atomistic-simulation

Apple Silicon-native molecular dynamics and DFT runtime built on MLX and Metal — the GPU on your Mac, no CUDA or cloud.

For visual atomistic modeling

Format-extensible magnetic-moment post-processing for spin molecular dynamics; first adapter supports Extended XYZ and GPUMD.

ASE with Rust hot paths — 12× faster neighbor lists, 6× faster VASP IO, 3.5× faster extxyz. Drop-in: pip install ase-fast

Atomistic molecular dynamics compression and mechanical descriptor analysis for Hf-Nb-Ta-Ti-Zr refractory alloys.

Legacy-compatible Python generator for initial spiral-carbon atomistic structures used in mechanical, thermal, and interfacial simulations.

MLIPs as ASE calculators, benchmarked on BCC TiZrNb properties: a distilled compact potential vs pretrained foundation potentials, accuracy vs speed

C++ research code for constructing dislocations in atomistic simulations and generating dislocation-containing LAMMPS atomic configurations.

Molecular dynamics study of tensile deformation and layer-resolved fracture in graphene/MoSe2/graphene heterostructures using LAMMPS.

From-scratch SchNet-style message-passing neural network potential for BCC TiZrNb, benchmarked head to head against descriptor baselines at matched data and budget. Surrogate teacher labels (CHGNet), real-DFT anchors, leakage-safe splits, autograd forces, NVE stability.

Query-by-committee active learning for TiZrNb machine-learned interatomic potentials: a from-scratch deep ensemble, structural label-budget accounting, and an honest benchmark where random selection wins on a mixed pool and the committee earns a 2.2x label saving on a scarce one

From-scratch Behler-Parrinello neural network potential for BCC TiZrNb: original ACSF descriptors, autograd forces, and a leakage-disciplined benchmark against tuned linear and pair-potential baselines (surrogate teacher labels, not DFT)

This project demonstrates how atomistic simulations are used to predict and analyze the behavior of atoms and molecules, providing detailed tutorials and simulations in a Jupyter Notebook.

Atomistic simulation of Cu in ASE with the EMT potential, including equilibrium lattice-parameter determination across SC/BCC/FCC structures, extraction of FCC elastic constants from energy–strain relations, and assessment of the strain range over which linear elasticity remains valid using NVT Langevin MD benchmarks.

Kinetic Monte Carlo simulations of Al diffusion in Cu and Au diffusion in Si via the kick-out mechanism. Includes atomistic and coarse-grained models, time-dependent temperature ramps, and animated defect evolution.