Open Access
Add to BibSonomy
Abstract
We demonstrate InGaN-based semipolar 560 nm micro-light-emitting diodes with 2.5% EQE on high-quality and low-defect-density (20-21) GaN templates grown on scalable and low-cost sapphire substrates. Through transmission electron microscopy observations, we discuss how the management of misfit dislocations and their confinement in areas away from the active light-emitting region is necessary for improving device performance. We also discuss how the patterning of semipolar GaN on sapphire influences material properties in terms of surface roughness and undesired faceting in addition to indium segregation at the proximity of defected areas.
© 2020 Optical Society of America under the terms of the OSA Open Access Publishing Agreement
1. Introduction
Micro-light-emitting-diodes (microLEDs) based on InGaN are likely candidates for the upcoming generation of display technologies (e.g. near-eye displays and head-up displays, etc.). MicroLEDs are advantageous compared to the current commercial displays (e.g. Organic-LEDs, LCDs) in high brightness and luminous efficiency, long operating lifetime and robustness [1,2]. High-resolution display applications that require ultra-small microLEDs, a phosphor-free solution for red, green and blue microLEDs is desired. Although high performance c-plane microLEDs have been developed and demonstrated, they still suffer a large polarization induced electric field at the heterointerfaces [3,4], and wavelength blue-shift with increasing current density, which is problematic for any display application. An alternative approach is growing microLEDs on semipolar crystallographic orientations to reduce the polarization induced electric fields, resulting in higher recombination rates and thus increased quantum efficiency. Indeed, efficient semipolar LEDs on freestanding GaN have been demonstrated [5–7], however, the use of semipolar GaN substrates for microLED production remains impractical, due to the high substrate price and scalability limitations to accommodate the large number of microLEDs needed for display applications. Semipolar GaN materials grown on patterned substrates that employ defect management strategies to reduce or eliminate the high density of basal stacking faults (BSFs) and threading dislocations (TDs) [8–10], which are inherent to semipolar GaN films grown on planar substrates are indeed a viable alternative to freestanding substrates. Such templates have enabled several demonstrations towards efficient light emitting devices on scalable and low-cost substrates while benefiting from the intrinsic advantages of semipolar growth planes [11–13]. In this manuscript, we discuss 560 nm microLEDs grown on low-defect-density (20-21) templates grown on (22-43) patterned sapphire substrates. We investigate material characteristics in-depth through atom probe tomography (APT), transmission electron microscopy (TEM), secondary ion mass spectroscopy (SIMS) and atomic force microscopy (AFM) to generate an understanding on the template behavior under typical device growth conditions. Properties of the 560 nm microLEDs having sizes ranging from 80 μm to 20 μm are also discussed and possible correlations with the material quality are presented.
2. Epitaxy and fabrication procedure
The (20-21) template, over which the LED device is grown on, was fabricated on a patterned (22-43) sapphire substrate (with 5 μm period), which involves growing GaN at an inclined c-axis (on the etched and inclined sapphire c-facet close to 75°) as shown in Fig. 1(a). The final TD density of the template is around 5 × 108 cm-2 with a low BSF density. Details on the template process, fabrication and structural and optical quality can be found in Ref. [10]. The subsequent LEDs were grown on such a 4-inch (20-21) GaN/sapphire template by metal-organic chemical vapor deposition (MOCVD) at atmospheric pressure. Trimethylgallium (TMG), triethylgallium (TEG), trimethylindium (TMI), ammonia (NH3), disilane (Si2H6) and bis(cyclopentadienyl)magnesium (Cp2Mg) (Cp2Mg) were used as precursors and the dopants. The structure of the epitaxial stack consisted of a 2.5 μm n-type GaN grown at 1100°C at a growth rate of ≈ 9 Å/s with silicon concentration [Si] of 8 x1018 atoms/cm3, a 30 period In0.02Ga0.98N / GaN superlattice grown at 950°C, a single 4 nm 26% InGaN quantum well (growth rate at 1 Å/s), a 15 nm p-type AlGaN electron-blocking layer (EBL), and finally a 150 nm p-type GaN layer with magnesium concentration [Mg] of 2 × 1020 atoms/cm3 capped with a 10 nm p+ highly Mg-doped GaN grown at 920°C.
Fig. 1. (a) Schematic illustration of the (20-21) template grown on patterned sapphire and (b) the epitaxial stack and microLED contacts grown on the same template.
