Method of manufacturing a substrate comprising a relaxed layer of indium gallium nitride

Abstract

A method of manufacturing a substrate, comprising the steps of: providing a stack comprising an initial substrate, a gallium nitride layer, a doped indium gallium nitride layer (13), and an unintentionally doped indium gallium nitride layer (14); transferring the doped indium gallium nitride layer (13) and the unintentionally doped indium gallium nitride layer (14) to an anodization support (21) to form a second stack (20); immersing the second stack (20) and a counter electrode in an electrolyte solution and applying a voltage or a current between the doped indium gallium nitride layer (13) and the counter electrode to porous the doped indium gallium nitride layer (13) and to relax the unintentionally doped indium gallium nitride layer (14); transferring the doped indium gallium nitride layer (13) and the unintentionally doped indium gallium nitride layer (14) to a support of interest; and forming an indium gallium nitride layer by epitaxy on the unintentionally doped indium gallium nitride layer to obtain a relaxed epitaxially grown indium gallium nitride layer.

Description

Technical Field

[0001] This invention relates to the general field of color microdisplays.

[0002] The present invention relates to a method for manufacturing a substrate or pseudo-substrate comprising a relaxed indium gallium nitride (InGaN) layer.

[0003] The present invention also relates to a substrate or pseudo-substrate comprising a relaxed indium gallium nitride layer.

[0004] This invention has applications in many industrial fields, particularly in the field of color microdisplays based on microLEDs with a pitch of less than 10 μm. Background Technology

[0005] Color microdisplays consist of red, green, and blue pixels (RGB pixels).

[0006] Blue and green pixels can be made of nitride materials, while red pixels can be made of phosphor materials. These three types of pixels are typically combined on the same substrate using a technique called "pick and place." However, this technique is no longer feasible for microdisplays with pixels smaller than 10 μm due to alignment issues and the time required to implement such a scale of technology.

[0007] Another solution is to use quantum dots (QDs) or nanophosphors for color conversion. However, controlling the deposition of these materials on small pixels is difficult, and their flow resistance properties are not robust enough.

[0008] Therefore, the ability to naturally obtain three RGB pixels on the same substrate using the same material family is crucial. In this case, indium gallium nitride (IGN) is the most promising material. In fact, depending on the indium concentration, it can theoretically cover the entire visible spectrum. IGN-based blue micro-LEDs have already shown significantly higher brightness than their organic counterparts. To emit light at green wavelengths, the quantum well (QW) of the LED should contain at least 25% indium, while for emitting red light, at least 35% indium is required. Unfortunately, due to the low miscibility of indium (In) in gallium nitride (GaN) and the inherent high compressive stress of growing IGN active regions on gallium nitride, the quality of IGN materials with indium content exceeding 20% ​​degrades.

[0009] Therefore, it is crucial to reduce the overall stress of gallium nitride / indium gallium nitride-based structures.

[0010] Several solutions have been considered to address this problem.

[0011] One solution is to form nanostructures such as nanowires or pyramids that allow stress relaxation through free edges. Axial nanowires can be grown using molecular beam epitaxy (MBE). However, the low growth temperatures used in MBE growth result in low internal quantum efficiency (IQE). Pyramids are used to bend dislocations. Specifically, a complete pyramid has a semi-polar plane, which facilitates indium incorporation and reduces the internal electric field of the active region. For a truncated pyramid, the truncated facets allow quantum wells to grow in the c-plane, resulting in more uniform emission compared to emission along the semi-polar plane of a complete pyramid. Alternatively, growth can also occur planarly on planes other than the c-plane in wurtzite structures, such as on a semi-polar plane that is more favorable for indium incorporation.

[0012] Another solution is to reduce the stress in the active region of the LED structure by using a substrate or pseudo-substrate with lattice parameters closer to those of the indium gallium nitride alloy for quantum wells. Therefore, even with a planar structure, the indium doping rate in indium gallium nitride can be increased. It has been found that as the substrate lattice parameter increases, the internal electric field decreases compared to a stress layer with the same indium concentration, and quantum well emission shifts towards red. The resulting relaxed indium gallium nitride layer enables the growth of type III nitride heterostructures via metal-organic vapor phase epitaxy (MOVPE). However, to our knowledge, the only substrate to date that has achieved this demonstration is Soitec's pseudo-substrate InGaNOS. This pseudo-substrate was developed by implementing a Smart Cut... TM It is manufactured using technology. Relaxation of the indium gallium nitride layer is achieved through various heat treatments [1]. However, by this method, cracks may appear in the indium gallium nitride layer and / or the surface of the indium gallium nitride layer may lose its flatness.

[0013] Another solution to reduce the overall stress in gallium nitride / indium gallium nitride-based LED structures is to porosilicate the gallium nitride layer. In reference [2], a stack was first prepared consisting of a sapphire substrate covered with an unintentionally doped gallium nitride layer (uid GaN) and an n+ doped gallium nitride layer. The doped gallium nitride layer was used as the anode. The cathode was a platinum wire. Electrochemical porosilicate was performed by applying a voltage of 15 V for 30 minutes in an oxalic acid solution (0.2 M (molar concentration)) and then placing it in a KOH solution (0.06 M) under ultraviolet radiation and applying a voltage of 9 V for 30 minutes. The resulting porous gallium nitride layer enabled the growth of a multi-quantum well (MQW) LED structure consisting of an n+ gallium nitride layer, five gallium nitride / indium gallium nitride quantum wells (QWs), and a p gallium nitride contact layer. Significant stress relaxation resulted in better electrical and optical performance, especially in photoluminescence (PL).

