JPH08152327A - Semiconductor yaw rate sensor and its manufacture - Google Patents

Abstract

PURPOSE: To provide a small-sized high performance yaw rate sensor by employing a transistor type yaw rate sensor in which problems, i.e., deterioration of detection accuracy and erroneous detection of acceleration, are solved. CONSTITUTION: Movable electrodes 51-54 are formed integrally above a semiconductor substrate 1 and a displaceable weight 4, serving as a gate electrode, is supported by beams 31-34 and anchor parts 21-24. Source electrodes 71, 72 and drain electrodes 81, 82 are formed oppositely to the weight 4 on the substrate 1 in order to constitute transistors along with the weight 4 serving as the gate electrode. A lower electrode 9 is formed at a part facing the weight 4 on the substrate 1 except the regions for forming the source and drain electrodes. Fixed electrodes 61-64 for excitation are then disposed on the substrate 11 through the movable electrodes 51-54 and the weight 4 is excited, along with the substrate 1, in the horizontal direction at a predetermined period. Displacement of the weight 4 caused by yaw is detected differentially based on the variation in respective source-drain current.

Description

Detailed Description of the Invention

[0001]

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to a semiconductor yaw rate sensor suitable for vehicle body control, navigation, etc., and a method for manufacturing the same.

[0002]

2. Description of the Related Art A conventional yaw rate sensor for detecting a yaw rate or the like acting on a vehicle body of an automobile is disclosed in Japanese Patent Laid-Open No.
A vibration gyro as disclosed in Japanese Patent No. 223817 is known. In this vibrating gyro, a piezoelectric element is bonded to a specified surface of a metal square bar to form a vibrating body,
This is supported by a thin rod.

Further, Japanese Patent Application Laid-Open No. 4-142420 discloses an angular velocity sensor having a structure in which a piezoelectric element is bonded to a metal tuning fork. That is, in any of these devices that detect acceleration such as yaw rate, the piezoelectric element vibrates the main body, and the distortion caused by the Coriolis force generated by the yaw rate that is the measurement target is detected by the piezoelectric element. Trying to detect by the change of.

Here, the Coriolis force generated at the yaw rate is used as the detection principle.
Is expressed by the following equation. Fc = 2 mVω (1) m: mass V: velocity ω: angular velocity As can be seen from the equation (1), the Coriolis force Fc is the mass m.
When the angular velocity ω is generated when the object of is moving at the velocity V, it is perpendicular to the moving direction and the direction of the rotation axis of the angular velocity ω, and its magnitude is the mass m, the velocity V, and the angular velocity ω. Occurs in proportion. Therefore, if the Coriolis force can be measured, the angular velocity can be obtained.

By the way, the performance such as the detection sensitivity in this type of sensor mechanism depends on the supporting method and the processing accuracy, and a particularly high-performance sensor is expensive due to the difficulty in processing and assembling. Further, there is a problem that miniaturization of the sensor is not easy due to restrictions of processing and assembly.

On the other hand, the applicant of the present application has previously developed a new semiconductor yaw rate sensor that applies semiconductor technology and is small in size and can be manufactured at a low cost by utilizing a transistor type displacement detection mechanism. No. 5-311762 is proposed.

In this semiconductor yaw rate sensor, a weight portion is supported by a beam, a source electrode and a drain electrode are formed on a substrate portion facing the weight portion, and the weight portion is used as a gate electrode via an air gap. A transistor is formed, and an electrode for exciting the weight portion is formed on the weight portion and the substrate.

According to this structure, when a yaw rate is generated in the excited weight portion, the Coriolis force acts in the substrate direction to change the air gap, which changes the drain current of the transistor. The yaw rate can be detected by measuring.

The sensor having this structure can be manufactured by diverting the IC process, and has a feature that the size is reduced and the processing accuracy is dramatically improved. However, as a result of further studies, it was found that there are problems that accuracy is reduced due to the inability to detect differentially, erroneous detection occurs due to acceleration, and a circuit load for solving this is large. Further, in the method of forming a transistor, the method of setting the gate width using the diffusion of impurities has a problem that the diffusion width of the impurities is short, so that the gate width is small and a sufficient current value for detection cannot be secured.

[0010]

SUMMARY OF THE INVENTION The present invention has been made in view of the above points, and solves the deterioration of detection accuracy and erroneous detection of acceleration even if it is a transistor type, and is mounted on, for example, an automobile. It is an object of the present invention to provide a small-sized and high-performance semiconductor yaw rate sensor that can be effectively used for vehicle body control, navigation, etc., and a manufacturing method thereof.

[0011]

In order to achieve the above object, (1) a semiconductor yaw rate sensor according to the present invention comprises:
A semiconductor substrate and a semiconductor substrate, which are arranged above the semiconductor substrate with a predetermined space therebetween and are supported by a beam structure so as to be displaceable,
A movable electrode for excitation is integrally formed so as to project in parallel with the semiconductor substrate, and a weight that functions as a gate electrode of a transistor and a predetermined gap above the semiconductor substrate are provided with a gap between the movable electrode and the movable electrode. Fixed electrodes for excitation, which are arranged in a manner to vibrate the weight by using electrostatic force, and two or more pairs of source electrodes and drain electrodes formed at positions that partially oppose each other on the periphery of the weight of the semiconductor substrate, A lower electrode formed in at least a region where the source electrode and the drain electrode do not face the portion of the semiconductor substrate facing the weight, and a transistor is configured by the gate electrode by the weight and the source electrode and the drain electrode. When the weight is horizontally excited with the semiconductor substrate at a predetermined cycle by the excitation fixed electrode, the yaw rate is controlled. The displacement of the weight associated with is characterized in that so as to differential detected by the current change between the two or more sets of source and drain.

