JP2010117247A - Mems sensor and method for manufacturing the same - Google Patents

Abstract

<P>PROBLEM TO BE SOLVED: To provide a MEMS sensor for simplifying a structure, reducing a thickness of the sensor, electrically stabilized, and effectively increasing the mass of a mass section in comparison with a conventional sensor, and its manufacturing method. <P>SOLUTION: The MEMS sensor has first and second substrates 1, 2 formed from silicon, and oppositely disposed in the height direction. The first substrate 1 is formed with the mass section 4, and an anchor section 5 connected to the mass section 4. The anchor section 5 is fixed to and supported by the second substrate 2. A space 6 is provided between the mass section 4 and the second substrate 2. A metal 7 having a specific gravity larger than the silicon is embedded in the mass section 4. <P>COPYRIGHT: (C)2010,JPO&INPIT

Description

  The present invention relates to a MEMS sensor such as an acceleration sensor or an angular velocity sensor formed using MEMS (Micro-Electro-Mechanical Systems) technology.

  A MEMS (Micro-Electro-Mechanical Systems) sensor is configured, for example, by forming a mass part (weight part), a movable electrode, a fixed electrode, and the like on one silicon substrate constituting an SOI (Silicon on Insulator) substrate.

  Patent Document 1 below discloses an invention of a semiconductor device in which a mass portion (weight portion) is formed of a material having a specific gravity greater than that of a semiconductor substrate material (claim 1, column [0035] of FIG. 1, FIG. 1).

  Patent Document 2 discloses an invention of an angular velocity sensor in which the thickness of a mass part (weight part) is made thinner than the thickness of a support beam (claim 2, column [0010] of Patent Document 2).

  Patent Document 3 discloses an invention of an angular velocity sensor in which a metal is passed through a mass part (mass part) and the mass of the mass part is increased (claim 1, column [0018] in Patent Document 3). , FIG. 1 etc.).

The sensor sensitivity can be improved by increasing the mass of the mass part. In addition, the resonance frequency necessary for driving and the like can be reduced, and the effect of reducing the IC design load can be expected.
JP 2006-349525 A JP 2000-329561 A JP 2003-50249 A

However, the invention described in each of the above patent documents has the following problems.
First, there was a problem that could not contribute to the thinning of the sensor. In Patent Document 1, the “weight part 3” is provided so as to be convex downward, and in Patent Document 2, the thickness of the mass part is formed thick, so the height at the part of the mass part (weight part). The dimensions cannot be effectively thinned. Further, in Patent Document 3, a metal is made to penetrate the mass part, but the metal protrudes in a convex shape from the upper surface and the lower surface of the mass part as described in FIG. Effectively, the thickness of the mass portion cannot be reduced.

  Also, in the [0023] column of Patent Document 1, “the weight of the weight portion is reduced to reduce the thickness, or the entire size (volume) is equal to or smaller than the conventional size (volume). There is a description of "

  However, even if the weight portion can be reduced in thickness, the weight portion cannot be easily formed using the SOI substrate in the structure of Patent Document 1, as will be described later. Further, it seems that it is possible to form this weight portion on the upper surface side of “beam portion 1” shown in FIG. 1 of Patent Document 1, but in this case, it is difficult to form “piezoresistive element 4” shown in FIG. Become. Further, in this case, the upper surface of the “beam portion 1” is formed as an uneven surface (the same problem occurs in Patent Document 3). Then, as will be described below, it becomes difficult to use the upper surface as a formation region for a wiring layer or the like.

  Further, in the inventions described in Patent Document 1 and Patent Document 3, since the metal provided in the mass portion is exposed, there is a concern about a decrease in electrical stability. Since the metal is exposed, for example, when the wiring layer is provided so as to cover the mass part, the formation of the wiring layer is made complicated by bypassing the metal portion, In the design in which the wiring layers approach each other, further structural ingenuity is necessary to ensure electrical insulation.

  Furthermore, in the invention described in each patent document, it is considered that the production is complicated or the production is difficult. For example, in Patent Document 1, a normal SOI substrate cannot be used until the sensor is completed. In FIG. 2 of Patent Document 1, only the formation of the “beam portion 1” is disclosed. However, since a support mechanism that finally supports the “beam portion 1” is necessary, the manufacture of this support mechanism is also included. Together, the production is thought to be very complicated.

