Open Access
Add to BibSonomy
Abstract
Aluminum-rich p-AlGaN electron blocking layers (EBLs) are typically used for preventing overflow of electrons from the active region in AlGaN-based deep ultraviolet (DUV) laser diode (LD). However, these cannot effectively prevent electron leakage and form barrier layers, which affects the hole injection efficiency. Herein, the traditional p-AlGaN EBL in LD is replaced with an undoped BGaN EBL. The undoped BGaN EBL LD increases the effective barrier height of the conduction band to prevent the leakage of electrons and decreases the energy loss caused by the polarization induced electric field, enhancing the hole injection. The slope efficiency of the undoped BGaN EBL LD is 289% higher than that of the highly doped AlGaN EBL LD, and its threshold current is 51% lower. Therefore, the findings of this study provide insights for solving the problems of electron leakage and insufficient hole injection in high-performance and undoped EBL DUV LDs.
© 2022 Optica Publishing Group under the terms of the Optica Open Access Publishing Agreement
1. Introduction
Deep ultraviolet (DUV) laser diodes (LDs) have excellent potential applications in biological monitoring, medical applications, water purification, air sterilization in dust-free workshops, and formaldehyde treatment; they have attracted research interest in various related fields [1]. DUV-C spectroscopy (<280 nm) can be used for surface disinfection and nucleic acid destruction of microorganisms (viruses and pathogens) [2–4]. The demand for DUV-C luminescence devices has increased due to the outbreak of the COVID-19 pandemic [5]. However, there are still many challenges for LDs in the DUV spectrum. The major challenge is decreasing the threshold current density of LDs, which depends on the optical confinement factor (OCF), carrier injection efficiency, and the internal loss of an LD [6]. The carrier injection efficiency and internal loss of an LD also influence its radiation recombination efficiency, affecting its photoelectric performance. However, the effective mass of electrons is lighter due to the higher activation energy of the donor silicon, which further increases with the aluminum composition [7]. This phenomenon is more obvious in DUV LDs. The significant leakage of electrons decreases the concentration of electrons in quantum wells as well as the radiative recombination rate of carriers, thus increasing the threshold current density of LDs. Moreover, the value of p-type conductivity is lower than the value of n-type conductivity; therefore, the electrons enter the active region faster than the holes, resulting in the leakage of electrons from the active region and non-radiative recombination in the p-type region and the anode, which further deteriorates the luminous efficiency of the LD [8]. Several researchers have proposed different approaches to address these problems, such as using aluminum-rich p-AlGaN electron blocking layers (EBLs). The EBL structures of superlattice AlGaN/GaN [9], aluminum component gradient structure [10], aluminum component V-shaped structure [11], aluminum component double taper structure [12,13], and polarization doping layer [14] have been designed to suppress electronic leakage in LDs and decrease the threshold current density. Quaternary AlInGaN EBLs can decrease the polarization charge density in heterostructure interfaces, decreasing the band bending of EBLs [15]. The hole injection layer was inserted between an EBL and a multiple quantum well (MQW) to effectively alleviate the valence band bending caused by polarization [16]. However, while the EBL forms an electron barrier to decrease electron leakage, it also forms a hole barrier layer, which hinders the injection of holes. In addition, AlGaN materials rely heavily on p-type doping, which decreases the carrier concentration of the LD and the doping solubility, increases the acceptor activation energy, and leads to considerable impurity compensation, further decreasing the hole injection efficiency. Therefore, the design of appropriate EBLs to decrease the leakage of electrons and improve the injection of holes is crucial for decreasing the threshold current density of LDs.
BGaN, a boron-containing III-N alloy, is a novel wide bandgap material for optoelectronic and power devices. Researchers have successfully grown BGaN/AlGaN and BGaN/GaN superlattices [17–19]. Lachebi et al. calculated the structure and electronic properties of the zinc blend (ZB) phase of BGaN alloys by using the enhanced plane wave method based on the total potential linearization, demonstrating that the development of alloys with large amounts of boron is possible [20]. Liu and Said et al. calculated the spontaneous polarization (SP) and piezoelectric polarization (PZ) constants of the full range components of hexagonal BGaN [21,22]. A small amount of boron doping in GaN can significantly change the refractive index of the material [23]. Refractive index depends on the properties of the electronic energy band structure, and it is negatively correlated with the bandgap. In addition, the bandgap bending factor of BGaN strongly depends on the composition and lattice parameters of boron. The BGaN alloy has a relatively small lattice constant and can achieve good lattice matching with AlGaN solving the problem of mismatch between the upper and lower lattices of AlGaN due to the high aluminum composition. In addition, BN is characterized by a low refractive index and low electron mobility, making it suitable for use as an electronic barrier material.
