Open Access
Add to BibSonomyAbstract
We report significant improvement in optical and electrical properties of green InGaN/GaN light-emitting diodes (LEDs) by using Si-doped graded short-period InGaN/GaN superlattice (SiGSL) formed by so called indium-conversion technique. For comparison, a conventional LED without the superlattice (C-LED) and a LED with undoped graded superlattice (unGSL-LED) were prepared, respectively. The photoluminescence (PL) intensity of the SiGSL-LED was increased more than 3 times at room temperature (RT) as compared to C-LED. The PL intensity ratios of RT to 10K for the C-LED, unGSL-LED, and SiGSL-LED were measured to be 25, 40.9, and 47.5%, respectively. The difference in carrier lifetimes between 10K and RT for the SiGSL-LED is relatively small compared to that of the C-LED, which is consistent with the variation in PL intensity. The output power of a transistor-outline type SiGSL-LED was increased more than 2 times higher than that of the C-LED.
© 2016 Optical Society of America
1. Introduction
It has been reported that quantum efficiency of light-emitting diodes (LEDs) based on III-nitride semiconductors is drastically decreased above the emission wavelength of ~500 nm, so called “green gap” [1–4 ]. Low efficiency of green emission for InGaN/GaN LEDs is largely related to crystal quality of InGaN with high indium (In) content, quantum confined stark effect (QCSE), and auger process which also become stronger as increasing In composition [5,6 ]. In particular, strain alleviation in a high-In InGaN/GaN quantum well (QW) is equally or even more important issue than the reduction in the defect density in green LEDs, since the In composition is strongly localized. Recently, in order to improve crystal quality and reduce strain of green LEDs, many effective methods including design of InGaN/GaN multiple QW (MQW), superlattice (SL) structure, electron blocking layer, and p/n-doping techniques have been actively proposed [4,7–12 ]. Of many methods, a SL structure plays important roles in performances of GaN-based optoelectronic devices. Zhang et al. reported that an effective advantage of LED structures using a Mg-doped AlGaN/GaN SLs was the enhancement in hole concentration and the block for electrons escaping in an InGaN/GaN MQW by decreasing polarization-induced field [13]. Also, an undoped InGaN/GaN SL structure worked as a strain-relief layer resulting in the decrease of internal polarization fields in MQW [14]. Sizov et al. reported that an undoped InGaN/GaN SL structure positioned between n-GaN and InGaN/GaN MQW decreased defect concentration and dislocations in an active region. However, the increase in the leakage of carriers was observed with increasing injection level for this undoped SL [15]. Therefore, more discussion including design and doping effects of strained SL structures on device characteristics of green LEDs is necessary.
In the present work, we report the influences of an undoped graded short-period InGaN/GaN SL (GSL) and Si-doped GSL (SiGSL) structures on green InGaN/GaN LEDs. A conventional InGaN/GaN LED (C-LED) without the GSL structure was also prepared as a reference. The structural, optical, and electrical properties of the LEDs were studied by transmission electron microscope (TEM), field-emission scanning electron microscope (FE-SEM), high-resolution x-ray diffraction (HRXRD), photoluminescence (PL), time-resolved PL (TRPL), and electroluminescence (EL) spectroscopies.
