A material structure for improving electron mobility of GaN heterojunction and an epitaxial growth method thereof

Abstract

The application discloses a material structure for improving GaN heterojunction electron mobility and an epitaxial growth method thereof. The structure comprises, from bottom to top, a single crystal substrate, an AlN nucleation layer, a GaN buffer layer, an AlN insertion layer, a GaN isolation layer, a lower AlGaN barrier layer, an upper AlGaN barrier layer and an InGaN cap layer. The epitaxial growth method is characterized in that: the GaN isolation layer is introduced to suppress alloy disorder scattering; the lower barrier layer is grown by using a high V / III ratio low-speed process to reduce defects, and the upper barrier layer is grown by using a low V / III ratio high-speed process to improve the interface morphology; and finally, the InGaN cap layer is grown to relieve stress and reduce surface states. The application can synergistically suppress multiple scattering mechanisms, and the room temperature electron mobility is improved to 2270 cm 2 / V·s or above, and the surface roughness is significantly reduced, thereby providing a key material basis for high-performance GaN microwave devices.

Description

Technical Field

[0001] This invention belongs to the field of semiconductor epitaxial materials technology, specifically relating to a material structure for improving the electron mobility of GaN heterojunctions and its epitaxial growth method. Background Technology

[0002] Gallium nitride high electron mobility transistors (GaN HEMTs) exhibit significant advantages in the microwave power field due to their superior characteristics, such as wide bandgap, high critical breakdown electric field, and inherently high electron saturation velocity at high frequencies. This is driving the rapid development of micro base stations, wireless communication systems, and other applications towards higher power densities and higher operating frequencies. Further improvements in device performance fundamentally depend on optimizing the transport properties of the two-dimensional electron gas (2DEG) in the channel of its core material—AlGaN / GaN heterojunction. Improving the room-temperature electron mobility of the 2DEG is one of the most direct and effective techniques for enhancing the transconductance, cutoff frequency, and output power of the device.

[0003] Theoretically, due to the modulation doping effect, the two-dimensional electron gas in AlGaN / GaN heterojunctions is spatially separated from its parent ionized impurities, which makes it possible to obtain extremely high electron mobilities. However, in practical applications, especially for materials epitaxially grown on heterostructures (such as Si and SiC), room-temperature electron mobility has long stagnated at approximately 2100 cm⁻¹. 2 The performance of GaN microwave devices has been limited to the level of / V·s, making it difficult to achieve a breakthrough. This has become a key bottleneck restricting the performance of GaN microwave devices from reaching their theoretical limits.

[0004] This bottleneck primarily stems from the combined constraints imposed by multiple scattering mechanisms on channel electron transport. Despite the extremely high surface density of the two-dimensional electron gas (approximately 10⁻⁶), 13 cm -2The heterojunction has a certain shielding effect on some scattering, but the special structure of the heterojunction and the defects introduced by epitaxy make the following scattering mechanisms particularly prominent: (1) Alloy disorder scattering: The random distribution of aluminum and gallium atoms in the AlGaN barrier layer will form irregular potential energy fluctuations in the barrier region. Although ultrathin AlN insertion layers are commonly used to suppress tunneling of channel electrons to the barrier layer, the thickness of the insertion layer is usually less than 3 nm, and the shielding effect is limited. Alloy disorder scattering is still one of the main mechanisms affecting mobility. (2) Interface roughness scattering: The atomic-level fluctuations of the heterojunction interface will destroy the periodic potential field of the electron transport path. Its scattering intensity is directly related to the interface quality, and the interface quality is precisely controlled by the surface migration and nucleation dynamics during epitaxial growth. (3) Coulomb scattering: It mainly comes from two types of charged centers: one is the ionized surface state located on the surface (or near the surface) of the barrier layer; the other is the deep level acceptor formed inside the barrier layer due to growth defects (such as nitrogen vacancies). The Coulomb field generated by these fixed charges will scatter channel electrons. As devices evolve towards thinner barrier layers to pursue higher frequencies, the distance between surface states and the channel decreases, significantly enhancing the Coulomb scattering effect. (4) Defect scattering: This can be considered a type of Coulomb scattering, specifically referring to the scattering caused by randomly distributed Coulomb centers formed after the ionization of point defects (such as vacancies and impurities) in the epitaxial layer. Growth process conditions, especially the V / III ratio, have a decisive influence on the density of intrinsic point defects such as nitrogen vacancies.

