The effect of employing an AlGaN cap layer in the active region of green c-plane light-emitting diodes (LEDs) was studied. Each quantum well (QW) and barrier in the active region consisted of an InGaN QW and a thin Al0.30Ga0.70N cap layer grown at a relatively low temperature and a GaN barrier grown at a higher temperature. A series of experiments and simulations were carried out to explore the effects of varying the Al0.30Ga0.70N cap layer thickness and GaN barrier growth temperature on LED efficiency and electrical performance. We determined that the Al0.30Ga0.70N cap layer should be around 2 nm and the growth temperature of the GaN barrier should be approximately 75° C higher than the growth temperature of the InGaN QW to maximize the LED efficiency, minimize the forward voltage, and maintain good morphology. Optimized Al0.30Ga0.70N cap growth conditions within the active region resulted in high efficiency green LEDs with a peak external quantum efficiency (EQE) of 40.7% at 3 A/cm2. At a normal operating condition of 20 A/cm2, output power, EQE, forward voltage, and emission wavelength were 13.8 mW, 29.5%, 3.5 V, and 529.3 nm, respectively.
© 2018 Optical Society of America under the terms of the OSA Open Access Publishing Agreement
After the first demonstration of high-brightness blue light-emitting diodes (LEDs) in 1993 , III-nitride alloys have attracted significant attention as the next generation of materials for optoelectronics applications [2,3]. Remarkable progress in white LEDs for solid state lighting (SSL) has been driven by phosphor-converted violet-blue LEDs . High efficiency violet and blue (400-450 nm) [Al,In,Ga]N-based LEDs with wall plug efficiency (WPE) and external quantum efficiency (EQE) exceeding 80% have been reported on different substrates [5–7]. Color-mixed LEDs produce white light by mixing red, green, blue, and amber (RGBA) LEDs and have a fundamental efficiency limit of ~400 lm/W, which is higher than the fundamental efficiency limit of ~300 lm/W for phosphor-converted LEDs [3,8]. However, the poor performance and large efficiency droop of green-to-amber LEDs, known as the “green gap”, is the primary limitation for developing high efficiency color-mixed LEDs [8,9].
A direct correlation has been observed between an increase in indium content of InGaN quantum wells (QWs) and a drop in internal quantum efficiency (IQE) of polar InGaN-based LEDs . Although the origin of the green gap is debated, some recent studies point to the large electric fields, excess defect formation in the InGaN QWs, and an increased Auger recombination as the main causes of the green gap [5,10–17]. Polar III-nitrides suffer from polarization-induced electric fields, which increase with increasing indium concentration in the QWs. This causes a significant band-bending along the growth direction, which leads to a spatial separation of electron and hole wave functions within the QWs, resulting in a reduction in radiative recombination and a decrease in device efficiency [18,19]. The growth of high indium content InGaN QWs is required to achieve long wavelength emission LEDs. Low growth temperatures are necessary to incorporate more indium in the active region. This causes more impurity incorporation, increased surface roughness and a high density of V-defects, resulting in increased Shockley-Read-Hall (SRH) non-radiative recombination [20,21].
Recent progress has been made in long wavelength LEDs by implementing AlGaN cap layers in the active region and holds promise for a new generation of high-efficiency green-amber based InGaN LEDs. Large improvements in IQE for green laser diodes (LD) and LEDs have been demonstrated with InGaN/AlGaN/GaN based QWs [22–24]. In a previous study we reported high performance green LEDs employing Al0.30Ga0.70N cap layers with peak EQE of 44.6% and luminous efficacies of 239 lm/W, respectively . Possible reasons were suggested for the improvements in IQE after implementing AlGaN cap layers, including better control of the polarization field in the active region, reduced SRH nonradiative recombination, and strain compensation of the compressively strained InGaN QWs [24,26–28], although further studies were needed to better understand the effect of the AlGaN cap layers on device performance.
