The ABINIT Users Mailing List ( CLOSED )

["Matteo Giantomassi" <Matteo.Giantomassi@uclouvain.be>]

> hi
>
> is it correct that for GW using HGH pseudos the only way to write down the
> KSS
> file is by using kssform=3?
>

 Hi Manuel,

 Yes. In the default mode (kssform=1), one has to
 construct <k+G| H | k+G'> and, unfortunately, the code is not
 able to take into account properly the 3 different projectors
 per angular state employed in the HGH pseudos.

 The only way to generate a KSS file is through the conjugate gradient 
algorithm.
 The code should complain a bit about the
 presence of multiple projectors per angular state but, don't worry, the
results are OK.
 Just keep in ming that during the calculation of the screening, the
inclusion
 of the commutator [H,r] has to be switched off (inclvkb=0). The KSS
 file, indeed, does not contain all the matrix elements of the nonlocal
 part of the Hamiltonian thus the evaluation of the commutator will give
 unpredictable results.

 Bye,
 Matteo Giantomassi

> thanks
>
> Manuel Perez Jigato
> KU Leuven
>
> Disclaimer: http://www.kuleuven.be/cwis/email_disclaimer.htm
>
>