The LEDs were fabricated in squares with an edge length ranging from 20μm to 80μm [14,15].
First, 110 nm indium-tin oxide (ITO) was deposited as a transparent and ohmic p-contact layer using e-beam evaporation. The microLED mesas were defined using photolithography and etching to the n-GaN layer using silicon tetrachloride in a reactive-ion etching (RIE) chamber. An omnidirectional reflector (ODR) consists of silicon dioxide and tantalum pentoxide, followed by aluminum oxide as the capping layer. 50 nm of silicon dioxide was blanket deposited by atomic layer deposition (ALD) as a sidewall passivation layer, and buffered hydrofluoric acid was used to open a window for metal deposition. The metal contact consisting of Al/Ni/Au (700/100/700 nm) was deposited using e-beam evaporation. Finally, the microLEDs were diced, packaged onto silver headers, encapsulated using Dow Corning OE-6650 resin having a refractive index of 1.54, and measured in a calibrated integrating sphere. The full device stack and fabrication is illustrated in Fig. 1(b).
3. Material characterization
3.1. Atom probe tomography
The indium distribution in the QW and superlattice were investigated using atom probe tomography (APT). A FEI Helios 600 dual beam FIB instrument was used for the preparation of the needle shaped samples following a standard procedure [16]. APT experiments were performed with a Cameca 3000X HR Local Electrode Atom Probe (LEAP). The detail of the APT analysis conditions and data reconstruction can be found in Ref. [17].
Figure 2(a) shows the APT global tip reconstruction showing the SQW and the bottom InGaN/GaN periods of the superlattice. A significant amount of In atoms can be observed above the SQW in the QB. Figure 2(b) is a 1D concentration profile extracted from the APT reconstruction in Fig. 2(a) and showing the In III site fraction measured in the SQW. The peak In value measured in the SQW is 0.26 ± 0.01. In diffusion is observed in the barrier layer with a progressive decrease from 0.10 ± 0.01 to a steady value of 0.01 ± 0.01. Similar In profiles in QWs and QBs were already reported for green/yellow emitting InGaN grown on the (11-22) semipolar plane [18,19], and can be in principle be reduced by re-designing the barrier layer into multi-step growth (LT-GaN followed by HT-GaN) [20]. A similar 1D concentration profile was measured for the InGaN/GaN superlattice and is shown in Fig. 2(c). 18 periods could be resolved before the APT tip fractured. The In fraction measured in the InGaN layers is around 0.02 ± 0.01 with some In still detected in between the InGaN layers. Figure 2(d) is a 2D side view of the In distribution in the SQW. As expected from a disordered alloy, some In rich and poor regions are observed in the well. In incorporation above the QW in the barrier layer can also be clearly resolved. The spatial extend of the 2D map is about 65 nm which is, compared to the size of the complete device, extremely small. However, in Fig. 2(d), it is still possible to notice some fluctuations in the morphology of the QW. The QW is not perfectly flat and it appears to be thinner in the right side of the figure. Indeed, this has been previously observed on similar (20-21) templates for blue LEDs [21], and is likely associated to the formation of alternative semipolar and nonpolar facets instead of the (20-21) one, and will be shown experimentally as morphological faceting in section 4 of this work. Statistical distribution analysis (SDA) for the In alloy fluctuation is performed in the flat region of the QW and is shown in Fig. 2(e). The In distribution (red histogram) is compared to the binomial distribution (dotted line) as described in Ref. [22]. According to Fig. 2(e), In in the QW is randomly distributed as already demonstrated in most InGaN alloys grown along semipolar directions [18,21,23].
Fig. 2. (a) APT 3D volume showing the InGaN QW and the last grown periods of the below InGaN/GaN superlattice. (b) 1D concentration profile of In measured in the quantum well and perpendicular to the (20-21) plane. (c) 1D concentration profile of In measured in the superlattice and perpendicular to the (20-21) plane. (d) 2D side view of the In distribution in the quantum well. (e) Distribution of bin compositions of In in the QW with its comparison to the binomial distribution expected for a random alloy.
