Semiconductor substrate for transferring a layer by laser separation and associated manufacturing method

Abstract

The invention relates to a method (PR1; PR1') for manufacturing a substrate (SUB) comprising a layer (TL) to be transferred and designed to allow the layer (TL) to be transferred to be separated by light irradiation, the method (PR1) comprising the steps of: forming (12) a first layer of silicon nitride (SiNCar) on a carrier substrate (WafCar); forming (22) a second layer of silicon nitride (SiNDon) on a donor substrate (WafDon); assembling (30) the carrier substrate (WafCar) with the donor substrate (WafDon) by directly bonding the first layer of silicon nitride (SiNCar) to the second layer of silicon nitride (SiNDon); forming (32, 34) the layer (TL) to be transferred by removing a part (Waf'Don) from the donor substrate (WafDon), the layer (TL) to be transferred remaining attached to the carrier substrate via the first layer of silicon nitride (SiNCar) and the second layer of silicon nitride (SiNDon).

Description

Semiconductor substrate for single-layer transfer by laser separation and associated manufacturing process TECHNICAL FIELD OF THE INVENTION

[0001] The present invention relates to the field of the semiconductor industry and proposes a substrate designed to allow laser peeling of a thin film, which can be a single-crystal layer or a structured layer forming all or part of a semiconductor device. TECHNOLOGICAL BACKGROUND

[0002] The semiconductor industry is increasingly using thin film transfer techniques, which consist of forming a thin film on a first substrate called a "donor substrate", then transferring the thin film formed onto a second substrate, called a "carrier substrate" or "manipulation substrate".

[0003] The donor substrate can be chosen for its compatibility with the conditions for thin film formation, or the thin film can be formed within the mass of the material constituting this donor substrate.

[0004] The carrier substrate is adapted to the processing of this thin film, for example by being compatible with subsequent manufacturing steps, or by supporting a structure onto which the thin film will be integrated. Finally, a fabricated thin structure comprising the thin film, or only a portion of it remaining after the manufacturing steps, must be transferred to a third substrate, called the "receiving substrate." Alternatively, the thin film can simply be separated from the carrier substrate and handled as an independent component.

[0005] Among the thin-film transfer or detachment techniques used in the semiconductor industry, the laser lift-off (LLO) separation technique has become established as an effective method for separating a thin structure, possibly composed of multiple stacked films, from a substrate via a layer called the "separation layer." The separation is triggered by thermal energy from light irradiation, such as laser irradiation. This thermal energy degrades the separation layer, thus separating the thin structure from the substrate.

[0006] Typically, the separation layer is irradiated by a laser through the carrier substrate. The separation layer, which can be single-layer or multi-layer, must absorb a sufficient amount of light at the laser wavelength to promote separation by degradation (for example, by locally melting the separation layer or causing its dissociation into elementary components), while the carrier substrate must be substantially transparent at that wavelength.A balance between laser power, laser wavelength and choice of materials (for the separation layer and carrier substrate) must be found to allow separation while preserving the structure to be transferred from the impact of energy dissipation due to interaction processes with light: heat transfer due to absorption in the separation layer or direct absorption of part of the laser irradiation by the layer(s) to be transferred.

[0007] US patents 2014 / 327049, US 2023 / 207661, CN 113 097 124 each describe a method for manufacturing semiconductor devices involving separation facilitated by laser irradiation.

[0008] However, there is a need for improvement in terms of the quality and reliability of the conventional process described above. In particular, separation can be difficult under certain circumstances, for example, depending on the type of treatments to which the layer to be transferred and its substrate have been subjected.

[0009] It has been observed that during laser irradiation layer transfer processes, the transfer can present difficulties in effectively separating certain layers from their supporting substrates. This problem is exacerbated when the layers to be transferred and their supporting substrates have been subjected to heat treatments at relatively high temperatures, typically exceeding 700°C.

[0010] The inventors of the present invention have determined that the cause of these difficulties is that (i) species present in their gaseous form (such as dihydrogen), located in the vicinity of the layers where the separation takes place, between a layer to be transferred and its carrier substrate play an assisting role in the separation, by forming bubbles which generate a pressure tending to push back the layers surrounding them and (ii) heat treatments at excessively high temperatures tend to degas these layers, removing said species which could have assisted the separation.

[0011] Consequently, it is proposed to trap a source of gaseous species (in this case hydrogen, for example in the form of water monolayers) between layers sufficiently impermeable to prevent the phenomenon of degassing at high temperatures, typically above 700°C, and even above 1200°C.

[0012] A first object of the invention is a method for manufacturing a substrate comprising a transfer layer and designed to allow separation of the transfer layer by light irradiation. The method comprises the steps of: forming a first silicon nitride layer on a carrier substrate; forming a second silicon nitride layer on a donor substrate; assembling the carrier substrate to the donor substrate by direct bonding of the first silicon nitride layer to the second silicon nitride layer; and forming the transfer layer by removing a portion of the donor substrate, leaving the transfer layer attached to the carrier substrate via the first and second silicon nitride layers. The method further comprises the first and second silicon nitride layers, each formed to have a density greater than 1.8 g / cm³. 3or a density greater than 2 g / cm³ 3 .

[0013] The substrate formation process according to the first object of the invention makes it possible to trap one or more chemical species (water H2O, hydroxide ions OH) - etc.) at the interface between the two silicon nitride layers. The silicon nitride is sufficiently dense to prevent or at least limit the loss of hydrogenated species (water, OH). -, etc.) of these layers, even when a substantial thermal budget is imposed.

