Atomic spectrometry update: review of advances in atomic spectrometry and related techniques
E. Hywel Evans*a,
Jorge Pisonerob,
Clare M. M. Smithc and
Rex N. Taylord
aThe Open University, UK. E-mail: e.h.evans@open.ac.uk
bUniversity of Oviedo, Faculty of Science, Department of Physics, c/ Calvo Sotelo s/n, 33006 Oviedo, Spain
cSchool of Education, College of Social Sciences, University of Glasgow, University Avenue, Glasgow G12 8QQ, UK
dOcean and Earth Science, University of Southampton, NOC, Southampton SO14 3ZH, UK
First published on 21st April 2026
Abstract
This review of 184 references covers developments in ‘Atomic Spectrometry’ published in the twelve months from December 2024 to November 2025 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 theseries.2–11 A critical approach to the selection of material has been adopted, with only novel developments in instrumentation, techniques and methodology being included. 3D printing is being used more often to fabricate sample introduction and preparation manifolds, of particular note is the incorporation of pH and temperature responsive materials into 3D devices, so-called 4D printing. A liquid electrode discharge used for vapour generation is a step towards miniaturisation of the sample introduction system for low power plasmas. Likewise, a point discharge cryogenic imaging platform for elemental mapping in biological samples points the way towards more compact instrumentation. Single particle and cell analysis continues to advance with the emergence of more reference materials and improved methodologies and devices. Laser based spectroscopic methods continue to evolve with some novel advances such as acoustic levitation, used for substrate free preconcentration, and isotope analysis.
1 Liquids analysis
1.1 Sample pre-treatment
An advance on 3D printing is 4D printing, where stimuli responsive materials are also incorporated. For example, Kuo and Su13 fabricated an SPE column incorporating a 2-carboxyethyl acrylate (CEA) flexible photocurable resin to form two [H+]-responsive flow-actuated needle valves. The valves closed after loading an alkaline sample (pH > pKa of CEA) due to swelling of the stem, and opened upon loading an acidic eluent (pH < pKa of CEA). The valves could be operated sequentially depending on the pH of the sample and eluent. It is not clear from the paper how this worked in practice, but seems to be related to the pH-controlled response times of the valves, sample loading and elution rates. The system mimicked a sequentially actuated FIA-LOV SPE system, but without the need for a control interface. The device was used for semi-automatic SPE of Cd, Co, Cu, Mn, Ni, Pb and Zn in RMs with ICP-MS detection. Results agreed with certified values (relative errors from −2.2% to +2.0% at the p < 0.05 level). Spike recoveries in natural water and human urine samples were in the range from 96 to 104%. Workers in the same group14 also fabricated a device with four temperature-controlled and magnetically actuated switching valves that were actuated by a hammer-shaped cantilever. The device was fabricated with the inclusion of FeII and FeIII oxide nanoparticle-incorporated polylactic acid filaments. When the inner chamber of the cantilever was filled with warm water (above the glass transition temperature of the polylactic acid), the applied external magnetic field induced the bending of the cantilever to switch the flow direction, and vice versa when removing the external magnetic field. This enabled an automated SPE system to be coupled with ICP-MS for determination of Cd, Co, Cu, Mn, Ni, Pb and Zn, and ions in RMs and spiked samples. LODs ranged from 0.3 to 2.8 ng L−1. Concentrations in the RMs agreed with the certified values (relative error: −3.7% to +4.7% at the p < 0.05 level) and spike recoveries in environmental water and human urine samples were between 96% and 104%.
Figuerola et al.15 evaluated several materials used for 3D printing for leaching of metals. 3D devices (23 × 23 × 6.5 mm) were made using two different printing methods: fused deposition modelling (FDM) and stereolithography (SLA). For FDM, materials were polylactic acid (PLA), nylon (NY), and polypropylene (PP). For SLA, materials were clear resin (a widely used transparent resin), elastic resin (a flexible resin) and 10 K rigid resin (a SiO2 based resin for non-flexible printing). The devices were placed in 15 mL of HNO3 and stirred at 200 rpm for 24 h, then filtered and the HNO3 analysed by ICP-MS for 29 metals. Increasing acid concentration resulted in higher metal concentrations in the leachate. PLA showed the lowest total metal release (below 239 µg kg−1) for up to 17 metals at 20% HNO3. Leaching from PP varied between 172 µg kg−1 in water and up to 15![[thin space (1/6-em)]](https://www.rsc.org/images/entities/char_2009.gif)
000 µg kg−1 in acid. NY exhibited >1200 µg kg−1 leaching and structural degradation above 10% HNO3. SLA resins particularly exhibited Sn leaching up to 1500 µg kg−1, possibly from the print platform. One or more of the toxic metals As, Cd, Cr and Pb were observed to leach from all materials except PP.
Han et al.16 developed an automated microfluidic platform, which incorporated commercially available microfluidic syringe pumps and automated valves, for actinide separation and analysis. A polymethyl acrylate microfluidic chip was constructed in three parts: a bottom layer with a 17.5 mm × 1.4 mm × 0.8 mm channel milled into it and packed with resin; a middle layer to seal the bottom layer; and a top layer with holes for connections to the external mixing manifold. A second mixing chip was also constructed and included in-line. The microfluidic extraction chip was fixed to a square PCB (25 × 25 cm2) together with the various fluidic components. The system was coupled to ICP-MS and controlled using pre-programmed scripts for automated separation and determination of trace elements in a U matrix. Sample volumes were only 20 µL, which reduced consumption of and exposure to any potential actinide materials in samples. LODs ranged between 0.1 ng mL−1 for heavy elements and 49 ng mL−1 for K. The system was validated by the separation and determination of 25 trace elements from a simulated U matrix and 50 from uranium ore (CUP-2 UOC) samples. Each analysis took 30 to 50 min of unattended operational time.
Jones and Jones17 reported an improvement to multi-channel dilution analysis, which is an on-line version of standard dilution analysis. The improvement was effected by modifying the existing simple manifold design by incorporating three channels made from precut, commercially available peristaltic pump tubes, two of which were extended using approximately 10 and 20 cm of plastic tubing. Thus, dilutions of a standard were made as solution was pumped from an autosampler to ICP-AES. Complete calibration consisted of a baseline sample measurement plus three additional standard levels, which were collected in less than one minute. Analyte recoveries for ranging from 95 to 106% were achieved in 20% ethanol and in 1% Ca with RSDs of ∼1%. The method was validated by the analysis of 1566b Oyster Tissue, 1577b Bovine Liver, and 1573a Tomato Leaves, with recoveries ranging from 91 to 104%.
Jiang et al.18 reviewed progress in elemental tagging between 2015 and 2025. The review covered detection methods (elemental tagging, metal containing NPs and label-free detection), sensitivity enhancement (amplification with nucleic acids and single particle analysis) and multiplex detection and mass barcoding for high throughput analysis. They concluded that there are still areas for development, including coupling with separation techniques and better methods to interpret multiplex data.
Much research into tagging methods for the determination of proteins and biomarkers continues to be published. An immunoassay approach is frequently adopted, often in conjunction with immobilisation on magnetic beads or Au NPs. However, these tend to be variations of existing methods, but with a different target protein, often for cancer diagnosis, metabolic studies and toxin detection. Importantly, in vivo detection still presents a challenge, but Fang et al.19 have taken a step in this direction by developing a method for tagging Bruton's tyrosine kinase (BTK) inside live cells. They used a tag comprised of a cell-permeable monoclonal antibody (Loncastuximab) and a targeted covalent Eu tag (Ibt-DOTA-Eu) linked with a cathepsin B-cleavable linker. This enabled cancer cell recognition, internalisation, release of targeted covalent element tags, and specific labelling of Eu to intracellular BTK for ICP-MS quantification.
Stanberry et al.20 developed a method for the determination of lysozyme, myoglobin and human IgG by tagging with La, Gd and Tb respectively using (2,2′,2″,2‴(2-(4-isothiocyanatobenzyl)-1,4,7,10-tetraazacyclododecane-1,4,7,10-tetrayl) tetraacetic acid (p-SCN-Bn-DOTA)). Metal-tagging of the proteins was achieved by covalent labelling of lysine residues. The proteins were separated by gel electrophoreses and then transferred to western blot paper. The tagged proteins were extracted from the paper using a probe with a bevelled tip, which limited contact of the sample to the extraction zone. The probe head formed a seal on the surface of the blot paper and the tagged proteins were extracted for 30 s in 5% HNO3, then directed to the ICP-MS instrument for analysis. LODs of 564, 54, and 2.5 fg were achieved for La, Gd and Tb respectively.
The determination of DNA and RNA has spawned a plethora of essentially similar methods involving various combination of tagging, hybridisation probe reactions and separation using magnetic beads. One such, for the amplification and determination of microRNAs (miRNAs), was developed by He et al.21 They employed two DNA-functionalised ‘walker’ particles, where one moves around the other. The process is quite involved and utilised a DNA walking mechanism involved three different DNA-combined magnetic beads (MB-DNA). MB1-DNA consisted of a DNAzyme strand and a blocker DNA strand attached to MBs. MB2-DNA contained hairpin structures with locked second walking strands. MB3-DNA was functionalised with cleavage substrates and labelled with 195Pt, 193Ir and 103Rh tags. In the absence of miRNA targets, the DNAzyme strand in MB1-DNA hybridised with blocker DNA, preventing the DNAzyme strand from hybridising with the hairpins in MB2-DNA and inhibiting DNA walking process. In the presence of targets, blocker DNA hybridised with miRNAs, releasing DNAzyme in MB1-DNA as the first walking strands, which then hybridised with the loop region of the substrate hairpins in MB2-DNA. These were then cleaved by DNAzyme-induced hydrolysis catalysed by Mg2+. The hairpin DNA (https://www-sciencedirect-com.libezproxy.open.ac.uk/topics/biochemistry-genetics-and-molecular-biology/dna-hairpin) structure was consequently destroyed, exposing new DNAzyme strands in MB2-DNA. The adjacent walking strand in MB1-DNA could then bind to another substrate hairpin in MB2-DNA, initiating the relative motion from MB1-DNA to MB2-DNA and the step-by-step cleavage of substrate hairpin. Upon the consumption of most substrate hairpin on the MB2-DNA surface, MB1-DNA dissociated from MB2-DNA and interacted with a new one. The walker 1 generated a large number of new walking strands in MB2-DNA, which drove the walker 2 process through a similar mechanism as walker 1. In walker 2, MB2-DNA moves around MB3-DNA via the DNAzyme-induced ribonucleoide hydrolysis, cleaving the substrate strand, thus generating significant DNA fragments with element tags. The elements were then separated magnetically and quantified using ICP-MS. The method was used to determine miRNA-21, miRNA-199a, and miRNA-499 with LODs of 1.8, 1.1, and 1.5 fM, respectively, with high specificity. Related amplification strategies were developed by other workers. Zhang et al.22 reported miRNA determination using Y tagging and a catalytic hairpin assembly amplification. Detection using ICP-MS yielded an LOD of 0.24 fM. Wu et al.23 developed a method for the analysis of circulating tumour cells in lung cancer samples. The approach was based on the interaction between aptamers and the proteins PD-L1 and mucin 1, which are overexpressed on the cell surface. Amplification was achieved using Y-DNA nanospheres and multiple catalytic hairpin assembly reactions. In addition, the formation of a four-armed structure through biotin-C-Ag+-C and biotin-T-Hg2+-T hairpins enabled filtration through a 0.1 µm nylon 6 filter for separation of free Hg2+ (Ag+), T-Hg2+-T (C-Ag+-C), and SA-(biotin-T-Hg2+-T)4 (SA-(biotin-C-Ag+-C)4) and selective determination by ICP-MS.
Xu et al.24 used spICP-MS and a strand displacement amplification (SDA) with CRISPR/Cas12a (an RNA-endonuclease guided system used to modify DNA at specific sites) strategy for miRNA-21 detection. Gold NPs were hybridised with linker DNA, forming large aggregates. In the absence of target miRNA-21, these large aggregates produced high pulse signals with spICP-MS detection. In the presence of the target miRNA-21 the SDA reaction was triggered, and the products activated CRISPR/Cas12a to cleave the linker DNA, resulting in disassembly of the AuNP aggregates. These AuNP aggregates, with smaller size, displayed lower pulse signals with spICP-MS detection. Thus, a relationship between the average pulse signal intensity of AuNP aggregates and the concentration of miRNA-21 was obtained, with an LOQ of 0.5 fmol L−1.
1.2 Sample introduction
Recent developments in sample introduction have focused on improving transport efficiency (TE), expanding analytical capabilities to complex matrices and larger particulate systems, and broadening biological and environmental applications. Xu et al.25 reviewed recent advances in single-cell ICP-MS, highlighting improvements in sample introduction, quantification strategies, and analytical throughput for studying cellular heterogeneity and NP uptake. The review emphasised ongoing challenges, including transport efficiency optimisation and multiparametric detection, while also demonstrating the technique's potential in biomedical research and metal-based drug studies. This biological focus aligns with broader trends toward high-resolution and single-entity analytical techniques.Sandro et al.29 investigated four sample introduction configurations to extend the size range of microplastic particles measurable by ICP-TOFMS. By implementing alternative introduction strategies, including falling-tube and vertical ICP configurations, the authors successfully increased the upper size limit to approximately 20 µm, exceeding the limitations of conventional pneumatic nebulisation. These developments are particularly relevant for environmental monitoring, where larger microplastic particles remain analytically challenging and environmentally significant.
Hernández-Postigo et al.31 integrated IDA with spICP-MS to mitigate matrix effects during NP sizing and counting. The method enabled accurate determination of size and particle number concentration for 30 and 50 nm PtNPs in seawater, biological media and urine.
Traditionally, spICP-MS quantification relies on integrated peak areas. Bazo et al.32 proposed transient peak heights as an alternative analytical metric. Using polynomial modelling of cumulative intensity, the approach improved accuracy close to the LOD and reduced uncertainty compared with raw maximum intensity measurements. The method enhanced sizing accuracy for smaller NPs and improved threshold selection, representing a novel parameter for particle characterisation.
Improved TE remains critical for accurate single-event measurements. Wang et al.33 developed a high-efficiency sample-introduction system through computational modelling, 3D printing and iterative testing. Particles sized between 20 and 100 nm achieved TE above 18.8%, while larger particles were lost completely through deposition. Optimisation strategies, including elevated operating temperatures, increased efficiency to 61.1%. Priede et al.34 evaluated multiple nebuliser and spray-chamber configurations for sp- and single cell-ICP-MS. High-flow cyclonic systems achieved approximately five-fold higher sensitivity than low-flow total-consumption systems. TEs varied widely, reaching up to 90% depending on the setup and calibrant. These findings emphasise the need for appropriate calibrant selection for accurate TE determination.
Although spICP-MS is primarily applied to NPs, measurement of microparticles requires extended linear dynamic ranges. The TE may become particle-size dependent, and deviations from linearity can arise from incomplete vaporisation or detector limitations. By reducing instrument sensitivity by up to 269×, the linear dynamic range was significantly extended, allowing signals proportional to particle mass.35 While the minimum detectable particle size increased, complete vaporisation of SiO2 particles up to 5000 nm was demonstrated.
A microextraction (ME) probe-based spICP-MS method enabled direct sampling of NPs from surfaces in a single step.36 Performance depended on parameters such as flow rate, carrier solution, particle size and probe-head design. A newly developed PEEK probe head improved extraction efficiency compared with commercial designs. Although higher flow rates reduced efficiency, they shortened analysis time. Overall extraction efficiencies (4–10%) were comparable to conventional spICP-MS TE and allowed direct analysis of nano- and microparticles while preserving spatial information.
Grebneva-Balyuk et al.37 reviewed biological sample preparation strategies for spICP-MS, including dilution, filtration, enzymatic digestion and LA. Methods that preserved analyte integrity, particularly dilution and filtration, were highlighted. LA coupled with spICP-MS was identified as a promising technique for investigating spatial distributions of NPs in biological systems. Xing et al.38 reviewed recent advances in AgNP analysis in biological materials, comparing pre-treatment approaches and identifying ongoing analytical challenges and optimisation requirements. The impact of filtration and ultracentrifugation on NP recovery from complex matrices were evaluated by Paton et al.39 Most preparation methods caused substantial particle losses (≥90%). Surfactants such as Triton X-100 improved AuNP recovery (∼30%) but were ineffective for Fe-rich particles, which exhibited losses up to 99%. The findings highlight the critical role of sample preparation in achieving reliable quantitative environmental NP analysis.
Sikora et al.40 developed a multimethod platform combining both dynamic image analysis (DIA) and spICP-MS for determining the number concentration of small microplastics. Using 1–10 µm polystyrene particles, the study identified detection thresholds, transport-efficiency calibration and sample-introduction configuration as major contributors to accuracy. Under optimised conditions, both techniques produced consistent results for 5 µm particles. Measurement uncertainty was primarily associated with particle-count variability in (DIA) and both counting variability and transport-efficiency calibration in spICP-MS.
