Hello
I’ve read some articles about active brownian particles with help of LAMMPS. But I don’t know how to realize the self-propulsion part.I have no idea that which command refers to the self-propulsion force.
Best regards
Mark
Hi Mark,
As far as I know there are quite a few models for “self-propulsion”/organized motion out there. I don’t think there are any that are native to LAMMPS but I am guessing some one has probably done it by now. How to implement what you are trying to do, will matter a great deal on what you have in mind.
Hello
I've read some articles about active brownian particles with help of LAMMPS.
But I don't know how to realize the self-propulsion part.I have no idea that
which command refers to the self-propulsion force.
the obvious address to ask for this would be the authors of the papers
you have read.
if they have used an existing feature, they should have inputs and if
they have implemented new code, they should be interested to share it
.
axel.
Dear Mark, Eric, Axel,
> I've read some articles about active brownian particles with help of LAMMPS.
> But I don't know how to realize the self-propulsion part.I have no idea that
> which command refers to the self-propulsion force.the obvious address to ask for this would be the authors of the papers
you have read.
if they have used an existing feature, they should have inputs and if
they have implemented new code, they should be interested to share it
My understanding is that in this field, there is very little open-source
activity. The (hypothetized) reason is that a typical Brownian dynamics
integrator is rather simple to implement. I would be very pleased to be proven
wrong but I guess that most requests for code would end up in "not ready/clean
enough/whatever".
Mark, active motion can be done in many ways.
1. velocity-dependent friction: Erdmann et al, Eur. Phys. J. B 15, pp 105-113
(2000) doi:10.1007/s100510051104
2. fixed velocity, brownian orientation. This is a theoretical simplification
found in the reference above.
On top of the ideas above, you need to add an interparticle potential. This has
to be defined either by you or by a reference paper.
There are much more complex models that involve a full description of the
solvent and of the phoretic mechanism that generates the active motion.
An interesting review for Brownian models is Romanczuk et al, Eur. Phys. J.
Special Topics 202, 1–162 (2012) doi:10.1140/epjst/e2012-01529-y
In LAMMPS, you would have to implement a fix that realizes one of the models
above. For a starter, you need to use an "atom style" that possess an internal
angular degree of freedom. The fix would update the velocity as depending on
these coordinates.
Anyway, as Axel write, the starting point is a model that you want to simulate.
Then, you can look for an implementation.
Regards,
Pierre
Pierre,
I hope you are wrong. Playing with some of these models of collective motion would probably be the most fun I could have with LAMMPS.
I was thinking of cooking up some simple active matter stuff for LAMMPS but am still finalising my thesis (which is not on active matter) so this has been on the back-burner. There is some interest in this in our group now though, so it will probably get done in the coming months.
The basic “engine” should be simple enough as it’s just adding a constant force if you do Brownian dynamics.
The collectiveness of the motion might be more tricky but should definitely be doable, the easiest would probably be a fix that accesses a neighbour list and modifies the active component of the force accordingly.
If, before that, you make some progress, please do share. I think such a feature would be really great to have in LAMMPS, as it allows for massively parallel active matter simulations.
Don’t know if “active matter” includes birds, but I remember
looking at a couple papers on simulating the flocking
behavior of birds when they fly in a coordinated fashion.
And thinking that kind of model would be do-able
in LAMMPS. As I recall, it involves adjusting the
velocity of each particle (bird) each step to be more aligned with
its nearest-neighbor bird velocities.
Might enable a definitive answer to the vexing question
of what is the average air speed of an unladen coarse-grained
African swallow …
Steve