Inertial sensor, method for manufacturing inertial sensor, and inertial measurement unit

Abstract

To provide an inertial sensor with high reliability, and a method for manufacturing an inertial sensor.SOLUTION: An inertial sensor comprises: a substrate; an insulating film provided on a principal surface of the substrate; a first semiconductor layer and a second semiconductor layer provided on a surface of the insulating film on the opposite side of the substrate; a first oxide film provided on a first side face of the first semiconductor layer facing the second semiconductor layer; a second oxide film provided on a second side face of the second semiconductor layer facing the first semiconductor layer; a flattened insulating film embedded above the first oxide film and the second oxide film and between the first oxide film and the second oxide film; and wiring provided on the flattened insulating film and electrically connected with the second semiconductor layer.SELECTED DRAWING: Figure 3

Description

[Technical field]

[0001] The present invention relates to an inertial sensor, a method for manufacturing an inertial sensor, and an inertial measurement unit. [Background technology]

[0002] Patent Document 1 describes an angular velocity sensor in which a trench groove for filling with an insulator is formed in a silicon substrate, which is a semiconductor substrate, and the insulator filled in the trench groove electrically separates, for example, a fixed electrode from its outer periphery at its base. Patent Document 2 describes a semiconductor device in which an insulation isolation trench is formed in a silicon substrate, a sidewall insulating film is formed on the sidewall of the insulation isolation trench, and buried polysilicon as a conductive material is filled in the insulation isolation trench with the sidewall insulating film formed therein. [Prior art documents] [Patent documents]

[0003] [Patent Document 1] Japanese Patent Application Publication No. 11-248733 [Patent Document 2] JP 2003-45988 A Summary of the Invention [Problem to be solved by the invention]

[0004] However, when the technology described in Patent Document 2 is applied to the angular velocity sensor described in Patent Document 1, a conductive material is inserted between the fixed electrode and its outer periphery at its base, that is, between the semiconductors separated by the trench, which increases the parasitic capacitance between the semiconductors separated by the trench, raising concerns about degradation of the bias characteristics of the angular velocity sensor. Furthermore, for example, when a trench is filled with a thermal oxide film, irregularities remain on the surface of the thermal oxide film, which may cause cracks in the wiring when the wiring is formed in the upper layer of the trench. In other words, there is a need for a highly reliable inertial sensor, a method for manufacturing an inertial sensor, and an inertial measurement unit. [Means for solving the problem]

[0005] An inertial sensor according to one embodiment of the present application comprises a substrate, an insulating film provided on a main surface of the substrate, a first semiconductor layer and a second semiconductor layer provided on a surface of the insulating film opposite the substrate, a first oxide film provided on a first side surface of the first semiconductor layer facing the second semiconductor layer, a second oxide film provided on a second side surface of the second semiconductor layer facing the first semiconductor layer, a planarizing insulating film embedded above the first oxide film and the second oxide film and between the first oxide film and the second oxide film, and a wiring provided on the planarizing insulating film and electrically connected to the second semiconductor layer.

[0006] A manufacturing method of an inertial sensor according to one embodiment of the present application includes the steps of: preparing a base body having a substrate, an insulating film formed on a main surface of the substrate, and a semiconductor layer formed on a surface of the insulating film opposite the substrate; forming a trench portion and a first semiconductor layer and a second semiconductor layer opposing each other via the trench portion by removing a portion of the semiconductor layer; thermally oxidizing the first semiconductor layer and the second semiconductor layer to form a first oxide film and a second oxide film, and filling the trench portion by physically contacting the first oxide film and the second oxide film; forming a planarizing insulating film on the first oxide film and the second oxide film by a coating method; and forming wiring on the planarizing insulating film.

[0007] An inertial measurement device according to one aspect of the present application includes the inertial sensor described above and a control unit that controls the inertial sensor. [Brief description of the drawings]

[0008] [Figure 1] FIG. 1 is a plan view of an acceleration sensor according to a first embodiment. [Diagram 2] FIG. 2 is a cross-sectional view taken along line bb in FIG. 1 . [Diagram 3]FIG. 3 is an enlarged cross-sectional view of part c in FIG. 2 . [Figure 4] FIG. 4 is a flowchart showing a manufacturing method of the acceleration sensor. [Diagram 5] 1A to 1C are cross-sectional views of an insulating isolation portion according to one embodiment of the present invention during the manufacturing process. [Figure 6] 1A to 1C are cross-sectional views of an insulating isolation portion according to one embodiment of the present invention during the manufacturing process. [Figure 7] 1A to 1C are cross-sectional views of an insulating isolation portion according to one embodiment of the present invention during the manufacturing process. [Figure 8] 1A to 1C are cross-sectional views of an insulating isolation portion according to one embodiment of the present invention during the manufacturing process. [Figure 9] FIG. 6 is a cross-sectional view of an insulating separation part according to a second embodiment. [Figure 10] FIG. 11 is a plan view of an acceleration sensor according to a third embodiment. [Figure 11] FIG. 11 is an enlarged cross-sectional view of part f in FIG. [Figure 12] 12 is a cross-sectional view taken along line gg in FIG. 11 . [Figure 13] FIG. 11 is an exploded perspective view of an inertial measurement unit according to a fourth embodiment. [Figure 14] FIG. DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENTS

[0009] EMBODIMENT 1 ***Accelerometer Overview*** Fig. 1 is a plan view of the acceleration sensor according to embodiment 1. Fig. 2 is a cross-sectional view taken along line bb in Fig. 1. First, an acceleration sensor 1 as an example of an inertial sensor according to this embodiment will be described with reference to the drawings. In each of the following figures, an X-axis, a Y-axis, and a Z-axis are illustrated as three mutually orthogonal axes. The direction along the X-axis is called the "X-direction", the direction along the Y-axis is called the "Y-direction", and the direction along the Z-axis is called the "Z-direction". In addition, the tip side of the arrow in each axis direction is also called the "plus side", and the base side of the arrow is also called the "minus side". That is, for example, the Y-direction refers to both the plus side of the Y-direction and the minus side of the Y-direction. In addition, the plus side of the Z-direction is also called the "upper side", and the minus side of the Z-direction is also called the "lower side". In addition, in the following figures, dimensions and scales different from the actual ones may be used to make the explanation easier to understand.

[0010] The acceleration sensor 1 is a capacitance-type acceleration sensor that detects acceleration using a change in capacitance that depends on the distance between a movable part 2 and a fixed electrode part 3. As shown in FIGS. 1 and 2, the acceleration sensor 1 is configured using a base body 7 in which a substrate 4, an insulating film 5, and a semiconductor layer 6 are laminated in this order along the Z direction.

