soldado and J. R. Encinar, J. Proteomics, 2022, 256, 104499 CrossRef CAS PubMed.
- T. Liu, E. Bolea-Fernandez and F. Vanhaecke, Spectrochim. Acta, Part B, 2022, 198, 106546 CrossRef CAS.
- C. Duyck, R. R. Alves Peixoto, A. A. Rocha, H. Guilhermond de Souza Severino, P. V. Oliveira, R. Damasceno and R. Lorençatto, J. Anal. At. Spectrom., 2022, 37(3), 474–496 RSC.
- C. Duyck, R. R. A. Peixoto, A. A. Rocha, H. G. D. Severino, P. V. Oliveira, R. Damasceno and R. Lorencatto, J. Anal. At. Spectrom., 2022, 37(3), 474–496 RSC.
- E. Mavrakis, Z. Toprakcioglu, N. Lydakis-Simantiris, T. P. J. Knowles and S. A. Pergantis, Anal. Chim. Acta, 2022, 1229, 340342 CrossRef CAS PubMed.
- X. W. Tian, H. W. Jiang, L. G. Hu, M. Wang, W. B. Cui, J. B. Shi, G. L. Liu, Y. G. Yin, Y. Cai and G. B. Jiang, TrAC, Trends Anal. Chem., 2022, 157, 116746 CrossRef CAS.
- O. Geiss, I. Bianchi, G. Bucher, E. Verleysen, F. Brassinne, J. Mast, K. Loeschner, L. Givelet, F. Cubadda, F. Ferraris, A. Raggi, F. Iacoponi, R. Peters, A. Undas, A. Mueller, A. K. Meinhardt, B. Hetzer, V. Graef, A. M. R. Bustos and J. Barrero-Moreno, Nanomaterials, 2022, 12(4), 725 CrossRef CAS PubMed.
- Z. Zhou, A. Al Hejami, M. J. Burgener, J. Burgener and D. Beauchemin, J. Anal. At. Spectrom., 2022, 37(7), 1450–1454 RSC.
- A. R. M. Bustos, K. E. Murphy and M. R. Winchester, Anal. Chem., 2022, 94(7), 3091–3102 CrossRef PubMed.
- A. R. Montoro Bustos, K. E. Murphy and M. R. Winchester, Anal. Chem., 2022, 94(7), 3091–3102 CrossRef CAS PubMed.
- K. H. Chun and W. T. Chan, J. Anal. At. Spectrom., 2021, 36(6), 1261–1271 RSC.
- K. H. Chun, J. T. S. Lum and K. S. Y. Leung, Anal. Chim. Acta, 2022, 1192, 339389 CrossRef CAS PubMed.
- S. Harycki and A. Gundlach-Graham, Anal. Bioanal. Chem., 2022, 414(25), 7543–7551 CrossRef CAS PubMed.
- J. Kocic, D. N. Dirin, R. Kägi, M. V. Kovalenko, D. Günther and B. Hattendorf, J. Anal. At. Spectrom., 2022, 37(8), 1738–1750 RSC.
- A. Williams and D. Beauchemin, J. Anal. At. Spectrom., 2022, 37(4), 727–732 RSC.
- R. M. de Oliveira, D. L. G. Borges, P. Grinberg, Z. Mester and R. E. Sturgeon, J. Anal. At. Spectrom., 2021, 36(6), 1235–1243 RSC.
- J. Hu, H. J. Chen, X. M. Jiang and X. D. Hou, Anal. Chem., 2021, 93(32), 11151–11158 CrossRef CAS PubMed.
- Y. Zhen, H. Chen, M. Zhang, J. Hu and X. Hou, Appl. Spectrosc. Rev., 2022, 57(4), 318–337 CrossRef CAS.
- S. Musil, J. Vyhnanovsky and R. E. Sturgeon, Anal. Chem., 2021, 93(49), 16543–16551 CrossRef CAS PubMed.
- W. Zeng, Z. H. Hu, J. Luo, X. D. Hou and X. M. Jiang, Anal. Chim. Acta, 2022, 1191, 339361 CrossRef CAS PubMed.
- L. Dong, H. J. Chen, Y. Y. Ning, Y. W. He, Y. Yu and Y. Gao, Anal. Chem., 2022, 94(11), 4770–4778 CrossRef CAS PubMed.
- J. Y. Cai, S. Liu, Y. L. Yu and J. H. Wang, Anal. Chem., 2022, 94(4), 2288–2297 CrossRef CAS PubMed.
- M. T. Li, Y. Luo, Z. R. Zou, F. J. Xu, X. M. Jiang and X. D. Hou, J. Anal. At. Spectrom., 2021, 36(12), 2735–2743 RSC.
