JP2002071708A - Semiconductor dynamic quantity sensor - Google Patents

Abstract

PROBLEM TO BE SOLVED: To reduce errors of a sensor output by minimizing a difference in the parasitic capacitance generated between fixed electrodes in one pair in a semiconductor dynamic quantity sensor, which enables detection of a dynamic quantity generated in directions based on a change in the capacitance between a mobile electrode and pairs of fixed electrodes, by forming the mobile electrode displaceable in X and Y directions on a semiconductor board and the pairs of fixed electrodes respectively facing each other with a detection interval between the pairs thereof and the mobile electrode in the directions. SOLUTION: Leaders 41, 51, 61 and 71 electrically connected to respective fixed electrodes separately and pads 42, 52, 62 and 72 for external connection electrically connected to the leaders are formed on the outer circumference parts of the individual first and second fixed electrodes 30 and 40 and 50 and 60, as viewed from the mobile electrode 30 out of a semiconductor substrate 12. The lengths of the leaders at the respective fixed electrodes can be made equal to each other as much as possible.

Description

DETAILED DESCRIPTION OF THE INVENTION

[0001]

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to a semiconductor substrate having a movable electrode which can be displaced in two axial directions and a fixed electrode which is opposed to the movable electrode at a detection interval and which is generated in the two axial directions. The present invention relates to a semiconductor physical quantity sensor capable of detecting a physical quantity based on a capacitance change between both electrodes.

[0002]

2. Description of the Related Art In general, a movable electrode which can be displaced in a predetermined uniaxial direction on a semiconductor substrate and a fixed electrode which is opposed to the movable electrode with a detection interval are generally used to detect a mechanical quantity generated in one axial direction. There is a semiconductor dynamic quantity sensor that forms an electrode and detects an applied dynamic quantity based on a change in capacitance of both electrodes when a dynamic quantity such as acceleration is applied in the one axis direction.

However, in recent years, there has been an increasing need for a semiconductor dynamic quantity sensor (hereinafter, referred to as a two-axis sensor) capable of detecting dynamic quantities in two axial directions. For example, in an automobile airbag control system, it is necessary to detect accelerations (impacts) from two directions, a frontal collision and a side collision. Here, when an acceleration sensor capable of detecting only one axis direction is used for the system, at least two acceleration sensors are required.

[0004] Then, there arises a problem that the physical size of the whole system becomes large, for example, a space for disposing the sensors becomes large, and a control circuit must be provided for each sensor. In that regard, the use of the two-axis sensor is very advantageous in solving such a problem.

[0005]

Therefore, in view of the above circumstances, the present inventor studied a prototype of a two-axis sensor. FIG. 5 is a diagram showing a schematic plan configuration of a two-axis sensor that can detect acceleration in two-axis directions prototyped by the present inventors.

In FIG. 5, reference numeral 12 denotes a semiconductor substrate made of a semiconductor such as silicon. By forming a groove in the semiconductor substrate 12, the semiconductor substrate 12
A movable electrode 30 as a movable portion and fixed electrodes 40, 50, 60, and 70 as fixed portions are defined.

The movable electrode 30 is displaceable in a first direction and a second direction orthogonal to each other in a plane parallel to the semiconductor substrate 12 in response to the application of acceleration. In FIG. 5, the movable electrode 30 is elastically supported by the semiconductor substrate 12 via beams 33 and 34, and is movable in the X direction (first direction) in the figure.
When acceleration occurs in the X direction, the beam portion 33 allows displacement in the X direction. When acceleration occurs in the Y direction (second direction) orthogonal to the X direction, the beam portion 34 allows displacement in the Y direction. It has become.

Further, both ends in the X direction and both ends in the Y direction of the movable electrode 30 are formed as comb teeth (movable comb teeth) 36 projecting in a comb shape. In addition, the fixed electrode 40
Reference numerals 70 are arranged on the outer periphery of the movable electrode 30 at both ends in the X direction and the outer periphery at both ends of the movable electrode 30 in the Y direction, respectively, and are fixedly supported by the semiconductor substrate 12. Each of the fixed electrodes 40 to 70 has a comb shape, and is arranged so as to mesh with a gap between the movable comb portions 36 facing each other.

Here, assuming that a pair of fixed electrodes 40 and 50 formed on the outer peripheral portions at both ends of the movable electrode 30 along the X direction are first fixed electrodes, the first fixed electrodes 40 and 50 opposed to each other. A first detection capacitance is formed between the movable electrode 30 and the movable comb portion 36 of the movable electrode 30. The first detection capacitor is composed of one (upper side in the figure) first fixed electrode 40 and movable comb tooth 3
6 and a detection capacitor C1 between the other (lower in the figure) first fixed electrode 50 and the movable comb tooth portion 36.
S2.

[0010] A pair of fixed electrodes 60 and 70 formed on the outer peripheral portions at both ends of the movable electrode 30 along the Y direction are connected to the second electrode.
Of the second fixed electrodes 6 opposed to each other
0, 70 and the movable comb portion 36 of the movable electrode 30
Is formed. The second detection capacitor is composed of one (the left side in the figure) second fixed electrode 60 and movable comb-tooth portion 36.
Between the detection capacitor CS3 and the other (the right-hand side in the figure) the second
Capacitance CS between the fixed electrode 70 and the movable comb tooth portion 36 of FIG.
4 These detection capacitors CS1 to CS4 are:
This is indicated by a capacitor symbol in FIG.

Further, each of the fixed electrodes 40, 50, 60, 70
Lead portions 41, 51, 61, 71 and pad portions 42, 52, 62, 72 are formed corresponding to the fixed electrodes 4
Reference numerals 0 to 70 denote pad portions 42 through the lead portions 41 to 71.
To 72 are electrically connected. In addition, the movable electrode 30
A lead portion 37 and a pad portion 38 are formed correspondingly, and the movable electrode 30 is connected to the pad portion 3 via the lead portion 37.
8 is electrically connected.