  Further, in Patent Document 2, in order to make the thickness of the support beam thinner than the thickness of the mass part (weight part), for example, the photolithography technique is used to form the support beam part so that the film thickness is reduced. I need it. Furthermore, a high degree of alignment accuracy is required to cut out the portion of the support beam formed in this way from the substrate.

  Further, in the invention described in Patent Document 3, a “hole 14b” is formed in the “mass portion 14” shown in FIG. 2 of Patent Document 3, and a low melting point metal is filled in the “hole 14b”. It is described that the metal is bonded to the “mass portion 14” in a state of protruding from both the upper and lower directions of the “hole 14b” (column [0018] in Patent Document 3). There is also a possibility that a part does not stay in the “hole 14b” but falls downward and adheres to the upper surface of the “silicon substrate 16” shown in FIG.

  Accordingly, the present invention solves the above-described conventional problems, and in particular, effectively increases the mass of the mass part with a thin sensor and electrical stability, etc., with a simple structure as compared with the conventional one. It is an object of the present invention to provide a MEMS sensor that can be used and a manufacturing method thereof.

The MEMS sensor of the present invention is
The first substrate has a first substrate and a second substrate disposed to face each other in the height direction, and the first substrate has a mass portion and an anchor portion connected to the mass portion, The anchor portion is fixedly supported on the second substrate side,
A space is provided between the mass part and the second substrate, and a metal having a specific gravity larger than that of the first substrate is embedded in the mass part. It is.

  As a result, it is possible to effectively increase the mass of the mass portion with a simple structure and a reduction in thickness, electrical stability, and the like as compared with the conventional case.

  In the present invention, the first substrate is provided with a detection unit including a movable electrode and a fixed electrode for detecting a displacement amount of the mass unit, and the movable electrode is formed on the space in the space. It is formed integrally with a mass part, the fixed electrode is formed separately from the movable electrode, and the anchor part connected to the fixed electrode is fixedly supported on the second substrate. Is preferred. For example, the present invention can be effectively applied to acceleration sensors and angular velocity sensors having such an electrode structure.

  Moreover, the manufacturing method of the MEMS sensor in this invention has the following processes, It is characterized by the above-mentioned.

(A) forming a through hole in the first substrate;
(B) forming a bottom at the bottom of the through hole;
(C) a step of embedding a metal having a higher specific gravity than the first substrate in the through hole;
(D) forming a lid over the first substrate and the metal;
(E) etching the first substrate in a state where the second substrate is bonded to the lower surface via an insulating layer to form a mass portion from the first substrate and an anchor portion connected to the mass portion; At this time, a step of forming the mass part in which the metal is embedded,
(F) A step of leaving the insulating layer between the anchor portion and the second substrate and removing the insulating layer located between the mass portion and the second substrate to form a space.

  By the manufacturing method described above, the metal can be easily and appropriately embedded in the mass part. At this time, the thickness of the mass part can be made thinner than the conventional one, and the sensor can be made thinner. Further, since the metal is not exposed from the upper and lower surfaces of the mass part, it is not affected by the etching in the step of removing the insulating layer in the step (F), and a sensor having excellent electrical stability can be formed. Since the upper and lower surfaces can be formed as flat surfaces, a wiring layer or the like can be freely and easily formed on the surface of the mass portion.

Moreover, in this invention, it replaces with the said (A) process and the said (B) process,
(G) forming a bottomed recess in the first substrate;
May be performed. In the above manufacturing method, the step (B) is not particularly necessary, and the number of steps can be reduced.

  Further, in the present invention, before the step (D) is performed, a flattening process is performed on the metal from the first substrate, and then the lid portion in the step (D) is formed. Is preferred. Thereby, a cover part can be formed appropriately and the upper surface of a mass part can be appropriately formed in a flat surface.

Moreover, in this invention, it replaces with the said (E) process and the said (F) process,
(H) etching the first substrate in a state in which a second substrate having a depressed region is bonded to a region facing the first substrate to be a mass unit, the mass unit from the first substrate, and Forming an anchor portion to be connected to the mass portion, and forming the mass portion embedded with the metal in the space between the second substrate and the depression region;
It may have.

  According to the MEMS sensor of the present invention, it is possible to effectively increase the mass of the mass part with a thin structure and electrical stability as compared with the conventional sensor.

  Moreover, according to the manufacturing method of the MEMS sensor of the present invention, the metal can be easily and appropriately embedded in the mass portion. At this time, the thickness of the mass portion can be made thinner than the conventional one, and the sensor can be made thinner. In addition, since the metal is not exposed from the upper and lower surfaces of the mass part, it is not affected by etching in the process of removing the insulating layer, and a sensor having excellent electrical stability can be formed. Therefore, a wiring layer or the like can be freely and easily formed on the surface of the mass part.

