JP2011187373A - Switch, manufacturing method thereof, and relay - Google Patents

Abstract

<P>PROBLEM TO BE SOLVED: To smoothly form a contact surface of each of contact points, and to make contact resistance smaller between the contact points by surely contacting each of the contact points. <P>SOLUTION: A fixed-contact section 33 has an adhesion layer 43 and a wiring layer 44 laminated on a fixed-contact substrate 41, and a contact layer 45 is formed thereon. An end surface opposite to the movable-contact section 34 of the contact layer 45 serves as a fixed contact 46 (a contact surface); and the fixed contact 46 is protruded from the wiring layer 44, the adhesion layer 43, and the fixed-contact substrate 41. A movable-contact section 34 has the adhesion layer 53 and the wiring layer 54 laminated on a movable-contact substrate 51, and the contact layer 55 is formed thereon. The end surface opposite to the fixed-contact section 33 of the contact layer 55 serves as a movable contact 56 (the contact surface); and the movable contact 56 is protruded from the wiring layer 54, the adhesion layer 53, and the movable-contact substrate 51. The fixed contact 46 and the movable contact 56 act as the surface in contact with the side of a molding section during the growing process of the contact layers 45, 55 through vapor deposition, sputtering and the like. <P>COPYRIGHT: (C)2011,JPO&INPIT

Description

  The present invention relates to a switch, a manufacturing method thereof, and a relay. Specifically, the present invention relates to a switch in which a surface perpendicular to the moving direction of a movable contact portion is a contact, and a manufacturing method thereof, and further to a relay using the switch structure.

(Patent Document 1)
A MEMS (Micro Electrical-Mechanical Systems) switch in which a surface perpendicular to the moving direction of the movable contact portion is a contact (contact surface) is disclosed in Patent Document 1. In this switch 11, as shown in FIG. 1A, an insulating layer 13a is formed on the upper surface of a substrate 12a, a conductive layer 14a made of Al, Cu, or the like is formed thereon, and the upper surface of the conductive layer 14a is further formed. A movable contact portion 17 is formed by growing a plated layer 15a of Au or the like from the end surface to the end surface. Similarly, an insulating layer 13b is formed on the upper surface of the substrate 12b, a conductive layer 14b made of Al, Cu, or the like is formed thereon, and a plated layer 15b such as Au is grown from the upper surface to the end surface of the conductive layer 14b. Thus, the fixed contact portion 18 is formed. Then, the movable contact portion 17 is moved in the direction of the arrow, and as shown in FIG. 1B, the movable contact 16a that is the protruding region of the plating layer 15a and the fixed contact 16b that is the protruding region of the plating layer 15b are contacted, or The switching operation is performed between the movable contact portion 17 and the fixed contact portion 18 by separating them.

  However, such a switch 11 has a structure in which the surface of the movable contact 16a formed on the end surface of the conductive layer 14a and the surface of the fixed contact 16b formed on the end surface of the conductive layer 14b are contacted and separated. The movable contact 16a and the fixed contact 16b come into contact with each other on a surface (plating surface) perpendicular to the growth direction of the plating layers 15a and 15b. However, since the surface of the plating layer is rough and has fine irregularities, the substantial contact area when the movable contact 16a and the fixed contact 16b are brought into contact with each other is considerably small, and the contact resistance between the contacts is large.

  Further, when a voltage is applied to the conductive layers 14a and 14b for plating, the electric field strength is high (the electric lines of force are concentrated) on the end faces of the conductive layers 14a and 14b. In contact 16b), the growth rate of the plating film is large, and it is difficult to control the gap distance between the contacts. On the other hand, since the surface of the contact is rough and has irregular fine irregularities, discharge tends to occur between the contacts when the contacts approach each other due to the gap distance variation. Therefore, in such a switch 11, it is difficult to narrow the gap distance between the contacts.

(Patent Document 2)
Further, as an electrostatic relay in which the movable contact moves in parallel with the base substrate and the movable contacts come into contact with or separate from each other, there is one disclosed in Patent Document 2, for example. In this electrostatic relay 21, as shown in FIG. 2, by applying a voltage to the movable comb electrode 22a and the fixed comb electrode 23a, the lever 24a is elastically bent, and at the same time, the movable comb electrode 22b and The lever 24b is elastically bent by applying a voltage to the fixed comb-like electrode 23b, and the movable contact 25a formed at the tip of the lever 24a and the movable contact 25b formed at the tip of the lever 24b are brought into contact with each other to move. The contact 25a, 25b is closed. Further, the movable contacts 25a and 25b are separated from each other by releasing the applied voltage between the comb-shaped electrodes 22a and 23a and the movable comb-shaped electrodes 22b and 23b. In the electrostatic relay 21, the movable contacts 25a and 25b are formed by depositing a metal film by vapor deposition, sputtering or the like on the end portions of the levers 24a and 24b.

  The electrostatic relay 21 of Patent Document 2 also has a structure in which the surfaces of the movable contacts 25a and 25b formed on the end surfaces of the levers 24a and 24b are brought into contact with and separated from each other. Therefore, also in this electrostatic relay 21, the movable contacts 25a and 25b are in contact with each other on a plane perpendicular to the growth method of the deposited film or the like.

  However, the plane perpendicular to the growth direction of these contacts (contact surface) is considerably rough when viewed microscopically, and has irregular and fine irregularities. Therefore, when viewed finely, the contact area between the contacts is small, and the contact resistance when the contacts are closed is large. Furthermore, since it is difficult to obtain parallelism between the surfaces of the opposing contacts, the contact resistance between the contacts tends to increase excessively.

JP-T-2006-526267 Japanese Patent Laid-Open No. 9-251834

  The present invention has been made in view of such technical problems, and the object of the present invention is to form the contact surfaces of the contacts smoothly and to ensure that the contacts are brought into contact with each other. It is an object of the present invention to provide a switch that can reduce the contact resistance between them, a manufacturing method thereof, and a relay using the switch structure.

  The switch according to the present invention includes a first contact portion formed with a plurality of layers including a first contact layer above a first substrate, and a plurality including a second contact layer above the second substrate. And a second contact portion formed with a second contact portion, wherein an end surface of the first contact layer parallel to the growth direction of the contact layer is formed as a contact of the first contact portion. An end surface of the contact layer parallel to the growth direction at the time of film formation of the contact layer is used as a contact of the second contact portion, and at least one of the first contact portion and the second contact portion. The contact of the contact portion protrudes from the respective end surfaces of the contact layer other than the contact layer and the substrate of the contact portion, and the contact of the first contact portion and the contact of the second contact portion It is characterized in that both contacts are made to contact or separate from each other.

