Open Access
Add to BibSonomy
Abstract
This work reports a high-performance InGaN-based red-emitting LED with a strain-release interlayer (SRI) consisting of an InGaN stress-release layer (SRL) and an AlN dislocation confinement layer (DCL) in unintentionally doped GaN (u-GaN). The SRL introduces a tensile strain which could decrease the in-plane compressive stress of the u-GaN layer, while the DCL could reduce the dislocation density and thus improve the crystal quality of the u-GaN layer. Consequently, a high-efficiency InGaN-based red-emitting LED with a peak wavelength of 651 nm and an external quantum efficiency of 6.04% is realized. In addition, the room-temperature photoluminescence (PL) mapping emission wavelength is uniform across a 4-inch wafer with a standard deviation of 3.3 nm. Therefore, the proposed SRI offers good potential for mass-producing high-performance and long-wavelength InGaN-based red-emitting LEDs.
© 2024 Optica Publishing Group under the terms of the Optica Open Access Publishing Agreement
1. Introduction
Micro-LED displays represent next-generation display technology owing to their superior characteristics, such as high brightness, low power consumption, long lifetime, and high response speed [1,2], and have recently attracted considerable attention in the fields of visible light communication (VLC), virtual reality (VR), augmented reality (AR), and wearable smart devices [3,4]. Conventionally, micro-LED displays employ InGaN-based LEDs for blue and green emissions, while AlGaInP-based LEDs are applied as the red-emitting components. However, the external quantum efficiency (EQE) of the conventional AlGaInP-based red-emitting LEDs suffers from significant degradation as the LED chip size decreases to a scale of tens-of-microns, which is mainly due to long carrier lifetimes and surface recombination [5,6]. In contrast, InGaN-based red-emitting LEDs can use sidewall chemical treatment to reduce the chip size effect and thus maintain the EQE of micro-size LEDs. In addition, tricolor micro-LED displays uniformly fabricated using InGaN materials can provide a higher level of integration, better processing compatibility, a matched light angular distribution, and reduced cost [7,8].
Although InGaN-based red-emitting LEDs are quite desirable for the application in micro-LED displays, several challenges have yet to be overcome. First, InGaN-based red-emitting LEDs require a high indium content in the InGaN quantum wells (QWs) in the active region, often achieved by lowering the growth temperature of the InGaN QWs. This severely degrades the crystal quality of the InGaN QWs and thus results in low EQE of InGaN-based red-emitting LEDs. In addition, the large lattice mismatch between the high indium content InGaN QW and the underlying unintentionally doped GaN (u-GaN) layer leads to high compressive stresses in the active region, which significantly increases the quantum-confined Stark effect (QCSE) and further decreases the EQE of InGaN-based red-emitting LEDs [9–12]. Therefore, how to increase the emission wavelength and maintain the high EQE of InGaN-based red-emitting LEDs is the bottleneck to be overcome.
Many strategies have been employed to improve the EQE and emission wavelength of the InGaN-based red-emitting LEDs. For example, Hwang et al. reported a 629-nm InGaN-based red-emitting LED with an EQE of 2.9% by inserting an AlGaN interlayer with a 90% Al content on each quantum well [13]. Iida et al. developed an InGaN-based red LED with an emission peak wavelength of 633 nm and EQE of 1.6% by applying an 8-µm thick n-doped GaN (n-GaN) layer [14]. Chen et al. produced 629 nm InGaN-based red LEDs with a high EQE of 7.4% by introducing a composite AlN buffer layer that increased the lattice constant of the u-GaN layer [15]. In our previous work, we demonstrated 634-nm InGaN-based red LEDs with an EQE of 1.3% by introducing the high-In-content InGaN decomposition layer strain relaxed template to relax the strain in the active region [16]. Recently, we successfully achieved 619-nm InGaN-based red LEDs with an EQE of 5.9% by increasing the pressure from 200 Torr to 550 Torr during the growth of InGaN QWs [17]. Although much progress has been made to improve the EQE of InGaN-based red LEDs, the emission wavelength of InGaN-based red LEDs is still not ideal. Most of the reported emission wavelengths of InGaN-based red LEDs are within 635 nm. To further enhance the color gamut and improve the display quality of the InGaN-based tricolor display, a longer emission wavelength of InGaN-based red-emitting LEDs with a high EQE is highly expected.
