The ABINIT Users Mailing List ( CLOSED )

[Vincenzo Fiorentini <vincenzo.fiorentini@dsf.unica.it>]

Dear Hua Bao and all,

if I may interject an "archeology" comment: at least in scalar- 
relativistic calculations

- different pseudopotentials produce (slightly) different equilibrium  
volumes, and using the theoretical volume for each pseudo (the  
consistent thing to do, of course) generally produces consistently  
the same ordering of conduction states, i.e. roughly a similar band  
structure (I showed this when I was young, in Solid State Commun. 81,  
871 (1992));

- using theoretical volumes (using a wel-behaving, well accepted  
pseudopotential set such as BHS) generally produces quite decent band  
structures, pressure coefficients, and effective masses (as shown in  
another oldie of mine: Phys. Rev. B 46, 2086 (1992));

- finally for GaAs d states produce some (fairly benign: not so  
GaN...!) effects on the bands. With different levels of treatment  
such as a) including d-states, b) applying non-linear core  
correction, c) doing nothing, you find the lattice constant shrinks:  
10.60, 10.50, 10.39 bohr respectively, if I remember correctly (I  
showed this in The Physics of Semiconductors, ed. D. J. Lockwood,  
World Scientific, Singapore, 1995, p.137-140 - proceedings of ICPS 22).

best - Vincenzo Fiorentini