Method for manufacturing MEMS device and MEMS device

Abstract

The present invention provides a method for manufacturing a MEMS device capable of suppressing damage to a device part and a MEMS device. The method for manufacturing a resonant device (1) includes: irradiating a laser LL from the lower surface side of a second substrate (350) so as to form a modified area (MA2) inside the second substrate (350) along a dividing line (LN1, LN2) between a first substrate (330) having a device part formed on its upper surface and a second substrate (350) whose upper surface is bonded to the lower surface of the first substrate (330) via a bonding part (60); and applying stress to the modified area (MA2) and dividing the first substrate (330) and the second substrate (350) along the dividing line (LN1, LN2) S307, wherein the bonding part (60) is formed along the dividing line (LN1, LN2) to shield the laser (LL).

Description

Technical Field

[0001] The invention relates to a method for manufacturing a MEMS device and the MEMS device. Background Art

[0002] Conventionally, devices manufactured using MEMS (Micro Electro Mechanical Systems) technology have become popular. For example, after forming a plurality of devices on a collective substrate (wafer), the wafer is divided and individualized into individual devices (chips).

[0003] As a laser processing method for dividing a wafer, for example, a method is known that includes the following steps: irradiating the inside of the substrate 15 of the object 1 to be processed, which includes the substrate 15 and the laminated parts 17a and 17b provided on the surface of the substrate 15, with a laser beam directed at a focal point, to form a modified region 7 inside the substrate 15, and forming a cutting start region from the laser incident surface of the object 1 along the predetermined distance inside the cutting line of the object 1 through the modified region 7; and a step of applying stress to the object 1 from the laminated parts 17a and 17b side after the cutting start region is formed, thereby cutting the object 1 along the cutting line (see Patent Document 1). In this laser processing method, laser beam is irradiated from the back surface 21 side of the object 1 to form the modified region 7 inside the substrate 15, and then laser beam is irradiated from the surface 3 side of the object 1 to form the modified region 7 inside the laminated parts 17.

[0004] Patent document 1: Japanese Patent Application Laid-Open No. 2007-83309.

[0005] However, in the method of cited document 1, for example, a portion of the laser light irradiated from the surface 3 side is transmitted to the surface opposite to the incident side in the wafer (substrate) as the processing object 1. As a result, when a part of a device part, such as an electrode pad, is formed on the surface opposite to the wafer, the laser light not absorbed by the wafer may damage the device part formed on the surface opposite to the incident side. Summary of the invention

[0006] The present invention has been made in view of the above circumstances, and an object of the present invention is to provide a method for manufacturing a MEMS device and a MEMS device capable of suppressing damage to a device portion.

[0007] A method for manufacturing a MEMS device according to one aspect of the present invention includes: a process of irradiating a laser from the second main surface side of the second substrate to form a modified region inside the second substrate along a dividing line between the first substrate and the second substrate, wherein a device portion is formed on the first main surface of the first substrate, and the first main surface of the second substrate is joined to the second main surface of the first substrate via a joining portion; and a process of applying stress to the modified region to split the first substrate and the second substrate along the dividing line, wherein the joining portion is formed along the dividing line to shield the laser.

[0008] A MEMS device according to another aspect of the present invention comprises: a first substrate having a device portion formed on a first main surface; and a second substrate having a first main surface bonded to a second main surface of the first substrate via a bonding portion, wherein the material of the bonding portion is a metal that is transparent to the first substrate and the second substrate and can block laser light.

[0009] According to the present invention, damage to the device portion can be suppressed. BRIEF DESCRIPTION OF THE DRAWINGS

[0010] Figure 1 1 is a flowchart showing a method for manufacturing a resonance device according to the first embodiment of the present invention.

[0011] Figure 2 Yes means Figure 1 A cross-sectional view of the process shown.

[0012] Figure 3 Yes means Figure 1 A cross-sectional view of the process shown.

[0013] Figure 4 Yes means Figure 1 A cross-sectional view of the process shown.

[0014] Figure 5 Yes means Figure 4 A top view of the first substrate is shown.

[0015] Figure 6 Yes means Figure 1 A cross-sectional view of the process shown.

[0016] Figure 7 Yes means Figure 1 A cross-sectional view of the process shown.

[0017] Figure 8 Yes means Figure 1 A cross-sectional view of the process shown.

[0018] Fig. 9 It is a perspective view schematically showing the appearance of the resonance device according to the first embodiment of the present invention.

[0019] Fig.10It is schematically indicated Fig. 9 An exploded perspective view of the structure of the resonance device shown.

[0020] Fig.11 It is schematically indicated Fig.10 A top view of the resonator structure is shown.

[0021] Fig.12 It is schematically indicated Figures 9 to 11 The structure of the cross section of the resonant device shown is taken along the line XII-XII.

[0022] Fig.13 1 is a flowchart showing a method for manufacturing a resonance device according to the second embodiment of the present invention.

[0023] Fig.14 Yes means Fig.13 A cross-sectional view of the process shown.

[0024] Fig.15 Yes means Fig.13 A cross-sectional view of the process shown.

[0025] Fig.16 It is a cross-sectional view schematically showing the structure of a resonance device according to a second embodiment of the present invention. DETAILED DESCRIPTION

[0026] Hereinafter, embodiments of the present invention will be described with reference to the accompanying drawings. In the following description of the accompanying drawings, identical or similar reference numerals are used to represent identical or similar components. The accompanying drawings are illustrative, and the sizes and shapes of the various parts are schematic, and the technical scope of the present invention should not be limited to the embodiments for interpretation.

[0027] <First embodiment>

[0028] First, refer to Figures 1 to 8 , a method for manufacturing a MEMS device according to a first embodiment of the present invention is described. Figure 1 1 is a flowchart showing a method for manufacturing the resonance device 1 according to the first embodiment of the present invention. Figure 2 Yes means Figure 1 A cross-sectional view of step S301 is shown. Figure 3 Yes means Figure 1 A cross-sectional view of step S302 is shown. Figure 4 Yes means Figure 1 A cross-sectional view of step S303 is shown. Figure 5 Yes means Figure 4 FIG. 3 is a top view of the first substrate 330 shown in FIG. Figure 6 Yes means Figure 1 A cross-sectional view of step S305 is shown. Figure 7Yes means Figure 1 A cross-sectional view of step S306 is shown. Figure 8 Yes means Figure 1 The cross-sectional view of step S307 is shown in FIG. Figures 2 to 8 In the present invention, for convenience of explanation, a part of the plurality of resonance devices 1 manufactured by the manufacturing method is shown for explanation.

[0029] In the following description, as a MEMS device, a resonance device 1 including a lower cover 20, a resonator 10 (hereinafter, the lower cover 20 and the resonator 10 are collectively referred to as a "MEMS substrate 50"), and an upper cover 30 is described as an example. However, the first embodiment of the present invention is not limited to the resonance device 1.

[0030] In the following description, the positive Z-axis direction side is referred to as the upper side (or the front side), and the positive Z-axis direction side is referred to as the lower side (or the back side).

[0031] like Figure 1 As shown, first, a first substrate 330 corresponding to the upper cover 30 of the resonance device 1 is prepared ( S301 ).

[0032] The first substrate 330 is formed using a silicon (Si) substrate (hereinafter, referred to as a "Si substrate"). Figure 2 As shown, a recess 31 is formed on the back side of the first substrate 330. The first substrate 330 is formed of a silicon (Si) wafer (hereinafter referred to as "Si wafer") L3 of a predetermined thickness. The front surface, the back surface (the surface facing the resonator 10), and the side surfaces of the through electrodes V1 and V2 in the first substrate 330 are covered with a silicon oxide film L31. The silicon oxide film L31 is formed on the surface of the Si wafer L3 by, for example, oxidation of the surface of the Si wafer L3 or chemical vapor deposition (CVD: Chemical Vapor Deposition).

