JPH07167885A - Semiconductor acceleration sensor and manufacture thereof as well as detecting system of acceleration by the sensor - Google Patents

Abstract

PURPOSE:To obtain a microminiature semiconductor acceleration sensor using a pulse transform method of an electromagnetic conduction system. CONSTITUTION:An arm 2 having elasticity is supported on a support 4, in parallel to a silicon base 1. A movable electrode 6 in the shape of a rectangular pulse of a prescribed pitch (d) is supported in the vicinity of the fore end of the arm 2, in parallel to the base 1, and a flat-plate-shaped weight part 3 is provided at the fore end of the arm 2. A fixed electrode 8 in the shape of the rectangular pulse is formed on the base 1 oppositely to the movable electrode 6 and an insulating film 9 of SiN is formed on the surface of the base 1, while a recession 5 is formed in the surface of the base 1 so that the weight part 3 can be displaced swingably in directions of the axis X and the thickness thereof. A constant current is made to flow through the movable electrode 6 from an input terminal 10, and acceleration is detected by detecting the number of pulses of an induced current generated in the fixed electrode 8 by the acceleration applied.

Description

Detailed Description of the Invention

[0001]

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to a semiconductor acceleration sensor, a method of manufacturing the same, and an acceleration detection method using the acceleration sensor. Specifically, the present invention relates to a semiconductor acceleration sensor to which an electromagnetic induction type pulse conversion method is applied, a method for manufacturing the acceleration sensor, and an acceleration detection method using the acceleration sensor.

[0002]

2. Description of the Related Art As a conventional semiconductor acceleration sensor detection method, a piezo detection method using a piezoresistive element or a strain gauge, or a capacitance detection method using capacitance between electrodes There is. What is the piezo detection method?
A piezoresistive element, a strain gauge, or the like is provided in a sensor portion that easily bends due to acceleration, and the magnitude of acceleration is detected by detecting a change in resistance value caused by the bending caused by acceleration.

In recent years, there has been an increasing demand for miniaturization and high functionality of semiconductor acceleration sensors. In this piezo-detection type acceleration sensor, the resistance value changes when the piezoresistive element, strain gauge, etc. are made smaller. Since it becomes difficult to detect the, it is difficult to miniaturize the acceleration sensor. Further, for example, in a high-performance sensor in which an arithmetic processing unit (CPU) is incorporated in a signal processing circuit for obtaining the magnitude of acceleration, an arithmetic processing unit is used by converting an analog signal indicating the magnitude of resistance value into a digital signal. Since the A / D conversion for inputting into the input is required and the A / D conversion circuit must be added, the signal processing circuit for detecting the signal of the acceleration sensor is further complicated.

The electrostatic capacitance detection method utilizes that the movable weight is displaced by acceleration, and a fixed electrode is provided on a substrate or the like so as to face the movable electrode formed on the upper surface or the lower surface of the movable weight. The change in the distance between the two electrodes due to the displacement of the movable weight due to the acceleration is detected as a change in the capacitance between the two electrodes, and the magnitude of the acceleration is detected.

Even in this capacitance detection type acceleration sensor, the capacitance detection circuit must be made highly sensitive so as to detect a minute capacitance change as the sensor becomes smaller, which is difficult to cope with. It was a thing. Further, similar to the piezo detection method, the A / D conversion circuit makes the signal processing circuit such as the capacitance detection circuit complicated.

[0006]

SUMMARY OF THE INVENTION The present invention has been made in view of the drawbacks of the above-mentioned conventional examples, and an object thereof is to use the pulse conversion method of the electromagnetic induction system to solve the above problems. To solve.

[0007]

In a semiconductor acceleration sensor according to the present invention, a fixed electrode having a substantially rectangular pulse pattern of a constant pitch is provided on a substrate surface, and a displaceable portion having a conductive portion of a substantially rectangular pulse pattern of a constant pitch is provided. The movable electrode is opposed to the fixed electrode with a certain gap.

In this semiconductor acceleration sensor, the base of an arm having elasticity may be supported on the substrate, and the movable electrode may be provided on the tip side of the arm. Further, a weight portion may be provided at the tip of the arm.

In this semiconductor acceleration sensor, a strain detecting element can be provided at the base of the arm, and the strain detecting element can detect the acceleration component in the direction perpendicular to the surface of the substrate. Alternatively, by forming a capacitance forming electrode on the lower surface of the weight portion, providing another capacitance forming electrode on the substrate so as to face the capacitance forming electrode, and detecting a change in capacitance between the capacitance forming electrodes. It is also possible to be able to detect the acceleration component in the direction perpendicular to the substrate surface.

Further, in the semiconductor acceleration sensor, a plurality of sets of the fixed electrode and the movable electrode having different rectangular pulse pattern pitches can be provided.

Further, in the semiconductor acceleration sensor, a plurality of sets of the fixed electrodes and the movable electrodes are provided, and one set of the fixed electrodes and the movable electrodes and the other set of the fixed electrodes and the movable electrodes are arranged in directions orthogonal to each other. It can also be arranged.

Further, at least a part of the movable electrode may be made of polycrystalline silicon or single crystal silicon.

Further, an insulating film may be formed on at least one of a region including the fixed electrode on the surface of the substrate and a surface of the movable electrode facing the fixed electrode. In this case,
The fixed electrode can be formed by introducing impurities into the surface layer portion of the substrate made of silicon, and the insulating film can be formed by introducing carbon into the surface layer portion of the substrate shallower than the fixed electrode. Alternatively, the insulating film may be formed by depositing silicon and carbon on the surface of the substrate.

In the semiconductor acceleration sensor, a supporting film having a compressive stress may be formed on the lower surface of the movable electrode, and a supporting film having a tensile stress may be formed on the upper surface of the movable electrode. In the latter case, the support film can be formed of silicon nitride.

A method of manufacturing a semiconductor acceleration sensor according to the present invention is a method for manufacturing the semiconductor acceleration sensor described above, wherein a sacrificial layer is previously formed on the surface of the fixed electrode,
A movable electrode is formed by forming a structure made of polycrystalline silicon above the sacrifice layer, and forming a conductive region in the structure with a doping pattern or a metal pattern, and fixing by etching away the sacrifice layer. A feature is that a gap is formed between the electrode and the movable electrode.

In another method of manufacturing a semiconductor acceleration sensor according to the present invention, a conductive region is formed on a surface layer portion of a single crystal silicon substrate by a doping pattern or a metal pattern, and the single crystal silicon substrate is formed below the surface layer portion. An oxide film is formed by thermal oxidation of the formed porous silicon, and a sacrificial layer etching of the oxide film is performed to form a movable electrode on the surface layer portion and a gap on the lower surface of the movable electrode. It has a feature.

In still another method of manufacturing a semiconductor acceleration sensor according to the present invention, a conductive region is formed by a doping pattern or a metal pattern on a surface layer portion of a single crystal silicon substrate, and a single region is formed around a predetermined region including the dielectric region. The crystalline silicon substrate is trench-etched and then anisotropically etched below a predetermined region to form a movable electrode and a gap is formed on the lower surface of the movable electrode.

Another semiconductor acceleration sensor according to the present invention is
A fixed electrode having a substantially-rectangular pulse pattern with a constant pitch is provided on the surface of the substrate, and a movable electrode having a stretchable mesh structure having a conductive portion having a substantially-rectangular pulse pattern with a constant pitch is separated from the fixed electrode by a certain gap. And a weight portion is connected to the free end of the movable electrode while supporting one end of the movable electrode.

In still another semiconductor acceleration sensor according to the present invention, a pair of expandable and contractible mesh-like structures supported at one end are arranged on both sides of the weight portion, and the free ends of the mesh-like structures are attached to the weight portion. Connected to both sides of the movable electrode, and at least one of the mesh-like structures is provided with a conductive portion having a substantially rectangular pulse pattern with a constant pitch to form a movable electrode, which is fixed on the substrate surface with a constant gap below the movable electrode. It is characterized in that a fixed electrode having a substantially rectangular pulse-shaped pattern with a pitch is provided.

In still another semiconductor acceleration sensor according to the present invention, a pair of stretchable mesh-like structures supported at one end are arranged on three or four sides of the weight portion and the free ends of the mesh-like structures are arranged. Connected to the outer circumference of the weight part on three or four sides, at least one of the mesh structures arranged in one direction is provided with a conductive part having a substantially rectangular pulse pattern with a constant pitch to form a movable electrode. , At least one of the mesh structures arranged in the other direction is also provided with a conductive portion having a substantially rectangular pulse pattern of a constant pitch to form a movable electrode, and a fixed electrode is provided below each movable electrode. It is characterized in that fixed electrodes having a substantially rectangular pulse pattern with a constant pitch are provided on the substrate surface with a gap.

In the above semiconductor acceleration sensor, any of the mesh-like structures disposed on both sides of the weight portion may be used as the movable electrode. At least a part of the mesh structure may be made of polycrystalline silicon or polyimide.

An acceleration detection method using the semiconductor acceleration sensor of the present invention uses the semiconductor acceleration sensor, wherein either one of the fixed electrode and the movable electrode has a constant current or an alternating current with a constant frequency. And the acceleration is detected by detecting a change in the induced current generated in one of the other electrodes by the induced electromotive force. In this case, a current-voltage converting means for converting the induced current into a pulse voltage and outputting the pulse voltage may be provided.

Another acceleration detection method by the semiconductor acceleration sensor of the present invention detects acceleration in a direction parallel to the surface of the substrate by detecting a change in electrostatic capacitance between the movable electrode and the fixed electrode. The feature is that it is detected.

Further, in another acceleration detecting method by the semiconductor acceleration sensor of the present invention, a vibrating electrode is provided on the side surface of the weight portion provided at the tip of the arm, and an electrostatic force is applied to the vibrating electrode to apply the weight to the weight. Part is oscillated at a constant frequency and a constant current is applied to one of the movable and fixed electrodes to detect the frequency change of the induced current flowing to the other electrode, or between the movable and fixed electrodes. It is characterized in that the change in the capacitance of is detected.

Still another acceleration detecting method using the semiconductor acceleration sensor of the present invention is provided with a vibrating electrode on the side surface of a weight portion provided at the tip of the arm, and one of the movable electrode and the fixed electrode is provided. The detection signal generated by the acceleration is detected by detecting the frequency change of the induced current generated in the other electrode by applying an alternating current of a constant frequency to the other electrode, or by detecting the change in the capacitance generated between the movable electrode and the fixed electrode. It is characterized in that an electrostatic force is generated in the vibrating electrode so as to cancel the change of the above, and the acceleration is detected from the feedback signal input to the vibrating electrode.

