記住我
In this work, LA-MOCVD n-GaN growth on ammonothermal freestanding GaN substrates was investigated under different Rg regime. A theoretical mass-transport based model was proposed to understand the correlations between C incorporation and GaN growth rate Rg. Approaches are suggested based on the experimental results and the theoretical modeling to further control C impurity incorporation at fast Rg.
Samples were grown on ammonothermal GaN substrates (threading dislocation density TDD 4 cm−2) in a commercial R&D vertical rotating disk MOCVD reactor. The details of the LA-MOCVD system are introduced in Ref. 1010. Y. Zhang, Z. Chen, K. Zhang, Z. Feng, and H. Zhao, “ Laser-assisted metal–organic chemical vapor deposition of gallium nitride,” Phys. Status Solidi RRL 15(6), 2100202 (2021). https://doi.org/10.1002/pssr.202100202. The laser power was set at 250 W. A secondary ion mass spectroscopy (SIMS) epi-structure was designed to quantitatively characterize the impurity levels under different growth conditions. The TMGa molar flow rate for each sublayer was varied from 36.25 to 911.45 μmol/min while keeping the other growth conditions the same. NH3 molar flow was kept constant for all growth conditions at 0.179 mol/min. For each condition, both conventional MOCVD (noted as MOCVD for the rest of the content) and LA-MOCVD growth techniques were used to grow GaN sublayers to compare the impurity incorporations. The SiH4 molar flow rate was kept as a constant (0.7 nmol/min) for all sublayer growths. Figure 1(a) shows the SIMS depth profiles of [C], [O], [H], and [Si]. The detection limits (DL) for [C], [O], [H], and [Si] are 1 × 1015, 5 × 1015, 5 × 1015, and 1 × 1015 cm−3, respectively.As shown in Fig. 1(a), the [O] is below the SIMS DL under all different TMGa flows with or without laser irradiations. The [Si] and [C], on the other hand, show strong dependence on TMGa molar flow rate and laser irradiation. Figures 1(b) and 1(c) plot the extracted [Si] and [C] from Fig. 1(a) as a function of the TMGa molar flow rate for both MOCVD and LA-MOCVD growths. The [Si] is inversely proportional to the TMGa molar flow rate, for both MOCVD and LA-MOCVD growths. GaN growth rate increases as the TMGa flow rate increases; thus, Si incorporation in GaN reduces as TMGa flow rate increases.1212. D. D. Koleske, A. E. Wickenden, R. L. Henry, and M. E. Twigg, “ Influence of MOVPE growth conditions on carbon and silicon concentrations in GaN,” J. Cryst. Growth 242(1–2), 55–69 (2002). https://doi.org/10.1016/S0022-0248(02)01348-9 As the TMGa flow rate reaches 600 μmol/min, LA-MOCVD sublayers show slightly higher [Si] as compared to those of MOCVD sublayers. This is due to the slightly reduced growth rates in LA-MOCVD grown layers from gas phase reaction when TMGa flow rate is high.1010. Y. Zhang, Z. Chen, K. Zhang, Z. Feng, and H. Zhao, “ Laser-assisted metal–organic chemical vapor deposition of gallium nitride,” Phys. Status Solidi RRL 15(6), 2100202 (2021). https://doi.org/10.1002/pssr.202100202The [C] shows strong positive correlations with the TMGa flow rate, as shown in Fig. 1(c). The high TMGa flow rate increases the C/N ratio in the gas phase, and C is energetically preferred to occupy the N sites in GaN,1313. J. L. Lyons, A. Janotti, and C. G. Van De Walle, “ Effects of carbon on the electrical and optical properties of InN, GaN, and AlN,” Phys. Rev. B 89(3), 035204 (2014). https://doi.org/10.1103/PhysRevB.89.035204 which lead to the increased C incorporation as TMGa flow rate increases. The [C] in the LA-MOCVD grown GaN layers are much lower as compared to those in the MOCVD grown layers at different TMGa molar flow rate. The reduction in [C] is more prominent at higher TMGa flow rates. In addition, from Fig. 1(a), the [H] in the MOCVD grown sublayers increases significantly in the sublayers with high