JP5849398B2 - Method for manufacturing MEMS device and MEMS device - Google Patents

Description

  The present specification relates to a method of manufacturing a MEMS device having a pair of counter electrodes that are relatively movable in a direction along a substrate, and the MEMS device.

  In recent years, a technology for manufacturing a MEMS device having a three-dimensional structure using a semiconductor integrated circuit manufacturing technology has been developed. Various technologies such as a pressure sensor, an acceleration sensor, a gyroscope, an optical deflection device, an RF switch, and a variable capacitor are provided. The device is realized. In order to realize a three-dimensional structure by a semiconductor integrated circuit manufacturing technique, a technique of selectively and locally etching a laminated substrate including an insulator layer as a sacrificial layer is used. The MEMS device referred to in this specification refers to selective and local etching of a laminated substrate including an insulator layer which is a sacrificial layer by using a semiconductor integrated circuit manufacturing technique represented by a film forming technique and an etching technique. An apparatus having a three-dimensional structure realized by the above.

  Patent Document 1 discloses a MEMS device having a pair of counter electrodes that are relatively movable in a direction along a substrate. In this MEMS device, the upper end portion of one counter electrode is processed so as to be lower than the upper end portion of the other counter electrode.

JP 2008-294455 A

  Of the pair of counter electrodes, there is a case where it is desired to process the lower end portion of one counter electrode so as to be higher than the lower end portion of the other counter electrode. In the technique of Patent Document 1, the upper end portion of one counter electrode is processed so as to be lower in height than the upper end portion of the other counter electrode. The electrodes are the same height. A technique capable of processing the lower end portion of one counter electrode so as to be higher than the lower end portion of the other counter electrode is expected.

  In this specification, the technique which solves said subject is provided. In this specification, in a MEMS device having a pair of counter electrodes that are relatively movable in the direction along the substrate, the lower end of one counter electrode may be cut without cutting the lower end of the other counter electrode. Provide possible technology.

  The present specification discloses a method of manufacturing a MEMS device having a pair of counter electrodes that are relatively movable in a direction along the substrate. The manufacturing method includes a step of preparing a semiconductor wafer in which an insulator layer is laminated on a first conductor layer serving as a substrate and a second conductor layer is laminated on the insulator layer, and a lower surface of the semiconductor wafer Forming a first groove portion that penetrates the first conductor layer and the insulator layer from the first conductor layer to reach the second conductor layer, forming a protective portion made of a conductor in the first groove portion, Forming a resist pattern corresponding to the shape of the pair of counter electrodes on the upper surface, and selectively etching the second conductor layer from the upper surface of the semiconductor wafer by reactive ion etching to form the second conductor layer; A step of forming a pair of counter electrodes, a step of cutting the lower end of one counter electrode using ions repelled laterally by charges accumulated in the insulator layer by continuing the reactive ion etching method, and an insulator Selectively removing the layer It is provided. In the manufacturing method, the first groove portion is disposed outside both side surfaces of the other counter electrode in the direction in which the pair of counter electrodes face each other.

  In the above manufacturing method, the pair of counter electrodes is formed on the second conductor layer by selectively etching the second conductor layer by a reactive ion etching method. At this time, at the place where the second conductor layer is removed, the insulator layer is exposed, and the protective portion made of the conductor formed using the first groove is also exposed. Thereafter, when the reactive ion etching method is further continued, charges are accumulated on the surface of the insulator layer, and the ions irradiated by the charges are repelled to the side, and the lower end portion of the remaining second conductor layer. Is cut from the side. At this time, since the charge is not accumulated on the surface where the protective portion made of the conductor is exposed, the irradiated ions are hardly repelled to the side. That is, when the reactive ion etching method is continued, the protective part formed in the first groove part plays a role of protecting the lower end part of the second conductor layer above it. Since this protective part is arranged outside the both side surfaces of the other counter electrode, the lower end of the other counter electrode is hardly shaved, and only the lower end of one counter electrode can be shaved from the side surface. .

  In the above manufacturing method, the first conductor layer is made of single crystal silicon, the insulator layer is made of silicon oxide, and the second conductor layer is made of single crystal silicon. The protective part may be made of metal.

  Since the etching rate of metal is slower than that of silicon, in the above case, the protective portion does not disappear when the reactive ion etching method is continued and the lower end portion of one counter electrode is shaved. The lower end portion of the other counter electrode can be reliably protected.

  Alternatively, in the above manufacturing method, the first conductor layer is made of single crystal silicon, the insulator layer is made of silicon oxide, and the second conductor layer is made of single crystal silicon. The protective portion may be made of polysilicon.

