JP2009004543A - Mechanical quantity detecting device and manufacturing method thereof - Google Patents

Abstract

<P>PROBLEM TO BE SOLVED: To provide a method of manufacturing a mechanical quantity detecting device of small size. <P>SOLUTION: The mechanical quantity detecting device is fabricated from a laminate substrate comprising a base layer, a sacrifice layer, and an active layer. The manufacturing method thereof includes a process for anisotropic etching the active layer, a process for forming an upper sacrifice layer on the laminate substrate, a process for growing a circuit layer on the upper sacrifice layer, a process for forming a semiconductor circuit in the circuit layer, a process for forming a through hole in the circuit layer, a process for isotropic etching the sacrifice layer and the upper sacrifice layer by introducing an etching agent through the through hole, and a process for forming a wiring part which electrically connects a connection region to the semiconductor circuit through the through hole. The position of the through hole and the etching time in isotropic etching are set so that a movable part can move, the sacrifice layer connecting a fixed part to the base layer remains, the sacrifice layer connecting a support part to the base layer remains, and the upper sacrifice layer connecting the support part to the circuit layer remains. <P>COPYRIGHT: (C)2009,JPO&INPIT

Description

  The present invention relates to a mechanical quantity detection device.

  A technique for manufacturing a mechanical quantity detection device has been developed from a laminated substrate in which a base layer, a sacrificial layer, and an active layer are laminated in that order. Such a mechanical quantity detection device has a fixed part made of an active layer fixed to a base layer through a sacrificial layer, and is free from the base layer because it extends from the fixed part and the sacrificial layer is removed. The movable part which consists of an active layer is provided. When a mechanical quantity such as acceleration, angular velocity, angular acceleration, external force, pressure, or the like acts on the mechanical quantity detection device, the movable part is displaced with respect to the base layer. The mechanical quantity detection device has a function of detecting the displacement amount of the movable part, and detects the mechanical quantity acting on the mechanical quantity detection device based on the detected displacement amount.

  Patent Document 1 discloses a mechanical quantity detection device in which a dynamic quantity detection unit and a semiconductor circuit are formed on one base layer (silicon substrate). The mechanical quantity detection unit has a movable part (movable weight, elastic beam) and a fixed part (fixed beam). The fixing part is fixed on the base layer. The movable part is supported by the fixed part and is separated from the base layer. The fixed part is electrically connected to the semiconductor circuit. Therefore, the movable part is electrically connected to the semiconductor circuit via the fixed part. Further, a lower electrode (etching stopper layer) is formed on the base layer below the movable part in a non-contact positional relationship with the movable part. That is, a capacitor is formed by the movable part and the lower electrode. The lower electrode is electrically connected to the semiconductor circuit. That is, a capacitor formed by the movable part and the lower electrode is connected to the semiconductor circuit.

  When acceleration acts on the mechanical quantity detection device, the movable part is displaced with respect to the base layer. Then, the interval between the movable part and the lower electrode changes. Then, the capacitance of the capacitor formed by the movable part and the lower electrode changes. This change in capacitance is detected by the semiconductor circuit. That is, the acceleration acting on the movable part is detected by the semiconductor circuit as a change in the capacitance of the capacitor.

  In this mechanical quantity detection device, the mechanical quantity detection unit and the semiconductor circuit are formed on one base layer. Therefore, the mechanical quantity detection device can be reduced in size compared to the case where the mechanical quantity detection unit and the semiconductor circuit are formed on different substrates.

JP 2006-263902 A

  According to the technique of Patent Document 1, the mechanical quantity detection device can be reduced in size. However, further miniaturization is required.

  The present invention has been made in view of the above-described circumstances, and provides a method for manufacturing a smaller dynamic quantity detection device while a dynamic quantity detection unit and a semiconductor circuit are formed on the same chip.

In the method for manufacturing a mechanical quantity detection device of the present invention, a mechanical quantity having a fixed portion, a movable portion, a support portion, and a circuit layer is obtained from a laminated substrate in which a base layer, a sacrificial layer, and an active layer are laminated in that order. A detection device is manufactured. The fixed part of the manufactured mechanical quantity detection device is composed of an active layer fixed to the base layer via a sacrificial layer. The movable part of the manufactured mechanical quantity detection device includes an active layer that extends from the fixed part and is free from the base layer because the sacrificial layer is removed. The support portion of the manufactured mechanical quantity detection device includes an active layer fixed to the base layer via a sacrificial layer. The circuit layer of the manufactured mechanical quantity detection device is fixed on the support part, extends above the movable part in a non-contact positional relationship with the movable part, and is electrically connected to the fixed part. A semiconductor circuit is provided. The manufacturing method of the mechanical quantity detection device of the present invention includes an anisotropic etching step, an upper sacrificial layer forming step, a circuit layer growing step, a semiconductor circuit forming step, a through-hole forming step, and an isotropic etching step. And a wiring part forming step. In the anisotropic etching process, a trench reaching the sacrificial layer is formed by anisotropically etching the active layer along the contour of the movable portion in a range excluding the connection portion with the fixed portion. In the upper sacrificial layer forming step, an upper sacrificial layer covering the active layer is formed on the laminated substrate after the anisotropic etching step. In the circuit layer growth step, a circuit layer is grown on the upper sacrificial layer. In the semiconductor circuit formation step, a semiconductor circuit is formed in the circuit layer. In the through hole forming step, a through hole penetrating from the upper surface to the lower surface of the circuit layer is formed at least at a position above the fixed portion in the circuit layer. In the isotropic etching step, an etching agent is introduced from the through hole to perform isotropic etching of the sacrificial layer and the upper sacrificial layer. The position of the through-hole formed in the through-hole forming step and the etching time in the isotropic etching step are the following conditions:
Removing the sacrificial layer between the moving part and the base layer to separate the moving part and the base layer;
Removing the upper sacrificial layer between the moving part and the circuit layer to separate the moving part and the circuit layer;
-The upper sacrificial layer filling between the movable part and the fixed part and the upper sacrificial layer filling between the movable part and the support part are removed, and the movable part is movable with respect to both the fixed part and the support part. Do;
-A sacrificial layer connecting the fixed part and the base layer is left below the fixed part;
-A sacrificial layer that connects the fixed part and the base layer is left below the support part;
The upper sacrificial layer connecting the support and the circuit layer is left on the upper part of the support;
Set to satisfy the condition. In the wiring portion forming step, a wiring portion that passes through the through hole located above the fixing portion and electrically connects the fixing portion and the semiconductor circuit is formed.
The fixing part and the supporting part may be formed so that the fixing part and the supporting part are continuous (that is, the fixing part and the supporting part are integrated), or the fixing part and the supporting part are formed separately. You may do it.
When a plurality of fixed parts are formed on the base layer, there may be a fixed part where the movable part does not extend (that is, a fixed part that is separated from the movable part and does not support the movable part). .