(2) Alternatively, the semiconductor substrate, a weight disposed above the semiconductor substrate with a predetermined space therebetween, movably supported by a beam structure, and functioning as a gate electrode of a transistor, and the semiconductor substrate. And two or more sets of source electrodes and drain electrodes formed at positions facing each other in the periphery of the weight, and at least a region where the source electrode and the drain electrode are not provided in a portion of the semiconductor substrate facing the weight. An exciting lower electrode for vibrating the weight by using electrostatic force, a transistor is constituted by the gate electrode by the weight, the source electrode and the drain electrode, and the exciting lower electrode has a predetermined cycle. When the weight is excited in the direction perpendicular to the semiconductor substrate, the displacement of the weight due to the action of yaw rate is Characterized by being adapted to the differential detected by the current change between more source drain.

(3) In the semiconductor yaw rate sensor having a semiconductor structure, the yaw rate detecting section for detecting a yaw rate and the acceleration detecting section for detecting an acceleration are provided, and the acceleration detecting section detects the acceleration from the yaw rate detecting section. It is characterized in that the detected amount is canceled.

(4) Further, in the method for manufacturing the semiconductor yaw rate sensor according to the present invention, the first step of introducing the impurities into the surface of the semiconductor substrate to form the source electrode and the drain electrode, and the sacrifice on the main surface of the semiconductor substrate. Second forming layer
A third step of forming a beam-shaped weight on the sacrificial layer and a fixed mental electrode for the weight, and introducing an excitation impurity into the semiconductor substrate in a self-aligning manner with respect to the weight to form the source electrode and the drain. A fourth step of setting a gate width formed of electrodes, and a fifth step of etching away the sacrificial layer under the weight so that a change in current between the source and drain electrodes due to displacement of the weight can be detected. The weight is vibrated by the electrostatic force from the excitation fixed electrode and is displaced by the Coriolis force generated by the yaw rate.

[0015]

According to the configuration of (1), the weight is vibrated in the horizontal direction with respect to the substrate by the electrostatic force generated by applying a voltage at a frequency between the fixed electrode for excitation and the weight, and the vibration is perpendicular to the substrate. When a yaw rate with the direction as the axis is applied, the weight is displaced in the horizontal direction with respect to the substrate due to the Coriolis force, and the gate width in the transistor configuration portion is changed. Here, since the two transistors are arranged so that the gate width is displaced in the opposite phase, the drain current is also changed in the opposite phase, and the differential detection becomes possible. The differential detection doubles the amount of change in current, enabling highly accurate detection. According to this configuration, the drain current changes even when acceleration is applied in the direction perpendicular to the substrate, but since the two transistors change in phase and in the same amount, differential detection does not result in erroneous detection. Further, even if the drain current changes due to the temperature characteristics of the transistor, the same amount changes in the same phase, so that differential detection does not result in erroneous detection.

Further, according to the configuration of (2), the weight is vibrated in the vertical direction with respect to the substrate by the electrostatic force generated by applying a voltage at a frequency between the lower electrode for excitation and the weight, and this is perpendicular to the vibration direction. When a yaw rate with a horizontal axis is applied to the substrate, the weight is displaced in the horizontal direction with respect to the substrate due to the Coriolis force, and the gate width in the transistor component changes. Here, since the two transistors are arranged so that the gate width is displaced in the opposite phase, the drain current is also changed in the opposite phase, and the differential detection becomes possible. The differential detection doubles the amount of change in current, enabling highly accurate detection. According to this configuration, the drain current changes even when acceleration is applied in the direction perpendicular to the substrate, but since the two transistors change in phase and in the same amount, differential detection does not result in erroneous detection. Further, even if the drain current changes due to the temperature characteristic of the transistor, it is not erroneously detected by the differential detection. This configuration has a simpler sensor shape than the configuration of (1).

According to the configuration of (3), the operation for removing the acceleration component contained in the yaw rate detection signal becomes possible.
By going through the manufacturing process as described in (4), the drain currents of the two transistors used for the differential can be easily made equal, and adjustment work such as trimming can be omitted.

[0018]

Embodiments of the present invention will be described in detail below with reference to the drawings. FIG. 1 is a configuration diagram showing a plan view of a first embodiment of a yaw rate sensor according to the present invention. Reference numeral 1 is a semiconductor substrate (here, a P-type silicon substrate is used). Anchor portions 21 to 24 are formed on, for example, four positions on the semiconductor substrate 1.
The weight 4 is supported by beams 31 to 34, one ends of which are supported by 1 to 24, respectively. The reason why the beams 31 to 34 are provided with a bend is that the weight 4 is movable in the x and y directions in the drawing, which are the horizontal directions of the substrate.

The weight 4 is for gaining a displacement amount due to the yaw rate, and also serves as a gate electrode in the transistor. Exciting movable electrodes 51 to 54 for receiving vibration are formed at four corners of the weight 4 so as to project in the horizontal direction of the substrate.

The above anchor portions 21 to 24 and the beams 31 to 3
4, the weight 4, and the excitation movable electrodes 51 to 54 are integrally formed of a heat resistant metal such as polysilicon or tungsten. In this embodiment, a typical material, polysilicon, is used.

That is, the weight 4 and the excitation movable electrodes 51 to 54 formed integrally with the weight 4 are arranged at a predetermined interval on the main surface of the semiconductor substrate 1, and are anchored via the beams 31 to 34. Held by parts 21-24.

