Open Access
Add to BibSonomy
Abstract
An asymmetric multiple quantum well (MQW) without the first quantum barrier layer is designed, and its effect on the device performance of GaN-based blue LDs has been studied experimentally and theoretically. It is found that compared with LD using symmetrical multiple quantum well, device performance is improved significantly by using asymmetric MQW, i.e. having a smaller threshold current density, a higher output optical power and a larger slope efficiency. The threshold current density decreases from 1.28 kA/cm2 to 0.86 kA/cm2, meanwhile, the optical power increases from 1.77 W to 2.52 W, and the slope efficiency increases from 1.15 W/A to 1.49 W/A. The electroluminescence characteristics below the threshold current demonstrate that asymmetric MQW is more homogeneous due to the suppressed strain and piezoelectric field. Furthermore, theoretical calculation demonstrates that the enhancement of electron injection ratio and reduction in optical loss are another reason for the improvement of device performance, which is attributed to a smaller electron potential barrier and a more concentrated optical field distribution in the asymmetric structure, respectively. The new structure design with asymmetric MQW is concise for epitaxial growth, and it would also be a good possible choice for GaN-based LDs with other lasing wavelengths.
© 2022 Optica Publishing Group under the terms of the Optica Open Access Publishing Agreement
1. Introduction
For the last thirty years, GaN-based bule laser diodes (LDs) have made great progress since the first GaN-based blue-violet LD working under pulsed current injection was reported by Nichia Ltd. in 1996 [1–3]. In recent years, the market of GaN-based LDs penetrates into new applications including laser processing and underwater wireless communication. Taking laser processing as an example, compared to traditional infrared LDs, GaN-based blue LDs have great benefits in high quality and efficiency, such as copper welding, due to a higher absorption coefficient of nonferrous metal in the blue spectral range [4–6]. Moreover, underwater wireless communication using the blue laser source could provide a high transmission data rate, owing to the low absorption window of seawater in the blue-green spectral range [7–9]. The requirements for device performance of GaN-based blue LDs are pushed to a new high by these applications. Researchers have made impressive progress in material growth, device structure design and fabrication, including the growth of crack-free thick AlGaN layers [10], epitaxial lateral overgrowth [11] and the growth of stepped upper waveguide layer [11]. However, related physical mechanisms need to further investigated, and LDs’ performance such as threshold current, optical power and slope efficiency also needs to be further improved.
Multiple quantum well (MQW) as the active area of GaN-based laser diodes is essential for device performance, whose material quality and structural characteristics play a decisive role in the optical performance of LDs. Researchers have attempted to promote the luminous efficiency of MQW in GaN-based laser diodes, such as tailoring first-barrier doping to enhanced carrier confinement and radiative recombination (even difficult to control the doping level in thin layers) [12], using quaternary AlInGaN quantum barriers to increase optical gain (even epitaxial growth of quaternary AlInGaN is complicated and particularly difficult) [13], and growing p-type contact layers under low temperature to suppress the degradation of quantum wells (material quality of p-type layers grown under low temperature is generally poor) [14]. Moreover, a InGaN lower waveguide layer is often used to employ in the device structure for obtaining better optical confinement effect. However, in this case device performance may be deteriorated when its indium content is too high compared to the quantum barrier because the carriers would accumulate in the lower InGaN layer [15,16]. These results encourage that a new MQW structure should be designed to meet the requirements of high device performance, especially for the LDs with high-indium content InGaN lower waveguide layer. Therefore, in this work, a new device structure with asymmetric MQW is proposed to improve the performance of GaN-based laser diodes, in which the first quantum barrier layer is removed. Experimental and theoretical calculation results demonstrate that the device performance is really improved significantly using asymmetric MQW, which is mainly attributed to the improvement of material quality, enhancement of electron injection ratio and reduction of optical loss. Moreover, this new device structure is concise for epitaxial growth and also for device structure design.
