I did some uB3LYP DFT calculations on several closed and open-shell amine -Fe and Cu complexes. I realized I used the 6-31++g(d,p) basis set for the lighter atoms and for the valence electrons of Fe and Cu as well, while using the LANL2DZ pseudopotential for the core electrons.
Is that a bad or a good thing? i.e. I guess the 6-31++g** basis set for the valence electrons is better than LANL2DZ's D95 (is it?), but does it work well with LANL2DZ's ECP?
In any case, could this mistake have introduced a systematic error so formation energies, being differences of energies calculated the same way, have this error cancelled?
Thanks!
The "problem" you'll have is that the basis is not optimized for use with this ECP. There are other basis/ECP combinations you can use, there's a LANL2TZ if you like that ECP, or there are the Peterson-Puzzarini and the def2- basis sets for use with the Stutgard ECPs, These bases have been optimized for use with the accompanying ECPs so in that regard they'll be better.
Dear Fedirico,
Actually i want to convey some basics from my knowledge. An Organic compounds will well optimise under B3LYP and if metal containing organic complexes means you may go to hybrid of (B3LYP+LANL2D = GEN EPS) basis set. If halogens containing compounds means you may proceed to LANL2D basis set.
Herein i have pasted one link to understand the insights of computation Chemistry. Please go through it.
https://www.researchgate.net/topic/Basis-Sets
Thank You
- Badges
- Science topic
- Similar topics
- Methodology
- Research Methodology
- Research Papers