Composite structure designed for planar co-integration of electronic components with different functions

Description

[0001] The present invention relates to advanced substrates for the fields of microelectronics, photonics, and power electronics. In particular, the present invention concerns a composite structure, intended for the planar cointegration of electronic components with different functions, and its manufacturing process.

[0002] To improve the performance of electronic devices, the manufacturing technology for CMOS (Complementary Metal Oxide Semiconductor) electronic components has gradually shifted away from the use of bulk substrates towards thin-film substrates, achieved via a buried oxide layer. This allows for better electrostatic control of the MOS transistor channel by preventing leakage currents. However, this FDSOI (Fully Depleted Silicon On Insulator) technology is limited when multiple transistors are to be placed on the same substrate. Shallow Trench Isolation (STI) trenches must be created in the substrate between each transistor. Furthermore, when components with different functions are required, such as a vertical conduction transistor and an RF (Radio Frequency) sensor on the same substrate, the requirements for the integration substrate properties differ.The transistor will need a conductive substrate while the RF sensor will need an electrically insulating substrate, exhibiting good thermal conductivity and capable of limiting RF losses.

[0003] To address these constraints, a known method involves using discrete components on a dedicated substrate. Each component (RF and transistors) is then fabricated on a specific and separate substrate. Each component is then cut out, removed, transferred to a substrate, and connected, a process that is lengthy and expensive.

[0004] Reference is also made to the following documents: US 2017 / 062357 A1 (KAMGAING TELESHPOR [US] ET AL) March 2, 2017; US 2011 / 073236 A1 (LEE SANG IN [US]) March 31, 2011; US 2020 / 365532 A1 (MALE BARRY JON [US] ET AL) November 19, 2020.

[0005] One of the aims of the present invention is to overcome at least one of these drawbacks. The invention is defined by the technical features disclosed in the independent claims.

[0006] To this end, the present invention proposes a composite structure, intended for planar cointegration of electronic components with different functions, the composite structure comprising from its base to its surface: a support substrate in a first material, the support substrate having cavities each opening into an upper face of the support substrate, the cavities being filled by at least one composite material consisting of a matrix of a cross-linked pre-ceramic polymer, the matrix being loaded with inorganic particles, a thin film in a second material, the thin film being bonded to the upper face of the support substrate and to the composite material.

[0007] Thus configured, the composite structure of the present invention features a support substrate comprising alternating regions made of different materials with varying properties. The support substrate includes regions containing a composite material whose particles allow for the modulation of its properties, particularly its electrical conductivity. Furthermore, the underlying thin film enables the integration of several electronic components with different requirements and operating functions, depending on whether they are positioned above the primary material or the composite material. This optimizes the performance of each component and reduces manufacturing time compared to prior art methods where different components are manufactured on separate substrates and then transferred to a single support.Furthermore, choosing a composite material made from a preceramic polymer ensures that the structure will not disintegrate or decompose during subsequent manufacturing or high-temperature use. This is because, after the crosslinking treatment, the shaped preceramic polymer becomes infusible.

[0008] Advantageously, the cavities of the same substrate can be filled with different types of composite materials, both in terms of the preceramic polymer and the nature and combination of inorganic particles. Thus, the composite structure can be entirely designed by adapting the properties of the composite material within the cavities according to the requirements of the components to be integrated.

[0009] According to one provision, the cavities have a depth greater than or equal to 10 micrometers, in particular greater than or equal to 50 micrometers and for example about 100 micrometers.

[0010] This depth advantageously eliminates the effect of the substrate located directly beneath the cavities. These dimensions are far greater than those obtained with conventional microelectronic technologies (CVD, PVD evaporation, etc.) where thicknesses are generally less than 5 µm, making the composite structure of the present invention more efficient.

[0011] According to one possibility, the cavities have a width ranging from 100 micrometers to 1 mm. These dimensions are achievable using standard microelectronic technologies.

[0012] Advantageously, the cavities have a variable overall shape, such as a polygon, a cylinder... they can have a circular cross-section of variable dimensions, etc., the shape of the cavity being configured to make it possible to fill it from a liquid or solid filling formulation using conventional techniques.

