Learning Forum

[tade]

Hello

My area of work involves Carrier Density calculations using CHARGE software.When I adjust the range interval from 1V to 0.1V, the results differ. This discrepancy results in a significant difference in the reflection calculated using the Drude model. Can you assist me in fixing the issue or provide guidance on setting boundary conditions to achieve accurate results?

[Amrita Pati]

Hi Tade,

In CHARGE, the solution of the Poisson equation at a given voltage point depends upon the solution at the previous voltage point. Therefore, the smaller are the voltage steps, the easier it is to find a solution and also obtain convergence. We have often seen that making the voltage steps too large leads to the divergence of the simulation altogether. There are a set of steps that can be followed if the simulation diverges but that is not the case here. 

I would rather suggest you to perform convergence testing i. e. keep reducing the voltage steps until the results you are looking at, converge. This is the usual recommendation for determining the optimal value of a lot of parameters. At the end, you may choose an interval at which the results start approaching a fixed value. Maybe you can reduce the range of voltages and look at smaller intervals if it saves time. 

Regards,
Amrita