In this study, due to the serious lattice mismatch and polarization effect between the EBL with high aluminum content and the upper and lower layers, the problem of low electron leakage and hole injection efficiency of the LD cannot be fundamentally solved. Moreover, due to the heavy dependence of AlGaN on p-type doping concentration, the problem of magnesium diffusion cannot be solved. An undoped BGaN EBL LD was proposed. The output characteristics of the BGaN EBL device are completely independent of the p-type doping concentration. The undoped BGaN EBL LD effectively decreased the electron leakage and improved the hole injection efficiency decreasing the threshold current of the LD.
2. Proposed deep ultraviolet laser diode structure and parameters
Figure 1 shows a schematic of the structure of the proposed DUV LD with different EBLs of $Al_{0.9}Ga_{0.1}N$ and $B_{0.45}Ga_{0.55}N$. The structure is grown on a sapphire substrate, followed by a 1-$\mathrm{\mu}\text {m}$ thick $Al_{0.75}Ga_{0.25}N$ n-cladding layer (n-CL) and a 110-nm thick $Al_{0.68}Ga_{0.32}N$ n-waveguide layer (n-WG). The active region comprises of two $Al_{0.58}Ga_{0.42}N$ (6 nm each) quantum wells (QWs) and three $Al_{0.68}Ga_{0.32} N$ (10 nm each) quantum barriers (QBs) emitting at 267 nm. Above the last QB, there is a 110-nm thick $Al_{0.68}Ga_{0.32}N$ p-waveguide layer (p-WG), followed by a 10-nm thick $Al_{0.9}Ga_{0.1}N$ EBL. This is followed by a 400-nm thick $Al_{0.75}Ga_{0.25}N$ p-cladding layer (p-CL) and a 10-nm thick $Al_{0.68}Ga_{0.32}N$ p-buffer layer (p-BL). This structure was labeled sample A; the structure of sample B comprised a $B_{0.45}Ga_{0.55}N$ EBL instead of $Al_{0.9}Ga_{0.1}N$ EBL. A photonic integrated circuit simulator (PIC3D) was used for simulation calculation to explore the photoelectric properties of the above two LDs. PIC3D uses the Poisson equation, drift-diffusion model, current continuity equation, and polarization charge model to calculate the electrical stimulation of the LD. The optical simulation of the LD is calculated using the vector Helmholtz wave equation and the Adachi refractive index model. The transmission matrix method and the Schrodinger-Poisson self-consistent method are used to calculate the carrier transport and distribution characteristics. In samples A and B, the same growth conditions are used; the width is 1.5 $\mathrm{\mu}\text {m}$, cavity length is 534 $\mathrm{\mu}\text {m}$, and back loss is assumed to be 2400 $m^{-1}$. The emission wavelength of the LD is approximately 267 nm, and all the simulations are conducted at room temperature $\mathrm {T}=300 \mathrm {~K}$. The SP and PZ constants of BGaN and AlGaN were derived from Refs. [21,24]; the accuracy of their calculation results had been confirmed [25]. All n-type doping is silicon doping, and the doping concentration is 1E18 $cm^{-3}$. All p-type doping is magnesium doping, and the doping concentration is 1E19 $cm^{-3}$. The built-in interface charge is calculated according to 40% of the theoretical value [26,27]. The band shift ratio of the epitaxial layer of the LD is $\Delta \mathrm {Ec} / \Delta \mathrm {Ev}=0.7 / 0.3$, and the reflectivity before and after the cavity is 85% and 75%, respectively. The energy bandgap of $Al_xGa_{1-x}N$ and BGaN is estimated using Eq. (1) and Eq. (2), respectively.
where b is the bending factor (=0.94eV), $x$ is the aluminum content, $E_g (AlN)=6.138eV$, $E_g (GaN)=3.415eV$ [28], $y$ is boron content, and $E_g (BN)=5.8eV$ [29].
