PIMD with CP2K

[Yun An]

Dear all,

I want to perform PIMD with cp2k, according to the manual, one can add a PINT session to run path integral simulations; also, can use i-PI mode  DRIVER. What is the difference between those two, is that in principle they both can do PIMD? I tried a H2O molecule with run_type of PINT, then I got centroid-pos-1.xyz and centroid-vel.xyz, I thought the centroid-pos-1.xyz  is the atom positions of each replica, but what is the meaning of centroid-vel.xyz?  Also, I want to see the trajectories, is this directly given in centroid-pos-1.xyz or one need to use some script to transform, because from current centroid-pos-1.xyz , the atoms are not quantum at all.

Thank you all in advance for any help.

Best regards,

Yun

[Thomas Kühne]

Dear Yun, 

you can indeed conduct PIMD using the CP2K built-in PINT module, or using the driver mode 

via the i-Pi program. The latter more functionality for more sophisticated PIMD simulations, 

though the former already permits to conduct all sort of conventional PIMD simulations and has 

the advantage of having everything in one package. 

The centroid-pos-1.xyz contains not the coordinates of every bead, but only the ones of the centroid, 

i.e. the center of mass of the closed ring-polymer. The centroid-vel-1.xyz contains the corresponding 

velocities, but be aware that except the when centroid molecular dynamics is used, the dynamics of 

the ring-polymer is unphysical, only static ensemble averages are. 

Best, 

Thomas

--
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/3aa5af5d-4b37-471b-ae2b-18dc83d0b3b5%40googlegroups.com.
<H2O-centroid-pos-1.xyz><H2O-centroid-vel-1.xyz>

==============================

Thomas D. Kühne

Dynamics of Condensed Matter

Chair of Theoretical Chemistry

University of Paderborn

Warburger Str. 100

D-33098 Paderborn

Germany

tdku...@mail.upb.de

+49/(0)5251/60-5726

[Yun An]

Dear Thomas,

Thank you for the kind reply, it helps a lot. I still have some questions:

I use 4 beads to test h2o molecule, is it means  the trajectory is represented by the average replica positions of each atom, in the current case, is the all the atom positions of these four H2O-pos-?-1.xyz file?

Anther question is if I want to run MD simulations to see the quantum effects, for example, H2 and D2 diffusion in the metal surface, is it possible that cp2k combines PINT and metadynamics? Because the RUN_TYPE section to me is like either one can choose MD to perform normal MD or PINT to run PIMD, but not both at one calculations. So in such case, how should I consider quantum effects?

Thank you again for your help.

Best regards,

Yun

Thomas Kühne <tku...@gmail.com> 于2019年7月31日周三 下午4:16写道：

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/CKzhBph4yFc/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/C543F0FA-BD29-4238-B04B-C4E0CDAB9E00%40gmail.com.

[Thomas Kühne]

Dear Yun, 

yes the trajectories of the individual beads are in H2O-pos-*-1.xyz, whereas the 

centroid, which corresponds to the center of mass of the closed ring-polymer made 

up of the imaginary-time replica, is in centroid-pos-1.xyz. 

Conceptually, conducting PIMD-based metadynamics should be rather straightforward, 

however, since there is some bookkeeping is necessary to apply the bias potential 

to the individual beads I suspect it not to directly work as is. Yet, via i-Pi v2 in conjunction 

with PLUMED it is advertised to work: http://ipi-code.org/assets/pdf/manual.pdf

Best, 

Thomas

To view this discussion on the web visit https://groups.google.com/d/msgid/cp2k/CA%2BSzDiPznCF4%3Do%2BNOkE8x_Lau3ScWFHDjDYje05-ZJGT-SS68Q%40mail.gmail.com.
<H2O-pos-1-1.xyz><H2O-pos-2-1.xyz><H2O-pos-3-1.xyz><H2O-pos-4-1.xyz><h2o_test.inp>

[Yun An]

Dear Thomas,

Thank you very much for the kind reply.  I tried with PINT module in cp2k, the weird thing is,  I want to fix some atoms during the calculations, but it seems that the "CONSTRAINT" section didn't work, because the atoms which are supposed to be fixed moves a lot according to the structures it gives out. Do you know what is the problem and how to fix it?

I attached the input file, thanks a lot in advance.