Plasma mediated vapour generation (PMVG) has been the subject of some research activity over the last few years. In this technique, volatile species are generated using a plasma, with some mechanistic similarities to photochemical vapour generation (PVG). The species can be subsequently atomised using a separate atomisation source, or alternatively, atomised in situ by the plasma used to generate the volatile species. Coelho et al.42 developed a DBD PMVG source which was used for both generation and atomisation of volatile As species. The DBD was arranged much like a traditional quartz trap, aligned in the optical path of an AAS instrument. Operating parameters were 200 mL min−1 He + 10 mL min−1 H2 discharge gas and 4.1 kV discharge voltage. Samples (18 µL) were introduced as droplets via a side-arm using an automated capillary, the capillary withdrawn, then the sample dried within the DBD tube with the discharge ignited in He only. Atomisation of As species was effected by the introduction of H2. The LOD was 9 µg L−1. The authors also performed a mechanistic study using two separate sources in order to separate the vapour generation and atomisation steps. Their interpretation of the results was that AsH3 was generated in the DBD at low voltage (up to 2.1 kV) without any changes in its chemical structure. As the voltage was increased the AsH3 was atomised until complete atomisation at 2.9 kV. The free As atoms were observed to decay quickly, making it impossible for them to be transported over long distances, thus the integrated source presented an advantage over traditional methods involving a separate VG reactor and atomiser.
Photochemical vapour generation (PVG) is a well-established technique whereby free radicals and aqueous electrons are generated from a sample solution exposed to UV radiation. These undergo reactions to form volatile species such as hydride, alkyl or carbonyl compounds of analytes of interest. The specificity and extent of the reaction is influenced by the reaction medium and concomitant species, particularly the presence of organic acids and transition metals used as so-called ‘sensitisers’. Sturgeon et al.43 investigated the role of these sensitisers, specifically TM and noble metals. They included both a tutorial review of the research area and results of their own investigations of the relative concentrations of stable small molecules (CO, H2, CO2 and CH4) generated during UV photolysis of acetic and formic acids, and the impact of added TMs. Their premise was that since these gases are the products of radical precursors, they should provide an insight into the effects of TMs on radical production. They found that oxidation of formic and acetic acids (and corresponding radical production) was <1%, but since this translated to a 104-fold molar excess of active radicals compared to typical analyte concentrations (∼100 ng mL−1) generation of volatile species was still efficient. They postulated that the effect of added TMs was due to the in situ formation of (photo)catalytic TM/TMO nanoparticles and ligand-to-metal charge transfer reactions, which both contributed to the production of reducing radicals. They placed these conclusions in the context of other mechanistic studies and acknowledged that further work was necessary to investigate their conclusions, especially with the regard to the role of (photo)catalytic TM/TMO nanoparticles. The same group published several similar papers reporting on PVG of different elements, including As, Bi, Co, Ge, Pt and Rh. Mechanistic studies reported in these papers generally supported the view of Sturgeon et al.43 that nanoparticles of sensitiser elements are formed and may play a catalytic role.
Guselnikova44 developed a method for the volatilisation of Te using Cl2gas. Samples of high purity Te (100 to 500 mg) were placed into quartz cups in a quartz chamber heated to 240 °C. This was placed in-line with a flow reactor constructed from laboratory glassware. Chlorine gas was electrolytically generated and passed over the Te samples sufficient to completely volatilise them. The evolved gas was collected in a beaker containing 30% NaOH. Any residue in the quartz cups was dissolved in 3:1 HN03-HCl. White precipitate from the reactor walls was also collected and analysed using XRD. Spike recoveries for 39 trace elements, determined using ETAAS and ICP-AES, ranged from 80 to 120%. Clearly some elements will not be volatilised or will precipitate within the flow reactor so the method is far from universal. However, the authors cite an advantage compared with traditional vacuum distillation of Te such that sample masses were between 4 and 20 times smaller when volatilised as TeCl4. Also, the elements Cd, Cr, P, Rb and Y were retained using this method.
2 Solids analysis
2.1 Direct methods
Recent advances in direct solids micro- and nanoscale analytical techniques have focused on improving the accuracy, spatial resolution, and throughput of isotopic and elemental measurements using SIMS, NanoSIMS, EPMA and mass spectrometry imaging (MSI), with applications that include geology, biology, and nuclear forensics. Developments in spatially resolved mass spectrometry were reviewed by Körber et al.,45 who provided a comprehensive overview of MSI technologies, including SIMS, MALDI, desorption electrospray ionisation (DESI), LA-ICP-MS, and emerging multimodal and targeted imaging strategies across materials science and life-science applications.Expanding the capabilities of MALDI MSI, Stopka et al.46 demonstrated direct elemental imaging using MALDI MSI, enabling spatial mapping of elements such as Ca, Cl, Fe, Gd and Pt and alongside biomolecules and establishing MALDI MSI as a complementary approach to SIMS and LA-ICP-MS for metallomics and biomedical research.
Several groups have published work describing methods to overcome spectral interference, matrix effects and beam-induced elemental migration in EPMA. He et al.47 developed a refined EPMA protocol for F and Cl quantification in mafic silicate glasses, showing that matrix effects dominate Fe Lα-F Kα overlap corrections and achieving LODs of 46 ppm for F and 12 ppm for Cl under optimised conditions. Tu et al.48 addressed challenges in high-spatial-resolution EPMA of H2O-bearing aluminosilicate glasses. Alkali metal ions, particularly Na+ and K+, are prone to migration under typical EPMA conditions, causing analytical bias. Glasses with varying water contents were analysed and Na was found to be more mobile than K under equivalent conditions. Alkali mobility was dependant on beam conditions and internal glass characteristics. Optimised analytical parameters and correction strategies suitable for small melt inclusions and experimental glass products were proposed. Riche et al.49 demonstrated that EPMA can serve as a non-destructive alternative for mixed-element microparticle characterisation in nuclear forensics, achieving trace-level quantification of dopants in synthetic oxide particles when calibrated against high-precision ICP-MS data.
Solid analysis MS methods like LA-ICP-MS, GDMS, and SIMS routinely achieve LOD at the ng g−1 level, but each has limitations. Laser ablation and ionisation mass spectrometry (LAIMS) is a newer technique showing good potential for direct solids analysis.50 In this study, LAIMS performance was improved by integrating a fs laser and an ion-counting system. The high-frequency fs laser increases atomisation and ionisation efficiency and speeds up analysis. Combining time-to-digital converter (TDC) and analogue-to-digital converter (ADC) data extended the dynamic range to nine orders of magnitude while maintaining ng g−1 LODs. Spectral interferences contributed only about 1 × 10−5 of the total ion current. These improvements position LAIMS as a versatile and competitive alternative to established techniques such as LA-ICP-MS, GDMS and SIMS.
Progress has been made in correcting matrix effects and instrumental mass fractionation (IMF) during SIMS-based isotope analyses. Dong et al.51 introduced an online nanoSIMS matrix correction method for B IRA in tourmaline, demonstrating a strong correlation between IMF and Fe-Mn content, eliminating the need for offline EPMA data by simultaneously measuring major-element and isotopic signals. He et al.52 evaluated sulfur isotope microanalysis in calcite by SIMS, identifying substantial matrix effects related to organic matter content and validating a biogenic calcite as a homogeneous sulfur isotope RM.
Komarov and Stebelkov53 systematically compared 235U/238U isotope ratios in microparticles by SIMS using two instruments with different magnet sizes. The larger magnet instrument was capable of significantly better precision and LODs for minor isotopes compared with the smaller magnet instrument. Complementary work demonstrated that pre-screening U microparticles by SEM improves interpretation of SIMS isotope data by identifying particles representative of distinct source materials.54
Pressure and geometry have also been identified as critical control parameters in GD. Kowalczyk et al.57 reported that elevated-pressure SCGD operation suppressed N2 emission and increased atomisation efficiency, producing LODs lower than those of ICP-OES for alkali metals. Wang et al.58 systematically investigated anode geometry in SCGD-OES. Optimal performance was achieved with a 2.4 mm diameter rod and a 60° conical angle, yielding high emission stability, low RSD (<1.3%), and improved LODs. The use of incorrect geometries led to thermal instability or discharge spreading. The study provided practical design guidance for optimising SCGD systems.
Several papers focused on improving the fundamental understanding and diagnostic capability of GD. Nedić et al.59 validated end-on OES as a practical diagnostic tool for estimating electric field strength in Grimm-type discharges via Stark-broadened He lines, offering a simpler alternative to spatially resolved measurements. Sharma et al.60 explored the temporal evolution of the opto-galvanic effect in Ar GD, providing insight into excited-state dynamics and laser–plasma interactions. At a more detailed spatial level, Zheng et al.61 applied the Abel inversion to SCGD plasmas, revealing strong radial non-uniformities in emissivity and excitation temperature, with direct implications for optimising analytical observation zones. LODs for Ag, Ca, Cd, Cu, In, K, Mn and Rb were determined at the positions of central and maximum emissivity, showing improvements by a factor ranging from 1.1 to 4.0 at the extreme positions following the Abel inversion.
SCGD-OES has significant potential as a metal element detection technique; however, matrix effects remain a challenge. Standard dilution analysis was combined with SCGD-OES to address matrix effects in complex matrices by Zhang et al.62 This automated calibration was found to compensate for matrix-induced signal suppression, allowing accurate determination of Ca, Fe, and Zn in pharmaceutical solutions. Zheng et al.63 introduced an iterative shift difference fitting (ISDF) algorithm for baseline and interference correction in SCGD-OES. The method eliminated the need for repeated background measurements and adapted to element-specific spectral features. Calibration accuracy exceeded R2 = 0.995, with measurement errors reduced to ∼5%. The algorithm significantly improves quantitative accuracy in SCGD-OES under complex spectral conditions. A fundamental investigation of matrix interactions by Zhang et al.64 involved the examination of analyte transfer and excitation mechanisms in SCGD using AAS. The results provided information on the influence of common cations and anions, together with the addition of a common sensitivity enhancer, HCOOH, on analyte transfer mechanisms in the SCGD. This helped to understand the general working fundamentals of the technique and possible matrix effects when combined with OES detection.
Lang et al.65 demonstrated that liquid sampling APGD coupled with Orbitrap MS enabled highly sensitive detection of Br and I as atomic anions, overcoming ionisation inefficiencies and isobaric interferences common in ICP-MS. Optimisation of discharge and fragmentation conditions enabled detection limits of 50 pg for Br and 5 pg for I. High-precision isotope ratio measurements were achieved with minimal matrix effects from Na+, K+ and Mg2+. Bryars et al.66 described a submerged plasma for isotopic detection and elemental resolution (SPIDER) probe for use with APGD below the surface of liquids. Determination of transition and rare earth elements in molten salts and detection of the isotopic shift of the Hβ → Dβ emission line were demonstrated. Two excitation mechanisms (film explosion and droplet vaporisation) were identified, each producing distinct plasma characteristics and emission features.
2.2 Indirect methods
Alternative calibration concepts have also been applied to environmental particle analysis. Brunnbauer et al.71 proposed an in-house prepared polystyrene thin-film calibration strategy for sizing microplastics via LA-spICP-MS. Reliable sizing across multiple polymer types without certified particulate standards was demonstrated. Schoberl et al.72 demonstrated the effectiveness of monodisperse microdroplet calibration for non matrix-matched quantification in LA-ICP-TOFMS. Using microdroplet-based calibration, less than ±20% deviation from reference values across certified materials and a gelatin standard was obtained. In contrast, using NIST SRM 610 as an external standard to quantify gelatin produced larger errors, whereas microdroplet calibration provided improved accuracy relative to traditional external standards. In a review of the applications of LA-ICP-MS in the analysis of solid fuel waste, Yang et al.73 further highlighted the ongoing need for improved calibration approaches in complex matrices where matrix-dependent fractionation and lack of suitable RM continue to hinder quantification.
The reliability of LA-ICP-MS measurements is strongly influenced by the characteristics of laser-generated aerosols and ablation dynamics. Holá et al.74 provided a comprehensive overview of aerosol characterisation methods and demonstrated how particle size distribution, morphology and experimental conditions affect TE, signal stability and elemental fractionation. Both online approaches – which can provide real-time information – and offline approaches – which allow for detailed morphological and compositional analysis – were described. By integrating these perspectives, the review provided guidance for selecting appropriate strategies to study aerosol properties and for optimising LA protocols toward improved accuracy, reproducibility and interpretation of analytical results. Using LA-ICP-TOFMS, Kaser et al.75 evaluated elemental fractionation during ablation and plasma processes. Mass balance calculations enabled separation of laser-induced and ICP-induced fractionation without offline sample analysis. The results can inform an understanding of TE and fractionation mechanisms. Knowledge of these parameters aids understanding of the ablation behaviour of different matrices and may therefore improve nonmatrix-matched quantification using LA-ICP-MS. Further work examined matrix-specific ablation behaviour. Ke et al.76 analysed UV-laser ablation of CaF2 crystals, identifying optimal surface roughness and laser fluence conditions to avoid catastrophic ablation and allow determination of REE in the crystals.
Optimal sampler and skimmer cone orifice diameter have been well developed for solution based ICP-MS but are not fully understood for LA systems. Richards et al.77 investigated the impact of cone orifice on ICP-MS sensitivity using solution mode and a low-dispersion LA system. The findings confirmed the suitability of standard cone sets for nebulisation of homogeneous solutions whilst suggesting a slight reduction in cone orifice diameter improved ion transmission through the cones.
Recent methodological advances have enabled diverse applications in environmental monitoring, biomedical analysis and particle research. Modlitbová et al.78 reviewed the use of LA-ICP-MS and LIBS for microplastic detection and characterisation. The authors evaluated the analytical advantages, limitations, and emerging applications in the area, highlighting the complementary nature of both techniques and advocating their tandem use for comprehensive analysis of microplastics. Manard et al.79 demonstrated the feasibility of LA-ICP-TOFMS for rapid elemental and isotopic analysis of environmental aerosols, achieving high-throughput detection of Ru-containing particles.
3 Instrumentation, fundamentals and chemometrics
3.1 Instrumentation
A review of ‘key’ publications in atomic spectrometry was published by Chan et al.81 The article provided commentary on 1055 publications in the field, each of which were nominated by at least one current expert and supported by a narrative that justified inclusion. The paper is organised into coherent subsections and chronological order. The most important papers were represented, as well as some surprising inclusions. However, a consideration of the early development of speciation methods is one area which seemed lacking.Miniaturised and portable instruments continue to be developed for a variety of scenarios. Such instruments generally utilise a low-power atomisation and excitation source with the consequent necessity for a sample introduction system which delivers a dry sample. A portable analyser that integrated ETAAS and ETAES, enabling simultaneous multi-element detection with compact, low-power instrumentation suitable for on-site analysis, was described by Li et al.82 Compared with single-function AAS or AES with the same electrothermal tungsten coil, the dual function analyser was demonstrated to expand the range of detectable elements and their linear ranges and improve analytical efficiency. It was also proposed that the simultaneous AAS and AES mode could be applied to the study of atomisation and excitation mechanisms of analyte elements. The same group83 developed a miniaturied PD-AES instrument by arranging a hollow titanium tube electrode (0.9 mm i.d., 1.2 mm o.d., 40 mm in length) and a tungsten rod electrode (1 mm diameter, 30 mm in length) perpendicularly in a T-shaped quartz (1.3 mm i.d. × 2.5 mm o.d). The discharge gap between the two electrodes was 4 mm and the PD microplasma was generated within the quartz tube. Samples were subject to HG and introduced, after passing through a gas–liquid separator (GLS), through the hollow electrode into the PD region in a flow of Ar. Analyte emission was collected via a quartz plate and collimating lens into a fibre optic for transmission to a CCD spectrometer. LODs for As, Sb, Ge, Pb, Hg, Se, and Sn were 1, 0.4, 0.5, 0.06, 0.09, 6, and 0.2 µg L−1, respectively, and the relative standard deviations (RSDs) were all less than 3.3%. Shen et al.84 developed a dual-stage DBD-PD excitation source for use as an in situ analyser. A discharge chamber (50 mm × 30 mm × 5 mm) 3D-printed to include a gas channel (38 mm × 5 mm × 3 mm). Quartz plates, with Cu foils on the outside surface, were embedded on its upper and lower surfaces to act as electrode-dielectrics. Two W electrodes were placed symmetrically on the left and right sides of the discharge chamber as PD electrodes, with a discharge gap of 3 mm. The idea behind the dual source was for the DBD to act a pre-excitation source before subsequent excitation in the PD. Sample introduction was by HG and emission was collect using fibre optic-CCD. The authors claimed up to 16-fold increase in sensitivity compared with a single PD or DBD source, with LODs for Se and As of 0.8 and 0.2 µg L−1, respectively, and RSDs < 5%.
