Open Access
Add to BibSonomy
Abstract
Ultraviolet light emitting diodes (UV-LEDs) face the challenges including insufficient hole injection and severe electron leakage. Quantum dots (QDs) have been proven to provide three-dimensionally localized states for carriers, thereby enhancing carrier confinement. Therefore, UV-LEDs employing InGaN QDs are designed and studied in this paper. The APSYs software is used to simulate UV-LEDs. Simulation results indicate that the QDs effectively improve the electron and hole concentration in the active region. However, UV-LEDs with QDs experience efficiency droop due to serious electron leakage. What’s more, the lattice mismatch between last quantum barrier (LQB) and electron blocking layer (EBL) leads to the polarization field, which induces the downward band bending at the LQB/EBL interface and reduces effective barrier height of EBL for electrons. The AlInGaN/AlInGaN lattice matched superlattice (LMSL) EBL is designed to suppress electron leakage while mitigating lattice mismatch between LQB and EBL. The results indicate that the utilization of QDs and LMSL EBL contributes to increasing the electron and hole concentration in the active region, reducing electron leakage, enhancing radiative recombination rate, and reducing turn-on voltage. The efficiency droop caused by electron leakage is mitigated. When the injection current is 120 mA, the external quantum efficiency is increased to 9.3% and the output power is increased to 38.3 mW. This paper provides a valuable reference for addressing the challenges of insufficient hole injection and severe electron leakage.
© 2024 Optica Publishing Group under the terms of the Optica Open Access Publishing Agreement
1. Introduction
III-V semiconductor-based ultraviolet light emitting diodes (UV-LEDs) are considered better candidates for replacing traditional ultraviolet mercury lamps due to their advantages, such as long lifetime, low power consumption, and environmental friendliness [1,2]. UV-LEDs in the UV-A spectral range (320 nm-400 nm) have extensive applications, including water purification, UV curing, and gas sensing [3–5], because UV-A radiation can damage the genetic material of microorganisms, trigger photosensitive agents used for curing, and interact with specific gas molecules [6–8]. However, due to the weak confinement capability of multiple quantum wells (MQWs), the low p-doping efficiency of AlGaN with high Al composition, and the quantum-confined Stark effect (QSCE) induced by the polarization field, the development of UV-LEDs still faces significant challenges, including serious electron leakage, poor hole injection, and low radiative recombination rate [9,10]. Quantum dots (QDs) were discovered and synthesized by Ekimov, Brus, and Bawendi, leading to their recognition with the 2023 Nobel Prize in Chemistry [11–13]. QDs are the zero-dimensional material which have atom-like energy levels and island-like shapes [14]. The QDs can be used to improve the carrier injection efficiency because they provide three-dimensionally localized states for carriers [15]. Narukawa et al. and Seoim et al. demonstrated that partial light emission from InGaN quantum well originates from dot-like regions [16,17]. Andreev et al. demonstrated that InGaN QDs provides a pathway to achieve emission of InGaN/AlGaN light emitting devices in the near-UV band [18]. Park et al. prepared InGaN QDs to enhance the radiative recombination rate and the output power of InGaN/GaN UV-LEDs [19]. The above results indicate that InGaN QDs have the potential to enhance the performance of UV-LEDs by improving the carrier injection in the active region. Therefore, InGaN/AlGaN UV-LEDs with InGaN QDs are designed and studied in this paper.
In addition, the performance enhancement of QDs UV-LEDs is still limited by the electron leakage due to the high mobility of electrons and the polarization field resulting from lattice mismatch [20]. The leaked electrons have non-radiative recombination with holes in the p-type region, which reduces the hole injection efficiency and aggravates the efficiency droop [21]. To deal with the serious electron leakage, Xie et al. designed the AlGaN/AlGaN superlattice electron blocking layer (EBL), Yan et al. proposed the special AlGaN graded superlattice EBL, and Modal et al. introduced the step-graded superlattice EBL [22–24]. Nevertheless, the lattice constant of EBL is less than that of the last quantum barrier (LQB), resulting in a considerable lattice mismatch between LQB and EBL. The polarization field, arising from lattice mismatch, induces downward band bending at the LQB/EBL interface and reduces the effective barrier height of EBL for electrons [25]. Fortunately, Shatalov et al. demonstrated that the quaternary AlInGaN layers can achieve lattice matching with either AlGaN or AlInGaN through the proper selection of alloy compositions, and then they prepared the deep UV-LEDs with AlInGaN/AlInGaN MQWs [26]. Kim et al. prepared the lattice-matched AlInGaN EBL to enhance the output power of GaN-based LED [27]. Ahmad et al. demonstrated the AlInGaN can be used to fabricate high-efficiency UV LEDs [28]. Du et al. prepared the AlInGaN/AlGaN EBL to alleviate the band-bending effect and optimize the performance of InGaN/AlGaN UV-LEDs [29]. The above studies indicate that employing quaternary AlInGaN EBLs provides enhanced electron confinement while minimizing lattice mismatch.
