Aw mate. I read the paper and I must say I'm disappointed.
The theoretical treatment really doesn't say much; it just plots the rate of nucleation against a pore shape parameter. It can be summed up in the line from the paper that reads
The suggestion is that if the material has a wide distribution of pore shapes and sizes, a few will be just right for nucleation
which for me was a bit of a 'no shit, Einstein' moment.
What's perplexing is that the authors talk about a concave surface (think of the inside surface of a pore) whereas the very readable paper that they reference implies that concave surfaces are worse than flat or convex for nucleation, all other things being equal. I have never heard a crystallographer complain of too many hits in a crystal screen.
So while they may be correct that the ability of the pores to nucleation crystallization depends on the radius of curvature, it strikes me that using pores at all is a mug's game. In other words, don't buy this bioglass stuff. Try my Insect-o-Leg (patent pending) instead.
There's one more thing. When you publish a paper, although you might call it 'representative' you damn' well make sure that your best data goes into the figures. And figure 4 in this paper shows some fairly crappy crystals, to be honest. If these crystals were indeed the best of the four non-model protein crystals obtained, then I dread to consider the state of the rest: It's then perhaps not surprising that the authors do not report the results of what a crystallographer would consider the only thing that really matters - how well did the seeded crystals diffract?
