Data-efficient fine-tuning of foundational models for first-principles quality sublimation enthalpies

2024/05/30 by Harveen Kaur, Kaur, Harveen, Flaviano Della Pia +15 · 13 citations
Chemistry · Materials Science · #Chemical Physics (physics.chem-ph) #Chemical Thermodynamics and Molecular Structure #FOS: Physical sciences #Materials Science (cond-mat.mtrl-sci) #Thermal and Kinetic Analysis #thermodynamics and calorimetric analyses

paper · pdf · doi:10.48550/arxiv.2405.20217

openalex publication_date 2024/05/30 · openalex created_date 2025/10/10 · openalex updated_date 2026/07/28

Abstract

Calculating sublimation enthalpies of molecular crystal polymorphs is relevant to a wide range of technological applications. However, predicting these quantities at first-principles accuracy -- even with the aid of machine learning potentials -- is a challenge that requires sub-kJ/mol accuracy in the potential energy surface and finite-temperature sampling. We present an accurate and data-efficient protocol based on fine-tuning of the foundational MACE-MP-0 model and showcase its capabilities on sublimation enthalpies and physical properties of ice polymorphs. Our approach requires only a few tens of training structures to achieve sub-kJ/mol accuracy in the sublimation enthalpies and sub 1 % error in densities for polymorphs at finite temperature and pressure. Exploiting this data efficiency, we explore simulations of hexagonal ice at the random phase approximation level of theory at experimental temperatures and pressures, calculating its physical properties, like pair correlation function and density, with good agreement with experiments. Our approach provides a way forward for predicting the stability of molecular crystals at finite thermodynamic conditions with the accuracy of correlated electronic structure theory.

Cited by

• How accurate are foundational machine learning interatomic potentials for heterogeneous catalysis?
• On the role of nuclear quantum effects on the stability of peptides
• Comparing the latent features of universal machine-learning interatomic potentials
• Fine-Tuning Unifies Foundational Machine-learned Interatomic Potential Architectures at ab initio Accuracy
• Predicting Crystal Structures and Ionic Conductivities in Li₃YCl_6-xBr_x Halide Solid Electrolytes Using a Fine-Tuned Machine Learning Interatomic Potential
• Comparing fine-tuning strategies of MACE machine learning force field for modeling Li-ion diffusion in LiF for batteries
• Scalable Foundation Interatomic Potentials via Message-Passing Pruning and Graph Partitioning
• Reaction dynamics of lithium-mediated electrolyte decomposition using machine learning potentials
• Expanding the search space of high entropy oxides and predicting synthesizability using machine learning interatomic potentials
• aims-PAX: Parallel Active eXploration for the automated construction of Machine Learning Force Fields
• Multi-head committees enable direct uncertainty prediction for atomistic foundation models
• Machine Learning Phonon Spectra for Fast and Accurate Optical Lineshapes of Defects
• Autonomous thermodynamically informed database generation for machine-learned interatomic potentials and application to magnesium

Related

• Accurate and efficient machine learning interatomic potentials for finite temperature modeling of molecular crystals
• DMC-ICE13: ambient and high pressure polymorphs of ice from Diffusion Monte Carlo and Density Functional Theory
• Statistical mechanical approximations to more efficiently determine polymorph free energy differences for small organic molecules
• A probabilistic framework for crystal structure denoising, phase classification, and order parameters
• Machine learning for predictive condensed-phase simulation