TMGa flow. In contrast, the [H] in the LA-MOCVD grown sublayers maintain below the DL. This could indicate that part of the carbon incorporation in GaN is in the form of hydrocarbon groups [such as CH3 and (CH3)Ga]. Due to the limitation of the SIMS DL, it is inconclusive if [H] follows the same trend at lower TMGa flow rates. LA-MOCVD growth leads to more efficient cracking efficiency of NH3, resulting in an increase in the number of H-radicals that can react with active C-species (such as CH3) more readily to form stable compounds such as CH4. This can explain the lower [H] concentrations observed within the LA-MOCVD layers at the high TMGa flows as compared to the same conditions with the conventional MOCVD growth. The relatively higher H incorporation at faster growth rate conditions indicates TMGa is also the source of H radicals.In order to further understand the dependence of impurity/dopant incorporation on the GaN growth rates, the [Si] is plotted as a function of Rg, as shown in Fig. 2(a). The GaN growth rates Rg were estimated based on the thicknesses extracted from the SIMS profile and the corresponding growth time for each layer. The [Si] in both MOCVD and LA-MOCVD layers are inversely proportional to the Rg. A comparison of the two sets of data indicates that LA-MOCVD growth does not impact the Si incorporation as compared to the conventional MOCVD growth.Figure 2(b) plots the [C] as a function of Rg for both MOCVD and LA-MOCVD grown layers and compared them with representative data points from literature reports.8–10,14–208. T. Ciarkowski, N. Allen, E. Carlson, R. McCarthy, C. Youtsey, J. Wang, P. Fay, J. Xie, and L. Guido, “ Connection between carbon incorporation and growth rate for GaN epitaxial layers prepared by OMVPE,” Materials 12(15), 2455 (2019). https://doi.org/10.3390/ma121524559. Y. Zhang, Z. Chen, W. Li, A. R. Arehart, S. A. Ringel, and H. Zhao, “ Metalorganic chemical vapor deposition gallium nitride with fast growth rate for vertical power device applications,” Phys. Status Solidi A 218(6), 2000469 (2021). https://doi.org/10.1002/pssa.20200046910. Y. Zhang, Z. Chen, K. Zhang, Z. Feng, and H. Zhao, “ Laser-assisted metal–organic chemical vapor deposition of gallium nitride,” Phys. Status Solidi RRL 15(6), 2100202 (2021). https://doi.org/10.1002/pssr.20210020214. Y. Cao, R. Chu, R. Li, M. Chen, R. Chang, and B. Hughes, “ High-voltage vertical GaN Schottky diode enabled by low-carbon metal-organic chemical vapor deposition growth,” Appl. Phys. Lett. 108, 062103 (2016). https://doi.org/10.1063/1.494181415. G. Piao, K. Ikenaga, Y. Yano, H. Tokunaga, A. Mishima, Y. Ban, T. Tabuchi, and K. Matsumoto, “ Study of carbon concentration in GaN grown by metalorganic chemical vapor deposition,” J. Cryst. Growth 456, 137–139 (2016). https://doi.org/10.1016/j.jcrysgro.2016.08.03016. F. Kaess, S. Mita, J. Xie, P. Reddy, A. Klump, L. H. Hernandez-Balderrama, S. Washiyama, A. Franke, R. Kirste, A. Hoffmann, R. Collazo, and Z. Sitar, “ Correlation between mobility collapse and carbon impurities in Si-doped GaN grown by low pressure metalorganic chemical vapor deposition,” J. Appl. Phys. 120(10), 105701 (2016). https://doi.org/10.1063/1.496201717. K. Matsumoto, H. Tokunaga, A. Ubukata, K. Ikenaga, Y. Fukuda, T. Tabuchi, Y. Kitamura, S. Koseki, A. Yamaguchi, and K. Uematsu, “ High growth rate metal organic vapor phase epitaxy GaN,” J. Cryst. Growth 310(17), 3950–3952 (2008). https://doi.org/10.1016/j.jcrysgro.2008.06.00918. A. E. Wickenden, D. D. Koleske, R. L. Henry, M. E. Twigg, and M. Fatemi, “ Resistivity control in unintentionally doped GaN films grown by MOCVD,” J. Cryst. Growth 260(1–2), 54–62 (2004). https://doi.org/10.1016/j.jcrysgro.2003.08.02419. J. Yang, D. G. Zhao, D. S. Jiang, P. Chen, Z. S. Liu, L. C. Le, X. J. Li, X. G. He, J. P. Liu, S. M. Zhang, H. Wang, J. J. Zhu, and