[0013] According to one provision, the preceramic polymer is a silicon-based polymer. The preceramic polymer is preferably selected from the group comprising a polysiloxane, a polysilsesquioxane, a polycarbosiloxane, a polycarbosilane, a polysilylcarbodiimide, a polysilsesquicarbodiimide, a polysilsesquiazane, a polysilazane, a polyborosilazane, a polyborosilazane, a polyborosiloxane, and a combination of these polymers. The choice of the specific preceramic polymer depends on the desired properties of the composite structure and its compatibility with the substrate onto which it is deposited. It is also chosen so that, once filled and crosslinked, the resulting composite material exhibits a coefficient of thermal expansion close to that of the thin film, preferably composed of silicon.

[0014] Preceramic polymers exhibit excellent temperature resistance (for example, in adhesive form, they withstand temperatures exceeding 350°C). This allows the transferred thin film to undergo high-temperature heat treatments, such as the repair of defects generated during implantation in the donor substrate. Indeed, a preceramic polymer is an organic / inorganic polymer generally used to create ceramic objects, known as PDCs (Polymer-Derived Ceramics), after high-temperature heat treatment.

[0015] Such a preceramic polymer can be applied in various forms and can therefore easily be used to fill cavities of the aforementioned dimensions. It is available in liquid form or is soluble in organic solvents and can be deposited onto a substrate via liquid, solid, or vapor phase.

[0016] It is possible to use all polymer shaping techniques such as spin coating, dip coating, spraying or tape casting, screen printing, inkjet printing.

[0017] Depending on one possibility, the matrix is ​​loaded with one or more types of inorganic particles, chosen from: a thermal conductor, such as SiC, an electrical insulator, such as AIN, Si3N4, crystalline Al2O3 or BN, an electrically conductive metal, such as Cu, Ag, Au or Al, a magnetic material, such as NdFeB, ZnMn or NiZn ferrite, a material with optical properties, notably CdSe in the form of 2-8 nm nanoparticles, so as to obtain fluorescence properties or graphite for reflective properties, and a combination of these different types of inorganic particles.

[0018] This wide range of particle options allows access to a large number of composite materials with highly varied properties. This enables excellent adaptation to the needs of different electronic components to be integrated. Indeed, the choice of these inorganic particles allows the composite material to be given thermal conductivity, electrical insulation, magnetic properties, and / or optical properties that differ from those of the preceramic polymer and the initial material.

[0019] It is also possible to use matrices containing several sizes and shapes of inorganic particles to promote percolation pathways.

[0020] Preferably, the matrix is ​​obtained from a filling formulation, comprising between 20% to 50% by volume of preceramic polymer and a loading rate of inorganic particles ranging from 50% to 80% by volume relative to the total volume of the filling formulation.

[0021] In the case of inorganic SiC particles (density 3.21), the SiC mass ratio varies from 71% to 91% for a preceramic polymer weight ranging from 9% to 29%. These particles are preferably mixed with the matrix so that the filler formulation has a sufficient filler content to achieve the desired properties and to obtain an intact layer depending on the deposition thickness and working temperature.

[0022] In one scenario, the thin film is bonded to the top surface of the substrate and to the composite material via an adhesion primer. The adhesion primer is thus positioned between the top surface of the substrate (and an exposed area of ​​the composite material filling the cavities) and the thin film.

[0023] The adhesion primer of the present invention is a preceramic polymer with a viscosity adapted to the shaping method. It does not degrade or decompose at high temperatures and is compatible with use in the manufacture of electronic devices. A solvent can be added to modulate the viscosity. It can be deposited by inkjet or screen printing. When the preceramic polymer is solid (initially or partially crosslinked), it exhibits thermoplastic behavior. The adhesion primer can thus be softened, at least on the surface, which allows the thin film to bond to the donor substrate.

[0024] Advantageously, the bonding primer used is of the same type as the preceramic polymer used for the matrix. This bonding primer therefore has the advantage of being compatible with the matrix and promoting good adhesion. It is chosen from among the polysiloxanes, which are advantageously compatible with the surface of a silicon substrate and at high temperatures since they transform into ceramics.

