Dear Sergey,
if you calculate the projected DOS you can see which atoms are
involved in the states at the edges of the Valence and Conduction Bands.
moreover you can see which angular momenta are participating into the
frontier states...
this should tell you what's going on. whether the states involved
are not affected by the U part of the Hamiltonian or some more tricky
cancellation is happening
HTH
stefano
On 10/05/21 00:35, Sergey Lisenkov wrote:
Dear users and developers,
I have a question about transferability of U value computed with HP code.
I have two materials, lets call them "A" and "B". "B" contains Mn
atoms, so I calculated U self-consistently using HP code (U
calculation - relax with U - U calculation - relax with U until U is
converged within 0.01 eV). Without "U" band gap of material "B" is 2.6
eV, with U - 4.7 eV. Material "A" does not have transition metal
atoms, so I used plain DFT for this. Bandgap for "A" is ~4.6 eV.
Now I study "A"-"B" superlattices - layers of "A" material alternating
with layers of "B" material. What I found that band gap with and
without U correction I calculated before is almost the same - ~2.6 eV.
Now I'm curious to know why there is no difference in band gap in
those two calculations. Is it possible that I have to calculate U
correction for this new material - "A/B" superlattice?
Thanks,
Sergey
Physics Department,
USF
_______________________________________________
Quantum ESPRESSO is supported by MaX (www.max-centre.eu)
users mailing list [email protected]
https://lists.quantum-espresso.org/mailman/listinfo/users
_______________________________________________
Quantum ESPRESSO is supported by MaX (www.max-centre.eu)
users mailing list [email protected]
https://lists.quantum-espresso.org/mailman/listinfo/users
