[lammps-users] thermal conductivity of metal

Dear all,
In lammps, there is a command “fix thermal conductivity” which use the Muller-Plathe algorithm , and this algorithm is sometimes called a reverse non-equilibrium MD (reverse NEMD) approach to computing thermal conductivity. However, when I use this command to calculate iron thermal conductivity, the results can not compare with the experiment datas. And I have read some papers about calculating the thermal conductivity of Ar utilizing that command, and the results coincides with the experiment results.
In Ar crystal, heat is conducted only by phonons, but in metal, heat is conducted mainly by free electrons. So I want to know whether it is properate to simulate thermal conductivity of metal utilizing this command of lammps or by MD. Or is there relevant computation module to calculate electronic thermal conductivity in lammps?

In the context of electron driven thermal conductivity, you’ll have to do quantum calculation, which lammps is currently not capable of. Lammps is a classic MD simulator.


Hi Maosheng,

You can use the Wiedemann–Franz law (L = sigma*T/ke) to calculate the electronic thermal conductivity ke. When comparing with the experiment data, you should be aware that it is for lattice thermal conductivity or total thermal conductivity. The thermal conductivity from classic MD simulation only calculate the lattice thermal conductivity.