arXiv subjects

Ethan Berger

Publications and source records attributed to Ethan Berger.

8 recordsLinked to original sources

arXiv2026

qNEP: A highly efficient neuroevolution potential with dynamic charges for large-scale atomistic simulations

Although electrostatics can be incorporated into machine-learned interatomic potentials, existing approaches are computationally very demanding, limiting large-scale, long-time simulations of electrostatics-driven phenomena such as dielectric response, infrared activity, and field-matter coupling. Here, we extend the neuroevolution potential (NEP), a highly efficient machine-learned interatomic potential, to a charge-aware framework (qNEP) by introducing explicit, environment-dependent partial charges. Each ionic partial charge is represented by a neural network as a function of the local descriptor vector, analogous to the NEP site-energy model. This formulation enables the direct prediction of the Born effective charge tensor for each ion and, consequently, the polarization. As a result, dielectric properties, infrared spectra, and coupling to external electric fields can be evaluated within a unified framework. We derive consistent expressions for the forces and virials that explicitly account for the position dependence of the partial charges. The qNEP method has been implemented in the free-and-open-source GPUMD package, with support for both Ewald summation and particle-particle particle-mesh treatments of electrostatics. We demonstrate the accuracy and efficiency of the qNEP approach through representative applications to water, Li7La3Zr2O12, BaTiO3, and a magnesium-water interface. These results show that qNEP enables accurate atomistic simulations with explicit long-range electrostatics, scalable to million-atom systems on nanosecond time scales using consumer-grade GPUs.

physics.comp-ph↗

arXiv2025

Thermal Stabilization of Defect Charge States and Finite-Temperature Charge Transition Levels

Point defects introduce localized electronic states that critically affect carrier trapping, recombination, and transport in functional materials. The associated charge transition levels (CTLs) can depend on temperature, requiring accurate treatment of vibrational and electronic free-energy contributions. In this work, we use machine-learned interatomic potentials to efficiently compute temperature-dependent CTLs for vacancies in MgO, LiF, and CsSnBr3. Using thermodynamic integration, we quantify free-energy differences between charge states and calculate the vibrational entropy contributions at finite temperatures. We find that CTLs shift with temperature in MgO, LiF and CsSnBr3 from both entropy and electronic contributions. Notably, in CsSnBr3 a neutral charge state becomes thermodynamically stable above 60 K, introducing a temperature-dependent Fermi-level window absent at 0 K. We show that the widely used static, zero-kelvin defect formalism can miss both quantitative CTL shifts and the qualitative emergence of new stable charge states.

cond-mat.mtrl-sci↗

arXiv2025

Massive Discovery of Low-Dimensional Materials from Universal Computational Strategy

Low-dimensional materials have attractive properties that drive intense efforts for novel materials discovery. However, experiments are tedious for systematic discovery, and present computational methods are often tuned to two-dimensional (2D) materials, overlooking other low-dimensional materials. Here, we combined universal machine-learning interatomic potentials (UMLIPs) and an advanced, interatomic force constant (FC) -based dimensionality classification method to make a massive discovery of novel low-dimensional materials. We first benchmarked UMLIPs' first-principles-level accuracy in quantifying FCs and calculated phonons for 35,689 materials from the Materials Project database. We then used the FC-based method for dimensionality classification to discover 9139 low-dimensional materials, including 1838 0D clusters, 1760 1D chains, 3057 2D sheets/layers, and 2484 mixed-dimensionality materials, all of which conventional geometric descriptors have not recognized. By calculating the binding energies for the discovered 2D materials, we also identified 887 sheets that could be easily or potentially exfoliated from their parent bulk structures.

cond-mat.mtrl-sci↗

arXiv2025

Screening of material defects using universal machine-learning interatomic potentials

Finding new materials with previously unknown atomic structure or materials with optimal set of properties for a specific application greatly benefits from computational modeling. Recently, such screening has been dramatically accelerated by the invent of universal machine-learning interatomic potentials that offer first principles accuracy at orders of magnitude lower computational cost. Their application to the screening of defects with desired properties or to finding new stable compounds with high density of defects, however, has not been explored. Here, we show that the universal machine-learning interatomic potentials have reached sufficient accuracy to enable large-scale screening of defective materials. We carried out vacancy calculations for 86 259 materials in the Materials Project database and analyzed the formation energies in terms of oxidation numbers. We further demonstrate the application of these models for finding new materials at or below the convex hull of known materials and for simulated etching of low-dimensional materials.

