In one of the sessions on ‘crystal engineering’, Guy Orpen told us about his hesitation to use this term: is it really engineering? The problem is reproducibility. If an engineer sets out to make a bicycle, and ends up with a submarine, he won’t be happy. Now if a scientist sets out to make a bicycle and finds that he made a submarine, he’ll be delighted. Yes, yes, it’s not quite a bicycle, but hey it is a means of transportation, only in water, right? And now having discovered this one by serendipity, we’ll make better submarines! In fact, the role of serendipity has been widely acknowledged throughout the conference.
We saw further work on co-crystals, and it shouldn’t come as a surprise that they are becoming so widely investigated. As I mentioned yesterday, one of the problems of drug molecules is poor solubility; and not all drug molecules are ionizable. In fact, according to Ashwini Nangia, 80% of the compounds in the R&D pipeline pose a solubility problem. This can be improved by inserting the active pharmaceutical ingredient (API) into a lattice with a second component – the trick is not to change to the API’s desired properties in the process. And besides pharmaceutical applications, co-crystallizing two compounds represents a chemical challenge in itself. What makes two molecules crystallize into one lattice rather than form two separate homo-crystals? Are co-crystals really that different from salts? Well, yes. As Christer Aakeroy explained, a cation cannot exist without an anion, whereas a molecule is happy on its own. And so he has set up a ‘molecular dating agency’ in his lab, and looks at various combinations to determine some rules on what affinities make molecules pair up in a co-crystal.
Another intriguing point about the crystalline solid state is polymorphism – salts, solvates, co-crystals and various combinations of these exist, and to complicate things further they all exist in different polymorphs. The form a compound will adopt depends on a lot of factors, and is hard to calculate, predict, or even understand. For example, Ram Jetti (from Matrix Laboratories, Hyderabad) showed us that for one drug molecule, temazepam, with 3 polymorphs reported in the literature, he was able to find no less than 10 new crystalline forms.
Add to this the fact that solids are a lot less inert than it first seems – this much was made clear in the solid-state reactivity sessions – and that surfaces and defects also play crucial roles in solid state reactivity, as we heard from William Jones (“sorry to take away from the perfection of single crystals”)… “one needs to be fearless to go into this mess” as Dario Braga puts it.
But I’m running out of time again, and will tell you tomorrow about some other exciting points that came up.
Anne Pichon (Associate Editor, Nature Chemistry)