[0005] Therefore, existing single optimization methods (such as using only AlN insertion layers or optimizing the growth temperature of the barrier layer) are insufficient to systematically suppress the aforementioned multiple scattering mechanisms simultaneously. Innovative heterojunction structure design and synergistically optimized epitaxial growth processes, comprehensively mitigating the effects of alloy disorder, interface roughness, Coulomb scattering, and defect scattering, are essential pathways to achieving breakthroughs in electron mobility and unlocking the ultimate performance potential of GaN microwave devices. This approach holds significant scientific and engineering value. Summary of the Invention

[0006] To address the shortcomings of existing technologies, this invention provides a material structure and its epitaxial growth method for improving the electron mobility of GaN heterojunctions. First, a GaN isolation layer is introduced at the interface between the AlN insertion layer and the AlGaN barrier layer to suppress disordered scattering of the barrier layer alloy. Second, the AlGaN barrier layer employs a composite growth process, where the lower AlGaN barrier layer uses a high V / III ratio and low-speed process to suppress the formation of point defects such as nitrogen vacancies and reduce defect scattering; the upper AlGaN barrier layer uses a low V / III ratio and high-speed process to improve the surface morphology of the barrier layer. Finally, an InGaN cap layer is introduced to reduce the heterostructure lattice mismatch stress, suppress lattice scattering, and simultaneously reduce the surface state density of the material, suppressing Coulomb scattering of the surface charge.

[0007] This invention is achieved through the following technical solution:

[0008] A material structure for improving the electron mobility of GaN heterojunctions comprises, from bottom to top: a single-crystal substrate, an AlN nucleation layer, a GaN buffer layer, an AlN insertion layer, a GaN isolation layer, an AlGaN barrier layer, and an InGaN cap layer; wherein the AlGaN barrier layer comprises a lower AlGaN barrier layer and an upper AlGaN barrier layer stacked from bottom to top.

[0009] Preferably, the thickness of the GaN isolation layer is 0.5~1.5 nm.

[0010] Preferably, the thickness of the lower AlGaN barrier layer is 2~10 nm, and the thickness of the upper AlGaN barrier layer is 3~20 nm.

[0011] Preferably, the thickness of the InGaN cap layer is 0.5~2.0 nm.

[0012] The epitaxial growth method for improving the electron mobility of GaN heterojunctions described above includes the following steps:

[0013] Step 1) Provide a single-crystal substrate and epitaxially grow an AlN nucleation layer on it;

[0014] Step 2) Epitaxially grow a GaN buffer layer on the AlN nucleation layer obtained in step 1);

[0015] Step 3) Epitaxially grow an AlN insertion layer on the GaN buffer layer obtained in step 2);

[0016] Step 4) Epitaxially grow a GaN isolation layer on the AlN insertion layer obtained in step 3);

[0017] Step 5) Epitaxially grow an AlGaN barrier layer on the GaN isolation layer obtained in step 4);

[0018] Step 6) An InGaN cap layer is epitaxially grown on the AlGaN barrier layer obtained in step 5).

[0019] Preferably, in step 1), the epitaxial growth temperature of the AlN nucleation layer is 1100~1250℃, the growth pressure is 30~150 torr, and the growth thickness is 30~200 nm; in step 2), the epitaxial growth temperature of the GaN buffer layer is 900~1100℃, the growth pressure is 150~500 torr, and the growth thickness is 0.2~3.0 μm; in step 3), the epitaxial growth temperature of the AlN insertion layer is 950~1100℃, the growth pressure is 30~150 torr, and the growth thickness is 0.2~3.0 μm.