On the other hand, on the semiconductor substrate 1, the beams 31 to 3 are formed.
4, a weight 4, and a lower electrode 9 having a shape corresponding to the shape is formed at a portion facing the movable electrodes 51 to 54 for excitation. The lower electrode 9 is composed of a diffusion layer formed by introducing N-type impurities into the semiconductor substrate 1 made of P-type silicon by means of ion implantation or the like. Is prevented from being displaced in the substrate direction and coming into contact with the substrate 1.

Further, on the semiconductor substrate 1, source electrodes 71 and 72 and drain electrodes 81 and 82 are formed respectively under the pair of opposing sides of the weight 4, and a part of them overlaps the weight 4. Configure a transistor with a gap. The source electrodes 71 and 72 and the drain electrodes 81 and 82 are diffusion layers formed by introducing N-type impurities into the semiconductor substrate 1 made of P-type silicon by means such as ion implantation.

Further, on the semiconductor substrate 1, the excitation movable electrodes 51 to 54 integrally formed with the weight 4 and the excitation fixed electrodes 61 to 64 are arranged to face each other with a predetermined gap. These excitation fixed electrodes 61 to 64 are formed of a polysilicon material at the same time as the excitation electrodes 51 to 54 and the like, and have a potential difference at a predetermined frequency between the weight 4 and the excitation movable electrodes 51 to 54 integrally formed. Is given to the weight 4 to give vibration in the x direction shown in the figure.

Although not shown in detail, the weight 4 is connected to an external circuit via the beams 31 to 34 and the anchor portions 21 to 24 and aluminum wiring on the substrate 1. The lower electrode 9 is also connected to an external circuit via the aluminum wiring on the substrate 1. Further, the excitation fixed electrodes 61 to 64 are also connected to an external circuit for supplying a voltage signal of a predetermined frequency via the aluminum wiring on the substrate 1.

However, it is assumed that the phase of the voltage signal is inverted by 180 degrees in the set of 61 and 62 and the set of 63 and 64. Further, the source electrodes 71, 72 and the drain electrode 8
Reference numerals 1 and 82 are connected to an external current detection circuit via aluminum wiring on the substrate 1.

In the above structure, a more specific structure will be described below with reference to FIG. FIG. 2A is A of FIG.
A semiconductor substrate 1 showing a cross-sectional structure corresponding to line A
Is composed of P-type silicon and has N on its main surface.
A source electrode 71 and a drain electrode 81, which are diffusion layers of type impurities, are formed. At this time, since the weight 4 functions as a gate, the inversion layer 91 is formed between the source electrode 71 and the drain electrode 81. Weight 4 and semiconductor substrate 1
A gap 100 is set between and, and the weight 4 can be displaced three-dimensionally.

FIG. 2B shows a sectional structure corresponding to the line BB in FIG. 1. The insulating film 11 is formed on the semiconductor substrate 1.
0 is formed on which the weight 4 is attached to the beams 32, 33,
It is supported via the anchor portions 22 and 23. That is,
The insulating film 110 sets the gap 100, and
₂ and Si ₃ N ₄ etc. A lower electrode 9 made of an impurity diffusion layer is formed on the main surface of the semiconductor substrate 1.

The insulating film 110 includes the weight 4 and the beams 31 to 31.
4 and the semiconductor substrate 1, which is a sacrificial layer that sets the gap between the semiconductor substrate 1 and the semiconductor substrate 1. The gap 100 is formed by etching away the portions corresponding to the anchor portions 21 to 24.
During this etching, the polysilicon, which is the material forming the weight 4, the beams 31 to 34, and the anchor portions 21 to 24, and the substrate 1 are not etched, and the insulating film 1 that is a sacrificial layer is not etched.
An etchant is used which only etches 10.

FIG. 2C shows a sectional structure corresponding to the line CC of FIG. 1, in which a gap 100 is set between the weight 4 and the surface of the substrate 1. FIG. 2D is D of FIG.
The cross-sectional structure corresponding to the -D line is shown in FIG.
Drain electrodes 81 and 8 which are diffusion layers due to impurities
2. The lower electrode 9 is formed, and the movable electrode 4 is formed via the gap 100. The inversion layer 9 is provided between the source electrode 71 and the drain electrode 81 and between the source electrode 72 and the drain electrode 82 facing the weight 4 functioning as a gate.
1, 92 are formed. At this time, the source electrodes 71, 7
2. Since the drain electrodes 81 and 82 are formed so as to project from the end of the weight 4, the inversion layers 91 and 92 are the source electrodes 7 and 7.
1 is not formed in the entire area between the drain electrode 81 and the source electrode 72 and the drain electrode 82. The weight 4 is an inversion layer 91,
The region 92 is the same, and the source electrode 71 and the drain electrode 8
1. The positions are set so that the currents between the source electrode 72 and the drain electrode 82 are equal.

The operation of the yaw rate sensor thus configured will be described. When a voltage is applied between the weight 4 and the semiconductor substrate 1, the inversion layers 91 and 92 are formed, and the source electrodes 71 and 72 and the drain electrodes 81 and 8 are formed.
An electric current flows between the two. Further, the fixed electrodes for excitation 61 to
The excitation voltage having a frequency between 64 and the excitation movable electrodes 51 to 54 is applied to the excitation fixed electrodes 61 and 62, 63 and 6
The weight 4 is applied to the substrate 1 by shifting the phase by 180 degrees.
And vibrate horizontally (x direction in FIG. 1). Since the Coriolis force generated by the yaw rate is proportional to the speed of this vibration, in order to increase the vibration speed,
The frequency is preferably selected at the resonance point where the amplitude becomes large.

When a yaw rate having an axis perpendicular to the semiconductor substrate 1 is generated for the weight 4 which is excited and vibrating in this way, Coriolis force proportional to the vibration speed and the mass of the vibrating body is generated, and the weight 4 is generated. Is displaced in the y direction shown by 1 in the figure. When the weight 4 functioning as a gate is displaced in the y direction, the current between the source electrodes 71 and 72 and the drain electrodes 81 and 82 changes, and the yaw rate can be detected by this change in current.