[0014] Thus, this process makes it possible to obtain a substrate carrying a transfer layer capable of withstanding heat treatments of the type required for FEOL (Front End Of Line) type processes, which may require high-temperature heat treatments for the manufacture of components in the semiconductor industry.

[0015] Of course, the reliability of separating a layer to be transferred during a transfer operation is also improved in the case where the layer and its support are not subjected to high temperature treatments.

[0016] According to additional, non-limiting characteristics of the support according to the invention, considered individually or in any technically feasible combination:

[0017] - the first silicon nitride layer and the second silicon nitride layer can each be formed so as to have an optical index greater than 1.75 or greater than 1.8 for a wavelength of 550 nm;

[0018] - the substrate can be designed to allow separation of the layer to be transferred by light irradiation at a given wavelength, and the carrier substrate can be substantially transparent at the given wavelength;

[0019] - the process may further include the formation of at least one of: a first thermal insulation layer on the carrier substrate, so as to form a stack comprising the first thermal insulation layer interposed between the carrier substrate and the first silicon nitride layer; and a second thermal insulation layer on the donor substrate, so as to form a stack comprising the second thermal insulation layer interposed between the donor substrate and the second silicon nitride layer;

[0020] - the first thermal insulation layer and the second thermal insulation layer can each be independently formed from a silicon oxide layer with a thickness between 10 nm and 1000 nm;

[0021] - the process may further include a step of forming a light-absorbing layer in direct contact with one of the first silicon nitride layer and the second silicon nitride layer;

[0022] - the light absorption layer can be formed from a material chosen from silicon dioxide, titanium nitride, silicon nitride and germanium nitride;

[0023] - in which molecules including hydrogen atoms can be trapped between the first silicon nitride layer and the second silicon nitride layer following the step of fixing the carrier substrate to the donor substrate by direct bonding, which can be detected as a peak in concentrations of chemical bonds such as OH or Si-H bonds, representative of the presence of the interface and its hydrogen content at the interface.

[0024] A second object of the invention is a substrate comprising a transfer layer and designed to allow separation of the transfer layer by light irradiation, the substrate comprising a stack of, in this order: a carrier substrate; a first silicon nitride layer; a second silicon nitride layer in direct contact with the first silicon nitride layer; and the transfer layer, wherein the first silicon nitride layer and the second silicon nitride layer each have a density greater than 1.8 g / cm³ 3 or a density greater than 2 g / cm³ 3 .

[0025] This substrate, intended as a starting element in a manufacturing process for an element in the field of semiconductor technology, is particularly suitable for the treatment of the thin film to be transferred, for example according to FEOL type processes, before it is transferred onto a final host substrate by laser irradiation.

[0026] Indeed, for the reasons detailed above, its structure, trapping hydrogen at the level of a bilayer which is heated locally by light irradiation, improves the reliability of a layer transfer by light irradiation, also called LLO for Laser Lift-Off in English terminology.

[0027] In the substrate, hydrogen atoms can be trapped between the first silicon nitride layer and the second silicon nitride layer, which can be detected as a peak in chemical bond concentrations such as OH or Si-H bonds, representative of the presence of the interface and its hydrogen content at the interface.

[0028] According to additional, non-limiting characteristics of the support according to the invention, considered individually or in any technically feasible combination:

[0029] - the first silicon nitride layer and the second silicon nitride layer can each have an optical index greater than 1.75 or greater than 1.8 for a wavelength of 550 nm;

[0030] - the substrate can be designed to allow separation of the layer to be transferred by light irradiation at a given wavelength, and the carrier substrate can be substantially transparent at the given wavelength;

[0031] - the substrate may further comprise at least one of: a first thermal insulation layer interposed between the carrier substrate and the first silicon nitride layer; and a second thermal insulation layer interposed between the second silicon nitride layer and the layer to be transferred, the first thermal insulation layer and the second thermal insulation layer each being formed of silicon oxide;

[0032] - the substrate may further comprise a light-absorbing layer in direct contact with one of the first silicon nitride layers and the second silicon nitride layer;

[0033] - the light absorption layer can be formed from a material chosen from silicon dioxide, titanium nitride, silicon nitride and germanium nitride;

[0034] - the carrier substrate can be substantially transparent to light of a given wavelength, and at least one of the first silicon nitride layer, the second silicon nitride layer and, where applicable, the light-absorbing layer, is substantially absorbing to light of the given wavelength, so that the layer to be transferred can be separated from the carrier substrate when that of the first silicon nitride layer, the second silicon nitride layer and, where applicable, the light-absorbing layer, is irradiated by light of the given wavelength through the carrier substrate during a light-irradiation separation operation;

[0035] - the substrate can be obtained by the process according to the first aspect of the invention;

[0036] - the substrate may include the thermal insulation layer of silicon oxide interposed between the second layer of silicon nitride and the layer to be transferred, the carrier substrate being able to be formed of silicon.

[0037] A third object of the invention is a method for forming and separating a transfer layer, comprising providing a substrate containing the transfer layer and manufactured according to the first aspect of the invention, and further comprising the steps of: assembling the transfer layer to a receiving substrate so as to form an assembly, the transfer layer being interposed between the carrier substrate and the receiving substrate; and irradiating the assembly by means of light irradiation so as to separate the transfer layer from the carrier substrate.

