Inertial sensor, and manufacturing method of inertial sensor

Abstract

To provide an inertial sensor with high bonding strength between a substrate and a lid body and excellent long-term reliability.SOLUTION: An inertial sensor according to the present invention is an electrostatic capacity variation type inertial sensor including: a substrate; a lid body; a functional element provided between the substrate and the lid body; and a metal eutectic layer configured to bond the substrate and the lid body to each other on a periphery of the functional element. The metal eutectic layer comprises a plurality of first regions which are composed mainly of a first metal and have a face-centered cubic lattice structure, and second regions which are composed mainly of a second metal and have a diamond structure, and these regions are adjacent to each other.SELECTED DRAWING: Figure 3

Description

Technical Field

[0001] The present invention relates to an inertial sensor and a method for manufacturing the inertial sensor.

Background Art

[0002] Conventionally, a sensor device including a substrate having a cavity, a sensor element suspended in the cavity, and a lid for sealing the cavity has been known. The substrate and the lid were joined via a bonding material. The bonding material was required to have high bonding strength and high long-term reliability of sealing.

[0003] For example, Patent Document 1 discloses using an AlGe eutectic as the bonding material. According to the document, the concentration of Ge in the AlGe eutectic is uniform or a function of the distance from the lid or the substrate. In particular, when performing a long-time heat treatment, it is disclosed that the concentration of Ge becomes uniform.

Prior Art Documents

Patent Documents

[0004]

Patent Document 1

Summary of the Invention

Problems to be Solved by the Invention

[0005] However, in the technique of Patent Document 1, there is a risk that the bonding strength by the bonding material may decrease. Also, there is a risk that the long-term reliability of the sealing may decrease. Specifically, when the concentration of Ge decreases according to the distance from the lid, AlGe eutectification cannot be achieved on the substrate side and only the Al layer remains, which may reduce the bonding strength. In addition, although long-time heat treatment is preferable for forming a bonding layer with a uniform Ge concentration, it generates hillocks in other metal wirings, such as Al wirings for extracting detection signals, and reduces the long-term reliability. That is, there has been a demand for an inertial sensor having high bonding strength between a base body and a lid body and excellent long-term reliability, and a method for manufacturing the inertial sensor.

Means for Solving the Problems

[0006] An inertial sensor according to one aspect of the present application is a capacitance change type inertial sensor, and includes a base body, a lid body, a functional element provided between the base body and the lid body, and a metal eutectic layer that joins the base body and the lid body around the functional element. The metal eutectic layer includes a plurality of first regions mainly composed of a first metal and having a face-centered cubic lattice structure, and a plurality of second regions mainly composed of a second metal and having a diamond structure, and the first regions and the second regions are adjacent to each other.

[0007] A method for manufacturing an inertial sensor according to one aspect of the present application is a manufacturing method for joining a base body provided with a functional element and a lid body that covers the functional element in a joining region surrounding the functional element. The method includes a step of forming a first joining portion mainly composed of a first metal in the joining region of the base body, a step of forming a second joining portion mainly composed of a second metal in the joining region of the lid body, a step of aligning the first joining portion and the second joining portion so as to overlap each other and stacking the base body and the lid body to form a laminate, a heating step of heating the laminate, and a weighting step of applying a weight to the laminate.

Brief Description of the Drawings

[0008]

Figure 1

Figure 2

Figure 3

Figure 4

Figure 5

Figure 6

Figure 7

Figure 8

Figure 9

Figure 10

Figure 11

Figure 12

Figure 13

Figure 14

Figure 15

Mode for Carrying Out the Invention

[0009] Embodiment 1 ***Configuration of the Inertial Sensor*** FIG. 1 is a plan view of the inertial sensor according to Embodiment 1. FIG. 2 is a cross-sectional view of the inertial sensor taken along the b-b cross-section of FIG. 1. The configuration of the inertial sensor 100 according to the present embodiment will be described with reference to FIGS. 1 and 2.

[0010] The inertial sensor 100 is, for example, an acceleration sensor that detects acceleration in the vertical direction. In each figure, the X-axis, Y-axis, and Z-axis, which are three mutually orthogonal axes, are shown. In the present embodiment, the Z-axis direction is taken as the vertical direction, but it is not limited thereto. The direction along the X-axis is referred to as the "X direction", the direction along the Y-axis is referred to as the "Y direction", and the direction along the Z-axis is referred to as the "Z direction". Also, the tip side of the arrow in each axis direction is also referred to as the "plus side", and the base side of the arrow is also referred to as the "minus side". For example, the Y direction refers to both the Y direction plus side and the Y direction minus side. Also, the Z direction plus side is referred to as "up", and the Z direction minus side is referred to as "down". Also, in the following figures, for the sake of clarity of explanation, there may be cases where the dimensions and scales are described differently from the actual ones.

[0011] The inertial sensor 100 is a uniaxial acceleration sensor composed of a MEMS (Micro Electro Mechanical Systems) device. Note that it is not limited to an acceleration sensor, and any capacitive change type inertial sensor can be used, for example, an angular velocity sensor may also be used. As shown in FIG. 2, the inertial sensor 100 includes a substrate 10, a sensor element 80 disposed on the substrate 10, a wiring layer 7 drawn from the sensor element 80, a lid 30 covering the sensor element 80, and the like. The substrate 10 is a SOI (Silicon On Insulator) substrate, and the substrate 1, the insulating layer 2, the semiconductor layer 3, and the wiring layer 7 are laminated in this order along the Z direction. The substrate 1 is a single crystal silicon substrate, and an insulating layer 2 is provided on its upper surface. The insulating layer 2 is a buried insulating layer made of SiO2.

[0012] The substrate 1 is provided with a recess 5 dug from the peripheral portion. The recess 5 is a cavity and is a part that forms a storage space S for storing the sensor element 80. Due to the recess 5, the movable body 55 (FIG. 1) of the sensor element 80 can swing. In FIG. 2, an insulating layer 2 is provided on the bottom surface of the recess 5, but it may not be provided. The semiconductor layer 3 is, for example, a conductive silicon substrate doped with impurities such as phosphorus (P), boron (B), and arsenic (As). In a preferred example, the semiconductor layer 3 and the insulating layer 2 are joined by a Si - SiO2 fusion bond.