That is, when the weight 4 vibrates in the x direction in FIG. 1, there is no change in the inversion layer regions 91 and 92 corresponding to the gate width of the transistor, but when the yaw rate occurs, the weight 4 is caused by the Coriolis force. Are displaced in the y direction shown in FIG. At this time, the inversion layer regions 91 and 92 increase and decrease in reverse phase, and the source electrodes 71 and 72 and the drain electrodes 81 and 8 are formed.
Since the current between the two changes to the opposite phase, differential detection is possible.

The yaw rate detection circuit based on this differential detection is constructed as shown in FIG. In FIG. 3, 121,
Reference numeral 122 denotes a transistor for detecting the yaw rate (the gate electrode by the weight 4 in FIG. 1, the source electrode 7 on the semiconductor substrate 1 side).
1, 72 and drain electrodes 81, 82), 1
Reference numerals 31 and 132 are resistors required for current-voltage conversion, and 140 is a voltmeter for measuring voltage changes due to the yaw rate.

First, when the yaw rate is not generated, the resistance values of the resistors 131 and 132 are equal, and the transistor 12
Since the resistors 1 and 122 (inversion layer regions) are also configured to be equal to each other, the voltmeter 140 indicates 0V.

Next, when a yaw rate occurs, the resistances of the transistors 121 and 122 change to opposite phases, and the voltmeter 140
The potential difference is displayed on and the yaw rate is detected. In this case, since the potential difference is doubled as compared with the case where the transistor is used alone, highly accurate detection becomes possible.

When acceleration in the direction perpendicular to the substrate occurs, the gap 100 shown in FIG. 2A changes and the carrier concentration of the inversion layer 91 changes, so that the transistor 12 is formed.
The resistance of 1,122 changes, but this is transistor 1
Since there is the same phase and the same amount of change in 21 and 122, no potential difference occurs in the voltmeter 140. Therefore, erroneous detection due to acceleration can be prevented.

Further, regarding the resistance change due to the temperature characteristic generated in the transistors 121 and 122, since the transistors 121 and 122 have the same phase and the same amount, erroneous detection can be prevented.

FIG. 4 shows the manufacturing process of the source electrode 71 portion constituting the transistor section in four stages (A) to (D). The left side of the figure is a top view and the right side of the figure is a top view side. It is the AA sectional view taken on the line. The source electrode 72 and the drain electrodes 81 and 82 are completely the same as the source electrode 71, and are therefore omitted.

First, as shown in FIG. 4A, an impurity is introduced into the surface of the semiconductor substrate 1 by ion implantation to form a diffusion resistance 71 which becomes a source electrode. Next, FIG. 4 (B)
As shown in FIG. 3, a sacrificial layer 110 and a polysilicon layer are formed on the substrate 1, and the polysilicon layer is processed to form a weight 4 that functions as a gate. When forming the weight 4, an opening 150 not shown in FIG. 1 is provided. The opening 150 is formed on the end of the source electrode 71.

Thereafter, as shown in FIG. 4C, a resist 160 is formed by a photo process. At this time, the resist 160 has an opening 152 larger than the opening 150.
Is formed. Using this resist 160 as a mask,
The sacrifice layer 110 below the opening 150 is removed by etching to form the opening 151. At this time, the opening 151
Is preferably slightly larger than the opening 150.

Ions are implanted through the openings 152, 150 and 151 to form an amorphous electrically insulating region 170 on the semiconductor substrate 1. This insulating region 170
Is formed in the opening 150 of the weight 4 by self-alignment,
The final source electrode 71 is formed by partially insulating the source electrode 71. The insulating region 170 becomes larger than the opening 150 due to the scattering of the implanted ions, and the region 1 that is also insulated in the portion facing the weight 4
71 can be formed. The impurities used for ion implantation are Ar
And the like, and those which do not serve as a dopant are preferable.

Finally, as shown in FIG. 4D, the sacrifice layer is etched to form the gap 100 and make the weight 4 movable. Insulation region 171 that is larger than the opening 150
When the weight 4 is displaced in the y direction due to
The overlap between the source electrode 71 and the drain electrode 81 increases, the inversion layer region 91 expands, and the current increases. When the weight 4 is displaced in the opposite direction, the inversion layer region 9
Since 1 is narrowed, the current decreases.

By the way, when it is attempted to detect the horizontal displacement of the weight 4, a source-drain electrode having a region which does not overlap with a portion of the weight 4 which functions as a gate is formed. The gate width must be changed by the displacement of the weight 4.

In the case of such a structure, a normal I
After forming the gate done in the C process, the source,
A self-aligned FET manufacturing method for forming a drain cannot be adopted. Therefore, the gate is formed after forming the source and drain electrodes.

Therefore, due to the displacement of the gate that occurs during the process, the overlap between the gate and the source / drain electrodes changes, and the initial current value differs from the designed value, resulting in an offset.

On the other hand, according to the manufacturing method shown in FIG. 4, it is possible to always form a constant inversion layer width (corresponding to the gate width) 172 with respect to the width 41 shown in FIG. It is possible to prevent the occurrence of offset.

Further, in the formation of the gate width utilizing the diffusion of impurities as disclosed in Japanese Patent Application No. 5-311762, there is a problem in that a sufficient current value for detection cannot be secured because the gate width is very small. By increasing the width 41 of the weight 4 shown in FIG. 4D, a sufficient amount of current can be secured.