2. Fabrication and theoretical calculation of GaN-based LDs
Two GaN-based blue LDs are fabricated, i.e., LDI and LDII, which are grown in an AIXTRON 3 × 2 in. close-coupled showerhead reactor on c-plane GaN substrates. Only the MQW is different between LDI and LDII, one is symmetrical and another is asymmetrical. There is no first quantum barrier (FQB) layer in the asymmetrical MQW of LDII. The schematic diagram of device structure of LDI and LDII is shown in Fig. 1. It is consisted of a GaN substrate, a thick n-GaN layer, an n-type cladding layer (CL), a lower waveguide layer (LWG), a multiple quantum well, an upper waveguide layer (UWG), an electron blocking layer (EBL), a p-type cladding layer, and a p-type GaN layer. The indium content of the InGaN LWG layer and quantum barriers in MQWs are 2% and 4%, respectively, which are checked by X-ray diffraction (XRD) rocking curve at the (002) reflection measurement. It is noted that the indium content of InGaN LWG is much lower than that of quantum well layers, and thus has a much larger energy gap than the quantum wells. During growth, TMGa/TEGa, TMAl, TMIn, NH3, Cp2Mg and SiH4 are used as the Ga, Al, In, N, Mg and Si sources, respectively. After epitaxial growth, a 30-µm-wide ridge stripe was formed by dry etching along the <1-100 > direction, and a 1200-µm-long cavity was fabricated by cleaving along the {1-100} plane. Then, the front and rear cleaved cavity facets are coated, and the reflectivity is 10% and 90%, respectively. As a part of the vertical device structure, Ti/Pt/Au and Pd/Pt/Au are used as the n-type and p-type metal electrodes, respectively. The output optical power of LDI and LDII are obtained through a Si-based photodetector and Keithley source meter under direct current (DC) mode, and the stimulated spectra are measured by an Ocean Optics HR2000 spectrometer. In addition, electroluminescence (EL) spectra below the lasing threshold current of LDI and LDII were measured using a Keithley 2400 source meter, an optical lens and a HORIBA FHR640 high-resolution spectrometer.
In addition, as shown in Fig. 2, two light emitting diode (LED) samples, named as LEDI and LEDII, respectively, are also additionally prepared to further check how the LWG and first quantum barrier layer have influence on the material quality of MQW. In order to make a better comparison, a similar InGaN LWG layer is grown in these LED structure, and the MQW structure is different for LEDI and LEDII. In a similar way, LEDI is symmetrical and LEDII is asymmetrical, respectively. Actually, the growth conditions of MQW in LEDI and LEDII are the same as that of LDI and LDII, except that the GaN templates are grown on sapphire instead of GaN in the LED structure. The reciprocal space mapping (RSM) of LEDI and LEDII is measured through X-ray diffraction (XRD, Rigaku SmartLab), which equipped with a 2D detector and a 2-bounch monochromator of Ge (220) crystal. During XRD measurements, the operation voltage and current is 40 kV and 30 mA, respectively, and the 0.154 nm Cu Kα1 radiation is used.
Moreover, two GaN-based blue LDs, i.e. LD1 and LD2 are studied theoretically with the LASTIP program (Crosslight Software Inc.), which is powerful for calculating and analyzing the electrical and optical characteristics of LDs thermionically by self-consistently solving Poisson’s equations and the current continuity equations. In fact, LD1 and LD2 are designed based on the device structures of LDI and LDII, respectively. The structural parameters of each layer in LD1 and LD2 are listed in Table 1, and most of them are the same except the MQW. There is a first In0.02Ga0.98N barrier layer in LD1, but not in LD2. During the simulation, the screening factor in the LASTIP calculation is set to 25%, and the front and rear reflectivity are set to 20% and 90%, respectively. In addition, the refractive indexes of AlGaN and InGaN alloys are estimated according to the method proposed by Laws [17] and Piprek [18,19]. The stimulated wavelength of LD1 and LD2 is the same, i.e. 451 nm.
