activity
20242026
collaborators

6 papers

cond-mat.mtrl-sci2026

Neural surrogates for crystal growth dynamics with variable supersaturation: explicit vs. implicit conditioning

Matteo Rigoni, Daniele Lanzoni, Francesco Montalenti +1

Simulations of crystal growth are performed by using Convolutional Recurrent Neural Network surrogate models, trained on a dataset of time sequences computed by numerical integrati…

cond-mat.stat-mech2025

Learning Kinetic Monte Carlo stochastic dynamics with Deep Generative Adversarial Networks

Daniele Lanzoni, Olivier Pierre-Louis, Roberto Bergamaschini +1

We show that Generative Adversarial Networks (GANs) may be fruitfully exploited to learn stochastic dynamics, surrogating traditional models while capturing thermal fluctuations. S…

cond-mat.mtrl-sci2025

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

Andrea Fantasia, Daniele Lanzoni, Niccolò Di Eugenio +3

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 appr…

cond-mat.mes-hall2024

Phase-field Modelling of Anisotropic Solid-State Dewetting on Patterned Substrates

Emma Radice, Marco Salvalaglio, Roberto Bergamaschini

We present a phase-field model for simulating the solid-state dewetting of anisotropic crystalline films on non-planar substrates. This model exploits two order parameters to trace…

cond-mat.mtrl-sci2024

Interface energies of Ga2O3 phases with the sapphire substrate and the phase-locked epitaxy of metastable structures explained

Ilaria Bertoni, Aldo Ugolotti, Emilio Scalise +2

Despite the extensive work carried out on the epitaxial growth of Ga2O3, a fundamental understanding of the nucleation of its different metastable phases is still lacking. Here we…

cond-mat.mes-hall2024

Extreme time extrapolation capabilities and thermodynamic consistency of physics-inspired Neural Networks for the 3D microstructure evolution of materials via Cahn-Hilliard flow

Daniele Lanzoni, Andrea Fantasia, Roberto Bergamaschini +2

A Convolutional Recurrent Neural Network (CRNN) is trained to reproduce the evolution of the spinodal decomposition process in three dimensions as described by the Cahn-Hilliard eq…