In this work, we investigate the effects of varying the Al0.30Ga0.07N cap layer thickness and the GaN barrier growth temperature on the device characteristics of InGaN/AlGaN/GaN-based multiple-quantum well (MQW) green LEDs. Based on the outcome of a series of experiments and simulations, we conclude that the Al0.30Ga0.70N cap layer should be around 2 nm and the growth temperature of the GaN barrier should be approximately 75° C higher than the growth temperature of the InGaN QW and the Al0.30Ga0.07N cap layer to maximize the LED efficiency and minimize the forward voltage. Optimized active region growth conditions resulted in high efficiency green LEDs with a peak EQE of 40.7% and an output power, EQE, forward voltage, and emission wavelength of 13.8 mW, 29.5%, 3.5 V, and 529.3 nm at 20 A/cm2, respectively.
Green LEDs with different active region growth conditions were grown heteroepitaxially by atmospheric pressure metal organic chemical vapor deposition (MOCVD) on (0001) patterned sapphire substrates (PSS). The n-side of the device structure consisted of a 1 μm unintentionally doped (UID) GaN template layer, a 2 μm Si-doped n-GaN layer ([Si] = 5x1018 cm−3), a thirty period Si-doped ([Si] = 5x1018 cm−3) In0.05Ga0.95N (2.5nm)/GaN (5 nm) superlattice layer, and a 10 nm UID GaN layer prior to the growth of the active region. The active region consisted of an undoped five period MQW that was grown in two steps: the first step consisted of the growth of a 3 nm InGaN QW with ~24% indium content, followed immediately by the growth of an Al0.30Ga0.70N cap layer at the same temperature; the second step consisted of ramping up the temperature 100 °C higher than the QWs and the AlGaN cap in 60 seconds, annealing the AlGaN cap layer for 60 seconds and the growth of a 10 nm high temperature (HT) GaN barrier. The growth rates of the InGaN QW, AlGaN cap layer and HT GaN barrier for all samples in this study were 1 Aͦ/sec, 0.7 Aͦ/sec and 0.35 Aͦ/sec, respectively. The growth temperature of the n-GaN template, InGaN QW and p-GaN were 1200 °C, 800 °C and 920 °C, respectively. Finally, an 10 nm Mg-doped p-Al0.20Ga0.80N electron blocking layer (EBL) was grown on top of the last GaN barrier of the MQW, followed by a 200 nm Mg-doped p-GaN layer ([Mg] = 5x1019 cm−3) and a 10 nm p++-GaN contact layer. The green LED device structure is illustrated in the cross-sectional schematic shown in Fig. 1(a).
We optimized the Aluminum composition in the AlGaN cap and we found that 30% yielded the highest efficiency, which was consistent with Toshiba’s results . Following MOCVD growth, the AlGaN cap layer thickness, QW thickness and Al composition were determined by high-resolution x-ray diffraction (XRD) using a PANalytic MRD PRO diffractometer using separate calibration samples.
The samples were processed into LEDs using standard contact photolithography. A 110 nm Sn-doped In2O3 (ITO) current-spreading layer was deposited at 290°C by electron beam deposition on the p++-GaN contact layer. A rectangular mesa with an active area of 0.1 mm2 was formed using a methane-hydrogen-argon etch to remove ITO followed by a Cl2-based reactive ion etch to reach the n-GaN layer. A 20/100/100/100 nm Ti/Al/Ni/Au n-contact and a 20/100/300 nm Cr/Ni/Au p-contact were deposited by electron beam deposition. The LEDs were singulated by dicing and individual LEDs were mounted with silver paste onto a silver header, wire bonded, and encapsulated in silicone (n = 1.41). Electroluminescence measurements were conducted under continuous-wave operation in a calibrated integrating sphere at room-temperature.
3. Results and discussion
3.1 Influence of AlGaN cap thickness on LED performance
Four green LEDs with different AlGaN cap layer thicknesses (1 nm, 2 nm, 3 nm, and 4.5 nm) were grown to understand the effect of cap thickness on device efficiency and electrical properties. Figure 2(a), 2(b) and 2(d) shows the dependence of EQE, forward voltage and peak wavelength, respectively, on current density for LEDs with different cap layer thicknesses.