Table 1. Structural parameters of LD1 and LD2 for calculation with LASTIP
3. Results and discussion
The ridge-stripped LD chips were fabricated to be approximately 100 µm thick, and then were packaged in a C-mount with n-side up and bonded on a Cu heatsink by indium solder. Figure 3 (a) shows the optical spectra of stimulated emission of LDI and LDII. Figure 3 (b) shows the optical output power as a function of the forward direct current (P-I) of LDI and LDII at room temperature. The optical spectra of stimulated emission show that the lasing wavelengths of LDI and LDII are approximately 452 nm and 451 nm, respectively. Moreover, the P-I curves indicate that the threshold current of LDI and LDII is approximately 459 mA and 310 mA, respectively, and the corresponding threshold current densities are approximately 1.28 kA/cm2 and 0.86 kA/cm2, respectively. The optical power of LDI and LDII reaches to 1.77 W and 2.52 W, respectively, when the injection current is 2.0 A. In addition, the slope efficiency(SE) of LDI and LDII is 1.15 W/A and 1.49 W/A, respectively, which is obtained by the linear fitting calculation from the P-I curves above the threshold current. Compared to LDI, threshold current, optical power and slope efficiency of LDII have an improvement of 32.5%, 42.4% and 29.6%, respectively. This result demonstrates that removing the first quantum barrier layer in the new device structure is beneficial to improve the device performance of GaN-based blue LDs.
Fig. 3. (a) Optical spectra of stimulated emission of LDI and LDII under 1.8 A current injection at room-temperature. (b) Optical output power versus current (P-I) curves of LDI and LDII in the C-mount package under continuous-wave operation at room temperature, where SE represents the slope efficiency.
The reason for the improvement of device performance may be attributed to the better material quality while using asymmetric MQW. It is noted that the quantum wells are grown above the In0.04Ga0.96N lower waveguide layer when the first quantum barrier layer is removed. However, the quantum wells are grown above the In0.02Ga0.98N quantum barrier layer for the symmetrical MQW. Thus, the strain and the piezoelectric field of the quantum wells in LDII would be smaller than that in LDI due to two reasons. On one side, compared to LDI, the difference of indium content between the In0.04Ga0.96N lower waveguide layer and quantum well layer is smaller than the difference of indium content between the FQB and quantum well layer in LDI, thus the piezoelectric field of the quantum wells in LDII would be smaller. Meanwhile, the strain in LDII should also be smaller. To check the strain and the piezoelectric field in the quantum well layers, XRD mapping of LEDI and LEDI and electroluminescence (EL) spectra of LDI and LDII are measured. The discussions are taken below.
First, it is worth noting that the InGaN lattice was not relaxed for both LEDI and LEDII according reciprocal space mapping (RSM) mapping on (002) reflection, as presented in Fig. 4. It can be seen that the peaks from up to down come from GaN template layer, In0.04Ga0.96N lower waveguide layer and MQW, respectively, and they are well aligned along Qz axis. In Fig. 4 their alignment is marked with black straight lines. The difference in Qx position between these peaks is less than 0.001 nm. It indicates that the MQW of LEDI and LEDII is in an elastic strained state due to the existed lattice differences between the neighboring layers.
Furthermore, EL spectra of LDI and LDII are measured under the same conditions as a function of injection current before the lasing as shown in Figs. 5 (a) and (b), respectively, where the scale of emission intensity for two LDs are normalized for making comparison. Figures 5 (c) and (d) show the injection current-dependent peak wavelength and full width at half maximum (FWHM) for LDI and LDII, respectively. These data are obtained by Gauss fitting of the EL spectra. Figure 5 shows that the EL intensity of LDI is much smaller than that of LDII when the injection current density is 0.0092 kA/cm2 (i.e. at a total current of 3.4 mA). When the injection current density increases from 0.0092 kA/cm2 to 0.0267 kA/cm2, the EL peak wavelength of LDI decreases from 474.4 nm to 467.4 nm, and that of LDII decreases from 468.3 nm to 465.9 nm. The injection-current-dependent wavelength show that the blue shift of the EL peak wavelength of LDI (7.0 nm) is much larger than that of LDII (2.4 nm).
Fig. 5. (a) and (b) Electroluminescence spectra under direct current injection of LDI and LDII before lasing. (c) and (d) Emission peak wavelength and full width at half maximum (FWHM) of the electroluminescence spectra for LDI (orange) and LDII (blue).