[0033] A plurality of electrode pads 41A and 41B are formed on the upper surface of the first substrate 330. Specifically, a resin film L32 is stacked on the silicon oxide film L31, and a plurality of electrode pads 41A and 41B are formed on the silicon oxide film L31. Each electrode pad 41A and 41B is formed using a metal such as gold (Au) or aluminum (Al).

[0034] In addition, each electrode pad 41A is formed so as to contact the through electrode V1. Specifically, a part of the resin film L32 is removed by etching or other processing, and the electrode pad 41A is formed on the exposed through electrode V1 and the resin film L32. Similarly, each electrode pad 41B is formed so as to contact the through electrode V2. Specifically, a part of the resin film L32 is removed by etching or other processing, and the electrode pad 41B is formed on the exposed through electrode V2 and the resin film L32.

[0035] The through-electrodes V1 and V2 are formed by filling a conductive material in a through-hole formed in the first substrate 330. The conductive material filled is, for example, impurity-doped polysilicon (Poly-Si), copper (Cu), gold (Au), or impurity-doped single crystal silicon.

[0036] Return to Figure 1 Next, a second substrate 350 corresponding to the MEMS substrate 50 (resonator 10 and lower cover 20 ) of the resonance device 1 is prepared ( S302 ).

[0037] In the second substrate 350, Si substrates are bonded to each other. Alternatively, the second substrate 350 may be formed using an SOI substrate. Figure 3 As shown, the second substrate 350 includes a Si wafer L1, a Si substrate F2, and a silicon oxide (eg, SiO2) layer F21.

[0038] The Si wafer L1 has a surface ( Figure 3 A concave portion 21 is formed on the upper surface of the Si wafer L1. The thickness of the Si wafer L1 is, for example, about 150 μm, and the depth of the concave portion 21 is, for example, about 50 μm. In addition, the Si wafer L1 is formed of non-degenerate silicon, and its resistivity is, for example, not less than 16 mΩ·cm.

[0039] The Si substrate F2 is formed, for example, using a degenerate n-type silicon (Si) semiconductor having a thickness of about 6 μm, and as an n-type dopant, phosphorus (P), arsenic (As), antimony (Sb), etc. can be included. The resistance value of the degenerate silicon (Si) used in the Si substrate F2 is, for example, less than 16 mΩ·cm, and more preferably less than 1.2 mΩ·cm. In addition, on the lower surface of the Si substrate F2, as an example of a temperature characteristic correction layer, a silicon oxide (e.g., SiO2) layer F21 is formed. Thus, the temperature characteristics can be improved. In addition, the silicon oxide layer F21 can be formed on the upper surface of the Si substrate F2, or can be formed on both the upper surface and the lower surface of the Si substrate F2.

[0040] A bonding portion 60 is formed on the upper surface of the Si substrate F2, in other words, on the upper surface of the second substrate 350 along the dividing line LN1. In addition, the upper surface of the second substrate 350 is opposite to the lower surface of the second substrate 350, which is equivalent to an example of the "first main surface" of the present invention. The bonding portion 60 is formed, for example, using an alloy of aluminum (Al) and germanium (Ge). In addition, the bonding portion 60 can be formed by a gold (Au) film and a tin (Sn) film, or a combination of gold (Au) and silicon (Si), gold (Au) and gold (Au), copper (Cu) and tin (Sn), etc. In addition, in order to improve the adhesion, the bonding portion 60 can also be thinly sandwiched between the stacked layers of titanium (Ti), titanium nitride (TiN), tantalum nitride (TaN), etc.

[0041] The bonding portion 60 has an opening 65 formed along a portion of the dividing line LN1. After the bonding portion 60 is formed without an opening, a portion of the bonding portion 60 is removed by etching or the like, thereby forming the opening 65.

[0042] In addition to the bonding portion 60, a connection wiring 70 is formed on the upper surface of the Si substrate F2. The connection wiring 70 is electrically connected to the resonator 10 (not shown). The connection wiring 70 is formed using a metal such as aluminum (Al), gold (Au), or tin (Sn). By using the same type of metal as the bonding portion 60 as the material of the connection wiring 70, the manufacturing process can be simplified.

[0043] In this embodiment, an example is shown in which the bonding portion 60 and the connection wiring 70 are formed on the upper surface of the second substrate 350, but the present invention is not limited thereto. For example, at least one of the bonding portion 60 and the connection wiring 70 may be formed on the upper surface of the second substrate 350. Figure 2 The lower surface of the first substrate 330 shown in FIG. 3 is opposite to the upper surface of the second substrate 350, which is equivalent to an example of the “second main surface” of the present invention. In addition, when the bonding portion 60 is composed of multiple materials, a part of the material in the bonding portion 60, such as germanium (Ge), can be formed on the lower surface of the first substrate 330, and the remaining material in the bonding portion 60, such as aluminum (Al), can be formed on the upper surface of the second substrate 350. Similarly, when the connecting wiring 70 is composed of multiple materials, a part of the material in the connecting wiring 70 can also be formed on the lower surface of the first substrate 330, and the remaining material in the connecting wiring 70 can be formed on the upper surface of the second substrate 350.

[0044] Return to Figure 1 Next, the lower surface of the first substrate 330 and the upper surface of the second substrate 350 are bonded via the bonding portion 60 ( S303 ).

[0045] Specifically, the lower surface of the first substrate 330 and the upper surface of the second substrate 350 are eutectic-bonded via the bonding portion 60. Figure 4 As shown, the positions of the first substrate 330 and the second substrate 350 are aligned so that the through electrodes V1 and V2 of the first substrate 330 and the connecting wiring 70 of the second substrate 350 are in contact. After the alignment, the first substrate 330 and the second substrate 350 are clamped by a heater or the like, and a heating treatment for eutectic bonding is performed. The temperature in the heating treatment for eutectic bonding is above the temperature of the confocal point, for example, above 424°C, and the heating time is, for example, about 10 minutes to about 20 minutes. During heating, the first substrate 330 and the second substrate 350 are pressed by a pressure of, for example, about 5 MPa to about 25 MPa. In this way, the bonding portion 60 eutectically bonds the lower surface of the first substrate 330 and the upper surface of the second substrate 350.

[0046] like Figure 5 As shown, each joint 60 is formed into a roughly rectangular ring along the XY plane. The short side of each joint 60 extends along the dividing line LN1 parallel to the X-axis, and the long side of each joint 60 extends along the dividing line LN2 parallel to the Y-axis. The short side of each joint 60 configured on the positive direction side of the Y-axis and the short side of each joint 60 configured on the negative direction side of the Y-axis are separated by the dividing line LN1. Similarly, the long side of each joint 60 configured on the positive direction side of the X-axis and the long side of each joint 60 configured on the negative direction side of the X-axis are separated by the dividing line LN2. In addition, each joint 60 has a flow channel extending from a corner and connected to the corner of other opposing joints 60. The flow channel and the gap (gap) between the dividing lines LN1 and LN2 form an opening 65 of the joint 60.

[0047] The width of the dividing lines LN1 and LN2 on the upper surface of the first substrate 330 is, for example, not less than 5 μm and not more than 20 μm. Thus, when the first substrate 330 and the second substrate 350 are divided along the dividing lines LN1 and LN2 and singulated (chipped) into the resonance devices 1 (each chip), each resonance device 1 can be miniaturized and the number of resonance devices 1 that can be obtained can be increased.