Liu et al.85 reported on a liquid electrode discharge microplasma-induced vapor generation (MPI-VG) device for determination of total organic carbon by OES. The basic instrument was the usual arrangement of VG coupled with a PD-OES instrument. The novel part was the VG device, which was fabricated as a PMMA chip with a curved sample flow channel (i.d. 1 mm). Tungsten rod electrodes were inserted into the channel in contact with the reaction liquid and a voltage applied, thus generating a microplasma in an argon bubble between the two reaction liquids. Consequently PMVG was effected to produce CO2 which was swept into the PD. Emission of C at 193.0 nm yielded an LOD of 0.15 mg L−1 with a RSD of < 3.7%.
Cai et al.86 developed a portable analyser by coupling a mini ultrasonic nebuliser with PD-MIP-OES. The aerosol was heated in a ceramic tube to vaporise the solvent, which was subsequently separated from the dried aerosol particles in a condenser tube in an ice bath. Thus, subsequent detection by PD-MIP-MS was enhanced up to 17-fold due to the absence of solvent (93% desolvation efficiency). For 160 µL samples analysis was completed within 10 s, with LODs for Cd, Zn, Pb and Mn of 1.3, 1.2, 2.4 and 2.1 µg L−1, respectively.
A portable LA-MIP-MS instrument was developed by Farcy et al.87 for use in spaceflight. A low power (30 W), low pressure (<1 Torr) MIP using an Evenson cavity was operated using 50 mL min−1 of He. The LA system utilised a commercial 266 nm solid state Nd-YAG laser with a fluence of ∼6 J cm−2. Analysis of stainless steel samples was performed using NIST SRM 1154 as a reference, and yielded results within 1.4 to 4% of those determined using XRF, with precision ranging from ±9.1 to 22% (2σ). Isotope ratios for Cu and Ni were withing ±0.8 to 3% (2σm) precision and reproducibility ranging from 0.12% to 11.8%. The LODs were 21 ppm for 57Fe, 780 ppm for 54Fe and <240 ppm for Cr, Mn, and Ni. A comparison was made with the performance of instruments used in previous space flights, and the data indicated that LODs were an order of magnitude better compared to LIBS and APXS, with less integration time (∼90 s per transient analysis) and fewer calibration points (∼10–15).
Ambient ionisation is a technique whereby ions are directly volatilised from a solid surface. Guo et al.89 used an MPT to induce in situ surface release and ionisation of volatile species of Ag, As, Bi, Cu, Pb, Sb, Se and Te from solid samples under ambient conditions without any reagent consumption and sample pretreatment, with analysis by MS. The authors speculated that active hydrogen species in the MPT caused the generation of volatile species (e.g. metal hydrides), with subsequent surface release and ionisation. Quantification of As and Sb in soil samples was performed by using the CRM GBW07405 as a standard, with respective LODs of 1.4 and 0.35 mg g−1. Khoo et al.90 simultaneously imaged both elements and biomolecules in mouse brain tissue by splitting an LA flow. ICP-MS was used to determine Cu, Fe, Mg, Mo, P and Zn in one flow, and DBD ionisation MS (DBDI-MS) was used to determine lipids and other biomolecules in the second flow. The LODs for P, Cu and Zn were 6.5, 0.4 and 2.4 pg respectively using LA-ICP-MS. Using DBDI-MS LODs for adenine (m/z 136.06) cholesterol (m/z 369.35) and phosphatidylcholine (m/z 551.50) were 91, 97 and 698 pg respectively.
Wang et al.91 developed an interesting technique for imaging biological samples. They constructed a cryogenic DBD imaging platform using quartz plates as a sample stage and dielectric layer between a tungsten wire electrode and a copper plate electrode. The uppermost quartz plate served both as the dielectric barrier between the electrodes and as the sample holder, and the whole was placed onto a cryogenic stage on a displacement platform. The cryogenic sample holder ensured that there was no water loss when the DBD was ignited, which improved sample excitation and avoided destruction of the biological sample. The system was evaluated using carrot samples cut into 200 µm thick slices in the shape of letters cut with a press. The DBD was ignited at 40 V for 200 ms discharge duration and a 100 µm moving step size, with the signal monitored using a CCD spectrometer via an optical fibre.
In a two-part study, Chan first charcterised94 and then applied95 an intensified CCD for photon counting. Parameters such as detector dark noise, threshold setting for single-photon detection, intensifier gain, analog-to-digital conversion rate, and pre-amplifier settings were characterised and discussed. Intensifier gain was identified as the most critical parameter, which affected signal spike size and ion feedback. The authors proposed a systematic optimisation strategy to minimise false positives and maximise counting efficiency. This was then applied95 for spectral analysis using LIBS. They found that photon counting, using this approach, provided higher spectral resolution compared to a conventional analog detector, which was particularly evident in the line wings of spectral peaks. This facilitated better discrimination of isotopic lines – e.g. two U-isotopic peaks were clearly resolved. Photon counting did not directly improve resolution, rather it rectified the resolution lost by signal spreading in analog measurement. There were a few caveats: a unimodal model was used to compensate for detector coincidence losses, so it was necessary to work at low signal intensity; because of the slow frame rate, it was necessary to process detector images offline to search for signal spikes.
Tao et al.96 developed an HG-AFS instrument based on a planar array single photon counting imaging detector (based on a microchannel plate), a flat-field concave grating as the splitter, and an EDL as the excitation source. Detection of As and Bi lines between 180 and 320 nm was achieved without scattering interference. Cited advantages were simplicity, ambient operation, no moving parts and simultaneous spectral measurement.
Pomme and Boulyga97 presented a throughput model explaining non-linearity in discrete ion counters used in MS. Ion count rates were corrected for losses due to non-extended dead time, but not for the additional extending dead time generated by pileup of the detector pulses. Formulas were given for the true throughput rate, the inverse throughput to correct for count loss, counting uncertainty, error propagation of the characteristic dead time and pulse width, and the incomplete dead-time correction error. This approach was used to extend the linearity of the ion counter for the measurement of the 233U/235U IR signal count rate from ∼104 to 3 × 107 s−1. Pomme published a second paper98 which addressed non-linearity in count rate for variable signals. In this case, when the input rate changes during measurement the output rate does not vary linearly with it and a conventional dead time correction will not suffice. Hence, the authors presented a throughput formula with a correction factor to account for the non-linearity between the average input and output rates.
3.2 Fundamentals
100 and 16700 K and was determined using the Boltzmann two-line method applied to Nb II lines at 307.6857 nm and 308.0343 nm. Values between 5.1 to 8.3 pm were obtained for the Stark widths and −0.27 to +2.09 for Stark shifts. Nedić et al.100 determined Stark widths for five Bi I, thirteen Bi II and three Bi III lines the wavelength range of 298 to 570 nm, using LIPS in He at reduced pressure. Values of ne (0.3 × 1023 to 1.1 × 1023 m−3) and Te (14000 to 17000 K) were reported. Stark widths for the Bi II 514.45 nm and 520.93 nm lines were determined to be 105.7 ± 10.8 nm and 143.7 ± 14.6 nm, respectively, which compared well with other values from the literature. Gosse et al.101 determined Stark widths for the O 777 nm triplet, using both LIPS and a DC arc. Given that the triplet was unresolved each transition was determined by deconvolution of the instrumental function contribution from the FWHM of the Lorentzian component of the interpolation. The Stark broadening was then characterised using an empirical law based on Griem's theory. For ne = 9.1 × 1022 m−3 and Te = 12200 K, Stark widths (FWHM) were determined to be 4.72, 3.32 and 3.24 pm for the lines at 777.19, 777.42 and 777.54 nm respectively.Transition probabilities were determined by Sarma et al.102 for 15 Nd II lines in the spectral range 378 to 521 nm, using a modified HCL to generate a stable Nd plasma, using Ar as a carrier gas. They used the Boltzmann plot method with the assumption of partial LTE. Results agreed within 30% of previously published data, confirming the validity of the pLTE assumption. Kodangil103 determined transition probabilities of 16 Eu I lines and 44 Eu II lines using LIBS in the range 200 to 1200 nm. They used a Boltzmann plot and assumed LTE. Several of the values were compared with results of other workers and showed good agreement. Dojić et al.104 determined transition probabilities of for nine Bi II lines in a LIP using the Saha–Boltzmann plot method for three ionisation states of Bi, assuming LTE. The ne was estimated based on the Stark width of He I 388.86 nm. They applied an Abel transform method and ensured that the plasma was in LTE through the McWhirter criterion. Belmonte et al.105 determined transition probabilities and oscillator strengths for 92 Xe II lines in the UV region between 217 and 385 nm, 91 of which were measured for the first time. They assumed pLTE in a pulsed-discharge lamp, used a Boltzmann plot with exponential fitting and statistically calculated the uncertainties for the transition probabilities.
Nine new spectral lines for S were reported by Antonov et al.106 The lines were identified in an ICP and He, Ar and air DC plasmas. The new lines were observed at 912.052, 913.696, 934.864, 936.132, 936.571, 937.877, 938.856, 939.279 and 940.183 nm.
3.2.2.1 Plasmas. Diagnostic studies of plasmas provide insights into atomisation, ionisation and excitation processes, as well as the effect of interferences, so are useful for development of existing and new analytical sources. Han et al.107 performed an experimental investigation of the dynamics of an ICP tail flame. The pulse frequency of the NAZ (determined using a high-speed camera) was stable and synchronous whereas the tail-flame fluctuated due to ambient air entrainment. Emission intensities of alumina powder particles injected into the coolant gas were tracked inside the torch and the particle velocity was calculated by the time-of-flight method, yielding axial and radial velocities of between 1.7 to 2.5 ms−1 and between 2.9 and 4.1 ms−1 respectively. Nebulisation of an Er solution was used to generate Er+ clouds which could be tracked downstream of the NAZ, yielding axial velocities (Vz) which gradually reduced from ∼17 to 10 ms−1 between ∼17 and 70 mm above the torch outlet at 1100 W, with higher power corresponding to higher velocity. The experimental results were compared with simulations, and Vz was found to be significantly lower than the simulated value at the same nominal RF power, probably because the power coupling efficiency was lower than unity. Radial velocities (Vr) on the plasma axis were close to zero in the range of 10 mm < z < 30 mm, indicating that the ion cloud mainly moved along the axial direction (z). At z > 30 mm, the ion cloud had non-zero radial velocity with its position and magnitude moving upstream as RF power increased. Axial velocities determined using the simulation varied between ∼12 and 27 ms−1 for powers between 600 and 1200 W, which generally declined as radial position increased.
Polyatomic ion interferences have posed a problem in ICP-MS from its inception. Pupyshev et al.108 performed an experimental and theoretical investigation of ArM+ ions. The ArM+: M+ ion intensity ratios for V, M, Co and Cu were measured with varying power and flow rate. For all four elements, the ratio increased with decreasing power and increasing Ar flow rate. Calculations of the equilibrium composition for V and Co species were also performed using a quasi-equilibrium thermodynamic model and the TERRA software package, which is based on the principle of maximising the entropy of a thermodynamic system, and thermodynamic data from the IVTANTHERMO database plus recently published data for ArO, ArO+, Ar2, Ar2+, ArN, ArN+, ArH, ArH+, ArV+ and ArCo+. Calculations were performed for varying plasma temperature (T), Ar flow, analyte concentration, concentration of Na as an EIE, and type of acid in the sample solution. ArM+: M+ ratios were predicted to decrease with increasing plasma temperature from 3500 to 9000 K at any Ar flow rate, analyte concentration and type of acid. In the presence of Na, ArM+ was predicted to decrease with increase in T and decrease overall with Na concentration. Furthermore, ArM+:M+ ratios qualitatively correlated with the bond energies of the ArM+ ions and the experimental data. The authors proposed that, using such theoretical and experimental data, ICP operating conditions could be chosen to minimise ArM+ spectral interferences, though how this would vary from day to day and for different instruments remains a question.
Space charge effects in ICP-MS were investigated by Zhu et al.109 They studied Ar+ related space-charge effects on elements of groups 1, 2, and 13 for various conditions of power and the voltage of the first extraction electrode. The premise being that space charge effects have frequently been reported with respect to matrix elements but not the dominant plasma species, namely Ar+. Increased suppression of signal intensities was observed with increasing power over 1.1 kW. This was attributed to space charge effects due to the increase in 40Ar+ density. A more negative extraction electrode voltage resulted in a more pronounced space charge effect but this decreased with an increase in m/z ratio of the elements. Signal suppression was least between 0.8 and 0.9 kW, particularly for element with m/z ratio < 100. It is questionable whether this adds much to the canon of knowledge regarding space charge effects given that an optimisation strategy for power and extraction voltage has long been recognised as essential to maximise signal intensity.
3.2.2.2 Graphite furnaces. Butcher110 provided an overview of recent developments in GF-AAS and GF-MAS, describing the advances in direct analysis, speciation, preconcentration and solid/slurry sampling, illustrating how furnace techniques are expanding toward molecular detection. Imai et al.111 investigated the thermodynamic interactions between gaseous atoms and graphite surfaces, using Gibbs free energy calculations from atomic absorption signals to explain atom-surface binding mechanisms during atomisation. It was proposed that the interaction between M(g) and the pyrolytic graphite surface was caused by the binding to sites of top-C for Au, C–C bridge for Ag and Pb, a hexagonal hole of sixfold sites within C-ring for Co, Fe and Ni, a pair of terminal-C on zigzag edge for Cr, top-subedge for Mn, and active sites in adjacent graphite layers controlling intercalation for Cu.
3.3 Chemometrics
Duponchel et al.112 investigated the use of a large language model (LLM) via a smartphone to perform multivariate analysis of a hyper-spectral imaging data set obtained using LIBS. In particular, they highlighted the process as being performed in an ‘interactive conversational manner’ This type of data analysis is clearly destined to become more common, but also raises serious questions about the reliability of the analysis when the underlying algorithms are invisible, there is an absence of validation, and very little knowledge is required to implement it and critically assess the results.This issue is exemplified by the problem of how measurement uncertainty affects classification outcomes in multivariate analysis, as investigated by Carneiro et al.113 The authors used Monte Carlo methodology to estimate classification uncertainty by propagating spectral measurement noise in simulated 2D, XRF and LIBS data, using various models. They found that classification uncertainty was influenced by spectral similarity between classes, measurement error, model complexity and the geometric alignment of data structures relative to decision boundaries. When class distributions aligned parallel to decision boundaries, uncertainty was reduced, but perpendicular alignment amplified uncertainty. Simple classifications in XRF using partial least squares discriminant analysis (PLS-DA) models yielded confident predictions using a single latent variable, but spectrally similar classes required more complex modelling and resulted in higher uncertainty. Classification of LIBS analysis of Dalbergia species using support vector machine (SVM) kernels revealed that uncertainty varied across different SVM kernel types, even when classification accuracy was perfect. A polynomial kernel introduced greater uncertainty whereas linear and radial basis function (RBF) kernels yielded more reliable predictions.
Cernatic et al.114 addressed the issue of LOD as applied to 2D elemental imaging, where signals below the LOD often remain visually discernible. They proposed the just noticeable difference (JND) figure of merit, i.e. the smallest perceptible difference between two stimuli detectable by the human senses. They developed a web-based application which allowed 2D maps to be imported and to perform the image processing to calculate the LOD and JND metrics, and demonstrated the approach as applied to LA-ICP-MS and LIBS mapped images of Fe, La, Gd and Ba in structured thin films prepared by photolithography. In practice, images were rendered in different shades of grey so that notable spatial features and lower absolute values of measured signals could be detected by the JND calculation. Signals far below the LOD value in complex LIBS images of moderate noise could be detected. In low-noise LA-ICP-MS maps JND did not give an advantage over LOD.
4 Laser-based atomic spectrometry
Key fundamental studies and instrumental developments (published in 2025 and at the end of 2024) in laser-based atomic spectrometry are highlighted in this section. Atomic spectrometry techniques where the laser is used as either an intense energy source or a source of precise wavelength (e.g. LIBS, LIF or LIMS) are considered. However, studies related to LA-ICP-MS/AES, 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 describes the latest instrumental developments and fundamental studies related to LIBS. Reviews that cover detailed LIBS applications in imaging, classification or quantitative analysis of different kinds of samples can be found elsewhere. For instance, a LIBS method primer was presented by Palleschi et al.115 outlining key considerations for optimising measurements, acquiring representative spectra and ensuring proper data interpretation. The authors reviewed current high-impact applications and future developments that may position LIBS as a stronger competitor to mainstream analytical methods. A comprehensive overview of recent advances in chemical composition imaging was presented by Zhao et al.,116 including a description of operational principles, imaging modalities, data-processing workflow and emerging technological trends. Another review by Safi et al.117 presented the current state and prospects of Tag-LIBS, an emerging technique that improves LIBS performance with molecular specificity using elemental tags. The authors traced the development of Tag-LIBS from its early demonstrations using silicon and iron-oxide microparticle tags to more sophisticated systems employing lanthanide-doped NP, quantum dots, and polymeric carriers. The authors concluded that Tag-LIBS was positioned as a promising multimodal platform for biomedical, environmental and point-of-care analysis. Rizwan et al.118 presented a broad overview of DP-LIBS, from fundamentals to applications, highlighting the multiple geometric and temporal pulse configurations.Shevchenko et al.122 developed a plasmonic-ordered-structure-enhanced LIBS (POSELIBS) platform based on SiO2 microsphere arrays coated with gold films. The authors employed a vertically self-assembled, highly ordered photonic structure that enabled precise tuning of localised surface plasmon resonance through control of sphere radius and gold thickness. The results confirmed that resonance-matched structures (e.g. 115 nm spheres, 14 nm Au) provided strong enhancement in the sensitivity of LIBS for the detection of TiO2 NP (e.g. enhanced factor values up to 11.6 and a >10× reduction in RSD). Moreover, off-resonant substrates provided moderate signal enhancement at 355 and 532 nm, in line with modeling predictions.
