Anisotropic stress for pure compressed gas

Hi everyone,

Just wondering if anyone encounters an anisotropic stress profile in a pure gas simulation.
I have run a pure compressed gas simulation.(i.e. the simulation box is filled with compressed gas, and only one type of gas is in the simulation box.)
Then, I used “compute stress/atom” to calculate stress(xx), stress(yy) and stress(zz), but I noticed an anisotropic stress profile. I am not sure why this happens, I am assuming it should be an isotropic stress profile. If you have any advice on this, it would be highly appreciated.

Thanks in advance!
Regards,
Darren