  1 is a cross-sectional view conceptually showing a MEMS sensor according to the present embodiment, FIG. 2 is a partially enlarged cross-sectional view showing an enlarged mass part shown in FIG. 1, and FIG. 3 is a MEMS sensor showing a modification. FIG.

The MEMS sensor S shown in FIG. 1 includes, for example, a first substrate 1 and a second substrate (support substrate) 2 made of silicon, and SiO interposed between the first substrate 1 and the second substrate 2. _The insulating layer 3 formed of ₂ is provided.

  As shown in FIG. 1, a mass part (weight part) 4 and an anchor part 5 connected to the mass part 4 are formed on the first substrate 1. In FIG. 1, the mass portion 4 and the anchor portion 5 are not connected in the drawing, but the mass portion 4 and the anchor portion 5 are connected via a support beam (not shown) at a position different from the cross-sectional position in FIG. ing.

  As shown in FIG. 1, the anchor portion 5 is fixedly supported on the second substrate 2 via an insulating layer 3. On the other hand, the insulating layer 3 is not formed between the mass part 4 and the second substrate 2, and a space 6 is formed.

  The thickness of the mass part 4 matches the thickness of the first substrate 1, and the upper surface 4a and the lower surface 4b of the mass part 4 are flat surfaces.

  As shown in FIG. 1, a metal 7 is embedded in the mass portion 4. The metal 7 is not exposed from the side surface of the mass part 4 or from the upper surface 4a and the lower surface 4b. The metal 7 is made of a material having a specific gravity greater than that of silicon. Thereby, the mass of the whole mass part 4 can be enlarged, and the improvement of a sensor sensitivity etc. can be implement | achieved. For example, if the MEMS sensor S in FIG. 1 is an angular velocity sensor, the Coriolis force can be increased by increasing the mass of the mass unit 4.

  Further, as shown in FIG. 1, the metal 7 is embedded in the mass portion 4, thereby making the thickness h <b> 1 (see FIG. 2) of the mass portion 4 thinner than that in the past, and further, the mass portion 4. Even if the size in the planar direction is reduced, the mass of the entire mass part 4 can be effectively increased. Therefore, according to the present embodiment, it is possible to contribute to the reduction in thickness of the sensor and the downsizing of the entire sensor.

  Moreover, as shown in FIG. 1, since the metal 7 is embedded in the mass part 4, it is excellent also in electrical stability. For example, even when a displacement of the mass part 4 is detected by a parallel plate electrode structure, or when a detection element, a wiring layer or the like is formed so as to cover the upper surface 4 a of the mass part 4, electrical It is possible to appropriately ensure insulation. Moreover, as shown in FIG. 1, since the upper surface 4a of the mass part 4 is a flat surface, a wiring layer etc. can be freely and easily formed on this upper surface 4a.

  The condition is that the metal 7 has a higher specific gravity than the first substrate 1 made of silicon. For metal 7, tungsten (specific gravity about 8.2 times that of silicon), molybdenum (specific gravity about 4.4 times that of silicon), manganese (specific gravity about 3.2 times that of silicon), etc. can be used. is there.

  As shown in FIG. 2, when the mass part 4 is formed by a manufacturing method described later, the mass part 4 includes a first substrate 1 made of silicon and a through hole 1b formed in the first substrate 1. A first silicon layer 8 formed of silicon filling the bottom, the metal 7, and a second silicon layer (lid portion) 9 covering the upper surface of the first silicon layer 8 from the upper surface of the metal 7. Is done.

  In another embodiment shown in FIG. 3, a recessed region 10 is formed on the surface 2 a of the second substrate 2. The hollow region 10 is formed at a position facing the mass portion 4 in the height direction. Thereby, a space 11 is formed between the mass portion 4 and the second substrate 2.

  In the form of FIG. 3, the insulating layer 3 is formed on the entire surface 2 a of the second substrate 2. However, as in FIG. 1, the insulating layer 3 may have a configuration in which the insulating layer 3 is removed in a region facing the mass portion 4, that is, in the recessed region 10.

  Alternatively, the first substrate 1 and the second substrate 2 may be directly joined without forming the insulating layer 3 shown in FIG.

