Dear All,

I am working with the given example on the thermal conductivity calculation example using the Green-Kubo formula. I noticed an inconsistency between the equation and the unit conversion in the example.

considering the integral;

\frac{V}{3kb*T^2} \int \langle J \cdot J \rangle dt

Now, if we don’t divide the J_{computed} by the volume, which is required by the LAMMPS documentation to get the actual J, the GK formula is;

\frac{V}{3kb*T^2} \int \frac{1}{V^2} \langle J_{computed} \cdot J_{computed} \rangle dt =

\frac{1}{3V kbT^2} \int \langle J_{computed} \cdot J_{computed} \rangle dt

the variable `scale`

is putting volume in the denominator of the integral coefficient, which is for the case we have not already divided J_{computed} by the volume. However, the script has three lines doing so before the GK calculation.

My question is, am I missing something in my logic?

Amir Behbahanian

Mechanical & Aerospace Engineering

Utah State University

4110 Old Main Hill

Logan, Utah 84322