This study investigated the temperature-dependent electroluminescent (EL) performance of InGaN-based amber micro-light-emitting diodes (μLEDs) with a diameter of 40 μm using an epitaxial tunnel junction (TJ) contact for current spreading. The TJ-μLEDs could achieve a high electrical efficiency of 0.935 and a remarkable wall-plug efficiency of 4.3% at 1 A/cm2 at room temperature, indicating an excellent current injection efficiency of the TJ layers regrown by molecular beam epitaxy. Moreover, the current injection of the amber TJ-μLEDs at the forward bias could be further improved at elevated temperatures. The improvement can be explained by the enhanced tunneling probability and acceptor ionization in p-GaN based on the theoretical simulation. The redshift coefficient, which describes the temperature-dependent peak wavelength shift, is obtained as small as 0.05 nm/K, and the high-temperature-to-room-temperature EL intensity ratio is calculated as >0.56 even at a low current density of 0.5 A/cm2 at the temperatures up to 80 °C. This thermal droop behavior was attributed to the enhanced non-radiative recombination, which was confirmed by the shorter carrier lifetime measured at high temperatures.

Full-color micro-light-emitting diodes (μLEDs) are considered promising for next-generation display applications, including large-area displays and smart wearable consumer electronics.1,2 While there has been notable progress in the development of efficient blue and green III-nitride LEDs, red emission still encounter considerable challenges.3–11 AlGaInP-based red LEDs are highly desirable for the excellent efficiency and optical properties, but issues like high surface recombination velocity and long minority carrier diffusion lengths limit their development, especially as devices shrink.12–14 Moreover, the AlGaInP red LEDs also suffer from severe thermal droop for the increased Shockley–Read–Hall (SRH) non-radiative recombination and severe carrier leakage at elevated temperatures.12,15,16

Therefore, InGaN-based red μLEDs seem to be a good alternative to be employed as red emitters for μLEDs displays, due to their small size effect and thermal robustness. However, to achieve red emission, the quantum wells (QWs) have to be incorporated with a high In content exceeding 0.3.17 This high In content leads to a significant lattice mismatch and introduces a substantial density of defects. Various growth techniques have been adopted to improve the material quality in the active region, such as Al(Ga)N capping and barrier layer,18–20 V-pits,21,22 InGaN in situ decomposition layer,11 and varying thickness of n-type GaN23 or buffer layer.24 However, the defect density in the active region of red LEDs remains higher than that in blue and green LEDs. This disparity in defect density results in distinct electronic and optical properties, especially at elevated temperatures and low current densities.

To date, there are few reports on the performance of InGaN-based amber/red μLEDs at elevated temperatures. Zhuang et al. investigated the effect of device size on thermal droop in InGaN-based red LEDs with a large chip size of several hundred micrometers, which did not encompass the μLED size range.25 Li et al. explored the high-temperature electroluminescence (EL) properties of InGaN-based red 40 × 40 μm2 μLEDs,26 but the thermal droop they investigated was primarily at large current densities. At these current densities, the emission wavelength shifted away from the red emission region. In addition, the large current density could make the SRH recombination ineffective, which is not a common case for μLEDs that are operated at low current densities in displays. Therefore, investigating the thermal efficiency droop for amber/red LEDs at low current injection is necessary. Additionally, while several groups have demonstrated that InGaN-based LEDs with tunnel junction (TJ) contacts offer the advantages of improved current spreading and reduced optical loss,27,28 the impact of temperature on the performance of TJ contact at elevated temperatures has yet to be explored.

In this work, we investigated the high-temperature performance of the amber μLEDs with a diameter of 40 μm using an epitaxial TJ contact experimentally and theoretically, which could emit red at low current densities and offer temperature-dependent insight for red μLEDs. Current density–voltage curves were used to characterize the injection efficiency and reverse leakage currents at different temperatures. The band diagram and carrier distribution were simulated to understand the temperature-dependent electrical behaviors in the TJ-μLEDs. Additionally, the peak wavelength, EL intensity, and wall-plug efficiency (WPE) of the μLEDs at different current densities were also carefully studied at elevated temperatures. Finally, we measured the photoluminescent (PL) intensity and lifetime mapping to elucidate the thermal droop behaviors in the amber TJ-μLEDs.

