NPL - Nanoparticle Library

Overview

NPL is a Python library for the simulation and structural optimization of nanoparticles, specifically tailored for bimetallic nanoparticles. Built on the robust ASE (Atomic Simulation Environment), it enables users to easily set up and analyze complex nanoparticle structures across a range of chemical compositions and structures. NPL provides high-level abstractions, making it accessible for both beginners and experienced researchers aiming to perform detailed nanoparticle simulations.

Features

• Surrogate Model Training: Train surrogate energy models using both Topological Descriptors and Atomic Coordination Type descriptors, enabling accurate energy predictions with reduced computational costs.

• Chemical Ordering Optimization: Efficiently optimize the chemical ordering of bi- or multi-metallic nanoparticles with global optimization methods like Monte Carlo, Genetic Algorithms, and Optimal Exchange.

• Bimetallic and Multimetallic Nanoparticles: Specifically designed for complex bimetallic and multimetallic structures, supporting a variety of chemical compositions.

• ASE Integration and Extensibility: Built on ASE for simulation versatility, with modularity for custom extensions.

Contents:

• Installation Guide
  • Option 1: Clone the Repository
  • Option 2: Install via pip
  • Troubleshooting
• Examples
  • Training a Topological Descriptor Model
  • Optimization of the chemical ordering a nanoalloy
  • Optimization of the chemical ordering a multimetallic nanoparticle
  • Loading a Pretrained TOP Model and Performing Chemical Ordering Optimization