Open Access
Add to BibSonomyAbstract
The epitaxial structure design of low-temperature barriers has been adopted to promote strain relaxation in multiple quantum well (MQWs) and achieve high-efficient GaN-based light-emitting diodes (LEDs). With these barriers, the relaxation value of wells increases from 0 to 4.59%. The strain-relaxed mechanism of low-temperature barriers is also discussed. The LED chip with the barriers grown at the TMIn flow of 75 sccm and the growth temperature of 830 °C has an optimal strain relaxation value of 1.53% in wells, and exhibits the largest light output power of 63.83 mW at the injection current of 65 mA, which is higher than that of conventional LED (51.89 mW) by 23%. In-depth studies reveal that the optimal low-temperature barriers remarkably promote the strain relaxation in wells without forming large density of crystalline defects. This achievement of high-efficiency LEDs sheds light on the future solid-state lighting applications.
© 2016 Optical Society of America
I. Introduction
In recent years, GaN-based light-emitting diodes (LEDs) have achieved significant progresses [1–4]. A further penetration of LEDs in solid-state lighting (SSL) market requires the cost-per-lumen of packaged LEDs to rapidly decrease from the present average values of 2.3 $/klm to 0.3 $/klm by 2020 [5]. Improving the luminous efficiency for LEDs is very important in regard of the energy-efficiency and cost-efficiency for packaging processes by easing the challenge of heat dissipation, which is the best choice to cope with the large demands from SSL market. However, the state-of-the-art LEDs are suffering from efficiency droop at high injection current, which has not been fully resolved so far [6–8]. Efficiency droop is caused by various non-radiative carrier loss mechanisms, which become dominant at large injection current, such as polarization sheet charges [7, 9–11], electron leakage [7, 12, 13], poor hole injection [9, 14, 15], carrier delocalization [16, 17], Auger recombination [18, 19], current crowding [20] etc. Among these non-radiative carrier loss mechanisms, polarization sheet charge is a key issue. The polarization sheet charges at the quantum well (QW)/quantum barrier (QB) interfaces will cause quantum confined Stark effect (QCSE), separating the electron-hole wavefunction [21, 22] and facilitating electron leakage from the active region into the p-side of the LEDs [7, 9–11]. At present, LEDs are mainly prepared along the [0001] c-axis of hexagonal GaN, where strong spontaneous and strain-induced piezoelectric polarization fields exist in multiple quantum wells (MQWs). To address this problem, approaches based on material modification and epitaxial structure design have already been proposed [1, 6, 8, 13]. For example, growing non- or semi-polar GaN films is considered as an effective way to eliminate the polarization sheet charges. So far, non- or semi-polar LEDs have been fabricated on various substrates [23–27]. However, in most cases, the crystalline quality of non- or semi-polar LEDs are not comparable with that of the c-plane LEDs, which significantly deteriorates the device performance of non- or semi-polar LEDs [6, 12, 26]. In this regard, the approach based on epitaxial structure design seems to be more promising to solve the efficiency droop problem, because it can reduce the strain in c-plane LEDs to alleviate QCSE without deteriorating crystalline quality. Several approaches have been proposed to reduce the strain in MQWs by inserting various layers, such as InGaN underlying layer [28], short-period superlattice layer [29,30], step stage superlattice layer [31], etc. These studies have demonstrated that such insertion layers can act as the lattice-mismatched buffer between the underlying n-GaN and overlying MQWs, to release the residual strain in MQWs layer. However, such insertion layers have been becoming more and more complicated, which will increase the difficulty and cost to fabricate high-performance devices. Therefore, a simple solution to effectively reduce the strain in MQWs is desperately demanded.
In this work, we report a simple approach to significantly reduce the biaxial compressive strain in MQWs, i.e., growing MQWs with low-temperature barriers over a pre-strained MQWs. The strain condition, structural properties, optoelectronic properties of such LEDs have been carefully investigated to obtain the optimal design parameters. This achievement is of paramount importance for the future SSL applications.
II. Experimental procedure
The LED samples used in this study were all grown on 2-in c-face (0001) patterned sapphire substrates by a Veeco K465i metal-organic chemical vapor deposition (MOCVD) system. As shown in Fig. 1, the LED structure consisted of a 200-nm-thick GaN buffer layer, a 4-μm-thick unintentionally doped GaN layer, a 3-μm-thick Si-doped GaN layer, a 5-period 2-/6-nm-thick Si-doped In0.13Ga0.87N/GaN pre-strained MQWs (named as first MQW pairs), 7-period 3-/12-nm-thick In0.13Ga0.87N/GaN MQWs (named as second MQW pairs), a 20-nm-thick Mg-doped Al0.15Ga0.85N layer and a 150-nm-thick Mg-doped GaN layer. During the growth, trimethylaluminum (TMAl), trimethylgallium (TMGa), trimethyIn (TMIn), and ammonia (NH3) were used as aluminum (Al), gallium (Ga), indium (In), and nitrogen (N) sources, respectively. Bicyclopentadienyl magnesium (Cp2Mg) and silane (SiH4) were used as the p- and n-type doping sources, respectively. Nitrogen (N2) was used as the carrier gas during the growth of MQWs and hydrogen (H2) were used for the remaining layers. Five LEDs denoted as LEDs A-E were grown under the same condition except for growing the 7 barriers of the second MQW pairs. The 7 barriers of the second MQW pairs for LED A were grown without SiH4 flow at the growth temperature of 840 °C. The growth temperature of the 7 barriers of the second MQW pairs for LEDs B-E were set as 840, 830, 820, 810 °C, respectively, with the same TMIn flow of 75 sccm. The 5 barriers of the first MQW pairs were grown with the temperature of 840 °C. The growth temperature of wells for all the LEDs are 750 °C.