[0013] The sensor element 80 is a functional element and is formed by etching and patterning the semiconductor layer 3. In a preferred example, a deep etching technique using a Bosch process is used. The sensor element 80 is fixed to the substrate 1 at the fixing portion 65 (FIG. 1). The lid 30 uses a silicon substrate as a preferred example. The lid 30 is provided with a recess 35 dug from the peripheral portion. The recess 35 is a part that forms a storage space S for housing the sensor element 80. The recess 35 is provided with a stopper portion 31 which is a convex portion for restricting the over-swing of the movable body 55 of the sensor element 80. The base body 10 and the lid 30 are joined by a metal eutectic layer 20 at their peripheral portions. The storage space S is hermetically sealed with an inert gas such as nitrogen, helium, or argon in a preferred example. It is preferable that the pressure is approximately atmospheric pressure or in a vacuum state in a use temperature environment of about -40°C to 120°C. For example, when the sensor element 80 is an acceleration sensor, the storage space S preferably has a pressure close to atmospheric pressure, and when the sensor element 80 is an angular velocity sensor, the storage space S preferably has a vacuum pressure.

[0014] ***Configuration of the sensor element*** As shown in FIG. 1, the sensor element 80 as a functional element is an acceleration sensor that detects acceleration in the Z direction, and adopts a so-called single-sided seesaw structure in which the movable body 55 swings about the swing axis 61. The sensor element 80 includes a fixed portion 65, a movable body 55 that can swing around a swing axis 61 along the Y axis passing through the center of the fixed portion 65, a first torsion spring 54a, a second torsion spring 54b, etc. that connect the fixed portion 65 and the movable body 55. The fixed portion 65 is fixed to a pedestal portion (not shown) protruding from the substrate 1 (FIG. 2). The periphery of the pedestal portion is a recess 5 (FIG. 2), and the movable body 55 is configured to be able to swing. In FIG. 1, a line segment that is orthogonal to the swing axis 61 and passes through the center of the sensor element 80 along the X axis is defined as the center line 60.

[0015] The movable body 55 has a first bar 52a extending in the X plus direction from the first torsion spring 54a, a second bar 52b extending in the X plus direction from the second torsion spring 54b, and a third bar 53 connecting the first bar 52a and the second bar 52b. Four comb-shaped movable electrode groups 73a to 73d are provided on the third bar 53. The movable electrode group 73a is composed of six movable electrodes 71c extending in the +X direction from the third bar 53 on the -Y side of the center line 60. The movable electrode group 73b is composed of six movable electrodes 71c extending in the -X direction from the third bar 53 on the -Y side of the center line 60. Note that the number is not limited to six, and any plural number of movable electrodes 71c may be used. The movable electrode groups 73c and 73d are provided at positions that are line-symmetrical to the movable electrode groups 73a and 73b on the +Y side with the center line 60 as the axis of symmetry.

[0016] On the substrate 1 (Figure 2) side, fixed electrode groups 74a to 74d facing the movable electrode groups 73a to 73d are provided. The fixed electrode group 74a is composed of a support portion 75a fixed to the substrate 1 and seven fixed electrodes 72 extending in the -X direction from the support portion 75a. The fixed electrode group 74b is composed of a support portion 75b fixed to the substrate 1 and seven fixed electrodes 72 extending in the +X direction from the support portion 75b. Note that the number is not limited to seven, and any number corresponding to the number of movable electrodes 71c may be used. The fixed electrode groups 74c and 74d are provided at positions that are line-symmetrical to the fixed electrode groups 74a and 74b on the +Y side with the center line 60 as the axis of symmetry.

[0017] The detection units formed by the fixed electrode group 74a and the movable electrode group 73a, and the detection units formed by the fixed electrode group 74b and the movable electrode group 73b are collectively referred to as the N-type detection unit 76n. In the N-type detection unit 76n, a parallel plate capacitor is formed by the opposed fixed electrodes 72 and movable electrodes 71c. This capacitor changes in accordance with the change in the overlapping area between the fixed electrode 72 and the movable electrode 71c as the displacement of the movable electrode 71c due to acceleration changes. Similarly, the detection units formed by the fixed electrode group 74c and the movable electrode group 73c, and the detection units formed by the fixed electrode group 74d and the movable electrode group 73d are combined to form a P-type detection unit 76p. In the P-type detection unit 76p, a parallel-plate capacitor is formed by the fixed electrode 72c and the movable electrode 71 arranged opposite to each other. This capacitor changes according to the change in the overlapping area between the fixed electrode 72c and the movable electrode 71 as the displacement of the movable electrode 71 due to acceleration occurs.

[0018] The movable electrode 71c of the N-type detection unit 76n is thinner in the Z-direction thickness than the movable electrode 71 of the P-type detection unit 76p. Specifically, the movable electrode 71c is notched stepwise in the middle of the extending direction from the same thickness as the base third bar 53 and becomes thinner. As a result, in all 12 movable electrodes 71c, the thickness on the Z-plus side is thinner at the portion facing the fixed electrode 72. The fixed electrode 72c of the P-type detection unit 76p is thinner in the Z-direction thickness than the fixed electrode 72 of the N-type detection unit 76n. Specifically, the fixed electrode 72c is notched stepwise in the middle of the extending direction from the thickness at the base on the support part 75c, 75d side and becomes thinner. As a result, in all 14 fixed electrodes 72c, the thickness on the Z-plus side is thinner at the portion facing the movable electrode 71.

[0019] With such a configuration, when an acceleration occurs in the Z-plus direction, the overlapping area decreases in the N-type detection unit 76n, and the overlapping area is maintained in the P-type detection unit 76p. Also, when an acceleration in the Z-minus direction occurs, the overlapping area is maintained in the N-type detection unit 76n, and the overlapping area decreases in the P-type detection unit 76p. Based on such a correlation, in the sensor element 80, by differentially detecting the change in the overlapping area in the N-type detection unit 76n and the P-type detection unit 76p as a capacitance change, the acceleration in the Z-plus / minus direction can be detected.