Although it is possible that the width 41 shown in the figure changes and the inversion layer width 172 changes due to process variations, the inversion layer width formed between the source electrode 72 and the drain electrode 82 used for differential detection. Is expected to change similarly, it is possible to prevent the occurrence of offset.

In the yaw rate sensor described in the above embodiment, the number of beams is four, but the number of beams is not necessarily four. Further, the excitation electrodes are provided on both sides in the vibration direction, but this may of course be provided on one side, and even if two electrodes are provided on each side, one may be provided, and the excitation electrodes may be arranged on a side orthogonal to the oscillation direction. It may be provided throughout. Although the P-type semiconductor is used as the substrate in the description, it may be made of an N-type semiconductor. In this case, the diffusion electrode is of P type.

FIG. 5 shows the configuration of the second embodiment according to the present invention. However, differences from the first embodiment will be mainly described here. That is, in the first embodiment, the weight is vibrated in the horizontal direction with respect to the substrate and the yaw rate having the rotation axis in the vertical direction of the substrate is detected. In this embodiment, the weight is vibrated in the vertical direction with respect to the substrate, and the horizontal direction of the substrate is detected. The purpose is to detect the yaw rate having an axis in the direction.

In FIG. 5, a semiconductor substrate 1a and a weight 4a which also functions as a gate in a transistor are beams 31a and 32.
a, 33a, 34a, anchor portions 21a, 22a, 23
It is held at a predetermined interval with respect to the substrate 1a via a and 24a. A fixed electrode 9a for excitation is provided on the surface of the substrate 1a with a weight 4a and beams 31a, 32a, 33a, 34a.
Is provided in a portion facing to.

Further, on the surface of the semiconductor substrate 1a, the source electrodes 71a, 72a and the drain electrodes 81a, 8a of the transistors are provided at positions facing the pair of sides of the weight 4a, respectively.
2a is formed.

That is, in the first embodiment, the beams 31, 3 are
By providing the bent portions at 2, 33, and 34, the weight 4 enables the two-dimensional displacement of the substrate 1 in the horizontal direction, whereas in the second embodiment, the displacement of the weight 4a is caused in the vertical direction of the substrate (see FIG. 5). Since it is the middle z direction) and the substrate horizontal direction (x direction in FIG. 5),
The beams 31a, 32a, 33a, 34a have no bent portion. Further, in the first embodiment, the weight 4 and the beams 31, 32, 3 are
The lower electrode 9 for preventing the generation of electrostatic force is formed on the surfaces of the substrates 1 and 3 facing each other, but in the second embodiment, the fixed electrode 9a for excitation is formed instead of the lower electrode 9. There is.

The actual operation is as follows. First, when a voltage is applied between the weight 4a and the excitation fixed electrode 9a at a predetermined frequency, the weight 41 vibrates in the direction perpendicular to the substrate 1a (z direction in FIG. 5). Due to this vibration, the weight 4a (1)
The velocity V of the equation occurs.

On the other hand, when the yaw rate is generated around the y direction shown in FIG. 5, the Coriolis force acts in the x direction in the figure, and the weight 4a is displaced in the x direction. At this time, between the source electrode 71a and the drain electrode 81a, the source electrode 72a
The drain current flowing between the drain electrode 82a and the drain electrode 82a changes to the opposite phase. Therefore, the yaw rate is detected by the differential detection described above. The second embodiment can have a simpler structure than the first embodiment.

FIG. 6 shows the configuration of the third embodiment according to the present invention. However, in FIG. 6, the same parts as those in FIG. 5 are denoted by the same reference numerals, and here, the differences from the second embodiment will be mainly described.

That is, in the second embodiment, the substrate 1a where the weight 4a and the beams 31a, 32a, 33a and 34a are opposed to each other.
The fixed electrode 9a for excitation was provided on the entire surface of the lower electrode 9a in this embodiment as in the first embodiment.
b, and only a part of the substrate 1a facing the center of the weight 4a is used as the excitation fixed electrode 9c. The lower electrode 9b is held at the same potential as the weight 4a in order to suppress the generation of electrostatic force.

With this structure, the magnitude of the electrostatic force can be set by the area of the excitation fixed electrode 9c in addition to the applied voltage between the excitation fixed electrode 9c and the weight 4a.
This makes it possible to adjust the applied voltage during excitation to a desired value by adjusting the area of the excitation fixed electrode 9c.

Particularly, in the sensor structure as shown in this embodiment, the distance between the movable electrode 4a and the substrate 1a is on the order of μm, and a very large electrostatic force is generated by a minute voltage. Therefore, when the movable electrode 4a is excited. It is necessary to make a fine adjustment of the applied voltage in order to prevent the substrate 1a from being adsorbed, but as in this embodiment, the excitation fixed electrode 9c is provided on a part of the surface of the substrate 1a facing the center of the weight 4a. If formed, the area where the electrostatic force is generated can be suppressed, and fine adjustment of the applied voltage becomes unnecessary.

FIG. 7 shows the configuration of the fourth embodiment according to the present invention. The purpose of this embodiment is to provide a yaw rate detecting section and an acceleration detecting section, measure the acceleration component generated in the yaw rate detecting section by the acceleration detecting section, and accurately detect the yaw rate.

In FIG. 7, reference numeral 1b is a semiconductor substrate, and a yaw rate detecting portion 20a and an acceleration detecting portion 20b are arranged on the semiconductor substrate 1b. First, the yaw rate detection unit 2
It will be described from 0a. 21b, 22b, 23b and 24b are anchor parts, 31b, 32b, 33b and 34b are beams, 4
b is a weight for earning a displacement amount by the yaw rate, 42
a and 42b are movable electrodes which also serve as gates of transistors, and 51b, 52b, 53b and 54b are weights 4b.
It is an electrode for excitation that gives vibration to.