Statistical comparison of LIBS signals generated by fs- and ps- LA was performed by Lednev et al.123 The authors demonstrated that ps-LIBS provided stable, Gaussian-like statistics independent of plasma temperature fluctuations, indicating that variations in atomised mass dominated signal instability. In contrast, ps-LIBS deviated from normality and showed stronger temperature fluctuations and greater influence of laser–plasma interaction, which distorted the signal distribution more than in the fs case. It was observed that fs ablation produced higher emissivity and more uniform plasma, while ps pulses generated hotter plasmas with reduced atomised mass, consistent with partial plasma shielding.
Song et al.124 investigated how ambient gas properties, including specific heat ratio (γ), molar mass (M) and ionisation energy (IE), governed both repeatability and intensity in LIBS measurements. By using gas mixtures that differed from air in only one of these three properties, the study overcame limitations associated with comparing pure gases, whose physical parameters are inherently coupled. Systematic plasma diagnostics (e.g. emission spectra, fast imaging and shadowgraphy) revealed that γ and M predominantly influenced shockwave strength and plasma morphological stability; while IE determined the initial absorption location of laser energy and thereby modified the early expansion dynamics of the plasma. The authors concluded that higher γ consistently enhanced both signal intensity and repeatability by reducing plasma heat dissipation and weakening back-pressure effects. Higher M increased laser absorption and thus intensity, but at the cost of reduced repeatability due to stronger shockwave-driven instabilities. A moderate increase in IE stabilised plasma formation and favoured energy deposition into the ablated material.
Kou et al.125 investigated the origins of signal uncertainty in LIBS applied to combustion diagnostics. The authors analysed how plasma formation and evolution differed across low temperature premixed gases and high temperature reactive flame regions. Signal fluctuations were shown to primarily arise from strong pulse-to-pulse variations in plasma morphology, which depended on local gas properties and incident laser energy. In low temperature premixed regions, the plasma became optically thick, leading to strong plasma shielding. These disturbances increased RSD despite increased signal intensity. Conversely, in high-temperature reactive zones, reduced gas density resulted in optically thin plasmas that absorbed only a small fraction of laser energy. Under these conditions, the authors identified the formation of a two-lobe plasma structure, generated by localised energy absorption at the edges of an initial plasma kernel. Interaction and merging/separation of these lobes produced large spatial fluctuations that severely degraded LIBS repeatability.
The formation, transport, and re-deposition of ablated material during micro-LIBS imaging was investigated by Motto-Ros et al.126 Using a brass alloy embedded in epoxy resin as a model system, the authors combined high-resolution optical microscopy, SEM analyses, and time-resolved plasma imaging to trace the fate of material ejected during each laser shot. Three classes of particles were identified: native debris produced directly by laser ablation; micron-scale particles, formed at the plasma–air interface during plume confinement and rapid cooling; and nanometric particles, generated within the dense early plasma through electrostatic growth. Moreover, shockwave dynamics were shown to play a central role in redistributing the materials.
A detailed investigation of microwave plasma torch (MPT) assistance as a multi-mechanism enhancement strategy for LIBS was performed by Wu et al.127 The authors showed that the MPT simultaneously lowered the breakdown threshold, increased the ablation mass, injected microwave energy into the early-stage plasma, and extended plasma lifetime by over two orders of magnitude. These combined effects provided much stronger enhancement than Ar flow-confinement alone, with intensity gains up to ∼40× and significant improvements in tolerance to defocusing, making the method attractive for irregular or rough targets. Some limitations were also exposed, including: destabilisation of MPT performance at higher laser energies where shockwave–torch interactions disrupted plasma confinement; dependence of enhancement on precise Ar flow and microwave power; and increased system complexity compared to simpler confinement or dual-pulse schemes. Another hybrid analytical approach that integrated fs-LIBS with a MPT was developed by Wei et al.128 to overcome short plasma lifetimes, weak emission, and high breakdown thresholds. By simultaneously irradiating the sample with ultrashort laser pulses and a 2.45 GHz Ar MPT, the system dramatically reduced the laser energy required for plasma generation from the mJ to the µJ regime. As a result, 30 µJ MPT-fs-LIBS produced stronger spectral emission and higher SNR than conventional 1.5 mJ fs-LIBS, while significantly improving measurement stability (RSD reduction by up to 4×). Additionally, LODs for trace metals in aluminium alloys improved by up to 130-fold at equal laser energies, and even at 30 µJ, detection sensitivity remained 17-times superior to 1.5 mJ fs-LIBS.
The transformation of a conventional Gaussian laser beam into an annular beam, using an axicon combined with a spherical lens, was demonstrated by Dai et al.129 to improve the analytical performance of LIBS. When applied to an alloy-steel target using a Q-switched Nd:YAG laser, the annular beam generated a larger and more spatially uniform plasma with increased spectral stability. The authors reported a 2 to 3-fold improvement in spectral stability, a 2.1× enhancement in sensitivity and a 38.5% reduction in the LOD, emphasising that the flatter spatial energy distribution of annular beams promoted more stable plasma formation and more efficient energy coupling into the sample. Complementarily, a numerical investigation of the annular–point dual-pulse (DP) laser excitation scheme was presented by Chen et al.130 In this configuration, a first annular pulse generated a toroidal plasma that converged radially toward the centre, followed by a second point pulse directed at the central region. The authors used the FLASH radiation-hydrodynamics code to compare this geometry with conventional single-pulse LIBS across different inter-pulse delays and fluences for boron targets under low-pressure conditions. The simulations showed that a 20 ns delay provided the strongest enhancement, sustaining elevated Te and ne within the critical 100 to 200 ns diagnostic window, where continuum background and Stark broadening were minimised.
An in situ dynamic correction strategy for mitigating ablation effects in LIBS was developed by Li et al.131 The methodology integrated LIBS and Raman spectroscopy with deep-learning modelling. The authors showed that consecutive LIBS pulses induced oxide accumulation and spatial confinement within the heat-affected zone, causing strong spectral instability. Raman spectroscopy provided a real-time proxy for ablation-induced chemical and morphological changes by monitoring oxide vibrational signatures in the strong, transitional, and weak oxidation regions. Data were employed to feed a deep convolutional neural-network correction model capable of iteratively compensating oxidation and confinement effects. Classification accuracy increased from 81.4% (uncorrected) to 90.2% after dynamic correction, and to above 99.3% when combined with data augmentation and feature-level fusion of Raman and LIBS signals.
A sealed fibre optic LIBS probe designed for underwater elemental analysis was developed and characterised by Sun et al.132 The system integrated a coaxial Ar gas-flow delivery that formed a stable solid–gas interface on submerged targets. This configuration produced a plasma with thermodynamic properties comparable to those generated under atmospheric conditions. Moreover, the use of axial plasma imaging through multimode fibres allowed real-time probe alignment in dynamic underwater conditions. Quantitative analysis of trace Cr in steel samples was achieved using internal standardisation, achieving an LOD of 95 ppm and an LDR from 100 to 13800 ppm. This work represents an advance toward robust, remotely deployable LIBS instrumentation for complex aquatic or radioactively hazardous environments, with implications for nuclear-debris characterisation, underwater archaeology, and in situ geochemical sensing.
A polarisation resolved LIBS (PR LIBS) configuration for accurate Cr quantification in soil was developed by Xu et al.,133 supported by a theoretical correction model derived from Fresnel equations and Malus's law. By modulating the incident laser polarisation and analyzing polarisation-dependent plasma emission using a Glan–Thompson prism, this method enhanced energy coupling, suppressed unpolarised background and matrix-induced interferences, and improved the SNR of Cr I 425.433, 427.481 and 428.973 nm lines. Using 150 datasets from soils spiked with five Cr concentrations, the authors demonstrated that PRLIBS improved linearity and analytical stability, increasing SNR by an average of 27.8% and reducing background by 71.5%.
Acoustic-levitation-assisted LIBS was investigated by Shetty et al.134 for the quantification of trace B in aqueous samples. Microliter-scale droplets were acoustically levitated to be free from any surface contact and uniformly preconcentrated through controlled IR laser evaporation, eliminating matrix–substrate interactions and improving reproducibility. Systematic delay gate mapping showed that optically thin behavior was achieved only for delay times ≥ 750 ns and gate times ≥ 1000 ns. Under these optimised conditions an LOD for H3BO3 (expressed as HBO2) of 28.7 mg L−1 was achieved. Moreover, the authors provided reliable measurements of trace B in commercial mineral water, with a deviation of ∼12% from the certified value. The method quantified B as low as ∼25 ng per droplet and demonstrated that, even in the absence of full LTE, the branching ratio approach (theoretical intensity ratio between B I 249.67 nm and 249.77 nm) provided a robust criterion for ensuring suitable plasma conditions.
A novel approach for isotopic analysis of liquid lithium, laser produced vapour (LPV) for LIBS, was developed by Tran et al.136 A dual-beam Nd:YAG configuration was employed for vapour generation and plasma formation, respectively. The 6Li/7Li isotopic shift (∼15.7 pm) was resolved with linear response and a precision of between 2.5 and 5.2%, addressing long-standing limitations of direct liquid-phase LIBS caused by bubble formation, plasma confinement and self-absorption. Laser induced plasmas in a LiAlO2 target at Ar pressures from 0.1 to 100 Torr were investigated by Karim et al.,137 using fast-gated imaging and spatially integrated OES. The authors assessed the relationship between Li atom distributions, emission gradients and self-reversal in the Li I line profiles. The results showed that self-reversal of the Li I 670.8 nm line became significant at delays beyond 5 µs for pressures between 1 and 100 torr due to the accumulation of ground-state Li atoms at the plume–background interface. Conversely, at pressures of 0.1 Torr or lower, rapid plasma expansion without a well defined interface prevented the formation of self-reversal features.
A morphology based calibration strategy for LIBS was developed by Pei et al.138 to link ablation crater 3D reconstruction with matrix effect correction. The authors showed that accurate crater volume measurements, obtained through a calibrated industrial CCD microscope system, correlated with laser energy, pulse duration, wavelength and sample physicochemical properties. A nonlinear model was developed to reduce matrix-induced deviations and improve quantitative accuracy.
A heterogeneous ensemble learning (HEL) framework for full-spectrum quantitative LIBS analysis was developed by Fang et al.139 by combining four fundamentally different regression models (CNN, Lasso, Boosting and FNN) using a Bayesian weighting strategy. The authors overcame the limitations of single model regressors, which struggle with the high dimensionality, instability and nonlinearity characteristic of LIBS spectra. Using the Steel and ChemCam datasets, the authors showed that HEL improved prediction accuracy compared with the individual models, homogeneous ensembles and previously proposed heterogeneous combinations. Reductions in root mean square error (RMSE) of up to ∼40% and more stable performance across elements were achieved.
A multimodal approach that integrated LIBS with laser induced plasma acoustic (LIPA) signals was proposed by Gao et al.140 to classify Cu alloys. The results showed that acoustic features provided valuable complementary information on plasma dynamics and substantially improved classification accuracy compared with LIBS alone. Using PCA for dimensionality reduction, support vector machine (SVM) for classification, and shapley additive explanations (SHAP) for interpretability, the authors highlighted the dominant contribution of LIPA features. Similarly, a quartz-tuning-fork-based acoustic correction method was developed by Zheng et al.141 to enhance stability and quantitative accuracy. The tuning fork, owing to its narrowband resonance, high Q-factor, strong noise immunity, and very low cost, provided a robust acoustic sensing alternative to microphones and polymer-fiber sensors traditionally used in acoustic LIBS. Using standard steel samples the authors demonstrated that acoustic correction reduced RSDs and improved calibration curve determination coefficients for Cr, Cu, Mn and Ni. Zhou et al.142 converted LIPA data into high-dimensional acoustic spectrograms using continuous wavelet transform. The authors extracted physically meaningful features (spectrogram energy and area) that correlated with ablation mass and plasma expansion dynamics. These acoustic descriptors were combined with plasma temperature, electron density, and elemental interference parameters obtained from the LIBS spectra, to construct a spectral deviation mapping model grounded in radiation physics. This strategy was demonstrated for four distinct metallic matrices (Al, Fe, Ni, and Ti), showing improvements in quantitative performance for the analytes studied.
Bosáková et al.143 investigated how laser wavelength, focusing conditions, and surface properties influenced both the optical and acoustic responses of LIPS. Experiments using 1064 nm and 266 nm Nd:YAG lasers showed that shorter wavelengths improved ablation efficiency, while longer wavelengths better coupled energy into the plasma, enhancing optical emission. Laser Induced Plasma Acoustics (LIPA) signals increased with fluence and were strongly affected by material dependent breakdown thresholds, enabling material discrimination under defocused, low fluence conditions and matrix effect minimisation at exact focus. Acoustic and LIBS mapping of aluminum surfaces coated with a thin copper layer confirmed that LIPA was sensitive to subtle differences in composition and surface finish, though with lower contrast than optical emission.
The progressive degradation of calibration accuracy caused by long-term instrumental drift and fluctuating experimental conditions was addressed by Xu et al.144 Instead of relying on a single calibration model, the authors constructed multiple calibration models using LIBS spectra acquired on different days under nominally identical conditions. To allow dynamic model selection during analysis, each calibration model was ‘tagged’ with the intensity of selected weak Fe lines, which were influenced by temporal variations in plasma conditions rather than changes in sample composition. During the analysis of steel, the optimal calibration model was selected for quantitative analysis by characteristic matching. Using Cr, Mn, Mo and V in alloy steel as test analytes, ten daily calibration models were created and validated over five days. The results demonstrated that this characteristic line-guided multi-model calibration approach significantly improved both accuracy and long-term reproducibility relative to conventional single-model calibration.
The use of resonant-excitation LIBS (RE-LIBS) was investigated by Yang et al.145 to improve the detection of gaseous ammonia, which is a long-standing challenge due to weak plasma emission, fast dissipation and low molecular density. By combining a tunable CW CO2 laser with conventional LIBS, the authors optimised gas flow-rate and resonant excitation energy, achieving enhancement factors up to ∼6.6 for H I lines and ∼2.4 for N I lines, respectively. The authors demonstrated that the enhancement mechanism was dominated by energy absorption and molecular vibrational coupling, rather than by thermal effects from CO2-laser irradiation. Preliminary tests with ammonia-ethylene mixtures supported the feasibility of RE-LIBS for elemental gas analysis.
Zhang et al.146 demonstrated that minimising signal uncertainty, rather than maximising the SNR was the more effective strategy for improving quantitative LIBS performance. Using brass samples measured under three ambient pressures of 100 kPa (baseline), 60 kPa (maximum SNR), and 5 kPa (minimum signal uncertainty) the authors showed that optimisation of SNR did not yield the best analytical results. Although SNR was strongly enhanced at 60 kPa, both univariate and multivariate statistical models exhibited reduced accuracy due to exacerbated matrix effects and larger variations between samples in plasma temperature and electron density. Conversely, at 5 kPa, where the SNR remained close to atmospheric levels, but the RSD of spectral intensities was minimal, both accuracy and precision improved substantially.
Guerrini et al.147 evaluated different denoising strategies tailored for ultrafast kHz-rate µLIBS imaging. While kHz lasers accelerated data acquisition, allowing >2 million spectra to be recorded per image, they also imposed harsh constraints (e.g., low-energy pulses that produced weaker plasmas, and high-speed sCMOS detection that reduced photon flux per pixel). Together, these factors significantly degraded SNR, especially for minor or biologically relevant trace elements. To address this, the authors compared five spectral denoising methods: Savitzky–Golay smoothing; FFT filtering, wavelet threshold denoising (hard/soft); Whittaker smoothing; and PCA. The study was focused on for Au, Fe, and P in rat kidney tissue containing exogenous gold nanoparticles. The authors demonstrated that PCA-based denoising outperformed the other approaches, improving SNR by a factor of ≈5, preserving emission-line shapes, and enabling more accurate chemical-image reconstruction.