The InGaN amber LED epitaxial wafers were grown on conventional c-plane patterned sapphire substrates by metal–organic chemical vapor deposition (MOCVD). The LED structures consisted of an unintentionally doped (UID) GaN layer (4 μm), a Si-doped n+-GaN (2.5 μm), 24 pairs In0.07GaN0.93/GaN (10/2 nm) superlattices, a single low-In-content quantum well (QW) with a 2.5-nm In0.2Ga0.8N well layer, a 1.5-nm Al0.2Ga0.8N cap layer, and 20-nm GaN/Al0.09Ga0.91N/GaN barrier layers, two pairs of InGaN amber QWs with a 3-nm In0.3Ga0.7N well layer, a 2-nm Al0.2Ga0.8N cap layer, and a 16-nm GaN/16-nm Al0.09Ga0.91N barrier layer, a 100-nm Mg-doped p+-GaN ( 7 × 1019 cm−3), and a 15-nm heavily-doped p++-GaN contact layer (3.5  × 1020 cm−3). After the MOCVD growth, the epitaxial wafer was cleaned in acetone, ethyl alcohol, and hydrofluoric acid to remove the surface organic pollutants and eliminate residual O. The prepared sample was then loaded into a buffer chamber for degassing at 450 °C for 30 min and then transferred to the plasma-assisted molecular beam epitaxy (PA-MBE) chamber to regrow a TJ, which comprised an ultra-thin UID InGaN layer (1 nm), a 30-nm heavily doped n++-GaN layer (1.8  × 1020 cm−3), and a 200-nm moderately doped n+-GaN layer (4.5  × 1019 cm−3).

The InGaN amber LED epitaxial structures with the regrown TJ were measured by cross-sectional transmission electron microscopy (TEM), as shown in Fig. 1(a). The clear and sharp contrast between each layer indicated the high crystalline quality of the amber LED and regrown TJ. Furthermore, we used the high-angle annular dark-field (HAADF) scanning TEM (STEM) to examine the regrown interface of the TJ on the p-GaN of the amber LED. The inserted UID InGaN layer was observed in the STEM image [Fig. 1(b)], as we designed. This layer presented a lower bandgap and large band bending due to the strong polarization field at the heterointerface, which was able to increase the tunneling probability.29 In addition, Fig. 1(b) also reveals a highly ordered atomic arrangement of the regrown TJ layer, and no deterioration in the crystalline quality of the as-grown structure was observed after the TJ regrowth.

FIG. 1.

(a) TEM images of the InGaN amber LED. (b) HAADF-STEM image of the TJ with an inserted UID InGaN layer. (c) STEM image of the hybrid QWs and their corresponding EDS mappings of (d) In and (e) Al elements. (f) SEM image of the TJ-μLED with a diameter of 40 μm.

FIG. 1.

(a) TEM images of the InGaN amber LED. (b) HAADF-STEM image of the TJ with an inserted UID InGaN layer. (c) STEM image of the hybrid QWs and their corresponding EDS mappings of (d) In and (e) Al elements. (f) SEM image of the TJ-μLED with a diameter of 40 μm.

Close modal

Figures 1(c)–1(e) show the high-resolution cross-sectional STEM image and its corresponding energy-dispersive spectroscopy (EDS) elemental mapping of the hybrid multiple QWs. The EDS elemental mappings show the distribution of In and Al atoms in the active region, which confirmed good compositional uniformity for each epitaxial layer. In addition, the AlGaN cap and barrier layer in the low-In-content QW acted as hole-blocking layers to suppress the hole injection.7 In the InGaN amber QWs active region, an AlGaN capping layer was adopted to avoid In evaporation during the high-temperature growth of barriers, while the AlGaN barrier layer has been demonstrated to be useful for band engineering and strain compensation.6,7

The epitaxial wafer was fabricated into μLED devices with a diameter of 40 μm by our standard fabrication processes.20 The TJ served as the current spreading layer instead of indium tin oxide transparent contact. To remove the dry etching damages at the sidewalls of μLEDs, a wet chemical surface treatment by potassium hydroxide and a SiO2 passivation layer was utilized after the mesa etching. Finally, the metal contact windows were opened on the SiO2 layer to expose the TJ and n-GaN layers, followed by an e-beam deposition of Ti/Al/Ni/Au as contact pads. The scanning electron microscope (SEM) was employed to examine the morphology of the amber TJ-μLED with a diameter of 40 μm, as shown in Fig. 1(f). The contact pads on the n-GaN and TJ layer were both designed in a circle shape to ensure uniform injection into the μLED.

