Generalized gradient approximation for the exchange-correlation hole of a many-electron system

We construct a generalized gradient approximation (GGA) for the density ${\mathit{n}}_{\mathrm{xc}}$(r,r+u) at position r+u of the exchange-correlation hole surrounding an electron at r, or more precisely for its system and spherical average 〈${\mathit{n}}_{\mathrm{xc}}$(u)〉=(4\ensuremath{\pi}${)}^{\mathrm{\ensuremath{-}}1}$\ensuremath{\int}d${\mathrm{\ensuremath{\Omega}}}_{\mathit{u}}$ ${\mathit{N}}^{\mathrm{\ensuremath{-}}1}$\ensuremath{\int}${\mathit{d}}^{3}$r n(r)${\mathit{n}}_{\mathrm{xc}}$(r,r+u). Starting from the second-order density gradient expansion, which involves the local spin densities ${\mathit{n}}_{\mathrm{\ensuremath{\uparrow}}}$(r),${\mathit{n}}_{\mathrm{\ensuremath{\downarrow}}}$(r) and their gradients \ensuremath{\nabla}${\mathit{n}}_{\mathrm{\ensuremath{\uparrow}}}$(r),\ensuremath{\nabla}${\mathit{n}}_{\mathrm{\ensuremath{\downarrow}}}$(r), we cut off the spurious large-u contributions to restore those exact conditions on the hole that the local spin density (LSD) approximation respects. Our GGA hole recovers the Perdew-Wang 1991 and Perdew-Burke-Ernzerhof GGA's for the exchange-correlation energy, which therefore respect the same powerful hole constraints as LSD. When applied to real systems, our hole model provides a more detailed test of these energy functionals, and also predicts the observable electron-electron structure factor. \textcopyright{} 1996 The American Physical Society.