[0020] ***Planar shape of the bonding region*** As shown in FIG. 1, the base body 10 is substantially rectangular, and the short side in the X minus direction is an overhanging portion 11 that protrudes from the short side of the lid body 30. Terminals 91 to 94 for external connection are provided on the overhanging portion 11. Terminal 91 is a movable electrode terminal and is electrically connected to all the movable electrodes 71, 71c by a wiring 81. Terminal 92 is an N-type fixed electrode terminal and is electrically connected to all the fixed electrodes 72 of the N-type detection unit 76n by a wiring 82. Terminal 93 is a P-type fixed electrode terminal and is electrically connected to all the fixed electrodes 72c of the P-type detection unit 76p by a wiring 83. Terminal 94 is a GND terminal and is electrically connected to the metal eutectic layer 20 by a wiring 84. The details of the connection form between the terminal 94 and the metal eutectic layer 20 will be described later.

[0021] The base body 10 and the lid body 30 are joined at a square annular joining region 20a that surrounds the periphery of the sensor element 80. The joining region 20a is a square annular region that is slightly smaller than the outer peripheral edge of the lid body 30. A metal eutectic layer 20 is formed in the joining region 20a. Further, the metal eutectic layer 20 is formed so as to intersect the wirings 81 to 83 in a plan view. In other words, the capacitance change type inertial sensor 100 includes a base body 10, a lid body 30, a sensor element 80 as a functional element provided between the base body 10 and the lid body 30, a wiring layer 7 laminated on the sensor element 80, and a metal eutectic layer 20 that joins the base body 10 and the lid body 30 around the sensor element 80.

[0022] ***Details of the metal eutectic layer*** FIG. 3 is an enlarged view of part c in FIG. 2. FIG. 4 is a cross-sectional view of the main part of the base body and the lid body before joining and corresponds to FIG. 2. FIG. 3 shows a cross-section of a portion where the joining region 20a overlaps the wiring 82 connected to the terminal 92. On the base body 10, an insulating layer 6, a wiring layer 7 including the wiring 82, an insulating layer 8, a barrier layer 12, a metal eutectic layer 20, and a lid body 30 are laminated in this order. The insulating layer 6 is an interlayer insulating layer, and in a preferred example, it is a SiO2 layer. Note that the insulating layer 6 may also be a SiN layer. The same applies to the insulating layer 8.

[0023] The wiring layer 7 is formed of multiple layers. For example, it has a four-layer structure laminated in the order of Ti, TiN, AlCu, and TiN from the bottom. Here, the barrier layer 12 refers to a two-layer structure of Ti and TiN, and it is what remains of the underlying layer of the first joint 15 (Figure 4) on the substrate 10 side before the formation of the metal eutectic layer 20. The Ti layer plays a role in enhancing the adhesion with the insulating layer 8, and the TiN layer plays a role in preventing the diffusion of Al from AlCu. If the adhesion between the barrier layer 12 and the insulating layer 8 is good, the Ti layer can be omitted.

[0024] As shown in Figure 4, before the formation of the metal eutectic layer 20, the substrate 10 is provided with a first joint 15, and the lid 30 is provided with a second joint 16. The first joint 15 has a two-layer structure of the barrier layer 12 and the first metal layer 13. The barrier layer 12 has a two-layer structure of Ti and TiN, and the first metal layer 13 is an AlCu layer. The second joint 16 is a Ge layer as the second metal. However, the Cu in the AlCu layer is mixed for the purpose of preventing electromigration, and the content is low as described later. Therefore, the main component of the first metal layer 13 is Al. On the other hand, the second joint 16 is a Ge layer as the second metal. The second joint 16 is also referred to as the second metal layer 16. In other words, the main component of the first metal is Al, and the main component of the second metal is Ge. The metal eutectic layer 20 is a eutectic layer formed by eutectic bonding the first joint 15 and the second joint 16 by the bonding method described later. Eutectic generally refers to an alloy formed by solidification from a liquid phase state in which two or more kinds of metals are mixed.

[0025] The metal eutectic layer 20 shown in Figure 3 is a faithful trace of the microscopic photograph of the eutectic layer. As a result of elemental analysis, as shown in FIG. 3, the metal eutectic layer 20 is formed in a state where a first region 21 mainly composed of AlCu, which is the first metal, and a second region 22 mainly composed of Ge, which is the second metal, are adjacent to each other. The content rate of the first metal in the first region 21 is higher than the content of the first metal in the second region 22. The content rate of the second metal in the second region 22 is higher than the content of the second metal in the first region 21. The second region 22 extends widely along the lid body 30, and a part of it reaches the boundary with the base body 10. For example, in FIG. 3, the extending portions 22a and 22b reach the base body 10. The boundary between the first region 21 and the second region 22 has a lot of entry and exit and is intricately intertwined. Also, in the portion where the second region 22 extends, the second region 22 is larger than the first region 21.

[0026] The distribution of Ge in the metal eutectic layer 20 is not uniform, and it exists relatively more in the second region 22, and is uniform and has no concentration gradient within that region. However, Ge also exists uniformly in the first region 21 although in a small amount. The first region 21 and the second region 22 are in contact with each other without gaps, and the contact area is larger than the planar area of the bonding region 20a. That is, the first region 21 and the second region 22 are randomly interlocked, and the bonding strength is very high. In other words, the contact area between the first region 21 and the second region 22 is larger than the area of the bonding region 20a where the base body 10 and the lid body 30 are bonded by the metal eutectic layer 20.

