Press "Enter" to skip to content

Machine Learning Can Predict Electron Densities with DFT Accuracy

A machine learning model has bypassed the necessity to use wavefunction or density functional theory (DFT) calculations to find out electron densities. It will enable chemists to rapidly determine properties that depend upon the electron density of enormous systems similar to van der Waals forces, halogen bonding, and C-H–π interactions. These non-covalent interactions can maintain insight into the binding of host-guest systems or favored enantiomers within reaction pathways where subtle attractions could also stabilize intermediates and transition states.

The electron density distribution is among the most reliable tools at the disposal of a computational chemist. From the electron density, properties similar to charges, dipoles, and electrostatic interaction energies could be determined. Precisely accounting for these is essential for the predictive power of many quantum chemistry techniques similar to computing infrared intensities or determining non-covalent interactions.

Computing the electron density could be difficult and time consuming for large systems using traditional wavefunction or DFT strategies. To overcome this concern, Clémence Corminboeuf, Michele Ceriotti and colleagues at the Swiss Federal Institute of Technology (EPFL) have developed a machine studying model that can predict the electron density from only atomic coordinates. ‘The breakthrough is to be able to accurately predict, in a couple of minutes at most, the electron density of complicated molecules without any quantum chemical computation,’ explains team member Alberto Fabrizio.

‘I feel it’s an exciting approach, both in terms of prediction errors and transferability to small and large systems,’ comments Natalie Fey who researches computational inorganic chemistry at the University of Bristol, UK.

Be First to Comment

Leave a Reply

Your email address will not be published. Required fields are marked *