Thermally insulated microsystem

Abstract

Microsystem comprising a porous thermal insulation body (5) composed of particles having a thermal conductivity of less than 10 W / (m·K), wherein the porous thermal insulation body is subjected to negative pressure and is enclosed by a wall on all sides so that uncoated particles are immobilely enclosed, and the wall on all sides is obtainable by filling the particles through an opening (19) into the recess (20) of a substrate (1) of the microsystem and closing this opening after application of a coating by applying a sealing layer (4) and / or by forming a solid coating at least at the opening by means of a coating (3) of the particles.

Description

Field of the invention

[0001] The present invention relates to a microsystem comprising a porous thermal insulation body composed of particles of low thermal conductivity. State of the art

[0002] Microelectromechanical sensors (MEMS sensors) based on thermal measurement principles, such as flow meters and gas sensors, are becoming increasingly important for a wide variety of applications. While flow meters have previously been used primarily for monitoring gaseous and liquid media in technical systems, gas sensors were used to detect flammable or explosive gases. These include volatile organic compounds (fuel gases, methane, and ethanol), toxic gases (CO, NO x , H2S) or flue gases. However, there are now also many consumer applications, such as CO2 monitoring in conference rooms and the detection of odors such as cigarette smoke or kitchen fumes.

[0003] Calorimetric gas sensors and flow meters are based on so-called hotplates. Generally speaking, a hotplate consists of a thin-film heating element on a thermally insulated structure made of a material with the lowest possible thermal conductivity. The term "hotplate" is used synonymously with "heating element" below. Flow meters require temperatures up to 200°C. Often, several heating elements or temperature sensors are arranged on the membrane. The operating temperature of gas sensors is 300–600°C. Depending on the operating principle, the hotplate can be coated with additional coatings. The main problem with calorimetric components is the energy consumption required to maintain the operating temperature. Minimizing power consumption is particularly important for mobile applications. Another problem is the fragility of the self-supporting MEMS structure forming the hotplate.The hotplate can be located on a closed membrane made of silicon oxide and nitride. The latter is created by etching the silicon from the back of the substrate. Alternatively, freestanding support structures can be used for the hotplate. Freestanding hotplate structures are created by etching out the underlying silicon from the front of the substrate. Both variants have advantages and disadvantages. With membrane-based hotplates, convection effects or pressure fluctuations in the volume below the membrane can distort the measurement, and pressure surges can destroy the structure. In freestanding hotplate structures, which span an open channel and are surrounded by the medium flowing on both sides, the thermal insulation to the substrate is influenced by the medium. Therefore, unlike membrane-based components, such flow meters are not suitable for liquids.

[0004] To maintain a temperature of 400°C continuously, conventional sensors require approximately 15 mW of power to maintain the operating temperature. If the same sensors are switched on for only 2-3 seconds per minute, energy consumption is significantly reduced. However, this is still too high for certain applications.

[0005] The high power consumption results from the specific nature of micromechanical structures. Since the area-to-cross-section ratio of the heated region is very high, heat losses, unlike macroscopic bodies, are dominated not by thermal conduction in the volume, but by heat transfer to the ambient medium. Heat conduction through the membrane accounts for only about 40% of the losses and, due to its very thin thickness, is only slightly dependent on the material used. Heat losses via the ambient air, in contrast, amount to about 60%. A large portion of this is lost through the cavity beneath the sensor. Consequently, a further reduction in heat losses is hardly achievable through design and material optimization of a conventional sensor structure.

[0006] US 2011 / 0 279 996 A1, DE 10 2009 043 413 B3, US 2009 / 0 184 100 A1, DE 101 30 379 A1, US 5 593 526 A, US 6 261 469 B1, US 2004 / 0 016 962 A1 and US 2013 / 0 075 930 A1 disclose microsystems with a porous insulation body composed of particles with a low thermal conductivity. Problems to be solved by the invention

[0007] The state of the art does not offer a satisfactory solution to the main problem of calorimetric devices. This main problem is the energy consumption required to maintain operating temperature. Another problem not satisfactorily addressed in the state of the art is the fragility of self-supporting structures in microsystems.

[0008] Therefore, the object of the present invention is to provide a microsystem in which the task of thermal insulation and stabilization is solved. Summary of the invention

[0009] The present invention describes a novel possibility for the thermal insulation of microsystems, such as MEMS components, which results in both a reduction in energy consumption and an increase in their mechanical stability.

[0010] The problem was solved by providing a microsystem that has a porous thermal insulation body.

[0011] The subject of the present invention is a microsystem comprising a porous thermal insulation body (5) composed of particles having a thermal conductivity of less than 10 W / (m·K), wherein the porous thermal insulation body is subjected to negative pressure and is enclosed by a wall on all sides so that uncoated particles are immobilely enclosed, and the wall on all sides is obtainable by filling the particles through an opening (19) into the recess (20) of a substrate (1) of the microsystem and closing this opening after application of a coating by applying a sealing layer (4) and / or by forming a solid coating at least at the opening by means of coating (3) the particles.

[0012] Preferred embodiments of the microsystem are defined in the dependent claims. Advantages and application of the invention

[0013] The microsystems according to the invention combine the possibilities of conventional superinsulations with microtechnological manufacturing processes.

