A streamlined algorithm for two-dimensional bandgaps and defect-state energy variations in InGaN-based micro-LEDs.

Bandgaps and defect-state energies are key electrical characteristics of semiconductor materials and devices, thereby necessitating nanoscale analysis with a heightened detection threshold. An example of such a device is an InGaN-based light-emitting diode (LED), which is used to create fine pixels in augmented-reality micro-LED glasses. This process requires an in-depth understanding of the spatial variations of the bandgap and its defect states in the implanted area, especially for small-sized pixelation requiring electroluminescence. In this study, we developed a new algorithm to achieve two-dimensional mappings of bandgaps and defect-state energies in pixelated InGaN micro-LEDs, using automated electron energy-loss spectroscopy integrated with scanning transmission electron microscopy. The algorithm replaces conventional background subtraction-based methods with a linear fitting approach, enabling enhanced accuracy and efficiency. This novel method offers several advantages, including the independent calculation of the defect energy (E_d) and bandgap energy (E_g), reduced thickness effects, and improved signal-to-noise ratio by eliminating the need for zero-loss spectrum calibration. These advancements allow us to reveal the relationship between the bandgap, defect states, microstructure, and electroluminescence of the semiconductor under ion-implantation conditions. The streamlined analysis achieves a spatial resolution of approximately 5 nm and an exceptional detection limit. Additionally, ab initio calculations indicate gallium vacancies as the predominant defects.