Still another acceleration detection method using the semiconductor acceleration sensor of the present invention is provided with means for mounting a heat-sensitive deformation element for vibrating the arm on a side surface of the arm, and electrically heating the heat-sensitive deformation element. An alternating current with a constant frequency is applied to one of the electrodes and the fixed electrode to detect the frequency change of the induced current generated in the other electrode, or the change in the capacitance between the movable electrode and the fixed electrode. Is detected, the heat-sensitive deformable element is energized and heated so as to cancel the change in the detection signal caused by acceleration, and the acceleration is detected from the energized signal.

Still another acceleration detection method by the semiconductor acceleration sensor of the present invention is a method in which a constant current or an alternating current having a constant frequency is applied to either one of the movable electrode and the fixed electrode, and either one is induced by an induced electromotive force. The change in the induced current generated in the other electrode is used as an output signal, and the acceleration is detected from the difference between the output signals of the movable electrodes or the fixed electrodes arranged on both sides of the weight portion.

According to another acceleration detecting method by the semiconductor acceleration sensor of the present invention, a vibrating electrode is provided on a side surface of the weight portion opposite to the side surface connected to the movable electrode, and either the movable electrode or the fixed electrode is provided. An alternating current with a constant frequency is applied to one electrode to detect the frequency change of the induced current generated in the other electrode, or to detect the change in the capacitance between the movable electrode and the fixed electrode, which is caused by acceleration. An electrostatic force is generated in the vibration electrode so as to cancel the change in the detection signal, and the acceleration is detected from the feedback signal input to the vibration electrode. In this case, movable electrodes are respectively connected to two side surfaces of the weight portion in a direction substantially orthogonal to each other, and vibration electrodes are respectively provided on the side surfaces opposite to the side surfaces to which the movable electrodes of the weight portion are connected. The acceleration may be detected.

In still another acceleration detecting method by the semiconductor acceleration sensor of the present invention, a mesh-like structure connected to one side surface of the weight portion with the weight portion interposed therebetween is used as a movable electrode, and a mesh-like structure connected to the other side surface thereof. The structure is the actuator,
An alternating current with a constant frequency is applied to one of the movable and fixed electrodes to detect the frequency change of the induced current generated in the other electrode, or the capacitance generated between the movable and fixed electrodes. Is detected, the actuator is driven so as to cancel the change in the detection signal caused by the acceleration, and the acceleration is detected from the drive signal of the actuator. In this case, the movable electrodes are connected to two side surfaces of the weight portion in a direction substantially orthogonal to each other, and the actuator is connected to a side surface opposite to the side surface to which the movable electrodes of the weight portion are connected, respectively. The acceleration may be detected.

[0030]

In the semiconductor acceleration sensor of the present invention, the movable electrode having the rectangular pulse pattern is displaceable with respect to the fixed electrode having the rectangular pulse pattern provided on the surface of the substrate. As a result of the movement of the movable electrode, it is possible to detect acceleration or a change in acceleration from an electrical signal generated in the fixed electrode or the movable electrode.

For example, by providing the movable electrode on the tip side of the elastic arm whose base is supported, the movable electrode can be displaced or vibrated together with the arm under the influence of acceleration. Especially, if a weight is provided at the tip of the arm,
Low acceleration can be detected with high sensitivity.

Further, by forming a movable electrode having a substantially rectangular pulse-shaped conductive portion in a stretchable mesh structure and providing a weight portion at its free end, the movable electrode is stretched and contracted by the influence of acceleration. The movable electrode can be displaced with respect to the fixed electrode. In this case, if the structure part of the movable electrode is made of polycrystalline silicon or polyimide, the elasticity of the movable electrode can be improved.

In order to detect acceleration with this semiconductor acceleration sensor, for example, a movable electrode (or a fixed electrode) is used.
If a constant current or a constant frequency current is applied to the fixed electrode (or the movable electrode), the induced current is generated in pulses in the fixed electrode (or the movable electrode) when the movable electrode is displaced due to the influence of acceleration. May change,
By detecting it, the acceleration and its change can be detected. In this case, if the induced current is converted into a pulse voltage and output, it can be directly processed by the arithmetic processing unit (CPU).

Further, the acceleration can be detected from the change of the electrostatic capacitance between the movable electrode and the fixed electrode, and in this case also, unlike the conventional capacitance type acceleration sensor, the change of the electrostatic capacitance is a pulse signal. The acceleration can be detected from the change in the number of pulses or the frequency.

If the vibrating electrode provided on the side surface of the weight portion, the heat-sensitive deformable element provided on the side surface of the arm, or the actuator formed on one mesh structure, the weight portion is vibrated at a constant frequency. Since a pulsed output is obtained and the output frequency changes due to the influence of acceleration, the acceleration and the change can be detected from the frequency change. Alternatively,
If feedback control is performed so as to cancel the change in the output frequency due to the influence of acceleration by the vibration electrode provided on the side surface of the weight portion, the heat-sensitive deformable element provided on the side surface of the arm, and the actuator formed on one mesh structure, Acceleration can be detected from the feedback signal, and a servo-type semiconductor acceleration sensor can be obtained.

Further, in a semiconductor acceleration sensor of a type using a mesh-shaped movable electrode, a stretchable mesh-like structure is provided on both sides of the weight portion, and a conductive portion is provided in any one of the mesh-like structures. If the movable electrode, the weight portion is less likely to be displaced in the lateral direction orthogonal to the expansion and contraction direction of the mesh structure,
The other axis sensitivity can be reduced. In particular, if the mesh-like structures on both sides are made movable electrodes, a larger output signal can be obtained and the sensitivity can be improved. In addition, the mesh structure is arranged on three or four sides of the weight part,
If the movable electrodes are formed in two directions that are orthogonal to each other, acceleration or a change in acceleration can be detected in two directions parallel to the substrate, and a two-dimensional semiconductor acceleration sensor can be obtained.

Further, in a semiconductor acceleration sensor of a type in which the movable electrode is supported by an arm, if a plurality of movable electrodes are provided and arranged in directions orthogonal to each other, the two parallel electrodes can be formed.
A two-dimensional semiconductor acceleration sensor can be manufactured by detecting the directional acceleration. Furthermore, if a strain detecting element is provided at the base of the arm so that the acceleration in the direction perpendicular to the substrate is also detected, a two-dimensional or three-dimensional semiconductor acceleration sensor can be obtained. Further, even if a capacitance forming electrode is formed on the lower surface of the weight portion and another capacitance forming electrode is provided so as to face the capacitance forming electrode and the change in the capacitance between both capacitance forming electrodes is detected, the three-dimensional semiconductor acceleration sensor Can be obtained.

If a plurality of sets of fixed electrodes and movable electrodes having different pitch dimensions of the rectangular pulse pattern are provided, output signals of different frequencies can be obtained, and the respective frequency characteristics are complemented to stabilize the output. Therefore, it is possible to detect vibration in a wide band.

If the movable electrode is made of single crystal silicon or polycrystalline silicon, the movable electrode can be manufactured by a semiconductor manufacturing process, and a fine and precise rectangular pulse pattern can be manufactured.

Further, if an insulating film is provided between the movable electrode and the fixed electrode, there is no risk of a short circuit between the movable electrode and the fixed electrode even if the movable electrode contacts the substrate. In particular, if the fixed electrode is formed by introducing impurities into the substrate and carbon is introduced shallower than the fixed electrode to form the insulating film, the insulating film can be formed on the substrate side. Alternatively, the insulating film can be formed on the substrate side by depositing silicon and carbon on the substrate surface. By forming the insulating film on the substrate side in this manner, it is possible to prevent the movable electrode from becoming heavy due to the insulating film.

Further, by providing a supporting film having a compressive stress on the lower surface of the movable electrode or a supporting film of silicon nitride having a tensile stress on the upper surface of the movable electrode, the movable electrode moves toward the substrate side by its own weight. It can be prevented from hanging down, and interference between the movable electrode and the substrate or fixed electrode can be avoided.

Further, according to the method of manufacturing the semiconductor acceleration sensor of the present invention, after the movable electrode is formed on the sacrificial layer, the sacrificial layer is removed by etching to fix the movable electrode made of a material different from the substrate. It can be easily formed at a distance from the electrode. Further, the oxide film formed by thermal oxidation of porous silicon is subjected to sacrifice layer etching, whereby the substrate and the movable electrode can be integrally formed from a single single crystal silicon substrate. Alternatively, when the trench etching method is used, the substrate and the movable electrode can be formed as an integral structure by a single side process using one single crystal silicon substrate, and the manufacturing process can be simplified.

When the movable electrodes having a mesh structure are provided on both sides of the weight portion, the acceleration can be detected from the difference between the output signals from the movable electrodes or the fixed electrodes on both sides.

[0044]

According to the present invention, since the detection signal indicating the acceleration or the change in acceleration is output as a pulse signal, it can be processed as a digital signal, and even when signal processing is performed by a signal processing circuit of a digital signal processing system, A digital output can be obtained without using a digital / analog conversion processing circuit or the like unlike the conventional analog type acceleration sensor. Therefore, a digital output type semiconductor acceleration sensor can be manufactured with a simple structure, and the signal processing circuit can be simplified. Also, because it outputs a digital signal, it is possible to improve the detection accuracy,
It is possible to manufacture a new compact semiconductor acceleration sensor that is resistant to noise and has a small measurement error.

Further, unlike the conventional capacitance type acceleration sensor, large air damping does not occur between the movable electrode and the fixed electrode and the acceleration detection sensitivity is not deteriorated.
It is possible to manufacture a high-sensitivity semiconductor acceleration sensor having a wider detection range than the conventional capacitance type or strain detection type acceleration sensor.

Also, various acceleration detection methods are possible due to the new structure of the acceleration detection section, and a detection method according to the application can be adopted. For example, the movable electrode and the fixed electrode which are not the detection side. By applying an alternating current to the electrode of or, or detecting the phase of the output signal, the acceleration can be detected with higher accuracy than the amount of movement of the movable electrode. Further, the accuracy of the acceleration sensor can be improved by using a servo detection type acceleration sensor. Further, in order to further improve the detection accuracy or to broaden the sensitivity band, a plurality of movable electrodes and fixed electrodes having different pitches can be provided. Moreover, a two-dimensional acceleration sensor and a three-dimensional acceleration sensor can be easily configured with one unit.