Next, a method for manufacturing the MEMS sensor S shown in FIG. 1 will be described with reference to FIG.
4A to 4G are cross-sectional views of the MEMS sensor S during the manufacturing process.

In the process of FIG. 4A, for example, the first substrate 1 and the second substrate 2 made of silicon, and SiO ₂ interposed between the first substrate 1 and the second substrate 2 are used. An SOI substrate 20 composed of the formed insulating layer 3 is prepared.

  Next, a resist pattern 21 is formed on the upper surface 1 a of the first substrate 1. A through hole 21 a is formed in the resist pattern 21. The through hole 21 a is formed in a region facing a region where the metal 7 is embedded in the first substrate 1.

  Next, in the step of FIG. 4B, the first substrate 1 not covered with the resist pattern 21 is removed by using deep RIE (Deep RIE).

  In this embodiment, a through hole 1b is formed in the first substrate 1, and the upper surface 3a of the insulating layer 3 is exposed from the through hole 1b. However, it is also possible to finish the RIE at the position of the dotted line B in FIG. 4B and form a bottomed recess in the first substrate 1 by RIE time control or the like. In such a case, the next step of FIG. 4C is not necessary, and the process proceeds to the step of FIG.

  After forming the through hole 1b, in the step of FIG. 4C, a bottom 8a made of silicon is formed at the bottom of the through hole 1b. For example, a polysilicon film is formed by a CVD method. In the step of FIG. 4C, the first silicon layer 8 is formed from the side wall of the through hole 1b to the upper surface 1c of the first substrate 1 including the bottom 8a.

  Next, in the step of FIG. 4D, the metal 7 is embedded in the through hole 1 b formed in the first substrate 1. In the step of FIG. 4D, for example, a metal paste is printed in the through hole 1b and then baked. In addition, since the bottom 8b made of silicon is formed between the through hole 1b and the insulating layer 3, the embedded metal 7 does not contact the insulating layer 3.

  Next, in the step shown in FIG. 4E, the first silicon layer 8 is formed on the upper surface 1a of the first substrate 1 (in this embodiment, the first silicon layer 8 is formed on the upper surface 1a of the first substrate 1). A planarization process is performed from the upper surface of the layer 8 to the upper surface of the metal 7 by using, for example, a CMP technique. Subsequently, from the flattened upper surface 1a of the first substrate 1 (when the first silicon layer 8 is left on the upper surface 1a, the upper surface of the first silicon layer 8 corresponds) to the upper surface of the metal 7 Then, a second silicon layer (lid portion) 9 is formed. Similar to the first silicon layer 8, the second silicon layer 9 is formed by depositing, for example, polysilicon by a CVD method.

  Next, in the step of FIG. 4F, a resist pattern (not shown) is formed on the second silicon layer 9. This resist pattern has the same shape as the mass part 4 and the anchor part 5 formed on the first substrate 1. Then, the second silicon layer 9, the first silicon layer 8, and the first substrate 1 that are not covered with the resist pattern are removed by using deep RIE (Deep RIE). Thereby, the mass part 4 and the anchor part 5 can be formed. The formation position of the resist pattern can be adjusted with an alignment key (not shown), and the mass portion 4 in which the metal 7 is embedded can be formed with high accuracy.

  4G, the insulating layer 3 between the mass part 4 and the second substrate 2 is removed by an isotropic etching process using wet etching or dry etching. Thereby, a space 6 can be formed between the mass part 4 and the second substrate 2.

In deep RIE, anisotropic etching is mainly performed using sulfur hexafluoride (SF ₆ ). In this embodiment, since the metal 7 is embedded in the mass part 4, the metal 7 can be prevented from being directly affected by deep RIE. In addition, before the step of removing the insulating layer 3, a large number of fine holes leading to the insulating layer 3 are formed in the mass portion 4 so that the insulating layer 3 can be removed by isotropic etching through the fine holes. It is also possible to keep it. However, the fine hole is formed avoiding the formation position of the metal 7. Further, in this step, the insulating layer 3 under the anchor portion 5 is left, and the anchor portion 5 remains fixed and supported on the second substrate 2.

  As shown in FIG. 4, in this embodiment, the MEMS sensor S in which the metal 7 is embedded in the mass portion 4 can be formed simply and appropriately using the SOI substrate 20. In the present embodiment, since the metal 7 is embedded, the thickness of the mass portion 4 can be reduced as compared with the conventional case, and the sensor can be thinned. Further, the size of the mass portion 4 in the planar direction can be reduced, which can contribute to the downsizing of the entire sensor.