- Q. He, Y. F. Qiao, M. G. Zhao and J. Zhang, Spectrochim. Acta, Part B, 2022, 191, 106396 CrossRef CAS.
- Q. He, C. C. Li, X. Y. Kong, M. G. Zhao and J. Zhang, J. Anal. At. Spectrom., 2022, 37(9), 1876–1882 RSC.
- X. M. Pan, Y. Lin, Y. B. Su, J. H. Yang, L. B. He, Y. R. Deng, X. D. Hou and C. B. Zheng, Anal. Chem., 2021, 93(23), 8257–8264 CrossRef CAS PubMed.
- Z. Q. Cai and Z. Wang, Anal. Chim. Acta, 2022, 1203, 339724 CrossRef CAS PubMed.
- Y. Zhen, T. Lin, H. Chen, J. Hu and X. Hou, Anal. Chem., 2022, 94(44), 15176–15182 CrossRef CAS PubMed.
- J. Kratzer, M. Lacko, K. Dryahina, T. Matousek, P. Spanel and J. Dedina, Anal. Chim. Acta, 2022, 1190, 339256 CrossRef CAS PubMed.
- A. Lourido-Grovas, I. Lavilla, C. Bendicho and I. de la Calle, Spectrochim. Acta, Part B, 2022, 189, 106373 CrossRef.
- N. Hazel, J. Orejas Ibanez and S. Ray, J. Anal. At. Spectrom., 2022, 37(6), 1229–1239 RSC.
- P. C. Zheng, Y. J. Luo, J. M. Wang, Q. Hu, Y. Yang, X. F. Mao and C. H. Lai, J. Anal. At. Spectrom., 2021, 36(6), 1228–1234 RSC.
- P. Zheng, Y. Yang, J. Wang, H. I. A. Qazi, M. Wu, Y. He, Q. Hu and N. Ding, J. Anal. At. Spectrom., 2022, 37(9), 1806–1814 RSC.
- J. Yu, L. J. Cui, Q. F. Lu, J. L. Li, Y. Feng, B. Wang and X. J. Ma, Spectrochim. Acta, Part B, 2021, 186, 106311 CrossRef CAS.
- Z. Q. Cai, L. Qian, X. X. Peng and Z. Wang, Anal. Chem., 2021, 93(44), 14701–14707 CrossRef CAS PubMed.
- V. Hoffmann, B. Gebel, R. Heller and T. Gemming, J. Anal. At. Spectrom., 2022, 37(6), 1223–1228 RSC.
- K. Swiderski, K. Greda, P. Pohl and P. Jamroz, J. Anal. At. Spectrom., 2022, 37(3), 517–527 RSC.
- A. V. Kuptsov, N. S. Medvedev, O. V. Lundovskaya, A. I. Saprykin and V. A. Labusov, J. Anal. At. Spectrom., 2021, 36(12), 2669–2674 RSC.
- A. V. Kuptsov, A. V. Volzhenin, V. A. Labusov and A. I. Saprykin, J. Anal. At. Spectrom., 2021, 36(4), 829–835 RSC.
- S. Mushtaq, J. Anal. At. Spectrom., 2022, 37(5), 985–993 RSC.
- S. Mushtaq, J. Anal. At. Spectrom., 2022, 37(3), 584–591 RSC.
- J. Merten, Spectrochim. Acta, Part B, 2022, 189, 106358 CrossRef CAS.
- H. J. Pan, L. X. Feng, Y. L. Lu, Y. C. Han, J. P. Xiong and H. M. Li, TrAC, Trends Anal. Chem., 2022, 156, 116710 CrossRef CAS.
- A. Jerse, K. Mervic, J. T. van Elteren, V. S. Selih and M. Sala, Analyst, 2022, 147(23), 5293–5299 RSC.
- J. Chen, R. Wang, M. Ma, L. Gao, B. Zhao and M. Xu, Anal. Bioanal. Chem., 2022, 414(24), 7023–7033 CrossRef CAS PubMed.
- J. Lin, Y. S. Liu, A. Yang, W. Chen, Y. Zhu and Z. C. Hu, J. Anal. At. Spectrom., 2022, 37(3), 592–602 RSC.
- O. Laurent, M. Guillong, C. A. Heinrich, K. Neubauer and C. Stephan, J. Anal. At. Spectrom., 2021, 36(10), 2043–2050 RSC.
- C. Neff, P. Becker and D. Gunther, J. Anal. At. Spectrom., 2022, 37(3), 677–683 RSC.
- C. Neff, P. Becker and D. Günther, J. Anal. At. Spectrom., 2022, 37(3), 677–683 RSC.