[0047]

1A is a schematic perspective view of a semiconductor acceleration sensor B1 according to an embodiment of the present invention (the insulating film 9 is omitted), and FIG. 1B is X1 of FIG. 1A. FIG. 1C is a cross-sectional view taken along line X2-X2, taken along the line X1-X1. The entire surface of the single crystal silicon substrate 1 is S
The silicon substrate 1 is covered with the insulating film 9 of iN.
On the upper surface of the arm 2, an elastic rod-like arm 2 is arranged substantially parallel to the surface of the insulating film 9 and the base end portion of the arm 2 is supported by a support base 4 provided on the insulating film 9. It is cantilevered. Further, a flat weight portion 3 is provided at the tip of the arm 2 to facilitate detection of low acceleration. The weight portion 3 provided on the arm 2 can swing in the X-axis direction (lateral direction) and in the thickness direction of the weight portion 3, so that the displacement of the weight portion 3 is not hindered. The upper surface is provided with a recess 5 larger than the normal maximum displacement of the weight portion 3. A movable electrode 6 is provided near the tip of the arm 2 near the weight portion 3 so as to project laterally. The movable electrode 6 is formed with a gap 15 between the insulating film 9 on the surface of the silicon substrate 1 and the silicon. It is arranged parallel to the surface of the substrate 1. The movable electrode 6 is shown in FIG.
As shown in (b) and (c), a metal electrode film 7b of Al or the like is formed on the upper surface of a structure 7a made of polycrystalline silicon, and is formed in a rectangular pulse shape with a constant pitch d. Both ends of the metal electrode film 7b of the movable electrode 6 are connected to the arm 2
And a pair of input terminals 10 provided on the insulating film 9 at the end portion of the silicon substrate 1 via connection wires 13 that pass through the surface of the support base 4 and are laid on the surface of the insulating film 9. .

The structure 7a does not necessarily have a rectangular pulse shape, and the metal electrode film 7b may have a rectangular pulse shape with a constant pitch, but as in this embodiment,
It is possible to reduce the weight of the movable electrode 6 by making the entire movable electrode 6 into a rectangular pulse shape, and to prevent the movable electrode 6 from bending downward due to its own weight and coming into contact with the silicon substrate 1. Further, as described in the manufacturing method, the structure 7a of the movable electrode 6, the arm 2, the support 4, and the weight portion 3 are integrally made of polycrystalline silicon.

A fixed electrode 8 is provided on the surface of the silicon substrate 1 so as to face the movable electrode 6 provided on the arm 2, and a fixed gap 15 is provided between the movable electrode 6 and the fixed electrode 8. It is provided. The fixed electrode 8 is wired on the surface of the silicon substrate 1 by a metal thin film such as Al and has a rectangular pulse shape with the same pitch d as the movable electrode 6. Both ends of the fixed electrode 8 are connected to the output terminal 11 provided at the end portion of the silicon substrate 1 via the connection wiring 14 provided on the surface of the silicon substrate 1. Further, an insulating film 9 of SiN is formed on the entire surface of the silicon substrate 1 so as to cover the fixed electrode 8 and the connection wiring 14, and the output terminal 11 is exposed from an opening provided in the insulating film 9. .

Input terminal 1 connected to movable electrode 6
A power circuit (not shown) is connected to 0 and 10,
A constant direct current (or alternating current) I1 can be passed through the movable electrode 6 by a power supply circuit. A resistance element 12 for converting the induced current I2 flowing through the fixed electrode 8 into a voltage signal is connected between the output terminals 11 connected to the fixed electrode 8. The voltage across the resistance element 12 is The output voltage is detected by a detection circuit (not shown).

The acceleration sensor B1 detects acceleration acting in the X-axis direction of FIG. 1, and when acceleration is applied in the X-axis direction, the arm 2 is elastically deformed due to inertial force, and is provided on the arm 2. The movable electrode 6 is easily displaced parallel to the silicon substrate 1 in the X-axis direction. The amount of displacement of the movable electrode 6 is almost proportional to the magnitude of the change in acceleration. When the movable electrode 6 moves, an induced electromotive force due to electromagnetic induction is generated in the fixed electrode 8, and a rectangular pulse-shaped induced current I2 is generated. Flows. FIG. 2 shows the relationship between the direct current I1 flowing through the movable electrode 6 and the induced current I2 flowing through the fixed electrode 8. Now, consider a case where a constant DC current I1 is flowing in the movable electrode 6 in the direction of the arrow from left to right in FIG. First, when the movable electrode 6 is not displaced when acceleration is not applied to the acceleration sensor B1 (or is maintained in a state where a constant acceleration is applied), as shown in FIG. Does not flow the induced current I2. Next, the acceleration sensor B1
When the movable electrode 6 moves in the + X direction due to the change in the acceleration applied to the fixed electrode 8 as described below.
And the induced current I with a pulse number proportional to the displacement of the movable electrode 6
2 will flow. For example, the movable electrode 6 is the fixed electrode 8
In the period in which the movable electrode 6 moves from the position displaced by ¼ pitch (d / 4) in the −X direction to the position displaced by ¼ pitch in the + X direction,
An induced current I2 flows through the fixed electrode 8 in the direction of arrow B in FIG. Further, during the period in which the movable electrode 6 moves from the position displaced by 1/4 pitch in the + X direction to the position displaced by 3/4 pitch (3d / 4), the fixed electrode 8 moves in the direction of arrow A in FIG. The induced current I2 flows. Movable electrode 6 total d
When the movable electrode 6 and the fixed electrode 8 are completely overlapped with each other when displaced by (1 pitch), the acceleration applied to the movable electrode 6 changes and the movable electrode 6 is displaced in the X-axis direction. Every time the movable electrode 6 moves by one pitch d, the induction current I2 for one cycle or one pulse flows through the fixed electrode 8. The induced current I2 flowing through the fixed electrode 8 is the output terminal 1
The resistance element 12 connected to 1 performs current-voltage conversion into a voltage pulse, and this voltage pulse is monitored by the detection circuit and processed by the arithmetic processing unit (CPU). In this way, the amount of displacement of the movable electrode 6, that is, the amount of change in acceleration is detected from the number of pulses of the induced current I2.
The number of pulses of the induced current I2 generated in the fixed electrode 8 is
Since the pitch d of the movable electrode 6 and the fixed electrode 8 is smaller, the change in acceleration can be detected with higher sensitivity and accuracy by reducing the pitch d. Further, since the induced electromotive force generated in the fixed electrode 8 is proportional to the magnitude of the direct current I1 flowing in the movable electrode 6, the induced current I2 can be increased by increasing the direct current I1 flowing in the movable electrode 6. Furthermore, if the pulse phase difference between the direct current I1 flowing through the movable electrode 6 and the induced current I2 flowing through the fixed electrode 8 is detected, the acceleration can be detected with high accuracy of 1 pitch d or less.

When the acceleration sensor B1 detects mechanical vibration, impact, etc., the movable electrode 6 also vibrates in the X-axis direction due to the vibration of the weight portion 3. At this time, when the vibration of the weight portion 3 is intense, the moving speed of the movable electrode 6 increases, so that the pulse frequency of the induced current I2 output from the fixed electrode 8 increases, and when the vibration becomes gentle, the movable electrode 6 moves.
Also, the moving speed of is reduced, so that the pulse frequency of the induced current I2 output from the fixed electrode 8 is reduced. Therefore, the intensity of vibration can also be detected by monitoring the pulse frequency of the induced current I2 flowing through the fixed electrode 8.

An alternating current I1 having a frequency f ₀ may be passed through the movable electrode 6 by a power supply circuit. In this case, even if the movable electrode 6 does not move, the induced current I2 of the frequency f ₀ flows through the fixed electrode 8. When acceleration is applied to the pressure sensor B1 and the movable electrode 6 moves, the frequency of the induced current I2 flowing through the fixed electrode 8 changes by Δf according to the acceleration, and the induced current I2 with the pulse frequency f = f ₀ + Δf flows. Therefore, in this case, the acceleration applied to the acceleration sensor B1 can be known by detecting the frequency f of the induced current I2 flowing through the fixed electrode 8. In particular, if the high-frequency alternating current I1 is passed through the movable electrode 6, the acceleration can be detected with high accuracy even when the displacement of the movable electrode 6 is small.

Therefore, in this acceleration sensor B1, acceleration and vibration can be detected directly from the pulse number and pulse frequency of the pulse voltage generated in the resistance element 12 by the induced current I2 flowing through the fixed electrode 8. A detection signal can be output as a digital signal, and highly accurate detection can be performed. Moreover, since it is a digital type, it has a characteristic of being resistant to noise unlike the conventional analog signal output type acceleration sensor, and an analog-digital (A / D) conversion circuit or a current is used even when processing is performed by an arithmetic processing device. -Direct and easy signal processing is possible without the need for a voltage conversion circuit. Therefore, even if the acceleration sensor B1 is miniaturized, it is possible to detect acceleration, vibration and the like with high accuracy, and it is possible to manufacture a small-sized, highly accurate and highly sensitive acceleration sensor B1.

The insulating film 9 is formed on the surface of the silicon substrate 1.
Therefore, even if the movable electrode 6 comes into contact with the insulating film 9, the friction of the movable electrode 6 can be reduced and the movement of the movable electrode 6 is not hindered.

In the above embodiment, a direct current or an alternating current I1 is made to flow through the movable electrode 6 and an induced current I2 is generated at the fixed electrode 8, but this may be reversed, and the fixed current may be fixed. A direct current or an alternating current may be passed through the electrode 8 and acceleration, vibration or the like may be detected from the change in the induced current generated in the movable electrode 6.

Next, a method of manufacturing the acceleration sensor B1 will be described in detail with reference to the schematic cross-sectional views of FIGS. 4A to 4K (not showing a specific structure or shape). . First, a metal thin film 22 of Al is formed on the entire surface of a silicon substrate 21 made of single crystal silicon (silicon wafer) prepared as shown in FIG. 4A (FIG. 4B), and the metal thin film 22 is etched. And the like to form the rectangular pulse-shaped fixed electrode 8, the metal wiring 14, and the output terminal 11 on the surface of the silicon substrate 21 (FIG. 4).
(C)). Next, as shown in FIG. 4D, plasma CVD is performed from above the fixed electrode 8, the connection wiring 14, and the output terminal 11.
S is formed on the entire surface of the silicon substrate 1 by the (PCVD) method.
The insulating film 9 is formed by depositing iN. Then, the SiO2 film 23 is formed on the entire surface of the insulating film 9 (Fig. 4 (e)), the arm 2 and the weight portion 3, SiO ₂ so as to leave the SiO ₂ film 23 in the region to form the movable electrode 6 The film 23 is partially removed by etching to leave the SiO 2 left on the insulating film 9.
A sacrificial layer 24 for forming the gap 15 is formed between the insulating film 9 and the movable electrode 6, the arm 2 or the like by the _two films 23 (FIG. 4F). Once the sacrificial layer 24 is formed, FIG.
As shown in FIG. 4G, a polycrystalline polysilicon layer 25 is deposited on the upper surface of the sacrificial layer 24 and the upper surface of the insulating layer 9 exposed from the sacrificial layer 24, and the polysilicon layer 25 is formed as shown in FIG. A metal thin film 26 is formed by depositing Al on the surface of 25. Thereafter, the metal thin film 26 is patterned by etching to form the rectangular pulse-shaped metal electrode film 7b of the movable electrode 6, the connection wiring 13 and the input terminal 10 (FIG. 4).
(I)). Then, the polysilicon layer 25 is etched to integrally form the structure 7a of the movable electrode 6, the arm 2, the support 4, and the weight portion 3 (FIG. 4 (j)). Finally, as shown in FIG. 4K, the sacrificial layer 24 is removed by etching to form a gap 15 between the insulating layer 9 and the movable electrode 6, the arm 2, etc., and the acceleration sensor B1 as shown in FIG. To create. In this way, the acceleration sensor B1 having the structure shown in FIG. 1 can be easily manufactured by using the surface micromachining technique.