  Since polysilicon can easily form and fill the groove, in the above case, the protective part can be easily formed in the first groove. Moreover, since the difference in thermal expansion coefficient between polysilicon and single crystal silicon is small, the occurrence of warpage in the substrate of the MEMS device to be manufactured can be suppressed.

  In the manufacturing method described above, a step of forming a second groove that reaches the second conductor layer from the lower surface of the semiconductor wafer through the first conductor layer and the insulator layer before performing the reactive ion etching method. And a step of forming an induction portion made of an insulator having a dielectric constant higher than that of the insulator layer in the second groove portion, wherein the second groove portion is formed on the one side with respect to a direction in which a pair of counter electrodes face each other. It is preferable that it is arrange | positioned on the outer side of the both sides | surfaces of this counter electrode.

  In the above manufacturing method, when the reactive ion etching method is continued, the induction part is exposed on the lower outside of both side surfaces of one counter electrode, and the protection is provided on the outer lower side of both side surfaces of the other counter electrode. The part is exposed. Since the induction portion is made of an insulator having a dielectric constant higher than that of the insulator layer, a large amount of electric charge is accumulated on the surface, and the irradiated ions are easily repelled sideways. As a result, the lower end of one counter electrode is quickly scraped. The manufacturing process can be speeded up.

  In the above manufacturing method, the first conductor layer is made of single crystal silicon, the insulator layer is made of silicon oxide, and the second conductor layer is made of single crystal silicon. The guiding portion may be made of silicon nitride.

  Since silicon nitride can easily form and fill the groove, in the above case, the guiding portion can be easily formed in the second groove.

  In this specification, a substrate formed on the first conductor layer and a second conductor layer stacked on the first conductor layer with the insulator layer interposed therebetween are relative to the direction along the substrate. A MEMS device having a pair of movable counter electrodes is disclosed. In the MEMS device, the lower end portion of one counter electrode is scraped from the side surface, a first groove portion penetrating the substrate is formed in a direction perpendicular to the substrate, and the other is provided in the direction in which the pair of counter electrodes face each other. The 1st groove part is arrange | positioned on the outer side of the both sides | surfaces of this counter electrode.

  In the above MEMS device, a protective part made of a conductor is formed in the first groove part at the time of manufacture, and the reaction continues even after a pair of counter electrodes are formed on the second conductor layer by a reactive ion etching method. It can be easily manufactured by continuing the reactive ion etching method.

  In the MEMS device described above, the protective portion made of a conductor may be left in the first groove portion. Or the protection part formed in the 1st groove part may be removed finally.

  In the MEMS device described above, the second groove portion penetrating the substrate is formed in a direction perpendicular to the substrate, and the second groove portion is formed outside the both side surfaces of one counter electrode in the direction in which the pair of counter electrodes are opposed to each other. It is preferable that they are arranged.

  In the MEMS device described above, when the reactive ion etching method is continued by forming an induction portion made of an insulator having a dielectric constant higher than that of the insulator layer in the second groove portion during manufacture, The lower end portion of the electrode can be quickly shaved. The manufacturing process can be speeded up.

  In the MEMS device described above, an induction portion made of an insulator having a dielectric constant higher than that of the insulator layer may be left in the second groove portion. Or the induction | guidance | derivation part formed in the 2nd groove part may be removed finally.

  According to the technology disclosed in this specification, in a MEMS device having a pair of counter electrodes that are relatively movable in a direction along the substrate, the one of the counter electrodes can be formed without scraping the lower end of the other counter electrode. The lower end can be shaved.

