TWM629298U - Gallium nitride epitaxial substrate having 2D material interposer - Google Patents

Abstract

This creation discloses a gallium nitride epitaxial substrate with a 2D material interposer, including a polycrystalline AlN substrate; a _SiO2 bonding layer on the polycrystalline AlN substrate; a c-plane sapphire bonding layer on the _SiO2 bonding layer; c-plane sapphire A polycrystalline 2D material interposer is grown on the bonding layer. The polycrystalline 2D material interposer has at least one top layer, and the lattice constant of the top layer is highly matched with AlN, AlGaN or GaN; on the polycrystalline 2D material interposer, van der Waals epitaxy is used A GaN single crystal epitaxial layer is grown, or, by means of van der Waals epitaxy, an AlN or AlGaN nucleation auxiliary layer is grown, and a GaN single crystal epitaxial layer is formed on the AlN or AlGaN nucleation auxiliary layer. This creation avoids the 2D material interlayer transfer process and possible quality effects, effectively overcomes the defect quality problem of the gallium nitride layer caused by the mismatch of the heteroepitaxial lattice, and can alleviate some of the thermal stress problems caused by different thermal expansion coefficients; To carry out the growth of high-quality AlN, AlGaN and GaN epitaxial layers for the production of wide energy gap optoelectronic and semiconductor components such as GaN systems.

Description

GaN epitaxial substrate with 2D material interposer

This creation involves a gallium nitride epitaxial substrate with a 2D material interposer.

In the manufacturing process of optoelectronic and semiconductor components, epitaxy has an important impact on the quality of products. The impact on quality includes component performance, yield, reliability and life. Generally, the material of the substrate hopes to minimize the defect density of the single crystal material. Only when the crystal structure, lattice constant (lattice constant), coefficient of thermal expansion (CTE, coefficient of thermal expansion) match with the epitaxial material can it be avoided as much as possible in the epitaxy process. affect crystal quality. In recent years, the third-generation semiconductor technology and market have developed rapidly with the demand for power and high-frequency semiconductor components. The basis for quality improvement depends on the supply of high-quality epitaxial substrates of silicon carbide and gallium nitride, the two main players of the third-generation semiconductor materials. Unlike gallium nitride based LEDs that use sapphire substrates, according to current technology, the most commonly used gallium nitride substrates are gallium nitride on silicon (GaN-on-Si) and gallium nitride on silicon carbide (GaN-on-Si) on-SiC) two substrates.

The main reason comes from the current cost and size constraints of the development of GaN single crystal technology. The melting points of both aluminum nitride and gallium nitride are above 2500 degrees Celsius and there is a problem of high vapor pressure. In other words, if you want to directly produce single crystal substrates of the above two materials by the method of melting crystal growth, not only the manufacturing cost If it is higher, it will generate more waste heat and cause inevitable pollution to the environment. For the vapor phase method, the current GaN crystal growth method is the Hydride Vapor Phase Epitaxy (HVPE) method to produce single crystal GaN substrates. Due to the limitations of production cost and yield conditions, currently The mass production technology reaches 4-inch substrates and the cost is extremely high. In fact, the defect density of the above-mentioned vapor phase method is still higher than that of other liquid phase crystallization processes, but it is limited by the slow growth rate of the remaining processes and the higher cost of mass production. Considering the trade-off of quantity, the mainstream of commercial transformation is still limited to the HVPE method. The literature points out that the vapor-phase GaN crystal growth rate is still possible to increase several times and maintain good crystallinity, but due to the deterioration of defect density, it has not been used as an orientation to reduce the cost of GaN substrates. As for the aluminum nitride crystal growth technology, physical vapor transport (PVT), one of the vapor-phase methods, is used to produce single-crystal aluminum nitride substrates. Due to the limitations of production technology and yield, there are only two companies in the world. Manufacturers have mass production capacity. At present, the mass production technology only reaches 2-inch substrates and the cost is extremely high, and the production capacity is all occupied by a few manufacturers and cannot be widely supplied to the market. Due to the chemical characteristics of aluminum nitride itself and the limitation of physical vapor transport method hardware components, the existence of carbon (C) and oxygen (O) impurities in the finished single crystal product is inevitable, which also affects the component characteristics to a certain extent.