H. Yang, “ Investigation on the compensation effect of residual carbon impurities in low temperature grown mg doped GaN films,” J. Appl. Phys. 115(16), 163704 (2014). https://doi.org/10.1063/1.487395720. T. Narita, M. Horita, K. Tomita, T. Kachi, and J. Suda, “ Why do electron traps at EC –0.6 EV have inverse correlation with carbon concentrations in N-type GaN layers?,” Jpn. J. Appl. Phys., Part 1 59(10), 105505 (2020). https://doi.org/10.35848/1347-4065/abb9ca There is a significant difference of [C] incorporation as a function of Rg between MOCVD and LA-MOCVD. With similar growth rates, [C] in LA-MOCVD grown layers are about 50% to 90% lower than that in MOCVD grown layers. Two possible mechanisms could contribute to the significant reduction: first, NH2, and NH have smaller adsorption energies as compared to NH3, which enable higher adsorption on the growth surface and less C incorporation. Second, the laser-assisted higher NH3 decomposition creates more H radicals, which reacts with hydrocarbon radicals such as CH3Ga and CH3, forming more chemically stable CH4, and thus, reduces the C incorporation in GaN. It is noted that as compared with our previous studies of MOCVD GaN on GaN-on-sapphire templates, the LA-MOCVD method shows more prominent [C] reduction in GaN grown on freestanding GaN substrates. This indicates that the threading dislocation density could play an important role on the C incorporation in GaN MOCVD. Higher dislocation density in GaN grown on GaN-on-sapphire templates could facilitate higher C incorporation. As illustrated in Fig. 2(b), in contrast to [Si], the [C] extracted from this study shows a more complex relationship with Rg. For relatively slow Rg (μm/h), the [C] increases linearly with Rg. With the increase in Rg between 3 and 10 μm/h, the [C] shows a much steeper increase, as Rg increases for both MOCVD and LA-MOCVD growths. It is also seen that despite different growth conditions, the lower bound of [C] from different reports shows a similar trend that [C] follows a non-linear relationship with Rg. It suggests different C incorporation mechanisms under different growth rate regimes and poses the challenge to lower [C] at fast Rg.To facilitate a quantitative understanding, a theoretical model is proposed to elaborate the correlations between TMGa flow rate, [C], and Rg. Previously, Burton, Prim, and Slichter developed a mass-transport based model (BPS model) to describe the process of solute element incorporation into bulk during crystal growth.2121. J. A. Burton, R. C. Prim, and W. P. Slichter, “ The distribution of solute in crystals grown from the melt. Part I. Theoretical,” J. Chem. Phys. 21(11), 1987–1991 (1953). https://doi.org/10.1063/1.1698728 The effective distribution coefficient k is defined as where CS (CL) refers to the ratio of the impurity concentration to the total atomic concentration in the solid (in the bulk of the melt/solution). For the case k k=k0k0+1−k0 exp (−VδD ),(2)where k0 is the equilibrium distribution coefficient, V is the growth velocity at the growth interface, δ is the boundary layer thickness, and D is the bulk diffusion coefficient of the impurity atoms. Here, k0 is defined as the ratio of impurity concentration in the solid and on the adsorption layer (C0), It is apparent that typical [C] in GaN epilayer is lower than the C concentration in the gas-phase during the MOCVD growth. Therefore, Eq. (2) is valid in this case. For MOCVD GaN growth with step-flow growth mode, the adatoms, including both source material atoms and impurities, incorporate into the solid through the growth of atomic steps on the growth surface. The macroscopic growth rate, Rg, represents the vertical component of the step velocity Vstep.2222. L. Jiang, J. Liu, A. Tian, X. Ren, S. Huang, W. Zhou, L. Zhang, D. Li, S. Zhang, M. Ikeda, and