The weight 4b, the movable electrodes 42a and 42b, and the excitation electrodes 51b to 54b are arranged above the semiconductor substrate 1b with a predetermined space therebetween, and are supported by the beams 31b to 34b and the anchor portions 21b to 24b. The anchor portions 21b to 24b, the beams 31b to 34b, the weight 4b, the movable electrodes 42a and 42b, and the excitation electrodes 51b to 54b are integrally configured by a heat resistant metal such as polysilicon or tungsten. In this embodiment, polysilicon is used as a typical material.

Reference numerals 71b, 72b, 81b and 82b denote a source electrode and a drain electrode which are diffusion layers formed by introducing N type impurities into the P type silicon substrate by means such as ion implantation, and these are movable electrodes 42a. , 4
2b and a transistor having an air gap.

Reference numerals 61a, 62a, 63a and 64a denote fixed electrodes for excitation, which are electrodes 51 for excitation formed integrally with the weight 4b.
A potential difference is applied between b to 54b at a predetermined frequency, and the weight 4b is vibrated in the x direction shown in the figure. The excitation fixed electrodes 61a to 64a are formed of a polysilicon material at the same time as the excitation electrodes 51b to 54b and the like.

Reference numeral 9d is a lower electrode composed of a diffusion layer formed by introducing N-type impurities into the P-type silicon substrate by means of ion implantation or the like. When the weight 4b is made to have the same potential, the weight 4b is electrostatically charged. Prevents the substrate from being displaced toward the substrate and coming into contact with the substrate 1b.

The movable electrodes 42a and 42b are connected to an external circuit via a weight, a beam, an anchor portion and an aluminum wiring (not shown). Further, excitation fixed electrodes 61a to 64a
Is connected via an aluminum wiring to an external circuit to which a voltage signal of a predetermined frequency is supplied. However, 61a and 62
In the set of a and the set of 63a and 64a, the phase of the voltage signal is 1
It is flipped 80 degrees. Further, the source electrodes 71b, 72
The b and drain electrodes 81b and 82b are connected to an external current detection circuit (not shown) via aluminum wiring.

The yaw rate detecting section 2 having such a configuration
The operation of 0a will be described. First, the movable electrodes 42a,
When a voltage is applied to the semiconductor substrate 1b with respect to 42b, between the source electrode 71b and the drain electrode 81b,
An inversion layer (not shown) is formed between the source electrode 72b and the drain electrode 82b, and the source electrode 72b and the drain electrode 82b are formed between the source electrode 71b and the drain electrode 81b.
Current flows between and.

Further, the excitation voltage having a frequency between the excitation fixed electrodes 61a to 64a and the excitation electrodes 51b to 54b is applied to the excitation fixed electrodes 61a and 62a, 63a and 64.
A phase shift of 180 degrees is applied at a and the weight 4b is vibrated in the horizontal direction of the substrate 1b (x direction in FIG. 7). Since the Coriolis force generated by the yaw rate is proportional to the speed of this vibration, it is preferable to select the frequency at the resonance point where the amplitude becomes large in order to increase the vibration speed.

The weight 4b excited and vibrating in this way
On the other hand, when a yaw rate in the ω direction shown in the figure is generated in the horizontal direction with respect to the semiconductor substrate 1b, a Coriolis force proportional to the vibration speed and the mass of the vibrating body is generated, and the weight 4b moves in the direction perpendicular to the substrate 1b shown in FIG. Displace. Then, the movable electrodes 42a and 42b functioning as gates are displaced in the direction perpendicular to the substrate 1b. As a result, the electric field strength applied to the inversion layer (not shown) changes, and the source electrode 71b and the drain electrode 81b are changed.
And the current flowing between the source electrode 72b and the drain electrode 82b change.

Here, the movable electrode 42 is caused by the Coriolis force.
When a and 42b are displaced in the direction away from the substrate 1b, the two currents decrease by the same amount, and when they approach each other, the two currents increase by the same amount. Therefore, the yaw rate can be detected from this current change.

Next, the acceleration detector 20b will be described. 21c, 22c, 23c, 24c are anchor parts, 3
1c, 32c, 33c and 34c are beams, 4c is a weight for gaining a displacement amount due to the yaw rate, and 43a and 43b are movable electrodes which also serve as gates of transistors.

Here, the beams 31c, 32c, 33c, 34
c is the beams 31b, 32b, 33b, 3 of the yaw rate detector.
It has the same shape as 4b. In addition, the weight 4c and the movable electrode 43
The mass including a and 43b is the weight 4b of the yaw rate detection unit 20a, the movable electrodes 42a and 42b, and the excitation electrode 51.
It is equal to the mass of b, 52b, 53b, and 54b added. The weight 4c integrally formed with the movable electrodes 43a and 43b is arranged above the semiconductor substrate 1b with a predetermined space therebetween, and
It is supported by 1c to 34c and anchor portions 21c to 24c.

These anchor portions 21c to 24c and the beam 31
c to 34c, the weight 4c, and the movable electrodes 43a and 43b are the anchor portions 21b to 24b, the beams 31b to 34b, the weight 4b, the movable electrodes 42a and 42b, of the yaw rate detection unit 20a.
At the same time as the excitation electrodes 51b to 54b, they are integrally formed of a heat-resistant metal such as polysilicon or tungsten. In this embodiment, polysilicon is used as a typical material.

Reference numerals 71c, 72c, 81c and 82c denote a source electrode and a drain electrode which are diffusion layers formed by introducing N-type impurities into the P-type silicon substrate by means such as ion implantation, and these are movable electrodes 43a. , 4
3b and a transistor having an air gap.