Typical efficiency droop behavior is observed for all samples, where the peak EQE occurs at low current density (1–2 A/cm2). Even though the samples have nearly the same EQE peak values, they have different droop ratios as the current density increases. The droop ratio was defined as (EQE20 ̶ EQE80)/ EQE20, where the subscript represents the current density in A/cm2. We note that the efficiency droop becomes larger as the AlGaN cap thickness increases. At current densities between 20 and 80 A/cm2, the droop ratio is 34%, 35.1% and 41.2% for the LEDs with AlGaN cap layers of 2 nm, 3 nm, and 4.5 nm, respectively. The cause of the increased droop can be attributed to the increase in electric field in the QWs as the AlGaN thickness increases and/or the suppression of interwell transport as the AlGaN thickness increases, as discussed in more detail below. The 1 nm AlGaN cap sample is off-trend from the others due to degradation of the QWs during the high temperature GaN barrier growth as a result of insufficient AlGaN cap layer thickness.
Table 1 shows the forward voltage, output power, EQE, full width at half maximum (FWHM), and peak wavelength of all samples at 20 A/cm2. The sample with a 1 nm cap exhibits a significant blueshift with a peak wavelength at 495 nm, while other samples have wavelengths around 525–530 nm. Figure 2(c) shows the electroluminescence (EL) spectra of all samples at 20 A/cm2 drive current. Since the only difference between the samples is the AlGaN cap thickness, it is expected that the peak wavelength should be close for all samples. This implies that the 1 nm AlGaN cap is too thin to protect the QW from In desorption during the growth of the HT GaN barrier. This assumption is also supported by noting that the FWHM of the sample with a 1 nm cap is much larger than all other samples. The large value of FWHM of the 1 nm sample indicates that the QWs had large potential profile fluctuations due to degradation of the QWs during the growth of the higher temperature GaN barrier. Previous work also has shown that InGaN QWs of a green c-plane LD with a thinner GaN cap (1.2 nm) exhibited a larger FWHM and shorter emission wavelength compared to a sample with a thicker GaN cap .
The LEDs with a 2 nm AlGaN cap show the highest efficiency, lowest FWHM, and lowest operating voltage at 20 A/cm2, as illustrated in Table 1. We note that the FWHM becomes larger as the AlGaN cap increases from 2 nm to 4.5 nm. We speculate that this may be related to the increased surface roughness as the thickness of the low temperature AlGaN cap increases, although further studies are needed to investigate the effect of AlGaN cap thickness on the surface morphology of the active region. Figure 2(b) shows the forward voltage as function of current density for all samples. We observed that the forward voltage increased with increasing AlGaN cap thickness. The sample with a 1 nm cap exhibited lower efficiency from the others, which probably was caused by the effect of the high temperature GaN barrier, as discussed above. The series resistance was measured between 20 and 40 A/cm2 and was 21.8, 17, 31, and 35.5 Ω for samples with 1 nm, 2 nm, 3 nm, and 4.5 nm AlGaN cap. The increase in forward voltage and series resistance can be attributed to an increase in the effective potential barrier between QWs with increasing the AlGaN cap layer thickness.
To understand the results from the AlGaN thickness series, simulations were performed for simplified single quantum well (SQW) LED structures using SiLENSe. The simulated structures consisted of a 100 nm n-GaN layer with [Si] = 5x1018 cm−3, a 10 nm UID GaN layer, a 3 nm UID In0.30Ga0.70N QW, a UID Al0.30Ga0.07N cap layer of varying thickness, a 10 nm UID GaN barrier and a 100 nm p-GaN layer with [Mg] = 1x1019 cm−3. Similar to the experimental structures described above, four different AlGaN cap layer thicknesses (1 nm, 2 nm, 3 nm, 4.5 nm) were considered for the simulations. All UID layers were given an n-type background doping level of 1x1016 cm−3, which we believe approximates the background concentration of oxygen in our samples. The simulated LED structures are illustrated in the cross-sectional schematic shown in Fig. 1(b).