Table 1 Material crystal structure Lattice constant Thermal expansion coefficient a c × ^10-6 ×K ^{- 1} Gallium Nitride GaN Wurtzite 0.31885 0.5185 α _a 5.59 α _c 3.17 Aluminum Nitride AlN Wurtzite 0.31106 0.49795 α _a 4.15 α _c 5.27 Zinc oxide ZnO Wurtzite 0.32496 0.52065 α _a 4.31 α _c 2.49 Silicon carbide SiC 6H Wurtzite 0.30806 1.51173 α _a 4.3 α _c 4.7 Sapphire Sapphire Rhombohedral 0.4765 1.2982 α _a 6.66 α _c 5 Si Si DiamondDiamond 0.5431 2.6

A similar situation also exists in the current silicon carbide (SiC) single crystal. Silicon carbide substrates are currently the substrate materials for high-performance power semiconductors and high-end light-emitting diodes. The single crystal growth process is physical vapor transport in the vapor phase method. PVT (Physical Vapor Transport, PVT), high-quality and large-size silicon carbide single crystal growth technology is difficult, high-end mass production technology is in the hands of a few manufacturers, and there is still a lot of room for improvement in the application cost. Gallium Nitride on Silicon Carbide (GaN-on-SiC) is a high-quality GaN epitaxial substrate, but for the above reasons, large-scale substrates have problems such as high price, limited supply and technology in the hands of a few manufacturers; relatively In other words, due to the large size, low cost, high productivity and stable quality of silicon substrates, the development of gallium nitride (GaN-on-Si) substrates on silicon wafers is more commonly concerned by relevant manufacturers.

Two substrate technologies, GaN-on-Si and GaN-on-SiC, both belong to heterojunction epitaxy in terms of epitaxy process, and heteroepitaxial needs to overcome different The lattice matching problem between materials and the thermal stress problem caused by the difference in thermal expansion coefficient between the epitaxial layer and the substrate, the higher quality of GaN-on-SiC than GaN-on-Si is precisely because of the lattice mismatch of GaN-on-SiC The degree of (lattice mismatch) is smaller than that of GaN-on-Si; another important feature is that the gallium nitride layer has significant tensile stress on the silicon surface. When the thickness of the gallium nitride layer is increased, the stress is higher, resulting in bending deformation of the substrate or even The gallium nitride layer may crack, and the associated effect becomes more severe as the wafer size increases. Relevant technical difficulties have resulted in a generally low yield rate of GaN-on-Si, and it is mostly used in power supply products. Currently, mass production is still dominated by six inches, and the advantages of large silicon wafers cannot be fully utilized.

Two-dimensional (2D) materials is a rapidly developing emerging field. The earliest and most well-known material in the 2D material family is graphene, which has a two-dimensional layered structure with special or Excellent physical/chemical/mechanical/optical properties, there is no strong bond between layers, only by van der Waals force, which also means that there is no dangling bond on the surface of the layered structure ) exists, and graphene has been confirmed to have extensive and excellent application potential; graphene research and development work is generally carried out around the world, and it also drives the research and development of more 2D materials, including hexagonal boron nitride hBN (hexagonal Boron Nitride), transition metal Dichalcogenides TMDs (transition metal dichalcogenides) and black phosphorus are also among the 2D material families that have accumulated more research and development achievements. The above materials each have specific material properties and application potential, and the development of manufacturing technology for related materials is also active. in progress. In addition to excellent optoelectronic properties, graphene, hBN and MoS ₂ , one of the TMDs materials, are considered to have excellent diffusion barrier properties and varying degrees of high temperature stability, especially hBN has excellent chemical passivation Inertness and high temperature oxidation resistance.

Due to the above-mentioned nature of the layered structure and the bonding characteristics of the Van der Waals forces between layers, the technical feasibility of fabricating two or more materials in the 2D material family into layered stacked hetero-structures (hetero-structures) is greatly open. It is possible to create new application characteristics or make new components, and research and development in the field of optoelectronics and semiconductors is currently very active. Specifically, it can be a mechanical composition stack, and it can also be physical or chemical vapor deposition.

