[Wien] Which exchange funtional used in CA-LDA?

[Peter Blaha]

> May be a silly question. The LSDA functional CA-LDA in
> Wien2K is parameterized by Perdew-Wang 92. As I
> understand, Perdew-Wang 92 gave only the correlation
> funtional. So, which exchange funtional is used in
> CA-LDA? Slater, Perdew-Wang 91, or something else?

"yes". The exchange in "LDA" is completely defined as  V_x ~ alpha*rho**(1/3)
with alpha fixed at 2/3. There is no "fredom" in LDA for exchange.

(For correlation only a numerical simulation (Ceperly-Alder) can give the
"exact"-LDA correlation, and this has been fitted by several authors.
Since these fits differ slightly, one may specify the correlation in LDA,
but never the exchange.

                                      P.Blaha
--------------------------------------------------------------------------
Peter BLAHA, Inst.f. Materials Chemistry, TU Vienna, A-1060 Vienna
Phone: +43-1-58801-15671             FAX: +43-1-58801-15698
Email: blaha at theochem.tuwien.ac.at    WWW: http://info.tuwien.ac.at/theochem/
--------------------------------------------------------------------------