3.2. Transmission electron microscopy
Transmission electron microscopy (TEM) was employed to investigate the defects present in the structure. The sample was prepared using Focused Ion Beam (FIB) technique with a FEI Helios Dualbeam Nanolab 650. The electron diffraction contrast imaging was carried out with a FEI Tecnai G2 Sphera Microscope, operated at 200 kV. Figures 3(a) and (b) are two beam electron diffraction contrast images in [1-210] cross-section of the LED stack. Figure 3(a) is acquired in g = 000-2, and Fig. 3(b) is taken in g = 10-10. Although the overall density of BSFs on the (20-21) template is significantly low, we detected one generated from substrate, which has the tendency to extend to the surface. It is not visible in g = 000-2 due to g dot b equal to an integer, and therefore only shows contrast in g = 10-10, and indicated in Fig. 3(b). MDs on the other hand (which are mainly a-type dislocations that are invisible in g = 000-2 as shown in Fig. 3(a) [24].), mainly lying along [1-210] direction, were observed at three different interfaces:
Fig. 3. Two beam electron diffraction contrast images in [1-210] cross-section of the LED stack. (a) is acquired in g = 000-2, and (b) is taken in g = 10-10.
1) at the regrowth interface with the template, which may be a result of surface modification as a result of the chemo-mechanical-polishing that is performed after the template is grown and before the device regrowth; 2) at the bottom interface of superlattice and 3) at the bottom interface of AlGaN EBL, which is an indication that the shear stresses on the inclined basal (0001) plane do not vanish, leading to the onset of relaxation processes in semipolar III-nitride heterostructures via dislocation glide and to the formation of misfit dislocations. This has been discussed in-depth by Romanov et al. [25]. No MDs have been observed at the interfaces of the SQW active region. By allowing the majorinty of MDs to form at the superlattice heterointerfaces, the constraints on the active region due to critical thickness for MD formation can be avoided, which has also aided in extending the emission wavelength to 560 nm. This was previously not possible due to nonradiative recombination at defected heterointerfaces in the vicinity of the active region [13,26–28].
3.3. Atomic force miscroscopy
Atomic Force Microscopy (AFM) was used to investigate the surface morphology after the growth of the device. AFM observations have been performed using an Asylum MFP-3D system operating in tapping mode. Figures 4(a) and (b) show 5 μm x 5 μm scans and their corresponding line scans for the LED grown on (20-21) freestanding GaN and on the (20-21) templates, respectively. Both samples have been co-loaded in the same MOCVD growth run and similarly do not display a step-flow growth mode but rather an alternating non-periodic zig-zag consisting of nonpolar m-planes and semipolar (10-11) planes, as it can be concluded from the analysis of the kinetic Wulff plots calculated for GaN [29]. Indeed, similar observations on the same (20-21) orientation have been previously reported [21,30,31]. Although both, freestanding and the GaN template exhibit consistent faceting behavior under a wide range of growth conditions, it is evident that the effect observed on the latter is more significant and and varies up to 9 nm, while the growth on bulk fluctuates around 1 nm or less (Fig. 4(b)). This can be owed to the fact that (20-21) freestanding substrates are cut from a single crystal and are therefore less prone to significant surface faceting upon regrowth, while the templates, although polished prior to regrowth, remain a series of coalesced adjacent crystals. It is also worth noting that even by taking into account the substrate thickness difference (350 μm for bulk GaN and approximately 650 μm for the sapphire) and the thermal conductivity of each substrate, a variation to compensate the growth temperature different does not yield any remarkably different results. Furthermore, simulations by Ivanov et al. [32] show that, besides the material degredation, nanofaceting of the QW interfaces has a profound influence on the properties of the intrinsic field causing it to no longer be perpendicular to the interfaces.
Fig. 4. 5 μm × 5 μm AFM images and a corresponding line scan profile of devices grown on (a) (20-21) bulk and (b) (20-21) templates on sapphire
3.4. Secondary ion mass spectrometry
In order to further understand the materials behavior in the QW, SIMS imaging was performed on a Cameca IMS 7f-Auto. A 10 kV-impact oxygen beam with a 150 pA current and an approximate width of 1.5 μm was rastered over a 150 μm x 150 μm area. Each image was accumulated for 60 seconds, during which approximately 2 Å of depth was ablated.
Remarkably, the acquired SIMS image in the SQW of the LED grown on the template shown in Fig. 5 shows a periodic alignment of indium, which corresponds to the same period used to fabricate the templates (i.e. 5 µm). Generally, similar templates have coalesced individual bands that are divided into two areas: as shown in the cross-sectional schematic and the plan-view panchromatic CL image in Figs. 5(a) and (b), respectively: one is highly defected as a result of the TDs that bend upwards after nucleating at the inclined sapphire c-facet (red dotted rectangle), and another area which is almost defect free (blue dotted rectangle), and corresponds to the area grown after all the generated TDs have been bent. More details on the TD bending mechanism for similar structures is explained in [9].