[0038] By using a substrate comprising a transfer layer such as that fabricated according to the first object of the invention, the process of separating the transfer layer is made more reliable and easier. Indeed, this structure allows for the trapping of species including hydrogen, which facilitates the separation of the transfer layer by forming bubbles that generate pressure at the interface between the two silicon nitride layers. This pressure, combined with the decomposition of a layer such as the silicon nitride layers, facilitates and improves the reliability of the separation by light irradiation.

[0039] Moreover, thanks to the use of silicon nitride layers, this advantage is retained even in situations where the layer to be transferred and its support have been subjected to heat treatments at relatively high temperatures, typically above 700°C.

[0040] According to additional, non-limiting characteristics of the support according to the invention, considered individually or in any technically feasible combination:

[0041] - the process may include the formation of the second thermal insulation layer of silicon oxide, on the donor substrate, the carrier substrate being formed of silicon, and the light irradiation being carried out through the carrier substrate and using a wavelength between 9 and 11 µm;

[0042] - the process may further include a step of preparing the layer to be transferred including a heat treatment at a temperature above 700°C or above 1000°C before the step of irradiating the assembly. FIGURES

[0043] Other features and advantages of the invention will become apparent from the detailed description of the invention which follows with reference to the accompanying figures in which:

[0044] Laillustre a manufacturing process for a substrate comprising a layer intended to be separated from its support;

[0045] Laillustre a process of transferring the layer from its carrier substrate to a host substrate;

[0046] Laillustre a variant of the process of the;

[0047] Laillustre is a variant of the process of the.

[0048] This illustrates a variant of the processes shown in Figures 1 and 2.

[0049] DETAILED DESCRIPTION OF A SPECIFIC EMBODIMENT OF THE INVENTION

[0050] First method of implementing the invention

[0051] A first embodiment of the invention is explained with the aid of Figures 1 and 2, which illustrate (i) the fabrication of a SUB substrate designed to enable laser transfer of a TL layer, for example a layer of a material such as single-crystal silicon, and (ii) a process of transferring the TL layer onto a receiving substrate.

[0052] Laillustrates the steps of a PR1 process for manufacturing a SUB substrate, which includes a SEP separation layer and is designed to allow laser transfer of a TL layer onto a Waf receiving substrate Rec , according to a PR2 process allowing cleavage of the SUB structure at the level of the SEP layer in response to laser irradiation. The PR2 process will be described in detail below using the.

[0053] Specifically, the SEP separation layer comprises a pair of silicon nitride SiN layers Car and SiN Donfacing each other, at the interface of which are trapped species capable of degassing under the effect of laser irradiation and aiding in the cleavage of the SUB substrate. It is advantageous to form nitride layers that tend to trap species, such as hydrogen, capable of degassing, and thus to form these layers in such a way that they exhibit a behavior impermeable to the diffusion of these species.

[0054] According to the PR1 process, a Waf carrier substrate Car is prepared in steps 10 to 12, a Waf donor substrate Don is prepared in steps 20 to 24, a TL layer of the donor substrate is transferred onto the carrier substrate and finalized in steps 30 to 34.

[0055] Preparation of the Waf carrier substrate Car

[0056] In step 10, a Waf carrier substrate Caris provided. The carrier substrate can be a substrate conventionally used in the semiconductor industry. The carrier substrate has a top surface Car and a lower surface area Bot Car which form two parallel flat surfaces, as illustrated in (A1).

[0057] The Waf carrier substrate Car can be chosen to be substantially transparent to light of a given wavelength which will be used during the PR2 process to transfer the TL layer onto the Waf host substrate Rec definitive, considering that light irradiation through Waf can then be carried out Car The Waf carrier substrate Car can be considered substantially transparent if 90% or more of the given wavelength light is transmitted through the carrier substrate Waf Car .

[0058] In this example, we assume that a wavelength in the ultraviolet range will be used for the transfer of the TL layer onto its Waf host substrate. Rec final during the implementation of the PR2 process. Therefore, a substrate that is substantially transparent to ultraviolet light can be chosen for the Waf carrier substrate. Car such as a glass or sapphire plate.

[0059] At step 12 following step 10, a layer of thermal insulation ThIns Car then a layer of silicon nitride SiN Car are formed successively, in this order, on the Top surface Car of the supporting substrate. Figure (A2) illustrates the supporting substrate following this step. The thermal insulation layer ThIns Car can be omitted.

[0060] The ThIns thermal insulation layer Carcan be formed by thermal oxidation of the substrate when it is made of silicon, to form a silicon oxide layer with a thickness between 10 nm and 200 nm. The ThIns layer Car It can also be formed by the PECVD (Plasma-Enhanced Chemical Vapor Deposition) process using tetraethyl orthosilicate, often identified by its abbreviation TEOS. Alternatively, the thermal insulation layer can be formed by any known thin-film deposition method, for example, by plasma-enhanced chemical vapor deposition, or PE-CVD for Plasma Enhanced Chemical Vapor Deposition.

[0061] The silicon nitride layer SiN Carcan be formed to a thickness of 10 nm to 200 nm by deposition of silicon nitride SiN, for example by PE-CVD. Preferably, in order to form a layer tending to be impermeable to the diffusion of outgassing species located between the two outgassing silicon nitride layers, the deposition parameters are such that the silicon nitride SiN layer Car is formed in such a way as to have a density greater than 1.8 g / cm³ 3 preferably a density greater than 2 g / cm³ 3Such densities guarantee low porosity and a sufficient level of impermeability to prevent or slow the diffusion of species degassed by laser irradiation for cleavage. These densities can correspond to optical indices greater than 1.75 or greater than 1.8 for a wavelength of 550 nm, respectively. The deposition techniques for obtaining such a layer are known; for example, see the articles by Moussi (Structural, optical and electrical properties of Si-rich and N-rich PECVD silicon nitride films, Scientific Reports, (2025) 15:33646) or Kaloyeros (Review—Silicon Nitride and Silicon Nitride-Rich Thin Film Technologies: State-of-the-Art Processing Technologies, Properties, and Applications, Alain E. Kaloyeros et al 2020 ECS J. Solid State Sci. Technol. 9 063006).

