Semiconductor structure
By growing the AlGaN base layer on the substrate and forming an InGaN quantum dot layer group, the dislocations are encapsulated by quantum dots and column structures, the problem of dislocation diffusion in nitride semiconductors is solved, and the crystal quality of the semiconductor structure and the performance of electronic equipment are improved, especially the high-frequency application of GaN HEMT.
Patent Information
- Application Number
- CN202480006110.0
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- Priority Date
- 2023-01-27
- Filing Date
- 2024-01-25
- Publication Date
- 2025-08-01
AI Technical Summary
There is a dislocation problem in existing nitride semiconductors, which affects device quality and performance. It is difficult to control dislocations when growing nitride semiconductors on non-nitride substrates.
The AlGaN base layer is grown on the substrate and the InGaN quantum dot layer group is formed thereon. Through the self-assembly of quantum dots and lattice mismatch mechanism, the vertical diffusion of dislocations is encapsulated and reduced by the lateral development of quantum dots. The dislocation bending propagates in the lateral direction by combining the column and nanowire structures.
Effectively reduce the vertical diffusion of dislocations, improve the crystal quality of semiconductor structures, and promote the high switching frequency and high electron mobility of high frequency electronic devices such as GaN HEMT, which are suitable for frequency applications of ultra-large-scale integrated devices and broadband cellular networks.
Smart Images
Figure CN120418929A_ABST
Abstract
Description
Technical Field
[0001] The inventive concept generally relates to semiconductor structures. Background Art
[0002] Nitride semiconductors (i.e., compounds of elements from Group III of the periodic table and nitrogen) offer many advantages over silicon. Devices based on them provide promising candidates for replacing silicon-based electronic devices. For example, such devices can offer faster switching speeds, increased electron mobility, lower resistance, greater breakdown voltage, etc.
[0003] However, nitride semiconductors typically include dislocations. Various methods have been developed to reduce such dislocations, but further improvements are possible. Summary of the Invention
[0004] An object of the inventive concept is to provide a semiconductor structure that facilitates high-quality nitride semiconductor epitaxial growth. Another object of the inventive concept is to facilitate high-quality electronic devices, such as fast, reliable, and inexpensive electronic devices.
[0005] These and other objects of the inventive concept are at least partially met by the invention as defined in the independent claims. Preferred embodiments are set forth in the dependent claims.
[0006] In this text, abbreviations of natural elements will be used according to the periodic table. For example, Al means aluminum, Ga means gallium, As means arsenic, Mg means magnesium, O means oxygen, and Si means silicon.
[0007] In this text, compounds including multiple elements can be referred to by combinations of element symbols. For example, gallium nitride can be called GaN, and aluminum nitride can be called AlN. Some compounds can be provided in terms of the ratio of their relative element compositions. For example, aluminum gallium nitride can be called Al
[0007] , 0.5 , 1-x ,
[0010] ,
[0009] ,
[0008] , , 0.5 , x , , Ga 1-x N, where x is a variable that can assume values from 0 to 1. The subscript suffix indicates the relative ratio of the preceding element. If, for example, x = 0.5 or Al 0.5 Ga 0.5 N, it can be understood that 50% is aluminum and 50% is gallium in terms of the Group III atomic content, and 100% is nitrogen in terms of the Group V atomic content. Only AlGaN (without a ratio) can alternatively refer to aluminum gallium nitride of different compositions. Group III and Group V in this text refer to Group III and Group V of the periodic table, respectively.
[0008] According to a first aspect, there is provided a semiconductor structure, comprising:
[0009] a substrate;
[0010] The base layer, the base layer includes an AlGaN layer, and the base layer is disposed above the substrate;
[0011] At least one layer group, the at least one layer group is disposed above the base layer, wherein each of the at least one layer group includes:
[0012] The first layer, the first layer is an InGaN layer and includes quantum dots;
[0013] The second layer, the second layer is an InGaN layer, the second layer is disposed on and above the first layer, wherein the second layer laterally surrounds the quantum dots of the first layer, and wherein the material composition of the InGaN of the second layer is different from the material composition of the InGaN of the first layer.
[0014] It is recognized that during the epitaxial growth of the nitride semiconductor, the quantum dots can encapsulate dislocations. Therefore, the quantum dots (QDs) can prevent or reduce the diffusion of dislocations above the second layer. The quantum dots can prevent or reduce the diffusion of dislocations in a direction orthogonal to the substrate above the second layer. Hereinafter, the direction orthogonal to the substrate (i.e., the growth direction) will be referred to as the vertical direction. In addition, the direction parallel to the substrate will be referred to as the lateral direction.
[0015] The substrate can be a non-nitride substrate, such as a silicon substrate, a sapphire substrate, or a silicon carbide substrate. Growing a nitride semiconductor on such a non-nitride substrate may be advantageous because substrates of nitride semiconductor materials may be difficult to produce and / or expensive to produce, or in some cases impossible to produce. However, the use of a non-nitride substrate may result in dislocations, such as dislocations in the base layer. Such vertical diffusion of dislocations can be prevented, and thus such dislocations can be prevented from reaching the device layer above the second layer.
[0016] A silicon substrate can be particularly advantageous because silicon substrates are inexpensive and compatible with common semiconductor processing methods. In particular, the silicon substrate can have a top surface with Miller indices of {111}.
[0017] It should be understood that the present invention also provides advantages when the substrate is a nitride substrate (e.g., a GaN substrate). Dislocations may also be formed when using a nitride substrate, in which case it is advantageous to encapsulate the dislocations.