2. Experiment
The samples used in this study were grown on patterned sapphire substrates with lens-shaped patterns by Thomas-Swan vertical-metal-organic chemical-vapor deposition system. Trimethygallium, trimethylindum, and ammonia were used for gallium (Ga), In, and nitrogen sources, respectively. Disilane (Si2H6) and bis-cyclopentadienyl magnesium (CP2Mg) were used as n-type and p-type doping sources, respectively. Hydrogen and nitrogen were used as carrier gases. The structure of the C-LED consists of a 2.5 μm-thick undoped-GaN buffer layer, a 3 μm Si-doped n-GaN layer, In0.27Ga0.73N/GaN MQW active region, and a 200 nm p-GaN cladding layer. The InGaN/GaN MQW consists of five-periods with 3-nm thick InGaN well layers and 9-nm thick GaN barriers. Undoped and Si-doped twelve-period InGaN/GaN GSL structures fabricated by so called In-conversion technique were inserted below the active MQW layer for the undoped GSL-LED (unGSL-LED) and SiGSL-LED, respectively. For the fabrication of the GSL structure by In-conversion technique, the InGaN/GaN layer was formed by cyclic deposition with subsequent growth interruption in each cycle at the condition of N2:H2 = 7:3 for 25 seconds, after the growth of 2 nm-thick initial In0.02Ga0.98N layer. During the growth interruption, GaN without In or InGaN with extremely low In was realized from the surface of the InGaN layer by In-etching phenomenon under hydrogen ambient. The growth temperature was linearly decreased during the GSL growth. As a result, In composition of InGaN was increased with increasing the number of periods during the GSL growth. According to secondary ion mass spectrometry result (not shown here), the carrier concentration for the SiGSL was measured to be 5 x 1017 cm−3. The LED samples were fabricated by using a standard process with a die size of 600x1000 μm2. Indium-tin-oxide with a thickness of 150 nm was deposited as a transparent conducting layer, and chromium and gold were deposited for p-type and n-type electrodes, respectively. The LED chips were then packaged into transistor-outline type LED lamps.
The structural properties of the LEDs were measured by using TEM (JEM-2100F of JEOL Ltd.) and HRXRD (PANalytical X’Pert PRO). Also, FE-SEM (HITACHI SU8010) was used to measure etch-pit density (EPD) of the LED samples. For PL measurements, a He–Cd laser with a wavelength of 325 nm was used as an excitation source. A time-correlated single photon counting technique was used for TRPL measurements. The excitation source was a picosecond laser (LDH-D-C-405 of Picoquant GmbH Berlin, Germany). The laser had pulse duration of less than 70 pico-seconds at a 40 MHz repetition rate. The collected emission was detected with a micro-channel plate photomultiplier tube. An integrating sphere detector was used for EL measurements for the packaged LEDs.
3. Results and discussion
Figure 1 shows the cross-sectional TEM image for the SiGSL-LED sample with a twelve-period GSL structure. The inset shows the magnified image for the GSL structure. Since In atoms can be easily desorbed from surface of the InGaN layer during the growth interruption under hydrogen ambient for the formation of the GSL structure by In-conversion technique [16], there is enhanced possibility for Ga atoms to move to vacant In sites. As a result, GaN or InGaN with extremely low In composition can be formed for the GSL structure. Since the formation procedure for the GSL structure is different from a conventional SL grown by alternately depositing two different epitaxial layers, the GSL interface between GaN and InGaN with different In composition cannot be defined as clearly as the conventional SL. The thicknesses of the GaN and InGaN layers at the GSL structure are not uniform due to local migration behaviors of In and Ga atoms [17]. In addition, since the substrate temperature was continuously decreased during the growth of the GSL, In composition and thickness of InGaN were increased with increasing the periods. According to the energy dispersive spectroscopy measurements (not shown here), In composition of the InGaN layer at the GSL structure was increased from 2 to 13%. The average thicknesses of the GaN and InGaN layers for the GSL structure were measured to be 1.0 and 1.2 nm, respectively.
Figure 2 shows the surface images with etch pits for the LED samples, obtained from FE-SEM measurements. For EPD measurements, selective wet-etching process by using molten KOH solution was carried out. The EPD of the C-LED, unGSL-LED, and SiGSL-LED were measured to be 7.9x107, 5.5x107, and 4.2x107 cm−2, respectively. The EPD of the SiGSL-LED was relatively low compared to that of the C-LED. If we consider that the device structure and growth conditions for the unGSL-LED and SiGSL-LED were exactly same as those of the C-LED except for the GSL structure, the defects properties may be largely influenced by the GSL structure. That is, the EPD reduction for the unGSL-LED and SiGSL-LED compared to that of the C-LED can be explained by partial blocking effect of threading dislocation by inserting the GSL structure. Even though only small fraction of threading dislocations causes the formation of etch pits observed at the surface, EPD is strongly depending on existence of threading dislocation [18,19 ]. In addition, the EPD of the SiGSL-LED was lower than that of the unGSL-LED. This is largely attributed to the influence of Si atoms at the GSL structure on dislocation propagation, which will be discussed later.