[0014] For example, the thermal insulation of a hot plate using a porous body offers decisive advantages compared to conventional MEMS components.

[0015] A first advantage is the mechanical stiffening of the hotplate. The application possibilities of conventional membrane-based components are significantly limited by the fragility of the thin membrane, which can be easily damaged, for example, by pressure surges, particle bombardment, or thermally induced stress at operating temperature.

[0016] A second advantage compared to state-of-the-art flow sensors is that the porous structure suppresses convection effects and pressure fluctuations in the cavity below the sensor membrane, which can lead to a distortion of the measurement.

[0017] A third advantage is that a silicon substrate with embedded porous structures can be further processed like a normal wafer. Using established semiconductor technology, vertical electrical feedthroughs (through silicon vias, TSV) can be easily integrated to spatially separate the electrical connections from the sensor area. Using such feedthroughs, the flow of the medium over the sensor surface is virtually unaltered.

[0018] The reduced heating power thanks to the improved thermal insulation in combination with mechanical stabilization of the sensor structure and spatial separation of the electrical connections enables the development of new areas of application.

[0019] If the microsystems according to the invention are gas sensors, they can be used much more easily in arrays for electronic noses or in so-called "sensor skins." The latter, in particular, require a flat design and the possibility of flush mounting.

[0020] Flow sensors could be used for wireless sensor nodes monitoring drinking water pipes in supply networks for real-time detection of leaks, consumption, or unauthorized use. Flow measurements on the outer skin of vehicles and aircraft, or in turbines, are also possible. The possibility of simple, seamless integration into the walls of pipes or conduits of any shape would also enable mass use in household appliances for flow measurement (coffee machines, air conditioners, showers, toilet flushes, etc.).

[0021] At a certain size of the particles contained in the heat insulation body (2 µm-20 µm), the IR light emitted by the heated surfaces is also scattered, so that the losses due to radiation transport are also minimized.

[0022] The technology described here is not limited to the production of hot plates for chemical sensors or thermal flow sensors, but is always advantageous when the best possible thermal insulation is required in a very tight installation space.

[0023] One obvious option is the construction of miniaturized thermal IR sources that also have to be operated at temperatures of several 100°C.

[0024] But this technology is also suitable for the construction of miniaturized atomic clocks, gyroscopes or magnetometers, because here the smallest gas cells must be kept at a temperature of over 100°C in order to keep the alkali metal (usually rubidium) contained therein in a gaseous state.

[0025] Last but not least, the best possible thermal insulation plays a key role in miniaturized heat engines, reducing losses due to heat conduction and increasing overall efficiency. One example of this is the design of pyroelectric harvesters. Description of the characters Fig. 1a and Fig. Figure 1b shows a schematic cross-section through a microsystem according to the invention. It is a hotplate arrangement in which the rear cavity was filled with a porous body after completion of the MEMS component. In Fig. 1a all particles are coated, in Fig. 1b only part of the particles are coated. Fig. 2a to 2c show the schematic cross-section through a microsystem according to the invention with a hotplate arrangement, which is only realized after generation and planarization of the thermally insulating porous regions in the substrate. Fig. 3a and Fig. 3b show schematic representations of the assembly and connection technology for a flowmeter with a conventional design (a) and a design according to the invention (b). The latter is based on an embedded, porous body for thermal insulation and features vertical electrical feedthroughs (TSV). Fig. 4a and Fig. 4b show schematic cross-sections through two possible arrangements according to the invention, based on thermal insulation made of solidified particles and in which the electrical connections are spatially separated from the sensor area. Accordingly, the medium to be monitored is located on the top side of the sensor chip in (a) and on the bottom side in (b). Fig. Figure 5 shows an embodiment in which an insulated reflector made of a thin gold layer additionally helps reduce radiation losses. To further reduce heat dissipation, this reflector is also thermally insulated and isolated by an evacuated cavity on its back, which is also filled with particles. This design is particularly interesting for hotplates with high temperatures > 500°C. Fig. 6a to 6e illustrate a method for producing a microsystem according to the invention. Fig. 7a to 7e illustrate a method for producing a microsystem according to the invention. Fig. 8a to 8f illustrate a method for producing a microsystem according to the invention. Fig. 9a and Fig. 9b shows the schematic cross-section through a microsystem according to the invention. Embodiments of the inventionMicrosystem

[0026] The microsystem according to the invention has dimensions in the range of micrometers to millimeters. Preferably, the dimensions are in the range of 1 µm to 10 mm, more preferably in the range of 10 µm to 5 mm, and even more preferably in the range of 50 µm to 2 mm.

[0027] The microsystem can be a MEMS structure.

[0028] In particular, the MEMS structure can be a calorimetric MEMS device based on a hot plate, e.g., a flow meter, a catalytic gas sensor, a MOS gas sensor (MOS stands for semiconducting metal oxide), or a radiation sensor (bolometer). Thermal insulation body

[0029] The thermal insulation body, sometimes referred to simply as a "porous structure" or "porous body" below, is arranged in the microsystem to provide thermal insulation. For this purpose, the thermal insulation body is located between two elements that have a temperature difference, for example, between two elements of the microsystem according to the invention. Alternatively or additionally, the thermal insulation body is arranged between an element of the microsystem and the environment. The environment can be an element of another structure or the ambient fluid, i.e., a liquid or a gas.