[0020] Preferably, the growth thickness of the GaN isolation layer in step 4) is controlled within the range of 0.5~1.5 nm.

[0021] Preferably, step 5) of the epitaxial growth step of the AlGaN barrier layer includes:

[0022] On the GaN isolation layer, an AlGaN barrier layer is epitaxially grown at a first V / III ratio N1 and a first growth rate V1.

[0023] On the lower AlGaN barrier layer, an upper AlGaN barrier layer is epitaxially grown at a second V / III ratio N2 and a second growth rate V2.

[0024] Wherein, the first V / III ratio N1 is greater than the second V / III ratio N2, and the first growth rate V1 is less than the second growth rate V2.

[0025] Preferably, the first V / III ratio N1 ranges from 2000 to 6000, and the first growth rate V1 ranges from 1 to 3 nm / min; the second V / III ratio N2 ranges from 400 to 1300, and the second growth rate V2 ranges from 6 to 15 nm / min.

[0026] Preferably, the epitaxial growth temperature of the InGaN cap layer in step 6) is 700~900℃, the growth pressure is 100~400 torr, and the growth thickness is controlled within the range of 0.5~2.0 nm.

[0027] The beneficial effects of this invention are as follows:

[0028] (1) This invention does not optimize a single scattering source, but rather uses a multi-level design of "structural isolation - bulk quality optimization - surface / interface engineering" to synergistically suppress major scattering mechanisms such as alloy disorder scattering, defect scattering, interface roughness scattering, lattice scattering, and surface Coulomb scattering. This systematic solution overcomes the limitations of traditional single technical means and achieves a comprehensive improvement in the electron transport environment.

[0029] (2) In this invention, a GaN isolation layer of 0.5~1.5 nm is inserted between the AlN insertion layer and the AlGaN barrier layer. Without significantly changing the two-dimensional electron gas surface density, this effectively increases the spatial distance between the channel electron wave function and the disordered region of the barrier layer alloy. This significantly weakens the interaction between the penetration of the electron wave function into the AlGaN barrier layer and the fluctuations of the alloy disordered potential, thereby directly reducing the constraint of alloy disordered scattering on mobility.

[0030] (3) This invention employs a unique double-layer AlGaN barrier growth strategy: the lower barrier layer (high V / III ratio, low-speed growth) grows at a low speed in a nitrogen-rich environment, which greatly suppresses the formation of intrinsic point defects such as nitrogen vacancies, significantly reduces the charge Coulomb scattering center density within the barrier layer, and improves the intrinsic quality of the material. The upper barrier layer (low V / III ratio, high-speed growth) grows at a high speed under metal-rich conditions, which promotes atomic surface migration, facilitates obtaining an atomically smooth surface morphology, significantly reduces the interface roughness of the heterojunction, and thus effectively suppresses interface roughness scattering. This composite process enables the barrier layer to simultaneously optimize the two often mutually restrictive characteristics of "low defect mass" and "smooth interface".

[0031] (4) The present invention grows an ultrathin InGaN cap layer on the surface of the barrier layer, which plays a dual role: firstly, stress compensation. The lattice constant of InGaN is greater than that of GaN. Its introduction can partially compensate for the tensile strain borne by the AlGaN barrier layer due to its smaller lattice constant, alleviate lattice mismatch stress, and reduce lattice scattering caused by lattice distortion. Secondly, surface passivation. The InGaN cap layer effectively reduces the density of states on the AlGaN surface and reduces the number of surface state charges, thereby weakening the Coulomb scattering effect of surface charges on channel electrons.