In the above manufacturing method, the depression 15
The etching step and the step of opening an opening in the insulating film 9 are omitted (the same applies hereinafter).

5A to 5G are schematic sectional views showing a method of manufacturing the acceleration sensor B2 according to another embodiment of the present invention. The acceleration sensor B2 manufactured by this manufacturing method is one in which the fixed electrode 8, the movable electrode 6, etc. are formed by a conductive layer made of impurities without forming the fixed electrode 8, the movable electrode 6, etc. by using a metal thin film. .
Hereinafter, this embodiment will be described with reference to FIGS. First, a mask formed in the surface of the single crystal silicon substrate 1 prepared as shown in FIG. 5A in a region where the fixed electrode 8, the connection wiring 14 and the output terminal 11 are to be formed, according to the pattern. Alternatively, a passivation film (not shown) is used to form an n ⁺ or p ⁺ conductive layer 27 by ion implantation or impurity diffusion, and the conductive layer 27 is applied to the surface layer portion of the silicon substrate 1 as shown in FIG. 5B. The fixed electrode 8, the connection wiring 14 and the output terminal 11 are formed.
After removing the mask or passivation film, FIG.
As shown in (c), SiO 2 is formed on the entire surface of the silicon substrate 1.
_{The 2} film 23 is deposited, and the SiO ₂ film 23 is partially etched away as shown in FIG. 5D to form the sacrificial layer 24. Next, as shown in FIG. 5E, a conductive polysilicon layer 28 doped with an n-type or p-type impurity is formed on the surface of the silicon substrate 21 from above the sacrificial layer 24, and then, as shown in FIG. ), The polysilicon layer 28 is patterned by etching to form the movable electrode 6, the arm 2, the support 4, the connection wiring 13, the input terminal 10 and the weight portion 3 which have a rectangular pulse shape by the polysilicon layer 28. . Finally, the sacrificial layer 24 is removed by etching to lift the arm 2, the weight portion 3, and the movable electrode 6 from the surface of the silicon substrate 21 to manufacture an acceleration sensor B2 as shown in FIG. In order to form a circuit for passing a current through the movable electrode 6, for example, two arms 2 and two support bases 4 may be provided. Alternatively, after the structure of the movable electrode 6, the arm 2, the support 4 and the like are formed from the polysilicon layer 28 not doped with impurities, the polysilicon layer 28 is doped with impurities to render the movable electrode 6 conductive. At the same time, the connection wiring 13 and the like may be circuitized. If the fixed electrode 8 and the movable electrode 6 are formed by ion implantation or impurity diffusion without using a metal thin film as in this embodiment, the manufacturing process of the acceleration sensor B2 can be reduced.
The manufacturing cost can be reduced.

In the acceleration sensor B2 as shown in FIG. 5, the insulating film 9 may be formed on the fixed electrode 8.
In the structure like the acceleration sensor B2, if the insulating film 9 is provided on the fixed electrode 8, it is possible to reduce the friction when the movable electrode 6 bends due to its own weight and touches the silicon substrate 1, and at the same time, the movable electrode 6 and the fixed electrode can be reduced. It is possible to prevent an electrical short-circuit accident between 8 and. FIGS. 6A and 6B are schematic sectional views of two types of acceleration sensors B3 and B4 having a structure for this purpose. In the acceleration sensor B3 of FIG. 6 (a),
After forming the fixed electrode 8 or the like made of the conductive layer 27 by ion implantation or impurity diffusion in the surface layer portion of the silicon substrate 1, carbon (C) is ion-implanted in the surface layer portion of the silicon substrate 1 shallower than the conductive layer 27, An insulating film 35 is formed on the surface layer of the silicon substrate 1. In the acceleration sensor B4 shown in FIG. 6B, after the fixed electrode 8 and the like made of the conductive layer 27 are formed on the surface layer portion of the silicon substrate 1 by ion implantation, impurity diffusion, etc., Si is formed on the surface of the silicon substrate 1 by the ion assist method. And C are simultaneously deposited to form the insulating film 36. A SiN film may be used as the insulating film 36.

Further, FIG. 7 is a partially broken sectional view of another acceleration sensor B5, and an insulating film 9 is formed from the lower surface of the movable electrode 6 to the lower surface of the arm 2 as shown in FIG. It was done. This acceleration sensor B5
Is a SiN passivation film 29 on the upper surface of the silicon substrate 1 on which the sacrificial layer 23 is formed as shown in FIG.
Is further formed, polysilicon is deposited on the upper surface of the passivation film 29 while doping n-type ions or p-type ions to form a polysilicon layer 28, and then the deposited passivation film 29 and polysilicon layer 28 are deposited.
After patterning, the sacrificial layer 24 can be removed by etching.

FIGS. 8A to 8G show a method of manufacturing the acceleration sensor B6 which is another embodiment. First, as shown in FIG. 8A, a silicon substrate 31 made of single crystal silicon is prepared, and an ion implantation method or impurities is applied to a region of the surface of the silicon substrate 31 where the fixed electrode 8, the connection wiring 14 and the output terminal 11 are formed. The n ⁺ type or p ⁺ type conductive layer 32 is formed relatively deeply by the diffusion method (FIG. 8B). Next, as shown in FIG. 8C, the conductive layer 32 is formed in a region wider than the region where the movable electrode 6, the arm 2 and the like are to be formed.
A p-type or n-type conductive region 33 having a conductivity type opposite to that of the above is formed, and an n ⁺ type or p ⁺ type region is formed in the conductive region 33 by ion implantation or the like by ion implantation or the like. A mold conductive layer 34 is formed (FIG. 8D). Then, a passivation film (not shown) is formed on the upper surface of the silicon substrate 31, and the passivation film is selectively etched so that it is left only on the upper surface of the conductive layer 34 to expose the conductive region 33 from the passivation film. After that, when a current is passed between the silicon substrate 31 and the counter electrode (not shown) in an aqueous solution of HF or HF + NH ₄ F, the exposed conductive region 33 is made porous, and FIG.
A porous silicon region 35 is formed as shown in FIG. Next, when the silicon substrate 31 is heat-treated, the porous silicon in the porous silicon region 35 is thermally oxidized to form a silicon oxide region (sacrificial layer) 36 (FIG. 8).
(F)). Finally, by removing the silicon oxide region 36 by etching, the movable electrode 6, the arm 2 and the weight portion 3 are integrally formed by the conductive layer 34, and the fixed electrode 8, the connection wiring 14, etc. are formed by the conductive layer 32. Thus, the gap 15 is formed on the lower surface of the movable electrode 6, the arm 2, etc., and the acceleration sensor B6 is manufactured. According to this method, an acceleration sensor can be manufactured from one silicon substrate 31 without forming a polysilicon layer or the like on the silicon substrate 31 (1).

9 (a) to 9 (l) are sectional views showing a further method of manufacturing the acceleration sensor B7 of the present invention. In this acceleration sensor B7, as shown in FIG. 9 (l), a glass cover 40 is overlaid on a rectangular frame-shaped silicon substrate 1 (41), and the peripheral portion of the glass cover 40 is bonded to the silicon substrate 1. One end of an arm 2 is supported on the silicon substrate 1 so as to be swingable in the horizontal direction with respect to the silicon substrate 1.
A flat weight portion 3 is supported on the other end of the. Further, a movable electrode 6 having a rectangular pulse shape is provided near the tip of the arm 2.
Is supported, and the movable electrode 6 is a structure 7a made of single crystal silicon.
The doping layer (conductive layer) 7c is formed on the upper surface of the. The movable electrode 6 is formed on the inner surface of the glass cover 40.
A fixed electrode 8 made of a metal thin film of Al is provided so as to face with.

Next, a method of manufacturing the acceleration sensor B7 will be described with reference to FIGS. First, as shown in FIG. 9A, a silicon substrate 41 made of single crystal silicon is prepared, and an inner peripheral region of the silicon substrate 41 is etched to form a shallow recess 42 (FIG. 9B). This recess 42 is a gap 15 between the movable electrode 6 and the fixed electrode 8.
Will be formed. Then, as shown in FIG. 9C, the region of the inner surface of the recess 42 where the movable electrode 6, the arm 2 and the weight portion 3 are formed is doped with an n-type or p-type impurity to form an n + or p + type. The conductive region 43 is formed. SiN passivation film 4 is formed on the surface of silicon substrate 41.
4 is formed (FIG. 9D), and the passivation film 44 is patterned so as to be left only on the upper surfaces of the movable electrode 6, the arm 2 and the weight portion 3 and the upper surface of the outer peripheral region of the silicon substrate 41 (FIG. 9E). ). After this, the passivation film 4
4 is vertically etched by a reactive ion etching (RIE) method or the like in the region exposed from 4 to form a trench 45 in a region other than the conductive region 43 and the outer peripheral region of the silicon substrate 41 (FIG. 9F). ).
At this time, in the rectangular pulse-shaped pattern portion which becomes the movable electrode 6, as shown in FIG. 10, {110} of the silicon substrate 41 is formed.
It is patterned in a rectangular pulse shape along the plane direction.
Then, as shown in FIG. 10, at each inner corner of the rectangular pulse-shaped pattern or the like which becomes the movable electrode 6, about 1 / m of the movable electrode width w at an angle of about 45 degrees with respect to the {110} plane orientation.
A slit-shaped correction pattern 46 is opened in the passivation film 44 so as to extend over the region 2. Then, the silicon substrate 41 is anisotropically etched to form the floating structure 47 such as the movable electrode 6, the arm 2, and the weight portion 3 (FIG. 9G). At this time, the floating structure 47 such as the movable electrode 6, the arm 2, the weight portion 3 and the like is formed into an inverted triangular cross section by anisotropic etching, and in the inner corner portion where the correction pattern 46 is provided, the portion is corrected. A square notch 48 (FIG. 10) is etched through the pattern 46.
Is indicated by a chain double-dashed line). In this case, unless the correction pattern 46 is provided, the floating structure 47 cannot be formed at the inner corner portion of the rectangular pulse-shaped pattern or the like that becomes the movable electrode 6 because it cannot be separated from the underlying silicon substrate 41. Therefore, the movable pattern 6, the arm 2 and the like can be separated from the underlying silicon substrate 41 by providing the correction pattern 46 to remove the inner corner portion by etching and providing the notch 48, and the floating structure 47. Can be formed by etching from one side. Then, the passivation film 44 on the upper surface of the movable electrode 6 and the like is removed to obtain a silicon substrate 41 having a desired structure (FIG. 9 (h)).