In this work, we demonstrated an InGaN-based red-emitting LED with a peak wavelength of 651 nm and EQE of 6.04% at a current density of 1 A/cm2 by introducing an InGaN/AlN strain release interlayer (SRI). The SRI consists of an In0.08Ga0.92N stress-release layer (SRL) and an AlN dislocation confinement layer (DCL) in the middle of the u-GaN. Here, the In0.08Ga0.92N SRL can effectively release the compressive stress of u-GaN from the sapphire substrate, while the AlN DCL can block the penetration of threading dislocations and improve the crystal quality [18,19]. Consequently, the high-performance InGaN-based red LED beads with long-wavelength red emission and high EQE were fabricated.
2. Experiment
All the samples in this work were grown on 4-inch c-plane-patterned sapphire substrates (PSS) using a VEECO K465i metal-organic chemical vapor deposition (MOCVD) system and the epitaxial structures are shown in Fig. 1. The gallium, aluminum, indium, and nitrogen precursors included trimethylgallium (TMGa), triethylgallium (TEGa), trimethylaluminum (TMAl), trimethylindium (TMIn), and ammonia (NH3), respectively. Silane (SiH4) and bis-cyclopentadienyl magnesium (CP2 Mg) were used for growing the respective n-type and p-type materials, while high-purity nitrogen (N2) and hydrogen (H2) were used as carrier gases.
Fig. 1. Schematics illustrating the epitaxial structures of the proposed samples: (a) u-GaN layer without the InGaN/AlN SRI (sample A); (b) u-GaN layer with the InGaN/AlN SRI (sample B); (c) LED epitaxial structures based on sample A (sample C) and sample B (sample D).
The InGaN/AlN SRI consists of an InGaN SRL and an AlN DCL, which grew in u-GaN. The effect of the SRI on the properties of u-GaN was evaluated by sample A without the SRI (Fig. 1(a)) and sample B with the SRI (Fig. 1(b)). Here, sample A included only a 5-µm thick u-GaN, grown at a pressure of 150 Torr and a temperature of 1050 °C. Sample B included a 3.5-µm thick u-GaN layer grown under the same conditions as those of sample A, followed by the epitaxial growth of the 90-nm thick In0.08Ga0.92N SRL at a pressure of 200 Torr and a temperature of 800 °C. Then, a 1-nm thick GaN cap was grown using the same temperature and pressure as SRL to prevent SRL decomposition. The 5-nm thick AlN DCL was grown at 65 Torr and 1100 °C. Finally, a 1.5-µm thick u-GaN was grown using a conventional three-dimensional/two-dimensional (3D/2D) GaN growth technique [20], which further improved the crystal quality of the u-GaN layer. To investigate the effect of SRI on the LED characteristics, samples C and D (Fig. 1(c)) with the complete LED structure were fabricated based on samples A and B, including a 1-µm thick n-type GaN layer, hybrid MQWs structure consisting of 2 pairs In0.15Ga0.85N (3 nm)/GaN (8 nm) blue MQWs and 3 pairs In0.33Ga0.67N (3.5 nm)/Al0.1Ga0.9N (15 nm) red MQWs. The well and barrier layers of the blue MQWs were grown at 200 Torr, 785 °C, while the well and barrier layers of the red MQWs were grown at 450 Torr, 725 °C. The p-doped GaN (p-GaN) layer was 100-nm thick, while the heavily doped p + -GaN layer was 50-nm thick.