  In this embodiment, even if the removal process of the insulating layer 3 constituting the SOI substrate 20 is performed after the metal 7 is buried, the metal 7 can be prevented from being exposed to the etching gas or the etching solution. Furthermore, in this embodiment, the upper and lower surfaces of the mass part 4 can be formed as flat surfaces, and a wiring layer or the like can be freely and easily formed on the surface of the mass part 4.

  Although the SOI substrate 20 is used in FIG. 4, for example, the first substrate 1 and the second substrate 2 are prepared separately, and the process up to the step of FIG. Thereafter, the first substrate 1 and the second substrate 2 can be bonded together via the insulating layer 3, and then the steps of FIGS. 4F and 4G can be performed. For example, the surface of the second substrate 2 is thermally oxidized to form the insulating layer 3, and the first substrate 1 and the second substrate 2 are joined at room temperature via the insulating layer 3.

  If the bonding technique as described above is used, as shown in FIG. 5, the depression region 10 is formed in the second substrate 2 and the first substrate 1 and the second substrate 2 are bonded. You can also. The recessed region 10 is formed in a portion facing the region that becomes the mass portion 4 of the first substrate 1. Thereby, the MEMS sensor S of FIG. 3 can be formed. In FIG. 5, the metal 7 is already embedded in the first substrate 1. For example, the RIE is stopped at the position of the dotted line B with respect to the first substrate 1 in the step of FIG. If the first substrate 1 before embedding the metal 7 and the second substrate 2 provided with the recessed region 10 shown in FIG. It is also possible to perform steps (d), (e), and (f). In the case where the insulating layer 3 is formed on the surface 2a of the second substrate 2 as shown in FIG. 5, whether or not the insulating layer 3 in the recessed region 10 is removed in the step of FIG. It is.

  The configuration of this embodiment can be applied to, for example, the acceleration sensor shown in FIG. 6 and the angular velocity sensor shown in FIG.

  In the acceleration sensor 30 shown in FIG. 6, a mass part 31, an anchor part 32, a support beam 33, and a detection part 34 are formed.

  The detection unit 34 is formed in a curved structure in which comb-like movable electrodes 35 and comb-like fixed electrodes 36 are alternately arranged. The movable electrode 35 is formed integrally with the mass part 31. On the other hand, the fixed electrode 36 is formed separately from the movable electrode 35. An anchor portion 37 is integrally formed at the outer end portion of the fixed electrode 36.

  The acceleration sensor 30 also has a laminated structure of the first substrate 1, the second substrate 2 and the insulating layer 3 as shown in FIG. The anchor portions 32 and 37 described above are fixedly supported by the second substrate 2 through the insulating layer 3, but the insulating layer 3 is removed in other portions, and the space 6 shown in FIG. 1 is formed. Has been.

In the embodiment of FIG. 6, the mass unit 31 is displaced by receiving a force (inertial force) accompanying acceleration, and detects a change in acceleration and a magnitude of acceleration based on a capacitance change in the detection unit 34. Can do.
As shown in FIG. 6, the metal 7 is embedded in the mass portion 31.

  The angular velocity sensor 39 shown in FIG. 7 includes mass portions 40 and 41, excitation portions 42 and 43, support beams, and anchor portions 44 to 47. The excitation units 42 and 43 have a structure in which comb-shaped fixed drive electrodes and comb-shaped movable drive electrodes are alternately arranged.

  In addition, vibration transmission beams 48 to 51 for transmitting vibrations from the excitation units 42 and 43 to the mass units 40 and 41 are provided.

  The angular velocity sensor 39 also has a laminated structure of the first substrate 1, the second substrate 2, and the insulating layer 3, as shown in FIG. The anchor portions 44 to 47 and the anchor portion 52 that supports the fixed drive electrode are fixedly supported on the second substrate 2 through the insulating layer 3, but the insulating layer 3 is removed in other portions. The space 6 shown in FIG. 1 is formed.

  For example, in this embodiment, when alternating drive signals having opposite phases are applied to the excitation units 42 and 43, a Coulomb force is generated between the movable drive electrode and the fixed drive electrode of each excitation unit 42 and 43. Acts to exert driving force. And a pair of excitation parts 42 and 43 vibrate with an antiphase in a X-axis direction with the displacement to the X direction of each support beam.

