arXiv subjects

Hossein Tahmasbi

Publications and source records attributed to Hossein Tahmasbi.

8 recordsLinked to original sources

arXiv2025

Benchmarking Universal Machine Learning Interatomic Potentials on Elemental Systems

The rapid emergence of universal Machine Learning Interatomic Potentials (uMLIPs) has transformed materials modeling. However, a comprehensive understanding of their generalization behavior across configurational space remains an open challenge. In this work, we introduce a benchmarking framework to evaluate both the equilibrium and far-from-equilibrium performance of state-of-the-art uMLIPs, including three MACE-based models, MatterSim, and PET-MAD. Our assessment utilizes Equation-of-State (EOS) tests to evaluate near-equilibrium properties, such as bulk moduli and equilibrium volumes, alongside extensive Minima Hopping (MH) structural searches to probe the global Potential Energy Surface (PES). Here, we assess universality within the fundamental limit of unary (elemental) systems, which serve as a necessary baseline for broader chemical generalization and provide a framework that can be systematically extended to multicomponent materials. We find that while most models exhibit high accuracy in reproducing equilibrium volumes for transition metals, significant performance gaps emerge in alkali and alkaline earth metal groups. Crucially, our MH results reveal a decoupling between search efficiency and structural fidelity, highlighting that smoother learned PESs do not necessarily yield more accurate energetic landscapes.

cond-mat.mtrl-sci↗

arXiv2025

Scalable Machine Learning Model for Energy Decomposition Analysis in Aqueous Systems

Energy decomposition analysis (EDA) based on absolutely localized molecular orbitals provides detailed insights into intermolecular bonding by decomposing the total molecular binding energy into physically meaningful components. Here, we develop a neural network EDA model capable of predicting the electron delocalization energy component of water molecules, which captures the stabilization arising from charge transfer between occupied absolutely localized molecular orbitals of one molecule and the virtual orbitals of another. Exploiting the locality assumption of the electronic structure, our model enables accurate prediction of electron delocalization energies for molecular systems far beyond the size accessible to conventional density functional theory calculations, while maintaining its accuracy. We demonstrate the applicability of our approach by modeling hydration effects in large molecular complexes, specifically in metal-organic frameworks.

physics.chem-ph↗

arXiv2023

Machine Learning-Driven Structure Prediction for Iron Hydrides

We created a computational workflow to analyze the potential energy surface (PES) of materials using machine-learned interatomic potentials in conjunction with the minima hopping algorithm. We demonstrate this method by producing a versatile machine-learned interatomic potential for iron hydride via a neural network using an iterative training process to explore its energy landscape under different pressures. To evaluate the accuracy and comprehend the intricacies of the PES, we conducted comprehensive crystal structure predictions using our neural network-based potential paired with the minima hopping approach. The predictions spanned pressures ranging from ambient to 100 GPa. Our results reproduce the experimentally verified global minimum structures such as \textit{dhcp}, \textit{hcp}, and \textit{fcc}, corroborating previous findings. Furthermore, our in-depth exploration of the iron hydride PES at different pressures has revealed complex alterations and stacking faults in these phases, leading to the identification of several new low-enthalpy structures. This investigation has not only confirmed the presence of regions of established FeH configurations but has also highlighted the efficacy of using data-driven, extensive structure prediction methods to uncover the multifaceted PES of materials.

cond-mat.mtrl-sci↗

arXiv2023

Transferable Interatomic Potentials for Aluminum from Ambient Conditions to Warm Dense Matter

We present a study on the transport and materials properties of aluminum spanning from ambient to warm dense matter conditions using a machine-learned interatomic potential (ML-IAP). Prior research has utilized ML-IAPs to simulate phenomena in warm dense matter, but these potentials have often been calibrated for a narrow range of temperature and pressures. In contrast, we train a single ML-IAP over a wide range of temperatures, using density functional theory molecular dynamics (DFT-MD) data. Our approach overcomes computational limitations of DFT-MD simulations, enabling us to study transport and materials properties of matter at higher temperatures and longer time scales. We demonstrate the ML-IAP transferability across a wide range of temperatures using molecular-dynamics (MD) by examining the thermal conductivity, diffusion coefficient, viscosity, sound velocity, and ion-ion structure factor of aluminum up to about 60,000 K, where we find good agreement with previous theoretical data.

