Computational prediction of salt and cocrystal structures--does a proton position matter?

The lattice energy landscape is calculated for three pyridinium carboxylate salts and the corresponding pyridine·carboxylic acid cocrystals. Experimentally, one system crystallizes as a salt, another as a cocrystal and the acidic proton in the third is disordered across the N(arom)...O hydrogen bond vector. A novel structure of a 1:1 4-cyanopyridine·4-fluorobenzoic acid cocrystal (I) was characterized to provide the cocrystal as a system with an isolated carboxylic acid-pyridine heterosynthon. By contrast, the 4-dimethylaminopyridinium maleate salt (GUKVUE) shows the effects of an internal hydrogen bond, and the proton-disordered pyridine·isophthalic acid crystal (IYUPEX) shows the effects of competing intermolecular hydrogen bonds. All three crystal structures were found low in energy on the lattice energy landscape for the correct proton connectivity. For all three systems, comparing the salt and cocrystal energy landscapes shows the importance of the proton position for the relative stabilities of structures, despite the expected similarities between the ionized and neutral forms of the carboxylic acid-pyridine heterosynthon. The systems with additional hydrogen bonds have some hydrogen bonding motifs that are only favourable for the salt or for the cocrystal. This illustrates the sensitivity of the range of thermodynamically plausible crystal structures to whether the molecules are assumed to be ionized or neutral.