It is a top view of the acceleration sensor 10 of Example 1, and the acceleration sensor 50 of Example 2. FIG. It is the longitudinal cross-sectional view which looked at the II-II cross section of FIG. 1 about the acceleration sensor 10 of Example 1. FIG. It is the longitudinal cross-sectional view which looked at the III-III cross section of FIG. 1 about the acceleration sensor 10 of Example 1, and the acceleration sensor 50 of Example 2. FIG. (A) is the acceleration sensor 10 of Example 1, and is a figure which shows the state which the movable electrode 116 displaced below with respect to the fixed electrode 114, (B) is the acceleration sensor 10 of Example 1, and is the fixed electrode 114. It is a figure which shows the state which the movable electrode 116 displaced upwards with respect to FIG. It is a figure which shows the manufacturing process of the acceleration sensor 10 which concerns on Example 1. FIG. It is a figure which shows the manufacturing process of the acceleration sensor 10 which concerns on Example 1. FIG. It is a figure which shows the manufacturing process of the acceleration sensor 10 which concerns on Example 1. FIG. It is a figure which shows the manufacturing process of the acceleration sensor 10 which concerns on Example 1. FIG. It is a figure which shows the manufacturing process of the acceleration sensor 10 which concerns on Example 1. FIG. It is a figure which shows the manufacturing process of the acceleration sensor 10 which concerns on Example 1. FIG. It is a figure which shows the manufacturing process of the acceleration sensor 10 which concerns on Example 1. FIG. FIG. 6 is a diagram illustrating an example of a planar arrangement of a protection unit 208 in the acceleration sensor 10 according to the first embodiment. FIG. 6 is a diagram illustrating another example of a planar arrangement of the protection unit 208 in the acceleration sensor 10 according to the first embodiment. It is a figure which shows the structure of the protection part 212 in the acceleration sensor 10 which concerns on the modification of Example 1. FIG. FIG. 6 is a diagram illustrating a configuration of a protection unit 214 in an acceleration sensor 10 according to a modification example of Example 1. It is a figure which shows the structure of the protection part 208 in the acceleration sensor 10 which concerns on the modification of Example 1. FIG. It is a figure which shows the structure of the protection part 215 in the acceleration sensor 10 which concerns on the modification of Example 1. FIG. It is the longitudinal cross-sectional view which looked at the II-II cross section of FIG. 1 about the acceleration sensor 50 of Example 2. FIG. It is a figure which shows the manufacturing process of the acceleration sensor 50 which concerns on Example 2. FIG. It is a figure which shows the manufacturing process of the acceleration sensor 50 which concerns on Example 2. FIG. It is a figure which shows the manufacturing process of the acceleration sensor 50 which concerns on Example 2. FIG. It is a figure which shows the manufacturing process of the acceleration sensor 50 which concerns on Example 2. FIG. It is a figure which shows the manufacturing process of the acceleration sensor 50 which concerns on Example 2. FIG.

The features of the preferred embodiment are listed first.
(Feature 1) One counter electrode is a movable electrode, and the other counter electrode is a fixed electrode.

  Hereinafter, the structure of the acceleration sensor 10 according to the first embodiment will be described with reference to FIGS. As shown in FIGS. 2 and 3, the acceleration sensor 10 has a laminated structure of a first layer 20 made of a conductor, a second layer 30 made of an insulator, and a third layer 40 made of a conductor. Yes. Specifically, in the acceleration sensor 10 of the present embodiment, the first layer 20 is made of single crystal silicon to which impurities are added, the second layer 30 is made of silicon oxide, and the third layer 30 is a single crystal to which impurities are added. It has a so-called SOI (Silicon on Insulator) structure made of silicon.

  As shown in FIGS. 1 to 3, the acceleration sensor 10 includes a support substrate 100, a proof mass 102, a support beam 104, and a frame-like support portion 106. The support substrate 100 is formed on the first layer 20. The proof mass 102, the support beam 104, and the frame-like support portion 106 are formed in the third layer 40. The frame-like support part 106 is fixed to the support substrate 100 via an insulating support part 108 formed on the second layer 30. The support beam 104 connects the frame-shaped support portion 106 and the proof mass 102. The proof mass 102 is formed in a rectangular flat plate shape. The proof mass 102 is supported by the support beam 104 so as to be substantially parallel to the support substrate 100 with a gap from the support substrate 100. The support beam 104 is formed in an elongated shape so that bending rigidity and shear rigidity in the Y direction and the Z direction are low. Therefore, for example, when an inertia force in the Y direction acts on the proof mass 102, the support beam 104 is bent in the Y direction, and the proof mass 102 is relatively displaced in the Y direction with respect to the support substrate 100. When an inertia force in the Z direction acts on the proof mass 102, the support beam 104 is bent in the Z direction, and the proof mass 102 is relatively displaced in the Z direction with respect to the support substrate 100. The amount of bending of the support beam 104 at this time depends on the magnitude of the inertial force acting on the proof mass 102. In other words, the relative displacement amount of the proof mass 102 with respect to the support substrate 100 corresponds to the magnitude of acceleration acting on the proof mass 102. Therefore, by detecting the amount of displacement of the proof mass 102 relative to the support substrate 100, the acceleration acting on the acceleration sensor 10 can be detected.