Motivated by the above research, InGaN QDs and the lattice matched superlattice (LMSL) EBL are utilized in InGaN/AlGaN UV-LEDs to improve carrier injection, suppress electron leakage, and alleviate the polarization effect resulting from lattice mismatch. APSYs software is used to simulate the conventional UV-LED, the UV-LED with QDs, the UV-LED with QDs and superlattice EBL, and the UV-LED with QDs and AlInGaN/AlInGaN LMSL EBL. The results show that QDs effectively improve the electron and hole concentration in the active region. The LMSL EBL significantly enhances electron confinement and improves hole injection efficiency. The electron leakage is reduced and the band-bending induced by the serious polarization in the LQB/EBL interface is alleviated. Due to the enhanced carrier injection efficiency and improved wavefunction overlap in the MQWs, the radiative recombination rate is increased. The turn-on voltage is reduced due to the diminished polarization electric field and improved hole injection. The efficiency droop caused by electron leakage is mitigated. When the injection current is 120 mA, the external quantum efficiency is increased to 9.3% and the output power is increased to 38.3 mW. It is concluded that QDs and the LMSL EBL play a crucial role in increasing the radiative recombination rate, external quantum efficiency, and output power of InGaN/AlGaN UV-LEDs.
2. Structures and simulation parameters
Figure 1 shows the epitaxial layers of the InGaN/AlGaN UV-LED with an emission wavelength around 362 nm, which is used as reference structure [30,31]. The UV-LED is grown on the GaN/sapphire substrate. The n-type region is composed of a 4.5-µm-thick GaN (n-doping = 5${\times} $1018 cm-3). The active region consists of five 3-nm-thick In0.01Ga0.99N quantum wells and six 12-nm-thick Al0.08Ga0.92N quantum barriers. The p-type region is composed of a 20-nm-thick Al0.25Ga0.75N EBL (p-doping = 1${\times} $1019 cm-3), a 150-nm-thick GaN (p-doping = 1${\times} $1019 cm-3), and a 25-nm-thick p + -GaN (p-doping = 1${\times} $1020 cm-3). A 150-nm-thick GaN is used to provide holes within the quantum wells, and a 25-nm-thick p + -GaN is employed to establish ohmic contact with the electrode [32]. The reference structure is denoted as LED1.
The conventional models suggested for QWs LEDs are not suitable for LEDs with QDs, because they don’t consider the factors related to the lateral confinement of electrons and holes in the QDs [33]. To explore the effect of QDs on the transport of carriers and the performance of UV-LEDs, the dot-in-well model is used in the active region. The dot-in-well model is described by a cylindrical coordinate, which is effective approximation for the QDs in the InGaN-based LEDs proposed by Krestnikov [34]. Based on the dot-in-well model proposed by Krestnikov and referring to the QDs model established by Xia in the APSYs [35], we initially constructed a model for QDs and performed calculations. Subsequently, we constructed the LED structure, incorporating the QDs model into the active region of the LED structure. Finally, we integrated the computational results from the QDs model established in the first step into the active region of the LED structure for simulation. In the dot-in-well model, Krestnikov et al. indicates that the diameter of InGaN quantum dots falls within the range of 3-5 nm; Lai's calculations, along with Nistor's observational results, suggest that the diameter of InGaN QDs ranges from 2 nm to 5 nm, with a height of 2 nm [36,37]. Xia et al. utilized InGaN QDs with diameters of 3.6 nm and 5 nm in their theoretical calculations. Therefore, to reflect the three-dimensional confinement on the carriers, we employ the QDs with the height of 2 nm, the density of 2.8${\times} $1010 cm-2, and the diameters of 3.5 nm and 5 nm. Figures 2(a) and (b) depict the structures of QDs with diameters of 3.5 nm and 5 nm embedded in the quantum wells, respectively. The UV-LED with QDs with diameters of 3.5 nm and 5 nm is denoted as LED2.