[0018] The base layer is disposed above the substrate, for example, disposed directly above the substrate. The base layer can be an AlGaN layer of any composition from AlN to GaN. A base layer including an AlN layer (e.g., an AlN layer directly above the substrate, e.g., an AlN layer directly above a silicon substrate) is advantageous. For example, an AlN layer sputtered directly above a silicon substrate is advantageous because the defect density at the interface between the sputtered AlN and silicon can be low.
[0019] At least one layer group is disposed above the base layer, for example, disposed directly above the base layer. Thus, the first layer is disposed above the base layer, for example, disposed directly above the base layer. At least one layer group may include more layers than the first layer and the second layer. For example, the first layer may be an intermediate layer between, for example, the zero layer and the second layer.
[0020] The quantum dots of the first layer may have a height (i.e., the dimension along the vertical direction) of 1 nm to 20 nm. The quantum dots of the first layer may have a width (i.e., the dimension along the lateral direction) of 10 nm to 20 nm. The quantum dots of the first layer may be shaped as pyramids, truncated pyramids or domes. The facets of the quantum dots may form inclined planes, for example, planes at an angle to the lateral direction.
[0021] The first layer may have a volume corresponding to 2 to 10 monolayers (e.g., 8 to 10 monolayers). For example, when 8 to 10 monolayers that are lattice mismatched with the underlying layer are grown, the monolayers may be rearranged during the self-assembly process to form quantum dots.
[0022] In some embodiments of the present invention, the quantum dots form islands, which may be truncated pyramids, pyramids including vertices, spheres, discs or elliptical discs. The quantum dots of the first layer may be quantum dots formed by self-assembly at a high density. The quantum dots of the first layer may be, for example, Stranski-Krastanov quantum dots. Thus, in addition to the quantum dots, the first layer may include a wetting layer. Alternatively, the quantum dots of the first layer may be Volmer-Weber quantum dots. Self-assembled quantum dots may provide a high quantum dot density and / or small quantum dots. This may be beneficial for providing effective encapsulation of dislocations.
[0023] The second layer is disposed above the first layer (e.g., disposed directly above the first layer), and the second layer laterally surrounds the quantum dots of the first layer. Thus, the quantum dots of the first layer may be partially or completely embedded in the second layer.
[0024] During the epitaxial growth of the second layer, dislocations may be encapsulated. For example, vertically propagating dislocations may be encapsulated. One mechanism for encapsulation may be the development of the facets of the quantum dots. The development of these facets may bend the dislocations towards the lateral direction. This may occur at least under some growth conditions, as understood from the application as a whole.
[0025] When the base layer includes pillars or nanocolumns, the above also applies. For example, during the epitaxial growth of the second layer, dislocations propagating vertically at the top and above the pillars or nanocolumns may be encapsulated, while the dislocation-free regions formed between the pillars continue to grow dislocation-free. As described, one mechanism for encapsulation may be the development of the facets of the quantum dots, bending the dislocations towards the lateral direction.
[0026] In a first example, when a second layer is grown on the side of a quantum dot, a dislocation propagating in the vertical direction (e.g., from a base layer and passing through the quantum dots of the first layer) can bend in the lateral direction. Thus, the dislocation can be encapsulated in the second layer and prevented from further propagating in the vertical direction.
[0027] In a second example, dislocations generated when the first layer overgrows the second layer can also be forced to propagate in the lateral direction. For example, it is forced into the lateral direction by the development of the side of the quantum dot.
[0028] The encapsulation of dislocations can be promoted additionally or alternatively by dislocation coalescence. For example, a dislocation originating from a first quantum dot can merge with a dislocation originating from a second quantum dot. Thus, the two dislocations can be terminated and prevented from further propagating.
[0029] Thus, there is high-quality semiconductor material above the quantum dots of the first layer, e.g., high crystalline quality. For example, a portion of the second layer above the quantum dots of the first layer has high quality, and / or the layer above the second layer has high quality.
[0030] Thus, the second layer or the layer above the second layer can be used in a device, such as a transistor or a light-emitting diode. Thus, the second layer or the layer above the second layer can include portions of a transistor, such as the source, body, or drain of a transistor. However, it should be understood that even if the semiconductor structure does not yet include portions of a transistor or a light-emitting diode, the semiconductor structure can still have advantages. For example, the semiconductor structure can be sold as a template (e.g., an epitaxy-ready template) on which further epitaxial growth can be performed. Such a template can facilitate the production of high-quality devices (e.g., high-quality transistors or light-emitting diodes).
[0031] As an example, the semiconductor structure can include or facilitate GaN complementary metal-oxide-semiconductor (CMOS) devices or More-than-Moore CMOS devices. Thus, the semiconductor structure is advantageous for using GaN in very large-scale integration (VLSI) device manufacturing.
[0032] As another example, the semiconductor structure can include or facilitate the growth of a heterojunction above the quantum dots of the first layer. A heterojunction can be an interface between two different semiconductor materials. The heterojunction can be configured to form a two-dimensional electron gas (2DEG). Such a heterojunction can be used to fabricate high electron mobility transistors (HEMTs). Such a heterojunction can be very sensitive to dislocations, and thus it can be particularly advantageous to have a first layer including quantum dots below the heterojunction.
[0033] In addition to encapsulating dislocations, quantum dots can have other advantages. For example, it is recognized that quantum dots can assist in charge carrier recombination and thereby improve devices above the quantum dots, such as transistors. For example, a HEMT can be improved by quantum dots at a distance of, for example, at least 1 micron from the 2DEG. Such quantum dots can be configured to have a bandgap of 1.6 eV to 1.8 eV. This bandgap can be in the middle of the bandgap of the surrounding material. The quantum dots can be configured as recombination centers. This can prevent light from being emitted in a "non-radiative" form due to high p-doping. In other words, the quantum dots can be resistive and / or non-conductive to electrons.