- X. Ding, C. Geng, S. Zhai, X. Cao, Z. Shi and K. Liu, Analyst, 2021, 146(24), 7537–7544 RSC.
- T. Van Acker, S. J. M. Van Malderen, T. Van Helden, C. Stremtan, M. Sala, J. T. van Elteren and F. Vanhaecke, J. Anal. At. Spectrom., 2021, 36(6), 1201–1209 RSC.
- T. Van Helden, S. Braeuer, T. Van Acker, O. Leroux, D. Van Der Straeten and F. Vanhaecke, J. Anal. At. Spectrom., 2022, 37(7), 1455–1461 RSC.
- D. Metarapi, J. T. van Elteren and M. Sala, J. Anal. At. Spectrom., 2021, 36(9), 1879–1883 RSC.
- T. R. Holbrook, D. Gallot-Duval, T. Reemtsma and S. Wagner, J. Anal. At. Spectrom., 2021, 36(10), 2107–2115 RSC.
- A. V. Kuptsov, N. S. Medvedev, E. V. Polyakova, A. I. Saprykin and V. A. Labusov, Spectrochim. Acta, Part B, 2022, 194, 106475 CrossRef CAS.
- P. Maung and D. Beauchemin, J. Anal. At. Spectrom., 2021, 36(6), 1104–1111 RSC.
- Y. R. Zhang, J. X. Liu, X. F. Mao, G. Y. Chen and D. Tian, TrAC, Trends Anal. Chem., 2021, 144, 116437 CrossRef CAS.
- S. Y. Gu, X. D. Huang, M. L. Chen, J. X. Liu, X. F. Mao, X. Na, G. Y. Chen and Y. B. Shao, Anal. Chem., 2021, 93(45), 15063–15071 CrossRef CAS PubMed.
- T. Lin, J. Hu, Y. J. Deng, W. Zeng, X. M. Jiang and X. D. Hou, Anal. Chim. Acta, 2022, 1229, 340324 CrossRef CAS PubMed.
- T. Lin, J. Hu, Y. Deng, W. Zeng, J. Chen, X. Jiang and X. Hou, J. Anal. At. Spectrom., 2022, 37(11), 2294–2299 RSC.
- J. H. Yang, Y. Lin, L. B. He, Y. B. Su, X. D. Hou, Y. R. Deng and C. B. Zheng, Anal. Chem., 2021, 93(45), 14923–14928 CrossRef CAS PubMed.
- Y. J. Deng, J. Hu, M. T. Li, L. He, K. Li, X. D. Hou and X. M. Jiang, Anal. Chim. Acta, 2021, 1163, 338502 CrossRef CAS PubMed.
- L. He, P. X. Li, K. Li, T. Lin, J. Luo, X. D. Hou and X. M. Jiang, J. Anal. At. Spectrom., 2021, 36(6), 1193–1200 RSC.
- H. X. Yang, H. Yuan, S. S. Li, W. Wang and D. Z. Yang, Appl. Sci., 2022, 12(10), 4939 CrossRef CAS.
- Y. H. Huang, D. Hirose, J. Minami, M. J. Wang and Y. Takamura, Anal. Chem., 2022, 94(23), 8209–8216 CrossRef CAS PubMed.
- S. White, K. Zheng, J. Tanen, J. E. Lesniewski and K. Jorabchi, J. Anal. At. Spectrom., 2022, 37(4), 870–882 RSC.
- T. Jiang, F. Jiang, Y. Zhong, L. Zhu, Z. Huang, Z. Zhou, J. Zou, J. Zeng, Z. Zhu and Z. Yang, Anal. Bioanal. Chem., 2022, 414(20), 6115–6126 CrossRef CAS PubMed.
- Y. Ikeda, Opt. Express, 2022, 30(17), 29868–29884 CrossRef CAS PubMed.
- H. B. Andrews, L. R. Sadergaski and K. G. Myhre, J. Anal. At. Spectrom., 2022, 37(4), 768–774 RSC.
- H. B. Andrews, L. R. Sadergaski and K. G. Myhre, J. Anal. At. Spectrom., 2022, 37(5), 1161 RSC.
- Y. Naoi, M. Iwata, D. Yokota, G. Gaigalas, D. Kato, I. Murakami, H. A. Sakaue, Y. Sekiguchi, M. Tanaka, H. Tanuma, S. Wanajo and N. Nakamura, Appl. Sci., 2022, 12(4), 2219 CrossRef CAS.
- S. Irvine, H. Andrews, K. Myhre, K. Goldstein and J. Coble, J. Quant. Spectrosc. Radiat. Transfer, 2022, 286, 108184 CrossRef CAS.
