No time to figure this out (am in a bit of a rush), but I remembered that this guy out in California did some work on this topic a while back and managed to find the citation via Google which may help:
C.S. Yoo, et al (1999) Phys. Rev. Lett. 83 (26): 5527-5530.
I would think a face-centered cubic structure would be OK as a first try, though. As I had to explain to one of my students a while back, you see the difference between the strength of weak chemical interactions versus covalent interactions when comparing carbon dioxide (dry ice) and silicon dioxide (quartz) quite clearly...