After the epitaxial growth, standard chip processing has been carried out for chip fabrication. The Ga-polarized surfaces of the as-grown LED wafers were partially ICP etched until the n-type GaN layer was exposed, followed by a 250-nm-thick transparent conductive indium tin oxide (ITO) layer deposited by electron beam evaporation. Cr/Pt/Au contacts (50/150/2000 nm) are subsequently evaporated as the n- and p-type electrodes, respectively. Finally, the LED wafers are divided into chips with an area of 750 × 220 um2.
The structural properties and strain condition of the LED wafers were characterized by high-resolution X-ray diffraction (HRXRD) (Bruker D8 X-ray diffractometer with Cu Kα1 X-ray source, λ = 0.15406 nm). Cross-sectional high-resolution transmission electron microscopy (HRTEM) (JEOL 3000F field emission gun TEM, working voltage = 300 kV, point to point resolution = 0.17 nm) was used to study the structural properties of the MQWs. The electroluminescence (EL) properties of the LEDs were investigated using the GAMMA Scientific GS-1190 RadoMA-Lite KEITHLEY 2400 system. A 405 nm laser (Y-Wafer GS4-GaN-R-405) was deployed as exciting source with an output power of 20 mW to study the photoluminescence (PL) properties of MQWs.
III. Results and discussion
To investigate the strain condition in the MQWs of the as-grown LEDs, reciprocal space mappings (RSMs) in the vicinity of the GaN (10-15) plane have been carried out, as shown in Fig. 2. We can observe that from LEDs A to E, the satellite peaks of MQWs gradually split into two sets in reciprocal space, resulting from the increasing average In composition of the MQWs. For LED A, the satellite peaks of the 12-period MQWs align GaN peak at the perpendicular direction, Fig. 2(a), indicating that the MQWs are fully stressed. However, for LEDs B-E, only the first MQW pairs align GaN peak at the perpendicular direction and the slightly tilted alignment of the second MQW pairs satellite peaks is clearly visible up to the 2nd order, revealing that the first MQW pairs are fully stressed while the second MQW pairs are partially relaxed [17, 32–35]. Moreover, from LEDs A to E, the satellite peak alignment of the second MQW pairs is becoming more and more tilted, indicating the enhancement of strain relaxation by using low-temperature barriers. In order to quantitatively compare the strain relaxation within the second MQW pairs for all LEDs, the layer relaxation value within wells, R, is adopted. The method to obtain R has been demonstrated by many previous studies [36–39]. This method requires for the structural parameters of MQWs, such as the barriers and wells’ thicknesses and In compositions.
Fig. 2 RSMs for the GaN (10-15) diffraction of LEDs (a)A, (b) B, (c) C, (d) D, and (e) E, respectively.
The structural properties of these five LEDs were studied by HRXRD ω-2θ symmetrical (0002) scans with the simulation software LEPTOS using genetic theory. Figure 3(a) shows the ω-2θ symmetrical (0002) scan for LED C along the growth direction. The substrate peak originates from the (0002) plane of the underlying GaN and the sharp satellite peaks come from the MQWs. Clear and sharp Pendellösung fringes can be observed. As we know, Pendellösung fringes arise from interference between coherent X-ray waves reflected within multilayer structure, typically in MQWs [40, 41]. Any composition fluctuation or interfacial imperfection will decrease the phase coherence and eliminate the Pendellösung fringes. In other words, the appearance of pronounced and clear Pendellösung fringes is the symbol of outstanding crystalline quality for MQWs [39, 41]. Evidently, the clear and sharp satellite peaks and Pendellösung fringes observed in LED C suggest that as-grown MQWs have abrupt InGaN/GaN or InGaN/InxGa1-xN interfaces with excellent layer periodicity, and the composition and the thickness of each barrier and well layer are well controlled [42]. From the inset of Fig. 3(a), we can note 10 clear fingers between the 0th and −1st order satellite peaks, indicating that the period of MQWs is 12 [43], which is consistent with the value we have designed during the growth.
Fig. 3 (a) HRXRD ω-2θ symmetrical (0002) scan of the MQWs of LED C with a simulation using a genetic theory; (b) HRXRD ω-2θ symmetrical (0002) scan of the MQWs for LEDs A to E; (c) HRTEM image and (d) high-magnification HRTEM image about MQWs region of LED C.
