2025/03/07 by Thomas Froitzheim, Marcel Müller, Andreas Hansen +1 · 1 voice
Chemistry · Materials Science · #Bond #Business #Charge (physics) #Chemical Thermodynamics and Molecular Structure #Chemical bond #Chemical physics #Chemistry #Computational chemistry #Electrochemical Analysis and Applications #Electronegativity #Formal charge #Machine Learning in Materials Science #Materials science #Organic chemistry #Physics #Quantum mechanics #Thermodynamics
paper · pdf · doi:10.26434/chemrxiv-2025-1nxwg
openalex publication_date 2025/03/07 · openalex created_date 2025/10/10 · openalex updated_date 2026/07/14
The accurate and efficient assignment of atomic partial charges is crucial for many applications in theoretical and computational chemistry, including polarizable force fields, dispersion corrections, and charge-dependent basis sets. Classical charge models struggle to distinguish between neutral and zwitterionic fragments because, unlike quantum mechanical methods, there are no discrete electronic states. This limitation can lead to either reduced or additional artificial charge transfer (CT) at different interfragment distances. To address this issue, we propose a new version of a bond capacity electronegativity equilibration (EEQBC) model, which limits artificial CT between distant fragments in the simple EEQ framework. EEQBC offers excellent agreement with DFT-based reference charges for elements up to lawrencium (Z = 103) with mean absolute errors as low as 0.02 and 0.07 e− for random PubChem molecules and "mindless" molecules (MLMs), respectively. Thanks to its computational efficiency for both atomic charges and their analytical nuclear gradients, EEQBC is highly suitable as an initial charge guess for next-generation tight-binding methods. For seamless accessibility, EEQBC is implemented in the freely available multicharge program at: github.com/thfroitzheim/multicharge/tree/eeq-bc.