[Pw_forum] Pseudopotential for Li

[Bertrand SITAMTZE]

Dear PW users,

I am using the PW code for the study of an ionic compound. As I know (Please tell me if I get wrong), one should know the chemical environment of the element in the compound and generate the pseudopotential according to it. But, if the generated pseudopotential is well transferable, there reference configuration is of lesser importance.
I want to use a pseudopotential for Li+ ion and I am facing the following problems with the ld1.x code:
1)  The Starting configuration 1s2 2s0 does'nt allow non linear core correction. Here is the section of the error message:

Computing core charge for nlcc: 

 %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
     from set_rho_core : error #         1
     error in bisection
 %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%

     stopping ...

 why this problem?

2) Using the same starting configuration without nlcc, I get a positive pseudo-2s function, which is exactly the opposite of the expected result (when I plot the opposite, it seems to be correct).
Can somebody explain this to me?

3) Considering nlcc is only possible if I start with the configuration 1s2 2s1.0 2p0.0, which is far from Li+. Despite this, I still get a positive pseudo-2s wawefunction. why?

Please I am looking forward for your help



*********************************
Bertrand SITAMTZE
PhD student
Department of physics
University of Yaounde I-Cameroon
*********************************