The mission peak and peak shift are the joint effect of the quantization effect of MQW structure, band-filling effect and quantum-confined Stark effect (QCSE). Since the thickness and indium content of quantum wells are unchanged for LDI and LDII, thus band-filling effect and QCSE are the main factors to influence the emission wavelength and blue shift in LDI. It is reported that carrier injection and its screening effect will compensate the QCSE during EL measurements, which will result in blue shift of emission peak [20–22]. In addition, band-filling happens and the transition energy of carrier recombination will raise up with increase of current injection, thus a blue shift of EL peak happens [23–25]. The value of blue shift is a good way to estimate how large is the strain [26–30]. In this work, it can be seen that the blue shift of LDI is larger than of LDII, indicating a stronger band-filling effect and a more suppressed QCSE in LDI. Therefore, it implies that the strain and piezoelectric field of the quantum wells in LDI is indeed larger than that of LDII, and it supports our previous suggestion on the reason for the better performance of LDII . Actually, for GaN-based LDs, the early investigations on the influence of strain have shown that a smaller strain is beneficial to improve the device performance, which can be attributed to the reduction of defects or increase of wave function overlap between electron and hole [31–33]. These reports give us a clue that, the suppressed strain and piezoelectric field in the asymmetric MQW could be one of the reasons for the improvement of device performance.
It is also worth noting that FWHM of EL spectra for LDI decreases from 29.3 nm to 27.6 nm when the current density increases from 0.0092 kA/cm2 to 0.0267 kA/cm2, and FWHM of EL spectra for LDII decreases from 26.6 nm to 24.3 nm. Actually, values of FWHM presents the homogeneity of MQW in optical transition energy. It means that the smaller FWHM, the more homogeneous of MQW. As expected, the value of EL spectral FWHM of LDII is smaller than that of LDI. Therefore, the material quality of MQW should be improved after the first quantum barrier is removed, which is beneficial to the improvement of device performance for LDII. Moreover, it is found that FWHM of EL spectra for LDI decreases first then increases, then decreases again. However, FWHM of EL spectra of LDII decreases monotonously when the injection current increases. On one side, it is reported that carrier injection will weaken the polarization effect and suppress the band tilt of quantum well, then the quantum energy level difference decreases, and then FWHM of EL spectra decreases when the injection current increases [34]. On the other side, more localized states will participate in carrier recombination with increase injection current and the energy band fill effect will happen, and then FWHM of EL spectra will increase [22,35]. It means that both of above-mentioned effects may play role and thus result in different injection current-related behaviors of the LD diodes. A larger strain results in a stronger energy band fill effect, thus an increase of FWHM would be more obvious in the EL spectra. Therefore an increase of FWHM for LDI is observed, indicating that a significantly larger strain exists in LD1. This result is consistent with the discussion on the blue shift of EL spectra.
To check whether there are other factors influencing the device performance in addition to the factor of material quality improvement during epitaxial growth, the optical performance and electrical injection of two LDs, i.e., LD1 and LD2, are calculated by a simulation through LASTIP. The device structures of LD1 and LD2 are designed as those of LDI and LDII, respectively, and the structure parameters are listed in Table 1. The conduction band diagrams under a 1200 mA injection current is shown in Fig. 6, which helps to verify the reason of higher injection ratio for LDII and LD2. Here, the electron barrier height ΔE is defined as the energy difference between the energy level in the lower waveguide layer and the first quantum barrier which is marked with the red double-headed arrows in Fig. 8. The value of ΔE represents the effective blocking barrier height for the electrons in the lower waveguide layer. The lower the electron barrier height, the easier the injection of electrons into the active region. According to the calculated result, the electron barrier height of LD1 and LD2 is 148 meV and 110 meV, respectively, being much lower in LD2 than in LD1, and then the electrons in LD2 can be more easily injected into the quantum well region. A higher injection ration would beneficial to improve the device performance, including in threshold current, optical power and slope efficiency, which is the other reason for the improved performance of LDII using asymmetric MQW. Detailed discussions on the injection ratio will be taken below according to the calculation of the electron concentration and electron current density distribution, which is shown in Fig. 8.
Fig. 6. Conduction band diagrams at an injection current of 1200 mA for LD1 and LD2. The dash lines mark the quasi-Fermi levels of electron.
