arXiv subjects

Niccolò Di Eugenio

Publications and source records attributed to Niccolò Di Eugenio.

5 recordsLinked to original sources

arXiv2026

Machine-Learned Interatomic Potentials for Structural and Defect Properties of YBa$_2$Cu$_3$O$_{7-δ}$

High-Temperature Superconductors (HTS) such as YBa2Cu3O7-delta (YBCO) are essential for next-generation Tokamak fusion reactors, where Rare-Earth Barium Copper Oxides (REBCO) form the functional layers in HTS magnets. Because YBCO's superconductivity depends strongly on oxygen stoichiometry and defect structure, atomistic simulations can provide crucial insight into radiation-damage mechanisms and pathways to maintain material performance. In this work, we develop and benchmark four Machine-Learned Interatomic Potentials (MLPs) for YBCO: the Atomic Cluster Expansion (ACE), the Message-Passing Atomic Cluster Expansion (MACE), the Gaussian Approximation Potential (GAP), and the Tabulated Gaussian Approximation Potential (tabGAP), trained on an extensive Density Functional Theory (DFT) database explicitly designed to include irradiation-damaged-like configurations. The resulting models achieve DFT-level accuracy across a wide range of atomic environments, faithfully capturing the interatomic forces relevant to radiation damage processes. Among the tested models, MACE delivers the highest accuracy, although at greater computational cost, while ACE and tabGAP provide an excellent balance between efficiency and fidelity. These machine-learned potentials establish a robust foundation for large-scale molecular dynamics simulations of radiation-induced defect evolution in complex superconducting materials

cond-mat.supr-con↗

arXiv2026

Insights Into Radiation Damage in YBa$_2$Cu$_3$O$_{7-δ}$ From Machine-Learned Interatomic Potentials

Accurate prediction of radiation damage in YBa$_2$Cu$3$O${7-δ}$ (YBCO) is essential for assessing the performance of high-temperature superconducting (HTS) tapes in compact fusion reactors. Existing empirical interatomic potentials have been used to model radiation damage in stoichiometric YBCO, but fail to describe oxygen-deficient compositions, which are ubiquitous in industrial Rare-Earth Barium Copper Oxide conductors and strongly influence superconducting properties. In this work, we demonstrate that modern machine-learned interatomic potentials enable predictive modelling of radiation damage in YBCO across a wide range of oxygen stoichiometries, with higher fidelity than previous empirical models. We employ two recently developed approaches: an Atomic Cluster Expansion (ACE) potential and a tabulated Gaussian Approximation Potential (tabGAP). Both models accurately reproduce Density Functional Theory (DFT) energies, forces, and threshold displacement energy distributions, providing a reliable description of atomic-scale collision processes. Molecular dynamics simulations of 5 keV cascades predict enhanced peak defect production and recombination relative to a widely used empirical potential, indicating different cascade evolution. By explicitly varying oxygen deficiency, we show that total defect production depends only weakly on stoichiometry, offering insight into the robustness of radiation damage processes in oxygen-deficient YBCO. Finally, fusion-relevant 300 keV cascade simulations reveal amorphous regions with dimensions comparable to the superconducting coherence length, consistent with electron microscopy observations of neutron-irradiated HTS tapes. These results establish machine-learned interatomic potentials as efficient and predictive tools for investigating radiation damage in YBCO across relevant compositions and irradiation conditions.

cond-mat.supr-con↗

arXiv2025

Beyond dpa: an atomistic framework for a quantitative description of radiation damage in YBa2Cu3O7

Radiation damage in high-temperature cuprate superconductors represents one of the main technological challenges for their deployment in harsh environments, such as fusion reactors and accelerator facilities. Their complex crystal structure makes modeling irradiation effects in this class of materials a particularly demanding task, for which existing damage models remain inadequate. In this work, we develop an atomistic-based approach for describing primary radiation damage in YBa2Cu3O7, by coupling Molecular Dynamics and Binary Collision Approximation simulations in a way that makes them complementary. When integrated with Primary Knock-on Atom spectra obtained from Monte Carlo codes, our results establish a framework for multiscale modeling of radiation damage, enabling quantitative estimates of several damage descriptors, such as defect production, defect clustering, and the effective damaged volume for any specific irradiation conditions where collision cascades dominate. This computational approach is suitable for the prediction of irradiation effects in any complex functional oxide, with applications ranging from aerospace to nuclear fusion and high-energy physics.

cond-mat.supr-con↗

arXiv2025

The diffusion-driven orthorhombic to tetragonal transition in YBa$_2$Cu$_3$O$_7$ derived with a machine learning interatomic potential

Defects in high temperature superconductors such as YBa$_2$Cu$_3$O$_7$ (YBCO) critically influence their superconducting behavior, as they substantially degrade or even suppress superconductivity. With the renewed interest in cuprates for next-generation superconducting magnets operating in radiation-harsh environments such as fusion reactors and particle accelerators, accurate atomistic modeling of defects and their dynamics has become essential. Here, we present a general-purpose machine-learning interatomic potential for YBCO, based on the Atomic Cluster Expansion (ACE) method and trained on Density Functional Theory (DFT) data, with particular emphasis on defects and their diffusion mechanisms. The potential is validated against DFT calculations of ground-state properties, defect formation energies of oxygen Frenkel pairs and diffusion barriers for their formation. Remarkably, the potential captures the diffusion-driven orthorhombic to tetragonal transition at elevated temperatures, a transformation that is difficult to describe with empirical potentials, elucidating how the formation of oxygen Frenkel pairs in the basal plane governs this order-disorder transition. The ACE potential introduced here enables large-scale, predictive atomistic simulations of defect dynamics and transport processes in YBCO, providing a powerful tool to explore its stability, performance, and functionality under realistic operating conditions. Moreover, this work proves that machine learning interatomic potentials are suitable for studies of quaternary oxides with complex chemistry.

cond-mat.mtrl-sci↗

arXiv2025

Unified machine-learning framework for property prediction and time-evolution simulation of strained alloy microstructure

We introduce a unified machine-learning framework designed to conveniently tackle the temporal evolution of alloy microstructures under the influence of an elastic field. This approach allows for the simultaneous extraction of elastic parameters from a short trajectory and for the prediction of further microstructure evolution under their influence. This is demonstrated by focusing on spinodal decomposition in the presence of a lattice mismatch eta, and by carrying out an extensive comparison between the ground-truth evolution supplied by phase field simulations and the predictions of suitable convolutional recurrent neural network architectures. The two tasks may then be performed subsequently into a cascade framework. Under a wide spectrum of misfit conditions, the here-presented cascade model accurately predicts eta and the full corresponding microstructure evolution, also when approaching critical conditions for spinodal decomposition. Scalability to larger computational domain sizes and mild extrapolation errors in time (for time sequences five times longer than the sampled ones during training) are demonstrated. The proposed framework is general and can be applied beyond the specific, prototypical system considered here as an example. Intriguingly, experimental videos could be used to infer unknown external parameters, prior to simulating further temporal evolution.

cond-mat.mtrl-sci↗