In the anisotropic etching step, a trench is formed along the contour of the movable portion in a range excluding the connection portion with the fixed portion, and the movable portion, the fixed portion, and the support portion are formed in the active layer. When separating the fixed part and the support part, a trench is also formed between the fixed part and the support part. In the upper sacrificial layer forming step, an upper sacrificial layer covering the active layer is formed. That is, the upper sacrificial layer is formed in the trench (ie, in the trench between the movable part and the fixed part, in the trench between the fixed part and the support part, and in the trench between the movable part and the support part) and the active layer. Form on top. In the circuit layer forming step, a circuit layer is formed on the upper sacrificial layer. As a result, a stacked structure in which the base layer, the sacrificial layer, the active layer, the upper sacrificial layer, and the circuit layer are stacked in this order is formed (however, the upper sacrificial layer is also present in the trench). In the semiconductor circuit formation step, a semiconductor circuit is formed in the circuit layer. In the through hole forming step, a through hole penetrating from the upper surface to the lower surface of the circuit layer is formed at a position above the fixing portion. In addition, you may form a through-hole in the circuit layer of the position above a movable part as needed. In the isotropic etching step, an etching agent is introduced from the through hole formed in the through hole forming step, and the sacrificial layer and the upper sacrificial layer are isotropically etched. The position of the through-hole formed in the through-hole forming step and the etching time in the isotropic etching step are set so as to satisfy the above conditions. Accordingly, after the isotropic etching step, each part is in the following state. First, the fixed part and the support part are maintained in a state of being fixed to the base layer via the sacrificial layer. Since the circuit layer is connected to the support portion and separated from the movable portion by the upper sacrificial layer, the circuit layer extends above the movable portion in a non-contact positional relationship with the movable portion. The movable portion is in a state where the surrounding sacrificial layer and the upper sacrificial layer are removed, and the movable portion is connected to the fixed portion only at the connection portion. That is, the movable part is in non-contact with the circuit layer, extends in a plane parallel to the base layer from the fixed part, and is free from the base layer. Further, since the upper sacrificial layer is removed from the through hole formed above the fixed portion, the upper surface of the fixed portion is exposed below the through hole. In the wiring portion forming step, a wiring portion that passes through the through hole located above the fixing portion and electrically connects the fixing portion and the semiconductor circuit is formed. Thereby, the mechanical quantity detection device is completed. There may be a fixed part that is not connected to the movable part. Further, the upper sacrificial layer above the fixed portion may be removed in the entire range, or only the vicinity of the through hole may be removed. Further, the support part and the fixing part may be continuous or separated.
As described above, according to the manufacturing method of the mechanical quantity detection device, the fixed portion fixed to the base layer, the movable portion extending from the fixed portion and free from the base layer, and the sacrificial layer A support portion made of an active layer fixed to the base layer via a circuit layer, and a circuit layer fixed on the support portion and extending above the movable portion in a non-contact positional relationship with the movable portion. A mechanical quantity detection device can be manufactured. In the manufactured mechanical quantity detection device, a semiconductor circuit is formed in a circuit layer stacked above the mechanical quantity detection unit (movable unit and fixed unit). Therefore, according to this manufacturing method, it is possible to manufacture a smaller mechanical quantity detection device than a conventional mechanical quantity detection device (a semiconductor circuit is formed next to the mechanical quantity detection unit).

In the method for manufacturing the mechanical quantity detection device described above, it is preferable to perform a step of planarizing the upper surface of the upper sacrificial layer before the circuit layer growth step.
According to such a configuration, the circuit layer can be formed flat in the circuit layer forming step. Therefore, a semiconductor circuit can be suitably formed in the circuit layer.

In the manufacturing method of the mechanical quantity detection device described above, the step of forming the insulating film on the upper surface of the circuit layer and the wall surface of the through hole is performed after the through hole forming step and before the wiring portion forming step. Is preferred.
According to such a configuration, it is possible to prevent the wiring portion from being electrically connected to the circuit layer at an unintended location.

In the manufacturing method of the mechanical quantity detection device described above, it is necessary to leave a sacrificial layer connecting the fixed part and the base layer at the lower part of the fixed part. On the other hand, it is necessary to remove the sacrificial layer and the upper sacrificial layer around the movable part to separate the movable part from the base layer, the circuit layer, the fixed part, and the support part. Therefore, a wider range of sacrificial layers must be etched around the movable part than around the fixed part.
Therefore, in the manufacturing method of the mechanical quantity detection device described above, it is preferable that a plurality of through holes are dispersed and formed in the circuit layer located above the movable portion in the through hole forming step.
According to such a configuration, the sacrificial layer and the upper sacrificial layer around the movable part can be removed earlier in the isotropic etching process. Therefore, even when a mechanical quantity detection device having a large area of the movable part is manufactured, the sacrificial layer and the upper sacrificial layer around the movable part are removed before the sacrificial layer below the fixed part is completely removed. Can be removed.

In the manufacturing method of the mechanical quantity detection device described above, it is preferable that in the anisotropic etching step, a plurality of communication holes for communicating the upper surface and the lower surface are dispersed and formed in the active layer forming the movable part.
When the communication holes are dispersedly formed in the active layer of the movable part, etching proceeds in the thickness direction of the active layer from the communication holes arranged in a dispersed manner in the isotropic etching process. Therefore, the sacrificial layer below the fixed portion and the support portion can remain, and the sacrificial layer below the movable portion can be removed.