cond-mat.mtrl-sci↗

arXiv2021

Large Scale Structure Prediction of Near-Stoichiometric Magnesium Oxide Based on a Machine-Learned Interatomic Potential: Novel Crystalline Phases and Oxygen-Vacancy Ordering

Using a fast and accurate neural network potential we are able to systematically explore the energy landscape of large unit cells of bulk magnesium oxide with the minima hopping method. The potential is trained with a focus on the near-stoichiometric compositions, in particular on suboxides, i.e., Mg$_x$O$_{1-x}$ with $0.50<x<0.60$. Our extensive exploration demonstrates that for bulk stoichiometric compounds, there are several new low-energy rocksalt-like structures in which Mg atoms are octahedrally six--coordinated and form trigonal prismatic motifs with different stacking sequences. Furthermore, we find a dense spectrum of novel non-stoichiometric crystal phases of Mg$_x$O$_{1-x}$ for each composition of $x$. These structures are mostly similar to the rock salt structure with octahedral coordination and five--coordinated Mg atoms. Due to the removal of one oxygen atom, the energy landscape becomes more glass-like with oxygen-vacancy type structures that all lie very close to each other energetically. For the same number of magnesium and oxygen atoms our oxygen-deficient structures are lower in energy if the vacancies are aligned along lines or planes than rock salt structures with randomly distributed oxygen vacancies. We also found the putative global minima configurations for each composition of the non-stoichiometric suboxide structures. These structures are predominantly composed of (111) slabs of the rock salt structure which are terminated with Mg atoms at the top and bottom, and are stacked in different sequences along the $z$-direction. Like other Magnéli-type phases, these structures have properties that differ considerably from their stoichiometric counterparts such as low lattice thermal conductivity and high electrical conductivity.

cond-mat.mtrl-sci↗

arXiv2021

An automated approach for developing neural network interatomic potentials with FLAME

The performance of machine learning interatomic potentials relies on the quality of the training dataset. In this work, we present an approach for generating diverse and representative training data points which initiates with \it{ab initio} calculations for bulk structures. The data generation and potential construction further proceed side-by-side in a cyclic process of training the neural network and crystal structure prediction based on the developed interatomic potentials. All steps of the data generation and potential development are performed with minimal human intervention. We show the reliability of our approach by assessing the performance of neural network potentials developed for two inorganic systems.

cond-mat.mtrl-sci↗

arXiv2019

FLAME: a library of atomistic modeling environments

FLAME is a software package to perform a wide range of atomistic simulations for exploring the potential energy surfaces (PES) of complex condensed matter systems. The range of methods include molecular dynamics simulations to sample free energy landscapes, saddle point searches to identify transition states, and gradient relaxations to find dynamically stable geometries. In addition to such common tasks, FLAME implements a structure prediction algorithm based on the minima hopping method (MHM) to identify the ground state structure of any system given solely the chemical composition, and a framework to train a neural network potential to reproduce the PES from $\textit{ab initio}$ calculations. The combination of neural network potentials with the MHM in FLAME allows a highly efficient and reliable identification of the ground state as well as metastable structures of molecules and crystals, as well as of nano structures, including surfaces, interfaces, and two-dimensional materials. In this manuscript, we provide detailed descriptions of the methods implemented in the FLAME code and its capabilities, together with several illustrative examples.

physics.comp-ph↗

arXiv2017

A two-dimensional hexagonal sheet of TiO$_2$

We report on the ab initio discovery of a novel putative ground state for quasi two-dimensional TiO$_2$ through a structural search using the minima hopping method with an artificial neural network potential. The structure is based on a honeycomb lattice and is energetically lower than the experimentally reported lepidocrocite sheet by 7~meV/atom, and merely 13~meV/atom higher in energy than the ground state rutile bulk structure. According to our calculations, the hexagonal sheet is stable against mechanical stress, it is chemically inert and can be deposited on various substrates without disrupting the structure. Its properties differ significantly from all known TiO$_2$ bulk phases with a large gap of 5.05~eV that can be tuned through strain engineering.

cond-mat.mtrl-sci↗