Journal of Analytical Atomic Spectrometry最新文献

Heavy metal concentrations in soils near smelters are critical indicators for assessing soil quality, ecological risks, and potential health threats. However, accurate monitoring remains challenging due to soil matrix complexity and limited labeled spectral data. This study presents a semi-supervised learning framework based on a teacher–student model combined with GraphSAGE. The approach incorporates intra-group consistency constraints to map laser-induced breakdown spectroscopy (LIBS) spectra to the concentrations of Cr, Cu, Cd, Pb, and Zn. Spectral data were preprocessed using Savitzky–Golay filtering, normalization, feature selection, and PCA. The resulting components served as inputs to the model. Under the fully labeled dataset, the GraphSAGE-based framework outperformed conventional sequence models (LSTM and Transformer), achieving lower mean absolute percentage error (MAPE), generally reduced root mean squared error of prediction (RMSEP), and improved precision reflected by a lower mean relative standard deviation (mean RSD) on the labeled test set. By integrating unlabeled samples via the semi-supervised strategy, the teacher–student framework further improved model robustness and predictive stability, lowering MAPE to 5.23% (Cu), 2.73% (Cr), 5.82% (Pb), 4.90% (Zn), and 6.29% (Cd), with corresponding reductions in RMSEP and mean RSD. Finally, the optimized model mapped heavy metal distributions across the study area. Concentrations were high near the smelter and peaked in downwind zones. These patterns align with the atmospheric transport of smelter-derived particulates, confirming their dominant role in dispersion. The proposed method offers a practical tool for environmental monitoring and supports precision remediation strategies.

Numerous methods and techniques have been developed over time in gamma spectrometry to analyze spectra using specialized software. Review-type publications in gamma spectroscopy generally do not present certain methods that nevertheless exist and are used. This article presents a collection of various approaches, both old and new, applied at each stage of gamma spectrum analysis to identify and quantify the radioisotopes present in a sample, while also discussing the strengths and limitations of each method. The objective is to offer the reader a clear and structured view of the available tools, thereby facilitating the development of more efficient and accurate analysis software. It covers the entire process, from the raw spectrum to the final data interpretation, including peak detection, centroid fitting, peak shape correction, 511 keV peak correction, coincidence correction, radioisotope identification, and activity calculation. Finally, several data resources developed by our team, including source code, are freely available to support developers in designing or improving spectral analysis software, particularly for HPGe detectors.

Underwater laser-induced breakdown spectroscopy (underwater LIBS) is gaining increasing attention as an on-site analytical technique available in water-confined environments, such as the deep sea. Use of a solid substrate in underwater LIBS is beneficial for detecting dissolved elements. In this study, we demonstrate the first use of porous silicon (Si) as a substrate for underwater LIBS, which is fabricated by metal-assisted etching (metal-assisted chemical etching) using silver nanoparticles. A long-nanosecond laser (30 ns) operating at a wavelength of 1064 nm and a low energy of 1.5 mJ was focused onto a substrate immersed in a 0.6 mol per L sodium chloride aqueous solution containing 2.5 ppm (mg L⁻¹) lithium (Li). By using porous Si instead of flat Si as the substrate, the spectral line intensity of Li was enhanced by 8.0 times. A linear calibration curve with a coefficient of determination of 0.999 was obtained using porous Si in the range of Li concentration from 0.5 to 10.0 ppm. This study suggests a new potential approach for utilizing porous Si and extends the applicability of LIBS to liquid analysis.

The rapid advancement of artificial intelligence (AI), particularly machine learning (ML), is revolutionizing stable isotope geochemistry, enhancing mass spectrometry-based analytical techniques and unlocking transformative capabilities in data interpretation and geochemical process modelling. This critical review offers a comprehensive synthesis of the integrated applications of ML in this field, encompassing a systematic survey of the commonly employed algorithms—such as random forests, support vector machines, and neural networks—along with an in-depth examination of their geochemical uses. We classify and evaluate these methods, highlighting their roles in improving the data processing efficiency, prediction accuracy, and mechanistic insight across diverse applications including provenance studies, paleoclimate reconstruction, and environmental forensics. Nevertheless, several pressing challenges impede their broader implementation, such as data scarcity for non-traditional isotope systems, limited model interpretability, and the persistent risk of geochemically implausible predictions. We argue that overcoming the “black-box” nature of ML demands the integration of domain knowledge through physics-informed neural networks (PINNs), improved explainable AI (XAI) frameworks, and strengthened interdisciplinary collaboration. Looking ahead, we emphasize the need to optimize the analytical accuracy through intelligent instrumentation, develop standardized data infrastructures, and foster algorithm innovation tailored to geochemical principles. This review aims to provide an authoritative reference by synthesizing the recent advances, openly addressing the current limitations, and outlining pragmatic research directions to accelerate the adoption of ML in stable isotope geochemistry. By tackling these priorities, ML stands to not only refine the existing methodologies but also open new scientific frontiers in understanding the Earth's dynamic systems, ultimately revolutionizing isotope-enabled discovery.