Figure 7 shows the calculated output optical power versus current of LD1 and LD2. The threshold current of LD1 and LD2 is 132 mA and 110 mA, respectively, and the output optical powers of LD1 and LD2 reach 1.27 W and 1.45 W, respectively, when the injection current is 1.2 A. In addition, the slope efficiency (SE) of LD1 and LD2 is 1.18 W/A and 1.35 W/A, respectively. The threshold current, optical power, and slope efficiency of LD2 were improved by 16.7%, 14.2% and 14.4%, respectively, compared to those of LD1. It indicates that the device performance of LD2 with asymmetric quantum wells is much better than that of LD1 with symmetrical quantum wells. Furthermore, this calculated result of LD1 and LD2 is consistent with the discussion on LDI and LDII, which means that using asymmetric quantum wells without FQB is indeed beneficial to improve the device performance of GaN-based LDs.
Fig. 7. Optical power versus injection current of LD1 (orange) and LD2 (blue). SE presents the slope efficiency.
Fig. 8. (a) Electron concentration and (b) electron current density distribution around the MQW active region of LD1 (orange) and LD2 (blue) at an injection current of 1200 mA.
The electrical characteristics of LD1 and LD2 are also examined to verify the influence of the electron barrier height ΔE on the injection ratio. Figure 8 shows the distributions of electron concentration and electron current density around the MQW active region, where the distance abscissa is the position along the growth direction, and the injection current is fixed at 1200 mA. In Fig. 8(a), the electron concentration in the lower waveguide layer of LD1 is larger than that in LD2. This means that more electrons accumulate in the lower waveguide layer when the FQB is located between lower waveguide layer and the first quantum well layer, and they do not have contribution to the stimulation emission. It is noted that the In0.04Ga0.96N lower waveguide layer acts as the quantum barrier in LD2 instead of the In0.02Ga0.98N layer, its energy level is lower, thus the electrons accumulated in the lower waveguide layer of LD2 is less than in LD1, indicating a less waste of injected carriers. Figure 8(b) shows the electron current distribution in LD1 and LD2 around the MQW active region. The electron injection current is 1200 mA. Here, we consider the injection ratio of electron current which is defined as the ratio of the electron current in the last quantum barrier and injection current before the first quantum well. The calculated result shows that the electron injection current in the last quantum barrier of LD2 is 170 A/cm2 larger than LD1, which means the injection ratio of electron current in LD2 is larger than LD1. This result suggests a higher injection ratio for asymmetric MQW due to a lower electron barrier height ΔE, and this may be another factor why the device performance of LD2 or LDII is better than that of LD1 or LDI.
Furthermore, optical characteristics of LD1 and LD2 are also calculated. Figure 9 shows the calculated optical field distribution of LD1 and LD2 along the growth direction of LD1 and LD2, where the injection current is 1200 mA. It shows that for LD1 and LD2, the peak position of the optical field distribution is 2.122 µm and 1.233 µm, respectively, and the full width at half maximum of the optical field distribution is 0. 296 µm and 0. 227µm, respectively. A more concentrated optical field is beneficial to reduce the optical loss and increase the optical confinement. As shown in Fig. 9(b), the optical confinement factor of LD1 and LD2 is 1.6% and 2.1%, respectively, and the optical loss is 11.4 cm-1 and 10.0 cm-1, respectively. The optical characteristics indicate that the optical confinement and optical loss are improved while using asymmetric MQW. Actually, for the asymmetric MQW structure LD2, the refractive index of the In0.04Ga0.96N lower waveguide layer is larger than that of the In0.02Ga0.98N quantum barrier layer, meanwhile LWG becomes closer to the quantum wells after removing the first quantum barrier well. Therefore, the optical field distribution moves closer to n-type region and far away from the p-type area, thus optical confinement is enhanced and optical loss is reduced. This result indicates that the enhancement of optical confinement and reduction of the optical loss are the other reasons for the improvement of LDs using asymmetric MQW.
Fig. 9. (a) Optical field distribution around the active area along the growth direction of LD1 (orange) and LD2 (blue) at an injection current of 1200 mA. (b) The optical loss and the optical confinement factor of LD1 and LD2. The connecting lines are a guide to the eye.