Fig. 1. Schematic cross-sectional structure of DUV LD EBLs: $Al_{0.9}Ga_{0.1}N$ (sample A) and $B_{0.45}Ga_{0.55}N$ (sample B).
3. p-BGaN EBL laser diode
When the EBL material was changed from AlGaN (sample A) to BGaN (sample B), the whole band of the device was rearranged. As shown in Figs. 2(a) and (b), compared with sample A, the effective barrier height (defined as $\phi _{e}=E_{c}-E_{f e}$ [7]) of the conductive band of sample B increased from 587.5 meV to 898.2 meV due to the low refractive index and low electron mobility of BGaN materials and the special band arrangement of BGaN/AlGaN [23]. The larger values also imply that BGaN EBL devices had lower electron overflow. However, the valence band effective barrier height (defined as $\phi _{h}=E_{f h}-E_{v}$ [7]) had also increased. To further examine the hole injection efficiency, we calculated the electric field distribution in the EBL of sample A and B (Fig. 2(c)); it shows that the BGaN EBL LD had lower electric field intensity as well as energy loss of the hole during transmission through the EBL. Therefore, sample B promoted hole transmission to the active region. To verify the accuracy of the above procedure, we calculated the electron current density and hole current density of the two samples (Fig. 3); these findings are consistent with those presented in Fig. 2. Sample B had lower electron current density and higher hole current density. Thus, the BGaN EBL can decrease the electron leakage of the LD as well as decrease the energy loss of the hole during the transmission process. Therefore, the BGaN EBL shows excellent potential in improving the carrier concentration of the LD.
The calculation results of band energy and carrier current density show that BGaN EBL can improve the carrier concentration of LD; the carrier concentration of the device in the active region can more effectively confirm the results [30]. As shown in Figs. 4(a) and (b), compared with sample A, the electron concentration and hole concentration of sample B in the active region increased by approximately 36.7% and 34%, respectively. The electron concentration of sample B increased because the BGaN EBL effectively decreased the overflow of electrons, which was attributed to the significantly lower electron concentration of sample B in the p-type region (Fig. 4(a)). This is in turn attributed to the increase in the effective barrier height of the conduction band of sample B at the critical point of EBL (Figs. 2(a) and (b)), which decreased the overflow of electrons. The hole concentration in sample B increased because the BGaN EBL effectively decreased the electric field intensity of the device in the EBL region (Fig. 2(c)), which resulted in lower energy loss when the hole passes through the EBL and improved hole injection efficiency. Figure 4(c) shows the radiative recombination efficiency of samples A and B in QWs. Compared with sample A, the radiative recombination efficiency of sample B increased by approximately 31.8%, indicating that the electron-hole radiative recombination mode in sample B accounted for approximately 93.5% of the total recombination mode. This high radiative recombination rate indicates that the BGaN EBL can improve the concentration of carriers, increase the wave function overlap of electrons and holes in the QWs, and effectively decrease non-radiative recombination.
Fig. 4. Calculated (a) electron concentration, (b) hole concentration, and (c) radiative recombination rate for samples A and B.
The most important index for LD to achieve continuous wave (CW) output is lower threshold current ($I_{th}$) and operating voltage ($V_{th}$) during laser oscillation. LD threshold current is influenced by multiple parameters, but among these, the OCF ($\Gamma$) is important [31]. It affects the $I_{th}$ of the LD as well as is an excellent indicator of the optical properties of the LD. As shown in Fig. 5(a), the estimated values of $\Gamma$ of samples A and B are 19.33% and 41.33% (1.13 times higher), respectively, which shows that sample B has higher optical properties as well as lower $I_{th}$, which is consistent with the calculation results in Fig. 5(b). The $I_{th}$ of sample B is approximately 14.3 mA, which is 49% lower than that of sample A. This is attributed to the high radiation recombination rate of sample B, which decreases the internal loss of the device, as well as its high OCF value. The output power increased to 127 mW, and the slope efficiency ($SE$) increased by 2.84 times. This is because sample B had higher carrier concentration, which increased the radiative recombination rate in QWs (Fig. 4) and further improved the photoelectric properties of the LD; sample B also had lower $V_{th}$ (Fig. 5(c)). According to the photoelectric properties of the two samples, using $B_{0.45}Ga_{0.55}N$ as EBL led to improved optical properties of the LD and output power and lower threshold current.
