JP2773698B2 - Capacitive acceleration sensor and method of manufacturing the same - Google Patents

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 for detecting vibration, impact, and the like, and a method for manufacturing the same. More particularly, the present invention relates to an electrostatic sensor comprising a movable electrode formed on a weight portion of a sensor substrate and a fixed electrode formed on an upper and lower stopper substrate. The present invention relates to a capacitance type acceleration sensor using a capacitance and a method of manufacturing the same.

[0002]

2. Description of the Related Art Conventionally, this kind of semiconductor acceleration sensor has
It is fixed to a moving body or the like and is used to measure the amount of acceleration of the moving body. Depending on the detection method, such a semiconductor acceleration sensor is classified into a piezoresistive sensor utilizing the piezoresistive effect of silicon, and an electrostatic capacitance sensor utilizing a change in capacitance.

FIG. 4 is a sectional view of a conventional piezoresistive acceleration sensor. As shown in FIG. 4, such a piezo-type sensor includes a sensor substrate 1 and a lower stopper 4 and an upper stopper 5 sandwiching the sensor substrate 1 from above and below.
The sensor substrate 1 has a rim portion 2 and a beam portion 7 and a weight portion 6 thinned by anisotropic etching. The sensor substrate 1 is formed with a groove 8 that penetrates the sensor substrate 1 by anisotropic etching or dry etching so that the weight portion 6 can be easily moved. A piezoresistive element 27 of the opposite type (for example, p-type when the substrate is n-type) is formed by diffusion or ion implantation at the edge of the portion 7.

On the other hand, upper and lower stoppers 5 and 4 adhesively fixed to the upper and lower sides of the sensor substrate 1 are provided with gaps 12 and 13 for allowing the weight portion 6 to move, and an excessive impact is applied to the sensor substrate 1. When the weight 6 touches the upper and lower stoppers 5 and 4, the beam 7
It also has the function of preventing destruction.

A pad 11 connected to the piezoresistive element 27 is formed at an end of the circuit forming surface of the sensor substrate 1.

FIG. 5 is a partially cutaway perspective view of the sensor substrate shown in FIG. As shown in FIG. 5, the piezoresistive element 27 is provided on the sensor substrate 1 at the edge of the beam portion 7 on the circuit forming surface 28, and is electrically connected to the pad 11, as described above.

In the operation of such a piezoresistive acceleration sensor, when an impact is applied to the sensor substrate 1, the edge of the beam 7 is bent by inertia at that time, and the weight 6 is displaced up and down. At this time, the resistance of the piezoresistive element 27 provided at the edge of the beam portion 7 changes under stress. By taking out the change in the resistance value through the pad 11 to the outside, the magnitude of the impact can be detected.

However, in the case of a piezoresistive acceleration sensor, a so-called piezoresistive effect is used.
It has low detection sensitivity and is not suitable for detecting low acceleration. Further, since the piezoresistive element itself has a temperature dependency, there is a problem that the temperature dependency of the sensitivity is large.

FIG. 6 is a sensitivity / temperature characteristic diagram of the sensor shown in FIG. As shown in FIG. 6, in the above-described sensor, the surface impurity concentration of the piezoresistive element 27 is set to 3 × 10 ¹⁸
[Cm / cm ³ ], and shows sensitivity temperature characteristics when 5 V constant voltage driving is performed. For example, when the sensitivity at 25 ° C. is used as a reference, the sensitivity changes by 13% at −40 ° C. and by −11% at 85 ° C.

The correction for such a change in the sensitivity of the sensor must be performed by an external peripheral circuit.

Hitherto, in order to improve the problem of the piezo sensor, a capacitance type acceleration sensor has been proposed. Hereinafter, this capacitance type acceleration sensor and a method of manufacturing the same will be described.

FIGS. 7A to 7F are cross-sectional views of a conventional capacitive acceleration sensor and a sensor shown in the order of steps for explaining a method of manufacturing the same.

