Optical generation rate object and unknown error


#1

I’m trying to simulate the PD presented in the attached paper. However when in the device simulation I increase the reverse bias voltage to anything above -4 I get the following error when the simulation reaches 98%:

MININT-BS5KS26(process 0): The program terminated due to an error: The simulated result is incomplete. Please review the log files for details.
MININT-BS5KS26(process 0): Error: there was an unknown parallel error. The error code is 9002, the process number is 0
I18N Runtime Warning:
Missing ICU data file detected while processing directory containing libut.{dll|so|sl|dylib}.
Hint: Check for a misconfigured environment or installation.
I18N Runtime Warning:
Misconfigured ICU_DATA directory path.
Hint: Check for a misconfigured environment or installation.

I cannot figure out what the issue is since it works as long as the voltage is below -4.

Another problem I have is when the generation rate .mat file is imported into the DEVICE from FDTD, the Optical GenerationA high-responsivity photodetector absent metalgermanium.pdf (2.4 MB) object position is changed. I set the coordinates and the span in FDTD, and I expect it should be the same when I import it into the DEVICE however, the imported generation_rate object is usually shifted. Any suggestions on fixing this?


#2

Since the simulation is working for voltages below -4 V, one possible reason for the solver to fail at higher voltage could be that the change in electric field is becoming very sharp in certain regions and you may need a finer mesh in regions where the Efield is peaking. You can choose “save and quit” to check the electric field up to the last bias point where the simulation converges and figure out these regions. Another thing to check would be your doping objects. If you are using constant doping objects then I would recommend using diffusion doping objects instead.

The shift is most likely due to x,y,z values of the (generation) source. The x, y, z values of the source can be used to shift the imported generation data. If these values are nonzero then the generation profile you have imported will be shifted. Usually when you add the import generation object, it centers around the center point in the current CAD window and if this point is not (0,0,0) then it is possible that the x,y,z values of the import generation objects are nonzero.