[0011] The substrate 4 has a first surface F1 which is the main surface of the substrate 4, and a second surface F2 which is opposite to the first surface F1. In this embodiment, the first surface F1 is the surface of the substrate 4 on the positive side in the Z direction, and is also referred to as the upper surface of the substrate 4. The second surface F2 is the surface of the substrate 4 on the negative side in the Z direction, and is also referred to as the lower surface of the substrate 4. A bottomed recess 8 is formed in the center of the substrate 4 in a plan view in the Z direction. The recess 8 has an opening on the first face F1 of the substrate 4 and is recessed from the first face F1 toward the second face F2. The recess 8 is also called a cavity. In this embodiment, the substrate 4 is a semiconductor substrate, and more specifically, a single crystal silicon substrate.

[0012] An insulating film 5 is formed on a first face F1, which is the upper face of the substrate 4. In this embodiment, the insulating film 5 is made of silicon oxide. In this embodiment, the insulating film 5 is also formed on the side and bottom faces of the recess 8. In this embodiment, it is sufficient that the insulating film 5 is formed on the first face F1, and the insulating film 5 on the side and bottom faces of the recess 8 may be removed. In this embodiment, no insulating film is formed on a second face F2, which is the lower face of the substrate 4, but an insulating film may be formed on the second face F2. A semiconductor layer 6 is formed on the opposite side of the insulating film 5 to the substrate 4. The semiconductor layer 6 is bonded to a third face F3, which is the face of the insulating film 5 opposite to the substrate 4, in the peripheral portion of the recess 8. The third face F3 of the insulating film 5 is also referred to as the upper face of the insulating film 5.

[0013] The semiconductor layer 6 has a fourth face F5 which is the face opposite to the insulating film 5. The fourth face F5 of the semiconductor layer 6 is also referred to as the upper face of the semiconductor layer 6. The semiconductor layer 6 is formed of a semiconductor such as silicon, germanium, or silicon germanium. The semiconductor layer 6 is preferably formed of a single crystal semiconductor doped with impurities such as boron (B) or phosphorus (P). By doping with impurities, carriers are generated in the semiconductor layer 6, and the resistivity can be reduced. In this embodiment, the semiconductor layer 6 is formed of single crystal silicon doped with boron to set the resistivity to 0.001 to 100 Ωcm. In this embodiment, the base 7 is an SOI (Silicon On Insulator) substrate having a recess 8 which is a cavity.

[0014] The semiconductor layer 6 is used to form the movable portion 2, the fixed electrode portion 3, the outer frame portion 9, and the elastic portion 13. In this embodiment, in a plan view in the Z direction, the movable portion 2 is formed inside the cavity recessed portion 8. A plurality of fixed electrode portions 3 are formed on both sides of the movable portion 2 in the X direction, sandwiching the movable portion 2 therebetween. The outer frame portion 9 is a substantially rectangular frame body surrounding the movable portion 2 and the fixed electrode portion 3. The outer frame portion 9 is formed on the third face F3, which is the upper surface of the insulating film 5, in the peripheral portion of the cavity recessed portion 8. The movable portion 2 and the outer frame portion 9 are connected via an elastic portion 13.

[0015] The movable section 2 has a movable electrode support section 10 and a plurality of movable electrode fingers 11 supported by the movable electrode support section 10. In this embodiment, the movable section 2 is a vibrator that is displaceable in the Y direction. The movable electrode support part 10 has a rectangular shape with its longer side in the Y direction. Elastic parts 13 are formed on both ends of the movable electrode support part 10 in the Y direction. Movable electrode fingers 11 are formed on both side surfaces of the movable electrode support part 10 in the X direction. The movable electrode fingers 11 have a cantilever shape with their free ends extending from the movable electrode support part 10 toward the fixed electrode part 3.

[0016] The fixed electrode portion 3 has a fixed electrode support portion 15 and fixed electrode fingers 16 . The fixed electrode support portion 15 is formed on the third face F3, which is the upper surface of the insulating film 5, in the periphery of the cavity recess 8. The fixed electrode fingers 16 have a cantilever shape with their free ends extending from the fixed electrode support portion 15 toward the movable portion 2. Movable electrode fingers 11 extending in the positive X-direction and fixed electrode fingers 16 extending in the negative X-direction are arranged opposite each other with a gap therebetween. Similarly, movable electrode fingers 11 extending in the negative X-direction and fixed electrode fingers 16 extending in the positive X-direction are arranged opposite each other with a gap therebetween. When movable part 2 is stationary, there is a predetermined gap between the side surfaces of movable electrode fingers 11 and fixed electrode fingers 16.

[0017] An electrode pad 27 is provided at one corner in the Y minus direction of the outer frame portion 9. A wiring 29 extends in the Y plus direction from the electrode pad 27. The extending wiring 29 branches in the direction of the fixed electrode support portion 15 (the X plus direction) and connects to a contact portion 26. 2, the contact portion 26 is electrically connected to the fixed electrode support portion 15 via the planarization insulating film 24. In other words, the electrode pad 27 is electrically connected to the fixed electrode support portion 15 via the wiring 29. In addition, an electrode pad 27 is provided at the other corner in the Y minus direction of the outer frame portion 9, and similarly, the electrode pad 27 is electrically connected to the fixed electrode support portion 15 via the wiring 29 and the contact portion 26.

[0018] The outer frame portion 9 is, for example, electrically grounded by wiring (not shown), and a potential difference is generated between the outer frame portion 9 and the fixed electrode support portion 15 . 2, an insulating separation part 20 is provided between the outer frame part 9 and the fixed electrode support part 15 to insulate them from each other. As described above, in the conventional technology, there was a risk of cracks occurring in the wiring 29 on the upper surface due to the influence of unevenness occurring on the surface of the insulating separation part 20, but this concern is eliminated by the insulating separation part 20 of this embodiment. Details of the insulating separation part 20 will be described later.

[0019] Such an acceleration sensor 1 can detect acceleration as follows: In this embodiment, the acceleration sensor 1 detects acceleration in the Y direction. When acceleration in the Y direction is applied, the movable part 2 is displaced in the Y direction relative to the base 7. As a result, the capacitance between the movable electrode fingers 11 of the movable part 2 and the fixed electrode fingers 16 of the fixed electrode part 3 changes. The acceleration in the Y direction can be detected based on this change in capacitance. Note that the detection of acceleration is not limited to the Y direction, and it is possible to detect acceleration in any direction. For example, acceleration in the X direction can be detected by rotating the acceleration sensor 1 by 90 degrees in a plane.