Further, a pad portion 80 which is electrically independent of these electrodes is formed in the peripheral portion of the semiconductor substrate 12, and this pad portion 80 fixes the potential of the peripheral portion of the semiconductor substrate 12. It is configured as a substrate potential fixing pad section for storing. Although not shown, the pad portions 38, 42 to 72, and 80 are connected to an external circuit or a wiring member by wire bonding or the like, and are electrically connected.

The movable electrode 3 is moved in response to the application of acceleration.
When 0 is displaced in the X direction (first direction), the applied acceleration is detected based on the change in the first detection capacitors CS1 and CS2, and the movable electrode 30 becomes Y in response to the applied acceleration.
When displaced in the direction (second direction), the second detection capacitance C
The applied acceleration is detected based on the changes in S3 and CS4.

However, in the above-mentioned prototype, the general arrangement of the pads in the conventional semiconductor dynamic quantity sensor, that is, the arrangement in which the pads are collected on one side of the semiconductor substrate 12, is as follows. The problem of a large detection error occurs. FIG. 6 schematically shows an equivalent circuit of the acceleration sensor shown in FIG. 5 described above, and a problem of a detection error will be described by taking acceleration detection in the X direction as an example.

Here, since the movable electrode 30 and the pair of first fixed electrodes 40 and 50 and the peripheral portion of the semiconductor substrate 12 are electrically partitioned via the groove, as shown in FIG.
In addition to the first detection capacitors CS1 and CS2, each electrode 3
Parasitic capacitances CP <b> 1 to CP <b> 3 exist between 0, 40, 50 and the periphery of semiconductor substrate 12.

The parasitic capacitance CP1 is a parasitic capacitance formed between one of the first fixed electrodes 40 and the periphery of the semiconductor substrate 12, and the parasitic capacitance CP2 is a parasitic capacitance formed between the other first fixed electrode 50 and the semiconductor substrate 12. Parasitic capacitance formed between the
The parasitic capacitance CP3 is a parasitic capacitance formed between the movable electrode 30 and a peripheral portion of the semiconductor substrate 12.

In such a circuit configuration, between the pair of first fixed electrodes 40 and 50 (between the pad portions 42 and 52).
, The acceleration in the X direction is output as a voltage value V0 as shown in the following equation 1.

[0018]

V0 = ｛(CS1-CS2) + (CP1-CP
2) · CP3｝ · V / Cf Here, Cf in Equation 1 is a feedback capacitance of a switched capacitor circuit provided on the detection circuit side for converting a detected change in capacitance into a voltage value. . in this way,
The acceleration is detected based on the difference between the detected capacitances (CS1-CS2). To reduce the output error (offset) of the sensor, an error due to the parasitic capacitance (CP1-CP2) is required.
2) It is necessary to reduce CP3.

However, in the above-mentioned prototype, the pad portions are arranged on one side of the semiconductor substrate 12 so that the lengths of the lead portions 41 and 51 of the pair of first fixed electrodes 40 and 50 are different from each other. Are different. Since the parasitic capacitances CP1 and CP2 of the fixed electrodes 40 and 50 are capacitances including the respective lead portions 41 and 51, it is difficult to equalize the parasitic capacitances CP1 and CP2.

That is, the pair of first fixed electrodes 40 and 50
, The peripheral portion of the semiconductor substrate 12 and the respective lead portions 4
By making the parasitic capacitances generated between the first fixed electrodes 40 and 5 equal to each other, the pair of first fixed electrodes 40 and 5
It is necessary to make the parasitic capacitances CP1 and CP2 between zero equal. This means that a pair of second fixed electrodes 6 for detecting the acceleration in the Y direction in the prototype are used.
The same applies to 0 and 70.

In the above prototype, the semiconductor substrate 1
In the case where a supporting substrate that supports the second substrate 2 is provided, the same problem as described above occurs with respect to the parasitic capacitance formed between the pair of fixed electrodes and the supporting substrate. In any case, when a parasitic capacitance is formed between these lead portions and the other electrically partitioned portions, a difference in parasitic capacitance occurs due to a difference in length of the lead portions of the pair of fixed electrodes, An error in the sensor output also becomes a problem.

In view of the above problems, the present invention forms a movable electrode that can be displaced in two axial directions on a semiconductor substrate and forms a pair of fixed electrodes opposed to the movable electrode in each direction with a detection interval. In a semiconductor dynamic quantity sensor formed and capable of detecting a dynamic quantity generated in the two axial directions based on a capacitance change between a movable electrode and each pair of fixed electrodes, a parasitic quantity generated between the pair of fixed electrodes. An object of the present invention is to reduce a difference in capacitance and reduce an error in a sensor output.

[0023]

According to the present invention, there is provided a two-axis detection type semiconductor dynamic quantity sensor capable of detecting in two orthogonal directions orthogonal to each other as shown in the above-mentioned trial product. Focusing on making the length of each lead-out portion of a pair of fixed electrodes approximately the same for each of them, and making the parasitic capacitance generated between the pair of fixed electrodes approximately the same, It was done.

That is, according to the first aspect of the present invention, in the two-axis detection type semiconductor dynamic quantity sensor, the semiconductor substrate (12)
Of the first and second fixed electrodes (30, 40, 50, 60) as viewed from the movable electrode (30), the lead portions (41, 41) electrically connected to each of the fixed electrodes.
51, 61, 71) and an external connection pad portion (42, 52, 62, 7) electrically connected to the lead portion.
2) is formed.

According to the present invention, the fixed electrodes are respectively arranged in four directions around the movable electrode along the first direction and the second direction, and the fixed electrodes are further disposed outside the individual fixed electrodes. The shape is such that the drawer and the pad are arranged. Therefore, the fixed electrode and the pad portion can be arranged close to each other for each fixed electrode, and the distance between the fixed electrode and the pad portion in each fixed electrode can be easily made substantially equal to each other.