[0030] Furthermore, the thermal insulation body can be arranged in the space between at least two elements of the microsystem or in a recess or cavity of a single element of the microsystem.

[0031] The thermal insulation body is enclosed on all sides by other structures. This all-round enclosure of the thermal insulation body allows a portion of the particles to remain unconsolidated, i.e., loose. The thermal insulation body is preferably sealed using a sealing layer (9).

[0032] In any case, the structures are designed and connected to each other in such a way that a closed space is created from which existing loose particles cannot escape and thus be lost to the thermal insulation body. particles

[0033] The particles have an average size in the range of nanometers to micrometers, preferably from 10 nm to 100 µm, more preferably from 100 nm to 30 µm, even more preferably from 1 µm to 20 µm.

[0034] The material forming the particles preferably has low thermal conductivity. However, a mixture of particles made of different materials can also be used, with one material having a relatively high thermal conductivity of, for example, 40 W / m·K. Since the particles only touch at specific points, the effective thermal conductivity of a corresponding porous body is lower.

[0035] The particles can, for example, consist of silicon oxide, which has a thermal conductivity of approximately 1.2 to 1.4 W / (m·K). Other metal oxides with low thermal conductivity include yttrium oxide (Y2O3) or zirconium oxide (zirconia, ZrO2).

[0036] Under standard conditions, aluminum oxide has a thermal conductivity of 35.6-39 W m -1 ·K -1 (monocrystalline corundum: 40 W m -1 ·K -1 , dense ceramic with 96 % Al2O3 approx. 25 W·m -1 ·K -1 ), which increases sharply with decreasing temperature and with increasing temperature at 1000 °C to about 5 W m -1 ·K -1 sinks.

[0037] Glass has different thermal conductivity depending on its composition, for example Zerodur has 1.46 W / (m·K) and soda-lime glass has 0.80 W / (m·K).

[0038] Example metals have the following thermal conductivities: tungsten 167 W / (m·K), Ni 85 W / (m·K), iron 80 W / (m·K), tin 67 W / (m·K), tantalum 54 W / (m·K), titanium 22 W / (m·K), bismuth 8.4 W / (m·K).

[0039] Typical thermal insulation materials such as kaolin, chalk and talc have a thermal conductivity of less than 2 W / (m·K).

[0040] Of course, organic polymers can also be used, for example: polyurethane compact (PUR) 0.245 W / (m·K); polytetrafluoroethylene (PTFE) 0.25 W / (m·K); polyvinyl chloride (PVC) 0.17; polypropylene (PP) 0.23 W / (m·K); polycarbonate 0.20 W / (m·K); epoxy resin (EP) 0.20 W / (m·K); polymethyl methacrylate (PMMA, Plexiglas) 0.19 W / (m·K); polysiloxanes (silicone) 0.2...0.3 W / (m·K).

[0041] The particles preferably have a relatively small diameter of less than 20 µm, preferably less than 10 µm and particularly preferably less than 5 µm.

[0042] To avoid damage to the carrier element and / or the particles during coating, it is advantageous if the particles and the substrate are heat-resistant up to a temperature of at least 60°C, preferably 150°C or 300°C. The higher the heat resistance, the fewer restrictions there are regarding the coating processes that can be used. At the same time, the high heat resistance enables faster implementation of the process without the risk of damage to the particles and / or the carrier element due to overheating.

[0043] In order to avoid a loss or reduction of the porous properties, the particles are preferably not substantially irreversibly deformed when bonded together.

[0044] In a particularly preferred embodiment, a particle mixture comprising particles made of at least two different materials is introduced. The materials preferably differ in at least one physical or chemical property. For example, the hardness, electrical conductivity, thermal conductivity, thermal expansion coefficient, and / or the electrochemical parameters of the particles can be selected differently. By using different materials with different mixing ratios and, if appropriate, different chemical and physical properties, the properties of the three-dimensional structure can be precisely and reliably adapted and adjusted to the desired specifications.

[0045] For example, the thermal expansion coefficient of the particles or particle mixture can be adapted to the thermal expansion coefficient of the carrier element. Likewise, the thermal expansion coefficient of the particles or particle mixture can also be adapted to the component in which the bonded particles are to be arranged.

[0046] The particles can also be a mixture of different materials in particulate form. For example, particles can be used to adjust the effective thermal expansion coefficient.

[0047] In some cases, it may also be advantageous to add short fibers to the particles in order to increase the mechanical stability of the entire structure.

[0048] Depending on the intended application, it may therefore be advantageous to adapt the particle size to the expected operating temperature of the sensor element. Thermal conductivity

[0049] The thermal conductivity of the particles is less than 10 W / (m·K), preferably less than 5 W / (m·K), more preferably less than 3 W / (m·K), and even more preferably less than 1 W / (m·K). The thermal conductivity data refer to measurements at 20 °C, 1013 hPa, and 50% relative humidity.

[0050] The thermal conductivity value of the particles may also refer to a mixture of particles made of different materials with a total conductivity of less than 10 W / (m·K) or a total conductivity within the preferred ranges mentioned. Coating

[0051] In a preferred embodiment, at least a portion of the particles are bonded together by means of a coating.