The structural parameters of the first and the second MQW pairs are obtained by fitting ω-2θ symmetrical (0002) scan. As shown in Fig. 3(a), the fitted simulation curve agrees well with the experiment curve. From the fitted simulation results, we can obtain that the wells and barriers’ thicknesses for the first MQW pairs are 2.03, 5.98 nm, respectively, and the In composition of the wells is 13.11%. The thicknesses of the wells and barriers for the second MQW pairs are also calculated as 2.97, 12.02 nm, respectively, and the corresponding In compositions are 13.11%, and 0.93%, respectively. The structural parameters of LEDs A, B, D, and E are obtained in the same way. As shown in Table 1, the structural parameters about the second MQW pairs of the five LEDs are almost the same except that the In composition of barriers increases with the decrease in growth temperature. Based on these structural parameters, the strain relaxation value within MQWs, R, can be calculated and is listed in Table 1.
Table 1. Structural parameters of the second MQW pairs, the FWHMs of −1st satellite peak and the relaxation values of wells, R, for LEDs A-E.
Figure 3(b) shows the ω-2θ symmetrical (0002) scans for the MQWs in the five LEDs. Additional structural properties can be deduced by the satellite peaks and Pendellösung fringes between these LEDs. First, it’s observed in Fig. 3(b) that the Pendellösung fringes between the 0th and the −1st satellite peaks for LEDs A to E gradually fade out. Moreover, the + 3rd satellite peak of LED E is hardly observable and its Pendellösung fringes also become obscure. These results reveal that the well/barrier interfaces of MQWs gradually become rough with an increasing relaxation value [41]. Second, the full widths at half maximum (FWHMs) of −1st satellite peaks for LEDs A to E increases from 89.78 to 97.77 arcsec, as shown in Table 1. The obviously observed satellite peak broadening for LEDs D and E also suggests that more crystalline defects may be induced into the MQWs when the In composition of barriers increases [41]. Actually, the TEM measurement is also used to study the InGaN/GaN MQWs. Figure 3(c) shows the TEM image about the MQWs region of LED C, where the first and the second MQW pairs can be clearly observed. The high-magnification TEM image of the second MQW pairs proves the high accuracy of the fitted thicknesses, Fig. 3(d).
The stress relaxation mechanism of the wells promoted by low-temperature barriers lies in the following two aspects. Stress in the as-grown MQWs is composed by internal stress and thermal stress [44–46]. On the one hand, the internal stress is mainly determined by the lattice mismatch between the two neighboring layers [47]. Because the lattice parameter of InN is larger than that of GaN, the lattice mismatch between the stacked In0.13Ga0.87N wells and InxGa1-xN barriers (x<0.13) induces biaxial compressive internal stress into the wells. When the barriers of the second MQW pairs are grown at lower growth temperature, the In atom incorporation into barriers is facilitated, and therefore the lattice mismatch between wells and barriers is reduced, which is significant to reduce the biaxial compressive internal stress in the wells. On the other hand, the thermal stress is induced by the difference in the thermal expansion of wells and barriers when cooling down from the growth temperature to the room temperature [48, 49]. The coefficients of thermal expansion (CTEs) of GaN and InN are 5.6 × 10−6 [50], and 2.7 × 10−6 K−1 [51], respectively. A well is surrounded by two barriers, and a smaller CTE of the well will produce biaxial compressive thermal stress at a lower temperature in comparison with its growth temperature, and produce biaxial tensile thermal stress at a higher temperature. Therefore, when the barriers of the second MQW pairs are grown at lower temperature, the thermal stress in the wells is favorable to be reduced.
Figure 4 exhibits the stress changing trend during the growth process and the cooling down process. When the temperature remains at 750 °C (the growth temperature of the wells), there is no thermal stress in the In0.13Ga0.87N well. As the temperature increases during the growth of barrier, EBL, and p-GaN (upper layer), In0.13Ga0.87N well suffers from biaxial tensile thermal stress, and this stress increases with the increasing temperature until the growth of upper layer is finished. Subsequently, the epitaxial wafer is gradually cooled down, and the biaxial tensile thermal stress decreases correspondingly and finally turns into biaxial compressive thermal stress. Its corresponding force situation at room temperature is shown in Fig. 4(b). The well is applied by two biaxial compressive thermal stresses produced by two stacked barriers. As a pair of interaction forces, the biaxial compressive thermal stress applied in well is equal to the biaxial tensile thermal stress applied in barriers. When the growth temperature of barriers decreases, the barriers’ In composition increases and its CTE becomes approximate to the well’s. Moreover, the difference of the growth temperature and the room temperature also reduces. These two factors result in a smaller biaxial tensile thermal stress applied in barriers. That is to say, the well’s biaxial compressive thermal stress also decreases.
Fig. 4 (a) The stress changing trend of In0.13Ga0.87N well during the growth process and cooling down process and (b) its force situation at room temperature.
