[Pw_forum] fhi98PP lloc

Andrea Floris afloris at physik.fu-berlin.de
Wed Apr 26 19:51:48 CEST 2006


Hi everybody,
I generated a NCPP with fhi98PP for Ca with semicore states in the
valence, with lmax=lloc=2 (variables values in fhi98PP).
In the UPF conversion it results  (correctly, because of the
Kleinman-Bylander form) that:

Max angular momentum component = 1
Number of Projectors           = 2
Wavefunctions          nl  l   occ
                       3S  0  2.00
                       3P  1  6.00
                       3D  2  0.00

Moreover the <PP_LOCAL>  field is the correct conversion of the local
pseudo that I've chosen in the generation.

Question: in the output of pwscf for Ca results lloc=0. I do not
understand this, I was expecting lloc=2...
Does it have any wrong consequence? I
plotted my bands
and they seems reasonable compared with all electron bands.

thanks in advance for any explanation,
Andrea



>> In the .UPF pseudopotential my "Max angular momentum component"
>> in <<PP_HEADER>> is 1 instead of 2.

>correct: the UPF file contains PPs in separable (Kleinman-Bylander)
>form, so you have l=0 and l=1 projectors, l=2 as the local potential.

>P.



More information about the Pw_forum mailing list