First, as shown in FIG. 7 (a), an n-type impurity such as phosphorus or arsenic is applied to an n-type semiconductor substrate 30 from both sides thereof so as to be used as a vertically movable electrode and a wiring of a weight portion described later. The n ⁺ diffusion layer 31 is formed by diffusion. Next, a protective film 32 serving as a mask material for performing anisotropic etching is coated on both surfaces of the semiconductor substrate 30. Thereafter, the protective film 32 where a thin film is desired to be formed is removed by etching using a general photolithography technique to form a protective film removing portion 33. Here, as the protective film 32, a silicon oxide film or a silicon nitride film is used.

Next, as shown in FIG. 7 (b), the semiconductor substrate 30 is immersed in an anisotropic etching solution to etch silicon, thereby reducing the thickness of the beam portion 7, the thick weight portion 6, and the rim portion. Form 2 As the anisotropic etching solution, an aqueous solution of potassium hydroxide, saturated hydrazine, EDP solution, or the like is used, and a desired beam thickness is obtained by controlling the etching rate and determining the pulling time.

Next, as shown in FIG. 7C, in order to form a groove, the protective film 32 is removed from the surface opposite to the surface on which the silicon etching has been performed, and a removed portion 34 is formed. In this case, the protective film on the portion where the groove is to be formed in a part of the beam portion 7 is removed using a general photolithography technique.

Next, as shown in FIG. 7D, silicon etching is performed from the protective film removing portion 34 by anisotropic etching or dry etching similar to the aforementioned anisotropic etching, and the upper and lower surfaces are penetrated. A groove 8 is formed.

Further, as shown in FIG. 7E, after removing the protective films 32 on both surfaces, a metal pad 11 is adhered on the upper surface by a sputtering technique or the like, and the sensor substrate 1 is formed.

Finally, as shown in FIG. 7 (f), the lower stopper 4 and the upper stopper 5, which have been separately prepared, are adhered and fixed to the above-mentioned sensor substrate 1 from both the upper and lower surfaces, thereby obtaining the capacitance. Type acceleration sensor is completed. At this time, the lower stopper 4 and the upper stopper 5 sandwiching the sensor substrate 1 are respectively connected to the lower fixed electrode 9 connected to the outside via the through hole 17 and the fixed electrode connected to the outside via the through hole 16. 10 and gaps 13 and 12 are formed to allow the weight 6 of the sensor substrate 1 to move up and down. The fixed electrodes 9 and 10 have substantially the same size as the weight.
In response to these, the n ⁺ diffusion layers remaining on the upper and lower surfaces of the weight portion 6 of the sensor substrate 1 form the upper movable electrode 14 and the lower movable electrode 15.
Becomes

FIG. 8 is a partially cutaway perspective view of the sensor substrate of the capacitive sensor shown in FIG. As shown in FIG. 8, the sensor substrate 1 has pads 11 provided on the rim portion 2. The AA ′ cross section shown here is a cross section of the sensor substrate 1 in FIG.

FIG. 9 is an equivalent circuit diagram of the capacitance type acceleration sensor shown in FIG. As shown in FIG. 9, in operation of such a sensor, the upper fixed electrode 10 corresponds to the terminal T1 side (upper side) of the capacitor C1, and the upper movable electrode 14
Corresponds to the terminal T3 side (lower side) of the capacitor C1, the lower movable electrode 15 corresponds to the terminal T3 side (upper side) of the capacitor C2, and the lower fixed electrode 9 corresponds to the terminal T2 side (lower side) of the capacitor C2. Corresponding to

When the sensor substrate 1 receives an impact, the weight 6 moves up and down due to its inertia, as in the case of a piezo sensor. When the weight 6 moves upward, the distance between the upper fixed electrode 10 and the upper movable electrode 14 becomes shorter, so that the capacitance of the capacitor C1 increases. vice versa,
Since the distance between the lower fixed electrode 9 and the lower movable electrode 15 increases, the capacitance of the capacitor C2 decreases. This makes it possible to convert the magnitude of the impact into a change in capacitance and detect it.

In such a capacitive acceleration sensor, unlike the above-described piezoresistive sensor, there is no temperature-dependent factor, so that the sensitivity-temperature characteristics are improved.

However, the formation of the beam portion 7 itself is as follows.
Because it is performed by controlling the etching time,
The beam thickness varies between wafer lots, or the beam thickness also varies within the wafer surface due to variations in the etching bath due to the etching solution and concentration, and variations in the wafer thickness. For this reason, there is a difference in the easiness of displacement of the weight portion 6 between the elements, and as a result, the sensitivity varies.

