My Community

Posted: **Thu Nov 01, 2012 5:30 pm**

Hello All,

I would like to follow a GGA+U calculation with a G0W0 or GW0 calculation. Is it possible to do this correctly in VASP?
If so, should I include the LDAU=.TRUE. and other DFT+U variables in my INCAR for the GW part of the calculation?

Thank you,
David

Posted: **Mon Nov 05, 2012 2:37 pm**

GGA+U can be combined correctly with non-selfconsistent G0W0 calculation

Posted: **Mon Nov 05, 2012 11:50 pm**

Thank You. I assume one retains the +U specifications in the INCAR file when G0W0 is run, as this treatment is consistent with the eigenvalues (and wavefunctions) in the input WAVECAR file. A few tests show the results of this make more sense than doing the G0W0 run with the +U lines removed.

My Community

how to combine LDA+U/DFT+U/GGA+U/PBE+U and GW approximation (GWA,GW0)

how to combine LDA+U/DFT+U/GGA+U/PBE+U and GW approximation (GWA,GW0)

how to combine LDA+U/DFT+U/GGA+U/PBE+U and GW approximation (GWA,GW0)

how to combine LDA+U/DFT+U/GGA+U/PBE+U and GW approximation (GWA,GW0)