GaN has garnered significant attention due to its wide bandgap, high electron mobility, high saturation drift velocity, high breakdown electric field, and excellent thermal conductivity. Group Ⅲ nitride semiconductors, including GaN, AlN, and InN, are direct bandgap materials that demonstrate theoretically high electro-optical conversion efficiency, making them well-suited for the design and manufacture of semiconductor lasers. Since the introduction of the first GaN-based blue-violet laser diode (LD) by Nichia Ltd. in 1996, GaN-based LDs have seen remarkable advancements over the past few decades. Notably, the successful development of GaN-based blue and green LDs has surpassed the limitations of second-generation semiconductors, which are unable to produce blue and green lasers. Today, GaN-based LDs are utilized across a wide range of applications, including laser displays, laser manufacturing, visible light communication, and quantum information technology. Recently, driven by increasing market demand, GaN-based LDs have emerged as a leading focus in the research of third-generation semiconductor materials and optoelectronic devices, positioning them as a key area for industrial development.

This study reviews the current state of development of GaN-based LDs. The highest reported wall-plug efficiencies for blue and green edge-emitting lasers have reached 52.4% and 24.2%, respectively, whereas that for GaN-based vertical-cavity surface-emitting lasers has exceeded 20%. Despite these achievements, enhancing the performance and reliability of these diodes remains a major challenge in both scientific research and industry. Because LDs involve closely interlinked optical, electrical, and thermal processes, their study encompasses a range of complex physical phenomena. Effective device design must focus on confining optical fields and precisely regulating the generation, transport, and recombination of carriers. Key factors for improving device performance and extending their lifespans include the rational design of LD structures, enhancing crystal quality, reducing leakage current, and establishing reliable ohmic contacts.

This study summarizes the significant research advancements made by our group over the past five years. To tackle the primary bottlenecks hindering performance improvements in GaN-based LDs, our group has explored various aspects of edge-emitting and novel GaN-based lasers, focusing on device design, material epitaxy, and semiconductor processing.

Obtaining large-scale, uniform GaN films with low dislocation density on foreign substrates is a significant challenge. To address this, we introduced a novel hexagonal 3D serpentine mask that reduces dislocation density in both the window and coalescence regions. This mask was designed, optimized, and fabricated on a sapphire substrate. Following the metal organic chemical vapor deposition (MOCVD) process for dislocation-controlled epitaxy, we successfully produced a wafer-scale, surface-flat GaN film. Cathodoluminescence images indicated an average dislocation density of 1.7×10⁷ cm^-2 on the GaN surface. This approach effectively diminishes regions with high dislocation density, thereby enhancing the overall material quality and uniformity.

In terms of device physics, we developed an analytical model to analyze leakage current in GaN-based lasers, concentrating on the carrier transport processes in the device-scale epitaxial direction. The model accuracy was validated by fitting experimental and finite element simulation data, revealing insights into the mechanism of the electron blocking layer.

The crystal quality of the active region is also crucial for LD lifespans. We found that the crystallinity restoring (CR) layer effectively repairs surface morphology. We hypothesize that the luminescence of the quantum well can be affected by the diffusion of nitrogen vacancies generated in the high-temperature CR layer. Thus, positing the CR layer directly above the quantum well allows the GaN spacer to encapsulate the largest number of nitrogen vacancies. Utilizing the CR layer during the epitaxy process, we achieve ultra-thick multiple quantum wells.

These studies lay the groundwork and provide technical support for the development of high-performance GaN-based LDs.

As market demand for GaN-based lasers continues to increase, scientific research, technological innovation, and process improvements aimed at achieving high-power, high-efficiency, and high-reliability GaN-based lasers are becoming prevalent trends. The complex structure of GaN-based lasers, combined with the difficulties and lengthy cycles of fabrication processes, presents numerous opportunities for in-depth research at every stage. High-performance devices depend on high-quality materials, whereas large-scale industrial applications require regular reductions in material costs. Our experimental results confirm that the proposed hexagonal 3D serpentine mask technique offers unique advantages and significant potential for improving the growth quality of heteroepitaxial materials, demonstrating promising prospects for large-scale industrial applications.

Several bottleneck issues remain in advancing blue lasers toward higher power and extending green lasers to longer wavelengths. Recent theoretical advancements offer new insights and inspiration for laser structure design. In high-power edge-emitting lasers, catastrophic optical mirror damage is the primary cause of failure, which limits both the maximum output power and the device lifespan. To increase the catastrophic optical mirror damage threshold and improve long-term facet stability, investigating the thermodynamics of device degradation, understanding the microscopic mechanisms of defect generation and evolution, and improving facet protection techniques are critical. In addition, the introduction of new GaN polarization theories and topological-cavity surface-emitting lasers is expected to help address the current technological bottlenecks in semiconductor lasers from a fundamental perspective.