N₂O grown high Al composition nitrogen doped β-(AlGa)₂O₃/β-Ga₂O₃ using MOCVD

β-Ga₂O₃ has been extensively studied as a promising ultrawide bandgap semiconductor for power electronics applications owing to its large bandgap of ∼4.9 eV (Ref. 1) and predicted high critical breakdown field of ∼6–8 MV/cm.² Over the last decade, significant progress has been made in the epitaxial growth of high-purity Ga₂O₃ films^3–5 and the demonstration of power electronic devices with encouraging results.^6–11 The progress is intensified by the availability of high-quality melt grown Ga₂O₃ substrates,¹² donor dopants with wide doping ranges (∼10¹⁴ to >10²⁰ cm⁻³),^13–18 deep acceptor dopants for efficient compensation,^19–22 and the development of epitaxial growth techniques used to grow high-purity materials.^4,5,20 One of the key achievements in the development of β-Ga₂O₃ is the formation of β-(Al_xGa_1–x)₂O₃ by alloying β-Ga₂O₃ with corundum phase Al₂O₃ to tune device’s electrical performance through the bandgap engineering of the surface¹⁴ or backside (AlGa)₂O₃ layers.²³ 

β-(Al_xGa_1–x)₂O₃/β-Ga₂O₃ heterostructure interface enables carrier confinement, forming two-dimensional electron gas (2DEG) with high electron mobility.^24,25 High Al composition β-(Al_xGa_1–x)₂O₃ is needed to increase the conduction band offsets and achieve higher 2DEG charges in transistor structures, such as modulation-doped field effect transistors (MODFETs). According to the equilibrium Al₂O₃–Ga₂O₃ phase diagram prediction by Hill et al., >60% Al incorporation can be achieved (at temperatures >800 °C)²⁶ in the monoclinic β phase. However, despite substantial efforts to grow high Al composition β-(Al_xGa_1–x)₂O₃ on various orientations of β-Ga₂O₃ substrates using MBE (Refs. 27–30) and MOCVD,^14,31–35 there are still challenges to achieving phase pure high Al content β-(AlGa)₂O₃ regardless of the growth techniques.

Most of the β-(AlGa)₂O₃ films have been grown on the (010) oriented β-Ga₂O₃ substrates.^{14,27,31–33,36} This orientation is the preferred growth orientation, and β-Ga₂O₃ films with superior materials quality and device performances have been realized on it.^4,5,10 As a result, several groups have reported the successful demonstration of (010) β-(Al_xGa_1–x)₂O₃/β-Ga₂O₃ MODFETs (x ≈ 8%–26%) with 2DEG charge density varying from ∼0.2 to 1.1 × 10¹³ cm⁻².^24,37–40 However, the consistent realization of high-quality phase pure β-(Al_xGa_1–x)₂O₃ films with Al content >20% in this orientation remains challenging due to the second phase and extended defect formation in high Al content films.³¹ Other β-Ga₂O₃ substrate orientations, such as (100), $( 2 ¯ 01)$⁠, and (001), have also been explored for the growth of β-(Al_xGa_1–x)₂O₃ by both MBE and MOCVD methods.^28,30,34,35 The (100) and $( 2 ¯ 01)$ substrate orientations have proven suitable for the growth of β-(Al_xGa_1–x)₂O₃ with higher Al incorporation. For example, Oshima et al. demonstrated (100) β-(AlGa)₂O₃ films with 61% Al content by growing the layer using MBE.²⁸ Likewise, using MOCVD, Bhuiyan et al. reported β-(Al_xGa_1–x)₂O₃ films with Al compositions of up to 48% and 52%, respectively, on $( 2 ¯ 01)$ oriented³⁴ and (100) oriented³⁵ substrates.