- B. Shafique, R. Ali, S. U. Haq, M. Rafique and M. A. Baig, Spectrochim. Acta, Part B, 2022, 193, 106434 CrossRef CAS.
- S. Saakyan, L. D'Yachkov, S. Klimov, V. Sautenkov and B. B. Zelener, Spectrochim. Acta, Part B, 2022, 195, 106503 CrossRef CAS.
- Z. Weiss, Spectrochim. Acta, Part B, 2022, 195, 106500 CrossRef CAS.
- X. Han, P. Wu, Z. Y. Chen, J. Y. Qi, Y. M. Wang, S. Y. Lu, Y. W. Cui, H. B. Tang and J. B. Cao, Spectrochim. Acta, Part B, 2022, 194, 106436 CrossRef CAS.
- Y. Xie, Y. Su, Z. Li, L. Zhai, X. Lan, W. Wang and J. Xu, J. Anal. At. Spectrom., 2022, 37(7), 1554–1564 RSC.
- T. Vonderach and D. Gunther, J. Anal. At. Spectrom., 2021, 36(12), 2617–2630 RSC.
- H. Wiltsche, M. Wolfgang and F. Hallwirth, J. Anal. At. Spectrom., 2022, 37(6), 1298–1308 RSC.
- S. Y. Shi, K. Finch and G. Gamez, J. Anal. At. Spectrom., 2021, 36(5), 1055–1073 RSC.
- T. Volker and I. B. Gornushkin, J. Anal. At. Spectrom., 2022, 37(10), 1972–1974 RSC.
- A. C. Tibere-Inglesse, S. D. McGuire, P. Mariotto and C. O. Laux, Plasma Sources Sci. Technol., 2021, 30(12), 125019 CrossRef CAS.
- M. C. Garcia, C. Yubero and A. Rodero, Spectrochim. Acta, Part B, 2022, 193, 106437 CrossRef CAS.
- T. Gajo, Z. Mijatovic and S. Djurovic, Spectrochim. Acta, Part B, 2022, 194, 106484 CrossRef CAS.
- J. D. Wu, H. Zheng, Y. F. Wang, Q. L. Cao, F. Z. Zhou, J. J. Zhang and X. J. Huang, Phys. Plasmas, 2021, 28(9), 093501 CrossRef CAS.
- W. Xiao, Y. Liao, F. Wang, Z. Zhang, H. Zhu, Y. Yang and K. Huang, IEEE Trans. Plasma Sci., 2022, 50(6), 1781–1789 CAS.
- N. V. Nedić, N. V. Ivanović, I. R. Videnović, D. Spasojević and N. Konjević, J. Anal. At. Spectrom., 2022, 37(6), 1318–1326 RSC.
- F. Ruiz, Z. Benzo, A. Garaboto, J. Salas and J. L. Brito, J. Anal. At. Spectrom., 2022, 37(3), 668–676 RSC.
- A. N. Popova, V. S. Sukhomlinov and A. S. Mustafaev, Appl. Sci., 2021, 11(23), 11237 CrossRef CAS.
- A. N. Popova, V. S. Sukhomlinov and A. S. Mustafaev, Appl. Sci., 2022, 12(13), 6575 CrossRef CAS.
- S. W. Smith, R. W. Martin, N. Hanks, P. A. Creed, K. Kovalcik, R. A. Wilson, K. Kubachka, J. A. Brisbin, J. A. L. Figueroa and J. T. Creed, J. Anal. At. Spectrom., 2022, 37, 898–909 RSC.
- F. V. Nakadi, M. C. Garcia-Poyo, C. Pecheyran and M. Resano, J. Anal. At. Spectrom., 2021, 36(11), 2370–2382 RSC.
- H. Wang, L. Si, C. Li, Z. Zhou and D. Tian, J. Anal. At. Spectrom., 2022, 37(8), 1715–1721 RSC.
- O. N. Grebneva-Balyuk, J. Anal. Chem., 2022, 77(1), 66–80 CrossRef CAS.
- Q. H. Zhang and Y. Z. Liu, At. Spectrosc., 2022, 43(2), 174–185 CAS.
- K. Liu, C. He, C. W. Zhu, J. Chen, K. P. Zhan and X. Y. Li, TrAC, Trends Anal. Chem., 2021, 143, 116357 CrossRef CAS.
- Z. H. Khan, M. H. Ullah, B. Rahman, A. I. Talukder, M. Wahadoszamen, K. M. Abedin and A. Haider, J. Spectrosc., 2022, 2022, 3887038 Search PubMed.