This update covers the literature published between approximately June 2024 and April 2025 and is the latest part of a series of annual reviews. It is designed to provide the reader with an overview of the current state of the art with respect to the atomic spectrometric analysis of various metals, polymers, electronic, nano and other materials. Data processing continues to be the major focus for LIBS and TOF-SIMS analyses, mainly to provide reliable analyte quantification data. A variety of machine learning algorithms and statistical approaches have been used for this, often in multiple steps. Although these algorithms have been used for some years, their use is expanding into new areas. Another development is the combination of complementary techniques on the same instrument platform. This enables data from the two techniques to be obtained simultaneously and from the same spot on the sample. The analysis of polymers and nanomaterials continues to develop, with the prominent platforms used being SP-ICP-MS, SP-ICP-TOF-MS and X-ray based techniques. In addition, efforts are now being accelerated to produce nanomaterial CRMs and RMs, the lack of which has hampered truly robust method validation. For electronic materials XPS, GIXRF, GEXRF and TOF-SIMS remain dominant for surface and depth profiling, whilst for bulk composition LIBS, ICP-MS, and XRF remain prominent. Work in this area is also focussing on the development of advanced sample preparation and microextraction approaches that expand the scope of laser-based spectroscopy.

Single particle inductively coupled plasma-mass spectrometry (sp-ICP-MS) produces time series—time scans—that require processing to extract meaningful information regarding the nanoparticles that are analysed. In this work, we present a stochastic algorithm to generate such time scans. We also introduce an open-source implementation of it, , a Python library designed to generate synthetic, yet realistic sp-ICP-MS time scans for a single mass-to-charge ratio. We argue that our library is an efficient and reliable testbed on which future studies can build to assess existing or new data processing strategies.

Niobium (Nb) is a critical strategic metal essential for advanced industrial and aerospace technologies. As aeschynite represents one of the primary natural sources of Nb, its geochronological characterization is of significant interest. In situ microbeam geochronology of aeschynite can provide a direct approach to constraining the timing of niobium mineralization and regional geological evolution. Accurate and precise in situ microbeam dating using SIMS or LA-ICP-MS requires well-characterized matrix-matched reference materials to correct for matrix-induced elemental fractionation. However, a critical gap exists due to the current absence of natural aeschynite reference materials. To address this deficiency, we propose Hidra aeschynite-(Y) from the Urstad Feldspar Mine, Norway, as a candidate of natural reference material for Pb–Pb and Lu–Hf dating. Multiple SIMS spot analyses demonstrate that it has a homogeneous age, with an arithmetic mean ²⁰⁷Pb/²⁰⁶Pb age of 902.1 ± 21.3 Ma (2SD; n = 46), corroborated by three LA-ICP-MS/MS analytical sessions (arithmetic mean ²⁰⁷Pb/²⁰⁶Pb age = 902.5 ± 22.5 Ma, 2SD; n = 55). Six ID-TIMS analyses yield an arithmetic mean ²⁰⁷Pb/²⁰⁶Pb age of 912.8 ± 6.9 Ma (2SD), recommended as the best estimate of the crystallization age. In addition, in situ LA-ICP-MS/MS Lu–Hf analyses of Hidra aeschynite-(Y) calibrated with NIST SRM 610 yield an isochron age of 912.5 ± 6.5 Ma (2σ; MSWD = 1.4; n = 68). The convergence of microbeam ²⁰⁷Pb/²⁰⁶Pb ages with the high-precision ID-TIMS benchmark, combined with reproducible Lu–Hf results, establishes Hidra aeschynite-(Y) as the first potential matrix-matched reference material for microbeam in situ Pb–Pb and Lu–Hf geochronology of aeschynite.