[0025] Preferably, the bonding primer is chosen from SILRES ®< H62 C (phenyl silicone resin) and SILRES ®< MK (methyl silicone resin) available from the supplier Wacker Chemie AG.

[0026] According to another aspect, the present invention proposes a planar cointegrating microelectronic system comprising: a composite structure as previously described, comprising a plurality of electronic components with different functions on the thin film, including: o passive electronic components, such as RF components, for example sensors, formed on first regions of the exposed surface of the thin film located above the composite material filling the cavities of the support substrate, o active electronic components, such as transistors associated with the RF components, formed on second regions of the exposed surface of the thin film located above the first material of the support substrate, the inorganic particles of the matrix are made of SiC and / or Si 3 N 4 to give the composite material filling the cavities a thermally conductive and electrically insulating character.

[0027] Thus, the composite material is made from a material compatible with the thin film in terms of its coefficient of thermal expansion (CTE) (this property also depending on the deposition thickness) and whose electrical and thermal properties can be adjusted according to the added load. This will allow for the consideration of technological operations involving significant thermal budgets for the resulting structure, including temperatures that can reach up to 1000°C or even higher.

[0028] The composite material can be different depending on the cavities filled, so the integration of other electronic components with different functions is conceivable on the composite structure.

[0029] According to other features, the structure of the present invention comprises one or more of the following optional features considered alone or in combination: The first material, i.e., the substrate material, is a semiconductor material, such as silicon. The first material is monocrystalline silicon. The first material is polycrystalline or amorphous silicon. The first material comprises SiC, Ge, InP, and / or GaAs. The substrate has a thickness between 10 micrometers and 1 mm. The substrate is a standard silicon wafer. It typically has a thickness of 725 micrometers and a diameter of 200 mm. The depth of each cavity is strictly less than the thickness of the substrate. The donor substrate or thin film is made from a second single-crystal, semiconductor material, such as silicon, SiC, Ge, InP, or GaAs, or one exhibiting piezoelectric properties, such as PZT, BaTiO3, LiNbO3, quartz, or LiTaO3. The crystalline quality of the support substrate is lower than that of the thin film.The thin film thickness ranges from 10 nm to 2 micrometers when obtained by creating a weakened plane in a donor substrate, followed by fracture. The thin film thickness is at least 100 micrometers when obtained by grinding a thicker donor substrate (a term commonly used by those skilled in the art to refer to grinding or rectifying). The second material in the thin film serves as a seed layer for epitaxial growth during the partial or complete fabrication of components. The inorganic particles loaded into the matrix have diameters ranging from nanometers to several micrometers. The thin film is bonded by direct bonding, i.e., molecular adhesion, to both the support substrate and the composite material. Layers of SixOy and / or SixNy are deposited at the bonding interface to facilitate direct bonding.These layers have a thickness between 0.1 and 5 micrometers. The thin film is bonded to the substrate and composite material using a metallic bonding process. Metallic layers of Cu, Ag, and / or W are deposited at the bonding interface to promote metallic bonding. The thin film is bonded to the substrate and composite material using a hybrid bonding process. The thin film is bonded to the substrate and composite material using SAB (ultra-high vacuum ion activation) bonding. Before contact for bonding, the surfaces undergo chemical treatment to remove residual particles and native oxides. In the case of direct bonding, plasma (O₂ or N₂ plasma), UV / ozone, and Megpie techniques are used.

[0030] According to another aspect, the invention proposes a method for manufacturing a composite structure intended for planar cointegration, the method comprising the steps of: a) provision of a support substrate in a first material, comprising a top face, the support substrate comprising cavities opening into the top face, b) filling of the cavities by a filling formulation comprising a matrix of a preceramic polymer, the matrix being loaded with inorganic particles, c) application of a crosslinking treatment so as to obtain a composite material consisting of the matrix of the crosslinked preceramic polymer, loaded with inorganic particles, d) bonding of a donor substrate in a second material to the top face of the support substrate and to the composite material, and e) thinning of the donor substrate to form a thin film.

[0031] Cavity filling with the filling compound is achieved using any standard microelectronic technique. This can be done by screen printing, inkjet printing, or any other method appropriate to the cavity dimensions. If the filling compound is in solid form, particularly as granular powders or granules, it can be dispensed into the cavities by extrusion using a syringe (a process similar to robocasting).

