1. Introduction

The Spectral Energy Density (SED) technique employed by pySED is a method that can directly predict phonon dispersion relations and their corresponding lifetimes from atomic velocities obtained in large-scale molecular dynamics (MD) simulations. SED represents the intensity of the kinetic energy distribution of a system across different phonon modes, effectively providing a dispersion weighted by the kinetic energy carried by each phonon.

pySED can read and process large-scale simulation trajectories from both LAMMPS and GPUMD, enhancing its versatility and applicability in various research scenarios.

1.1. Main Features

• SED Visualization: Generates high-quality SED images for detailed analysis of phonon dispersion and lifetimes.

• Lorentzian Fitting: Utilizes Lorentzian fitting to accurately model SED peaks.

• Comprehensive Peak Fitting: Capable of fitting all peaks simultaneously to output phonon lifetimes or fitting them individually as needed.

• User-Friendly Interface: Designed for simplicity and ease of use.

• Qualitative Phonon Lifetimes: Provides qualitative insights into phonon lifetimes based on SED analysis.

• Parallel Processing: Supports parallel execution to accelerate SED computations and obtain results quickly.