Back to Search Start Over

Modeling Molecular Crystals by QM/MM: Self-Consistent Electrostatic Embedding for Geometry Optimizations and Molecular Property Calculations in the Solid.

Authors :
Bjornsson R
Bühl M
Source :
Journal of chemical theory and computation [J Chem Theory Comput] 2012 Feb 14; Vol. 8 (2), pp. 498-508. Date of Electronic Publication: 2012 Jan 09.
Publication Year :
2012

Abstract

We present an approach to model molecular crystals using an adaptive quantum mechanics/molecular mechanics (QM/MM) based protocol. The molecule of interest (or a larger cluster thereof) is described at an appropriate QM level and is embedded in a large array of MM atoms built up from crystal structure information. The nonbonded MM force field consists of atom-centered point charges and Lennard-Jones potentials using van der Waals parameters from the UFF force field. The point charges are initially derived from a single molecule DFT calculation and are then updated self-consistently in the field of point charges. Additional charges are fitted around the MM cluster to correct for missing long-range electrostatic effects. The geometry of the central complex can then be relaxed by quantum chemical calculations in the surrounding MM reaction field, hence capturing solid-state effects on the geometry. We demonstrate the accuracy of this approach for geometry optimization by successful modeling of the huge gas-to-solid bond contraction of HCN-BF3, the ability to reproduce periodic-DFT quality local geometries of solid VOCl3, and the geometry of [Ru(η(5)-Cp*)(η(3)-CH2CHCHC6H5)(NCCH3)2](2+), a difficult ruthenium allyl complex in the solid state. We further show that this protocol is well suited for subsequent molecular property calculations in the solid state (where accurate relaxed geometries are often required) as exemplified by transition metal NMR and EFG calculations of VOCl3 and a vanadium catechol complex in the solid state.

Details

Language :
English
ISSN :
1549-9618
Volume :
8
Issue :
2
Database :
MEDLINE
Journal :
Journal of chemical theory and computation
Publication Type :
Academic Journal
Accession number :
26596600
Full Text :
https://doi.org/10.1021/ct200824r