Reference numeral 9e is a lower electrode composed of a diffusion layer formed by introducing N-type impurities into the P-type silicon substrate by means of ion implantation or the like. When the weight 4c is made to have the same potential, the weight 4c is electrostatically charged. It is prevented to displace in the substrate direction and come into contact with the substrate 1b.

The movable electrodes 43a and 43b are connected to an external circuit via a weight, a beam, an anchor portion and an aluminum wiring not shown, and the lower electrode 9e is also connected to an external circuit via an aluminum wiring not shown. Further, the source electrodes 71c and 72c and the drain electrodes 81c and 82c are connected to an external current detection circuit (not shown) via aluminum wiring.

The acceleration detecting section 20b thus configured
The operation of will be described. First, the movable electrodes 43a, 43
When a voltage is applied to the semiconductor substrate 1b with respect to b,
Inversion layers (not shown) are formed between the source electrode 71c and the drain electrode 81c and between the source electrode 72c and the drain electrode 82c, and the source electrode 71c and the drain electrode 81 are formed.
An electric current flows between the source electrode 72c and the drain electrode 82c, respectively.

When acceleration in the vertical direction with respect to the substrate 1b occurs in such a state, the weight 4c is displaced in the vertical direction with respect to the substrate 1b. The movable electrode 4 that functions as a gate
By displacing 3a and 43b in the direction perpendicular to the substrate 1b, the electric field strength applied to the inversion layer (not shown) changes and flows between the source electrode 71c and the drain electrode 81c and between the source electrode 72c and the drain electrode 82c. The current changes.

Here, the movable electrodes 43a, 43
When b is displaced in the direction away from the substrate 1b, the two currents decrease by the same amount, and when they approach each other, the two currents increase by the same amount. The acceleration can be detected by this change in current.

By using the yaw rate detecting section 20a and the acceleration detecting section 20b, the yaw rate detecting circuit as shown in FIG. 8 is constructed. In FIG. 8, 123 is a transistor for detecting the yaw rate, 124 is a transistor for detecting the acceleration, and 141 is a voltmeter for measuring a voltage change due to the yaw rate.

First, when the yaw rate is not generated, the voltmeter 141 shows a constant value. When a yaw rate occurs in this, the resistance of the yaw rate detection transistor 123 changes, so the voltage changes, and the yaw rate can be detected.

Next, when only acceleration occurs, the resistance of the yaw rate detecting transistor 123 changes, but the resistance of the acceleration detecting transistor 124 also changes by the same amount, so that the voltmeter 141 does not change. Therefore, erroneous detection due to acceleration can be prevented.

When the yaw rate and the acceleration occur at the same time, only the yaw rate component is detected as a voltage change in the above manner, and a highly accurate yaw rate sensor which does not erroneously detect the acceleration component can be realized.

Although FIG. 7 shows the case where the beam is a straight line and the displacement in the vertical direction of the substrate is detected by the transistor, it is also possible to detect the displacement in the horizontal direction of the substrate shown in FIG. In this case, the acceleration detection unit excluding the excitation electrodes 51 to 54 and the excitation fixed electrodes 61 to 64 from the structure of the yaw rate sensor shown in FIG. 1 and the yaw rate sensor shown in FIG. It is formed.

The yaw rate detection circuit in this case has a structure as shown in FIG. In FIG. 9, 125 and 126 are transistors of the acceleration detection unit, 127 and 128 are transistors of the yaw rate detection unit, and 142 is a voltmeter for measuring voltage changes.

Here, due to the acceleration, the transistor 12
The detectors are designed so that the resistance values of the transistors 5, 127 change in phase and in the same amount.
The detection unit is designed so that the resistance value of 1 changes in phase and in the same amount due to acceleration. However, the transistors 125 and 1
In Nos. 27, 126 and 128, the resistance change due to acceleration is in the opposite phase.

With such a structure, no voltage change occurs when acceleration occurs, and only the resistance change due to the yaw rate can be detected by the voltmeter 142. Also,
It is considered that the temperature characteristics generated in the transistor can be solved. That is, the resistance of the transistor normally changes depending on the temperature, which causes a false detection of the yaw rate. Therefore, if the structural parameters of the transistors are made equal to each other so that the temperature characteristics become equal, erroneous detection due to the temperature characteristics like acceleration can be solved.

[0089]

As described above, according to the semiconductor yaw rate sensor of the present invention, even if it is a transistor type, it is possible to solve the deterioration of the detection accuracy and the erroneous detection of acceleration, and to realize the miniaturization and high performance. For example, it can be mounted on an automobile or the like and effectively used for vehicle body control, navigation, etc.
Further, such a semiconductor yaw rate sensor has a conventional I
It can be easily manufactured by applying the C manufacturing process, and in particular, the transistor structure portion for detecting the yaw rate can be manufactured with high accuracy due to the member displaced by the action of the yaw rate, so that it is highly sensitive and reliable. A rich yaw rate sensor can be easily obtained.

[Brief description of drawings]

FIG. 1 is a first semiconductor yaw rate sensor according to the present invention.
FIG. 3 is a configuration diagram showing a configuration of an embodiment.

FIG. 2 is a sectional view showing a sectional structure of each part of the embodiment.

FIG. 3 is a circuit diagram showing a configuration of a yaw rate detection circuit of the same embodiment.

FIG. 4 is a cross-sectional view showing stepwise a manufacturing process of a source electrode portion which constitutes the transistor portion of the embodiment.

FIG. 5 is a configuration diagram showing a configuration of a second embodiment according to the present invention.

FIG. 6 is a configuration diagram showing a configuration of a third embodiment according to the present invention.

FIG. 7 is a configuration diagram showing a configuration of a fourth embodiment according to the present invention.