  The acceleration sensor 10 includes a Y direction detection unit 110 and a Z direction detection unit 124. The Y-direction detection unit 110 includes a fixed electrode support unit 112, a comb-shaped fixed electrode 114 extending from the fixed electrode support unit 112, and a comb-shaped movable electrode 116 extending from the proof mass 102. The fixed electrode support portion 112, the fixed electrode 114, and the movable electrode 116 are formed on the third layer 40. The fixed electrode support part 112 is fixed to the support substrate 100 via an insulating support part 113 formed on the second layer 30. The fixed electrode 114 and the movable electrode 116 are arranged to face each other in the Y direction. Between the fixed electrode 114 and the movable electrode 116, a capacitance according to the facing area and distance between the two is formed. When the movable electrode 116 is displaced in the Y direction with respect to the fixed electrode 114, the distance in the Y direction between the fixed electrode 114 and the movable electrode 116 changes, and the capacitance between the fixed electrode 114 and the movable electrode 116 is changed accordingly. The size of changes. By detecting this change in capacitance, it is possible to detect the amount of displacement of the proof mass 102 in the Y direction relative to the support substrate 100. In the acceleration sensor 10 of this embodiment, the Y-direction detection unit 110 disposed on one side (for example, the upper side in FIG. 1) and the other side (for example, the lower side in FIG. 1) with respect to the proof mass 102 Further, by combining the outputs of the Y direction detection unit 110, it is possible to correct a capacity error caused by manufacturing tolerances.

  As shown in FIGS. 2 and 3, the Z-direction detection unit 124 includes a support substrate 100 that functions as a fixed electrode and a proof mass 102 that functions as a movable electrode. Between the support substrate 100 and the proof mass 102, an electrostatic capacity corresponding to the opposing area and distance between the two is formed. When the proof mass 102 is displaced in the Z direction with respect to the support substrate 100, the magnitude of the capacitance between the support substrate 100 and the proof mass 102 changes. By detecting this change in capacitance, the amount of displacement of the proof mass 102 in the Z direction relative to the support substrate 100 can be detected.

  In the acceleration sensor 10 of the present embodiment, the proof mass 102, the support beam 104, the frame-like support portion 106, and the movable electrode 116 are seamlessly integrated by trimming the single crystal silicon of the third layer 40. Is formed. Therefore, the proof mass 102, the support beam 104, the frame-like support portion 106, and the movable electrode 116 are kept at the same potential.

  The acceleration sensor 10 detects the potential of the frame-shaped support portion 106 (which is also the potential of the proof mass 102 that is the movable electrode of the Z direction detection unit 124 and the potential of the movable electrode 116 of the Y direction detection unit 110). The surface electrode 118, the second surface electrode 120 that detects the potential of the support substrate 100 (that is, the potential of the fixed electrode of the Z direction detection unit 124), and the third that detects the potential of the fixed electrode 114 of the Y direction detection unit 110. A surface electrode 122 is provided. Based on the output from these surface electrodes, an arithmetic circuit (not shown) performs arithmetic processing, whereby acceleration in the Y direction and Z direction acting on the acceleration sensor 10 can be detected.

  As shown in FIG. 2, in the acceleration sensor 10 of the present embodiment, the upper end of the movable electrode 116 is lower than the upper end of the fixed electrode 114, and the lower end of the movable electrode 116 is higher than the lower end of the fixed electrode 114. A Y direction detection unit 110 is formed. Therefore, even when the proof mass 102 is relatively displaced in the direction approaching the support substrate 100 with respect to the Z direction, the facing area of the movable electrode 116 and the fixed electrode 114 does not change as shown in FIG. Further, even when the proof mass 102 is relatively displaced in the direction away from the support substrate 100, the facing area of the movable electrode 116 and the fixed electrode 114 does not change as shown in FIG. Therefore, it is possible to suppress the displacement of the proof mass 102 in the Z direction on the detection result of the Y direction detection unit 110.

  Further, in the acceleration sensor 10 of the present embodiment, since the lower end of the movable electrode 116 is formed to be higher than the lower end of the fixed electrode 114, a distance between the support substrate 100 and the movable electrode 116 is ensured and the support is supported. The parasitic capacitance between the substrate 100 and the movable electrode 116 is kept small. In particular, as shown in FIG. 4A, even when the proof mass 102 sinks with respect to the support substrate 100, the distance between the support substrate 100 and the movable electrode 116 is ensured, and the support substrate 100 and the movable electrode are secured. The parasitic capacitance between 116 is kept small. According to the acceleration sensor 10 of the present embodiment, the parasitic capacitance between the support substrate 100 and the movable electrode 116 is always kept small, and the influence of this parasitic capacitance on the detection value of the acceleration sensor 10 can be suppressed.

  Further, in the acceleration sensor 10 of the present embodiment, the movable electrode 116 is formed so that the width becomes narrower as it approaches the lower end, but the fixed electrode 114 is formed so that the width is constant from the upper end to the lower end. Yes. Therefore, even when the movable electrode 116 is displaced in the Z direction with respect to the fixed electrode 114, the distance between the movable electrode 116 and the fixed electrode 114 does not change. Variation in capacitance between the movable electrode 116 and the fixed electrode 114 can be suppressed.

