Connecting Theory and Practice in Optoelectronics

Efficiency droop caused by carrier-induced strain shift?

srep17227-f2We still don’t fully understand the efficiency droop observed in GaN-based light-emitting diodes (LEDs). That is why I keep looking for new ideas and I just found one at [1]. The authors of this paper performed Raman measurements and observed that the GaN strain turns from compressive to tensile with increasing current density while the efficiency droops by 80% (see figure). Based on first-principle calculations, they claim that the strain reversal is caused by carrier accumulation within the InGaN quantum wells. But how can the rising density of electrons and holes expand the InGaN lattice constant in such dramatic way?

Anyway, the rising carrier density also triggers enhanced Auger recombination which many believe is the main origin of the efficiency droop. But since their simulations show only 5% droop, the authors summarily exclude Auger recombination and self-heating as cause of the measured 80% droop.  In my view, the Auger coefficient employed in these simulations was simply too small. Unfortunately, I frequently find such mistakes in the literature. Numerical results are often taken at face value without realizing how strongly they depend on underlying models and parameters.

However, even if the measured strain shift does not cause the efficiency droop,  how can it be explained? The initial GaN strain at the LED surface seems to come from the sapphire substrate. Sapphire exhibits not only a lower lattice constant than GaN , but also a lower thermal expansion coefficient. Thus, current-induced self-heating should make the GaN surface strain even more compressive, doesn’t it?

Despite my initial excitement, I find that this paper leaves too many questions open and I wonder who was checking it before publication.

[1] Zheng, J. et al., Sci. Rep. 5, 17227; doi: 10.1038/srep17227 (2015).


One response to “Efficiency droop caused by carrier-induced strain shift?

  1. Felix Nippert 2015-12-18 at 05:28

    It seems to me that the shift of the E2high is entirely consistent with the given change in junction temperature. See e.g.

Leave a Reply

Fill in your details below or click an icon to log in: Logo

You are commenting using your account. Log Out / Change )

Twitter picture

You are commenting using your Twitter account. Log Out / Change )

Facebook photo

You are commenting using your Facebook account. Log Out / Change )

Google+ photo

You are commenting using your Google+ account. Log Out / Change )

Connecting to %s