The current density–voltage (J–V) characteristics of the amber TJ-μLEDs were first investigated at different stage temperatures from room temperature (25 °C) to 100 °C, as shown in Fig. 2(a). At the forward bias region, the current density of the TJ-μLEDs gradually rose as the temperature increased. This positive temperature-dependent coefficient is primarily attributed to the reduction in series resistance, a consequence of the enhanced tunneling probability in the TJ and the ionization of Mg acceptors in p-GaN. As a result, the forward voltage of our amber μLED decreased from 2.1 to 1.99 V at 1 A/cm2 as the temperature increased from 25 to 100 °C. Notably, these values are much lower compared to the previous reports.30,31 Moreover, the series resistance of the amber TJ-μLED decreased from 0.21 to 0.085 Ω cm2 as the temperature increased to 100 °C, indicating an efficient electrical injection for our amber TJ-μLED.

FIG. 2.

(a) |J|–V curves and (b) the corresponding electrical efficiency and ideality factor of InGaN amber μLED from 25 to 100 °C. (c) Simulated |J|–V curves of the tunnel junction. Simulated (d) energy band diagrams and (e) the distribution of hole concentration of the TJ-μLEDs at different temperatures. (f) Simulated reverse leakage current density when the carrier trap lifetime decreased from 0.9 to 0.05 s.

FIG. 2.

(a) |J|–V curves and (b) the corresponding electrical efficiency and ideality factor of InGaN amber μLED from 25 to 100 °C. (c) Simulated |J|–V curves of the tunnel junction. Simulated (d) energy band diagrams and (e) the distribution of hole concentration of the TJ-μLEDs at different temperatures. (f) Simulated reverse leakage current density when the carrier trap lifetime decreased from 0.9 to 0.05 s.

Close modal

Figure 2(b) shows the calculated electrical efficiency (EE) at different temperatures, which is defined as EE =  h c q · 1 λ V. λ and V represent the peak wavelength [Fig. 3(b)] and forward voltage of the amber TJ-μLEDs at 1 A/cm2 [Fig. 2(a)], respectively. The EE value could reach as high as 0.935 when the device was operated at room temperature. As the temperature increased, the EE value also slightly increased, approaching the ideal carrier injection.

FIG. 3.

(a) EL spectra of an amber TJ-μLED in a diameter of 40 μm. Temperature dependence of (b) peak wavelength and (c) EL intensity for this TJ-μLED. (d) Temperature dependence of WPE for this TJ-μLED after packaged. (e) EL emission images of the TJ-μLED at different stage temperatures.

FIG. 3.

(a) EL spectra of an amber TJ-μLED in a diameter of 40 μm. Temperature dependence of (b) peak wavelength and (c) EL intensity for this TJ-μLED. (d) Temperature dependence of WPE for this TJ-μLED after packaged. (e) EL emission images of the TJ-μLED at different stage temperatures.

Close modal

We also calculated the ideality factor of our amber TJ-μLED by n ideality = q k T ( lnI V ) 1, from the slope of the linear region of logarithmic J–V curves. The ideality factor reflects the carrier transport mechanism, which is determined by trap-assisted tunneling, carrier leakage, and other imperfect injections.32  Figure 2(b) shows the obtained nideality at various temperatures. The value of the ideality factor decreases from 3.68 to 3.5 and then increases to 5.83 as the temperature rises from 25 to 100 °C. The reduction in the ideality factor with the temperature reflects an improved carrier injection process, while the increase in the ideality factor indicates the dominant current leakage at temperatures above 80 °C. These results are consistent with our discussion about the J–V curves and EE in Figs. 2(a) and 2(b).

In addition, a theoretical simulation was carried out to explore the carrier transport in our amber TJ-μLEDs at different temperatures. Figure 2(c) shows the simulated J–V curves of the bare TJ at different temperatures. The positive voltage was applied to the n-GaN [the inset image in Fig. 2(c)] because the TJ operates in its reverse bias region as the p-contact for TJ-μLEDs. Clearly, the slope of the J–V curves increases with the temperature, indicating an enhanced tunneling probability for the TJ at elevated temperatures. This is one of the reasons for the increased EE discussed above.