  Below, the manufacturing method of the acceleration sensor 10 of a present Example is demonstrated, referring FIGS. 5-11. 5 to 11 correspond to the cross section taken along the line II-II of FIG. 1, that is, the cross section of FIG.

  First, as shown in FIG. 5, a laminated structure of a first layer 20 made of single crystal silicon doped with impurities, a second layer 30 made of silicon oxide, and a third layer 40 made of single crystal silicon doped with impurities. Then, a shallow wide groove 204 is formed on the surface of the third layer 40 by reactive ion etching (RIE). The position of the shallow and wide groove portion 204 corresponds to the position of the movable electrode 116 of the acceleration sensor 10 to be manufactured, and the depth of the shallow and wide groove portion 204 corresponds to the height of the upper end of the movable electrode 116. The first layer 20 corresponds to the support substrate 100 of the acceleration sensor 10 to be manufactured.

  Next, as shown in FIG. 6, deep and narrow grooves 206 are formed in the first layer 20 from the back side of the SOI wafer 202 by the deep RIE method. The deep and narrow groove 206 penetrates the first layer 20 and reaches the interface with the second layer 30. The positions of the deep and narrow grooves 206 are set so as to be arranged on both sides of the fixed electrode 114 of the acceleration sensor 10 to be manufactured.

  Next, as shown in FIG. 7, the deep and thin groove 206 is extended by the oxide film RIE method, and penetrates through the second layer 30 to reach the interface with the third layer 40.

  Next, as shown in FIG. 8, the deep and narrow groove 206 is filled with metal by a chemical vapor deposition (CVD) method to form a protective portion 208. For example, aluminum or copper can be used as the metal for forming the protection portion 208.

  Next, as shown in FIG. 9, a resist pattern 210 is formed on the surface of the third layer 40, and the third layer 40 is trimmed by a deep RIE method. As a result, the proof mass 102, the support beam 104, the frame-like support portion 106, the fixed electrode support portion 112, the fixed electrode 114, and the movable electrode 116 are formed on the third layer 40.

  Next, as shown in FIG. 10, the deep RIE method is continued as it is (over-etching), and the lower end portion of the movable electrode 116 is removed. After the third layer 40 is etched by the deep RIE method and the second layer 30 is exposed, when the deep RIE method is further continued, charges are accumulated on the exposed surface of the second layer 30. Due to the charge accumulated on the surface of the second layer 30, the irradiated ions are repelled sideways, and the lower end of the movable electrode 116 is shaved from the side surface. In this embodiment, overetching is continued until the lower end of the movable electrode 116 reaches a predetermined height.

  Note that, during the above-described over-etching, no charge is accumulated on the surface of the second layer 30 at the portion where the protective portion 208 is exposed on the surface of the second layer 30. Accordingly, even when over-etching is performed, irradiated ions are not repelled sideways in the vicinity of the protection portion 208. In other words, the protection unit 208 plays a role of protecting the third layer 40 located in the vicinity of the upper portion during overetching. In this embodiment, since the protective portions 208 are disposed on both sides of the fixed electrode 114, it is possible to prevent the lower end portion of the fixed electrode 114 from being scraped during overetching.

  In the case of the over-etching described above, the lower end portion of the edge portion of the frame-like support portion 106 and the fixed electrode support portion 112 where the protection portion 208 is not disposed is shaved by ions repelled sideways. However, since the area to be scraped is very small compared to the size of the frame-like support part 106 and the fixed electrode support part 112 and does not affect the function of the frame-like support part 106 and the fixed electrode support part 112, it is shown here. Is omitted.

  Next, as shown in FIG. 11, the resist pattern 210 is removed, and the silicon oxide of the second layer 30 which is a sacrificial layer is selectively removed. As a result, the portion excluding the insulating support portion 108 and the insulating support portion 113 is removed from the second layer 30.

  And the acceleration sensor 10 shown in FIGS. 1-3 can be obtained by removing the protection part 208. FIG. The protection unit 208 can be left as it is without being removed.

  In the above embodiment, the case where the solid protective portion 208 is formed by filling the deep and thin groove portion 206 with metal has been described. Alternatively, the hollow protective portion 208 may be formed by depositing a metal only on the inner surface of the deep and narrow groove 206.

  As for the planar arrangement of the protective portion 208 that protects the lower end of the fixed electrode 114 during overetching, for example, as shown in FIG. As shown in FIG. 13, the fixed electrodes 114 may be discretely arranged outside both side surfaces.