[0048] On the other hand, the width AW of the narrowest portion of the opening 65 on the upper surface of the second substrate 350 is, for example, 20 μm or less.

[0049] The electrode pads 41A to 41D are formed above each corner of the joint 60. When looking down at the upper surface of the first substrate 330, each electrode pad 41A to 41D has a rectangular shape extending in the Y-axis direction. In addition, each electrode pad 41A to 41D is arranged across the dividing lines LN1 and LN2. That is, the electrode pad 41A and the electrode pad 41B are arranged across the dividing line LN1, and the electrode pad 41C and the electrode pad 41D are arranged across the dividing line LN1. In addition, the electrode pads 41A and the electrode pad 41C are arranged across the dividing line LN2, and the electrode pads 41B and the electrode pad 41D are arranged across the dividing line LN2. In addition, the electrode pads 41A to 41D are equivalent to an example of the "device portion" of the present invention, and the upper surface of the first substrate 330 is equivalent to an example of the "first main surface" of the present invention.

[0050] The distance PD between each electrode pad 41A to 41D and the dividing lines LN1 and LN2 on the upper surface of the first substrate 330 is greater than 1 / 2 of the width AW of the opening 65. In other words, for example, when the width AW of the opening 65 is 20 μm, the distance PD between each electrode pad 41A to 41D is greater than 10 μm.

[0051] In addition, the distance PD between each electrode pad 41A to 41D and the dividing lines LN1 and LN2 on the upper surface of the first substrate 330 is greater than 1 / 6 of the thickness (length in the Z-axis direction) of the first substrate 330. In other words, for example, when the thickness of the first substrate 330 is 150 μm, the distance PD between each electrode pad 41A to 41D is greater than 50 μm.

[0052] Back to Figure 1 Next, a dicing tape DT is attached to the lower surface of the second substrate 350 (S304). The dicing tape DT is formed by applying an adhesive to a substrate such as polyvinyl chloride (PVC), polyethylene terephthalate (PET), or polyolefin (PO). In addition, the lower surface of the second substrate 350 is equivalent to an example of the "second main surface" of the present invention.

[0053] In this embodiment, as a method of dividing the first substrate 330 and the second substrate 350, a stealth cutting technique that can make the width of the dividing lines LN1 and LN2 relatively small (narrow) is used for cutting. In the stealth cutting technique, a modified region (modified layer) is formed inside each of the first substrate 330 and the second substrate 350 as the dividing object, and the first substrate 330 and the second substrate 350 are cut with each modified region as a starting point.

[0054] Next, laser LL is irradiated from the upper surface side of the first substrate 330 to form a modified area MA1 inside the first substrate 330 along the division lines LN1 and LN2 ( S305 ).

[0055] like Figure 6 As shown, the laser LL is irradiated in a manner of focusing (condensing) at a predetermined position in the thickness direction (direction parallel to the Z axis) of the first substrate 330. The laser LL is transparent to the first substrate 330. When the first substrate 330 is, for example, a Si substrate, the wavelength of the laser LL is about 1000nm to 2000nm. A portion of the irradiated laser LL is absorbed by the inside of the first substrate 330, and a modified area MA1 is formed around the focal position.

[0056] Furthermore, by moving the laser LL and at least one of the first substrate 330 along the dividing line LN1 in a direction parallel to the X-axis direction, a modified region MA1 along the dividing line LN1 can be formed inside the first substrate 330. Similarly, by moving the laser LL and at least one of the first substrate 330 along the dividing line LN2 in a direction parallel to the Y-axis direction, a modified region MA1 along the dividing line LN2 can be formed inside the first substrate 330.

[0057] As described above, the distance PD between each electrode pad 41A to 41D and the dividing lines LN1 and LN2 on the upper surface of the first substrate 330 is greater than 1 / 6 of the thickness (length in the Z-axis direction) of the first substrate 330. Thus, each electrode pad 41A to 41D can be prevented from interfering with the laser light LL irradiated from the upper surface side of the first substrate 330 along the dividing lines LN1 and LN2.

[0058] The modified region MA1 includes a crack (not shown) in the thickness direction (direction parallel to the Z axis) of the first substrate 330. The crack extends toward the upper surface and the lower surface of the first substrate 330 by applying stress. Therefore, by applying stress to the modified region MA1 formed inside the first substrate 330, the first substrate 330 can be divided along the dividing lines LN1 and LN2.

[0059] The laser light LL which is not absorbed inside the first substrate 330 tries to penetrate the first substrate 330 and enter the upper surface of the second substrate 350. The laser light LL is partially blocked by the joint 60 formed along the dividing lines LN1 and LN2. By sufficiently reducing the width AW of the opening 65 relative to the wavelength of the laser light LL, the amount of laser light LL passing through the opening 65 of the joint 60 can be reduced.

[0060] In addition, Figure 6 , an example is shown in which the modified region MA1 is formed at one location inside the first substrate 330, but the present invention is not limited thereto. The number of modified regions MA1 varies depending on the thickness of the first substrate 330. Thus, for example, a plurality of modified regions can be formed in the thickness direction (direction parallel to the Z axis) of the first substrate 330 by irradiating lasers having different focal positions.

[0061] Back to Figure 1 Next, laser LL is irradiated from the lower surface side of the second substrate 350 to form a modified area MA2 inside the second substrate 350 along the dividing lines LN1 and LN2 (S306).

[0062] like Figure 7 As shown, the laser LL is irradiated in a manner of focusing (condensing) at a predetermined position in the thickness direction (direction parallel to the Z axis) of the second substrate 350. The laser LL is also transparent to the second substrate 350. When the second substrate 350 is, for example, a Si substrate, the wavelength of the laser LL is about 1000nm (1μm) to 2000nm (2μm). A portion of the irradiated laser LL is absorbed inside the second substrate 350, and a modified area MA2 is formed around the focal position.

[0063] Furthermore, by moving the laser LL and at least one of the second substrate 350 along the dividing line LN1 in a direction parallel to the X-axis direction, a modified region MA2 along the dividing line LN1 can be formed inside the second substrate 350. Similarly, by moving the laser LL and at least one of the second substrate 350 along the dividing line LN2 in a direction parallel to the Y-axis direction, a modified region MA2 along the dividing line LN2 can be formed inside the second substrate 350.

[0064] The modified region MA2 includes a crack (crack) not shown in the thickness direction (direction parallel to the Z axis) of the second substrate 350. The crack extends toward the upper surface and the lower surface of the second substrate 350 by applying stress. Therefore, by applying stress to the modified region MA2 formed inside the second substrate 350, the second substrate 350 can be divided along the dividing lines LN1 and LN2.

[0065] The laser light LL which is not absorbed inside the second substrate 350 tries to penetrate the second substrate 350 and enter the lower surface of the first substrate 330. The laser light LL is partially shielded by the joint 60 formed along the dividing lines LN1 and LN2. By sufficiently reducing the width AW of the opening 65 relative to the wavelength of the laser light LL, the amount of laser light LL passing through the opening 65 of the joint 60 can be reduced.

[0066] In this way, the joint portion 60 formed along the dividing lines LN1 and LN2 shields the laser light LL irradiated from the lower surface side of the second substrate 350, thereby reducing the amount of laser light LL that is not absorbed inside the second substrate 350 and enters the first substrate 330. Therefore, damage to the device parts such as the electrode pads 41A to 41D formed on the first substrate 330 caused by the laser light LL can be suppressed.