[0034] Therefore, an object of the present invention is that the quantum dots can also be provided as recombination centers by quantum dots acting as artificial atoms. Currently, the recombination centers of high-frequency GaN HEMT power amplifiers (PAs) are provided by iron atoms, which act as electron trap states and acceptors for electrons near the conduction band edge. Iron dopants in the GaN crystal structure can generate gallium vacancy states that are deep donor states 0.70 eV above the valence band. Since the quantum dots have a bandgap in the middle of the bandgap of GaN, the quantum dots exhibit a state that can act as both an acceptor and a donor.
[0035] Therefore, preferably, the quantum dots include InGaN, which has a bandgap lower than that of GaN. Due to the miscibility gap between InN and GaN, a preferred embodiment of the present invention is that the indium concentration is in the range of 5% to 20% or in the range of 5% to 10%.
[0036] Such a semiconductor structure can promote the high switching frequency of the HEMT. GaN HEMT devices have short turn-on times due to their low on-resistance. However, at high frequencies, the device can still significantly slow down the turn-off time, thereby preventing high-frequency applications. By promoting the recombination of electron and hole charge carriers at quantum dots that can be optimized by band engineering, the inventors have recognized that frequencies in the range of 24 GHz to 56 GHz suitable for the fifth-generation technology standard of broadband cellular networks (5G) can be provided. This will also be beneficial for providing energy-efficient GaN-on-silicon HEMT devices below 6 GHz as an alternative to silicon LDMOS technology.
[0037] At least one layer group can include a plurality of layer groups. Therefore, the semiconductor structure can include a plurality of layers containing quantum dots. For example, the semiconductor structure can include:
[0038] A first layer group above the base layer, the first layer group including a first layer and a second layer. The first layer of the first layer group is a layer of InGaN and includes quantum dots, and the second layer of the first layer group is a layer of InGaN disposed above the first layer of the first layer group, wherein the second layer of the first layer group laterally surrounds the quantum dots of the first layer of the first layer group; and
[0039] A second layer group above the first layer group, the second layer group including a first layer and a second layer, the first layer of the second layer group being a layer of InGaN and including quantum dots, the second layer of the second layer group being a layer of InGaN disposed above the first layer of the first layer group, wherein the second layer of the second layer group laterally surrounds the quantum dots of the first layer of the first layer group.
[0040] By using a stack of several self-assembled quantum dot layers, dislocations can be encapsulated more effectively compared to using a single layer stack. For example, a dislocation that has propagated through the first layer stack (e.g., by passing between the quantum dots) can be encapsulated by the second layer stack.
[0041] It should be understood that the layer stacks in the plurality of layer stacks may not necessarily be the same. They may differ, for example, in the size or composition of the quantum dots.
[0042] The semiconductor structure may be configured such that for at least one in at least one layer stack: the indium content of the InGaN in the first layer is higher than the indium content of the InGaN in the second layer.
[0043] Thus, the first layer may include a material having a larger lattice constant than the second layer. Thus, the first layer may include a material having a smaller bandgap than the second layer. This can provide effective encapsulation of dislocations.
[0044] The semiconductor structure may be configured such that for at least one in at least one layer stack:
[0045] The material composition of the InGaN in the first layer is In x Ga 1-x N, where x is a number in the range of 0 to 0.2 or 0.2 to 0.4. It should be recognized that such a composition provides particularly effective encapsulation of dislocations.
[0046] For illustration, the semiconductor structure may be (from bottom to top):
[0047] Substrate / AlGaN / InGaN / InGaN-QD / AlGaN, where the InGaN-QD has a maximum indium composition of 50% (i.e., x ≤ 0.5).
[0048] For further illustration, the semiconductor structure may be (from bottom to top):
[0049] Silicon substrate / AlGaN / InGaN / In 0.30 Ga 0.70 N-QD / AlGaN.
[0050] The semiconductor structure may be configured such that for at least one in at least one layer stack:
[0051] The density of the quantum dots in the first layer is between 10 9 cm -2 and 10 13 cm -2 .
[0052] The semiconductor structure can be configured such that for at least one in at least one layer group:
[0053] The density of the quantum dots in the first layer is at least 10 10 cm -2 .
[0054] It should be recognized that the above density of the quantum dots provides a particularly effective encapsulation of dislocations. A high density of quantum dots can be beneficial for providing an effective encapsulation of already formed dislocations. At the same time, too high a density of quantum dots may form new dislocations. In this sense, a density of the quantum dots between 10 9 cm -2 and 10 13 cm -2 can be beneficial. However, it should be understood that in principle the density of the quantum dots can be higher than 10 13 cm -2 .
[0055] It should be understood that the density of the quantum dots refers to the number of quantum dots per unit area in the plane including the quantum dots.
[0056] The semiconductor structure can be configured such that the quantum dot layer (i.e., the first layer) has a lattice mismatch with the layer below or above.
[0057] In a first example of lattice mismatch, consider the case where the quantum dot layer (i.e., the first layer) has a lattice mismatch with the layer below:
[0058] The semiconductor structure can be configured such that for at least one in at least one layer group:
[0059] At room temperature, the lattice constant of the material in the first layer is at least 0.5% larger than the lattice constant of the material in the layer directly below the first layer.
