Issue with kmcmultiple

[张恒悦]

Hello,

A bug was found. In the xtp tutorials of KMC, 1000 seconds of KMC simulation was scheduled to run. However, after checking the output, I found that only 1000 steps of KMC was executed. According to the manual, if 'runtime' is larger than 100, it may be steps.

What's more, another issue relating to KMC made me a bit nervous. I want to operate a kmc simulation to calculate charge mobility. I made this calculation with the calculator kmcmultiple. From the article [1], two methods can be applied to calculate the mobility. One is with the Einstein relationship, and another one is with the equation μ=μE/E (as shown in the following figure).

However, during the calculation, I changed the strength of electric field, but μE dosen't change a little. And I have no idea on how to calculate this with Einstein relationship for I found that only overall mobility was ouputed but the mobility of every direction cannot be found.

Sorry to disturb you! Thx a lot!

Best regards,

Hengyue Zhang

[1]Reggiani, L.; Brunetti, R.; Normantas, E. Diffusion Coefficient of Holes in Silicon by Monte Carlo Simulation. Journal of Applied Physics 1986, 59 (4), 1212–1215. https://doi.org/10.1063/1.336507.

[Jens Wehner]

Hello,

so if you want more than 100s kmcmultiple will assume that you want steps so you just got steps out. (Maybe we should fix that interface a bit. )

It could be that there is a bug, but KMC calculations are not easily converged so, it really depends on your energy landscape as well. 

Would be nice if you could post a minimal example.