There’s a new and very useful paper out on the “molecular sponge” technique for crystallography (first blogged about here, with updates here and here). It’s from the Clardy group at Harvard with collaboration from Argonne, in Acta Crystallographica, and you can tell by reading it that it’s intended to put the whole method on a firmer footing.
That it does. Some of the data sets produced so far haven’t really been up to the quality standards that most crystallographers feel comfortable standing behind, but the paper notes that synchrotron sources (as is often the case!) are a far better bet for useful structures than lab-scale equipment using Mo K-alpha X-rays. The paper also contains detailed advice on the production and handling of the MOF crystals themselves, how best to approach the structure refinement of the soaked guests, and much more. It’s essential reading for anyone looking at this method. It’s still not a casual stroll to high-quality structures, though:
Despite the described synthetic and crystallographic guidelines, it is imperative to note that the crystalline sponge method must be used judiciously, and that the results obtained are not always unambiguous or ‘crystal clear’, per se. Great care must be taken in interpreting the residual electron density for the guest molecules, especially in cases where the structure is not completely known, or if it exhibits conformational flexibility and thus disorder. With excessive disorder, poor data, over-modeling and/or making erroneous assumptions based upon misguided optimism, the disastrous outcome of drawing incorrect conclusions is very real. . .
Spoken like a crystallographer, for sure. These are early days for the whole MOF structure field, and it wouldn’t surprise me at all to find the current “Zn-MOF” framework superseded by something with wider applicability. (Indeed, I think its inventors, the Fujita group, are busy trying to supersede it right now). One of the biggest limitations, which I’ve had a chance to explore personally, is the apparently complete incompatibility of the current frameworks with basic amines and/or heterocycles. But the idea has tremendous promise, and I’m happy to see this amount of work being put into it.
Update: forgot to add the link to the paper!