The role of solvent in mechanochemical and sonochemical cocrystal formation: a solubility-based approach for predicting cocrystallisation outcome

Mechanochemical liquid-assisted grinding (LAG) and sonochemical (SonicSlurry) techniques have been compared as methods to construct four model pharmaceutical cocrystals involving theophylline and caffeine as pharmaceutical ingredients and L-malic or L-tartaric acid as pharmaceutical cocrystal formers. For these model systems, the results are interpreted using the parameter η, the ratio of solvent volume to sample weight. Each if the four cocrystals was studied in four different solvents using LAG at η = 0.25 and 10 µL mg−1, as well as SonicSlurry experiments at η = 2, 6 and 12 µL mg−1. The formation of the cocrystal is observed in all standard LAG experiments when η = 0.25 µL mg−1. Cocrystal formation by neat grinding, i.e. with no liquid added, was observed only for the cocrystal of theophylline and L-malic acid. LAG experiments at very low η values (below 0.5 µL mg−1) revealed that the rate of cocrystal formation depended on the choice of the liquid and increases with η. SonicSlurry experiments performed at higher η values of 2, 6 and 12 µL mg−1 provided three different outcomes: the pure cocrystal, a mixture of the cocrystal with a cocrystal component, or a single cocrystal component. LAG experiments at η = 10 µL mg−1 produced results consistent with the SonicSlurry experiments at η = 12 µL mg−1. Measuring approximate solubilities of individual cocrystal components revealed that product formation is not dictated by the specific processing method but by saturation levels of reactants. An experimental approach based on approximate solubilities of cocrystal components has been developed to qualitatively predict the outcome of cocrystallization experiments at different η values. As a general guideline, cocrystal formation is expected under conditions in which all cocrystal components remain saturated.