[0052] Preferably, this proportion is at least 5%, at least 10%, at least 20%, at least 50% or at least 90%.

[0053] In a further preferred embodiment, at least 95%, more preferably at least 98% and even more preferably all particles are bonded together by means of a coating.

[0054] The thermal insulation body is surrounded by a wall on all sides. This wall can enclose a space initially provided with an opening for filling the particles. The opening can then be sealed by coating the particles to form a solid coating, at least at the opening. This coating then accounts for a proportion of 3 to 50%, preferably 5 to 30%, and more preferably 10 to 20% of the total amount of particles. Thermal insulation

[0055] The term thermal insulation as used herein means the reduction of heat transfer by conduction, preferably the reduction of heat transfer by conduction and at least one type of heat transfer selected from radiation and convection.

[0056] The reduction in heat conduction is achieved by using a material with low thermal conductivity.

[0057] Convection reduction is achieved by using a material with bonded particles. This material contains or consists of the thermal insulation body contained in the microsystem. negative pressure

[0058] The thermal insulation body is subjected to negative pressure. This means that the pressure inside is lower than the pressure outside the body. Preferably, the pressure is reduced in at least one part of the thermal insulation body where pores are located.

[0059] The term “negative pressure” also includes vacuum.

[0060] The negative pressure is less than 1013 hPa, preferably less than 1000 hPa, more preferably less than 700 hPa, even more preferably less than 500 hPa.

[0061] The most preferred negative pressure is in the range of 10 -3 to 300 hPa, or particularly preferably in the range from 1 to 100 hPa.

[0062] The term “vacuum” as used herein means that the pressure is less than 300 hPa, preferably less than 100 hPa, more preferably less than 10 hPa and even more preferably less than 1 hPa.

[0063] The negative pressure has the advantage that the thermal conductivity of the thermal insulation body is further reduced. porosity

[0064] The thermal insulation body contained in the microsystem according to the invention is porous, i.e. it has a large number of pores between the particles.

[0065] Porosity indicates the ratio of the void volume to the total volume of the thermal insulation body. The porosity of the thermal insulation body is high for weight reasons. Due to the negative pressure or vacuum within the thermal insulation body, high porosity results in low weight while maintaining good thermal insulation.

[0066] Examples of porosity ranges are 1 to 70%, 5 to 40%, or 10 to 50%. Other examples are porosities of more than 1%, more than 5%, more than 10%, more than 20%, more than 30%, and more than 50%; and each of these values ​​optionally combined with an upper limit of 80%, or each of these values ​​optionally combined with an upper limit of 70%.

[0067] Porosity can be controlled. Reducing porosity is possible through additional coating. Higher porosities can be achieved if the particles are already porous or, for example, have a branched shape. Sealing layer

[0068] Depending on the embodiment, a sealing layer (9), also called a cover layer, can be applied at least partially to the arrangement of particles and carrier element. This sealing layer essentially completely seals cavities near the surface and / or forms a closed layer on an uppermost layer of the particles. The cover layer is preferably subsequently planarized, for example by grinding or polishing. This allows a particularly smooth layer to be created.

[0069] The thermal insulation body is surrounded by a wall on all sides. This wall can enclose a space that is initially provided with an opening for filling the particles. The opening can then be closed by applying a sealing layer (9).

[0070] The sealing layer is applied after optionally a portion or all of the particles have been bonded by coating. First preferred method for producing the porous structure

[0071] In the present invention, the heat insulating body (porous structure) is preferably manufactured by arranging a plurality of particles in a recess to form a porous structure.

[0072] A general process for producing porous structures is described in DE 10 2011 010 899 A1.

[0073] In this process, particles are first applied to or onto a carrier element. A multitude of at least partially interconnected cavities are formed between the particles. These cavities are also called pores. The particles come into contact with each other at points of contact. Application can be achieved, for example, by doctoring. The particles can be of any shape and size.

[0074] After the particles have been inserted or applied, they are preferably bonded together. This bonding occurs at the points of contact. For this purpose, the particles are coated, with the cavities being at least partially penetrated by the layer created during the coating process.

[0075] Depending on the design, the cavities can remain open during coating, meaning that while the coating creates a connection between the particles, the majority of the cavities between the particles remain connected. Alternatively, at least some of the cavities can be closed by the coating, so that the cavities form closed pores.

[0076] It is also possible that only a part of the particles are bonded during coating, i.e. both partial and complete penetration of the cavities with the layer created during coating is possible.

[0077] If essentially all particles are connected to one another and thus essentially all cavities are penetrated, a layer also forms on the area of ​​the carrier element in which the particles are applied or introduced.

[0078] Depending on the application, electrically conductive or non-conductive particles can be used. Electrically conductive or non-conductive materials can also be used for the coating.

[0079] A relatively uncomplicated and uniform application of the coating is possible using a CVD (chemical vapor deposition) process, and in particular, using atomic layer deposition. In atomic layer deposition, two different components are alternately introduced in the gaseous state, each of which deposits on the surface of the particles or the carrier element. With each deposition process, a substantially continuous one-atom-thick layer is formed. This creates a particularly uniform layer. The layer thickness is primarily adjusted by how often the application of a first component alternates with that of a second component.

