Dear LAMMPS users,
I have been trying to implement RNEMD on an LJ particle system with repulsive walls. However, I have been unsuccessful. Half of the simulation shears slower than the other half, creating an asymmetric velocity profile. How can I solve this problem? Below is my input file. Thank you for your time.
Regards,
Michelle Len
Graduate Student
Mechanical Engineering
University of California, Merced
What do you mean by uneven? Is the variation in shear rate transient or spatial? If its spatial, I’d suggest reading the documentation.
Hi Eric,
Thank you for your reply. Visually, the particles do not shear evenly in the x-direction when the dump file is viewed in OVITO. I also checked the last timestep’s output of the fix ave/spatial command, found in the output file “vxprofile.” Indeed, the velocity profile is not symmetric as it would be in a simple RNEMD system without any repulsive walls. I have also tried increasing the run time from 10000 to 2000000 timesteps. While the slopes in the velocity profile become much more linear, the values of the slopes are still uneven, indicating that half of the system is shearing slower than the other half. I’ve attached an image of the velocity profile at the end of 10000 timesteps. I hope it helps. Thank you for your time.
Regards,
Michelle Len
Graduate Student
Mechanical Engineering
University of California, Merced