[0062] Waf donor substrate preparation Don

[0063] In step 20, a Waf donor substrate Donis provided. The donor substrate can be a substrate conventionally used in the semiconductor industry, such as a semiconductor wafer, for example, a silicon wafer. The donor substrate has a top surface Don and a lower surface area Bot Don which form two parallel flat surfaces, as illustrated in (B1).

[0064] At step 22 following step 20, a layer of thermal insulation ThIns Don then a layer of silicon nitride SiN Don are formed successively, in this order, on the Top surface Car of the supporting substrate. Figure (B2) illustrates the supporting substrate following this step. The thermal insulation layer ThIns Don can be omitted.

[0065] The optional ThIns thermal insulation layer Don and the silicon nitride layer SiN Doncan be formed using the same method(s) as the optional ThIns thermal insulation layer Car and the silicon nitride layer SiN Car , respectively. The characteristics, such as thickness, of each of these layers can be determined independently of those of the other layers.

[0066] At step 24 following step 22, in order to define a layer of the material forming the donor substrate and to prepare its separation from the rest of the donor substrate, an ion implantation process Imp is implemented: a light ion species, hydrogen H + and / or helium He + is introduced into the Waf donor substrate Don through the top surface Don , and therefore through the ThIns layers Don and SiN Don , to form a plan for weakening Frgl Don within the donor substrate. The embrittlement plane is substantially parallel to the top surface.Don , as illustrated in (B3). The volume of the donor substrate between the upper surface Top Don The embrittlement plan Frgl defines a thin layer TL which will subsequently be transferred to the host substrate and then, if necessary, to a receiving substrate. The thin layer TL will be detached from the rest of the donor substrate Waf Don by Smart Cut technology TM .

[0067] The introduction of ions can correspond to the implantation of hydrogen and / or helium, that is to say, an ion bombardment of the upper surface. Don The donor substrate is coated with hydrogen and / or helium ions. Generally, the type, dose, and energy of the implanted species are chosen based on the desired thickness of the layer to be transferred and the physicochemical properties of the donor substrate. These parameters can be adjusted to define the thickness of the thin film (TL), for example, between 5 nm and 2 µm. The Smart Cut processTM allows obtaining a thin TL film with high crystallinity, exhibiting both high thickness uniformity and low roughness.

[0068] Assembly and finalization of the SUB substrate

[0069] After the preparation of the donor substrate (step 12) and the carrier substrate (steps 22 and 24), they are joined together in a bonding step 30: the silicon nitride layers covering the upper surfaces of the Waf donor substrate Don and the Waf carrier substrate Car are brought into intimate contact and undergo molecular adhesion and / or electrostatic bonding between them, as mentioned, for example, in French patent application No. 2,914,492. This is referred to as direct nitride-nitride bonding between the two silicon nitride (SiN) layers. Car and SiN Don, but this type of bonding includes the presence of water and / or hydroxide ions at the interface between the two nitride layers.

[0070] In this embodiment, the surfaces in close contact can be prepared by cleaning, brushing, drying, polishing, humidification, or plasma activation, as is known in the field. In particular, the presence of water and / or hydroxide ions on the surfaces of one or both SiN nitride layers can be promoted. Car and SiN Don by means of known processes, such as brushing accompanied by the application of an ammonia, alkaline, or fluoride solution. However, it is preferable to avoid subjecting the two surfaces to be joined to a plasma, so as to prevent promoting the transformation of SiN into SiO2 by the action of water. The resulting assembly comprises the following elements, stacked in this order: carrier substrate Waf Car ThIns thermal insulation layer Car, silicon nitride layer SiN Car , silicon nitride layer SiN Don ThIns thermal insulation layer Don , and Waf donor substrate Don As illustrated by (C1), each element of this stack is in direct contact with its immediate neighbor(s). One or both of the thermal insulation layers could be omitted.

[0071] As is known, during a molecular adhesion process, the exposed surfaces of the insulating layer and the donor substrate, which are perfectly clean, flat, and smooth, are brought into intimate contact to promote electrostatic bonding or the development of molecular bonds, such as van der Waals or covalent bonds. The two bodies are then joined without the use of an adhesive. The bonding process may involve the application of a low-temperature heat treatment (e.g., between 50 and 300°C, typically 100°C) to heal crystalline defects in the donor substrate and sufficiently increase the bonding energy to allow for a possible subsequent thinning step or other processing.

[0072] At a step 32 following step 30, a fractionation process is carried out to divide the Waf donor substrate Don at the level of the Frgl weakening plane so as to leave the TL layer fixed to the Waf carrier substrateCar The structure obtained after the split is the semiconductor substrate SUB illustrated by (C2).

