Learning Forum

[Kryzchel Anne Dela Cruz]

Hi,

I would like to ask for help regarding MQW Gain Solver.

The peak wavelength in the MQW gain solver is far off from our experimental data. What parameters should I be taking into account for in this difference? Please note that the experimental data is backed by theoretical assumption. The peak wavelength difference is far off by about ~60 nm (Experiment PL=1320 nm, Simulated PL=1261 nm).

 

The material set in the simulation is quaternary alloy with which x and y compositions were specified. We have also already double-checked the material interpolation. Thickness and strain were also specified accordingly. Everything else was set as default. Carrier density was also varied but PL is still far off.

[Guilin Sun]

based on the very limited amount of information from your message, I can recommend a few things:

1. Try default strain (need to specify reference material, for example, InP, or GaAs). Adding Strain to the MQW Layers – Ansys Optics
2. Try user defined valence band offsets (VBO). These can often be found in the literature as CBO to VBO ratio. Sometime these values are not accurate enough in the default material database (which is mostly based on interpolation). Understanding band offsets in heterostructures – Ansys Optics
3. Check that the band structure and wavefunctions look correct. Do you see the expected number of subbands (e.g. 1 or 2 in the CB)?