[0027] FIG. 5 is a table showing the relationship of surface energy per unit area. Note that the source of Table 19 in FIG. 5 is by Yokota et al., Precision Engineering, Vol. 31, No. 10 (1965), pp. 828-835, and H.W. Sheng et al., PhysRevB. 83.134118 (2011). Generally, it is known that Ge has a diamond structure and Al has a face-centered cubic lattice structure. If Ge is abundant as the main component of the eutectic layer, a solid solution with a diamond structure is formed, and if Al is abundant, a solid solution with a face-centered cubic lattice structure is formed. A solid solution refers to a substance in which two elements are melted together and the whole is in a solid phase with a relatively uniform concentration. However, each solid solution has a different component ratio within the solid solubility limit. That is, in the second region 22 with a high Ge content, a solid solution of a diamond structure is realized, and in the first region 21 with a high Al content, a solid solution of a face-centered cubic lattice structure is realized. When cutting the crystal to expose the surface, the surface energy serves as a reference. As shown in Table 19 of FIG. 5, it is known that the surface energy per unit area of Ge is higher than that of Al in any crystal plane orientation. In other words, in the metal eutectic layer 20, there are a plurality of first regions 21 mainly composed of the first metal and having a face-centered cubic lattice structure, and second regions 22 mainly composed of the second metal and having a diamond structure, and they are adjacent to each other.

[0028] As shown in FIG. 3, a part of the second region 22 reaches the boundary with the substrate 10. And the second region 22 extends from the lid 30 to the substrate 10. In other words, the second region 22 reaches the boundary with the substrate 10. That is, the second region 22 reaches the substrate 10 regardless of the distance from the lid 30. At the same time, the second region 22 with a high Ge content contains Al within a range not exceeding the solid solubility limit for Ge. And the component ratio of Ge and Al in the second region 22 is relatively uniform and does not depend on the distance from the lid 30. From the perspective of surface energy, it is better for the second region 22 with a high Ge content to reach the boundary with the substrate 10 in terms of joint strength. More preferably, it is better if there are more portions where the second region 22 with a high Ge content extends from the lid 30 to the substrate 10. On the other hand, at the boundary between the lid 30 and the metal eutectic layer 20 in FIG. 3, since the second joint portion 16 made of Ge is directly formed on the lid 30 (FIG. 4), Ge diffuses into the silicon which is the lid 30. That is, the second region 22 mainly composed of Ge forms a fine uneven shape at the boundary portion with the lid 30 (the dotted line in FIG. 3), increasing the contact area and thus increasing the joint strength.

[0029] ***Bonding method of the bonding region*** FIG. 6 is a flowchart showing the process of the bonding method. Here, the bonding method between the substrate 10 and the lid 30 will be described mainly with reference to FIG. 6, appropriately referring to other drawings as well.

[0030] In step S10, a first joint portion 15 is formed on the base body 10. Prior to step S10, it is described that the base body 10 including the sensor element 80 has been formed. As shown in FIG. 4, the first joint portion 15 has a two-layer structure of a barrier layer 12 and a first metal layer 13. In a preferred example, the barrier layer 12 has a two-layer structure of Ti and TiN. The first metal layer 13 is an AlCu layer. In a preferred example, the Cu content concentration is 0.1 to 1.0 wt%, and the layer thickness is about 10,000 Å. The first joint portion 15 is formed, for example, by film formation using a DC sputtering method and then patterning according to the joint region 20a.

[0031] In step S11, a second joint portion 16 is formed on the lid body 30. Prior to step S11, it is described that the lid body 30 including the recess 35 has been formed. Also, the formation of the base body 10 and the lid body 30 may be performed in parallel in separate processes. As shown in FIG. 4, the second joint portion 16 is a Ge layer as the second metal. In the present embodiment, the second joint portion 16 made of Ge is directly formed on the silicon substrate constituting the lid body 30. In a preferred example, after forming a Ge layer about 2000 to 30000 Å using a DC sputtering method, it is patterned according to the joint region 20a to form the second joint portion 16. Note that the formation of the first joint portion 15 and the second joint portion 16 may use an RF sputtering method as long as a desired film thickness is obtained.

[0032] FIG. 7 is a side view showing a schematic configuration of the bonding apparatus. For convenience of explanation, the state in which the base body 10 and the lid body 30 are overlapped before bonding is referred to as a laminate 99. The bonding device 48 is arranged in a chamber (not shown) and is composed of a heating jig 41, a weighting jig 42, etc. The heating jig 41 includes a metal stage 41s on which the laminate 99 is placed, and a heating heater 41h such as a sheathed heater. The weighting jig 42 includes a stage 42s facing the stage 41s, and a heating heater 42h such as a sheathed heater. Note that the heating heater is not limited to a sheathed heater, and any heater that can heat is acceptable. For example, it may be a carbon heater.

[0033] In step S12, after aligning the positions of the bonding regions 20a and making the laminate 99 in a state where the base body 10 and the lid body 30 are overlapped, it is set on the heating jig 41 of the bonding device 48. Specifically, as shown in FIG. 7, the laminate 99 with the base body 10 facing down is placed on the stage 41s of the heating jig 41. Note that it may be placed on the heating jig 41 with the lid body 30 facing down. Also, after aligning the positions of the bonding regions 20a, the base body 10 and the lid body 30 may be temporarily fixed with fixing pins (not shown). In this step S12, the weighting jig 42 and the laminate 99 are not in contact with each other and are in a separated state. Note that the laminate 99 may be in the state of a large-sized substrate on which a plurality of sensor elements 80 are arranged. Therefore, the base body 10 and the lid body 30 may be of the same size and match. Also, it is preferable that the inside of the chamber is made into an environment in which moisture removal is performed as much as possible and filled with an inert gas such as N2 or Ar. Alternatively, it may be in an atmospheric pressure state, or may be depressurized to a vacuum state. For example, when the sensor element 80 is an acceleration sensor, the inside of the chamber is preferably in an atmospheric pressure state. This is because the damping effect by atmospheric pressure may improve the characteristics of the acceleration sensor. Also, when the sensor element 80 is an angular velocity sensor, a vacuum state with a pressure of 0.1 to 10 Pa is preferable. This is because the vibration characteristics become good in such a vacuum state.

[0034] FIG. 8 is a graph showing an example of a temperature profile in the heating process. The horizontal axis represents the elapsed time (minutes), the left vertical axis represents the temperature (°C), and the right vertical axis represents the weight (arbitrary unit). In FIG. 8, graph 43 shows the set temperature of the heating jig 41, and graph 44 shows the set temperature of the weight jig 42. Graph 45 shows the temperature of the laminate 99. Graph 47 shows the applied state of the weight.