3. Results and discussion
The defect characteristics of sample B (Fig. 1(b)) with SRI can be analyzed by the cross-sectional transmission electron microscopy (TEM) images. Figure 2(a) and Fig. 2(b) present TEM images of sample B taken along the u-GaN [1-100] crystal orientation at diffraction vectors g = [0002] and g = [11-20], respectively. We note that the In0.08Ga0.92N SRL and AlN DCL are observable between the u-GaN. According to the previous reports [21,22], screw-type and mixed-type dislocations are observable with g = [0002], while edge-type and mixed-type dislocations are observable with g = [11-20]. As shown in Fig. 2(a) and Fig. 2(b), a large number of edge-type dislocations appear with g = [11-20], and a few mixed-type dislocations are observable with both reflections g = [0002] and g = [11-20], while almost no dislocations can be observed with g = [0002], indicating the amount of pure screw-type dislocations is very small and can be neglected. It is worth noting that in the same area region above and below the SRI in Fig. 2(b), the defect density under the SRI is estimated to be about 1.56 × 108 cm-2, while the defect density above the SRI is reduced to 1.08 × 108 cm-2. This indicates that the AlN DCL can effectively reduce the dislocation density of u-GaN. The AlN DCL leads to compressive stress in the u-GaN. It has been proposed that the bending of edge-type and mixed-type dislocations occurs when the film is growing under compressive stress. The bending will enhance the interaction and annihilation of dislocations. As a result, the dislocation density in the top u-GaN layer decreases significantly [23–25]. According to the higher magnification TEM image of sample B near the SRI region shown in Fig. 2(c), we can clearly distinguish the u-GaN, InGaN SRL and AlN DCL layers.
Fig. 2. Dark-field cross-sectional TEM images of sample B viewed along the [1-100] zone axis with reflections (a) g = [0002] and (b) g = [11-20]. (c) Higher magnification TEM image of sample B near SRI region.
The residual stress of the u-GaN layers can be analyzed by Raman scattering spectra. Figure 3(a) and Fig. 3(b) show the Raman spectra of samples A and B, respectively. The blue dashed line at a wave number of 567.5 cm−1 represents the strain-free GaN E2(high) phonon peak [26]. The E2(high) phonon peak of sample A and sample B are observed at 571.7 cm−1 and 570.2 cm−1, respectively. the in-plane stress ($\sigma$) of samples A and B can be calculated according to the following equation [27]:
where $\Delta {\omega _{{E_{2({high} )}}}}$ is the difference of the E2(high) phonon peaks between measured samples and position of strain-free GaN; and ${k_2}$ = 2.56 cm-1/GPa is the biaxial strain factor for the E2(high) phonon mode [28]. According to the calculation, the compressive stress of samples A and B are 1.641 GPa and 1.055 GPa, respectively. The results indicate that InGaN/AlN SRI can effectively release the compressive stress of GaN grown on sapphire substrate.Figure 4(a) and Fig. 4(b) show the 5 µm × 5 µm atomic force microscopy (AFM) images of samples A and B, respectively. The analysis root-mean-square (RMS) roughness values for sample A and sample B are 0.329 nm and 0.348 nm, respectively, indicating that the InGaN/AlN SRI has almost no effect on the u-GaN surface morphology. It is noteworthy that in the AFM image, there are some pit-like features in sample A, while such feature rarely exists in sample B. We expect that it is mainly due to the growth of AlN DCL, which is conductive for pits coalescence during the growth of the overlying GaN layer. In our previous work, we used an InGaN decomposition template above the u-GaN to release the compressive stress of the u-GaN [16]. Although the InGaN decomposition template could effectively release the stress in the underlying u-GaN, it caused serious degradation of the surface morphology and the crystal quality. In contrast, the InGaN/AlN SRI effectively releases the stress in the underlying u-GaN while almost not affecting its surface morphology or defect density in u-GaN. Figure 4(c) shows the room-temperature photoluminescence (PL) mapping of sample D (full 4-inch wafer), excited by a 375-nm laser with the excitation power at 10 mW. The results confirmed that sample D containing the InGaN/AlN SRI has good emission uniformity obtained across the full width of the 4-inch wafer, with a PL emission wavelength standard deviation (STD) of 3.3 nm. The low PL mapping STD demonstrates the possibility of commercializing InGaN/AlN SRI for InGaN-based red-emitting LED wafers. Figure 4(d) shows the typical PL emission spectrum of sample D extracted from the PL mapping result. The PL emission spectrum of the sample D shows a single peak at around 645 nm.
Fig. 4. 5 × 5 µm2 AFM images of (a) sample A, (b) sample B. (c) Room-temperature photoluminescence (PL) mapping of sample D. (d) Room-temperature PL spectra of sample D.