[0080] The thickness of the applied layer can particularly preferably be between one atomic layer and 5 µm, between 5 nm and 1 µm, or between 50 nm and 300 nm. Thin layer thicknesses have the advantage of being able to be produced within a short time. Layer thicknesses in the upper range of the specified ranges, on the other hand, enable a particularly strong and reliable bond.

[0081] The coating preferably occupies a negligible proportion of the volume of the cavities formed during the insertion or application of the particles. The volume of the layer can preferably be less than 1% of the volume of the cavities.

[0082] Depending on the design, either only a portion of the particles or essentially all of the particles inserted or applied are bonded together. Bonding essentially all of the particles is particularly advantageous when a structure of the carrier element is to be molded.

[0083] The particles can be arranged in loose contact with one another before coating. Thus, no additional process steps are required to create a bond between the particles before applying the coating. Substrates made of silicon, glass, or ceramic have proven particularly suitable. The substrate can be in the form of a wafer or plate, for example. Silicon wafers are particularly well-suited for applications in the microelectronics sector.

[0084] Based on Fig. 2a and Fig. 6a to 6e illustrate an embodiment of the method.

[0085] In a first process step, a substrate 1 is provided into which a recess 20 is formed. Particles are filled into the recess 20, e.g., by sprinkling the particles and / or by doctor blade coating. The arrangement and orientation of the particles relative to one another is arbitrary.

[0086] In a subsequent process step, a coating with a thickness of 10 nm to 300 nm is applied using a CVD (chemical vapor deposition) process, more precisely, using atomic layer deposition. The coating material not only forms a layer at the bottom of the recess or next to the recess, but also adheres to the particles. Since the particles are in contact with each other at points of contact after being filled into the recess 20, the deposition of the coating material bonds the particles to one another, forming a coherent body of solidified particles. During coating, the particles and the substrate element are heated to approximately 60°C to 300°C. Accordingly, the material for the particles and substrate element is designed to be temperature-resistant.

[0087] Fig. shows another variant in which the particles are not solidified to the bottom of the depression. The corresponding process is described in Fig. 7a to 7e.

[0088] Fig. shows a third variant, in which the Fig. Loose particles are selectively etched out before sealing. This can be done using a suitable etching gas through the pores between the solidified particles 17. If the unsolidified particles 18 consist of silicon dioxide, for example, they can be removed in the XeF2 gas phase. The process is described in Fig. 8a to 8f.

[0089] A further embodiment of the method according to the invention is explained below ( Fig. 9a and Fig. 9b). In this embodiment, a substrate with a filled recess 20, as in connection with Fig. 2a or Fig. 2b. In this exemplary embodiment, however, the process parameters of the ALD method can also be selected such that the ALD layer does not completely penetrate the cavities, i.e., not all of the particles necessarily have to be connected to one another. Subsequently, the substrate 1 is partially removed, so that the compacted particles are only attached to the substrate at the edge, while in a central region they form a self-supporting membrane. If some particles are not connected, they are also removed in this process step. The thickness of the self-supporting membrane, which can be approximately 40 µm, for example, is then determined by the penetration depth of the ALD layer into the recess 20 filled with particles.This can, in particular when substantially all particles have been bonded and the coating material has penetrated the cavities to the bottom of the recess 20, still be covered on its back side with the substantially closed ALD layer that has formed on the surface of the substrate 1 during the application of the coating.

[0090] This ALD layer, if present, is removed in the subsequent process step. Removal can be achieved, for example, by ion beam etching.

[0091] Alternatively, the ALD shell (3) is retained and covered from below with a sealing layer. This makes the porous body Fig. 9a under vacuum (in contrast to Fig. 9b).

[0092] Preferred embodiments of the method are given below: [1] A method for producing a three-dimensional structure comprising the steps of: - applying or introducing particles onto or into a substrate (1), wherein a plurality of at least partially interconnected cavities is formed between the particles and wherein the particles come into contact with each other at contact points, - Connecting the particles at the contact points by coating the particles, whereby the cavities are at least partially penetrated by the layer produced during the coating. [2] The method according to point [1], characterized in that the coating is carried out by means of a CVD process and in particular an atomic layer deposition. [3] The method according to any one of the preceding points, characterized in that the layer is formed with a thickness between one atomic layer and 5 µm, preferably between 5 nm and 1 µm and particularly preferably with a thickness between 50 nm and 300 nm. [4] The method according to any one of the preceding points, characterized in that the particles and / or the substrate (1) are heat-resistant up to a temperature of at least 60°C, preferably 150°C or 300°C. [5] The method according to any one of the preceding points, characterized in that the particles do not deform substantially irreversibly when bonded together. [6] The method according to one of the preceding points, characterized in that a particle mixture with particles of at least two different materials is introduced. [7] The method according to any one of the preceding points, characterized in that substantially all the particles applied or introduced are bonded together by the coating. [8] The method according to any one of the preceding points, characterized in that the particles are arranged in loose contact with each other before coating. [9] The method according to any one of the preceding points, characterized in that a substrate is used which is made of silicon, glass or ceramic.