As a method for suppressing such a variation in beam thickness, an etching method using an anodic oxidation method is generally known.

FIG. 10 is a sectional view of an etching tank using such a conventional anodic oxidation method. As shown in FIG.
Protective films 41 and 43 for an etchant 45 are formed on an n-type substrate n-epi layer semiconductor substrate (hereinafter referred to as an n / p semiconductor substrate) 20, and protection of a portion to be etched using a general photolithography technique Remove the film. A positive potential is applied to the n-epi layer side and a negative potential is applied to the reference electrode 44 while immersing the n / p semiconductor substrate 20 in an anisotropic etching solution 45 such as an aqueous potassium hydroxide solution. In particular, on the n-epi layer side, high-concentration n-type diffusion is performed to improve the contact property of the voltage application portion and reduce the potential variation in the wafer surface, and the n ⁺ diffusion layer 31 is formed on the contact portion and the surface of the wafer. It is formed in the entire inner surface or in a mesh shape.

This etching proceeds from the p-type substrate side to the n-epi layer. Since a reverse bias is applied between the n-epi layer and the reference electrode 44 at first,
No current flows. However, when etching progresses to the region of the depletion layer 40 between the n-type epitaxial layer and the p-type substrate which is generated by applying a reverse bias between the p-type substrate and the n-type epitaxial layer, a current starts to flow. The silicon oxide film 42 is formed in the portion, and the etching stops.

Therefore, even if the etching rate varies in the wafer surface, the etching of the portion where the etching is completed first does not proceed any more, so that a uniform beam thickness can be obtained. When the etching is completed on the entire surface of the wafer, the current stops flowing again, and the etching is completed.

In short, the thickness of the beam is determined by the thickness of the n-epi layer and the remaining thickness on the p-type substrate side caused by the spread of the depletion layer 40, and is uniform in the substrate plane.

FIGS. 11 (a) to 11 (f) are cross-sectional views of a sensor in the order of steps for explaining a method of manufacturing a capacitance type acceleration sensor using a conventional anodic oxidation method.

First, as shown in FIG.
The semiconductor substrate 20 is diffused with n-type impurities such as phosphorus or arsenic, which will be movable electrodes and wirings of a weight portion described later,
An n ⁺ diffusion layer 31 is formed on both sides. Next, a protective film 49 serving as a mask material for performing anisotropic etching is formed on the substrate 20.
On both sides. Thereafter, using a general photolithography technique, the protective film 49 where a thin film is to be formed is removed by etching, and a protective film removing portion 46 is formed.
At the same time, the protection film 49 on the n-epi layer is removed to take out the electrode when performing anodic oxidation, thereby forming a protection film removal part 48. Further, in order to short-circuit the p-type substrate of the substrate 20 and the n ⁺ diffusion layer 31 formed on the p-type substrate in a later step, a protective film is formed on a portion corresponding to the lower movable electrode. Is removed to form a protective film removing portion 47. In this case, the protective film includes
An oxide film or a nitride film similar to the above-described example is used.

Next, as shown in FIG.
The p semiconductor substrate 20 is immersed in an anisotropic etching solution, and silicon is etched using the anodic oxidation method described above. When this etching is completed, a thinned beam portion 7 having a uniform thickness, a thick weight portion 6 and a rim portion 2 are formed.
Further, in this step, the removal of the protective film (formation of the protective film removing portion 47) performed on a part of the portion corresponding to the lower movable electrode of the weight portion 6 requires the p ⁺
The V-shaped etching proceeds so as to enter the mold substrate, and the contact hole 50 on the back surface of the weight is formed.
The electrode removal protective film removing portion 48 on the n-epi layer side includes:
A jig or the like prevents contact with the etching solution.
Here, the thickness of the beam portion 7 is the sum of the thickness of the n-epi layer and the thickness of the depletion layer that has entered the p-type substrate, as described in the anodic oxidation method described above.

