TOTEN in the EXX step (EXX+RPA calculation)
Posted: Wed Nov 25, 2015 7:38 pm
Hi there,
I am trying to perform EXX+RPA calculations following the online instruction.
In the step 2 (the EXX step), I obtained the OUTCAR as
Free energy of the ion-electron system (eV)
---------------------------------------------------
alpha Z PSCENC = 10.79902750
Ewald energy TEWEN = -326.85184009
-Hartree energ DENC = -242.20535843
-exchange EXHF = 79.22129233
-V(xc)+E(xc) XCENC = 0.00000000
PAW double counting = 561.13412586 -575.06334815
entropy T*S EENTRO = -0.04334204
eigenvalues EBANDS = -90.75405794
atomic energy EATOM = 555.18254431
Solvation Ediel_sol = 0.00000000
---------------------------------------------------
free energy TOTEN = -28.58095663 eV
Is the TOTEN above the total energy that I should add the RPA correlation energy to? My concern is that the TOTEN obtained above still has the double-counting part, since "-V(xc)+E(xc)" is zero.
This means that the TOTEN above is not the correct total energy for EXX, but contains the <vxc,rho> term. Then we cannot add RPA correlation energy to TOTEN.
Thanks a lot for your insightful comments.
Best,
Chen Huang
Department of Scientific Computing
Florida State University
I am trying to perform EXX+RPA calculations following the online instruction.
In the step 2 (the EXX step), I obtained the OUTCAR as
Free energy of the ion-electron system (eV)
---------------------------------------------------
alpha Z PSCENC = 10.79902750
Ewald energy TEWEN = -326.85184009
-Hartree energ DENC = -242.20535843
-exchange EXHF = 79.22129233
-V(xc)+E(xc) XCENC = 0.00000000
PAW double counting = 561.13412586 -575.06334815
entropy T*S EENTRO = -0.04334204
eigenvalues EBANDS = -90.75405794
atomic energy EATOM = 555.18254431
Solvation Ediel_sol = 0.00000000
---------------------------------------------------
free energy TOTEN = -28.58095663 eV
Is the TOTEN above the total energy that I should add the RPA correlation energy to? My concern is that the TOTEN obtained above still has the double-counting part, since "-V(xc)+E(xc)" is zero.
This means that the TOTEN above is not the correct total energy for EXX, but contains the <vxc,rho> term. Then we cannot add RPA correlation energy to TOTEN.
Thanks a lot for your insightful comments.
Best,
Chen Huang
Department of Scientific Computing
Florida State University