[0014] However, the crystal quality of the LED layer depends on the pore size and porosity of the porous gallium nitride layer, as well as the required thickness. Therefore, suitable parameters need to be found each time, which complicates the industrialization of this process. Summary of the Invention

[0015] The object of the present invention is to provide a method for obtaining an at least partially or fully relaxed epitaxially grown indium gallium nitride layer from a gallium nitride / indium gallium nitride substrate to manufacture, for example, red, green and blue pixels.

[0016] Therefore, the present invention provides a method for fabricating a relaxed epitaxially grown indium gallium nitride layer from a gallium nitride / indium gallium nitride substrate, the method comprising the following steps:

[0017] a) Providing a first stack, the first stack sequentially comprising an initial substrate, a gallium nitride layer, a doped indium gallium nitride layer, and an unintentionally doped indium gallium nitride layer,

[0018] b) Transferring a doped indium gallium nitride (IGaN) layer and an unintentionally doped IGaN layer onto an anodized support, wherein the unintentionally doped IGaN layer contacts the anodized support to form a second stack.

[0019] c) Connect the doped indium gallium nitride layer and the counter electrode to a voltage generator or a current generator.

[0020] d) Immerse the second stack and the counter electrode in the electrolyte solution.

[0021] e) Applying a voltage or current between the doped indium gallium nitride layer and the counter electrode to porosilicate the doped indium gallium nitride layer.

[0022] f) Transferring the porous doped indium gallium nitride layer and the unintentionally doped indium gallium nitride layer to the support of interest.

[0023] g) A relaxed epitaxially grown indium gallium nitride layer is obtained by forming an indium gallium nitride layer on an unintentionally doped indium gallium nitride layer.

[0024] The fundamental difference between this invention and the prior art lies in the step of making the doped indium gallium nitride layer porous through an electrochemical means.

[0025] Porosity enables the surface structuring of the indium gallium nitride layer and improves extraction efficiency, and allows for the incorporation of more indium through stress relaxation.

[0026] It is easy to adjust the aperture size to achieve the required relaxation percentage for the desired wavelength by adjusting the doping of the indium gallium nitride layer, the applied voltage, and / or the selected electrolyte (nature and / or concentration).

[0027] The first transition enables a porosification step to be performed on the surface of the nitrogen (N) polar indium gallium nitride layer, thus making it easier to obtain porosity. The second transition enables gallium (Ga) polarity on the front side of the substrate, which allows for subsequent epitaxial regeneration.

[0028] The undoped or weakly doped indium gallium nitride layer is electrically insulating. It is not porousized in step e). Therefore, it can be used as an epitaxial regeneration layer.

[0029] Finally, this method yields an InGaNOX (“Indium Gallium Nitride on Substrate X”) type structure with an undoped indium gallium nitride layer, which does not require heat treatment and has a full plate (i.e., no mesa needs to be formed).

[0030] In epitaxial regeneration methods, the growth temperature used (typically 900°C to 1000°C) allows for improved porosity of the layer, particularly by increasing the porosity of the layer to provide greater freedom while maintaining suitable lattice parameters for the epitaxial regeneration layer. Therefore, this epitaxial regeneration layer allows for further relaxation of the unintentionally doped indium gallium nitride layer. This results in at least a partially relaxed, and preferably fully relaxed, indium gallium nitride layer.

[0031] Preferably, the anodized support and / or the support of interest includes, for example, a support layer of sapphire or silicon (Si) and a buried oxide layer.

[0032] According to a particular embodiment, preferably according to SmartCut TM A method of type b) is used to perform step b), which includes the following steps:

[0033] - Implant atomic species to form embrittlement regions at a depth in the doped indium gallium nitride layer that is close to the final thickness of the layer to be retained in the final substrate;

[0034] - The first stack is bonded to the anodized support by contacting the undoped indium gallium nitride layer with the anodized support; and

[0035] - Heat is transferred to the embrittled region to separate the doped indium gallium nitride layer from the gallium nitride layer.

[0036] Preferably, the first stack further includes a highly doped gallium nitride layer disposed between an unintentionally doped gallium nitride layer and a doped indium gallium nitride layer.

[0037] According to this embodiment, step b can be performed according to two preferred alternatives.

[0038] According to the first alternative embodiment, step b) includes the following steps:

[0039] - Electrochemical anodizing of the heavily doped gallium nitride layer to make it embrittled.

[0040] - Separate the doped indium gallium nitride layer and the unintentionally doped indium gallium nitride layer from the highly doped gallium nitride layer by thermal activation and / or mechanical action.

[0041] According to a second preferred alternative embodiment, step b) includes the following steps: electrochemically anodizing the heavily doped gallium nitride layer until it dissolves, thereby separating the doped indium gallium nitride layer and the unintentionally doped indium gallium nitride layer from the doped gallium nitride layer. The gallium nitride layer can be dissolved before or after transferring the indium gallium nitride layer to the anodized support.

[0042] Preferably, the method includes a subsequent step: for example, structuring the doped indium gallium nitride layer and the unintentionally doped indium gallium nitride layer by photolithography to form indium gallium nitride mesa. The formation of the mesa allows for the introduction of additional relaxation at the free edges of the mesa.