An integrated substrate engineering strategy, that simultaneously enhanced plasma generation and suppressed the coffee-ring effect (CRE), was introduced by Wang et al.148 for heavy-metal detection in aqueous samples. The authors fabricated dual functional Cu-based substrates (e.g., super hydrophilic CuO/Cu nanosheet arrays, and super hydrophobic D-CuO/Cu films, respectively). The CuO nanosheets acted as efficient anti-reflection micro/nano absorbers, lowering the ablation threshold by >40% and enabling up to 46-fold enhancement of Cu plasma emission through improved laser–matter coupling, higher Te (∼1.8×), and increased ne (∼2.2×). Suppression of the CRE was achieved by two mechanisms: capillary wicking on super-hydrophilic surfaces, which provided uniform deposition; and Marangoni-driven convection on superhydrophobic D-CuO/Cu, which concentrated analytes into an ultra-small 0.03 mm2 region. The authors demonstrated that, combined with droplet-constraining PET films, the D-CuO/Cu substrate enabled ultrasensitive detection of Cr and Pb, achieving LODs of 0.08 mg L−1 and 0.09 mg L−1 respectively.
4.2 Laser induced fluorescence (LIF)
A two-line LIF thermometry method for spatially resolved temperature diagnostics of a LIP was developed by Beglaryan et al.149 Building on three-level excitation kinetics, the authors adapted the classical two-line LIF approach widely used in combustion diagnostics, to the harsher conditions of LIP, where high electron densities, strong background emission and risks of secondary breakdown typically hinder conventional fluorescence thermometry. Under the experimental conditions, stimulated processes dominated over spontaneous decay by several orders of magnitude, allowing fluorescence saturation within ∼10−13 to 10−12 s, far shorter than the probe-pulse duration. Using spatial scanning of the probe beam across the plasma cross-section, a 2D temperature map was reconstructed with sub-millimeter resolution. The authors demonstrated that the LIP exhibited a central, quasi-uniform core at ∼4000 K surrounded by a hotter peripheral region reaching between 6000 and 8000 K, consistent with shock-wave propagation mechanisms that either heat the surrounding gas directly or transport excited species outward.4.3 Laser atomic absorption spectroscopy (LAAS)
Wala et al.150 conducted a detailed time resolved absorption spectroscopy study of uranium laser-produced plasmas, providing quantitative insight into charge state evolution, excitation/kinetic temperatures and electron densities at late stages of plasma cooling. The authors used a high-resolution tunable Ti:Sapphire probe to measure ten U I and six U II transitions between 763 and 835 nm, tracking their absorption lineshapes from 3.5 to >1000 µs after ablation, in 50 Torr Ar. Voigt-profile fits provided column densities, Doppler temperatures and collisional broadening components, enabling the separation of van der Waals and Stark contributions. It was observed that the U II:U I column density ratio exceeded 14 at 15 µs, and only reached unity near 50 µs, far from Saha equilibrium predictions. This result indicated that the uranium plasma remained strongly ionised for tens of microseconds. Moreover, electron densities extracted from Stark widths decreased from ∼1015 to ∼1013 cm−3 between 4 and 25 µs, revealing plasma properties inaccessible by emission-based measurements. The main broadening mechanisms affecting LAAS (Stark, Doppler and pressure) and their impact on spectral resolution during laser-produced plasma evolution were reviewed by Chenshen et al.151 The authors outlined how shifts in plasma temperature, density and pressure modulated line shapes, which directly influence key analytical outputs such as electron density, temperature determination and isotope selectivity.5 Isotope analysis
5.1 Reviews
A new instrumental development in IRMS was presented by Retzmann et al.152 Instead of using an Ar plasma as an ion source for an MC-ICP-MS they used an N2 MICAP. It was found that levels of ionisation were lower than in an Ar plasma, however, background Ar-derived species such as Ar+, ArH+, ArC+, ArN+, ArO+ and Ar2+ were absent. Low mass oxygen species (O+, OH+ and H2O+) were less intense than in Ar ICP. This study was the first examination of a MICAP ion source combined with a multi-collector system. For this test Sr isotopes were selected as they have no significant influence from either Ar- or N-based interferences. Results indicated that instrumental mass bias was comparable in magnitude and correction strategy between the two plasma types. Furthermore, the precision and accuracy achieved by the MC-MICAP-MS were equivalent, with intermediate precision of 0.710256 ± 0.000009 (2SD, n = 70) for 87Sr/86Sr on NIST SRM 987. Future developments of this interface are predicted to be in the isotopic measurement of elements such as Ca, Fe, K and Se.A review of the developments in tandem ICP-QMS was made by Zhu et al.153 This focussed on the measurement of REE and radionuclides and the strategies for isolating analyte isotopes from their spectral interferences. REE measurements were made with reaction gases including O, N2O and NH3, the latter being effective at separating Lu+ from HfNH3+. Pu isotope determinations were also described including an assessment of the reduction in spectral interferences by the use of O, NH3 and CO2 in the collision-reaction cell. Each of the techniques highlighted in the review were accompanied by a useful table providing information on the reference source, isotope/element of interest, sensitivity, concentration, and the gas species/parameters used in the determinations.
A review of nuclear forensic analysis and interpretation was completed by Li et al.154 This provided a good introduction to the concept of analytical methods in nuclear systems, with coverage of U and Pu isotopic measurement techniques. It includes a detailed recap of Pu isotope measurement by TIMS using carburised filament loading. However, the section does not include any information or data on how the different Pu or U isotope ratios have been used to discriminate between different sources of nuclear material, which is a surprise given the word ‘application’ in the title of the paper. Age determination of nuclear materials was covered with details of the separation and measurement techniques for each radiochronometer. A useful table with the spikes used in analysis, sample quotients, methods, uncertainties and references of each radiochronometer was also provided. Rare earth elements can also be used to characterise where uranium ore samples are from, either by a visualisation of a normalised REE pattern or by using ratios of individual REE concentrations from across the atomic number spectrum. The section on the signatures of REE in the review indicates the unfortunate disconnect between analytical chemistry workers and Earth scientists. Geology and geochemical investigations make extensive use of REEs, and a wealth of techniques have been developed for their separation and measurement, yet this knowledge base is largely passed over in the review. Indeed, the paper cited as an example of U ore discrimination by REE (Balboni et al., 2017; https://doi.org/10.1016/j.apgeochem.2017.10.007) is misrepresented as the Li et al.154 paper. Figure 4 in the article shows an ore/altered ore comparison rather than the chondrite normalised patterns from the original paper. A similar criticism can be levelled at the review's coverage of isotopic ratios, such as 143Nd/144Nd, 87Sr/86Sr and 20xPb/204Pb as ore discriminants.
Matthew155 set out a critical assessment of Pu isotope measurement and precision using total evaporation TIMS analysis, which is the benchmark technique. This study examined the nature of fractionation during the evaporation of Am, Ga, Pu and U from TIMS filaments, and assessed the use of the double isotope ratio technique to estimate the half-life of 241Pu. Overall, the uncertainty of 240Pu/239Pu by the total evaporation technique was found to be in the range ± 0.024 to 0.030% (2SD) for the RM CRM 136, 137, 138 and 126-A.
5.2 Radiogenic isotope ratio analysis
Laser ablation systems coupled with ICP-MS instruments are an effective method of isotope analysis of solid samples. However, isotopic fractionation during ablation and transfer from the material to the spectrometer is difficult to avoid. Weiss et al.156 developed an improved LA method by ablation in water, and used it to collect ablated material for radiogenic isotope ratio analysis of diamonds. This methodology has the advantage of capturing essentially all the ablated material and hence largely avoids ablation fractionation. Water with ablated material was dried, the residue dissolved in HF-HNO3 and put through column chemistry to isolate Nd, Pb and Sr for isotopic determination using TIMS. The Pb isotopes were fractionation corrected using a 204Pb-207Pb double spike. In this process around 5 to 10 mg of diamond were ablated in around 10 to 20 hours, which was enough analyte to produce high precision isotopic data. This technique has the potential to measure inclusions in diamonds and be applied to other minerals where spatial microsampling could constrain isotopic variation.Varga et al.157 used MC-ICP-MS/MS to investigate its potential for measuring 230Th/232Th for geological and nuclear forensic applications. A key issue in measuring the minor 230Th in the presence of a large 232Th signal is the tailing from m/z 232 onto m/z 230 (230Th). Typically, this measurement is best performed on TIMS instruments where tailing and other spectral interferences are reduced. However, it was found that the system used here, which includes a secondary electron multiplier with a dedicated retardation potential quadrupole filter (RPQ), could reduce the tailing onto 230Th to below 1 ppb. This resulted in a 10-fold improvement in the precision of the 230Th/232Th measurement. It was noted that if the mass resolution of the instrument was switched from low resolution (typical for most inorganic isotope ratio determinations) to medium resolution the abundance sensitivity did not change. However, other spectral interferences were reduced with this switch, which may significantly improve analyses where LA systems are coupled to this instrument.
Ariga et al.158 assessed the use of different cone configurations in the MC-ICP-MS measurement of 87Sr/86Sr and ∂88Sr/86Sr. A standard sampling cone paired with a ‘H’ skimmer cone was compared with a Jet sampler and ‘X’ skimmer combination. The sensitivity enhancement of the Jet/X combination is well known, and their study demonstrated a >4 times signal increase over the standard/H arrangement. Results indicated that precise and accurate results (NIST SRM 987 0.710250 ± 0.00004 2sd) were generated for 87Sr/86Sr using the standard/H arrangement at 200 ppb Sr, but similar results were achieved for 50 ppb Sr with the Jet/X arrangement (NIST SRM 987 0.710250 ± 0.00003 2sd). A marked degradation in data precision and accuracy was noted when standard/H was used with a 50 ppb Sr solution. Using the addition of Zr at 100 ppb to correct for Sr instrumental mass bias, and 50 ppb Sr with the Jet/X, resulted in a measurement for ∂88Sr/86Sr of ±0.023. Tests were made with similar solutions doped with Ca to check for the effects of spectral interferences which were found to have no effect on the precision and accuracy of the ratios.
Liu et al.159 developed an alternative method to the potentially explosive Carius tube for sample preparation for Re-Os elemental and isotopic determinations. An issue with Os isotopes is an incomplete equilibration between sample and spike due to variable oxidation states in the system. High-temperature and pressure conditions of the Carius tube can overcome this, but in this study it was tackled by addition of a strong reducing reagent. An optimised reductant of 0.05 mol L−1 NH2OH HCl + 0.01 mol L−1 HBr demonstrated exceptional isotopic homogenisation capability between 188Os-190Os spikes and the Os salts from samples. Analysis by N-TIMS for this reductant method produced results which were comparable to the Carius tube technique. This methodology was expected to produce lower blanks due to fewer reagents used and increased efficiency due to the ambient condition processing.
Israel et al.160 used TIMS to test the precision that could be achieved for Nd isotope analysis using a multicollector array. Using 14 of the 16 available Faraday detectors they constructed a 5 mass-jump measurement sequence which developed dynamic Nd isotope ratios. This produced a significant improvement in precision for the key isotope ratios of 142Nd/144Nd (±3.2 ppm) and 143Nd/144Nd (±2.0 ppm). Such high precision on 143Nd/144Nd is certainly beneficial to geological studies where subtle changes in this radiogenic ratio need to be defined. In the case of 142Nd/144Nd it was considered that it is now possible to detect minute changes that may reflect processes during Solar System formation and Earth's early differentiation events.
5.3 Geological studies
Chen et al.161 examined the potential of a columbite megacryst to become a reference material for U-Pb dating. They used LA-ICP-MS combined with ID TIMS to measure the U-Pb systematics. A date of 908.6 ± 1.4 Ma (2SD) was recommended as the best estimated crystallisation age, and it was concluded that the material was suitable to constrain the age of ore concentrates.Bau et al.162 synthesised two sulfide mineral standards for Pb isotope measurement by LA-MC-ICP-MS. These were a galena and a pyrite produced by a fast hot-press sintering method. Ablation of these samples by ns-LA resulted in matrix effects causing deviation from expected ratios when using silicate standardisation, hence requiring sulfide matrix matching. In contrast, fs-LA did produce precise and accurate results from both silicate and sulfide standardisation. Fragments of these standards are available from the authors on request.
Niki and Hirata163 investigated elemental fractionation between Th and Pb during LA using a high-time-resolution MC-ICP-MS. This system can distinguish signal events from multiple particles, and it was found that 232Th/208Pb ages of monazite varied with particle type. Coarse, Pb depleted particles originated from the molten rim of the ablation pit, whereas the volatile and Pb-rich particles were generated by central vapour plume. This elemental fractionation was the cause of systematically discrepant ages for standard monazites, with accurate ages only being achieved by characterisation of each particle signal event using high-time-resolution.
Eensoo et al.164 investigated the measurement of S isotope ratios in pyrite and pyrrhotite using LA-ICP-MS/MS. Their method used N2O as a primary reaction gas mixed with He in the collision cell to measure S16O+ signals. Optimised conditions resulted in precision of 1 to 1.5‰ and accuracy within ±0.3‰ of reference values for ∂34S in pyrite RMs.
5.4 Stable isotope ratio studies
Kipp et al.165 assessed a method for measuring Se isotopes using MC-ICP-MS. They used CRC-MC-ICP-MS with a He–N2 gas mixture, which was found to effectively remove Ar dimers at m/z 74, 76. 78 and 80 from the mass spectrum. For comparison, the instrument was also used in its high-energy ion path mode, i.e. without the CRC. Combining the low-energy CRC system with a 74Se:77Se double spike produced the best precision and accuracy with rock standard SCo-1 producing δ82/76Se = −0.1 ± 0.24‰ (2SD). Furthermore, with the CRC active, all of the Se isotope ratios, including δ80/78Se with the largest argon dimer isobaric interference, showed systematic mass-dependent fractionation. This demonstrated that high precision could be achieved by incorporation of 80Se into the double spike deconvolution and with analyte levels of between 1 and 5 ng Se. Hence, this CRC method was expected to open the door to the isotopic characterisation of low-selenium reservoirs.Standish et al.166 assessed the potential for B isotope measurements by LA-MC-ICP-MS/MS. Two issues with B isotope analysis by LA are isobaric interferences and variable mass bias related to mass loading of the plasma. Part of the isobaric effect comes from an enhanced and variable baseline around m/z 10, primarily related to scattered quadruply charged Ca and Ar ions, that can result in inaccuracies of more than 20‰ δ11B. The study used a MC-ICP-MS/MS instrument in full transmission mode i.e. with the double Wein filter activated but without a reaction/collision gas. Scans across m/z 10 when ablating carbonate samples in full transmission demonstrated that Faraday cup baselines around this mass were essentially zero, whereas without Wein filtration the baselines were raised to ∼0.15 mV. Mass loading and bias effects were minimised by matching laser parameters between sample and RM, and by minimising the mass of ablated material. Results indicated reproducibility was better than 1‰ with 11B signal intensities >40 mV. A further aspect from this methodology was the use of 12C as a stoichiometric proxy for Ca concentration in CaCO3. If 12C was measured in the same Faraday cup array as 10B and 11B then a reliable B/Ca ratio determination could be made.
A method to measure ∂18O/16O in water using ICP-MS/MS was detailed by Lancaster et al.167 A central part of their method was to use D2 as a reaction gas. This had the effect of generating 16O2D3+ and 18O2D3+ ions (m/z = 22 and 24) which are shifted to higher m/z than the interfering 16OH+, 16OH2+ and 16OH3+ ions. Their method also used the mass filtration quadrupole to eliminate Na and Mg isobaric interferences and the post collision cell quadrupole to isolate the mass of interest from e.g. 16O1H22D+. Another interesting facet of the method was using the collision quadrupole to vary the signal intensity of each oxygen isotope such that they could be measured in the same detector mode. This was possible because the ultimate ∂18O/16O determinations were made using the same conditions for the samples and the bracketing standards. Overall, the combined uncertainties of this method were ±0.7‰ for ∂18O/16O, which, although an order of magnitude higher than reference methods, is still useful for water determinations which have a wide spread of ∂18O/16O. Liu et al.168 also investigated oxygen isotopes using MC-ICP-MS without a reaction cell. The approach used variable mass resolution, oxide formation and cup positioning to minimize interference from Ar and other polyatomic ion species. Atmospheric oxygen was accounted for by blank signal intensity determination using a 3 × 10−9 Ω amplifier and using measurement of oxygen at 16O18O+ and 16O16O+. Contamination was also minimised using a positive shield gas pressure preventing the atmosphere from entering the ICP area.
As there are only three stable Mg isotopes there is no conventional method to correct for mass bias using a double spike. However, the critical mixture double spike technique,185 which relies on an optimised sample-double spike mixture, was applied to accurate Mg isotope measurements by Wang et al.169 and Zhang et al.170 Wang et al.169 used this technique to determine the Mg isotope systematics of halite samples with sub-microgram quantities of Mg. It was found that the critical double spike method could be used with less than 1 µg of Mg and produced ∂26Mg of ± 0.05. This enabled an evaluation of ∂26Mg in an evaporite drill core revealing highly variable ratios likely to reflect dry-wet climatic changes during halite formation.