- C. Marina-Montes, V. Motto-Ros, L. V. Perez-Arribas, J. Anzano, M. Millan-Martinez and J. O. Caceres, Anal. Chim. Acta, 2021, 1181, 338947 CrossRef CAS PubMed.
- Z. L. Hu, D. Zhang, W. L. Wang, F. Chen, Y. B. Xu, J. F. Nie, Y. W. Chu and L. B. Guo, TrAC, Trends Anal. Chem., 2022, 152, 116618 CrossRef CAS.
- Z. Wang, M. S. Afgan, W. L. Gu, Y. Z. Song, Y. Wang, Z. Y. Hou, W. R. Song and Z. Li, TrAC, Trends Anal. Chem., 2021, 143, 116385 CrossRef CAS.
- N. Zhang, T. X. Ou, M. Wang, Z. J. Lin, C. Lv, Y. Z. Qin, J. M. Li, H. Yang, N. Zhao and Q. M. Zhang, Front. Phys., 2022, 10, 887171 CrossRef.
- N. Bordel, L. J. Fernandez-Menendez, C. Mendez-Lopez, C. Gonzalez-Gago and J. Pisonero, Plasma Phys. Controlled Fusion, 2022, 64(5), 054010 CrossRef.
- L. J. Fernandez-Menendez, C. Mendez-Lopez, C. Abad, J. Fandino, C. Gonzalez-Gago, J. Pisonero and N. Bordel, Spectrochim. Acta, Part B, 2022, 190, 106390 CrossRef CAS.
- Y. Chen, Y. Liu, Q. Wang, S. Li, Y. Jiang, A. Chen and M. Jin, J. Anal. At. Spectrom., 2022, 37(1), 82–88 RSC.
- T. Delgado, F. J. Fortes, L. M. Cabalin and J. J. Laserna, Spectrochim. Acta, Part B, 2022, 192, 106413 CrossRef CAS.
- D. S. Vogt, S. Schroder, S. Frohmann, P. B. Hansen, F. Seel, M. Gensch and H. W. Hubers, Spectrochim. Acta, Part B, 2022, 187, 106326 CrossRef CAS.
- S. X. Ma, L. B. Guo and D. M. Dong, J. Anal. At. Spectrom., 2022, 37(3), 663–667 RSC.
- M. Gaft, L. Nagli, A. Gorychev and Y. Raichlin, Spectrochim. Acta, Part B, 2022, 192, 106426 CrossRef CAS.
- I. B. Gornushkin, V. P. Veiko, Y. Y. Karlagina, A. A. Samokhvalov and D. S. Polyakov, Spectrochim. Acta, Part B, 2022, 193, 106449 CrossRef CAS.
- N. Li, J. J. Guo, J. J. Song, W. Q. Ye, Y. Lu, Y. Tian and R. E. Zheng, J. Anal. At. Spectrom., 2021, 36(12), 2660–2668 RSC.
- N. Li, N. Nishi, R. Zheng and T. Sakka, J. Anal. At. Spectrom., 2021, 36(6), 1170–1179 RSC.
- A. S. Zakuskin and T. A. Labutin, Spectrochim. Acta, Part B, 2022, 190, 106394 CrossRef CAS.
- G. Arora, J. Thomas and H. C. Joshi, J. Anal. At. Spectrom., 2022, 37(5), 1119–1125 RSC.
- H. C. Wu, C. Li, D. Wu, S. M. Liu, Z. H. Hu, C. L. Feng, R. Hai, F. Ding, G. N. Luo and H. B. Ding, J. Anal. At. Spectrom., 2022, 37(10), 2069–2081 RSC.
- I. Urbina, F. Bredice, C. Sanchez-Ake, M. Villagran-Muniz and V. Palleschi, Spectrochim. Acta, Part B, 2022, 195, 106489 CrossRef CAS.
- Y. Tang, X. Peng, Z. Hu, J. Li, Z. Hao, X. Yang, S. Ma, R. Yuan and L. Guo, J. Anal. At. Spectrom., 2022, 37(7), 1535–1540 RSC.
- A. Taleb, V. Motto-Ros, M. J. Carru, E. Axente, V. Craciun, F. Pelascini and J. Hermann, Anal. Chim. Acta, 2021, 1185, 339070 CrossRef CAS PubMed.
- C. Méndez-López, L. J. Fernández-Menéndez, C. González-Gago, J. Pisonero and N. Bordel, J. Anal. At. Spectrom., 2023, 38(1), 80–89 RSC.
- M. Hu, S. Shi, M. Yan, E. Wu and H. Zeng, J. Anal. At. Spectrom., 2022, 37(4), 841–848 RSC.