The conventional EBL is replaced by superlattice EBL to suppress the electron leakage, which consists of five pairs of Al0.22Ga0.78N/Al0.28Ga0.72N. The UV-LED with QDs and superlattice EBL is denoted as LED3. LMSL EBL is proposed to reduce the lattice mismatch between EBL and LQB. The lattice constant (a) of AlInGaN is calculated by the following equation,
Table 1. The lattice constants of Al0.22InxGa0.78-xN and Al0.28InyGa0.72-yN with different In compositions
The APSYs software is used to simulate the performance of UV-LEDs. In the simulation, device geometry is fabricated into a rectangular shape of 300${\times} $300 µm2. The band offset of the materials is set to 0.7/0.3. The Shockley-Read-Hall (SRH) lifetime, Auger coefficient, and background loss are set to 100 ns, 5${\times} $10−30 cm6/s, and 500 m-1, respectively. The temperature is set to 300 K. The polarization charge caused by spontaneous and piezoelectric polarization effect is calculated by the method proposed by Fiorentini [39]. And the polarization-induced charge density is assumed to be 50% of the theoretical value considering the screening effect of defects [40]. The other simulation parameters can be found in the paper published by Vurgaftman [41].
3. Results and discussion
The carrier concentration in the active region and the p-type region has a significant effect on the performance of devices. Therefore, the carrier distribution of the four LEDs is analyzed in this paper. Figures 3(a)-(d) show the electron concentration in the active region, the hole concentration in the active region, the electron current density in the p-type region, and the hole current density in the p-type region, respectively. The active region is the area where radiative recombination occurs. The increase of carrier concentration in the active region has a positive impact on the radiative recombination rate of the UV-LEDs. The electron current density in the p-type region represents electron leakage. The leaked electrons have non-radiative recombination with holes, which is the critical factor leading to efficiency droop. The hole current density in the p-type region represents hole injection. The insufficient hole injection caused by low p-type doping efficiency and serious electron leakage is the significant challenges for the optimization of output power [42]. As noted above, the purpose of research is to achieve higher carrier concentration in the active region, lower electron current density in the p-type region, and higher hole current density in the p-type region. Figures 3(a) and (b) indicate that, compared to LED1, the electron concentration and the hole concentration in the active region of LED2 have been significantly improved. The phenomenon arises because the quantum wells can only confine carriers from one direction, and its restrictive effect is relatively weak, especially for electrons with high mobility. As a result, UV-LEDs with QWs face insufficient carrier injection. In contrast, the QDs with small sizes exhibit a three-dimensional confinement effect on carriers, trapping more electrons and holes within the active region [43]. Therefore, LED2 shows a higher carrier concentration within the active region compared to LED1. Four figures show that compared to LED1, LED2, and LED3, LED4 has the lowest electron leakage and the highest carrier injection. It indicates that LMSL EBL enhances electron confinement, leading to the increased electron injection and the reduced electron leakage. The reduction in electron leakage results in more holes avoiding non-radiative recombination, thus increasing hole injection. Therefore, LED4 has the highest hole injection current. The previous analysis and research results indicate that the QDs and LMSL EBL offer promising solutions to the severe electron leakage and the insufficient hole injection efficiency.
Fig. 3. (a) The electron concentration in the MQWs, (b) the hole concentration in the active region, (c) the electron current density in the p-type region, and (d) the hole current density in the p-type region.