MOCVD of β-(Al_xGa_1–x)₂O₃ films is typically conducted using aluminum and gallium metalorganic precursors as sources for Al and Ga and pure oxygen for oxidation. However, pure oxygen is very reactive with aluminum metalorganic sources, increasing the gas phase reaction that depletes the Al source and reduces its incorporation into β-(AlGa)₂O₃. Lower reactor pressure (<60 Torr) is used to minimize the gas phase reaction of the aluminum metalorganic sources, but achieving higher Al incorporation remains challenging. In this study, we report on the MOCVD growth of high-quality β β-(Al_xGa_1–x)₂O₃ films grown on (010) and (100) oriented β-Ga₂O₃ substrates using nitrous oxide (N₂O) as an oxygen source. The N₂O reacts less aggressively with Al precursors than pure oxygen, resulting in higher incorporation of Al into the β-(AlGa)₂O₃films. (010) β-(Al_xGa_1–x)₂O₃ and (100) β-(Al_yGa_1–y)₂O₃ layers with a record high Al composition of x = 30% on the (010) substrate and up to y = 45% on the (100) oriented substrate were demonstrated.

Ga₂O₃ and other oxides’ (e.g., ZnO) growth with N₂O requires an entirely different set of growth conditions than those typically used for growth with pure oxygen.^20,41 The N₂O based growth for Ga₂O₃ generally requires a higher growth pressure (>75 Torr) and temperature (>700 °C) to ensure the efficient decomposition of N₂O. In addition to its use to grow Ga₂O₃ thin films, N₂O can also dope the layers with nitrogen, and the concentration of N incorporation into the films strongly depends on process conditions, including N₂O flow rate, growth pressure, and temperature.^20,42 Considering this, we also studied the doping of β-(Al_xGa_1–x)₂O₃layers with nitrogen (N) and the effects of process conditions on its incorporation efficiency. β-(AlGa)₂O₃ films with [N] ranging from ∼6 × 10¹⁷ to ∼2 × 10¹⁹ cm⁻³ are reported in this work.

β-(AlGa)₂O₃ layers were grown using an Agnitron Technology’s Agilis^TM 100 MOCVD reactor with trimethylaluminum (TMAl), triethylgallium (TEGa), and nitrous oxide (5 N) as precursors, and nitrogen as the carrier gas. Fe doped or unintentionally doped (UID) β-Ga₂O₃ substrates with (010) and (100) orientations (Novel Crystal Technology) were coloaded into the reactor during the growth of the films. Between the β-(AlGa)₂O₃ layer and the substrates, a pure oxygen-grown unintentionally doped Ga₂O₃ (∼30 nm thick) layer was introduced regardless of the substrate’s orientation. The N₂, O₂, and N₂O gasses were purified using point-of-use purifiers to reduce the impurity to below ppm level. The flow of nitrous oxide (N₂O) was set to 1000 SCCM, while the growth pressure and substrate temperature were varied from 150 to 200 Torr and 700–900 °C, respectively. The gas phase [TMAl]/([TMAl]+[TEGa]) molar flow rate ratio (x_g) introduced into the reactor was varied as 9.86%, 18.20%, and 32.2%, which were obtained by introducing a constant TEGa molar flow rate of 9.69 μmol/min and varying the TMAl molar flow rate as 1.06, 2.16, and 4.60 μmol/min, respectively. The aluminum composition and surface roughness for β-(AlGa)₂O₃ layers were analyzed using high resolution XRD (HRXRD) and atomic force microscopy (AFM). To identify the fraction of Al estimated for the β-(AlGa)₂O₃ layers grown on (010) and (100) substrates by coloading them into the reactor, we used x and y notations. As such, the fraction of Al in (AlGa)₂O₃ films grown on (010) is identified as (010) β-(Al_xGa_1–x) O₃, while the film grown on (100) is denoted as (100) β-(Al_yGa_1–y)O₃ substrates. HRXRD reciprocal space mapping (RSM) was also conducted on selected samples to evaluate the strain characteristics in the β-(AlGa)₂O₃ films with respect to the underlying (010) or (100) substrates. Nitrogen doping and its incorporation dependence on process conditions were studied using secondary ion mass spectroscopy (SIMS, EAG Eurofins) measurements.