[0073] More specifically, after the 30th bonding step, the TL layer is detached from the rest of the donor substrate. WafDon by fracture at the level of the weakening plane Frgl and is therefore transferred to the load-bearing substrate Waf Car This detachment step may involve applying heat treatment at a temperature between 350°C and 1000°C to detach the TL layer from the donor substrate and complete its transfer to the Waf carrier substrate. Car As an alternative or complement to heat treatment, this step may consist of applying a blade or jet of gaseous or liquid fluid, or any other mechanical force to the embrittlement plane Frgl.

[0074] After the detachment step, which produces the SUB structure illustrated by (C2), a stabilizing heat treatment can optionally be applied to the SUB substrate in a post-fragmentation step. This stabilizing heat treatment heals crystalline defects in the TL thin film and helps consolidate the bond between the TL thin film and the carrier substrate, for example, by heating the Sub substrate to a temperature between 300°C and 800°C for a period of 30 minutes to 10 hours. This heat treatment is preferably carried out by exposing the free face of the TL thin film to a neutral gaseous atmosphere.

[0075] In step 34 following step 32, a smoothing step of the free surface of the TL layer is performed, for example by CMP (Chemical Mechanical Polishing) and / or wet etching. Figures C3 and C1 illustrate the SUB substrate after step 34. At this stage, the SUB semiconductor substrate can be ready for any processing intended by the user of the SUB substrate. In particular, heat treatment at a temperature above 700°C can be considered because the titanium nitride layers are insulated from oxygen by the silicon nitride layers.

[0076] Alternatively, the semiconductor substrate can be shipped to the user without the smoothing step having been performed, allowing the user to carry out a smoothing step of their choice. This is because the SUB semiconductor substrate is intended for use by various parties, and the manufacturing unit where it was produced will generally differ from the manufacturing unit where the active material layer will be processed to form a functional semiconductor device or a component thereof.

[0077] Furthermore, instead of creating a weakening plane and fracturing the donor substrate at this plane, the donor substrate can be thinned after assembly with the carrier substrate, for example by etching, grinding, and / or chemical-mechanical polishing (CMP). This thinning can be used to remove the donor substrate entirely or only partially, for example, to achieve the situation illustrated by (C2).

[0078] It should be noted that the highest temperatures in the temperature ranges mentioned above are only accessible thanks to the particular structure of the SUB substrate, trapping species present in their gaseous form (such as dihydrogen) and preventing their escape even in the case of high-temperature treatments, so that a subsequent transfer step by light irradiation can be safely carried out.

[0079] In the SUB substrate, the interface between the two layers of silicon nitride SiN Don and SiN Car can be detected by transmission electron microscopy (TEM) or by scanning transmission electron microscopy (STEM) applied to a cross-section of the SUB substrate.

[0080] In addition, this interface includes chemical species containing hydrogen, so it can also be detected by physico-chemical characterization: in a cross-section of a SUB substrate, peaks of chemical bond concentrations such as OH or Si-H bonds, representative of the presence of the interface and its hydrogen content at the interface, can be detected, for example by electron energy loss spectroscopy (EELS), energy-dispersive X-ray spectroscopy (EEDS), Fourier-transform infrared spectroscopy (FTIR) or secondary ion mass spectroscopy (SIMS).

[0081] Example of using the SUB substrate

[0082] Laillustrates the steps of a PR2 manufacturing process consisting of transferring a TL layer from the SUB substrate to the Waf receiving substrate by laser detachment Rec .

[0083] In this example, the TL layer is made of monocrystalline silicon, obtained from the Waf donor substrate Don which was a single-crystal silicon wafer. The TL layer is suitable for applications in fields such as SRAM (static random access memory), HBM (high bandwidth memory) or logic circuits, and is compatible with FEOL (Front End Of Line) processes.

[0084] At step 100, the Waf receptor substrate Rec is provided. The substrate WafRec It can be any type of flat substrate with a smooth surface, such as substrates commonly used in the semiconductor industry: Si, SiC, sapphire, glass, AlN, or any combination of these materials. The Waf receiving substrate Rechas a top surface Rec , as illustrated in (A1). The receiving substrate must be compatible with a bonding process for the transfer of the TL layer and with the intended use of a device to be fabricated from the TL layer.

[0085] At step 200, the SUB substrate obtained by the PR1 manufacturing process is supplied, as illustrated in (B1).

[0086] In a step 202 following step 200, the SUB substrate undergoes a manufacturing process that includes modification, treatment, or structuring of the TL layer, and possibly the addition of other elements such as electrical traces, doping zones, or electrically insulating layers. Step 202 may include heat treatment of the SUB substrate at a temperature exceeding 700°C, 900°C, 1100°C, or 1200°C.

[0087] At step 300 following steps 100 and 202, the thin TL layer is bonded to the top surface.Rec of the Waf receptor substrate Rec as illustrated in Figure (C1). The same process as in step 30 of direct bonding can be implemented. Although not shown in Figure 1, layers (monolayer or multilayer) promote bonding between the TL thin film and the Waf receiving substrate. Rec may eventually be formed on one, the other, or both of the SUB substrate and the Waf support Rec For example, in the case of oxide-oxide bonding, an oxide layer can be deposited on the thin layer TL and an oxide layer can be formed on the top surface. Rec of the Waf receptor substrate Rec .

[0088] At a step 302 following step 300, light irradiates the separation layer Sep, here formed of silicon nitride layers SiN Don and SiN Car , through the Waf carrier substrate Caras illustrated by Irr light irradiation in the. Irr light irradiation can be emitted by a laser source at a wavelength in the ultraviolet and absorbed by the silicon nitride of the Sep layer, for example around 250 nm, after passing through the Waf carrier substrate Car formed from a plate of glass or sapphire.

