[ADF-LIST] DFT basis sets

Foerster, A.T.L. a.t.l.foerster at vu.nl
Fri Jul 16 14:06:29 CEST 2021

Dear Pascal,

G-31G* would correspond to DZP.  For B3LYP and TZ2P vs. def-TZVP, you can find such a comparison in


Note, that such a comparison is difficult. You can usually not expect quantitative agreement. Even if the basis sets have the same number of functions, they have different shapes and the basis set errors will be different. For DZ, agreement will most likely be worse.

It is better to compare values obtained at the complete basis set limit to each other. For various systems, you can find such reference values for example here:


You might compare them to ADF (for the functional of your choice) using QZ4P. At least for GGAs and hybrids this should be very reliable.

Alternatively, you can also check out the paper above and decide based on the data therein what functional works best for the systems you are interested in.

Best regards,

On 16 Jul 2021, at 12:07, Brault Pascal <pascal.brault at univ-orleans.fr<mailto:pascal.brault at univ-orleans.fr>> wrote:

Dear developers
I have to carry out DFT simulation and need to have a level of comparison between various methods.
Often in publication are used GTOs for example 6-31G* , 6-31++G, 6-31G**, etc.
AMS is using STOs. My question is: If I wish to reproduce for example B3LYP calculations with 6-31G* basis set how can I choose similar basis set. similar means including same or nearly same information and interactions.
Is there somewhere a table or at least a discussion, in an article or book, on how all these basis sets compare among them.
thanks a lot

ADFlist mailing list
ADFlist at scm.com<mailto:ADFlist at scm.com>

-------------- next part --------------
An HTML attachment was scrubbed...
URL: <https://lists.scm.com/pipermail/adflist/attachments/20210716/a8fc3fa6/attachment.html>

More information about the ADFlist mailing list