[0020] ***Insulation isolation section configuration*** FIG. 3 is an enlarged cross-sectional view of a portion c in FIG. 3, a trench portion 40 is formed in the semiconductor layer 6. The trench portion 40 is a bottomed groove with the insulating film 5 as the bottom. The trench portion 40 divides the semiconductor layer 6 into an outer frame portion 9 as a first semiconductor layer and a fixed electrode support portion 15 as a second semiconductor layer. The insulating separation portion 20 is composed of a first oxide film 21, a second oxide film 22, a planarizing insulating film 24, etc.

[0021] The first oxide film 21 and the second oxide film 22 are thermal oxide films that fill the trench portion 40. The first oxide film 21 is provided along a first side surface W1 of the trench portion 40 on the outer frame portion 9 side, with the insulating film 5 as a bottom, and extends onto the upper surface of the outer frame portion 9. The oxide film extending onto the upper surface of the outer frame portion 9 is referred to as a first oxide film 21b. As will be described in detail later, a thickness t1 of the first oxide film 21b is thinner than a thickness t2 of the first oxide film 21. The second oxide film 22 is provided along the second side surface W2 of the trench portion 40 on the side of the fixed electrode support portion 15, with the insulating film 5 serving as the bottom, and extends to the upper surface of the fixed electrode support portion 15. The oxide film extending to the upper surface of the fixed electrode support portion 15 is referred to as the second oxide film 22b. As with the first oxide film 21 side, the thickness t1 of the second oxide film 22b is thinner than the thickness t2 of the second oxide film 22. In other words, the first oxide film 21 and the second oxide film 22 are thermally-oxidized silicon.

[0022] In the trench portion 40, the first oxide film 21 and the second oxide film 22 are in physical contact with each other. A void portion 43 is formed at the portion where the first oxide film 21 and the second oxide film 22 are joined to the insulating film 5. The void portion 43 is formed in a portion surrounded by the insulating film 5, the first oxide film 21, and the second oxide film 22. When the void portion 43 is present, the dielectric constant in the void portion 43 becomes the dielectric constant ε0 of a vacuum. The dielectric constant ε0 of a vacuum is 8.85×10 -12 [F / m]. In general, the relative dielectric constant of an oxide film with respect to the dielectric constant ε0 of a vacuum is 3.8 to 3.9, so the fringe capacitance is reduced by the presence of void 43, which has a relative dielectric constant of 1. In other words, the parasitic capacitance between the first semiconductor layer and the second semiconductor layer can be reduced.

[0023] Moreover, the thickness of the insulating film 5 directly below the void 43 is thicker than the surrounding film thickness. The reason why the insulating film 5 is thick in this portion is that, during oxidation by thermal oxidation treatment, oxygen atoms penetrate and diffuse into the insulating film 5, and react with the substrate 4 to become a thermally oxidized film. The thick insulating film 5 in this portion can reduce the parasitic capacitance between the first semiconductor layer and the second semiconductor layer. Additionally, the upper and lower corners of the outer frame portion 9 are rounded. Similarly, the upper and lower corners of the fixed electrode support portion 15 are also rounded. In other words, the upper peripheral portion and the lower peripheral portion of the trench portion 40 are rounded. These rounded corners are formed by carrying out a thermal oxidation treatment for a sufficient period of time, which improves the symmetry of parasitic capacitances such as fringe capacitance.

[0024] A recess 41 is formed at the boundary portion on the upper surface of the first oxide film 21 and the second oxide film 22. Just as the corner of the outer frame portion 9 and the fixed electrode support portion 15 have a rounded corner, the boundary portion on the upper surface of the first oxide film 21 and the second oxide film 22 also has a curve. The gap gradually narrows from the upper surfaces of the first oxide film 21 and the second oxide film 22 to the region where they physically contact each other. From another perspective, the recess 41 can be expressed as a V-groove, a seam, a notch, etc. Such a recess 41 having a certain depth is filled with the planarization insulating film 24 of the upper layer. That is, the planarization insulating film 24 is uniformly present at the boundary portion of the first oxide film 21 and the second oxide film 22, and no void exists. In other words, the planarization insulating film 24 is embedded above the first oxide film 21 and the second oxide film 22 and between the first oxide film 21 and the second oxide film 22. In a preferred embodiment, the planarization insulating film 24 is a low dielectric constant layer. The low dielectric constant layer is an insulating layer formed of a material having a relative dielectric constant of 3.9 or less of a thermal oxide film, and preferably uses a spin-on-glass (SOG) material. The SOG material is a low dielectric constant material that can be applied and formed into a film by a spin coating method using an SOG solution containing siloxane as a main component, and examples of the SOG material that can be used include silica glass, alkylsiloxane polymers, alkylsilsesquioxane polymers, hydrogenated silsesquioxane polymers, and hydrogenated alkylsilsesquioxane polymers. The upper surface of the planarization insulating film 24 is planarized.

[0025] Wiring 29 is formed on the upper surface of the planarization insulating film 24. As a suitable example, wiring 29 is formed of a metal multilayer film. In other words, wiring 29 is provided on the planarization insulating film 24 and is electrically connected to the fixed electrode support portion 15. The metal multilayer film is preferably formed of a material that can ensure adhesion to the planarization insulating film 24, and for example, a metal multilayer film containing titanium nitride, aluminum, copper, or the like as a main component is used. The electrode pads 27, 28 (FIG. 1) are also made of a similar material.

[0026] A contact portion 26 that electrically connects the wiring 29 and the fixed electrode support portion 15 is provided between them. The contact portion 26 penetrates the second oxide film 22b and the planarization insulating film 24 on the upper surface of the fixed electrode support portion 15 to connect the wiring 29 and the fixed electrode support portion 15. The contact portion 26 is a via hole, and is made of a conductive material including a metal such as aluminum, copper, or tungsten.

[0027] ***How ​​acceleration sensors are manufactured*** Fig. 4 is a flow chart showing a method for manufacturing the acceleration sensor, Figs. 5 to 8 are cross-sectional views of the insulating separation part in the manufacturing process, and correspond to Fig. 3. Here, the method for manufacturing the acceleration sensor 1 will be described, focusing mainly on the insulating separation section 20.

[0028] In step S10, a base 7 (FIG. 1) is prepared. As described above, the base 7 is an SOI substrate having a semiconductor layer 6 and a recess 8 which is a cavity. In other words, in this step, the base 7 is prepared, which has a substrate 4, an insulating film 5 provided on a first face F1 which is a main face of the substrate 4, and a semiconductor layer 6 formed on the surface of the insulating film 5 opposite to the substrate 4.

