Specimen identifier in pdCIF?

[rowlesmr]

From here, to hook into the diffraction experiment identifier, we need to define our own data name in the DIFFRN category corresponding to specimen identifier.

How much of cif core do we need to reproduce?

_diffrn.crystal_id appears to be the direct analogue we would want to copy inform our new data item. But that is a pointer to _exptl_crystal.id, which is it's own thing. Do we need to hook in PD_SPEC, PD_PREP, and PD_CHAR? If we do, do we need to worry about specimens changing (ie insitu) affecting how we can identify them?

[jamesrhester]

I think PD_SPEC is the logical correlate of exptl_crystal, so it should be sufficient to create a specimen identifier and then the corresponding data name in _diffrn.id.

As far as specimens changing goes, the non-committal answer is that whenever anything in PD_SPEC changes, then the specimen is different. Perhaps a "time in beam before experiment started" data name is useful as a PD_SPEC data name, and then if this changes then you have a new specimen, but that also implies a new experiment (_diffrn.id). Alternatively you could just have a series of experiments on the "same" specimen - I'm thinking here of battery cycling measurements, where the external specimen is the same but the cycling is causing internal changes.