A Physics-Chemistry-Materials-Informed Neural Digital Twin for Autonomous Design, Characterization, and Optimization of Green-Synthesized Antibacterial Silver Nanoparticles

This repository contains the datasets and source codes accompanying the publication "Physics-Constrained Multi-Informed Neural Network Digital Twin for Green Synthesis of Silver Nanoparticles." It includes a physics-consistent virtual dataset of 2,500 simulated green synthesis experiments, the implementation of the proposed Physics-Constrained Multi-Informed Neural Network Digital Twin (PCMINN-DT), baseline neural network models, training and evaluation scripts, pretrained model weights, and the outputs used to generate the results presented in the paper. The repository enables full reproducibility of the study and provides a foundation for researchers to validate, benchmark, and extend the proposed framework for AI-driven nanomaterials design, scientific machine learning, and digital twin applications.