Best regards,

Yun

Thomas Kühne <tku...@gmail.com> 于2019年8月1日周四 下午10:55写道：

To view this discussion on the web visit https://groups.google.com/d/msgid/cp2k/AF2E4F64-1C56-4B51-9B8C-0CB8120FAE3C%40gmail.com.

[Thomas Kühne]

Dear Sun, 

to the best of my knowledge fixed atoms constraints are simply 

not implemented within PINT, which must be enforced on the centroid. 

The individual beads are still allowed to move (as long as the centroid 

is fixed) and will do so at a very high classical kinetic energy due to 

the presence of a rather high quantum kinetic energy, which can easily

be on the order of a few hundred kelvin! 

Cheers, 

Thomas

To view this discussion on the web visit https://groups.google.com/d/msgid/cp2k/CA%2BSzDiMnadRpxU0rg5d7t3uCVu-3vJ4k4Q4KHaZOj4PTmS_Kew%40mail.gmail.com.
<test_PINT.inp>

[Jinggang Lan]

Constraint atoms lead to nonphysical velocity assignment. It has been fixed in I-PI but not in CP2K as Thomas mentioned.

To unsubscribe from this group and stop receiving emails from it, send an email to cp...@googlegroups.com.

To view this discussion on the web visit https://groups.google.com/d/msgid/cp2k/3aa5af5d-4b37-471b-ae2b-18dc83d0b3b5%40googlegroups.com.
<H2O-centroid-pos-1.xyz><H2O-centroid-vel-1.xyz>

==============================

Thomas D. Kühne

Dynamics of Condensed Matter

Chair of Theoretical Chemistry

University of Paderborn

Warburger Str. 100

D-33098 Paderborn

Germany

tdku...@mail.upb.de

+49/(0)5251/60-5726

--
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/CKzhBph4yFc/unsubscribe.

To unsubscribe from this group and all its topics, send an email to cp...@googlegroups.com.

To view this discussion on the web visit https://groups.google.com/d/msgid/cp2k/C543F0FA-BD29-4238-B04B-C4E0CDAB9E00%40gmail.com.

--
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 cp...@googlegroups.com.

To view this discussion on the web visit https://groups.google.com/d/msgid/cp2k/CA%2BSzDiPznCF4%3Do%2BNOkE8x_Lau3ScWFHDjDYje05-ZJGT-SS68Q%40mail.gmail.com.
<H2O-pos-1-1.xyz><H2O-pos-2-1.xyz><H2O-pos-3-1.xyz><H2O-pos-4-1.xyz><h2o_test.inp>

--
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/CKzhBph4yFc/unsubscribe.

To unsubscribe from this group and all its topics, send an email to cp...@googlegroups.com.

[Chenghan]

Hi,

This is Chenghan. Thank Thomas and Jinggang for the information. That's very helpful!

I also want to use PLUMED in PIMD. A question I have:

Does i-Pi act on the bias forces from PLUMED on the centroid position or treat each bead independently and calculate bias forces for each bead seperately?

Best Regards,

Chenghan

[Thomas Kühne]

Dear Chenghan, 

yes, but please have a look at section 4.5 of: https://arxiv.org/pdf/1808.03824.pdf 

for details. 

Cheers, 

Thomas

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/0dfde21d-2e32-4951-823b-8f0c5827f365n%40googlegroups.com.

[Chenghan Li]

Hi Tomas,

Thanks a lot for the paper!

I can see that the bias forces were acted on the centroid in the example of H5O2+ in Section 4.5 of the paper.

I am wondering if it is possible to calculate and apply bias forces on each PIMD bead instead?

Best Regards,

Chenghan

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/A29FF3FB-D268-4509-AAC4-AFCE9A44EBA8%40gmail.com.

[Thomas Kühne]

Dear Chenghan, 

this definitely possible, but I suspect not implemented. Applying the bias to 

the centroid is not only easier, but apart from unusual scenarios IMHO also 

equivalent. In any case, this seems more like a topic for the i-Pi marling list. 

Greetings, 

Thomas

To view this discussion on the web visit https://groups.google.com/d/msgid/cp2k/77F8C577-BBE1-4DD1-91B6-5AA738B93346%40gmail.com.

[Chenghan Li]

Dear Tomas,

Thank you very much! I’ll ask there and stop spamming here.

Best Regards,

Chenghan

To view this discussion on the web visit https://groups.google.com/d/msgid/cp2k/72F7FA4A-2B7D-4396-99C4-F68FCC4C082A%40gmail.com.