FIG. 8 is a circuit diagram showing a yaw rate detection circuit configuration using the same embodiment.

FIG. 9 is a circuit diagram showing a yaw rate detection circuit configuration when the configuration of FIG. 1 is used in a fourth embodiment.

[Explanation of symbols]

1, 1a, 1b ... Semiconductor substrate, 21-24, 21a-2
4a, 21b-24b, 21c-24c ... Anchor part,
31-34, 31a-34a, 31b-34b, 31c
~ 34c ... beams, 4, 4a, 4b, 4c ... weights, 42a,
42b, 43a, 43b ... Movable electrodes 51-54, 51
b-54b ... Exciting movable electrodes, 61-64, 61a-6
4a ... Fixed electrode for excitation, 71, 72, 71a, 72a,
71b, 72b, 71c, 72c ... Source electrode, 81,
82, 81a, 82a, 81b, 82b, 81c, 82
c ... drain electrode, 9, 9b, 9d, 9e ... lower electrode,
91, 92 ... Inversion layer regions, 9a, 9c ... Exciting fixed electrodes, 100 ... Gap, 110 ... Insulating films, 121, 12
2, 123, 127, 128 ... Transistors for yaw rate detection, 124, 125, 126 ... Transistors for acceleration detection, 131, 132 ... Resistors, 140, 141, 14
2 ... Voltmeter, 150-152 ... Opening part, 160 ... Resist, 170,171 ... Insulating region, 41, 171, 172
... width, 20a ... yaw rate detecting section, 20b ... acceleration detecting section.

Claims (10)

[Claims] 1. A semiconductor substrate, and a semiconductor substrate disposed above the semiconductor substrate with a predetermined space therebetween.
A weight that is displaceably supported by a beam structure and integrally formed with a movable electrode for excitation so as to project in parallel with the semiconductor substrate, and that functions as a gate electrode of a transistor, and a predetermined distance above the semiconductor substrate. And a fixed electrode for excitation, which is disposed with a gap between the movable electrode and the movable electrode and which vibrates the weight by using electrostatic force, and is formed at a position facing a part of the periphery of the weight of the semiconductor substrate. A pair of source electrodes and drain electrodes, and a lower electrode formed on at least a region of the semiconductor substrate facing the weight without the source electrode and the drain electrode. A transistor is formed by the electrode and the drain electrode, and the weight is formed on the semiconductor substrate by the excitation fixed electrode at a predetermined cycle. A semiconductor yaw rate sensor, wherein a displacement of the weight due to the action of a yaw rate is differentially detected by a current change between the two or more sets of source and drain when being excited in the horizontal direction. 2. A semiconductor substrate and a semiconductor substrate, which is disposed above the semiconductor substrate with a predetermined space therebetween.
A weight that is displaceably supported by a beam structure and that functions as a gate electrode of a transistor; and two or more pairs of source electrodes and drain electrodes that are formed at positions on the semiconductor substrate that partially oppose each other around the weight. A lower electrode for excitation, which is formed in at least a region where the source electrode and the drain electrode do not exist in a portion of the semiconductor substrate facing the weight, and vibrates the weight by using electrostatic force; and a gate electrode by the weight. A transistor is formed by the source electrode and the drain electrode, and when the weight is excited in a direction perpendicular to the semiconductor substrate by the excitation lower electrode in a predetermined cycle, the displacement of the weight due to the action of a yaw rate is set to the two sets. A semiconductor yawley characterized by performing differential detection based on the above current change between the source and drain. Sensor. 3. The semiconductor substrate is equipotential to the weight in at least a region where the source electrode, the drain electrode, and the lower electrode for excitation are not provided in the facing portion of the movable electrode, and electrostatic force between the weight and the substrate is applied. The semiconductor yaw rate sensor according to claim 2, wherein the semiconductor yaw rate sensor has a lower electrode for suppressing the generation thereof. 4. A semiconductor yaw rate sensor having a semiconductor structure, comprising: a yaw rate detecting section for detecting a yaw rate; and an acceleration detecting section for detecting acceleration, and a detection result of the acceleration detecting section is detected from a detection result of the yaw rate detecting section. A semiconductor yaw rate sensor characterized by being canceled. 5. The semiconductor yaw rate sensor according to claim 4, wherein the yaw rate detector and the acceleration detector are on the same substrate. 6. The yaw rate detection unit and the acceleration detection unit are equal in detection method to each other.
The semiconductor yaw rate sensor described. 7. The semiconductor yaw rate sensor according to claim 4, wherein the yaw rate detecting section and the acceleration detecting section have equal sensitivity to acceleration. 8. The semiconductor yaw rate sensor according to claim 4, wherein the yaw rate detecting section and the acceleration detecting section have beams and weights having the same shape. 9. The semiconductor yaw rate sensor according to claim 4, wherein a transistor is used for each of the yaw rate detecting section and the acceleration detecting section. 10. A first step of forming a source electrode and a drain electrode by introducing impurities into a surface of a semiconductor substrate, a second step of forming a sacrificial layer on a main surface of the semiconductor substrate, and a beam on the sacrificial layer. A third step of forming a shape weight and an excitation fixed electrode for the weight; and a fourth step of performing ion implantation on the semiconductor substrate in a self-aligning manner with respect to the weight and setting a gate width formed of the source electrode and the drain electrode. And a fifth step of etching away the sacrificial layer under the weight so that a change in current between the source and drain electrodes due to displacement of the weight can be detected. A method for manufacturing a semiconductor yaw rate sensor, characterized in that the semiconductor yaw rate sensor is vibrated by an electrostatic force from an electrode and is displaced by a Coriolis force generated by a yaw rate.