  In the above embodiment, the case where the upper end of the protection unit 208 is located at the interface between the second layer 30 and the third layer 40 has been described. In contrast to this, for example, in FIG. 7, after the deep and narrow groove 206 is extended to the interface between the second layer 30 and the third layer 40, the deep and thin groove 206 is further formed into the third layer 40 by the deep RIE method. By extending, as shown in FIG. 16, the upper end of the protection portion 208 can be formed to be higher than the lower end of the fixed electrode 114. In the case of this configuration, when the lower end portion of the movable electrode 116 is processed by over-etching, the area where the protection portion 208 is exposed is increased, so that the lower end portion of the fixed electrode 114 can be reliably protected. The protection unit 208 can be finally removed or can be left as it is. When the protection unit 208 is left as it is, the protection unit 208 functions as a mechanical stopper against the displacement of the movable electrode 116 in the Y direction.

  In the above embodiment, the acceleration sensor 10 is removed from the SOI wafer 202 having a laminated structure of the first layer 20 made of a conductive material, the second layer 30 made of an insulating material, and the third layer 40 made of a conductive material. The case of manufacturing has been described. Unlike this, for example, as shown in FIG. 17, the acceleration sensor 10 can be manufactured by a similar manufacturing method from a wafer having a laminated structure including more layers. FIG. 17 shows a stacked structure of a conductive first layer 216, an insulating second layer 218, a conductive third layer 220, an insulating fourth layer 222, and a conductive fifth layer 224. The case where the acceleration sensor 10 is manufactured from the laminated wafer 226 having the above is shown.

  In the above embodiment, the case where the protective portion 208 having a uniform thickness in the vertical direction is formed has been described. Unlike this, for example, as shown in FIG. 14, a protective portion 212 having a shape in which the thickness of the upper end portion is larger than the thickness of the lower portion may be formed. When the deep and narrow groove 206 is extended in FIG. 7, such a protective part 212 replaces the silicon oxide of the second layer 30 from the tip of the deep and thin groove 206 by isotropic etching instead of the oxide film RIE method. It can form by etching toward the direction. With such a structure, the protective portion 212 can be reliably formed immediately below both side surfaces of the fixed electrode 114. When the lower end portion of the movable electrode 116 is processed by overetching, the lower end of the fixed electrode 114 is formed. The part can be reliably protected. Also in this case, the solid protective portion 212 may be formed by filling the deep and thin groove portion 206 with metal, or the metal may be formed only on the inner surface of the deep and thin groove portion 206 to form the hollow protective portion 212. May be formed. Note that the protection unit 212 can be finally removed, or can be left as it is without being removed. In the case where the protection part 212 is left without being removed, the fixed electrode 114 and the protection part 212 are electrically connected. It is preferable to have a separate structure for insulating between the protective portions 212.

  In the above embodiment, the configuration in which the pair of protection portions 208 are disposed on both sides of the fixed electrode 114 has been described. Unlike this, for example, as shown in FIG. 15, a protective part 214 having a width wider than the fixed electrode 114 and a size protruding from both sides of the fixed electrode 114 may be formed. Also in this case, the solid protective portion 214 may be formed by filling the deep and thin groove portion 206 with metal, or the metal is formed only on the inner surface of the deep and thin groove portion 206 to form the hollow protective portion 214. May be formed. The protective part 214 can be finally removed, or can be left as it is without being removed. In the case where the protective part 214 is left without being removed, the fixed electrode 114 and the protective part 214 are electrically connected, so that the other part of the support substrate 100 (at least, the part serving as the fixed electrode of the Z-direction detection unit 124). It is preferable to have a separate structure for insulating between the protective portions 214.

  In the above embodiment, the case where the protection unit 208 is made of metal has been described. Since the etching rate of metal is slower than that of silicon, the protective portion 208 is not lost during over-etching to remove the lower end portion of the movable electrode 116, and the lower end portion of the fixed electrode 114 is reliably protected. be able to. Unlike this, as the material of the protective portion 208, other conductive materials can be used. For example, when polysilicon doped with impurities is used as the protective portion 208, film formation and filling are easy, and the difference in thermal expansion coefficient from the single crystal silicon of the first layer 20 is small. There is an advantage that it is difficult to cause warping. In this case as well, the deep and thin groove 206 may be filled with polysilicon to form the solid protective portion 208, or the polysilicon may be formed only on the inner surface of the deep and thin groove 206 to protect the hollow portion. The portion 208 may be formed. Moreover, the protection part 212 as shown in FIG. 14 and the protection part 214 as shown in FIG. 15 can also be formed of polysilicon.

  The acceleration sensor 50 of the present embodiment has substantially the same configuration as the acceleration sensor 10 of the first embodiment shown in FIGS. 1 to 3, but as shown in FIG. This is different from the acceleration sensor 10 of the first embodiment in that not only the deep and thin groove portion 206 is formed but also the deep and thin groove portion 228 is formed in the substrate 100 below the movable electrode 116.