H. Yang, “ Influence of substrate misorientation on carbon impurity incorporation and electrical properties of p-GaN grown by metalorganic chemical vapor deposition,” Appl. Phys. Express 12(5), 055503 (2019). https://doi.org/10.7567/1882-0786/ab0da2Therefore, Rg can be expressed as follows:2222. L. Jiang, J. Liu, A. Tian, X. Ren, S. Huang, W. Zhou, L. Zhang, D. Li, S. Zhang, M. Ikeda, and H. Yang, “ Influence of substrate misorientation on carbon impurity incorporation and electrical properties of p-GaN grown by metalorganic chemical vapor deposition,” Appl. Phys. Express 12(5), 055503 (2019). https://doi.org/10.7567/1882-0786/ab0da2 where θ is the off-cut angle of the substrate. Based on the above analysis, V in Eq. (2) denotes Vstep in the step flow growth mode. The boundary layer thickness corresponds to the thickness of the adsorption layer. The boundary layer thickness is roughly estimated within previous work to be within the range of one atomic layer.2323. T. Nishinaga, C. Sasaoka, and K. Pak, “ Study of nitrogen inhomogeneity in LPE GaP by spatially resolved photoluminescence,” Jpn. J. Appl. Phys., Part 1 28(5R), 836 (1989). https://doi.org/10.1143/JJAP.28.836 The diffusion constant, D, is on the order of 10−13 cm2/s, as measured by concentration-depth profiles.2424. X. A. Cao, R. G. Wilson, J. C. Zolper, S. J. Pearton, J. Han, R. J. Shul, D. J. Rieger, R. K. Singh, M. Fu, V. Scarvepalli, and J. A. Sekhar, “ Redistribution of implanted dopants in GaN,” J. Electron. Mater. 28(3), 261–265 (1999). https://doi.org/10.1007/s11664-999-0025-yCombining Eqs. (1), (2), and (4), one can obtain Cs=CLk0k0+1−k0 exp (−Rg·δD· sin θ ).(5)In the case of step flow growth, CL is the ratio of the C impurity concentration to the total atomic concentration at the growth interface, which can be considered as the impurity that can be adsorbed onto the surface from the gas phase.
Therefore, where fTMGa and ftotal represent the TMGa and total gas molar flow rate, respectively. ftotal includes the flow of TMGa, NH3 as well as the carrier gas (H2). The constant 3 represents the three -CH3 structures attached to one Ga atom within a single TMGa molecule. n is defined as the C adsorption probability during the adsorption process. There are different C species in the gas phase [CH4, CH3, (CH3)Ga, etc.], in which only a small portion of the C species [CH3, (CH3)Ga] participate in the adsorption process.2525. Ö. Danielsson, X. Li, L. Ojamäe, E. Janzén, H. Pedersen, and U. Forsberg, “ A model for carbon incorporation from trimethyl gallium in chemical vapor deposition of gallium nitride,” J. Mater. Chem. C 4(4), 863–871 (2016). https://doi.org/10.1039/C5TC03989D Considering that in GaN MOCVD, the V/III ratio is normally above 100, we can consider ftotal a constant if TMGa is the only variable.From the experiment, we have established the relationship between the carbon concentration, TMGa flow rate, and Rg. It is well recognized that many parameters can impact the Rg, including the growth temperature and pressure. Nevertheless, the TMGa flow rate is the most significant factor that directly impacts GaN Rg. Typical GaN MOCVD growths are under N-rich condition, thus, Rg is limited by transport of Ga species. Here, one can assume Rg is proportional to the TMGa flow rate, where b is defined as the growth efficiency coefficient. Eq. (5) can be further derived as [C]=3n·Rgb·ftotalρGaNk0k0+1−k0exp(−Rg·δD· sin θ ).(8)As shown in Fig. 3(a), the growth efficiency b is extracted as 3.16 × 10−2 and 2.23 × 10−2 cm/mol for MOCVD and LA-MOCVD growths, respectively. ρGaN is the molecular density of GaN and is considered as 8.9 × 1022 cm−3. Next, [C] as a function of Rg as obtained from SIMS results and shown in Fig. 3(b) are fitted by Eq. (8). Table I lists the parameters used for the fitting and the extracted fitting results.