Germanium has a range of oxidations states in natural materials, and its isotopes are known to show variations, particularly in geological mineral deposits such as sphalerite and other sulfide ores. This is likely the result of kinetic isotope fractionation during ore formation. Wölfer et al.171 examined the use of two different MC-ICP-MS procedures to measure Ge isotopes. They used both hydride generation and, for the first time in Ge analysis, desolvation nebulisation to introduce the samples to the plasma. A 70Ge-73Ge double spike was added to samples to correct for instrumental mass fractionation. Results for solution reference materials, terrestrial basalts and meteoritic samples all produced reproducible ∂74/70Ge to around ± 0.09‰. In context, this level of precision was set within an observed variation of ∼2‰ and hence indicates that these techniques are appropriate to investigate a range of geochemical and cosmochemical processes.
Jiang et al.172 investigated measurement of Ca isotopes in Sr-rich carbonates by LA-MC-ICP-MS. Sr is a problem in Ca isotope measurement because of the isobaric interference of double charged Sr2+ in the Ca region. Their study used a rigorous procedure to correct for isobaric Sr. It was found that mass fractionation of Sr2+ ions was different from Sr+, hence it was better to utilise a modified Ca fractionation coefficient for the correction. Using their correction, it was found that both ∂44/42Ca and ∂43/42Ca could be determined to within 0.1 to 0.2‰ (2SD) in materials with Sr/Ca of up to ∼0.06.
In the review section above, the study of Retzmann et al.152 was featured which used a novel N2 plasma in a MC-MICAP-MS instrument. This mass spectrometer was also trialled by Retzmann and Wieser 2025 (ref. 173) to measure Ca isotopes. Ca measurement using an Ar plasma is hampered by the direct isobaric overlap between Ca and Ar. Since Ar is eliminated in the N2 plasma, Ca isotopes were measured directly at low mass resolution enabling an intermediate precision of <±0.1‰ (2SD) on ∂44/40Ca.
Li et al.174 provided a benchmark study which examined the measurement of Ga isotope composition in various reference materials and geological samples. Ga only has two stable isotopes, 69Ga and 71Gahence cannot be measured by double spike methodology. Various elemental and oxide ion interferences (e.g. Al, Cu, Cr, Al, Zn) are present on the Ga isotopes. Their effects and correction procedures were tested by doping at various levels up to interfering element/Ga ∼1. The highlighted elements were found to be correctable at these concentrations, but more significant correction was required for Ce2+ and Ba2+. It was found that the only way to reduce the measurement errors induced by Ce and Ba was to separate these elements chromatographically prior to mass spectrometry. The study contributed tables and figures which document and compare the new data with previous Ga isotope determinations on RMs. As such, it provided clear evidence for terrestrial variance in Ga isotopic compositions.
Unlike most of the other rare earth elements, Eu exists in two valence states in Earth environments: EuII and EuIII, giving it the potential for use as a redox proxy. However, it has only two isotopes and hence the double spike cannot be used to produce high precision determinations of their ratios. Wu et al.175 investigated the correction of Eu mass bias by internal normalisation to Nd isotopes, with sample-standard bracketing as a secondary strategy. Their study trialled a number of Nd isotope pairs to test the precision generated for ∂153/151Eu and the effects of Sm interference. It was found that 146Nd/144Nd normalisation produced the most consistent ∂153/151Eu across a range of Sm concentrations and resulted in ∂153/151Eu precision of ± 0.03‰ (2SD). Measurement of geological standards using this technique recognised that basaltic standards BHVO-2 and BCR-2 have ∂153/151Eu ∼0‰ relative to the syenite standard SY-3 which has ∂153/151Eu −0.14‰. This demonstrated the potential for Eu to be used in the geochemical investigations related to magmatic and hydrothermal processes.
5.5 Nuclear forensics
The isotope ratios of 242Am/241Am and 243Am/241Am can trace the origin and history of nuclear-related activities such as weapons testing, reprocessing nuclear fuels, nuclear waste disposal and nuclear accidents. Wang et al.176 assessed the use of MC-ICP-MS for accurate determination of Am isotope ratios and verified the results by comparison with the reference method of total evaporation TIMS. Mass fractionation was corrected using 235U/238U which was found to have a similar β-factor to Pu and Am. The use of a desolvation device meant that about 1 ng of 241Am was consumed in the MC-ICP-MS to determine 242Am/241Am ratios of ∼10−5 and 243Am/241Am ratios 10−4 with RSD of 0.2%.Han et al.177 determined Pu isotope ratios using an automated microfluidic separation system coupled with MC-ICP-MS. This type of in-line separation system is used where isolation of Pu from interfering elements such as U is required without separate time-intensive chromatography. AG® 1-X8 resin was use in the separation loop to minimize U interference. However, results indicated that the level of separation was not sufficient to eliminate U from the mass spectrum, and as such measurement of 238Pu/239Pu was compromised by isobaric 238U. Sharma et al.178 contributed a review of Pu separation and measurement. This detailed the purification by ion exchange and extraction chromatography, HPLC, liquid–liquid extraction and cloud point extraction. Measurement techniques, including liquid scintillation, alpha spectrometry, SIMS, RIMS, TIMS ICP-MS and AMS were reviewed. A systematic review of the precision of Pu isotope ratios generated by each of these techniques was not provided in their study.
Peng et al.179 examined the potential for ICP-MS/MS collision gasses to enable the measurement of ultra low-level Pu isotopes including 238Pu. U needed to be removed from the Pu measurement solution via a single TK200 resin column, but remained at ppt levels. A CRC was used, with C2H4 and He, and it was found that this reduced the signal of 238U to less than 1 count per second with up to 50 ppt U present. This enabled accurate determination of 238Pu and hence ratios with the heavier Pu isotopes. It was hoped that this will provide a useful source of data for nuclear fingerprinting and source identification.
Zhang et al.180 developed a method for correcting the 238U isobaric interference on 238Pu during total evaporation TIMS analysis. This methodology utilised the difference in thermal evaporation behaviour between U and Pu and incorporated a U interference indicator into the Pu solution. This permitted a linear model of m/z 238 intensity relative to 238Pu/239Pu and 238U/235U to be developed. It was demonstrated that precision on 238Pu/239Pu was 0.2% (1SD) with picogram levels of Pu.
Uranium isotopes were analysed by Varga and Wallenius181 using LA-MC-ICP-MS The objective was to investigate how this new technique could assist in nuclear forensic measurements of confiscated material to determine its potential threat, intended use and origin. Laser ablation is a key method because it has minimal sample preparation and only consumes a small amount of material. It does however involve a variable and extensive matrix which can potentially interfere with the U or Pu isotopic analysis. In this study no collision gas was used, but the combination of pre-filtration by the Wien filters and the use of medium resolution produced precise and accurate results. The medium mass resolution (∼7100) gave slightly worse precision than the low resolution (∼2000) but results from both (<0.053% 2sd) were an order of magnitude better than produced by single collector ICP-MS. Materials used in the experiments were analysed spatially and found to be heterogeneous in their U isotopes. This indicated that this strategy could be extremely useful in characterising solid samples without alteration of the physical evidence.
Stanberry et al.182 utilised ICP-TOF-MS to analyse U isotopes in single uranium oxide particles. Time of flight systems are particularly suited to SP analysis because they can be used to simultaneously measure all elements across the periodic table. They injected suspensions of micron-scale uranium oxide particles into the ICP-TOF-MS and found that differences of 235U/238U could be accurately detected in populations of depleted and natural compositions with ratio differences of <10%. It was expected that this would provide a rapid and accurate throughput method for particulate analysis that could replace existing TIMS and SIMS techniques.
A technique used in nuclear forensic examination is to determine the age of uranium enrichment from 230Th/234U. This relies on the purification of U from Th at the time of manufacture, such that 230Th is negligible. Subsequent decay of 234U to 230Th with a half-life of 2.46 × 105 years permits an age to be calculated from measured 230Th/234U. Zirakparvar et al.183 considered the measurement of 230Th and 234U using MC-ICP-MS with 1013Ω and 1011Ω amplifiers respectively in a Faraday array. Other U isotopes were measured in suitably positioned high-mass Faradays for the simultaneous determination of U isotope ratios. Analyses were made of NBS CRM U630 and U850 which have specified purification dates of June 1989 and December 1957 respectively. 234U signal level was found to require a signal of >0.5 V to achieve a within run precision of ∼<2% on 230Th/234U. Measured ages broadly agreed with the purification ages which indicated that this all-Faraday method has the potential to obtain 230Th/234U ages.
Another method of nuclear fuel characterisation for provenance determination in forensic investigations is by tagging fuel with perturbed isotope compositions. In particular, an Fe double spike is used as a taggant which has the properties of remaining identifiable regardless of the level of dilution with natural Fe isotope compositions. To test what level of taggant could be recognised in fuel cycles, Bowden et al.184 investigated the precision and accuracy with which Fe isotopes could be measured by ICP-MS and MC-ICP-MS. They found both instruments were capable of distinguishing addition of a 75:25 57Fe:54Fe taggant down to 104× dilution with natural Fe using ∂57Fe. However, only MC-ICP-MS could identify a 25:75 57Fe:54Fe at 104× dilution. Due the large uncertainties measuring 54Fe both instruments could only resolve down to 103 using the ∂54Fe parameter. The key message was that the more widely available and lower cost ICP-MS instrument could be used for taggant characterisation.
6 Abbreviations
| AAS | atomic absorption spectrometry |
| AC | alternating current |
| AES | atomic emission spectrometry |
| AMS | accelerator mass spectrometry |
| CC | collision cell |
| CCD | charge coupled detector |
| CCP | capacitively coupled plasma |
| CID | charge injection detector |
| CMOS | complementary metal oxide semiconductor |
| CRC | collision reaction cell |
| CRM | certified reference material |
| CVG | chemical vapour generation |
| DBD | dielectric barrier discharge |
| DC | direct current |
| DCP | direct current plasma |
| DNA | deoxyribonucleic acid |
| DOTA | 1,4,7,10-tetraazacyclododecane-1.4.7.10-tetraacetic acid |
| EI | electron ionisation |
| EIE | easily ionizable element |
| EPMA | electron probe microanalysis |
| ESI | electrospray ionisation |
| ETA | electrothermal atomisation |
| ETAAS | electrothermal atomic absorption spectrometry |
| ETAES | electrothermal atomic emission spectrometry |
| ETV | electrothermal vaporisation |
| FAAS | flame atomic absorption spectrometry |
| FIA | flow injection analysis |
| FWHM | full width at half maximum |
| GD | glow discharge |
| GF | graphite furnace |
| HCL | hollow cathode lamp |
| HG | hydride generation |
| ICP | inductively coupled plasma |
| ID | isotope dilution |
| IDA | isotope dilution analysis |
| IDMS | isotope dilution mass spectrometry |
| IE | ionisation energy |
| IMF | instrumental mass fractionation |
| IR | infra-red |
| IRA | isotope ratio analysis |
| IRMS | isotope ratio mass spectrometry |
| ISO | International Organization for Standardisation |
| IT | ion trap |
| LA | laser ablation |
| LAAS | laser atomic absorption spectroscopy |
| LAMIS | laser ablation molecular isotopic spectrometry |
| LDR | linear dynamic range |
| LIBS | laser induced breakdown spectroscopy |
| LIF | laser induced fluorescence |
| LIMS | laser ionisation mass spectrometry |
| LIP | laser induced plasma |
| LOD | limit of detection |
| LOV | lab on a valve |
| LTE | local thermodynamic equilibrium |
| MALDI | matrix-assisted laser desorption ionisation |
| MAS | molecular absorption spectrometry |
| MB | magnetic bead |
| MC | multi-collector |
| MIP | microwave induced plasma |
| MPT | microwave plasma torch |
| MS | mass spectrometry |
| MS/MS | tandem mass spectrometry |
| NAZ | normal analytical zone |
| Nd:YAG | neodymium doped: yttrium aluminium garnet |
| NE | nanoparticle enhanced |
| ne | electron number density |
| NIST | National Institute of Standards and Technology |
| NP | nanoparticle |
| OES | optical emission spectroscopy |
| PCA | principal components analysis |
| PCB | printed circuit board |
| PCR | principal components regression |
| PEEK | polyetheretherketone |
| PET | polyethyleneterephthalate |
| PMT | photomultiplier |
| PMVG | plasma mediated vapour generation |
| PN | pneumatic nebuliser |
| PVG | photochemical vapour generation |
| QMS | quadrupole mass spectrometry |
| REE | rare earth element |
| RF | radiofrequency |
| RIMS | resonance ionisation mass spectrometry |
| RM | reference material |
| RNA | ribonucleic acid |
| RSD | relative standard deviation |
| SAGD | solution anode glow discharge |
| SBR | signal to background ratio |
| SC | solution cathode |
| SCGD | solution cathode glow discharge |
| SD | standard deviation |
| SEM | scanning electron microscopy |
| SI | Système International d’Unités |
| SIMS | secondary ion mass spectrometry |
| SNR | signal to noise ratio |
| SP | single particle |
| SPE | solid phase extraction |
| SRM | standard reference material |
| TE | transport efficiency |
| Te | electron temperature |
| TIMS | thermal ionisation mass spectrometry |
| TM | transition metal |
| TOF | time-of-flight |
| TOFMS | time-of-flight mass spectrometry |
| UV | ultra-violet |
| VG | vapour generation |
| XRD | X-ray diffraction |
Conflicts of interest
There are no conflicts to declare.Data availability
There is no additional data associated with this article.Acknowledgements
J. Pisonero acknowledges funding from project MCINN-24-PID2023-149335NB-I00 funded by the Spanish Ministerio de Ciencia e Innovación and the Agencia Estatal de Investigación.References
- E. H. Evans, J. Pisonero, C. M. M. Smith and R. N. Taylor, J. Anal. At. Spectrom., 2025, 40(5), 1136–1157 RSC.
- M. Patriarca, N. Barlow, A. Cross, S. Hill, D. Milde and J. Tyson, J. Anal. At. Spectrom., 2025, 40(3), 541–664 RSC.
- A. Agarwal, E. Bolea-Fernandez, R. Clough, A. Fisher, B. Gibson and S. Hill, J. Anal. At. Spectrom., 2025, 40(11), 2982–3022 RSC.
- W. R. L. Cairns, O. T. Butler, O. Cavoura, C. M. Davidson, J.-L. Todolí-Torró and M. v. d. Au, J. Anal. At. Spectrom., 2025, 40(1), 11–69 RSC.
- R. Clough, C. F. Harrington, S. J. Hill, Y. Madrid and J. F. Tyson, J. Anal. At. Spectrom., 2025, 40(7), 1615–1644 RSC.
- C. Vanhoof, A. Cross, U. E. A. Fittschen and L. Vincze, J. Anal. At. Spectrom., 2025, 40(9), 2275–2289 RSC.
- M. Patriarca, N. Barlow, A. Cross, S. Hill, A. Robson and J. Tyson, J. Anal. At. Spectrom., 2024, 39(3), 624–698 RSC.
- R. Clough, C. F. Harrington, S. J. Hill, Y. Madrid and J. F. Tyson, J. Anal. At. Spectrom., 2024, 39, 1629–1664 RSC.
- C. Vanhoof, J. R. Bacon, U. E. A. Fittschen and L. Vincze, J. Anal. At. Spectrom., 2024, 39, 2152–2164 RSC.
- E. Bolea-Fernandez, R. Clough, A. Fisher, B. Gibson and B. Russell, J. Anal. At. Spectrom., 2024, 39(11), 2617–2693 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., 2024, 39(1), 11–65 RSC.
- L. Cheng, M. He, B. Chen and B. Hu, Anal. Chem., 2025, 97(37), 20389–20398 CrossRef CAS PubMed.
- C. H. Kuo and C. K. Su, Anal. Chim. Acta, 2025, 1336, 343536 CrossRef CAS PubMed.
- J. Y. Jian and C. K. Su, Talanta, 2026, 298(Pt A), 128955 CrossRef CAS.
- A. Figuerola, M. Marí, L. Ferrer and E. Palacio, Microchem. J., 2025, 216, 114653 CrossRef CAS.
- S. Y. Han, B. J. Treves Brown, M. A. Higginson, P. Kaye, C. A. Sharrad and S. L. Heath, J. Chromatogr. A, 2025, 1742, 465646 CrossRef CAS PubMed.
- W. B. Jones and B. T. Jones, Microchem. J., 2025, 208, 112423 CrossRef CAS.
- M. Jiang, C. Zhang, Y. Chen, M. Sun, R. Liu and Y. Lv, TrAC, Trends Anal. Chem., 2025, 191, 118317 CrossRef CAS.