  In the switch of the present invention, since the end faces parallel to the growth direction of the first and second contact layers are used as the contact of the first contact portion and the contact of the second contact portion, respectively, the MEMS technology is used. Thus, when the first and second contact layers are formed, the surface that becomes each contact can be smoothed, and the parallelism between the contacts can be improved. Therefore, the substantial contact area between the contacts can be increased, and the contact resistance between the contacts can be decreased. Further, since the contact surfaces of both the contacts become smooth, the contacts are less likely to be welded to each other, and the open / close life of the switch is increased. Furthermore, since the distance between the contacts can be reduced, it is possible to open and close the contacts by driving the actuator with a low voltage.

  Further, in the switch of the present invention, in at least one of the first contact part and the second contact part, the contact of the contact part is a layer other than the contact layer in the contact part and the contact. Since the first contact portion and the second contact portion are in contact with each other, the layers other than the contact layer and the substrate are in contact with each other. The contact between the contact of the part and the contact of the second contact part is not hindered. Further, when the contacts are in contact with each other, contact between layers other than the contact layer can be prevented, and adhesion between layers other than the contact layer can be prevented to extend the contact life.

  In an embodiment of the switch according to the present invention, the first and second contact layers are formed of any one of a noble metal, an alloy, a conductive Si-based material, or a conductive oxide. It is characterized by that. According to such an embodiment, the first and second contact layers can be formed of a material having high hardness and a relatively small specific resistance.

  In another embodiment of the switch according to the present invention, the first contact portion is formed with a first wiring layer above the first substrate, and the first wiring layer is formed on an upper surface of the first wiring layer. A contact layer is formed, and the second contact portion is formed with a second wiring layer above the second substrate, and the second contact layer is formed on an upper surface of the second wiring layer. It is characterized by being. According to this embodiment, since the wiring layer for wiring and the contact layer having contacts for opening and closing can be separated, it is possible to select optimum materials for the wiring layer and the contact layer, respectively.

  In another embodiment of the switch according to the present invention, in the at least one contact portion where the contact protrudes from the respective end surfaces of the layer other than the contact layer and the substrate, the end surface of the wiring layer of the contact portion is It is characterized in that it is an inclined surface that gradually retracts from the edge of the contact portion on the side in contact with the contact layer toward the direction close to the substrate of the contact portion. According to such an embodiment, the protruding portions of the contact layer can be supported by the wiring layers while avoiding contact between the wiring layers.

  In still another embodiment of the switch according to the present invention, the first and second wiring layers are formed of any one of a noble metal, an alloy, a conductive Si-based material, or a conductive oxide. It is characterized by having. According to such an embodiment, it is possible to form the first and second wiring layers with a material having a small specific resistance and a relatively high hardness.

  A first manufacturing method of a switch according to the present invention is to form a plurality of layers including a contact layer above the substrate by growing a plurality of layers including the contact layer in the thickness direction of the substrate above the substrate. A step of forming a mold part having a predetermined pattern on the uppermost surface; and etching the plurality of layers including the contact layer using the mold part as a mask to divide the plurality of layers including the contact layer into a plurality of regions. Forming a contact surface by the etched surface of the contact layer, and subjecting the surface of the substrate to isotropic etching between the divided regions of the plurality of layers including the contact layer. Forming a recess and anisotropically etching the substrate between a plurality of divided regions including the contact layer, thereby dividing the plurality of layers including the contact layer. A step of dividing the substrate into a plurality of regions according to a region, and etching at least one of the divided regions by etching a layer other than the contact layer to form the end surface of the layer other than the contact layer. And a step of retracting from the surface to be the contact point. The contact layer is formed by a deposition method such as vapor deposition, sputtering, MBE, CVD, plating, spray method, sol-gel method, ink jet method, or screen printing.

  In the first manufacturing method of the switch according to the present invention, when the contact layer is etched and divided, the etched surface becomes a contact, so that the surface that becomes each contact can be smoothed. The parallelism between the contacts can also be improved. Therefore, the contact resistance between the contacts which can increase the substantial contact area between the contacts can be reduced. Further, since the contact surfaces of both the contacts become smooth, the contacts are less likely to be welded to each other, and the open / close life of the switch is increased. Furthermore, since the distance between the contacts can be reduced, it is possible to open and close the contacts by driving the actuator with a low voltage.

  Further, in the manufacturing method, each contact protrudes from the layer other than the contact layer and the respective end surfaces of the substrate of the contact portion. Therefore, before the contacts of the contact portions contact each other, Layers and substrates are in contact with each other, and contact between contacts is not hindered. Further, when the contacts are in contact with each other, contact between layers other than the contact layer can be prevented, and adhesion between layers other than the contact layer can be prevented to extend the contact life.

  The second manufacturing method of the switch according to the present invention includes forming a mold part having a predetermined pattern above the substrate, and including a plurality of contact layers in a plurality of regions excluding the region where the mold part is formed above the substrate. Forming a plurality of layers including the contact layer above the substrate by growing the layer in the thickness direction of the substrate, removing the mold part, and contacting the side of the mold part of the contact layer Forming a surface to be a contact by the surface, forming a recess in the surface of the substrate by isotropically etching the surface of the substrate between the separated regions of the plurality of layers including the contact layer, The substrate is divided into a plurality of regions corresponding to the divided regions of the plurality of layers including the contact layer by anisotropically etching the substrate between the separated regions of the plurality of layers including the contact layer. And, in at least one region of the separated regions, a step of etching back a layer other than the contact layer to recede an end surface of the layer other than the contact layer from a surface serving as a contact of the contact layer It is characterized by having. The contact layer is formed, for example, by any film forming method of vapor deposition, sputtering, PLD, MBE, ALD, MOCVD, thermal CVD, plating, spray method, sol-gel method, inkjet method, or screen printing.

  In the second manufacturing method of the switch according to the present invention, since the surface of the contact layer that is in contact with the mold portion becomes a contact, the surface that becomes each contact can be smoothed, and the contacts are The degree of parallelism can also be improved. Therefore, the contact resistance between the contacts which can increase the substantial contact area between the contacts can be reduced. Further, since the contact surfaces of both the contacts become smooth, the contacts are less likely to be welded to each other, and the open / close life of the switch is increased. Furthermore, since the distance between the contacts can be reduced, it is possible to open and close the contacts by driving the actuator with a low voltage.

  Further, in the manufacturing method, each contact protrudes from the layer other than the contact layer and the respective end surfaces of the substrate of the contact portion. Therefore, before the contacts of the contact portions contact each other, Layers and substrates are in contact with each other, and contact between contacts is not hindered. Further, when the contacts are in contact with each other, contact between layers other than the contact layer can be prevented, and adhesion between layers other than the contact layer can be prevented to extend the contact life.

