JP3528539B2 - Semiconductor device and manufacturing method thereof - 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 device formed by using semiconductor micromachining technology and a method for manufacturing the same.

[0002]

2. Description of the Related Art A semiconductor device using a semiconductor surface micromachining technique is disclosed in, for example, Japanese Patent Laid-Open No. 7-24541.
There is an acceleration detecting device described in No. 6. In this device, a movable part that is displaced by the application of acceleration and a fixed part that faces the movable part are formed in a semiconductor substrate, and a comb-teeth-shaped movable electrode and a fixed electrode are provided in each, and both electrodes have a minute gap. They are arranged so as to mesh with each other and are configured to detect the magnitude of acceleration based on a change in electrostatic capacitance between both electrodes when the movable portion is displaced by application of acceleration. The reason why the comb-teeth electrode is used in the capacitance detection type as described above is to increase the facing area to increase the capacitance and to improve the sensitivity and accuracy.

In the structure as described above, it is necessary to insulate each fixed electrode formed in a comb shape from the surroundings and to form a wiring for connecting each fixed electrode to the outside. In the above-mentioned conventional example, in order to insulate each fixed electrode from the surrounding structure, a trench groove is provided, and the groove is insulated so as not to contact the surrounding portion. Therefore, the wiring structure for taking out from each fixed electrode to the outside floats in the groove and cannot be fixed.By forming the wiring structure on another substrate and attaching it to the substrate, each fixing It has a structure in which wiring is connected to the electrodes.

[0004]

As described above, in the conventional device, since the structure is formed by providing the trench groove in the semiconductor substrate to form the structure, and is also insulated from the surroundings,
The wiring cannot be directly drawn out from each part constituting the structure within the main surface of the substrate, and thus it is necessary to use a special structure in which a wiring structure is formed on another substrate and bonded as described above. Since it is complicated and the area is large, miniaturization is difficult and the manufacturing cost is high.
There was a problem.

As a method of solving the above problem, for example, a method of filling the trench groove with an insulator (dielectric) and forming an electrode thereon can be considered. However, this method has the following problems. Hereinafter, description will be given with reference to FIG.

FIG. 10 is a diagram showing an outline of the above method,
(A) is a schematic cross-sectional view of a main part of a structure, and (b) is a schematic perspective view of a main part. As shown in (a), when the groove 101 is formed in the semiconductor substrate 100 and filled with an insulator, if the groove 101 is formed by etching from the substrate surface, it is possible to form a groove whose wall surface is completely vertical. It is difficult, and the etched surface is slightly inclined. That is, the cross-sectional shape of the groove is an inverted trapezoid (a shape in which the upper bottom is wider than the lower bottom). In order to fill the groove with an insulator and separate the electrodes 102 from each other, as shown in (b), when a process of removing a portion between the electrodes 102 (portion 104) by etching is performed, the reverse of the above is performed. Since the lower end portion of the trapezoid is shaded when viewed from the upper surface, the etching residue 103 is likely to occur at that portion. When such an etching residue occurs, as shown in (b), the electrodes may come into contact with each other to cause a short circuit, which causes a problem that the manufacturing yield is significantly deteriorated.

As described above, the structure of the conventional example has a problem that the wiring structure is complicated and the area is large, so that miniaturization is difficult and the manufacturing cost is high.
In order to solve the problem, if an insulator is simply buried in the groove to form a wiring structure on the insulator, there is a problem that the short circuit between electrodes occurs and the manufacturing yield is deteriorated.

The present invention has been made to solve the problems of the prior art as described above, and has a simple wiring structure, enables miniaturization, a high manufacturing yield, and a low cost semiconductor device. It is an object to provide a manufacturing method thereof.