Figure 3(a) shows the simulated energy band diagrams and electron and hole wavefunctions for SQW LEDs with 1 nm and 4.5 nm AlGaN cap thicknesses under forward bias (J = 20 A/cm2). Increasing the AlGaN cap thickness from 1 nm to 4.5 nm increases the electric field in the QW, as shown in Fig. 3(b). This causes a larger band-bending along the growth direction, resulting in a further spatial separation of electron and hole wavefunctions within the QW. The resulting reduction in the wavefunction overlap with increasing AlGaN cap thickness reduces the radiative recombination and Auger recombination coefficients . This leads to an increase in carrier density for a given current density, which increases the droop for devices with thicker AlGaN cap layers. Previous reports attributed the correlation between droop and the emission wavelength of c-plane LEDs to a decrease in wavefunction overlap in devices with higher indium contents in their active region .
Figure 4 shows the dependence of the output power on AlGaN cap thickness at 20 A/cm2 for the MQW LEDs and compares it to the simulated square of the electron hole wavefunction overlap for the simplified SQW LED structures. As the cap thickness was increased from 2 nm to 4.5 nm, both the simulated wavefunction overlap and observed output power decreased. This implies that the decrease in efficiency in the samples with thicker AlGaN cap layers can be attributed to the increase in the electric field in the QWs with increasing Al0.30Ga0.70N cap layer thickness.
Although the sample with 1 nm AlGaN cap exhibited the highest wavefunctions overlap, it had the lowest peak efficiency. This noticeable reduction in efficiency occurs because the 1 nm cap is not thick enough to protect the QWs during the growth of the HT GaN barrier, as discussed above.
3.2 Effect of GaN barrier growth temperature on LED performance
The effect of the growth temperature of GaN barrier on the device performance was explored following the optimization of the AlGaN cap layer thickness. The AlGaN cap layer was annealed during the temperature ramp-up prior to the HT GaN barrier growth. Four different green LEDs with varying GaN barrier growth temperatures in the active region were considered. The growth temperature differences between the QW/AlGaN cap layer and the GaN barrier (∆T = THT GaN – TAlGaN) were 0 °C, 50 °C, 75 °C and 100 °C. The LED structures grown for this series were identical to the LED with an optimized 2 nm AlGaN cap layer.
Figure 5 shows EQE curves, forward voltage, wall-plug efficiency (WPE), and EL spectra as a function of current density for LEDs with various GaN barrier growth temperatures. The efficiency of the LEDs significantly increased when the ∆T increased from 0°C to 75°C, as shown in Fig. 5(a). All samples exhibited efficiency droop behavior except for the LED with ∆T = 0 °C, which peaked at ~5% EQE and maintained the same value at higher current densities. Previous work has shown that annealing the AlGaN cap layer reduces SRH-related nonradiative recombination . The LED with ∆T = 100 °C demonstrated lower EQE compared to other LEDs, which is most likely caused by thermal damage of the QWs during the relatively high GaN barrier growth temperature.
The forward voltage decreased when the growth temperature of GaN barrier increased, as illustrated in Fig. 5(b). The series resistances measured between 20 and 40 A/cm2 were 26.6, 25.7, 20.5 and 19.1 Ω for samples with ∆T = 0 °C, 50 °C, 75 °C and 100 °C, respectively. The improvement in forwarded voltage as ∆T increased is most likely due to the improved layer quality and reduced resistivity when the GaN barrier layers are grown at higher temperatures.
Additionally, we measured the WPE for all samples. WPE is defined as the ratio of emitted optical power to injected electrical power (WPE = P/IV), where P is the output power, I is the current and V is the forward voltage. The sample with ∆T = 75 °C exhibited the highest efficiency compared to the other LEDs due to higher output power, as shown in Fig. 5 (c).
