RE: [ccp4bb]: smallest diffracting crystals?

["Bernhard Rupp" <br@llnl.gov>]

***  For details on how to be removed from this list visit the  ***
***          CCP4 home page http://www.ccp4.ac.uk         ***

> crystallites. If that assumption is correct wouldn't a single 
> crystallite then be the relevant volume that determines line 
> broadening.

That is the fact, as far as I know from small molecule/alloy
work. You are approaching single domain size at such
dimensions/number of atoms. 

> I don't know my quantum stuff but even in a perfectly aligned 
> crystal I don't think a photon interacts with the whole 
> crystal volume. 

The scattering coherence length of one photon must be larger than the
domain size otherwise you would not notice the broadening effect. 
It becomes distinct at sub um sizes for materials
around 10 A size. This was the basis for the 1000 molecule
extimate. It is also an approximate domain size limit for other 
materials (corresponding to grain boundaries = mosaic) where 
coherent long range effects (ferromagnetism etc) change.     
 
> The point I am getting at is that we only have to start 
> worrying about line broadening once the crystal volume gets 
> significant smaller than the "diffraction volume" experienced 
> by a photon in a typical larger crystal. 

That is the precisely the case, I am very sure. I don't have 
my q-phys books here in TX to give an estimate for the 
max photon coherence length. 

> One should also 
> remember that spot size depends on the irradiated crystal 
> volume, mosaicity, beam divergence, and point spread function 
> of the detector. Broadening due to too few unit cells only 
> becomes relevant when it starts to dominate the other 
> contributing factors.

True. It always convolutes like any other spread function 
and inevitably will become important. 
 
> To make this a useful discussion we should have some idea 
> about the magnitudes of the effects we are talking about.

The effect can quite pronounced. At below 0.1 um (10-5 cm) even
a refractory material like Aluminiumoxid does not diffract
very high due to line broadening 
(look at http://www-structure.llnl.gov/powder.gif).

best regards, br