Comparison of Ga₂O₃ Films Grown on m- and r-plane Sapphire Substrates by MOCVD

Ga₂O₃ films were grown on 1 cm × 1 cm single-sided polished m-plane and r-plane sapphire substrates by the MOCVD system independently developed by Xidian University. Triethylgallium (TEGa) and high-purity oxygen were respectively used as the precursors of Ga and O, and high-purity nitrogen was used as carrier gas to transport TEGa and oxygen into the reaction chamber. TEGa was always placed in a constant temperature water at 24 °C to maintain a constant vapor pressure. TEGa and oxygen were respectively set to 70 and 3000 sccm, and the growth temperature was maintained at 800 °C.

Figures 1a–1c show the AFM images of Ga₂O₃ films grown on m-plane sapphire substances at 30, 40 and 50 Torr, respectively, and the grain size progressively decreases and grain density increases with the increasing growth pressure. It can be attributed that the higher growth pressure increases the concentration of molecules in the reaction chamber, leading to intensified collisions between molecules, which in turn increases the nucleation and the density of crystal grains. The fierce collision reduces the kinetic energy of atom migration on the surface, and the crystal grains cannot be fully coalesced. The shape of grain trasforms from chunky rock-like structure to small granular structure, and a number of granular grains tightly gather together to form some large and relatively flat regions at 50 Torr. The RMS roughness of Ga₂O₃ film decreases from 43 nm at 30 Torr to 29.7 nm at 40 Torr and then further slightly decreases to 29 nm at 50 Torr (as seen in Fig. 1g) because of the degradation of surface undulation. Figures 1d–1f present the AFM images of those grown on r-plane sapphire substances, and similar phenomena to the Ga₂O₃ films grown on m-plane appare. It's just that at 50 Torr, the grains are not as they do on the m-plane, but are evenly distributed on the surface. Interestingly, the RMS roughness of β-Ga₂O₃ film increases from 27.7 nm to 31.1 nm when the growth pressure increases from 30 Torr to 40 Torr and drastically reduce to 16.2 nm at 50 Torr. This can be explained as the increase in pressure leads to uneven reduction in grain size, which increases the degree of surface undulation. When the growth pressure is further increased to 50 Torr, all the grains on the surface have a granular structure and are uniformly distributed, so the degree of surface fluctuation and the RMS roughness decreases.

Zoom In Zoom Out Reset image size

Figure 2a exhibits the height distributions of Ga₂O₃ films deposited on m-plane sapphire substances. It can be noticed that the height distribution becomes more and more concentrated with the increasing growth pressure, which reduces the surface fluctuations and RMS roughness. Figures 2c, 2e and 2g show the Gaussian fitting curve of Fig. 2a. The height distribution at 30 Torr is basically symmetrical but not a Gaussian distribution, and it can be decomposed into 3 Gaussian peaks (as seen in Fig. 2c). The height distribution curve transforms into a strictly symmetric Gaussian curve at 40 Torr, however, the height distribution curve continues to become asymmetric at 50 Torr, indicating that the film grown at 40 Torr has a randomly undulated surface without a special preferred height, and the larger or smaller growth pressure will cause the crystal grains to transform from random orientation growth to preferential orientation growth. ^36,37 Figure 2b shows the height distributions of those prepared on r-plane sapphire substances. When the growth pressure is increased from 30 Torr to 40 Torr, the height distribution curve is slightly broadened, so the degree of surface undulation and the RMS roughness increase. Continue to increase the growth pressure to 50 Torr, and the height distribution curve becomes more concentrated and changes from an asymmetric curve to a Gaussian curve, so the RMS roughness drops sharply and the grains become randomly oriented growth.