That is, since the length of the lead-out portion in each fixed electrode can be made as short as possible, and the length of each lead-out portion can be made almost the same as each other, the parasitic capacitance generated in each lead-out portion can be reduced. The values can be made small and close to each other.

As described above, according to the present invention, it is possible to reduce the difference between the parasitic capacitances generated in the respective lead portions of the pair of fixed electrodes. As a result, the parasitic capacitance generated between the pair of fixed electrodes can be reduced. The difference can be reduced, and the error of the sensor output can be reduced.

According to a second aspect of the present invention, there is provided a specific structure of the semiconductor dynamic quantity sensor according to the first aspect, wherein the semiconductor substrate (12) has a rectangular opening which is opened on one side. The movable electrode (30) is positioned at a substantially central portion of the opening, and each of the first and second fixed electrodes is supported by one side of the supporting substrate (20) having the (21). And placed on each side of the first and second
The lead portions (41-71) and the pad portions (42-72) corresponding to each of the fixed electrodes (40-70) are also arranged separately on each side of the opening.

In an actual sensor, in addition to the parasitic capacitance difference, a difference in the wiring resistance of the lead portion also causes an error in the sensor output. In this regard, according to the third aspect of the present invention, the first and second fixed electrodes (40 to 70), the lead portions (41 to 71) corresponding to the first and second fixed electrodes, respectively, and The pad portions (42 to 72) are symmetrically arranged in four directions around the movable electrode (30). Therefore, the wiring resistance of each lead-out portion can be made substantially the same, and the error of the sensor output can be suppressed at a higher level.

Further, according to the fourth aspect of the present invention, the lead portions (41, 51) and the pad portions (42, 52) corresponding to each of the first fixed electrodes (40, 50) are formed in the first fixed electrode (40, 50).
And the second fixed electrode (60,
70), the drawer portions (61, 71) and the pad portions (62, 72) corresponding to the respective components (70) are arranged along the second direction (Y).

According to this configuration, in the symmetrical arrangement according to the third aspect, the lead portion and the pad portion corresponding to each of the first and second fixed electrodes are arranged in the first direction or the second direction.
, The distance between each fixed electrode and the pad portion can be made substantially the shortest distance.
Therefore, the wiring resistance of each lead-out part and the parasitic capacitance generated in each lead-out part can be suppressed to the minimum level, which is preferable.

According to the fifth aspect of the present invention, the semiconductor substrate is formed in a rectangular plate shape, and pad portions (3) for wire bonding are provided at least at two or more corners of the semiconductor substrate.
8, 80), one of the at least two pad portions is a movable electrode pad portion (38) electrically connected to the movable electrode (30), and the other is to fix the potential of the semiconductor substrate. For fixing the substrate potential (80).

In this kind of semiconductor dynamic quantity sensor,
A pad for the movable electrode and a pad for fixing the substrate potential are required. In such a case, stress is generated in the semiconductor substrate by these formed pad portions. Therefore, when these pad portions are formed on the semiconductor substrate, it is preferable that the balance of the stress be made uniform.

In this regard, according to the fifth aspect of the present invention, the movable electrode and the pad for fixing the substrate potential are formed in at least two of the four corners of the rectangular plate-shaped semiconductor substrate. The balance of the stress generated in the semiconductor substrate by the pad portion can be made relatively uniform, which is preferable.

According to the sixth aspect of the present invention, the movable electrode (30) includes a rectangular first weight (31) and a second weight protruding from four corners of the first weight. And a weight portion (32). According to this, since the second weight is further provided at the four corners of the first weight, the weight of the entire movable electrode can be effectively increased.

The reference numerals in parentheses of the above means are examples showing the correspondence with specific means described in the embodiments described later.

[0037]

BRIEF DESCRIPTION OF THE DRAWINGS FIG. 1 is a block diagram showing a first embodiment of the present invention. In the present embodiment, the present invention is applied to a differential capacitance type semiconductor acceleration sensor as a two-axis detection type semiconductor dynamic quantity sensor. FIG. 1 shows a plane configuration of the semiconductor acceleration sensor 100, and FIG.
4 shows a schematic cross-sectional structure along the line A.

The semiconductor acceleration sensor (hereinafter simply referred to as a sensor) 100 is formed by subjecting a semiconductor substrate to a known micromachining process. In this example, the sensor 10
0 is the first semiconductor substrate as shown in FIG.
It is formed on a rectangular plate-shaped SOI (silicon-on-insulator) substrate 10 having an oxide film 13 as an insulating layer between a silicon substrate 11 and a second silicon substrate 12 as a second semiconductor substrate.

Here, the second silicon substrate 12 is the semiconductor substrate referred to in the present invention, and the first silicon substrate 11 and the oxide film 13 are formed on the supporting substrate 2 supporting the second silicon substrate 12.
It is configured as 0. The support substrate 20 has an opening 21 formed on one side (the oxide film 13 side).

A groove is formed in the second silicon substrate 12 so that the movable electrode 30 as a movable portion and the fixed electrodes 40 and 50 as fixed portions electrically partitioned from the movable electrode 30 are formed. , 60, 70 are formed. Here, the opening 2
In this example, 1 is configured as a rectangular hole that penetrates the region of the support substrate 20 corresponding to the beam structures 30 to 70 in the thickness direction.

The movable electrode 30 has a first direction X and a second direction Y orthogonal to each other in a plane parallel to the second silicon substrate (semiconductor substrate) 12 in response to the application of acceleration (each direction is 1 (shown by arrows X and Y in FIG. 1).
In the present example, the movable electrode 30 is located substantially at the center of the rectangular opening 21, and has a rectangular first weight 31 and a second weight 31 protruding from four corners of the first weight 31. And a weight portion 32.

The movable electrode 30 is connected to the second weight 32
, Is elastically supported by the edge of the opening 21 in the support substrate 20. In the present example, at the edges of the four corners of the opening 21, each of the second weight portions 32 is elastically deformable in the first direction X and the first beam portion 33 and elastically deformable in the second direction Y. The anchor portions 35a, 35 are provided via the second beam portions 34.
b, 35c, 35d.