[0035] In step S13, a heating process for heating the bonding device 48 is performed. In the heating process, both the heating jig 41 and the weight jig 42 are heated at a first set temperature of 400°C or higher. The eutectic temperature of AlGe is 420°C. FIG. 8 shows an example of a temperature profile in a preferred example. As shown in graph 43, in the heating jig 41, 425°C is set as the first set temperature. Note that it is not limited to 425°C, and any temperature between 415°C and 425°C is acceptable. Also, as shown in graph 44, in the weight jig 42, 445°C is set as the first set temperature. Note that it is not limited to 445°C, and any temperature between 420°C and 450°C is acceptable.

[0036] In step S14, it is determined whether the temperature of the laminate 99 has stabilized due to heating at the first set temperature. If the temperature has stabilized, the process proceeds to step S15. If the temperature has not stabilized, the heating at the first set temperature in step S13 is continued. In the example of FIG. 8, as shown in graph 45, since the temperature stabilized at about 435°C, the process proceeds to step S15. At this time, although the temperature of the laminate 99 exceeds the eutectic temperature of AlGe, which is 420°C, no weight is applied, so the formation of the metal eutectic layer 20 of AlGe does not progress significantly and is limited. By setting 425°C for the heating jig 41 and 445°C for the weight jig 42 as the first set temperature, sufficient heat can be transferred to the laminate 99.

[0037] In step S15, the heating setting is changed to a second set temperature lower than the first set temperature. In the case of FIG. 8, the heating jig 41 is changed from 425°C of the first set temperature to 415°C of the second set temperature. Note that it is not limited to 415°C, and any temperature between 410°C and 420°C is acceptable. Similarly, the weighting jig 42 is changed from 445°C of the first set temperature to 425°C of the second set temperature. Note that it is not limited to 425°C, and any temperature between 410°C and 430°C is acceptable. In this step S15, the reason for changing the heating setting to a second set temperature lower than the first set temperature is to prevent the temperature of the laminate 99 from rising excessively when the weighting jig 42 is brought into contact with the laminate 99. It is presumed that the excessive temperature rise is partly due to the lack of a heat dissipation path due to the close contact of the weighting jig 42 with the laminate 99. If the temperature of the laminate 99 rises excessively, the metal eutectic layer 20 may rapidly liquefy and protrude beyond the designed bonding region 20a. By setting the second set temperature lower than the first set temperature in this way, the rapid liquefaction phenomenon and protrusion of the metal eutectic layer 20 can be prevented.

[0038] In step S16, the weighting jig 42 is lowered and brought into contact with the laminate 99, and a predetermined load is applied for a predetermined time. When the upper side of the laminate 99 is the lid 30, the weighting jig 42 contacts the lid 30, and when the upper side of the laminate 99 is the base 10, the weighting jig 42 contacts the base 10. As shown in the graph 47 of FIG. 8, the weighting is applied after the set temperature has dropped. In the case of FIG. 8, a load of approximately 100 is applied in arbitrary units for a predetermined time. Also, as shown in the graph 45, the temperature of the laminate 99 drops to approximately 425°C before the weighting is applied, and the temperature gradually rises in synchronization with the application of the weighting and reaches approximately 433°C.

[0039] In step S17, the heating setting is changed to a third set temperature lower than the second set temperature. In the example of FIG. 8, the heating jig 41 is changed from 415° C. of the second set temperature to 100° C. of the third set temperature. Note that it is not limited to 100° C., and for example, room temperature may be used. Similarly, the weighting jig 42 is changed from 425° C. of the second set temperature to 100° C. of the third set temperature. Note that it is not limited to 100° C., and for example, room temperature may be used. At this time, after the temperature of the laminate 99 falls below the eutectic temperature 420° C. of AlGe, the weighting is released. That is, the predetermined time of weighting ends after the temperature becomes 420° C. or lower of the eutectic temperature. By doing so, the eutectic reaction can be suppressed in a short time, and hillocks generated in the wiring layer 7 including the wirings 81 to 83 can be suppressed. Therefore, the inertial sensor 100 having long-term reliability can be provided.

[0040] By the above bonding method, the metal eutectic layer 20 shown in FIG. 3 is formed in the bonding region 20a, the base body 10 and the lid body 30 are firmly bonded, and the inertial sensor 100 is completed. In particular, by performing the heating process and the weighting process according to the above temperature 4 profile, a good eutectic bond can be realized without leaving an unreacted layer of Al or Ge in the eutectic layer.

[0041] In other words, it is a manufacturing method of bonding a base body 10 provided with a sensor element 80 and a lid body 30 covering the sensor element 80 in a bonding region 20a surrounding the sensor element 80, including a step of forming a first bonding portion 15 mainly composed of a first metal in the bonding region 20a of the base body 10, a step of forming a second bonding portion 16 mainly composed of a second metal in the bonding region 20a of the lid body 30, a step of overlapping the base body 10 and the lid body 30 to form a laminate 99 so that the first bonding portion 15 and the second bonding portion 16 overlap, a heating step of heating the laminate 99, and a weighting step of applying a weight to the laminate 99. In the heating step, heating is started at the first set temperature, and after the temperature of the laminate 99 stabilizes, the heating setting is changed to a second set temperature lower than the first set temperature, and then the weighting step is performed. Then, when the weighting step ends, the temperature setting is changed to a third set temperature lower than the second set temperature.

[0042] ***Potential of the cover body*** FIG. 9 is a perspective view of a main part showing the electrical wiring structure of the metal eutectic layer, and is an enlarged perspective view around the terminal 94 in FIG. 1. As shown in FIG. 9, the metal eutectic layer 20 is electrically connected to the terminal 94 which is a GND terminal by the protruding portion 15b of the first joint portion 15 (FIG. 4). The protruding portion 15b is a wiring pattern formed together with the first joint portion 15, and protrudes from the joint region 20a toward the terminal 94 side. Since there is no opposing portion on the side of the second joint portion 16 (FIG. 4) in the protruding portion 15b, it functions as an electrical wiring drawn out from the metal eutectic layer 20.

