I am calculation gas (NO2, NH3) binding with graphene using VASP. At first i used GGA_PBE but it gave no binding. Now i am using LDA_CA which gives good binding and i want to include GRIMME (D2 D3) or TS correction.
IVDW-flag (given in manual)to invoke this correction is for PBE functional if i am not mistaken.
Can any one suggests how can i invoke van der wall's correction while using LDA functional ?
(I'm new VASP user )
I think you can formally use the IDVW option with any exchange-correlation potential, in the sense that VASP won't protest and will not crash. But I doubt it makes much sense to use van der Waals correction with LDA_CA. The default parameters of the DFT-D2/3 methods have been optimized for PBE. If, as you write, LDA_CA gives better binding, it is probably because it partially compensates for the missing vdW interaction, so as soon as you include van der Waals, you should rather abandon LDA.
I used IVDW flags and @Martin you are right i am looking for the parameters to invoke DFT-D2/D3 because most of the work has done on PBE.
Its seems that its not wise to use Van der Wall's correction for LDA as it gives already higher binding than GGA and in most of the cases even experimental binding value.
D2 or D3 method is based on GGA exchange correlation functional
(Grimme method). In this method the exchange-correlation functional from GGA has been combined with a correlation term. So, in general, using D3 with LDA is wrong!
- Badges
- Science topic
- Similar topics
- Biological Science
- Molecular Biology
- Biomolecules
- Proteins