To further investigate the carrier injection in amber LEDs at elevated temperatures, we simulated the energy band diagrams and the distribution of hole concentration within the active region of the TJ-μLED at 25 and 100 °C in Figs. 2(d) and 2(e), respectively. The quasi-Fermi level of electrons (EFn) in the p-GaN region at 100 °C was lower compared to its position at 25 °C, which indicated an enhanced ionization of acceptors in p-GaN. Moreover, the hole concentration in the second amber QW from the p-side was significantly higher at 100 °C in Fig. 2(e), demonstrating the efficient hole injection process at high temperatures. Thus, these are also the reasons for the high EE observed at elevated temperatures.

However, we noticed that the reverse current of the amber TJ-μLEDs became significant when the temperature was higher than 80 °C, as shown in Fig. 2(a). This phenomenon illustrated that many leakage channels were activated at high temperatures. We presumed that these leakage channels were caused by the activation of deep-level defects, which assisted the tunneling of carriers across the p–n junction of the amber μLEDs at the reverse bias. In addition, a “dip” located at the reverse bias was observed and shifted to a lower voltage with the increase in temperature. This shift is primarily attributed to the increased trap-assisted tunneling probability at high temperatures.

To explore the influence of the activation of deep-level defects, we simulated the reverse current density of the device by setting different trap lifetimes. The trap lifetime will become shorter when the deep-level defects activate at specific high temperatures.33 Notably, all the InGaN and AlGaN ternary alloys in our amber TJ-LED structure were replaced by GaN to improve the convergence during the simulation at reverse bias. Figure 2(f) shows that the reverse leakage current density increased as the trap lifetime gradually decreased at the reverse biases from −7 to −10 V, aligning well with the observed large leakage currents at high temperatures above 80 °C in Fig. 2(a). Therefore, we believed that the more defects in the amber TJ-μLEDs gave rise to the larger leakage currents at high temperatures.

The electroluminescence (EL) of the amber TJ-μLEDs was then investigated at elevated temperatures under a microscope system. We first measured the EL spectra of the μLED at 0.1 to 10 A/cm2 at RT, a typical operation range of the current density for μLED displays. As shown in Fig. 3(a), the peak wavelength of the amber TJ-μLED exhibits a significant blue shift from 617 to 581 nm. The blue shift is caused by the screening of the large piezoelectric field in InGaN amber QWs with high In composition. In addition, no additional peaks were observed in the spectra, indicating that the InGaN amber QWs in the active region are free from strong phase separation.

Figure 3(b) shows the temperature dependence of the peak wavelength of our amber TJ-μLEDs at 0.5, 1, and 10 A/cm2. When the temperature increased from 25 to 80 °C, the peak wavelength showed a redshift with an estimated coefficient of 0.05 nm/K regardless of the current density. This value is much lower than the value of 0.137 nm/K in AlInGaP-based red LEDs,34 revealing the good wavelength stability at high temperatures for InGaN-based amber μLEDs. However, a more pronounced 7-nm redshift at 1 A/cm2 was observed when the temperature rose up to 100 °C, which was still lower than the redshift value (10 nm) of AlInGaP-based red LEDs.

However, the EL intensity of the amber TJ-μLEDs reduced gradually as the stage temperature increased, especially at the lower current densities. This behavior can be attributed to the severe SRH recombination, which competed with the radiative recombination and is mainly determined by the defect density. Generally, amber/red μLEDs tend to possess a relatively high density of defects due to the higher In composition in QWs. Therefore, more SRH recombination centers exist in the amber μLEDs and become effective at elevated temperatures. These SRH recombination centers are incompletely suppressed at low current densities, leading to a more significant thermal droop in the EL intensity of the amber TJ-μLEDs.

The ratio of the high-temperature EL intensity to room-temperature EL intensity remains above 0.56, even at a low current density of 0.5 A/cm2, when the temperature increases up to 80 °C. Although the ratio is lower than that of blue/green μLEDs,35,36 it has been better than that of AlGaInP red μLEDs with small sizes.12 However, we noticed that the EL intensity dropped dramatically when the temperature exceeded 80 °C. This sharp drop that we guessed originated from the pronounced current leakage at high temperatures in our amber TJ-μLEDs in Fig. 2(a).