Fig. 5. (a) Measured near-field optical distribution, (b) emitted power, and (c) voltage–current characteristics of samples A and B.
4. Undoped BGaN EBL laser diode
The weak electric activation of magnesium dopants leads to unshielded polarization charge between the EBL and the upper and lower layers, which decreases the efficiency of the LD, so it is worth paying attention to the magnesium doping concentration in the design of high-performance LD structures [32]. To evaluate the effect of different doping levels of the EBL on LD performance, we designed a series of magnesium-doped EBLs. The doping concentration of magnesium in the EBLs of samples A and B was 1E19 $cm^{-3}$. We also designed AlGaN EBL and BGaN EBL LDs with undoped EBL and doping concentration of 1E20 $cm^{-3}$. All other parameters of these structures were the same. Figures 6(a) and (b) show the radiative recombination rates of AlGaN EBL and BGaN EBL LD structures, respectively, with undoped EBLs and magnesium doping concentrations of 1E19 $cm^{-3}$ and 1E20 $cm^{-3}$, respectively. Figure 6 shows that the doping concentration considerably influences the radiative recombination rate of the AlGaN EBL device. The radiative recombination rate of the 1E19 $cm^{-3}$ device is 33% lower than that of the 1E20 $cm^{-3}$ device, and the radiative recombination rate of the undoped device is lower. For the BGaN EBL devices, the radiative recombination rates of the three doping levels differ only slightly. However, the radiative recombination rate of the undoped EBL devices in the first quantum well was higher than that of the high doped BGaN EBL devices. This is because in the undoped BGaN EBL, the magnesium diffusion problem and the influence of AlGaN polarization are avoided [33]. However, the radiative recombination efficiency of BGaN EBL devices is significantly higher than that of AlGaN EBL devices (Fig. 4). This is because the BGaN EBL device can effectively reduce the electron leakage and improve the hole injection efficiency (Fig. 3), so that the carrier concentration in the quantum well is increased (Fig. 4), thereby improving the radiative recombination efficiency and optoelectronic performance of the device (Fig. 5).
Fig. 6. Calculated radiative recombination rate of (a) AlGaN EBL LD and (b) BGaN EBL LD with different doping concentrations.
To further investigate the effect of different magnesium doping concentrations on the photoelectric properties of the devices, we calculated the optical field distribution and the output power of the LDs using the injection current curves (Fig. 7). Table 1 shows the values of $\Gamma$, $P_{max}$, $SE$, $I_{th}$ and $V_{th}$. The optical properties of the AlGaN EBL devices are considerably affected by the magnesium doping concentration, especially the values of $\Gamma$ of the undoped EBL devices are as low as 7.56%. Other electrical properties of this structure show the same trend. For the device with doping concentration of 1E19 $cm^{-3}$, $P_{max}$ is 10.7% lower than that of the 1E20 $cm^{-3}$ device, and the difference between the $SE$ and $I_{th}$ is small. The optical and electrical properties of the undoped AlGaN EBL structure are considerably lower than those of the device with doping concentration of 1E19 $cm^{-3}$. This is because aluminum-rich p-AlGaN EBL forms a barrier layer that hinders hole transport, and high magnesium doping can reduce the energy level of this barrier layer, thereby increasing the hole injection efficiency, but this in turn reduces the ability to limit electrons [34]. Therefore, the AlGaN EBL is heavily dependent on the magnesium doping concentration. The optical properties of the BGaN EBL devices are less affected by the magnesium doping concentration. The values of $\Gamma$, $P_{max}$, and $SE$ of the undoped BGaN EBL devices are higher than those of the highly doped BGaN EBL devices, and the value of $I_{th}$ is lower. This is because in the undoped BGaN EBL, the influence of magnesium doping was avoided. Because uncontrolled diffusion of magnesium dopants can degrade device performance [33]. The calculation results in Table 1 show that compared with the 1E19 $cm^{-3}$ AlGaN EBL device, the optical properties of the undoped BGaN EBL device were 113% higher, the $SE$ increased by 289%, and the $I_{th}$ decreased by 51%. This is consistent with the calculation structure presented in Fig. 6. This significant change is attributed to the use of undoped BGaN EBL LDs that can decrease the electron slope, improve the hole injection ability, thereby increasing the carrier concentration, and avoid the problems occurring due to high magnesium doping.