[0043] Preferably, the method includes the following steps: a doping step performed on an indium gallium nitride mesa by implantation or metal-organic vapor phase epitaxy (possibly using different doping types). For example, different doping types can be used to implant silicon (or n-type dopant) for one mesa and another. This results in pixels (e.g., three pixels) with different doping levels, and therefore different relaxation percentages and different emission wavelengths. This alternative embodiment is advantageous for forming multispectral devices, such as LEDs of different colors or multicolor microdisplays in a simplified manner.

[0044] According to another preferred alternative, the method includes a step between step b) and step c): for example, structuring a doped indium gallium nitride layer and an unintentionally doped indium gallium nitride layer by photolithography to form an indium gallium nitride mesa. This mesa is thus formed prior to the porousing step, allowing the mesa to be porous through both its sides and its upper surface in contact with the electrolyte solution.

[0045] According to this other preferred alternative, the step preferably includes the steps preceding step c): a doping step (possibly using different doping types) performed on the indium gallium nitride mesa by implantation or metal-organic vapor phase epitaxy. For example, selective implantation of n-dopants (e.g., silicon) or p-dopants (e.g., magnesium) can be performed to obtain more or fewer doped mesa, which will result in more or fewer relaxed mesa during the porosification step.

[0046] According to other preferred alternatives, the injection step can be performed before the formation of the platform.

[0047] Preferably, the thickness of the mesa is less than 100 nm. Therefore, despite the high indium concentration, the defect density in the mesa remains limited.

[0048] Preferably, the thickness of the undoped or weakly doped indium gallium nitride layer is less than 3 nm.

[0049] Preferably, the indium content in the indium gallium nitride layer is greater than or equal to 8%. This ensures abundant indium and high-quality epitaxial regeneration.

[0050] This method has many advantages:

[0051] -Easy to implement

[0052] - Suitable for mesa with a small thickness (typically less than 100nm),

[0053] -Structured countertops provide compliance benefits.

[0054] - This caused overall stress relaxation, which reduced piezoelectric polarization compared to a stress layer with the same indium concentration, especially with increased indium doping under the same growth conditions.

[0055] - It provides a "bottom-up" method for manufacturing μLEDs and μ displays: regardless of pixel size, the growth of optical structures (N, QW, P) can be achieved after mesa pixelation, and alignment problems in "pick-and-place" methods can be eliminated.

[0056] - The pixel etching process has no impact on the efficiency of micro LEDs, which makes it possible to produce pixels at the micrometer or even submicrometer scale.

[0057] This method allows for an indium content of up to 40% in the trap, and the quantum efficiency (EQE) of red light can exceed 2.9%.

[0058] The present invention also relates to a substrate, the substrate comprising, in sequence:

[0059] -The supporting structure that we are concerned about

[0060] - A doped porous indium gallium nitride layer, wherein the porosity of the doped porous indium gallium nitride layer is preferably greater than 1% and more preferably 5% to 70%.

[0061] - An unintentionally doped indium gallium nitride layer.

[0062] The unintentionally doped indium gallium nitride layer is not porous.

[0063] Preferably, the support of interest includes a support layer and a buried oxide layer, such as sapphire, silicon carbide (SiC) or silicon.

[0064] The present invention also relates to a light-emitting diode structure, which sequentially comprises a substrate and a stack of epitaxial regenerated lines.

[0065] The substrate includes:

[0066] -The supporting structure that we are concerned about

[0067] - A doped porous indium gallium nitride layer, wherein the porosity of the doped porous indium gallium nitride layer is preferably greater than 1% and more preferably 5% to 70%.

[0068] - An unintentionally doped indium gallium nitride layer,

[0069] The epitaxially regenerated stack, starting from the unintentionally doped indium gallium nitride layer on the substrate, sequentially includes:

[0070] - An epitaxially grown indium gallium nitride layer doped with a first conductivity type of relaxed material.

[0071] - The active region has one or more InGaN / (Ga,In)N quantum wells that emit red, green, or blue light.

[0072] - An indium gallium nitride layer doped with a second conductivity type different from the first conductivity type.

[0073] Other features and advantages of the invention will become apparent from the description appended below.

[0074] Needless to say, this additional description is given only by way of illustrative purposes and should not be construed as limiting those purposes. Attached Figure Description

[0075] The invention will be better understood by reading the description of exemplary embodiments given for purposes purely illustrative and not limiting, with reference to the accompanying drawings:

[0076] Figure 1A , Figure 1B , Figure 1C as well as Figure 1D The different steps of a method for manufacturing a substrate comprising at least a partially relaxed indium gallium nitride layer according to a particular embodiment of the present invention are schematically illustrated by cross-sectional views.

[0077] Figure 2A , Figure 2B , Figure 2C , Figure 2D as well as Figure 2E The different steps of a method for manufacturing a substrate comprising at least a partially relaxed indium gallium nitride layer according to a specific embodiment of the present invention are schematically illustrated by cross-sectional views.

[0078] Figure 3The graphs are based on specific embodiments of the invention and illustrate different phenomena occurring during the anodizing step as a function of doping rate and applied potential (pre-breakdown, porosification, and electropolishing).