To package the μLED chip, the fabricated 40-μm-diameter amber μLEDs underwent initial dicing, followed by mounting on an aluminum stage and subsequent encapsulation with epoxy resin. To regulate the ambient temperature of this packaged TJ-μLED, the device was positioned on a ceramic heating plate. The temperature of the heating plate was meticulously controlled by adjusting the applied voltage. This regulated temperature was then conveyed to the packaged device through the aluminum stage. Subsequently, the temperature-controllable TJ-μLED underwent measurement in an integrating sphere. Figure 3(d) shows the temperature dependence of WPE at different current densities. A high value of WPE, approximately 4.3% measured at 1 A/cm2 at RT, was achieved for our TJ-μLED. This high WPE is comparable with our previous work with ITO as the current spreading layer and better than other InGaN-based amber/red μLEDs,20 implying good carrier injection of our regrown TJ and crystal quality of our amber μLEDs. The WPE also decreased as the temperature increased. The thermal droop in WPE became smaller when the current density increased, which was consistent with the EL intensity curves in Fig. 3(c).

We captured the emission images of the amber TJ-μLEDs at 10 A/cm2 from 25 to 100 °C in Fig. 3(e). Most μLED areas exhibited uniform luminescence in amber-red color at 25 °C. When the temperature increased, the brightness of the μLED gradually dimmed, and more dark spots were observed. This behavior agreed well with the results in Figs. 3(b)–3(d) and those dark spots might be related to the activated defects in the amber TJ-μLEDs, as we discussed above.

Finally, we conducted PL intensity and carrier lifetime mappings at elevated temperatures to investigate the mechanism of the thermal droop in the amber TJ-μLEDs.37,38 The PL intensity mappings were measured using a 385-nm LED source with an average excitation power of 0.056 kW/cm2. The carrier lifetime mappings were obtained using a fluorescence lifetime imaging microscope (FLIM) with a 375-nm pulsed laser. The lifetime resolution is 100 ps, and the PL excitation power is around 0.027 kW/cm2. These measurements were performed at low excitation power, corresponding to the scope of low current density injection for μLEDs.

Figures 4(a) and 4(b) show the PL intensity mappings of the amber TJ-μLED at 25 and 100 °C, respectively. A reduction in PL intensity over the whole TJ-μLED area was observed at the high temperature, consistent with the EL emission trends illustrated in Fig. 3(e). The corresponding carrier lifetime mappings at 25 and 100 °C are presented in Figs. 4(c) and 4(d), respectively. Upon comparing Fig. 4(a) with 4(c) and Fig. 4(b) with 4(d), we noticed that regions with higher emission intensity, typically indicating fewer non-radiative recombination centers, tended to exhibit longer carrier lifetime.39 In other words, more non-radiative recombination centers lead to a shorter carrier lifetime, which support the assumption in the earlier analysis of Fig. 2(f) that a high density of defects corresponds to a decreased trap lifetime. When the temperature increases, the carrier lifetime becomes shorter in Fig. 4(d). The shorter lifetime implies that more non-radiative recombination centers were activated at high temperatures, which could explain the leakage current in Fig. 2(a) and the thermal droop in the emission intensity.

FIG. 4.

PL intensity mapping images of a 40-μm-in-diameter TJ-μLED at (a) 25 and (b) 100 °C. The carrier lifetime mappings of this TJ-μLED at (c) 25 and (d) 100 °C.

FIG. 4.

PL intensity mapping images of a 40-μm-in-diameter TJ-μLED at (a) 25 and (b) 100 °C. The carrier lifetime mappings of this TJ-μLED at (c) 25 and (d) 100 °C.

Close modal

In summary, we demonstrated InGaN-based amber μLEDs with a diameter of 40 μm using an epitaxial TJ contact. The EE and WPE of the amber TJ-μLEDs could be obtained as high as 0.935 and 4.3% at 1 A/cm2, which are attributed to the near-ideal electrical injection and good crystal quality of the amber QWs. The carrier injection at the forward bias could be further improved at elevated temperatures. This improvement can be explained by the enhanced tunneling probability and acceptor ionization in p-GaN. In addition, our amber μLEDs exhibited a redshift coefficient of 0.05 nm/K and a high-temperature/room-temperature EL intensity ratio of >0.56 even at a low current density of 0.5 A/cm2 at the temperature up to 80 °C. The thermal droop in EL intensity was caused by the SRH recombination due to the activation of more defects at high temperatures, which could be explained by the shorter carrier lifetime. Nonetheless, the InGaN TJ-μLEDs exhibit superior high-temperature performance compared to AlGaInP red μLEDs. These findings provide valuable insight into the thermal droop behavior of the amber TJ-μLED at low current densities and pave the way for developing efficient and temperature-tolerated amber/red μLED in the future.

