Hello,
I’m trying to calculate the thermal conductivity of a material that I’m modeling and found within the documentation an example using a green-kubo method for Argon. I decided to test the script to see how it is working and noticed that the resulting value for thermal conductivity at room temperature (25 C) is incorrect. The only thing that I have changed from the example script was the temperature from 70 K to 298.15 K. The average value resulting in the simulation is 0.306 W/mK when the reported value for Ar at that temperature is about 0.016 W/mK; this about 19 times the value of what it should be. I’m not expecting it to be exactly the same, but this is an unreasonable result. Am I not understanding something about how this works or what the result of the average conductivity represents?
Thanks,
The only thing you changed was the temperature. You also need to equilibrate the system at this new temperature, probably with different settings.
Green-Kubo is an equilibrium method; it only gives meaningful results when the system is in equilibrium. I bet that your system was not in equilibrium if all you did was change the temperature, you also need to let the volume and pressure equilibrate beforehand.
Drew Rohskopf
Atomistic Simulation & Energy Group
Massachusetts Institute of Technology
(404)403-0313
That 0.016 W/mK value is also from a gas at certain conditions (maybe STP?) but it sounds like you started with the same volume as the solid Ar example. So you basically calculated the thermal conductivity of a very condensed gas.
Drew Rohskopf
Atomistic Simulation & Energy Group
Massachusetts Institute of Technology
(404)403-0313