  In another embodiment of the first or second manufacturing method of the switch according to the present invention, the plurality of layers including the contact layer are formed with contact layers on the upper surface of the wiring layer formed above the substrate. It is characterized by being. According to this embodiment, since the wiring layer for wiring and the contact layer having contacts for opening and closing can be separated, it is possible to select optimum materials for the wiring layer and the contact layer, respectively. This wiring layer can be formed by a deposition method such as vapor deposition, sputtering, MBE, CVD, plating, spray method, sol-gel method, ink jet method or screen printing.

  Further, in this embodiment, in the step of retracting the end face of the layer other than the contact layer from the face to be a contact of the contact layer, the end face of the wiring layer is increased as it goes from the contact layer side to the substrate. It can be tilted back. According to such an embodiment, the protruding portions of the contact layer can be supported by the wiring layers while avoiding contact between the wiring layers.

  The relay according to the present invention includes a switch according to the present invention, at least one of the first contact portion and the second contact portion as a contact between the first contact portion and the second contact portion. And an actuator for moving the contacts in a direction perpendicular to the contact surfaces to bring the contacts into contact with or apart from each other. In such a relay, since the contact surface between the contacts of the first contact portion and the second contact portion can be formed smoothly, the contact resistance when the contacts come into contact with each other is reduced. Further, since the contact surfaces of both the contacts become smooth, it becomes difficult to discharge when the contacts are close to each other, and welding between the contacts is difficult to occur. As a result, the life of the relay is increased. In addition, since the contact protrudes from the end face of the other layer or the end face of the substrate, the layers other than the contact layer and the substrate come into contact with each other before the contacts come into contact with each other. The contact with the contact of the contact portion is not hindered.

  The means for solving the above-described problems in the present invention has a feature in which the above-described constituent elements are appropriately combined, and the present invention enables many variations by combining such constituent elements. .

1A and 1B are cross-sectional views showing a MEMS switch disclosed in Patent Document 1. FIG. FIG. 2 is a perspective view of the electrostatic relay disclosed in Patent Document 2. As shown in FIG. 3A and 3B are cross-sectional views showing the structure of the switch according to Embodiment 1 of the present invention. 4A to 4D are schematic cross-sectional views illustrating a first manufacturing method of the switch of the first embodiment. FIG. 5A to FIG. 5D are schematic cross-sectional views showing a process following FIG. 6A to 6D are schematic cross-sectional views illustrating a second manufacturing method of the switch of the first embodiment. FIGS. 7A to 7D are schematic cross-sectional views showing a process following FIG. 6D. 8A to 8D are schematic cross-sectional views for explaining a third manufacturing method of the switch of the first embodiment. FIG. 9 is a cross-sectional view showing the structure of the switch according to Embodiment 2 of the present invention. 10A to 10D are schematic cross-sectional views illustrating a first method for manufacturing the switch of the second embodiment. FIGS. 11A to 11C are schematic cross-sectional views showing a process following FIG. 12A to 12D are schematic cross-sectional views illustrating a second manufacturing method of the switch of the second embodiment. FIGS. 13A to 13C are schematic cross-sectional views showing a process following FIG. FIG. 14 is a plan view showing an electrostatic relay according to Embodiment 3 of the present invention. FIG. 15 is an enlarged perspective view showing a portion A of FIG. FIG. 16 is a schematic cross-sectional view along the line BB in FIG.

  Hereinafter, preferred embodiments of the present invention will be described with reference to the accompanying drawings. However, the present invention is not limited to the following embodiments, and various design changes can be made without departing from the gist of the present invention.

[First Embodiment]
(Construction)
FIG. 3A is a cross-sectional view showing the structure of the switch according to Embodiment 1 of the present invention. The switch 31 includes a fixed contact portion 33 and a movable contact portion 34. The fixed contact portion 33 has its lower surface fixed to the upper surface of the base substrate 32 via the insulating film 42, and the movable contact portion 34 floats from the upper surface of the base substrate 32 and is parallel to the upper surface of the base substrate 32 by the actuator ( Move in the direction indicated by the white arrow). For example, the switch of the present invention can also be used for a MEMS switch having a structure as disclosed in Patent Document 1.

  The fixed contact portion 33 is obtained by providing a wiring pattern portion 48 on the upper surface of the fixed contact substrate 41. The wiring pattern portion 48 includes an adhesion layer 43 positioned on the upper surface of the fixed contact board 41, and a wiring layer 44 and a contact layer 45 laminated thereon. Further, the movable contact portion 34 is obtained by providing a wiring pattern portion 58 on the upper surface of the movable contact substrate 51. The wiring pattern portion 58 includes an adhesion layer 53 positioned on the upper surface of the movable contact substrate 51, and a wiring layer 54 and a contact layer 55 laminated thereon.

The adhesion layer 43 is a layer for increasing the adhesion strength (peeling strength) between the wiring layer 44 and the fixed contact substrate 41. The adhesion layer 53 is a layer for increasing the adhesion strength (peeling strength) between the wiring layer 54 and the movable contact substrate 51. The adhesion layers 43 and 53 have a two-layer structure made of, for example, lower layer Cr / upper layer Au, and are formed by a method such as CVD, vapor deposition, sputtering, electrolytic plating, or electroless plating. The wiring layers 44 and 54 are preferably formed of a material having a small specific resistance and a high hardness, and doped with a noble metal such as Pt, Rh, Pd, Au, an alloy, polysilicon (Poly-Si), or an impurity. It is made of a Si-based material such as doped silicon or doped polysilicon, or a conductive oxide such as AgO or SrRuO ₃ . The contact layers 45 and 55 are also preferably made of a material having a small specific resistance and a high hardness, such as noble metals such as Pt, Rh, Pd, Au, polysilicon, doped silicon, doped polysilicon. Or a conductive oxide such as AgO or SrRuO ₃ . The wiring layers 44 and 54 and the contact layers 45 and 55 are formed by a deposition method such as vapor deposition, sputtering, MBE, CVD, plating, spray method, sol-gel method, ink jet method, or screen printing.

  However, the contact layers 45 and 55 are layers for forming a fixed contact and a movable contact that are in contact with or away from each other, and a sticker (adhesion) is less likely to occur when the contacts are in contact with each other because of the high hardness material. Therefore, it is desirable to select the material of the contact layers 45 and 55 with priority on high hardness. On the other hand, the wiring layers 44 and 54 are layers for transmitting signals, and are not in direct contact with each other, and can be expected to have an effect of reducing the impact at the time of contact between the contacts even if they are somewhat soft. The wiring layers 44 and 54 are preferably selected with priority given to being a low-resistance material rather than having a high hardness. Accordingly, the wiring layers 44 and 54 and the contact layers 45 and 55 are preferably made of a material having low resistance and high hardness, but generally the wiring layers 44 and 54 are made of a material having a lower specific resistance than the contact layers 45 and 55. The contact layers 45 and 55 are formed of a material having a higher hardness than the wiring layers 44 and 54.