  Below, the manufacturing method of the acceleration sensor 10 of a present Example is demonstrated, referring FIGS. 19-23. 19 to 23 correspond to the section taken along the line II-II of FIG. 1, that is, the section of FIG.

  First, steps similar to those shown in FIGS. 5 to 8 of the first embodiment are performed, and a first layer 20 made of single crystal silicon doped with impurities, a second layer 30 made of silicon oxide, and impurities A shallow wide groove portion 204 and a protective portion 208 are formed in an SOI wafer 202 having a laminated structure of the third layer 40 made of single crystal silicon to which is added.

  Next, as shown in FIG. 19, a deep and narrow groove 228 is formed in the first layer 20 from the back side of the SOI wafer 202 by a deep RIE method. The deep and narrow groove 228 passes through the first layer 20 and reaches the interface with the second layer 30. The positions of the deep and narrow grooves 228 are set so as to be arranged on both sides of the movable electrode 116 of the acceleration sensor 50 to be manufactured. Further, the deep and narrow groove 228 is extended by the oxide film RIE method, and reaches the interface with the third layer 40 through the second layer 30.

  Next, as shown in FIG. 20, silicon nitride is filled in the deep and narrow groove portion 228 by chemical vapor deposition (CVD) to form the guiding portion 230.

  Next, as shown in FIG. 21, a resist pattern 210 is formed on the surface of the third layer 40, and the third layer 40 is trimmed by a deep RIE method. As a result, the proof mass 102, the support beam 104, the frame-like support portion 106, the fixed electrode support portion 112, the fixed electrode 114, and the movable electrode 116 are formed on the third layer 40.

  Next, as shown in FIG. 22, the deep RIE method is continued as it is (over-etching), and the lower end portion of the movable electrode 116 is removed. At this time, a large amount of electric charge is accumulated in the portion where the induction portion 230 is exposed on the surface of the second layer 30, so that the irradiated ions are laterally present compared to the portion where the induction portion 230 is not formed. It becomes easy to be played. As a result, the lower end of the movable electrode 116 is quickly shaved. On the other hand, in the location where the protection part 208 is exposed, the charge is hardly accumulated and the irradiated ions are not repelled sideways. Thereby, it is possible to prevent the lower end portion of the fixed electrode 114 from being scraped. In this embodiment, overetching is continued until the lower end of the movable electrode 116 reaches a predetermined height.

  Next, as shown in FIG. 23, the resist pattern 210 is removed, and the silicon oxide of the second layer 30 which is a sacrificial layer is selectively removed. As a result, the portion excluding the insulating support portion 108 and the insulating support portion 113 is removed from the second layer 30.

  Then, the acceleration sensor 50 can be obtained by removing the protection unit 208 and the guide unit 230. Note that the protection unit 208 and the guide unit 230 can be left as they are without being removed.

  Similarly to the description of the protection part 208 in the first embodiment, the guide part 230 may be formed in a solid structure by filling the deep and narrow groove part 228 with silicon nitride, or the inner surface of the deep and narrow groove part 228. Alternatively, silicon nitride may be deposited only to form a hollow structure.

  Similarly to the description of the protection unit 208 in the first embodiment, the guide unit 230 may be continuously arranged on the outer sides of the both sides of the movable electrode 116, or discretely on the outer sides of the both sides of the movable electrode 116. You may arrange in.

  Similarly to the description of the protection part 208 in the first embodiment, the guide part 230 may be formed in a shape having a uniform thickness in the vertical direction, and the thickness of the upper end part is smaller than the thickness of the lower part. You may form in a big shape. In such a guiding portion 230, when the deep and narrow groove portion 228 is extended in the process of FIG. 19, the silicon oxide of the second layer 30 is made isotropically etched by the isotropic etching instead of the oxide film RIE method. It can form by etching toward the side.

  Similarly to the description of the protection unit 208 in the first embodiment, a pair of the guide units 230 may be arranged on both sides of the movable electrode 116, or the width of the guide unit 230 may be larger than that of the movable electrode 116. It is good also as a structure arrange | positioned so that it may protrude.

  In the above embodiment, the case where the guiding portion 230 is made of silicon nitride has been described. However, as the material of the guiding portion 230, any material having a dielectric constant higher than that of the silicon oxide of the second layer 30 can be used. May be used.

  As mentioned above, although the Example of this invention was described in detail, these are only illustrations and do not limit a claim. 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.