  And vibration is transmitted to mass parts 40 and 41 via vibration transmission beams 48-51. Thereby, the mass parts 40 and 41 excite with a reverse phase in the Y-axis direction with the displacement of the support beam supporting the mass parts 40 and 41 in the Y direction.

At this time, when the angular velocity Ω is applied to the angular velocity sensor 39 with the X-axis direction as the rotation axis, the mass portions 40 and 41 are displaced in the Z-axis direction due to the Coriolis force. Note that the mass part 40 and the mass part 41 are displaced in opposite phases. As a result, the angular velocity Ω can be detected based on the capacitance change between the electrodes (not shown). Here, a structure in which comb-shaped movable electrodes and comb-shaped fixed electrodes are alternately arranged or a parallel plate structure can be applied to the electrode structure.
As shown in FIG. 7, metal 7 is embedded in the mass portions 40 and 41.

  FIG. 6 and FIG. 7 are one form, and are not limited to these. Further, in FIGS. 1, 6, and 7, one metal 7 is embedded in the mass portion, but a plurality of metals 7 may be embedded. For example, in the case where a fine hole is formed in the mass part in order to remove the insulating layer 3 below the mass part as described above, the fine hole is formed uniformly throughout the mass part and avoiding the metal 7. For this, it is better to bury the metal 7 in a plurality of parts.

Sectional drawing which showed notionally the MEMS sensor in this embodiment, The partial expanded sectional view which expanded and showed the mass part shown in FIG. Sectional drawing of the MEMS sensor which shows a modification, Sectional drawing which shows the manufacturing process of the MEMS sensor of this embodiment, Sectional drawing for demonstrating the manufacturing method of the MEMS sensor of FIG. The top view of the acceleration sensor in this embodiment, The top view of the angular velocity sensor in this embodiment,

Explanation of symbols

DESCRIPTION OF SYMBOLS 1 1st board | substrate 1b Through-hole 2 2nd board | substrate 3 Insulating layer 4, 31, 40, 41 Mass part 5, 32, 44-47, 52 Anchor part 6, 11 Space 7 Metal 8 1st silicon layer 8a Bottom 9 Second silicon layer (lid)
DESCRIPTION OF SYMBOLS 10 Depression area | region 20 SOI substrate 21 Resist pattern 30 Acceleration sensor 34 Detection part 35 Movable electrode 36 Fixed electrode 39 Angular velocity sensors 42 and 43 Excitation part

Claims (6)

The first substrate has a first substrate and a second substrate disposed to face each other in the height direction, and the first substrate has a mass portion and an anchor portion connected to the mass portion, The anchor portion is fixedly supported on the second substrate side,
A space is provided between the mass portion and the second substrate, and a metal having a specific gravity larger than that of the first substrate is embedded in the mass portion. Sensor.   The first substrate is provided with a detection unit including a movable electrode and a fixed electrode for detecting a displacement amount of the mass unit, and the movable electrode is integrated with the mass unit in the space. The MEMS sensor according to claim 1, wherein the fixed electrode is formed separately from the movable electrode, and an anchor portion connected to the fixed electrode is fixedly supported by the second substrate. A method for manufacturing a MEMS sensor, comprising the following steps.
(A) forming a through hole in the first substrate;
(B) forming a bottom at the bottom of the through hole;
(C) a step of embedding a metal having a higher specific gravity than the first substrate in the through hole;
(D) forming a lid over the first substrate and the metal;
(E) etching the first substrate in a state where the second substrate is bonded to the lower surface via an insulating layer to form a mass portion from the first substrate and an anchor portion connected to the mass portion; At this time, a step of forming the mass part in which the metal is embedded,
(F) A step of leaving the insulating layer between the anchor portion and the second substrate and removing the insulating layer located between the mass portion and the second substrate to form a space. Instead of the step (A) and the step (B),
(G) forming a bottomed recess in the first substrate;
The manufacturing method of the MEMS sensor of Claim 3 which performs.   The flattening process is performed on the metal from the first substrate before performing the step (D), and then the lid portion in the step (D) is formed. The manufacturing method of the MEMS sensor of description. Instead of the step (E) and the step (F),
(H) etching the first substrate in a state in which a second substrate having a depressed region is bonded to a region facing the first substrate to be a mass unit, the mass unit from the first substrate, and Forming an anchor portion to be connected to the mass portion, and forming the mass portion embedded with the metal in the space between the second substrate and the depression region;
A method for manufacturing a MEMS sensor according to claim 3, comprising:

Images