[0009]

In order to achieve the above object, the present invention is constructed as described in the claims. That is, in the invention described in claim 1, the active layer substrate having the through hole filled with the first dielectric and the supporting substrate are integrated with the first dielectric.
The through holes are joined together through the formed second dielectric, and the through holes are trapezoidal in which the lower bottom is longer than the upper bottom in cross section, and the lower bottom of the trapezoid is arranged on the side of the supporting substrate. SOI
Using a substrate, at least one conductive region formed so as to be insulated from the surroundings by the trench groove and the through hole provided so as to reach from the surface to the bottom surface of the active layer substrate at a position different from the through hole . At least part of
In a semiconductor device having a structure held by the supporting substrate via a second dielectric, electrical wiring for connecting the conductive region to the outside of the insulated range is connected to the SO
It is arranged so as to cross the first dielectric filled in the through hole on the main surface of the I substrate.

According to the above structure, the cross-sectional shape of the through hole is
When the active layer substrate is trench-etched to form a conductive region, the trapezoid has a longer bottom than the top and the side to be etched has a slope that is open upward (slope that is open when viewed from the etching direction). In addition, there is no shaded portion when viewed from the upper surface where etching is performed, and the semiconductor active layer region on the side surface of the through hole can be completely etched away. Therefore, a short circuit portion does not occur between the conductive regions, and the manufacturing yield can be improved. Further, in this structure, since the electric wiring can be provided on the main surface of the SOI substrate on the dielectric filled in the through hole, the structure is simplified and miniaturization becomes possible.

The invention described in claim 2 is the same as claim 1.
In the structure described in (1), a movable part that is displaced according to the application of a physical quantity, a plurality of comb-teeth movable electrodes that interlock with the movable part, and face the comb-teeth movable electrodes with a minute gap on the longitudinal side surface. It is applied to a physical quantity detection sensor (for example, an acceleration sensor) having a plurality of comb-teeth fixed electrodes. As described above, in the structure of the present invention, there is no possibility of causing a short circuit between the comb-teeth fixed electrodes or the comb-teeth movable electrodes, and the wiring structure is simple, so that it is possible to further reduce the distance between the electrodes. Become. Therefore, the detection capacitance can be increased.
Since the detection capacitance affects the detection accuracy, the sensitivity and S / N can be improved by increasing the detection capacitance in this way. Also, miniaturization is possible. This configuration corresponds to, for example, the first embodiment described below.

Further, in the invention according to claim 3, two or more independent comb tooth fixed electrodes are arranged for one comb tooth movable electrode, and the displacement of the movable portion and the comb tooth movable electrode is It is configured to detect as a differential value of the electrostatic capacitance between the comb-teeth movable electrode and the plurality of comb-teeth fixed electrodes. In the structure of the present invention, the wiring can be drawn out from each comb-teeth fixed electrode with a simple structure. Therefore, if the above configuration is adopted and the capacitance between the comb-teeth fixed electrodes is detected as a differential value, , The sensitivity can be improved. This configuration corresponds to, for example, the first embodiment described below.

According to a fourth aspect of the present invention, two or more independent comb tooth fixed electrodes are arranged for one comb tooth movable electrode, and at least one comb tooth fixed electrode is a control electrode. It is configured to be used as. This control electrode can be used as, for example, a shield electrode, an input acceleration compensation electrode for generating an electrostatic attractive force for compensating an input acceleration, or a pseudo acceleration input electrode for inputting a pseudo acceleration by an electrostatic attractive force. Note that this configuration is, for example, the second
Corresponding to the embodiment.

Further, claim 5 and claim 6 show a method of manufacturing the semiconductor device according to any one of claims 1 to 4, wherein claim 5 is such that the active layer substrate and the support substrate have a dielectric therebetween. The subsequent steps are shown on the premise that an SOI substrate bonded by the above method is used. Further, claim 6 shows the entire manufacturing method including the manufacturing process of the SOI substrate as described above.

Incidentally, in the acceleration sensor using polycrystalline silicon as described in Japanese Patent Publication No. 6-44008, the electrodes of polycrystalline silicon are deposited by the chemical vapor deposition method. It is difficult to control the physical properties of such polycrystalline silicon, and it is particularly difficult to control the residual internal stress that affects the sensitivity. Therefore, the characteristics tend to vary from individual to individual. Moreover, since the thickness of the electrode formed by deposition is at most about several μm, it is difficult to increase the area of the facing surface of the comb-teeth electrode to increase the capacitance. In this respect, in the present invention, since an active layer substrate made of single crystal silicon can be used, it is easy to control the material constants such as residual internal stress, and it is possible to suppress the variation between solids.
Moreover, since the structure such as the comb-teeth electrode is formed by performing trench etching on the active layer substrate, the electrode thickness can be appropriately set according to the thickness of the active layer substrate. Therefore, the area of the facing surface of the comb-teeth electrode can be increased to increase the detection capacitance, and high sensitivity and high accuracy can be realized.