[0067] In addition, as described above, the distance PD between each electrode pad 41A to 41D and the dividing lines LN1 and LN2 on the upper surface of the first substrate 330 is greater than 1 / 2 of the width AW of the opening 65. Therefore, since each electrode pad 41A to 41D is arranged outside the opening 65 relative to the dividing lines LN1 and LN2, even if the laser LL irradiated along the dividing lines LN1 and LN2 from the lower surface side of the second substrate 350 passes through the opening 65, the possibility of acting on each electrode pad 41A to 41D formed on the upper surface of the first substrate 330 or the resin film L32 formed thereunder can be reduced. Therefore, damage to each electrode pad 41A to 41D formed on the first substrate 330 caused by the laser LL can be further suppressed.

[0068] In addition, Figure 7 , an example is shown in which the modified region MA2 is formed at one location inside the second substrate 350, but the present invention is not limited thereto. The number of modified regions MA2 varies depending on the thickness of the second substrate 350. Thus, for example, a plurality of modified regions can be formed in the thickness direction (direction parallel to the Z axis) of the second substrate 350 by irradiating lasers having different focal positions.

[0069] Back to Figure 1 Next, the first substrate 330 and the second substrate 350 are divided along the dividing lines LN1 and LN2 ( S307 ).

[0070] Specifically, if Figure 8 As shown by the black arrows in the middle, the dicing tape DT is extended in the directions parallel to the positive direction of the Y axis and the negative direction of the Y axis. By extending the dicing tape DT, a layer of Figure 8 The tensile stress indicated by the white arrow in the middle is transmitted to the inside of the second substrate 350 and applied to the crack in the modified area MA2. As a result, the crack is extended and the second substrate 350 is divided. In addition, by further extending the dicing tape DT, a tensile stress is generated on the lower surface of the first substrate 330, and the tensile stress is transmitted to the inside of the first substrate 330 and applied to the crack in the modified area MA1. As a result, the crack is extended and the first substrate 330 is divided.

[0071] In this way, by dividing the first substrate 330 and the second substrate 350 along the dividing lines LN1 and LN2 , the first substrate 330 and the second substrate 350 are singulated (chipped) into individual resonant devices 1 including the upper cover 30 and the MEMS substrate 50 (the lower cover 20 and the resonator 10 ).

[0072] Here, in the case where the joint portion 60 does not have an opening 65, the joint portion 60 is formed across (crosses) the dividing lines LN1 and LN2. Therefore, when the first substrate 330 and the second substrate 350 are divided along the dividing lines LN1 and LN2, a division defect sometimes occurs in the joint portion 60. As a result of this division defect, a product defect such as a protrusion (burr) may occur in the monolithic resonant device 1. Therefore, by having the joint portion 60 have an opening 65 formed along a portion of the dividing lines LN1 and LN2, it is possible to suppress the division defect that may occur when the first substrate 330 and the second substrate 350 are divided along the dividing lines LN1 and LN2.

[0073] In addition, as described above, the width AW of the opening 65 on the upper surface of the second substrate 350 is 20 μm or less. This can achieve both shielding of the laser light LL and suppression of division failure.

[0074] Next, refer to Figures 9 to 12 , a resonance device 1 manufactured using the method for manufacturing a MEMS device according to the first embodiment of the present invention will be described. Fig. 9 It is a perspective view schematically showing the appearance of the resonance device 1 according to the first embodiment of the present invention. Fig.10 It is schematically shown Fig. 9 The structure of the resonance device 1 is shown in an exploded perspective view. Fig.11 It is schematically shown Fig.10 A top view of the structure of the resonator 10 is shown. Fig.12 It is schematically shown Figures 9 to 11 The structure of the cross section of the resonance device 1 along the line XII-XII is shown.

[0075] In the following description, the side of the resonance device 1 where the upper cover 30 is provided is referred to as the upper side (or the front side), and the side where the lower cover 20 is provided is referred to as the lower side (or the back side).

[0076] The resonance device 1 includes a MEMS substrate 50 (a lower cover 20 and a resonator 10) and an upper cover 30. That is, the resonance device 1 is configured by laminating the MEMS substrate 50, the above-described bonding portion 60, and the cover 30 in this order.

[0077] The resonator 10 is a MEMS vibrator manufactured using MEMS technology. The resonator 10 and the upper cover 30 are bonded via a bonding portion 60 .

[0078] The upper cover 30 extends in a flat plate shape along the XY plane, and a flat cubic recess 31 is formed on its back. The recess 31 is surrounded by a side wall 33 and forms a part of the vibration space, which is the space where the resonator 10 vibrates. In addition, the upper cover 30 may have a flat plate structure without the recess 31.

[0079] Four electrode pads 41A, 41B, 41C, and 41D are provided on the surface of the upper cover 30 . Each of the electrode pads 41A, 41B, 41C, and 41D is electrically connected to the resonator 10 .

[0080] The lower cover 20 has a bottom plate 22 in a rectangular flat plate shape arranged along the XY plane, and a side wall 23 extending from the peripheral portion of the bottom plate 22 in the Z-axis direction, that is, the stacking direction of the lower cover 20 and the resonator 10. The lower cover 20 has a recessed portion 21 formed by the surface of the bottom plate 22 and the inner surface of the side wall 23 on the surface facing the resonator 10. The recessed portion 21 forms a part of the vibration space of the resonator 10. In addition, the lower cover 20 may be a flat plate structure without the recessed portion 21.

[0081] like Fig.11 As shown, the resonator 10 is a MEMS vibrator manufactured using MEMS technology. Fig.11 Out-of-plane vibration is performed in the XY plane in the orthogonal coordinate system of the resonator 10. In addition, the resonator 10 is not limited to a resonator using an out-of-plane bending vibration mode. The resonator of the resonance device 1 can use, for example, an extended vibration mode, a longitudinal thickness vibration mode, a Lamb wave vibration mode, an in-plane bending vibration mode, and a surface wave vibration mode. These vibrators are used, for example, in timing devices, RF filters, duplexers, ultrasonic transducers, gyroscope sensors, acceleration sensors, etc. In addition, it can also be used for piezoelectric mirrors with actuator functions, piezoelectric gyroscopes, piezoelectric microphones with pressure sensor functions, ultrasonic vibration sensors, etc. It can also be used for electrostatic MEMS elements, electromagnetically driven MEMS elements, and piezoresistive MEMS elements.

[0082] The resonator 10 includes a vibrating portion 120 , a holding portion 140 , and a holding arm 110 .

[0083] The holding portion 140 is formed in a rectangular frame shape so as to surround the outer side of the vibration portion 120 along the XY plane. For example, the holding portion 140 is formed integrally with a prism-shaped frame body. In addition, the holding portion 140 only needs to be provided at least partially around the vibration portion 120, and is not limited to a frame shape.

[0084] The holding arm 110 is provided inside the holding portion 140 , and connects the vibration portion 120 and the holding portion 140 .

[0085] The vibration part 120 is provided inside the holding part 140, and a space is formed between the vibration part 120 and the holding part 140 at a predetermined interval. Fig.11In the example shown, the vibration part 120 includes a base 130 and four vibration arms 135A to 135D (hereinafter, collectively referred to as "vibration arms 135"). The number of vibration arms is not limited to four, and can be set to any number of one or more, for example. In the present embodiment, each vibration arm 135A to 135D is formed integrally with the base 130.

