soprano missing many isotopes

[joelvdavies]

For silicon for instance, only have 29Si, which by the looks of it isn't the most abundant (28Si). These also have different spins so (1/2 vs 0) so muspinsim may behave in an unexpected way.

[joelvdavies]

As a further thing to be careful of, currently 19F is mislabelled as a spin 1/2. This is fine when not including quadrupole moments as it will behave as expected, but since soprano only currently includes NMR active nuclei, since the electric quadrupole moment tensor of a spin 1/2 nuclei is 0, the quadrupolar moment its actually referring to is an excited state with spin 5/2. So in effect, any quadrupole interactions for 19F will be calculated incorrectly. This is currently avoided due to raising an error for such spin 1/2 nuclei but if soprano were to label it correctly then 19F would refer to the excited state in muspinsim instead of the ground state which may be unexpected.

[joelvdavies]

For future reference: There are issues over on mendeleev here: lmmentel/mendeleev#97 and lmmentel/mendeleev#100 for potentially improving mendeleev and using it in soparano.

The last time I heard from Kane he was unsure what approach to take between waiting for and fixing mendeleev vs creating their own package/updating soprano, I gave him my thoughts but haven't heard back since, it would be worth checking if its progressed.