ExPaNDS-eu / ExPaNDS-experimental-techniques-ontology

EU Photon and Neutron Ontologies (task 3.2)
8 stars 3 forks source link

Scattering vs. diffraction #80

Open gerrit-guenther opened 1 year ago

gerrit-guenther commented 1 year ago

Maybe you could think about making the relation between scattering and diffraction in the ontology more clear. To my understanding, you can (1) treat them as synonyms or (2) define diffraction as scattering where the sample is ordered and, thus, results in a ('elastic') diffraction pattern (but I guess you would like to avoid any sample dependency here).

On one hand, in the 'defined by experimental physical process' branch you have 'diffraction' as a subClassOf 'elastic scattering' and in 'diffraction' e.g. 'small angle neutron scattering' which would point towards the handling as synonyms; on the other hand, there is a separate term for 'neutron scattering' (as a non-elastic process) beside 'neutron diffraction' (as an elastic process) which implies that, here, the sample dependency plays a role. Perhaps you could consider to either use diffraction/scattering as synonyms or list separate terms, e.g. 'small angle neutron diffraction' as elastic diffraction process.

spc93 commented 1 year ago

This needs some thought!

I would say that diffraction is a subclass of scattering. Scattering just means that something goes in and something of the same type comes out. So every 'diffraction' technique is a 'scattering' technique but not vice versa.

Coherent scattering is more tricky. It just required preserving phase information, which is lost if another excitation is involved. I think the complication is that there are two types of incoherent scattering. First, scattering that is intrinsically incoherent (e.g. Compton scattering). Secondly, scattering which is not intrinsically incoherent but where the properties of the sample mean that the scattering forms such a fine speckle pattern that it is hard (or impossible) to resolve.

Consider a hypothetical crystal containing two isotopes, where the different isotopes are found at periodic sites within the crystal. The combined scattering of the two isotopes is then completely coherent, whereas there would be 'incoherent' scattering with a random mix. So the scattering process hasn't changed, just the sample.

It's a bit tricky though and requires some thought!