The electroluminescence (EL) spectra obtained at current densities of 0.1-10 A/cm2 for packaged LED beads based on samples C and D using an integrating sphere are presented in Fig. 5(a) and Fig. 5(b), respectively. As can be seen, the peak emission wavelengths for both samples shifted toward shorter wavelengths as the current densities increased. When the current density is increased from 0.1 A/cm2 to 10 A/cm2, the peak wavelength of sample C decreased from 668 nm to 603 nm, and sample D decreased from 676 nm to 614 nm. The blueshift of sample C and sample D are 65 nm and 62 nm, respectively. Despite a longer emission wavelength, the reduced blueshift observed for sample D indicates that this LED exhibited a relatively smaller QCSE, which can be attributed to the decreased compressive stress of the u-GaN caused by the inserted InGaN/AlN SRI [29]. At the current density of 1 A/cm2, the peak wavelength of sample C and sample D were 634 nm and 651 nm, respectively. Compared with sample C, sample D had a 17 nm redshift. We believe this is because the InGaN SRL effectively reduces the in-plane compressive stress in the underlying u-GaN layer of sample D. It would increase the indium incorporation of the red MQWs, hence enhancing the emission wavelength [7,14]. The corresponding insets present images of the actual emissions of the LEDs obtained at a current density of 1 A/cm2. Compared with sample C, sample D shows a deeper red coloration in the insets, which is consistent with the EL spectra. Figure 5(c) shows the peak emission wavelengths FWHM of samples C and D as a function of the current density from 0.1 to 10 A/cm2. The FWHM of samples C and D have similar trends, but the FWHM of sample D is slightly higher than that of sample C. We believe that this is related to In fluctuations due to higher In content in the QWs of sample D.
Fig. 5. EL spectra of LED beads obtained at current densities of 0.1-10 A/cm2: (a) sample C; (b) sample D. The insets present images of LED emissions obtained at a current density of 1 A/cm2. (c) FWHMs of samples C and D at current densities of 0.1-10 A/cm2.
The absolute I - V curves for sample C and sample D are plotted in Fig. 6(a). Absolute current and absolute current density are plotted as left y-axis and right y-axis, respectively. Both absolute current and current density are plotted on a log scale. The reverse leakage current increases from a reverse voltage of about -1 V. At -5 V, the reverse leakage currents of samples C and D are about 4.8 × 10−7 A (9.6 × 10−5 A/cm2) and 5.2 × 10−10 A (1.1 × 10−7A/cm2), respectively. In this interval, the reverse leakage current of sample D is significantly lower than that of sample C. And the leakage current level is positively correlated with the defects in QWs [30,31]. This indicates that InGaN/AlN SRI can effectively reduce the dislocations in the active region. The forward voltage of sample C is lower than that of sample D. We expect that it may be because the SRI structure can reduce the V-pits in the active region of sample D, which decreases in the holes injected through V-pits and results in an increase in the forward voltage [32–34].
Fig. 6. EL characterization of samples C and D: (a) absolute current and current densities at different bias voltages; (b) EQE as a function of current density.
The EQE obtained as a function of the current density for LEDs based on samples C and D is plotted in Fig. 6(b). As can be seen, the peak EQEs of both sample C and sample D appear at the current density of 1 A/cm2 with values reaching 7.42% and 6.04%, respectively. The trends in the EQE of samples C and D differ substantially from those observed for previously reported InGaN-based red-emitting LEDs [15–17]. There are two phases of EQE rise, and the reasons for the second phase of EQE rise need to be further investigated. Compared with sample C, the EQE of sample D slightly decreased. Sample D with the InGaN/AlN SRI achieved a longer emission wavelength, which indicated an increase of indium content in InGaN red MQWs. The high indium content InGaN red MQWs will lead to indium compositional fluctuations and phase separation, which lowers the crystal quality of sample D of red MQWs and leads to a decrease in EQE. It should be noted that the EQE of sample D decreased but is still higher than 6% when the peak wavelength is over 650 nm. Therefore, the InGaN/AlN SRI can effectively increase the emitting wavelength and meanwhile prevent the EQE droop dramatically.
