Quantum Efficiency Improvement of InGaN Near Ultraviolet LED Design by Genetic Algorithm

Abstract

A near-ultraviolet (367-nm) InGaN light-emitting diode (LED) with 5.75 nm quantum well depth was designed and both internal/external quantum efficiency (IQE/EQE) values were optimized considering the effects of non-radiative recombination rates and possible fabrica-tion errors. Firstly, the IQE of the design was enhanced by a genetic algorithm code which was developed particularly for this study. Distributed Bragg Reflectors and optional ultra-thin 1nm AlN interlayer were also used to increase overall light extraction efficiency. Then, alloy and doping concentration effects on wavelength-dependent optical and structural parameters were analyzed via the CASTEP software package based on density functional theory to pre-sent a more detailed and realistic optimization. The relatively great values of 42.6% IQE and 90.2% LEE were achieved. The final structure with 1.00 mm × 1.00 mm surface area requires only 200 mW input power to operate at 3.75 V.