The peak EL spectra at 20 A/cm2 for these devices ranged between 540 nm and 543 nm, as shown in Fig. 6(d). We note that the sample with ∆T = 0 °C has a sizable shoulder at ~420 nm. This is most likely due to the higher density of V-defects within the active region, resulting in the growth of QWs on the sidewalls of the V-defects that have different thickness and indium composition than the c-plane QWs [20–31].
For further characterization, we investigated the effect of different annealing temperatures on the morphology of the AlGaN cap layer. A series of SQW samples was grown by varying the AlGaN cap annealing temperature. The growth was interrupted after annealing the AlGaN cap layer and before growth of the HT GaN barrier. Like the previous experiment, four different temperatures were considered: ∆T = 0 °C, 50 °C, 75 °C and 100 °C. An Asylum MFP-3D atomic force microscope (AFM) was used in tapping mode to characterize the surface morphology. Figure 6 shows the AFM images of the growth surface for all samples with different AlGaN cap annealing temperatures. The root-mean-squared (RMS) roughness was 2.72 nm, 1.95 nm, 1.25 nm and 1.52 nm for the samples with ∆T = 0°C, 50°C, 75°C and 100°C, respectively. Increasing ∆T from 0°C to 75°C resulted in a significant improvement in the surface morphology of the AlGaN cap layer. This may explain the higher device performance when the ∆T of the HT GaN barrier increased, as illustrated in Fig. 5. Additionally, previous work showed that increasing the growth temperature of AlGaN layers in the active region, reduces the point defect density and improves device efficiency .
The sample with ∆T = 75 °C was further examined by high angle annular dark field scanning transmission electron microscopy (HAADF-STEM) and atom probe tomography (APT). HAADF-STEM images were taken at 300 kV using an FEI Titan FEG microscope. A Cameca Local Electrode Atom Probe 3000X HR was used for the APT measurements. A Nd:YAG laser (532 nm second harmonic and 120 ps pulse width) was pulsed at 200 kHz with a pulse energy of 0.1 nJ for the APT experiments. The STEM and APT samples were prepared by focused ion beam (FIB) with a Helios 600 Dual Beam instrument.
The APT and HAADF-STEM image in Fig. 7 shows a cross section of the LED structure, which distinctly show the layers of the active region. The STEM image in Fig. 7(b) shows the heterointerfaces between the AlGaN cap layers (darker contrast) and the InGaN QWs (lighter contrast). From the STEM image, the layer thicknesses of the InGaN QWs, Al0.30Ga0.70N cap layers, and GaN barriers were 3.3, 1.9 and 10 nm, respectively. Figure 7(a) shows the APT data as a reconstructed atom map of the active region. The In and Al ions are shown as purple and blue dots that indicate the location of the InGaN and AlGaN layers, respectively. We observed that the AlGaN cap layers are thicker than InGaN QWs, which contradicts the thicknesses of the layers from the STEM image. This can be attributed to the different evaporation behaviors of In, Ga, and Al atoms, which sometimes results in inaccurate thickness measurements by APT .
3.3 LEDs with optimized AlGaN cap thickness and annealing temperature
A final green LED structure was grown that incorporates the optimized AlGaN cap layer thickness and annealing conditions. The structure of the device was similar to the LED with a 2 nm AlGaN cap and ∆T = 75 °C, but with minor modifications to the EBL and p-GaN thickness. A 10 nm Mg-doped p-Al0.15Ga0.87N EBL (instead of a UID Al0.20Ga0.80N EBL) was grown on top of the last GaN barrier of the MQW followed by a 100 nm Mg-doped p-GaN layer (instead of a 250 nm p-GaN layer). For improved light extraction efficiency, the LEDs were packaged in a transparent vertical stand geometry .
Figure 8(a) shows the dependence of light output power and forward voltage on current density and the inset shows an optical micrograph of the packaged green LED. At 20 A/cm2, the output power and forward voltage were 13.8 mW and 3.5 V, respectively. Figure 8(b) shows the dependence of EQE and WPE of the device on current density and the inset shows the EL peak wavelength dependence on current density. The peak EQE was 40.7% at 3 A/cm2. At 20 A/cm2, the EQE and peak wavelength were 29.1% and 529 nm, respectively. The FWHM of the LED was only 33 nm, indicating that the QWs had minimal potential profile fluctuations during the growth of the GaN barriers at higher temperatures.
