TC calculation of Nano-fluid

Dear Steve and Lammps users
I try to simulate the Thermal Conductivity of Nano-fluid of Cu-Ar use GK formula in NVT ensemble with EMD method.
I have confirmed the GK formula for 2 component with it in single liquid Ar, just to verify my method.
When I put a nano particle Cu with radius 1.0nm with fix/rigid command, with 1% volume load the result is not satisfied.
I looked into mailing list before, and find Steve said when use fix rigid should think compute heat/flux make sense when comindividual atoms are part of rigid body.
I want to know how to put a nano particle with fix rigid or some command can make all cu atom be a nanoparticle and use compute heat/flux correctly.
Please forgive me if my question is not a lammps Q but a concept one.
I really appreciate for any comment.
Best regards.

I want to know how to put a nano particle with fix rigid or some command can make all cu atom be a >nanoparticle and use compute heat/flux correctly.

I don’t know the answer to this. At a minimum the kinetic energy term in what

comptue heat/flux calculates would be invalid for the atoms in a rigid body.

Possibly Reese or Jeremy can comment.

Steve