cond-mat.mtrl-sci↗

arXiv2024

Raman spectra of amino acids and peptides from machine learning polarizabilities

Raman spectroscopy is an important tool in the study of vibrational properties and composition of molecules, peptides and even proteins. Raman spectra can be simulated based on the change of the electronic polarizability with vibrations, which can nowadays be efficiently obtained via machine learning models trained on first-principles data. However, the transferability of the models trained on small molecules to larger structures is unclear and direct training on large structures in prohibitively expensive. In this work, we first train two machine learning models to predict polarizabilities of all 20 amino acids. Both models are carefully benchmarked and compared to DFT calculations, with neural network method found to offer better transferability. By combining machine learning models with classical force field molecular dynamics, Raman spectra of all amino acids are also obtained and investigated, showing good agreement with experiments. The models are further extended to small peptides. We find that adding structures containing peptide bonds to the training set greatly improves predictions even for peptides not included in training sets.

physics.comp-ph↗

arXiv2023

Polarizability Models for Simulations of Finite Temperature Raman Spectra from Machine Learning Molecular Dynamics

Raman spectroscopy is a powerful and nondestructive method that is widely used to study the vibrational properties of solids or molecules. Simulations of finite-temperature Raman spectra rely on obtaining polarizabilities along molecular dynamics trajectories, which is computationally highly demanding if calculated from first principles. Machine learning force fields (MLFF) are becoming widely used for accelerating molecular dynamics simulations, but machine-learning models for polarizability are still rare. In this work, we present and compare three polarizability models for obtaining Raman spectra in conjunction with MLFF molecular dynamics trajectories: (i) model based on projection to primitive cell eigenmodes, (ii) bond polarizability model, and (iii) symmetry-adapted Gaussian process regression (SA-GPR) using smooth overlap of atomic positions. In particular, we investigate the accuracy of these models for different systems and how much training data is required. Models are first applied to boron arsenide, where the first- and second-order Raman spectra are studied as well as the effect of boron isotopes. With MoS$_2$ we study the applicability of the models for highly anisotropic systems and for simulating resonant Raman spectra. Finally, inorganic halide perovskites CsPbBr$_3$ and CsSnBr$_3$ are studied with a particular interest in simulating the spectra across phase transitions and the evolution of the central peak. All models can be used to efficiently predict polarizabilities and are applicable to large systems and long simulation times, and while all three models were found to perform similarly for BAs and MoS$_2$, only SA-GPR offers sufficient flexibility to accurately describe complex anharmonic materials like the perovskites.

cond-mat.mes-hall↗

arXiv2023

Controlled defect production in monolayer MoS2 via electron irradiation at ultralow accelerating voltages

Control on spatial location and density of defects in 2D materials can be achieved using electron beam irradiation. Conversely, ultralow accelerating voltages (less than or equal to 5kV) are used to measure surface morphology, with no expected defect creation. We find clear signatures of defect creation in monolayer (ML) MoS2 at these voltages. Evolution of E' and A1' Raman modes with electron dose, and appearance of defect activated peaks indicate defect formation. To simulate Raman spectra of MoS2 at realistic defect distributions, while retaining density-functional theory accuracy, we combine machine-learning force fields for phonons and eigenmode projection approach for Raman tensors. Simulated spectra agree with experiments, with sulphur vacancies as suggested defects. We decouple defects, doping and carbonaceous contamination using control (hBN covered and encapsulated MoS2) samples. We observe cryogenic PL quenching and defect peaks, and find that carbonaceous contamination does not affect defect creation. These studies have applications in photonics and quantum emitters.

cond-mat.mes-hall↗

arXiv2022

Raman Spectra of Titanium Carbide MXene from Machine-Learning Force Field Molecular Dynamics

MXenes represent one of the largest class of 2D materials with promising applications in many fields and their properties tunable by the surface group composition. Raman spectroscopy is expected to yield rich information about the surface composition, but the interpretation of measured spectra has proven challenging. The interpretation is usually done via comparison to simulated spectra, but there are large discrepancies between the experimental and earlier simulated spectra. In this work, we develop a computational approach to simulate Raman spectra of complex materials that combines machine-learning force-field molecular dynamics and reconstruction of Raman tensors via projection to pristine system modes. The approach can account for the effects of finite temperature, mixed surfaces, and disorder. We apply our approach to simulate Raman spectra of titanium carbide MXene and show that all these effects must be included in order to properly reproduce the experimental spectra, in particular the broad features. We discuss the origin of the peaks and how they evolve with surface composition, which can then be used to interpret experimental results.

physics.comp-ph↗