Figure 3 shows the HRXRD spectra of the (0002) diffractions for the C-LED, unGSL-LED, and SiGSL-LED, where the strongest peak originates from a GaN buffer layer. The satellite peaks up to the fifth order were clearly defined for the LED samples. As shown in the inset, the positions of the satellite peaks for the unGSL-LED and SiGSL-LED samples are shifted toward the GaN peak as compared to the C-LED, indicating that the amount of strain in the MQW region was reduced. In addition, the shift in peak positions for the SiGSL-LED was relatively large compared to that of the unGSL-LED, which is due the Si doping for the GSL structure. As reported by Xu et al., strain for GaN thin films grown on sapphire substrates was significantly relaxed by low Si doping (<7x1017 cm−3) via the mechanism of bending dislocation [20,21 ]. The additional strain relaxation for the SiGSL-LED compared to unGSL-LED resulted in the reduction in the density of dislocations leading to the decrease in the EPD at the surface as shown in the FE-SEM image.
Fig. 3 HRXRD spectra of (0002) diffractions for the LED samples, where the inset is a magnified view.
Figure 4(a) shows the PL spectra for the C-LED, unGSL-LED, and SiGSL-LED measured at RT. The inset shows the PL spectra of the LED samples measured at 10K. The peak wavelengths of the C-LED, unGSL-LED, and SiGSL-LED are 539 (533), 531 (525), and 522 nm (519 nm) at RT (10K), respectively. The peak wavelengths of unGSL-LED and SiGSL-LED were blue-shifted compared to that of the C-LED, due to the strain relaxation by the GSL structure. That is, this strain relaxation resulted in reduction in the QCSE at the InGaN/GaN MQW [22,23 ]. In addition, the amount of blue-shift for the SiGSL-LED (17 nm) from the C-LED is larger than that of the unGSL-LED (8 nm). This is mainly related to the additional contribution of Si-doping to the compensation of the QCSE via Coulomb screening [24]. Furthermore, the PL intensities for the unGSL-LED and SiGSL-LED were stronger than that of the C-LED both at 10K and RT. The RT-PL intensity of unGSL-LED and SiGSL-LED increases more than 2.2 and 3 times, respectively, as compared to C-LED, which can be explained again by the reduction in dislocation density and the alleviation of the QCSE by inserting the GSL structure. As discussed before, the amount of the QCSE for the SiGSL-LED was decreased compared to the C-LED, resulting that the overlap integral between electron and hole wave-functions at the MQW region was increased. As a result, radiative recombination component was enhanced. Figure 4(b) shows the PL intensity ratios of RT to 10K, which were measured to be 25, 40.9, and 47.5% for the C-LED, unGSL-LED, and SiGSL-LED samples, respectively. The intensity ratio of the SiGSL–LEDs was relatively higher than that of unGSL-LED and C-LED, which is related to the reduction of dislocation density.
Fig. 4 (a) PL spectra at RT, where the inset is the 10K PL spectra, and (b) PL intensity ratios of RT to 10K for the LED samples, where a dotted line is only a guide for the eyes.
Figure 5 shows the TRPL decay curves at the peak wavelength of the LED samples. In Fig. 5(a), the decay curves for the unGSL-LED and SiGSL-LED samples were relatively faster than that for the C-LED at 10K. The carrier lifetime (τ) estimated from the decay curve by using a single-exponential function of the form I(t) = Aexp(-t/τ), where A is a pre-exponential constant contributing to PL intensity, was measured as 23.83 ns for the SiGSL-LED at 10K. This value was relatively short compared to that of the C-LED with 28.19 ns.
