Open Access

Atomistic simulations are integral to the knowledge and design of glasses, but are nonetheless challenging due to slow structural dynamics, complex workflows, and issues of data management. With advances in high-throughput modeling and machine-learned interatomic potentials, it is worth increasing the Findability, Accessibility, Interoperability, and Reusability (FAIR) of these simulations. Here we present a Python package for the automated setup, execution, and analysis of atomistic simulations of glasses. The software package has semi-to-fully automated modules for the setup of molecular dynamics simulations, preparation of glassy systems, calculation of properties (elastic moduli, viscosity), and structural analysis of the resulting atomic configurations across short- and medium-range order. The computational and workflow management capability is based on the pyiron framework. Showcases of the automated glass preparation, property calculation, and analysis will be presented. The package includes a dedicated web application programming interface for use by large language models. We therefore provide an extensible platform for standardized, FAIR, AI-assisted atomistic simulation of glasses.

Accurate extraction of the complex dielectric constant in the terahertz (THz) band is essential for material characterization and non-destructive evaluation yet remains challenging due to the ill-posed nature of electromagnetic inverse problems and the limited availability of reliable reference data. In this work, a field-material couple neural network (FMCNN) is proposed to retrieve the complex dielectric constant directly from THz measurements. The FMCNN consists of a field neural network and a material neural network that are strongly coupled through the frequency-domain Maxwell equations in the form of a Helmholtz equation, with the governing physics enforced by partial differential equation and boundary condition constraints. This formulation enables zero-shot physics-informed learning inversion, requiring only measured test data as input. The extracted dielectric constants are validated by comparison with results from a one-dimensional normal-incidence model and the Drude–Lorentz model, showing good agreement over a broad frequency range, particularly above 0.2 THz. These results demonstrate that the FMCNN provides a physics-consistent and data-efficient approach for material parameter extraction in the THz band, offering an alternative to conventional model-based methods.

Despite EU-driven standardization efforts, the lack of harmonized analytical workflows still limits the compa rability of microplastic (MP) measurements across laboratories. This study evaluates three infrared (IR)-based techniques (µ-FTIR, FPA-FTIR, and QCL-LDIR) using a metrologically characterized polyethylene terephthalate (PET) representative test material (RTM) containing 1318 ±73 particles per tablet (20–500 µm). Independent laboratories applied their routine analytical workflows, while µ-Raman spectroscopy served as the orthogonal reference method. The results identify 50 µm as a critical metrological threshold influencing IR analytical performance. Above 50 µm, IR-based methods showed closer agreement with Raman values and reduced dispersion compared to smaller fractions, with recoveries ranging from 28–54% for QCL-LDIR, 60–73% for FPA-FTIR, and 83–108% for µ-FTIR. Below 50 µm, recoveries decreased (µ-FTIR: 41–44%; FPA-FTIR: 36–54%; QCL-LDIR: 59–95%), although the broad interlaboratory range observed for QCL-LDIR indicates that optimized analytical workflows can sub stantially improve performance. Z-score and statistical analyses further showed that analytical performance was strongly size-dependent and influenced by both instrumental characteristics and laboratory-specific workflows. Supplementary experiments using a low-load PET RTM (96 ±14 particles per tablet, 20–500 µm) improved agreement with the Raman reference, confirming that particle crowding may contribute to analytical bias. Overall, the study shows that high internal precision may coexist with systematic bias, emphasizing that reproducibility alone is insufficient to ensure metrological reliability and interlaboratory comparability. These findings support size-resolved performance evaluation and the harmonization of analytical workflows to achieve robust and comparable microplastic measurements.

A distributed fiber optic sensing system combining Raman, Rayleigh, and Brillouin scattering mechanisms is proposed for simultaneous strain, temperature, and vibration measurement. Preliminary experiments demonstrate its excellent sensing performance.

The report describes preparation, analysis and certification of the certified copper reference material BAM-M380, which is intended for establishing and checking the calibration of optical emission spectrometers for the analysis of samples of similar matrix composition.