[0060] Thus, the first layer can have a lattice mismatch with the layer directly below the first layer. The lattice mismatch can be at least 0.5%. Such a lattice mismatch can be a driving force for the formation of self-assembled quantum dots. Thus, the quantum dots in the first layer can be formed due to the lattice mismatch.
[0061] In a second example of lattice mismatch, consider the case where the quantum dot layer (i.e., the first layer) has a lattice mismatch with the layer above:
[0062] The semiconductor structure can be configured such that for at least one in at least one layer group:
[0063] At room temperature, the lattice constant of the material in the first layer is at least 0.5% larger than the lattice constant of the material in the second layer.
[0064] Accordingly, the first layer can be lattice mismatched with the second layer. The lattice mismatch can be at least 0.5%.
[0065] It should be understood that when two crystalline materials are combined, one or both materials can be strained and at least partially accommodate the lattice constant of the other material. Accordingly, it should be understood that the above lattice constants are nominal lattice constants, i.e., the lattice constants of the materials when unstrained. In other words, the lattice constants are non-strained lattice constants.
[0066] The semiconductor structure can be configured such that the base layer includes:
[0067] A primary layer, which is a layer of semiconductor material and includes columns extending in a direction perpendicular to the substrate;
[0068] A secondary layer, which is a layer of semiconductor material, wherein the secondary layer is configured to laterally surround the columns of the primary layer.
[0069] The primary layer can include AlGaN (e.g., AlN). The primary layer can include sputtered AlN. The secondary layer can include Al x Ga 1-x N, where 0 ≤ x ≤ 0.95. The secondary layer can include Al x Ga 1-x N, where 0 ≤ x ≤ 0.40. The secondary layer can include Al x Ga 1-x N, where 0 ≤ x ≤ 0.20. In a preferred embodiment of the secondary layer, the secondary layer is GaN. The diameter of the columns can be in the range of 1 nm to 100 nm, preferably in the range of 5 nm to 50 nm, and more preferably in the range of 10 nm to 30 nm. The columns can be nanowires.
[0070] The secondary layer can grow as a shell on the columns of the primary layer, thereby surrounding the columns. A semiconductor layer structure including columns surrounded by another layer can facilitate the encapsulation and accumulation of dislocations in the shell along the columns, and thus reduce the amount of material defects and cracks propagating to the surface of the semiconductor layer structure.
[0071] Dislocations can be particularly prone to propagating laterally between the columns (e.g., by epitaxial growth along the length of the nanowires), such that the shells from different nanowires coalesce to form a common thin film. When the dislocations are laterally bent laterally, the further propagation of the dislocations can be prevented. The quantum dots or the third layer grown on the second layer can be silicon-doped.
[0072] As an example of the benefits of a base layer including the primary and secondary layers described above, consider the following:
[0073] The primary layer can be deposited, for example, on a silicon substrate (e.g., on a silicon substrate having a top surface with Miller indices {111}) by sputtering AlN (or AlGaN). The sputtered AlN is not necessarily single crystal and can form grain boundaries at least under some growth conditions. Then, vertical pillars can be etched from the AlN layer, and thus the vertical pillars can include grain boundaries. When the pillars overgrow laterally, by depositing the secondary layer, Al(x)Ga(1-x)N growth can be seeded through these grain boundaries. At least under some growth conditions, dislocations can particularly tend to propagate laterally between the pillars along the M direction of the wurtzite crystal structure, and thus propagate laterally (since the substrate is a Si{111} substrate, the M direction is the lateral direction). Thus, dislocations can coalesce laterally with dislocations from other nanowire structures to form a dislocation-free film. Thereby, defects can be filtered out at least to some extent, reducing the amount of defects and cracks that propagate to the surface of the semiconductor layer structure. When the pillar shells combine laterally into a common film and then the film begins to grow along the C direction, dislocations can be trapped and prevented from expanding along the C direction (i.e., prevented from expanding vertically). Thus, by first forming dislocations, the strain caused by growth on the silicon substrate can be accommodated, but at least some dislocations can be encapsulated and prevented from further expanding into the vertical direction. Then, during the growth of a layer stack including quantum dots, dislocations that escape the base layer above the nanowires can be encapsulated. Additionally or alternatively, other types of dislocations (e.g., dislocations associated with the combination of the pillar shells between and above the nanowires) can be encapsulated during the growth of the layer stack including quantum dots.
[0074] It is to be recognized that the combination of pillars and quantum dots can be particularly useful for encapsulating dislocations that are not filtered to the direct height of the nanowires. Many dislocations may already be encapsulated in the base layer. Dislocations that escape the base layer and propagate into at least one layer stack can be encapsulated there by the quantum dots. The pillars and quantum dots can be complementary to each other and encapsulate different types of dislocations.
[0075] According to a second aspect, there is provided a method for producing a semiconductor structure of the first aspect, the method comprising:
[0076] Providing a substrate;
[0077] Forming a base layer above the substrate;
[0078] Forming at least one layer stack above the base layer by forming a first layer and forming a second layer, wherein forming at least one layer stack includes: for at least one of the at least one layer stack, forming quantum dots of the first layer by self-assembly.
[0079] The layers forming the semiconductor structure can be accomplished by any type of epitaxial growth (e.g., by metalorganic vapor phase epitaxy (MOVPE) or molecular beam epitaxy (MBE)).