[0032] According to one provision, the crosslinking treatment includes a heat treatment.

[0033] When the filler formulation includes a solvent, step c) of applying a crosslinking treatment includes prior heat treatment for evaporation at a temperature lower than that of the heat treatment for crosslinking so as to allow evaporation of the solvent from the filler formulation.

[0034] In one embodiment, the application of the crosslinking treatment includes the application of a temperature ramp to carry out the evaporation heat treatment. The temperature ramp is implemented with or without steps depending on the nature and quantity of the solvent used, so as to evaporate all of the solvent.

[0035] According to one possibility, step d) of bonding is preceded by step i) of planarizing the upper surface of the support substrate and / or an exposed surface of the composite material, so that the exposed surface and the upper surface extend in a single plane and form a surface suitable for bonding.

[0036] Step i) of planarization includes a grinding and / or chemical polishing step to achieve a surface roughness of the desired type, depending on the bonding method. For example, the roughness will be less than 0.5 nm RMS for direct bonding.

[0037] According to one provision, step d) of bonding is preceded by step j) of deposition, which includes the formation of an adhesion primer on the upper surface of the support substrate and on the exposed surface of the composite material and / or on the donor substrate, so as to facilitate bonding. In this case, step i) of planarization is less demanding; it is even sometimes possible to omit it depending on the initial roughness of the surfaces.

[0038] The primer applied to the surface(s) before bonding is less than 10 micrometers thick. This allows for surface leveling without negatively impacting the properties of the composite structure.

[0039] According to one possibility, the process includes carrying out a step k) of preparing the upper surface between step b) and step c) comprising: the arrangement and maintenance of a covering element against the upper face of the support substrate, covering at least the cavities, the turning of the support substrate, so that the filling formulation comes by gravity to the surface of the upper face of the support substrate.

[0040] Once positioned, the matrix either rests against the inner wall of the covering element or remains in its initial position, flush with the plane of the upper surface of the substrate before curing, depending on the cavity filling. Under these conditions, shrinkage of the composite material due to solvent evaporation or curing does not affect the difference in level between the plane of the exposed surface of the composite material and that of the upper surface of the substrate.

[0041] It is understood in this document that the reversal leads to the inversion of the position of the upper face with that of the base of the supporting substrate.

[0042] Once the crosslinking has been completed according to step c), the substrate is returned to its initial position, with the upper surface facing upwards. The covering element is removed. The exposed surface of the composite material and the upper surface form a single plane, creating a surface suitable for bonding. Indeed, during the polymer crosslinking, interactions occurred between the matrix and the cavity surfaces, such that the composite material remains in the position it assumed at the time of crosslinking, even after any subsequent shrinkage.

[0043] According to one possibility, at least one through-hole is provided beforehand in the support substrate so as to allow the evaporation of any solvent from the composite formulation, from the cavity towards the base of the support substrate oriented upwards.

[0044] The covering element includes, for example, an adhesive film or a backing plate, held against the upper face of the supporting substrate.

[0045] The bonding surface thus formed is planarized according to step i) if necessary. Application of an adhesion primer according to step j) is also possible, either separately or in combination.

[0046] Other aspects, objects, and advantages of the present invention will become clearer upon reading the following description of the various embodiments thereof, given by way of non-limiting example and with reference to the accompanying drawings. The figures do not necessarily have to be to scale for all the elements shown in order to improve their legibility. In the remainder of the description, for the sake of simplicity, identical, similar, or equivalent elements of the different embodiments are referred to by the same numerals in which: [ Fig. 1 ] represents a step in the process according to a first embodiment of the composite structure 100 according to the present invention. Fig. 2 ] represents a step in the process according to the first embodiment of the present invention. Fig. 3 ] represents a step in the process according to the first embodiment of the present invention. Fig. 4 ] represents a step in the process according to the first embodiment of the present invention. Fig. 5 ] represents a step in the process according to the first embodiment of the present invention. Fig. 6 ] represents a step in the process according to the first embodiment of the present invention. Fig. 7 ] illustrates a step of the process according to a second embodiment of the composite structure 100 of the present invention. Fig. 8 ] illustrates a second step of the process according to the second embodiment of the present invention. Fig. 9 ] illustrates a third step of the process according to the second embodiment of the present invention.