[0086] In a plan view, the base 130 has long sides 131a and 131b in the X-axis direction and short sides 131c and 131d in the Y-axis direction. The long side 131a is one side of the surface of the front end of the base 130 (hereinafter, also referred to as "front end 131A"), and the long side 131b is one side of the surface of the rear end of the base 130 (hereinafter, also referred to as "rear end 131B"). In the base 130, the front end 131Ah and the rear end 131B are arranged to face each other.

[0087] The base 130 is connected to the vibration arm 135 at the front end 131A and to the holding arm 110 described later at the rear end 131B. Fig.11 In the example shown, the base 130 has a substantially rectangular shape when viewed from above, but is not limited to this. The base 130 only needs to be formed to be roughly plane-symmetrical with respect to a virtual plane P defined by a perpendicular bisector along the long side 131a. For example, the base 130 can be a trapezoid in which the long side 131b is shorter than 131a, or can be in the shape of a semicircle with the long side 131a as the diameter. In addition, the surfaces of the base 130 are not limited to flat surfaces, but can also be curved surfaces. In addition, the virtual plane P is a plane passing through the center of the direction in which the vibration arms 135 are arranged in the vibration portion 120.

[0088] In the base 130, the longest distance between the front end 131A and the rear end 131B in the direction from the front end 131A to the rear end 131B, that is, the base length, is about 35 μm. In addition, the longest distance between the side ends of the base 130 in the width direction perpendicular to the base length direction, that is, the base width, that is, is about 265 μm.

[0089] The vibration arms 135 extend along the Y-axis direction and have the same size. The vibration arms 135 are respectively arranged between the base 130 and the holding portion 140 in parallel with the Y-axis direction, one end of which is connected to the front end 131A of the base 130 to become a fixed end, and the other end becomes an open end. In addition, the vibration arms 135 are respectively arranged in parallel at a predetermined interval in the X-axis direction. In addition, the width of the vibration arm 135 in the X-axis direction is about 50 μm, and the length in the Y-axis direction is about 465 μm.

[0090] The vibration arm 135 is wider in the X-axis direction than other parts of the vibration arm 135 at a portion about 150 μm away from the open end, for example. The portion where the width is widened is called a weight portion G. The weight portion G is, for example, 10 μm wider than other parts of the vibration arm 135 in the left and right widths in the X-axis direction, and the width in the X-axis direction is about 70 μm. The weight portion G is formed integrally with the vibration arm 135 through the same process as the vibration arm 135. By forming the weight portion G, in the vibration arm 135, the weight per unit length of the fixed end side of the open end side is heavier than the weight per unit length of the fixed end side. Therefore, by having each vibration arm 135 have a weight portion G on the open end side, the amplitude of the vibration in the up and down directions in each vibration arm can be increased.

[0091] A protective film 235 described later is formed on the surface of the vibration part 120 (the surface facing the upper cover 30) so as to cover the entire surface. In addition, a frequency adjustment film 236 is formed on the surface of the protective film 235 at the front end of the open end side of the vibration arms 135A to 135D. The resonant frequency of the vibration part 120 can be adjusted by the protective film 235 and the frequency adjustment film 236.

[0092] In addition, in the present embodiment, substantially the entire surface of the resonator 10 (the surface on the side opposite to the upper cover 30) is covered by the protective film 235. Furthermore, substantially the entire surface of the protective film 235 is covered by the parasitic capacitance reducing film 240. The protective film 235 only needs to cover at least the vibration arm 135, and is not limited to a structure that covers substantially the entire surface of the resonator 10.

[0093] like Fig.12 As shown, in the resonator device 1, the holding portion 140 of the resonator 10 is joined to the side wall 23 of the lower cover 20, and the holding portion 140 of the resonator 10 is joined to the side wall 33 of the upper cover 30. In this way, the resonator 10 is held between the lower cover 20 and the upper cover 30, and a vibration space in which the vibration arm 135 vibrates is formed by the lower cover 20, the upper cover 30, and the holding portion 140 of the resonator 10. In addition, electrode pads 41A and 41B are formed on the upper surface of the upper cover 30 (the surface opposite to the surface facing the resonator 10). The electrode pads 41A and 41B and the resonator 10 are electrically connected through the through electrodes V1 and V2, the connection wiring 70, and the contact electrodes 76A and 76B.

[0094] The upper cover 30 is bonded to the holding portion 140 of the resonator 10 at its peripheral portion (side wall 33) via the bonding portion 60. The lower surface of the upper cover 30 facing the resonator 10, the upper surface, and the side surfaces of the through electrodes V1 and V2 are covered with the silicon oxide film L31.

[0095] In addition, the getter layer 34 is formed on the surface of the side facing the resonator 10 in the recessed portion 31 of the upper cover 30. The getter layer 34 is formed of, for example, titanium (Ti) or the like, and absorbs exhaust gas generated in the vibration space. Since the getter layer 34 is formed on almost the entire surface of the surface facing the resonator 10 in the recessed portion 31 in the upper cover 30 of the present embodiment, it is possible to suppress the reduction of the vacuum degree of the vibration space.

[0096] The through-electrode V1 of the upper cover 30 functions as a wiring that electrically connects the electrode pad 41A and the connection wiring 70 . The through-electrode V2 of the upper cover 30 functions as a wiring that electrically connects the electrode pad 41B and the connection wiring 70 .

[0097] The bottom plate 22 and the side wall 23 of the lower cover 20 are integrally formed by the Si wafer L1. In addition, the lower cover 20 is bonded to the holding portion 140 of the resonator 10 via the upper surface of the side wall 23.

[0098] The holding portion 140, the base portion 130, the vibration arm 135, and the holding arm 110 in the resonator 10 are integrally formed by the same process. The resonator 10 forms a piezoelectric film F3 on a Si substrate F2 to cover the Si substrate F2, and further stacks a metal layer E2 on the piezoelectric film F3. Furthermore, the piezoelectric film F3 is stacked on the metal layer E2 to cover the metal layer E2, and the metal layer E1 is stacked on the piezoelectric film F3. A protective film 235 is stacked on the metal layer E1 to cover the metal layer E1, and a parasitic capacitance reduction film 240 is stacked on the protective film 235.

[0099] On the lower surface of the Si substrate F2, a silicon oxide (eg, SiO2) layer F21 is formed as an example of a temperature characteristic correction layer.

[0100] The metal layers E1 and E2 have a thickness of, for example, about 0.1 μm to about 0.2 μm, and after film formation, they are patterned into a desired shape by etching or the like. The metal layers E1 and E2 use a body-centered structure metal. Specifically, the metal layers E1 and E2 are formed using Mo (molybdenum), tungsten (W), or the like.

[0101] The metal layer E1 is formed to realize the role of an upper electrode on the vibrating portion 120 , for example. Also, the metal layer E1 is formed to realize the role of a wiring for connecting the upper electrode to an AC power source provided outside the resonator 10 on the supporting arm 110 and the supporting portion 140 .

[0102] On the other hand, the metal layer E2 is formed to realize the role of a lower electrode on the vibration part 120. In addition, the metal layer E2 is formed to realize the role of a wiring for connecting the lower electrode to the circuit provided outside the resonator 10 on the holding arm 110 and the holding part 140.

[0103] The piezoelectric film F3 is a piezoelectric film that converts an applied voltage into vibration. The piezoelectric film F3 is formed of a material having a wurtzite-type hexagonal crystal structure, and may be composed mainly of nitrides or oxides such as aluminum nitride (AlN), scandium aluminum nitride (ScAlN), zinc oxide (ZnO), gallium nitride (GaN), and indium nitride (InN). Among them, scandium aluminum nitride is obtained by replacing part of the aluminum in aluminum nitride with scandium, and scandium may be replaced by two elements such as magnesium (Mg) and niobium (Nb), or magnesium (Mg) and zirconium (Zr). The piezoelectric film F3 has a thickness of 1 μm, for example, but a thickness of about 0.2 μm to 2 μm may also be used.

