Workflow Overview

Conceptual Pipeline

The ML Doping Workflow implements a fully automated, multi-stage surrogate pipeline for the exploration of doped crystalline materials.

Pipeline Structure

Enumeration → Ranking → Relaxation → Filtering → Band Gap → Formation Energy → Database

Stages

1. Symmetry-reduced dopant enumeration

2. ML-based energy ranking (M3GNet)

3. Full cell relaxation (M3GNet + FIRE)

4. Energy-window filtering

5. Band gap prediction (ALIGNN)

6. Formation energy evaluation using ML chemical potentials

7. Database assembly