Dear all,
I recently had the opportunity to compute a shear viscosity with the Green-Kubo approach, and I wanted to implement the method discussed in the supporting information of the paper in reference below, where all 9 terms of the deviatoric stress tensor are used (section S13, page 15) :
https://pubs.acs.org/doi/suppl/10.1021/acs.nanolett.5c03950/suppl_file/nl5c03950_si_001.pdf
However I came upon a practical problem: in LAMMPS, compute pressure only provides “A symmetric pressure tensor, stored as a 6-element vector”. I was wondering: is the stress tensor expected to be diagonal at any time, or only on average?
If instantaneously the stress tensor is not diagonal: I’m willing to investigate in the code, but in case someone already knows the answer I dare to ask: does LAMMPS simply ignore one of each pair of non-diagonal terms?
I’m using LAMMPS (29 Aug 2024 - Update 2) version.
Best wishes,
Laurent