[0104] The piezoelectric film F3 expands and contracts in the in-plane direction of the XY plane, i.e., in the Y-axis direction, according to the electric field applied to the piezoelectric film F3 through the metal layers E1 and E2. Due to the expansion and contraction of the piezoelectric film F3, the vibration arm 135 displaces its free end toward the inner surface of the lower cover 20 and the upper cover 30 and vibrates in an out-of-plane bending vibration mode.

[0105] In the present embodiment, the phase of the electric field applied to the outer vibration arms 135A and 135D and the phase of the electric field applied to the inner vibration arms 135B and 135C are set to be opposite to each other. As a result, the outer vibration arms 135A and 135D and the inner vibration arms 135B and 135C are displaced in opposite directions. For example, if the free ends of the outer vibration arms 135A and 135D are displaced toward the inner surface of the upper cover 30, the free ends of the inner vibration arms 135B and 135C are displaced toward the inner surface of the lower cover 20.

[0106] The protective film 235 prevents oxidation of the upper electrode for piezoelectric vibration, that is, the metal layer E2. Preferably, the protective film 235 is formed of a material whose mass reduction rate based on etching is slower than that of the frequency adjustment film 236. The mass reduction rate is represented by the etching rate, that is, the product of the thickness removed per unit time and the density. For example, in addition to piezoelectric films such as aluminum nitride (AlN), scandium aluminum nitride (ScAlN), zinc oxide (ZnO), gallium nitride (GaN), and indium nitride (InN), the protective film 235 is also formed of insulating films such as silicon nitride (SiN), silicon oxide (SiO2), and aluminum oxide (Al2O3). The thickness of the protective film 235 is, for example, about 0.2μm.

[0107] The frequency adjustment film 236 is formed on substantially the entire surface of the vibration part 120, and then formed only in a predetermined area by etching or other processing. The frequency adjustment film 236 is formed of a material whose quality is reduced faster than that of the protective film 235 by etching. Specifically, the frequency adjustment film 236 is made of a metal such as molybdenum (Mo), tungsten (W), gold (Au), platinum (Pt), nickel (Ni), and titanium (Ti).

[0108] In addition, if the relationship between the mass reduction rates of the protective film 235 and the frequency adjustment film 236 is as described above, the magnitude relationship of the etching rates is arbitrary.

[0109] The parasitic capacitance reduction film 240 is formed of tetraethyl orthosilicate (TEOS). The thickness of the parasitic capacitance reduction film 240 is about 1 μm. It reduces the parasitic capacitance in the detour wiring portion, and has the function of an insulating layer when wirings of different potentials cross each other, and the function of isolation for expanding the vibration space.

[0110] The bonding portion 60 bonds the MEMS substrate 50 and the upper cover 30 to seal the vibration space of the resonator 10. Thus, the vibration space is hermetically sealed and a vacuum state is maintained.

[0111] The material of the joint 60 is a metal that is transparent to the upper cover 30 and the MEMS substrate 50 (the lower cover 20 and the resonator 10) and can block the laser light LL. As a result, during the separation, the laser light LL irradiated from the lower surface side of the MEMS substrate 50 (the lower cover 20 and the resonator 10) is blocked by the joint 60, and the amount of light incident on the upper cover 30 can be reduced. Therefore, damage to the device parts such as the electrode pads 41A to 41D formed on the upper cover 30 caused by the laser light LL can be suppressed.

[0112] exist Fig.12 In the example shown, the aluminum (Al) layer 61 , the germanium (Ge) layer 62 , and the aluminum (Al) layer 63 are described as independent layers, but in reality, these interfaces are eutectic-bonded.

[0113] The joint 60 is disposed on the upper surface of the MEMS substrate 50 (the lower cover 20 and the resonator 10) at a first distance D1 from the outer edge. This can suppress product defects of the resonator 1 such as protrusions (burrs) caused by poor division that may occur when the joint 60 is not at the first distance D1.

[0114] The first distance D1 of the bonding portion 60 is, for example, 20 μm or less. This can achieve both shielding of the laser light LL and suppression of product failure of the resonance device 1 .

[0115] Each electrode pad 41A to 41D is arranged on the upper surface of the upper cover 30 at a second distance D2 greater than the first distance D1 from the outer edge (the first distance D1 < the second distance D2). Thus, since each electrode pad 41A to 41D is arranged on the inner side of the bonding portion 60 relative to the outer edge, the possibility of the laser LL irradiated from the lower surface side of the MEMS substrate 50 (the lower cover 20 and the resonator 10) acting on each electrode pad 41A to 41D during the division can be reduced. Therefore, the damage to each electrode pad 41A to 41D formed on the upper cover 30 caused by the laser LL can be further suppressed.

[0116] The second distance D2 between the electrode pads 41A to 41D is, for example, not less than 1 / 6 of the thickness (length in the Z-axis direction) of the upper cover 30. This prevents the electrode pads 41A to 41D from interfering with the laser light LL irradiated from the upper surface side of the upper cover 30 during division.

[0117] The connection wiring 70 is electrically connected to the electrode pad 41A via the through electrode V1 and to the contact electrode 76A. In addition, the connection wiring 70 is electrically connected to the electrode pad 41B via the through electrode V2 and to the contact electrode 76B.

[0118] The contact electrode 76A is formed to contact the metal layer E1 of the resonator 10 and electrically connect the connection wiring 70 and the resonator 10. The contact electrode 76B is formed to contact the metal layer E2 of the resonator 10 and electrically connect the connection wiring 70 and the resonator 10. Specifically, when connecting the contact electrode 76A and the metal layer E1, a part of the piezoelectric film F3, the protective film 235, and the parasitic capacitance reduction film 240 stacked on the metal layer E1 is removed, and a through hole V3 is formed to expose the metal layer E1. The inside of the formed through hole V3 is filled with the same material as the contact electrode 76A, thereby connecting the metal layer E1 and the contact electrode 76A. Similarly, when connecting the contact electrode 76B and the metal layer E2, a part of the piezoelectric film F3 and the parasitic capacitance reduction film 240 stacked on the metal layer E2 is removed, and a through hole V4 is formed to expose the metal layer E2. The inside of the formed through hole V4 is filled with the contact electrode 76B, thereby connecting the metal layer E2 and the contact electrode 76B. The contact electrodes 76A and 76B are made of metal such as aluminum (Al), gold (Au), tin (Sn), etc. In addition, it is preferred that the connection position between the metal layer E1 and the contact electrode 76A and the connection position between the metal layer E2 and the contact electrode 76B are in the area outside the vibration part 120, and in the present embodiment, they are connected by the holding part 140.

[0119] <Second embodiment>

[0120] Next, refer to Figure 13 to Figure 16 , a method for manufacturing a MEMS device and a MEMS device according to a second embodiment of the present invention are described. In addition, the same or similar reference numerals are added to the same or similar structures as those in the first embodiment. The following describes the differences from the first embodiment. In addition, the same effects based on the same structure are not mentioned in sequence.

[0121] First, refer to Figure 13 to Figure 15 , a method for manufacturing a MEMS device according to a second embodiment of the present invention is described. Fig.13 1 is a flowchart showing a method for manufacturing the resonance device 1 according to the second embodiment of the present invention. Fig.14 Yes means Fig.13 A cross-sectional view of step S311 is shown. Fig.15 Yes means Fig.13 The cross-sectional view of step S312 shown in FIG. Fig.14 as well as Fig.15 In the present invention, for convenience of explanation, a part of the plurality of resonance devices 1 manufactured by the manufacturing method is shown for explanation.