To further verify the conclusion, we fabricated another LED (denoted as sample E) with the same epitaxial structure as sample C by lowering the growth temperature of three pairs of red MQWs (about 10 °C). Sample E can get a similar emission wavelength as sample D. The normalized EL spectra of samples C, D, and E at the current density of 1 A/cm2 are presented in Fig. 7(a), where analysis yielded peak wavelengths of 634 nm, 651 nm, and 656 nm, and full-width-at-half-maximum (FWHM) values of 57 nm, 59 nm, and 64 nm, respectively. As can be seen, the peak EL intensity for sample D is about 76% of that for sample C, whereas the intensity for sample E is only 43% of that for sample C. The EQE values obtained for samples C, D, and E are plotted in Fig. 7(b) as a function of the peak wavelength under current densities in the range of 0.5-10 A/cm2, where EQE values obtained under current densities of 1 A/cm2 were 7.42%, 6.04%, and 3.70% of samples C, D, and E, respectively. Although the emission wavelength of sample E is similar to that of sample D, the EL intensity and EQE of sample E drop significantly compared with sample D. This result further confirms the effect of the InGaN/AlN SRI on the increase of the emitting wavelength and the prevention of the EQE droop dramatically of the InGaN-based red-emitting LEDs.
Fig. 7. Performance characterization of LEDs based on samples C, D, and E: (a) EL spectra at a current density of 1 A/cm2; (b) EQE versus peak wavelength under current densities in the range 0.5-10 A/cm2.
The Peak EQE or wall-plug efficiency (WPE) values and corresponding peak wavelengths for state-of-the-art InGaN-based red-emitting LEDs are plotted in Fig. 8 [5,6,15,35–37]. The current density at the occurrence of peak EQE (WPE) is labeled in the figure. As can be seen, by combining the EQE value with peak wavelength comparison, the samples in this work are among the best reports of InGaN-based red-emitting LED. Sample D with SRI achieved an EQE of over 6% and emission wavelengths greater than 650 nm, indicating that the proposed InGaN/AlN SRI is an effective method to increase the emission wavelength and maintain the high EQE of the InGaN-based red-emitting LEDs.
Fig. 8. Peak EQE or WPE values and corresponding peak wavelengths for state-of-the-art InGaN-based red-emitting LEDs.
4. Conclusion
In summary, the present work has demonstrated the development of an InGaN-based red-emitting LED with a peak wavelength of 651 nm and an EQE of greater than 6% by inserting an InGaN/AlN SRI in the middle of the u-GaN layer, which effectively reduced the u-GaN in-plane compressive stress. Moreover, the PL emission wavelength of the LED was demonstrated to be uniform across a 4-inch wafer with an STD of 3.3 nm. The combined high EQE and the peak wavelength of the proposed structure confirm that the proposed InGaN/AlN SRI is a promising approach contributing toward the commercialization of high-efficiency InGaN-based red-emitting LEDs.
Funding
National Natural Science Foundation of China (62204073); Natural Science Foundation of Anhui Province (2208085QF210); National Key Research and Development Program of China (2022YFB3604701); Fundamental Research Funds for the Central Universities; Pangna Micro Semiconductor Technology Co. Ltd.
Disclosures
The authors declare no conflicts of interest.
Data availability
The data that support the findings of this study are available from the corresponding author upon reasonable request.