Then, as shown in FIG. 11C, in order to form a groove, a groove corresponding to a part of the beam 7 is formed on the protective film on the n-epi layer by using a general photolithography technique. The protective film on the formation region is removed, and a protective film removing portion 5 is formed.
Form one. At this time, at the same time, in order to short-circuit the n-epi layer and the p-type substrate in a later step, the protection film is removed from a part of the portion corresponding to the upper movable electrode, and the protection film removal part 52 is formed.

Then, as shown in FIG. 11D, silicon etching is performed from the protective film removing portion 51 using the same anisotropic etching or dry etching as described above.
A groove 8 penetrating through the upper and lower surfaces is formed. At this time, the etching from the protective film removing part 52 also proceeds, and
Contact hole 5 on the top of the weight over the die substrate
3 is formed.

Further, as shown in FIG. 11 (e), after removing the protective films 49 on both surfaces, the metal pad 11 and the weight portion upper surface contact electrode 54 are formed by sputtering or the like.
And the weight portion lower surface contact electrode 55 are formed.
Thus, the sensor substrate 1 is completed. Here, the weight portion upper surface contact electrode 54 has a function that the sputtered film enters the weight portion upper surface contact hole 53 and short-circuits the n-epi layer and the p-type substrate. Similarly, the weight portion lower surface contact electrode 55 has a function that the sputtered film enters the weight portion lower surface contact hole 50 and short-circuits the n ⁺ diffusion layer 31 and the p-type substrate.

Finally, as shown in FIG. 11F, the lower stopper 4 and the upper stopper 5 separately formed on the upper and lower surfaces of the sensor substrate 1 are adhered and fixed to complete the capacitance type acceleration sensor 56. The structure of the lower stopper 4 and the upper stopper 5 is the same as that of the sensor shown in FIG.

The capacitance type acceleration sensor 56 includes an n-epi layer, a p-type substrate, and an n ⁺ diffusion layer 3.
1 is electrically short-circuited so that a depletion layer region due to p / n junction is not generated.

FIG. 12 is an equivalent circuit diagram of the electrostatic sensor shown in FIG. As shown in FIG.
The positional relationship between C1 and C2 is the same as that of the capacitor in the sensor of FIG. 9 described above, but in such an electrostatic sensor, a resistor R1 is added. This resistance R1 is
It is synthesized from the substrate resistance of the p-type substrate, the contact resistance between the weight upper surface contact electrode 54 and the p-type substrate, and the contact resistance between the weight lower surface contact electrode 55 and the p-type substrate. Therefore, the capacitor C2
Has a time constant between C2 and R1.

The contact hole 53 on the upper surface of the weight portion
Also, as the opening diameter of the weight portion lower surface contact hole 50 is larger, the contact resistance and the substrate resistance can be reduced, but on the other hand, the electrode area is reduced and the capacitance sensitivity is lowered. Therefore, usually, the opening diameter is 100 μm.
m, but in this case, the area of the opening is small, and furthermore, the sputtered film is poorly wrapped around the contact hole, resulting in poor contactability and increased contact resistance. . As a result, the substrate resistance also increases. Usually, the substrate resistance of the p-type substrate is 30 to 5
When the resistance is 0 Ω · cm and the opening diameter is 100 μm, a resistance R1 of several hundred kΩ to several tens MΩ is generated at room temperature.

Since this resistor R1 has temperature dependency,
The time constants of R1 and C2 also have temperature dependence. As a result, the sensitivity of the sensor also has temperature dependence, and the advantage of the original capacitance type acceleration sensor that there is no temperature dependence of sensitivity is lost. Specifically, the sensitivity temperature change rate under the condition of the opening diameter of 100 μm described above is almost the same as the temperature characteristic of the piezoresistive acceleration sensor of FIG. 6, and eventually, the temperature correction of the sensitivity by an external circuit is required.

[0040]

The above-mentioned conventional acceleration sensor and the method of manufacturing the same, when the silicon anisotropic etching for forming a beam is performed only by controlling the etching rate and time, the wafer-to-wafer or wafer-to-wafer There is a drawback that the beam thickness varies in the plane and the detection sensitivity varies.

