analiza aiqbench
This research introduces the Thermodynamic Interatomic Potential (TIP), a machine learning model that extends static energy calculations to full Gibbs free energy models using automatic differentiation. By training on diverse fidelity levels from quasi-harmonic approximations to molecular dynamics, TIP enables high-throughput discovery of finite-temperature phase stability and dynamically stabilized materials states. The method effectively bridges the gap between computationally expensive ensemble averages and rapid potential evaluations for crystalline systems.
Universal Thermodynamic Interatomic Potentials for Crystalline Materials
arXiv:2608.14502v1 Announce Type: cross Abstract: Free energies govern solid-state phase stability, yet computational materials discovery still relies largely on ground-state energies because free energy calculations require ensemble averages. We introduce the thermodynamic interatomic potential (TIP), which extends an interatomic potential from its static energy to a thermodynamically consistent Gibbs free energy model, with thermodynamic responses following from temperature and pressure by automatic differentiation. We implement TIP[UMA] using the universal potential UMA, train it on free energies from quasi-harmonic to molecular dynamics fidelity, and calibrate it to higher-resolution calculations or experiment. From a single evaluation, it returns the equation of state of a crystal and locates phase transitions among competing branches, including dynamically stabilized phases. Fine-tuning extends the model to alloy solubility limits and miscibility gaps. TIP makes the free energy as accessible as the potential energy, opening finite-temperature phase stability to high-throughput discovery.
Abstract:Free energies govern solid-state phase stability, yet computational materials discovery still relies largely on ground-state energies because free energy calculations require ensemble averages. We introduce the thermodynamic interatomic potential (TIP), which extends an interatomic potential from its static energy to a thermodynamically consistent Gibbs free energy model, with thermodynamic responses following from temperature and pressure by automatic differentiation. We implement TIP[UMA] using the universal potential UMA, train it on free energies from quasi-harmonic to molecular dynamics fidelity, and calibrate it to higher-resolution calculations or experiment. From a single evaluation, it returns the equation of state of a crystal and locates phase transitions among competing branches, including dynamically stabilized phases. Fine-tuning extends the model to alloy solubility limits and miscibility gaps. TIP makes the free energy as accessible as the potential energy, opening finite-temperature phase stability to high-throughput discovery.
Materials Science (cond-mat.mtrl-sci); Statistical Mechanics (cond-mat.stat-mech); Artificial Intelligence (cs.AI); Machine Learning (cs.LG); Chemical Physics (physics.chem-ph)
arXiv:2608.14502 [cond-mat.mtrl-sci]
(or arXiv:2608.14502v1 [cond-mat.mtrl-sci] for this version)
https://doi.org/10.48550/arXiv.2608.14502 arXiv-issued DOI via DataCite (pending registration)
Text integral sub licența deschisă a publicației · 168 cuvinte
Subiecte conexe
FAIRChem v2 UMA for Multidomain Atomistic Simulation across Molecules, Catalysts, Materials, Vibrations, and Molecular Dynamics
MarkTechPost · 26 iul. 2026
Thermodynamic Effects Emerge in Snyder-de Sitter Einstein Crystals
Quantum Zeitgeist · 15 aug. 2026
Columbia & Cambridge unlock quantum AI for new material models
Quantum Zeitgeist · 11 sept. 2026
PhononBench:A Large-Scale Phonon-Based Benchmark for Dynamical Stability in Crystal Generation
arXiv · 12 iun. 2026