In summary, we have investigated the effects of an AlGaN cap layer on the active region of green c-plane LEDs. The active region consisted of InGaN QWs combined with AlGaN cap layers and high temperature GaN barriers. We analyzed, using experiments and simulations, the effect of AlGaN cap thickness on green LED optical and electrical properties. From this analysis, we concluded that employing 2 nm AlGaN cap thickness is optimal to maximize the EQE, reduce the forward voltage, and minimize the EL FWHM. However, increasing the AlGaN cap thickness further increases the electric field within the active region and reduces the electron-hole wavefunction overlap, which increases droop. We also investigated the effect of varying the annealing temperature of the AlGaN layer and growth temperature of the GaN barrier on the device performance. Annealing the AlGaN cap layer prior the growth of the GaN barrier results in a significant improvement due to the reduction of SRH nonradiative recombination. Optimized AlGaN cap growth conditions within the active region resulted in high efficiency green LEDs with a peak EQE of 40.7% at 3 A/cm2. At 20 A/cm2, the output power, emission wavelength, forward voltage and EQE were 13.8 mW, 529.3 nm, 3.5 V and 29.5%, respectively.
King Abdulaziz City for Science and Technology (KACST) Technology Innovations Center (TIC) program and the KACST-KAUST-UCSB Solid State Lighting Program.
Additional support was provided by the Solid State Lighting and Energy Electronics Center (SSLEEC) at UCSB. A portion of this work was done in the UCSB nanofabrication facility, part of the NSF NNIN network (ECS-0335765), as well as the UCSB MRL, which is supported by the NSF MRSEC program (DMR-1121053).
References and links
1. S. Nakamura, T. Mukai, and M. Senoh, “P-GaN / N-InGaN / N-GaN Double-Heterostructure Blue-Light-Emitting Diodes,” Jpn. J. Appl. Phys. 32(2), L8–L11 (1993). [CrossRef]
3. S. Pimputkar, J. S. Speck, S. P. DenBaars, and S. Nakamura, “Prospects for LED lighting,” Nat. Photonics 3(4), 180–182 (2009). [CrossRef]
4. K. A. Bulashevich, A. V. Kulik, and S. Y. Karpov, “Optimal ways of colour mixing for high-quality white-light LED sources,” Phys. Status Solidi 212(5), 914–919 (2015). [CrossRef]
5. S. Kimura, H. Yoshida, K. Uesugi, T. Ito, A. Okada, and S. Nunoue, “Performance enhancement of blue light-emitting diodes with InGaN/GaN multi-quantum wells grown on Si substrates by inserting thin AlGaN interlayers,” J. Appl. Phys. 120(11), 113104 (2016). [CrossRef]
6. Y. Narukawa, M. Ichikawa, D. Sanga, M. Sano, and T. Mukai, “White light emitting diodes with super-high luminous efficacy,” J. Phys. D 43(35), 354002 (2010). [CrossRef]
7. C. A. Hurni, A. David, M. J. Cich, R. I. Aldaz, B. Ellis, K. Huang, A. Tyagi, R. A. Delille, M. D. Craven, F. M. Steranka, and M. R. Krames, “Bulk GaN flip-chip violet light-emitting diodes with optimized efficiency for high-power operation,” Appl. Phys. Lett. 106(3), 31101 (2015). [CrossRef]
8. M. R. Krames, O. B. Shchekin, R. Mueller-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]
9. M. H. Crawford, “LEDs for solid-state lighting: Performance challenges and recent advances,” IEEE J. Sel. Top. Quantum Electron. 15(4), 1028–1040 (2009). [CrossRef]
10. J. T. Chen, U. Forsberg, and E. Janzén, “Impact of residual carbon on two-dimensional electron gas properties in AlxGa1-xN/GaN heterostructure,” Appl. Phys. Lett. 102(19), 193506 (2013). [CrossRef] [PubMed]
11. N. Okada, K. Tadatomo, K. Yamane, H. Mangyo, Y. Kobayashi, H. Ono, K. Ikenaga, Y. Yano, and K. Matsumoto, “Performance of InGaN / GaN light-emitting diodes grown using NH3 with oxygen-containing impurities with oxygen-containing impurities,” Jpn. J. Appl. Phys. 53(8), 81001 (2014). [CrossRef]
12. E. Kioupakis, Q. Yan, and C. G. Van De Walle, “Interplay of polarization fields and Auger recombination in the efficiency droop of nitride light-emitting diodes,” Appl. Phys. Lett. 101(23), 231107 (2012). [CrossRef]
13. S. Chichibu, T. Azuhata, T. Sota, and S. Nakamura, “Recombination of Localized Excitons in InGaN Single- and Multiquantum-Well Structures,” MRS Proc. 449, 653–658 (1996).