To further elucidate the rationale for the carrier distribution, an investigation into the energy band of LEDs is undertaken. The effective barrier height is defined as the potential difference between the band edge and its corresponding Quasi-Fermi level, which plays a significant role in influencing the migration of carriers. A higher effective barrier height in the conduction band restricts the movement of electrons, whereas a lower effective barrier height in the valence band facilitates the hole injection [44]. Figures 4(a)-(d) show the diagrams of energy band and Quasi-Fermi level of four LEDs. The effective barrier heights for the electrons in the conduction band of LED1, LED2, LED3, and LED4 are 225 meV, 225 meV, 247 meV, and 293 meV, respectively. The effective barrier heights for the holes in the valence band of four LEDs are 228 meV, 228 meV, 269 meV, and 223 meV, respectively. In accordance with the data presented in Figs. 4(a) and (b), it is evident that conventional EBLs exhibit relatively weak electron confinement capabilities and introduce certain obstacles to hole injection. Hence, it is imperative to optimize the EBLs to enhance electron confinement and improve hole injection. The superlattice EBL is employed as a replacement for the conventional EBLs. A comparison of the data in Figs. 4(b) and (c) reveals that the superlattice EBL can elevate the effective barrier height for electrons in the conduction band. However, it enhances the effective barrier height for holes in the valence band, which does not facilitate hole injection. What’s more, significant lattice mismatch exists between LQB and EBL. The polarization field resulting from lattice mismatch induces the downward band bending at the LQB/EBL interface and the reduction of effective barrier height for electrons in the conduction band. The downward band bending at the LQB/EBL interface forms traps that capture electrons. The captured electrons undergo non-radiative recombination with holes injected into the active region, which weakens hole injection efficiency. And the reduction in effective barrier height for electrons exacerbates electron leakage. Considering the reasons mentioned above, the LMSL EBL is proposed to replace the superlattice EBL to mitigate the effects of lattice mismatch. Figures 4(e) and (f) depict the electric field between LQB and EBL of LED2, LED3 and LED4 and the diagram of band bending at the LQB/EBL interface of three LEDs. In Fig. 4(f), we shifted the energy bands of LED2 and LED3 to align them with LED4 for the convenience of observing the effects of traditional EBL, SL EBL, and LMSL EBL on the energy bands. Figure 4(e) shows that the electric field between LQB and EBL of LED4 is effectively reduced. The phenomenon indicates that the polarization field resulting from lattice mismatch has been alleviated by the use of LMSL EBL. Therefore, as depicted in Fig. 4(f), the band bending at the LQB/EBL interface of LED4 is alleviated. Compared to LED1, LED2, and LED3, the effective barrier height for electrons of LED4 is increased by 68 meV, 68 meV, and 46 meV, respectively; the effective barrier height for holes is reduced by 5 meV, 5 meV, and 46 meV, respectively. The above results show a decrease in the band bending at the LQB/EBL interface, an increase in the effective barrier height for electrons, and a decrease in the effective barrier height for holes of LED4. It is concluded that the LMSL EBL can effectively alleviate lattice mismatch, enhance electron confinement, and facilitate hole injection.
Fig. 4. The diagrams of energy band and Quasi-Fermi level of (a) LED1, (b) LED2, (c) LED3, and (d) LED4; (e) the electric field between LQB and EBL of LED2, LED3, and LED4 and (f) the diagram of band bending at the LQB/EBL interface of three LEDs.
The recombination processes for electrons and holes include Shockley-Read-Hall (SRH) recombination, radiative recombination, and Auger recombination, which can be calculated by the following equation [45],
where, A is the SRH parameter, B is the radiative recombination coefficient, C is the Auger recombination coefficient, n is the electron concentration, and p is the hole concentration. Non-radiative recombination releases energy in the form of phonons or thermal energy, while radiative recombination releases energy in the form of photons. Hence, enhancing the radiative recombination rate is of paramount importance in optimizing the optical performance of UV-LEDs. Figures 5(a) and (b) depict the radiative recombination rate and electroluminescence spectra of four LEDs. The results demonstrate that, in comparison to LED1, LED2 exhibits an improved radiative recombination rate and spontaneous emission rate. Equation (2) indicates that an increase in carrier concentration has a positive impact on the improvement of radiative recombination rate. What’s more, for UV-LEDs with QWs, the presence of a polarization electric field leads to the separation of electron and hole wave functions, negatively impacting the radiative recombination rate. QDs with small sizes can confine carriers in three dimensions. It has been reported that the isotropic confinement in QDs would mitigate the effect of the piezoelectric field compared to the case of QWs, which increases the overlap of electron and hole wave functions [46]. Therefore, the enhancement in radiative recombination rate of LED2 is attributed to the increased carrier concentration within the active region and the enhanced overlap of electron and hole wavefunctions resulting from the use of QDs [47]. LMSL EBL enhances electron confinement capability and improves hole injection efficiency. Consequently, the radiative recombination rate of LED4 experiences a significant enhancement. It is concluded that QDs and LMSL EBL effectively optimize the radiative recombination rate of UV-LEDs.Fig. 5. (a) The radiative recombination rate of four LEDs and (b) electroluminescence spectra of four LEDs.