DESCRIPTION OF SYMBOLS 10, 50 Acceleration sensor 20 1st layer 30 2nd layer 40 3rd layer 100 Support substrate 102 Proof mass 104 Support beam 106 Frame-shaped support part 108 Insulation support part 110 Y direction detection part 112 Fixed electrode support part 113 Insulation support part 114 Fixed electrode 116 Movable electrode 118 First surface electrode 120 Second surface electrode 122 Third surface electrode 124 Z direction detection unit 202 SOI wafer 204 Shallow wide groove 206 Deep thin groove 208, 212, 214 Protective part 210 Resist pattern 216 First layer 218 2nd layer 220 3rd layer 222 4th layer 224 5th layer 226 Laminated wafer 228 Deep and narrow groove 230 Inductor

Claims (10)

A method of manufacturing a MEMS device having a pair of counter electrodes that are relatively movable in a direction along a substrate,
Preparing a semiconductor wafer in which an insulator layer is laminated on a first conductor layer serving as a substrate, and a second conductor layer is laminated on the insulator layer;
Forming a first groove from the lower surface of the semiconductor wafer through the first conductor layer and the insulator layer and reaching the second conductor layer;
Forming a protective part made of a conductor in the first groove part;
Forming a resist pattern corresponding to the shape of the pair of counter electrodes on the upper surface of the semiconductor wafer;
Selectively etching the second conductor layer from the upper surface of the semiconductor wafer by reactive ion etching to form a pair of counter electrodes on the second conductor layer;
Continue the reactive ion etching method, using the ions repelled to the side by the charge accumulated in the insulator layer, scraping the lower end of one counter electrode,
A step of selectively removing the insulator layer;
The method of manufacturing a MEMS device, wherein the first groove portion is disposed outside both side surfaces of the other counter electrode in a direction in which the pair of counter electrodes oppose each other.   The first conductor layer is made of single crystal silicon, the insulator layer is made of silicon oxide, the second conductor layer is made of single crystal silicon, and the protection portion is made of metal. The method for manufacturing a MEMS device according to claim 1, comprising:   The first conductor layer is made of single crystal silicon, the insulator layer is made of silicon oxide, the second conductor layer is made of single crystal silicon, and the protection portion is made of polycrystal silicon. 2. The method of manufacturing a MEMS device according to claim 1, wherein the method is made of silicon. Before performing the reactive ion etching method, the following two steps are performed:
Forming a second groove from the lower surface of the semiconductor wafer through the first conductor layer and the insulator layer and reaching the second conductor layer;
Further comprising the step of forming an induction portion made of an insulator having a dielectric constant higher than that of the insulator layer in the second groove portion,
2. The method of manufacturing a MEMS device according to claim 1, wherein the second groove portion is disposed outside both side surfaces of the one counter electrode in a direction in which the pair of counter electrodes face each other.   The first conductor layer is made of single crystal silicon, the insulator layer is made of silicon oxide, the second conductor layer is made of single crystal silicon, and the induction portion is nitrided 5. The method of manufacturing a MEMS device according to claim 4, wherein the method is made of silicon. A substrate formed on the first conductor layer;
A MEMS device having a pair of counter electrodes formed on a second conductor layer stacked on a first conductor layer with an insulator layer interposed therebetween and relatively movable in a direction along the substrate,
The upper end of one counter electrode is lower than the upper end of the other counter electrode,
The lower end of one counter electrode is higher than the lower end of the other counter electrode ,
First groove penetrating the substrate in a direction perpendicular to the base plate and is formed,
A MEMS device, wherein a first groove portion is disposed outside both side surfaces of the other counter electrode in a direction in which the pair of counter electrodes oppose each other. A substrate formed on the first conductor layer;
A MEMS device having a pair of counter electrodes formed on a second conductor layer stacked on a first conductor layer with an insulator layer interposed therebetween and relatively movable in a direction along the substrate ,
First groove penetrating the substrate in a direction perpendicular to the base plate and is formed,
With respect to the direction in which the pair of counter electrodes are opposed to each other, the first groove portion is disposed outside the both side surfaces of the other counter electrode,
A protective part made of a conductor is formed in the first groove part,
A MEMS device, wherein an upper end portion of the protection portion is higher than a lower end portion of the other counter electrode. The MEMS device according to claim 6 or 7, wherein the width of the one counter electrode becomes narrower toward the lower end. A second groove portion penetrating the substrate in a direction perpendicular to the substrate is formed;
The MEMS device according to any one of claims 6 to 8, wherein a second groove portion is disposed outside both side surfaces of one counter electrode in a direction in which the pair of counter electrodes oppose each other. The MEMS device according to claim 9, wherein an induction portion made of an insulator having a dielectric constant higher than that of the insulator layer is formed in the second groove portion.

Images