[0029] In step S11, a trench portion 40 is formed in the semiconductor layer 6. More specifically, as shown in FIG. 5, a hard mask 48 having an opening in the portion that will become the trench portion 40 is formed, and the trench portion 40 is formed by using a DRIE (Deep Reactive Ion Etching) method. As a result, as shown in FIG. 5, a trench portion 40 is formed that divides the semiconductor layer 6 into two parts with the insulating film 5 as the bottom. In a preferred example, the hard mask 48 is formed by forming a thermal oxide film on the surface of the semiconductor layer 6 and patterning the thermal oxide film. Alternatively, a mask may be formed by forming a CVD (Chemical Vapor Deposition) film and patterning the CVD film. The size of the trench portion 40 is, for example, about 2 um wide x 30 um deep. In other words, in this step, a part of the semiconductor layer 6 is removed to form the trench portion 40, and the outer frame portion 9 as the first semiconductor layer and the fixed electrode support portion 15 as the second semiconductor layer that face each other through the trench portion 40. The trench portion 40 is formed in a portion where the cavity recess 8 does not exist.

[0030] In step S12, the trench portion 40 is filled with oxide. In a preferred embodiment, a wet oxidation process is performed. For example, a pyrogenic oxidation process is used to perform a thermal oxidation process at 1100° C. for 10 hours or more. For example, when an oxide film having a thickness of 1 um or more is formed from the first side W1 and the second side W2 of the trench portion 40 having a width of 2 um by the thermal oxidation process, the oxide films formed from both sides come into physical contact with each other, and the trench portion 40 can be filled. As a result, as shown in FIG. 6, the trench portion 40 is filled with the first oxide film 21 and the second oxide film 22. At this time, the first oxide film 21 and the second oxide film 22 are formed on the upper surface of the semiconductor layer 6 with substantially the same thickness as the thickness on the first side W1 and the second side W2.

[0031] In addition, a recess 47 is formed at the junction between the first oxide film 21 and the second oxide film 22 in the upper part of the trench portion 40. For example, when the trench portion 40 with a width of 2 um is oxidized by pyrogenic oxidation at 1100°C for 18.5 hours, a recess with an opening width of 0.6 um and a depth of 1.29 um is formed. At this time, the above-mentioned void portion 43, a portion with a thick film thickness of the insulating film 5, and a corner R are also formed by the thermal oxidation treatment. In a preferred example, the hard mask 48 remaining at the time of forming the trench portion 40 is not removed, and the thermal oxidation treatment is performed as it is. It is known that this causes the opening width to expand and the shape to be easily filled. For example, when the trench portion 40 with a width of 2 um is subjected to thermal oxidation treatment under the same oxidation conditions while leaving the hard mask 48 of about 0.88 um, a recess with an opening width of 2.33 um and a depth of 2.23 um is formed. Using the above process, the outer frame portion 9 and the fixed electrode support portion 15 are oxidized to form a first oxide film 21 and a second oxide film 22, and the first oxide film 21 and the second oxide film 22 are brought into physical contact with each other, thereby filling the trench portion 40.

[0032] In step S13, the first oxide film 21 and the second oxide film 22 formed on the upper surface of the trench portion 40 and the semiconductor layer 6 are etched back. In a preferred embodiment, the first oxide film 21 and the second oxide film 22 are dry etched. If wet etching is used, the recess 47 (FIG. 6) will be opened widely because it will be isotropic etching, but by using dry etching, it is possible to etch back while maintaining the opening width. The first oxide film 21 and the second oxide film 22 become thinner by the etch back, and the position of the recess 47 (FIG. 6) is lowered accordingly, forming the recess 41 (FIG. 7). As a result, the first oxide film 21b and the second oxide film 22b, which are thinned, are formed on the upper surface of the semiconductor layer 6 as shown in FIG. 7.

[0033] In the present embodiment, as a preferred example, the thicknesses of the first oxide film 21 and the second oxide film 22 on the first side W1 and the second side W2 are set as the side thickness t2 of the thermal oxide film, and the thicknesses of the first oxide film 21b and the second oxide film 22b after the etch-back are set as the top surface thickness t1 of the thermal oxide film, so as to satisfy the following formula (1). If the thermal oxide film is not etched back, the first oxide film 21 and the second oxide film 22 are too thick, which may cause contact failure in the contact portion 26 described later. Conversely, if the etch-back is excessive and the semiconductor layer 6 is exposed, a step is formed, which may cause unevenness to remain for a long time.

[0034] 0 < top thickness of thermal oxide film t1 < side thickness of thermal oxide film t2 ……Equation (1)

[0035] In step S14, the planarizing insulating film 24 is formed on the first oxide film 21b and the second oxide film 22b. In detail, an SOG solution mainly composed of an alkylsilsesquioxane polymer is applied by spin coating and cured by heat treatment. The heating temperature is preferably 300°C or more and 500°C or less, and more preferably 300°C or more and 450°C or less. In a preferred embodiment, the application and baking process are performed multiple times. As a result, as shown in FIG. 8, the gap between the first oxide film 21b and the second oxide film 22b can be filled. Even if the recess 41 has a curve and the gap gradually narrows until it reaches the region where the recess 41 physically contacts with the first oxide film 21b, the recess 41 can be filled with the planarizing insulating film 24 even if the recess 41 has a certain depth such as a V-groove, a seam, or a notch. That is, the planarizing insulating film 24 is uniformly present at the boundary between the first oxide film 21 and the second oxide film 22, and a configuration can be achieved in which no voids exist. Furthermore, the recesses 41 at the thermal oxide film boundary are reliably filled, and the upper surface of the planarization insulating film 24 can be made flat. Although the relative dielectric constant of the planarization insulating film 24 is 2.9 to 3.4, it is sufficient if it is 3.9 or less. If an SOG material with a dielectric constant of 2.0 or less is used, the mechanical strength of the planarization insulating film 24 may be reduced. In addition, with organic SOG materials having methyl groups, outgassing may affect device performance, so it is necessary to use them according to the application.

[0036] In step S15, the wiring 29 is formed on the planarization insulating film 24. First, prior to the formation of the wiring 29, the contact portion 26 (FIG. 3) is formed. Then, a wiring layer is formed on the planarization insulating film 24 and the contact portion 26 by using, for example, a CVD method or a vapor deposition method, and the wiring layer is patterned to form the wiring 29 shown in FIG. 3 (FIG. 1). When the wiring 29 is formed, other wirings (not shown) and electrode pads 27, 28 (FIG. 1) are also formed at the same time.

[0037] In step S16, the outer shapes of the movable part 2 (FIG. 1), fixed electrode part 3, outer frame part 9, etc. are formed. As described above, the movable part 2, fixed electrode part 3, outer frame part 9, etc. are formed using the semiconductor layer 6. In this embodiment, unnecessary parts of the semiconductor layer 6 are removed using a dry etching method or the like according to the outer shapes of each part, such as the movable part 2, fixed electrode part 3, outer frame part 9, etc. In some cases, the insulating layer 5, which becomes unnecessary, may also be removed. In this way, the acceleration sensor 1 shown in FIG. 1 is formed.

