With the increasing production and release of new chemicals, it is essential to have efficient and accurate theoretical tools for predicting environmentally important properties such as partition coefficients or vapor pressure of compounds. Over the years, many methods have been proposed for this purpose, and popular
methods such as quantitative structure-activity relationships (QSAR) provide reasonable results for common compounds. However, these methods are often highly parameterized and can be unreliable outside their parameterization regime. Moreover, conformational flexibility, i.e., different 3D structures in a conformer ensemble (CE), is often ignored. We present an advanced multi-level quantum mechanical (QM) workflow termed CRENSO, that thoroughly takes into account the CE as well as solvation and thermostatistical effects. The CRENSO workflow consists of two main parts. The creation of an CE using the CREST program and the refinement of this ensemble at a higher level of theory using the CENSO algorithm based on accurate DFT calculations. We have recently shown that this workflow is able to accurately compute the partition coefficients K_OW, K_HdA, and K_OAfor very flexible compounds with errors in the range of 0.5 log units in a reasonable computation time. One of the main advantages of the workflow, besides the high accuracy, is its general applicability to basically arbitrary classes of compounds. Furthermore, similar accuracy for other properties, such as vapor pressure is expected without the need for reparameterization or adjustment.
Examples from a wide range of different, environmentally relevant molecules are discussed.

Marcel Stahn

Research Assistant, University of Bonn

Appearances