I love reading papers that describe crystallization conditions and I make it a habit to read the experimental section. As you are probably well aware, Professor Fujita published some thought-provoking papers some time ago that described what he calls “crystal-free crystallography”. In this method, a molecule that is reluctant to form single crystals suitable for X-ray analysis, is coaxed into forming inclusion compounds that diffract reasonably well. This process has a number of far-reaching implications, particularly in structure determination of natural products. I have to admit that, for me, this process has always been reminiscent of what people routinely do in protein crystallography. In fact, my lab has been involved in experiments of this type in collaboration with our colleagues at the SGC (I refer to soaking experiments I mentioned several times in the past). The main difference is that the lattice is based on metal/ligand complexes in the Fujita technology. Another aspect is that when we run protein/small molecule soaking experiments, we already know which molecules we put in (or do we?). Fujita’s trick is to use well-defined inorganic materials that provide a “surrogate” lattice, so to say, and enable diffraction data to be collected for the guests that have been entrapped in nanocompartments. There was a lot of press surrounding this methodology, some of which hinted at some difficulties encountered in attempts to repeat the procedure, which the authors later admitted, aiming to work on process improvements. I was glad to see a Nature Protocols published by the Fujita team that provides a step-by-step recipe for how to run these crystallization experiments. It appears that making the metal-organic framework is a piece of cake and sounds like a lot of fun: you take a test tube and set up diffusion of a metha
nol solution of zinc diiodide layered onto a nitrobenzene/methanol solution that contains 2,4,6-tri(4-pyridyl)-1,3,5-triazine. Single crystals of the networked material form at the boundary (I enjoyed looking at the images). 
The rest is no different from how one would soak a small molecule into a protein crystal.

http://www.nature.com/nprot/journal/v9/n2/full/nprot.2014.007.html