[0047] As illustrated on the figure 1 A support substrate 1 made of a first semiconductor material, such as a silicon wafer 725 micrometers thick, is first provided according to step a) of the process. It comprises a plurality of cavities 2, each opening into the upper face 3 of the support substrate 1.

[0048] These cavities 2 were formed beforehand, for example by a process comprising a masking step involving the deposition of photolithography resin on the regions of the support substrate 1 to be preserved, followed by a deep etching step by RIE of the unmasked regions. Alternatively, the cavities 2 are obtained using a laser until a depth of approximately 100 micrometers and a cross-section of 1x1 mm² are reached.

[0049] As illustrated in the figure 2 The cavities 2 are then filled with a filling formulation comprising a matrix 4 of a preceramic polymer loaded with inorganic particles, according to step b) of the process. The filling of the cavities 2 is carried out while the upper surface 3 of the substrate 1 still contains the photolithography resin so as to protect the surface. In one variant, this protection is achieved by a sticky cutting film or a screen printing stencil.

[0050] According to one possibility, the filler formulation includes the polysiloxane polymer SILRES ®< H62 C. According to another arrangement, the polymer is SMP-10 polycarbosilane supplied by Starfire.

[0051] The inorganic particles consist of Si3N4 and / or AIN to impart thermal and electrical insulation properties to the composite material. The filler content is approximately 60% of the total formulation volume.

[0052] A solvent, such as xylene, butan-2-one or Diestone DLS, is added to matrix 4 to adjust the viscosity and obtain a liquid filler formulation.

[0053] The filling of cavities 2 by the filling formulation is then obtained according to step b) of the screen printing process.

[0054] A crosslinking treatment according to step c) of the process is carried out by applying a heat treatment at approximately 200°C when the polymer is SILRES® < H62 C. This heat treatment also contributes to solvent evaporation (visible figure 3 ). A composite material 5 is then obtained in the cavities 2.

[0055] In one scenario, the filling formulation contains no more than 30% solvent by weight to achieve excellent flowability. However, such a high solvent content results in significant shrinkage (or volume loss) of the composite material 5 once the solvent has evaporated and the preceramic polymer has cured. To achieve satisfactory filling of the cavities 2, a sequence involving heat treatment at a temperature lower than that of curing and filling the cavity with the filling formulation is repeated several times, so as to fill at least 95% of the cavity volume 2 with the composite material 5. This filling rate reduces the need for subsequent planarization or surface preparation steps for optimal bonding with the thin film.

[0056] According to one variant not shown, the filling formulation is solid. The preceramic polymer is SILRES®< MK, which is dissolved in a solvent and mixed with Si3N4 particles. The mixture is then dried to obtain a powder (filling formulation) used to fill cavities 2. A crosslinking heat treatment produces the composite material 5. According to another variant, the solid preceramic polymer is heated to its melting temperature (lower than its crosslinking temperature) and mixed with the inorganic particles. Once the formulation has cooled and solidified, it is ground and introduced into cavities 2.

[0057] As depicted on the figure 4 After removal of the photolithography resin or adhesive film, a bonding primer 6 is deposited on the upper surface 3 of the support substrate 1 and the exposed surface of the composite material 5 according to step j) of the process, in order to avoid very significant grinding / planarization steps of the surfaces before bonding. This bonding primer 6 consists of a pre-ceramic polymer with thermoplastic behavior, so that once deposited and heated to its softening temperature, it allows bonding to a donor substrate according to step d) of the process.

[0058] According to a possibility illustrated in the figure 5 The adhesion primer 6 is also deposited on the donor substrate 7 before step d) of bonding. Alternatively, it may be deposited only on this donor substrate and not on the upper surface of the support substrate 1.

[0059] According to another possibility not illustrated, the bonding according to step d) is carried out by the prior deposition of a thin layer of amorphous silicon bombarded by argon (SAB bonding).