[10] The method according to any one of the preceding points, characterized in that elements which have at least ten times the average volume of a particle are embedded and bonded to the particles and the substrate (1) during coating.

[11] The method according to one of the preceding points, characterized in that after the coating of the particles, at least partially a cover layer (sealing layer 9) is applied, which preferably completely closes cavities near the surface and which is particularly preferably subsequently planarized.

[12] The method according to one of the preceding points, characterized in that after coating and bonding the particles, the substrate (1) and optionally the particles remaining uncoated are at least partially removed, wherein the removal is preferably carried out by etching. Second preferred method for producing the porous structure

[0093] In this method, the thermal insulation body is preferably produced by arranging a plurality of particles in a recess so that a porous structure is formed.

[0094] Bonding a portion of the particles prevents the remaining particles from escaping, allowing the reactor to operate reproducibly and sustainably.

[0095] According to one embodiment, a device comprises a substrate having a recess and a plurality of particles arranged in the recess.

[0096] The term “device” used below may be the microsystem according to the invention, be contained therein or contain it as a component.

[0097] The term “porous structure” used below can be the thermal insulation body contained in the microsystem according to the invention, be contained therein or contain it as a component.

[0098] In a preferred embodiment, a first portion of the particles is bonded to a porous structure by means of a coating and a second portion of the particles is not bonded to a porous structure by means of the coating.

[0099] The first portion of particles is arranged closer to an opening in the recess than the second portion of particles, preventing the second portion of particles from escaping from the recess through the opening. This allows the properties of the uncoated portion of particles to be little or not affected by the coating. At the same time, the arrangement of particles allows for a high surface area and thus high porosity to be achieved.

[0100] By bonding the particles, they are solidified and form a composite structure. In one embodiment, the entire particles are incorporated into the composite structure. The production of the composite structure is described below.

[0101] According to a further embodiment, the first portion of particles comprises a plurality of cavities arranged between the particles of the first portion, which are at least partially interconnected, wherein the porous structure is immovably connected to the substrate. One advantage of this embodiment is the stabilization of the substrate.

[0102] According to a further embodiment, the opening of the recess borders a cavity of the device or the surrounding area. This is advantageous in that the second portion of particles can be held in place by the first portion of particles. With respect to the cavity of the device, for example, an internal volume, the second portion of particles can thus be prevented from falling out or trickling.

[0103] According to another embodiment, a method for manufacturing a device comprises providing a substrate having a recess and introducing a plurality of particles into the recess. The method further comprises coating a portion of the plurality of particles such that the first portion is bonded to form a porous structure.

[0104] The coating is applied by a coating process that has a penetration depth starting from an opening of the recess into the recess. The penetration depth is adjusted such that a second portion of the particles is not bonded by the coating, the first portion of particles is arranged between the second portion of particles and a surrounding area of ​​the recess, and movement of the second portion of particles toward the surrounding area of ​​the recess is substantially prevented. This means that the first portion of particles can be arranged facing the opening of the recess.

[0105] Fig. Figure 2b shows a schematic side sectional view of a device comprising a substrate 1. The substrate 1 comprises a recess 20, for example, a blind hole, a trench, or a depression. A plurality of particles are arranged in the recess 20. A first portion 17 of the particles is bonded to a porous structure 5 by means of a coating. A second portion 18 of the particles is not bonded by means of the coating.

[0106] The first portion 17 of the particles is arranged closer to an opening 19 of the recess 20 than the second portion 18 of the particles. Escape of the second portion 18 of particles from the recess through the opening 19 thereof is prevented by means of the porous structure 5, i.e. the first portion 16 of the particles. The porous structure 5 can be firmly connected to the substrate 1, for example by the first portion 17 and the substrate 1 having the same coating and the coating connecting both the particles of the first portion 16 to one another and the porous structure 5 to the substrate 1. Alternatively, it is also conceivable for the porous structure 5 to be connected to the substrate 1 in a different way, for example by a printing, pressing or adhesive process.

[0107] The porous structure 5 is designed to have a plurality of cavities that are at least partially interconnected. The cavities can be obtained by having particles brought into contact with one another with cavities arranged therebetween, which remain at least partially unclosed by the coating.

[0108] A coating of the first portion 16 can be carried out by means of any suitable method (coating process), for example an atomic layer deposition (ALD), a physical vapor deposition (PVD) or a chemical vapor deposition (CVD).

[0109] For example, during a manufacturing process, a large number of particles can be arranged in the recess 20, wherein the particles can be of different types or of the same type. Subsequently, the first portion 16 can be coated by selecting the coating parameters accordingly, or the coating can be ended or aborted when a sufficient number of particles have been coated so that the first portion 17 can form the porous structure 5. The coating process can have a penetration depth 21 that describes an effectiveness (coating of particles) of the coating process from the opening 19 along a depth direction 22 into the recess 20. Put simply, particles of the first portion 16 are essentially arranged at a distance from the opening 19 that is less than or equal to the penetration depth 21 of the coating process.