The effects of substrate orientation and reactor pressure on the Al incorporation and surface roughness were studied by growing β-(AlGa)₂O₃ films on (010) and (100) oriented β-Ga₂O₃ substrates using a constant substrate temperature of 850 °C, an N₂O flow rate of 1000 SCCM, and a gas phase [TMAl]/[TMAl+TEGa] molar flow rate ratio (x_g) of ∼9.86%. Figures 1(a) and 1(b) show the HRXRD ω–2θ scans for β-(AlGa)₂O₃ films grown, respectively, on (010) and (100) substrates by coloading them into the reactor. The films were grown at reactor pressure values of 150 and 200 Torr. The HRXRD ω–2θ scans shown in Fig. 1 are based on signals originating from (020) and (400) reflection planes, respectively, for (010) β-(Al_xGa_1–x) O₃ [Fig. 1(a)] and (100) β-(Al_yGa_1–y)O₃ [Fig. 1(b)] grown on the corresponding β-Ga₂O₃ substrates. The Al content (x) in the β-(AlGa)₂O₃ films grown on (010) β-Ga₂O₃ substrates was estimated by comparing the XRD peak separation between (020) β-Ga₂O₃ and (020) β-(Al_xGa_1–x)₂O₃.⁴³ Similarly, for the β-(AlGa)₂O₃ films grown on (100) β-Ga₂O₃ substrates, the Al content in the film (y) was estimated by comparing the peak separation between (400) β-Ga₂O₃ and (400) β-(Al_yGa_1–y)₂O₃.³⁰ Accordingly, x/y values of 11.70%/12.70% and 4.82%/5.23% were estimated for the β-(AlGa)₂O₃ films grown on (010)/(100) oriented β-Ga₂O₃ substrates at growth pressures of 150 and 200 Torr (see Table I).

Higher Al incorporation was obtained for the β-(AlGa)₂O₃ films grown at lower reactor pressure (150 Torr) for both substrate orientations. This is due to the oxidizing gas (N₂O), which decomposes efficiently at high pressure and temperature.^20,41 At a reactor pressure of 200 Torr, the Al composition in the β-(AlGa)₂O₃ layers grown on each substrate orientation is ∼46%–51% lower than the gas phase Al fraction (x_g= 9.86%) introduced to the reactor to grow the layers. This suggests the presence of a strong gas phase reaction that depletes the Al precursor before reaching the surface of the growing film due to the efficient decomposition of N₂O at this pressure. When the layer is grown at lower pressure (150 Torr), the Al composition measured in the grown films for both substrate orientations (x/y = 11.70%/12.70%) is more than the gas phase Al fraction (x_g= 9.86%) introduced into the reactor, suggesting a lower gas phase reaction. Low reactor pressure generally reduces the gas phase reaction even when pure oxygen is employed to grow the (AlGa)₂O₃ layers. However, here, the effect of lower pressure is substantial due to the lower decomposition efficiency of N₂O at this pressure. The same phenomenon is observed at low substrate temperatures (vide infra). On the other hand, regardless of the growth conditions, the Al composition of the (100) β-(Al_yGa_1–y)₂O₃ films grown on the (100) β-Ga₂O₃ substrate is slightly higher than that of the (010) β-(Al_xGa_1–x)₂O₃ films grown on the (010) β-Ga₂O₃ substrate (i.e., y > x), even though the substrates were coloaded into the reactor. Such a high Al composition in MOCVD grown (100) β-(AlGa)₂O₃ layers had been reported by Bhuiyan et al.³⁵ previously by growing the layers using pure oxygen. The incorporation of higher Al in the layer grown on the (100) β-Ga₂O₃ substrates could be attributed to the greater stability of the (AlGa)₂O₃ layers owing to the lower surface energies of the (100) β-Ga₂O₃.⁴⁴ 

The thickness of the β-(AlGa)₂O₃ films was estimated from the XRD fringes for films grown on both substrate orientations (see Table I). The estimated thickness values for the films grown on both orientations under the same growth conditions are comparable, suggesting a comparable film growth rate. However, comparing the growth rate of the films at the two growth pressure values, the growth rate at 150 Torr is ∼13% higher than that at 200 Torr, further confirming the presence of the gas phase reaction at higher growth pressure. The surface morphologies of the films were analyzed using AFM measurements. Figures 2(a) and 2(b) compare the 2D AFM images taken from a scan area of 5 × 5 μm² for β-(AlGa)₂O₃ films grown on (010) and (100) at 200 Torr (sample A). The films have shown noticeably different surface morphologies, but both are smooth layers with the root mean square surface roughness of ∼0.65 nm.