The present invention also provides a new mechanical quantity detection device. This mechanical quantity detection device includes a base layer, a fixed part, a movable part, a support part, a circuit layer, and a wiring part. The fixed part is made of an active layer fixed to the base layer via a sacrificial layer. The movable part consists of an active layer extending from the fixed part and free from the base layer. The support part is made of an active layer fixed to the base layer via a sacrificial layer. The circuit layer is fixed to the support portion via the upper sacrificial layer, extends above the movable portion in a non-contact positional relationship with the movable portion, and penetrates from the upper surface to the lower surface in a range located above the fixed portion. A through-hole is formed, and a semiconductor circuit is formed. The wiring portion passes through the through hole and electrically connects the fixed portion and the semiconductor circuit.
According to such a configuration, a smaller mechanical quantity detection device can be provided.

  According to the present invention, a smaller mechanical quantity detection device can be provided.

The main features of the embodiments described in detail below are listed first.
(Feature 1) The movable part has a beam that extends from the fixed part and can swing within a plane parallel to the base layer, and a vibration part (weight part) connected to the tip of the beam.
(Feature 2) The semiconductor circuit has a displacement amount detection circuit that detects the displacement amount of the vibration part.
(Feature 3) The semiconductor circuit has a drive signal output circuit that outputs an electrical signal that vibrates the vibration part.

  An embodiment in which the present invention is applied to a method of manufacturing an angular velocity detection device will be described with reference to the drawings. First, the angular velocity detection device manufactured by the manufacturing method of the present embodiment will be described.

FIG. 1 shows a partial cross-sectional view of an angular velocity detection device 10 of the embodiment. As illustrated, the angular velocity detection device 10 has a stacked structure in which a base layer 100, a sacrificial layer 102, an active layer 104, an upper sacrificial layer 106, and a circuit layer 108 are stacked. The active layer 104 is free from the base layer 100, the first part (fixed part 110) fixed to the base layer 100, the second part (support part 111) fixed to the base layer 100. There is a part (movable part 112). A sacrificial layer 102 is formed in the lower portion of the fixing part 110, and is fixed to the base layer 100 by the sacrificial layer 102. A sacrificial layer 102 is formed in the lower portion of the support portion 111 and is fixed to the base layer 100 by the sacrificial layer 102. The movable portion 112 does not have the sacrificial layer 102 formed on the lower portion. That is, a space is formed between the movable part 112 and the base layer 100. As will be described in detail later, the active layer 104 constituting the movable portion 112 is connected to the active layer 104 constituting the fixed portion 110 within a range not shown. Thereby, the movable part 112 is supported in a state of being separated from the base layer 100.
The upper sacrificial layer 106 is not formed on the fixed portion 110. The upper sacrificial layer 106 is not formed on the movable portion 112. An upper sacrificial layer 106 is formed on the support portion 111. The support part 111 is connected to the circuit layer 108 via the upper sacrificial layer 106. The circuit layer 108 is a layer having substantially the same area as the base layer 100 and faces the base layer 100. The circuit layer 108 is fixed to the support portion 111 by the upper sacrificial layer 106. The circuit layer 108 is supported in a non-contact positional relationship with respect to the fixed portion 110 and the movable portion 112. The circuit layer 108 and the fixing part 110 are connected by a wiring part 122.
FIG. 2 shows an enlarged view of the circuit layer 108. As illustrated, a semiconductor circuit 130 is formed on the upper surface side of the circuit layer 108. Note that reference numeral 132 in FIG. 2 indicates the gate electrode 132 constituting the semiconductor circuit 130. The semiconductor circuit 130 is covered with an interlayer insulating film 134. A passivation film 136 is formed on the interlayer insulating film 134.
As shown in FIG. 1, a plurality of through holes 120 penetrating from the upper surface to the lower surface of the circuit layer 108 are formed in a range of the circuit layer 108 that is not in contact with the upper sacrificial layer 106 on the support portion 111. ing. Each through hole 120 is formed at a position where it does not interfere with the semiconductor circuit 130. As shown in FIG. 2, an interlayer insulating film 134 is formed on the wall surface of the through hole 120. Of the plurality of through holes 120, the wiring part 122 is formed so as to pass through the through hole 120 located above the fixed part 110. The wiring part 122 electrically connects the lower fixing part 110 and the semiconductor circuit 130. Hereinafter, a region of the fixing unit 110 that is connected to the wiring unit 122 is referred to as a connection region 124.

The base layer 100, the active layer 104, and the circuit layer 108 are made of single crystal silicon. Impurities are introduced into the single crystal silicon. Therefore, the base layer 100, the active layer 104, and the circuit layer 108 are conductive. In particular, the active layer 104 is highly doped with impurities and has a very high conductivity. The sacrificial layer 102 is made of SiO ₂ and is an insulator. The upper sacrificial layer 106 is made of SiO and is an insulator. Therefore, the active layer 104 is insulated from the base layer 100. The active layer 104 is electrically connected to the circuit layer 108 only in the connection region 124.

  FIG. 3 is a plan view of the angular velocity detection device 10 with the circuit layer 108 removed, as viewed from the top side. In FIG. 3, the horizontal direction in FIG. 3 is referred to as the X direction, and the vertical direction in FIG. 3 is referred to as the Y direction. The base layer 100 has the same shape as the outer peripheral shape of the angular velocity detection device 10 shown in FIG. On the base layer 100, the above-described fixed portion 110, support portion 111, and movable portion 112 are formed. In FIG. 3, the fixed part 110 is indicated by a diagonal line pattern, the support part 111 is indicated by a vertical line pattern, and the movable part 112 is indicated by a dot pattern.

  As illustrated, the support portion 111 is formed along the outer peripheral edge of the base layer 100. As described above, the circuit layer 108 having substantially the same area as the base layer 100 is fixed to the upper portion of the support portion 111. That is, a box-shaped structure is formed by the base layer 100, the support portion 111, and the circuit layer 108. A fixed portion 110 and a movable portion 112 are formed in the space inside the box-shaped structure.