Fig. 7. Calculated (a) wave intensity and (b) emitted power of AlGaN EBL LDs and BGaN EBL LDs with different doping concentrations.
Table 1. Output characteristics of AlGaN EBL LDs and BGaN EBL LDs with different doping levels.
5. Conclusion
This study presents an undoped BGaN EBL to improve the performance of DUV LDs. First, a p-BGaN EBL is used instead of the traditional p-AlGaN EBL LD. The DUV LD with p-BGaN EBL can inhibit the leakage of electrons and achieve efficient hole injection. The suppression of electron leakage is attributed to the improvement of the effective barrier height of the conduction band, and the enhancement of hole injection is attributed to the proposed polarization self-shielding property of the p-BGaN EBL, which decreases the polarization induced electric field and the energy loss during hole transmission through the EBL. Therefore, the p-BGaN EBL can significantly improve the radiative recombination rate in the MQW region, thereby decreasing the threshold current of DUV LD and improving its photoelectric properties. Next, the effects of AlGaN and BGaN EBLs with various doping concentrations were systematically examined on the output performance of DUV LDs. The photoelectric properties of the undoped AlGaN EBL devices decreased considerably, and the high doping level in AlGaN EBL is crucial for improving the radiative recombination rate and photoelectric properties. While the BGaN EBL is not sensitive to doping concentration, the radiation recombination rate and photoelectric performance of the undoped BGaN EBL devices were better and the threshold current was lower. Thus, we propose a high-performance undoped BGaN EBL structure, which can decrease electron leakage, improve hole injection efficiency, and alleviate the p-doping problem in traditional AlGaN EBL, thus simplifying the epitaxial process. Therefore, the DUV LD using undoped BGaN as EBL had a high OCF of 42.21% and threshold current of 12.2 mA. The findings of this study provide insights for designing high-performance DUV LDs without the problems associated with high p-doping.
Funding
National Natural Science Foundation of China (62174148); National Key Research and Development Program of China (2016YFE0118400, 2022TFE0112000); Zhengzhou 1125 Innovation Project (ZZ2018-45); Ningbo 2025 Key Innovation Project (2019B10129).
Acknowledgments
We thank Muhammad Nawaz Sharif and Yongjie Zhou for helpful discussions.
Disclosures
The authors declare no conflicts of interest.
Data availability
Data underlying the results presented in this paper are not publicly available at this time but may be obtained from the authors upon reasonable request.