  The adhesion layers 43 and 53, the wiring layers 44 and 54, and the contact layers 45 and 55 are formed by growing respective materials in the thickness direction (arrow direction α in FIG. 3). Among the opposing end surfaces of the contact layer 45 and the contact layer 55, the opposing end surface of the contact layer 45 becomes a fixed contact 46 (electrical contact surface), and the opposing end surface of the contact layer 55 becomes a movable contact 56 (electrical contact surface). It has become. Therefore, the fixed contact 46 is an end surface parallel to the growth direction α of the contact layer 45 or a surface perpendicular to the surface of the contact layer 45. The movable contact 56 is also an end surface parallel to the growth direction α of the contact layer 55 or a surface perpendicular to the surface of the contact layer 55. The fixed contact 46 and the movable contact 56 are parallel to each other, and both are formed smoothly. However, the fixed contact 46 and the movable contact 56 do not necessarily have to be flat surfaces, and may be curved surfaces.

  On the surface facing the movable contact portion 34, the fixed contact 46 protrudes in the horizontal direction from the end surface of the fixed contact substrate 41 and the end surface of the adhesion layer 43. The end of the upper surface of the wiring layer 44 is aligned with the fixed contact 46 or is recessed more than the fixed contact 46, and the end surface 49 of the wiring layer 44 is retracted away from the movable contact portion 34 as it approaches the fixed contact substrate 41 side. It is out. Similarly, on the surface facing the fixed contact portion 33, the movable contact 56 protrudes in the horizontal direction from the end surface of the movable contact substrate 51 and the end surface of the adhesion layer 53. The end of the upper surface of the wiring layer 54 is aligned with the movable contact 56 or is recessed from the movable contact 56, and the end surface 59 of the wiring layer 54 is retracted away from the fixed contact portion 33 as it approaches the movable contact substrate 51 side. It is out.

  In the fixed contact portion 33 and the movable contact portion 34, an insulating layer may be formed between the adhesion layers 43 and 53 and the substrates 41 and 51.

  In this switch 31, when the movable contact portion 34 is moved in a direction parallel to the upper surface of the base substrate 32 by an actuator or the like, as shown in FIG. 3B, the fixed contact 46 of the fixed contact portion 33 and the movable contact portion 34. The movable contact 56 comes into contact, and the space between the fixed contact 46 and the movable contact 56 is electrically closed. Moreover, since the contact layers 45 and 55 protrude in the horizontal direction from the end surfaces of the wiring layers 44 and 54 and the adhesion layers 43 and 53 and the end surfaces of the substrates 41 and 51, the fixed contact 46 and the movable contact 56 come into contact with each other. The contact between the fixed contact 46 and the movable contact 56 is hindered by the contact between the wiring layers 44 and 54, the contact between the adhesion layers 43 and 53, or the contact between the substrates 41 and 51. There is nothing. Further, since the contact between the fixed contact 46 and the movable contact 56 prevents the wiring layers 44 and 54 and the contact layers 43 and 53 from contacting each other, the wiring layers 44 and 54 and the contact layers 43 and 53 have low hardness. Even when a material is used, the wiring layers 44 and 54 and the adhesion layers 43 and 53 do not stick to each other, and the contact life is not affected.

  Further, since the end surfaces 49 and 59 of the wiring layers 44 and 54 are inclined so as to protrude upward, the contact layers 45 and 55 are prevented by the wiring layers 44 and 54 while avoiding contact between the wiring layers 44 and 54. Each of the protruding portions can be supported.

  Moreover, if it is a structure like this switch 31, various manufacturing methods are employable as follows.

(First manufacturing method)
The switch 31 is manufactured using MEMS technology. 4A to 4D and FIGS. 5A to 5D illustrate an example of a manufacturing process of the switch 31. FIG.

  FIG. 4A shows a state in which the adhesion layer A3 is formed on the upper surface of the substrate A1 made of Si by a method such as vapor deposition or sputtering. For the adhesion layer A3, a material having high adhesion, such as Cr or Ti, is used for the lower layer, and a low resistance material such as Au, Cu, or Al is further formed thereon. After the adhesion layer A3 is formed on the upper surface of the substrate A1, a photoresist is applied to the upper surface of the adhesion layer A3, and the photoresist is patterned by a photolithography technique. As shown in FIG. The mold portion A2 is provided in a region other than a region where the wiring pattern portions 48 and 58 are to be formed.

  Next, as shown in FIG. 4C, the wiring layer material is deposited on the adhesion layer A3 by a method such as vapor deposition, sputtering, or electrolytic plating, and the wiring layer is formed in the region where the wiring pattern portions 48 and 58 are to be formed. Laminate A4. Subsequently, the material of the contact layer is deposited on the wiring layer A4 by a method such as vapor deposition, sputtering, or electrolytic plating, and the contact layer A5 is laminated in the region where the wiring pattern portions 48 and 58 are to be formed.

  Thereafter, when the mold part A2 is peeled off by being immersed in a stripping solution, the wiring layers 44 and 54 and the contact layers 45 and 55 are formed in the regions where the wiring pattern parts 48 and 58 are to be formed, as shown in FIG. Is formed. Further, the end surfaces of the contact layers 45 and 55 that have been in contact with the mold part A2 are formed in a smooth and parallel manner to become a fixed contact 46 and a movable contact 56, respectively.

  Next, the contact layer A3 is selectively etched using an etching solution with which the wiring layers 44 and 54, the contact layers 45 and 55, and the substrate A1 are resistant, and as shown in FIG. The regions exposed from the layers 45 and 55 are removed and the adhesion layer A3 is over-etched so that the edge of the adhesion layer A3 is retracted from the edge of the wiring layers 44 and 54, and the adhesion layers 43 and 53 are patterned.

  Thereafter, in the region A6 between the contact layer 45 and the contact layer 55, the substrate A1 is isotropically etched using the contact layers 45 and 55 as a mask. At this time, as shown in FIG. 5B, the contact layers 45 and 55, the wiring layers 44 and 54, and the adhesion layers 43 and 53 use an etching method having corrosion resistance, and the upper surface of the substrate A1 is between the adhesion layers 43 and 53. The substrate A1 is over-etched so as to be etched with a width wider than the opening width, and a recess A7 is provided on the upper surface of the substrate A1. As a method for isotropically etching the substrate A1, for example, RIE (Reactive Ion Etching) is performed using sulfur hexafluoride and perfluorocyclobutane as gas species (for example, under conditions of a pressure of 10 to 100 Pa and a high frequency power of 50 to 200 W). ). As other methods for isotropic etching, there are a dry etching method using xenon gas as a gas species and a wet etching method using a hydrofluoric acid solution.