References
1. X. Zhao, Bin Tang, Liyan Gong, et al., “Rational construction of staggered InGaN quantum wells for efficient yellow light-emitting diodes,” Appl. Phys. Lett. 118(18), 182102 (2021). [CrossRef]
2. P. Li, Hongjian Li, Haojun Zhang, et al., “Size-independent peak external quantum efficiency (>2%) of InGaN red micro-light-emitting diodes with an emission wavelength over 600 nm,” Appl. Phys. Lett. 119(8), 081102 (2021). [CrossRef]
3. X. Zhao, Ke Sun, Siyuan Cui, et al., “Recent Progress in Long-Wavelength InGaN Light-Emitting Diodes from the Perspective of Epitaxial Structure,” Adv. Photonics Res. 4(9), 2300061 (2023). [CrossRef]
4. T.-Y. Lee, Li-Yin Chen, Yu-Yun Lo, et al., “Technology and Applications of Micro-LEDs: Their Characteristics, Fabrication, Advancement, and Challenges,” ACS Photonics 9(9), 2905–2930 (2022). [CrossRef]
5. S. Zhang, Jianli Zhang, Jiangdong Gao, et al., “Efficient emission of InGaN-based light-emitting diodes: toward orange and red,” Photon. Res. 8(11), 1671–1675 (2020). [CrossRef]
6. P. Li, Hongjian Li, Yifan Yao, et al., “Significant Quantum Efficiency Enhancement of InGaN Red Micro-Light-Emitting Diodes with a Peak External Quantum Efficiency of up to 6%,” ACS Photonics 10(6), 1899–1905 (2023). [CrossRef]
7. D. Iida and K. Ohkawa, “Recent progress in red light-emitting diodes by III-nitride materials,” Semicond. Sci. Technol. 37(1), 013001 (2022). [CrossRef]
8. Z. Zhuang, Daisuke Iida, Kazuhiro Ohkawa, et al., “InGaN-based red light-emitting diodes: from traditional to micro-LEDs,” Jpn. J. Appl. Phys. 61(SA), SA0809 (2022). [CrossRef]
9. P. Chan, Steven P. DenBaars, Shuji Nakamura, et al., “Growth of highly relaxed InGaN pseudo-substrates over full 2-in. wafers,” Appl. Phys. Lett. 119(13), 131106 (2021). [CrossRef]
10. P. Kirilenko, Zhe Zhuang, Daisuke Iida, et al., “Investigation of a Separated Short-Wavelength Peak in InGaN Red Light-Emitting Diodes,” Crystals 11(9), 1123 (2021). [CrossRef]
11. Z. Liu, Mingtao Nong, Yi Lu, et al., “Effect of the AlN strain compensation layer on InGaN quantum well red-light-emitting diodes beyond epitaxy,” Opt. Lett. 47(23), 6229–6232 (2022). [CrossRef]
12. D. Iida, Kazumasa Niwa, Satoshi Kamiyama, et al., “Demonstration of InGaN-based orange LEDs with hybrid multiple-quantum-wells structure,” Appl. Phys. Express 9(11), 111003 (2016). [CrossRef]
13. J.-I. Hwang, Rei Hashimoto, Shinji Saito, et al., “Development of InGaN-based red LED grown on (0001) polar surface,” Appl. Phys. Express 7(7), 071003 (2014). [CrossRef]
14. D. Iida, Zhe Zhuang, Pavel Kirilenko, et al., “633-nm InGaN-based red LEDs grown on thick underlying GaN layers with reduced in-plane residual stress,” Appl. Phys. Lett. 116(16), 162101 (2020). [CrossRef]
15. Z. Chen, Bowen Sheng, Fang Liu, et al., “High-Efficiency InGaN Red Mini-LEDs on Sapphire Toward Full-Color Nitride Displays: Effect of Strain Modulation,” Adv. Funct. Mater. 33(26), 2300042 (2023). [CrossRef]
16. K. Xing, Zhihu Xia, Guangxia Xie, et al., “Achieving InGaN-based red light-emitting diodes by increasing the growth pressure of quantum wells,” IEEE Photon. Technol. Lett. 35(24), 1439–1442 (2023). [CrossRef]
17. J. Hu, Kun Xing, Zhihu Xia, et al., “Wafer-scale emission uniformity of InGaN-based red light-emitting diodes on an in situ InGaN decomposition template,” Appl. Phys. Lett. 123(11), 111107 (2023). [CrossRef]
18. K. Anazawa, Sodabanlu Hassanet, Katsushi Fujii, et al., “Growth of strain-compensated InGaN/AlN multiple quantum wells on GaN by MOVPE,” J. Cryst. Growth 370, 82–86 (2013). [CrossRef]
