Schred V2: Simulating n-type substrates and Vfb determination
Using the compiled Schred V2 code and supplied sample input files I am able to successfully model structures with a uniformly doped p-type Si substrate. I assume that to model an n-type substrate I need to change the sign of the parameter “Na”, which specifies the doping concentration. However, when I do this, I receive NaN values in the output “cv.dat” file for the capacitance in strong accumulation. The rest of the C-V curve looks alright. Am I doing something wrong here, or is this a bug in the code? Also, is there an easy way to determine the value of the flatband voltage from the simulation output? At the moment the best option I can see is to interpolate the values of fi_sc vs Vg to zero in the “av_dist.dat” file, but this is not precise.
0 Like 0 Dislike
Gokula Kannan @ on
The SCHRED V2.0 solves the schrodinger wave equation(SWE) only for the electrons, so the code supports only p-type substrate. The previous version of SCHRED I guess supports the hole SWE as well. The boundary conditions are defined at the start of the gate and the end of the substrate, where we force the potential (voltage) – Dirichlet boundary conditions. So, unfortunately there is no way of forcing fi_sc to zero to find the exact value of Vfb. So i suggest you take smaller steps of Vg once you determine the range at which fi_sc approaches zero, and you should get a much accurate plot of fi_Sc Vs Vg. That should be give you a reasonably accurate value of fi_sc. Hope that helps!
gokul
Reply Report abuse
Please login to answer the question.