On the other hand, a thin glass cover 40 is prepared (FIG. 9 (i)), and a metal thin film 49 of Al is vapor-deposited on the surface of the glass cover 40 as shown in FIG. 9 (j). 49 is patterned to form the fixed electrode 8, the connection wiring 14, etc. (FIG. 9K).

Thereafter, the silicon substrate 41 and the glass cover 40 are overlapped with each other so that the movable electrode 6 and the fixed electrode 8 are opposed to each other, and the upper surface of the peripheral portion of the silicon substrate 41 is covered with the glass cover 4.
The acceleration sensor B7 can be manufactured by anodic bonding to 0.

In such an acceleration sensor B7,
Since the airtight structure can be easily achieved, the space in which the movable electrode 6 is housed can be evacuated or filled with nitrogen, and the environment resistance to exhaust gas of automobiles can be improved.

FIG. 11 is a partially cutaway side view of an acceleration sensor B8 according to another embodiment. In the acceleration sensor B8, the support film 52 having tensile stress (initial stress) is formed on the upper surface of the movable electrode portion 6. For example, the support film 52 is formed by depositing SiN on the upper surface of the metal electrode film 7b on the structure 7a made of polysilicon. When the structure 7a of the movable electrode 6 is formed of polysilicon, the movable electrode 6 is easily bent downward due to its own weight, and a short circuit accident occurs by contact with the fixed electrode 8 or the movement of the movable electrode 6 is hindered. However, by generating tensile stress on the upper surface of the movable electrode 6 by the support film 52 on the upper surface of the movable electrode 6, it is possible to bend the free end portion of the movable electrode 6 so as to pull it upwards. It is possible to prevent contact with the fixed electrode 8. Further, by forming the support film 52 also in the tip end region of the arm 2 supporting the weight portion 3, the weight of the weight portion 3 causes the arm 2 to bend downward, so that the weight portion 3 and the bottom surface of the recess 5 are formed. Can also be prevented.

Further, like the acceleration sensor B9 shown in FIG. 12, an insulating support film 53 having a compressive stress (initial stress) is provided on the lower surface of the movable electrode 6 formed by introducing impurities into the structure 7a, The free end of the movable electrode 6 may be prevented from bending downward.

FIG. 13A is a perspective view of an acceleration sensor B10 which is still another embodiment, and FIG.
It is sectional drawing which cut | disconnected by the X3-X3 line of (a). The acceleration sensor B10 has almost the same structure as that of the acceleration sensor B1 of the first embodiment, and the vibration electrodes 55a and 56a are provided on both side surfaces of the weight portion 3 provided at the tip of the arm 2, and the vibration electrode B10 is provided. Vibration electrodes 55b and 56b are provided on both inner wall surfaces of the recess 5 so as to face the 55a and 56a, respectively. Vibration electrodes 55a, 5 on both sides of the weight portion 3
Each of 6a is connected to a vibration input terminal 57a provided on the silicon substrate 1. Further, the vibration electrodes 55b and 56b in the recess 5 are connected to the vibration input terminal 57b provided on the silicon substrate 1. Then, by passing a constant current I1 through the movable electrode 6 and applying an appropriate AC voltage or pulse voltage between the vibration input terminals 57a and 57b, the weight portion 3 has a constant frequency f in the X-axis direction.
It is forced to vibrate at ₁ . The frequency f ₁ is the output frequency f of the induced current I2 generated in the fixed electrode 8.
_It is preferable to set ₀ so that it becomes equal to the resonance frequency of the signal processing circuit. As a method for vibrating the weight portion 3, for example, the vibrating electrodes 55a and 56a are kept at a constant positive potential or a negative potential, and the vibrating electrode 55 is
It is advisable to apply an alternating voltage to b and 56b with a phase difference of 180 °. Alternatively, the vibration electrodes 55a and 56a
It is also possible to provide two vibration input terminals 57a so that each can independently apply a voltage, and to alternately generate electrostatic attraction between the vibration electrodes 55a and 55b and between the vibration electrodes 56a and 56b. .

However, when acceleration or vibration is applied to the acceleration sensor B10, the vibration of the movable electrode 6 changes, and the frequency of the induced current I2 flowing through the fixed electrode 8 is the normal frequency f ₀ (for example, a signal processing circuit). (Resonance frequency of), acceleration and vibration can be detected from this frequency change.

Further, an acceleration sensor having a similar structure to the acceleration sensor B10 shown in FIG. 13 can be used to form an acceleration sensor of another principle. For example, the servo type semiconductor acceleration sensor B11 can be obtained as follows. FIG. 14 is a control block diagram of the acceleration sensor B11. 111 is a power supply circuit for flowing an alternating current I1 having a constant frequency f _{0 in} the movable electrode 6, and 112 is a signal for detecting an induced current I2 flowing in the fixed electrode 8. A processing circuit 113 is a weight portion drive circuit for applying a voltage to the vibration electrodes 55a, 55b, 56a, 56b to vibrate the weight portion 3. When acceleration is applied to the acceleration sensor B11, the output frequency f ₀ of the induced current I2 is changed, the change of the output frequency f ₀ is detected at the signal processing circuit 112. When the signal processing circuit 112 detects a change in the output frequency, the signal processing circuit 112 outputs a feedback signal to the weight drive circuit 113 so as to cancel the change in the output frequency. Then, the vibration electrode 5
5a, 55b and 56a, 56b are applied by an electrostatic force to the weight portion 3 so that the output frequency from the fixed electrode 8 becomes equal to the original frequency f ₀ (that is, the movable electrode 6 is displaced). Control). That is, the output frequency output from the fixed electrode 8 is feedback-controlled so as to always have a constant value f ₀ even if the acceleration changes. On the other hand, the signal processing circuit 112 obtains the acceleration from the feedback signal output to the weight drive circuit 113. With this method, the weight 3 is hardly displaced, and the arm 2 is less likely to be damaged by elastic fatigue.

Further, in the acceleration sensor of FIG. 13, the alternating current I1 is made to flow in the movable current 6 and the acceleration is detected from the change of the induced current I2 flowing in the fixed electrode 8, but the weight portion 3 is fixed. The acceleration may be detected by vibrating at the frequency and detecting the frequency of the change in the capacitance between the movable electrode 6 and the fixed electrode 8.
Further, in the servo type semiconductor acceleration sensor, the feedback signal may be output so that the frequency of capacitance change between the movable electrode and the fixed electrode becomes constant.

FIG. 15A shows another acceleration sensor B.
12 is a perspective view, and FIG. 15 (b) is X4-X in FIG. 15 (a).
It is sectional drawing in which the 4 line | wire was partially broken. In the acceleration sensor B12, the base both side surfaces 58a, 58 of the arm 2 are
The heat-sensitive deformable elements 59a and 59b in the form of a thin film such as a shape memory alloy such as TiNi (Nitinol) or a bimetal are joined to b. The heat-sensitive deformable elements 59a and 59b are wiring 60
Are connected to the connection terminals 61a, 61b and 61c, 61d by an energization heating circuit (not shown) to energize between the connection terminals 61a, 61b or 61c, 61d to heat the heat-sensitive deformable elements 59a, 59b. The arm 2 can be bent rightward or leftward (X-axis direction).

The acceleration sensor B12 can also constitute a servo type acceleration sensor. That is,
An alternating current I1 having a frequency f ₀ is passed through the movable electrode 6, and an induced current I2 having a frequency f ₀ is generated in the fixed electrode 8. When acceleration acts on the acceleration sensor B12, the output frequency from the fixed electrode 8 changes, but the signal processing circuit outputs a feedback signal to the energization heating circuit so as to cancel the change in the output frequency, and the thermosensitive deformation elements 59a and 59b. The arm 2 is bent by electrically heating. Then, the signal processing circuit detects acceleration or a change in acceleration from the feedback signal at this time.

FIG. 16 is a perspective view of another acceleration sensor B13 of the present invention. This acceleration sensor B13
In this case, the base portions of the two arms 2 and 62 extending in two directions orthogonal to each other are supported in a cantilever manner by the support base 4, and the weight portions provided at the tips of the arms 2 and 62. 3, 63 are located in the depressions 5, 65 of the silicon substrate 1. A movable electrode 6 in the shape of a rectangular pulse is extended from one arm 2 in a direction orthogonal to the arm 2 so as to float above the silicon substrate 1, and from the other arm 62 in a direction orthogonal to the arm 62. A movable electrode 66 having a rectangular pulse shape is extended toward. Therefore,
Both movable electrodes 6 and 66 are also orthogonal to each other. Further, on the upper surface of the silicon substrate 1, the movable electrode 6 is provided below the movable electrode 6.
The fixed electrode 8 is provided in parallel with the movable electrode 66, and the fixed electrode 68 is provided below the movable electrode 66 in parallel with the movable electrode 66.

The movable electrodes 6 and 66 are both the connection wiring 1
3 is connected to the input terminal 10 by the input terminal 1
A constant current or an alternating current having a constant frequency can be applied to both movable electrodes 6 and 66 from 0 at the same time. The fixed electrode 8 is connected to the output terminal 11a by the connection wiring 14.
The fixed electrode 68 is connected to the output terminal 11b by the connection wiring 64, and the induced current I2a generated in the fixed electrode 8 and the induced current I2 generated in the fixed electrode 68 are connected to each other.
b can be detected individually.

Therefore, in the acceleration sensor B13, the acceleration of the silicon substrate 1 in the X-axis direction can be detected by monitoring the induced current I2a generated in the fixed electrode 8, and the acceleration of the silicon substrate 1 in the Y-axis direction is fixed. The two-dimensional acceleration sensor can be detected by monitoring the induced current I2b generated in the electrode 68 and can simultaneously detect the acceleration in the X-axis direction and the acceleration in the Y-axis direction.

