For more than a decade, the LED industry has sought to understand why LED efficiency drops as current increases — a phenomenon called droop. Rensselaer Polytechnic Institute (RPI) professors and students have published a paper that identifies electron leakage as the culprit, and believe that understanding the problem will lead to a solution at the component level that in turn would lower the cost of solid-state lighting (SSL) products.
The team has published their research in the Applied Physics Letters scientific journal. The explanation is quite simple. Under higher current, some of the electrons injected from the n-type semiconductor layer pass through the active quantum well region and into the p-type layer without combining with holes. It's that combination of electrons and holes in the quantum well that generates photons.
While the tone of the RPI press release is authoritative, the theory is one of several such authoritative claims to understanding droop. For example, researchers from University of California, Santa Barbara (UCSB), and the École Polytechnique in France, said back in April that the Auger effect is the root cause of droop. The Auger effect occurs when an electron, a hole, and another charge-carrying particle (such as another electron) combine to produce a high-energy electron but no light. We covered the Auger effect and initial UCSB research on the subject in a report from Semicon West two years ago.
In an article on the UCSB research, IEEE Spectrum explains the experiment in which the LED was placed in a vacuum and the energy of the electrons that escaped the active region was measured with a spectrometer. But the article also questioned the conclusion that Auger loss was the cause. Indeed, the article included the opinion from a Boston University professor that the cause in the UCSB study was also electron leakage, perhaps with an assist from the Auger effect.
The point is that multiple top-notch researchers have come up with different theories with none receiving universal acceptance in their explanation of the droop phenomenon. But much work has been done to minimize the impact of droop. For example, UCSB also revealed research last May on new crystal structures that could increase efficiency at high current.
RPI's lead researcher E. Fred Schubert, founding director of the university's National Science Foundation-funded Smart Lighting Engineering Research, discussed past and future approaches to droop. "In the past, researchers and LED manufacturers have made progress in reducing efficiency droop, but some of the progress was made without understanding what causes the droop," said Schubert. "I think now we have a better understanding of what causes the droop and this opens up specific strategies to address it."
The RPI team is now working on a new structure for LEDs. But in reality droop is not as significant of a roadblock as it was just a few years ago when it was seen as a major hurdle to lower-cost SSL products and broad LED deployment in general lighting applications. In a presentation at Strategies in Light in Feb 2012, Mark McClear, global director of applications engineering at Cree, specifically said that efficacy gains in LEDs had largely eliminated droop as a concern. Marginally better efficiency could lessen the cost of LED-based lighting, but McClear said that once component efficacy hit 100 lm/W, the droop hurdle has essentially been cleared.