Recently I am trying to calculate the atomization energy of G-2 test set using vasp and try to match
the result with the following paper http://aip.scitation.org/doi/pdf/10.1063/1.1926272.
Generally atomization energy is calculated as the energy difference between the molecule and its individual constituents.
For individual atom calculation the vasp will output almost zero energy because of that result with respect to the energy
at which pseudopotential is calculated. I have tried H2O molecule. I have attached INCAR, POSCAR, POTCAR, KPOINTS and OUTCAR files.
At OUTCAR it gives energy 14.22 ev.
Could any one please help me how to find out atomization energy here?
vasp subtracts the atomic energy of the constituents from the total energy of the molecule or solid. that atomic energy is for the non-spin polarized neutral atom. for oxygen (and hydrogen) that is not very accurate because of spin-polarization being non-zero for both constituents. so you'll have to account for that. the spin-pol energy of symmetry-broken oxygen is -1.9 eV (as calculated by vasp in a large cell). also, in most cases atomization energies are not very interesting since one will compare some stable or metastable configurations of the same atoms, such as 2 H2O O2 +2 H2, that is to say, use chemical potentials different than the atomic energies. another thing is of course the dependence on the functional (LDA, PBE, BLYP, whatever: many produce overbinding anyway). clearly your cutoff and cell size have to be converged, and the structure be optimized.
Thank you Sir for your help. Now, I understood the problem according to your instruction.
- Badges
- Science topic
- Similar topics
- Mathematical Sciences
- Graphs