Interpret Current Density Results from DEVICE CHARGE



I am trying to understand my current density simulation results from DEVICE CHARGE. The structure is a simple reverse biased PN junction. Below shows the result: (right side is P, left side is N)
[1] Jn:

[2] Jp:

From the perspective of current continuity, the integration of current density over current flow cross section should be constant. Even the current flow direction is not straightforward in my case, I still have trouble understanding why it shows such an abrupt change of current density - basically it seems there is no current flowing in neutral P section and N section.

I also notice that the position of Jn and Jp are slightly different. So I am guessing that Lumerical only solves certain types of current (diffusion for example?) at certain position (quasi-equilibrium boundary of pn junction for example?). Or am I getting something completely wrong?



Hi @luyazhang, the visualizer in DEVICE does some interpolation/averaging when visualizing the current density data. This is why the values in the plots can sometime look inaccurate. The values shown by the visualizer gets affected by the mesh size and that is why you see a larger value in the depletion region (since you have finer mesh there). If you plot the data with some other third party tools like Matlab, you will get a better picture. Here is a KX post that talks about how you can get your data in Matlab for plotting purpose: The post talks about 3D data but the same approach will work for 2D as well. Another option would be to interpolate the data on to a rectangular grid and then plot it. Take a look at this other post to see how you can interpolate the finite element data saved by DEVICE onto a rectangular grid: