Gradient Corrections

From: Andrew Horsfield <andrew.horsfield-: at :-materials.ox.ac.uk>
Subject: Gradient Corrections
Date: Mon, 13 Mar 2000 15:59:38 +0000 (GMT)

Hi,
 I am thinking about adding gradient corrections to a density functional
 code that works with numerical atomic type orbitals. Given that there are
 a number of functionals in the literature, I would welcome suggestions of
 where to start. Note that I cannot include any exact exchange since it is
 too expensive to evaluate the integrals with numerical orbitals.
 In particular, questions that come to mind are:
 1) Is exchange better described by gradient corrections than correlation
    (i.e. would implementing just Becke exchange be a worthwhile move)?
 2) Is the correct choice of functional very problem dependent? Is there
    one choice that gives better *overall* results than the others?
 3) Are there other methodologies for improving on LDA besides gradient
    corrections? If so, what (dis)advantages do they have?
 Cheers,
 Andrew
   +----------------------------------------------------+
    Andrew Horsfield       e-mail: horsfield-: at :-fecit.co.uk
      FECIT, 2 Longwalk Road, Stockley Park, Uxbridge,
           Middlesex UB11 1AB, United Kingdom.
    phone: +44-(0)181-606-4653  FAX: +44-(0)181-606-4422
   +----------------------------------------------------+