The growth of β-(AlGa)₂O₃/β-Ga₂O₃ superlattices (SLs) on both (010) and (100) oriented substrates was also demonstrated using N₂O as an oxygen source. The SL structures were grown at 200 Torr by coloading the two substrates into the reactor and using the gas phase [TMAl]/[TMAl+TEGa] molar flow rate ratio (x_g) of ∼9.86%, substrate temperature of 850 °C, and N₂O flow rate of 1000 SCCM. Both SL structures were grown on a ∼30 nm thick UID β-Ga₂O₃ layer, and the β-(AlGa)₂O₃ barrier and β-Ga₂O₃ well thicknesses were 5 and 10 nm, respectively, and the SL period was 8. Figures 3(a) and 3(b) show the HRXRD ω–2θ scans of SL structures around the β-Ga₂O₃ (020) and (400) planes for the layers grown, respectively, on (010) and (100) oriented substrates. The XRD pattern for both samples shows superlattices with satellite peaks of −1 and +1 orders, and the Al content in the β-(AlGa)₂O₃ barrier is (x ≈ y∼5%) as estimated from a single layer (Table I, sample A).

The gas phase [TMAl]/[TMAl+TEGa] molar flow rate ratio (x_g) and growth temperature effects on the incorporation efficiency of Al were studied by growing β-(AlGa)₂O₃ films on (010) and (100) oriented β-Ga₂O₃ substrates at substrate temperature ranging 700850 °C (Table I, samples B–G). The reactor pressure and N₂O flow rate were set at 150 Torr and 1000 SCCM, but x_g was varied as 9.86% (sample B), 18.20% (samples C, D, and E), and 32.2% (samples F, G, and H). Figures 4(a) and 4(b) show the HRXRD ω–2θ scans for (020) and (400) reflection planes, respectively, for (010) β-(Al_xGa_1–x)O₃ and (100) β-(Al_yGa_1–y)O₃ grown at variable substrate temperature and x_g. For the β-(AlGa)₂O₃ films grown at 850 °C, increasing the x_g from 9.86% (black trace, sample B) to 18.20% (red trace, sample C) led to an increase in the Al incorporation into the films for each substrate orientation. This is expected since introducing more Al source into the MOCVD reactor leads to higher Al incorporation into the film. On the other hand, reducing the substrate temperature from 850 to 700 °C for a constant x_g of 18.20% led to the increase in the Al incorporation into the β-(AlGa)₂O₃ films as seen from the shift in the XRD peak position of the (020) or (400) reflection planes for β-(AlGa)₂O₃ films grown on each substrate orientation. Figure 4(c) shows the Al composition in the β-(AlGa)₂O₃ films estimated from the HRXRD patterns shown in Figs. 4(a) and 4(b) versus substrate temperature. The Al content in the films increases when the substrate temperature decreases for both substrate orientations. For the films grown on the (010) substrate orientation, the highest Al composition obtained was 30% at a substrate temperature of 800 °C. This is the record Al composition in a phase pure β-(AlGa)₂O₃ film grown on (010) oriented substrates. In an oxygen based MOCVD epitaxy, the highest Al composition in phase pure (010) β-(AlGa)₂O₃ is <27%, and attempts to increase the Al composition beyond this value have led only to the second phase or extended defect formation.³¹ The increase in Al incorporation at lower substrate temperature is due to the decrease in the gas phase reaction because of the lowering of the decomposition efficiency of N₂O at low substrate temperature and pressure, as discussed above.