  When the recess A7 is formed on the upper surface of the substrate A1 as shown in FIG. 5B, the substrate A1 is further anisotropically etched from the recess A7 side using the contact layers 45 and 55 as a mask, as shown in FIG. In addition, the substrate A1 is divided into the fixed contact substrate 41 and the movable contact substrate 51 by anisotropic etching, the end surface of the fixed contact substrate 41 is retracted from the fixed contact 46, and the end surface of the movable contact substrate 51 is more than the movable contact 56. Retract. As a method of anisotropic etching, for example, DRIE (Deep Reactive Ion Etching) is performed using sulfur hexafluoride as a gas species (for example, performed under conditions of a pressure of 3 to 10 Pa and a high frequency power of 200 to 800 W). In addition to this, there are other methods of performing anisotropic etching, such as ion milling, wet etching using a KOH aqueous solution and a TMAH solution. Further, the distance between the fixed contact substrate 41 and the movable contact substrate 51 after anisotropic etching (or the extent to which the end surfaces of the fixed contact substrate 41 and the movable contact substrate 51 are retracted) is the width of the recess A7 in FIG. Can be controlled by.

  Further, the end faces 49 and 59 of the wiring layers 44 and 54 are inclined by etching (etching back) the end faces of the wiring layers 44 and 54. In FIG. 5D, the ends of the upper surfaces of the wiring layers 44 and 54 are aligned with the fixed contact 46 and the movable contact 56, but may be retracted from the fixed contact 46 and the movable contact 56. Note that the end surfaces 49 and 59 of the wiring layers 44 and 54 do not have to be inclined surfaces, and as long as they are retracted from the fixed contact 46 and the movable contact 56, they may be vertical surfaces parallel to both the contacts 46 and 56.

  Thus, one block becomes the fixed contact portion 33 in which the fixed contact substrate 41, the adhesion layer 43, the wiring layer 44, and the contact layer 45 are laminated. The fixed contact portion 33 is fixed to the upper surface of the base substrate 32 through an insulating film 42. The other block is the movable contact portion 34 in which the movable contact substrate 51, the adhesion layer 53, the wiring layer 54, and the contact layer 55 are laminated. The movable contact 34 is finally separated from the base substrate 32 by etching away the insulating film on the lower surface. As a result, the switch 31 (MEMS switch) is manufactured.

  In the switch 31 thus manufactured, the surfaces to be the fixed contact 46 and the movable contact 56 are parallel to the growth direction of the contact layers 45 and 55, and are formed by both side surfaces of the mold part A2. Therefore, it can form smoothly compared with the surface of the contact layers 45 and 55, and parallelism also improves. Therefore, both the contacts 46 and 56 can be reliably brought into contact with each other, and the contact resistance between the contacts can be reduced. Further, since the contact surfaces of both the contacts 46 and 56 can be made smooth, it is difficult to discharge when the contacts are close to each other, the welding of the fixed contact 46 and the movable contact 56 is less likely to occur, and the open / close life of the switch 31 is increased. Increase.

  Furthermore, according to such a manufacturing method, the distance between the fixed contact 46 and the movable contact 56 can be accurately determined by the width of the mold part A2, and discharge occurs between the contacts as described above. Therefore, the distance between the fixed contact 46 and the movable contact 56 can be reduced, and the actuator can be driven with a low voltage to open and close the contacts.

(Second manufacturing method)
Further, the switch 31 can also be manufactured by the steps shown in FIGS. 6A to 6D and FIGS. 7A to 7D. Hereinafter, the second manufacturing method will be described.

  First, as shown in FIG. 6A, an adhesion layer A3 is formed on the upper surface of a substrate A1 made of Si by a method such as vapor deposition or sputtering, and a wiring layer A4 and a contact layer A5 are further stacked on the upper surface.

  Next, a photoresist is applied on the contact layer A5 and patterned to form a mold portion A2 in a region where the wiring pattern portions 48 and 58 are to be formed as shown in FIG. 6B. When the mold part A2 is patterned, as shown in FIG. 6C, the exposed region of the contact layer A5 is selectively etched using the mold part A2 as a mask, and the contact layers 45 and 55 are formed on the upper surface of the wiring layer A4. Patterning is performed, and a fixed contact 46 and a movable contact 56 are formed on the end faces of the contact layers 45 and 55, respectively. Further, the etching solution is changed, and as shown in FIG. 6D, the exposed region of the wiring layer A4 is selectively etched using the mold part A2 as a mask, and the wiring layers 44 and 54 are patterned on the adhesion layer A3. .

  Further, the contact layer A3 is selectively etched using an etchant with which the contact layers 45 and 55 and the substrate A1 are resistant, and exposed from the contact layers 45 and 55 of the contact layer A3 as shown in FIG. Then, the adhesion layer A3 is over-etched so that the edge of the adhesion layer A3 is withdrawn from the edge of the wiring layers 44 and 54, and the adhesion layers 43 and 53 are patterned. Next, the mold part A2 is peeled off by dipping in a stripping solution.

  Thereafter, through steps similar to those of FIGS. 5B to 5D of the first manufacturing method, a recess A7 is formed on the upper surface of the substrate A1 by isotropic etching (FIG. 7B), and the substrate A1 is formed. Is etched into a fixed contact substrate 41 and a movable contact substrate 51 (FIG. 7 (c)), and end faces 49 and 59 of the wiring layers 44 and 54 are etched back (FIG. 7 (d)). 31 is produced.

  Even in such a method, the fixed contact 46 and the movable contact 56 are formed by etching the contact layer A5 using the mold portion A2 as a mask, so that the fixed contact 46 and the movable contact 56 are made smooth and parallel to each other. Can be formed. Further, the distance between the contact points of the fixed contact 46 and the movable contact 56 can be determined with high accuracy.

(Third production method)
Further, the switch 31 can also be manufactured by the steps shown in FIGS. 4A to 4D and FIGS. 8A to 8D. Also in the third manufacturing method, first, the adhesion layer A3 is formed on the upper surface of the substrate A1 by the steps of FIGS. 4A to 4D, and the regions other than the regions where the wiring pattern portions 48 and 58 are to be formed are formed. After providing the mold part A2 and laminating the wiring layer A4 and the contact layer A5 on the adhesion layer A3 in the region where the wiring pattern parts 48 and 58 are to be formed, the mold part A2 is peeled off with a stripping solution. Since the steps of FIGS. 4A to 4D have already been described, description thereof will be omitted.