- Y. Li, G. X. Zhang, Z. Q. Cai and Z. Wang, Anal. Chem., 2022, 94(16), 6281–6288 CrossRef CAS PubMed.
- B. P. Xu, S. M. Liu, B. Y. Lei, Y. H. Liu, W. F. Zhang, J. Tang, Y. S. Wang, W. Zhao and Y. X. Duan, J. Anal. At. Spectrom., 2022, 37(6), 1350–1359 RSC.
- Z. W. Sun and Z. Alwahabi, Appl. Opt., 2022, 61(2), 316–323 CrossRef CAS PubMed.
- C. J. He, W. H. Huang, Y. Q. Mao, A. Z. Li, K. M. Xu, H. Cui, L. R. Qiu and W. Q. Zhao, Spectrochim. Acta, Part B, 2022, 188, 106340 CrossRef CAS.
- J. X. Lv, C. W. Zhu, Z. Y. Tang, Q. Z. Li, K. Liu, W. Zhang, K. Liu and X. Y. Li, J. Anal. At. Spectrom., 2021, 36(12), 2756–2762 RSC.
- M. Abdel-Harith, A. Elhassan, Z. Abdel-Salam and M. F. Ali, Anal. Chim. Acta, 2021, 1184, 339024 CrossRef CAS PubMed.
- B. P. Xu, Y. H. Liu, B. Y. Lei, J. Wang, W. F. Zhang, Y. S. Wang, W. Zhao, Y. X. Duan and J. Tang, Spectrochim. Acta, Part B, 2022, 197, 106528 CrossRef CAS.
- J. Buday, D. Prochazka, A. Zadera, V. Kana, P. Porizka and J. Kaiser, Spectrochim. Acta, Part B, 2022, 194, 106476 CrossRef CAS.
- Y. Chen, M. Dong, J. Cai, H. Chen, Z. Shang and J. Lu, J. Anal. At. Spectrom., 2022, 37(5), 1126–1133 RSC.
- R. Finotello, M. Tamaazousti and J. B. Sirven, Spectrochim. Acta, Part B, 2022, 192, 106418 CrossRef CAS.
- M. Weiss, Z. Gajarska, H. Lohninger, M. Marchetti-Deschmann, G. Ramer, B. Lendl and A. Limbeck, Anal. Chim. Acta, 2022, 1195, 339422 CrossRef CAS PubMed.
- F. Chang, J. H. Yang, H. L. Lu and H. X. Li, J. Anal. At. Spectrom., 2021, 36(5), 1007–1017 RSC.
- S. Chen, H. Pei, J. Pisonero, S. Yang, Q. Fan, X. Wang and Y. Duan, J. Anal. At. Spectrom., 2022, 37(3), 508–516 RSC.
- Z. Liu, R. Zheng, Y. Tian, B. Wang, J. Guo and Y. Lu, J. Anal. At. Spectrom., 2022, 37(5), 1134–1140 RSC.
- N. H. Cheung, Spectrochim. Acta, Part B, 2022, 194, 106473 CrossRef CAS.
- C. Ursu, P. Nica, G. B. Rusu, C. Vitelaru, G. Popa and C. Focsa, Spectrochim. Acta, Part B, 2022, 196, 106510 CrossRef CAS.
- N. D. Greber and K. van Zuilen, CHIMIA, 2022, 76(1–2), 18–25 CrossRef CAS.
- P. A. Penanes, A. R. Galan, G. Huelga-Suarez, J. A. Rodriguez-Castrillon, M. Moldovan and J. I. G. Alonso, J. Anal. At. Spectrom., 2022, 37(4), 701–726 RSC.
- E. Freres, D. Weis, K. Newman, M. Amini and K. Gordon, Geostand. Geoanal. Res., 2021, 45(3), 501–523 CrossRef CAS.
- H. Q. Huang, M. Guillong, Y. Hu and C. Spandler, J. Anal. At. Spectrom., 2021, 36(4), 836–844 RSC.
- T. H. Luu, P. Gutierrez, E. C. Inglis, D. Roberts and C. Chauvel, Chem. Geol., 2022, 611, 121078 CrossRef CAS.
- D. Rösel and T. Zack, Geostand. Geoanal. Res., 2022, 46(2), 143–168 CrossRef.
- L. Martin, A. Galy, G. Barbotin, F. Claverie, E. Pons-Branchu, C. Tribolo, N. Mercier and C. Pecheyran, Anal. Chem., 2022, 94(7), 3046–3055 CrossRef CAS PubMed.
- L. A. Rush, A. M. Duffin and C. S. Menoni, J. Anal. At. Spectrom., 2022, 37(9), 1902–1914 RSC.