As shown in FIG. 3, base portions 50 a to 50 d that are the fixing portions 110 are formed on the base layer 100. The bases 50a and 50b are formed near one end of the base layer 100 (the upper end in FIG. 3). The bases 50c and 50d are formed near the other end of the base layer 100 (the end in the downward direction in FIG. 3). The base portion 50 c is connected to the connection region 124 a by the extension portion 58.
The movable part 112 is connected to the bases 50a to 50d. As a result, the movable portion 112 is supported in a floating state on the base layer 100. The movable portion 112 includes beams 30a to 30d, sub frames 24a and 24b, beams 28a to 28d, main frames 22a and 22b, beams 26a to 26d, and a vibrator (weight portion) 20.

  The beams 30a and 30b extend in the Y direction (upward in FIG. 3) from the base portions 50a and 50b. A subframe 24a is connected to the distal ends of the beams 30a and 30b. From the subframe 24a, beams 28a and 28b extend in the Y direction (downward in FIG. 3). A main frame 22a is connected to the distal ends of the beams 28a and 28b. As shown, the beams 30a, 30b, 28a, 28b are long in the Y direction and narrow in the X direction. Therefore, the beams 30 a, 30 b, 28 a, 28 b can be bent in a plane parallel to the base layer 100. Accordingly, the main frame 22a can be displaced in the X direction with respect to the base layer 100.

  The beams 30c and 30d extend in the Y direction (downward in FIG. 3) from the base portions 50c and 50d. A subframe 24b is connected to the distal ends of the beams 30c and 30d. From the subframe 24b, beams 28c and 28d extend in the Y direction (upward in FIG. 3). A main frame 22b is connected to the distal ends of the beams 28c and 28d. As illustrated, the beams 30c, 30d, 28c, and 28d are long in the Y direction and narrow in the X direction. Therefore, the beams 30 c, 30 d, 28 c, 28 d can be bent in a plane parallel to the base layer 100. Accordingly, the main frame 22b can be displaced in the X direction with respect to the base layer 100.

As illustrated, the vibrator 20 is located between the main frame 22a and the main frame 22b.
One end portion (upper end portion in FIG. 3) of the vibrator 20 is connected to the main frame 22a by a beam 26a and a beam 26b extending in the X direction. The other end portion (the lower end portion in FIG. 3) of the vibrator 20 is connected to the main frame 22b by beams 26c and 26d extending in the X direction. The beams 26a to 26d are long in the X direction and narrow in the Y direction. Therefore, the beams 26 a to 26 d can be bent in a plane parallel to the base layer 100. Since the beams 26a to 26d can be bent, the vibrator 20 can be displaced in the Y direction with respect to the main frames 22a and 22b.

  Next, the capacitor structure formed in each part of the angular velocity detection device 10 will be described.

The vibrator 20 includes a body portion 20a and arms 20b to 20e extending from the body portion 20a. Four plate-like portions 32a extending in the X direction are formed in the vicinity of the distal end and the proximal end of the arm 20b. A plate-like portion 52a fixed to the base layer 100 is formed at a position facing each plate-like portion 32a. That is, the capacitor 80a is formed by the four plate-like portions 32a and the four plate-like portions 52a. The plate-like portion 52a is connected to the connection region 124b by the extending portion 54a.
Similarly, a capacitor 80b is formed in the vicinity of the arm 20c by the four plate-like portions 32b and the four plate-like portions 52b. The plate-like part 52b is connected to the connection region 124c by the extension part 54b.
Similarly, a capacitor 80c is formed in the vicinity of the arm 20d by the four plate-like portions 32c and the four plate-like portions 52c. The plate-like portion 52c is connected to the connection region 124d by the extending portion 54c.
Similarly, a capacitor 80d is formed in the vicinity of the arm 20e by the four plate-like portions 32d and the four plate-like portions 52d. The plate-like portion 52d is connected to the connection region 124e by the extending portion 54d.

A plurality of plate-like portions 34a extending in the X direction are formed at one end in the X direction of the main frame 22a. A plate-like portion 62a fixed to the base layer 100 is formed at a position facing each plate-like portion 34a. Thereby, a capacitor 82a is formed. The plate-like portion 62a is connected to the connection region 124f by the extending portion 64a.
Similarly, a capacitor 82b is formed at the other end in the X direction of the main frame 22a by a plurality of plate-like portions 34b and a plurality of plate-like portions 62b. The plate-like part 62b is connected to the connection region 124g by the extension part 64b.
Similarly, a capacitor 82c is formed by a plurality of plate-like portions 34c and a plurality of plate-like portions 62c at one end in the X direction of the main frame 22b. The plate-like portion 62c is connected to the connection region 124h by the extending portion 64c.
Similarly, a capacitor 82d is formed by the plurality of plate-like portions 34d and the plurality of plate-like portions 62d at the other end in the X direction of the main frame 22b. The plate-like portion 62d is connected to the connection region 124i by the extending portion 64d.

As described above, each of the connection regions 124 (that is, the connection regions 124a to 124i) is electrically connected to the semiconductor circuit 130 of the circuit layer 108 by the wiring portion 122 (therefore, in FIG. Although only one is shown, nine wiring portions 122 are formed in the acceleration detection device 10). The semiconductor circuit 130 includes a drive signal output circuit 130a and a capacitance detection circuit 130b. As shown in FIG. 3, the connection regions 124f to 124i are connected to the drive signal output circuit 130a. The connection regions 124a to 124e are connected to the capacitance detection circuit 130b.
The drive signal output circuit 130a outputs an AC signal having a predetermined frequency between the connection regions 124f to 124g and between the connection regions 124h to 124i.
The capacitance detection circuit 130b detects the capacitances of the capacitors 80a to 80d between the connection regions 124a to 124b, between the connection regions 124a to 124c, between the connection regions 124a to 124d, and between the connection regions 124a to 124e.

