Atomic Forces too Low

[Lucas Bandeira]

Dear cp2k community,

I am trying to analyze k-points convergence for a Au bulk, but I am facing a problem: the atomic forces are too small. They are of the order of 1E-13 Ha. Then, when I try to compute the atomic forces between two consecutive grids, they are of the order of 1E-14 Ha, much lower than the threshold I use, 4.5E-4 Ha (the default maximum atomic force used in cp2k geometry optimization). So basically it means that the results will be the same no matter the k-point mesh I choose, from a 1x1x1 to a 15x15x15 mesh. Can somebody help me to solve this problem? Attached are one of the inputs that I have used and the corresponding output.

Yours sincerely,

Lucas Bandeira

[Fabian Ducry]

Hi Lucas

The forces are zero because of symmetry (the 1E-13 Ha is noise). You should check the pressure (stress tensor) instead. Alternatively, you can perform a volume optimization and check the convergence of the lattice parameter wrt. k-point grid.

Cheers Fabian

--
You received this message because you are subscribed to the Google Groups "cp2k" group.
To unsubscribe from this group and stop receiving emails from it, send an email to cp2k+uns...@googlegroups.com.
To view this discussion on the web visit https://groups.google.com/d/msgid/cp2k/CAM1%3DHZm0WxJwKD4LrhCCER75TMOYryC6Ka685NydKzP6S8eyQQ%40mail.gmail.com.

[Lucas Bandeira]

Thank you for your reply, Fabian.

But I have performed the same kind of calculations for a Cu bulk, and the forces printed are much larger. I am attaching the input and output files. In addition, I am analyzing cutoff and relative cutoff convergence for a CO molecule. With the GPW, the printed forces were too low, but with the GAPW the forces are much higher. Could you help me with that? 

To view this discussion on the web visit https://groups.google.com/d/msgid/cp2k/80e6f8fc-f5a9-e2b8-80c2-406eeac7b1b2%40gmail.com.

[Lucas Bandeira]

Sorry, I have sent the email by accident without finishing it. 

Attached (in the previous email) are the input and corresponding output files that I have used for the calculations with GAPW and GPW.

Yours faithfully, 

Lucas Bandeira