[0016]

According to the present invention, since the wiring can be drawn out within the main surface of the substrate, the electric wiring can be realized in each part of the microstructure by the usual semiconductor technology. Therefore, unlike the conventional example, it is not necessary to use a special technique such as joining with an insulating substrate on which wiring is formed or embedded wiring, and an effect that the manufacturing cost can be suppressed can be obtained.

Further, when it is applied to a sensor or the like for detecting the displacement of a microstructure by a change in capacitance, it is possible to improve the disposition density of displacement detection electrodes and increase the detection capacitance. You can Therefore, the detection S / N ratio of the sensor can be improved. Further, since the same detection capacitance can be secured in a smaller area and the structure is simple, the manufacturing cost can be suppressed and the size can be reduced.
Furthermore, if the same detection capacitance is used, a larger change in the detection capacitance can be obtained, so that it is possible to obtain the effect that the accuracy of the sensor can be improved.

[0018]

DETAILED DESCRIPTION OF THE INVENTION

(First Embodiment) FIG. 1 is a schematic plan view showing a first embodiment of the present invention, FIG. 2 is an enlarged schematic plan view of an electrostatic capacitance portion in FIG. 1, and FIG. A'section schematic diagram, FIG.
2 is a schematic sectional view taken along the line BB ′ of FIG. 1. First, the configuration will be described. The present embodiment is an example of an acceleration sensor realized by an SOI substrate. The SOI substrate used in this embodiment is one in which a semiconductor active layer substrate (for example, a single crystal silicon substrate) 48 and a semiconductor support substrate 47 are bonded via a dielectric 46. The semiconductor active layer substrate 48 has the through hole 4
9 has a structure in which an oxide film 43 and a dielectric 46 are embedded, and the through hole 49 has a trapezoidal sectional shape (semiconductor active layer substrate 4
When viewed from the side with 8 facing up, the upper base is a trapezoid that is narrower than the lower base (see FIG. 3).

The acceleration sensor structure is realized by forming the trench groove 28 in the semiconductor active layer substrate 48. This acceleration sensor structure has a movable mass 3
0, and a supporting portion 31 having one end connected to the movable mass 30,
The fixed portion 32 to which the other end of the support portion 31 (the end portion on the side opposite to the connection portion to the movable mass 30) is connected, the comb-teeth movable electrode 38 (there are a plurality) connected to the movable mass 30, and the comb. Fixed-toothed comb electrodes 39 and 40 (there are a plurality of electrodes) facing the tooth movable electrode 38 with a minute gap on the longitudinal side surface thereof.
And a fixing portion 34 for fixing the comb-teeth fixed electrodes 39, 40 to the support substrate 47.

Further, as shown in FIG. 2, an etching hole 5 for forming a gap described below is formed in the movable mass 30.
Many 3 are arranged. As a result, by using the method described in the manufacturing process example described later, as shown in FIGS. 3 and 4, a gap 45 is formed between the dielectric body 46 and immediately below the structure other than the fixing portions 32 and 34. A movable space is secured. The fixing portions 32 and 34 are fixed to the dielectric 46 by the oxide film 43.

Further, as shown in FIG. 2, two comb-tooth fixed electrodes are opposed to one comb-tooth movable electrode at equal intervals. Contact holes 37 and 42 are formed in the insulating film 44 formed on the surfaces of the fixing portions 32 and 34. The conductive wiring 26 deposited by the contact holes is fixed to the fixing portion 34, and the wiring 36 is fixed to the fixing portion. It is electrically connected to 32. Protective films 41 and 35 are formed on the surfaces of the wirings 26 and 36. The wirings 26 and 36 are drawn from the fixing portions 34 and 32 to the frame portion 50 across the surface portion 29 of the through hole 49 in which the oxide film 43 and the dielectric 46 are embedded. In particular, the wiring 26 drawn out from the fixed portion 34 of the comb-teeth fixed electrode is shown in FIG.
The ones 40 facing each other in the upper part and the 39 parts facing each other in the lower part are connected to each other. A jumper wire 27 using a conductive wiring is used to realize this connection.