Fig. 5. (a) A cross sectional schematic on the (20-21) template on patterned sapphire showing the TD bending resulting in defect and defect-free areas and (b) A panchromatic plan-view CL image showing five coalesced bands with the red and blue dotted rectangles highlighting the defective and defect-free areas on each band, and the corresponding (c) SIMS image acquired in the InGaN SQW showing local variations in the indium concentrations in similar periods corresponding to the patterns used to fabricate the template. (Note the different scales)
The origin of the different indium uptake in between the highly defective and low defect density material is a topic of ongoing investigation. However, it may be caused by surface roughness difference, step density, and preferential indium incorporation at the TD proximity. As discussed by Fiore et al., each dislocation is associated with a strain field determined by its Burgers vector. Because the In atom is larger than the host Ga atom, it is expected that if the In atoms are sufficiently mobile during growth, segregation will occur to the tensile part of the dislocation strain field [33]. Based on the strong periodic correlation between the CL and SIMS images shown in Figs. 5(b) and (c), respectively, it can be inferred that the local concentration of In surrounding dislocations is likely to be very different compared to the bulk [34]. It is worth noting that due to the difference in the spatial resolution between APT and SIMS, such variations are very difficult to detect with the former technique. Furthermore, this segregation could significantly affect electronic properties of dislocations and carrier injection into the indium-rich areas, and hence affect device performance [34]. It is important to understand the implications of this behavior on device performance, from a materials and electrical perspective (i.e. leakage, carrier recombination, etc.).
4. Device results & discussion
The optical and electrical properties of the fabricated microLED devices on the (20-21) templates are shown in Fig. 6. Electroluminescence (EL) spectra at current densities ranging from 10 to 100 A/cm2 of 20 μm x 20 μm microLEDs are shown in Fig. 6(a) indicating yellow luminescence around 560 nm. Figure 6(b) shows the forward voltage as a function of the current density for the microLEDs ranging from 80 μm x 80 μm down to 20 μm x 20 μm. The forward voltage at 20 A/cm2 was 2.4 V for the 20 μm x 20 μm microLED and increases slightly to 2.6 V for the larger ones. Figure 6(c) plots the EL peak wavelength as a function of current density (from 10 to 100 A/cm2) for the 20 μm x 20 μm devices indicating a blue shift of 15 nm. The relatively high blueshift observed here in the yellow microLEDs in comparison to our previous reports on green (11-22) microLEDs with 1 nm blueshift within a similar current density range is expected as the wavelength gets longer [19].
Fig. 6. Top row showing luminous images of the fabricated microLEDs from 20 µm to 80 µm (a) EL spectra of the 560 nm LED from 10 - 100 A/cm2 and (b) the I-V curves of the fabricated microLEDs from 80 μm x 80 μm down to 20 μm x 20 μm and (c) a plot of the wavelength shift as a function of current density.
Devices exhibit a peak EQE of 2.5%. Generally, the peak EQE drops as the microLEDs shrink in size due to non-radiative recombination generated from sidewall damage and surface recombination [35]. The size-independence in peak EQE in our microLEDs indicates that the efficiency is limited by other non-radiative recombination sites and is not constrained by the sidewall damage and surface recombination, thanks to the applied surface passivation. The surface faceting and indium segregation on the dislocated areas discussed earlier may be a factor contributing to the degradation of device performance. Even though the EQE remains lower in comparison to devices grown on semipolar bulk GaN (18%) [36] or the commercial mature c-plane technology [37], further improvements in epitaxial design for enhanced defect management of MDs and TDs would further increase the efficiency for semipolar yellow microLEDs on scalable and mass-producible foreign substrates. Furthermore, due to the patterned substrates beneath the fabricated devices, one of the observed characteristics of all the devices grown on such templates, is the emitted light interaction with the template patterns as shown in Fig. 6. As the devices are electrically injected, the emitted light from the mesa is scattered out of the microLEDs exclusively along the [10-14] direction (i.e. also the direction along the projection of the c-axis) and not along the [11-22] (parallel to the direction of the bands). This effect has also been observed for semipolar microLEDs on patterned substrates of different orientations (e.g. (11-22)) and wavelengths (e.g. 440 nm, and 530 nm) [19]. Depending on the intended application, this may indeed be a beneficial feature for controlled and directional light extraction.
