Intelligent optimization design of electron barrier layer for AlGaN-based deep-ultraviolet light-emitting diodes

AlGaN-based deep-ultraviolet light-emitting diodes (DUV LEDs) are widely used in sterilization, sensing,water purification, medical treatment, non-line of sight (NLOS) communication and many other fields.Especially it has been reported that the global novel coronavirus (COVID-19) can be effectively inactivated bythe DUV light with a wavelength below 280 nm (UVC) within a few seconds, which has also attracted greatattention. However, the external quantum efficiency (EQE) of UVC LED is still at a low level, generally notmore than 10%. As an important component of EQE, internal quantum efficiency (IQE) plays a crucial role inrealizing high-performance DUV-LED. In order to improve the IQE of AlGaN-based DUV-LED, this workadopts an electron blocking layer (EBL) structure based on InAlGaN/AlGaN superlattice. The results showthat the superlattice EBL structure can effectively improve the IQE compared with the traditional single-layerand double-layer EBL structure for the DUV-LED. On this basis, the optimization method based on JAYAintelligent algorithm for LED structure design is proposed in this work. Using the proposed design method, theInAlGaN/AlGaN superlattice EBL structure is further optimized to maximize the LED' s IQE. It isdemonstrated that the optimized superlattice EBL structure is beneficial to not only the suppression of electronleakage but also the improvement of hole injection, leading to the increase of carrier recombination in the activeregion. As a result, the IQE of the DUV-LED at 200 mA injection current is 41.2% higher than that of thesingle-layer EBL structure. In addition, the optimized structure reduces IQE at high current from 25% to 4%.The optimization method based on intelligent algorithm can break through the limitation of the current LEDstructure design and provide a new method to improve the efficiency of AlGaN-based DUV-LED.