Further, in the above-described conventional example, when silicon anisotropic etching is performed by anodic oxidation in order to suppress variations in the beam thickness, the distance between the n-epi layer and the p-type substrate and the p-type substrate are reduced. And the n ⁺ diffusion layer must be short-circuited. At this time, if the diameter of the silicon etching opening for short-circuiting is small, the contact area decreases, so that the contact resistance and the substrate resistance increase, and there is a disadvantage that the temperature characteristics of the resistance cause temperature dependency in sensitivity.

Conversely, in order to reduce the substrate resistance,
Increasing the aperture diameter has the disadvantage that the electrode area is reduced, and the desired capacitance sensitivity cannot be obtained.

An object of the present invention is to reduce the variation in the beam thickness and to realize an element with uniform sensitivity, and to reduce an extra parasitic resistance to obtain a good sensitivity / temperature characteristic. It is to provide a sensor and a manufacturing method thereof.

[0044]

A capacitive acceleration sensor according to the present invention comprises a rim portion, a thinned beam portion, an upper movable electrode, and a lower movable electrode which are obtained by applying anisotropic etching to an n / p type semiconductor substrate. A sensor substrate formed with a thick weight portion provided with: and a gap for allowing the weight portion to move, which is adhesively fixed to the upper and lower surfaces of the sensor substrate, and the upper movable portion formed in the weight portion An upper stopper and a lower stopper each provided with a fixed electrode for detecting a change in capacitance at a position facing the electrode and the lower movable electrode, wherein the upper movable surface of the rim portion of the sensor substrate and the upper movable portion of the weight portion are provided; electrodes configured the lower movable electrode and the tapered portion surfaces are connected by n + -type semiconductor layer.

Further, according to the method of manufacturing a capacitance type acceleration sensor of the present invention, phosphorus or arsenic is diffused on both sides of an n / p type semiconductor substrate obtained by epitaxially growing an n type silicon layer on the upper surface of a p type semiconductor substrate. and forming an n ⁺ -type diffusion layer, anisotropic etching using an anodic oxidation method from the first protection film coated on the n / p-type semiconductor substrate formed with the n ⁺ -type diffusion layer Forming regions to be weight portions, rim portions, and beam portions of the sensor substrate by performing
After the protective film is removed, a step of diffusing impurities such as phosphorus or arsenic from both sides of the n / p type semiconductor substrate to form a thick n ⁺ type diffusion layer, Forming a groove by removing a portion of the beam by anisotropic etching or dry etching to facilitate movement of the weight, after coating the second protective film;
Removing a pad from the n ⁺ -type diffusion layer of the n / p-type semiconductor substrate exposed by removing the second protective film by sputtering or the like to complete the sensor substrate;
The weight portion of the sensor substrate is provided with a space in which the weight portion can move, and a lower stopper and an upper stopper formed with fixed electrodes for detecting capacitance at positions opposed to the upper and lower surfaces of the weight portion, respectively. Bonding to the upper and lower surfaces.

[0046]

Next, embodiments of the present invention will be described with reference to the drawings.

FIG. 1 is a sectional view of a capacitance type acceleration sensor according to an embodiment of the present invention. As shown in FIG. 1, the sensor according to the present embodiment includes
The sensor substrate 1 includes a rim portion 2 and a beam portion 7 and lower and upper stoppers 4 and 5 adhesively fixed to upper and lower surfaces of the sensor substrate 1. The cross section of the sensor substrate 1 is the same as that of the conventional example shown in FIG. A cross section similar to the -A 'cross section is shown.

In particular, the sensor substrate 1 is formed so that the upper and lower surfaces are covered with the n ⁺ diffusion layer 3 and the tapered portion 18 of the weight 6 is also covered with this layer. Therefore, the surface of the weight portion 6 on the n-type epi layer side becomes the upper movable electrode 14, and the opposite surface side becomes the lower movable electrode 15, which are connected to each other by the n ⁺ diffusion layer 3. That is, these upper movable electrodes 14,
The lower movable electrode 15 and the pad 11 are commonly wired by an n-type impurity layer. In short, the sensor substrate 1 is covered with an n-type layer in which the front and back of the weight portion 6 and the rim portion 2 are connected to each other.