5. Summary
InGaN-based semipolar 560 nm microLEDs on high-quality and low-defect-density (20-21) GaN templates grown on scalable and low-cost sapphire substrates are demonstrated. We show that MD confinement at hetero-interfaces away from the active region aid in improving device performance. Conversely however, we discuss issues specific to semipolar templates on patterned sapphire such as surface faceting and indium segregation at the defected areas, which likely contribute to degrading device performance. It is therefore necessary to manage all aspects of the device architecture simultaneously to achieve more efficient devices.
Funding
CREST; Simons Foundation (601952, JS); Solid State Lighting and Energy Electronics Center, University of California Santa Barbara.
Acknowledgments
The authors acknowledge the UCSB-Collaborative Research in Engineering, Science and Technology (CREST) Malaysia project and Solid State Lighting and Energy Electronics Center (SSLEEC) at UCSB for funding. A portion of this work was done in the UCSB nanofabrication facility. This work was supported by a grant from the Simons Foundation (601952, JS).
Disclosures
Authors declare no conflicts of interest
References
1. J. Day, J. Li, D. Lie, C. Bradford, J. Lin, and H. Jiang, “Iii-nitride full-scale high-resolution microdisplays,” Appl. Phys. Lett. 99(3), 031116 (2011). [CrossRef]
2. V. W. Lee, N. Twu, and I. Kymissis, “Micro-led technologies and applications,” Inf. Disp. 32(6), 16–23 (2016). [CrossRef]
3. M. Leroux, N. Grandjean, M. Laügt, J. Massies, B. Gil, P. Lefebvre, and P. Bigenwald, “Quantum confined stark effect due to built-in internal polarization fields in (al, ga) n/gan quantum wells,” Phys. Rev. B 58(20), R13371 (1998). [CrossRef]
4. P. Waltereit, O. Brandt, A. Trampert, H. Grahn, J. Menniger, M. Ramsteiner, M. Reiche, and K. Ploog, “Nitride semiconductors free of electrostatic fields for efficient white light-emitting diodes,” Nature 406(6798), 865–868 (2000). [CrossRef]
5. K. Fujito, K. Kiyomi, T. Mochizuki, H. Oota, H. Namita, S. Nagao, and I. Fujimura, “High-quality nonpolar m-plane gan substrates grown by hvpe,” Phys. Status Solidi A 205(5), 1056–1059 (2008). [CrossRef]
6. T. Hashimoto, F. Wu, J. S. Speck, and S. Nakamura, “A gan bulk crystal with improved structural quality grown by the ammonothermal method,” Nat. Mater. 6(8), 568–571 (2007). [CrossRef]
7. I. Grzegory and S. Porowski, “Gan substrates for molecular beam epitaxy growth of homoepitaxial structures,” Thin Solid Films 367(1-2), 281–289 (2000). [CrossRef]
8. 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]
9. M. Khoury, P. Vennéguès, M. Leroux, V. Delaye, G. Feuillet, and J. Zúñiga-Pérez, “Defect blocking via laterally induced growth of semipolar (1 0 1 1) gan on patterned substrates,” J. Phys. D: Appl. Phys. 49(47), 475104 (2016). [CrossRef]
10. B. Leung, D. Wang, Y.-S. Kuo, K. Xiong, J. Song, D. Chen, S. H. Park, S. Y. Hong, J. W. Choi, and J. Han, “Semipolar (20 2¯ 1) gan and ingan quantum wells on sapphire substrates,” Appl. Phys. Lett. 104(26), 262105 (2014). [CrossRef]
11. F. Brunner, F. Edokam, U. Zeimer, W. John, D. Prasai, O. Krüger, and M. Weyers, “Semi-polar (112¯ 2)-gan templates grown on 100 mm trench-patterned r-plane sapphire,” Phys. Status Solidi B 252(5), 1189–1194 (2015). [CrossRef]