[0043] The protruding portion 15b is provided so as to overlap with the wiring 84 connected to the terminal 94 via the insulating layer 8. A conductive contact portion 18 is provided at a portion where the wiring 84 and the protruding portion 15b overlap in the insulating layer 8. Thereby, the terminal 94 and the metal eutectic layer 20 are electrically connected via the wiring 84, the contact portion 18, and the protruding portion 15b. Note that it is not limited to the GND potential, and any electrically stable potential may be used. For example, a fixed potential including the power supply potential may be used. In other words, the cover body 30 is electrically connected to the power supply wiring of the base body 10 via the metal eutectic layer 20.

[0044] In addition, in the above description, the sensor element 80 has been described as a Z-direction acceleration sensor, but it is not limited thereto, and any electronic component that requires an airtight environment may be used. For example, an MEMS device such as an angular velocity sensor or a timing device such as a crystal oscillator or a ceramic oscillator may be used.

[0045] As described above, according to the inertial sensor 100 of the present embodiment and the manufacturing method of the inertial sensor 100, the following effects can be obtained. The inertial sensor 100 is a capacitance change type inertial sensor, and includes a base body 10, a lid body 30, a sensor element 80 as a functional element provided between the base body 10 and the lid body 30, and a metal eutectic layer 20 that joins the base body 10 and the lid body 30 around the sensor element 80. The metal eutectic layer 20 includes a plurality of first regions 21 mainly composed of a first metal and having a face-centered cubic lattice structure, and a plurality of second regions 22 mainly composed of a second metal and having a diamond structure, and the first regions 21 and the second regions 22 are adjacent to each other.

[0046] According to this, since the base body 10 and the lid body 30 are joined by a metal eutectic layer in which the first regions 21 having a face-centered cubic lattice structure and the second regions 22 having a diamond structure are randomly interlocked, a high joining strength can be obtained. Further, since the entire joining region 20a is AlGe eutectified, the reliability is excellent. Therefore, it is possible to provide an inertial sensor 100 having a high joining strength between the base body 10 and the lid body 30 and excellent long-term reliability.

[0047] Further, the second region 22 reaches the boundary with the base body 10. According to this, since the second region 22 with a high Ge content reaches the boundary with the base body 10, the joining strength in the joining region 20a is increased.

[0048] Further, the second region 22 extends from the lid body 30 to the base body 10. According to this, the joining strength in the joining region 20a is further increased.

[0049] Further, in the portion where the second region 22 extends, the second region 22 is more than the first region 21. According to this, since the extending portion of the second region 22 with a high Ge content is large, the joining strength in the joining region 20a is increased.

[0050] Further, the contact area between the first region 21 and the second region 22 is larger than the area of the joining region 20a where the base body 10 and the lid body 30 are joined by the metal eutectic layer 20. According to this, since the first region 21 and the second region 22 are randomly mated with a large contact area, their bonding strength is very high.

[0051] Further, the lid body 30 is electrically connected to the power supply wiring of the base body 10 via the metal eutectic layer 20. According to this, since the lid body 30 becomes the power supply potential and is electrically stabilized, it is less likely to be affected by noise, and the operation of the sensor element 80 can be stabilized.

[0052] Also, the first metal is Al and the second metal is Ge. According to this, a metal eutectic layer 20 with high bonding strength can be formed.

[0053] The manufacturing method of the inertial sensor 100 is a manufacturing method in which the base body 10 provided with the sensor element 80 and the lid body 30 covering the sensor element 80 are joined in a joining region 20a surrounding the sensor element 80, and a first joining portion 15 mainly composed of a first metal is formed in the joining region 20a of the base body 10. a step of forming a second joining portion 16 mainly composed of a second metal in the joining region 20a of the lid body 30; a step of aligning the first joining portion 15 and the second joining portion 16 so as to overlap, and overlapping the base body 10 and the lid body 30 to form a laminate 99; a heating step of heating the laminate 99; and a weighting step of applying a weight to the laminate 99.

[0054] According to this manufacturing method, the temperature and weight required for the formation of the eutectic layer can be applied by the heating step and the weighting step. Therefore, a metal eutectic layer 20 with high bonding strength can be formed in the joining region 20a.

[0055] Also, after the temperature of the laminate 99 is stabilized by the heating step set at the first set temperature, the temperature setting is changed to a second set temperature lower than the first set temperature, and then the weighting step is performed. According to this, an excessive temperature rise when the weighting jig 42 is brought into contact with the laminate 99 can be prevented. Therefore, the metal eutectic layer 20 can be formed at an appropriate heating temperature.

[0056] Further, when the weighting process is completed, the temperature setting is changed to a third set temperature lower than the second set temperature. According to this, the metal eutectic layer 20 can be formed at an appropriate heating temperature.

[0057] Also, when the laminate 99 is placed on the heating jig 41 with the base 10 at the bottom, a weighting jig 42 is disposed above the lid 30 in the laminate 99. In the heating process, the heating jig and the weighting jig generate heat. In the weighting process, the weighting jig abuts against the lid 30 to apply a weight. According to this, the metal eutectic layer 20 can be formed by an appropriate method using the bonding device 48, and the base 10 and the lid 30 can be bonded together. Conversely, when the laminate 99 is placed on the heating jig 41 with the lid 30 at the bottom, a weighting jig 42 is disposed above the base 10 in the laminate 99. In the heating process, the heating jig and the weighting jig generate heat. In the weighting process, the weighting jig abuts against the base 10 to apply a weight.

[0058] Embodiment 2 *** Different Modes of Bonding Layer - 1 *** FIG. 10 is a cross-sectional view of the main part of the base and the lid before bonding according to Embodiment 2, corresponding to FIG. 4. In the above embodiment, it has been described that the second joint portion 16 is patterned in accordance with the joint region 20a, but the present invention is not limited thereto, and patterning may not be performed. Hereinafter, the same parts as those in the above embodiment are given the same reference numerals, and redundant descriptions are omitted.

[0059] In this embodiment, a Ge layer is directly formed on the entire surface of the lid 30, and the Ge layer is used as the bonding layer 26. Specifically, as shown in FIG. 10, the portion overlapping the first joint portion 15 of the bonding layer 26 becomes the second joint portion 16b, and the first joint portion 15 can be joined to form the metal eutectic layer 20. Other configurations are the same as those described in Embodiment 1. Even with this configuration, it is possible to provide an inertial sensor 100 having a high bonding strength between the base body 10 and the lid body 30 and excellent long-term reliability. Furthermore, since patterning of the second joint portion is not required, the number of processing steps is reduced, and the manufacturing cost can be reduced.