  Although not shown in FIG. 3, a portion of the movable portion 112 excluding width details (beams 26 a to 26 d, beams 28 a to 28 d, beams 30 a to 30 d, plate-like portions 32 a to 32 d, plate-like portions 34 a to 34 d). A plurality of communication holes are formed in the. FIG. 4 shows an enlarged view of a part of the movable portion 112. As illustrated, the movable portion 112 has a plurality of communication holes 150 formed therein. The communication hole 150 communicates the upper surface and the lower surface of the movable part 112 (active layer 104). The communication hole 150 has a substantially square cross-sectional shape. The communication holes 150 are formed at substantially equal intervals in the X direction and the Y direction.

  Next, the operation of the angular velocity detection device 10 will be described. When the angular velocity detection device 10 is used, the drive signal output circuit 130a outputs an AC signal having a predetermined frequency between the connection regions 124f to 124g and between the connection regions 124h to 124i. As a result, charge and discharge of charges are repeated in the capacitors 82a to 82d. Then, the movable part 112 receives the force which vibrates in the X direction by the electrostatic force of the electric charge charged and discharged by the capacitors 82a to 82d. As described above, the movable portion 112 is supported so as to be displaceable in the X direction with respect to the base layer 100. Therefore, the movable part 112 vibrates in the X direction by the electrostatic force of electric charges.

When the angular velocity detecting device 10 is moved at a predetermined angular velocity (angular velocity around an axis perpendicular to the paper surface of FIG. 3) in a state where the movable portion 112 is vibrating in the X direction, the vibrator 20 is moved in the Y direction. Coriolis force acts. The magnitude and direction of the Coriolis force vary according to the magnitude of the angular velocity and the vibration of the movable portion 112 in the X direction. Further, as described above, the vibrator 20 is supported so as to be displaceable in the Y direction with respect to the main frames 22a and 22b. Therefore, the vibrator 20 vibrates in the Y direction due to the variation of the Coriolis force. The amplitude of vibration in the Y direction of the vibrator 20 is an amplitude corresponding to the acting angular velocity. When the vibrator 20 vibrates in the Y direction, the capacities of the capacitors 80a to 80d change.
The capacitance detection circuit 130b detects the capacitances of the capacitors 80a to 80d (that is, changes in capacitance) between the connection regions 124a and 124b, between the connection regions 124a and 124c, between the connection regions 124a and 124d, and between the connection regions 124a and 124e. . The capacitance detection circuit 130b detects the amplitude of vibration in the Y direction of the vibrator 20 from the detected capacitance. Then, the angular velocity acting on the angular velocity detection device 10 is calculated from the detected amplitude.

  Next, a method for manufacturing the angular velocity detection device 10 will be described. FIG. 5 shows a manufacturing process of the angular velocity detection device 10. The angular velocity detection device 10 is manufactured from a multilayer substrate 300 having a cross-sectional structure shown in FIG. FIG. 6 shows a cross section corresponding to the cross section of the angular velocity detection device 10 shown in FIG. The multilayer substrate 300 includes a base layer 100, a sacrificial layer 102 stacked on the base layer 100, and an active layer 104 stacked on the sacrificial layer 102.

In the anisotropic etching step S2, the active layer 104 is etched by reactive ion etching (RIE). The active layer etching step S2 is performed by a method in which the active layer 104 (ie, silicon) is etched and the sacrificial layer 102 (ie, SiO ₂ ) is not etched. The active layer etching step S2 is performed by selecting an etching range, for example, by forming an etching mask on the active layer 104. In the active layer etching step S <b> 2, the active layer 104 is etched in a range (a range that is not hatched in FIG. 3) except for the range that becomes the fixed portion 110, the support portion 111, and the movable portion 112. Further, a minute region corresponding to the communication hole 150 in FIG. These etching ranges are etched until the sacrificial layer 102 is reached (that is, a trench is formed in the multilayer substrate 300). According to RIE, etching proceeds substantially perpendicular to the active layer 104. As a result, the outer shape of the fixed portion 110, the support portion 111, and the movable portion 112 is formed, and the communication hole 150 is formed in a range that becomes the movable portion 112. Therefore, the laminated substrate 300 is formed as shown in FIG.

  When the anisotropic etching process is completed, a material of the same type as the sacrificial layer 102 is grown on the laminated substrate 300 by the CVD method or the like (upper sacrificial layer forming process S4). As a result, the upper sacrificial layer 106 is formed on the multilayer substrate 300. In the upper sacrificial layer growth step S4, as shown in FIG. 8, the upper sacrificial layer 106 is grown until the trench formed by the anisotropic etching step is filled with the upper sacrificial layer 106 and the active layer 104 is covered with the upper sacrificial layer 106. Let

  When the upper sacrificial layer growth step S4 is completed, the upper surface of the upper sacrificial layer 106 is etched. As a result, as shown in FIG. 9, the upper surface of the upper sacrificial layer 106 is flattened (upper sacrificial layer flattening step S6). Note that the upper surface of the upper sacrificial layer 106 may be planarized by polishing or the like.

  After the upper surface of the upper sacrificial layer 106 is planarized, a circuit layer 108 made of single crystal silicon is grown on the upper sacrificial layer 106 by a CVD method or the like (circuit layer growth step S8). Since the upper surface of the upper sacrificial layer 106 is flattened, the circuit layer 108 is formed in a substantially planar shape as shown in FIG.

  After the circuit layer 108 is formed, a semiconductor circuit 130 is formed on the circuit layer 108 (semiconductor circuit forming step S10). That is, an integrated circuit including transistors and wirings is formed, and the semiconductor circuit 130 including the drive signal output circuit 130a and the capacitance detection circuit 130b is formed. The semiconductor circuit 130 is formed so as not to interfere with a through hole 120 to be formed later. Since the formation of the semiconductor circuit 130 can be performed by a conventionally known method, description thereof is omitted.

