LAMMPS WWW Site - LAMMPS Documentation - LAMMPS Mailing List Archives
Re: [lammps-users] Getting the DPD thermostat right
[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

Re: [lammps-users] Getting the DPD thermostat right


From: Peter Vanya <pv278@...690...>
Date: Fri, 18 Aug 2017 11:16:10 +0100

Dear Axel,

thanks a lot for the clarification. So I understand that the input below fully describes the DPD in LAMMPS:

pair_style dpd T rc seed
pair_coeff * * a gamma rc
fix 1 all nve

I am not sure if I am the only one, but I think if would be very useful if these lines were stated in the DPD section of the manual.

To improve my understanding I am now trying to reproduce a more complex system -- phase separation of a symmetric diblock copolymer, as described in this paper. Polymerisation  N = 10, like bead interaction 25 and cross interaction 50. In theory I should be easily getting a lamellar phase after as few as 20k steps, but in practice I am see only a disordered phase. However, I was able to get this phase with the DL_MESO package. I am sending the screenshots and input file attached.

Can you please suggest what I am doing wrong? I have tried everything I could think of.

Thank you in advance.

Kind regards,

On 17 August 2017 at 18:07, Axel Kohlmeyer <akohlmey@...24...> wrote:


On Thu, Aug 17, 2017 at 12:16 PM, Peter Vanya <pv278@...690...> wrote:
Dear all,

I have spent a considerable amount of time trying to make DPD with the thermostat work correctly, yet my results are clearly wrong. Can you please help?

Reading the manual, I understand that there are in general two options for thermostatting, starting from temperature T, the cutoff, interaction parameter a, friction gamma, and the seed:

1. 
pair_style dpd T cutoff seed
pair_coeff * * a gamma
fix 1 all nve
fix 2 all langevin T T gamma seed

2. 
pair_style hybrid/overlay dpd T cutoff seed dpd/tstat T T cutoff seed
pair_coeff * * dpd a gamma
pair_coeff * * dpd/tstat gamma
fix 1 all nve

​both of these inputs make no sense.
the effect of pair style dpd/tstat is *already included* in pair style dpd.​ so there is no point in using hybrid/overlay. the reason for dpd/tstat is to apply it to other potentials.

similarly, it is not correct to use fix langevin in combination with pair style dpd. you are applying a thermostat twice with conflicting approaches. that is likely to be bogus.

​axel.​

​​

As far as I understand, the first option makes the particles interact with an implicit solvent and not each other and so is not a proper DPD. Hence, only the second option remains.

Trying the second option on a pure DPD liquid, the temperature goes up to 3.5 and does not settle (it should settle at 1.0), and the RDF is totally incorrect close to r = 0.

Can I please ask what exactly wrong with my input, and how to get it right? I tried to follow the manual as closely as it gets. Attached I am sending the data and input file together with the RDF.

Thank you in advance.

Kind regards,
--
Peter Vanya

PhD candidate | Dept. of Materials Science and Metallurgy | University of Cambridge
+44 74 037 505 20 | pv278@...3639...0... | mml.msm.cam.ac.uk

------------------------------------------------------------------------------
Check out the vibrant tech community on one of the world's most
engaging tech sites, Slashdot.org! http://sdm.link/slashdot
_______________________________________________
lammps-users mailing list
lammps-users@...655....net
https://lists.sourceforge.net/lists/listinfo/lammps-users




--
Dr. Axel Kohlmeyer  akohlmey@...24...  http://goo.gl/1wk0
College of Science & Technology, Temple University, Philadelphia PA, USA
International Centre for Theoretical Physics, Trieste. Italy.



--
Peter Vanya

PhD candidate | Dept. of Materials Science and Metallurgy | University of Cambridge
+44 74 037 505 20 | pv278@...690... | mml.msm.cam.ac.uk

Attachment: diblock.in
Description: Binary data

Attachment: vmdscene_dlms.png
Description: PNG image

Attachment: vmdscene_lmp.png
Description: PNG image