Computation of the local HGH pseudopotential

Asked by Tommi Höynälänmaa on 2019-07-08

I'm doing computations with periodic boundary conditions and no PAW. It seems that the erf term of the local HGH pseudopotential (equation (1) in arXiv:cond-mat/9803286v1) is not computed but some calculations involving Gaussian functions are done instead. These calculations occur in subroutine createIonicPotential in file init/ionicpot.f90 lines 949 and 950 (BigDFT version 1.8.2). I added some debug output to the BigDFT code and it shows that procedure sum_erfcr is not called at all in the test computation.

What is happening here?

Question information

English Edit question
BigDFT Edit question
No assignee Edit question
Last query:
Last reply:
Revision history for this message
Thierry Deutsch (thierry-deutsch) said :

Instead of using erfc with collocation method, we prefer to use a gaussian charge and use Poisson solver with is similar to apply an erfc function.
The advantage is to have the proper multipoles and reduce the collocation error.

 See J. Chem. Phys. 129, 014109, p. 9

Can you help with this problem?

Provide an answer of your own, or ask Tommi Höynälänmaa for more information if necessary.

To post a message you must log in.