specification in the NMR example

[Victor Volkov]

Dear Dr. Iannuzzi:

good morning.

I wish to ask you two questions concerning your introduction to NMR under cp2k

https://www.cp2k.org/_media/events:2015_cecam_tutorial:iannuzzi_spectroscopy_b.pdf

1) In your example, Iteration Tolerance for spin 1 converged in 698 iterations

100 0.2560E-01
....
600 0.3288E-03

Is it possible to instruct cp2k
not to search for convergence better than 0.1E-03?

2) In your example, you define
&KIND N
LEBEDEV_GRID 100
RADIAL_GRID  200

May the Grid numbers be different for other atoms?

If so, are there any rules to anticipate optimal numbers?

I would be thankful if you comment.

Best regards,

Victor

[Marcella Iannuzzi]

Dear Victor

The iterations you are pointing to are those for the localisation of the MOS. 

There is a keyword in the localisation section to  control the convergence. 

The Lebedev grids can be adapted for each kind individually. 

The optimal number of points in general depends on the maximum angular momentum in the basis set.

 100/200 are good values, you can easily check whether anything changes using larger values.

Regards

Marcella

[Victor Volkov]

Dear Marcella

thank you very much for the kind reply.

:-)

With best regards,

Victor

--
You received this message because you are subscribed to a topic in the Google Groups "cp2k" group.
To unsubscribe from this topic, visit https://groups.google.com/d/topic/cp2k/09quQ7dq0ow/unsubscribe.
To unsubscribe from this group and all its topics, send an email to cp2k+uns...@googlegroups.com.
To view this discussion on the web visit https://groups.google.com/d/msgid/cp2k/92b14daa-6a71-4bdc-b164-9c58a0541082n%40googlegroups.com.