14. S. F. Chichibu, A. Uedono, T. Onuma, B. A. Haskell, A. Chakraborty, T. Koyama, P. T. Fini, S. Keller, S. P. Denbaars, J. S. Speck, U. K. Mishra, S. Nakamura, S. Yamaguchi, S. Kamiyama, H. Amano, I. Akasaki, J. Han, and T. Sota, “Origin of defect-insensitive emission probability in In-containing (Al,In,Ga)N alloy semiconductors,” Nat. Mater. 5(10), 810–816 (2006). [CrossRef] [PubMed]
15. A. E. Romanov, T. J. Baker, S. Nakamura, and J. S. Speck, “Strain-induced polarization in wurtzite III-nitride semipolar layers,” J. Appl. Phys. 100(2), 23522 (2006). [CrossRef]
16. F. Bernardini and V. Fiorentini, “Spontaneous versus Piezoelectric Polarization in III-V Nitrides: Conceptual Aspects and Practical Consequences,” Phys. Status Solidi 216(1), 391–398 (1999). [CrossRef]
17. F. Nippert, S. Y. Karpov, G. Callsen, B. Galler, T. Kure, C. Nenstiel, M. R. Wagner, M. Straßburg, H.-J. Lugauer, and A. Hoffmann, “Temperature-dependent recombination coefficients in InGaN light-emitting diodes: Hole localization, Auger processes, and the green gap,” Appl. Phys. Lett. 109(16), 161103 (2016). [CrossRef]
18. H. Masui, H. Asamizu, T. Melo, H. Yamada, K. Iso, S. C. Cruz, S. Nakamura, and S. P. DenBaars, “Effects of piezoelectric fields on optoelectronic properties of InGaN/GaN quantum-well light-emitting diodes prepared on nonpolar (1 0 −1 0) and semipolar (1 1 −2 2) orientations,” J. Phys. D Appl. Phys. 42(13), 135106 (2009). [CrossRef]
19. N. Grandjean, B. Damilano, S. Dalmasso, M. Leroux, M. Laügt, and J. Massies, “Built-in electric-field effects in wurtzite AlGaN/GaN quantum wells,” J. Appl. Phys. 86(7), 3714–3720 (1999). [CrossRef]
20. X. H. Wu, C. R. Elsass, A. Abare, M. Mack, S. Keller, P. M. Petroff, S. P. DenBaars, J. S. Speck, and S. J. Rosner, “Structural origin of V-defects and correlation with localized excitonic centers in InGaN/GaN multiple quantum wells,” Appl. Phys. Lett. 72(6), 692–694 (1998). [CrossRef]
21. D. D. Koleske, S. R. Lee, M. H. Crawford, K. C. Cross, M. E. Coltrin, and J. M. Kempisty, “Connection between GaN and InGaN growth mechanisms and surface morphology,” J. Cryst. Growth 391, 85–96 (2014). [CrossRef]
22. Y.-D. Lin, S. Yamamoto, C.-Y. Huang, C.-L. Hsiung, F. Wu, K. Fujito, H. Ohta, J. S. Speck, S. P. DenBaars, and S. Nakamura, “High quality InGaN/AlGaN multiple quantum wells for semipolar InGaN green laser diodes,” Appl. Phys. Express 3(8), 82001 (2010). [CrossRef]
23. 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]
24. T. Shioda, H. Yoshida, K. Tachibana, N. Sugiyama, and S. Nunoue, “Enhanced light output power of green LEDs employing AlGaN interlayer in InGaN/GaN MQW structure on sapphire (0001) substrate,” Phys. Status Solidi 209(3), 473–476 (2012). [CrossRef]