[0038] As described above, according to the acceleration sensor 1 and the manufacturing method for the acceleration sensor 1 of this embodiment, the following effects can be obtained. The acceleration sensor 1 as an inertial sensor comprises a substrate 4, an insulating film 5 provided on a first face F1 which is the main face of the substrate 4, an outer frame portion 9 as a first semiconductor layer and a fixed electrode support portion 15 as a second semiconductor layer provided on the face of the insulating film 5 opposite the substrate 4, a first oxide film 21 provided on a first side W1 of the outer frame portion 9 facing the fixed electrode support portion 15, a second oxide film 22 provided on a second side W2 of the fixed electrode support portion 15 facing the outer frame portion 9, a planarizing insulating film 24 embedded above the first oxide film 21 and the second oxide film 22 and between the first oxide film 21 and the second oxide film 22, and a wiring 29 provided on the planarizing insulating film 24 and electrically connected to the fixed electrode support portion 15.

[0039] According to this, the space between the first semiconductor layer and the second semiconductor layer is filled with a first oxide film 21 and a second oxide film 22, which are thermal oxide films. Therefore, unlike conventional inertial sensors in which a conductive material is inserted between semiconductors separated by a trench, which increases the parasitic capacitance between the semiconductors and causes deterioration of bias characteristics, etc., it is possible to reduce the parasitic capacitance, such as the fringe capacitance, between the first semiconductor layer and the second semiconductor layer, and therefore it is possible to provide an acceleration sensor 1 with good bias characteristics. Furthermore, the recess 41 occurring at the boundary portion on the upper surfaces of the first oxide film 21 and the second oxide film 22 is filled with the planarization insulating film 24, and the upper surface of the planarization insulating film 24 is planarized, so that even if wiring 29 is formed in the upper layer, there is no concern that defects such as cracks will occur. Therefore, it is possible to provide a highly reliable acceleration sensor 1.

[0040] In addition, when the thicknesses of the first oxide film 21 and the second oxide film 22 on the first side W1 and the second side W2 are defined as the side thickness t2 of the thermal oxide film, and the thicknesses of the first oxide film 21b and the second oxide film 22b after etch-back are defined as the top surface thickness t1 of the thermal oxide film, they are set to satisfy formula (1). If the thermal oxide film is not etched back, the first oxide film 21 and the second oxide film 22 will be too thick, which may cause poor contact at the contact portion 26. Conversely, if the film is etched back too much and the semiconductor layer 6 is exposed, a step will be formed, which may cause unevenness to remain for a long time. This makes it possible to optimize the thickness t1 of the first oxide film 21b and the second oxide film 22b, thereby ensuring the reliability of the wiring 29 formed in the upper layer.

[0041] The planarization insulating film 24 has a relative dielectric constant of 3.9 or less. According to this, the planarization insulating film 24 has the necessary strength and planarization performance, and is therefore highly reliable.

[0042] The first semiconductor layer and the second semiconductor layer are made of single crystal silicon, and the first oxide film 21 and the second oxide film 22 are made of thermally-oxidized silicon. Thermal silicon oxide is an insulator with higher quality and higher withstand voltage than silicon oxide produced by, for example, a CVD method, and therefore the insulation characteristics can be improved, and the reliability of the insulating isolation section 20 can be increased.

[0043] Also, there is a gap 43 surrounded by the insulating film 5, the first oxide film 21, and the second oxide film 22. With this, the fringe capacitance can be reduced due to the presence of the gap 43 whose relative dielectric constant is 1. Therefore, the parasitic capacitance between the first semiconductor layer and the second semiconductor layer is reduced, and an acceleration sensor 1 with good bias characteristics can be provided.

[0044] A manufacturing method of the acceleration sensor 1 includes the steps of preparing a base 7 having a substrate 4, an insulating film 5 provided on a first face F1 which is a main face of the substrate 4, and a semiconductor layer 6 formed on the surface of the insulating film 5 opposite the substrate 4; removing a portion of the semiconductor layer 6 to form a trench portion 40, and an outer frame portion 9 as a first semiconductor layer and a fixed electrode support portion 15 as a second semiconductor layer which face each other via the trench portion 40; thermally oxidizing the outer frame portion 9 and the fixed electrode support portion 15 to form a first oxide film 21 and a second oxide film 22, and filling the trench portion 40 by physically contacting the first oxide film 21 and the second oxide film 22; forming a planarizing insulating film 24 on the first oxide film 21 and the second oxide film 22 by a coating method; and forming wiring 29 on the planarizing insulating film 24.

[0045] According to this method, the space between the first semiconductor layer and the second semiconductor layer can be filled with the first oxide film 21 and the second oxide film 22, which are thermal oxide films, by thermal oxidation. Therefore, unlike conventional inertial sensors in which a conductive material is inserted between semiconductors separated by a trench, which increases the parasitic capacitance between the semiconductors and causes deterioration of bias characteristics, etc., it is possible to reduce the parasitic capacitance, such as the fringe capacitance, between the first semiconductor layer and the second semiconductor layer, and therefore it is possible to provide an acceleration sensor 1 with good bias characteristics. Furthermore, by applying an SOG material, the recesses 41 generated at the boundary between the upper surfaces of the first oxide film 21 and the second oxide film 22 can be filled with the planarization insulating film 24. Since the upper surface of the planarization insulating film 24 is planarized, there is no concern that defects such as cracks will occur even if the wiring 29 is formed in the upper layer. Therefore, it is possible to provide a manufacturing method capable of manufacturing a highly reliable acceleration sensor 1.

[0046] Moreover, the method for manufacturing the acceleration sensor 1 further includes, after the step of filling the trench portion 40, a step of etching back the first oxide film and the second oxide film formed on the upper surfaces of the first semiconductor layer and the second semiconductor layer. According to this, the thickness t1 of the first oxide film 21b and the second oxide film 22b can be optimized, so that the reliability of the wiring 29 formed in the upper layer can be ensured.

[0047] The planarization insulating film 24 is formed by applying an SOG material and then performing a heat treatment, and the temperature of the heat treatment is 300° C. or more and 500° C. or less. This makes it possible to reliably fill the recess 41 at the thermal oxide film boundary and to flatten the upper surface of the planarization insulating film 24. This ensures the reliability of the wiring 29 formed in the upper layer.

[0048] EMBODIMENT 2 ***Different configurations of insulating isolation parts*** FIG. 9 is a cross-sectional view of an insulating isolation portion according to the second embodiment, and corresponds to FIG. In the above embodiment, the trench portion 40 is formed using the DRIE method, but this is not limited to this method, and any method can be used as long as it can form a trench portion that divides the first semiconductor layer and the second semiconductor layer with the insulating film 5 as the bottom. For example, in this embodiment, the Bosch process is used to form the trench portion 44. Hereinafter, the same parts as those in the above embodiment are denoted by the same reference numbers, and duplicated explanations will be omitted.