[0080] As previously mentioned, the quantum dots of the first layer can be self-assembled quantum dots. The quantum dots of the first layer can be, for example, Stranski-Krastanov quantum dots. Thus, in addition to the quantum dots, the first layer can include a wetting layer. Alternatively, the quantum dots of the first layer can be Volmer-Weber quantum dots. Self-assembled quantum dots can provide a high quantum dot density and / or small quantum dots. This can be beneficial for providing effective encapsulation of dislocations. In addition, combining the self-assembly process with a patterning process can be advantageous. For example, if the quantum dots formed by self-assembly are combined with a base layer including pillars formed by patterning (e.g., lithographic patterning), the self-assembly process can, for example, compensate for the non-uniformities in the patterning process.
[0081] According to the above, the self-assembly formation of quantum dots can be driven by the lattice mismatch between the first layer and the second layer.
[0082] Forming at least one layer group can include: for at least one of the at least one layer group, forming the quantum dots of the first layer by Stranski-Krastanov growth.
[0083] Stranski-Krastanov growth can be a growth mode that provides a suitable quantum dot size and / or quantum dot density.
[0084] Stranski-Krastanov growth can be a growth mode suitable for forming quantum dots, which can facilitate the encapsulation of dislocations propagating from below without forming a large number of new dislocations. Thus, the net effect of such quantum dots can be to reduce the dislocations further propagating above the first layer including the quantum dots.
[0085] Forming at least one layer group can include: for at least one of the at least one layer group,
[0086] interrupting the gas flow for at least 5 seconds between forming the first layer and forming the second layer, wherein the gas flow is a gas flow including nitrogen.
[0087] For example, between forming the first layer and forming the second layer, the gas flow can be interrupted for 10 seconds to 20 seconds. The gas flow including nitrogen can be, for example, a gas flow including nitrogen gas and / or a gas flow including ammonia.
[0088] Such interruptions can provide a suitable quantum dot size and / or quantum dot density. For example, the first layer can be formed by growing 2 to 10 monolayers (e.g., 8 to 10 monolayers) planar. An interruption including a gas flow of nitrogen can be regarded as stopping the group V supply. As an alternative, the group III supply can be interrupted, for example, interrupting the flow of group III precursor gas, such as interrupting trimethylgallium and / or triethylgallium in a nitrogen or ammonia atmosphere. Group III and group V herein refer to group III and group V of the periodic table of elements.
[0089] The interruption can stop further growth. During the interruption, the planar monolayers can be rearranged to form quantum dots, such as rearranged to form a wetting layer and quantum dots.
[0090] Forming at least one layer group can include: for at least one of the at least one layer group,
[0091] ending the layer formation of the first layer based on the photoluminescence signal of the quantum dots from the first layer.
[0092] The photoluminescence spectrum can depend on the quantum dot size and / or quantum dot density. Therefore, photoluminescence can be used to fine-tune the quantum dot size and / or quantum dot density.
[0093] As a first example: when the photoluminescence spectrum indicates that the quantum dots have reached a threshold size, the formation of the first layer can be ended. The threshold size can be a threshold indicating quantum dots with a height of 1 nm to 20 nm and / or quantum dots with a width of 10 nm to 20 nm.
[0094] As a second example: when the photoluminescence spectrum indicates that the quantum dots have reached a threshold density, the formation of the first layer can be ended. The threshold density can indicate that the density of the quantum dots is above 10 10 cm -2 or the density of the quantum dots is between 10 9 cm -2 and 10 13 cm -2 .
[0095] Forming the base layer can include:
[0096] forming a primary layer, the primary layer being a layer of semiconductor material, the primary layer including columns extending in a direction perpendicular to the substrate;
[0097] forming a secondary layer, the secondary layer being a layer of semiconductor material, wherein the secondary layer is configured to laterally surround the columns of the primary layer;
[0098] wherein the primary layer is formed in an atmosphere including hydrogen and nitrogen; and / or wherein the secondary layer is formed in an atmosphere including nitrogen.
[0099] The base layer can be an intermediate layer of AlN or AlGaN. A preferred embodiment of the present invention uses the strain of InGaN growth on the AlN or AlGaN intermediate layer as the driving force for quantum dot formation. Reducing the dislocations at these interfaces can alleviate the cracking of the GaN HEMT epitaxial wafer. The intermediate layer can have a thickness of 10 nm to 30 nm and is grown until a growth interruption of 5 seconds to 20 seconds, on which quantum dots are formed. Herein, the base layer as the intermediate layer can be interpreted as the base layer as the layer between the quantum dot layer (the first layer) and another layer.
[0100] The primary layer can be formed in an atmosphere comprising hydrogen and nitrogen, wherein the molar ratio of hydrogen to nitrogen is between 2 to 1 (i.e., the amount of hydrogen is twice the amount of nitrogen) and 1 to 2 (i.e., the amount of nitrogen is twice the amount of hydrogen). In particular, the primary layer can be formed in an atmosphere comprising hydrogen and nitrogen, wherein the molar ratio of hydrogen to nitrogen is 1 to 1, i.e., equal amounts of hydrogen and nitrogen. BRIEF DESCRIPTION OF THE DRAWINGS
[0101] The above and additional objects, features, and advantages of the inventive concept will be better understood from the following illustrative and non - limiting detailed description with reference to the accompanying drawings. In the drawings, like reference numerals will be used for like elements unless otherwise noted.
[0102] Figure 1 is a cross - sectional view of a semiconductor structure.
[0103] Figure 2 is a cross - sectional view of a semiconductor structure.
[0104] Figure 3 is a cross - sectional view of a semiconductor structure.
[0105] Figure 4 is a cross - sectional view of a semiconductor structure.
[0106] Figure 5 is a cross - sectional view of a semiconductor structure.
[0107] Figure 6 is a cross - sectional view of a semiconductor structure.