[0032] (5) The synergistic effect of the above technical measures ultimately leads to a systematic reduction in the main scattering resistance faced by heterojunction channel electrons during transport. Therefore, this invention can significantly improve the room temperature electron mobility without sacrificing the high areal density advantage of the two-dimensional electron gas, providing a core material basis for the fabrication of GaN microwave devices with higher frequency, higher power, and lower loss, and has important industrial application value. Attached Figure Description

[0033] Figure 1 This is a schematic diagram of the AlGaN / GaN heterojunction material in Example 1;

[0034] Figure 2 This is a schematic diagram of the AlGaN / GaN heterojunction material in Comparative Example 1;

[0035] Figure 1 , 2 In the middle: 1. Single crystal substrate; 2. AlN nucleation layer; 3. GaN buffer layer; 4. AlN insertion layer; 5. GaN isolation layer; 6. Lower AlGaN barrier layer; 7. Upper AlGaN barrier layer; 8. InGaN cap layer; 9. Single-layer AlGaN barrier layer;

[0036] Figure 3 For comparison of the conduction band structure of each group of AlGaN / GaN heterojunction materials in Test Example 1;

[0037] Figure 4 For comparison of the yellow-band photoluminescence curves of each group of AlGaN / GaN heterojunction materials in Test Example 1 at room temperature;

[0038] Figure 5 The surface morphology of each group of AlGaN / GaN heterojunction materials in Test Example 1 is shown in the following figures: (a) Example 1; (b) Comparative Example 1.

[0039] Figure 6 The electron mobility data of each AlGaN / GaN heterojunction material in Test Example 1 are the on-chip five-point test data: (a) is Example 1; (b) is Comparative Example 1. Detailed Implementation

[0040] The present invention will now be described in further detail with reference to the accompanying drawings and specific embodiments.

[0041] Unless otherwise specified, the technical means used in the following embodiments are all conventional means well known to those skilled in the art, and the experimental methods without specific conditions are all conventional methods in the art.

[0042] Unless otherwise specified, the equipment and materials used in the following embodiments are commercially available.

[0043] Example 1

[0044] This embodiment provides a material structure for improving the electron mobility of GaN heterojunctions, such as... Figure 1 As shown, the structure, from bottom to top, includes: a single crystal substrate 1, an AlN nucleation layer 2, a GaN buffer layer 3, an AlN insertion layer 4, a GaN isolation layer 5, a lower AlGaN barrier layer 6, an upper AlGaN barrier layer 7, and an InGaN cap layer 8.

[0045] This embodiment also provides an epitaxial growth method for the above-mentioned material structure, the specific steps of which are as follows:

[0046] (1) Substrate preparation and baking

[0047] A 4-inch silicon carbide (SiC) single-crystal substrate was selected and placed on the base of a metal-organic chemical vapor deposition (MOCVD) apparatus. The reaction chamber temperature was raised to 1070°C and the pressure was set to 100 Torr. The substrate was baked in a hydrogen (H2) atmosphere for 10 min to remove surface contaminants.

[0048] (2) Growth of AlN nucleation layer

[0049] The reaction chamber temperature was raised to 1150℃, the pressure was maintained at 100 Torr, and ammonia (NH3) gas was introduced at a flow rate of 10 slm and trimethylaluminum (TMAl) gas at a flow rate of 300 sccm to grow an AlN nucleation layer with a thickness of 50 nm. After growth was completed, the TMAl source was turned off.

[0050] (3) Growth of GaN buffer layer

[0051] In an atmosphere continuously purging NH3, the reaction chamber temperature was lowered to 1000℃, the pressure was increased to 250 Torr, and a trimethylgallium (TMGa) source was introduced to grow a GaN buffer layer with a thickness of 2.0 μm. After growth was complete, the TMGa source was turned off.

[0052] (4) Growth of AlN insertion layer

[0053] In an NH3 atmosphere, the reaction chamber temperature was set to 1040℃ and the pressure to 100 Torr. TMAl was introduced to grow an AlN insertion layer with a thickness of 1 nm. After growth was complete, the TMAl source was turned off.

[0054] (5) Growth of GaN isolation layer

[0055] Under constant temperature (1040℃) and pressure (100 Torr) and continuous NH3 atmosphere, TMGa was introduced to grow a GaN isolation layer with a thickness of 1.0 nm. After growth was completed, the TMGa source was turned off.

