This should consider different possibility, and will determine the logic of the VibroWorkchain. The VibroWorkChain may also be splitted in the standard aiida-vibroscopy ones, referring to https://github.com/mikibonacci/aiidalab-qe-vibroscopy/issues/22.
[x] Specift that we are in first order non-resonant Raman scattering regime (maybe only when Raman is selected).
[x] specify what properties can be computed (bands, dos, INS...)
The properties will be grouped in this way:
[x] Raman, IR, Dielectric, INS**, Phonons (bands, pdos, thermal): HarmonicWorkChain - heavier computational cost, as Raman and IR are then computed in supercell approach (also q!=0 phonons);
[x] Raman, IR, Dielectric: primitive cell approach. We always compute both (IRamanWorkChain, dielectric.property="raman"), as the third order derivatives does not introduce too large additional computational cost;
[x] INS**, Phonons (bands, pdos, thermal): supercell approach, can be both polar (HarmonicWorkChain, with dielectric properties) or not (PhononWorkChain);
[x] Dielectric properties only: compute up to third order (DielectricWorkChain, dielectric.property="raman");
** The inelastic neutron scattering (INS) is just a post processing from phonons.
This should consider different possibility, and will determine the logic of the
VibroWorkchain. TheVibroWorkChainmay also be splitted in the standardaiida-vibroscopyones, referring to https://github.com/mikibonacci/aiidalab-qe-vibroscopy/issues/22.The properties will be grouped in this way:
dielectric.property="raman"), as the third order derivatives does not introduce too large additional computational cost;dielectric.property="raman");** The inelastic neutron scattering (INS) is just a post processing from phonons.