[0079] Figure 4A , Figure 4B , Figure 4C , Figure 4D and Figure 4E The different steps of a method for manufacturing a substrate comprising at least a partially relaxed indium gallium nitride layer according to a specific embodiment of the present invention are schematically illustrated by cross-sectional views.

[0080] Figure 5 A micro LED according to a specific embodiment of the present invention is schematically shown in a cross-sectional view.

[0081] To make the accompanying drawings easier to read, the different parts shown in the drawings need not be drawn to a uniform scale.

[0082] It should be understood that different possibilities (alternatives and embodiments) are not excluded and that they can be combined with each other.

[0083] Furthermore, in the following description, the terms of orientation based on the structure, such as "above," "below," etc., will be applied by considering the orientation of the structure as shown in the accompanying drawings. Detailed Implementation

[0084] While not restrictive, this invention is particularly applicable in the field of color microdisplays, especially in the manufacture of red, green, and blue pixels. However, on the one hand, because indium gallium nitride absorbs the entire visible spectrum, and on the other hand, because the valence and conduction bands of indium gallium nitride are near the water stability range—the thermodynamic conditions necessary for water decomposition—this invention can also be used in the photovoltaic or water decomposition fields. This invention is also significant for the manufacture of LEDs or long-wavelength emitting lasers.

[0085] This method enables the acquisition of a substrate or dummy substrate comprising at least partially relaxed, unintentionally doped indium gallium nitride layer 14 on a doped indium gallium nitride layer 13, the method comprising the step of porousening the doped indium gallium nitride layer 13 by electrochemical anodizing.

[0086] The term "partial relaxation" refers to the fact that the lattice parameter a of the unintentionally doped indium gallium nitride layer 14 is greater than the lattice parameter a of gallium nitride on the sapphire layer, thereby allowing more indium to be incorporated.

[0087] The relaxation percentage corresponds to:

[0088] Δa / a =(a c2 -ac1 ) / a c1

[0089] Among them, a c1 The lattice parameters of the initial layer, and

[0090] a c2 represents the lattice parameters of the relaxation layer.

[0091] If a c2 Corresponding to the lattice parameters of the solid material, the layer is 100% relaxed.

[0092] when a c1 =a c2 At that time, the layer is considered to be under stress.

[0093] Partial relaxation refers to a relaxation percentage exceeding 50%.

[0094] More specifically, the method includes the following steps:

[0095] a) Provide a first stack 10, the first stack comprising, in sequence, an initial substrate 11, a gallium nitride layer 12, a doped indium gallium nitride layer 13, and an unintentionally doped indium gallium nitride layer 14;

[0096] b) Transfer the doped indium gallium nitride layer 13 and the unintentionally doped indium gallium nitride layer 14 to the anodized support 21, wherein the unintentionally doped indium gallium nitride layer 14 contacts the anodized support 21 to form a second stack 20.

[0097] c) Connect the doped indium gallium nitride layer 13 and the counter electrode to a voltage generator or a current generator;

[0098] d) Immerse the second stack 20 and the counter electrode in the electrolyte solution;

[0099] e) Apply a voltage or current between the doped indium gallium nitride layer 13 and the counter electrode to make the doped indium gallium nitride layer 13 porous.

[0100] f) Transfer the porous doped indium gallium nitride layer 13 and the unintentionally doped indium gallium nitride layer 14 to the support of interest 31;

[0101] g) Obtaining a relaxed epitaxially grown indium gallium nitride layer by epitaxy on an unintentionally doped indium gallium nitride layer; and

[0102] h) Optionally, micrometer-scale mesa (e.g., from 1 μm to tens of micrometers, preferably 5 μm to 10 μm) can be formed by relaxing via the free edges of the mesa to increase the relaxation percentage.

[0103] Step h) can be performed between steps b) and c), or after step g).

[0104] First refer to Figures 1A to 1D .

[0105] exist Figure 1A The diagram illustrates the first stack 10 provided in step a), which includes an initial substrate 11 (e.g., sapphire, silicon carbide, silicon, or glass). The thickness of the substrate 11 ranges, for example, from 350 μm to 1.5 mm.

[0106] The thickness of the gallium nitride layer 12 ranges, for example, from 30 nm to 4 μm. The gallium nitride layer is unintentionally doped. Unintentionally doped gallium nitride refers to a doping concentration of less than 5 × 10⁻⁶. 17 / cm 3 .

[0107] The doped indium gallium nitride layer 13 is n-doped. Its thickness is, for example, 10 nm to 200 nm. Doped indium gallium nitride refers to an electron concentration of 2*10⁻⁶. 18 Up to 2*10 19 / cm 3 The doped indium gallium nitride layer 101 is conductive and is porousized in step e).

[0108] The doped indium gallium nitride layer 13 has two main surfaces: a first main surface 13a and a second main surface 13b. The first main surface 13a has nitrogen (N) polarity. This first main surface is disposed on the gallium nitride layer 12. The second main surface 13b has gallium (Ga) polarity. This second main surface is configured to contact the unintentionally doped indium gallium nitride layer 14.

[0109] The thickness of the unintentionally doped indium gallium nitride layer 14 is preferably less than 3 nm, for example, 1 nm. Unintentionally doped indium gallium nitride refers to a doping concentration of less than 5 × 10⁻⁶. 17 / cm 3 The layer is electrically insulating. The layer is not porous in step e). At the end of step e), the unintentionally doped indium gallium nitride layer 14 is non-porous (i.e., with a porosity of less than 0.01%, preferably less than 0.001%).