[0049] In the insulating isolation portion 25 of this embodiment shown in FIG. 9, the trench portion 44 is formed by the Bosch process. The Bosch process is a type of DRIE, and is a process in which etching and deposition of a protective layer are alternately performed. When the semiconductor layer 6 is anisotropically etched by this process, shell-shaped irregularities called scallops are formed on the first side surface W1 and the second side surface W2 of the trench portion 44. When a thermal oxidation process is performed with this irregularity formed, the shape is inherited by the thermal oxide film.

[0050] 9, in insulating isolation portion 25, a plurality of scallops 49 are formed at the boundary between first oxide film 21 and second oxide film 22. Note that, apart from the presence of scallops 49, the description is the same as that in FIG. Furthermore, even if scallops 49 are present, the recesses 41 that occur at the boundary between the first oxide film 21 and the second oxide film 22 can be filled by applying an SOG solution, and a planarized insulating film 24 with a planarized surface can be formed.

[0051] As described above, according to the acceleration sensor 1 and the manufacturing method of the acceleration sensor 1 of this embodiment, in addition to the effects of the first embodiment, the following effects can be obtained. Even for insulating isolation parts 25 having scallops 49, a planarized insulating film 24 with a planarized surface can be formed by applying an SOG solution. Furthermore, the insulating performance of insulating isolation parts 25 and the reliability of wiring 29 can be secured to be comparable to those of insulating isolation part 20 in FIG. 3. Therefore, a highly reliable acceleration sensor 1 can be provided. Also, a manufacturing method capable of manufacturing a highly reliable acceleration sensor 1 can be provided.

[0052] EMBODIMENT 3 ***Application of insulating isolation parts*** FIG. 10 is a plan view of the acceleration sensor according to the third embodiment. In the above embodiment, an example was described in which the insulating isolation parts 20, 25 are used for the purpose of insulating and isolating semiconductor elements and ensuring the reliability of wiring, but the use is not limited to this. For example, the insulating isolation parts may be used for three-dimensional intersections of wiring.

[0053] ***Accelerometer Overview*** An acceleration sensor 51 of this embodiment shown in FIG. 10 is an acceleration sensor that detects acceleration in the Z-axis direction, and employs a so-called one-sided seesaw structure in which a movable body 55 oscillates around an oscillating shaft 61. The acceleration sensor 51 is composed of a fixed part 80, a movable body 55 that is swingable around a swing axis 61 that passes through the center of the fixed part 80 and is aligned along the X-axis, and a first rotating spring 54a, a second rotating spring 54b, and the like that connect the fixed part 80 and the movable body 55. The fixed part 80 is fixed to a base that protrudes from a base body (not shown). The periphery of the base is a cavity that allows the movable body 55 to swing.

[0054] The movable body 55 has a first bar 52a extending in the Y positive direction from the first rotating spring 54a, a second bar 52b extending in the Y positive direction from the second rotating spring 54b, and a third bar 53 connecting the first bar 52a and the second bar 52b. The third bar 53 is provided with four comb-shaped movable electrodes 73a to 73d. Movable electrode group 73a is composed of seven movable electrodes 71c extending in the positive Y direction from third bar 53 on the negative X side of center line 60. Movable electrode group 73b is composed of seven movable electrodes 71c extending in the negative Y direction from third bar 53 on the negative X side of center line 60. Note that the number of movable electrodes 71c is not limited to seven, and it is sufficient that the number of movable electrodes 71c is more than one. The movable electrode groups 73c and 73d are provided at positions symmetrical to the movable electrode groups 73a and 73b on the X-positive side with respect to the center line 60 as an axis of symmetry.

[0055] The base body is provided with a group of fixed electrodes 74a to 74d that face the group of movable electrodes 73a to 73d. The fixed electrode group 74a is composed of a support portion 75a fixed to the base and six fixed electrodes 72 extending in the negative Y direction from the support portion 75a. The fixed electrode group 74b is composed of a support portion 75b fixed to the base and six fixed electrodes 72 extending in the positive Y direction from the support portion 75b. Note that the number of fixed electrodes 72 is not limited to six, and may be any number corresponding to the number of movable electrodes 71c. The fixed electrode groups 74c and 74d are provided at positions symmetrical to the fixed electrode groups 74a and 74b on the X-positive side with the center line 60 as the axis of symmetry.

[0056] The detection section formed by the fixed electrode group 74a and the movable electrode group 73a and the detection section formed by the fixed electrode group 74b and the movable electrode group 73b are collectively referred to as an N-type detection section 76n. In the N-type detection section 76n, a parallel plate type capacitance is formed by the fixed electrode 72 and the movable electrode 71c arranged opposite to each other. The capacitance changes according to the change in the overlapping area between the fixed electrode 72 and the movable electrode 71c as the movable electrode 71c is displaced due to acceleration. Similarly, a detection section formed by fixed electrode group 74c and movable electrode group 73c and a detection section formed by fixed electrode group 74d and movable electrode group 73d are collectively referred to as P-type detection section 76p. In P-type detection section 76p, a parallel plate type capacitance is formed by fixed electrode 72c and movable electrode 71 arranged opposite to each other. The capacitance changes in accordance with a change in the overlapping area with fixed electrode 72c as movable electrode 71 is displaced due to acceleration.

[0057] The movable electrode 71c of the N-type detection portion 76n is thinner in the Z direction than the movable electrode 71 of the P-type detection portion 76p. More specifically, the movable electrode 71c is thinned by being cut out in a stepped shape midway along the extension direction from the same thickness as the third bar 53 at its base. As a result, the thickness of the seven movable electrodes 71c on the Z positive side is thinner in the portion facing the fixed electrode 72. The fixed electrodes 72c of the P-type detection portion 76p are thinner in the Z direction than the fixed electrodes 72 of the N-type detection portion 76n. More specifically, the fixed electrodes 72c are thinned by being cut out in a stepped shape midway along the extension direction from the thickness at the base on the support portions 75c, 75d side. As a result, the thickness of each of the six fixed electrodes 72c on the Z positive side is thinner in the portion facing the movable electrode 71.

[0058] With this configuration, when acceleration occurs in the positive Z direction, the overlapping area of ​​the N-type detection portion 76n decreases, while the overlapping area of ​​the P-type detection portion 76p is maintained. Also, when acceleration occurs in the negative Z direction, the overlapping area of ​​the N-type detection portion 76n is maintained, while the overlapping area of ​​the P-type detection portion 76p decreases. Based on this correlation, the acceleration sensor 51 can detect acceleration in the positive / negative Z directions by detecting a change in the overlapping area of ​​the N-type detection portion 76n and the P-type detection portion 76p as a change in capacitance.

