98%
921
2 minutes
20
Large-scale atomistic simulations rely on interatomic potentials, providing an efficient representation of atomic energies and forces. Modern machine-learning (ML) potentials provide the most precise representation compared to electronic structure calculations, while traditional potentials provide a less precise but computationally much faster representation and, thus, allow simulations of larger systems. We present a method to combine a traditional and a ML potential into a multi-resolution description, leading to an adaptive-precision potential with an optimum of performance and precision in large, complex atomistic systems. The required precision is determined per atom by a local structure analysis and updated automatically during simulation. We use copper as demonstrator material with an embedded atom model as classical force field and an atomic cluster expansion (ACE) as ML potential, but, in principle, a broader class of potential combinations can be coupled by this method. The approach is developed for the molecular-dynamics simulator LAMMPS and includes a load-balancer to prevent problems due to the atom dependent force-calculation times, which makes it suitable for large-scale atomistic simulations. The developed adaptive-precision copper potential represents the ACE-forces with a precision of 10 me V/Å and the ACE-energy exactly for the precisely calculated atoms in a nanoindentation of 4 × 106 atoms calculated for 100 ps and shows a speedup of 11.3 compared with a full ACE simulation.
Download full-text PDF |
Source |
---|---|
http://dx.doi.org/10.1063/5.0245877 | DOI Listing |
ACS Appl Mater Interfaces
September 2025
School of Materials Science and Engineering, Beihang University, Beijing 100191, P. R. China.
Nanostructured cubic boron nitride (NS-cBN) has attracted significant attention due to its high hardness and excellent thermal stability, yet a systematic strategy to balance strength and toughness through atomically structural design remains elusive. Here, we integrate plasticity theory with large-scale atomistic simulations to elucidate the size-dependent roles of internal defects, i.e.
View Article and Find Full Text PDFJ Chem Theory Comput
September 2025
Molecular Microbiology and Structural Biochemistry (MMSB), UMR 5086 CNRS & Université Claude Bernard Lyon 1, Lyon 69367, France.
The Martini model is a coarse-grained force field allowing simulations of biomolecular systems as well as a range of materials including different types of nanomaterials of technological interest. Recently, a new version of the force field (version 3) has been released that includes new parameters for lipids, proteins, carbohydrates, and a number of small molecules, but not yet carbon nanomaterials. Here, we present new Martini models for three major types of carbon nanomaterials: fullerene, carbon nanotubes, and graphene.
View Article and Find Full Text PDFACS Appl Mater Interfaces
September 2025
Department of Chemical Engineering, National Taiwan University, No. 1, Sec. 4, Roosevelt Road, Taipei 106319, Taiwan.
Metal-organic frameworks (MOFs), known for their highly versatile nature, show considerable promise as adsorbents and membranes for water-related applications such as water harvesting and water filtration. One of the key factors that may influence their efficiency is the diffusion of water within MOFs. However, the behaviors and mechanisms of water diffusion in MOFs remain relatively underexplored.
View Article and Find Full Text PDFCommun Biol
August 2025
Research Unit of Structural Chemistry & Computational Biophysics, Leibniz-Forschungsinstitut für Molekulare Pharmakologie, Berlin, Germany.
Mammalian cyclic nucleotide-gated (CNG) ion channels play a fundamental role in signal transduction within the visual and olfactory sensory cells, converting external stimuli into electrical signals. Here, using large-scale atomistic molecular dynamics (MD) simulations of three different constructs under applied transmembrane voltages, we uncover the atomistic mechanism of monovalent cation permeation in the homotetrameric CNGA1 channel. Owing to the high plasticity and large dimensions of its selectivity filter (SF), monovalent cation binding within the SF of the CNGA1 channel is more dynamic and diffuse compared to that in potassium-selective and hyperpolarization-activated cyclic nucleotide-gated (HCN) channels.
View Article and Find Full Text PDFJ Am Chem Soc
September 2025
Department of Chemistry, Queen Mary University of London, Mile End Road, London E1 4NS, U.K.
Dopant size is known to influence oxygen vacancy-mediated conduction pathways and ionic conductivity in doped ceria, yet the underlying atomic-scale mechanisms remain unclear. Here, we combine neutron total scattering and large-scale atomistic simulations to analyze the local defect structures of two representative doped ceria systems: CeGdO (GDC) and CeNdO (NDC). The local structure of GDC, a commercially used ion conductor, is investigated for the first time using neutron total scattering on Gd-enriched samples.
View Article and Find Full Text PDF