When the semiconductor circuit 130 is formed, through holes 120 are formed in the circuit layer 108 by RIE as shown in FIG. 11 (through hole forming step S12). The through hole 120 is formed according to at least the following conditions.
-It is not formed in the range above the support part 111 among the circuit layers 108.
-It forms in the position above the connection area | region 124 (124a-124i) of the fixing | fixed part 110 among the circuit layers 108. FIG.
In the circuit layer 108, a plurality of through holes 120 are dispersedly formed in a range above the movable portion 112.
A plurality of through holes 120 are dispersedly formed in the active layer removal range (the range 160 in FIG. 11 and the non-hatched range in FIG. 3) of the circuit layer 108 where the active layer 104 is removed. .
The through-hole 120 is formed so as to satisfy the conditions of an isotropic etching process described later in addition to the above rules. The position, size, and shape of the through hole 120 are set so as to satisfy these rules. In this embodiment, the through hole 120 is formed so that the cross-sectional shape is a circle having a diameter of about 1 μm.

After the through holes 120 are formed, hydrofluoric acid (HF) gas (etching agent) is introduced into each through hole 120 to etch the sacrificial layer 102 and the upper sacrificial layer 106 (isotropic etching step S14). According to the hydrofluoric acid gas, the sacrificial layer 102 (ie, SiO ₂ ) and the upper sacrificial layer 106 (ie, SiO) are etched, but the base layer 100, the active layer 104, and the circuit layer 108 (ie, single crystal silicon). Is hardly etched. Etching of the sacrificial layer 102 and the upper sacrificial layer 106 with hydrofluoric acid gas proceeds isotropically. In the isotropic etching process, the etching time is set so as to satisfy the following conditions.
The sacrificial layer 102 between the movable part 112 and the base layer 100 is removed, and the movable part 112 and the base layer 100 are separated.
The upper sacrificial layer 106 between the movable part 112 and the circuit layer 108 is removed, and the movable part 112 and the circuit layer 108 are separated.
The upper sacrificial layer 106 that fills the space between the movable portion 112 and the fixed portion 110 and the upper sacrificial layer 106 that fills the space between the movable portion 112 and the support portion 111 are removed, and the movable portion 112 and the fixed portion 110 are removed. The movable portion 112 is movable with respect to the support portion 111 (that is, the movable portion 112 is connected to the fixed portion 110 only at the connection portion directly connected to the base portions 50a to 50d and is separated from the support portion 111). And).
The sacrificial layer 102 that connects the fixing part 110 and the base layer 100 is left below the fixing part 110.
The sacrificial layer 102 that connects the support part 111 and the base layer 100 is left below the support part 111.
The upper sacrificial layer 106 that connects the support part 111 and the circuit layer 108 is left on the support part 111.
-The upper sacrificial layer 106 above the connection regions 124a to 124i is removed.
That is, in the through hole forming step S12, each through hole 120 is formed in a positional relationship that can satisfy this condition. Then, the sacrificial layer 102 and the upper sacrificial layer 106 are etched so as to satisfy the above conditions by adjusting the etching time in the isotropic etching step S14.

The isotropic etching step S14 will be described in more detail. FIG. 12 shows a state of the multilayer substrate 300 during the isotropic etching step S14.
As shown in the drawing, in the active layer removal range 160, etching proceeds isotropically from the through hole 120 in the active layer removal range 160. That is, etching proceeds both in the thickness direction and in the lateral direction. Therefore, the sacrificial layer 102 and the upper sacrificial layer 106 in the active layer removal range 160 are removed in a short time.
Etching proceeds from the through hole 120 above the movable portion 112 around the movable portion 112. At this time, etching proceeds in the thickness direction through the communication hole 150. Therefore, the etching proceeds both in the thickness direction and in the lateral direction around the movable portion 112. Therefore, the sacrificial layer 102 and the upper sacrificial layer 106 around the movable portion 112 are removed in a short time.
In the periphery of the fixing portion 110, etching proceeds from the through hole 120 (the through hole 120a in FIG. 12) above the connection region 124. Etching from the through hole 120a does not proceed in the thickness direction because the fixing portion 110 exists. Therefore, the etching from the through hole 120a proceeds in the lateral direction. Accordingly, the upper sacrificial layer 106 above the fixed portion 110 is removed in a short time, but the lower sacrificial layer 102 below the fixed portion 110 is difficult to remove. The sacrificial layer 102 below the fixed portion 110 is etched only from the side. Therefore, as shown in FIG. 13, the sacrificial layer 102 remains under the fixed portion 110.
The through hole 120 is not formed above the support portion 111. Therefore, the etching only proceeds from the through hole 120 of the adjacent active layer removal range 160 around the support portion 111. Therefore, the sacrificial layer 102 and the upper sacrificial layer 106 around the support portion 111 are difficult to remove. Therefore, as shown in FIG. 13, the sacrificial layer 102 remains in the lower portion of the support portion 111, and the upper sacrificial layer 106 remains in the upper portion of the support portion 111.

  As described above, the sacrificial layers (sacrificial layer 102 and upper sacrificial layer 106) around the support portion 111 and the sacrificial layer 102 below the fixed portion 110 are in a positional relationship that is difficult to remove. Therefore, by adjusting the etching time of the isotropic etching step S14, the sacrificial layers (sacrificial layer 102 and upper sacrificial layer 106) around the support portion 111 and the sacrificial layer 102 below the fixed portion 110 are left, and the like. The sacrificial layer 102 and the upper sacrificial layer 106 in the range can be removed. That is, the sacrificial layer 102 and the upper sacrificial layer 106 can be removed while satisfying the conditions of the isotropic etching step S14 described above. When the isotropic etching step S14 is completed, the laminated substrate 300 is in a state shown in FIG.

When the isotropic etching step S14 is completed, an interlayer insulating film 134 (see FIG. 2) is formed on the circuit layer 108 and the wall surface of the through hole 120 by CVD (interlayer insulating film forming step S16). At this time, the interlayer insulating film 134 is not formed in the connection region of the semiconductor circuit 130 (region for electrically connecting to the wiring portion 122 or the outside) in the upper surface of the circuit layer 108.
Note that when the interlayer insulating film 134 is formed, the through-hole 120 may be filled with the formed interlayer insulating film 134. In addition, an interlayer insulating film 134 may be formed on the connection region 124 of the fixed part 110. In such a case, only the central portion of the through hole 120a above the connection region 124 is selectively dry etched. By performing etching in this way, the interlayer insulating film 134 can remain on the wall surface of the through hole 120a and the buried through hole 120a can be opened. Further, the connection region 124 is also etched by etching the central portion of the through hole 120a. Therefore, the interlayer insulating film 134 formed on the connection region 124 can be removed.

