Obviously, the answer depends on numerous factors; how much time can you
afford, crystal quality, composition. Using a conventional source equipped
with a CCD detector, organics can be collected which are plate like habits 30
micron minor dimension with little difficulty; needles with the two minor
dimensions being around 50 microns also with little problem. I really have
not pushed it (I could use much longer frame times and probably go lower). I
have little experience with inorganics, although I saw a talk presented at the
ACA this year by Peter Burns (Notre Dame Univ.) where he was solving
structures on crystals which were less than 5 microns for some heavy metal
inorganic minerals using his conventional source and CCD. Certainly the
synchrotrons can go significantly lower.
Sincerely,
Greg