The I-V curves of the four LEDs are depicted in Fig. 6. The results indicate that changes in the EBLs significantly affect the I-V properties of the LEDs. Compared to LED1 and LED2, LED3 exhibits a notable increase in turn-on voltage, while LED4 shows a distinct decrease in turn-on voltage. The preceding analysis indicates that the superlattice EBL in LED3 introduces a significant lattice mismatch with the LQB, resulting in a strong polarization field. To overcome this polarization field, a higher bias voltage is required, resulting in a noticeable increase in turn-on voltage. The LMSL EBL in LED4 reduces the lattice mismatch between EBL and LQB, consequently lowering the polarization field. Therefore, LED4 exhibits the lowest turn-on voltage. Furthermore, the use of LMSL EBL diminishes the hindrance of the EBL to hole injection, which contributes to the reduction in turn-on voltage [48,49]. The reduction in turn-on voltage has a positive impact on the improvement of external quantum efficiency [50,51]. The results indicate that LMSL EBL has the capability to reduce the turn-on voltage of UV-LEDs and optimize their I-V properties.
The conversion efficiency of UV-LEDs can be reflected by external quantum efficiency. External quantum efficiency can be calculated by the Eq. (3) [52],
Output power is also a critical performance of UV-LEDs. Figure 7(b) depicts the P-I curves of four LEDs. When the injection current is 120 mA, the output powers of LED1, LED2, LED3, and LED4 are 18.8 mW, 23.6 mW, 33.1 mW, and 38.3 mW, respectively. Compared to LED1, LED2, and LED3, the output power of LED4 is increased by 103.7%, 62.3%, and 15.7%, respectively. Due to the reduced electron leakage and the enhanced radiative recombination rate, more photons are produced at the same injection current. The three-dimensional confinement effect provided by QDs prevents carriers from undergoing non-radiative recombination at dislocations [55]. More electrical energy is converted into light energy, resulting in the improved output power. The increase in output power directly correlates with the overall performance of UV-LEDs. Higher output power means the devices can provide more intense and effective ultraviolet radiation, which is crucial for numerous applications. Powerful UV-LEDs also reduce the need for hazardous materials, such as mercury, which is commonly used in traditional UV lamps. The results demonstrate that the QDs and LMSL EBL enhance the output power of UV-LEDs, which expands their potential applications and provides ecological advantages.
4. Conclusions
Though the remarkable progress has been achieved in UV-LEDs, they still face the challenges such as insufficient hole injection and severe electron leakage. QDs have been proven to play a crucial role in the luminescence of InGaN-based LEDs. Therefore, UV-LEDs employing InGaN QDs are designed and studied in this paper. The APSYs software is used to simulate UV-LEDs. Simulation results indicate that QDs effectively enhance the concentration of electrons and holes in the active region. Compared to QWs, QDs can mitigate the impact of the polarization electric field on radiative recombination and improve external quantum efficiency. As a result, UV-LEDs with QDs demonstrate a significant enhancement in output power. However, UV-LEDs with QDs experience efficiency droop due to the severe electron leakage caused by the high mobility of electrons and the weak confinement of EBL. To further reduce electron leakage, the superlattice EBL is proposed to enhance electron confinement through the multi-reflection effects on the electron wave function. But the polarization field resulting from lattice mismatch between LQB and EBL induces the downward band bending at the LQB/EBL interface and reduces effective barrier height of EBL for electrons. The AlInGaN/AlInGaN LMSL EBL is designed to suppress electron leakage while mitigating lattice mismatch. The results indicate that the UV-LED with QDs and LMSL EBL has the increased electron and hole concentration in the active region, the reduced electron leakage, the enhanced radiative recombination rate and the reduced turn-on voltage. Meanwhile, the efficiency droop caused by electron leakage is mitigated. When the injection current is 120 mA, the external quantum efficiency is increased to 9.3% and the output power is increased to 38.3 mW. In summary, the utilization of QDs and LMSL EBL contributes to addressing the issues of severe electron leakage and insufficient hole injection, which provides a valuable reference for the preparation of high-efficiency UV-LEDs.
Funding
National Natural Science Foundation of China (62174148); National Key Research and Development Program of China (2016YFE0118400, 2022YFE0112000); Key Program for International Joint Research of Henan Province (231111520300); Science and Technology Innovation 2025 Major Project of Ningbo (2019B10129); Zhengzhou 1125 Innovation Project (ZZ2018-45).