The Van der Waals force binding properties of 2D materials have also attracted attention for the use of epitaxial substrates applied to traditional 3D materials. of thermal expansion) must be very well matched with the substrate material, but in reality, it is often encountered that the subject of the present invention lacks suitable substrate materials, or the ideal substrate material is expensive or difficult to obtain. At this time, 2D materials are suitable for heteroepitaxial substrates. Another solution is provided, the so-called van der Waals Epitaxy. The mechanism that van der Waals epitaxy may be beneficial to heteroepitaxy comes from the fact that the direct chemical bond of the traditional epitaxial interface is replaced by the van der Waals force bond, which will make the stress or strain energy from the lattice and thermal expansion mismatch in the epitaxy process obtain a certain amount. The degree of relaxation, so that the quality of the epitaxial layer can be improved, or through the introduction of 2D materials and van der Waals epitaxy, some hetero-epitaxial technologies that were not practical before become possible. Relevant studies also pointed out that when the above-mentioned 2D materials are stacked with each other in a heterostructure, the interaction force is dominated by the Van der Waals force; and the epitaxial time of the 3D material on the 2D material is due to the dangling bonds of the 3D material on the interface ( dangling bond) while contributing to the cohesion of the interface, this epitaxy is not essentially pure van der Waals Epitaxy or, more precisely, Quasi van der Waals Epitaxy; whether In any case, the matching degree of lattice and thermal expansion undoubtedly still plays a certain role in the final epitaxy quality. Both the 2D material interposer and the substrate material contribute to the overall matching degree. The above-mentioned 2D layered material has a hexagon or honeycomb structure, and is considered to be structurally compatible with Wurtzite and Zinc-Blende structural materials in epitaxial time, the related field of the present invention The main epitaxial materials belong to this type of structure.

Based on the use of epitaxial substrates, single crystal is one of the requirements to ensure epitaxial quality. Generally, the growth of 2D materials tends to correlate with the crystal orientation of the crystalline substrate during the nucleation stage. It belongs to a polycrystalline structure, and the 2D material has already formed an inconsistent direction in the nucleation stage. After the nuclei aggregate into a continuous film with the growth, there are still domains with different orientations instead of single crystals; when the substrate is a single crystal material such as sapphire, it is still Due to the symmetry correlation between the two structures, the possible specific nucleation direction is not unique, and it is impossible to form a single crystal continuous film. Recent studies have found that by improving the existing process, when the copper foil is heat-treated to form a copper foil with a specific lattice orientation, the anisotropic lattice domains formed during the growth of 2D materials graphene and hexagonal boron nitride (hBN) can be eliminated. ) feature while growing into continuous thin films of single-crystal graphene and hexagonal boron nitride.

In recent years, studies have pointed out that layered MoS ₂ , WS ₂ , MoSe ₂ , WSe ₂ and other TMD materials with good crystallinity can be grown on the c-plane sapphire surface of single crystal by CVD or other methods. There are two types of TMD materials grown. (0º and 60º) crystal orientation (Reference: Nature 2019, v.567, 169-170). For the AlGaN and GaN materials concerned by the present invention, the crystal structure has hexagonal symmetry on the epitaxial junction. Although the above-mentioned TMD layer does not constitute a single crystal layer, theoretically it does not hinder AlGaN and GaN epitaxy when used as an epitaxial substrate. The layer forms a single crystal. At present, when applying on the surface of substrates other than sapphire, the two-dimensional material layer grown on the surface of high-quality sapphire is often transferred to other substrate surfaces through a transfer process; the two-dimensional material layer grown on the surface of high-quality sapphire is directly applied without transfer. In theory, the possible quality effects of the two-dimensional material layer including defects, wrinkles and surface contamination residues from the transfer process can be avoided.

The existing process gallium nitride on silicon wafer (GaN-on-Si), as shown in Figure 1. Heteroepitaxy needs to overcome the lattice matching problem between different materials, as well as the thermal stress problem caused by the different thermal expansion coefficients between the epitaxial layer and the substrate. The degree of GaN-on-Si lattice mismatch is high, resulting in epitaxy During the process, the defect density of the gallium nitride layer is relatively high; another important feature is that the gallium nitride layer has significant tensile stress on the silicon surface. When the thickness of the gallium nitride layer is increased, the stress is higher, resulting in the bending deformation of the substrate or even the gallium nitride. Layers may crack, and the associated effects become more severe as wafer size increases. Relevant technical difficulties have resulted in a generally low yield rate of GaN-on-Si, and it is mostly used in power supply products. Currently, mass production is still dominated by six inches, and the advantages of large silicon wafers cannot be fully utilized.