Further, FIG. 17 shows an acceleration sensor B14 which is still another embodiment, FIG. 17 (a) is a perspective view thereof, and FIG. 17 (b) is a partially broken sectional view thereof. . In addition to the configuration of the embodiment of FIG. 16 described above, the acceleration sensor B14 includes the capacitance detection electrode 7 on the lower surface of the weight portion 3 (or 63).
1 is provided, another capacitance detection electrode 73 is formed on the inner surface of the recess 5 with a minute gap 72 between the capacitance detection electrode 71, and a capacitor is formed between both capacitance detection electrodes 71, 73. The capacitance forming electrodes 71 and 73 are connected to the connection wiring 74 and
A pair of output terminals 75, 75 provided on the silicon substrate 1 are connected by 74. Therefore, the acceleration in the Z-axis direction perpendicular to the silicon substrate 1 is not affected by the capacitance forming electrode 7.
Since it can be detected by the change in the electrostatic capacitance between No. 1 and No. 73, it is possible to detect 3 in the X-axis, Y-axis and Z-axis directions as a whole.
The directional acceleration can be detected at the same time.

FIG. 18 is a sectional view of an acceleration sensor B15 which is another embodiment. In the acceleration sensor B15, in addition to the configuration of the embodiment shown in FIG. 16, a strain detecting element 77 such as a piezoresistor or a strain gauge is provided on the upper surface of the tip region 76 of the arm 2. When acceleration is applied in the Z-axis direction of the acceleration sensor B15, the tip region 76 of the arm 2 bends up and down due to the displacement of the weight portion 3. Therefore, by detecting this with the strain detecting element 77, the acceleration in the Z-axis direction is detected. it can. Therefore, in this embodiment as well, as in the embodiment of FIG.
Dimensional acceleration can be detected.

FIG. 19 shows an acceleration sensor B16 which is still another embodiment of the present invention, in which movable electrodes 6a, 6b,
Figure 6 shows a plan view of the arm 2 provided with 6c. The arm 2 is made to project to the left and right sides, and the pitch d
The movable electrode 6a having a rectangular pulse shape of 1, the movable electrode 6b having a rectangular pulse shape of the pitch d2, and the movable electrode 6c having a rectangular pulse shape of the pitch d3 are arranged in parallel (where d1 <d2. <D3). Although not shown, each movable electrode 6a, 6b, 6c is formed on the surface of the silicon substrate 1.
There are provided three kinds of fixed electrodes having a rectangular pulse shape with the same pitch d1, d2, d3 facing each other, and the movable electrodes 6a, 6b, 6c and the fixed electrode having the same pitch face each other with a gap therebetween. Has been made.

If a plurality of movable electrodes 6a, 6b, 6c and fixed electrodes having different pitches are provided in this way, when the acceleration sensor B16 receives acceleration, output signals of different frequencies are output from each fixed electrode. Therefore, it is possible to compensate for different frequency characteristics and stabilize the detection accuracy, and it is possible to detect a wide range of acceleration compared to the frequency band of the frequency detection circuit in the signal processing circuit. Can be widened.

FIG. 20 is a perspective view of an acceleration sensor B17 which is still another embodiment of the present invention. In the acceleration sensor B17, the rectangular pulse-shaped movable electrode 6 provided on the arm 2 is connected to the terminal 80 by the connection wiring 82, and the rectangular pulse-shaped movable electrode 6 facing the movable electrode 6 is provided on the upper surface of the silicon substrate 1. The fixed electrode 8 is connected to the terminal 81 by the connection wiring 83. Then, when the arm 2 is bent by the acceleration in the X-axis direction and the movable electrode 6 moves in the X-axis direction, the electrostatic capacitance C between the movable electrode 6 and the fixed electrode 8 changes as shown in FIG. That is, the movable electrode 6
Is displaced by that one pitch d, the electrostatic capacitance is also one wavelength Δ
It changes by t. Therefore, the change in acceleration can be detected by counting the wave number of the change in the capacitance C.

Since the conventional electrostatic capacitance detection method utilizes the fact that the distance between two flat electrodes is changed, air damping occurs between the electrodes and the detection accuracy of the acceleration sensor deteriorates. However, in this acceleration sensor B17, since the movable electrode 6 and the fixed electrode 8 move in parallel, there is no air damping (the air damping between the weight portion 3 and the recess 5 can be reduced by widening the recess 5), and the acceleration The detection accuracy can be improved. Further, although the conventional capacitance type acceleration sensor provides an analog output, the acceleration sensor B17 provides a digital output, so that the detection accuracy becomes higher. When the movable electrode is displaced in the Z-axis direction, the electrostatic capacitance C between the movable electrode 6 and the fixed electrode 8 changes, but the acceleration in the X-axis direction is detected not by the electrostatic capacitance change but by the wave number of the change. Therefore, the detection accuracy in the X-axis direction is not affected by the vibration in the Z-axis direction.

FIG. 22 (a) is a perspective view showing a main portion of an acceleration sensor B18 according to still another embodiment of the present invention, FIG.
22B is a sectional view taken along line X5-X5 of FIG. In the acceleration sensor B18, the silicon substrate 9
One end of a movable electrode 93 having a mesh structure having a constant mesh pitch d / 2 is supported on a support base 92 on the first movable electrode 93. The movable electrode 93 is separated from a fixed electrode 95 on a silicon substrate 91 by a minute gap 98. ing. A flat weight 94 is attached to the other end of the movable electrode 93 in parallel with the silicon substrate 91.

The mesh structure 96a of the movable electrode 93 is integrally formed with the support base 92 and the weight portion 94 from polycrystalline silicon, polyimide resin, single crystal silicon or the like.
The mesh structure 96a of the movable electrode 93 can be expanded and contracted by deforming the mesh structure.
When acceleration is applied to the mesh structure 96a, the weight 94 is displaced by expanding and contracting the mesh structure 96a. In addition, as shown in FIG. 23, the movable electrode 93 is formed by patterning a metal thin film such as Al on the surface of the mesh structure 96a or by doping the mesh structure 96a made of silicon with impurities. The conductive portions 96b having a rectangular pulse pattern with a constant pitch are formed, and the conductive portions 96b are provided on the silicon substrate 91 via the support base 92 and the connection wiring 97 provided on the surface of the silicon substrate 91. It is connected to an input terminal (not shown).
On the upper surface of the silicon substrate 91, a fixed electrode 95 having a rectangular pulse pattern with a constant pitch d is formed of a metal thin film such as Al so as to face the movable electrode 93.
It is connected to a pair of output terminals (not shown) provided on the silicon substrate 91.

Even in the acceleration sensor B18, however, for example, a constant DC current or an alternating current having a constant frequency is supplied to the movable electrode 93 through the input terminal. Then, acceleration is applied in the X-axis direction of FIG.
Moves in the X-axis direction, the conductive portion 96b of the movable electrode 93 is moved.
Is displaced across the fixed electrode in the X-axis direction, the pulse number or frequency change of the induced current generated in the fixed electrode 94 can be detected, and the acceleration in the X-axis direction can be detected. Alternatively, it is also possible to employ the acceleration detection method as described above, such as detecting the acceleration from the change in the electrostatic capacitance between the fixed electrode 94 and the movable electrode 93.

FIG. 24 is a schematic plan view showing an acceleration sensor B19 which is still another embodiment of the present invention, in which movable electrodes 93a and 93b, a weight portion 94 and the like (the silicon substrate 91 is omitted). It represents. In this acceleration sensor B19, movable electrodes 93a and 93b in which conductive portions 96b are formed on a mesh-like structure 96a having a constant mesh pitch d / 2 on both left and right sides of a flat weight 94 are formed on a silicon substrate 91. The movable electrodes 93a and 93b are arranged with a constant gap 98 therebetween, and the base ends of the movable electrodes 93a and 93b are supported by the support bases 92a and 92b.
And the free ends of the movable electrodes 93a and 93b are connected to the weight portion 94. The movable electrodes 93a and 93b supported by the two supports 92a and 92b can freely expand and contract due to their elastic deformation, and when the weight portion 94 senses acceleration,
The weight 94 and the weight 94 are displaced in the X-axis direction. On the upper surface of the silicon substrate 91, two fixed electrodes (not shown) having a rectangular pulse pattern are provided as movable electrodes 93a and 93, respectively.
It is provided so as to face b.

Therefore, even the acceleration sensor B19 having such a structure can detect the acceleration in the X-axis direction, but the weight portion 94 is further provided with the movable electrode 93 from both sides.
Since it is supported by a and 93b, the weight portion 94 is less likely to be displaced in the direction orthogonal to the X-axis direction, and the sensitivity of other axes other than the X-axis direction can be reduced. Further, since the outputs can be obtained from the fixed electrodes on both sides, the detection sensitivity of the acceleration sensor B19 can be further increased.

In the embodiment of FIG. 24, the movable electrodes 93a and 93b arranged on both sides of the weight portion 94 are formed by forming a conductive portion 96b formed of a metal thin film or a doping region on the mesh structure 96a. There is a weight 94
May be supported by the movable electrode 93a and the mesh structure 96a. That is, the weight portion 94 is supported by the two mesh structures 96a, 96a, and the conductive portion 96b is provided on one of the mesh structures 96a to form the movable electrode 9.
3a, and the other may be left as the mesh structure 96a without providing the conductive portion 96b. Therefore, in this case, there is only one fixed electrode.

FIG. 25 shows a plan view of an acceleration sensor B20 which is another embodiment. This acceleration sensor B20
In this case, the recessed portion 101 for disposing the weight portion 94 and the movable electrodes 93a, 93b, 93c, 93d is formed on the silicon substrate 91, and the support portion 92 is provided around the recessed portion 101.
Is provided. A weight 94 is provided in the center of the recess 101.
Are arranged, and the movable electrodes 93a, 93b, 93c, 93 in which conductive parts 96b are provided on the mesh structure 96a are arranged on the four sides.
and the movable electrodes 93a, 93b, 93c, 9 are arranged.
The base end portion of 3d is supported by the support portion 92, and the free ends of the movable electrodes 93a, 93b, 93c, and 93d are the weight portion 9.
4 are integrally connected to each side surface in four directions. The movable electrodes 93a, 93b, 93 are provided on the silicon substrate 91.
Four fixed electrodes each having a rectangular pulse pattern are formed so as to face c and 93d.

Thus, by providing the movable electrodes 93a, 93b, 93c, 93d in the X-axis direction and the Y-axis direction in this way, it is possible to detect a two-dimensional acceleration in the X-axis and Y-axis directions. The sensor B20 can be manufactured.

Note that any one movable electrode may be eliminated from the acceleration sensor having the structure shown in FIG. 25, and three movable electrodes may be used. Further, there is no problem in obtaining the two-dimensional acceleration sensor even if only one of the mesh-like structures arranged on both sides of the weight portion is used as the movable electrode.

FIG. 26 is a partially cutaway plan view showing an acceleration sensor B21 according to another embodiment. In this acceleration sensor B21, one side surface of the weight portion 94 is attached to the support base 9
The vibrating electrode 102 is formed on the other side surface 99 of the weight portion 94 while being supported by the movable electrode 93 supported by the second electrode. Another vibrating electrode 103 is formed on the side surface 105 of another supporting base 104 on the silicon substrate 91 so as to face the vibrating electrode 102 of the weight portion 94. Movable electrode 93
A DC constant current I1 is flowing through the
An AC voltage having a constant frequency f ₁ is applied between 2 and 103, and when acceleration is not applied, the weight portion 94 vibrates at a frequency f ₁ and the fixed electrode 95 has a constant frequency f ₀ . Induction current I2 is flowing.