These anchor portions 35a to 35d
As shown in the figure, at the edge of the opening 21, it is supported and fixed to the first silicon substrate 11 via the oxide film 13. As a result, the movable electrode 30 and each of the beams 33 and 34 face the opening 21.

Each of the beam portions 33 and 34 is formed by connecting a plurality of beams in a folded shape, and the connected beams are elastically deformed by bending in the longitudinal direction of the beams. . The movable electrode 30 is connected to each of the beams 33 and 34.
Is displaceable as follows.

That is, the movable electrode 30 is displaced in the first direction X when receiving the acceleration including the component in the first direction X, and restores the original state in accordance with the disappearance of the acceleration.
On the other hand, the movable electrode 30 is displaced in the second direction Y when receiving an acceleration including a component in the second direction Y, and restores the original state according to the disappearance of the acceleration.

Here, due to the structure of each beam portion 33, 34, the movable electrode 30 is suppressed from being displaced in both directions X and Y (that is, obliquely). That is,
The movable electrode 30 is substantially displaced only in the first direction X or only in the second direction Y according to the magnitude of the acceleration component.

The first direction X in the movable electrode 30
Are formed as a comb-tooth portion (hereinafter referred to as a movable comb-tooth portion) 36 protruding in a comb-tooth shape. In the present example, the second weight 3 in the first weight 31
On each side between the two, four protruding rod-shaped portions are formed, whereby the movable comb-shaped portion 36 is formed.

The fixed electrodes 40 to 70 are formed on the outer peripheral portions of both ends of the movable electrode 30 along the first direction X in the second silicon substrate 12, respectively. A pair of first capacitors forming the detection capacitors CS1 and CS2
Are formed on the outer periphery of both ends of the movable electrode 30 along the second direction Y, respectively.
A pair of second fixed electrodes 60 and 70 that form second detection capacitors CS3 and CS4 between the first and second detection capacitors CS3 and CS4.

The first and second fixed electrodes 40, 5
In this example, each of 0, 60, and 70 has a comb shape, and is arranged on each side of the opening 21. Each of the fixed electrodes 40 to 70 is supported and fixed in a cantilever manner to the first silicon substrate 11 via the oxide film 13 at the edge of the opening 21, similarly to the anchor portion 35 a shell 5 d. 21.

The fixed electrodes 40 to 70 are arranged so as to mesh with the gaps of the movable comb teeth 36 facing each other, and the movable electrode 3 is placed between the side faces of the meshed comb teeth.
0 and the detection capacitors CS1 to C between the fixed electrodes 40 to 70.
S4 will be formed. Each of these detection capacitors CS1
ＣＳCS4 is indicated by a capacitor symbol in FIG.

Here, the first fixed electrodes 4 opposed to each other
0, 50 and the movable comb tooth 36 of the movable electrode 30.
Of the detection capacitors CS1 and CS2 are formed. This first detection capacitor is connected to one (upper side in FIG. 1) first fixed electrode 40.
And the movable capacitor 36 and a detection capacitor CS2 between the other (lower in FIG. 1) first fixed electrode 50 and the movable comb 36.

Further, the second fixed electrodes 6 opposed to each other
0, 70 and the movable comb portion 36 of the movable electrode 30
Of the detection capacitors CS3 and CS4 are formed. The second detection capacitor is connected to one (left side in FIG. 1) second fixed electrode 60.
And a movable capacitor 36 and a detection capacitor CS3 between the second fixed electrode 70 and the movable comb 36 (the right side in FIG. 1).

Next, the configuration around the pad portion of the sensor 100, which is the main feature of the present embodiment, will be described.

When viewed from the movable electrode 30 side of the second silicon substrate (semiconductor substrate) 12, the individual fixed electrodes 40 to 70 are formed on the outer peripheral portions of the individual first and second fixed electrodes 40, 50, 60, and 70. Each of the lead portions 41, 51, 61, 71 electrically connected to each of the fixed electrodes 40 to 70 is formed, and further, for each of the lead portions 41 to 71,
Pad portions 42, 52, 62, 72 for external connection electrically connected to the respective lead portions 41 to 71 are formed.

That is, the fixed electrodes 40 to 70 are electrically connected to the pad portions 42 to 72 through the lead portions 41 to 71, respectively. In this example, the lead portions 41 to 7 corresponding to the first and second fixed electrodes 40 to 70, respectively.
Similarly to the fixed electrode 40 shell 70, the 1 and the pad portions 42 to 72 are also allocated to each side of the opening 21 described above.

Thus, the first and second fixed electrodes 40 to
70, leading portions (leading portions for fixed electrodes) 41 to 41 corresponding to the first and second fixed electrodes 40 to 70, respectively.
71 and pad portions (fixed electrode pad portions) 42 to 72
Are arranged symmetrically about the movable electrode 30.

Further, in this example, the first fixed electrodes 40, 50
Are arranged along the first direction X, and correspond to the second fixed electrodes 60 and 70, respectively. The fixed electrode lead portions 61 and 71 and the fixed electrode pad portions 62 and 72 are arranged along the second direction Y.

Further, a lead-out portion (lead-out portion for movable electrode) 37 and a pad portion (pad portion for movable electrode) 38 are also formed for the movable electrode 30, and the movable electrode 30 is provided with the lead-out portion 37 for movable electrode. Through the movable electrode pad 38
Is electrically connected to This movable electrode pad 3
Reference numeral 8 in this example is electrically connected to the movable electrode 30 via a lead portion 37 formed continuously from one anchor portion 35b.

Further, a pad portion 80 which is electrically independent from the movable and fixed electrodes 30 to 70 is formed in the peripheral portion of the second silicon substrate 12. Each electrode 30 to 70 of the silicon substrate 12
It is configured as a substrate potential fixing pad portion for fixing the potential of the other portion, that is, the peripheral portion of the second silicon substrate 12.