12. M. Khoury, H. Li, P. Li, Y. C. Chow, B. Bonef, H. Zhang, M. S. Wong, S. Pinna, J. Song, J. Choi, and J. S. Speck, “Polarized monolithic white semipolar (20–21) InGaN light-emitting diodes grown on high quality (20–21) GaN/sapphire templates and its application to visible light communication,” Nano Energy 67, 104236 (2020). [CrossRef]
13. M. Khoury, H. Li, H. Zhang, B. Bonef, M. S. Wong, F. Wu, D. Cohen, P. De Mierry, P. Vennéguès, J. S. Speck, and S. Nakamura, “Demonstration of Electrically Injected Semipolar Laser Diodes Grown on Low-Cost and Scalable Sapphire Substrates,” ACS Appl. Mater. Interfaces 11(50), 47106–47111 (2019). [CrossRef]
14. D. Hwang, A. Mughal, C. D. Pynn, S. Nakamura, and S. P. DenBaars, “Sustained high external quantum efficiency in ultrasmall blue iii-nitride micro-leds,” Appl. Phys. Express 10(3), 032101 (2017). [CrossRef]
15. M. S. Wong, D. Hwang, A. I. Alhassan, C. Lee, R. Ley, S. Nakamura, and S. P. DenBaars, “High efficiency of iii-nitride micro-light-emitting diodes by sidewall passivation using atomic layer deposition,” Opt. Express 26(16), 21324–21331 (2018). [CrossRef]
16. K. Thompson, D. Lawrence, D. Larson, J. Olson, T. Kelly, and B. Gorman, “In situ site-specific specimen preparation for atom probe tomography,” Ultramicroscopy 107(2-3), 131–139 (2007). [CrossRef]
17. B. Bonef, M. Catalano, C. Lund, S. P. Denbaars, S. Nakamura, U. K. Mishra, M. J. Kim, and S. Keller, “Indium segregation in n-polar ingan quantum wells evidenced by energy dispersive x-ray spectroscopy and atom probe tomography,” Appl. Phys. Lett. 110(14), 143101 (2017). [CrossRef]
18. H. Li, M. Khoury, B. Bonef, A. I. Alhassan, A. J. Mughal, E. Azimah, M. E. Samsudin, P. De Mierry, S. Nakamura, J. Speck, and S. P. DenBaars, “Efficient semipolar (11-22) 550 nm yellow/green ingan light-emitting diodes on low defect density (11-22) gan/sapphire templates,” ACS Appl. Mater. Interfaces 9(41), 36417–36422 (2017). [CrossRef]
19. H. Li, M. S. Wong, M. Khoury, B. Bonef, H. Zhang, Y. Chow, P. Li, J. Kearns, A. A. Taylor, P. De Mierry, Z. Hassan, S. Nakamura, and S. P. DenBaars, “Study of efficient semipolar (11-22) ingan green micro-light-emitting diodes on high-quality (11-22) gan/sapphire template,” Opt. Express 27(17), 24154–24160 (2019). [CrossRef]
20. 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]
21. M. Khoury, H. Li, B. Bonef, L. Y. Kuritzky, A. J. Mughal, S. Nakamura, J. S. Speck, and S. P. DenBaars, “Semipolar gan templates on sapphire: 432 nm ingan light-emitting diodes and light extraction simulations,” Appl. Phys. Express 11(3), 036501 (2018). [CrossRef]
22. M. P. Moody, L. T. Stephenson, A. V. Ceguerra, and S. P. Ringer, “Quantitative binomial distribution analyses of nanoscale like-solute atom clustering and segregation in atom probe tomography data,” Microsc. Res. Tech. 71(7), 542–550 (2008). [CrossRef]
23. R. Shivaraman, Y. Kawaguchi, S. Tanaka, S. DenBaars, S. Nakamura, and J. Speck, “Comparative analysis of 20 2¯ 1 and 20 2¯ 1¯ semipolar gan light emitting diodes using atom probe tomography,” Appl. Phys. Lett. 102(25), 251104 (2013). [CrossRef]
24. P. S. Hsu, E. C. Young, A. E. Romanov, K. Fujito, S. P. DenBaars, S. Nakamura, and J. S. Speck, “Misfit dislocation formation via pre-existing threading dislocation glide in (11 2¯ 2) semipolar heteroepitaxy,” Appl. Phys. Lett. 99(8), 081912 (2011). [CrossRef]
