Hi all,
I wanted to compare MD results for a confined fluid flow through two parallel plates to analytical solutions. I applied some interaction parameters between fluid and wall so that no-slip B.C is happen.
Now, Could you tell me how to relate the external force which is given by LAMMPS with the pressure gradient? As you know in analytical solution we have pressure gradient and fluid viscosity, and to compare MD results with exact solution, I need to related external force applied to each atom with pressure gradient through the channel.
Any helps would be really appreciated.
Thanks,
Omid
Hi all,
I wanted to compare MD results for a confined fluid flow through two
parallel plates to analytical solutions. I applied some interaction
parameters between fluid and wall so that no-slip B.C is happen.
Now, Could you tell me how to relate the external force which is given by
LAMMPS with the pressure gradient? As you know in analytical solution we
have pressure
"as you know" is a strong statement in this community. 
most people
probably do not know (and care) about continuum model solutions.
gradient and fluid viscosity, and to compare MD results with exact
solution, I need to related external force applied to each atom with
pressure gradient through the channel.
two comments on this statement:
- *both*, your analytical model and MD are computing *exact* solutions.
they are just different models with different resolution based on different
assumptions about the nature of matter.
- relating microscopic to macroscopic properties is not exactly a simple
thing because of the difference in representation and assumption of length
(and time) scales. i suspect you need to spend much more time to study the
literature describing how to do that. a starting point could be to review
the papers related to the atoms-to-continuum package (AtC) in LAMMPS and
specifically the fundamental publication or textbook knowledge used in it.
axel.
Axel, thanks for replying soon.
I want to only compare the results for a no-slip B.C problem with MD in non-dimensional form.
So the final results should be the same, and MD should have similar results as analytical if not that might not be a reliable method!
I have already worked with MD in LAMMPS so much, and validated results with previous studies. But, this time I want to validate MD results in a very basic example with analytical solution in pressure driven flow. My problem is to relate the applied gradient pressure along the channel to the external force term applied to each atom in MD-LAMMPS.
Do you know any primary MD study validated with analytical solutions in basic flows?
Omid
Axel, thanks for replying soon.
I want to only compare the results for a no-slip B.C problem with MD in
non-dimensional form.
So the final results should be the same, and MD should have similar
results as analytical if not that might not be a reliable method!
i disagree with that assessment.
I have already worked with MD in LAMMPS so much, and validated results with
previous studies. But, this time I want to validate MD results in a very
basic example with analytical solution in pressure driven flow. My problem
is to relate the applied gradient pressure along the channel to the
external force term applied to each atom in MD-LAMMPS.
you are basically just repeating what you stated before. thus i will save
you repeating my answer. just re-read my previous e-mail.
Do you know any primary MD study validated with analytical solutions in
basic flows?
that is something that *you* should have researched with a proper
literature search *before* even starting a simulation. this is just a
basic good practice in research to survey the available literature before
re-inventing wheels or repeating mistakes.
axel.
Yes, you are right that I should have checked primary steps in MD validation with analytical solutions (in continuum at least)
I started my research from validating my scripts results with MD papers, not anayltical ones.
But for a simple case of pressure driven flow in no-slip B.C and non-dimensional form in which continuum assumption might be correct, then MD results should be same as analytical. Am I right?
Yes, you are right that I should have checked primary steps in MD
validation with analytical solutions (in continuum at least)
I started my research from validating my scripts results with MD papers,
not anayltical ones.
But for a simple case of pressure driven flow in no-slip B.C and
non-dimensional form in which continuum assumption might be correct, then
MD results should be same as analytical. Am I right?
this is yet another variant of the same question and an ill-formed
question to boot. my initial answer also applies to this version of your
inquiry. before asking yet another version of the same question, please go
and do your homework and research in the published literature or suitable
text books how to properly relate microscopic behavior to macroscopic
expressions. i still have the impressions that you are underestimating how
different those two views of condensed matter are.
axel.