The lower stopper 4 and the upper stopper 5 are provided with cavities 13 and 12 so that the weight 6 of the sensor substrate 1 can move up and down. The upper fixed electrode 10 and the lower fixed electrode 9 are formed at the same and opposite inner positions. The upper fixed electrode 10 is drawn out to the upper surface through a through-hole portion 16 provided on the upper stopper 5, and can make contact with the outside. Similarly, the lower fixed electrode 9 is drawn out to the lower surface through the through-hole portion 17 and can make contact with the outside. These stoppers can be realized, for example, by bonding and fixing to a ceramic substrate or the like using a conductive paste or the like.

Next, the operation of the capacitance type acceleration sensor according to the present embodiment will be described. The equivalent circuit of this sensor is the same as the equivalent circuit of the conventional sensor shown in FIG. However, the diffusion resistance of the n-type impurity for wiring from the pad 11 to the upper movable electrode 14 and wiring from the pad 11 to the lower movable electrode 15 is as small as several hundred Ω or less, and is ignored in this case. . 9, the upper fixed electrode 10 corresponds to the upper electrode of the capacitor C1, and the upper movable electrode 14 corresponds to the lower electrode of the capacitor C1. Similarly, the lower movable electrode 15 is connected to the capacitor C2.
The lower fixed electrode 9 corresponds to the upper electrode of the capacitor C2.
Corresponding to the lower electrode.

Here, when the sensor substrate 1 receives an impact,
The weight 6 moves up and down due to its inertia. When the weight 6 moves upward, the distance between the upper fixed electrode 10 and the upper movable electrode 14 becomes shorter, so that the capacitance increases. Conversely, the distance between the lower fixed electrode 9 and the lower movable electrode 15 increases, so that the capacitance decreases. Similarly, when the weight 6 moves downward, the spread of the electrodes is in the opposite relationship, and the capacitance is also in the opposite relationship. This makes it possible to convert the magnitude of the impact into a change in capacitance and detect it.

In the above-described capacitance type acceleration sensor, since the beam thickness is controlled by using the anodic oxidation method, the variation of the beam thickness between wafers or within the wafer surface is small, and the beam thickness at the top and bottom of the weight portion is small. Since there is no variation in ease of movement, a product with uniform sensitivity can be realized. Moreover,
In such a sensor, since the factors (members) having temperature dependency are eliminated, the rate of change in sensitivity due to temperature is extremely small.

FIG. 2 is a sensitivity / temperature characteristic diagram of the sensor in FIG. As shown in FIG. 2, when the temperature change rate of the sensitivity when the above-mentioned capacitance type acceleration sensor is driven at a constant voltage of 5 V with reference to 25 ° C. is less than ± 5%, the conventional anodic oxidation It can be understood that it is greatly improved even when compared with the change rate of the capacitance type acceleration sensor using the method.

FIGS. 3A to 3D are cross-sectional views of a sensor in the order of steps for explaining an embodiment of a method of manufacturing a capacitive acceleration sensor according to the present invention.

First, as shown in FIG. 3 (a), an n / n layer composed of a p-type substrate 19A and an n-type epi layer 19B is formed.
In order to form a movable electrode and a wiring of a weight portion in the p semiconductor substrate 20, n ⁺ impurities such as phosphorus or arsenic are diffused to form n ⁺ diffusion layers 21 on both surfaces. Furthermore, n
/ P semiconductor substrate 20 is coated on both sides with a protective film 22 serving as a mask material when performing anisotropic etching, and then etched using a general photolithography technique to form a thin film portion. The protection film is removed, and a protection film removal part 23 is formed. At the same time as the formation of the removal portion 23, the protection film 22 is partially removed from the n-epi layer 19B to form an electrode extraction window for performing anodic oxidation, and the protection film removal portion 24 is formed. . Note that a silicon oxide film, a silicon nitride film, or the like is used as the protective film 22 at this time.

