The curious case of (caffeine)·(benzoic acid): how heteronuclear seeding allowed the formation of an elusive cocrystal

Dejan-Krešimir Bučar, Graeme M. Day, Ivan Halasz, Geoff G. Z. Zhang, John R. G. Sander, David G. Reid, Leonard R. MacGillivray, Melinda J. Duer, William Jones
  • Chemical Science, January 2013, Royal Society of Chemistry
  • DOI: 10.1039/c3sc51419f

Discovery a predicted caffeine cocrystal

What is it about?

The cocrystal of caffeine with benzoic acid was previously thought not to form, as it failed to be observed after extensive screening. However, modern crystal structure prediction calculations showed that it should be stable. Templating experiments were designed based in the most probable predicted crystal structure, which led to the material being formed. This is a good example of calculations informing experiments, and leading to materials that might otherwise have never been discovered.

Why is it important?

Crystal structure prediction methods have been extended beyond prediction of the most likely structures of a compound, to predicting whether two molecules will form a multicomponent crystal (cocrystal or solvate) or not, as well as its energetically preferred composition (see and for methods). This is an example of the discovery of a cocrystal that was predicted to exist, and how the computational results guide the experiement to produce this new structure.

Read Publication

The following have contributed to this page: Professor Graeme Day, Dr Ivan Halasz, and Dr Dejan-Kresimir Bucar