4. Conclusion
Asymmetric MQW structure without the first quantum barrier layer is beneficial to improve the device performance of GaN-based LDs. Experimental and theoretical investigations show that there are three reasons for the improvement of device performance when the asymmetric MQW is used. One is that a more homogeneous MQW is obtained during the epitaxial growth after removing the first quantum barrier, which is helpful to suppress the strain and piezoelectric field of the quantum wells. In addition, the energy barrier height for the injection electrons is reduced in the asymmetric MQW device, which could enhance the electron injection efficiency. Furthermore, the optical field is more concentrated, resulting in an increase of optical confinement and a reduction of the optical loss. This new structure would be also better suitable for other LDs with longer lasing wavelengths because of the fact that the difference of the energy bang gap and refractive index between the lower waveguide layer and quantum barrier layer may be even larger.
Funding
National Key Research and Development Program of China (2021YFF0307403); National Natural Science Foundation of China (61874175, 61904172, 61974162, 62034008, 62074140, 62074142, 62127807); Beijing Nova Program (202093); Strategic Priority Research Program of Chinese Academy of Sciences (XDB43030101); Beijing Municipal Science& Technology Commission, Zhongguancun Science Park (Z211100004821001); Youth Innovation Promotion Association of the Chinese Academy of Sciences (2019115).
Disclosures
The authors declare that there are no conflicts of interest related to this article.
Data availability
Data underlying the results presented in this paper are not publicly available at this time but may be obtained from the authors upon reasonable request.
References
1. S. Nakamura, S. Masayuki, N. Shin-ichi, I. Naruhito, Y. Takao, M. Toshio, K. Hiroyuki, and S. Yasunobu, “InGaN-Based Multi-Quantum-Well-Structure Laser Diodes,” Jan,” J. Appl. Phys. 35(1B), L74–L76 (1996). [CrossRef]
2. S. Nakamura, S. Masayuki, N. Shin-ichi, I. Naruhito, Y. Takao, M. Toshio, K. Hiroyuki, S. Yasunobu, K. Tokuya, U. Hitoshi, S. Masahiko, and C. Kazuyuki, “Violet InGaN/GaN/AlGaN-Based Laser Diodes with an Output Power of 420 mW,” Jan,” J. Appl. Phys. 37(6A), L627–L629 (1998). [CrossRef]
3. I. Akasaki, S. Sota, H. Sakai, T. Tanaka, M. Koike, and H. Amano, “Shortest wavelength semiconductor laser diode,” Electron. Lett. 32(12), 1105 (1996). [CrossRef]
4. H. Z. Wang, Y. Kawahito, R. Yoshida, Y. Y. Nakashima, and K. Shiokawa, “Development of a high-power blue laser (445 nm) for material processing,” Opt. Lett. 42(12), 2251 (2017). [CrossRef]
5. M. Baumann, A. Balck, J. Malchus, R. V. Chacko, S. Marfels, U. Witte, D. Dinakarana, S. Ocylok, M. Weinbach, C. Bachert, A. Kösters, V. Krause, H. König, A. Lell, B. Stojetz, M. Ali, and U. Strauss, “1000 W blue fiber-coupled diode laser emitting at 450 nm,” Proc. of SPIE 10900, 3 (2019). [CrossRef]
6. W. B. Simon, S. LaurensS, M. Thomas, and R. Markus, “Blue high-power laser sources for processing solutions in e-mobility and beyond,” Procedia CIRP 94, 592–595 (2020). [CrossRef]