- Z. Fang, X. Zhao, R. Hou, Y. Zhao, L. Wang, L. Yi, L. Yang, X. Yan and Q. Wang, Anal. Chem., 2025, 97(23), 12125–12132 CrossRef PubMed.
- J. S. Stanberry, R. Meeks, R. W. Peterson, H. B. Andrews, A. N. Bible, S. Szakas, B. C. Sanders and B. T. Manard, J. Anal.At. Spectrom., 2025, 40(4), 961–966 RSC.
- T. He, D. Li, Y. Qin, Y. Xiang, Y. Tang, J. Tian and Y. Tang, Talanta, 2025, 293, 128069 CrossRef CAS PubMed.
- X. W. Zhang, Y. Y. Li, Z. W. Wu, J. H. Liu and B. Lin, Talanta, 2026, 298(Pt A), 128879 CrossRef CAS PubMed.
- Y. Wu, Y. Wang, L. Yan, K. Huang, K. Chang and P. Chen, Small Methods, 2025, 9(8), e2500314 CrossRef PubMed.
- Y. Xu, B. Chen, M. He, G. Yuan and B. Hu, Anal. Chem., 2025, 97(1), 811–817 CrossRef CAS PubMed.
- T. Xu, X. W. Tian, Y. W. Liu, Y. Y. Guo, L. G. Hu, Y. G. Yin, Q. H. Zhang, Y. Cai and G. B. Jiang, Chin. J. Anal. Chem., 2024, 52(10), 1403–1412 CAS.
- L. Cheng, X. Ou, M. He, B. Chen and B. Hu, Anal. Chem., 2025, 97(6), 3504–3514 CrossRef CAS PubMed.
- N. H. Erfurth, M. M. Jones, C. M. Pincock and E. M. Hoadley, J. Anal. At. Spectrom., 2025, 40(8), 2243–2250 RSC.
- G. Kajner, Á. Bélteki, M. Cseh, Z. Geretovszky, I. Szenti, Á. Kukovecz, T. Ajtai and G. Galbács, Spectrochim. Acta, Part B, 2025, 223, 107078 CrossRef CAS.
- F. Sandro, H. Bodo and G. Detlef, J. Anal. At. Spectrom., 2025, 40(1), 276–285 RSC.
- D. Bartczak, S. C. Nunez, A. Kubicka, D. Ojeda, A. S. Cachero, S. Cowen, S. Ellison, G. Holcombe and H. Goenaga-Infante, Anal. Bioanal. Chem., 2025, 417(12), 2655–2667 CrossRef CAS PubMed.
- M. Hernández-Postigo, A. Sanchez-Cachero, M. Jimenez-Moreno and R. C. Rodriguez Martin-Doimeadios, Mikrochim. Acta, 2024, 192(1), 28 CrossRef PubMed.
- A. Bazo, L. Lopez-Villellas, M. Mataloni, E. Bolea-Fernandez, A. Rua-Ibarz, M. Grotti, M. Aramendia and M. Resano, Anal. Chim. Acta, 2025, 1378, 344694 CrossRef CAS PubMed.
- D. Wang, J. Zhang, C. Fan, X. Li, L. Liu, X. Yan, Y. Li, B. He, Y. Yin, L. Hu and G. Jiang, Anal. Chem., 2025, 97(3), 1792–1798 CrossRef CAS PubMed.
- A. Suarez Priede, M. Corte-Rodriguez, H. Godde, M. Montes Bayon and J. Bettmer, Talanta, 2025, 295, 128372 CrossRef CAS PubMed.
- M. Lomax-Vogt, L. M. Carter, J. Wielinski, S. Kutuzov, G. V. Lowry, R. Sullivan, P. Gabrielli and J. W. Olesik, J. Anal. At. Spectrom., 2025, 40(3), 848–859 RSC.
- J. S. Stanberry, H. B. Andrews, C. V. Thompson, B. W. Ticknor and B. T. Manard, Anal. Chem., 2025, 97(3), 1688–1694 CrossRef CAS.
- O. N. Grebneva-Balyuk, M. S. Kiseleva and I. V. Kubrakova, J. Anal. Chem., 2025, 80(4), 608–622 CrossRef CAS.
- G. H. Xing, L. H. Liu, J. H. Zhang, B. He, Y. G. Yin, L. G. Hu and G. B. Jiang, Chin. J. Anal. Chem., 2024, 52(10), 1413–1423 CAS.
- L. Paton, S. Kiesel, G. Steinhoefel, M. Elinkmann, T. T. Moro, R. Gonzalez de Vega, P. Bohleber and D. Clases, J. Anal. At. Spectrom., 2025, 40(10), 2897–2908 RSC.
- A. Sikora, D. Ojeda, D. Bartczak and H. Goenaga-Infante, Anal. Chem., 2025, 97(47), 26054–26061 CrossRef CAS PubMed.
- W. Khan, P. Dvorak, M. Mrkvickova, M. Svoboda and J. Kratzer, Anal. Chim. Acta, 2025, 1347, 343787 CrossRef CAS PubMed.
- G. S. Coelho Junior, J. Dedina and J. Kratzer, Anal. Chim. Acta, 2025, 1372, 344444 CrossRef CAS PubMed.
- R. E. Sturgeon, E. Pagliano, G. S. Lopes, R. S. A. Neto and J. K. S. Brito, J. Anal. At. Spectrom., 2025, 40(1), 70–97 RSC.
- T. Y. Guselnikova, At. Spectrosc., 2025, 46(01), 24–32 CrossRef CAS.
- A. Körber, I. G. M. Anthony and R. M. A. Heeren, Anal. Chem., 2025, 97(29), 15517–15549 CrossRef PubMed.
- S. A. Stopka, C. Bodineau, G. Baquer, J. R. Lin, M. A. Hossain, C. J. Martinez, M. S. Regan, D. Ruiz, S. M. Fletcher, O. Pourquie, S. Lutsenko, C. N. Payne, J. N. Agar, R. Mazitschek, N. J. McDannold, P. K. Sorger, S. Santagata and N. Y. R. Agar, Anal. Chem., 2025, 97(45), 25334–25345 CrossRef CAS PubMed.
- P.-L. He, X.-L. Huang, Y.-Q. Zhang, W.-H. Lu, Y.-Z. Liu and Y. Yu, J. Anal. At. Spectrom., 2025, 40(11), 3221–3235 RSC.
- C. Tu, X.-Y. Gao, O. G. Safonov, T.-T. Xiao, V. O. Yapaskurt and W.-C. Li, J. Anal. At. Spectrom., 2025, 40(11), 3150–3160 RSC.
- A. T. Riche, P. McSwiggen, K. Harvey, S. Bowden, R. A. Bergin, S. M. Scott, B. J. Foley and K. M. Samperton, Microsc. Microanal., 2025, 31(2), ozaf022 CrossRef CAS PubMed.
- Y. Wang, D. Luo, Q. Zhang, Z. Lin, Z. Xu and W. Hang, Anal. Chem., 2025, 97(18), 10085–10093 CrossRef CAS PubMed.
- S. Dong, Y. Chen, J.-F. Gao, X. Bi and R. Hu, J. Anal. At. Spectrom., 2025, 40(11), 3294–3305 RSC.
- M. He, W. Deng, Z. Cui, Q. Yang, Y. Zhang and G. Wei, J. Anal. At. Spectrom., 2025, 40(9), 2290–2295 RSC.
- Y. A. Komarov and V. A. Stebelkov, J. Anal. Chem., 2024, 79(13), 1956–1961 CrossRef CAS.
- Y. A. Komarov and V. A. Stebelkov, J. Anal. Chem., 2024, 79(13), 1918–1922 CrossRef CAS.
- A. Bengtson, D. Malmstrom, R. Quardokus and J. Russell, Appl. Spectrosc., 2025, 79(10), 1549–1558 CrossRef CAS PubMed.
- J. Yu, X. Li, K. Wang, Y. Zhou, X. Fang, C. Zhang, C. Liang and Q. Lu, Microchem. J., 2025, 216, 114727 CrossRef CAS.
- K. Kowalczyk, K. Greda and P. Pohl, Talanta, 2026, 298(Pt A), 128902 CrossRef CAS PubMed.
- J. Wang, W. Li, P. Zheng, B. Li, B. Zhang, L. Guo, H. Tian and D. Dong, J. Anal. At. Spectrom., 2025, 40(2), 346–353 RSC.
- N. V. Nedić, N. V. Ivanović, I. R. Videnović, D. Spasojević and N. Konjević, J. Anal. At. Spectrom., 2025, 40(6), 1591–1601 RSC.
- R. C. Sharma, B. K. Das, G. Sharma, V. K. Saraswat and S. N. Thakur, Spectrosc. Lett., 2024, 58(4), 319–327 CrossRef.
- P. Zheng, J. Zhou, J. Wang, J. Liu, W. Li, B. Li, L. Guo, H. Tian and D. Dong, J. Anal. At. Spectrom., 2025, 40(5), 1284–1296 RSC.
- B. Zhang, P. Zheng, J. Wang, X. Luo, J. Li, J. Xiang, L. Guo, H. Tian and D. Dong, J. Anal. At. Spectrom., 2026, 41(1), 271–281 RSC.
- P. Zheng, J. Xiang, J. Wang, B. Li, G. Chen, X. Luo, B. Zhang and T. Hu, Spectrochim. Acta, Part B, 2025, 233, 107291 CrossRef CAS.
- Y. Zhang, J. Orejas, M. d. M. González del Campo, J. Pisonero and N. Bordel, Spectrochim. Acta, Part B, 2025, 229, 107216 CrossRef CAS.
- D. A. Lang, J. V. Goodwin, C. J. Stouffer, B. T. Manard and R. K. Marcus, J. Anal. At. Spectrom., 2025, 40(10), 2625–2632 RSC.
- D. Bryars, M. Jahan, K. Hahn, A. Jugan, A. Leong, A. Williams and A. Bataller, Appl. Spectrosc., 2026, 80(1), 71–82 CrossRef CAS PubMed.
- H. Yuan, S.-Q. Li, J.-P. Liang, Z.-L. Cui, D.-Z. Yang and R. S. Rawat, J. Anal. At. Spectrom., 2026, 41(1), 252–260 RSC.
- A. Zhirkov and V. Yagov, Spectrochim. Acta, Part B, 2026, 235, 107382 CrossRef CAS.
- A. Bazo, E. Bolea-Fernandez, K. Billimoria, A. Rua-Ibarz, M. Aramendía, P. Menero-Valdés, J. Morley, S. Neves, A. Sánchez-Cachero, H. Goenaga-Infante and M. Resano, J. Anal. At. Spectrom., 2025, 40(10), 2673–2681 RSC.
- V. Boger, P. Pirkwieser, N. Orth, M. Koehler and V. Somoza, Anal. Chim. Acta, 2025, 1355, 343999 CrossRef CAS PubMed.
- L. Brunnbauer, L. Kronlachner, E. Foisner and A. Limbeck, J. Anal. At. Spectrom., 2025, 40(3), 753–761 RSC.
- T. Schoberl, M. Bachmann and D. Gunther, J. Anal. At. Spectrom., 2025, 40(9), 2438–2446 RSC.
- D. Yang, K. Tanilas, O. Jarvik and A. Konist, Talanta, 2026, 298(Pt B), 128951 CrossRef CAS PubMed.
- M. Holá, J. Ondráček, M. Bahelková and T. Vaculovič, Appl. Spectrosc. Rev., 2025, 1–42 CrossRef.
- D. Käser, T. Van Acker, J. Koch, B. Hattendorf and D. Günther, J. Anal. At. Spectrom., 2025, 40(9), 2408–2417 RSC.
- Y. Ke, H. Hu, J. Zhou, H. Li, Y. Zhang, C. Tu, H. Wen and Y. Sun, Spectrochim. Acta, Part B, 2025, 224, 107091 CrossRef CAS.
- C. A. Richards, M. A. Turner and A. J. Managh, J. Anal. At. Spectrom., 2025, 40(7), 1726–1732 RSC.
- P. Modlitbová, L. Brunnbauer, G. Kalcikova, A. Fazlic, A. Limbeck, P. Porizka and J. Kaiser, J. Anal. At. Spectrom., 2025, 40(11), 3044–3062 RSC.
- B. T. Manard, S. E. Szakas, J. S. Stanberry, B. W. Ticknor, L. O'Brien, M. Boris, J. T. Hewitt, P. Cable-Dunlap and H. B. Andrews, J. Anal. At. Spectrom., 2025, 40(5), 1241–1248 RSC.
- Y. Wang, S. J. Kienast and D. Beauchemin, Anal. Chem., 2025, 97(35), 19168–19176 CrossRef CAS PubMed.
- G. C. Chan, G. M. Hieftje, N. Omenetto, O. Axner, A. Bengtson, N. H. Bings, M. W. Blades, A. Bogaerts, M. A. Bolshov, J. A. C. Broekaert, W. Chan, J. M. Costa-Fernandez, S. R. Crouch, A. De Giacomo, A. D'Ulivo, C. Engelhard, H. Falk, P. B. Farnsworth, S. Florek, G. Gamez, I. B. Gornushkin, D. Gunther, D. W. Hahn, W. Hang, V. Hoffmann, N. Jakubowski, V. Karanassios, D. W. Koppenaal, R. K. Marcus, R. Noll, J. W. Olesik, V. Palleschi, U. Panne, J. Pisonero, S. J. Ray, M. Resano, R. E. Russo, A. Scheeline, B. W. Smith, R. E. Sturgeon, J. L. Todoli, E. Tognoni, F. Vanhaecke, M. R. Webb, J. D. Winefordner, L. Yang, J. Yu and Z. Zhang, Appl. Spectrosc., 2025, 79(4), 481–735 CrossRef CAS PubMed.
- K. Li, W. Zeng, C. Li, X. Jiang and X. Hou, Anal. Chem., 2025, 97(10), 5439–5444 CrossRef CAS PubMed.
- W. Zeng, T. Lin, Q. Tang, X. Jiang and X. Hou, J. Anal. At. Spectrom., 2025, 40(7), 1652–1657 RSC.
- H. Shen, J. Y. Cai, J. H. Wang, Y. L. Yu and S. Liu, Talanta, 2025, 286, 127540 CrossRef CAS PubMed.
- Z. Liu, J. Zhang and C. Zheng, At. Spectrosc., 2025, 46(01), 1–9 CrossRef CAS.
- J. Y. Cai, J. M. Xu, H. Shen, J. H. Wang and Y. L. Yu, Talanta, 2025, 285, 127373 CrossRef CAS PubMed.
- B. J. Farcy, J. Graham, M. Raith, R. Arevalo Jr, M. Taghioskoui, S. Budinoff, A. McAdam, J. Lee, R. M. Danell, D. A. Kaplan, C. Gundersen and W. F. McDonough, Rapid Commun. Mass Spectrom., 2026, 40(3), e10152 CrossRef CAS PubMed.
- A. Raab, H. Badiei and J. Feldmann, J. Anal. At. Spectrom., 2025, 40(7), 1689–1699 RSC.
- Y. Guo, T. Shi, D. Wu, L. Wang, H. Guo, W. Liu and C. Leng, J. Anal. At. Spectrom., 2025, 40(8), 1946–1953 RSC.
- H. H. Khoo, A. Kubota, T. Matsukawa and T. Hirata, J. Anal. At. Spectrom., 2025, 40(8), 2016–2025 RSC.
- X.-S. Wang, Y. Wang, C.-X. Wu, Y. Shu, X. Wei, M.-L. Chen and J.-H. Wang, Microchem. J., 2025, 219, 115846 CrossRef CAS.
- L. Han, L. Yin, Y. Sun, J. Chen, Y. Xu, Y. Zhou and L. Wang, Appl. Spectrosc. Rev., 2024, 60(5), 404–430 CrossRef.
- L. Han, L. Yin, Y. Zhou, Y. Sun, J. Chen, Y. Xu and L. Wang, Spectrochim. Acta, Part B, 2025, 233, 107288 CrossRef CAS.
- G. C. Y. Chan, Spectrochim. Acta, Part B, 2025, 229, 107194 CrossRef CAS.
- G. C. Y. Chan, Spectrochim. Acta, Part B, 2025, 229, 107195 CrossRef CAS.
- C. Tao, H. J. Zhang, C. S. Li, L. P. He, Z. Y. Ma, B. Chen and R. Zhang, Chin. J. Anal. Chem., 2025, 53(2), 187–194 CAS.
- S. Pommé and S. F. Boulyga, J. Anal. At. Spectrom., 2025, 40(8), 2073–2082 RSC.
- S. Pommé, J. Anal. At. Spectrom., 2025, 40(10), 2937–2944 RSC.
- J. Manrique, D. M. Díaz Pace, J. A. Aguilera and C. Aragón, Spectrochim. Acta, Part B, 2025, 234, 107337 CrossRef CAS.
