[Pw_forum] Generating Pseudo potential

[Vi Vo]

Dear PWSCF users,

I am learning how to generate a pseudo potential (PP) from the tutor notes.  I 
don't know based on what criteria to chose the "rlderiv" at which ""nld" 
logarithmic derivatives are calculated.  The reason why I have this question 
because when I generate a PP for Yb (even though I know that the current DFT, 
e.g. PBE or LDA is not good enough), in plotting the log derivative, the atomic 
and the pseudo log derivatives do not match well at the tail.  However, by 
changing "rlderiv", I can make the match better.   I don't know if there is any 
criteria for choosing or adjusting this parameter (rlderiv).  I have read a few 
papers,  e.g. by  Hamann, Kleiman Bylander, X. Gonze, but I still could not 
figure out the rule of thumb for this.   Could you help me to understand this.

Thank you very much,

Vi Vo



      
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.quantum-espresso.org/pipermail/users/attachments/20100731/8844f8ec/attachment.html>