[0059] ***Application to wiring of insulating separation parts*** Here, a description will be given of an example of a three-dimensional intersection of wiring to which the insulating separation section 20 is applied. Note that, although many other wirings are actually provided, the following description will focus on wirings to which the insulating separation section 20 is applied. As described above, the fixed electrode groups 74a and 74b in the N-type detection unit 76n form a pair, and are therefore electrically connected to each other by the wiring 81. The wiring 81 passes through the outer frame 65 in a large crank shape from the support portion 75a of the fixed electrode group 74a, extends in the Y positive direction along the center line 60, and connects to the support portion 75b of the fixed electrode group 74b. The wiring 81 also branches midway and connects to an electrode pad 92. The electrode pad 92 outputs a voltage detected by the N-type detection unit 76n.

[0060] Here, an insulating separation portion 20 is provided between the support portion 75a and the outer frame 65, and the wiring 81 is formed on the insulating separation portion 20. Similarly, an insulating separation portion 20 is provided between the support portion 75b and the semiconductor portion 96, and the wiring 81 is formed on the insulating separation portion 20. The semiconductor portion 96 is connected to the outer frame 65 at a portion along the center line 60. In this manner, the insulating separation portion 20 is used in each portion of the acceleration sensor 51.

[0061] A semiconductor portion 90 is provided in the negative X direction of support portion 75a. Semiconductor portion 90 is, for example, a stopper portion, and is a portion that limits the in-plane displacement of movable body 55. The same potential as that of movable body 55 is applied to semiconductor portion 90 by wiring 82a. The wiring 82a passes from the semiconductor portion 90 through the outer frame 65, straddles the wiring 81, and is connected to the wiring 82b, which is connected to the electrode pad 91. An insulating separation portion 20 is provided between the semiconductor portion 90 and the outer frame 65, and the wiring 82a is formed on the insulating separation portion 20.

[0062] Fig. 11 is an enlarged view of a portion f in Fig. 10. Fig. 12 is a cross-sectional view taken along line gg in Fig. 11. Here, a case of a three-dimensional intersection using the insulating separation part 20 between the wiring 81 and the wirings 82a and 82b will be described. As shown in Fig. 11, the insulating isolation part 20 in this embodiment is formed in an elliptical shape and in a closed loop shape. Inside the ellipse is an intersection of the wiring 81 and wirings 82a and 82b. As a result, as shown in Fig. 12, the semiconductor part 68 defined by the insulating isolation part 20 is separated from the semiconductor layer of the surrounding outer frame 65, and becomes an electrically independent part. Furthermore, by forming the insulating isolation part 20 in a closed loop shape, the number of insulating isolation ends can be reduced, and the yield can be improved.

[0063] When two wirings are crossed over each other, the semiconductor portion 68 is used as the wiring. More specifically, as shown in FIG. 12, an electrical conduction path is formed by the path of the wiring 82a, the contact portion 26, the semiconductor portion 68, the contact portion 26, and the wiring 82b. Such a crossing configuration is called a crossing portion 20c. The crossing portion 20c is not limited to be applied to the wirings 81 and 82, but can be applied to a portion where two wirings with different potentials cross each other. Furthermore, as shown in FIG. 11, even if multiple wirings pass over the top of the insulating separation portion 20, no cracks are generated in the wiring, and the reliability is excellent. Note that the insulating separation portion 25 may be applied instead of the insulating separation portion 20, and the same effect can be obtained.

[0064] Return to Figure 10. The acceleration sensor 51 has a substantially symmetrical configuration with the center line 60 as the axis of symmetry, and the same parts are given the same numbers. A semiconductor section 90 is also provided on the X-positive side (right side) of the center line 60, and the same potential is applied to the semiconductor section 90 on the X-negative side (left side). The right-side wiring 82b electrically connected to the semiconductor section 90 passes through the outer frame 65 and connects to wiring 83c via the crossing 20c. The wiring 83c connects to wiring 83b via the crossing 20c. The wiring 83b connects to wiring 83a via the crossing 20c, and the wiring 83a connects to the left-side wiring 82b via the crossing 20c. In this way, the crossing 20c is used in multiple places. Moreover, the wiring 83a branches midway, passes over the insulating separation part 20, and is also connected to the fixed part 80. The fixed part 80 is connected to the first bar 52a via the first rotating spring 54a. In other words, the electrode pad 91 applies a potential to the movable body 55 via the wiring 83a. Note that in FIG. 10, the wiring 83c also branches midway, passes over the insulating separation part 20, and is connected to the fixed part 80, but electrically, it is sufficient that either one of the wirings 83a or 83c is connected to the fixed part 80.

[0065] As described above, according to the acceleration sensor 51 of this embodiment, in addition to the effects of the above-mentioned embodiment, the following effects can be obtained. According to this, the insulating separation parts 20 and 25 can also be applied to the acceleration sensor 51 that detects acceleration in the Z-axis direction. Furthermore, the insulating separation parts 20 and 25 can realize a three-dimensional intersection at the portion where two wirings with different potentials intersect. Therefore, it is possible to provide a highly reliable acceleration sensor 51. It is also possible to provide a highly reliable three-dimensional wiring crossing structure.

[0066] EMBODIMENT 4 ***Inertial Measurement Unit Overview*** Fig. 13 is an exploded perspective view of the inertial measurement unit, and Fig. 14 is a perspective view of the circuit board. The inertial measurement unit 2000 of this embodiment is equipped with the above-mentioned acceleration sensor 1.

[0067] The inertial measurement unit 2000 is a device that detects the inertial momentum of the attitude, behavior, etc. of a moving body such as an automobile, a robot, etc. The inertial measurement unit 2000 includes inertial sensors such as an acceleration sensor and an angular velocity sensor, and functions as a so-called motion sensor.

[0068] 13, the inertial measurement unit 2000 has a rectangular parallelepiped shape that is a substantially square in plan view. The inertial measurement unit 2000 has an outer case 301, a joint member 310, and a sensor module 325 in which an inertial sensor is mounted.

[0069] The external shape of outer case 301 is a rectangular parallelepiped with a substantially square planar shape, similar to the overall shape of inertial measurement unit 2000, and screw holes 302 are formed near two vertices located diagonally of the square. By passing two screws through these two screw holes 302, inertial measurement unit 2000 can be fixed to the mounting surface of an object to which it is to be mounted, such as an automobile.

[0070] Further, outer case 301 is box-shaped and houses sensor module 325 therein. Specifically, sensor module 325 is inserted into outer case 301 with joint member 310 interposed therebetween.