TABLE I. Constant parameters used in Eq. (8) to fit the experimental data in Fig. 3 and the resulting fitted parameters.ParameterMOCVDLA-MOCVDδ (cm)2 × 10–8Constant valuesθ (°)0.3ftotal (mol/min)0.178b (cm/mol)3.16 × 10–22.23 × 10–2Fitted from Eq. (7)k00.06900.1340Fitted from Eq. (8)n8.22 × 10–41.36 × 10–4D (cm2/s)3.085 × 10–133.109 × 10–13It is compelling that the developed theoretical model fits well with the experimental data. The diffusion coefficient D for both MOCVD and LA-MOCVD is around 3 × 10−13 cm2/s, indicating that the laser excitation has no significant impact on the diffusion properties of C atoms in the lattice. LA-MOCVD, however, shows a significant lower C adsorption probability n. Two possible mechanisms can affect the C incorporation in LA-MOCVD GaN: (i) More H radicals are generated through laser-assisted NH3 decomposition, which react with active C-species, such as CH3 and MMGa (CH3Ga), turning them into more inert species such as CH4. (ii) The cracked species of ammonia, NH2, and NH have lower adsorption energies as compared to that of NH3.29,3029.
H. R. Golgir,
Y. S. Zhou,
D. Li,
K. Keramatnejad,
W. Xiong,
M. Wang,
L. J. Jiang,
X. Huang,
L. Jiang,
J. F. Silvain, and
Y. F. Lu, “
Resonant and nonresonant vibrational excitation of ammonia molecules in the growth of gallium nitride using laser-assisted metal organic chemical vapour deposition,” J. Appl. Phys.
120(10), 105303 (2016). https://doi.org/10.1063/1.496242630.
H. Suzuki,
R. Togashi,
H. Murakami,
Y. Kumagai, and
A. Koukitu, “
Ab-initio calculation for an initial growth process of GaN on (0001) and (000 1¯) surfaces by vapor phase epitaxy,” Phys. Status Solidi C
6(S2), S301–S304 (2009). https://doi.org/10.1002/pssc.200880805 This enables a higher adsorption on the growth surface and, thus, reduces C incorporation. It is currently unclear as to which mechanism dominates, and if the dominant mechanism depends on the growth rate. However, the overall effect of either of these mechanisms results in a reduction in the C adsorption probability, which results in a reduction in the value of n. It is worth noting that LA-MOCVD results in a higher k0. Under the same equilibrium condition, less C is absorbed onto the surface under LA-MOCVD, thus the denominator in Eq. (3) is less, which gives a larger k0.In MOCVD growth, arriving adatoms, including both source and material atoms and impurities, move along the surface until they typically reach a step-edge or kink site. Some adatoms get incorporated within the crystal lattice through adsorption, while other adatoms that arrive at the surface go through desorption and do not get incorporated. In the case of step flow growth mode, adatoms, including both source material atoms and impurities, are incorporated at the step edges and kinks where the incorporation energies are lower.2626.