[0060] Embodiment 3 *** Different modes of the bonding layer - 2 *** FIG. 11 is a cross-sectional view of the main part of the base body and the lid body before bonding according to Embodiment 3, corresponding to FIG. 4. FIG. 12 is an enlarged view around the metal eutectic layer after bonding, corresponding to FIG. 3. In the above embodiment, the second joint portion 16 has been described as being directly formed on the lid body 30, but the present invention is not limited thereto, and a barrier layer 27 may be provided as a base. Further, a sealing hole 36 may be provided in the lid body 30. Hereinafter, the same parts as those in the above embodiment are given the same reference numerals, and redundant descriptions are omitted.

[0061] In the present embodiment, the second joint portion 28 has a two-layer structure. Specifically, the second joint portion 28 is composed of a barrier layer 27 made of a TiN layer and a second metal layer 16 made of a Ge layer. The barrier layer 27 is provided on the silicon substrate constituting the lid body 30. The barrier layer 27 may be a material having a melting point higher than the eutectic point of the metal eutectic layer 20, and may be Ti, Mo, W, Co, Pt, Ta, TiN, or an alloy thereof. In a preferred example, the second joint portion 28 is formed by depositing each layer using a DC sputtering method and then patterning in accordance with the bonding region 20a to form the second joint portion 28. Note that an RF sputtering method may also be used.

[0062] Further, the lid body 30 is provided with a sealing hole 36 penetrating the lid body. The sealing hole 36 serves to communicate the outside air with the inside of the storage space S when the base body 10 and the lid body 30 are joined. The sealing hole 36 is formed in a recess recessed from the upper surface of the lid body 30, and after the base body 10 and the lid body 30 are joined, it can be sealed with, for example, a solder ball 37. Alternatively, the sealing hole 36 may be directly melted by laser light for sealing. Other than these configurations, the configuration in Embodiment 1 is the same. By providing the sealing hole 36, it is not necessary to seal the inside of the storage space S during the formation of the metal eutectic layer 20, so the manufacturing efficiency can be improved.

[0063] FIG. 12 is a faithful tracing of a micrograph of the eutectic layer in the metal eutectic layer 20 formed by joining the first joint portion 15 and the second joint portion 28 by the joining method of FIG. 6. As shown in FIG. 12, the metal eutectic layer 20 is formed in a state where a first region 21 mainly composed of Al as the first metal and a second region 22 mainly composed of Ge as the second metal are adjacent to each other. The second region 22 extends widely along the lid body 30, and a part of it reaches the boundary with the base body 10. That is, similar to the description in FIG. 3, a joint with high joint strength is realized by the metal eutectic layer 20. Furthermore, by providing the barrier layer 27, diffusion of Ge into the lid body 30 is prevented, and electrical contact with the silicon constituting the lid body 30 is improved. That is, an ohmic contact between the metal eutectic layer 20 and the lid body 30 can be realized by the barrier layer 27.

[0064] Embodiment 4 ***Application to an inertial measurement device***

[0065] FIG. 13 is a plan view of the inertial sensor according to Embodiment 4 and corresponds to FIG. 1. In the above embodiment, the inertial sensor 100 has been described as housing one sensor element 80, but it is not limited thereto, and it may house a plurality of sensor elements. Hereinafter, the same parts as those in the above embodiment are given the same reference numerals, and duplicate descriptions are omitted.

[0066] As shown in FIG. 13, the inertial sensor 110 of the present embodiment includes, in addition to the aforementioned sensor element 80, a sensor element 85 and a sensor element 86. The sensor element 85 is a capacitance change type acceleration sensor that detects acceleration in the X direction. The sensor element 86 is a capacitance change type acceleration sensor that detects acceleration in the Y direction. That is, the inertial sensor 110 is a three-axis acceleration sensor capable of detecting three-axis acceleration in the XYZ directions. Similar to the inertial sensor 100, the inertial sensor 110 has a structure in which the base body 10 and the lid body 30 are joined by the metal eutectic layer 20 in the joining region 20a, and has an accommodation space S inside. In the accommodation space S, three sensor elements 80, 85, and 86 are accommodated in a state where they can detect oscillation. In FIG. 13, the joining region 20a is a square ring region that is slightly smaller than the outer peripheral edge of the lid body 30, but is not limited thereto. The joining region 20a only needs to surround and close the sensor elements 80, 85, and 86, and may be a polygon or an ellipse. However, it is configured to intersect with the lead wiring (not shown) in a plan view.

[0067] FIG. 14 is an exploded perspective view of the inertial measurement device. FIG. 15 is a perspective view of the substrate. The inertial measurement device 2000 of the present embodiment shown in FIG. 14 is equipped with the inertial sensor 110. The inertial measurement device 2000 is a rectangular parallelepiped having a substantially square planar shape. The inertial measurement device 2000 is an inertial measurement sensor unit (IMU: Inertial Measurement Unit) that detects the posture and behavior of a mounted object such as an automobile or a robot. The inertial measurement device 2000 functions as a so-called six-axis motion sensor equipped with a three-axis acceleration sensor and a three-axis angular velocity sensor.

[0068] The inertial measurement device 2000 includes an outer case 301, a joining member 310, and a sensor module 325 on which the inertial sensor is mounted. The outer shape of the outer case 301 is a rectangular parallelepiped having a substantially square planar shape, similar to the overall shape of the inertial measurement device 2000, and screw holes 302 are formed near two vertices located in the diagonal direction of the square. The inertial measurement device 2000 can be fixed to the mounting surface of a mounted object such as an automobile by passing two screws through the two screw holes 302.

[0069] In addition, the outer case 301 is box-shaped, and the sensor module 325 is housed therein. Specifically, the sensor module 325 is inserted inside the outer case 301 with the joining member 310 interposed therebetween.

