arXiv subjects

Edmanuel Torres

Publications and source records attributed to Edmanuel Torres.

5 recordsLinked to original sources

arXiv2026

Machine-Learning Potentials for sodium-potassium chloride mixtures: Predicting thermophysical properties and phase behavior of multicomponent salts

Predicting the properties of multicomponent molten salts using density functional theory (DFT) remains challenging because the spatial and temporal scales required to evaluate transport properties and phase behavior are computationally prohibitive. In this work, we develop a moment tensor potential trained using a a DFT dataset of NaCl, KCl, NaCl-KCl mixtures, and the NaK alloy, enabling large-scale molecular dynamics simulations across wide ranges of temperatures and compositions. We systematically evaluate the effect of D3 dispersion corrections and apply the resulting potential to predict liquid densities, diffusion coefficients, radial distribution functions, heat capacities, thermal conductivities, and the NaCl-KCl phase diagram. The model successfully reproduces many temperature- and composition-dependent trends. However, systematic deviations in several absolute properties persist, highlighting the importance of experimental validation and calibration. These findings support a hybrid modeling framework in which first-principles-informed machine-learning potentials provide transferable predictive capability and mechanistic insight, while experimental data incorporated during model development or subsequent engineering assessments is necessary to improve quantitative accuracy.

cond-mat.mtrl-sci↗

arXiv2025

A Kokkos-Accelerated Moment Tensor Potential Implementation for LAMMPS

We present a Kokkos-accelerated implementation of the Moment Tensor Potential (MTP) for LAMMPS, designed to improve both computational performance and portability across CPUs and GPUs. This package introduces an optimized CPU variant--achieving up to 2x speedups over existing implementations--and two new GPU variants: a thread-parallel version for large-scale simulations and a block-parallel version optimized for smaller systems. It supports three core functionalities: standard inference, configuration-mode active learning, and neighborhood-mode active learning. Benchmarks and case studies demonstrate efficient scaling to million-atom systems, substantially extending accessible length and time scales while preserving the MTP's near-quantum accuracy and native support for uncertainty quantification.

cond-mat.mtrl-sci↗

arXiv2025

Small-Cell-Based Fast Active Learning of Machine Learning Interatomic Potentials

Machine learning interatomic potentials (MLIPs) are often trained with on-the-fly active learning, where sampled configurations from atomistic simulations are added to the training set. However, this approach is limited by the high computational cost of ab initio calculations for large systems. Recent works have shown that MLIPs trained on small cells (1-8 atoms) rival the accuracy of large-cell models (100s of atoms) at far lower computational cost. Herein, we refer to these as small-cell and large-cell training, respectively. In this work, we iterate on earlier small-cell training approaches and characterize our resultant small-cell protocol. Potassium and sodium-potassium systems were studied: the former, a simpler system benchmarked in detail; the latter, a more complex binary system for further validation. Our small-cell training approach achieves up to two orders of magnitude of cost savings compared to large-cell (54-atom) training, with some training runs requiring fewer than 120 core-hours. Static and thermodynamic properties predicted using the MLIPs were evaluated, with small-cell training in both systems yielding strong ab initio agreement. Small cells appear to encode the necessary information to model complex large-scale phenomena--solid-liquid interfaces, critical exponents, diverse concentrations--even when the training cells themselves are too small to accommodate these phenomena. Based on these tests, we provide analysis and recommendations.

cond-mat.mtrl-sci↗

arXiv2013

Extension of the B3LYP - Dispersion-Correcting Potential Approach to the Accurate Treatment of both Inter- and Intramolecular Interactions

We recently showed that dispersion-correcting potentials (DCPs), atom-centered Gaussian-type functions developed for use with B3LYP (J. Phys. Chem. Lett. 2012, 3, 1738-1744) greatly improved the ability of the underlying functional to predict non-covalent interactions. However, the application of B3LYP-DCP for the {\beta}-scission of the cumyloxyl radical led a calculated barrier height that was over-estimated by ca. 8 kcal/mol. We show in the present work that the source of this error arises from the previously developed carbon atom DCPs, which erroneously alters the electron density in the C-C covalent-bonding region. In this work, we present a new C-DCP with a form that was expected to influence the electron density farther from the nucleus. Tests of the new C-DCP, with previously published H-, N- and O-DCPs, with B3LYP-DCP/6-31+G(2d,2p) on the S66, S22B, HSG-A, and HC12 databases of non-covalently interacting dimers showed that it is one of the most accurate methods available for treating intermolecular interactions, giving mean absolute errors (MAEs) of 0.19, 0.27, 0.16, and 0.18 kcal/mol, respectively. Additional testing on the S12L database of complexation systems gave an MAE of 2.6 kcal/mol, showing that the B3LYP-DCP/6-31+G(2d,2p) approach is one of the best-performing and feasible methods for treating large systems dominated by non-covalent interactions. Finally, we showed that C-C making/breaking chemistry is well-predicted using the newly developed DCPs. In addition to predicting a barrier height for the {\beta}-scission of the cumyloxyl radical that is within 1.7 kcal/mol of the high-level value, application of B3LYP-DCP/6-31+G(2d,2p) to 10 databases that include reaction barrier heights and energies, isomerization energies and relative conformation energies gives performance that is amongst the best of all available dispersion-corrected density-functional theory approaches.

physics.chem-ph↗

arXiv2013

Chirp-wave Expansion of the Electron Wavefunctions in Atoms

The description of the electron wavefunctions in atoms is generalized to the fractional Fourier series. This method introduces a continuous and infinite number of chirp basis sets with linear variation of the frequency to expand the wavefunctions, in which plane-waves are a special case. The chirp characteristics of each basis set can be adjusted through a single parameter. Thus, the basis set cutoff can be optimized variationally. The approach is tested with the expansion of the electron wavefunctions in atoms, and it is shown that chirp basis sets substantially improve the convergence in the description of the electron density. We have found that the natural oscillations of the electron core states are efficiently described in chirp-waves.

physics.comp-ph↗