However, when acceleration is applied to the acceleration sensor B21, the frequency of the induced current I2 flowing through the fixed electrode changes to f = f ₀ + Δf according to the magnitude of the acceleration applied to the acceleration sensor B21. Therefore, the magnitude of acceleration can be known by detecting the frequency change of the induced current I2 flowing through the fixed electrode.

Further, as another acceleration detection method of the acceleration sensor B21 having the structure shown in FIG.
An alternating current I1 having a constant frequency f ₀ is caused to flow through the fixed electrode 3, and an induction current I2 having the same frequency f ₀ is detected by the fixed electrode. When the frequency of the induced current I2 flowing through the fixed electrode changes due to acceleration, the vibration electrode 10 is canceled so as to cancel the frequency change and keep the frequency constant.
It is also possible to output a feedback signal to a driving power supply that applies an AC voltage between 2 and 103 and detect the acceleration from this feedback signal.

FIG. 27 is a partially broken plan view showing an acceleration sensor B22 which is still another embodiment of the present invention.
In this embodiment, the side surfaces of the weight portion 94 in the X-axis direction and the Y-axis direction are supported by the movable electrodes 93a and 93b, and the movable electrodes 93a and 93b are opposite to the connected side surfaces 99a and 99b. The vibration electrodes 102
a and 102b are provided. In addition, the side surface 1 of the support portion 92
Also on 05a and 105b, vibration electrodes 103a and 103b are formed so as to face the vibration electrodes 102a and 102b. Such an acceleration sensor B22 is a two-dimensional acceleration sensor that can simultaneously detect accelerations in the X and Y axis directions by the same detection method as the acceleration sensor B21 of FIG.

FIG. 28 is a partially cutaway plan view of an acceleration sensor B23 which is still another embodiment. The acceleration sensor B23 has substantially the same structure as that of the acceleration sensor B19 shown in FIG. 24, and mesh-like mesh-like structures 96a and 96a are formed on both left and right sides of a flat plate-like weight portion 94 so as to be expandable and contractible. A movable electrode 93 having a conductive portion 96b having a rectangular pulse-shaped wiring pattern is formed on the mesh structure 96a on the right side of the weight portion 94, and the movable electrode 93 is formed on the silicon substrate 91 so as to face the movable electrode 93. A fixed electrode (not shown) is formed on the wiring pattern.

FIG. 29 (a) is a plan view showing the actuator 106 of the acceleration sensor B23, and FIG. 29 (b) is a partially enlarged plan view. In the acceleration sensor B23, the mesh structure 96a is provided on one side surface of the weight portion 94.
A movable electrode 93 having a conductive portion 96b formed therein is connected thereto, and an actuator 106 having a mesh structure as shown in FIGS. 29A and 29B is provided on the other side surface of the weight portion 94. In this actuator 106, two conductive parts 107a and 107b are formed on a net-like structure 96a similar to that used for the movable electrode 93 by a doping layer and a metal thin film such as Al.
Is formed. As shown in FIG. 29 (a),
Each of the two conductive portions 107a and 107b has a comb-shaped pattern shape and is meshed from both sides of the mesh structure 96a so as not to contact each other. Moreover, each of the conductive portions 107a and 107b is Reticulated structure 9
As shown in FIG. 29 (b), it is bent zigzag along the line 6a in a triangular wave shape.

The conductive portions 107a and 107b are connected to connection terminals (not shown) on the silicon substrate 91 by connection wirings 108 and 108, respectively. One conductive portion, for example 107a, is held at a constant potential (positive potential or negative potential), and the oscillator is connected to the other conductive portion 107b. Therefore, the conductive portion 107a,
Electrostatic attraction force and repulsive force alternately act on the portions of 107b that are substantially parallel to each other, and the actuator 106, which is the mesh structure 96a, contracts or extends, causing stretching vibration at the same frequency as the output frequency of the oscillator. However, the weight portion 94 can be forcibly displaced in the X-axis direction.

Therefore, in the acceleration sensor B23, the actuator 106 can be used similarly to the pair of vibration electrodes 102 and 103 in the acceleration sensor B21 of FIG. That is, the constant current I1 is made to flow through the movable electrode 93, the weight portion 94 is vibrated by the actuator 106 at a constant frequency, the induced current I2 flowing in the fixed current is detected, and the induced current when acceleration is applied to the weight portion 94 is detected. The acceleration can be obtained from the frequency change of I2. Alternatively, a constant current I1 is passed through the movable electrode 93a, and a force is applied to the weight portion 94 by the actuator 106 so that the cycle of the induced current I2 output from the fixed electrode does not change even when acceleration is applied to induce the fixed electrode. It is also possible to detect the acceleration from a feedback signal input to the actuator 106 based on the current I2.

In the acceleration sensor B22 having the structure as shown in FIG. 27, there are two vibration electrodes 102a and 102a.
If each of b, 103a, and 103b is replaced with the actuator 106 as described above, the actuator 106
It is possible to obtain an acceleration sensor that is capable of detecting acceleration in a two-dimensional direction using.

[Brief description of drawings]

1A is a perspective view showing an acceleration sensor according to an embodiment of the present invention, FIG. 1B is a sectional view taken along line X1-X1 in FIG. 1A, and FIG. 1C is X2-X2 in FIG. It is sectional drawing in which the wire was partially broken.

FIG. 2 is an explanatory diagram showing how an induced current is generated in the acceleration sensor of the above.

3A is a diagram showing an induced current in a state where no acceleration is applied, and FIG. 3B is a diagram showing an induced current in a state where acceleration is applied.

4A to 4K are cross-sectional views showing a method of manufacturing the acceleration sensor of the above.

5A to 5G are sectional views showing a method of manufacturing an acceleration sensor according to another embodiment of the present invention.

FIG. 6A is a sectional view showing an acceleration sensor according to yet another embodiment of the present invention, and FIG. 6B is a sectional view showing an acceleration sensor according to yet another embodiment.

FIG. 7 is a sectional view showing an acceleration sensor according to still another embodiment of the present invention.

8A to 8G are cross-sectional views showing a method of manufacturing an acceleration sensor which is still another embodiment of the present invention.

9A to 9L are cross-sectional views showing a method for manufacturing an acceleration sensor which is still another embodiment of the present invention.

FIG. 10 is a diagram showing a mask pattern in the above manufacturing method.

FIG. 11 is a partially cutaway sectional view showing an acceleration sensor according to still another embodiment of the present invention.

FIG. 12 is a partially cutaway sectional view showing an acceleration sensor according to still another embodiment of the present invention.

13A is a perspective view showing an acceleration sensor according to still another embodiment of the present invention, and FIG. 13B is a cross-sectional view taken along line X3-X3 of FIG. 13A.
It is a line sectional view.

FIG. 14 is a schematic block diagram showing an acceleration sensor according to still another embodiment of the present invention.

15A is a perspective view showing an acceleration sensor according to still another embodiment of the present invention, and FIG. 15B is a sectional view taken along line X4-X4 of FIG.
It is a line sectional view.

FIG. 16 is a perspective view showing an acceleration sensor which is still another embodiment of the present invention.

17A is a perspective view showing an acceleration sensor according to still another embodiment of the present invention, and FIG. 17B is a partially cutaway sectional view thereof.

FIG. 18 is a partially cutaway sectional view showing an acceleration sensor according to still another embodiment of the present invention.

FIG. 19 is a plan view showing a main part of an acceleration sensor provided with a plurality of movable electrodes according to still another embodiment of the present invention.

FIG. 20 is a perspective view showing an acceleration sensor according to still another embodiment of the present invention.

FIG. 21 is a diagram showing a change in electrostatic capacitance of a capacitor included in the above acceleration sensor.

22 (a) is a partially cutaway perspective view showing an acceleration sensor according to still another embodiment of the present invention, and FIG. 22 (b) is (a).
6 is a cross-sectional view taken along line X5-X5 in FIG.

FIG. 23 is a diagram showing a wiring pattern of a movable electrode of the above acceleration sensor.

FIG. 24 is a partially omitted plan view of an acceleration sensor according to still another embodiment of the present invention.

FIG. 25 is a plan view in which an acceleration sensor according to still another embodiment of the present invention is partially omitted.

FIG. 26 is a partially omitted plan view of an acceleration sensor according to still another embodiment of the present invention.

FIG. 27 is a partially omitted plan view of an acceleration sensor according to still another embodiment of the present invention.

FIG. 28 is a partially omitted plan view of an acceleration sensor according to yet another embodiment of the present invention.

FIG. 29 (a) is a diagram showing a wiring pattern of a movable electrode of the above acceleration sensor, and FIG. 29 (b) is an enlarged diagram showing a part thereof.

[Explanation of symbols]

2 arm 3 weight part 6 movable electrode 9 insulating film 15 gap 24 SiO ₂ film 45 trenches 52, 53 support film 59a, 59b thermosensitive deformation element 66 movable electrode 68 fixed electrode 92 support base 93, 93a, 93b, 93c, 93d movable electrode 95 fixed electrode 102, 103 vibration electrode 106 actuator

Claims (33)

