We present an automated procedure for the fast and reliable sampling of the conformational space of typical organic and metalorganic molecules. Knowledge about a molecules' conformational space is important for the computational prediction of many physical properties, such as NMR spectra or macroscopic pKa values, where conformer ensembles have to be used to obtain properly averaged, i.e., Boltzmann-weighted data. In our approach we use the GFN<i>n</i>-xTB semi-empirical tight-binding based quantum chemistry methods in the framework of meta-dynamics to find low energy structures and conformer ensembles. The underlying meta-dynamics descriptor is a Gaussian-shaped biasing potential that uses the root-mean-square-deviation (RMSD) in Cartesian space as a metric for the collective variables. Combining the use of the Cartesian RMSD as a collective variable and QC driven meta-dynamics enables a robust and efficient screening of a molecules potential energy surface that can be extended with other conformational search procedures, such as genetic algorithms, to produce even better conformer ensembles.

Meta-dynamics-based conformational sampling with semi-empirical tight-binding methods

2019 Fall National Meeting

The official archive and repository for all ACS Meetings & Expositions abstracts and select posters or presentations.