J. E. Ayers,
T. Kujofsa,
P. Rago, and
J. Raphael, Heteroepitaxy of Semiconductors: Theory, Growth, and Characterization, 2nd ed. (
CRC Press, 2016). Therefore, the impurity incorporation during crystal growths depends on the adsorption and desorption rates of the impurity species, as well as the step velocity. Both Rg and C concentration in gas phase are proportional to TMGa flow rate. At low Rg, the step velocity is low, the C species on the surface are within the adsorption-desorption equilibrium. Under this condition, increasing Rg does not significantly increase the C incorporation through the step edges. Higher TMGa flow rate increases the C-containing-species partial pressure in the gas phase, thus [C] shows a linear correlation with Rg. While further increasing the TMGa flow rate, the step velocity increases to a certain extent where some of the surface C species are captured by the step edges before being desorbed. Under this condition, both higher C in the gas phase and the lower desorption of surface C species contribute to the increased C incorporation. Therefore, [C] shows a steeper increase with Rg as shown in Fig. 2(b). When further increasing Rg by continuing to increase the TMGa flow rate, the step velocity is fast enough that most of the surface C species are captured by the step edges before being desorbed.From Eq. (8), [C] inevitably increases with Rg, which has also been observed from experiments. Meanwhile, there exist several strategies to suppress [C] even at high Rg regimes. First, as shown in Fig. 3(b), the C adsorption probability n can be greatly suppressed by using the LA-MOCVD technique. Second, based on the developed theoretical model, the increase in the off-cut angle θ can further suppress [C] incorporation. With a larger θ, lower step velocity is required to reach the same Rg, and more surface C species can be desorbed before reaching the step edges, resulting in lower [C] incorporation.Figure 4(a) plots the calculated [C] as a function of θ under different Rg. As the off-cut angle increases, the general trend shows that [C] decreases. This effect is more prominent at fast growth rates. The simulated results assumes that step-flow growth is maintained for the full range of Rg values, and the overall growth conditions are the same as the experimental conditions utilized for MOCVD growth in this work. Indeed, previous experimental studies have revealed that impurity level is negatively correlated with θ.22,27,2822.
L. Jiang,
J. Liu,
A. Tian,
X. Ren,
S. Huang,
W. Zhou,
L. Zhang,
D. Li,
S. Zhang,
M. Ikeda, and
H. Yang, “
Influence of substrate misorientation on carbon impurity incorporation and electrical properties of p-GaN grown by metalorganic chemical vapor deposition,” Appl. Phys. Express
12(5), 055503 (2019). https://doi.org/10.7567/1882-0786/ab0da227.
A. Tanaka,
Y. Ando,
K. Nagamatsu,
M. Deki,
H. Cheong,
B. Ousmane,
M. Kushimoto,
S. Nitta,
Y. Honda, and
H. Amano, “m-plane GaN Schottky Barrier Diodes fabricated with MOVPE layer on several off-angle m-plane GaN substrates,” Phys. Status Solidi A
215(9), 1700645 (2017). https://doi.org/10.1002/pssa.20170064528.
K. Nagamatsu,
Y. Ando,
T. Kono,
H. Cheong,
S. Nitta,
Y. Honda,
M. Pristovsek, and
H. Amano, “Effect of substrate misorientation on the concentration of impurities and surface morphology of an epitaxial GaN layer on n-polar GaN substrate by MOVPE,” J. Cryst. Growth
512, 78–83 (2019). https://doi.org/10.1016/j.jcrysgro.2019.02.013 Therefore, by combining the use of substrates with large off-cut angles and the LA-MOCVD technique, one can expect lower [C] incorporation, especially at the fast Rg regime.Figure 4(b) plots [C] as a function of GaN Rg for both conventional MOCVD and LA-MOCVD growths (experimental data), as well as the simulated [C] employing different off-cut angles and laser excitation power. Using the GaN substrate with a small off-cut angle (θ = 0.3°) and a conventional MOCVD growth technique, [C] is estimated as ∼5 × 1017 cm−3 at a Rg of 10 μm/h. In contrast, by utilizing the LA-MOCVD growth technique and θ = 3°, [C] is expected to reduce by more than one magnitude ([C]∼3 × 1016 cm−3) at a similar growth rate. Using the same laser excitation power, the further increase in the off-cut angle from θ = 3° to 10° does not lead to significant reduction of [C]. However, by increasing the laser excitation power, here we assume n is reduced from 1.36 × 10−4 to 0.68 × 10−4, [C] can be further reduced to 1 × 1016 cm−3 at Rg of 10 μm/h.In conclusion, we demonstrated a high quality GaN LA-MOCVD epitaxy on a free-standing GaN substrate with a fast growth rate and low [C], critical for vertical power device applications. [C] in LA-MOCVD GaN is reduced by 50%–90% as compared to the conventional MOCVD growth within a wide Rg range between 1 and 16 μm/h. [C] in both conventional and LA-MOCVD grown GaN shows a non-linear increase with Rg. A mass-transport based model is developed to understand the C incorporation at different GaN Rg regimes, which reveals that LA-MOCVD can effectively suppress C incorporation by reducing the active C species in the gas phase. Moreover, high step velocity in step flow growth mode can facilitate the C incorporation at fast Rg, exhibiting steeper C increase. Our investigation suggests that [C] can be further suppressed to below 1016 cm−3 with a fast Rg of 10 μm/h by utilizing high power LA-MOCVD technique and GaN substrates with relatively large off-cut angles.