In FIG. 1, the wiring drawn from the fixed portion 32 of the movable mass 30 to the frame portion 50 is C, and the fixed comb electrode 40 facing the movable comb electrode in FIG.
Is indicated by U, and the wiring from the comb-teeth fixed electrode 39 facing the lower side in FIG. 1 is indicated by D. Wiring C, U, D is 2
They are provided one by one, and they are commonly connected and input to the subsequent signal processing circuit (the signal processing circuit is not shown).

Next, FIGS. 5 to 7 are sectional views showing an example of a manufacturing process of the above-mentioned device. This cross section corresponds to FIG. It should be noted that although the steps (a) to (j) in FIGS. 5 to 7 are a series of steps, they are described separately in three for convenience of illustration. Hereinafter, this manufacturing process will be described. (A) Semiconductor active layer substrate (eg, silicon single crystal substrate)
The V-shaped groove 51 is formed by performing anisotropic etching on 48 using an alkaline etching solution or the like. (B) An oxide film 43 is formed on the surface of the semiconductor active layer substrate 48. (C) A dielectric 46 is deposited on the oxide film 43 on the surface of the semiconductor active layer substrate 48. Polycrystalline silicon deposited by a chemical vapor deposition method, B-Si-O glass deposited by a flame deposition method, or the like can be applied as the dielectric. (D) After flattening the surface of the dielectric 46, the semiconductor active layer substrate 48 is inverted so that the dielectric 46 faces downward, and the semiconductor active layer substrate 48 is overlaid and bonded onto the semiconductor supporting substrate 47.

(E) The semiconductor active layer substrate 48 is filled with the dielectric 46 and the oxide film 43 by grinding the surface of the semiconductor active layer substrate 48 opposite to the bonding surface to a desired thickness reaching the V-shaped groove. The SOI substrate having the through holes 49 formed is completed. The cross-sectional shape of the through hole 49 is trapezoidal (a trapezoid whose upper bottom is narrower than its lower bottom when viewed from the side with the semiconductor active layer substrate 48 facing up). (F) An insulating film 44 is formed on the SOI substrate. (G) The contact hole 42 is formed in the insulating film 44.
After that, a conductor layer is deposited and patterned to form the wiring 26 so as to cross the through hole portion 49. (H) The insulating film 44 is patterned to form a wiring protective film 41.
Are deposited and patterned.

(I) Trench etching mask layer 52
Are deposited and patterned for the acceleration sensor structure. Although many patterns of the etching holes 53 are formed in the movable mass 30, they are not shown. However, in the trench etching pattern near the comb-teeth fixed electrode, as shown in FIG. 2, an overlapping region 33 with the surface 29 of the through hole is provided. After that, the trench groove 28 reaching the oxide film 43 is formed in the semiconductor active layer substrate 48. (J) A part of the oxide film 43 is removed with a hydrofluoric acid-based solution,
The movable space gap 45 is formed. A movable space gap 45 is also formed directly below the movable mass 30 via an etching hole 53 (not shown).

Further, when the above manufacturing process is used, FIG.
The portion indicated by DD 'of FIG. 8 is realized as shown in FIGS. In FIG. 8, (a) the state immediately before the step of FIG. 7I, in which the etching mask 52 is deposited on the patterned insulating film 44. (B) The etching mask 52 is patterned according to the structure. (C) Trench etching is performed to reach the oxide film 43. At this time, since the side surface of the through hole 49 is inclined upward, there is no shaded portion when viewed from the top surface for etching, and the semiconductor active layer region on the side surface of the groove can be completely removed by etching. . Therefore, no short-circuit portion is generated between the comb-teeth fixed electrodes. (D) Corresponding to the step of FIG. 7J, a part of the oxide film 43 is removed with a hydrofluoric acid-based solution to form the movable space gap 45.