[0108] Figure 7 is a top - view of a semiconductor structure.
[0109] Figure 8 is a flowchart of a method
[0110] Figure 9 is a flowchart of a method
[0111] Figure 10 is an energy band diagram
[0112] Figure 11 is an energy band diagram DETAILED DESCRIPTION
[0113] In conjunction with the accompanying drawings, the technical content and detailed description of the present invention are described according to preferred embodiments that are not intended to limit the scope of the claims. The present invention can be implemented in many different forms and should not be construed as limited to the embodiments set forth herein; rather, these embodiments are provided to be exhaustive and complete and to fully convey the scope of the present invention to those skilled in the art.
[0114] Figures 1 to 6 A semiconductor structure 1 according to the present invention is shown. The semiconductor structure 1 is shown in a cross-sectional view. Each semiconductor structure 1 includes:
[0115] A substrate 2;
[0116] A base layer 10, the base layer 10 includes an AlGaN layer, and the base layer 10 is disposed above the substrate 2;
[0117] At least one layer group 20, the at least one layer group 20 is disposed above the base layer 10, wherein each of the at least one layer group 20 includes:
[0118] A first layer 21, the first layer 21 is an InGaN layer and includes quantum dots 24;
[0119] A second layer 22, the second layer 22 is an InGaN layer, and the second layer 22 is disposed above the first layer 21
[0120] above, wherein the second layer 22 laterally surrounds the quantum dots 24 of the first layer 21, and wherein the material composition of the InGaN of the second layer 22 is different from the material composition of the InGaN of the first layer 21.
[0121] The quantum dots 24 of the first layer 21 may be self-assembled quantum dots. The quantum dots 24 of the first layer 21 may be, for example, Stranski-Krastanov quantum dots. In addition to the quantum dots 24, the first layer 21 may further include a wetting layer 26, as Figure 1 shown. Alternatively, the quantum dots 24 of the first layer 21 may be Volmer-Weber quantum dots 24. The Volmer-Weber quantum dots 24 may be formed directly on the base layer 10 without any wetting layer 26, as Figure 2 shown. Thus, there may be gaps between the quantum dots 24, wherein the layer below the first layer 21 (the base layer 10 in Figure 2 is in contact with the second layer 22.
[0122] Figure 3Shows how the quantum dots 24 can bend the dislocations 4 towards the lateral direction. The figure shows two dislocations 4 of the quantum dots 24 that propagate in the vertical direction from the interface between the substrate 2 and the base layer 10, pass through the base layer 10 and through the first layer 21. Then, each shown dislocation 4 bends in the lateral direction in the second layer 22.
[0123] Figure 4 Shows a semiconductor structure 1 including a plurality of layer groups 20 (in this case three layer groups 20). Each layer group 20 herein includes a first layer 21 and a second layer 22. The quantum dots of different layer groups do not necessarily need to be aligned. Thus, the quantum dots of the first layer group can be arranged at lateral positions where there are no quantum dots in the second layer group.
[0124] By using several layer groups, the dislocations can be effectively encapsulated. Dislocations that have propagated through the first layer group (e.g., by passing between the quantum dots) can be encapsulated by the second layer group.
[0125] Figure 5 Shows a semiconductor structure 1 including a heterojunction 30 above the quantum dots 24 of the first layer 21. The heterojunction 30 is an interface between two different semiconductor materials. In Figure 5 , the heterojunction 30 is an interface between the second layer 22 and the barrier layer 32. In this article, the barrier layer 32 can have a larger bandgap than the underlying layer. In this case, the barrier layer 32 can have a larger bandgap than the second layer 22. The barrier layer 32 can be an AlGaN layer, and the underlying layer (in this case the second layer 22) can be a GaN layer.
[0126] The heterojunction 30 can be configured to form a 2DEG. Such a heterojunction 30 can be used to fabricate a HEMT. Figure 5 Can be regarded as an example of such a HEMT, where the source electrode 40, the gate electrode 42, and the drain electrode 44 are arranged on the surface of the semiconductor structure 1 above the heterojunction, in this case on top of the barrier material.
[0127] Figure 6 Is a cross-sectional view taken from the side of the semiconductor structure 1, where the base layer 10 includes:
[0128] A primary layer 11, the primary layer 11 is a layer of semiconductor material, and the primary layer 11 includes columns 14 extending in a direction perpendicular to the substrate 2;
[0129] A secondary layer 12, the secondary layer 12 is a layer of semiconductor material, wherein the secondary layer 12 is configured to laterally surround the columns 14 of the primary layer 11. Figure 7 Is a cross-sectional view taken from the top of the same semiconductor structure 1, and this cross-section is along the Figure 6 Line A in. The columns 14 can be arranged in an array, asFigure 7 as shown in the cross-sectional view taken from the top. Alternatively, the pillars 14 may be randomly distributed. It should be understood that during the maskless overgrowth of GaN on the pillars 14, dislocations may propagate vertically, and the dislocations are filtered directly above the height of the pillars to form a dislocation-free layer. In some embodiments, the quantum dots may be provided as a dislocation filter for vertically propagating dislocations on top of the nanocolumns. In other embodiments, one or more quantum dot layers adjacent to the AlN or Al(x)Ga(1-x)N intermediate layer may be provided, and the intermediate layer provides a driving force for the formation of the quantum dots.
