Phase-Dependence of Molecular Structures Arising from Weak London Dispersion Interactions

ConspectusThe structures of molecules can be different in different phases. Intermolecular forces, even those of weak noncovalent interactions (WNCIs), can lead to a preference for quite different conformations in the solid, the gas, and the liquid phases. WNCIs can cause variations in bond lengths, angles, and torsional angles. Since structure is a fundamental concept in chemistry, the knowledge of structural changes with phase is important to understand the source and effects of distorting contributions from WNCIs but also as a predictive tool for the design and stabilization of new bonding situations.X-ray crystallography is ubiquitous and now mostly straightforward to perform, but facilities for the determination of accurate gas-phase structure determination are rare, and gas-phase work is laborious and time-consuming. There are currently about 1.25 million crystal structures and more than 12 500 experimental gas-phase structures, but the intersection of the two data sets that can tell us about the structural differences of the same molecule in different phases is surprisingly small.In this Account, we describe several cases of WNCI-dominated systems for which accurate experimental structure determinations exist for both the gas phase and the solid state and, in one case, also for solution. The examples include aryl-aryl, aryl-alkyl, and alkyl-alkyl interactions; systems with chalcogen and halogen bonding; and fluorine-based interactions in arylboranes. We work out the role of WNCIs in stabilizing large, strained, or sterically overloaded molecules. We will show how flexible molecules will fold under the action of WNCIs when isolated in the gas and how they fold or unfold when they are embedded in an environment of neighbors in crystals. We will show how they can vary in strength when the substitution patterns in aryl groups are changed by different halogens and how intramolecular WNCIs, such as those forming rings, change when such systems experience additional intermolecular WNCIs.Overall, we hope that this Account will give the reader an idea of the type and magnitude of structural changes that can be expected from a free molecule in the gas phase or a single molecule calculated by quantum chemistry compared with one embedded in a crystal. This should define the limits of comparability and provide some predictive concepts of the distortions and variations to be expected.