Exploring Coformer Substitution in Cocrystallization: Griseofulvin and Phenol Derivatives

Abstract

This study investigates the cocrystallization of griseofulvin with phenolic coformers, highlighting its feasibility and variability. In addition to the previously reported cocrystal of griseofulvin with 4-t-butylphenol (1:1), the experimental screening identified three new cocrystals: with phenol (2:5), 4-t-amylphenol (1:1), and 2,4,6-trichlorophenol (2:3). Phenols with carbon substituents in the ortho or meta positions failed to form cocrystals, likely due to steric hindrance and electron-donating effects. In contrast, phenols with chlorine substituents, particularly in the ortho and para positions, demonstrated enhanced cocrystallization potential, driven by the electron-withdrawing effects that promote hydrogen bonding. The 2:5 phenol cocrystal required optimized conditions for isolation and exhibited instability under ambient conditions due to coformer sublimation, a tendency also observed for the other cocrystals. While challenging, sublimation facilitated the determination of stoichiometric ratios, which varied from 1:1 to 2:3 and 2:5. Furthermore, this study provides a data set of cocrystal-forming and noncocrystal-forming combinations as a rigorous test case for virtual cocrystal prediction. Among the tested methods, crystal structure prediction proved the most reliable, identifying all observed cocrystal combinations and, together with powder X-ray diffraction, offering insights into the experimental coformer and cocrystal structures. Future integration of CSP with machine learning could accelerate prediction speed and accommodate a broader range of stoichiometric ratios. Overall, this work highlights the complexity and potential of cocrystallization.

This study demonstrates the cocrystallization feasibility between griseofulvin and selected phenolic coformers, identifying three new cocrystals with phenol, 4-t-amylphenol, and 2,4,6-trichlorophenol. The results emphasize the impact of substituent effects, sublimation tendencies, and stoichiometric variability on cocrystal formation and stability. The dataset serves as a critical benchmark for virtual prediction tools, showcasing both the potential and limitations of in silico cocrystal screening approaches.