[Claims] 1. A fixed electrode having a substantially rectangular pulse-shaped pattern with a constant pitch is provided on a substrate surface, and a movable electrode having a conductive portion having a substantially rectangular pulse-shaped pattern with a constant pitch is provided with a fixed gap with the fixed electrode. A semiconductor acceleration sensor characterized in that they are opposed to each other. 2. The semiconductor acceleration sensor according to claim 1, wherein a base of an elastic arm is supported on the substrate, and the movable electrode is provided on a tip side of the arm. 3. The semiconductor acceleration sensor according to claim 2, wherein a weight portion is provided at a tip portion of the arm. 4. A strain detecting element is provided at the base of the arm,
3. The strain detecting element is adapted to detect an acceleration component in a direction perpendicular to the substrate surface.
Alternatively, the semiconductor acceleration sensor described in 3. 5. A capacitance forming electrode is formed on a lower surface of the weight portion, another capacitance forming electrode is provided on the substrate so as to face the capacitance forming electrode, and a change in capacitance between the capacitance forming electrodes is detected. The acceleration component in the direction perpendicular to the substrate surface can be detected by performing the above.
The semiconductor acceleration sensor described in 1. 6. The semiconductor acceleration sensor according to claim 1, wherein a plurality of sets of the fixed electrode and the movable electrode having different rectangular pulse pattern pitches are provided. 7. A plurality of sets of fixed electrodes and movable electrodes are provided, and one set of fixed electrodes and movable electrodes and the other set of fixed electrodes and movable electrodes are arranged in directions orthogonal to each other. The semiconductor acceleration sensor according to claim 1, 2, 3, 4, 5 or 6. 8. The semiconductor acceleration sensor according to claim 1, wherein at least a part of the movable electrode is made of polycrystalline silicon or single crystal silicon. 9. The insulating film is formed on at least one of a region including a fixed electrode on the surface of the substrate and a surface of the movable electrode facing the fixed electrode.
The semiconductor acceleration sensor according to 2, 3, 4, 5, 6, 7 or 8. 10. The fixed electrode is formed by introducing impurities into the surface layer of the substrate made of silicon, and the insulating film is formed by introducing carbon into the surface layer of the substrate shallower than the fixed electrode. The semiconductor acceleration sensor according to claim 9, wherein. 11. The semiconductor acceleration sensor according to claim 9, wherein the insulating film is formed by depositing silicon and carbon on the surface of the substrate. 12. The support film having a compressive stress is formed on the lower surface of the movable electrode, as set forth in claim 1,
The semiconductor acceleration sensor according to 3, 4, 5, 6, 7, 8, 9, 10 or 11. 13. The support film having a tensile stress is formed on the upper surface of the movable electrode.
The semiconductor acceleration sensor according to 2, 3, 4, 5, 6, 7, 8, 9, 10 or 11. 14. The semiconductor acceleration sensor according to claim 13, wherein the support film is made of silicon nitride. 15. Claims 1, 2, 3, 4, 5, 6, 7,
A method for manufacturing the semiconductor acceleration sensor according to 8, 9, 10, 11, 12, 13 or 14, wherein a sacrificial layer is formed on the surface of the fixed electrode in advance, and a sacrificial layer is formed above the sacrificial layer. A movable electrode is formed by forming a structure made of crystalline silicon and forming a conductive region in the structure by a doping pattern or a metal pattern, and removing the sacrificial layer by etching between the fixed electrode and the movable electrode. A method of manufacturing a semiconductor acceleration sensor, characterized in that a gap is formed in the gap. 16. The method according to claim 1, 2, 3, 4, 5, 6, 7,
A method for manufacturing the semiconductor acceleration sensor according to 8, 9, 10, 11, 12, 13 or 14, wherein a conductive region is formed on a surface layer portion of a single crystal silicon substrate by a doping pattern or a metal pattern, An oxide film is formed by thermal oxidation of the porous silicon formed on the single crystal silicon substrate below the surface layer portion, and a sacrificial layer etching is performed on the oxide film to form a movable electrode on the surface layer portion. A method for manufacturing a semiconductor acceleration sensor, comprising forming a gap on a lower surface of a movable electrode. 17. The method according to claim 1, 2, 3, 4, 5, 6, 7,
A method for manufacturing the semiconductor acceleration sensor according to 8, 9, 10, 11, 12, 13 or 14, wherein a conductive region is formed on a surface layer portion of a single crystal silicon substrate by a doping pattern or a metal pattern, It is possible to form a movable electrode by trench-etching the single-crystal silicon substrate around a predetermined region including a conductive region, and then anisotropically etching the lower side of the predetermined region to form a gap on the lower surface of the movable electrode. A method for manufacturing a characteristic semiconductor acceleration sensor. 18. A fixed electrode having a substantially rectangular pulse-shaped pattern with a constant pitch is provided on a substrate surface, and a movable electrode having a stretchable mesh-like structure having a conductive portion having a substantially rectangular pulse-shaped pattern with a constant pitch is used as the fixed electrode. 3. The movable electrode is arranged to face each other with a constant gap therebetween, supports one end of the movable electrode, and connects a weight portion to the other free end of the movable electrode.
The semiconductor acceleration sensor described in 1. 19. A pair of stretchable mesh-like structures supported at one end are arranged on both sides of the weight portion, and free ends of each mesh-like structure are connected to both sides of the weight portion, and at least one mesh-like shape is formed. A fixed electrode having a substantially rectangular pulse-shaped pattern with a constant pitch on the surface of the substrate with a constant gap provided under the movable electrode to form a movable electrode by providing a conductive portion having a substantially-rectangular pulse-shaped pattern with a constant pitch on the structure. The semiconductor acceleration sensor according to claim 1, further comprising: 20. A pair of expandable and retractable mesh-like structures, one end of which is supported, are arranged on three sides or four sides of the weight portion, and the free ends of each mesh-like structure are connected to three or four outer surfaces of the weight portion, At least one of the mesh structures arranged in one direction is provided with a conductive portion having a substantially rectangular pattern with a constant pitch to form a movable electrode, and the mesh structure arranged in the other direction. At least one of the mesh structures is also provided with a conductive portion having a substantially rectangular pattern with a constant pitch to form a movable electrode, and a substantially rectangular pulse having a constant pitch on the substrate surface with a constant gap below each movable electrode. A fixed electrode having a striped pattern is provided.
The semiconductor acceleration sensor described in 1. 21. The semiconductor acceleration sensor according to claim 19, wherein the mesh-like structures disposed on both sides of the weight portion are sandwiched between the movable electrodes. 22. At least a part of the mesh structure is composed of polycrystalline silicon or polyimide.
The semiconductor acceleration sensor described in 1. 23. Claims 1, 2, 3, 4, 5, 6, 7,
8, 9, 10, 11, 12, 13, 14, 18, 19,
An acceleration detection method using the semiconductor acceleration sensor according to 20, 21, or 22, wherein a constant current or an alternating current having a constant frequency is applied to one of the fixed electrode and the movable electrode to generate an induced electromotive force. An acceleration detection method using a semiconductor acceleration sensor, wherein acceleration is detected by detecting a change in an induced current generated in one of the other electrodes. 24. The acceleration detection method according to claim 23, further comprising a current-voltage conversion means for converting the induced current into a pulse voltage and outputting the pulse voltage. 25. Claims 1, 2, 3, 4, 5, 6, 7,
8, 9, 10, 11, 12, 13, 14, 18, 19,
An acceleration detection method using the semiconductor acceleration sensor according to 20, 21, or 22, wherein a direction parallel to the substrate surface is detected by detecting a change in capacitance between the movable electrode and the fixed electrode. Acceleration detection method using a semiconductor acceleration sensor, which is characterized by detecting acceleration in the. 26. Claims 3, 4, 5, 6, 7, 8, 9,
An acceleration detection method using the semiconductor acceleration sensor according to 10, 11, 12, 13 or 14, wherein a vibration electrode is provided on a side surface of a weight portion provided at the tip of the arm, and an electrostatic force is applied to the vibration electrode. The weight portion is vibrated at a constant frequency by applying a constant current to one of the movable electrode and the fixed electrode to detect the frequency change of the induced current flowing to the other electrode, or An acceleration detection method using a semiconductor acceleration sensor, which is configured to detect a capacitance change between a fixed electrode and a fixed electrode. 27. Claims 3, 4, 5, 6, 7, 8, 9,
An acceleration detection method using the semiconductor acceleration sensor according to 10, 11, 12, 13 or 14, wherein a vibration electrode is provided on a side surface of a weight portion provided at a tip portion of the arm, and a movable electrode and a fixed electrode are provided. An alternating current with a constant frequency is applied to one of the electrodes to detect the frequency change of the induced current generated in the other electrode, or to detect the change in the capacitance between the movable electrode and the fixed electrode to determine the acceleration. An acceleration detection method using a semiconductor acceleration sensor in which an electrostatic force is generated in a vibration electrode so as to cancel a change in the detection signal caused by the above, and acceleration is detected from a feedback signal input to the vibration electrode. 28. Claims 2, 3, 4, 5, 6, 7, 8,
An acceleration detection method using the semiconductor acceleration sensor according to 9, 10, 11, 12, 13 or 14, wherein a heat-sensitive deformable element for vibrating the arm is attached to a side surface of the arm, and the heat-sensitive deformable element is electrically heated. Means for detecting the frequency change of the induced current generated in the other electrode by applying an alternating current having a constant frequency to one of the movable electrode and the fixed electrode, or between the movable electrode and the fixed electrode. An acceleration detection method using a semiconductor acceleration sensor that detects a change in generated capacitance, electrically heats the heat-sensitive deformable element so as to cancel a change in the detection signal caused by acceleration, and detects acceleration from the current-carrying signal. . 29. An acceleration detection method using the semiconductor acceleration sensor according to claim 21, wherein a constant current or an alternating current having a constant frequency is applied to one of the movable electrode and the fixed electrode, A semiconductor acceleration sensor in which a change in an induced current generated in one of the other electrodes due to an induced electromotive force is used as an output signal, and acceleration is detected from a difference between output signals of movable electrodes or fixed electrodes arranged on both sides of the weight portion. Acceleration detection method by. 30. An acceleration detection method using the semiconductor acceleration sensor according to claim 18, 19 or 20, wherein a vibration electrode is provided on a side surface of the weight portion opposite to a side surface connected to the movable electrode. , A fixed frequency alternating current is applied to one of the movable and fixed electrodes to detect the frequency change of the induced current generated in the other electrode, or electrostatic generated between the movable and fixed electrodes. With a semiconductor acceleration sensor that detects a change in capacitance, generates an electrostatic force in the vibration electrode so as to cancel the change in the detection signal caused by acceleration, and detects the acceleration from a feedback signal input to the vibration electrode. Acceleration detection method. 31. A movable electrode is connected to each of two side surfaces of the weight portion in a direction substantially orthogonal to each other, and a vibration electrode is provided on each side surface of the weight portion opposite to the side surface to which the movable electrode is connected. 31. The acceleration detection method by the semiconductor acceleration sensor according to claim 30, wherein the acceleration is detected. 32. An acceleration detection method using the semiconductor acceleration sensor according to claim 19 or 20, wherein a mesh-like structure connected to one side surface of the weight portion with the weight portion interposed therebetween is used as a movable electrode, The mesh-like structure connected to the other side surface is used as an actuator, and an alternating current having a constant frequency is applied to one of the movable electrode and the fixed electrode to detect a frequency change of the induced current generated in the other electrode, Alternatively, by detecting a change in capacitance generated between the movable electrode and the fixed electrode, the actuator is driven so as to cancel the change in the detection signal caused by acceleration, and the acceleration is detected from the drive signal of the actuator. Acceleration method using the semiconductor acceleration sensor. 33. The movable electrodes are respectively connected to two side surfaces of the weight portion in a direction substantially orthogonal to each other, and the actuator is connected to a side surface opposite to the side surface to which the movable electrodes of the weight portion are connected, respectively. 33. An acceleration detection method using a semiconductor acceleration sensor according to claim 32, wherein the acceleration of the sensor is detected.