GaN-on-Sapphire of the existing process is shown in Figure 2. Heteroepitaxy needs to overcome the lattice matching problem between different materials, as well as the thermal stress problem caused by the different thermal expansion coefficients between the epitaxial layer and the substrate. The high degree of GaN-on-Sapphire lattice mismatch leads to the epitaxy process. The defect density of the medium GaN layer is at a certain level, and it still has an irreplaceable position under the long-term commercial technology development of light-emitting diodes; however, in the field of high-frequency and power semiconductors, due to the low thermal conductivity of sapphire, GaN on sapphire is caused. (GaN-on-Sapphire) applications are hindered.

In order to solve the problems existing in the existing process, this creation provides a gallium nitride epitaxial substrate with a 2D material interposer.

The authored solution is as follows:

A gallium nitride epitaxial substrate with a 2D material interposer includes a polycrystalline AlN substrate; the polycrystalline AlN substrate has a SiO ₂ bonding layer; the SiO ₂ bonding layer has a c-plane sapphire bonding layer; the c-plane sapphire bonding layer is on the A polycrystalline 2D material interposer is grown. The polycrystalline 2D material interposer has at least one top layer, and the lattice constant of the top layer is highly matched with AlN, AlGaN or GaN; GaN is grown on the polycrystalline 2D material interposer by means of van der Waals epitaxy. A single crystal epitaxial layer, or a polycrystalline 2D material interlayer, is grown with an AlN or AlGaN nucleation auxiliary layer by means of van der Waals epitaxy, and then a GaN single crystal epitaxial layer is formed on the AlN or AlGaN nucleation auxiliary layer.

The thickness of the 2D material interposer is greater than 0.5 nm.

The thickness of the c-plane sapphire bonding layer is greater than 10 nm.

The 2D material interposer is a 2D layer suitable for GaN, AlGaN or AlN epitaxy.

The 2D material interposer is a single-layer structure with only a top layer, and the top layer is a 2D material suitable for GaN, AlGaN or AlN epitaxy.

The 2D material interlayer is a composite layer structure formed by a top layer and a bottom layer, the top layer is a 2D material suitable for GaN, AlGaN or AlN epitaxy, and the bottom layer is a 2D material suitable for a single crystal base layer.

The top layer adopts WS ₂ or MoS ₂ ; the bottom layer adopts hBN.

The single-layer structure or composite-layer structure of the 2D material interposer has a top lattice constant a that does not match AlN, AlGaN or GaN by more than 20% and is suitable for AlN, AlGaN or GaN epitaxy.

At least the top layer of the polycrystalline 2D material interposer is composed of two crystalline domains whose orientations are matched at an angle of 60 degrees to each other.

After adopting the above scheme, in this work, the high-quality single-crystal c-plane (c-plane) sapphire lamination thin layer is cut and then bonded to the surface of the polycrystalline AlN substrate, which can directly grow layers with good crystallinity and there are two types (0 ° and 60°) polycrystalline oriented 2D material interposer, which avoids the transfer process and possible quality effects; the formation of a substrate whose surface lattice constant is highly matched with AlN, AlGaN and GaN can effectively overcome the Heteroepitaxial lattice mismatch leads to defect quality problems in gallium nitride layers; the characteristics of van der Waals junctions can alleviate some of the thermal stress problems caused by different thermal expansion coefficients. Since only the sapphire bonding layer is used and the bonding body is a polycrystalline AlN substrate with excellent thermal conductivity, the overall substrate structure and the heat dissipation performance of the components can be maintained at a good level. Therefore, the substrate structure of the present invention is beneficial to be used for the production of high-quality AlN, AlGaN and GaN epitaxial layers for the fabrication of GaN-based optoelectronic and semiconductor components with wide energy gaps.

The present creation will be further described in detail below in conjunction with the accompanying drawings and specific embodiments.

Please refer to FIG. 3 to FIG. 6 , which are embodiments of the present invention of a gallium nitride epitaxial substrate with a 2D material interposer, including a polycrystalline AlN substrate 1 . The polycrystalline AlN substrate 1 has a SiO ₂ bonding layer 2 on it. The SiO ₂ bonding layer 2 has a c-plane sapphire bonding layer 3 . The preferred design of the c-plane sapphire bonding layer 3 is that the thickness is greater than 10 nm. A polycrystalline 2D material interposer is grown on the c-plane sapphire bonding layer 3 . The 2D material interposer is a 2D layer suitable for GaN, AlGaN or AlN epitaxy. The preferred design of the 2D material interposer is that the thickness is greater than 0.5 nm. The polycrystalline 2D material interposer has at least one top layer 41, and the lattice constant of the top layer 41 is highly matched with AlN, AlGaN or GaN. A GaN single crystal epitaxial layer 5 is directly grown on the top layer 41 of the polycrystalline 2D material interposer by means of van der Waals epitaxy, or AlN or AlGaN is first grown on the top layer 41 of the polycrystalline 2D material interposer by means of van der Waals epitaxy. The nucleation auxiliary layer 6 is further provided with a GaN single crystal epitaxial layer 5 on the AlN or AlGaN nucleation auxiliary layer 6 .

