Learning to Converge: Warm-Starting DFTB Self-Consistent Charges with Machine Learning
Maximilian L. Ach, Karsten Reuter, Chiara Panosetti
arXiv:2607.09304·cond-mat.mtrl-sci·Published 2026-07-10
Semiempirical electronic structure methods such as Density-Functional Tight-Binding (DFTB) offer a computationally efficient approach to molecular and materials simulations, bridging the gap between first-principles accuracy and classical force field speed while retaining full access to electronic properties. However, DFTB calculations based on self-consistent charge (SCC) schemes can still suffer from slow convergence, particularly for complex molecular and materials systems, making the iterative procedure a significant bottleneck in large-scale simulations and high-throughput workflows. We present a machine learning approach that accelerates DFTB simulations by predicting optimal initial atomic charges. Using element-specific models based on the Smooth Overlap of Atomic Positions descriptor and kernel ridge regression, we train charge models on reference calculations and demonstrate that ML-predicted initial charges consistently and significantly improve SCC convergence across diverse chemical systems including organic molecules, biomolecules, water clusters, transition metal oxides and solid electrolytes.
TopicsMaterials Science & Condensed Matter
Tagselectronic-structure
arXiv categoriescond-mat.mtrl-sci
arXiv abstract pagePDF