25. A. E. Romanov, E. C. Young, F. Wu, A. Tyagi, C. S. Gallinat, S. Nakamura, S. P. DenBaars, and J. S. Speck, “Basal plane misfit dislocations and stress relaxation in iii-nitride semipolar heteroepitaxy,” J. Appl. Phys. 109(10), 103522 (2011). [CrossRef]
26. A. Tyagi, F. Wu, E. C. Young, A. Chakraborty, H. Ohta, R. Bhat, K. Fujito, S. P. DenBaars, S. Nakamura, and J. S. Speck, “Partial strain relaxation via misfit dislocation generation at heterointerfaces in (al, in) gan epitaxial layers grown on semipolar (11 2¯ 2) gan free standing substrates,” Appl. Phys. Lett. 95(25), 251905 (2009). [CrossRef]
27. P. Shan Hsu, M. T. Hardy, F. Wu, I. Koslow, E. C. Young, A. E. Romanov, K. Fujito, D. F. Feezell, S. P. DenBaars, J. S. Speck, and S. Nakamura, “444.9 nm semipolar (11 2¯ 2) laser diode grown on an intentionally stress relaxed ingan waveguiding layer,” Appl. Phys. Lett. 100(2), 021104 (2012). [CrossRef]
28. M. Khoury, H. Li, L. Y. Kuritzky, A. J. Mughal, S. Nakamura, J. S. Speck, and S. P. DenBaars, “444 nm ingan light emitting diodes on low-defect-density gan templates on patterned sapphire,” Appl. Phys. Express 10(10), 106501 (2017). [CrossRef]
29. Q. Sun, C. D. Yerino, T. S. Ko, Y. S. Cho, I.-H. Lee, J. Han, and M. E. Coltrin, “Understanding nonpolar gan growth through kinetic wulff plots,” J. Appl. Phys. 104(9), 093523 (2008). [CrossRef]
30. M. Sawicka, C. Cheze, H. Turski, J. Smalc-Koziorowska, M. Kryśko, S. Kret, T. Remmele, M. Albrecht, G. Cywiński, I. Grzegory, and C. Skierbiszewski, “Growth mechanisms in semipolar (202¯ 1) and nonpolar m-plane (101¯ 0) algan/gan structures grown by pambe under n-rich conditions,” J. Cryst. Growth 377, 184–191 (2013). [CrossRef]
31. L. Marona, J. Smalc-Koziorowska, E. Grzanka, M. Sarzynski, T. Suski, D. Schiavon, R. Czernecki, P. Perlin, R. Kucharski, and J. Domagala, “Suppression of extended defects propagation in a laser diodes structure grown on (20-21) gan,” Semicond. Sci. Technol. 31(3), 035001 (2016). [CrossRef]
32. R. Ivanov, S. Marcinkevičius, Y. Zhao, D. Becerra, S. Nakamura, S. DenBaars, and J. Speck, “Impact of carrier localization on radiative recombination times in semipolar (20 2¯ 1) plane ingan/gan quantum wells,” Appl. Phys. Lett. 107(21), 211109 (2015). [CrossRef]
33. N. F. Fiore and C. L. Bauer, “Binding of solute atoms to dislocations,” Prog. Mater. Sci. 13, 85–134 (1968). [CrossRef]
34. M. K. Horton, S. Rhode, S.-L. Sahonta, M. J. Kappers, S. J. Haigh, T. J. Pennycook, C. J. Humphreys, R. O. Dusane, and M. A. Moram, “Segregation of in to dislocations in ingan,” Nano Lett. 15(2), 923–930 (2015). [CrossRef]
35. F. Olivier, A. Daami, C. Licitra, and F. Templier, “Shockley-read-hall and auger non-radiative recombination in gan based leds: A size effect study,” Appl. Phys. Lett. 111(2), 022104 (2017). [CrossRef]
36. S. Yamamoto, Y. Zhao, C.-C. Pan, R. B. Chung, K. Fujito, J. Sonoda, S. P. DenBaars, and S. Nakamura, “High-efficiency single-quantum-well green and yellow-green light-emitting diodes on semipolar (2021) gan substrates,” Appl. Phys. Express 3(12), 122102 (2010). [CrossRef]
37. F. Jiang, J. Zhang, L. Xu, J. Ding, G. Wang, X. Wu, X. Wang, C. Mo, Z. Quan, X. Guo, C. Zheng, S. Pan, and J. Liu, “Efficient ingan-based yellow-light-emitting diodes,” Photonics Res. 7(2), 144–148 (2019). [CrossRef]