- Y. Masuda, T. Yokoyama, S. Okabayashi, A. Ishikawa and H. Iwamori, Geostand. Geoanal. Res., 2022, 46(2), 205–222 CrossRef CAS.
- E. Yobregat, C. Fitoussi, E. Pili and M. Touboul, Int. J. Mass Spectrom., 2022, 476, 116846 CrossRef CAS.
- C. A. Parendo, S. B. Jacobsen, J. I. Kimura and R. N. Taylor, Earth Planet. Sci. Lett., 2022, 578, 117291 CrossRef CAS.
- D. Bevan, C. D. Coath, J. Lewis, J. Schwieters, N. Lloyd, G. Craig, H. Wehrs and T. Elliott, J. Anal. At. Spectrom., 2021, 36(5), 917–931 RSC.
- Y. K. Di, E. Krestianinov, S. Zink and Y. Amelin, Chem. Geol., 2021, 582, 120411 CrossRef CAS.
- H. C. Liu, Y. H. Chen, C. F. You and C. H. Chung, J. Anal. At. Spectrom., 2021, 36(11), 2322–2329 RSC.
- J. Vogl, O. Rienitz, A. Pramann and L. Flierl, Geostand. Geoanal. Res., 2022, 46(4), 773–787 CrossRef.
- S. C. An, X. L. Luo and W. Q. Li, Rapid Commun. Mass Spectrom., 2022, 36(11), e9289 CrossRef CAS PubMed.
- W. Li, M. Cui, Q. Pan, J. Wang, B. Gao, S. Liu, M. Yuan, B. Su, Y. Zhao, F.-Z. Teng and G. Han, Sci. China, Ser. D: Earth Sci., 2022, 65(8), 1510–1521 CrossRef CAS.
- X.-K. Wang, X.-M. Liu and H. Chen, J. Anal. At. Spectrom., 2022, 37(11), 2410–2419 RSC.
- X.-Y. Zheng, X.-Y. Chen, W. Ding, Y. Zhang, S. Charin and Y. Gérard, J. Anal. At. Spectrom., 2022, 37(6), 1273–1287 RSC.
- H. O. Gu and H. Sun, J. Anal. At. Spectrom., 2021, 36(11), 2545–2552 RSC.
- J. H. Bai, F. Liu, Z. F. Zhang, J. L. Ma, L. Zhang, Y. F. Liu, S. X. Zhong and G. J. Wei, J. Anal. At. Spectrom., 2021, 36(12), 2695–2703 RSC.
- M. Buisson, P. Louvat, C. Thaler and C. Rollion-Bard, J. Anal. At. Spectrom., 2021, 36(10), 2116–2131 RSC.
- I. V. S. Chanakya and S. Misra, J. Anal. At. Spectrom., 2022, 37(6), 1327–1339 RSC.
- K. Chen, Z. Bao, H. Yuan and N. Lv, J. Anal. At. Spectrom., 2022, 37(2), 249–263 RSC.
- T. Lei, M. Li, Z. C. Wang, Y. T. Zhu, Z. C. Hu, Y. S. Li and X. N. Chai, At. Spectrosc., 2022, 43(3), 214–222 Search PubMed.
- J. Wang, D. M. Tang, B. X. Su, Q. H. Yuan, W. J. Li, B. Y. Gao, K. Y. Chen, Z. A. Bao and Y. Zhao, J. Anal. At. Spectrom., 2022, 37(9), 1869–1875 RSC.
- L. Xu, L. W. Xie, H. Wang, Y. H. Yang, C. Huang, J. H. Yang and S. T. Wu, Spectrochim. Acta, Part B, 2022, 189, 106374 CrossRef CAS.
- G. M. Poole, R. Stumpf and M. Rehkamper, J. Anal. At. Spectrom., 2022, 37(4), 783–794 RSC.
- K. Y. Chen, Z. A. Bao, P. Liang, X. J. Nie, C. L. Zong and H. L. Yuan, Spectrochim. Acta, Part B, 2022, 188, 106344 CrossRef CAS.
