ABINIT Discussion Forums

Dear colleagues,

We have a problem with interpreting my results. We have calculated the delithiated form of a layered solid Li2MnSiO4 (in essence 2D inorganic framework MnSiO4) with PBE, PW91 and PBESol functional. We found that for PBESol there is a strong repulsion between layers so that they move apart each other. Could anyone explain the true form of functional we should use and what is the reason of such behavior. Does it physical or a fake phenomena. Any references will be also appreciated.

Thank in advance,
M. Arsentev
PhD ISC RAS,
St-Petersburg, Russia.

I do not see a good reason for a strong difference (how much?) - your question is more about physics and xc than about the code. The PBEsol should only change a little the exchange part of pbe to be better for 3D solids (instead of molecules as in PBE): your layers are intermediate in dimensionality.

Perhaps you should try vdw corrections as well if your system is really layered? GGA is quite bad for graphite and this sort of 2D bulk materials. Check the vdw_xc and related input variables.