References
1. Y. Muramoto, M. Kimura, and S. Nouda, “Development and Future of Ultraviolet Light-Emitting Diodes: UV-LED Will Replace the UV Lamp,” Semicond. Sci. Technol. 29(8), 084004 (2014). [CrossRef]
2. M. Kneissl, T. Y. Seong, and J. Han, “The emergence and prospects of deep-ultraviolet light-emitting diode technologies,” Nat. Photonics 13(4), 233–244 (2019). [CrossRef]
3. H. Hirayama, N. Maeda, and S. Fujikawa, “Recent progress and future prospects of AlGaN-based high-efficiency deep-ultraviolet light-emitting diodes,” Jpn. J. Appl. Phys. 53(10), 100209 (2014). [CrossRef]
4. H. X. Jiang and J. Y. Lin, “Hexagonal boron nitride for deep ultraviolet photonic devices,” Semicond. Sci. Technol. 29(8), 084003 (2014). [CrossRef]
5. N. Mahanta, V. Saxena, and L. M. Pandey, “Performance study of a sterilization box using a combination of heat and ultraviolet light irradiation for the prevention of COVID-19,” Environ. Res. 198(6), 111309 (2021). [CrossRef]
6. Z. Zhang, M. Kushimoto, M. Horita, N. Sugiyama, and H. Amano, “Space charge profile study of AlGaN-based p-type distributed polarization doped claddings without impurity doping for UV-C laser diodes,” Appl. Phys. Lett. 117(15), 152104 (2020). [CrossRef]
7. Z. H. Zhang, J. Kou, S. Chen, H. Shao, J. Che, and C. Chu, “Increasing the hole energy by grading the alloy composition of the p-type electron blocking layer for very high-performance deep ultraviolet light-emitting diodes,” Photonics Res. 7(4), B1–B6 (2019). [CrossRef]
8. Z. H. Zhang, L. Li, and Y. Zhang, “On the electric-field reservoir for III-nitride based deep ultraviolet light-emitting diodes,” Opt. Express 25(14), 16550 (2017). [CrossRef]
9. Y. K. Kuo, F. M. Chen, B. C. Lin, J. Y. Chang, Y. H. Shih, and H. C. Kuo, “Simulation and experimental study on barrier thickness of superlattice electron blocking layer in near-ultraviolet light-emitting diodes,” IEEE J. Quantum Electron. 52(8), 1–6 (2016). [CrossRef]
10. Y. Zhang, L. Yu, K. Li, H. Pi, J. Diao, X. Wang, Y. Shen, C. Zhang, W. Hu, W. Song, and S. Li, “The improvement of deep-ultraviolet light emitting diodes with gradually decreasing Al content in AlGaN electron blocking layers,” Superlattices Microstruct. 82, 151–157 (2015). [CrossRef]
11. X. Fan, H. Sun, X. Li, H. Sun, C. Zhang, Z. Zhang, and Z. Guo, “Efficiency improvements in AlGaN-based deep ultraviolet light-emitting diodes using inverted-V-shaped graded Al composition electron blocking layer,” Superlattices Microstruct. 88, 467–473 (2015). [CrossRef]
12. M. M. Satter, Y. S. Liu, T. T. Kao, Z. Lochner, X. Li, J. H. Ryou, S. C. Shen, T. Detchprohm, R. D. Dupuis, and P. D. Yoder, “Theoretical analysis of strategies for improving p-type conductivity in wurtzite III-nitride devices for high-power opto-and microelectronic applications,” Phys. Status Solidi C 11(3-4), 828–831 (2014). [CrossRef]
13. Y. Wang and M. I. Niass, “Reduction of Electron Leakage in a Deep Ultraviolet Nitride Laser Diode with a Double-Tapered Electron Blocking Layer,” Chin. Phys. Lett. 36(5), 057301 (2019). [CrossRef]
14. W. Hu, P. Qin, W. Song, C. Zhang, R. Wang, L. Zhao, C. Xia, S. Yuan, Y. Yin, and S. Li, “Ultraviolet light-emitting diodes with polarization-doped p-type layer,” Superlattices Microstruct. 97, 353–357 (2016). [CrossRef]
15. G. Alahyarizadeh, M. Amirhoseiny, and Z. Hassan, “Effect of different EBL structures on deep violet InGaN laser diodes performance,” Opt. Laser Technol. 76, 106–112 (2016). [CrossRef]
16. L. J. Wu, S. T. Li, C. Liu, H. L. Wang, T. P. Lu, K. Zhang, and X. D. Yang, “Simulation study of blue InGaN multiple quantum well light-emitting diodes with different hole injection layers,” Chin. Phys. B 21(6), 068506 (2012). [CrossRef]