  In the third manufacturing method, after forming the wiring layers 44 and 54 and the contact layers 45 and 55 on the adhesion layer A3 by the steps of FIGS. 4A to 4D, as shown in FIG. The adhesion layer A3 is selectively etched using an etchant with which the contact layers 45 and 55 and the substrate A1 are resistant. As a result, the regions exposed from the contact layers 45 and 55 of the adhesion layer A3 are removed and the adhesion layer A3 is over-etched so that the edge of the adhesion layer A3 is retracted from the edge of the wiring layers 44 and 54.

  Thereafter, in the region A6 between the contact layer 45 and the contact layer 55, the substrate A1 is anisotropically etched from the upper surface side using the contact layers 45 and 55 as a mask, and the substrate A1 is fixedly contacted as shown in FIG. The substrate 41 and the movable contact substrate 51 are divided. As a method of anisotropic etching, for example, DRIE is performed using sulfur hexafluoride as a gas species. In addition to this, there are other methods of performing anisotropic etching, such as ion milling, wet etching using a KOH aqueous solution and a TMAH solution.

  Next, the fixed contact substrate 41 and the movable contact substrate 51 are isotropically etched from above using the contact layers 45 and 55 as a mask, and the upper surface angles of the fixed contact substrate 41 and the movable contact substrate 51 are shown in FIG. A recess A7 is formed in the part. At this time, the contact layers 45 and 55, the wiring layers 44 and 54, and the adhesion layers 43 and 53 are etched using an etching method, and the upper surface of the substrate A1 is etched with a width wider than the opening width between the adhesion layers 43 and 53. The substrate A1 is over-etched so that As a method of isotropically etching the fixed contact substrate 41 and the movable contact substrate 51, for example, RIE is performed using sulfur hexafluoride and perfluorocyclobutane as gas species. As other methods for isotropic etching, there are a dry etching method using xenon gas as a gas species and a wet etching method using a hydrofluoric acid solution.

  Further, the end surfaces of the wiring layers 44 and 54 are etched (etched back), and the end surfaces 49 and 59 of the wiring layers 44 and 54 are inclined as shown in FIG. In FIG. 5D, the ends of the upper surfaces of the wiring layers 44 and 54 are aligned with the fixed contact 46 and the movable contact 56, but may be retracted from the fixed contact 46 and the movable contact 56. Further, when the end faces 49 and 59 of the wiring layers 44 and 54 are etched back, the fixed contact board 41 and the movable contact board 51 are further etched simultaneously, and the end face of the fixed contact board 41 is retracted from the end face of the adhesion layer 43. At the same time, it is desirable to retract the end surface of the movable contact substrate 51 from the end surface of the adhesion layer 53.

[Second Embodiment]
(Construction)
FIG. 9 is a sectional view showing the structure of a switch 31A according to Embodiment 2 of the present invention. In this switch 31 </ b> A, the contact layer 45 is formed directly on the adhesion layer 43 formed on the upper surface of the fixed contact substrate 41 to form the fixed contact portion 33, which is formed on the upper surface of the movable contact substrate 51. The contact layer 55 is formed directly on the adhesion layer 53 to constitute the movable contact portion 34. The point which the end surfaces which the contact layers 45 and 55 mutually oppose becomes the fixed contact 46 and the movable contact 56 is the same as Embodiment 1. Therefore, compared with the switch 31 of the first embodiment, the switch 31A does not have the wiring layers 44 and 54, the wiring pattern portion 48 has a two-layer structure of the adhesion layer 43 and the contact layer 45, and the wiring pattern portion 58 also adheres closely. The layer 53 and the contact layer 55 have a two-layer structure, and the contact layers 45 and 55 have both a function of bringing the contacts into contact with each other and a function of transmitting a signal (function of the wiring layer).

(First manufacturing method)
10A to 10D and FIGS. 11A to 11C show an example of the manufacturing process of the switch 31A.

  FIG. 10A shows a state in which the adhesion layer A3 is formed on the upper surface of the substrate A1 made of Si by a method such as vapor deposition or sputtering. For the adhesion layer A3, a material having high adhesion, such as Cr or Ti, is used for the lower layer, and a low resistance material such as Au, Cu, or Al is further formed thereon. After the adhesion layer A3 is formed on the upper surface of the substrate A1, a photoresist is applied to the upper surface of the adhesion layer A3, and the photoresist is patterned by a photolithography technique. As shown in FIG. The mold portion A2 is provided in a region other than a region where the wiring pattern portions 48 and 58 are to be formed.

  Next, as shown in FIG. 10C, a contact layer material is deposited on the adhesion layer A3 by a method such as vapor deposition, sputtering, or electrolytic plating, and the contact layer is formed in the region where the wiring pattern portions 48 and 58 are to be formed. Laminate A5.

  Thereafter, when the mold portion A2 is removed, contact layers 45 and 55 are formed in regions where the wiring pattern portions 48 and 58 are to be formed, as shown in FIG. As a result, the end surfaces of the contact layers 45 and 55 that have been in contact with the mold part A2 are formed smoothly and in parallel with each other to become the fixed contact 46 and the movable contact 56, respectively.

  Next, the adhesion layer A3 is selectively etched using an etchant with which the contact layers 45 and 55 and the substrate A1 are resistant, and exposed from the contact layers 45 and 55 of the adhesion layer A3 as shown in FIG. Then, the adhesion layer A3 is over-etched so that the edge of the adhesion layer A3 is withdrawn from the edge of the contact layers 45 and 55, and the adhesion layers 43 and 53 are patterned.

  Thereafter, in the region A6 between the contact layer 45 and the contact layer 55, the substrate A1 is isotropically etched using the contact layers 45 and 55 as a mask. At this time, as shown in FIG. 11B, an etching method in which the contact layers 45 and 55 and the adhesion layers 43 and 53 are resistant is used, and the upper surface of the substrate A1 is wider than the opening width between the adhesion layers 43 and 53. The substrate A1 is over-etched so as to be etched with a width, and a recess A7 is provided on the upper surface of the substrate A1. As a method for isotropically etching the substrate A1, for example, RIE is performed using sulfur hexafluoride and perfluorocyclobutane as gas species (for example, performed under conditions of a pressure of 10 to 100 Pa and a high frequency power of 50 to 200 W). As other methods for isotropic etching, there are a dry etching method using xenon gas as a gas species and a wet etching method using a hydrofluoric acid solution.