Specifically, as shown in the first embodiment shown in FIG. 3 and the third embodiment shown in FIG. 5 , the 2D material interposer has a single-layer structure and only has a top layer 41 , and the top layer 41 is suitable for GaN, AlGaN or AlN epitaxy 2D material. In the first embodiment, the GaN single crystal epitaxial layer 5 is directly grown on the top layer 41 by means of van der Waals epitaxy. In the third embodiment, an AlN or AlGaN nucleation auxiliary layer 6 is first grown on the top layer 41 by van der Waals epitaxy, and then a GaN single crystal epitaxial layer 5 is formed on the AlN or AlGaN nucleation auxiliary layer 6 .

In the second embodiment shown in FIG. 4 and the fourth embodiment shown in FIG. 6 , the 2D material interposer is a composite layer structure formed by a top layer 41 and a bottom layer 42 , and the top layer 41 is suitable for GaN, AlGaN or AlN epitaxy 2D material, the bottom layer 42 is a 2D material suitable as a single crystal base layer. In the third embodiment, the GaN single crystal epitaxial layer 5 is directly grown on the top layer 41 by means of van der Waals epitaxy. In the fourth embodiment, an AlN or AlGaN nucleation auxiliary layer 6 is first grown on the top layer 41 by van der Waals epitaxy, and then a GaN single crystal epitaxial layer 5 is formed on the AlN or AlGaN nucleation auxiliary layer 6 .

The top layer 41 described in this creation is a 2D layer that meets requirements such as lattice constant matching, such as using WS ₂ or MoS ₂ , and the bottom layer 42 is a 2D layer suitable as a base layer, such as using hBN, but not limited to the above materials.

Table 2 Material Lattice constant a (nm) Hexagonal Boron Nitride hBN 0.25 graphene graphene 0.246 WS ₂ 0.318 MoS ₂ 0.3161 WSe ₂ 0.3297 MoSe ₂ 0.3283

Whether the 2D material interposer has a single-layer structure or a composite-layer structure, the lattice constant a of the top layer 41 does not match AlN, AlGaN or GaN by more than 20% and is suitable for AlN, AlGaN or GaN epitaxy. At least the top layer 41 of the polycrystalline 2D material interposer is composed of two crystalline domains with matching directions at an angle of 60 degrees.

This invention creates a method for preparing a gallium nitride-on-silicon GaN-on-Si epitaxial substrate with a 2D material interposer. The steps are as follows:

Step 1, using the surface-polished polycrystalline AlN substrate 1 as the starting material, through appropriate process treatment (including thin film evaporation, chemical mechanical polishing, spin-on-glass and heat treatment, etc.) to make the substrate surface highly flattened, as a follow-up Preparation of manufacturing procedures; SiO ₂ bonding layer 2 is coated on polycrystalline AlN substrate 1;

Step 2, using the existing process technology, transfer and bond the c-plane sapphire bonding layer 3 from the surface of the c-plane sapphire wafer to the SiO ₂ bonding layer 2 on the surface of the aforementioned polycrystalline AlN substrate 1;

Step 3, using the existing manufacturing process, grow a polycrystalline 2D material interposer on the surface of the c-plane sapphire wafer;

Step 4, using the Van der Waals epitaxy technology, the subsequent GaN epitaxy can be continued on the polycrystalline AlN substrate 1 with the polycrystalline 2D material intermediary layer on the surface in Step 3; or the AlN or AlGaN nucleation layer coating can be carried out first and then continue GaN epitaxy is performed.