In this embodiment, three substrate potential fixing pad portions 80 are provided, and at least one of these three pad portions 80 is actually used. . The four pad portions of the movable electrode pad portion 38 and the substrate potential fixing pad portion 80 are symmetrically arranged at the four corners of the second silicon substrate (semiconductor substrate) 12 having a rectangular plate shape.

It is needless to say that each of the pad portions and the lead portions is electrically connected to the corresponding electrode.
As shown in FIG. 1, electricity is supplied through a groove (air isolation, which is indicated by hatching for convenience in FIG. 1 and the oxide film 13 is visible in FIG. 1) formed in the second silicon substrate 12. Electrically insulated. Each of the lead portions and the pad portions is supported on the support substrate 20.

The pad portions 38, 42 to 7
2, 80 are made of, for example, aluminum;
Although not shown, each of the pad portions is connected to an external circuit or a wiring member by wire bonding or the like, and is electrically connected.

The above is the configuration around the pad section in this embodiment. Next, an example of a method for manufacturing the sensor 100 according to the present embodiment based on the above configuration will be described.

FIG. 3 is a schematic sectional view corresponding to FIG. 2 showing a manufacturing process of the sensor 100. First, as shown in FIG. 3A, the first and second silicon substrates 11 made of a single-crystal silicon wafer whose surface orientation is set to (100), for example, via the silicon oxide film 13 described above, An SOI substrate 10 to which the substrates 12 are bonded is prepared (showing a state where the SOI substrate 10 is not processed).

Next, an electrode pad forming step shown in FIG. 3B is performed. In this step, aluminum is deposited on the entire surface of the second silicon substrate 12 so as to have a thickness of, for example, about 1 μm, and the aluminum film is patterned by using a photolithography technique and an etching technique. Pads 38, 42-72, 8
0 (not shown in FIG. 3 except for the movable electrode pad section 38 and one substrate potential fixing pad section 80).

From this state, the dimension adjusting step shown in FIG. In this step, the first silicon substrate 11
By performing cutting / polishing on the surface (the surface opposite to the oxide film 13), the thickness of the substrate 11 is adjusted to, for example, 300 μm, and the processed surface is mirror-finished.

The reason why the thickness of the first silicon substrate 11 is reduced to 300 μm as described below is as follows.
This is because when the opening 21 is formed by anisotropic etching, the etching depth is reduced, thereby preventing an increase in chip design dimensions due to the anisotropic etching.

Next, a mask forming step shown in FIG. In this step, a silicon nitride film is deposited on the entire surface (mirror-finished surface) of the first silicon substrate 11 to a thickness of about 0.5 μm by, for example, a plasma CVD method, and then the silicon nitride film is formed. Is patterned using a photolithography technique and an etching technique to form a mask M1 when the opening 21 is formed by etching.

Thereafter, the trench (trench) shown in FIG.
Perform a forming step. In this step, a resist (not shown) having dry etching resistance is formed as a mask on the second silicon substrate 12 and the pad portions 38, 42 to 72, and 80, and anisotropic dry etching is performed by a dry etching apparatus. Thereby, a trench T1 reaching the silicon oxide film 13 is formed in the second silicon substrate 12.

At this time, the second silicon substrate 12 is
Movable electrode 30 and fixed electrodes 40 to 70 as shown in FIG.
A pattern of a beam structure composed of the fixed electrodes 40 to 70 (not shown) and a groove (air isolation pattern) for electrically partitioning each pad portion and lead-out portion are formed.

From this state, the first etching step shown in FIG. In this first etching step, the first silicon substrate 12 is exposed to the surface (silicon oxide film 1) using a mask M1 and using, for example, a KOH aqueous solution.
Selective etching is performed from the side opposite to (3).

For example, if the thickness of the first silicon substrate 11 is 1
The etching time is controlled with the aim of remaining about 0 μm. Although not specifically illustrated, the surface of the SOI substrate 10 is covered with a resist before the execution of the first etching step. It will be removed after finishing.

Next, a second etching step shown in FIG. In this second etching step, the first
By performing dry etching using, for example, a plasma etching apparatus from the front side of the silicon substrate 11, the silicon oxide film 13 is formed in the first etching step.
First silicon substrate 11 having a thickness of about several μm left between
Is removed, so that the back surface (lower surface) of the silicon oxide film 13 is exposed. Note that, with such dry etching, the mask M1 is also removed at the same time.

Next, a third etching step (release step) is performed. In this third etching step, HF
By performing etching with a system etchant,
The silicon oxide film 13 is removed. According to the execution of the third etching step, the opening 21 is formed, and the movable electrode 30 and the fixed electrodes 40 to 70 are released. Thus, the sensor 100 shown in FIGS. 1 and 2 is completed.

The operation of the sensor 100 will be described. As described above, in the sensor 100, when the movable electrode 30 is displaced in the first direction X in response to the application of the acceleration, the applied acceleration is detected based on the change in the first detection capacitors CS1 and CS2. When the movable electrode 30 is displaced in the second direction Y in response to the application of the acceleration, the applied acceleration is detected based on a change in the second detection capacitors CS3 and CS4.

For example, assuming that the acceleration is applied in the right direction (+ X direction) in FIG. 1 along the first direction X, the movable electrode 30 moves in the left direction (−in FIG. 1) along the first direction X.
X direction). At this time, one side (upper side in FIG. 1)
The distance between the first fixed electrode 40 and the movable comb portion 36 is increased, and the detection capacitance CS1 is reduced. The other (lower in FIG. 1) first fixed electrode 50 and the movable comb portion 36 are connected to each other. The interval decreases and the detection capacitance CS2 increases.

Here, the equivalent circuit for detecting acceleration in the present sensor 100 is the same as that in FIG.
2-CS1) is converted to C-V by an external detection circuit (not shown).
By performing the conversion, the output is obtained as the voltage value V0, as shown in the above equation (1). At this time, the change in the second detection capacitors CS3 and CS4 is very small and can be ignored. Here, a mathematical expression similar to the above mathematical expression 1 is shown again as the following mathematical expression 2.

