First-Principles Calculations of Auger Recombination in Optoelectronic Materials

Abstract

Lighting currently uses around 15% of global energy expenditure. Reducing this energy usage would be an important part of reducing the effects of global climate change. LEDs (light-emitting diodes) could be a more efficient light source than traditional incandescent or compact-fluorescent bulbs. However, they currently suffer from ‘efficiency droop’ — as the power through the device is increased, the efficiency goes down.

While the exact cause of droop is not known completely, one effect is the Auger recombination process. Auger recombination involves an electron and hole recombination which transfers energy to another carrier (rather than emitting a photon as in the desired radiative process). Auger recombination comes in many forms and can be assisted by phonons and alloys which allow Auger to occur when it would otherwise be prevented by momentum conservation. Reducing the Auger recombination process in LED materials would allow for manufacturers to use fewer LEDs in their bulbs, reducing costs and encouraging widespread LED adoption.

The challenge to reducing Auger is quantifying its impact. Experimental determination of Auger can be confounded by many factors and reported rates for Auger often vary over orders of magnitude. Theoretical determination of Auger recombination is also difficult because of the complexity of the equations involved. In this situation, computation can be an important tool to understand the underlying physics. In particular, this thesis focuses on first-principles calculations, which solve Schödinger’s equation directly rather than relying on existing experimental data. This allows these calculations to be predictive of experiment and can guide future research on more efficient materials.

While many open-source and commercial options exist for codes that will solve Schödinger’s equation using Density Functional Theory, there is not an available code to solve for recombination rates. As part of this thesis, code used to find the Auger recombination rates was expanded and improved. This was then used to study various optoelectronic materials.

The group-III nitrides are widely used for making LEDs. GaN is popular in making blue LEDs, but other materials in this family have uses. InN has a band gap in the infrared, and could be used for telecommunication purposes. We studied the Auger process in InN and found that it is dominated by the direct Auger process, as expected for its small band gap. We also found that at high carrier densities, Auger was primarily reduced by carrier screening rather than phase-space filling. On the other hand, AlGaN alloys could be used to create UV LEDs, with applications in sterilization and sensing. We studied Auger in AlGaN alloys of three compositions and modeled expanding these Auger values throughout the entire alloy spectrum. We expected to find the maximum alloy-assisted Auger at the 50/50 alloy, but found the opposite trend. This unintuitive result warrants further study.

Scintillators are another type of device that emits light when struck by radiation. Auger affects these devices by suppressing light output and making it difficult to identify what the elemental source of the original radiation was. We studied Auger in NaI and found that the phonon-assisted process dominates as expected by the large gap of the material.

This methodology and code has shed light on nonradiative carrier combination in optoelectronic materials and devices and can continue to be used in future studies.