Strain condition is a critical influential factor to the properties of MQWs. Figures 5(a) and (b) exhibits the EL spectra of LEDs A and E. The MQWs of LED A are almost fully stressed while the MQWs of LED E are the most stress-relaxed. We observe a great blue shift of 1.7 nm in peak EL wavelength of LED A when the injected current increases from 2 to 9 mA. In contrast, the blue shift in the EL peak wavelength of LED E is only 0.5 nm at the same range of injection current. Generally, blue shift in EL peak wavelength for the GaN-based MQWs LED at small injection current is ascribed to the screening of the piezoelectric filed [52–54]. This phenomenon can be explained as the following. Originally, as the QWs are strained, their transition energy is smaller than that of the unstrained QWs because the band alignment of QWs is tilted by piezoelectric field, so-called QCSE [52, 53, 55]. When the LED is operated with a forward current, the piezoelectric fields in the InGaN strained QWs are screened by free carriers injected into the QWs, thus weakening the QCSE. Increasing the forward current further weakens the QSCE and increases the transition energy. This is why blue shift occurs [53, 56]. For this reason, blue shift of the EL peak wavelength at low injection current is an indication of QCSE and also for the strain condition.
Fig. 5 EL spectra of LEDs (a) A and (b) E, and (c) the blue shifts of EL peak wavelength for LEDs A to E with the injection current ranging from 2 to 9 mA.
Figure 5(c) shows the blue shifts of the EL peak wavelength for LEDs A-E, respectively. From LED A to E, the blue shift of EL peak wavelength with the injection current ranging from 2 to 9 mA decreases from 1.7 to 0.5 nm. The reduced blue shift of the EL peak wavelength with increasing current suggests the mitigation of QCSE. The change of QCSE is caused by the change of the stress applied in QWs. That is to say, through stress control by low-temperature barriers, the compressive strain in the wells reduces, which in return mitigates the QCSE. The result is consistent with the strain condition calculated from the RSMs for the (10-15) diffraction.
The mitigation of the QCSE in the wells is supposed to improve the luminous performance of LEDs. Figure 6(a) shows the light output power of LEDs A-E as the injection current increases from 0 to 65 mA. Thorough the introduction of low-temperature barriers, the gain on light output power by the mitigation of QCSE is very remarkable. LED C which has a medium relaxation value in the wells of 1.53%, owns the largest light output power of 63.83 mW, which is a 23% improvement compared to LED A (51.89 mW) at the injection current of 65 mA. Figure 6(b) shows the I-V characteristics for LEDs A to E. At the injection current of 20 mA, the forward voltages are 3.28, 3.30, 3.34, 3.37, 3.39V for LEDs A to E, respectively. The increasing electric resistance observed in Fig. 6(b) may be attributed to the degraded crystalline quality of MQWs for LEDs A to E. But the increment of the electric resistance is very little, and its influence on the performance of LED will be very limited. Figure 6(c) exhibits the wall-plug efficiency (WPE) for LEDs A to E with the injection current ranging from 0 to 65 mA. The WPE droops for LEDs A to E are 34.69%, 33.66%, 33.48%, 22.83% and 15.79%, respectively. From this result, we can see that the LED with larger relaxation in MQWs owns the weaker WPE droop, which is consistent with the mitigation of the QCSE observed in Fig. 5. However, as shown in Fig. 6(d), further increase on the relaxation value unexpectedly degrades the LED’s luminous performance. When the relaxation value is over 1.53%, LED’s light output power suffers distinct droop. At small injection current (I < 20 mA), the light output power of LED A is the largest among LEDs A, D, and E. At high injection current (I > 50 mA), the gain on LED E’s light output power is very small in comparison with LEDs B-D. These unexpected results reveal that negative effect is also brought in MQWs by low-temperature barriers.
Fig. 6 (a) Light output power vs current, (b) current vs voltage and (c) WPE vs current for LEDs A to E with the injection current ranging from 0 to 65 mA; (d) Relaxation value of the wells vs In composition of barriers and light output power vs In composition of barriers at the injection current of 65 mA.
Considering the HRXRD satellite peak broadening of LEDs D and E observed in Fig. 2(a), the negative effect is believed to be the crystalline degradation induced by strain relaxation. The light output power degradation resulted from crystalline defects can be explain by the enhanced non-radiative carrier recombination process. Competitively, radiative and non-radiative recombination processes all take place inside the LED active region, including Shockley-Read-Hall (SRH) recombination, radiative recombination and Auger recombination [6, 57, 58]. As non-radiative recombination processes, SRH and Auger recombination are associated with crystalline defects [6, 12, 57, 58]. Therefore, the increasing crystalline defects will enhance non-radiative recombination processes, weaken the performance improvement brought by the mitigation of QCSE, and finally result in the light output power degradation for LEDs D and E. In order to confirm the crystalline degradation in MQWs, room temperature PL spectra have been adopted. Figures 7(a) and 7(b) show the room-temperature PL spectra and the PL FWHMs of LEDs A-E. The blue shift of PL peak wavelength of LEDs A-E can be observed in Fig. 7(a) and implies the mitigation of QCSE, which is consistent with the EL measurement. However, the FWHMs of PL peaks, which are the symbol of crystalline quality of LEDs [59, 60], are quite different for these LEDs. It is found that the PL FWHM decreases as the relaxation value increases. Figure 7(b) exhibits that the minimum FWHM of 19.2 nm occurs in LED A, which is 8.6 nm smaller than the maximum FWHM of 27.8 nm in LED E. This reveals that more crystalline defects are induced by strain relaxation as the relaxation value increases. Such an observation agrees well with the HRXRD results shown in Fig. 3(b).