[0060] According to yet another variant, the bonding does not use a primer 6. The face of the substrate 1 and an exposed surface of the composite material 5 are planarized by grinding and chemical-mechanical polishing (CMP) steps to form a single plane, creating a surface suitable for subsequent bonding. For example, in preparation for direct bonding, this surface is polished to a roughness of less than 0.5 micrometers RMS. The work-hardened area from grinding has been removed.

[0061] As seen at the figure 5 The donor substrate 7 has a weakening plane 8 obtained prior to bonding by implantation of ionic species according to Smart Cut® technology. Once bonding is complete, the thin film 9 is separated from the negative of the donor substrate 7 by applying a thermal fracture budget according to step e) of the process. This budget can be reduced by the simultaneous application of mechanical stress at the weakening plane 8.

[0062] Preferably the donor substrate 7 and / or the thin film 9 is made of a single-crystal material.

[0063] SixOy oxide layers are deposited by PEVCD onto the surfaces to be bonded, with a thickness of approximately 100 nm, for direct bonding of the support substrate 1 and the donor substrate 7. A CMP step is performed on the oxide layers to achieve the surface roughness required for direct bonding, for example, 0.3 micrometers RMS. After the surfaces are brought into contact, a bonding anneal is applied to enhance the bonding energy between the donor substrate 7 and the support substrate 1. This anneal advantageously provides the thermal budget necessary to fracture the donor substrate 7 at the level of the embrittlement plane 8 and releases the thin film 9 bonded to the support substrate 1. The surface of the thin film 9 is polished and cleaned for the fabrication of the electronic components 10, 11 ( figure 6 )

[0064] According to another alternative, the unweakened donor substrate 7 is thinned by a grinding step, which may be combined with chemical etching, until the desired thickness of the thin film 9 is reached. Then, a mechanochemical polishing is performed, notably to obtain a surface of the thin film 9 suitable for the epitaxial growth of a new material. The composite structure 100 thus obtained is capable of planar cointegration of different electronic components on regions with different properties depending on whether they are located on the composite material 5 or on the first material. (Microelectronic system 200 with planar cointegration on the figure 6 ).

[0065] According to a variant of the process illustrated in figures 7 à 9 , once the filling formulation is introduced into the cavities 2, a covering element 12, such as a backing plate, is positioned against the upper face 3 of the support substrate 1 so as to cover the cavities 2. The support substrate 1 is then inverted so that its upper face 3 is oriented downwards and its base is oriented upwards, the backing plate 12 holding the formulation in the cavity 2 ( figure 7 ). In this position, the filling formulation 4 comes into contact with the inner wall of the backing plate 12 so as to be flush with the plane of the upper face 3 of the support substrate 1. The crosslinking treatment is applied so that the polymer of the composite formulation is crosslinked in this position. Once the support substrate 1 is returned to its initial position, the exposed surface of the composite material 5 is flush with the plane of the upper face 3 ( figure 8 ). A single plane is thus formed, which limits the planarization or surface preparation steps before bonding ( figure 9 ).

[0066] According to a possibility illustrated in the figure 7 , through holes 13 are provided beforehand through the thickness of the support substrate 1 so as to allow the evaporation of the solvent of the filling formulation, if necessary, of the cavities 2 towards the outside of the support substrate 1. The counter plate 12 is then removed, a direct bonding with layers of oxide 14 is carried out so as to obtain the composite structure 100. Different electronic components are formed on the different regions of the thin film 9 depending on whether the regions are above the cavities 2 or of the first material.

[0067] The composite structure 100 of the present invention thus facilitates the co-integration of components with different functions while improving their performance thanks to an underlying substrate configured to exhibit the appropriate properties. These structures 100 make it possible to achieve more compact systems 200, requiring fewer manufacturing steps than those known to date.

Claims

1. A composite structure (100) intended for planar co-integration of electronic components with different functions, the composite structure (100) comprising, from its base to its surface: - a support substrate (1) made of a first semiconductor material, the support substrate (4) including cavities (2) each opening onto an upper face (3) of the support substrate (1), the cavities (2) being filled by at least one composite material (5) consisting of a matrix (4) of a crosslinked preceramic polymer, the matrix (4) being loaded with inorganic particles, - a thin film (9) made of a second semiconductor or piezoelectric material, the thin film (9) being bonded to the upper face (3) of the support substrate (1) and to the composite material (5).

