Keep PACE with GRACE
Towards quantum accuracy in atomistic modeling of materials
Matous Mrovec, Minaam Qamar, Yanyan Liang, Sergei Starikov, Anton Bochkarev,
Yury Lysogorskiy, Ralf Drautz
The atomic cluster expansion (ACE) provides a general and mathematically complete representation of the properties of interacting atoms that fulfills fundamental invariances required to obtain a physically meaningful representation of energy. Its recent extension, graph atomic cluster expansion (GRACE), enables to capture also semi-local interactions and is applicable to multi-component material systems with unlimited number of interacting elements. Existing ACE models for elements as well as multicomponent systems have been shown to be superior or comparable to the best currently available machine learning (ML) potentials in terms of accuracy while being significantly more computational performant (PACE=performant implementation of ACE). These models can be developed either as general-purpose parametrizations with a broad transferability or as purpose-specific parametrizations tailored to a specific application. We will present several examples of ACE models and their applications for various classes of materials including covalent elements, metals and alloys, oxides, and molecular systems.
R. Drautz, Atomic cluster expansion for accurate and transferable interatomic potentials, Phys. Rev. B 99, 014104 (2019)
Y. Lysogorskiy, C.v.d. Oord, A. Bochkarev, et al. Performant implementation of the atomic cluster expansion (PACE) and application to copper and silicon, npj Comput Mater 7, 97 (2021)
A. Bochkarev, Y. Lysogorskiy, S. Menon, et al, Efficient parametrization of the atomic cluster expansion, Phys. Rev. Materials 6, 013804 (2022)
M. Qamar, M. Mrovec, Y. Lysogorskiy, et al. Atomic Cluster Expansion for Quantum-Accurate Large-Scale Simulations of Carbon, J. Chem. Theory Comput. 19, 5151 (2023)
M. Rinaldi, M. Mrovec, A. Bochkarev, et al. Non-collinear magnetic atomic cluster expansion for iron, npj Comput Mater 10, 12 (2024)
A. Bochkarev, Y. Lysogorskiy, R. Drautz, Graph Atomic Cluster Expansion for Semilocal Interactions beyond Equivariant Message Passing, Physical Review X 14, 021036 (2024)
A. Ngoipala, C. Schott, V. Briega‐Martos, M. Qamar, M. Mrovec, et al., Hydride‐Induced Reconstruction of Pd Electrode Surfaces: A Combined Computational and Experimental Study, Advanced Materials 37, 2410951 (2024)
B. Bienvenu, M. Todorova, J. Neugebauer, D. Raabe, M. Mrovec, et al., Development of an atomic cluster expansion potential for iron and its oxides, npj Computational Materials 11, 81 (2025)