Next, the operation of this embodiment will be described. The basic operation of the acceleration sensor is the same as that of the conventional example. In FIG. 1, the movable mass 30 and the comb-tooth movable electrode 38 are displaced along with the input of acceleration in the x-axis direction.
The displacement is caused by the comb-teeth movable electrode 38 and the comb-teeth fixed electrodes 39, 40.
It is detected as a change in capacitance between them. The two capacitance changes have opposite signs, that is, when the capacitance between the comb-tooth movable electrode 38 and the comb-tooth fixed electrode 39 increases, the comb-tooth movable electrode 38 is increased.
The capacitance between the fixed electrode 40 and the comb-teeth fixed electrode 40 decreases, and the latter increases when the former decreases. As described above, since the comb-teeth fixed electrodes 39 and 40 are wired independently of each other, the differential connection enables detection by the differential capacitance.

The above operation will be described below in comparison with the above-mentioned conventional example (JP-A-7-245416). In the conventional example, each comb-teeth fixed electrode is connected to one wiring. In such a structure, as shown in FIG. 11, it is necessary to dispose adjacent comb-teeth movable electrodes and comb-teeth fixed electrodes at different intervals. Temporarily, each comb tooth movable electrode 106a-10
6c and the comb-teeth fixed electrodes 108a to 108c are arranged at the same intervals, when the movable portion 105 is displaced in the direction of arrow A, for example, the comb-teeth movable electrode 106a is moved to the comb-teeth fixed electrode 10.
The distance from the comb-teeth fixed electrode 108a is the same as the distance closer to 8b, and the change in the overall capacitance is offset. Therefore, as shown in FIG. 11, if one interval is k and the other is m, then k ≠ m. As described above, in order to arrange the adjacent comb-teeth movable electrodes and comb-teeth fixed electrodes at different intervals, it is necessary to increase the inter-electrode distance, so that the arrangement density of the comb-teeth electrodes in the main surface of the substrate is improved. There is a limit to this, and there is a limit to the improvement of the detection capacitance value.

Further, as described above, even if the adjacent comb-teeth movable electrodes and comb-teeth fixed electrodes are arranged at different intervals, they are separated from the other electrode by the amount close to one electrode, so that the detected electrostatic capacitance is obtained. Changes are offset and become smaller. The details will be described below. In FIG. 11, for example, the comb-teeth movable electrode 1
The distance between 06a and the comb-teeth fixed electrode 108a is k, and the distance between the comb-teeth movable electrode 106a and the comb-teeth fixed electrode 108b is m,
If the former capacitance is Ck and the latter capacitance is Cm, the total detected capacitance is composed of the sum of Ck and Cm.

When the movable portion 105 is displaced by x in the direction of the arrow A, the gap m is separated by the amount that the gap k is close to each other, so that the electrostatic capacitance changes of Ck and Cm have opposite signs and cancel each other out. If m = n × k (where n is an arbitrary number), the detected capacitance (Ck after displacement by x)
+ Cm) _x is represented by the following (Formula 1). (Ck + Cm) _x = (εS / k) · (1 + 1 / n) · [1- (x / k) · (1-1 / n)] (Equation 1) where ε: permittivity x: displacement S : Electrode facing area This shows that the rate of change decreases to (1-1 / n) as compared with the case of the detected capacitance facing only at the gap k. For example, if n = 3, the rate of change is reduced by 33%. That is, the sensitivity is lowered. Also,
As described above, increasing the ratio n between k and m has a limit in view of the electrode arrangement density, and even if the electrode arrangement density is sacrificed and increased to n = 5, the above 33% reduction occurs. Since it is only 20% reduction, it is not so effective.

As described above, in the structure in which each comb-teeth fixed electrode is connected to one wiring, it is necessary to dispose the adjacent comb-teeth movable electrode and comb-teeth fixed electrode at different intervals. There is a limit in improving the arrangement density of the comb-teeth electrodes in the inside, and there is a limit in improving the detection capacitance value. Further, even if the adjacent comb-teeth movable electrode and the comb-teeth fixed electrode are arranged at different intervals, since the distance from the other electrode is increased due to the proximity to one electrode, the change in the detected capacitance is offset and becomes small. .