The authors acknowledge the financial support from the National Key R&D Program of China (No. 2023YFB3610300), the National Nature Science Foundation of China (Nos. 62274083, 62204073, 62074077, 62004104, and 61974126), the Leading-edge Technology Program of the Jiangsu Natural Science Foundation (No. BE2021008-2), the Fundamental Research Funds for the Central Universities (No. 021014380192), the Collaborative Innovation Center of Solid-State Lighting and Energy-Saving Electronics, the Foundation of Lohua Chip-Display Technology Development Company, Ltd. (No. 2022320116000031), the King Abdullah University of Science and Technology Research Funding (Nos. ORA-2022-5313 and BAS/1/1676-01-01), the Natural Science Foundation of Fujian Province of China (No. 2021J06009), and the National Nature Science Foundation of Anhui Province (No. 2208085QF210).

The authors thank Tianjin SimuCal Technology Co., Ltd. for supporting numerical simulation.

The authors have no conflicts to disclose.

Yimeng Sang: Data curation (equal); Formal analysis (equal); Writing – original draft (equal). Zhe Zhuang: Formal analysis (equal); Funding acquisition (equal); Writing – review & editing (equal). Kun Xing: Resources (equal). Dongqi Zhang: Resources (equal). Jinjian Yan: Data curation (equal). Zhuoying Jiang: Data curation (equal). Chenxue Li: Investigation (equal). Kai Chen: Investigation (equal). Yu Ding: Investigation (equal). Tao Tao: Funding acquisition (equal). Daisuke Iida: Resources (equal). Ke Wang: Resources (equal). Cheng Li: Resources (equal). Kai Huang: Resources (equal). Kazuhiro Ohkawa: Resources (equal). Rong Zhang: Resources (equal). Bin Liu: Conceptualization (equal); Funding acquisition (equal); Writing – review & editing (equal).

The data that support the findings of this study are available within the article.

