Page 1 of 1

ENCUT Convergence

Posted: Fri Apr 21, 2023 6:08 am
by Mausumi
I am a beginner in VASP, please help me in clarifying the error .We are trying to do dft calculations on N,S dual doped graphene in vasp. The ENCUT values are not getting converged for set of values ranging from 200 to 650 eV. We tried with Monkhorst pack grid 2*2*1 and Gamma centered pack grid 13*13*1 (as per the literature). Could you please look into this and give some appropriate suggestions?

Re: ENCUT Convergence

Posted: Mon Apr 24, 2023 6:59 am
by alex
Hello Mausumi,

if I got your question correctly you are after basis set convergence wrt. basis set size?!

In that case you need to know that VASP extrapolates to the basis set limit by adding a cut-off dependend energy per atom to correct for basis set incompleteness. You'll find this in your(!) OUTCAR.

PAW_PBE C 08Apr2002 :
energy of atom 1 EATOM= -147.1560
kinetic energy error for atom= 0.0854 (will be added to EATOM!!)
PAW_PBE S 06Sep2000 :
energy of atom 2 EATOM= -276.8230
kinetic energy error for atom= 0.0051 (will be added to EATOM!!)
PAW_PBE N 08Apr2002 :
energy of atom 3 EATOM= -264.5486
kinetic energy error for atom= 0.2274 (will be added to EATOM!!)


If you remove these energies you'll find a nice behaviour E_tot = f(ENCUT).

Hth,

alex