[0110] The particles of the second portion 18 can be arranged at a distance from the opening 19 that is greater than the penetration depth 21 of the coating process. Adjusting the penetration depth 21 can comprise adjusting a parameter of the coating process, such as pressure, time, temperature, or the like. For example, adjusting the penetration depth 21 can be such that, for an atomic layer deposition (coating process), the residence time of a first or second reactant is set such that it penetrates the particles from the opening 19 or a cover surface (fill level) of the particles only up to the penetration depth 21. Coating is achieved, for example, by both reactants (sequentially one after the other) coming into contact with the particle and, for example, depositing themselves on the particle. For example, a reactant can be introduced and portions not arranged on the particles can be removed again (suctioned off) before another reactant is introduced.If both reactants come into contact with each other on the particles, a coating can form there. If the residence time is adjusted so that at least one of the reactants does not exceed the penetration depth 21, coating of particles that are further away from the opening 19 than the penetration depth 21 can be reduced or avoided. Alternatively, for example, a coating time of a CVD deposition can be adjusted so that coating of particles that are further away from the opening 19 than the penetration depth 21 is essentially or completely prevented. For example, when using Al2O3, a "penetration depth" within which the particles are coated can be adjusted within wide limits. Aspect ratios of over 1000 can be achieved.

[0111] This means that a chemical or physical property of the second portion 18 of the plurality of particles may remain substantially unchanged during a manufacturing process.

[0112] The coating can be carried out starting from one side of the area surrounding the recess 20 in a direction of a volume of the recess. This allows particles arranged adjacent to the opening 19 of the recess (the first portion 17) to be coated, while particles spaced apart from the opening 19 remain uncoated.

[0113] The ratio of the number of particles in the second portion to the number of particles in the first portion can be 0.01, 0.1, 0.2, 0.5, 1, 2, 10, or greater. This means that there can be more uncoated particles than coated particles, or vice versa. In one embodiment, all particles are coated.

[0114] The substrate 1 can be, for example, a semiconductor material. The semiconductor material can be, for example, a silicon material or a gallium arsenide material.

[0115] Alternatively or additionally, the substrate 1 can comprise a glass material, a ceramic material, a glass-ceramic material, and / or a plastic material. The substrate 1 can be doped. For example, the semiconductor material can be doped, such as with a boron material. For example, the substrate 1 is a MEMS wafer. The substrate 1 can have a planar or non-planar shape (e.g., curved).

[0116] If the coating process (especially the residence time of the reactants) is modified, it can be achieved that the introduced particles are only coated to a certain depth.

[0117] Falling out of the uncoated, ie, loose, particles can be prevented by bonding the porous structure to the substrate 22 or by preventing the porous structure 5 from detaching from the substrate. This allows materials in powder form, ie, the second portion of the particles, to be enclosed in a miniaturized container, ie, the recess.

[0118] In one embodiment, the introduced particles are only coated superficially in the corresponding steps. To avoid interaction of the reactive particles with the environment, ALD processes can be used that are highly effective even at low temperatures, as is the case with Al2O3, for example.

[0119] The embodiments described above are independent of the size (e.g. diameter or lateral dimension) of the particles.

[0120] Although the previously described embodiments were described in such a way that the porous structure has cavities that are at least partially interconnected, so that the porous structure 5 forms a porous membrane, the coating can also be used, particularly with a particle size of <2 µm, to seal the porous structure 5, i.e., to close off the cavities between the particles. Thus, a recess can also be closed by means of the coating.

[0121] Preferred embodiments of the device (microsystem) and the method for producing the same are described below. [1] A device comprising: a substrate having a recess; and a plurality of particles arranged in the recess; wherein a first portion of the particles is bonded to a porous structure by means of a coating and wherein a second portion of the particles is not bonded to a porous structure by means of the coating; and wherein the first portion of particles is arranged closer to an opening of the recess than the second portion of particles, so that escape of the second portion of particles from the recess through the opening is prevented. [2] The device according to item [1], wherein the first portion of the particles comprises a plurality of cavities arranged between the particles of the first portion of the particles, the cavities being at least partially connected to one another and the porous structure being immovably connected to the substrate. [3] Device according to one of the preceding points, wherein the substrate comprises a material selected from a semiconductor material, a glass material, a ceramic material, a glass-ceramic material and a plastic material. [4] A method for manufacturing a device comprising the following steps: Providing a substrate having a recess; introducing a plurality of particles into the recess; Coating a first portion of the plurality of particles such that the first portion is bonded to a porous structure using a coating process having a penetration depth from an opening of the recess along a depth direction into the recess; and wherein the penetration depth of the coating process into the recess is adjusted such that a second portion of the particles is not connected by means of the coating, and that the first portion of particles is arranged between the second portion of particles and an environment of the recess and a movement of the second portion of particles towards the environment of the recess is substantially prevented. [5] Method according to item [4], in which the coating is carried out from a side of the environment of the recess in the direction of a volume of the recess, so that particles of the first portion arranged adjacent to the opening are reached and coated by the coating process based on the penetration depth of the coating process and particles of the second portion spaced from the opening are not coated. [6] A method according to item [4] or [5], in which the coating step is carried out by means of chemical vapor deposition, atomic layer deposition or atomic vapor deposition. [7] A process according to any one of items [4] to [6], wherein a chemical or physical property of the second portion of the plurality of particles remains unchanged during the process. Examples of implementation

[0122] Using the process described in DE 10 2011 010 899 A1, a membrane-based hotplate is provided with a porous, vacuum-sealed support. Since the particles of such a porous body only touch each other at specific points and there is no ambient medium, the hotplate is far better insulated from below than conventional components without a porous structure (thermal insulation body).