- F. Moynier, W. Dai, T. Yokoyama, Y. Hu, M. Paquet, Y. Abe, J. Aleon, C. M. O. Alexander, S. Amari, Y. Amelin, K. I. Bajo, M. Bizzarro, A. Bouvier, R. W. Carlson, M. Chaussidon, B. G. Choi, N. Dauphas, A. M. Davis, T. Di Rocco, W. Fujiya, R. Fukai, I. Gautam, M. K. Haba, Y. Hibiya, H. Hidaka, H. Homma, P. Hoppe, G. R. Huss, K. Ichida, T. Iizuka, T. R. Ireland, A. Ishikawa, M. Ito, S. Itoh, N. Kawasaki, N. T. Kita, K. Kitajima, T. Kleine, S. Komatani, A. N. Krot, M. C. Liu, Y. Masuda, K. D. McKeegan, M. Morita, K. Motomura, I. Nakai, K. Nagashima, D. Nesvorny, A. Nguyen, L. Nittler, M. Onose, A. Pack, C. Park, L. Piani, L. Qin, S. S. Russell, N. Sakamoto, M. Schoenbaechler, L. Tafla, H. Tang, K. Terada, Y. Terada, T. Usui, S. Wada, M. Wadhwa, R. J. Walker, K. Yamashita, Q. Z. Yin, S. Yoneda, E. D. Young, H. Yui, A. C. Zhang, T. Nakamura, H. Naraoka, T. Noguchi, R. Okazaki, K. Sakamoto, H. Yabuta, M. Abe, A. Miyazaki, A. Nakato, M. Nishimura, T. Okada, T. Yada, K. Yogata, S. Nakazawa, T. Saiki, S. Tanaka, F. Terui, Y. Tsuda, S. I. Watanabe, M. Yoshikawa, S. Tachibana and H. Yurimoto, Geochem. Perspect. Lett., 2022, 24(1), 1–6 Search PubMed.
- W. Dai, F. Moynier, M. Paquet, J. Moureau, B. Debret, J. Siebert, Y. Gerard and Y. Zhao, Chem. Geol., 2022, 590, 120688 CrossRef CAS.
- D. Razionale, R. N. Bryant and C. L. Blattler, Rapid Commun. Mass Spectrom., 2022, 36(7), e9249 CrossRef CAS PubMed.
- X. Hu, X. Y. Nan, H. M. Yu and F. Huang, J. Anal. At. Spectrom., 2021, 36(12), 2744–2755 RSC.
- C. Huang, H. Wang, L. W. Xie, N. X. Nie, Y. H. Yang, X. M. Zhao, J. Li, H. C. Tian, S. T. Wu, L. Xu and J. H. Yang, J. Anal. At. Spectrom., 2022, 37(10), 2165–2175 RSC.
- J. Li, S. H. Tang, X. K. Zhu, J. X. Ma and X. M. Zhao, Acta Geol. Sin. (Engl. Ed.), 2022, 96(2), 517–524 CrossRef CAS.
- J. Li, Q. H. Zhong, L. Zhang, H. X. Yu, Z. Y. Wang and H. Y. Bao, J. Anal. At. Spectrom., 2022, 37(4), 849–860 RSC.
- R. Lin, J. Lin, K. Zong, A. Yang, K. Chen, Y. Liu and Z. Hu, Anal. Chem., 2022, 94(20), 7200–7209 CrossRef CAS PubMed.
- X. Liu, L. J. Chen, H. Peng, G. Wang, N. S. Belshaw, H. T. Zheng, S. H. Hu and Z. L. Zhu, Anal. Chim. Acta, 2022, 1215, 339980 CrossRef CAS PubMed.
- N. Sun, X. Q. Chen, L. L. Tian and F. Huang, J. Anal. At. Spectrom., 2022, 37(2), 390–398 RSC.
- H.-G. Zhu, J.-M. Zhu, D. Tan, X. Lin, K. Lu and W. Yang, J. Anal. At. Spectrom., 2022, 37(5), 1063–1075 RSC.
- H. Bosco, L. Hamann, N. Kneip, M. Raiwa, M. Weiss, K. Wendt and C. Walther, Sci. Adv., 2021, 7(44), eabj1175 CrossRef CAS PubMed.
- M. G. Jin, Y. N. Jiao, Y. Liu, Q. F. Shao, X. G. Kong and Y. J. Wang, Chin. Sci. Bull., 2022, 67(22), 2651–2661 CrossRef.
- P. Lindahl, G. Olszewski and M. Eriksson, J. Anal. At. Spectrom., 2021, 36(10), 2164–2172 RSC.
- Y. Y. Ni, W. T. Bu, X. T. Ding, K. Xiong, H. L. Wang, C. T. Yang, S. Hu and W. Men, J. Anal. At. Spectrom., 2022, 37(4), 919–928 RSC.
- A. Quemet and S. Baghdadi, J. Anal. At. Spectrom., 2022, 37(2), 240–248 RSC.
- S. Wakaki, J. Aoki, R. Shimode, K. Suzuki, T. Miyazaki, J. Roberts, H. Vollstaedt, S. Sasaki and Y. Takagai, Sci. Rep., 2022, 12(1), 1151 CrossRef CAS PubMed.
| This journal is © The Royal Society of Chemistry 2023 |