Freeze Ga d electron in the core, as it is done in most pseudo-potentials currently in use. No virtual-crystal approximation is applicable to atoms that have different occupied shells. This is *NOT* the case for Ga and Al, which are amongst the pairs of elements for which the VGA makes the most sense. The CVA is conceptually totally agnostic of the XC functional in use. If you have reliable PP'S of two isovalent elements, you can mix them independently of the XC functional used to construct them. HTH — Stefano
> On 8 Jun 2019, at 12:57, Tuoc Vu N. (HUST) <[email protected]> wrote: > > Dear QE experts, > How can we generate the pseudo-potential (by virtual.x distribution of > Quantum Espresso) to simulate properties of random miscible alloys such as > AlxGa1-xN as the specific system of interest. > The valence configuration of Alluminium and Gallium is different (with and > without d electron) so I've met a false. How can I solve it. > Further, is is possible that wit QE we could use VCA with the PBEsol > exchange-correlation (xc) functional then (if it is possible? ) use VCA with > the TB09-MGGA xc-functional for bandstructure derived calculation. > Is any references? > Many thanks and regards, > Tuoc Vu > > _______________________________________________ > Quantum ESPRESSO is supported by MaX (www.max-centre.eu/quantum-espresso) > users mailing list [email protected] > https://lists.quantum-espresso.org/mailman/listinfo/users — Stefano Baroni - Trieste — http://stefano.baroni.me <http://stefano.baroni.me/>
_______________________________________________ Quantum ESPRESSO is supported by MaX (www.max-centre.eu/quantum-espresso) users mailing list [email protected] https://lists.quantum-espresso.org/mailman/listinfo/users