On the other hand, in the structure of the present invention, the wiring can be easily drawn from each electrode. Therefore, as shown in the embodiment, for one comb-tooth movable electrode,
If two or more independent comb-teeth fixed electrodes are arranged, when the electrostatic capacitance between the comb-teeth movable electrode and one comb-teeth fixed electrode increases due to the displacement of the movable portion, the other comb-teeth fixed electrode of the other comb-teeth fixed electrode Although the electrostatic capacitance will decrease, by differentially connecting the two, there is no cancellation of the detected electrostatic capacitance change, and a large electrostatic capacitance change can be obtained. Therefore, the comb-teeth movable electrode and the comb-teeth fixed electrode which are adjacent to each other can be arranged at equal intervals, so that the arrangement density of the comb-teeth electrodes in the main surface of the substrate can be improved. Note that, as a specific example of differentially connecting the electrostatic capacitance between the two comb-teeth fixed electrodes, the wiring drawn from the two comb-teeth fixed electrodes may be respectively connected to the two input terminals of the differential amplifier.

The following effects can be obtained in the embodiment of the present invention. (1) In an acceleration sensor structure formed by trench etching using an SOI substrate, electrical wiring from each component can be realized on the main surface of the substrate. Therefore, the wiring can be formed by an ordinary semiconductor manufacturing technique. As a result, there is no need for special techniques such as joining with another substrate as in the conventional example, embedded wiring, and routing of wiring from the embedded wiring to the main surface of the substrate, and the structure is simple and miniaturization is possible. The manufacturing cost can be suppressed. (2) Since wiring is possible within the main surface of the substrate, when the displacement of the movable portion is detected by electrostatic capacitance, it is possible to detect the displacement using the differential capacitance, and it is possible to realize high precision detection. (3) Since the differential capacitance can be detected, the arrangement density of the detection capacitance within the main surface of the substrate can be improved. Therefore, if the area is the same, a larger emission capacitance can be obtained and the sensitivity can be improved. Further, if the sensitivities are the same, miniaturization and miniaturization are possible. (4) Since the differential capacitance can be detected, it is possible to avoid the offset of the electrostatic capacitance change due to the adjacent comb-teeth electrodes, and to secure a larger detected electrostatic capacitance change.

(Second Embodiment) FIG. 9 is a schematic plan view showing a second embodiment. The basic structure of this embodiment is the same as that of the first embodiment, and only the electrode connecting method is different. Only the parts of the structure that differ from the first embodiment will be described. In FIG.
One of the two comb-teeth fixed electrodes 54 and 55 (corresponding to 39 and 40 in FIG. 1) facing the comb-teeth movable electrode 38 connected to the movable mass 30 (55) functions as a control electrode.

Next, the operation will be described. The basic operation is the same as that of the first embodiment, but each comb tooth movable electrode 38 and each comb tooth fixed electrode 54 (U, D) with respect to the displacement of the movable mass 30.
Only changes in capacitance between them are detected. And the control electrode 5
5 (CU, CD) has the same potential as the movable mass 30 and serves as a shield electrode, an input acceleration compensation electrode for generating an electrostatic attractive force for compensating the input acceleration, or a pseudo acceleration input electrode for inputting a pseudo acceleration by the electrostatic attractive force. To use.

The effects obtained are the same as those of the first embodiment, but the following effects can also be obtained by providing the control electrode 55. (1) When the control electrode 55 is the shield electrode, it is possible to avoid the offset of the capacitance change due to the adjacent comb-teeth electrodes, and it is possible to secure a larger capacitance change than the conventional example. (2) When the control electrode 55 is the input acceleration compensation electrode, the capacitance detection electrode and the input acceleration compensation electrode are independent, so the detection signal and the compensation signal can be separated, and the system configuration can be simplified. . (3) When the control electrode 55 is the pseudo acceleration input electrode, the capacitance detection electrode and the pseudo acceleration input electrode are independent, so that the detection signal and the pseudo input signal can be separated and the system configuration can be simplified.