Also, the decay curve for the SiGSL-LED was faster than that of the unGSL-LED with 25.05 ns. When the ambient temperature of 10K is assumed to be in a thermal equilibrium state, the carriers disappear largely due to the radiative recombination process with non-radiative recombination caused by defects. Therefore, the relatively short carrier lifetime for the SiGSL-LED is likely to the reduction in dislocation density and the increase in the overlap integral between electron and hole wave-functions caused by the reduction in the QCSE in InGaN/GaN MQW, which is in accordance with the blue-shift in emission wavelength and the increase in PL intensity for the SiGSL-LED. However, as shown in Fig. 5(b), the decay curves for the unGSL-LED and SiGSL-LED samples were relatively slower than that for the C-LED at RT. The RT carrier lifetime analyzed by a two-exponential function [25] was evaluated to be 4.6 ns for the SiGSL-LED, which is longer than that for the C-LED (3.18 ns). The relatively slow decay curve for the SiGSL-LED indicates that the non-radiative component of carrier lifetime is elongated because of the reduction in dislocation [26]. In addition, the difference in carrier lifetime for the SiGSL-LED between 10K and RT is relatively small compared to that of the C-LED, indicating that the radiative recombination rate is relatively stable with the suppressed activation of non-radiative recombination processes over the whole temperature range. These results are well agreed with the variation in the PL intensity for SiGSL-LED with increasing temperature compared with C-LED. From this result, the highly efficient LED with high brightness can be obtained by using the SiGSL structure.
Figure 6(a) shows the current-voltage (I-V) characteristic curves of LED samples. At the injection current of 20 mA, it was found that the forward voltages were 2.82, 2.8, and 2.74 V for the C-LED, unGSL-LED, and SiGSL-LED, respectively. The slopes of the unGSL-LED and SiGSL-LED after turned-on voltage were relatively steeper than that of the C-LED. The resistances estimated from the I-V curves were 5.2, 4, and 3.5 Ω for the C-LED, unGSL-LED, and SiGSL-LED, respectively. The reduction in the turn-on voltage and the resistance for the SiGSL-LED is due to relatively low resistivity in the SiGSL structure by Si doping [27] and the improvement in crystal quality. In Fig. 6(b), the output power of unGSL-LED and SiGSL-LED at injection current of 120 mA were measured to be 28.9 and 38.9 mW, which is 1.5 and 2.0 times higher than that of the C-LED (19.3 mW), respectively. This result can be explained again by reduction in dislocation density and an alleviation of the QCSE due to strain relaxation by the GSL structure. The enhancement in output power is also partially due to the increase in carrier injection efficiency of unGSL-LED and SiGSL-LED compared to that of the C-LED, because energy-band structure between the n-GaN and InGaN/GaN MQWs was modulated by inserting the GSL structure [28]. In addition, the increasing rate of light output power with current for the SiGSL-LED is relatively higher than that of the unGSL-LED, which is attributed to enhanced current spreading effect due to the relatively low resistance and additional contribution of Si-doping to the compensation of the QCSE.
Fig. 6 (a) I-V curves for LED samples, and (b) light output powers for C-LED, unGSL-LED and SiGSL-LED at injection current from 2 to 200 mA.
Figure 7 shows the EL peak wavelengths of the LED samples with respect to current ranging from 2 to 200 mA. The inset shows the EL spectra of the LED samples at the injection current of 20 mA, where the peak wavelengths were measured to be 536, 533, and 531 nm for the C-LED, unGSL-LED, and SiGSL-LED, respectively. The peak wavelengths of all the LED samples were blue-shifted with increasing injection current, which is attributed to the band-filling effect and compensation of the QCSE [29,30 ]. The amount of blue-shift for the SiGSL-LED at the current ranging from 2 to 100 mA was measured to be 6.6 nm, which is relatively small compared to the C-LED (9.2 nm) and unGSL-LED (8.2 nm). Since the injection current is same, the contribution of band-filling effect to the blue-shift is considered as same for all the LED samples. Therefore, the relatively small blue-shift for the SiGSL-LED sample is mainly related to the additional contribution of Si-doping on the Coulomb screening of the QCSE. In addition, the strain reduction in the MQW for the SiGSL-LED by the GSL structure, as shown in the HRXRD spectra, is partially responsible for the relatively small blue-shift in peak wavelength. Usually, when injection current is further increased, a red-shift in the emission wavelength for LEDs is observed due to heat generation by current injection [31]. For the C-LED and unGSL-LED samples, the emission wavelengths were changed to red-shift at the injection current of 120 and 160 mA, respectively. For the SiGSL-LED, the red-shift in the peak wavelength was not observed at the injection current of below 200 mA, which is relatively higher level compared to those of the C-LED, indicating to more stable in electrical properties. This can be explained by the decrease in heat generation probability at a relatively high injection current due to the low total resistance of the SiGSL-LED, which was confirmed from I-V curves shown in Fig. 6(a). As a result, the probability for current spreading into the MQW region could be improved.