[0122] like Fig.13 As shown, the manufacturing method of the resonant device 1 involved in the second embodiment of the present invention is different from the manufacturing method of the resonant device 1 involved in the first embodiment of the present invention in that after the dicing tape DT is pasted on the lower surface of the second substrate 350 in step S304, the order of forming the modified area MA1 of the first substrate 330 and the modified area MA2 of the second substrate 350 is replaced.

[0123] That is, first, laser LL is irradiated from the lower surface side of the second substrate 350 to form the modified area MA2 inside the second substrate 350 along the dividing lines LN1 and LN2 ( S311 ).

[0124] like Fig.14 As shown, part of the laser light LL irradiated from the lower surface side of the second substrate 350 is absorbed inside the second substrate 350, and a modified area MA2 is formed around the focal position.

[0125] Furthermore, by moving the laser LL and at least one of the second substrate 350 along the dividing line LN1 in a direction parallel to the X-axis direction, a modified region MA2 along the dividing line LN1 can be formed inside the second substrate 350. Similarly, by moving the laser LL and at least one of the second substrate 350 along the dividing line LN2 in a direction parallel to the Y-axis direction, a modified region MA2 along the dividing line LN2 can be formed inside the second substrate 350.

[0126] Here, when the modified region MA2 is formed, stress is generated starting from the modified region MA2. Fig.14 As shown by the white arrow in the middle, the stress acts in the direction of widening the second substrate 350 (the direction parallel to the Y axis). If the modified region MA1 along the dividing lines LN1 and LN2 is first formed inside the first substrate 330 as in the first embodiment, the first substrate 330 may be warped due to the stress in the direction of widening the first substrate 330. Therefore, a dicing tape DT is attached to the lower surface of the second substrate 350, and before the modified region MA1 is formed inside the first substrate 330, a laser is irradiated from the lower surface side of the second substrate to form the modified region MA2 inside the second substrate 350, so that the modified region MA2 can be formed inside the second substrate 350. Fig.14The adhesive force of the dicing tape DT indicated by the black arrow in the middle counteracts the stress in the direction of widening the second substrate 350 starting from the modified area MA2. Therefore, the warping of the second substrate 350 can be suppressed, and the focus measurement error caused by the warping when irradiating the laser LL can be reduced.

[0127] The resonator device 1 according to the second embodiment of the present invention is different from the resonator device 1 according to the first embodiment of the present invention in that the joint portion 60A does not have an opening.

[0128] The laser light LL which is not absorbed in the second substrate 350 attempts to pass through the second substrate 350 and enter the lower surface of the first substrate 330. The laser light LL is entirely blocked by the joint portion 60A formed along the dividing line LN1.

[0129] Back to Fig.13 Next, laser LL is irradiated from the upper surface side of the first substrate 330 to form a modified area MA1 inside the first substrate 330 along the dividing lines LN1 and LN2 (S312).

[0130] like Fig.15 As shown, part of the laser light LL irradiated from the upper surface side of the first substrate 330 is absorbed inside the first substrate 330 , thereby forming a modified region MA1 .

[0131] Furthermore, by moving the laser LL and at least one of the first substrate 330 along the dividing line LN1 in a direction parallel to the X-axis direction, a modified region MA1 along the dividing line LN1 can be formed inside the first substrate 330. Similarly, by moving the laser LL and at least one of the first substrate 330 along the dividing line LN2 in a direction parallel to the Y-axis direction, a modified region MA1 along the dividing line LN2 can be formed inside the first substrate 330.

[0132] In this way, after laser LL is irradiated from the lower surface side of the second substrate 350 to form the modified region MA2 inside the second substrate 350, laser LL is irradiated from the upper surface side of the first substrate 330 to form the modified region MA1 inside the first substrate 330 along the dividing lines LN1 and LN2. Compared with the case where the modified region MA2 is formed inside the second substrate 350 before the modified region MA1 is formed inside the first substrate 330 as in the first embodiment, the bending of the first substrate 330 can be suppressed. In addition, by applying stress to the modified region MA1 of the first substrate 330, the first substrate 330 can be easily divided.

[0133] The laser light LL which is not absorbed in the second substrate 350 attempts to pass through the second substrate 350 and enter the lower surface of the first substrate 330. The laser light LL is entirely blocked by the bonding portion 60A formed along the dividing line LN1.

[0134] Next, refer to Fig.16 , a MEMS device according to a second embodiment of the present invention is described. Fig.16 It is a cross-sectional view schematically showing the structure of a resonance device 1 according to a second embodiment of the present invention. Fig.16 is with Fig.12 The same cross-sectional view.

[0135] like Fig.16 As shown, the bonding portion 60A is different from the bonding portion 60 of the resonance device 1 according to the first embodiment of the present invention in that it is arranged on the upper surface of the MEMS substrate 50 (the lower cover 20 and the resonator 10) at a distance from the outer edge. That is, in this embodiment, Fig.12 The first distance D1 of the illustrated joint 60 is zero.

[0136] The above is a description of the exemplary embodiment of the present invention. However, in a method for manufacturing a resonant device according to one embodiment of the present invention, the joint is formed along the dividing line to shield the laser irradiated from the lower surface side of the second substrate. As a result, the amount of laser light that is not absorbed inside the second substrate and enters the first substrate can be reduced. Therefore, damage to the device part such as the electrode pad formed on the first substrate caused by the laser can be suppressed.

[0137] In addition, in the aforementioned manufacturing method, a dicing tape is attached to the lower surface of the second substrate, and before the modified region is formed inside the first substrate, a laser is irradiated from the lower surface side of the second substrate to form the modified region inside the second substrate, thereby enabling Fig.14 The adhesive force of the dicing tape indicated by the black arrow in the middle counteracts the stress in the direction of expansion of the second substrate starting from the modified region, thereby suppressing the bending of the second substrate and reducing the focus measurement error during laser irradiation caused by the bending.

[0138] In addition, in the aforementioned manufacturing method, after laser irradiation is performed from the lower surface side of the second substrate to form a modified region inside the second substrate, laser irradiation is performed from the upper surface side of the first substrate to form a modified region inside the first substrate along the dividing line. Thus, compared with the case where the modified region is formed inside the first substrate before the modified region is formed inside the second substrate, the bending of the first substrate can be suppressed. In addition, by applying stress to the modified region of the first substrate, the first substrate can be easily divided.

[0139] In the above-mentioned manufacturing method, the joint portion has an opening formed along a portion of the dividing line. This can suppress division failure that may occur when the first substrate and the second substrate are divided along the dividing line.

[0140] In addition, in the above-mentioned manufacturing method, the width of the dividing line on the upper surface of the first substrate is greater than 5 μm and less than 20 μm. Thus, when the first substrate and the second substrate are divided along the dividing line and monolithicized (chip-ized) into each resonant device (each chip), each resonant device can be miniaturized and the number of resonant devices can be increased. In addition, the width of the opening on the upper surface of the second substrate is less than 20 μm. Thus, both the shielding of the laser and the suppression of poor division can be achieved.

[0141] In addition, in the aforementioned manufacturing method, the distance between each electrode pad and the dividing line on the upper surface of the first substrate is greater than 1 / 2 of the width of the opening. Thus, since each electrode pad is arranged on the outer side of the dividing line relative to the opening, even if the laser irradiated along the dividing line from the lower surface side of the second substrate passes through the opening, the possibility of acting on each electrode pad formed on the upper surface of the first substrate or the resin film formed thereunder can be reduced. Therefore, damage to each electrode pad formed on the first substrate caused by the laser can be further suppressed.