In the above-described embodiments, only the acceleration sensor has been described. However, the intended content of the present invention is to use the acceleration sensor as long as the electrical wiring is pulled out from the structure formed by trench etching. Not limited to
It can be applied in the same way and the same effect can be obtained.

[Brief description of drawings]

FIG. 1 is a schematic plan view showing a first embodiment of the present invention.

FIG. 2 is an enlarged schematic view of an essential part of FIG.

3 is a cross-sectional view taken along the line A-A ′ in FIG.

4 is a sectional view taken along the line B-B ′ of FIG.

FIG. 5 is a cross-sectional view showing a part of the manufacturing process of the present invention.

FIG. 6 is a cross-sectional view showing another part of the manufacturing process of the present invention.

FIG. 7 is a cross-sectional view showing another part of the manufacturing process of the present invention.

FIG. 8 shows another part of the manufacturing process of the present invention (D in FIG. 2).
A sectional view showing (corresponding to -D ').

FIG. 9 is a schematic plan view showing a second embodiment of the present invention.

10A and 10B are views for explaining a problem in a conventional example, FIG. 10A is a schematic cross-sectional view of a main part, and FIG. 10B is a schematic perspective view of a main part.

FIG. 11 is a cross-sectional view for explaining a capacitance change between a comb-teeth movable electrode and a comb-teeth fixed electrode.

[Explanation of symbols]

26 ... Wiring 27 ... Jumper wire 28 ... Trench groove 29 ... Through hole surface part 30 ... Movable mass 31 ... Support part 32 ... Fixed part 33 ... Pattern overlapping region 34 ... Fixed part 35 ... Protective film 36 ... Wiring 37 ... Contact hole 38 Comb-tooth movable electrodes 39, 40 ... Comb-tooth fixed electrode 41 ... Protective film 42 ... Contact hole 43 ... Oxide film 44 ... Insulating film 45 ... Movable space gap 46 ... Dielectric 47 ... Semiconductor support substrate 48 ... Semiconductor active layer substrate 49 ... through hole 50 ... frame portion 51 ... V-shaped groove 52 ... etching mask 53 ... etching hole 54 ... comb-teeth fixed electrode 55 ... control electrode 100 ... semiconductor substrate 101 ... groove 102 ... electrode 103 ... etching residue 104 ... removed by etching Part 105 ... Movable parts 106a to 106c ... Comb-tooth movable electrode 107 ... Fixed parts 108a to 108c ... Comb-tooth fixed electrode C ... Possible Wiring U drawn from fixed portion 32 of mass 30 to frame portion 50 ... Comb-tooth movable electrode 38 and wiring D from comb-teeth fixed electrode 40 facing upward in FIG. 1 ... Comb-tooth movable electrode 38 and in FIG. 1. The wiring CU from the fixed comb tooth electrode 39 facing below the wiring CU ... The wiring CD from the control electrode facing the upper comb tooth movable electrode 38 in FIG. Wiring from control electrode

─────────────────────────────────────────────────── ─── Continuation of the front page (56) References JP-A-6-123628 (JP, A) JP-A-4-127479 (JP, A) (58) Fields investigated (Int.Cl. ⁷ , DB name) H01L 29/84 G01L 1/14 G01P 15/125

Claims (6)