V. Conclusion
High-efficiency blue GaN-based LEDs have been successfully achieved by employing low-temperature barriers to control the strain condition of the wells. The LED which has a proper strain relaxation value of 1.53% in wells owns the largest light output power of 63.83 mW at the injection current of 65 mA, which is a 23% improvement compared to the conventional LED (51.89 mW). In order to clarify the mechanism behind the performance improvement, four LEDs that have seven low-temperature barriers which are grown at the temperature of 840, 830, 820, and 810 °C, are systematically investigated accompanied by the LED with normal GaN barriers (growth temperature = 840 °C). The HRXRD ω-2θ symmetrical (0002) scans of MQWs reveal that as the growth temperature decreases, the In composition of low-temperature barriers increases with the value of 0.60%, 0.93%, 1.28%, and 1.89%, respectively. The thickness of barriers is also calculated as about 12 nm, which is proven by the HRTEM observation. The RSMs in the vicinity of the GaN (10-15) plane reveal that wells are stress relaxed with the values of 0.96%, 1.53%, 2.88%, and 4.59%. The strain relaxation of the wells notably mitigates the QCSE, which is confirmed by the enlarged blue shift of EL peak wavelength at the injection current ranging from 2 to 9 mA. However, the LED with a larger relaxation value suffers unexpected power degradation. The broadening of satellite peaks in HRXRD ω-2θ symmetrical (0002) scans and the enlarged FWHM of PL spectra reveal that the origin of the power degradation is the increasing crystalline defects in wells with large strain relaxation. This work would be very helpful for the SSL application of LEDs in future.
Acknowledgments
This work is financially supported by National Science Fund for Excellent Young Scholars of China (No. 51422203), National Natural Science Foundation of China (No. 51002052, 51372001), Key Project in Science and Technology of Guangdong Province (Nos: 2014B010119001, 2014B010121004 and 2011A080801018), Distinguished Young Scientist Foundation of Guangdong Scientific Committee (No. S2013050013882), Strategic Special Funds for LEDs of Guangdong Province (Nos. 2011A081301010, 2011A081301012 and 2012A080302002), and Key Scientific Research Project in State Key Laboratory of Luminescent Materials and Devices (No: C7160050).
References and links
1. S. Nakamura and M. R. Krames, “History of gallium-nitride-based light-emitting diodes for illumination,” Proc. IEEE 101(10), 2211–2220 (2013). [CrossRef]
2. G. Li, W. Wang, W. Yang, and H. Wang, “Epitaxial growth of group III-nitride films by pulsed laser deposition and their use in the development of LED devices,” Surf. Sci. Rep. 70(3), 380–423 (2015). [CrossRef]
3. G. Li, W. Wang, W. Yang, Y. Lin, H. Wang, Z. Lin, and S. Zhou, “GaN-based light-emitting diodes on various substrates: a critical review,” Rep. Prog. Phys. 79(5), 056501 (2016). [CrossRef] [PubMed]
4. D. H. Kim, D. S. Shin, and J. Park, “Enhanced light extraction from GaN based light-emitting diodes using a hemispherical NiCoO lens,” Opt. Express 22(S4Suppl 4), A1071–A1078 (2014). [CrossRef] [PubMed]
5. United States Department of Energy, Solid-State Lighting R&D Plan, http://www.energy.gov/eere/ssl/downloads/solid-state-lighting-rd-plan (2015).