17. M. Kurimoto, T. Takano, J. Yamamoto, Y. Ishihara, M. Horie, M. Tsubamoto, and H. Kawanishi, “Growth of BGaN/AlGaN multi-quantum-well structure by metalorganic vapor phase epitaxy,” J. Cryst. Growth 221(1-4), 378–381 (2000). [CrossRef]
18. K. Ueyama, H. Mimura, and Y. Inoue, “Effect of substrate offcut angle on BGaN epitaxial growth,” Jpn. J. Appl. Phys. 55(5S), 05FD05 (2016). [CrossRef]
19. H. Srour, J. P. Salvestrini, and F. A. Ahaitou, “Solar blind metal-semiconductor-metal ultraviolet photodetectors using quasi-alloy of BGaN/GaN superlattices,” Appl. Phys. Lett. 99(22), 221101 (2011). [CrossRef]
20. A. Lachebi, H. Abid, and M. Driz M, “First-principles study of cubic BxGa1−xN alloys,” Turk. J. Phys. 33(6), 81–90 (2008). [CrossRef]
21. K. Liu, H. Sun, F. AlQatari, W. Guo, X. Liu, J. Li, C. G. Torres Castanedo, and X. Li, “Wurtzite BAlN and BGaN alloys for heterointerface polarization engineering,” Appl. Phys. Lett. 111(22), 222106 (2017). [CrossRef]
22. A. Said, M. Debbichi, and M. Said, “Theoretical study of electronic and optical properties of BN, GaN and BxGa1−xN in zinc blende and wurtzite structures,” Optik 127(20), 9212–9221 (2016). [CrossRef]
23. S. Watanabe, T. Takano, and K. Jinen, “Refractive indices of BxAl1−xN (x = 0 − 0.012) and ByGa1−yN (y = 0 − 0.023) epitaxial layers in ultraviolet region,” Phys. Status Solidi 7(7), 2691–2694 (2003). [CrossRef]
24. S. P. Wan, J. B. Xia, and K. Chang, “Effects of piezoelectricity and spontaneous polarization on electronic and optical properties of wurtzite III-V nitride quantum wells,” J. Appl. Phys. 90(12), 6210–6216 (2001). [CrossRef]
25. C. E. Dreyer, A. Janotti, and D. Van, “Correct Implementation of Polarization Constants in Wurtzite Materials and Impact on III-Nitrides,” Phys. Rev. X 6(2), 021038 (2016). [CrossRef]
26. I. Vurgaftman and J. R. Meyer, “Band parameters for nitrogen-containing semiconductors,” J. Appl. Phys. 94(6), 3675–3696 (2003). [CrossRef]
27. Z. Zhang, M. Kushimoto, and T. Sakai, “A 271.8 nm deep-ultraviolet laser diode for room temperature operation,” Appl. Phys. Express 12(12), 124003 (2019). [CrossRef]
28. N. Nepal, J. Li, and M. L. Nakarmi, “Temperature and compositional dependence of the energy band gap of AlGaN alloys,” Appl. Phys. Lett. 87(24), 242104 (2005). [CrossRef]
29. T. Malinauskas, D. Dobrovolskas, and J. Mickevicius, “Optical and structural properties of BGaN layers grown on different substrates,” J. Phys. D: Appl. Phys. 48(46), 465307 (2015). [CrossRef]
30. C. Chu, K. Tian, and J. Che, “On the origin of enhanced hole injection for AlGaN-based deep ultraviolet light-emitting diodes with AlN insertion layer in p-electron blocking layer,” Opt. Express 27(12), A620–A628 (2019). [CrossRef]
31. S. Tanaka, Y. Ogino, and K. Yamada, “Low-threshold-current ( 85 mA) of AlGaN-based UV-B laser diode with refractive-index waveguide structure,” Appl. Phys. Lett. 14(9), 094009 (2021). [CrossRef]
32. M. L. Nakarmi, N. Nepal, and J. Y. Lin, “Photoluminescence studies of impurity transitions in Mg-doped AlGaN alloys,” Appl. Phys. Lett. 94(9), 091903 (2009). [CrossRef]
33. P. P. Michalowski, S. Zlotnik, and J. Sitek, “Oxygen-induced high diffusion rate of magnesium dopant in GaN/AlGaN based UV LED heterostructures,” Phys. Chem. Chem. Phys. 20(20), 13890–13895 (2018). [CrossRef]
34. C. Y. Su, C. G. Tu, and W. H. Liu, “Enhancing the Hole-Injection Efficiency of a Light-Emitting Diode by Increasing Mg Doping in the p-AlGaN Electron-Blocking Layer,” IEEE Trans. Electron Devices 64(8), 3226–3233 (2017). [CrossRef]