Fig. 7 Summary on the peak wavelengths for the LED samples with respect to currents ranging from 2 to 200 mA, where the inset is the EL spectra for the LED samples at injection current of 20 mA.
4. Conclusion
In conclusion, we discussed the influences of unGSL and SiGSL structures realized by In conversion technique on device characteristics of green InGaN/GaN LEDs. The results indicate that the SiGSL structure can be an effective way to improve optical and electrical properties of green InGaN/GaN LEDs.
Acknowledgments
This work was supported by a National Research Foundation of Korea (NRF) funded by the Ministry of Science, ICT and Future Planning (BRL. No. 2015042417), by the Ministry of Education (No. 2015R1D1A1A01060681), and by Business for Cooperative R&D between Industry, Academy, and Research Institute funded Korea Small and Medium Business Administration in 2015 (Grant No. C0267277).
References and links
1. M. R. Krames, O. B. Shchekin, R. M. Mach, G. O. Mueller, L. Zhou, G. Harbers, and M. G. Craford, “Status and future of high-power light-emitting diodes for solid-state lighting,” J. Disp. Technol. 3(2), 160–175 (2007). [CrossRef]
2. T. Langer, A. Kruse, F. A. Ketzer, A. Schwiegel, L. Hoffmann, H. Jonen, H. Bremers, U. Rossow, and A. Hangleiter, “Origin of the “green gap”: increasing nonradiative recombination in indium-rich GaInN/GaN quantum well structures,” Phys. Status Solidi., C Curr. Top. Solid State Phys. 8(7–8), 2170–2172 (2011). [CrossRef]
3. M. H. Crawford, “LEDs for solid-state lighting: performance challenges and recent advances,” IEEE Sel. Top. Quantum Electron. 15(4), 1028–1040 (2009). [CrossRef]
4. S. Saito, R. Hashimoto, J. Hwang, and S. Nunoue, “InGaN light-emitting diodes on c-face sapphire substrates in green gap spectral range,” Appl. Phys. Express 6(11), 111004 (2013). [CrossRef]
5. M. A. Hoffbauer, T. L. Williamson, J. J. Williams, J. L. Fordham, K. M. Yu, W. Walukiewicz, and L. A. Reichertz, “In-rich InGaN thin films: progress on growth, compositional uniformity, and doping for device applications,” J. Vac. Sci. Technol. B 31(3), 03C114 (2013). [CrossRef]
6. K. T. Delaney, P. Rinke, and C. G. Van de Walle, “Auger recombination rates in nitrides from first principles,” Appl. Phys. Lett. 94(19), 191109 (2009). [CrossRef]
7. R. Hashimoto, J. I. Hwang, S. Saito, and S. Nunoue, “High-efficiency green-yellow light-emitting diodes grown on sapphire (0001) substrate,” Phys. Status Solidi., C Curr. Top. Solid State Phys. 10(11), 1529–1532 (2013). [CrossRef]
8. H. J. Li, P. P. Li, J. J. Kang, Z. Li, Y. Y. Zhang, Z. Li, J. Li, X. Y. Yi, J. M. Li, and G. H. Wang, “Quantum efficiency enhancement of 530 nm InGaN green light-emitting diodes with shallow quantum well,” Appl. Phys. Express 6(5), 052102 (2013). [CrossRef]