  When the recess A7 is formed on the upper surface of the substrate A1 as shown in FIG. 11B, the substrate A1 is further anisotropically etched from the recess A7 side using the contact layers 45 and 55 as a mask, as shown in FIG. The substrate A1 is divided into the fixed contact substrate 41 and the movable contact substrate 51 by anisotropic etching, and etching is performed until the end surface of the fixed contact substrate 41 and the end surface of the movable contact substrate 51 are retracted from the ends of the adhesion layers 43 and 53, respectively. To do. As a method of anisotropic etching, for example, DRIE is performed using sulfur hexafluoride as a gas species (for example, performed under conditions of a pressure of 3 to 10 Pa and a high frequency power of 200 to 800 W). In addition to this, there are other methods of performing anisotropic etching, such as ion milling, wet etching using a KOH aqueous solution and a TMAH solution. Further, the distance between the fixed contact substrate 41 and the movable contact substrate 51 after anisotropic etching (or the extent to which the end surfaces of the fixed contact substrate 41 and the movable contact substrate 51 are retracted) is the width of the recess A7 in FIG. Can be controlled by.

  Thus, one block becomes the fixed contact portion 33 in which the fixed contact substrate 41, the adhesion layer 43, and the contact layer 45 are laminated. The fixed contact portion 33 is fixed to the upper surface of the base substrate 32 through an insulating film 42. The other block is the movable contact portion 34 in which the movable contact substrate 51, the adhesion layer 53, and the contact layer 55 are laminated. The movable contact 34 is finally separated from the base substrate 32 by etching away the insulating film on the lower surface. In this way, the switch 31A (MEMS switch) is manufactured.

  Even in the thus-produced switch 31A, the surfaces to be the fixed contact 46 and the movable contact 56 are formed by the both side surfaces of the mold part A2, and therefore, they should be formed more smoothly than the surfaces of the contact layers 45 and 55. In addition, parallelism is improved. Therefore, both the contacts 46 and 56 can be reliably brought into contact with each other, and the contact resistance between the contacts can be reduced. Further, since the contact surfaces of both the contacts 46 and 56 can be made smooth, it is difficult to discharge when the contacts are close to each other, the welding of the fixed contact 46 and the movable contact 56 is less likely to occur, and the open / close life of the switch 31A is increased. Increase.

  Furthermore, according to such a manufacturing method, the distance between the fixed contact 46 and the movable contact 56 can be manufactured with high accuracy without variation depending on the width of the mold part A2, and discharge occurs between the contacts as described above. Therefore, the distance between the fixed contact 46 and the movable contact 56 can be reduced, and the actuator can be driven with a low voltage to open and close the contacts.

(Second manufacturing method)
Further, the switch 31A can also be manufactured by the steps shown in FIGS. 12 (a) to 12 (d) and FIGS. 13 (a) to 13 (c). Hereinafter, the second manufacturing method will be described.

  First, as shown in FIG. 12A, an adhesion layer A3 is formed on the upper surface of a substrate A1 made of Si by a method such as vapor deposition or sputtering, and a contact layer A5 is formed on the upper surface.

  Next, a photoresist is applied on the contact layer A5 and patterned to form a mold portion A2 in a region where the wiring pattern portions 48 and 58 are to be formed as shown in FIG. When the mold part A2 is patterned, as shown in FIG. 12C, the exposed region of the contact layer A5 is selectively etched using the mold part A2 as a mask to pattern the contact layers 45 and 55 and the contact layer 45 , 55 are formed with a fixed contact 46 and a movable contact 56, respectively.

  Further, the adhesion layer A3 is selectively etched using an etchant having resistance to the contact layers 45 and 55 and the substrate A1, and exposed from the contact layers 45 and 55 of the adhesion layer A3 as shown in FIG. Then, the adhesion layer A3 is over-etched so that the edge of the adhesion layer A3 is withdrawn from the edge of the contact layers 45 and 55, and the adhesion layers 43 and 53 are patterned.

  Thereafter, in the region A6 between the contact layer 45 and the contact layer 55, the substrate A1 is isotropically etched using the mold part A2 as a mask. At this time, as shown in FIG. 13A, an etching method in which the contact layers 45 and 55 and the adhesion layers 43 and 53 are resistant is used, and the upper surface of the substrate A1 is wider than the opening width between the adhesion layers 43 and 53. The substrate A1 is over-etched so as to be etched with a width, and a recess A7 is provided on the upper surface of the substrate A1.

  Thus, when the recess A7 is formed on the upper surface of the substrate A1 as shown in FIG. 13A, the substrate A1 is further anisotropically etched from the recess A7 side using the mold part A2 as a mask, and the difference as shown in FIG. The substrate A1 is divided into the fixed contact substrate 41 and the movable contact substrate 51 by isotropic etching, and the end surface of the fixed contact substrate 41 and the end surface of the movable contact substrate 51 are retracted from the ends of the adhesion layers 43 and 53, respectively.

  Thereafter, the mold part A2 on the wiring layers 44 and 54 is peeled off by the peeling liquid, and the switch 31A is manufactured.

  In the switch 31A manufactured in this way, since the surfaces to be the fixed contact 46 and the movable contact 56 are formed by etching, they can be formed more smoothly than the surfaces of the contact layers 45 and 55, and in parallel. The degree is also improved. Therefore, both the contacts 46 and 56 can be reliably brought into contact with each other, and the contact resistance between the contacts can be reduced. Further, since the contact surfaces of both the contacts 46 and 56 can be made smooth, it is difficult to discharge when the contacts are close to each other, the welding of the fixed contact 46 and the movable contact 56 is less likely to occur, and the open / close life of the switch 31A is increased. Increase. Further, the distance between the fixed contact 46 and the movable contact 56 can be reduced, and the actuator can be driven with a low voltage to open and close the contacts.

  In the switch 31, both the fixed contact 46 and the movable contact 56 protrude from the end surfaces of the substrates 41 and 51, the adhesion layers 43 and 53, and the wiring layers 44 and 54. In the switch 31A, the fixed contact 46 and the movable contact 56 are fixed. Although both the contact 46 and the movable contact 56 protrude from the end faces of the substrates 41 and 51 and the adhesion layers 43 and 53, only one of the fixed contact 46 and the movable contact 56 protrudes, and the other The contacts may be aligned with the edges of the substrate or adhesive layer.

[Third Embodiment]
Next, the structure of a high-frequency electrostatic relay 31B according to Embodiment 3 of the present invention will be described. FIG. 14 is a plan view showing the structure of the electrostatic relay 31B. FIG. 15 is an enlarged perspective view showing a portion A of FIG. 14, and FIG. 16 is a schematic cross-sectional view taken along the line BB of FIG.