25. A. I. Alhassan, R. M. Farrell, B. Saifaddin, A. Mughal, F. Wu, S. P. DenBaars, S. Nakamura, and J. S. Speck, “High luminous efficacy green light-emitting diodes with AlGaN cap layer,” Opt. Express 24(16), 17868–17873 (2016). [CrossRef] [PubMed]
26. T. Doi, Y. Honda, M. Yamaguchi, and H. Amano, “Strain-Compensated Effect on the Growth of InGaN / AlGaN Multi-Quantum Well by Metalorganic Vapor Phase Epitaxy,” Jpn. J. Appl. Phys. 52(8S), 08JB14 (2013). [CrossRef]
27. K. Lekhal, B. Damilano, H. T. Ngo, D. Rosales, P. De Mierry, S. Hussain, P. Vennéguès, and B. Gil, “Strain-compensated (Ga,In)N/(Al,Ga)N/GaN multiple quantum wells for improved yellow/amber light emission,” Appl. Phys. Lett. 106(14), 142101 (2015). [CrossRef]
28. D. D. Koleske, A. J. Fischer, B. N. Bryant, P. G. Kotula, and J. J. Wierer, “On the increased efficiency in InGaN-based multiple quantum wells emitting at 530–590nm with AlGaN interlayers,” J. Cryst. Growth 415, 57–64 (2015). [CrossRef]
29. A. Tian, J. Liu, L. Zhang, Z. Li, M. Ikeda, S. Zhang, D. Li, P. Wen, F. Zhang, Y. Cheng, X. Fan, and H. Yang, “Green laser diodes with low threshold current density via interface engineering of InGaN/GaN quantum well active region,” Opt. Express 25(1), 415–421 (2017). [CrossRef] [PubMed]
30. A. David and M. J. Grundmann, “Influence of polarization fields on carrier lifetime and recombination rates in InGaN-based light-emitting diodes,” Appl. Phys. Lett. 97(3), 33501 (2010). [CrossRef]
31. S. J. Leem, Y. C. Shin, E. H. Kim, C. M. Kim, B. G. Lee, Y. Moon, I. H. Lee, and T. G. Kim, “Optimization of InGaN/GaN multiple quantum well layers by a two-step varied-barrier-growth temperature method,” Semicond. Sci. Technol. 23(12), 125039 (2008). [CrossRef]
32. A. Armstrong, A. A. Allerman, T. A. Henry, and M. H. Crawford, “Influence of growth temperature on AlGaN multiquantum well point defect incorporation and photoluminescence efficiency,” Appl. Phys. Lett. 98(16), 162110 (2011). [CrossRef]
33. L. Rigutti, L. Mancini, D. Hernández-Maldonado, W. Lefebvre, E. Giraud, R. Butté, J. F. Carlin, N. Grandjean, D. Blavette, and F. Vurpillot, “Statistical correction of atom probe tomography data of semiconductor alloys combined with optical spectroscopy: The case of Al0.25Ga0.75N,” J. Appl. Phys. 119(10), 105704 (2016). [CrossRef]
34. C. C. Pan, I. Koslow, J. Sonoda, H. Ohta, J. S. Ha, S. Nakamura, and S. P. DenBaars, “Vertical stand transparent light-emitting diode architecture for high-efficiency and high-power light-emitting diodes,” Jpn. J. Appl. Phys. 49(8), 80210 (2010). [CrossRef]