The β-(AlGa)₂O₃ films grown on (100) oriented substrates consistently showed a higher Al composition compared to the film grown on the (010) substrate despite being coloaded into the reactor during growth. For example, the Al composition in the (100) β-(Al_yGa_1–y)₂O₃ film grown on the (100) substrate by coloading it with the (010) substrate that showed a record high Al composition of x = 30% was y ∼ 39% (Table I). The highest Al composition of ∼45% was realized in the (100) oriented β-(AlGa)₂O₃ films grown at a substrate temperature of 700 °C. However, at the same temperature, the film grown on the (010) oriented substrate showed no signature of beta or any other (AlGa)₂O₃ phases, such as the gamma phase that typically grows when Al incorporated into the film is higher.⁴⁵ Suspecting the presence of gamma phase (AlGa)₂O₃ in the film grown at 700 °C on the (010) oriented substrate [Fig. 4(a), sample D], we performed a wide angle XRD scan (full scale data not shown), but there is no peak observed.

The strain characteristics of the selected (010) β-(Al_xGa_1–x)₂O₃ and (100) β-(Al_yGa_1–y)₂O₃ films grown on both (010) and (100) substrates (samples C and D) were investigated by measuring the RSMs in HRXRD (Fig. 5). The measurements were performed for the off-axis (420) and $( 80 1 ¯)$ reflections in the asymmetrical geometry of (010) β-(Al_xGa_1–x)₂O₃ and (100) β-(Al_yGa_1–y)₂O₃, respectively. Figures 5(a) and 5(b) show the asymmetrical RSMs for (420) reflections of (010) β-(Al_xGa_1–x)₂O₃/β-Ga₂O₃ with Al compositions, x, of 26.7% and 30.0%. For the layer with an Al composition of 26.7%, the β-(Al_xGa_1–x)₂O₃ and β-Ga₂O₃ peaks appear to have the same in-plane lattice constant, indicating that (010) β-(Al_xGa_1–x)₂O₃ is coherently strained to the (010) β-Ga₂O₃ substrate. However, for the layer with an Al composition of 30% [Fig. 5(b)], the (420) peak position for (010) β-(Al_xGa_1–x)₂O₃ is not aligned with the peak for (010) β-Ga₂O₃, indicating the growth of a relaxed film. Similarly, Figs. 5(c) and 5(d) show the asymmetrical RSMs for $( 80 1 ¯)$ reflections of (100) β-(Al_yGa_1–y)₂O₃/β-Ga₂O₃ with Al compositions, y, of 34.4% and 39.1%. Both layers show coherently strained (100) β-(Al_yGa_1–y)₂O₃ layers grown on (100) oriented β-Ga₂O₃ substrates as seen from the well aligned position of the epitaxial layers and the substrate peaks.

With the aim to further increase the incorporation of Al, additional (AlGa)₂O₃ films were grown on both (010) and (100) oriented substrates by increasing only the gas phase [TMAl]/[TMAl+TEGa] molar flow rate ratio (x_g) to 32.20% (Table I, samples F, G, and H) but keeping the other growth conditions the same with that used in Fig. 4. Figures 6(a) and 6(b) show the HRXRD ω–2θ scans for (020) and (400) reflection planes, respectively, for (010) β-(Al_xGa_1–x)O₃ and (100) β-(Al_yGa_1–y)O₃ films grown at variable substrate temperature from 700 to 850 °C. The XRD patterns for the films grown on the (010) substrate show no peaks showing β phase (010) (Al_xGa_1–x)O₃ at each substrate temperature used to grow the layers. However, for the layers grown at 800 and 850 °C substrate temperatures, the XRD peak was observed at 2θ ∼65.5°, corresponding to the γ phase (010) (Al_xGa_1–x)O₃, consistent with reports in the literature.⁴⁵ For the corresponding films grown on (100) substrates at substrate temperatures of 850 (sample F) and 800 °C (sample G), the XRD patterns showed two peaks (doublets) in the scan range that correspond to the β phase (100) (AlGa)₂O₃. The splitting of the XRD peaks may have resulted from the presence of local Al segregation in the film leading to the β phase (100) (AlGa)₂O₃ with two separate Al compositions rather than the growth of other (AlGa)₂O₃ phases. The tendency of the second phase formation in the (100) (AlGa)₂O₃ with high Al composition has also been observed.³⁵ On the other hand, like the (AlGa)₂O₃ layer grown on (010) substrates [Fig. 4(a), sample D], the films grown on both (010) and (100) substrates at 700 °C (sample H) did not show peaks that correspond to any (AlGa)₂O₃ phase.

