As a first approximation, you can look at the Green-Kubo formula you're using to compute thermal conductivity in equilibrium molecular dynamics. The thermal conductivity is directly proportional to the integral (over infinite time lag) of the heat flux density autocorrelation . Meaning that if you have a high conductivity material like graphene, the autocorrelation function will probably decay slowly with time lag - which means that you will need a long correlation time to get a converged result. And I think that's about all you can guess about the value of your correlation time without running some computations.
Turns out that for graphene, decays extremely slowly. See for instance PhysRevB.92.195404 . The order of magnitude is (at RT) a couple of nanoseconds.
In practice, you'll want to fit the tail of the autocorrelation function with say a power law, and extrapolate to high time lag values - up to the point where your thermal conductivity is converged to a certain tolerance.
As a first approximation, you can look at the Green-Kubo formula you're using to compute thermal conductivity in equilibrium molecular dynamics. The thermal conductivity is directly proportional to the integral (over infinite time lag) of the heat flux density autocorrelation . Meaning that if you have a high conductivity material like graphene, the autocorrelation function will probably decay slowly with time lag - which means that you will need a long correlation time to get a converged result. And I think that's about all you can guess about the value of your correlation time without running some computations.
Turns out that for graphene, decays extremely slowly. See for instance PhysRevB.92.195404 . The order of magnitude is (at RT) a couple of nanoseconds.
In practice, you'll want to fit the tail of the autocorrelation function with say a power law, and extrapolate to high time lag values - up to the point where your thermal conductivity is converged to a certain tolerance.
Dear Arthur
I would like to thank you for your answer.
It really helps me to a first approximation that high conductivity material, high correlation length .
- Badges
- Science topic
- Similar topics
- Physical Sciences
- Materials Science
- Materials
- Carbon Materials
- Graphite
- Graphene