On Jan 15, 2008, at 10:03 , Nicola Marzari wrote: > These u_nk are what is stored in evc (I hope - I'm not that familiar > with the innards of PWscf), in their *plane wave expansion*.
confirmed > Note that if u_n0(r) has inversion symmetry, then the coefficients > of the plane wave expansion are also real - albeit this has not been > exploited, afaik, since early calculations (1991-1992) with the > Cambridge CASTEP codes. once upon a time there was a version of the PWscf code that exploited inversion symmetry and used real wavefunctions and hamiltonian matrix (at the time, the hamiltonian was still stored in a matrix). I think it was also ported to separable pseudopotentials (i.e. hamiltonian stored as an operator, not as a matrix). It disappeared a few years ago for no other reason than the difficulty to keep 2N+1 versions of the code alive at the same time. It should be relatively straightforward to implement it again, though, adapting the existing Gamma-only technology. It could be a good and useful exercise for anybody who would like to learn how plane-wave codes work Paolo --- Paolo Giannozzi, Dept of Physics, University of Udine via delle Scienze 208, 33100 Udine, Italy Phone +39-0432-558216, fax +39-0432-558222