[0110] Preferably, the first stack 10 consists of the aforementioned layers. In other words, the first stack does not include any other layers.

[0111] Preferably, these layers are deposited on the initial substrate 11 in a full-plate manner.

[0112] In step b), the doped indium gallium nitride layer 13 and the unintentionally doped indium gallium nitride layer 14 are transferred to the anodic oxide support 21. Figure 1B ).

[0113] Preferably, for example, the free surface of the unintentionally doped indium gallium nitride layer 14 is transferred to the oxide layer 23BOX of the anodic oxide support 21 by direct bonding or molecular bonding.

[0114] The anodic oxide substrate 21 preferably includes a support layer 22 and a layer referred to as BOX (“buried oxide layer”) 23 (sometimes also referred to as a continuous oxide layer).

[0115] The support layer (or support body) 22 is, for example, sapphire, silicon, or glass. The thickness of the support layer 22 ranges, for example, from 350 μm to 1.5 mm.

[0116] The thickness of the oxide layer 23BOX ranges from, for example, 100 nm to 4 μm.

[0117] According to a preferred embodiment, Smart Cut is used. TM The technology performs this transition. The method includes the following steps:

[0118] - Implant atomic material to form an embrittlement region at a depth in the doped indium gallium nitride layer, the depth being close to the final thickness of the doped indium gallium nitride layer to be retained in the final substrate;

[0119] - Bond the first stack 10 to the anodized support 21.

[0120] - Transfer heat energy to separate the doped indium gallium nitride layer in the embrittled region.

[0121] The initial substrate 11 and the unintentionally doped gallium nitride layer 12 of the first stack 10 are separated from the doped indium gallium nitride layer 13 by splitting along the plane of the embrittlement region. This separation is performed, for example, by heat treatment at a temperature range of 400°C to 600°C for several minutes to several hours. This splitting allows the doped indium gallium nitride layer 13 and the unintentionally doped indium gallium nitride layer 14 to be transferred to the anodized support 21.

[0122] According to a particular embodiment, the initial substrate 11 and the unintentionally doped gallium nitride layer 12 can be detached from the doped indium gallium nitride layer 13 by, for example, etching the sacrificial intermediate layer using a laser (also known as a "lift-off" technique).

[0123] Thus, at the end of step b), a second stack 20 is obtained, which sequentially includes a support layer 22, a buried oxide layer 23, an unintentionally doped indium gallium nitride layer 14, and a doped indium gallium nitride layer 13. Figure 1B Therefore, the first surface 13a of N polarity can be reached.

[0124] Injecting Smart Cut TMAfterwards and before the anodizing step, a silicon implantation step can be performed to make the doped indium gallium nitride layer 13 more conductive.

[0125] according to Figures 2A to 2E In an alternative embodiment of the method shown, the first stack 10 may further include a highly doped gallium nitride layer 15, referred to as n++ gallium nitride, between the gallium nitride layer 12 and the doped indium gallium nitride layer 13.

[0126] The height of the highly doped gallium nitride layer 15 ranges, for example, from 10 nm to 200 nm. Highly doped gallium nitride refers to gallium with a doping level of 10⁻⁶. 19 / cm 3 Or higher.

[0127] Advantageously, according to this alternative embodiment, in step b)( Figure 2B After the first stack 10 is transferred to the anodized support 21, the highly doped gallium nitride layer 15 is separated from the doped indium gallium nitride layer 14 by electrochemical anodizing, thereby making the highly doped gallium nitride layer porous. Figure 2C The highly doped gallium nitride layer 15 was used as a thin film. The different doping rates allowed for high selectivity between porousification and electropolishing. Figure 3 ).

[0128] Advantageously, separation is facilitated by thermal and / or mechanical activation.

[0129] Alternatively, for example, if it is necessary to apply mechanical stress to separate the doped gallium nitride layer, an anodizing step of the doped gallium nitride layer 15 can be performed before transfer step b).

[0130] When a porosimetry step is performed on the doped gallium nitride layer 15 prior to step b), the porosimetry rate can be controlled to maintain consistency throughout the stack. After the transition, thermal and / or mechanical stress is applied to separate the device into two parts, preferably at the most brittle interface, i.e., between the highly doped gallium nitride layer and the doped indium gallium nitride layer.

[0131] The porousification step of the highly doped gallium nitride layer 15 can be performed using the same electrolyte solution (properties and / or concentration) and / or the same voltage (or applied current) used in step (step e) of the porous doped indium gallium nitride layer 13.

[0132] according to Figures 4A to 4E The alternative embodiment shown includes a first stack comprising a highly doped gallium nitride layer 15 disposed between a gallium nitride layer 12 and a doped indium gallium nitride layer. Figure 4A Step b) is performed as follows:

[0133] Connect the doped gallium nitride layer 15 and the counter electrode to a voltage generator or a current generator.

[0134] - Immerse the first stack 10 and the counter electrode in the electrolyte solution.

[0135] - A voltage or current is applied between the highly doped gallium nitride layer 15 and the counter electrode to etch the highly doped gallium nitride layer 15, thereby separating the unintentionally doped indium gallium nitride layer 14 and the doped indium gallium nitride layer 13 from the gallium nitride layer 12. Figure 4B ).

[0136] This produced an indium gallium nitride thin film.

