Dear All,
I'd like to ask for help/advise on the following issue:
what basis set(s) and DFT functional(s) would be the best choice for calculations of Eu-containing compounds? I am using Gaussian software to study complexes of Eu with carbon nanotubes.
Thank you in advance for your help!
For Eu-containing compounds I'd strongly advise you to employ large basis set(s) and account for relativistic effects.
For example: I'd employ SARC2-QZVP-DKH2 for the Eu atom and def2-TZVP for the lighter atoms.
To do so, you'll have to:
1) Use the keyword "int=dkh2" for the relativistic computation of two-electron integrals, more info on https://gaussian.com/integral/
2) Define the SARC2-QZVP-DKH2 basis set for Eu, you can find the basis set on https://www.basissetexchange.org/ and the syntax to define it in https://gaussian.com/gen/
3) Define def2-tzvp for the lighter atoms in a similar fashion as you did in step 2.
Of course, you could use different basis sets for the lighter atoms and for Eu, but you should always try to go beyond triple zeta and always include relativistic effects as well.
First thing is to consider the specialized literature. However, a general insight when selecting a basis set and functionals for calculations involving Eu-containing compounds, it's important to consider the electronic structure and properties of the system you're studying so here are some ideas that can serve as a starting point for your calculations of Eu complexes with carbon nanotubes. Firstly, for Basis set and in the case of lanthanide elements like Eu, relativistic effects can be significant. Therefore, it is advisable to use basis sets that include relativistic effects, such as the Stuttgart–Dresden (SDD) pseudopotential basis set. It provides a good balance between accuracy and computational cost for lanthanide elements. Secondly, for exchange-correlation functionals you have several options. For instance, Hybrid functionals, in the case of Lanthanide compounds which often exhibit strong electron correlation effects such functionals could be used. Hybrid functionals, which combine a fraction of exact exchange with density functionals, can provide improved accuracy for these systems. some popular hybrid functionals include B3LYP, PBE0, and M06. Maybe B3LYP is the most famous with high accuracy. Moreover, you can also consider Range-separated functionals, such as CAM-B3LYP or ωB97X-D. These functionals treat short-range and long-range electron-electron interactions separately, which can be beneficial for lanthanide compounds. Finally, because you have carbon nanotubes which are known to exhibit van der Waals interactions. Including a dispersion correction, such as the Grimme's D3 correction, can improve the accuracy of your calculations when studying interactions between Eu complexes and carbon nanotubes. It is worth mentioning that the choice of basis set and DFT functional may depend on the specific properties you are interested in, the size of the system, and the computational resources available to you. Additionally, it is always a good practice to perform benchmark calculations and compare the results with experimental data, if available, to validate the chosen methodology.
Best wishes and happy computing.
You should visit the website (https://www.basissetexchange.org/) for download the basis set for Eu along with references for particular basis function.
Good Luck !!
Hi!
You do not need QZVP basis sets. Maybe they can be used for property calculations, but, would be an overkill for geometry optimization.
SARC-ZORA-TZVP is what we normally. I do not know if the ZORA method is available in Gaussian. Alternatively, for geometry optimization def2-ECP basis sets can be trusted for all lanthanides.
Additionally, GGA functionals offer the best cost to accuracy ratio for these systems.
For more information you can see our recent publication 10.1021/acs.jpca.3c03027.
Hope this helps
- Badges
- Science topic
- Similar topics
- Computer Science
- Program