I am performing particle dynamics simulations to model biological cell and tissue mechanics using the LAMMPS 29 October 2020 version.
The cytoplasmic environment of the cell is highly viscous, hence the motion of particles is overdamped. I am using the following command lines in the input script:
fix 1 all langevin 300 300 6.20343e-10 236509
fix 2 all bd 6.20343e-10
In the system of units, time is in seconds and temperature is in Kelvin.
I also use the line:
compute_modify thermo_temp dynamic/dof yes
since the number of particles change during the simulations.
The temperature that I get as thermo output (0.0005 K) is much less than that specified as Tstart and Tstop (300K) in the fix langevin command and the contribution of kinetic energy to total energy is very small. Is it a possible observation in overdamped systems or is there a probability that the thermo output is not measuring the desired temperature?
Difficult to say since you are using a fix that is not part of the canonical LAMMPS distribution so we don’t know what it does and whether it has an impact on the kinetic energy.
Thank you so much for your response. The code for fix bd can be found here:
The system of units, named “cell” is here:
Instead of fix nve from the LAMMPS canonical distribution, fix bd is used to account for changing number of particles. Does taking a look at this help us in understanding why the temperature value is so small compared to Tstart and Tstop?
Sorry, but that explanation makes no sense. Fix nve does not care whether particles are added or removed. However, I do see in the fix bd source code that it sets velocities to 0 at the pre_force step.
Bottom line, you have to talk to the developers of this package.
Thank you so much for your response and pointing out that the velocities are set to zero at the pre_force step.
“Fix nve does not care whether particles are added or removed” – Is the number of particles not expected to remain constant in a micrcanonical (NVE) ensemble? N keeps on increasing in our simulations of (biological) cell growth and proliferation. Can we use fix nve along with fix langevin in our simulations to implement Brownian dynamics?
Please read the documentation. Using fix nve does not mean you have an NVE ensemble. It just does plain time integration. That will give you an NVE ensemble provided there are no other manipulations of the system. Adding atoms is one of them.
Again, what you are asking is specific to your model and cannot be answered in general. Since your model depends on code we know nothing about, you have to discuss with the authors/maintainers of that code.