[0137] Then the doped indium gallium nitride layer 13 and the unintentionally doped indium gallium nitride layer 14 are transferred to the anodic oxide support 21 (step b- Figure 4C ).

[0138] In this embodiment, the n++ gallium nitride layer 15 is etched laterally. Preferably, this embodiment of the method does not perform additional steps for separating the indium gallium nitride layers 13, 14 from the other layers of the first stack 10. This alternative embodiment is preferably used for small-sized samples (typically a few square centimeters at most). For larger areas, electrolyte pathways can be pre-formed within the doped gallium nitride layer 15.

[0139] After the transition between the doped indium gallium nitride layer 13 and the undoped indium gallium nitride layer 14, the doped indium gallium nitride layer 13 is subjected to electrochemical anodic oxidation. This anodic oxidation can partially relax the doped indium gallium nitride layer 13.

[0140] Therefore, in step c), the second stack 20 and the counter electrode (CE) 500 are electrically connected to a voltage generator or a current generator. The second stack 20 serves as a working electrode (WE). It is referred to as a voltage generator below, but it can also be a current generator that applies current between the device and the counter electrode.

[0141] The counter electrode 500 is made of a conductive material such as a metal like platinum.

[0142] In step d), the electrode is immersed in an electrolyte, also known as an electrolyte bath or electrolyte solution. The electrolyte can be acidic or alkaline. An example of an electrolyte is oxalic acid. It can also be potassium hydroxide (KOH), hydrogen fluoride (HF), nitric acid (HNO3), or sulfuric acid (H2SO4).

[0143] Step d) can be performed before step c).

[0144] In step e), a voltage is applied between the device and the counter electrode 500. This voltage can range from 1V to 100V, preferably from 3V to 100V. The duration of this voltage application ranges from, for example, several seconds to several hours. When the current becomes zero at the applied potential, the anodic oxidation reaction ends: in this case, there is no more charge transfer and the electrochemical reaction stops.

[0145] The electrochemical anodizing step can be performed under ultraviolet (UV) light. The method may also include a first electrochemical anodizing without UV radiation and a second electrochemical anodizing with UV radiation.

[0146] Preferably, the entire volume of the doped indium gallium nitride layer 13 is porous.

[0147] At the end of the porosimetry step, the porosity of the doped indium gallium nitride layer 13 is at least 1%. Preferably, the porosity ranges from 5% to 70%.

[0148] The maximum size (height) of the pores can vary from a few nanometers to a few micrometers. The minimum size (diameter) of the pores can vary from a few nanometers to several hundred nanometers, particularly from 10 nm to 70 nm, for example from 30 nm to 70 nm, and preferably from 15 nm to 40 nm.

[0149] The obtained porosity (porosity and pore size) depends on the doping amount of the indium gallium nitride layer 13 and the process parameters (applied voltage, time period, electrolyte properties and concentration). By changing the porosity, the incorporation / segregation rate can be controlled. Subsequently, the porosity, especially the pore size, can be changed according to the applied temperature during epitaxial regrowth.

[0150] The porousing step allows the unintentionally doped indium gallium nitride layer 14 on the indium gallium nitride layer 13 to be fully or partially relaxed, after which the indium gallium nitride layer 13 is lightly doped or even no longer doped.

[0151] To make the second side 13b of the metal (gallium) polarity face the front, the second stack 20 is transferred to the support 31 of interest, which includes, for example, a support layer 32 preferably made of silicon, silicon carbide, glass or sapphire and an oxide layer BOX 33 (step f).

[0152] Step f) can be performed using SmartCut™ technology, following these steps:

[0153] - Inject atomic material to form an embrittlement region at a depth in the unintentionally doped indium gallium nitride layer, which is close to the final thickness of the layer to be retained in the final substrate;

[0154] - Bonding an unintentionally doped indium gallium nitride layer to the support of interest 31; and

[0155] - Transfer thermal energy to separate unintentionally doped indium gallium nitride layers in embrittled regions.

[0156] Therefore, at the end of this method, an InGaNOX (“Indium Gallium Nitride on Substrate X”) type substrate 30 is obtained, which includes (from the back side to the front side) Figure 1C , Figure 2D , Figure 4D ):

[0157] - The support 31 of interest is formed, for example, by a support layer 32 preferably made of sapphire and an oxide layer BOX 33.

[0158] - A porous doped indium gallium nitride layer 13, the porous doped indium gallium nitride layer having gallium polarity on its front side,

[0159] - An unintentionally doped indium gallium nitride layer 14 that is at least partially or completely relaxed without pores.

[0160] According to an alternative embodiment, the method may include structuring a porous indium gallium nitride layer and an unintentionally doped indium gallium nitride layer to form an indium gallium nitride mesa. Figure 1D The subsequent steps (step h) of (2E, 4E).

[0161] According to a preferred alternative embodiment, the platform can be formed prior to the porousing step.

[0162] The mesa can be constructed and then doped. In this alternative embodiment, the method may include the following steps prior to the porosification step:

[0163] -Structuring of indium gallium nitride (INH) layers deposited in a monolithic manner and unintentionally doped INH layers to form INH mesa.

[0164] - Preferably, the space between the countertops is filled with a suitable material, and the resulting whole is planarized to make the surface flat.

[0165] - Locally implant dopants into the mesa with the same or different concentrations; for example, multiple implantation operations can be performed using different sets of masks.