[0071] The sensor module 325 includes an inner case 320 and a substrate 315 . Inner case 320 is a member that supports substrate 315, and substrate 315 is bonded to the bottom surface of inner case 320 via an adhesive.

[0072] Moreover, inner case 320 is shaped to fit inside outer case 301. Inner case 320 is formed with recess 331 for preventing contact with substrate 315 and opening 321 for exposing connector 316 (described later). Inner case 320 is joined to outer case 301 via joining member 310.

[0073] Next, the substrate 315 on which the inertial sensor is mounted will be described. 14, acceleration sensor 1, connector 316, angular velocity sensor 317z that detects angular velocity around the Z axis, etc. are mounted on the top surface of substrate 315, which faces inner case 320. Angular velocity sensor 317x that detects angular velocity around the X axis and angular velocity sensor 317y that detects angular velocity around the Y axis are mounted on the side surface of substrate 315.

[0074] Incidentally, the acceleration sensor 1 may be an acceleration sensor capable of detecting acceleration in two directions, the X direction and the Y direction, or an acceleration sensor capable of detecting acceleration in three directions, the X direction, the Y direction, and the Z direction, as necessary.

[0075] Furthermore, a control IC 319 is mounted as a control unit on the underside of substrate 315 facing outer case 301. Control IC 319 is an MCU (Micro Controller Unit) that incorporates a storage unit including a non-volatile memory, an A / D converter, and the like, and controls each unit of inertial measurement unit 2000. The storage unit stores a program that defines the sequence and content for detecting acceleration and angular velocity, a program that digitizes the detected data and incorporates it into packet data, associated data, and the like. Note that a plurality of other electronic components are also mounted on substrate 315.

[0076] According to the inertial measurement unit 2000, since the acceleration sensor 1 is used as an example of the inertial sensor described above, it is possible to provide an inertial measurement unit 2000 that enjoys the effects associated with the acceleration sensor 1. [Explanation of symbols]

[0077] 1...accelerometer, t1...thickness, t2...thickness, 2...movable portion, 3...fixed electrode portion, 4...substrate, 5...insulating film, 6...semiconductor layer, 7...base, 8...recess, 9...outer frame portion, 10...movable electrode support portion, 11...movable electrode finger, 13...elastic portion, 15...fixed electrode support portion, 16...fixed electrode finger, 20...insulating separation portion, 20c...multi-level intersection portion, 21...first oxide film, 21b...first oxide film, 22...second oxide film, 22b...second oxide film, 24...planarization Insulating film, 25...insulating separation portion, 26...contact portion, 27, 28...electrode pad, 29...wiring, 40...trench portion, 41...recess, 43...void portion, 44...trench portion, 47...recess, 48...hard mask, 49...scallop, 51...accelerometer, 52a...first bar, 52b...second bar, 53...third bar, 54a...first rotating spring, 54b...second rotating spring, 55...movable body, 60...center line, 61...oscillating axis, 65... Outer frame, 68... semiconductor portion, 71... movable electrode, 71c... movable electrode, 72... fixed electrode, 72c... fixed electrode, 73a to 73d... movable electrode group, 74a to 74d... fixed electrode group, 75a to 75c... support portion, 76n... N-type detection portion, 76p... P-type detection portion, 80... fixed portion, 81... wiring, 82... wiring, 82a... wiring, 82b... wiring, 90... semiconductor portion, 91... electrode pad, 92... electrode pad, 96... semiconductor portion, 301... outer case base, 302...screw hole, 310...jointing member, 315...board, 316...connector, 317x...angular velocity sensor, 317y...angular velocity sensor, 317z...angular velocity sensor, 320...inner case, 321...opening, 325...sensor module, 331...recess, 2000...inertial measurement unit, 319...control IC, F1...first surface, F2...second surface, F3...third surface, F5...fourth surface, W1...first side, W2...second side.

Claims

1. A substrate, an insulating film provided on a main surface of the substrate, a first semiconductor layer and a second semiconductor layer provided on a surface of the insulating film opposite to a surface on the substrate side, 2 semi conductor layers, a first oxide film provided on a first side surface of the first semiconductor layer on the second semiconductor layer side, a second oxide film provided on a second side surface of the second semiconductor layer on the first semiconductor layer side, a planarization insulating film provided above the first oxide film and the second oxide film and embedded between the first oxide film and the second oxide film, and a wiring provided on the planarization insulating film and electrically connected to the second semiconductor layer including, 、 an inertial sensor.

2. In Claim 1, when the thicknesses of the first oxide film and the second oxide film on the first side surface and the second side surface are defined as the side surface thickness t2 of the thermal oxide film, and the thicknesses of the upper portions of the first oxide film and the second oxide film are defined as the upper surface thickness t1 of the thermal oxide film, 0 < upper surface thickness t1 of the thermal oxide film < side surface thickness t2 of the thermal oxide film is satisfied, an inertial sensor.

3. In Claim 1, the relative permittivity of the planarization insulating film is 3.9 or less, an inertial sensor.

4. In Claim 1, the first semiconductor layer and the second semiconductor layer are single crystal silicon, the first oxide film and the second oxide film are thermal oxide silicon, an inertial sensor.

5. In Claim 1, a trench portion is provided between the first semiconductor layer and the second semiconductor layer, the first oxide film and the second oxide film are embedded in the trench portion, the trench portion is provided with a void portion surrounded by the insulating film, the first oxide film, and the second oxide film, an inertial sensor.

6. A step of preparing a substrate having a substrate, an insulating film formed on a main surface of the substrate, and a semiconductor layer formed on a surface of the insulating film opposite to the substrate side; a step of forming a trench portion, a first semiconductor layer and a second semiconductor layer facing each other through the trench portion by removing a part of the semiconductor layer; a step of thermally oxidizing the first semiconductor layer and the second semiconductor layer to form a first oxide film and a second oxide film, and embedding the trench portion by physically contacting the first oxide film and the second oxide film; a step of forming a planarization insulating film on the first oxide film and the second oxide film by a coating method; a step of forming a wiring on the planarization insulating film; including, a method for manufacturing an inertial sensor. ​ ​ ​ ​ ​ ​ ​ ​ **Claim 7** In claim 6, after the step of embedding the trench portion, a step of etching back the first oxide film and the second oxide film formed on the upper surfaces of the first semiconductor layer and the second semiconductor layer, further included, method for manufacturing an inertial sensor. **Claim 8** In claim 7, the planarization insulating film is formed by applying an SOG material and then performing a heat treatment, the temperature of the heat treatment is 300°C or higher and 500°C or lower, method for manufacturing an inertial sensor. **Claim 9** an inertial sensor according to any one of claims 1 to 5, a control unit for controlling the inertial sensor, included, inertial measurement device.

Images