(57) [Claims] 1. An active layer substrate having a through hole filled with a first dielectric and a support substrate are integrated with the first dielectric.
The trapezoid has a trapezoidal cross-sectional shape in which the lower bottom is longer than the upper bottom and the lower bottom of the trapezoid is located on the side of the support substrate. The S
An OI substrate is used, and is insulated from the surroundings by the trench and the through hole provided so as to reach from the surface to the bottom of the active layer substrate at a position different from the through hole , and
At least a part of the surface has a movable space gap and the second guide.
At least one conductive region formed so as to face the supporting substrate via an electric body is held by the supporting substrate via the first and second electric conductors. In the semiconductor device, electrical wiring connecting from the conductive region to the outside of the insulated region is arranged so as to cross the first dielectric filled in the through hole on the main surface of the SOI substrate. A semiconductor device characterized by the above. 2. A movable part which is displaced according to the application of a physical quantity,
A plurality of comb-teeth movable electrodes that interlock with the movable portion and a plurality of comb-teeth fixed electrodes that face the comb-teeth movable electrodes with a minute gap on the side surfaces in the longitudinal direction are formed on the SOI substrate, and the comb-shaped electrodes are formed. A semiconductor device which serves as a physical quantity detection sensor having a movable tooth electrode and a fixed comb tooth electrode as the conductive regions, and an electrical connection between the movable tooth comb electrode and the fixed comb electrode to the outside of the insulated range. 2. The semiconductor device according to claim 1, wherein a physical wiring is arranged so as to cross the first dielectric filled in the through hole on the main surface of the SOI substrate. 3. Two or more independent comb-tooth fixed electrodes are arranged for one comb-tooth movable electrode, and the displacement of the movable portion and the comb-tooth movable electrode is determined by a plurality of comb-tooth movable electrodes and a plurality of comb-tooth movable electrodes. The semiconductor device according to claim 2, wherein the semiconductor device is configured to detect as a differential value of electrostatic capacitance between the comb-teeth fixed electrodes. 4. Two or more independent comb-teeth fixed electrodes are arranged for one comb-teeth movable electrode, and at least one comb-teeth movable electrode is arranged.
The semiconductor device according to claim 2, wherein the comb-teeth fixed electrode of the book is configured to be used as a control electrode. 5. A semiconductor active layer substrate having a V-shaped groove on the surface, an oxide film provided on the surface inside and outside the V-shaped groove, and a dielectric provided on the oxide film. , An SOI substrate bonded to a semiconductor support substrate via the dielectric is used, and a surface opposite to the bonding surface of the semiconductor active layer substrate is connected to the V
By grinding to a desired thickness reaching the V-shaped groove, the semiconductor active layer substrate portion has a trapezoidal cross-sectional shape filled with the dielectric and the oxide film, and the lower bottom of the trapezoidal shape is on the supporting substrate side. Having through-holes formed so that
A step of forming a substrate, a step of forming an insulating film on the SOI substrate, and a step of forming a contact hole in a predetermined portion of the insulating film and depositing and patterning a conductor layer to cross the through hole. To form wiring, patterning the insulating film, further depositing and patterning a wiring protective film, and trench etching
After depositing the mask layer, pattern it to fit the structure
And reaches the oxide film on the semiconductor active layer substrate.
A step of forming a pit and a trench groove, and then introducing a hydrofluoric acid based solution from the etching hole.
Part of the oxide film is removed by etching
5. A method of manufacturing a semiconductor device according to claim 1, further comprising: forming a movable space gap between the semiconductor active layer substrate and the dielectric . 6. A step of forming a V-shaped groove by anisotropically etching the semiconductor active layer substrate, and forming an oxide film on the surface of the semiconductor active layer substrate including the inner surface of the V-shaped groove. A step of forming, a step of depositing a dielectric on the oxide film on the surface of the semiconductor active layer substrate, and a step of flattening the surface of the dielectric, and then forming the semiconductor active layer substrate with the dielectric interposed therebetween. A step of forming an SOI substrate by stacking and bonding the semiconductor active substrate on the semiconductor supporting substrate;
By grinding to a desired thickness reaching the V-shaped groove, the semiconductor active layer substrate portion has a trapezoidal cross-sectional shape filled with the dielectric and the oxide film, and the lower bottom of the trapezoidal shape is on the supporting substrate side. Having through-holes formed so that
A step of forming a substrate, a step of forming an insulating film on the SOI substrate, and a step of forming a contact hole in a predetermined portion of the insulating film and depositing and patterning a conductor layer to cross the through hole. So as to form a wiring, patterning the insulating film, further depositing and patterning a wiring protective film, and after depositing a trench etching mask layer, the structure is formed.
It is patterned together and the above is applied to the semiconductor active layer substrate.
Form etching holes and trenches that reach the oxide film
And then introducing a hydrofluoric acid-based solution through the etching hole.
Part of the oxide film is removed by etching
5. A method of manufacturing a semiconductor device according to claim 1, further comprising: forming a movable space gap between the semiconductor active layer substrate and the dielectric .