Valérie Cappuyns

Antimony (Sb) is a critical raw material whose extraction, processing, and use contribute to environmental contamination and to its enrichment in soils, dusts, slags, and mining residues. Human exposure to Sb occurs through, among others, incidental ingestion of contaminated particles, making oral bioaccessibility an important parameter for human health risk assessment. This study investigated the in vitro oral bioaccessibility of Sb in solid environmental matrices and evaluated the influence of geochemical properties, environmental matrix type, and analytical methodology. A meta-analysis was conducted using data from 20 peer-reviewed studies published between 2009 and 2025, comprising 175 samples of soils, dusts, and mining waste analyzed using established in vitro bioaccessibility methods. Total Sb concentration was identified as the strongest predictor of both gastric and intestinal Sb bioaccessibility. Differences in gastric and intestinal bioaccessibility were observed between soils, dust and mining waste samples, while no conclusions could be drawn for intestinal bioaccessibility. The results demonstrate that Sb bioaccessibility is governed primarily by total Sb concentration, while methodological and mineralogical factors contribute to additional variability. Relative gastric bioaccessibility was also influenced by the in vitro method used, with Method 1340 and the UBM generally yielding lower values than the PBET, SBET, and SBRC. However, without a paired design, any causal conclusion about the optimality of a particular test is premature.

The origin and properties of dissolved organic matter in geothermal fluids have been little studied so far. It can be assumed that it originates from the geothermal reservoir rocks (natural organic matter, NOM) and from anthropogenic additives such as scaling and corrosion inhibitors (synthetic organic matter, SOM) mixed with the geothermal fluid during operation. In this study, Liquid Chromatography-Organic Carbon Detection (LC-OCD) was used to characterize NOM and SOM contributions to the geothermal brine. NOM leachability from reservoir rocks was investigated by batch leaching experiments using 3.5 mol/L NaCl at 125◦C and 70◦C. SOM contribution was assessed by studying the thermal degradation of three inhibitor types (acrylates, phosphonates, and a commercial inhibitor mixture) diluted in 3.5 mol/L NaCl, also at 125◦C and 70◦C. The results show that the leachable NOMfraction (3-10%) is primarily composed of hydrophilic low molecular weight organics and macromolecules, with minor hydrophobic components. Only part of this leachable NOM is mobile, since the ground rock samples do not reflect the intact rock matrices. In the latter, large polymers and particulate organic matter (POM) can be physically retained. At high temperature and salinity, SOM from antiscalants and chemical inhibitors degrades, releasing diverse organic molecules. They can form organic precipitates that may clog filters in geothermal installations. Compared with organic matter in natural geothermal brine, results reveal that its molecular weight distribution is the same as that of NOM and SOM, where low molecular weight organics predominate. Characterizing commercial inhibitor mixtures remains challenging due to undisclosed formulations, as well as analytical limitations. Low resolution chromatographic methods alone cannot reliably distinguish betweenNOM and SOM.

Pyrometallurgical production of non-ferrous metals generates substantial slag volumes, yet potential metal leaching remains a critical barrier to sustainable valorisation. In principle, appropriate control of metallurgical processing can tailor slag characteristics to achieve acceptable leachability and reduced environmental risk. Understanding this behaviour requires systematic consideration of slag chemistry, mineralogy, and microstructure. This is explored herein in synthetic Fe,Al-silicate slags, representative of those produced in non-ferrous metallurgy. Chemistry, oxygen partial pressure, and cooling rates were varied to assess their influence on the distribution and leaching of As, Sb, and Mo. Slags were characterised by XRF, ICP-OES, QXRD, SEM/EDX, and EPMA, while leaching was evaluated (EN 12457-2 test) to establish processing–leaching relationships. Rapid cooling was found promoting amorphous Fe–Si matrices formation, immobilising As and Sb, whereas slower cooling favoured fayalite crystallisation, resulting in increased leaching of these elements. Mo exhibited the lowest leaching in slowly cooled slags, likely due to the formation of Mo-rich phases. Arsenic (As) and Sb preferentially partitioned into the amorphous phase, independent of the Fe/Si ratio; however, higher Fe/Si ratios diminished the stabilising capacity of this phase increasing their leaching. Oxygen partial pressure played a critical role in elemental partitioning: higher pO2 (10⁻10 atm) favoured the retention of As, Sb, and Mo within the amorphous matrix, whereas lower pO2 (10⁻12 atm) promoted the formation of metallic Fe particles enriched in these elements, leading to elevated leaching. Overall, the findings herein provide insights into how to steer process parameters during slag formation and effectively minimise leaching.

The expansion of lithium mining in Europe is critical for securing strategic raw materials but poses growing challenges for the long-term management of mine tailings. Arsenic (As), a potentially harmful element present in trace amount in lithium ores, may report to some residues and tailings during lithium beneficiation, yet its mineral hosts and solid-phase speciation remain poorly constrained due to low bulk concentrations and diffuse distribution across multiple phases. This uncertainty limits reliable prediction of arsenic behaviour during processing, leaching, and storage, while current regulatory assessments largely rely on bulk concentrations and standardized leaching tests that may overlook key mineralogical controls. Here, we combine quantitative mineral deportment analysis with arsenic K-edge X-ray absorption near-edge structure (XANES) spectroscopy to characterise arsenic distribution and solid-phase speciation in ores and process residues from three major European lithium projects. Results show that arsenic is distributed among sulphide, silicate, oxide, and phosphate phases, many of which host arsenic at trace concentrations but occur in high modal abundance. Least-squares linear combination fitting reveals the predominance of As(-I) and As(V), with minor As(III) contributions in a subset of samples. XANES analyses of leaching residues demonstrate oxidation of arsenic toward As(V), showing that low arsenic release does not necessarily reflect stability, as sorption processes can mask underlying reactivity. This mineral-centric approach enables the estimation of the relative contributions of As(-I,III,V)-bearing minerals and their control on arsenic release across contrasting geochemical environments. The proposed framework supports evidence-based decisions for targeted tailings management solutions, thereby strengthening environmental strategies for cleaner lithium production.

In vitro bioaccessibility tests are used to estimate the release of contaminants from environmental samples during simulated digestion, making them available for intestinal absorption. In most cases, the samples are fine-grained materials with varying chemical, physical, and mineralogical properties, but it is not always clear how these properties influence the bioaccessibility of elements. The present study focusses on the bioaccessibility of lead (Pb) and arsenic (As) in mining waste and mining-affected soils. From the literature, data from mining waste and mining-affected soil samples were used to investigate the relation between chemical (element composition, pH, organic carbon content), physical (grain size distribution), and mineralogical properties of the samples and the gastric and intestinal bioaccessibility of Pb and As. Mean gastric As bioaccessibility was significantly lower in acidic samples than neutral and alkaline samples. A significant difference was also found between As and Pb bioaccessibility in mining residues and mining-affected soil samples. Overall, total Pb an As concentrations and pH were the most significant predictors of Pb and As bioaccessibility. Due to the lack of (quantitative) mineralogical data in many papers, it was not possible to make precise predictions of As and Pb bioaccessibility based on mineralogical sample composition. Despite the challenging nature of quantitative mineralogical characterization, it can contribute to a more precise estimation of the bioavailability of Pb and As in mining waste. Given their significant impact on the bioavailability of metal(loid)s, pH and the (quantitative) mineralogical sample composition should be more systematically determined and reported.