[0123] A first embodiment is shown in Fig. 1. The cavities below the membrane at the substrate level are filled from the back with particles of a thermally poorly conductive material (e.g., SiO2) and solidified with an Al2O3 ALD layer. The porous bodies are then sealed by applying a sealing layer to the back of the substrate, e.g., using PECVD, under vacuum. Removing any of the particles or the substrate is not mandatory.

[0124] In contrast to Fig. 1a illustrated Fig. 1b the case that the particles below the membrane are not completely solidified.

[0125] When executed in Fig. 2a, the thermally insulating regions are created before the sensor-specific structures, i.e., integrated into the sensor manufacturing process. The substrate with the embedded porous regions is optionally planarized before further processing, e.g., by grinding and polishing. This removes at least some of the solidified particles. A covering layer is then applied to seal the porous regions. Only then are the sensor-specific structures processed. Fig. 2b are the particles (analogous to Fig. 1b) is not fully solidified. Moreover, the unsolidified particles can be etched out through the porous structure of the solidified area before the sealing layer is applied.

[0126] The thermal insulation of a hotplate by means of a porous body according to Fig. 1 or Fig. 2 offers decisive advantages compared to conventional MEMS components.

[0127] Advantages of this design are the mechanical stiffening of the hot plate and the suppression of convection effects and pressure fluctuations in the cavity below the sensor membrane, which can lead to a falsification of the measurement.

[0128] A further advantage is the processability of the silicon substrate with embedded porous structures like a normal wafer. Using established semiconductor technology, vertical electrical feedthroughs (Through Silicon Vias, TSV) can be easily integrated to spatially separate the electrical connections from the sensor area. By relocating the electrical connections to the back of the sensor chip, its assembly and connection technology is significantly simplified. With conventional components, Fig. 3a The bond wires are potted with a suitable plastic (glob top) if the medium to be monitored is corrosive or contains moisture. A cover is often mounted to protect the electrical leads on the carrier. In the case of a flowmeter, the potted bond wires impair the flow of the medium to be monitored. Components with TSV may require a cover and sealing of the gaps, see Fig. 3b. The flow of the medium over the sensor surface is almost not distorted.

[0129] Fig. Figure 4 shows two possible device designs with a thermally insulating porous region below the hotplate and a spatial separation of the electrical connections from the sensor area. The latter can also be achieved without TSV, see Fig. 4b. The hotplate is created on the bottom of a KOH-etched cavity with sloped sides. The electrical connections to the heater are routed to the surface via the sloped sides of the cavity. The cavity is then filled with particles. After applying a sealing layer, the surface can optionally be planarized. Finally, the silicon of the substrate is removed from the back to expose the heater / temperature sensor. This can be achieved by full-surface removal (grinding and polishing) followed by mask etching. Re-thinning the substrate by grinding and polishing from the back is also possible with the design according to Fig.4a is possible. These innovative insulation elements combine the capabilities of conventional superinsulation with microtechnological manufacturing processes. The loose particles within the insulation ensure minimal contact surfaces between the individual particles, which in turn greatly reduces their contribution to thermal conductivity. Furthermore, since there is a vacuum within the insulation, convection is also absent, and heat conduction via molecular motion is also eliminated. List of reference symbols 1 substrate 2 Bond pad 3 ALD layer (shell) 4 Sealing layer 5 thermal insulation bodies 6 heaters 7 electrical supply line 8 carriers 9 Cover 10 Membran 11 Bonding wire 12 Glob Top 13 electrical feedthroughs (Through Silicon Vias, TSV) 14 Pads 15 Gold reflector 16 mounting surfaces 17 first portion of the particles 18 second part of the particles 19 Opening 20 recess 21 Penetration depth 22 Depth direction

Claims

[1] Microsystem comprising a porous thermal insulation body (5) composed of particles having a thermal conductivity of less than 10 W / (m·K), wherein the porous thermal insulation body is subjected to negative pressure and is enclosed by an all-round wall so that uncoated particles are immobilely enclosed, and the all-round wall is obtainable in that the particles are filled through an opening (19) into the recess (20) of a substrate (1) of the microsystem and this opening is closed after application of a coating by applying a sealing layer (4) and / or by forming a solid coating at least at the opening by means of coating (3) the particles. [2] Microsystem according to claim 1, wherein the particles have a thermal conductivity of less than 3 W / (m·K). [3] Microsystem according to claim 1 or 2, wherein at least a portion of the particles are interconnected by means of a coating (3). [4] Microsystem according to claim 3, wherein all particles are connected to one another by means of a coating (3). [5] Microsystem according to one of the preceding claims, wherein the heat insulation body (5) is arranged between at least two elements which have different temperatures during operation of the microsystem and / or between at least one element which is heated during operation of the microsystem and the environment. [6] Microsystem according to claim 5, wherein the element which is heated during operation of the microsystem is a heating element (6). [7] Microsystem according to one of the preceding claims, wherein the microsystem is a flow sensor, a gas sensor, a bolometer or an infrared light source.

Images