[0089] By absorbing the energy from the Irr light irradiation, the silicon nitride layers are locally heated and degrade. This action, combined with the formation of hydrogen bubbles and their expansion due to heating, releases the thin film TL. Through mechanical traction Trac, the carrier substrate Waf Car is far from the Waf receptor substrate Rec , leaving the thin TL layer bound to the Waf receptor substrate Rec .

[0090] ThIns thermal insulation layers Don and ThIns CarThese layers help limit the dispersion of heat generated by light irradiation absorbed by the nitride layers, facilitating heating in the space between these thermal insulation layers. The ThIns layer Don limits the heating of the TL layer, preventing or limiting its degradation under the effect of the generated heat.

[0091] La(C2) illustrates the result of the separation of the Waf carrier substrate Car and the Waf receptor substrate Rec at step 302: irregular parts of the Sep separation layer and possibly the ThIns layer Car and the Waf carrier substrate Car , may remain as R remnants, and it is generally preferable to clean these R remnants, so as to expose and access the thin TL layer.

[0092] Thus, at a step 304 following step 302, the TL layer is cleaned of residues R from the remaining layers above it, for example by CMP (Chemical Mechanical Polishing) and / or wet etching. Figure (C3) illustrates the Waf receiving substrate Rec at the end of step 304.

[0093] Second method of implementing the invention

[0094] A second embodiment of the invention is illustrated by Figures 3 and 4 and is very similar to the first embodiment.

[0095] In the first implementation, illustrated in Figures 1 and 2, the SEP separation layer is formed from two layers of silicon nitride SiN Car and SiN Don capable of absorbing a wavelength of light located in the ultraviolet spectrum. The two functions of light absorption and separation are fulfilled by each of the two silicon nitride (SiN) layers. Car and SiNDon During the PR2 process, the irradiation of the SEP separation layer is done through the WafCar carrier substrate, requiring it to be transparent at this wavelength.

[0096] In the second implementation, the SEP separation layer comprises, in addition to the two silicon nitride layers SiN Car and SiN Don , an ABS absorption layer is formed adjacent to and is preferably in direct contact with one of the two silicon nitride (SiN) layers Car and SiN Don .

[0097] Laillustrates a PR1' process of the second implementation method, which is a variant of the PR1 process of the first implementation method. Except for the points specified below, reference may be made to the text describing the PR1 process.

[0098] In the PR1' process, during step 22, an ABS layer Don light absorption is formed on the Waf donor substrate Don , between the ThIns layersDon and SiN Don In this implementation, the SEP separation layer is therefore formed from the three ABS layers. Don ThIns Don and SiN Don The two functions necessary for the transfer of the TL layer by light irradiation are distributed between different elements: the ABS layer Don is designed to absorb incident light radiation and produce heating in response, the SiN layer Don is designed to degrade under the effect of the heat produced by the ABS layer Don , in addition to its hydrogen-trapping function. The ThIns layer Don serves to confine heat within the SiN layer D on .

[0099] The presence of a layer specifically dedicated to light absorption allows the wavelength to be chosen for irradiation by selecting the material forming the ABS layer. DonThis also allows for greater flexibility in choosing the material for the Waf load-bearing substrate. Car : it no longer necessarily has to be transparent to UV light (in the case of irradiation through substrate).

[0100] For example, the ABS absorbent layer Don can be formed from titanium nitride TiN or silicon dioxide SiO2, absorbing light wavelengths close to 1.9 µm and 10 µm respectively, and the carrier substrate Waf Car can be made of monocrystalline silicon, transparent to these wavelengths. Thus, the ABS absorbing layer DonIt can be formed from any type of material that absorbs a given wavelength of light (silicon dioxide, titanium nitride, silicon nitride, germanium nitride, etc.), provided that the substrate material is transparent to that wavelength and that the absorbing layer generates sufficient heat for its own degradation or that of the nitride layer. Any known method can be used to form the ABS absorbing layer. Don , which can for example be formed by plasma-enhanced chemical vapor deposition, or PE-CVD for Plasma Enhanced Chemical Vapor Deposition in English terminology, over a thickness of 10 nm to 200 nm.

[0101] A layer can be considered absorbing when it absorbs at least 30%, preferably at least 50%, of the radiation that reaches it.

[0102] La(C3) illustrates the SUB substrate obtained by the PR1' process, which comprises the following elements, stacked in this order: Waf carrier substrate Car ThIns thermal insulation layer Car , silicon nitride layer SiN Car , silicon nitride layer SiN Don ABS light-absorbing layer Don ThIns thermal insulation layer Don and TL layer, each element of this stack being in direct contact with its immediate neighbor(s). Any one or both of the thermal insulation layers could be omitted.

[0103] Laillustrates a PR2' process of the second implementation method, which is a variant of the PR2 process of the first implementation method. Except for the points specified below, reference may be made to the text describing the PR1 process.

[0104] In the PR2' process, during step 302, the purpose of Irr light irradiation is to heat the ABS light absorption layer Don , which is designed to absorb the wavelength of the irradiation used, which can be 1.9 µm when the ABS layer Don is made of titanium nitride (TiN) and the carrier substrate is Waf Car is made of monocrystalline silicon. This heating propagates to the SiN layers Car and SiN Don which degrade, as explained for the PR2 process.

[0105] The result of the PR2' process, illustrated by (C3), is the same as that of the PR2 process.

[0106] An alternative to the PR2' process would be to form the ABS layer Don on the WafCar carrier substrate, preferably in direct contact with the SiNCar layer, between the ThInsCar and SiNCar layers.

