Dear all,

It’s introduced in the manual that the compute heat/flux command can calculate the heat flux based on the compute ke/atom, compute pe/atom and compute stress/atom commands. The heat flux calculation will include exactly the same terms in ke, pe and stress. I notice that the contributing authors Reese Jones (Sandia), Philip Howell (Siemens), Vikas Varsney (AFRL) didn’t offer any references related with this calculation method. So can anyone offer some references to demonstrate that this is method is right, especially when this method is used with many-body potentials?

Thank you very much!

Best wishes!

Sincerely yours

Shenghong Ju

I'll let Reese answer this. See also section 4.20 of the

manual, which overviews 3 competing methods.

This does indeed work for manybody potentials.

Steve