Pressure tensor in confined systems

Dear LAMMPS users

I am using LAMMPS version 22 Aug, 2018. I am simulating a 3d box containing aspherical particles defined by GB potential. I have confined the box using wall on both sides of box perpendicular to an axis and I am trying to calculate the pressure tensor in such a system. I have defined the pressure of this system using compute pr all pressure 1_temp where 1_temp is the temperature calculated from the fix nvt/asphere (also this temperature is calculated by subtracting ‘N’ degrees of freedom for N, no. of uniaxial ellipsoidal particles using compute_modify 1_temp extra/dof N ). Thus the pressure tensor is calculated as c_pr[1]=pxx, c_pr[2]=pyy,c_pr[3]=pzz.

My queries are as follows

  1. Should the pressure normal to the wall be always greater than the tangential pressure? That is not seen in my output.

  2. If the location of wall is changed from x-axis to y-axis or z-axis, accordingly, should the values of pressure tensor be changed? i.e. if wall is perpendicular to x-direction, the values of pyy and pzz should be overlapped (app. same) and pxx value should be different and if it is normal to y-axis, the pxx and pzz should be overlapped and pyy should be different. In my case the values of pxx, pyy are always greater than pzz independent of location of wall in x-axis or y-axis or z-axis.

Thank you in advance.