In addition to its use to grow β-(AlGa)₂O₃, N₂O can also be used to dope the layers with nitrogen deep acceptors. We have extensively studied the use of N₂O to dope Ga₂O₃ with nitrogen to achieve semi-insulating films due to the deep acceptor nature of N dopants. The concentration of N incorporated into the Ga₂O₃ film was generally MOCVD process condition dependent, but particularly the effect of the substrate temperature was the strongest.²⁰ Similarly, in this work, we studied the nitrogen doping of β-(AlGa)₂O₃ using the N₂O source. The influence of the substrate temperature and Al composition on the incorporation efficiency of N into the films was studied.

The effect of substrate temperature on the incorporation efficiency of N atoms into the β-(AlGa)₂O₃ layers was studied by growing the SIMS stack of N doped β-(AlGa)₂O₃ layers on the (010) β-Ga₂O₃ substrate by varying the substrate temperature while keeping the other growth conditions the same. The N₂O flow rate, growth pressure, and gas phase [TMAl]/[TMAl+TEGa] molar flow rate ratio (x_g) were kept at 1000 SCCM, 150 Torr, and 9.86%, respectively, whereas substrate temperatures of 700, 800, 850, and 900 °C were used to grow four different N doped β-(AlGa)₂O₃ layers. Figure 7(a) shows the schematics of the SIMS stack layers grown using the indicated substrate temperatures, where each of N doped β-(AlGa)₂O₃ layers (a, b, c, and d) is separated by UID Ga₂O₃ spacers that are grown using pure oxygen at a substrate temperature of 850 °C and a reactor pressure of 60 Torr. The N₂O grown β-(AlGa)₂O₃ layers and the UID spacers are 150 nm thick. Figure 7(b) shows the SIMS depth profiles for N and secondary ion intensity for Al in the SIMS stack layer shown in Fig. 7(a). The SIMS data confirm the doping of the (010) β-(AlGa)₂O₃ layers with nitrogen, and the N incorporation into the layers increases with the decrease in the substrate temperature, which is consistent with the behavior of nitrogen doping of Ga₂O₃ using N₂O.²⁰ The concentration of N impurities in the (010) β-(AlGa)₂O₃ increased from ∼7 × 10¹⁷ cm⁻³ at 900 °C (layer a) to ∼2 × 10¹⁹ cm⁻³ at 700 °C (layer d). Despite the constant gas phase [TMAl]/[TMAl+TEGa] molar flow rate ratio (x_g) of 9.86% introduced into the reactor during the growth of the SIMS stack, the secondary ion intensity for Al shows an increasing trend of the Al composition in the β-(AlGa)₂O₃ with the decrease in the substrate temperature. Figure 7(c) shows the dependence of peak N concentration and secondary ion intensity for Al on substrate temperature in which both Al and N incorporation increased with the decreasing substrate temperature.

The increase in the Al composition in the layers grown at lower substrate temperature is observed above (in Fig. 4) for the β-(AlGa)₂O₃ films grown on both (010) and (100) oriented substrates. However, the effect of Al concentration on the incorporation efficiency of N is unclear. To study this, we grew the SIMS stack layers of β-(Al_xGa_1–x)₂O₃ on the (010) β-Ga₂O₃ substrate by introducing a variable gas phase [TMAl]/[TMAl+TEGa] molar flow rate ratio (x_g) but keeping other process conditions the same. The N₂O flow rate, growth pressure, and substrate temperature were kept at 1000 SCCM, 150 Torr, and 850 °C, respectively. The x_g values, however, were varied to obtain N doped β-(Al_xGa_1–x)₂O₃ layers with estimated Al compositions (x) of 0% (i.e., N₂O grown Ga₂O₃), ∼6%, ∼12%, and ∼22%. Figure 8(a) shows the schematics of the SIMS stack layers grown using variable Al composition, where each of N doped β-(AlGa)₂O₃ layers (e, f, g, and h) are separated by UID Ga₂O₃ spacers that are grown using pure oxygen at a substrate temperature of 850 °C and a reactor pressure of 60 Torr. Figure 8(b) shows the SIMS depth profiles for N in the SIMS stack layer grown by varying the Al composition in β-(AlGa)₂O₃ [Fig. 8(a)]. No obvious effect of the Al composition on the amount of N atoms incorporated into the layers is observed for the β-(AlGa)₂O₃ layers with an Al composition up to 6%, as the N concentration for the layers with 0% (i.e., N₂O grown Ga₂O₃, layer e) and 6% (layer f) Al content remains to be constant at ∼2.5 × 10¹⁸ cm⁻³. However, when the Al composition increases from ∼6% (layer f) to 22% (layer h), N incorporated into (010) β-(AlGa)₂O₃ increased linearly [see the inset in Fig. 8(b)] by more than three times. Since the Al incorporation efficiency into the β-(AlGa)₂O₃ layer grown on (100) oriented substrates is higher than the one grown on (010) oriented substrates (Fig. 4), the N incorporation is expected to be higher in the β-(AlGa)₂O₃ layer grown on the (100) orientation than the layer grown on (010) orientation.