6. G. Verzellesi, D. Saguatti, M. Meneghini, F. Bertazzi, M. Goano, G. Meneghesso, and E. Zanoni, “Efficiency droop in InGaN/GaN blue light-emitting diodes: Physical mechanisms and remedies,” J. Appl. Phys. 114(7), 071101 (2013). [CrossRef]
7. M. Kim, M. F. Schubert, Q. Dai, J. K. Kim, E. F. Schubert, J. Piprek, and Y. Park, “Origin of efficiency droop in GaN-based light-emitting diodesm,” Appl. Phys. Lett. 91(18), 183507 (2007). [CrossRef]
8. W. Yang, W. Wang, Y. Lin, Z. Liu, S. Zhou, H. Qian, F. Gao, S. Zhang, and G. Li, “Deposition of nonpolar m-plane InGaN/GaN multiple quantum wells on LiGaO2(100) substrates,” J. Mater. Chem. C Mater. Opt. Electron. Devices 2(5), 801–805 (2014). [CrossRef]
9. I. V. Rozhansky and D. A. Zakheim, “Analysis of processes limiting quantum efficiency of AlGaInN LEDs at high pumping,” Phys. Status Solidi 204(1), 227–230 (2007). [CrossRef]
10. J. Xu, M. F. Schubert, A. N. Noemaun, D. Zhu, J. K. Kim, E. F. Schubert, M. H. Kim, H. J. Chung, S. Yoon, C. Sone, and Y. Park, “Reduction in efficiency droop, forward voltage, ideality factor, and wavelength shift in polarization-matched GaInN/GaInN multi-quantum-well light-emitting diodes,” Appl. Phys. Lett. 94(1), 011113 (2009). [CrossRef]
11. S. Ling, T. Lu, S. Chang, J. Chen, H. Kuo, and S. Wang, “Low efficiency droop in blue-green m-plane InGaN/GaN light emitting diodes,” Appl. Phys. Lett. 96(23), 231101 (2010). [CrossRef]
12. M. F. Schubert, S. Chhajed, J. K. Kim, E. F. Schubert, D. D. Koleske, M. H. Crawford, S. R. Lee, A. J. Fischer, G. Thaler, and M. A. Banas, “Effect of dislocation density on efficiency droop in GaInN/GaN light-emitting diodes,” Appl. Phys. Lett. 91(23), 231114 (2007). [CrossRef]
13. M. F. Schubert, J. Xu, J. K. Kim, E. F. Schubert, M. H. Kim, S. Yoon, S. M. Lee, C. Sone, T. Sakong, and Y. Park, “Polarization-matched GaInN/AlGaInN multi-quantum-well light-emitting diodes with reduced efficiency droop,” Appl. Phys. Lett. 93(4), 041102 (2008). [CrossRef]
14. I. A. Pope, P. M. Smowton, P. Blood, J. D. Thomson, M. J. Kappers, and C. J. Humphreys, “Carrier leakage in InGaN quantum well light-emitting diodes emitting at 480 nm,” Appl. Phys. Lett. 82(17), 2755–2757 (2003). [CrossRef]
15. J. Xie, X. Ni, Q. Fan, R. Shimada, U. Ozgur, and H. Morkoç, “On the efficiency droop in InGaN multiple quantum well blue light emitting diodes and its reduction with p-doped quantum well barriers,” Appl. Phys. Lett. 93(12), 121107 (2008). [CrossRef]
16. A. Y. Kim, W. Götz, D. A. Steigerwald, J. J. Wierer, N. F. Gardner, J. Sun, S. A. Stockman, P. S. Martin, M. R. Krames, and R. S. Kern, “Performance of high-power AlInGaN light emitting diodes,” Phys. Status Solidi 188, 15–21 (2001). [CrossRef]
17. W. K. Fong, K. K. Leung, and C. Surya, “Growth and characterization of GaN/InGaN multiple quantum wells on nanoscale epitaxial lateral overgrown layers,” Cryst. Growth Des. 11(6), 2091–2097 (2011). [CrossRef]
18. Y. C. Shen, G. O. Mueller, S. Watanabe, N. F. Gardner, A. Munkholm, and M. R. Krames, “Auger recombination in InGaN measured by photoluminescence,” Appl. Phys. Lett. 91(14), 141101 (2007). [CrossRef]
19. N. F. Gardner, G. O. Müller, Y. C. Shen, G. Chen, S. Watanabe, W. Götz, and M. R. Krames, “Blue-emitting InGaN-GaN double-heterostructure light-emitting diodes reaching maximum quantum efficiency above 200A/cm2,” Appl. Phys. Lett. 91(24), 243506 (2007). [CrossRef]
20. C. K. Li and Y. R. Wu, “Study on the current spreading effect and light extraction enhancement of vertical GaN/InGaN LEDs,” IEEE Trans. Electron Dev. 59(2), 400–407 (2012). [CrossRef]
21. 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), 033501 (2010). [CrossRef]
22. J. Hader, J. V. Moloney, B. Pasenow, S. W. Koch, M. Sabathil, N. Linder, and S. Lutgen, “On the importance of radiative and Auger losses in GaN-based quantum wells,” Appl. Phys. Lett. 92(26), 261103 (2008). [CrossRef]