  After the interlayer insulating film 134 is formed, a wiring portion 122 that passes through the through hole 120a above the connection region 124 (that is, the connection regions 124a to 124i) and connects the connection region 124 and the semiconductor circuit 130 is formed (wiring portion). Forming step S18). The wiring part 122 is formed by sputtering aluminum from the upper surface side of the multilayer substrate 300. That is, aluminum is grown on the connection region 124 by performing sputtering on the through hole 120a. Aluminum is grown until it reaches the upper opening of the through hole 120a. Then, an aluminum wiring is formed between the upper end of the grown aluminum and the connection region of the semiconductor circuit 130. Thus, by forming the wiring part 122, the connection region 124 and the semiconductor circuit 130 are electrically connected.

  After the wiring part 122 is formed, a passivation film 136 is formed on the circuit layer 108 by CVD (passivation film forming step S20). Thus, the angular velocity detection device 10 shown in FIGS. 1 and 3 is manufactured.

  As described above, according to this manufacturing method, the angular velocity detection device 10 can be manufactured. In the angular velocity detection device 10, a circuit layer 108 in which a semiconductor circuit 130 is formed is stacked above an angular velocity detection unit including a fixed unit 110 and a movable unit 112. Therefore, the angular velocity detection device 10 can be reduced in size by the area of the semiconductor circuit compared to the angular velocity detection device in which the semiconductor circuit is formed next to the angular velocity detection unit.

  In the method of manufacturing the angular velocity detecting device 10 described above, the upper sacrificial layer planarization step S4 for planarizing the upper surface of the upper sacrificial layer 106 is performed before the circuit layer growth step S8. Therefore, the circuit layer 108 can be formed flat in the circuit layer forming step S8. Therefore, the semiconductor circuit 130 can be preferably formed in the circuit layer 108.

  In the manufacturing method of the angular velocity detecting device 10 described above, the interlayer insulating film forming step S16 is performed after the through hole forming step S12 and before the wiring portion forming step S18. In the interlayer insulating film forming step S <b> 16, outside the connection region of the semiconductor circuit 130 on the upper surface of the circuit layer 108 (outside the region where the semiconductor circuit is electrically connected to the wiring portion 122 or the outside) and the wall surface of the through hole 120. An interlayer insulating film 134 is formed. Therefore, the wiring portion 122 can be prevented from conducting with the circuit layer 108 at an unintended location.

  In the manufacturing method of the angular velocity detecting device 10 described above, the plurality of through holes 120 are formed in a distributed manner in the circuit layer 108 in the range located above the movable portion 112 in the through hole forming step S12. Therefore, the sacrificial layer 102 and the upper sacrificial layer 106 around the movable portion 112 can be removed earlier when the isotropic etching step S14 is performed. That is, the sacrificial layer 102 and the upper sacrificial layer 106 around the movable portion 112 can be removed before the sacrificial layer 102 below the fixed portion 110 is completely removed. As described above, since the sacrificial layer 102 around the movable portion 112 can be removed quickly, the shape of the movable portion 112 can be freely designed.

  Further, in the manufacturing method of the angular velocity detecting device 10 described above, a plurality of communication holes 150 for communicating the upper surface and the lower surface are dispersed and formed in the active layer 104 forming the movable portion 112 in the anisotropic etching step S2. To do. Since the communication hole 150 is thus formed, the etching can proceed in the thickness direction around the movable portion 112 when the isotropic etching step S14 is performed. Therefore, the sacrificial layer 102 below the movable portion 112 can be removed more quickly. That is, the sacrificial layer 102 below the movable portion 112 can be removed before the sacrificial layer 102 below the fixed portion 110 is completely removed. As described above, since the sacrificial layer 102 around the movable portion 112 can be removed quickly, the shape of the movable portion 112 can be freely designed.

  In addition, in the acceleration detection apparatus 10 of the Example mentioned above, although the fixing | fixed part 110 and the support part 111 were isolate | separated, the fixing | fixed part 110 and the support part 111 may be connected.

  Further, in the acceleration detection device 10 of the above-described embodiment, the active layer removal range 160 (the range not hatched in FIG. 3) is a very wide range, but the active layer removal range 160 is a range with a smaller area. There may be. That is, the active layer removal range 160 may be set at least in a region along the outline of the movable part 112 other than the connection part with the fixed part 110. If at least the active layer in the region along the outline of the movable part 112 other than the connection part with the fixed part 110 is removed, the movable part 112 can be movable with respect to the fixed part 110 and the support part 111. it can. Therefore, in the anisotropic etching step S2, if at least the region along the outline of the movable portion 112 other than the connection portion with the fixed portion 110 is etched, the active layer 104 in a different range from the present embodiment is etched. May be.

  In the manufacturing method of the above-described embodiment, the sacrificial layer 102 and the upper sacrificial layer 106 in the active layer removal range 160 are all removed in the isotropic etching step S14. However, if at least the sacrificial layer 102 and the upper sacrificial layer 106 around the movable portion 112 are removed, the movable portion 112 can be movable with respect to the fixed portion 110 and the support portion 111. Therefore, the sacrificial layer 102 and the upper sacrificial layer 106 may be left in the active layer removal range 160 by removing the sacrificial layer 102 and the upper sacrificial layer 106 in a range where the movable portion 112 is movable. As described above, when the sacrificial layer 102 and the upper sacrificial layer 106 are left in the active layer removal range 160, the through holes 120 are dispersed in the circuit layer 108 in the active layer removal range 160 in the through hole forming step S12. It does not have to be formed.

  In the above-described embodiment, the upper sacrificial layer 106 is made of SiO, but may be made of other materials. For example, an oxide insulator such as P-SiO or SOG can be used.

  In the above-described embodiments, the manufacturing method of the angular velocity detection device has been described. However, the manufacturing method of the present invention can also be used in devices that detect other mechanical quantities. For example, the manufacturing method of the present invention can be used for manufacturing a mechanical quantity detection device having a movable part separated from a base layer, such as an acceleration detection device and an angular acceleration detection device.