[0078]

V0 = ｛(CS1-CS2) + (CP1-CP
2) · CP3｝ · V / Cf Here, V is a voltage applied between the pair of first fixed electrodes 40 and 50 (between the pad portions 42 and 52), as in the above-described formula 1. Is the feedback capacitance of the switched capacitor circuit provided on the detection circuit side. Also, CP1
Is a parasitic capacitance formed between one of the first fixed electrodes 40 and the peripheral portion of the second silicon substrate (semiconductor substrate) 12;
CP1 is a parasitic capacitance formed between the other first fixed electrode 50 and the peripheral portion of the second silicon substrate 12.

On the other hand, for example, assuming that acceleration is applied in the upward direction (+ Y direction) in FIG. 1 along the second direction Y, the movable electrode 30 moves downward in FIG. 1 along the second direction Y. (−Y direction). At this time, one side (FIG. 1)
The distance between the second fixed electrode 60 (middle, left) and the movable comb portion 36 is increased, and the detection capacitance CS3 is reduced.
The distance between the second fixed electrode 70 (middle, right) and the movable comb portion 36 is reduced, and the detection capacitance CS4 is increased.

FIG. 6 also shows an equivalent circuit for detecting the acceleration in the second direction Y. Then, the capacitance difference (CS4
−CS3) is similarly subjected to CV conversion, whereby an output is obtained as a voltage value V0 as shown in the following Expression 3.

[0081]

V0 = ｛(CS4-CS3) + (CP4-CP
5) · CP3｝ · V / Cf Here, V and Cf are the same as the above formulas 1 and 2,
Further, CP 4 is a parasitic capacitance formed between one second fixed electrode 60 and the peripheral portion of the second silicon substrate 12, C 4
P5 is a parasitic capacitance formed between the other second fixed electrode 70 and the peripheral portion of the second silicon substrate 12.

As can be seen from Equations 2 and 3,
In order to reduce the error of the sensor output, each of the first fixed electrodes 40 and 50 and the second fixed electrodes 60 and 70 is formed between the fixed electrode and the peripheral portion of the second silicon substrate 12. Difference between a pair of parasitic capacitances (CP1-CP
2) It is necessary to make (CP4-CP5) as similar as possible.

Here, according to the present embodiment, the fixed electrodes 30 are respectively arranged in four directions around the movable electrode 30 along the first direction X and the second direction Y. The fixed electrode lead portions 41, 51, 61, 71 and the fixed electrode pad portions 42, 52, 62, 72 are further arranged outside the reference numerals 40, 50, 60, 70.

For this reason, the fixed electrodes 40 to 70 are fixed to the fixed electrode 40 shell 70 in comparison with the arrangement in which the pads are collected on one side of the semiconductor substrate 12 as in the prototype shown in FIG. The electrode pad portions 42 to 72 can be arranged close to each other, and the distance between the fixed electrodes 40 to 70 and the fixed electrode pad portions 42 to 72 in each of the fixed electrodes 40 to 70 can be easily made equal to each other. Can be.

As a result, the length of the lead portions 41 to 71 in each of the fixed electrodes 40 to 70 can be made as short as possible.
Since the lengths of the lead portions 41 to 71 can be made substantially equal to each other as much as possible, the parasitic capacitance generated in each of the lead portions 41 to 71 can be made small and close to each other.

As described above, according to the present embodiment, the difference between the parasitic capacitances generated at the respective lead-out portions of the pair of fixed electrodes can be reduced, and as a result, the first and second fixed electrodes 40 are formed. 70 to 70, the difference in parasitic capacitance between the pair of fixed electrodes (CP1-CP2), (CP4-CP4).
CP5) can be reduced, and errors in the sensor output can be reduced.

In this embodiment, the portions of the second silicon substrate 12 where the fixed electrode lead portions 41 to 71 are formed are supported by the first silicon substrate 11 via the oxide film 13. Therefore, each drawer part 41-
A parasitic capacitance is also formed between the first silicon substrate 11 and the first silicon substrate 11, and this parasitic capacitance also affects the sensor output.

However, according to the present embodiment, the length of the pair of lead portions 41, 51 of the first fixed electrodes 40, 50 is made as short as possible and substantially equal to each other, and the length of the second fixed electrodes 60, 70 is reduced. The length of the pair of drawers 61 and 71 is as short as possible and substantially equal to each other. Therefore, the parasitic capacitance difference between the pair of fixed electrodes can be reduced with respect to the parasitic capacitance between the first silicon substrate 11 and the influence on the error of the sensor output is small.

In the sensor 100, in addition to the parasitic capacitance difference, a difference in wiring resistance between the lead portions 41 to 71 also causes a sensor output error. According to the present embodiment, as described above, each of the first and second fixed electrodes 4 as viewed from the movable electrode 30 of the second silicon substrate 12.
On the outer periphery of the fixed electrodes 0 to 70, the fixed electrode lead portions 41 to 71 and the fixed electrode pad portions 42 to 72 are provided for each fixed electrode.
Are formed, the arrangement of these lead portions and pad portions need not be symmetrical about the movable electrode 30.

However, in this case, since the lengths of the fixed electrode lead portions 41 to 71 are slightly different, each lead portion 4
A difference occurs in the wiring resistances 1 to 71. In that regard, in the above-described sensor 100, the first and the second
Lead-out portions 41-71 and pad portions 4 corresponding to the first and second fixed electrodes, respectively.
2 to 72 are symmetrically arranged in four directions around the movable electrode 30 (hereinafter, referred to as a symmetrically arranged configuration such as a pad portion).

According to the symmetric arrangement of the pads and the like,
The lengths of the fixed electrode lead portions 41 to 71 can be made closer to each other. Therefore, the wiring resistance of each of the fixed electrode lead portions 41 to 71 can be made substantially the same, and the error of the sensor output can be suppressed at a higher level.