7. C. H. Kang, A. Trichili, O. Alkhazragi, H. F. Zhang, R. C. Subedi, Y. J. Guo, S. Mitra, C. Shen, I. S. Roqan, T. K. Ng, M. S. Alouini, and B. S. Ooi, “Ultraviolet-to-blue color-converting scintillating-fibers photoreceiver for 375-nm laser-based underwater wireless optical communication,” Opt. Express 27(21), 30450 (2019). [CrossRef]
8. C. Shen, O. Alkhazragi, X. B. Sun, Y. J. Guo, T. K. Ng, and B. Boon, “Laser-based visible light communications and underwater wireless optical communications: a device perspective,” Proc. of SPIE 10939, 13 (2019). [CrossRef]
9. X. Q. Yang, Z. J. Tong, Y. Z. Dai, X. Chen, H. Zhang, H. W. Zou, and J. Xu, “100mfull-duplex underwater wireless optical communication based on blue and green lasers and high sensitivity detectors,” Opt. Commun. 498, 127261 (2021). [CrossRef]
10. I. Akasaki, S. Kamiyama, T. Detchprohm, T. Takeuchi, and H. Amano, “Growth of crack-free thick AlGaN layer and its application to GaN-based laser diode,” Mater. Res. Soc. Symp. Proc. 595, F99W6.8 (1999). [CrossRef]
11. T. Kamikawa, S. Gandrothula, M. Araki, H. J. Li, V. B. Oliva, F. Wu, D. Cohen, J. S. Speck, S. P. Denbaars, and S. Nakamura, “Realization of thin-film m-plane InGaN laser diode fabricated by epitaxial lateral overgrowth and mechanical separation from a reusable growth substrate,” Opt. Express 27(17), 24717 (2019). [CrossRef]
12. J. X. Liu, H. R. Qie, Q. Sun, M. X. Feng, J. Wang, X. J. Sun, X. Sheng, M. Ikeda, and H. Yang, “Enhanced carrier confinement and radiative recombination in GaN-based lasers by tailoring first-barrier doping,” Opt. Express 28(21), 32124 (2020). [CrossRef]
13. S. H. Park, H. M. Kim, and D. Ahn, “Optical gain in GaN quantum well lasers with quaternary AlInGaN barriers,” Jpn. J. Appl. Phys. 44(10), 7460–7463 (2005). [CrossRef]
14. S. H. Park, H. M. Kim, and D. Ahn, “Optical gain in GaN quantum well lasers with quaternary AlInGaN barriers,” Appl. Phys. Lett. 90(2), 023508 (2007). [CrossRef]
15. J. Yang, D. G. Zhao, D. S. Jiang, P. Chen, J. J. Zhu, Z. S. Liu, F. Liang, W. Liu, S. T. Liu, and M. Li, “Enhancing the performance of GaN based LDs by using low In content InGaN instead of GaN as lower waveguide layer,” Opt. Laser Technol. 111, 810–813 (2019). [CrossRef]
16. F. Liang, D. D. Zhao, D. S. Jiang, Z. S. Liu, J. J. Zhu, P. Chen, J. Yang, W. Liu, S. T. Liu, Y. Xiang, L. q. Zhang, W. J. Wang, M. Li, Y. T. Zhang, and G. T. Du, “Improvement of slope efficiency of GaN-Based blue laser diodes by using asymmetric MQW and InxGa1-xN lower waveguide,” Opt. Laser Technol. 731, 243–247 (2018). [CrossRef]
17. G. M. Laws, E. C. Larkins, I. Harrison, C. Molloy, and D. Somerford, “Improved refractive index formulas for the AlxGa1−xN and InyGa1−yN alloys,” J. Appl. Phys. 89(2), 1108–1115 (2001). [CrossRef]
18. J. Piprek, T. Peng, G. Qui, and J. O. Olowolafe, “Energy gap bowing and refractive index spectrum of AlInN and AlGaInN,” Proc. IEEE Conf. Compound Semiconductors 227, 227–230 (1997). [CrossRef]
19. T. Peng and J. Piprek, “Refractive index of AlGaInN alloys,” Electron. Lett. 32(24), 2285–2286 (1996). [CrossRef]
20. S. C. Tsai, C. H. Lu, and C. P. Liu, “Piezoelectric effect on compensation of the quantum-confined Stark effect in nGaN/GaN multiple quantum wells based green light-emitting diodes,” Nano Energy 28, 373–379 (2016). [CrossRef]
21. D. A. Miller, D. Chemla, T. Damen, A. Gossard, W. Wiegmann, T. Wood, and C. Burrus, “Band-edge electro absorption in quantum well structures: the quantum-confined stark effect,” Phys. Rev. Lett. 53(22), 2173–2176 (1984). [CrossRef]