[0166] Alternatively, doping can be performed before constructing the mesa. According to this alternative embodiment, the method may include the following steps prior to the porosimetry step:

[0167] - Local implantation of dopant into a doped indium gallium nitride (IGaN) layer deposited across the entire substrate, wherein the dopant concentration can vary along the IGaN layer to create more or fewer doped regions.

[0168] -Structuring the indium gallium nitride (IGN) layers deposited on the entire plate and the unintentionally doped IGN layers to form IGN mesa.

[0169] As an illustration, hard masks such as silicon nitride (SiN) or silicon oxide (SiO2) type masks can be used to perform implantation, and alignment marks (e.g., aluminum (Al), titanium (Ti), titanium nitride (TiN)) can be used to align the mesa with the implantation region. Using two different masks, two regions with different implantation energies can be fabricated via, for example, MOCVD, thus enabling the fabrication of two different doping types in addition to the initial doping of the doped substrate. Alignment marks are used to align the two masks used for implantation and the mask used for mesa etching.

[0170] Use the same set of masks for injection and platform.

[0171] The mesa can be constructed using photolithography.

[0172] Tabletops (also known as elevations) are relief elements. They are created, for example, by etching consecutive layers or several stacked consecutive layers, leaving only the layer or multiple "reliefs" of those layers. Etching is typically plasma or dry etching (e.g., RIE plasma). Reliefs can define pixels.

[0173] Preferably, the side of the mesa is perpendicular to the stacking of different layers of the substrate 30.

[0174] The dimensions (width and length) of the mesa range from 500 nm to 500 μm. The width and length mentioned above refer to the dimensions parallel to the surfaces stacked below.

[0175] The spacing between the two consecutive mesa 100 is 50 nm to 20 μm.

[0176] The indium gallium nitride mesa includes a doped indium gallium nitride layer 13 and an unintentionally doped indium gallium nitride layer 14. Preferably, the mesa consists of these two layers 13 and 14.

[0177] The thickness of the doped indium gallium nitride layer 13 on the mesa is preferably in the range of 10 nm to 200 nm. Thinner layers (less than 100 nm) can contain a high concentration of indium while maintaining good material quality (few defects). Thickness refers to the dimension of the mesa perpendicular to the stack below.

[0178] Alternatively, doping via implantation (particularly silicon or germanium doping) can be replaced by doping via metal-organic vapor deposition (MOCVD) epitaxy. For example, three consecutive epitaxial steps can be performed to obtain three different doping levels at the end of the process to form red, green, and blue mesa (RGB).

[0179] The method includes an unintentionally doped indium gallium nitride layer 14 across the entire plate. Figure 5 This is a subsequent step of fabricating an epitaxially regenerated length stack 40 on or on an unintentionally doped indium gallium nitride layer 14 on an indium gallium nitride mesa. This step is preferably performed to form an epitaxially regenerated length LED, particularly an all-indium gallium nitride red (or green) LED.

[0180] Since the unintentionally doped indium gallium nitride layer 14 is not porous during the electrochemical anodizing step, it is a continuous, non-porous 2D layer, which is beneficial for the epitaxial regeneration of long stacks 40 and the epitaxial layer has better resistance.

[0181] The epitaxial regeneration stack 40 of an all InGaN LED includes, for example, an unintentionally doped InGaN layer 14, sequentially comprising:

[0182] The indium gallium nitride layer 41 is n-doped, wherein the indium concentration of the n-doped indium gallium nitride layer is preferably the same as the indium concentration of the indium gallium nitride layer of the relaxation substrate.

[0183] -Active region 42, with one or more InGaN / (Ga,In)N quantum wells emitting red (or green) light.

[0184] -Based on p-doped aluminum gallium nitride (AlGaN) or the electronic barrier of gallium nitride 43,

[0185] The p-doped indium gallium nitride layer 44 preferably has an indium concentration that is the same as or lower than that of the n-doped indium gallium nitride layer.

[0186] The p++-doped indium gallium nitride layer 45 preferably has the same indium concentration as the p indium gallium nitride layer.

[0187] More specifically, the all-indium gallium nitride (IGaN) LED structure can include, in sequence:

[0188] -Indium gallium nitride substrate,

[0189] - A 350nm n-doped indium gallium nitride layer formed from 15x In0.03Ga0.97N / GaN (20nm / 1.8nm thickness),

[0190] - Multiple quantum wells (MQWs) formed from 5x In0.40Ga0.60N / In0.03Ga0.097N (thicknesses of 2.3nm / 5nm, 7nm, and 11nm),

[0191] -uid In 0.03 Ga 0.97 N-layer (10nm),

[0192] -Al 0.1 Ga 0.9 N:Mg layer (20nm),

[0193] -Mg-doped In 0.03 Ga 0.97 N-layer (125nm),

[0194] -p++ doped In 0.03 Ga 0.97 N-layer (25nm).

[0195] Different growth processes can be used to form epitaxially grown layers.

[0196] According to the first alternative embodiment, the regrowth is lateral, and the subsequent pixelation step is preferably performed by etching.

[0197] According to an alternative embodiment, regrowth occurs vertically above the platform. In this alternative embodiment, the pixel corresponds to the platform below.

[0198] Preferably, the passivation layer is deposited, for example, by atomic layer deposition (ALD) on the side surface of the mesa 100 and / or the epitaxially regenerated LED. The passivation layer can be aluminum oxide. The thickness of the passivation layer can be several nanometers, for example, 2 nm to 5 nm.