Further, as a more preferable form in the above-mentioned symmetrical arrangement of the pad portions and the like, in the sensor 100, the lead portions 41 to 71 and the pad portions corresponding to the first and second fixed electrodes 40 to 70 respectively. 42-72
Are arranged along the first direction X or the second direction Y.

According to this, the individual fixed electrodes 40 to 70
And the distance between the fixed electrode pad portions 42 to 72 can be made substantially the shortest distance. For this reason, the wiring resistance of each of the fixed electrode lead portions 41 to 71 and the parasitic capacitance generated in each of the fixed electrode lead portions 41 to 71 can be suppressed to the minimum level, which is preferable.

Further, according to the sensor 100, a total of four pad portions, one movable electrode pad portion 38 and three substrate potential fixing pad portions 80, are connected to the second silicon substrate 12 having a rectangular plate shape. Since these pads are symmetrically arranged at the four corners, the balance of the stress generated on the second silicon substrate 12 by these pad portions (aluminum or the like) can be made uniform.

In this embodiment, the case where there are four pads for the movable electrode and for fixing the substrate potential is shown. However, in addition to this embodiment, one pad for the movable electrode and the pad for fixing the substrate potential are provided. At least two pad parts are required, such as one pad part. In that case, pad portions for wire bonding are formed at least at two or more corners of the rectangular second silicon substrate 12, one of these at least two pad portions is a movable electrode pad portion, and the other is a substrate potential. It is preferable to configure as a fixing pad portion.

If at least two pad portions, which are configured as movable electrodes and pad portions for fixing the substrate potential, are formed at the four corners of the second silicon substrate 12 having a rectangular plate shape, the second silicon portion is formed by these pad portions. The balance of the stress generated in the substrate 12 can be relatively uniform, which is preferable.

Further, according to the present sensor 100, the movable electrode 30 is made up of the rectangular first weight 31 and the first weight 3
And a second weight portion 32 protruding from the four corners of the first weight portion 31. In addition to the first weight portion 31, the second weight portion 3
2, the weight of the entire movable electrode 30 can be effectively increased.

Further, in FIG. 1, each of the lead portions 41 to 71 for fixed electrodes has a convex shape in which a portion near the fixed electrodes 40 to 70 is wide and a portion near the pad portions 42 to 72 is narrow. Although it has a shape, the shape of each fixed electrode lead-out portion 41-71 is not limited to this. For example, as shown in FIG. 4, each of the lead portions 41 to 71 for the fixed electrodes is connected to the pad portions 42 from the fixed electrodes 40 to 70.
The shape having the same width from the side to the 72th side may be used.

(Other Embodiments) The sensor 100
In the above, the opening 21 formed in the supporting substrate is rectangular, but the opening 21 may not be rectangular but may have another geometric shape. The opening 21 does not have to penetrate the support substrate 20 in the thickness direction. For example, the oxide film 13 is removed by sacrifice layer etching or the like, and the first silicon substrate 11 is left to form a recess. The recess may be configured as an opening.

Further, the above-mentioned sensor 100 is
Although the semiconductor substrate 12 is supported above, the support substrate 20 may not be provided. For example, a single-layer silicon substrate may be used as the semiconductor substrate of the present invention, and each electrode, pad portion, and the like may be formed on the silicon substrate.

Further, the present invention can be applied to, for example, an angular velocity sensor and the like in addition to the acceleration sensor. For example, in the sensor 100, the movable electrode 30 is driven and vibrated in the second direction Y by using the movable comb portion 36 and the second fixed electrodes 50 and 60. When an angular velocity is applied around an axis orthogonal to the X and Y directions, a Coriolis force is generated in the movable electrode 30 in the first direction X.

The Coriolis force in the first direction X is equal to the capacitance between the movable comb portion 36 and the first fixed electrodes 40 and 50 (the first fixed electrode 40, 50).
Can be detected as a change. Based on this point, when the driving vibration direction is the first direction X, the Coriolis force generated in the second direction Y can be detected.

[Brief description of the drawings]

FIG. 1 is a schematic plan view of a semiconductor acceleration sensor according to an embodiment of the present invention.

FIG. 2 is a schematic sectional view taken along the line AA in FIG.

FIG. 3 is a process chart showing a method for manufacturing the sensor shown in FIG. 1;

FIG. 4 is a schematic plan view of a semiconductor acceleration sensor as a modification of the embodiment.

FIG. 5 is a schematic plan view of a prototype of a semiconductor acceleration sensor capable of detecting acceleration in two axial directions, which is prototyped by the present inventors.

FIG. 6 is an equivalent circuit diagram in a semiconductor acceleration sensor capable of detecting acceleration in two axial directions.

[Explanation of symbols]

12 ... second silicon substrate, 20 ... support substrate, 21 ... opening, 30 ... movable electrode, 31 ... first weight, 32 ... second weight, 38 ... movable electrode pad, 40, 50 ... 1st fixed electrode, 41, 51 ... Lead-out part for fixed electrodes corresponding to 1st fixed electrode, 42, 52 ... Pad part for fixed electrodes corresponding to 1st fixed electrode, 60, 70 ... 2nd fixing Electrodes, 61, 71: fixed electrode lead-out portions corresponding to the second fixed electrode, 62, 72: fixed electrode pad portions corresponding to the second fixed electrode, 80: substrate potential fixing pad portions.

────────────────────────────────────────────────── ───

[Procedure amendment]

[Submission date] September 29, 2000 (2009.2)
9)

[Procedure amendment 1]

[Document name to be amended] Statement

[Correction target item name] Claims

[Correction method] Change

[Correction contents]

[Claims]

[Procedure amendment 2]

[Document name to be amended] Statement

[Correction target item name] 0009

[Correction method] Change

[Correction contents]

Here, assuming that a pair of fixed electrodes 40 and 50 formed on the outer peripheral portions at both ends of the movable electrode 30 along the Y direction are first fixed electrodes, the first fixed electrodes 40 and 50 opposed to each other. A first detection capacitance is formed between the movable electrode 30 and the movable comb portion 36 of the movable electrode 30. The first detection capacitor is composed of one (upper side in the figure) first fixed electrode 40 and movable comb tooth 3
6 and a detection capacitor C1 between the other (lower in the figure) first fixed electrode 50 and the movable comb tooth portion 36.
S2.