Next, as shown in FIG. 3B, in order to form the weight portion 6 and the beam portion 7, the n / p semiconductor substrate 20 is immersed in an anisotropic etching solution to remove the protective film on the epi layer. A positive potential is applied to the portion 23 and a negative potential is applied to the reference electrode provided in the etching solution, and silicon is etched using an anodic oxidation method. An aqueous solution of potassium hydroxide, saturated hydrazine, an EDP solution, or the like is used as the anisotropic etching solution at this time. First, etching is started from the p-type substrate 19A side, and thereafter, when etching proceeds to a depletion layer region generated by applying a bias to the n-epi layer 19B and the p-type substrate 19A, the etching is automatically performed there. finish.
As a result, a thin beam portion 7 having a uniform thickness in the plane of the substrate and a thick weight portion 6 are formed.

The thickness of the thinned beam 7 is, for example, p
The substrate resistance of the mold substrate 19A is 30 to 50Ω · c.
When a 5 V bias is applied to a semiconductor substrate having an m and n epi layer 19B having a substrate resistance of 3 to 5 Ω · cm and a thickness of 8 μm and etching is performed using an anodic oxidation method, a p-type substrate 19A is formed. Since the depletion layer of about 4 μm spreads over the entire area, the thickness of the beam portion 7 becomes about 12 μm. Note that n
The protective film removing portion 24 for taking out the electrode on the epi layer side is prevented from coming into contact with the etching solution by a jig or the like.

Next, as shown in FIG. 3C, the protective film 22 left on the semiconductor substrate 20 is removed in order to perform n-type impurity diffusion in the next step.

Further, as shown in FIG. 3D, in order to connect the front and back of the p-type substrate 19A to each other with an n-type layer, an n-type impurity such as phosphorus or arsenic is applied from both sides of the n / p semiconductor substrate 20. To spread. In other words, at least in the thinned beam portion 7, the diffusion is performed to a depth at which the p-type layer left by the expansion of the depletion layer is changed to the n-type layer.

For example, the impurity concentration of phosphorus is 10 ²⁰ / cm
After diffusion at 950 ° C for 60 minutes at ³
When the indentation oxidation treatment is performed with C for about 180 minutes, diffusion of about 8 μm in depth can be performed, and the 4 μm-thick p-type layer left by the expansion of the depletion layer is changed to an n-type layer. Becomes possible.

[0061] Also, since n ⁺ diffusion layer 3 to the tapered portion of the weight portion formed by anisotropic etching is formed from the n ⁺ diffusion layer 3 formed on the n epitaxial layer, n epitaxial layer, A series of n-type impurity layers are formed up to the back surface of the substrate via the beam portion 7 and the tapered portion 18. Thereafter, both surfaces of the substrate are coated with the protective film 25 again.

Then, as shown in FIG. 3E, in order to form a groove for facilitating the movement of the weight portion 6, the protective film 25 on the n-epi layer is formed using a general lithography technique. In particular, the protection film 25 on the groove forming region corresponding to a part of the beam portion is removed to form a protection film removal part 26.

Then, as shown in FIG. 3 (f), the substrate is immersed in an anisotropic etchant or dry-etched to remove the protective film 26 in the same manner as described above.
Then, a groove 8 penetrating through the upper and lower surfaces is formed.

Further, as shown in FIG. 3 (g), after removing the protective films 25 on both surfaces of the substrate, a metal pad 11 for extracting an output signal to the outside is formed by sputtering or the like. Complete. If the lower and upper stoppers are adhesively fixed to the sensor substrate 1 from above and below, a capacitance type acceleration sensor can be configured.
The overall structure of the sensor substrate 1 is the same as the appearance of the above-described conventional example shown in FIG.

In short, in the present embodiment, the same steps as those in the related art can be adopted except that the weight portion 6 of the sensor substrate 1 is covered with the n-type layer.

[0066]

As described above, according to the capacitance type acceleration sensor and the method of manufacturing the same of the present invention, an anisotropic silicon etching using an anodizing method is performed on an n / p type semiconductor substrate and a weight portion and By forming a thinned beam portion and then performing n-type impurity diffusion to form n-type layers connected to each other on the front and back of the p-type semiconductor region, an element having a uniform beam thickness and uniform sensitivity can be obtained. There is an effect that it can be realized.

Further, according to the present invention, after the beam portion, the weight portion, and the rim portion are formed by anisotropic silicon etching, n-type impurity diffusion is performed from both sides of the substrate to entirely cover the n-type region. By arranging the pad and the upper and lower movable electrodes with n-type impurities, there is an effect that no extra parasitic resistance is added and a good sensitivity temperature characteristic can be obtained.