[0199] References

[0200] [1] Even et al, "Enhanced In incorporation in full InGaN heterostructure grown on relaxed InGaN pseudo-substrate", Appl. Phys. Lett. 110, 262103 (2017).

[0201] [2] Jang et al, "Electrical and structural properties of GaN films and GaN / InGaN light-emitting diodes grown on porous GaN templates fabricated by combined electrochemical and photoelectrochemical etching", Journal of Alloys and Compounds 589(2014)507-512.

Claims

1. A method for fabricating a relaxed epitaxially grown indium gallium nitride layer from a gallium nitride / indium gallium nitride substrate, the method comprising the following steps: a) A first stack (10) is provided, the first stack comprising, in sequence, an initial substrate (11), a gallium nitride layer (12), a doped indium gallium nitride layer (13), and an unintentionally doped indium gallium nitride layer (14), wherein the doping concentration of the doped indium gallium nitride layer (13) is 2*10⁻⁶. 18 Up to 2*10 19 / cm 3 The unintentionally doped indium gallium nitride layer (14) has a doping concentration of less than 5*10⁻⁶. 17 / cm 3 ; b) The doped indium gallium nitride layer (13) and the unintentionally doped indium gallium nitride layer (14) are transferred to the anodized support (21), wherein the unintentionally doped indium gallium nitride layer (14) is in contact with the anodized support (21) to form a second stack (20). c) Connect the doped indium gallium nitride layer (13) and the counter electrode to a voltage generator or a current generator; d) Immerse the second stack (20) and the counter electrode in the electrolyte solution. e) Apply a voltage or current between the doped indium gallium nitride layer (13) and the counter electrode to make the doped indium gallium nitride layer (13) porous; f) Transferring the porous doped indium gallium nitride layer (13) and the unintentionally doped indium gallium nitride layer (14) to the support of interest (31); and g) An indium gallium nitride layer is formed by epitaxy on the unintentionally doped indium gallium nitride layer to obtain a relaxed epitaxially grown indium gallium nitride layer.

2. The method according to claim 1, wherein, The anodic oxide support (21) and / or the support of interest (31) include a support layer (22, 32) and a buried oxide layer (23, 33).

3. The method according to claim 2, wherein, The support layers (22, 32) are made of sapphire or silicon.

4. The method according to claim 1, wherein, The first stack (10) further includes a highly doped gallium nitride layer (15) disposed between the unintentionally doped gallium nitride layer (12) and the doped indium gallium nitride layer (13), the highly doped gallium nitride layer (15) having a doping concentration of not less than 10. 19 / cm 3 .

5. The method according to claim 4, characterized in that, Perform step b) according to the following steps: - The highly doped gallium nitride layer (15) is subjected to electrochemical anodic oxidation to make the highly doped gallium nitride layer (15) embrittled; as well as - The doped indium gallium nitride layer (13) and the unintentionally doped indium gallium nitride layer (14) are separated from the highly doped gallium nitride layer (15) by thermal activation and / or mechanical action.

6. The method according to claim 4, wherein, Step b) includes the following steps: performing electrochemical anodizing on the highly doped gallium nitride layer (15) until the highly doped gallium nitride layer (15) dissolves, thereby separating the doped indium gallium nitride layer (13) and the unintentionally doped indium gallium nitride layer (14) from the doped gallium nitride layer.

7. The method according to any one of claims 1 to 4, wherein, Step b) is performed according to the process that includes the following steps: - Inject atomic material to form an embrittlement region at a depth in the doped indium gallium nitride layer (13) that is close to the final thickness of the doped indium gallium nitride layer (13) to be retained in the final substrate (30); - The first stack (10) is bonded to the anodic oxide support (21), and the unintentionally doped indium gallium nitride layer (14) is in contact with the anodic oxide support (21); as well as - Transfer heat energy to separate the doped indium gallium nitride layer (13) from the gallium nitride layer (12) in the embrittlement region.

8. The method according to any one of claims 1 to 6, wherein, The method includes the following steps: structuring the doped indium gallium nitride layer (13) and the unintentionally doped indium gallium nitride layer (14) to form an indium gallium nitride mesa.

9. The method according to claim 8, wherein, The doped indium gallium nitride layer (13) and the unintentionally doped indium gallium nitride layer (14) are structured by photolithography.

10. The method according to claim 8, wherein, The method includes the following steps: performing a doping step by implantation or metal-organic vapor phase epitaxy on the indium gallium nitride mesa.

11. The method according to any one of claims 1 to 6, wherein, Different types of doping are performed during the doping step.

12. The method according to any one of claims 1 to 6, wherein, The method includes the following steps between step b) and step c): structuring the doped indium gallium nitride layer (13) and the unintentionally doped indium gallium nitride layer (14) to form an indium gallium nitride mesa.

13. The method according to claim 12, wherein, The doped indium gallium nitride layer (13) and the unintentionally doped indium gallium nitride layer (14) are structured by photolithography.

14. The method according to claim 12, wherein, The method includes the following steps prior to step c): performing a doping step by implantation or metal-organic vapor phase epitaxy on the indium gallium nitride mesa.

15. The method according to any one of claims 1 to 6, wherein, Different types of doping are performed during the doping step.

Images