Impact of contact size on pn junction capacitance and transient response



I’m simulating the interleaved pn junction phase shifter of the knowledge base. However, I noticed that if I increase or decrease the width of the simulation region (the x span) which increases the portion/width of the Aluminum contacts inside the region, it affects the result (changes junction capacitance). What I understood that the mesh is applied to the semiconductor region only and the drift diffusion solver solves the equations only on the mesh. Could you please explain a little about the effect of contacts on the simulation? Would increasing the contact:semiconductor interface area affects the transient response as well?



Sorry for the late response. You should not see any difference in capacitance in this particular structure based on the overlap of the contact and the semiconductor. I have checked the example and have changed the x span of the simulation region to run the simulation with a little portion of the contacts as well as a large portion of the contacts. In both cases, the junction capacitance was almost the same. If you plot the two on the same graph, there is very slight difference. This difference is most likely coming from a coarse mesh. By using a fine enough mesh, you could make the results overlap more nicely.


Thanks @aalam your answer helps.