1.
P. J.
Parbrook
,
B.
Corbett
,
J.
Han
,
T.
Seong
, and
H.
Amano
,
Laser Photonics Rev.
15
,
2000133
(
2021
).
2.
T.
Wu
,
C.-W.
Sher
,
Y.
Lin
,
C.-F.
Lee
,
S.
Liang
,
Y.
Lu
,
S.-W.
Huang Chen
,
W.
Guo
,
H.-C.
Kuo
, and
Z.
Chen
,
Appl. Sci.
8
,
1557
(
2018
).
3.
H. S.
Wasisto
,
J. D.
Prades
,
J.
Gülink
, and
A.
Waag
,
Appl. Phys. Rev.
6
,
041315
(
2019
).
4.
B.
Damilano
and
B.
Gil
,
J. Phys. D
48
,
403001
(
2015
).
5.
P.
Li
,
H.
Li
,
M. S.
Wong
,
P.
Chan
,
Y.
Yang
,
H.
Zhang
,
M.
Iza
,
J. S.
Speck
,
S.
Nakamura
, and
S. P.
Denbaars
,
Crystals
12
,
541
(
2022
).
6.
J.-I.
Hwang
,
R.
Hashimoto
,
S.
Saito
, and
S.
Nunoue
,
Appl. Phys. Express
7
,
071003
(
2014
).
7.
D.
Iida
,
P.
Kirilenko
,
M.
Velazquez-Rizo
,
Z.
Zhuang
,
M. A.
Najmi
, and
K.
Ohkawa
,
AIP Adv.
12
,
065125
(
2022
).
8.
S. S.
Pasayat
,
C.
Gupta
,
M. S.
Wong
,
R.
Ley
,
M. J.
Gordon
,
S. P.
DenBaars
,
S.
Nakamura
,
S.
Keller
, and
U. K.
Mishra
,
Appl. Phys. Express
14
,
011004
(
2020
).
9.
A.
Dussaigne
,
P.
Le Maitre
,
H.
Haas
,
J.-C.
Pillet
,
F.
Barbier
,
A.
Grenier
,
N.
Michit
,
A.
Jannaud
,
R.
Templier
,
D.
Vaufrey
,
F.
Rol
,
O.
Ledoux
, and
D.
Sotta
,
Appl. Phys. Express
14
,
092011
(
2021
).
10.
Z.
Liu
,
M.
Nong
,
Y.
Lu
,
H.
Cao
,
S.
Yuvaraja
,
N.
Xiao
,
Z.
Alnakhli
,
R. R.
Aguileta Vázquez
, and
X.
Li
,
Opt. Lett.
47
,
6229
(
2022
).
11.
J.
Hu
,
K.
Xing
,
Z.
Xia
,
Y.
Sang
,
X.
Yang
,
T.
Tao
,
Z.
Zhuang
,
R.
Zhang
, and
B.
Liu
,
Appl. Phys. Lett.
123
,
111107
(
2023
).
12.
D.-H.
Lee
,
S.-Y.
Lee
,
J.-I.
Shim
,
T.-Y.
Seong
, and
H.
Amano
,
ECS J. Solid State Sci. Technol.
10
,
095001
(
2021
).
13.
J.-T.
Oh
,
S.-Y.
Lee
,
Y.-T.
Moon
,
J. H.
Moon
,
S.
Park
,
K. Y.
Hong
,
K. Y.
Song
,
C.
Oh
,
J.-I.
Shim
,
H.-H.
Jeong
,
J.-O.
Song
,
H.
Amano
, and
T.-Y.
Seong
,
Opt. Express
26
,
11194
(
2018
).
14.
S.
Sinha
,
F.-G.
Tarntair
,
C.-H.
Ho
,
Y.-R.
Wu
, and
R.-H.
Horng
,
IEEE Trans. Electron Devices
68
,
2818
(
2021
).
15.
M. S.
Kim
,
H. K.
Lee
, and
J. S.
Yu
,
Semicond. Sci. Technol.
28
,
025005
(
2013
).
16.
C.-H.
Oh
,
J.-I.
Shim
, and
D.-S.
Shin
,
Jpn. J. Appl. Phys., Part 1
58
,
012005
(
2019
).
17.
Z.
Zhuang
,
D.
Iida
, and
K.
Ohkawa
,
Jpn. J. Appl. Phys., Part 1
61
,
SA0809
(
2022
).
18.
D.
Iida
,
S.
Lu
,
S.
Hirahara
,
K.
Niwa
,
S.
Kamiyama
, and
K.
Ohkawa
,
J. Cryst. Growth
448
,
105
(
2016
).
19.
K.
Xing
,
Z.
Xia
,
G.
Xie
,
Z.
Pan
,
Z.
Zhuang
,
J.
Hu
,
Y.
Sang
,
T.
Tao
,
X.
Yang
,
B.
Liu
, and
R.
Zhang
,
IEEE Photonics Technol. Lett.
35
,
1439
(
2023
).
20.
Y.
Sang
,
Z.
Zhuang
,
K.
Xing
,
Z.
Jiang
,
C.
Li
,
F.
Xu
,
D.
Zhang
,
J.
Yu
,
J.
Zhao
,
T.
Zhi
,
T.
Tao
,
C.
Li
,
K.
Huang
,
K.
Ohkawa
,
R.
Zhang
, and
B.
Liu
,
IEEE Electron Device Lett.
45
