The question is pretty simple, is GGA better than LDA at predicting vibrational properties of molecules/materials ? What is the nuance and so forth.

  • 3
    $\begingroup$ GGA is better when it is better, but not when it is not $\endgroup$
    – Wesley
    Feb 1, 2022 at 18:36

1 Answer 1


This is a partial answer focusing on comparing phonons calculated with LDA or GGA in solids.

LDA tends to overbind solids, which means that LDA equilibrium volumes tend to be smaller than experimental equilibrium volumes. By contrast, GGAs like PBE tend to underbind solids, which means that the corresponding volumes tend to be larger than experimental equilibrium volumes. As a result of these different volumes, LDA phonon frequencies tend to be higher than PBE phonon frequencies because the LDA potential is steeper. An intermediate result is typically obtained with PBEsol, a GGA tuned to accurately reproduce experimental volumes, which then leads to phonon frequencies in-between those of LDA and PBE.

Overall, the volume difference is what tends to drive any differences in phonon frequencies when comparing LDA and GGAs. There is some work on using other types of functional (e.g. hybrids) and in that case effects other than volume differences can be observed to influence the predicted phonon frequencies. However, I would argue that the latter is still an open area of research, so I would recommend doing your own tests.

  • 1
    $\begingroup$ Love your videos ! $\endgroup$
    – Elie H
    Feb 7, 2022 at 21:45
  • $\begingroup$ @ElieH Thanks! :) $\endgroup$
    – ProfM
    Feb 8, 2022 at 8:19

You must log in to answer this question.

Not the answer you're looking for? Browse other questions tagged .