The information, data, or work presented herein was funded in part by the Advanced Research Projects Agency-Energy (ARPA-E), U.S. Department of Energy, under Award No. DE-AR0001036, and the U.S. Department of Energy's Office of Energy Efficiency and Renewable Energy (EERE) under the Advanced Manufacturing Office, FY18/FY19 Lab Call. The views and opinions of authors expressed herein do not necessarily state or reflect those of the United States Government or any agency thereof.
Conflict of Interest
The authors have no conflicts to disclose.
Author Contributions
Yuxuan Zhang and Vijay Gopal Thirupakuzi Vangipuram contributed equally to this work.
Yuxuan Zhang: Conceptualization (equal); Formal analysis (equal); Investigation (equal); Writing – original draft (lead); Writing – review & editing (equal). Vijay Gopal Thirupakuzi Vangipuram: Formal analysis (equal); Investigation (equal); Validation (equal); Writing – original draft (equal); Writing – review & editing (equal). Kaitian Zhang: Investigation (supporting). Hongping Zhao: Conceptualization (lead); Formal analysis (equal); Funding acquisition (lead); Investigation (equal); Methodology (equal); Project administration (lead); Resources (lead); Supervision (lead); Validation (equal); Writing – original draft (supporting); Writing – review & editing (equal).
The data that support the findings of this study are available from the corresponding author upon reasonable request.
REFERENCES
1. H. Ohta, K. Hayashi, F. Horikiri, T. Nakamura, and T. Mishima, “ 5.0 kV breakdown-voltage vertical GaN p-n junction diodes,” Jpn. J. Appl. Phys, Part 1 57(4S), 04FG09 (2018). https://doi.org/10.7567/JJAP.57.04FG09, Google ScholarCrossref2. L. Yates, B. P. Gunning, M. H. Crawford, J. Steinfeldt, M. L. Smith, V. M. Abate, J. R. Dickerson, A. M. Armstrong, A. Binder, A. A. Allerman, and R. J. Kaplar, “ Demonstration of >6.0-kV breakdown voltage in large area vertical GaN p-n diodes with step-etched junction termination extensions,” IEEE Trans. Electron Devices 69(4), 1931–1937 (2022). https://doi.org/10.1109/TED.2022.3154665, Google ScholarCrossref3. V. Talesara, Y. Zhang, Z. Chen, H. Zhao, and W. Lu, “ Design and development of 1.5 kV vertical GaN pn diodes on HVPE substrate,” J. Mater. Res. 36(24), 4919–4926 (2021). https://doi.org/10.1557/s43578-021-00435-8, Google ScholarCrossref4. V. Talesara, Y. Zhang, Z. Chen, H. Zhao, and W. Lu, “ GaN power p–n diodes on hydride vapor epitaxy GaN substrates with near-unity ideality factor and <0.5 mΩ cm2 specific on-resistance,” Phys. Status Solidi RRL 16(4), 2100599 (2022). https://doi.org/10.1002/pssr.202100599, Google ScholarCrossref5. H. S. Lee, Y. Zhang, Z. Chen, M. W. Rahman, H. Zhao, and S. Rajan, “ Design and fabrication of vertical GaN p-n diode with step-etched triple-zone junction termination extension,” IEEE Trans. Electron Devices 67(9), 3553–
留言 (0)