[0130] Figure 8 FIG. 5 shows a flowchart of a method 100 for producing a semiconductor structure 1. According to the method 100, first, a substrate 2 is provided at S102. Then, a base layer 10 is formed at S110 above the substrate 2. Subsequently, at least one layer stack 20 is formed at S120 above the base layer 10 by forming a first layer 21 and forming a second layer 22, wherein forming at least one layer stack 20 includes: for at least one of the at least one layer stack 20, quantum dots 24 of the first layer 21 are formed at S124 by self-assembly. The above can be achieved by placing the substrate 2 (e.g., a silicon substrate) in a reactor unit of a metalorganic vapor phase epitaxy (MOVPE) system and supplying precursor gases at appropriate pressures and temperatures to grow the layers of the semiconductor structure 1. Alternatively, the base layer 10 or a part of the base layer 10 may be formed before placing the substrate 2 in the MOVPE system. For example, the base layer 10 or a part of the base layer 10 may be formed by sputtering AlN or AlGaN on the substrate 2.
[0131] Forming the quantum dots 24 of the first layer 21 may be accomplished by Stranski-Krastanov growth. Thus, the material forming the quantum dots 24 may grow lattice-mismatched with the underlying layer, e.g., lattice-mismatched with the base layer, or lattice-mismatched with the second layer 22 of the previous layer stack.
[0132] In addition to forming the quantum dots 24 at S124, the step of forming at least one layer stack at S120 may include one or more optional steps, as Figure 9 seen in.
[0133] As an example, between forming the first layer 21 and forming the second layer 22, the gas flow may be interrupted at S125 for at least 5 seconds, wherein the gas flow is a gas flow including nitrogen. Thus, the gas flow of nitrogen or ammonia may be interrupted for at least 5 seconds. The interrupted gas flow may be the entire gas flow of the group-V supply in the MOVPE reactor unit. Alternatively, the interrupted gas flow may be a part of the entire gas flow of the group-V supply in the MOVPE reactor unit.
[0134] As another example, the layer formation of the first layer 21 in S126 can be terminated based on the photoluminescence signal from the quantum dots 24 in the first layer 21. In this document, the photoluminescence signal can be obtained by irradiating the semiconductor structure 1 with light (e.g., laser) and collecting the photoemission from the quantum dots 24. In this document, the photoemission can be collected, for example, using a microscope objective (e.g., a microscope objective with a long working distance). The collected light can be sent to a spectrometer, whereby a spectrum can be obtained. Each quantum dot can generate at least one spectral peak, which can be characteristic of the quantum dot size. High-energy (short spectral wavelength) spectral peaks can indicate small quantum dots. Low-energy (long spectral wavelength) spectral peaks can indicate large quantum dots. Many quantum dots can contribute to the photoluminescence signal. However, the same principle as for a single quantum dot can apply. If the photoluminescence signal is centered at a short spectral wavelength, most of the quantum dots can be considered high-energy quantum dots, i.e., small quantum dots. As the quantum dots become larger, the spectrum of the photoluminescence signal may shift towards longer wavelengths. When the spectral peak of the photoluminescence signal is centered at a threshold, this can be regarded as an indication that the quantum dots 24 have reached the threshold size, and the layer formation of the first layer 21 in S126 can be terminated. When the spectral peak of the photoluminescence signal has a specific width and / or position, this can be regarded as an indication that the quantum dots 24 have reached the threshold density, and the layer formation of the first layer 21 in S126 can be terminated.
[0135] As described above, the quantum dots 24 can be combined with the underlying layer 10 including columns. Thus, forming the underlying layer 10 in S110 can include forming a primary layer 11, the primary layer 11 including columns 14 extending in a direction perpendicular to the substrate 2. For example, a semiconductor material can be deposited on the substrate 2, and then the semiconductor material is lithographically patterned and etched to form the columns 14. Then, a secondary layer 12 can be formed, the secondary layer 12 being configured to laterally surround the columns 14 of the primary layer 11. The material for the primary layer 11 can be grown, for example, by MOVPE. Alternatively, the material for the primary layer 11 can be a sputtered semiconductor material. The material for the secondary layer 12 can be, for example, an epitaxially grown semiconductor material (e.g., grown by MOVPE).
[0136] The primary layer 11 can be formed in an atmosphere including hydrogen and nitrogen. The secondary layer 12 can be formed in an atmosphere including nitrogen. The primary layer 11 can be formed in an atmosphere including hydrogen and nitrogen, wherein the molar ratio of hydrogen to nitrogen is between 2 to 1 and 1 to 2. In particular, the primary layer can be formed in an atmosphere including hydrogen and nitrogen, wherein the molar ratio of hydrogen to nitrogen is 1 to 1. The above can apply to the case where the material for the primary layer 11 is grown by MOVPE and the case where the material for the primary layer 11 is sputtered.
[0137] As described above, the quantum dots 24 can form recombination centers. Thus, the quantum dots 24 can replace the iron dopants as recombination centers. Figures 10 to 11 Shows the difference between the iron dopants and the quantum dots 24.
[0138] Figure 10 Schematically shows the energy band diagram of an iron-doped semiconductor (e.g., GaN). The figure shows the valence band 70 and the conduction band 72. The figure also shows the energy level 74 of the iron dopant. The iron dopant in the GaN crystal structure can generate a gallium vacancy state as a deep donor state 0.70 eV above the valence band 70.
[0139] Figure 11 Schematically shows the energy band diagram of a semiconductor (e.g., GaN) including the energy structure 76 of the quantum dots 24. In this figure, the energy structures 76 of three quantum dots 24 can be seen. As shown, the quantum dots 24 can form potential wells in both the valence band 70 and the conduction band 72. As shown, holes can exist in the potential well of the valence band 70. Similarly, as shown in the figure, electrons can exist in the potential well of the conduction band 72. Thus, since the quantum dots 24 have a band gap in the middle of the band gap of GaN, the quantum dots 24 exhibit a state that can act as both an acceptor and a donor.