,
76
(
2024
).
21.
D.
Iida
,
Z.
Zhuang
,
P.
Kirilenko
,
M.
Velazquez-Rizo
,
M. A.
Najmi
, and
K.
Ohkawa
,
Appl. Phys. Lett.
116
,
162101
(
2020
).
22.
D.
Iida
and
K.
Ohkawa
,
Semicond. Sci. Technol.
37
,
013001
(
2021
).
23.
X.
Hou
,
S.-S.
Fan
,
H.
Xu
,
D.
Iida
,
Y.-J.
Liu
,
Y.
Mei
,
G.-E.
Weng
,
S.-Q.
Chen
,
B.-P.
Zhang
, and
K.
Ohkawa
,
Appl. Phys. Lett.
120
,
261102
(
2022
).
24.
Z.
Chen
,
B.
Sheng
,
F.
Liu
,
S.
Liu
,
D.
Li
,
Z.
Yuan
,
T.
Wang
,
X.
Rong
,
J.
Huang
,
J.
Qiu
,
W.
Liang
,
C.
Zhao
,
L.
Yan
,
J.
Hu
,
S.
Guo
,
W.
Ge
,
B.
Shen
, and
X.
Wang
,
Adv. Funct. Mater.
33
,
2300042
(
2023
).
25.
Z.
Zhuang
,
D.
Iida
, and
K.
Ohkawa
,
Appl. Phys. Lett.
116
,
173501
(
2020
).
26.
P.
Li
,
A.
David
,
H.
Li
,
H.
Zhang
,
C.
Lynsky
,
Y.
Yang
,
M.
Iza
,
J. S.
Speck
,
S.
Nakamura
, and
S. P.
DenBaars
,
Appl. Phys. Lett.
119
,
231101
(
2021
).
27.
Y.
Wu
,
B.
Liu
,
F.
Xu
,
Y.
Sang
,
T.
Tao
,
Z.
Xie
,
K.
Wang
,
X.
Xiu
,
P.
Chen
,
D.
Chen
,
H.
Lu
,
R.
Zhang
, and
Y.
Zheng
,
Photonics Res.
9
,
1683
(
2021
).
28.
D.
Hwang
,
A. J.
Mughal
,
M. S.
Wong
,
A. I.
Alhassan
,
S.
Nakamura
, and
S. P.
DenBaars
,
Appl. Phys. Express
11
,
012102
(
2018
).
29.
D.
Takasuka
,
Y.
Akatsuka
,
M.
Ino
,
N.
Koide
,
T.
Takeuchi
,
M.
Iwaya
,
S.
Kamiyama
, and
I.
Akasaki
,
Appl. Phys. Express
9
,
081005
(
2016
).
30.
P.
Li
,
H.
Li
,
H.
Zhang
,
C.
Lynsky
,
M.
Iza
,
J. S.
Speck
,
S.
Nakamura
, and
S. P.
DenBaars
,
Appl. Phys. Lett.
119
,
081102
(
2021
).
31.
R. C.
White
,
H.
Li
,
M.
Khoury
,
C.
Lynsky
,
M.
Iza
,
S.
Keller
,
D.
Sotta
,
S.
Nakamura
, and
S. P.
DenBaars
,
Crystals
11
,
1364
(
2021
).
32.
Y.
Liu
,
K.
Zhang
,
F.
Feng
,
K.
Chan
,
S.
Yeung
,
H.
Kwok
, and
Z.
Liu
,
J. Soc. Inf. Disp.
29
,
948
(
2021
).
33.
X. S.
Nguyen
,
K.
Lin
,
Z.
Zhang
,
B.
McSkimming
,
A. R.
Arehart
,
J. S.
Speck
,
S. A.
Ringel
,
E. A.
Fitzgerald
, and
S. J.
Chua
,
Appl. Phys. Lett.
106
,
102101
(
2015
).
34.
H. K.
Lee
,
D. H.
Lee
,
Y. M.
Song
,
Y. T.
Lee
, and
J. S.
Yu
,
Solid-State Electron.
56
,
79
(
2011
).
35.
I. E.
Titkov
,
S. Y.
Karpov
,
A.
Yadav
,
V. L.
Zerova
,
M.
Zulonas
,
B.
Galler
,
M.
Strassburg
,
I.
Pietzonka
,
H.-J.
Lugauer
, and
E. U.
Rafailov
,
IEEE J. Quantum Electron.
50
,
911
(
2014
).
36.
A.
David
,
N. G.
Young
,
C.
Lund
, and
M. D.
Craven
,
Appl. Phys. Lett.
115
,
223502
(
2019
).
37.
W.
Li
,
Q.
Wu
,
L.
Lu
,
Y.
Tian
,
H.
Luo
,
Y.
Yun
,
S.
Jiang
,
M.
Chen
, and
C.
Li
,
J. Mater. Chem. C
11
,
3314
(
2023
).
38.
J.
Zhang
,
B.
Hu
,
Z.
Guo
,
H.
Luo
,
R.
Wang
,
M.
Li
,
M.
Chen
, and
C.
Li
,
Appl. Surf. Sci.
628
,
157339
(
2023
).
39.
B.
Hu
,
J.
Zhang
,
Z.
Guo
,
L.
Lu
,
P.
Li
,
M.
Chen
, and
C.
Li
,
ACS Appl. Mater. Interfaces
14
,
15840
(
2022
).