The development of machine-learning models for atomic-scale simulations has greatly benefited from the large databases of materials and molecular properties, computed using electronic-structure calculations. Recently, these databases enabled the training of “universal” models that aim to make accurate predictions for arbitrary atomic geometries and compositions. However, many of these databases were originally designed for materials discovery, focusing primarily on equilibrium structures. Here, we introduce a dataset designed to train machine-learning models to make reasonable predictions for arbitrary structures. Starting with relatively small sets of stable structures, we built the dataset aiming to achieve “massive atomic diversity” (MAD) by aggressively modifying these structures and utilizing highly consistent electronic-structure settings for property calculations. Despite containing fewer than 100,000 entries, the MAD dataset has already enabled the training of universal interatomic potentials that rival those trained on datasets containing two to three orders of magnitude more data. We detail the design philosophy of the dataset and introduce low-dimensional structural latent space descriptors that can be used as a general-purpose materials cartography tool.
We present the experimental campaigns - namely, three per facility - carried out between 2018 and 2021 in the AKR-2 and CROCUS zero power reactors within the framework of the Horizon 2020 European project CORTEX. Their purpose was to produce high-quality and noise-specific experimental data for the validation of the neutron noise computational models developed in CORTEX. In both reactors, perturbations were induced by two devices, separately and altogether. In AKR-2, they consisted of a rotating absorber, i.e. an absorber of variable strength, and a linear oscillator, i.e. a vibrating absorber, both sets in horizontal channels close to the core. In CROCUS, the project benefited from the COLIBRI experimental program and its fuel rods oscillator set in the outer lattice; an additional vibrating absorber called POLLEN was set in a vertical air-channel at core center. The campaigns at both facilities consisted of neutron measurements with numerous detectors at reference static states, and with the addition of the mechanical perturbations to induce neutron reactivity modulation. The present article documents the experimental setups and measurements for each facility and perturbation type. A focus is set on the experimental designs and their evolution along the project, as well as motivations and learned lessons. Results are presented and discussed in details in associated papers.
This letter reports the result of the search for the decay mu(+ )-> e(+)gamma undertaken at the Paul Scherrer Institut in Switzerland with the MEG II experiment using the data collected in the 2021-2022 physics runs. The sensitivity of the branching ratio measurement in this search is 2.2 x 10(-13), a factor of 2.4 better than that of the full MEG dataset and obtained in a data taking period of about one fourth that of MEG, thanks to the superior performances of the new detector. The observed data are consistent with the expected background, yielding an upper limit on the branching ratio of B(mu(+ )-> e(+)gamma) < 1.5 x 10(-13 )(90% C.L.). Additional improvements are expected with the data collected during the years 2023-2024. The data-taking will continue in the coming years.
Cement production, a significant contributor to global CO _2 emissions (about 8 _2 footprint. Given a cement recipe (clinker and supplementary cementitious materials), thermodynamic modeling is used to predict the corresponding hydration products that can then be utilized to calculate homogenized mechanical properties at the paste scale, e.g. elastic moduli. This supports a robust training database. An Artificial Neural Network is then trained to predict elastic moduli directly from the cement recipe, greatly accelerating the evaluation of the cement properties. The optimization targets are set to minimize CO _2 emissions associated with the recipe, and maximize the bulk modulus of the resulting hydrated cement paste. The results show a sequence of optimal cement recipes, lying on a Pareto front, showcasing bulk modulus values of hydrated cement paste from 14.61 to 23.63 GPa and CO _2 equivalents (in kg per kg cement paste) ranging between 0.39 to 0.76. The developed methodology highlights the potential to reduce CO _2 emissions while maintaining or improving the mechanical properties of cementitious materials, enhancing sustainability in construction practice.
The US Magnet Development Program is leading an effort to design and manufacture a 20 T accelerator dipole magnet. Various designs are under development featuring cos-theta, block-coil, or common-coil geometries. To achieve the target field while maintaining cost-effectiveness the magnet cross-section includes high-temperature superconductor (HTS) inner coils and Nb $_{3}$ Sn outer coils, which are all powered in series. The quench protection of this class of high-field accelerator magnets is extremely challenging due to the high energy density, high current density, slow quench propagation, in particular in the HTS coils, and highly inhomogeneous thermal properties of HTS and Nb $_{3}$ Sn coils. The magnet quench discharge is simulated with the STEAM-LEDET program as coupled electro-magnetic and thermal transients. The peak temperature and voltage to ground in the magnet coils during the transient are presented. The performance of a CLIQ-based quench protection system applied to 1 m long model magnets in terms of peak temperature and voltage to ground is investigated. Furthermore, the scalability of the proposed solution to full-scale, 15 m long magnets is discussed.