9. C. H. Wang, S. P. Chang, P. H. Ku, J. C. Li, Y. P. Lan, C. C. Lin, H. C. Yang, H. C. Kuo, T. C. Lu, S. C. Wang, and C. Y. Chang, “Hole transport improvement in InGaN/GaN light-emitting diodes by graded-composition multiple quantum barriers,” Appl. Phys. Lett. 99(17), 171106 (2011). [CrossRef]
10. Y. A. Chang, Y. T. Kuo, J. Y. Chang, and Y. K. Kuo, “Investigation of InGaN green light-emitting diodes with chirped multiple quantum well structures,” Opt. Lett. 37(12), 2205–2207 (2012). [CrossRef] [PubMed]
11. C. Du, Z. Ma, J. Zhou, T. Lu, Y. Jiang, P. Zuo, H. Jia, and H. Chen, “Enhancing the quantum efficiency of InGaN yellow-green light-emitting diodes by growth interruption,” Appl. Phys. Lett. 105(7), 071108 (2014). [CrossRef]
12. N. Zhang, Z. Liu, T. B. Wei, L. Zhang, X. C. Wei, X. D. Wang, H. X. Lu, J. M. Li, and J. X. Wang, “Effect of the graded electron blocking layer on the emission properties of GaN-based green light-emitting diodes,” Appl. Phys. Lett. 100(5), 053504 (2012). [CrossRef]
13. Y. Y. Zhang and Y. A. Yin, “Performance enhancement of blue light-emitting diodes with a special designed AlGaN/GaN superlattice electron-blocking layer,” Appl. Phys. Lett. 99(22), 221103 (2011). [CrossRef]
14. C. Jia, T. Yu, H. Lu, C. Zhong, Y. Sun, Y. Tong, and G. Zhang, “Performance improvement of GaN-based LEDs with step stage InGaN/GaN strain relief layers in GaN-based blue LEDs,” Opt. Express 21(7), 8444–8449 (2013). [CrossRef] [PubMed]
15. V. S. Sizov, A. F. Tsatsulnikov, A. V. Sakharov, W. V. Lundin, E. E. Zavarin, N. A. Cherkashin, M. J. Hytch, A. E. Nikolaev, A. M. Mintairovd, Y. He, and J. L. Merz, “The use of short-period InGaN/GaN superlattices in blue-region light-emitting diodes,” Semiconductors 44(7), 924–930 (2010). [CrossRef]
16. A. F. Tsatsulnikov, W. V. Lundin, A. V. Sakharov, E. E. Zavarin, S. O. Usov, A. E. Nikolaev, N. V. Kryzhanovskaya, V. S. Sizov, M. A. Synitsin, E. V. Yakovlev, A. E. Chernyakov, A. L. Zakgeim, N. A. Cherkashin, and M. Hytch, “InGaN/GaN short-period superlattices: synthesis, properties, applications,” Phys. Status Solidi., C Curr. Top. Solid State Phys. 8(7–8), 2308–2310 (2011). [CrossRef]
17. N. Duxbury, U. Bangert, P. Dawson, E. J. Thrush, W. Van der Stricht, K. Jacobs, and I. Moerman, “Indium segregation in InGaN quantum-well structures,” Appl. Phys. Lett. 76(12), 1600 (2000). [CrossRef]
18. Y. Chen, T. Takeuchi, H. Amano, I. Akasaki, N. Yamada, Y. Kaneko, and S. Y. Wang, “Pit formation in GaInN quantum wells,” Appl. Phys. Lett. 72(6), 710 (1998). [CrossRef]
19. H. K. Cho, J. Y. Lee, G. M. Yang, and C. S. Kim, “Formation mechanism of V defects in the InGaN/GaN multiple quantum wells grown on GaN layers with low threading dislocation density,” Appl. Phys. Lett. 79(2), 215 (2001). [CrossRef]
20. P. Cantu, F. Wu, P. Waltereit, S. Keller, A. E. Romanov, U. K. Mishra, P. DenBaars, and J. S. Speck, “Si doping effect on strain reduction in compressively strained Al0.49Ga0.51N thin films,” Appl. Phys. Lett. 83(4), 674 (2003). [CrossRef]