[Procedure amendment 3]

[Document name to be amended] Statement

[Correction target item name] 0010

[Correction method] Change

[Correction contents]

Further, a pair of fixed electrodes 60 and 70 formed on the outer peripheral portions at both ends of the movable electrode 30 along the X direction are connected to the second electrode.
Of the second fixed electrodes 6 opposed to each other
0, 70 and the movable comb portion 36 of the movable electrode 30
Is formed. The second detection capacitor is composed of one (the left side in the figure) second fixed electrode 60 and movable comb-tooth portion 36.
Between the detection capacitor CS3 and the other (the right-hand side in the figure) the second
Capacitance CS between the fixed electrode 70 and the movable comb tooth portion 36 of FIG.
4 These detection capacitors CS1 to CS4 are:
This is indicated by a capacitor symbol in FIG.

[Procedure amendment 4]

[Document name to be amended] Statement

[Correction target item name] 0030

[Correction method] Change

[Correction contents]

Furthermore, according to the invention of claim 4, lead portions corresponding to each of the first fixed electrode (40, 50) (41, 51) and a pad portion (42, 52), second
Direction (Y) in along disposed, a second fixed electrode (60,
70), the lead-out portions (61, 71) and the pad portions (62, 72) corresponding to each of (70) are arranged along the first direction ( X ).

[Procedure amendment 5]

[Document name to be amended] Statement

[Correction target item name] 0048

[Correction method] Change

[Correction contents]

The fixed electrodes 40 to 70 are respectively formed on the outer peripheral portions of both ends of the movable electrode 30 along the second direction Y in the second silicon substrate 12, and the first electrode is provided between the fixed electrode 40 and the movable electrode 30. A pair of first capacitors forming the detection capacitors CS1 and CS2
Are formed on the outer periphery of both ends of the movable electrode 30 along the first direction X , respectively.
A pair of second fixed electrodes 60 and 70 that form second detection capacitors CS3 and CS4 between the first and second detection capacitors CS3 and CS4.

[Procedure amendment 6]

[Document name to be amended] Statement

[Correction target item name] 0057

[Correction method] Change

[Correction contents]

Further, in this example, the first fixed electrodes 40, 50
Are arranged along the second direction Y , and correspond to the second fixed electrodes 60 and 70, respectively. The fixed electrode lead portions 61 and 71 and the fixed electrode pad portions 62 and 72 are arranged along the first direction X.

Claims (6)

[Claims] 1. A semiconductor substrate (12) made of a semiconductor, and a first direction (X) formed on the semiconductor substrate and orthogonal to a plane parallel to the semiconductor substrate in response to application of a physical quantity.
And a movable electrode (30) displaceable in a second direction (Y)
And a pair of first electrodes formed on outer peripheral portions of both ends of the movable electrode along the first direction in the semiconductor substrate and forming a first detection capacitance (CS1, CS2) between the movable electrodes and the first electrode. One fixed electrode (40, 50) and a second detection capacitor (CS3) between the movable electrode and the outer periphery of both ends of the movable electrode in the semiconductor substrate along the second direction. , CS4), and a pair of first detection capacitors when the movable electrode is displaced in the first direction in response to the application of a mechanical quantity. An applied dynamic quantity is detected based on a change, and when the movable electrode is displaced in the second direction in response to the application of the dynamic quantity, the applied dynamic quantity is detected based on a change in the pair of second detection capacities. Semiconductor dynamic quantity sensor that detects In addition, on the outer peripheral portion of each of the first and second fixed electrodes as viewed from the movable electrode in the semiconductor substrate, a lead portion (41, 51, 51) electrically connected to each of the fixed electrodes is provided.
61, 71) and a pad portion (42, 52, 62, 72) for external connection electrically connected to the lead portion. 2. The support substrate (2), wherein the semiconductor substrate (12) has a rectangular opening (21) opening on one side.
0), the movable electrode (30) is located substantially at the center of the opening, and each of the first and second fixed electrodes (40 to 70)
The lead-out portions (41-71) and the pad portions (42-72), which are respectively allocated and arranged on each side of the opening, and correspond to each of the first and second fixed electrodes.
The semiconductor dynamic quantity sensor according to claim 1, wherein the semiconductor dynamic quantity sensors are also arranged separately on each side of the opening. 3. The first and second fixed electrodes (40 to 7).
0), the lead-out portions (41-71) and the pad portions (42-72) corresponding to each of the first and second fixed electrodes are symmetrically arranged about the movable electrode (30). 3. The method according to claim 1, wherein
4. A semiconductor dynamic quantity sensor according to claim 1. 4. The lead portion (41, 51) and the pad portion (42, 52) corresponding to each of the first fixed electrodes (40, 50) extend along the first direction (X). The lead portions (61, 71) and the pad portions (62, 62) corresponding to each of the second fixed electrodes (60, 70).
72. The semiconductor physical quantity sensor according to claim 3, wherein 72) is arranged along the second direction (Y). 5. The semiconductor substrate (12) has a rectangular plate shape, and wire bonding pad portions (38, 80) are formed at least at two or more corners of the semiconductor substrate. One of the pad portions is a movable electrode pad portion (38) electrically connected to the movable electrode (30), and the other is a substrate potential fixing pad portion for fixing the potential of the semiconductor substrate. The semiconductor physical quantity sensor according to any one of claims 1 to 4, wherein (80). 6. The movable electrode (30) has a rectangular first shape.
And a second weight (32) protruding from four corners of the first weight (31). Semiconductor dynamic quantity sensor according to the above description.