In summary, nitrogen-doped (AlGa)₂O₃ epitaxial films with high Al compositions were grown on (010) and (100) oriented β-Ga₂O₃ substrates by MOCVD using N₂O as an oxygen source. The Al composition, strain states, and phases of the (AlGa)₂O₃ layers grown on β-Ga₂O₃ substrates were evaluated using HRXRD. The incorporation efficiency of Al into the (AlGa)₂O₃ films depends on process conditions, including chamber pressure, growth temperature, and gas phase TMAl/(TMAl+TEGa) molar flow rate ratio (x_g). Layers grown at lower reactor pressure and substrate temperature, and higher x_g showed higher Al incorporation. Phase pure β-(AlGa)₂O₃ films with an Al composition of up to 45% for layers grown on the (100) β-Ga₂O₃ substrate and a record high Al composition of 30% for a film grown on a (010) β-Ga₂O₃ substrate were demonstrated. N₂O grown β-(AlGa)₂O₃/β-Ga₂O₃ superlattice structures with an Al composition of 5% were also demonstrated on both substrate orientations. With the increase in the gas phase molar flow rate ratio of Al (x_g), gamma phase (AlGa)₂O₃ was grown on (010) substrates, but the layers grown on (100) showed a β phase layer with two different compositions of Al. Nitrogen doping of β-(AlGa)₂O₃ was also achieved using N₂O, but the incorporation of N depends on the substrate temperature and Al compositions in the film. N concentration ranging between 6 × 10¹⁷ and 2 × 10¹⁹ cm⁻³ was achieved for (010) β-(AlGa)₂O₃ layers grown at a substrate temperature of 900–700 °C. A higher Al composition and lower substrate temperature lead to higher N incorporation.

The work at Agnitron was supported by the ONR STTR Phase II program N6833518C0192 (Program Manager: Mr. Lynn Petersen). The support for the work at UCSB was provided by the AFOSR Program FA9550-18-1-0479 (AFOSR GAME MURI, Dr. Ali Sayir Program Manager). Research at the Naval Research Laboratory was supported by the Office of Naval Research.

The authors have no conflicts to disclose.

Fikadu Alema: Conceptualization (lead); Data curation (equal); Formal analysis (equal); Funding acquisition (equal); Investigation (equal); Writing – original draft (lead); Writing – review & editing (equal). Takeki Itoh: Conceptualization (equal); Data curation (equal); Formal analysis (equal); Investigation (equal); Writing – review & editing (supporting). William Brand: Conceptualization (equal); Data curation (equal); Formal analysis (supporting). Marko Tadjer: Conceptualization (equal); Resources (equal); Writing – review & editing (equal). Andrei Osinsky: Conceptualization (equal); Formal analysis (equal); Funding acquisition (lead); Methodology (equal); Supervision (equal); Writing – review & editing (supporting). James S. Speck: Conceptualization (equal); Formal analysis (equal); Funding acquisition (lead); Methodology (supporting); Supervision (equal); Writing – review & editing (equal).

The data that support the findings of this study are available from the corresponding author upon request.