Zoom In Zoom Out Reset image size

Figure 3a presents the XRD pattern of Ga₂O₃ films grown on m-plane sapphire substrates. It can be observed that a number of diffraction peaks related to Ga₂O₃ appear, which indicates that polycrystalline Ga₂O₃ films were obtained on the m-plane sapphire substrates. It is worth mentioning that the diffraction peak at 64.4° comes from α-Ga₂O₃ (30–30) plane, and the remaining diffraction peaks are related to β-Ga₂O₃, consistent with our previous research. ³⁸ Interestingly, the diffraction peak intensities related to β-Ga₂O₃ and α-Ga₂O₃ (30–30) plane decreases, on the contrary, the intensity of β-Ga₂O₃ (−402) diffraction peak at 38.1° increases as the growth pressure increases (as can be seen in Fig. 3b). These two different phenomenons exclude the influence of film thickness, indicating that the higher growth pressure suppresses the polycrystallization of the films but is not beneficial to the growth of α-Ga₂O₃. However, the diffraction peaks related to Ga₂O₃ films grown on the r-plane sapphire substrates are not easily detected. Liu et al. also obtained the same result using the PE-MBE epitaxy method, however, they confirmed that β-Ga₂O₃ containing (100) and (001) oriented grains appeared on the r-plane sapphire substrate. ²⁶

Zoom In Zoom Out Reset image size

The thickness of Ga₂O₃ films measured by the ellipsometer as a function of growth pressure is presented in Fig. 4. The thickness of the film deposited on the m-plane sapphire substrate decreases approximately linearly with the increasing growth pressure, and the film grown at 30, 40 and 50 Torr is respectively 419.4, 354.6 and 300.8 nm. The thickness of the Ga₂O₃ film epitaxially on the r-plane sapphire substrate slightly increases from 496.2 nm to 497.4 nm, and then drops sharply to 304.8 nm at 50 Torr, indicating that the lower growth pressure has little effect on the deposition rate of Ga₂O₃ film on the r-plane sapphire substrate. With reference to each set of growth pressures, it can be observed that the deposition rate on the r-plane sapphire substrate is significantly higher than the m-plane under the lower growth pressure, and the increase in pressure forces the deposition rate on the two substrates to become consistent. Interestingly, the trend of growth thickness and surface roughness is consistent (as can be seen in Fig. 1g), indicating that there is a positive correlation between the two parameters.

Zoom In Zoom Out Reset image size

The absorption spectra of Ga₂O₃ films grown on m-plane and r-plane sapphire substrates are shown in Fig. 5. Here only the absorption spectra of some samples are presented, while other samples can not be measured due to some surface factors. The Ga₂O₃ film of Figs. 5a and 5b have obvious absorption edge near 270 and 280 nm, respectively. For Ga₂O₃ material, the relationship between its optical bandgap and absorption coefficient is ^28,31–33 :

$$\begin{eqnarray*}&&{\left(\alpha h\nu \right)}^{2}={\rm{C}}\left({\rm{h}}\nu -{\rm{Eg}}\right)\end{eqnarray*}$$

where − α is the absorption coefficient, h is the Planck's constant, ν is the frequency of the incident light, and C is a constant. Then, the optical bandgap (E_g) of the samples can be obtained by fitting curves of (αhν)² vs hν and extending the straight part of the curve to the x-axis. As shown in the inset of Fig. 5, the estimated bandgap values are respectively 4.61 and 4.78 eV, consistent with the previous reports. ^20,32,34

Zoom In Zoom Out Reset image size

We obtained Ga₂O₃ films respecyively epitaxially on the m- and r-plane sapphire substrates. The higher growth pressure slows down the deposition rate and reduces the surface grain size. XRD results indicated that the higher growth pressure is beneficial to the preferential growth of β-Ga₂O₃ deposited on the m-plane sapphire substrate, while the lower growth pressure is beneficial to α-Ga₂O₃.

We gratefully acknowledge the financial support from the National key Research and Development Program of China (grant no. 2018YFB040650), the Fundamental Research Funds for the Central Universities (grant no. JB181108), the National Natural Science Foundation of China (grant no. 61774116, 61904139, 61974112 and 61974115), the Wuhu and Xidian University special fund for industry-university-research cooperation (grant no. XWYCXY-012019002), the Natural Science Basic Research Program of Shaanxi (Program No. 2020JQ-315), the Key Research and Development program in Shaanxi Province (grant no. 2018ZDCXL-GY-01-02-02), and this work was also supported by the 111 Project (B12026).