Laser-induced breakdown spectroscopy (LIBS) suffers from shot-to-shot fluctuations that constrain signal intensity and stability. While spatial confinement effectively enhances emission signals, the quantitative relationship between plasma plume morphology and spectral characteristics remains unclear, limiting mechanistic understanding and optimization strategies. This study establishes a three-tier hierarchical correlation framework connecting spectral characteristics, global plasma morphology and local morphological features using Pearson and Spearman correlation analysis. We employed optical emission spectroscopy (OES), high-speed photography, and shadowgraphy to analyze with plate wall spatial confinement (PWSC), correlations between morphological parameters (axial length, radial length, plume area, total integrated image intensity, average pixel intensity, region of interest integrated intensity and axial-to-radial ratio) and spectral characteristics (intensity, stability, enhancement factors). Results show PWSC altered morphology–spectrum correlations: at 7 μs, spectral enhancement reached 1.88 fold with RSD reduced from ∼12% to 5.8%. Notably, the AL/RL-intensity correlation reversed from strong positive (0.94 to 0.997) to moderate negative (−0.38 to −0.46), while RL intensity correlation strengthened dramatically (0.13 to 0.95), highlighting RL dominant role under confinement. These findings provide mechanistic insights into confinement-induced plume dynamics and establish a foundation for correlation-weighted plasma image-spectrum fusion optimization in high-precision elemental analysis.

As a real-time in situ elemental detection technology, Laser-Induced Breakdown Spectroscopy (LIBS) has been applied to the detection and exploration of marine environments and mineral resources. However, due to the influence of high pressure, there have been no reported LIBS detection results for solid samples at depths exceeding 4000 m on the seabed. This study developed a novel elemental chemical sensor system using high-pressure gas ventilation, creating a high-pressure gas environment on the seabed to avoid the influence of seawater on the LIBS results. In 2024, the elemental chemical sensor system was deployed on the Haixing 6000 remotely operated vehicle (ROV) for three deep-sea trials, achieving the world's first spectral line results for carbon steel samples on the seabed at 6000.8 m depth. The research also investigated the influence of delay time on spectral results at 6000.8 m depth. Comparative analysis revealed that the gas-ventilation method effectively extends plasma lifetime compared to direct solid detection in aqueous environments. The above results demonstrate that the elemental chemical sensor possesses the capability for in situ detection of solid samples in deep-sea environments, providing a novel and effective solution for submarine geochemical research and submarine mineral exploration.

The determination of long-lived radionuclides by inductively coupled plasma tandem mass spectrometry (ICP-MS/MS) is a well-established approach. However, such determinations can still be hindered by isobaric interferences from stable isotopes of neighbouring elements. As such, investigations towards novel gas cell approaches for removing interfering ions are required in order to improve the reliability of the analysis. Nitrous oxide (N₂O) is a reaction gas that has been well studied for stable isotope analysis. Studies towards its applicability to radionuclide analysis have so far been limited. Here, the use of N₂O, as well as a mixture with ammonia (NH₃), have been evaluated for determinations of 10 radionuclides of interest for nuclear decommissioning: ⁴¹Ca, ⁶³Ni, ⁷⁹Se, ⁹⁰Sr, ⁹³Zr, ⁹³Mo, ⁹⁴Nb, ¹⁰⁷Pd, ¹³⁵Cs, and ¹³⁷Cs. Single element solutions of stable isotope analogues of the radionuclides, as well as solutions of the interfering ions, were used to observe the reactions with the ICP-MS/MS reaction cell gases. Abundance-corrected sensitivities were used to assess the achievable separation factors and sensitivities for the determination of the radionuclides of interest. The N₂O/NH₃ gas mixture was found to provide a significant enhancement in the removal of isobaric interferences, as well as instrument detection limits (given in brackets), compared to N₂O alone for determinations of ⁴¹Ca (0.50 pg g⁻¹ (0.0016 Bq g⁻¹)), ⁷⁹Se (0.11 pg g⁻¹ (5.4 × 10⁻⁵ Bq g⁻¹)), ⁹⁰Sr (0.11 pg g⁻¹ (0.56 Bq g⁻¹)), ⁹³Mo (0.12 pg g⁻¹ (0.0044 Bq g⁻¹)), ¹³⁵Cs (0.1 pg g⁻¹ (7.5 × 10⁻⁶ Bq g⁻¹)), and ¹³⁷Cs (0.1 pg g⁻¹ (0.33 Bq g⁻¹)).