[0107] A third embodiment of the invention is illustrated by the and is very similar to the first and second embodiments.

[0108] Laillustrates a PR1''-PR2'' process of the third implementation, which includes variants of the PR1 processes of the first and second implementations. Except for the points specified below, reference may be made to the text describing the processes of these implementations.

[0109] Laillustre provides an application example that incorporates the PR1, PR2, PR processes from Figures 1 and 2, applied to the specific case where the Waf substrate Car is made of silicon, preferably single-crystal, and the ThIns layer Don is composed of silicon oxide. Unless otherwise specified, for the elements indicated in the, reference may be made to the descriptions of the elements with the same identifiers in figures 1 to 4.

[0110] In this application example, the light irradiation, which in this example must pass through the Waf substrate Car as illustrated by the, must be of the wavelength chosen to pass through the Waf substrate Car and, preferably, to be absorbed by the ThIns layer Don which then fulfills the function of the ABS layer Don of the PR2' process illustrated by the figure. Thus, a wavelength in the infrared, on the order of 10 µm, for example between 9 and 11 µm, can be used, possibly for light irradiation emitted by a CO2 laser, advantageous in that its wavelength can be adjusted within this range of wavelengths and in that it is powerful enough to provide adequate energy to the ThIns layer. Don to generate the desired heat.

[0111] In this example, the ThIns layer Donis formed from a layer of silicon dioxide obtained, for example, by thermal oxidation of a silicon substrate Waf Don , or by a sol-gel process based on TEOS. This layer can fulfill both the function of thermal insulation to protect the transfer layer TL and the function of the ABS layer. Don light absorption in the PR1' and PR2' processes. To this end, it is preferable for the layer to be sufficiently thick, for example between 200 nm and 1000 nm, preferably between 400 nm and 700 nm. These ranges allow both sufficient absorption of incident light irradiation and good thermal protection of the TL layer, while remaining technologically and economically compatible with the other steps of the process.

[0112] The portion of the ThIns layer Don close to the interface between ThIns Don and SiN Don(approximately the first 100 nanometers) will absorb a high percentage of the energy from the incident radiation, advantageously promoting separation at the SEP layer stack. As one approaches the interface between ThIns Don and TL, the percentage of the amount of energy of the incident radiation absorbed decreases, and becomes negligible in the last nanometers of the ThIns layer Don close to the TL layer, advantageously protecting the TL layer from excessive heating. Therefore, the ThIns layer Don Silicon oxide advantageously plays the dual role of an absorption layer and a thermal protection layer for the TL layer.

[0113] SiN layers Don and SiN Car Each can independently have a thickness between 3 and 20 nm. Unlike the PR1 and PR2 processes, the ThIns layer Caris not formed on the, but it would be possible to form it, as previously described in the PR1 and PR2 process.

[0114] In this document, the layer designated as the transfer layer is a layer that can be intended to be separated from its carrier substrate after being fixed to a receiving substrate, to be separated from its carrier substrate and then fixed to a receiving substrate, to be separated from its carrier substrate and handled as an independent object, for example, to be transferred to another location or to be integrated into a receiving device, or to be separated from its carrier substrate and used in any manner of use known or unknown to a person skilled in the art.

[0115] The invention is not limited to the embodiments described above and variations thereof may be made without departing from the scope of the invention as defined by the claims.

[0116] Specifically, the implementation methods described each employ Irr irradiation to separate the transfer layer TL from its carrier substrate Waf Car at which is achieved through this Waf carrier substrate Car These modes are advantageous because they avoid or limit irradiation of the transfer layer TL, but irradiation through the carrier substrate Waf. Car but through the receiving substrate Waf Rec is also conceivable, meaning that the irradiation source can be located on the receiving substrate side of the SEP layer designed to absorb this irradiation. More generally, the invention is not limited to processes in which the Irr irradiation is carried out through the Waf carrier substrate. Car , and therefore it is not necessarily transparent to the wavelength of Irr irradiation.

