When running jobs similar to:
# opt freq ub3lyp/6-311g geom=connectivity
I often get this warning message:
**** Warning!!: The largest alpha MO coefficient is 0.20327268D+02
Is this a critical situation? What can/should I do to fix this problem?
Thanks for considering this question.
Dear Professor,
You can try your calculation with a smaller basis set or guess=mix.
Dear professor,
Using "guess=mix" is good advice and we must consider that "guess=mix" will result in only an approximate answer and the resulting energy will be lower than the ub3lyp result with "guess=mix".
This warning related to the values of coefficients and basis set and it's most relevant for calculations where the accuracy of the result is related to the accuracy of the MO integrals. When you have near linear dependencies in the basis set you will get large MO coefficients related to maintaining orthogonality and if one or more of your coefficients if very high the numerical precision on your transformed MO integrals will be compromised and you have to be careful with the result. It means coefficients are too large due to a basis set that is too linearly dependent (for example, due to diffuse functions).
And it's important that, for a large alpha MO coefficient, we must define the open-shell situation for molecules.
The way to circumvent this problem is reduce the linear dependency on your basis set with a smaller basis set.
This is a warning that can typically be ignored for normal DFT, but is relevant for post-HF methods (eg MP2, MP4) . It typically means coefficients are too large due to a basis set that is too linearly dependent (possibly due to diffuse functions etc). You can verify with a smaller basis sets that the results don't differ too much if you can.
It would be nice to see your input file - 6-311G isn't a very large or diffuse basis set, so I think there is possibly something else going on here as well -
Dear professor Hipps,
opt and freq both keywords have created a problem during the optimization, So I request you please not use both keywords at a time, First of all, use the keyword opt and after the optimization applies Freq keyword. Gaussian 09 only created this kind of problem in some time however, Gaussian 16 suitable for both keywords.
- Badges
- Science topic
- Similar topics
- Biological Science
- Kinesiology
- Running