[Brief description of the drawings]

FIG. 1 is a sectional view of a capacitance type acceleration sensor according to an embodiment of the present invention.

FIG. 2 is a sensitivity / temperature characteristic diagram of the sensor in FIG.

FIG. 3 is a sectional view of the sensor shown in the order of steps for explaining one embodiment of the method of manufacturing the capacitance type acceleration sensor of the present invention.

FIG. 4 is a cross-sectional view of a conventional piezoresistive acceleration sensor.

5 is a partially cutaway perspective view of the sensor substrate shown in FIG.

FIG. 6 is a sensitivity / temperature characteristic diagram of the sensor shown in FIG. 4;

FIG. 7 is a cross-sectional view of a conventional capacitance-type acceleration sensor and a sensor shown in the order of steps for explaining a method of manufacturing the same.

FIG. 8 is a partially cutaway perspective view of a sensor substrate of the capacitive sensor in FIG. 7;

FIG. 9 is an equivalent circuit diagram of the capacitance type acceleration sensor in FIG. 7;

FIG. 10 is a cross-sectional view of an etching tank using a conventional anodic oxidation method.

FIG. 11 is a sensor cross-sectional view shown in the order of steps for explaining a method of manufacturing a capacitance type acceleration sensor using a conventional anodic oxidation method.

12 is an equivalent circuit diagram of the electrostatic sensor shown in FIG.

[Explanation of symbols]

DESCRIPTION OF SYMBOLS 1 Sensor board 2 Rim part 3, 21 n ⁺ diffusion layer 4 Lower stopper 5 Upper stopper 6 Weight part 7 Beam part 8 Groove part 9 Lower fixed electrode 10 Upper fixed electrode 11 Pad 12, 13 Air gap 14 Upper movable electrode 15 Lower movable electrode 16 , 17 Through hole portion 18 Taper portion 19A p-type semiconductor substrate 19B n-type epilayer 20 n / p-type semiconductor substrate 22, 25 protective film 23, 24, 26 protective film removing portion

Claims (2)

(57) [Claims] A rim portion, a thinned beam portion, an upper movable electrode, and an anisotropically etched n / p type semiconductor substrate;
A sensor substrate having a thick weight portion provided with a lower movable electrode, and adhesively fixed to the upper and lower surfaces of the sensor substrate to form a gap for the weight portion to move, and to be formed in the weight portion. An upper stopper and a lower stopper each provided with a fixed electrode for detecting a change in capacitance at a position facing the upper movable electrode and the lower movable electrode, wherein a rim portion surface of the sensor substrate and a weight portion of the weight portion are provided; the upper movable electrode, an electrostatic capacitance type acceleration sensor, characterized in that the lower movable electrode and the tapered portion surfaces are connected by n + -type semiconductor layer. Forming an n + -type diffusion layer by diffusing phosphorus or arsenic on both sides of an n / p-type semiconductor substrate having an n-type silicon layer epitaxially grown on an upper surface of the p-type semiconductor substrate; A first protective film is coated on the n / p type semiconductor substrate having the n + type diffusion layer formed thereon, and then subjected to anisotropic etching using an anodic oxidation method, whereby a weight portion of the sensor substrate,
Forming a region to be a rim portion and a beam portion, and after removing the first protective film, diffusing impurities such as phosphorus or arsenic from both surfaces of the n / p type semiconductor substrate to form an n + type diffusion layer. Forming a second protective film on both surfaces of the n / p type semiconductor substrate, and then anisotropically etching or drying a part of the beam portion to facilitate movement of the weight portion. Removing a groove by etching to form a groove; and attaching a pad on the n + type diffusion layer of the n / p type semiconductor substrate exposed by removing the second protective film by sputtering or the like. A step of completing the substrate, and while providing a space in which the weight portion of the sensor substrate is movable, a lower stopper formed with a fixed electrode for detecting capacitance at a position opposed to each of the upper and lower surfaces of the weight portion, and Adhering and fixing an upper stopper to upper and lower surfaces of the sensor substrate.