how to choose the frequency at which stress is calculated

165 views
Skip to first unread message

lar...@lbl.gov

unread,
Sep 11, 2015, 2:47:06 PM9/11/15
to cp2k
Dear team,

I was wondering if there is any available setting in cp2k to set the frequency at which the stress is calculated in an AIMD simulation. I am currently using metaGGAs in my simulations and because the only way of calculating the pressure is numerically for this type of xc functionals the MD simulations are painfully slow. If there would be a way to not perform the numerical stress calculation every step that would make them manageable. Btw, just changing how often it's printed doesn't change anything. Thanks in advance!

Best,
Luis

Matt W

unread,
Sep 13, 2015, 8:53:08 AM9/13/15
to cp2k
Hi,

the simplest way might be to post process - use the REFTRAJ ensemble in the MD to rerun over the frames you like. The overhead of having to redo the SCF from not quite as good a guess should be negligible.

Matt

lar...@lbl.gov

unread,
Sep 15, 2015, 11:29:28 AM9/15/15
to cp2k
Thanks Matt. Interesting capability, I didn't know about it. I will try it.

Best,
Luis

lar...@lbl.gov

unread,
Oct 14, 2015, 12:27:04 PM10/14/15
to cp2k
Hi Matt,

I have been trying to use REFTRAJ to calculate the stress on some frames from a trajectory that I already ran but I don't seem to get it to work. The simulation in the REFTRAJ ensemble runs fine but the outputs are zero for the stress and other quantities in the *.ener file. I have attached here the input file as well as the outputs for energy and stress. I am not sure what I'm doing wrong so any pointer would be highly appreciated. Thanks in advance for your help!

Best,
Luis




On Sunday, September 13, 2015 at 5:53:08 AM UTC-7, Matt W wrote:
files.zip

Matt W

unread,
Oct 14, 2015, 4:35:53 PM10/14/15
to cp2k
Hi Luis,

seems like this might not be doable at the moment then, maybe only energy-force. Sorry if the suggestion was bad. I'd think it would be easy to add the extra ability to calculate stress tensor too, I'll try and have a look, but not sure I'll have time for a bit.

Matt 

Luis Ruiz Pestana

unread,
Oct 14, 2015, 4:54:19 PM10/14/15
to cp...@googlegroups.com
Thanks Matt.

It would definitely be useful to implement a way in cp2k to calculate stresses when using metaGGA's (either through the analytical calculation or indirectly through the REFTRAJ ensemble using the numerical stress calculation). Let me know if you get it done any time soon, it would be a life saver!

I really like cp2k and I highly recommend it to other users but in the current state I might be forced to change software to simulate systems with mGGA's.

Best,
Luis

--
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/ZHD9Y8dsH2E/unsubscribe.
To unsubscribe from this group and all its topics, send an email to cp2k+uns...@googlegroups.com.
To post to this group, send email to cp...@googlegroups.com.
Visit this group at http://groups.google.com/group/cp2k.
For more options, visit https://groups.google.com/d/optout.



--
Luis A. Ruiz Pestana, PhD
Chemist Postdoctoral Fellow
Lawrence Berkeley National Laboratory & UC Berkeley
Berkeley, CA 94720

S Ling

unread,
Oct 14, 2015, 6:47:11 PM10/14/15
to cp...@googlegroups.com
Hi Luis,

I seem to remember there is no analytical stress tensor implemented for meta-GGAs in CP2K because they use laplacian of the density, see:


In addition, the stress tensor from MD and single point energy calculations will be different, see earlier post by Prof Hutter:


If you decide to run the numerical stress tensor calculations, please keep in mind that you will need to use tighter convergence criteria (e.g. plane-wave cutoff, energy, force).

I believe you will be able to find more useful information if you search in the mailing list.

SL




--
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.
Reply all
Reply to author
Forward
0 new messages