23. Y. Jun Sun, O. Brandt, B. Jenichen, and K. H. Ploog, “In surface segregation in M-plane (In,Ga)N/GaN multiple quantum well structures,” Appl. Phys. Lett. 83(25), 5178–5180 (2003). [CrossRef]
24. M. S. A. J. Keun, “Growth and characterization of a-plane InGaN/GaN multiple quantum well LEDs grown on r-plane sapphire,” Semicond. Sci. Technol. 27(1), 015011 (2012). [CrossRef]
25. H. Jönen, H. Bremers, T. Langer, U. Rossow, and A. Hangleiter, “Large optical polarization anisotropy due to anisotropic in-plane strain in m-plane GaInN quantum well structures grown on m-plane 6H-SiC,” Appl. Phys. Lett. 100(15), 151905 (2012). [CrossRef]
26. F. Tendille, P. De Mierry, P. Vennéguès, S. Chenot, and M. Teisseire, “Defect reduction method in (11-22) semipolar GaN grown on patterned sapphire substrate by MOCVD: Toward heteroepitaxial semipolar GaN free of basal stacking faults,” J. Cryst. Growth 404, 177–183 (2014). [CrossRef]
27. J. F. Kaeding, M. Iza, H. Sato, S. P. Denbaars, J. S. Speck, and S. Nakamura, “Effect of substrate miscut on the direct growth of semipolar (10-1-1) GaN on (100) MgAl2O4 by metalorganic chemical vapor deposition,” Jpn. J. Appl. Phys. 45(21), L536–L538 (2006). [CrossRef]
28. Y. Kuo, J. Chang, M. Tsai, and S. Yen, “Advantages of blue InGaN multiple-quantum well light-emitting diodes with InGaN barriers,” Appl. Phys. Lett. 95(1), 011116 (2009). [CrossRef]
29. A. D. Bykhovski, B. L. Gelmont, and M. S. Shur, “Elastic strain relaxation and piezoeffect in GaN-AlN, GaN-AlGaN and GaN-InGaN superlattices,” J. Appl. Phys. 81(9), 6332–6338 (1997). [CrossRef]
30. 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]
31. 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]
32. R. M. Farrell, C. J. Neufeld, S. C. Cruz, J. R. Lang, M. Iza, S. Keller, S. Nakamura, S. P. DenBaars, U. K. Mishra, and J. S. Speck, “High quantum efficiency InGaN/GaN multiple quantum well solar cells with spectral response extending out to 520 nm,” Appl. Phys. Lett. 98(20), 201107 (2011). [CrossRef]
33. A. Ubukata, K. Ikenaga, N. Akutsu, A. Yamaguchi, K. Matsumoto, T. Yamazaki, and T. Egawa, “GaN growth on 150-mm-diameter (111) Si substrates,” J. Cryst. Growth 298, 198–201 (2007). [CrossRef]
34. S. R. Lee, D. D. Koleske, M. H. Crawford, and J. J. Wierer Jr., “Effect of interface grading and lateral thickness variation on x-ray diffraction by InGaN/GaN multiple quantum wells,” J. Cryst. Growth 355(1), 63–72 (2012). [CrossRef]
35. Y. Taniyasu, J. Carlin, A. Castiglia, R. Butté, and N. Grandjean, “Mg doping for p-type AlInN lattice-matched to GaN,” Appl. Phys. Lett. 101(8), 082113 (2012). [CrossRef]
36. H. Angerer, D. Brunner, F. Freudenberg, O. Ambacher, M. Stutzmann, R. Höpler, T. Metzger, E. Born, G. Dollinger, A. Bergmaier, S. Karsch, and H. J. Körner, “Determination of the Al mole fraction and the band gap bowing of epitaxial AlxGa1-xN films,” Appl. Phys. Lett. 71(11), 1504–1506 (1997). [CrossRef]
37. P. F. Fewster, N. L. Andrew, and C. T. Foxon, “Microstructure and composition analysis of group III nitrides by X-ray scattering,” J. Cryst. Growth 230(3-4), 398–404 (2001). [CrossRef]
38. X. Y. Liu, J. F. Fälth, T. G. Andersson, P. Holmström, P. Jänes, U. Ekenberg, and L. Thylén, “Structural and optical properties of GaN/AlN multiple quantum wells for intersubband applications,” J. Cryst. Growth 278(1-4), 397–401 (2005). [CrossRef]
39. W. Wang, H. Yang, and G. Li, “Growth and characterization of GaN-based LED wafers on La0.3Sr1.7AlTaO6 substrates,” J. Mater. Chem. C Mater. Opt. Electron. Devices 1(26), 4070–4077 (2013). [CrossRef]
40. Y. H. Cho, F. Fedler, R. J. Hauenstein, G. H. Park, J. J. Song, S. Keller, U. K. Mishra, and S. P. Denbaars, “High resolution x-ray analysis of pseudomorphic InGaN/GaN multiple quantum wells: Influence of Si doping concentration,” J. Appl. Phys. 85(5), 3006–3008 (1999). [CrossRef]
41. M. A. Moram and M. E. Vickers, “X-ray diffraction of III-nitrides,” Rep. Prog. Phys. 72(3), 1773–1774 (2009). [CrossRef]