2. The composite structure (100) according to claim 1, wherein the cavities (2) have a depth greater than or equal to 10 micrometers.

3. The composite structure (100) according to claim 1 or 2, wherein the preceramic polymer is a silicon-based polymer, and preferably selected from the group comprising a polysiloxane, a polysilsesquioxane, a polycarbosiloxane, a polycarbosilane, a polysilylcarbodiimide, a polysilsesquicarbodiimide, a polysilsesquiazane, a polysilazane, a polyborosilazane, a polyborosilane, a polyborosiloxane, and a combination of these polymers.

4. The composite structure (100) according to any of claims 1 to 3, wherein the matrix (4) is filled with one or more types of inorganic particles, selected from: - a thermal conductor, such as SiC, - an electrical insulator, such as AIN, Si3N4, crystalline Al2O3 or BN, - an electrically conductive metal, such as Cu, Ag, Au or Al, - a magnetic material, such as NdFeB, ZnMn ferrite or NiZn ferrite, - a material with optical properties, and a combination of these different types of inorganic particles.

5. The composite structure (100) according to any of claims 1 to 4, wherein the matrix (4) is obtained from a filling formulation, comprising between 20% and 50% by volume of preceramic polymer and a filler rate of inorganic particles ranging from 50 to 80% by volume relative to the total volume of the filling formulation.

6. The composite structure (100) according to any of claims 1 to 5, wherein the thin film (9) is bonded to the upper face (3) of the support substrate (1) and to the composite material (5) by means of an adhesion promoter (6) arranged on the upper face (3) of the support substrate (1) and on an exposed surface of the composite material (5) filling the cavities (2).

7. A microelectronic system (200) with planar co-integration comprising: - a composite structure (100) according to any of claims 1 to 6, - a plurality of electronic components with different functions on the thin film (9) comprising: ∘ passive electronic components (11), such as RF components, for example sensors, formed on first regions of the exposed surface of the thin film (9) located directly above the composite material (5) filling the cavities (2) of the support substrate (1), ∘ active electronic components (10), such as transistors associated with the RF components, formed on second regions of the exposed surface of the thin film (9) located directly above the first material of the support substrate (1), the inorganic particles of the matrix (4) are made of SiC and / or Si3N4 to impart a thermally conductive and electrically insulating character to the composite material (5) filling the cavities (2).

8. A method for manufacturing a composite structure (100) intended for planar co-integration, the method comprising the steps of: - a) providing a support substrate (1) made of a first material, comprising an upper face (3), the support substrate (1) comprising cavities (2) opening onto the upper face (3), - b) filling the cavities (2) with a filling formulation comprising a matrix (4) of a preceramic polymer, the matrix (4) being filled with inorganic particles, - c) applying a crosslinking treatment so as to obtain a composite material (5) consisting of the crosslinked preceramic polymer matrix (4), filled with inorganic particles, - d) bonding a donor substrate (7) made of a second material to the upper face (3) of the support substrate (1) and to the composite material (5), and - e) thinning the donor substrate (7) to form a thin film (9).

9. The manufacturing method according to claim 8, wherein step d) is preceded by a step i) of planarizing the upper face (3) of the support substrate (1) and / or an exposed surface of the composite material (5), such that the exposed surface and the upper face (3) extend in a single plane and form a surface suitable for bonding.

10. The manufacturing method according to claim 8 or 9, wherein step d) is preceded by a deposition step j) comprising the formation of an adhesion promoter (6) on the upper face (3) of the support substrate (1) and the exposed surface of the composite material (5) and / or on the donor substrate (7), such that the bonding of the donor substrate (7) to the support substrate (1) and to the composite material (5) is obtained by means of an adhesion promoter (6).

11. The manufacturing method according to any of claims 8 to 10, comprising performing a step k) of preparing the upper face (3) between step b) and step c) comprising: - arranging and holding a covering element (12) against the upper face (3) of the support substrate (1), covering at least the cavities (2), - turning the support substrate (1), so that the filling formulation is brought under gravity to the plane of the upper face (3) of the support substrate (1).