[0142] In addition, in the aforementioned manufacturing method, the distance between each electrode pad and the dividing line on the upper surface of the first substrate is at least 1 / 6 of the thickness (length in the Z-axis direction) of the first substrate. Thus, each electrode pad can be prevented from interfering with the laser irradiated along the dividing line from the upper surface side of the first substrate.

[0143] In the resonant device according to one embodiment of the present invention, the material of the joint portion 60 is a metal that is transparent to the upper cover 30 and the MEMS substrate 50 (the lower cover 20 and the resonator 10) and can shield the laser light LL. As a result, the laser light LL irradiated from the lower surface side of the MEMS substrate 50 (the lower cover 20 and the resonator 10) during the division is shielded by the joint portion 60, and the amount of light incident on the upper cover 30 can be reduced. Therefore, damage to the device parts such as the electrode pads 41A to 41D formed on the upper cover 30 caused by the laser light LL can be suppressed.

[0144] In addition, in the aforementioned resonant device, the joint portion is arranged on the upper surface of the MEMS substrate (lower cover and resonator) at a first distance from the outer edge. Thus, it is possible to suppress product defects of the resonant device such as protrusions (burrs) due to poor segmentation that may occur when the joint portion is not separated by the first distance.

[0145] In the above-described resonance device, the first distance D1 is equal to or less than 20 μm. This can achieve both shielding of laser light and suppression of product failure of the resonance device.

[0146] In addition, in the aforementioned resonant device, each electrode pad is arranged on the upper surface of the upper cover at a second distance greater than the first distance from the outer edge (the first distance < the second distance). Thus, since each electrode pad is arranged on the inner side of the joint relative to the outer edge, the possibility of the laser irradiated from the lower surface side of the MEMS substrate (lower cover and resonator) acting on each electrode pad during division can be reduced. Therefore, damage to each electrode pad formed on the upper cover caused by the laser can be further suppressed.

[0147] In the above-described resonance device, the second distance is equal to or greater than 1 / 6 of the thickness (length in the Z-axis direction) of the upper cover. This prevents the electrode pads from interfering with the laser light irradiated from the upper surface of the upper cover during division.

[0148] In addition, the above-described embodiments are for easy understanding of the content of the present invention, and are not for limiting the content of the present invention. The present invention can be changed / improved without departing from its purport, and the present invention also includes its equivalents. That is, as long as the method in which the technical personnel in this field appropriately applies the design changes to each embodiment has the features of the present invention, it is included in the scope of the present invention. For example, each element and its configuration, material, condition, shape, size, etc. possessed by the embodiment are not limited to the illustrated content and can be appropriately changed. In addition, the embodiment is illustrative, and of course, the partial replacement or combination of the structure shown in different embodiments can be performed, and these are also included in the scope of the present invention as long as they include the features of the present invention.

[0149] Description of Reference Numerals

[0150] 1 ... resonance device, 10 ... resonator, 20 ... lower cover, 21 ... back surface, 21 ... recessed portion, 22 ... bottom plate, 23 ... side wall, 30 ... upper cover, 31 ... recessed portion, 33 ... side wall, 34 ... gettering layer, 41A, 41B, 41C, 41D ... electrode pad, 50 ... MEMS substrate, 60, 60A ... bonding portion, 61 ... aluminum layer, 62 ... germanium layer, 63 ... aluminum (Al) layer, 65 ... opening, 70 ... connection wiring, 76A, 76B ... contact electrode, 110 ... holding arm, 120 ... vibration portion, 130 ... base portion, 131a ... long side, 131A ... front end, 131b ... long side, 131B ... rear end, 131c, 131d ... short side, 135, 135A, 135 B, 135C, 135D…vibration arm, 140…holding part, 235…protective film, 236…frequency adjustment film, 240…parasitic capacitance reduction film, 330…first substrate, 350…second substrate, AW…width, D1…first distance, D2…second distance, DT…dicing tape, E1, E2…metal layer, F2…Si substrate, F3…piezoelectric film, F21…silicon oxide layer, G…weighting part, L1…Si wafer, L3…Si wafer, L31…silicon oxide film, L32…resin film, LL…laser, LN1, LN2…dividing line, MA1, MA2…modified area, P…virtual plane, PD…distance, V1, V2…through electrodes, V3, V4…through holes.

Claims

1. A method for manufacturing a MEMS device, comprising: A step of irradiating a laser beam from the second main surface side of the second substrate so as to form a modified region inside the second substrate along a dividing line between the first substrate and the second substrate, wherein a device portion is formed on the first main surface of the first substrate, and the first main surface of the second substrate is bonded to the second main surface of the first substrate via a bonding portion; and a step of applying stress to the modified region to separate the first substrate and the second substrate along the separation line; The joining portion is formed along the dividing line to shield the laser. The device portion includes a plurality of electrode pads, and the plurality of electrode pads are arranged on the device portion to be separated from the dividing line on the first main surface of the first substrate. The plurality of electrode pads are arranged in a direction perpendicular to the first main surface so as not to overlap with the dividing line.

2. The method for manufacturing a MEMS device according to claim 1, wherein: Before the irradiation step, the method further includes a step of attaching a dicing tape to the second main surface of the second substrate.

3. The method for manufacturing a MEMS device according to claim 1, wherein: After the irradiation step, the method further includes irradiating the first substrate with the laser light from the first main surface side so as to form a modified region inside the first substrate along the dividing line.

4. The method for manufacturing a MEMS device according to claim 2, wherein: After the irradiation step, the method further includes irradiating the first substrate with the laser light from the first main surface side so as to form a modified region inside the first substrate along the dividing line.

5. The method for manufacturing a MEMS device according to any one of claims 1 to 4, wherein: The joining portion has an opening formed along a portion of the dividing line.

6. The method for manufacturing a MEMS device according to claim 5, wherein: The width of the dividing line on the first main surface of the first substrate is 5 μm or more and 20 μm or less, A width of the opening on the first main surface of the second substrate is 20 μm or less.

7. The method for manufacturing a MEMS device according to claim 6, wherein: The distance between each of the electrode pads and the dividing line on the first main surface of the first substrate is greater than 1 / 2 of the width of the opening.

8. The method for manufacturing a MEMS device according to claim 7, wherein: A distance between each of the electrode pads and the dividing line on the first main surface of the first substrate is greater than 1 / 6 of a thickness of the first substrate.

9. A MEMS device having: a first substrate having a device portion formed on a first main surface; and The second substrate has a first main surface bonded to the second main surface of the first substrate via a bonding portion, The material of the bonding portion is a metal that is transparent to the first substrate and the second substrate and can shield the laser. The device portion includes a plurality of electrode pads, and the plurality of electrode pads are arranged on the device portion to be separated from a dividing line between the first main surface of the first substrate and the second substrate. The plurality of electrode pads are arranged in a direction perpendicular to the first main surface so as not to overlap with the dividing line.

10. The MEMS device according to claim 9, wherein: The bonding portion is disposed on the first main surface of the second substrate at a first distance from an outer edge.

11. The MEMS device according to claim 10, wherein: The first distance is less than 20 μm.

12. The MEMS device according to claim 10 or 11, wherein: The electrode pads are respectively arranged on the first main surface of the first substrate at a second distance from an outer edge that is larger than the first distance.

13. The MEMS device according to claim 12, wherein: The second distance is greater than 1 / 6 of the thickness of the first substrate.

Images