[0070] The sensor module 325 has an inner case 320 and a substrate 315. The inner case 320 is a member that supports the substrate 315, and the substrate 315 is joined to the lower surface of the inner case 320 via an adhesive.

[0071] Also, the inner case 320 is shaped to fit inside the outer case 301. The inner case 320 is formed with a recess 331 for preventing contact with the substrate 315 and an opening 321 for exposing a connector 316 described later. The inner case 320 is joined to the outer case 301 via the joining member 310.

[0072] Next, the substrate 315 on which the inertial sensor 110 is mounted will be described. As shown in FIG. 15, an inertial sensor 110, a connector 316, an angular velocity sensor 317z for detecting the angular velocity about the Z axis, etc. are mounted on the surface of the substrate 315 on the side of the inner case 320, which is the upper surface. Angular velocity sensors 317x for detecting the angular velocity about the X axis and 317y for detecting the angular velocity about the Y axis are mounted on the side surface of the substrate 315. Note that an inertial sensor 100 may be mounted instead of the inertial sensor 110.

[0073] Also, a control IC 319 as a control unit is mounted on the surface of the outer case 301 side, which is the lower surface of the substrate 315. The control IC 319 is an MCU (Micro Controller Unit), which incorporates a storage unit including a non-volatile memory, an A / D converter, etc., and controls each part of the inertial measurement device 2000. The storage unit stores a program that defines the order and content for detecting acceleration and angular velocity, a program that digitizes the detection data and incorporates it into packet data, and accompanying data. In addition, a plurality of other electronic components are mounted on the substrate 315.

[0074] According to such an inertial measurement device 2000, since the inertial sensor 110 is used, it is possible to provide an inertial measurement device 2000 having excellent long-term reliability that enjoys the effects according to the above embodiment.

Explanation of Reference Numerals

[0075] 1... Substrate, 2... Insulating layer, 3... Semiconductor layer, 5... Recess, 6... Insulating layer, 7... Wiring layer, 8... Insulating layer, 10... Substrate body, 11... Protrusion, 12... Barrier layer, 13... First metal layer, 15... First joint, 15b... Protrusion, 16... Second joint, 16b... Second joint, 18... Contact portion, 19... Surface, 20... Metal eutectic layer, 20a... Joint region, 21... First region, 22... Second region, 22a, 22b... Extension portion, 26... Joint layer, 27... Barrier layer, 28... Second joint layer, 30... Cover body, 31... Stopper portion, 35... Recess, 36... Sealing hole, 37... Solder ball, 41... Heating jig, 41h... Heating heater, 41s... Stage, 42... Weighting jig, 42h... Heating heater, 42s... Stage, 43, 44, 45... Graph, 48... Joining device, 52a... First bar, 52b... Second bar, 53... Third bar, 54a... First rotary spring, 54b... Second rotary spring, 55... Movable body, 60... Center line, 61... Oscillation axis, 65... Fixed portion, 71, 71c... Movable electrode, 72, 72c... Fixed electrode, 73a~73d... Movable electrode group, 74a~74d... Fixed electrode group, 75a~75d... Support portion, 76n... N-type detection portion, 76p... P-type detection portion, 80... Sensor element, 81~84... Wiring, 85... Sensor element, 86... Sensor element, 91~94... Terminal, 99... Laminate, 100... Inertial sensor, 110... Inertial sensor, 301... Outer case, 302... Screw hole, 310... Joining member, 315... Substrate, 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 device.

Claims

1. A capacitance change type inertial sensor, comprising: a substrate; a lid; a functional element provided between the substrate and the lid; a metal eutectic layer that joins the substrate and the lid around the functional element, wherein the metal eutectic layer has a plurality of adjacent first regions mainly composed of a first metal and having a face-centered cubic lattice structure, and second regions mainly composed of a second metal and having a diamond structure; an inertial sensor.

2. The inertial sensor according to claim 1, wherein the second region reaches the boundary with the substrate. The inertial sensor according to claim 1.

3. The inertial sensor according to claim 2, wherein the second region extends from the lid to the substrate. The inertial sensor according to claim 2.

4. The inertial sensor according to claim 3, wherein in the portion where the second region extends, there is more of the second region than the first region. The inertial sensor according to claim 3.

5. The inertial sensor according to claim 3, wherein the contact area between the first region and the second region is larger than the area of the bonding region where the substrate and the lid are bonded by the metal eutectic layer. The inertial sensor according to claim 3.

6. The inertial sensor according to claim 2, wherein the lid is electrically connected to the power supply wiring of the substrate through the metal eutectic layer. The inertial sensor according to claim 2.

7. The inertial sensor according to claim 2, wherein the first metal is Al and the second metal is Ge. The inertial sensor according to claim 2.

8. A manufacturing method for joining a substrate provided with a functional element and a lid covering the functional element in a bonding region surrounding the functional element, comprising: forming a first bonding portion mainly composed of a first metal in the bonding region of the substrate; forming a second bonding portion mainly composed of a second metal in the bonding region of the lid; aligning the first bonding portion and the second bonding portion so that they overlap, and stacking the substrate and the lid to form a laminate; a heating step of heating the laminate; a weighting step of applying a weight to the laminate. A manufacturing method for an inertial sensor.

9. In the heating step, heating is started at a first set temperature, after the temperature of the laminate stabilizes, the heating setting is changed to a second set temperature lower than the first set temperature, and then the weighting step is performed. The manufacturing method for an inertial sensor according to claim 8.

10. When the weighting step is completed, the heating setting is changed to a third set temperature lower than the second set temperature. The manufacturing method for an inertial sensor according to claim 9.

11. The laminate is placed on a heating jig with the substrate facing down. Above the lid in the laminate, a weight jig is disposed. In the heating step, the heating jig and the weight jig generate heat. In the weighting step, the weight jig abuts against the lid to apply a weight. The method for manufacturing an inertial sensor according to claim 8.

12. The laminate is placed on the heating jig with the lid facing down. Above the base in the laminate, a weight jig is disposed. In the heating step, the heating jig and the weight jig generate heat. In the weighting step, the weight jig abuts against the base to apply a weight. The method for manufacturing an inertial sensor according to claim 8.

Images