Acknowledgments
We thank Li Zimo and Yin Xue 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. D. B. Li, K. Jiang, X. J. Sun, et al., “AlGaN photonics: recent advances in materials and ultraviolet devices,” Adv. Opt. Photonics 10(1), 43–110 (2018). [CrossRef]
2. S. Zhao, H. P. T. Nguyen, M. G. Kibria, et al., “III-Nitride nanowire optoelectronics,” Prog. Quantum Electron. 44, 14–68 (2015). [CrossRef]
3. Y. Z. Wang, X. F. Zheng, F. Dai, et al., “Influence of constant current stress on the conduction mechanisms of reverse leakage current in UV-A light emitting diodes,” Superlattices Microstruct. 112, 105–110 (2017). [CrossRef]
4. Z. H. Zhang, C. S. Chu, C. H. Chiu, et al., “UVA light-emitting diode grown on Si substrate with enhanced electron and hole injections,” Opt. Lett. 42(21), 4533–4536 (2017). [CrossRef]
5. H. C. Tao, S. R. Xu, Y. R. Cao, et al., “Enhanced performance of N-polar AlGaN-based ultraviolet light-emitting diodes with lattice-matched AlInGaN insertion in n-AlGaN layer,” IEEE Photonics J. 15(3), 1–5 (2023). [CrossRef]
6. E. V. Navarro, I. Levchuk, J. J. R. Marquez, et al., “Inactivation of simulated aquaculture stream bacteria at low temperature using advanced UVA- and solar-based oxidation methods,” Sol. Energy 227, 477–489 (2021). [CrossRef]
7. N. Masomboon, C. W. Chen, J. Anotai, et al., “A statistical experimental design to determine o-toluidine degradation by the photo-Fenton process,” Chem. Eng. J. 159(1-3), 116–122 (2010). [CrossRef]
8. S. A. D. L. T. Pari, J. C. R. Aquino, A. F. C. Chilo, et al., “Enhancing the photoconductivity and gas sensing performance of TiO2/SnO2 heterostructures tuned by the thickness of the SnO2 upper layer,” Appl. Surf. Sci. 613, 156028 (2023). [CrossRef]
9. A. Khan, K. Balakrishnan, and T. Katona, “Ultraviolet light-emitting diodes based on group three nitrides,” Nat. Photonics 2(2), 77–84 (2008). [CrossRef]
10. L. F. He, K. Zhang, H. L. Wu, et al., “Efficient carrier transport for 368 nm ultraviolet LEDs with a p-AlInGaN/AlGaN short-period superlattice electron blocking layer,” J. Mater. Chem. C 9(25), 7893–7899 (2021). [CrossRef]
11. A. I. Ekimov and A. A. Onushchenko, “Quantum size effect in three-dimensional microscopic semiconductor crystals,” JEPT Lett. 34(6), 345–348 (1981).
12. L. E. Brus, “Electron-electron and electron-hole interactions in small semiconductor crystallites: The size dependence of the lowest excited electronic state,” J. Chem. Phys. 80(9), 4403–4409 (1984). [CrossRef]
13. C. B. Murray, D. J. Noms, and M. G. Bawendi, “Synthesis and characterization of nearly monodisperse CdE (E = S, Se, Te) semiconductor nanocrystallites,” J. Am. Chem. Soc. 115(19), 8706–8715 (1993). [CrossRef]
14. L. Wang, D. Yang, Z. B. Hao, et al., “Metal-organic-vapor phase epitaxy of InGaN quantum dots and their applications in light-emitting diodes,” Chin. Phys. B 24(6), 067303 (2015). [CrossRef]
15. J. M. Gerard, O. Cabrol, and B. Sermage, “InAs quantum boxes: Highly efficient radiative traps for light emitting devices on Si,” Appl. Phys. Lett. 68(22), 3123–3125 (1996). [CrossRef]
16. Y. Narukawa, Y. Kawakami, M. Funato, et al., “Role of self-formed InGaN quantum dots for exciton localization in the purple laser diode emitting at 420 nm,” Appl. Phys. Lett. 70(8), 981–983 (1997). [CrossRef]
17. J. S. Im, S. Heppel, H. Kollmer, et al., “Evidence for quantum-dot-like states in GaInN/GaN quantum wells,” J. Cryst. Growth 189-190, 597–600 (1998). [CrossRef]
18. A. F. Ioffe and E. P. O’Reilly, “Theory of the gain characteristics of InGaN/AlGaN QD Lasers,” MRS Internet j. nitride semicond. res. 4(S1), 721–726 (1999). [CrossRef]