[0140] Above, the inventive concept has been mainly described with reference to a limited number of examples. However, as will be readily understood by those skilled in the art, other examples in addition to the examples disclosed above are equally possible within the scope of the inventive concept defined by the appended claims.
Claims
1. A semiconductor structure (1), comprising: a substrate (2); a base layer (10), the base layer (10) including an AlGaN layer, the base layer (10) being disposed above the substrate (2); at least one layer group (20), the at least one layer group (20) being disposed above the base layer (10), wherein each of the at least one layer group (20) includes: a first layer (21), the first layer (21) being a layer of InGaN and including quantum dots (24); a second layer (22), the second layer (22) being a layer of InGaN, the second layer (22) being disposed on and above the first layer (21), wherein the second layer (22) laterally surrounds the quantum dots (24) of the first layer (21), and wherein the material composition of InGaN of the second layer (22) is different from the material composition of InGaN of the first layer (21); a heterojunction (30), the heterojunction (30) being disposed above the quantum dots (24) of the first layer (21), the heterojunction (30) being configured to form a two-dimensional electron gas.
2. The semiconductor structure (1) according to claim 1, wherein, The at least one layer group (20) includes a plurality of layer groups (20).
3. The semiconductor structure (1) according to claim 1 or 2, wherein, For at least one of the at least one layer group (20), the indium content in the InGaN of the first layer (21) is higher than the indium content in the InGaN of the second layer (22).
4. The semiconductor structure (1) according to any one of the preceding claims, wherein, For at least one of the at least one layer group (20), the material composition of InGaN of the first layer (21) is In x Ga 1-x N, where x is a number in the range of 0.1 to 0.2 or 0.2 to 0.
4.
5. The semiconductor structure (1) according to any one of the preceding claims, wherein, For at least one of the at least one layer group (20), the density of the quantum dots (24) in the first layer (21) is between 10 9 cm -2 and 10 13 cm -2 .
6. The semiconductor structure (1) according to any one of the preceding claims, wherein, For at least one of the at least one layer group (20), the density of the quantum dots (24) of the first layer (21) is at least 10 10 cm -2 .
7. The semiconductor structure (1) according to any one of the preceding claims, wherein, For at least one of the at least one layer group (20), at room temperature, the lattice constant of the material in the first layer (21) is at least 0.5% larger than the lattice constant of the material in the layer directly below the first layer (21).
8. The semiconductor structure (1) according to any one of the preceding claims, wherein, For at least one of the at least one layer group (20), at room temperature, the lattice constant of the material in the first layer (21) is at least 0.5% larger than the lattice constant of the material in the second layer (22).
9. The semiconductor structure (1) according to any one of the preceding claims, wherein, The base layer (10) includes: a primary layer (11), the primary layer (11) being a layer of semiconductor material, the primary layer (11) including columns (14) extending in a direction perpendicular to the substrate (2); a secondary layer (12), the secondary layer (12) being a layer of semiconductor material, wherein the secondary layer (12) is configured to laterally surround the columns (14) of the primary layer (11); wherein the primary layer (11) or the secondary layer (12) forms the AlGaN layer of the base layer (10).
10. A method (100) for producing a semiconductor structure (1) according to any one of the preceding claims, the method (100) comprising: providing (S102) the substrate (2); forming (S110) the base layer (10) above the substrate (2); The at least one layer group (20) is formed (S120) over the base layer (10) by forming the first layer (21) and forming the second layer (22), wherein forming the at least one layer group (20) includes: for at least one of the at least one layer group (20), forming (S124) quantum dots (24) of the first layer (21) by self-assembly; A heterojunction (30) is formed over the quantum dots (24) of the first layer (21), and the heterojunction (30) is configured to form a two-dimensional electron gas.
11. The method (100) according to claim 10, wherein, Forming (S120) the at least one layer group (20) includes: for at least one of the at least one layer group (20), forming (S124) quantum dots (24) of the first layer (21) by Stranski-Krastanov growth.
12. The method (100) according to claim 10 or 11, wherein Forming (S120) the at least one layer group (20) includes: for at least one of the at least one layer group (20), interrupting (S125) a gas flow for at least 5 seconds between forming the first layer (21) and forming the second layer (22), wherein the gas flow is a gas flow including nitrogen.
13. The method (100) according to any one of claims 10 to 12, wherein, Forming (S120) the at least one layer group (20) includes: for at least one of the at least one layer group (20), ending (S126) layer formation of the first layer (21) based on a photoluminescence signal from the quantum dots (24) of the first layer (21).
14. The method (100) according to any one of claims 10 to 13, wherein, Forming (S110) the base layer (10) includes: Forming a primary layer (11), the primary layer (11) being a layer of semiconductor material, the primary layer (11) including columns (14) extending in a direction perpendicular to the substrate (2); Forming a secondary layer (12), the secondary layer (12) being a layer of semiconductor material, wherein the secondary layer (12) is configured to laterally surround the columns (14) of the primary layer (11); wherein the primary layer (11) is formed in an atmosphere including hydrogen and nitrogen; and / or wherein the secondary layer (12) is formed in an atmosphere including nitrogen.
15. The method (100) according to any one of claims 10 to 14, wherein, At least a portion of the base layer is formed by reactive sputtering of AlN or AlGaN.
16. The method (100) according to claim 15, wherein, The intermediate layer (10) has a thickness of 10 nm to 30 nm.