This is mostly a question for epi scientists working on phosphide- or arsenide-based LEDs.

Epi scientists who, like me, work on LEDs and lasers based on nitride materials (GaN, AlGaN, InGaN), tend to take as a given that their structure must have an EBL in order to have a good quantum efficiency. (Even though a few publications in the past few years have claimed the contrary.)

But is that also the case in similar structures made of different III-V compound materials? Or is it rather a characteristic of the nitrides, perhaps due to their strong polarization fields?

Note that I'm speaking about multiple-quantum-well structures, not double hetero-structures (which of course have an EBL!).

Could you please cite a reference from the literature that supports your answer?

Similar questions and discussions