III-Nitride Light-Emitting Devices

Chip-level mass detection for micro-LED displays based on regression analysis and deep learning

Though micro-light-emitting diode (micro-LED) displays are regarded as the next-generation emerging display technology, challenges such as defects in LED's light output power and radiation patterns are critical to the commercialization success. Here we propose an electroluminescence mass detection method to examine the light output quality from the on-wafer LED arrays before they are transferred to the display substrate. The mass detection method consists of two stages. In the first stage, the luminescent image is captured by a camera by mounting an ITO (indium-tin oxide) transparent conducting glass on the LED wafer. Due to the resistance of the ITO contact pads and on-wafer n-type electrodes, we develop a calibration method based on the circuit model to predict the current flow on each LED. The light output power of each device is thus calibrated back by multi-variable regression analysis. The analysis results in an average variation as low as 6.89% for devices predicted from luminescent image capturing and actual optical power measurement. We also examine the defective or non-uniform micro-LED radiation profiles by constructing a 2-D convolutional neural network (CNN) model. The optimized model is determined among three different approaches. The CNN model can recognize 99.45% functioning LEDs, and show a precision of 96.29% for correctly predicting good devices.

Electronic and optical properties of core-shell InAlN nanorods: a comparative study via LDA, LDA-1/2, mBJ, HSE06, G0W0 and BSE methods

Currently, self-induced InAlN core-shell nanorods enjoy an advanced stage of accumulation of experimental data from their growth and characterization as well as a comprehensive understanding of their formation mechanism by the ab initio modeling based on Synthetic Growth Concept. However, their electronic and optical properties, on which most of their foreseen applications are expected to depend, have not been investigated comprehensively. GW and the Bethe-Salpeter equation (BSE) are regarded as the state-of-the-art ab initio methodologies to study these properties. However, one of the major drawbacks of these methods is the computational cost, much higher than density-functional theory (DFT). Therefore, in many applications, it is highly desirable to answer the question of how well approaches based on DFT, such as e.g. the local density approximation (LDA), LDA-1/2, the modified Becke-Johnson (mBJ) and the Heyd-Scuseria-Ernzerhof (HSE06) functionals, can be employed to calculate electronic and optical properties with reasonable accuracy. In the present paper, we address this question, investigating how effective the DFT-based methodologies LDA, LDA-1/2, mBJ and HSE06 can be used as approximate tools in studies of the electronic and optical properties of scaled down models of core-shell InAlN nanorods, thus, avoiding GW and BSE calculations.

High brightness and broad modulation bandwidth InGaN-based red micro-LEDs integrated with plasmonic gratings

We propose red micro-LEDs integrated with plasmonic gratings, which demonstrate high efficiency and broad modulation bandwidth. The Purcell factor and external quantum efficiency (EQE) for an individual device can be improved up to 5.1 and 11%, respectively, due to the strong coupling between surface plasmons and multiple quantum wells. The cross talk effect between adjacent micro-LEDs can be efficiently alleviated as well, thanks to the high-divergence far-field emission pattern. Moreover, the 3-dB modulation bandwidth of the designed red micro-LEDs is predicted to be ∼ 528 MHz. Our results can be used to design high-efficiency and high-speed micro-LEDs for the applications of advanced light display and visible light communication.

Improving Output Efficiency of InGaN-Based MQW Green Laser Diodes by Modulating Indium Content of Quantum Barriers and Using Composite Lower Waveguide Layers

Potential barriers between the waveguide layer and MQW active region may influence injection efficiency significantly, which is important in improving output characteristics of GaN-based green laser diodes (LDs). In this study, potential barriers and injection efficiency of LDs are investigated by simulation methods. It is found that different indium content in quantum barrier layers results in different potential barrier heights, leading to different recombination rates in upper and lower waveguide layers, and the injection efficiency can be modulated effectively. An eclectic choice of indium content can suppress recombination in two waveguide layers, improving the output characteristics of green LDs. Additionally, a composite lower waveguide layer structure is proposed to reduce the negative effect of potential barriers. High output power and low threshold current are achieved owing to the reduction in electron injection blockage and hole leakage effects.

Deep-Ultraviolet Micro-LEDs Exhibiting High Output Power and High Modulation Bandwidth Simultaneously

Deep-ultraviolet (DUV) solar-blind communication (SBC) shows distinct advantages of non-line-of-sight propagation and background noise negligibility over conventional visible-light communication. AlGaN-based DUV micro-light-emitting diodes (µ-LEDs) are an excellent candidate for a DUV-SBC light source due to their small size, low power consumption, and high modulation bandwidth. A long-haul DUV-SBC system requires the light source exhibiting high output power, high modulation bandwidth, and high rate, simultaneously. Such a device is rarely reported. A parallel-arrayed planar (PAP) approach is here proposed to satisfy those requirements. By reducing the dimensions of the active emission mesa to micrometer scale, DUV µ-LEDs with ultrahigh power density are created due to their homogeneous injection current and enhanced planar isotropic light emission. Interconnected PAP µ-LEDs with a diameter of 25 µm are produced. This device has an output power of 83.5 mW with a density of 405 W cm^-2 at 230 mA, a wall-plug efficiency (WPE) of 4.7% at 155 mA, and a high -3 dB modulation bandwidth of 380 MHz. The remarkable high output power and efficiency make those devices a reliable platform to develop high-modulation-bandwidth wireless communication and to meet the requirements for bio-elimination.

Review of Low-Frequency Noise Properties of High-Power White LEDs during Long-Term Aging

Low-frequency noise investigation is a highly sensitive and very informative method for characterization of white nitride-based light-emitting diodes (LEDs) as well as for the evaluation of their degradation. We present a review of quality and reliability investigations of high-power (1 W and 3 W) white light-emitting diodes during long-term aging at the maximum permissible forward current at room temperature. The research was centered on the investigation of blue InGaN and AlInGaN quantum wells (QWs) LEDs covered by a YAG:Ce³⁺ phosphor layer for white light emission. The current-voltage, light output power, and low-frequency noise characteristics were measured. A broadband silicon photodetector and two-color (blue and red) selective silicon photodetectors were used for the LED output power detection, which makes it possible to separate physical processes related to the initial blue light radiation and the phosphor luminescence. Particular attention was paid to the measurement and interpretation of the simultaneous cross-correlation coefficient between electrical and optical fluctuations. The presented method enables to determine which part of fluctuations originates in the quantum well layer of the LED. The technique using the two-color selective photodetector enables investigation of changes in the noise properties of the main blue light source and the phosphor layer during the long-term aging.