Hi community,

I'm working on a drift-diffusion model which gives equations of the following type:

(a[0, 1] ' ' )[z] + 5.78 a[0, 1] [z] - 0.036 (a[0, 1] ' )[z] + (UnitStep[-1.63265 + z] - UnitStep[1.63265 + z]) == 0

wher a[ i , j ][z] are the expressions of interest (here as an example for i=0, j=1).

The most simple system of equations for a[0,1], a[1,1], a[1,2], a[2,2] can be found in the notebook.

The boundary condition should be n vanishes in both directions (it is a density).

I was trying NSolve, but unfortunately I only get oscillating results.

Is it possible that Mathematica has a problem with this type of DGL? I don't get any meaningful results for uncoupled equations either. What can I do ?

Best regards,

Michael