42. S. Yoshida, T. Yokogawa, Y. Imai, S. Kimura, and O. Sakata, “Evidence of lattice tilt and slip in m-plane InGaN/GaN heterostructure,” Appl. Phys. Lett. 99(13), 131909 (2011). [CrossRef]
43. S. C. Jain, M. Willander, J. Narayan, and R. V. Overstraeten, “III-nitrides: Growth, characterization, and properties,” J. Appl. Phys. 87(3), 965–1006 (2000). [CrossRef]
44. K. S. Chen, X. Zhang, and S. Y. Lin, “Intrinsic stress generation and relaxation of plasma-enhanced chemical vapor deposited oxide during deposition and subsequent thermal cycling,” Thin Solid Films 434(1-2), 190–202 (2003). [CrossRef]
45. F. Spaepen, “Interfaces and stresses in thin films,” Acta Mater. 48(1), 31–42 (2000). [CrossRef]
46. H. Köstenbauer, G. A. Fontalvo, J. Keckes, and C. Mitterer, “Intrinsic stresses and stress relaxation in TiN/Ag multilayer coatings during thermal cycling,” Thin Solid Films 516(8), 1920–1924 (2008). [CrossRef]
47. F. O. Adurodija, H. Izumi, T. Ishihara, H. Yoshioka, and M. Motoyama, “Effects of stress on the structure of indium-tin-oxide thin films grown by pulsed laser deposition,” J. Mater. Sci. Mater. Electron. 12(1), 57–61 (2001). [CrossRef]
48. L. Zhang, K. Cheng, S. Degroote, M. Leys, M. Germain, and G. Borghs, “Strain effects in GaN epilayers grown on different substrates by metal organic vapor phase epitaxy,” J. Appl. Phys. 108(7), 073522 (2010). [CrossRef]
49. C. Röder, F. Lipski, F. Habel, G. Leibiger, M. Abendroth, C. Himcinschi, and J. Kortus, “Raman spectroscopic characterization of epitaxially grown GaN on sapphire,” J. Phys. D Appl. Phys. 46(28), 285302 (2013). [CrossRef]
50. L. Liu and J. H. Edgar, “Substrates for gallium nitride epitaxy,” Mater. Sci. Eng. Rep. 37(3), 61–127 (2002). [CrossRef]
51. K. Wang and R. R. Reeber, “Thermal expansion and elastic properties of InN,” Appl. Phys. Lett. 79(11), 1602–1604 (2001). [CrossRef]
52. J. H. Ryou, W. Lee, J. Limb, D. Yoo, J. P. Liu, R. D. Dupuis, Z. H. Wu, A. M. Fischer, and F. A. Ponce, “Control of quantum-confined Stark effect in InGaN/GaN multiple quantum well active region by p-type layer for III-nitride-based visible light emitting diodes,” Appl. Phys. Lett. 92(10), 101113 (2008). [CrossRef]
53. J. H. Ryou, P. D. Yoder, J. Liu, Z. Lochner, H. Kim, S. Choi, H. J. Kim, and R. D. Dupuis, “Control of quantum-confined stark effect in InGaN-based quantum wells,” IEEE J. Sel. Top. Quantum Electron. 15(4), 1080–1091 (2009). [CrossRef]
54. C. F. Huang, C. Y. Chen, C. F. Lu, and C. C. Yang, “Reduced injection current induced blueshift in an InGaN/GaN quantum-well light-emitting diode of prestrained growth,” Appl. Phys. Lett. 91(5), 051121 (2007). [CrossRef]
55. D. Miller, D. Chemla, T. Damen, A. Gossard, W. Wiegmann, T. Wood, and C. Burrus, “Band-edge electroabsorption in quantum well structures: the quantum-confined stark effect,” Phys. Rev. Lett. 53(22), 2173–2176 (1984). [CrossRef]
56. T. Takeuchi, S. Sota, M. Katsuragawa, M. Komori, H. Takeuchi, H. Amano, and I. Akasaki, “Quantum-confined Stark effect due to piezoelectric fields in GaInN strained quantum wells,” Jpn. J. Appl. Phys. 36(Part 2, No. 4A), L382–L385 (1997). [CrossRef]
57. Z. Liu, T. Wei, E. Guo, X. Yi, L. Wang, J. Wang, G. Wang, Y. Shi, I. Ferguson, and J. Li, “Efficiency droop in InGaN/GaN multiple-quantum-well blue light-emitting diodes grown on free-standing GaN substrate,” Appl. Phys. Lett. 99(9), 091104 (2011). [CrossRef]
58. J. Piprek, “Efficiency droop in nitride-based light-emitting diodes,” Phys. Status Solidi 207(10), 2217–2225 (2010). [CrossRef]
59. S. J. Chang, W. C. Lai, Y. K. Su, J. F. Chen, C. H. Liu, and U. H. Liaw, “InGaN-GaN multiquantum-well blue and green light-emitting diodes,” IEEE J. Sel. Top. Quantum Electron. 8(2), 278–283 (2002). [CrossRef]
60. C. L. Hsiao, H. C. Hsu, L. C. Chen, C. T. Wu, C. W. Chen, M. Chen, L. W. Tu, and K. H. Chen, “Photoluminescence spectroscopy of nearly defect-free InN microcrystals exhibiting nondegenerate semiconductor behaviors,” Appl. Phys. Lett. 91(18), 181912 (2007). [CrossRef]