[0056] The introduction of a GaN isolation layer increases the spacing between the barrier layer and the channel well, which weakens the influence of disordered scattering by the barrier layer alloy on channel electrons and is beneficial to improving electron mobility. However, if the GaN isolation layer thickness is too small (<0.5 nm), the weakening effect on disordered scattering by the barrier layer alloy is weak; if the GaN isolation layer thickness is too large (>1.5 nm), parasitic channels are easily formed at the AlGaN barrier layer / GaN transition layer interface, affecting the performance of GaN heterojunction materials. Therefore, the thickness of the GaN isolation layer needs to be controlled within the range of 0.5~1.5 nm.

[0057] (6) Growth of the lower AlGaN barrier layer

[0058] Maintaining the same temperature and pressure in an NH3 atmosphere, TMGa and TMAl were introduced. By precisely adjusting the flow rates of NH3 and the organometallic source, the V / III ratio of the reaction atmosphere (i.e., the total molar flow ratio of NH3 to the group III metal source) was achieved to 3600, and the growth rate was controlled at 2.2 nm / min. Under these conditions, Al with an aluminum composition of 28% was grown. 0.28 Ga 0.72 An N-type barrier layer, 4 nm thick, is used as the lower barrier layer. After growth is complete, the metal-organic source is turned off.

[0059] The V / III ratio (N1) of the lower AlGaN barrier layer needs to be controlled within the range of 2000–6000, and the growth rate (V1) needs to be controlled within the range of 1–3 nm / min. Using a nitrogen-rich process with a high V / III ratio and maintaining a low growth rate in the lower AlGaN barrier layer is beneficial for suppressing the formation of nitrogen vacancies and weakening defect scattering.

[0060] (7) Growth of the upper AlGaN barrier layer

[0061] Maintaining the same temperature and pressure in an NH3 atmosphere, TMGa and TMAl were introduced again. Process parameters were adjusted, reducing the V / III ratio to 650 and increasing the growth rate to 10.6 nm / min. Under these conditions, Al with the same aluminum composition of 28% was continued to grow. 0.28 Ga 0.72 An N-type barrier layer, 7 nm thick, was used as the upper barrier layer. After growth was complete, the metal-organic source was turned off.

[0062] The V / III ratio (N2) of the upper AlGaN barrier layer needs to be controlled within the range of 400–1300, and the growth rate (V2) needs to be controlled within the range of 6–15 nm / min. Using a low V / III ratio aluminum-rich process and maintaining a high growth rate for the upper AlGaN barrier layer is beneficial for improving the surface morphology of the barrier layer and weakening interface roughness scattering.

[0063] (8) Growth of InGaN cap layer

[0064] In an NH3 atmosphere, the reaction chamber temperature was lowered to 800°C, the pressure was set to 300 Torr, and TMGa and trimethylindium (TMIn) were introduced. Indium was grown with an indium composition of 8% In. 0.08 Ga 0.92 The N-cap layer was 1.0 nm thick. After growth was complete, all metal-organic sources were turned off.

[0065] The lattice constant of AlGaN is smaller than that of GaN, resulting in a stretched lattice in the AlGaN barrier layer. This stretching becomes more pronounced with increasing Al content, leading to more severe lattice scattering. InGaN, on the other hand, has a larger lattice constant than GaN. Therefore, introducing an InGaN cap layer onto the AlGaN barrier layer can alleviate the stretching of the AlGaN lattice and weaken lattice scattering. However, if the InGaN cap layer thickness is too small (<0.5 nm), the weakening effect on lattice scattering is limited; if the InGaN cap layer thickness is too large (>2.0 nm), the enhanced anti-polarization effect leads to a decrease in the two-dimensional electron gas surface density of the channel. Therefore, the thickness of the InGaN cap layer needs to be controlled within the range of 0.5–2.0 nm. Furthermore, the introduction of the InGaN cap layer can reduce the surface state density of the AlGaN barrier layer, thereby weakening Coulomb scattering during surface state ionization.

[0066] (9) Cooling and removing the tablets

[0067] Under a continuous NH3 protective atmosphere, the temperature of the reaction chamber is reduced to room temperature, and the finally obtained epitaxial wafer (AlGaN / GaN heterojunction material) is removed.