19. I. K. Park, M. K. Kwon, S. B. Seo, et al., “Ultraviolet light-emitting diodes with self-assembled InGaN quantum dots,” Appl. Phys. Lett. 90(11), 111116 (2007). [CrossRef]
20. J. Brault, B. Damilano, A. Kahouli, et al., “Ultra-violet GaN/Al0.5Ga0.5N quantum dot based light emitting diodes,” J. Cryst. Growth 363, 282–286 (2013). [CrossRef]
21. J. Piprek and S. Li, “Electron leakage effects on GaN-based light-emitting diodes,” Opt. Quant. Electron 42(2), 89–95 (2010). [CrossRef]
22. F. Xie and F. X. Wang, “Improved performance of N-face AlGaN-based deep ultraviolet light-emitting diodes with superlattice electron blocking layer,” Appl. Phys. A 122(8), 781 (2016). [CrossRef]
23. X. Y. Yi, H. Q. Sun, Z. F. Li, et al., “Special AlGaN graded superlattice hole and electron blocking layers improved performance of AlGaN-based ultraviolet light-emitting diodes,” Opt. Laser Technol. 106, 469–473 (2018). [CrossRef]
24. R. K. Mondal, V. Chatterjee, and S. Pal, “Effect of step-graded superlattice electron blocking layer on performance of AlGaN based deep-UV light emitting diodes,” Physica E Low Dimens. Syst. Nanostruct. 108, 233–237 (2019). [CrossRef]
25. Y. K. Kuo, J. Y. Chang, and M. C. Tsai, “Enhancement in hole-injection efficiency of blue InGaN light emitting diodes from reduced polarization by some specific designs for the electron blocking layer,” Opt. Lett. 35(19), 3285–3287 (2010). [CrossRef]
26. M. Shatalov, J. Zhang, A. S. Chitnis, et al., “Deep ultraviolet light-emitting diodes using quaternary AlInGaN multiple quantum wells,” IEEE J. Select. Topics Quantum Electron. 8(2), 302–309 (2002). [CrossRef]
27. Y. J. Kim and S. N. Lee, “Lattice-matched AlInGaN electron-blocking layer embedded in high-performance GaN-based light-emitting devices,” J. Korean Phys. Soc. 72(10), 1194–1197 (2018). [CrossRef]
28. S. Ahmad, M. A. Raushan, H. Gupta, et al., “Performance enhancement of UV quantum well light emitting diode through structure optimization,” Opt. Quantum Electron. 51(7), 243 (2019). [CrossRef]
29. P. Du, X. Y. Zhao, Y. Z. Qian, et al., “Rational superlattice electron blocking layer design for boosting the quantum efficiency of 371 nm ultraviolet light-emitting diodes,” IEEE Trans. Electron Devices 68(12), 6255–6261 (2021). [CrossRef]
30. Y. K. Kuo, J. Y. Chang, M. C. Tsai, et al., “Advantages of blue InGaN multiple-quantum well light emitting diodes with InGaN barriers,” Appl. Phys. Lett. 95(1), 011116 (2009). [CrossRef]
31. Y. L. Wang, P. X. Li, S. Y. Jiang, et al., “Enhancing the luminous efficiency of ultraviolet light emitting diodes by adjusting the Al composition of pre-well superlattice,” IEEE Photon. J. 13(5), 8200106 (2021). [CrossRef]
32. Y. L. Wang, P. X. Li, S. R. Xu, et al., “Double superlattice structure for improving the performance of ultraviolet light-emitting diodes,” Chinese Phys. B 28(3), 038502 (2019). [CrossRef]
33. D. Barettin, M. A. Maur, A. Carlo, et al., “Influence of electromechanical coupling on optical properties of InGaN quantum-dot based light-emitting diodes,” Nanotechnology 28(1), 015701 (2017). [CrossRef]
34. I. L. Krestnikov, N. N. Ledentsov, A. Hoffmann, et al., “Quantum dot origin of luminescence in InGaN-GaN structures,” Phys. Rev. B 66(15), 155310 (2002). [CrossRef]
35. C. S. Xia, W. D. Hu, C. Wang, et al., “Simulation of InGaN/GaN multiple quantum well light-emitting diodes with quantum dot model for electrical and optical effects,” Opt. Quantum Electron. 38(12-14), 1077–1089 (2007). [CrossRef]
36. Y. L. Lai, C. P. Liu, and Z. Q. Chen, “Study of the dominant luminescence mechanism in InGaN/GaN multiple quantum wells comprised of ultrasmall InGaN quasiquantum dots,” Appl. Phys. Lett. 86(12), 121915 (2005). [CrossRef]