19. D. Iida, Shen Lu, Sota Hirahara, et al., “Enhanced light output power of InGaN-based amber LEDs by strain-compensating AlN/AlGaN barriers,” J. Cryst. Growth 448, 105–108 (2016). [CrossRef]
20. S. Kim, Jeongtak Oh, Joongseo Kang, et al., “Two-step growth of high quality GaN using V/III ratio variation in the initial growth stage,” J. Cryst. Growth 262(1-4), 7–13 (2004). [CrossRef]
21. L. Lu, Z. Y. Gao, B. Shen, et al., “Microstructure and origin of dislocation etch pits in GaN epilayers grown by metal organic chemical vapor deposition,” J. Appl. Phys. 104(12), 123525 (2008). [CrossRef]
22. K. Xing, Jiangtao Wang, Liancheng Wang, et al., “Epitaxial GaN films with ultralow threading dislocation densities grown on an SiO2-masked patterned sapphire substrate,” Appl. Phys. Express 12(10), 105501 (2019). [CrossRef]
23. W. Hui, Liang Hu, Wang Yong, et al., “Effects of AlGaN/AlN Stacked Interlayers on GaN Growth on Si (111),” Chinese Phys. Lett. 27(3), 038103 (2010). [CrossRef]
24. X. Weng, S. Raghavan, J.D. Acord, et al., “Evolution of threading dislocations in MOCVD-grown GaN films on (111) Si substrates,” J. Cryst. Growth 300(1), 217–222 (2007). [CrossRef]
25. D. M. Follstaedt, D. M. Follstaedt, S. R. Lee, et al., “Relaxation of compressively-strained AlGaN by inclined threading dislocations,” Appl. Phys. Lett. 87(12), 121112 (2005). [CrossRef]
26. S. Tripathy, S. J. Chua, P. Chen, et al., “Micro-Raman investigation of strain in GaN and Al x Ga 1− x N/GaN heterostructures grown on Si (111),” J. Appl. Phys. 92(7), 3503–3510 (2002). [CrossRef]
27. K. Xing, Guangxia Xie, Xueying Cheng, et al., “Non-polar a-plane AlN epitaxial films on r-plane sapphire with greatly reduced defect densities obtained by high-temperature annealing,” J. Cryst. Growth 597, 126855 (2022). [CrossRef]
28. J. Ryu, Y. S. Katharria, H. Y. Kim, et al., “Stress-relaxed growth of n-GaN epilayers,” Appl. Phys. Lett. 100(18), 181904 (2012). [CrossRef]
29. X. Hou, Tao Yang, Shao-Sheng Fan, et al., “Improvement of optical properties of InGaN-based red multiple quantum wells,” Opt. Express 31(11), 18567–18575 (2023). [CrossRef]
30. C. Huang, Yu-Ming Huang, Yun-Han Chang, et al., “High-bandwidth green semipolar (20–21) InGaN/GaN micro light-emitting diodes for visible light communication,” ACS Photonics 7(8), 2228–2235 (2020). [CrossRef]
31. P. Shubhra, “Demonstration of ultra-small (< 10 µm) 632 nm red InGaN micro-LEDs with useful on-wafer external quantum efficiency (> 0.2%) for mini-displays,” Appl. Phys. Express 14(1), 011004 (2021). [CrossRef]
32. F. Jiang, Jianli Zhang, Longquan Xu, et al., “Efficient InGaN-based yellow-light-emitting diodes,” Photon. Res. 7(2), 144–148 (2019). [CrossRef]
33. Q. Lv, Junlin Liu, Chunlan Mo, et al., “Realization of Highly Efficient InGaN Green LEDs with Sandwich-like Multiple Quantum Well Structure: Role of Enhanced Interwell Carrier Transport,” ACS Photonics 6(1), 130–138 (2019). [CrossRef]
34. Z. Zhuang, Daisuke Iida, Martin Velazquez-Rizo, et al., “606-nm InGaN Amber Micro-Light-Emitting Diodes With an On-Wafer External Quantum Efficiency of 0.56%,” IEEE Electron Device Lett. 42(7), 1029–1032 (2021). [CrossRef]
35. D. Iida, Pavel Kirilenko, Martin Velazquez-Rizo, et al., “Demonstration of 621-nm-wavelength InGaN-based single-quantum-well LEDs with an external quantum efficiency of 4.3% at 10.1 A/cm2,” AIP Adv. 12(6), 065125 (2022). [CrossRef]
36. P. Ayush, Yixin Xiao, Maddaka Reddeppa, et al., “A red-emitting micrometer scale LED with external quantum efficiency> 8%,” Appl. Phys. Lett. 122(15), 1103 (2023). [CrossRef]
37. H. Ray-Hua, Chun-Xin Ye, Po-Wei Chen, et al., “Study on the effect of size on InGaN red micro-LEDs,” Sci. Rep. 12(1), 1324 (2022). [CrossRef]