21. Z. Xu, J. Zhang, H. Duan, Z. Zhang, Q. Zhu, H. Xu, and Y. Hao, “Stress, structural and electrical properties of Si-doped GaN film grown by MOCVD,” J. Semicond. 30(12), 123003 (2009). [CrossRef]
22. N. Nanhui, W. Huaibing, L. Jianping, L. Naixin, X. Yanhui, H. Jun, D. Jun, and S. Guangdi, “Enhanced luminescence of InGaN/GaN multiple quantum wells by strain reduction,” Solid-State Electron. 51(6), 860–864 (2007). [CrossRef]
23. K.-M. Song, P.-G. Kang, H.-S. Shin, J.-M. Kim, W.-K. Park, C.-G. Ko, H.-W. Shim, and D. Ho Yoon, “Influence of AlGaN/GaN/InGaN superlattice on the characteristics of LEDs grown by metalorganic chemical vapor deposition,” J. Cryst. Growth 312(20), 2847–2851 (2010). [CrossRef]
24. L. W. Wu, S. J. Chang, T. C. Wen, Y. K. Su, J. F. Chen, W. C. Lai, C. H. Kuo, C. H. Chen, and J. K. Sheu, “Influence of Si-doping on the characteristics of InGaN-GaN multiple quantum-well blue light emitting diodes,” IEEE J. Quantum Electron. 38(5), 446–450 (2002). [CrossRef]
25. M. Y. Ryu, C. Q. Chen, J. S. Kim, and M. Asif Khan, “Optical characterization of quaternary AlInGaN epilayer and multiple quantum wells grown by a pulsed metalorganic chemical vapor deposition,” Curr. Appl. Phys. 11(2), 231–235 (2011). [CrossRef]
26. J. Ma, L. Wang, Z. Liu, G. Yuan, X. Ji, P. Ma, J. Wang, X. Yi, G. Wang, and J. Li, “Nitride-based micron-scale hexagonal pyramids array vertical light emitting diodes by N-polar wet etching,” Opt. Express 21(3), 3547–3556 (2013). [CrossRef] [PubMed]
27. J. K. Sheu, G. C. Chi, and M. J. Jou, “Enhanced output power in an InGaN-GaN multiquantum-well light-emitting diode with an InGaN current-spreading layer,” IEEE Photonics Technol. Lett. 13(11), 1164–1166 (2001). [CrossRef]
28. C. H. Wang, C. C. Ke, C. Y. Lee, S. P. Chang, W. T. Chang, J. C. Li, Z. Y. Li, H. C. Yang, H. C. Kuo, T. C. Lu, and S. C. Wang, “Hole injection and efficiency droop improvement in InGaN/GaN light-emitting diodes by band-engineered electron blocking layer,” Appl. Phys. Lett. 97(26), 261103 (2010). [CrossRef]
29. H. Masui, J. Sonoda, N. Pfaff, I. Koslow, S. Nakamura, and S. P. DenBaars, “Quantum-confined stark effect on photoluminescence and electroluminescence characteristics of InGaN-based light-emitting diodes,” J. Phys. D Appl. Phys. 41(16), 165105 (2008). [CrossRef]
30. T. Kuroda and A. Tackeuchi, “Influence of free carrier screening on the luminescence energy shift and carrier lifetime of InGaN quantum wells,” J. Appl. Phys. 92(6), 3071 (2002). [CrossRef]
31. C. Lu, L. Wang, J. Lu, R. Li, L. Liu, D. Li, N. Liu, L. Li, W. Cao, W. Yang, W. Chen, W. Du, C.-T. Lee, and X. Hu, “Investigation of the electroluminescence spectrum shift of InGaN/GaN multiple quantum well light-emitting diodes under direct and pulsed currents,” J. Appl. Phys. 113(1), 013102 (2013). [CrossRef]