[0068] Comparative Example 1

[0069] To verify the effectiveness of this invention, a comparative sample (AlGaN / GaN heterojunction material) was grown using conventional processes, and its structure is as follows. Figure 2 As shown, from bottom to top, the layers are: 1. SiC single crystal substrate; 2. 50 nm thick AlN nucleation layer; 3. 2.0 μm thick GaN buffer layer; 4. 1 nm thick AlN insertion layer; 5. 11 nm thick monolayer AlGaN barrier layer; 6. (Al 0.28 Ga 0.72 N).

[0070] Test Example 1

[0071] The samples prepared in Example 1 of this invention and the conventional samples prepared in Comparative Example 1 were characterized and analyzed, and the results are as follows:

[0072] 1. Structural Advantages Analysis

[0073] like Figure 3 As shown, compared to the conventional sample in Comparative Example 1, Example 1 of this invention introduces a 1.0 nm GaN isolation layer between the AlN insertion layer and the AlGaN barrier layer. This increases the spatial distance between the two-dimensional electron gas channel and the disordered region of the barrier layer alloy. Simultaneously, theoretical calculations show that the electron potential well depth at the AlGaN barrier layer / GaN isolation layer interface is +0.3 eV, higher than the Fermi level, preventing the formation of parasitic channels and effectively weakening the influence of alloy disorder scattering on channel electrons. Therefore, the introduction of the GaN isolation layer is beneficial for improving electron mobility.

[0074] 2. Defect suppression effect

[0075] like Figure 4 As shown, the room-temperature yellow band photoluminescence (YL) signal intensity of the sample in Example 1 of this invention decreased by 35% compared to the conventional sample in Comparative Example 1. Since the yellow band luminescence intensity is directly related to the point defect density such as nitrogen vacancies in the AlGaN barrier layer, this result indicates that the high V / III ratio and low-speed growth process used in the lower AlGaN barrier layer can effectively reduce the point defect density of the barrier layer, thereby suppressing defect scattering.

[0076] 3. Improved surface morphology

[0077] like Figure 5 As shown, the sample of Example 1 of the present invention ( Figure 5 (a) The surface of the upper AlGaN barrier layer, compared with the sample grown using conventional process in Comparative Example 1. Figure 5 Compared to (b), the surface roughness (RMS) within a 5 μm × 5 μm range is significantly reduced from 0.374 nm to 0.192 nm, and the surface pit defect density is also lower. This demonstrates that the low V / III ratio and high-speed growth process used in the upper AlGaN barrier layer can effectively improve the surface morphology of the barrier layer, thereby suppressing interface roughness scattering.

[0078] 4. Verification of the function of the cap layer

[0079] The areal density of the two-dimensional electron gas was measured before and after the introduction of the InGaN cap layer, and the results showed no significant change. However, the room-temperature electron mobility increased from 2194.82 cm⁻¹. 2 / V·s increased to 2270.13 cm 2 / V·s. This indicates that the introduction of the InGaN cap layer, while maintaining a high areal density, effectively suppresses lattice scattering and surface Coulomb scattering by alleviating lattice mismatch stress and reducing surface state density, thereby improving electron mobility.

[0080] 5. Overall performance improvement

[0081] like Figure 6 As shown, five-point sampling tests were performed on the entire epitaxial wafer. The average intra-wafer room-temperature electron mobility of the AlGaN / GaN heterojunction material prepared in Example 1 of this invention reached 2270.13 cm⁻¹. 2 / V·s( Figure 6 (a) Compared to the sample grown using conventional techniques in Comparative Example 1 ( Figure 6 Middle (b), mean 2089.62 cm 2The electron mobility of GaN heterojunctions has been significantly improved ( / V·s). This comprehensively demonstrates that the structural design and epitaxial method of this invention can effectively suppress multiple scattering mechanisms in a synergistic manner, ultimately achieving a breakthrough in the electron mobility of GaN heterojunctions.