37. L. Nistor, H. Bender, A. Vantomme, et al., “Direct evidence of spontaneous quantum dot formation in a thick InGaN epilayer,” Appl. Phys. Lett. 77(4), 507–509 (2000). [CrossRef]
38. A. J. Ghazai, S. M. Thahab, H. A. Hassan, et al., “The effects of quantum wells number and the built-in polarization on the performance of quaternary AlInGaN UV laser diode,” Optik 123(10), 856–859 (2012). [CrossRef]
39. V. Fiorentini, F. Bernardini, and O. Ambacher, “Evidence for nonlinear macroscopic polarization in III–V nitride alloy heterostructures,” Appl. Phys. Lett. 80(7), 1204–1206 (2002). [CrossRef]
40. J. Y. Chang, B. T. Liou, M. F. Huang, et al., “High-efficiency deep-ultraviolet light-emitting diodes with efficient carrier confinement and high light extraction,” IEEE Trans. Electron Devices 66(2), 976–982 (2019). [CrossRef]
41. I. Vurgaftman and J. R. Meyer, “Band parameters for nitrogen-containing semiconductors,” J. Appl. Phys. 94(6), 3675–3696 (2003). [CrossRef]
42. J. Piprek and Z. M. S. Li, “Sensitivity analysis of electron leakage in III-nitride light-emitting diodes,” Appl. Phys. Lett. 102(13), 131103 (2013). [CrossRef]
43. J. Brault, M. A. Khalfioui, S. Matta, et al., “UVA and UVB light emitting diodes with AlyGa1−yN quantum dot active regions covering the 305-335 nm range,” Semicond. Sci. Technol. 33(7), 075007 (2018). [CrossRef]
44. P. P. Li, H. J. Li, L. C. Wang, et al., “High quantum efficiency and low droop of 400-nm InGaN near-ultraviolet light-emitting diodes through suppressed leakage current,” IEEE J Quantum Electron. 51(9), 1–5 (2015). [CrossRef]
45. D. Barettin, M. A. Maur, A. Carlo, et al., “Carrier transport and emission efficiency in InGaN quantum-dot based light-emitting diodes,” Nanotechnology 28(27), 275201 (2017). [CrossRef]
46. I. K. Park, M. K. Kwon, J. O. Kim, et al., “Green light-emitting diodes with self-assembled In-rich InGaN quantum dots,” Appl. Phys. Lett. 91(13), 133105 (2007). [CrossRef]
47. J. Brault, S. Matta, T. H. Ngo, et al., “Investigation of AlyGa1-yN/Al0.5Ga0.5N quantum dot properties for the design of ultraviolet emitters,” Jpn. J. Appl. Phys. 55(5S), 05FG06 (2016). [CrossRef]
48. Y. H. Liang and E. Towe, “Low-threshold voltage ultraviolet light-emitting diodes based on (Al,Ga)N metal–insulator–semiconductor structures,” Appl. Phys. Express 10(12), 121005 (2017). [CrossRef]
49. K. Nagata, H. Makino, H. Miwa, et al., “Reduction in operating voltage of AlGaN homojunction tunnel junction deep-UV light-emitting diodes by controlling impurity concentrations,” Appl. Phys. Express 14(8), 084001 (2021). [CrossRef]
50. A. T. M. G. Sarwar, B. J. May, J. I. Deitz, et al., “Tunnel junction enhanced nanowire ultraviolet light emitting diodes,” Appl. Phys. Lett. 107(10), 101103 (2015). [CrossRef]
51. X. Liu, S. R. Xu, H. C. Tao, et al., “High efficiency deep ultraviolet light-emitting diodes with polarity inversion of hole injection layer,” IEEE Photonics J. 15(2), 1–5 (2023). [CrossRef]
52. J. Piprek, “Efficiency droop in nitride-based light-emitting diodes,” Phys. Status Solidi A 207(10), 2217–2225 (2010). [CrossRef]
53. M. Zhang, P. Bhattacharya, and W. Guo, “InGaN/GaN self-organized quantum dot green light emitting diodes with reduced efficiency droop,” Appl. Phys. Lett. 97(1), 011103 (2010). [CrossRef]
54. M. C. Teodorani and D. Barettin, “Optimization of InGaN quantum-dot based light-emitting diodes by means of cellular automata algorithms,” Opt. Quantum Electron. 54(7), 398 (2022). [CrossRef]
55. S. Matta, J. Brault, T. H. Ngo, et al., “Influence of the heterostructure design on the optical properties of GaN and Al0.1Ga0.9N quantum dots for ultraviolet emission,” J. Appl. Phys. 122(8), 085706 (2017). [CrossRef]