Claims

A process (PR1; PR1') for manufacturing a substrate (SUB) comprising a transfer layer (TL) and designed to allow separation of the transfer layer (TL) by light irradiation, process (PR1) comprising the steps of: - forming (12) a first silicon nitride (SiN) layer Car ) on a load-bearing substrate (Waf Car ) ;- form (22) a second layer of silicon nitride (SiN Don ) on a donor substrate (Waf Don ) ;- assemble (30) the supporting substrate (Waf Car ) to the donor substrate (Waf Don ) by direct bonding of the first layer of silicon nitride (SiN Car ) to the second silicon nitride layer (SiN Don ) ;- form (32, 34) the layer to be transferred (TL) by eliminating a part (Waf' Don ) of the donor substrate (Waf Don leaving the transfer layer (TL) attached to the carrier substrate via the first silicon nitride (SiN) layer Car) and the second silicon nitride layer (SiN Don ), in which the first layer of silicon nitride (SiN Car ) and the second silicon nitride layer (SiN Don ) are each formed in such a way as to have a density greater than 1.8 g / cm³ 3 or a density greater than 2 g / cm³ 3 . A method according to claim 1, wherein the first layer of silicon nitride (SiN) Car ) and the second silicon nitride layer (SiN Don ) are each formed in such a way as to have an optical index greater than 1.75 or greater than 1.8 for a wavelength of 550 nm. A method (PR1; PR1') according to claim 1 or 2, wherein the substrate (SUB) is provided to allow separation of the transfer layer (TL) by light irradiation at a given wavelength, and wherein the carrier substrate (Waf Car ) is substantially transparent at the given wavelength. A method (PR1; PR1') according to any one of claims 1 to 3, further comprising the formation of at least one of: - a first layer of thermal insulation (ThIns Car ) on the supporting substrate (Waf Car ), so as to form a stack comprising the first layer of thermal insulation (ThIns Car ) interposed between the supporting substrate (Waf Car ) and the first layer of silicon nitride (SiN Car ); and a second layer of thermal insulation (ThIns Don ) on the donor substrate (Waf Don ), so as to form a stack including the second layer of thermal insulation (ThIns Don ) interposed between the donor substrate (Waf Don ) and the second silicon nitride layer (SiN Don ). Method (PR1; PR1') according to claim 4, wherein the first thermal insulation layer and the second thermal insulation layer are each independently formed of a silicon oxide layer with a thickness between 10 nm and 1000 nm. A method (PR1') according to any one of claims 1 to 5, further comprising a step of forming a layer (ABS Don ) of light absorption in direct contact with one of the first silicon nitride (SiN) layers Car ) and the second silicon nitride layer (SiN Don ). Method (PR1') according to claim 6, wherein the layer (ABS Don ) of light absorption is formed from a material chosen from silicon dioxide, titanium nitride, silicon nitride and germanium nitride. Substrate (SUB) comprising a transfer layer (TL) and designed to allow separation of the transfer layer (TL) by light irradiation, the substrate comprising a stack including, in this order: - a carrier substrate (Waf Car ) ;- a first layer of silicon nitride (SiN Car ) ;- a second layer of silicon nitride (SiN Don ) in direct contact with the first silicon nitride layer (SiN Car ) ; and- the transfer layer (TL), in which the first silicon nitride (SiN) layer Car ) and the second silicon nitride layer (SiN Don ) each have a density greater than 1.8 g / cm³ 3 or a density greater than 2 g / cm³ 3 . The substrate according to claim 8, wherein the first layer of silicon nitride (SiN Car ) and the second silicon nitride layer (SiN Don) each have an optical index greater than 1.75 or greater than 1.8 for a wavelength of 550 nm. The substrate according to claim 8 or 9, wherein the substrate (SUB) is provided to allow separation of the transfer layer (TL) by light irradiation at a given wavelength, and wherein the carrier substrate (Waf Car ) is substantially transparent at the given wavelength. The substrate according to any one of claims 8 to 10, further comprising at least one of: - a first thermal insulation layer (ThIns Car ) interposed between the supporting substrate (Waf Car ) and the first layer of silicon nitride (SiN Car ); and a second layer of thermal insulation (ThIns Don ) interposed between the second layer of silicon nitride (SiN Don ) and the transfer layer (TL), the first thermal insulation layer (ThIns Car) and the second layer of thermal insulation (ThIns Don ) each being formed of silicon oxide. The substrate according to any one of claims 8 to 11, further comprising a light-absorbing layer (ABS) in direct contact with one of the first silicon nitride (SiN) layers Car ) and the second silicon nitride layer (SiN Don ). The substrate according to claim 12, in which the light-absorbing layer (ABS) is formed of a material selected from silicon dioxide, titanium nitride, silicon nitride and germanium nitride. The substrate (SUB) according to any one of claims 8 to 13, wherein: - the carrier substrate (Waf Car ) is substantially transparent to light of a given wavelength, and - at least one of the first silicon nitride (SiN) layer Car ), of the second silicon nitride layer (SiN Don) and, where applicable, the light-absorbing layer (ABS) when the substrate, according to any one of claims 10 and 11, is substantially absorbent to light of the given wavelength, so that the layer (TL) to be transferred can be separated from the carrier substrate (Waf Car ) when that of the first silicon nitride layer (SiN Car ), of the second silicon nitride layer (SiN Don ) and, where applicable, of the light-absorbing (ABS) layer when the substrate, according to one of claims 10 and 11, is irradiated by light of the given wavelength through the carrier substrate (Waf Car ) during a light irradiation separation operation. The substrate (SUB) according to claim 11, comprising the second thermal insulation layer (ThIns Don ) of silicon oxide interposed between the second layer of silicon nitride (SiN Don) and the layer to be transferred (TL), the carrier substrate (Waf Car ) being made of silicon. A method (PR2; PR2') for forming and separating a transfer layer, comprising supplying (200) a substrate (SUB) comprising the transfer layer (TL) according to any one of claims 8 to 15, and further comprising the steps of: assembling (300) the transfer layer to a receiving substrate (Waf Rec ) so as to form an assembly (ENS), the layer to be transferred being interposed between the carrier substrate (Waf Car ) and the receiving substrate (Waf Rec ) ; and- irradiate (302) the whole (ENS) by means of light irradiation so as to separate the layer to be transferred (TL) from the carrier substrate (Waf Car ). A method according to claim 16, claim 5 and claim 6, comprising the formation of the second thermal insulation layer (ThIns Don) of silicon oxide, on the donor substrate (Waf Don ), the supporting substrate (Waf Car ) being made of silicon, and the light irradiation taking place through the carrier substrate (Waf Car ) and using a wavelength between 9 and 11 µm. Method (PR2; PR2') according to claim 16 or 17, further comprising a preparation step (202) of the layer to be transferred including a heat treatment at a temperature above 700°C or above 1000°C before the step of irradiating (302) the assembly (ENS).

Images