Dear QE users,I have two fundamental questions regarding polarization calculated using the Berry Phase method.
1. How should one convert the raw polarization printed in the nscf step from C/m2 to pC/m?
For example, say for a monolayer system, the printed value of polarization is 1.0704332 (mod 1.0641566) C/m2, where the polarization quantum is Pq=1.0641566 C/m2. First, bring the raw polarization within the interval [-Pq/2,+Pq/2].
For this, I subtracted 1.0704332 - 1* 1.0641566 C/m2= 0.0062766 C/m2. Call it Pphysical.
Now, to get to pC/m, what should I do? Should I multiply 0.0062766 C/m2 by the cell height 20 angstroms (assume the cell height is 20 angstroms) or assume a physical thickness of 3-5 angstroms?
For a bilayer system, suppose the actual thickness of the slab is 3 angstrom (Zmax-Zmin=3 angstroms). Then physical thickness of the bilayer system will be roughly 6 angstroms (twice of 3 angstroms). so here should I multiply the polarization value (Pphysical: derived after bringing the raw polarization within [-Pq/2,+Pq/2]) by the cell height (assume cell height =20 angstroms) or should I multiply by the physical thickness of the slab (= 6 angstroms)??
2. Secondly, Which is more correct approach to find polarization of a system : Planar averaged charged density difference analysis ( which gives only the electronic contribution) OR the Berry phase method which gives both the ionic and electronic components of the polarization?
This is important for heterobilayer systems. Kindly help me understand these points. Thanking you. Sincerely Jainandan Modi Ph.D. Scholar Electrical Engineering IIT Kanpur India
_______________________________________________________________________________ The Quantum ESPRESSO Foundation stands in solidarity with all civilians worldwide who are victims of terrorism, military aggression, and indiscriminate warfare. -------------------------------------------------------------------------------- Quantum ESPRESSO is supported by MaX (www.max-centre.eu) users mailing list [email protected] https://lists.quantum-espresso.org/mailman/listinfo/users
