[lammps-users] Could you help me for the problem of compute heat/flux?


Hello, I used the compute heat/flux command in lammps to compute the thermal conductivity of Ar basefluid and Ar-Cu nanofluid ( units lj ), and I find only when the timestep is as small as 0.0005, the thermal conductivity of Ar is convergent to be about 0.132(2048 Ar particles), but if the timestep is 0.002, the result will be rather bad, whic would be larger and larger when the m (the parameter in the discrete expression for thermal conductivity) is increasing, which should not be that, because the result others report is what when the timestep is 0.002.

And as for Ar-Cu, the thermal conductivity is as large as ten times compared with the result in others’s articles.

I notice the words of German Samolyuk in the following , so I want to call for your help. Whether there are some errs in the current compute heat/flux of lammps? Or it’s not appropriate for liquid-solid nanofluid? Can you show me your compute_hf.cpp and compute_hf.h? Or can you give me some suggestions?

****************************************************************************************************
>>>>>>>>>> I was not sure if it’s a good idea to all lammps users, because it
>>> still in testing phase.
>>>>>>>>>> If you think it will be interesting to everybody I will do it.
>>>>>>>>>>
>>>>>>>>>> I’ve attached slightly modified style.h file. I added compute_hf.h
>>> to it
>>>>>>>>>> and compute_hf.cpp, computer_hf.h files. I’ve used
>>>>>>>>>> compute_group_group.cpp
>>>>>>>>>> as a prototype, as soon as it has cycle over pair in it.
>>>>>>>>>>
>>>>>>>>>> It should be compiled
>>>>>>>>>>
>>>>>>>>>> make yes-dpd,
>>>>>>>>>>
>>>>>>>>>> - it uses velocities of ghost atoms.
>>>>>>>>>>
>>>>>>>>>> I’ve also attached input file for Ar computation with two
>>>>>>>>>> temperatures
>>>>>>>>>> (in comment).
>>>>>>>>>> the results could be compared with results presented in chapter 2.18
>>>>>>>>>> of S. Yip (ed.), Handbook of Materials Modeling, 763?771. c 2005
>>> Springer. Printed in the Netherlands.
>>>>>>>>>> I’ve deleted first 1000 iterations and calculated averaged values
>>> over
>>>>>>>>>> 20000 samples with 50 steps shift.
>>>>>>>>>>
>>>>>>>>>> Known problems:
>>>>>>>>>>
>>>>>>>>>> 1) It’s working with pair potentials only,
>>>>>>>>>> 2) I do no know how to include Coulomb interaction.
>>>>>>>>>>
>>>>>>>>>> I would really appreciate any comments and suggestions …
>>>>>>>>>>
>>>>>>>>>> Best,
>>>>>>>>>> German
>>>>>>>>>>
>>>>>>>>>>
>>>>>>>>>> On Wed, Apr 8, 2009 at 1:10 AM, Mario Pinto
>>> <[email protected]…24…> wrote:
>>>>>>>>>>
>>>>>>>>>>> Dear German,
>>>>>>>>>>>
>>>>>>>>>>> I would be very grateful if you could share your code. Could you
>>> please
>>>>>>>>>>> mail
>>>>>>>>>>> it to me and also put it on the mailing list?
>>>>>>>>>>>
>>>>>>>>>>> Thanks!
>>>>>>>>>>> Mario Pinto
>>>>>>>>>>>
>>>>>>>>>>> On Wed, Apr 8, 2009 at 2:47 AM, German Samolyuk
>>>>>>>>>>> <[email protected]…24…>
>>>>>>>>>>> wrote:
>>>>>>>>>>>
>>>>>>>>>>>> I have some variant of code with calculation of j(t) - heat
>>> flux, actually energy flux.
>>>>>>>>>>>> For the calculation of lambda in the few components systems. I
>>> tested it on a linux cluster using 32 Xenon processors for Ar -
>>> solid and liquid.
>>>>>>>>>>>> It looks like it’s working OK. So I can share it. I will appreciate
>>>>>>>>>>>> any information of
>>>>>>>>>>>> mistakes in the code.
>>>>>>>>>>>>
>>>>>>>>>>>> But I do not know how to include Coulomb interactions in this code.
>>>>>>>>>>>> As I understand pair->single() returns only real space part of
>>> coulomb interaction. How I can calculate rest of the sum which
>>> is calculated in kspace?
>>>>>>>>>>>>
>>>>>>>>>>>> Thanks,
>>>>>>>>>>>> German Samolyuk

|