- N. V. Nedić, D. Dojić, M. Skočić and S. Bukvić, Spectrochim. Acta, Part B, 2025, 233, 107321 CrossRef.
- L. Gosse, A. Favre, A. Bultel, V. Morel, S. Djurović, N. Simić and L. Gavanski, Spectrochim. Acta, Part B, 2025, 230, 107222 CrossRef CAS.
- P. R. Sen Sarma, M. T. Belmonte, S. Llorente and S. Mar, Spectrochim. Acta, Part B, 2026, 235, 107342 CrossRef CAS.
- S. Kodangil, M. Tanaka, D. Kato, G. Gaigalas, H. Tanuma and N. Nakamura, J. Quant. Spectrosc. Radiat. Transfer, 2026, 348, 109692 CrossRef CAS.
- D. Dojić, N. V. Nedić, M. Skočić and S. Bukvić, Spectrochim. Acta, Part B, 2025, 224, 107114 CrossRef.
- M. T. Belmonte, S. Mar, S. Djurović, J. A. Menéndez and L. Gavanski, Spectrochim. Acta, Part B, 2025, 229, 107190 CrossRef CAS.
- D. Antonov, E. Silkis and V. Kolotov, Spectrochim. Acta, Part B, 2025, 229, 107189 CrossRef CAS.
- X. Han, Y. Su, Z. Li, Y. Wang, W. Wang, R. Xi, S. Zhou, G. Wei, S. Fang, Y. Wang, J. Xu and X. Lan, J. Anal. At. Spectrom., 2025, 40(8), 1916–1928 RSC.
- A. A. Pupyshev, P. V. Kel’, M. Y. Burylin, A. G. Abakumov, P. G. Abakumov, M. A. Maltcev, I. V. Morozov and E. L. Osina, J. Anal. Chem., 2025, 80(5), 864–876 CrossRef CAS.
- Y. B. Zhu, Y. Shikamori, K. Nakano and K. Yamaoka, At. Spectrosc., 2025, 46(1), 10–16 CAS.
- D. J. Butcher, Appl. Spectrosc. Rev., 2024, 60(5), 431–449 CrossRef.
- S. Imai, C. Sasada and Y. Yamamoto, Anal. Sci., 2025, 41(9), 1491–1499 CrossRef CAS PubMed.
- L. Duponchel, R. Rocha de Oliveira and V. Motto-Ros, Anal. Chem., 2025, 97(13), 6956–6961 CrossRef CAS PubMed.
- H. V. Carneiro, C. P. Celani and K. S. Booksh, J. Chemom., 2025, 39(7), e70048 CrossRef CAS.
- F. Cernatic, L. Brunnbauer, K. Mervic, J. Willner, A. Limbeck and M. Sala, Anal. Chem., 2025, 97(37), 20108–20116 CrossRef CAS PubMed.
- V. Palleschi, S. Legnaioli, F. Poggialini, F. O. Bredice, I. A. Urbina, N. Lellouche and S. Messaoud Aberkane, Nature Reviews Methods Primers, 2025, 5(1) DOI:10.1038/s43586-025-00388-w.
- S. Zhao, Y. Zhao, Y. Dai, Z. Liu, Z. Hou, X. Gao and Z. Wang, J. Anal. At. Spectrom., 2025, 40(3), 665–678 RSC.
- A. Safi and N. Melikechi, J. Anal. At. Spectrom., 2025, 40(11), 3031–3043 RSC.
- M. Rizwan, M. S. Afgan, S. Saleem, K. Kou, Z. Hou and Z. Wang, Spectrochim. Acta, Part B, 2025, 227, 107168 CrossRef CAS.
- M. Khalaji, S. H. Tavassoli and S. Karimi, Spectrochim. Acta, Part B, 2026, 235, 107358 CrossRef CAS.
- Z. Li, H. Wei, G. Gao, Z. Deng, S. Sun, Z. Liu, B. Ding, B. Hu and J. Shen, J. Anal. At. Spectrom., 2025, 40(7), 1711–1717 RSC.
- X. Zhang, X. Li, T. Ren, Y. Yuan and Z. Guo, J. Anal. At. Spectrom., 2025, 40(8), 2083–2094 RSC.
- M. A. Shevchenko, A. N. Maresev, M. O. Astafurov, N. V. Tcherniega, S. O. Klimonsky, S. G. Dorofeev, S. F. Umanskaya and V. V. Voronova, Spectrochim. Acta, Part B, 2025, 226, 107159 CrossRef CAS.
- V. N. Lednev, P. A. Sdvizhenskii, A. V. Rogachevskaya, P. A. Chizhov, A. A. Ushakov, V. V. Bulgakova, V. A. Zavozin, S. M. Pershin and A. F. Bunkin, Spectrochim. Acta, Part B, 2026, 235, 107393 CrossRef CAS.
- Y. Song, Z. Hou, X. Yu, W. Yao and Z. Wang, Anal. Chim. Acta, 2025, 1335, 343464 CrossRef CAS PubMed.
- K. Kou, Z. Hou and Z. Wang, Spectrochim. Acta, Part B, 2025, 228, 107186 CrossRef CAS.
- V. Motto-Ros, F. Pelascini, M. Dell'Aglio and A. De Giacomo, Spectrochim. Acta, Part B, 2025, 230, 107240 CrossRef CAS.
- S. Wu, Z. Lai, C. Yang, J. Yue, X. Duan, L. Li, Z. Cui, B. Wei, S. Sun, B. Hu and Z. Liu, Spectrochim. Acta, Part B, 2025, 224, 107110 CrossRef CAS.
- B. Wei, S. Liu, L. Zhao, Z. Zhu, J. Xie, Y. Chen, S. Shi, S. Wu, Q. Zeng, D. Qian, X. Liu, J. Shen and Z. Liu, Anal. Chem., 2025, 97(43), 24031–24039 CrossRef CAS PubMed.
- B. Dai, J. Wang, M. Ye, Y. Han and H. Li, J. Anal. At. Spectrom., 2025, 40(3), 879–887 RSC.
- X. Chen, R. Hai, Y. Wu and H. Ding, Spectrochim. Acta, Part B, 2025, 234, 107341 CrossRef CAS.
- Y. Li, L. Ding, Y. Wang, M. Shan, J. Cong, J. Lin and M. Cui, Talanta, 2025, 290, 127762 CrossRef CAS PubMed.
- H. Sun, Y. Zhou, J. Li, X. Guo, M. Shi, Z. Hu, J. Wu, X. Li and J. Li, Talanta, 2025, 289, 127706 CrossRef CAS PubMed.
- J. Xu, X. Wang, M. Yao and M. Liu, J. Anal. At. Spectrom., 2025, 40(9), 2556–2561 RSC.
- S. J. Shetty, J. C. Peralta, H. Tatarevska, I. A. Urbina, V. Contreras and P. Veis, Spectrochim. Acta, Part B, 2025, 230, 107218 CrossRef CAS.
- F. V. Nakadi, A. Garcia-Garcia, A. Rua-Ibarz and M. Resano, Talanta, 2025, 292, 127920 CrossRef CAS PubMed.
- T. N. Tran, D. Han, S. Shim and D. H. Choi, Spectrochim. Acta, Part B, 2025, 225, 107121 CrossRef CAS.
- N. Karim, M. P. Polek, A. M. Casella, D. J. Senor, S. S. Harilal and E. J. Kautz, Spectrochim. Acta, Part B, 2025, 223, 107081 CrossRef CAS.
- H. Pei, Q. Fan, Y. Duan and M. Zhang, Appl. Sci., 2025, 15(15), 8640 CrossRef CAS.
- X. Fang, H. Yu, Q. Huang, Z. Jiang, D. Pan and W. Gui, Chemometrics Intell. Lab. Syst., 2025, 257, 105321 CrossRef CAS.
- W. Gao, B. Han, Z. Sun, Y. Yan, Y. Ye, J. Feng, Y. Cai, A. Iroshan and Y. Liu, Talanta, 2025, 293, 128182 CrossRef CAS PubMed.
- P. Zheng, Z. Sun, J. Wang, G. Chen, Z. Yang and B. Li, Talanta, 2026, 297(Pt A), 128572 CrossRef CAS PubMed.
- J. Zhou, W. Huang, H. Sattar, Q. Hu, H. Liu, G. Wang, D. Zhang and L. Guo, Talanta, 2026, 296, 128453 CrossRef CAS PubMed.
- M. Bosáková, K. Novotný, J. Moros and J. Laserna, Spectrochim. Acta, Part B, 2025, 226, 107140 CrossRef.
- B. Xu, Z. Hao, Y. Wang, L. Liu, N. Zhang, Y. Rao, L. Wang, J. Shi and X. He, J. Anal. At. Spectrom., 2025, 40(8), 2038–2048 RSC.
- G. Yang, X. Jin, L. Liu, T. Gu, Y. Fu, H. Sun, C. Li and L. Yao, Talanta, 2025, 292, 127936 CrossRef CAS PubMed.
- K. Zhang, J. Ji, Z. Liu, Z. Hou and Z. Wang, Chemosensors, 2024, 12(12), 277 CrossRef.
- R. Guerrini, C. Alvarez-Llamas, L. Sancey, V. Motto-Ros and L. Duponchel, Spectrochim. Acta, Part B, 2025, 227, 107167 CrossRef CAS.
- B. Wang, Y. Lu, Q. Wei, Q. Shan, Z. Zhang, Y. Ling and W. Jia, Opt Laser. Technol., 2026, 193, 114262 CrossRef CAS.
- B. G. Beglaryan, A. S. Zakuskin, A. M. Popov and T. A. Labutin, Spectrochim. Acta, Part B, 2026, 235, 107381 CrossRef CAS.
- R. G. Wala, M. P. Polek, S. S. Harilal, R. J. Jones and M. C. Phillips, Spectrochim. Acta, Part B, 2025, 227, 107142 CrossRef CAS.
- E. Chenshen, J. Tan, B. Wang, E. Jiang, N. Zhao, S. Zheng, Z. Luo, J. Li and Q. Zhang, J. Anal. At. Spectrom., 2025, 40(8), 1902–1915 RSC.
- A. Retzmann, A. Menon and M. E. Wieser, J. Anal. At. Spectrom., 2025, 40(10), 2803–2815 RSC.
- Y. Zhu, G. Yang, A. Sakaguchi, T. Miura, Y. Shikamori and J. Zheng, J. Anal. At. Spectrom., 2025, 40(6), 1428–1446 RSC.
- Z. Li, Y. Shao, M. Luo, L. Ma and D. Xu, Forensic Chemistry, 2025, 43, 100646 CrossRef CAS.
- K. J. Mathew, J. Anal. At. Spectrom., 2025, 40(8), 1879–1901 RSC.
- Y. Weiss, G. Guadaño, J. M. Koornneef, Y. Kempe, O. Tirosh and G. R. Davies, Mineral. Petrol., 2025, 119(3), 449–464 CrossRef CAS.
- Z. Varga, M. Wallenius and K. Mayer, Talanta, 2025, 292, 127896 CrossRef CAS PubMed.
- T. Ariga, K.-C. Shin and T. Narukawa, Spectrochim. Acta, Part B, 2025, 225, 107134 CrossRef CAS.
- J.-H. Liu, X.-D. Deng, J.-W. Li, L.-P. Feng and T. J. Algeo, J. Anal. At. Spectrom., 2025, 40(10), 2641–2648 RSC.
- C. Israel, C. Chauvel, T. H. Luu, E. C. Inglis and D. Roberts, Chem. Geol., 2025, 680, 122686 CrossRef CAS.
- Z. Chen, X.-X. Ling, S.-T. Wu, S. L. Kamo, Y. Liu, D. Zhang, Q.-L. Li and X.-H. Li, J. Anal. At. Spectrom., 2025, 40(11), 3236–3244 RSC.
- Z. Bao, C. Zong, D. Peng and H. Yuan, J. Anal. At. Spectrom., 2025, 40(8), 2107–2116 RSC.
- S. Niki and T. Hirata, J. Anal. At. Spectrom., 2025, 40(10), 2716–2724 RSC.
- E. Eensoo, P. Paiste, K. Paiste, D. A. Fike and J. L. Houghton, J. Anal. At. Spectrom., 2025, 40(8), 2126–2137 RSC.
- M. A. Kipp, L. F. Piccirillo and D. Peters, J. Anal. At. Spectrom., 2025, 40(10), 2792–2802 RSC.
- C. D. Standish, J. A. Milton, R. M. Brown and G. L. Foster, J. Anal. At. Spectrom., 2025, 40(5), 1309–1322 RSC.
- S. T. Lancaster, J. Irrgeher, R. Dallmayr, E. Conrad, M. Horhold, P. Bohleber, M. Behrens, F. Camin, K. Zagar, P. Vreca and T. Prohaska, Anal. Chem., 2025, 97(38), 20788–20797 CrossRef CAS PubMed.
- Z. Liu, J. Lin, X. Jiang, X. Zhu, W. Liu, Y. Liu, W. Zhang and Z. Hu, J. Anal. At. Spectrom., 2025, 40(5), 1192–1202 RSC.
- Y. Wang, W. Yang, S. Ke, Y. He and Q. Fan, J. Anal. At. Spectrom., 2025, 40(8), 2197–2206 RSC.
- Y. Zhang, W. Yang, Y. Wang, Y. He, S. Ke, K. Qu and S. Zhang, Geostand. Geoanal. Res., 2025, 49(2), 315–327 CrossRef CAS.
- E. Wolfer, C. Burkhardt and T. Kleine, J. Anal. At. Spectrom., 2025, 40(4), 1023–1036 RSC.
- J. Jiang, W. Chen, J. Lu, Y. Liu, M. Li, J. Sun, K. Zhao, S. Jiang and Y. Liu, J. Anal. At. Spectrom., 2025, 40(4), 975–988 RSC.
- A. Retzmann and M. E. Wieser, Anal. Chim. Acta, 2026, 1383, 344865 CrossRef CAS PubMed.
- D. Li, Z. Zhang, X. Fan, T. Wu, Z. Huang, Q. Shen, X. Liu and S.-A. Liu, J. Anal. At. Spectrom., 2025, 40(5), 1383–1393 RSC.
- H. Wu, J.-H. Bai, X.-R. Liang, X.-B. Lu, Y.-Y. Li, Y.-N. Deng, Y. Zhao and G.-J. Wei, J. Anal. At. Spectrom., 2025, 40(3), 860–867 RSC.
- W. Wang, L. Feng, R. Xi, W. Wang, J. Xu, Y. Wang, S. Guo, P. Zhang, M. Zhang, Y. Zhang, F. Yu, Y. Wang, Y. Su, S. Fang, X. Yuang and Z. Li, Spectrochim. Acta, Part B, 2025, 223, 107077 CrossRef CAS.
- S. Y. Han, S. Hemming, B. J. Treves Brown, M. A. Higginson, P. Kaye, C. A. Sharrad and S. L. Heath, J. Radioanal. Nucl. Chem., 2025, 334(5), 3783–3796 CrossRef CAS PubMed.
- P. Sharma, J. Govindaraj, C. Poulin-Ponnelle and D. Lariviere, Microchem. J., 2025, 212, 113099 CrossRef CAS.
- C. Peng, X. Du, H. Wang, Y. Lu, J. Chen, F. Zhang, X. Gao, H. Ma, S. Xing, X. Hou and K. Shi, Anal. Chem., 2025, 97(30), 16338–16345 CrossRef CAS PubMed.
- P.-f. Zhang, Z.-m. Li, W.-l. Wang, L. Feng, J. Xu, M.-c. Zhang, R.-y. Xi, X.-p. Shen and W. Wang, Int. J. Mass Spectrom., 2025, 517, 117496 CrossRef CAS.
- Z. Varga and M. Wallenius, Spectrochim. Acta, Part B, 2025, 233, 107287 CrossRef CAS.
- J. S. Stanberry, S. E. Szakas, H. B. Andrews, L. Hendriks, B. W. Ticknor, R. A. Bergin, S. K. Tazik, P. Kegler, S. Neumeier, N. A. Zirakparvar, D. R. Dunlap and B. T. Manard, Talanta, 2026, 297(Pt A), 128578 CrossRef CAS PubMed.
- N. A. Zirakparvar, B. T. Manard, D. R. Dunlap, M. Darnell, D. Bostick, B. Ticknor and C. R. Hexel, Rapid Commun. Mass Spectrom., 2025, 39(17), e10078 CrossRef CAS PubMed.
- S. Bowden, L. R. Shultz-Johnson, M. O. Harrell, M. G. Bronikowski, C. Fitzgerald, C. R. Barrett, M. S. Wellons, K. M. Samperton, K. Koh and G. B. Green, J. Anal. At. Spectrom., 2025, 40(12), 3520–3529 RSC.
- C. D. Coath, T. Elliott and R. C. Hin, Chem. Geol., 2017, 451, 78–89 CrossRef CAS.
| This journal is © The Royal Society of Chemistry 2026 |