I'm simulating rare earth complexes using Gaussian and DFT. I've been using B3lyp/LanL2DZ for the metal and it should work for any element between La and Bi according to literature and the Gaussian's user's reference. However I can't run the job for praseodymium (Pr). Can anyone help me?
I think that the original Hay & Wadt publication on the 'Los Alomos' pseudopotentials do not go further than mercury, so do not include lanthanides or actinides.
On the website https://www.basissetexchange.org/ you will find that there are indeed no such pseudopotentials for the f-elements (except for La itself and U, Pu, Np).
You could try another (relativistic) pseudopotential such as CRENBL, or you are most welcome to explore ADF, which has all electron basis sets for the all the elements in the periodic table from single zeta to quadruple zeta quality. Inner electrons can be frozen - note that this is not a pseudopotential: it'll give you the proper potential and density.
ADF has been trusted for studying heavy element chemistry, including lanthanides and actinides, see e.g. https://www.scm.com/applications/heavy-elements/
Hope this helps,
Fedor
The range of applicability for LANL2DZ is : H, Li - La, Hf - Bi (see https://gaussian.com/basissets/ ). So, the Praseodymium (Pr, lanthanide , Z=59) is not included. I suggest the use of SDD or SDDall.
Choose suitable basis-set from https://www.basissetexchange.org (former bee.pnl.gov)
Cheapest option I think would be def2-TZVP that is available everywhere, but I am not sure how good are relativistic effects in it (I am afraid not at all). Main approaches to that are ZORA and DKH, but first one I think is not in Gaussian (free ORCA has both). Also take into account that relativism is unpleasantly expencive.
- Badges
- Science topic
- Similar topics
- Physical Sciences
- Quantum Physics
- Quantum Mechanics