About per-atom stress

Thanks for the reply.

Yes, it is correct in the sense of summing them up and you get the correct total pressure. I tested that, it’s true. But my concern is the distribution of the Pzz component (Pxx and Pyy seem to be uniform ) in some regions of the simulation box. My results did not stand against the correctness of the above mentioned aspect, as it is negative the lower region and positive in the upper region. If you sum them up, they cancels out. You still get the correct pressure. I’m just saying that the distribution is somewhat not physical.

Is it possible that fix rigid or restraining force or the pressure coupling I applied might be able to cause this?



I could definitely see how adding a local external force could give a
non-uniform stress profile. Why would that be unphysical?