[0082] The embodiments described above are only some, not all, of the embodiments of the present invention. The detailed description of the embodiments of the present invention is not intended to limit the scope of the claimed invention, but merely to illustrate selected embodiments. The scope of protection of the present invention is determined by the scope claimed in the claims. All other embodiments obtained by those skilled in the art based on the embodiments of the present invention without inventive effort are within the scope of protection of the present invention.

Claims

1. A material structure for improving the electron mobility of GaN heterojunctions, characterized in that, From bottom to top, it includes: a single crystal substrate, an AlN nucleation layer, a GaN buffer layer, an AlN insertion layer, a GaN isolation layer, an AlGaN barrier layer, and an InGaN cap layer; wherein, the AlGaN barrier layer includes a lower AlGaN barrier layer and an upper AlGaN barrier layer stacked from bottom to top.

2. The material structure for improving the electron mobility of GaN heterojunctions according to claim 1, characterized in that, The thickness of the GaN isolation layer is 0.5~1.5 nm.

3. The material structure for improving the electron mobility of GaN heterojunctions according to claim 1, characterized in that, The thickness of the lower AlGaN barrier layer is 2~10 nm, and the thickness of the upper AlGaN barrier layer is 3~20 nm.

4. The material structure for improving the electron mobility of GaN heterojunctions according to claim 1, characterized in that, The thickness of the InGaN cap layer is 0.5~2.0 nm.

5. An epitaxial growth method for a material structure that improves the electron mobility of a GaN heterojunction as described in any one of claims 1-4, characterized in that, Includes the following steps: Step 1) Provide a single-crystal substrate and epitaxially grow an AlN nucleation layer on it; Step 2) Epitaxially grow a GaN buffer layer on the AlN nucleation layer obtained in step 1); Step 3) Epitaxially grow an AlN insertion layer on the GaN buffer layer obtained in step 2); Step 4) Epitaxially grow a GaN isolation layer on the AlN insertion layer obtained in step 3); Step 5) Epitaxially grow an AlGaN barrier layer on the GaN isolation layer obtained in step 4); Step 6) An InGaN cap layer is epitaxially grown on the AlGaN barrier layer obtained in step 5).

6. The epitaxial growth method according to claim 5, characterized in that, Step 1) The epitaxial growth temperature of the AlN nucleation layer is 1100~1250℃, the growth pressure is 30~150 torr, and the growth thickness is 30~200 nm; Step 2) The epitaxial growth temperature of the GaN buffer layer is 900~1100℃, the growth pressure is 150~500 torr, and the growth thickness is 0.2~3.0 μm; Step 3) The epitaxial growth temperature of the AlN insertion layer is 950~1100℃, the growth pressure is 30~150 torr, and the growth thickness is 0.2~3.0 μm.

7. The epitaxial growth method according to claim 5, characterized in that, Step 4) The growth thickness of the GaN isolation layer is controlled within the range of 0.5~1.5 nm.

8. The epitaxial growth method according to claim 5, characterized in that, Step 5) The epitaxial growth step of the AlGaN barrier layer includes: On the GaN isolation layer, an AlGaN barrier layer is epitaxially grown at a first V / III ratio N1 and a first growth rate V1. On the lower AlGaN barrier layer, an upper AlGaN barrier layer is epitaxially grown at a second V / III ratio N2 and a second growth rate V2. Wherein, the first V / III ratio N1 is greater than the second V / III ratio N2, and the first growth rate V1 is less than the second growth rate V2.

9. The epitaxial growth method according to claim 8, characterized in that, The first V / III ratio N1 ranges from 2000 to 6000, and the first growth rate V1 ranges from 1 to 3 nm / min; the second V / III ratio N2 ranges from 400 to 1300, and the second growth rate V2 ranges from 6 to 15 nm / min.

10. The epitaxial growth method according to claim 5, characterized in that, Step 6) The epitaxial growth temperature of the InGaN cap layer is 700~900℃, the growth pressure is 100~400 torr, and the growth thickness is controlled within the range of 0.5~2.0 nm.

Images