Specific examples of the present invention have been described in detail above, but these are merely examples and do not limit the scope of the claims. The technology described in the claims includes various modifications and changes of the specific examples illustrated above.
The technical elements described in this specification or the drawings exhibit technical usefulness alone or in various combinations, and are not limited to the combinations described in the claims at the time of filing. In addition, the technology illustrated in the present specification or the drawings achieves a plurality of objects at the same time, and has technical utility by achieving one of the objects.

2 is a partial cross-sectional view of the angular velocity detection device 10. FIG. The enlarged view of the circuit layer 108 in sectional drawing of FIG. The top view of the angular velocity detection apparatus 10 of the state which removed the circuit layer 108. FIG. The enlarged view of the upper surface of the movable part 112. FIG. 5 is a flowchart showing a manufacturing process of the angular velocity detection device 10. FIG. 6 is a partial cross-sectional view of a multilayer substrate 300. The fragmentary sectional view of the multilayer substrate 300 after anisotropic etching process S2 implementation. The fragmentary sectional view of the laminated substrate 300 after upper sacrificial layer formation process S4 execution. The fragmentary sectional view of the laminated substrate 300 after upper sacrificial layer planarization process S6 execution. The fragmentary sectional view of the laminated substrate 300 after circuit layer growth process S8 implementation. The fragmentary sectional view of the multilayer substrate 300 after through-hole formation process S12 execution. The fragmentary sectional view of the laminated substrate 300 in process of isotropic etching process S14. The fragmentary sectional view of the laminated substrate 300 after isotropic etching process S14 implementation.

Explanation of symbols

10: Angular velocity detection device 20: vibrators 22a, 22b: main frames 24a, 24b: subframes 26a-26d: beams 28a-28d: beams 30a-30d: beams 50a-50d: bases 80a-80d: capacitors 82a-82d: Capacitor 100: Base layer 102: Sacrificial layer 104: Active layer 106: Upper sacrificial layer 108: Circuit layer 110: Fixed portion 111: Supporting portion 112: Movable portion 120: Through hole 122: Wiring portion 124: Connection regions 124a to 124i: Connection region 130: Semiconductor circuit 130a: Drive signal output circuit 130b: Capacitance detection circuit 134: Interlayer insulating film 136: Passivation film 150: Communication hole 160: Active layer removal range 300: Multilayer substrate

Claims (6)

From the laminated substrate in which the base layer, the sacrificial layer, and the active layer are laminated in that order,
A fixed portion made of an active layer fixed to the base layer via a sacrificial layer;
A movable part comprising an active layer extending from the fixed part and free from the base layer because the sacrificial layer is removed;
A support made of an active layer fixed to the base layer via a sacrificial layer;
A circuit layer that is fixed on the support portion and extends above the movable portion in a non-contact positional relationship with the movable portion, and includes a semiconductor circuit that is electrically connected to the fixed portion;
A method for manufacturing a mechanical quantity detection device having
An anisotropic etching step of forming a trench reaching the sacrificial layer by anisotropically etching the active layer along the outline of the movable portion in a range excluding the connection portion with the fixed portion;
An upper sacrificial layer forming step of forming an upper sacrificial layer covering the active layer on the laminated substrate after the anisotropic etching step;
A circuit layer growth step for growing a circuit layer on the upper sacrificial layer;
A semiconductor circuit forming step of forming a semiconductor circuit in the circuit layer;
A through hole forming step of forming a through hole penetrating from the upper surface to the lower surface of the circuit layer at least at a position above the fixed portion of the circuit layer;
An isotropic etching process in which an etching agent is introduced from the through hole to perform isotropic etching of the sacrificial layer and the upper sacrificial layer;
A wiring part forming step of forming a wiring part that electrically connects the fixing part and the semiconductor circuit through a through hole located above the fixing part;
The position of the through hole formed in the through hole forming step and the etching time in the isotropic etching step are as follows:
Removing the sacrificial layer between the moving part and the base layer to separate the moving part and the base layer;
Removing the upper sacrificial layer between the moving part and the circuit layer to separate the moving part and the circuit layer;
The upper sacrificial layer filling between the movable part and the fixed part and the upper sacrificial layer filling between the movable part and the support part are removed, and the movable part is movable with respect to both the fixed part and the support part. Do;
-A sacrificial layer connecting the fixed part and the base layer is left below the fixed part;
-A sacrificial layer connecting the support part and the base layer is left below the support part;
The upper sacrificial layer connecting the support and the circuit layer is left on the upper part of the support;
The manufacturing method of the mechanical quantity detection apparatus characterized by setting so that the conditions may be satisfy | filled.   2. The method of manufacturing a mechanical quantity detection device according to claim 1, wherein a step of flattening an upper surface of the upper sacrificial layer is performed before the circuit layer growth step.   3. The step of forming an insulating film on the upper surface of the circuit layer and the wall surface of the through hole is performed after the through hole forming step and before the wiring portion forming step. A method for manufacturing a mechanical quantity detection device.   The manufacturing of the mechanical quantity detection device according to any one of claims 1 to 3, wherein, in the through hole forming step, a plurality of through holes are dispersedly formed in the circuit layer located above the movable part. Method.   5. The method according to claim 1, wherein, in the anisotropic etching process, a plurality of communication holes for communicating the upper surface and the lower surface of the active layer are dispersed in the active layer forming the movable part. A manufacturing method of the described mechanical quantity detection device. The base layer,
A fixed portion made of an active layer fixed to the base layer via a sacrificial layer;
A movable part consisting of an active layer extending from the fixed part and free from the base layer;
A support made of an active layer fixed to the base layer via a sacrificial layer;
It is fixed to the support part via the upper sacrificial layer, extends above the movable part in a non-contact positional relationship with the movable part, and a through-hole penetrating from the upper surface to the lower surface is located above the fixed part. And a circuit layer in which a semiconductor circuit is formed, and
A wiring portion that passes through the through hole and electrically connects the fixed portion and the semiconductor circuit;
A mechanical quantity detection device comprising:

Images