This work uses the application of polycrystalline to 2D material heterojunction interposer and van der Waals epitaxy (VDWE) to directly grow the surface of sapphire single crystal layer to form a substrate whose surface lattice constant is highly matched with AlN, AlGaN and GaN. This creation effectively overcomes the defect quality problem of the gallium nitride layer caused by the mismatch of the heteroepitaxial lattice, and alleviates the thermal stress problem caused by the different thermal expansion coefficients. The overall substrate structure and component heat dissipation performance of this creation can maintain a good level, which is conducive to the growth of high-quality AlN, AlGaN and GaN epitaxial layers for the production of GaN-based optoelectronic and semiconductor components with wide energy gaps.

The above descriptions are only preferred embodiments of the present creation, and are not intended to limit the present creation. It should be pointed out that after reading this specification, the equivalent changes made by those skilled in the art according to the design ideas of this case all fall into the protection scope of this case.

1: Polycrystalline AlN substrate 2: SiO ₂ bonding layer 3: Sapphire bonding layer 41: Top layer 42: Bottom layer 5: GaN single crystal epitaxial layer 6: AlN or AlGaN nucleation auxiliary layer

FIG. 1 is a schematic diagram of a gallium nitride (GaN-on-Si) structure on a silicon wafer in the prior art; FIG. 2 is a schematic diagram of a GaN-on-Sapphire structure in the prior art; Fig. 3 is the structural representation of embodiment one of this creation; Fig. 4 is the structural representation of embodiment two of this creation; Fig. 5 is the structural representation of embodiment three of this creation; FIG. 6 is a schematic structural diagram of Embodiment 4 of the present invention.

1: Polycrystalline AlN substrate

2: SiO ₂ bonding layer

3: Sapphire lamination layer

41: top floor

5: GaN single crystal epitaxial layer

Claims (10)

A gallium nitride epitaxial substrate with a 2D material interposer, comprising: a polycrystalline AlN substrate; a _SiO2 bonding layer on the polycrystalline AlN substrate; a c-plane sapphire bonding layer on the _SiO2 bonding layer; the c A 2D material interposer with a polycrystalline orientation is grown on the sapphire bonding layer, and the 2D material interposer in the polycrystalline orientation has at least one top layer, and the lattice constant of the top layer is highly matched with AlN, AlGaN or GaN; A GaN single crystal epitaxial layer is grown on the 2D material interposer by van der Waals epitaxy, or an AlN or AlGaN nucleation auxiliary layer is grown on the polycrystalline 2D material interposer by van der Waals epitaxy, and then the AlN Or there is a GaN single crystal epitaxial layer on the AlGaN nucleation auxiliary layer. The gallium nitride epitaxial substrate with a 2D material interposer according to claim 1, wherein: the thickness of the 2D material interposer is greater than 0.5 nm. The gallium nitride epitaxial substrate with a 2D material interposer according to claim 1, wherein: the thickness of the c-plane sapphire bonding layer is greater than 10 nm. The gallium nitride epitaxial substrate with a 2D material interposer according to claim 1, wherein: the 2D material interposer is a 2D layer suitable for GaN, AlGaN or AlN epitaxy. The gallium nitride epitaxial substrate with a 2D material interposer according to claim 1, wherein: the 2D material interposer has a single-layer structure and only has the top layer, and the top layer is a 2D material suitable for GaN, AlGaN or AlN epitaxy . The gallium nitride epitaxial substrate with a 2D material interlayer according to claim 5, wherein: the top layer adopts WS ₂ or MoS ₂ . The gallium nitride epitaxial substrate with a 2D material interposer according to claim 1, wherein: the 2D material interposer is a composite layer structure formed by the top layer and a bottom layer, and the top layer is suitable for GaN, AlGaN or AlN epitaxy 2D material, the bottom layer is a 2D material suitable as a single crystal base layer. The gallium nitride epitaxial substrate with a 2D material interlayer according to claim 7, wherein: the top layer adopts WS ₂ or MoS ₂ ; the bottom layer adopts hBN. The gallium nitride epitaxial substrate with a 2D material interposer according to any one of claims 1 to 8, wherein: the top lattice constant a of the single-layer structure or composite layer structure of the 2D material interposer is related to AlN, AlGaN Or GaN mismatch not more than 20% and suitable for AlN, AlGaN or GaN epitaxy. The gallium nitride epitaxial substrate with a 2D material interposer according to any one of claims 1 to 8, wherein: at least the top layer of the 2D material interposer in a polycrystalline orientation is made of two kinds of materials that are matched at an angle of 60 degrees to each other oriented crystalline regions.

Images