22. T. Wang, D. Nakagawa, J. Wang, T. Sugahara, and S. Sakai, “Photoluminescence investigation of InGaN/GaN single quantum well and multiple quantum wells,” Appl. Phys. Lett. 73(24), 3571–3573 (1998). [CrossRef]
23. P. Debernardi and P. Fasano, “Quantum confined stark effect in semiconductor quantum wells including valence band mixing and coulomb effects,” IEEE J. Quantum Electron. 29(11), 2741–2755 (1993). [CrossRef]
24. L. H. Peng, C. W. Chuang, and L. H. Lou, “Piezoelectric effects in the optical properties of strained InGaN quantum wells,” Appl. Phys. Lett. 74(6), 795–797 (1999). [CrossRef]
25. G. B. Xu, G. Sun, Y. J. J. Ding, H. P. Zhao, G. Y. Liu, J. Zhang, and N. Tansu, “Investigation of large Stark shifts in InGaN/GaN multiple quantum wells,” J. Appl. Phys. 113(3), 033104 (2013). [CrossRef]
26. C. Y. Cho and S. J. Park, “Enhanced optical output and reduction of the quantum-confined Stark effect in surface plasmon-enhanced green light-emitting diodes with gold nanoparticles,” Opt. Express 24(7), 7488 (2016). [CrossRef]
27. D. M. Van Den Broeck, D. Bharrat, A. M. Hosalli, N. A. El-Masry, and S. M. Bedair, “Strain-balanced InGaN/GaN multiple quantum wells,” Appl. Phys. Lett. 105(3), 031107 (2014). [CrossRef]
28. M. Benaissa, W. Sigle, M. Korytov, J. Brault, P. Vennegues, and P. A. van Aken, “Plasmon energy from strained GaN quantum wells,” Appl. Phys. Lett. 103(2), 021901 (2013). [CrossRef]
29. L. Y. Chen, H. H. Huang, C. H. Chang, Y. Y. Huang, Y. R. Wu, and J. J. Huang, “Investigation of the strain induced optical transition energy shift of the GaN nanorod light emitting diode arrays,” Opt. Express 19(S4), A900 (2011). [CrossRef]
30. K. S. Ramaiah, Y. K. Su, S. J. Chang, B. Kerr, H. P. Liu, and I. G. Chen, “Characterization of InGaNÕGaN multi-quantum-well blue-light-emitting diodes grown by metal organic chemical vapor deposition,” Appl. Phys. Lett. 84(17), 3307–3309 (2004). [CrossRef]
31. A. Paliwal, K. Singh, and M. Mathew, “Strain-free GaN/InAlN chirped short-period superlattice electron-blocking layer for 450 nm InGaN laser diode,” Laser Phys. 29(5), 056204 (2019). [CrossRef]
32. Y. C. Yeo, T. C. Chong, and M. F. Li, “Uniaxial Strain Effect on the Electronic and Optical Properties of Wurtzite GaN-AlGaN Quantum-Well Lasers,” IEEE J. Quantum Electron. 34(11), 2224–2232 (1998). [CrossRef]
33. S. Nakamura, M. Senoh, S. Nagahama, N. Iwasa, T. Yamada, T. Matsushita, H. Kiyoku, Y. Sugimoto, T. Kozaki, H. Umemoto, M. Sano, and K. Chocho, “InGaN/GaN/AlGaN-based laser diodes with modulation-doped strained-layer superlattices grown on an epitaxially laterally overgrown GaN substrate,” Appl. Phys. Lett. 72(2), 211–213 (1998). [CrossRef]
34. S. Grzanka, G. Franssen, G. Targowski, R. Czernecki, A. Khachapuridze, I. Makarowa, R. Wisniewska, P. Mensz, P. Perlin, and T. Suski, “Carrier recombination mechanisms in nitride single quantum well light-emitting diodes revealed by photo- and electroluminescence,” J. Appl. Phys. 104(9), 094504 (2008). [CrossRef]
35. Y. J. Lee, C. H. Chiu, C. C. Ke, P. C. Lin, T. C. Lu, H. C. Kuo, and S. C. Wang, “Study of the Excitation Power Dependent Internal Quantum Efficiency in InGaN/GaN LEDs Grown on Patterned Sapphire Substrate,” IEEE J. Sel. Top. Quantum Electron. 15(4), 1137–1143 (2009). [CrossRef]