  The electrostatic relay 31B includes a fixed contact portion 33, a movable contact portion 34, a fixed electrode portion 35, a movable electrode portion 36 that supports the movable contact portion 34, an elastic spring on the upper surface of a base substrate 32 made of a Si substrate, a glass substrate, or the like. 37, a support portion 38 for supporting the elastic spring 37 is provided.

As shown in FIG. 16, in the fixed contact portion 33, the lower surface of the fixed contact substrate 41 made of Si is fixed to the upper surface of the base substrate 32 by an insulating film 42 (SiO ₂ ). As shown in FIG. 15, the upper surface of the fixed contact substrate 41 is made of a material having a high adhesion (for example, a material such as Cr or Ti) as a lower layer, and further a low resistance material (for example, Au, Cu, or the like). Adhesion layers 43a and 43b having a two-layer structure formed with a material such as Al are formed, and wiring layers 44a and 44b such as Pt and contact layers 45a and 45b are laminated on the adhesion layers 43a and 43b. Yes.

  As shown in FIGS. 14 and 15, the fixed contact board 41 extends in the width direction (X direction) at the end of the upper surface of the base board 32, and protrudes toward the movable contact part 34 at the center. An overhang portion 41a is formed, and pad support portions 41b and 41b are formed at both ends, respectively. The wiring pattern portions 48a and 48b are wired along the upper surface of the fixed contact board 41, and one end portions of the wiring pattern portions 48a and 48b are arranged parallel to each other on the upper surface of the overhanging portion 41a. The tip surfaces of the contact layers 45a and 45b protruding from the end surfaces are located in the same plane and are fixed contacts 46a and 46b (electrical contact surfaces), respectively. Metal pad portions 47a and 47b are formed on the upper surfaces of the pad support portions 41b and 41b at the other ends of the wiring pattern portions 48a and 48b. When the wiring pattern portion 48 has a three-layer structure of the adhesion layers 43a and 43b, the wiring layers 44a and 44b, and the contact layers 45a and 45b as in the switch 31 of FIG. 3, the contact layers 45a and 45b It is not always necessary to provide the wiring pattern portions 48 a and 48 b as a whole.

  The movable contact portion 34 is provided at a position facing the overhanging portion 41a. As shown in FIG. 15, the movable contact portion 34 has an adhesion layer 53 made of lower layer Cr / upper layer Au formed on the upper surface of a movable contact substrate 51 made of Si, and a wiring layer such as Pt on the adhesion layer 53. 54 and a contact layer 55 are laminated. As shown in FIG. 16, the end surface of the contact layer 55 facing the contact layers 45a and 45b protrudes from the front surface of the movable contact substrate 51 and is formed in parallel with the fixed contacts 46a and 46b. 56 (electrical contact surface). The movable contact 56 has a width substantially equal to the distance from the outer edge of the fixed contact 46a to the outer edge of the fixed contact 46b.

  In addition, the movable contact substrate 51 is supported in a cantilever manner by a support beam 57 protruding from the movable electrode portion 36. The lower surfaces of the movable contact substrate 51 and the support beam 57 float from the upper surface of the base substrate 32 and can move in parallel with the length direction (Y direction) of the base substrate 32 together with the movable electrode portion 36.

  In this electrostatic relay 31B, a main circuit (not shown) is connected to the metal pad portions 47a and 47b of the fixed contact portion 33, and the main circuit is closed by bringing the movable contact 56 into contact with the fixed contacts 46a and 46b. The main circuit can be opened by separating the movable contact 56 from the fixed contacts 46a and 46b. Further, the end surfaces of the wiring layers 44a, 44b, 54 are inclined so as to recede as they go downward, and the end surfaces of the overhanging portion 41a and the movable contact substrate 51 are also fixed from the fixed contacts 46a, 46b and the movable contact 56, respectively. Since it is retracted, the wiring layers 44a, 44b and the wiring layer 54 come into contact with each other when the contacts are closed, and the overhanging portion 41a and the movable contact substrate 51 come into contact with each other, so that the movable contact 56 and the fixed contacts 46a, 46b Prevents poor contact.

  An actuator for moving the movable contact portion 34 includes a fixed electrode portion 35, a movable electrode portion 36, an elastic spring 37, and a support portion 38.

  As shown in FIG. 14, a plurality of fixed electrode portions 35 are arranged on the upper surface of the base substrate 32 in parallel with each other. The fixed electrode portion 35 has branch-like electrode portions 67 extending in a branch shape from both surfaces of the rectangular pad portion 66 in the Y direction in plan view. The branch electrode portions 67 protrude from the branch electrode portions 67 so as to be bilaterally symmetric, and the branch portions 68 are arranged at a constant pitch in the Y direction.

  As shown in FIG. 16, in the fixed electrode portion 35, the lower surface of the fixed electrode substrate 61 is fixed to the upper surface of the base substrate 32 by an insulating film 62. In the pad portion 66, a fixed electrode 63 is formed on the upper surface of the fixed electrode substrate 61 with Cu, Al, or the like, and an electrode pad layer 65 is provided on the fixed electrode 63.

  As shown in FIG. 14, the movable electrode portion 36 is formed so as to surround each fixed electrode portion 35. Comb-like electrode portions 74 are formed in the movable electrode portion 36 so as to sandwich each fixed electrode portion 35 from both sides (a pair of comb-like electrode portions 74 form a branch shape between the fixed electrode portions 35). ing). The comb-like electrode portions 74 are symmetric with respect to each fixed electrode portion 35, and the comb-tooth portions 75 extend from each comb-like electrode portion 74 toward the gap between the branch portions 68. Yes. In addition, each comb tooth 75 has a distance from the branch 68 positioned adjacent to the comb tooth 75 and closer to the movable contact 34 so that the comb tooth 75 is adjacent to the comb tooth 75 from the movable contact 34. It is shorter than the distance to the branch portion 68 located on the far side.

  The movable electrode portion 36 is composed of a Si movable electrode substrate 71, and the lower surface of the movable electrode substrate 71 is lifted from the upper surface of the base substrate 32. Further, a support beam 57 projects from the center of the movable contact side end face of the movable electrode portion 36, and the movable contact portion 34 is held at the tip of the support beam 57.

  The support portion 38 is made of Si and extends long in the X direction at the other end portion of the base substrate 32. The lower surface of the support portion 38 is fixed to the upper surface of the base substrate 32 by an insulating film 39. Both end portions of the support portion 38 and the movable electrode portion 36 (movable electrode substrate 71) are connected by a pair of elastic springs 37 formed symmetrically by Si, and the movable electrode portion 36 is supported via the elastic springs 37. It is supported horizontally by the portion 38. The movable electrode portion 36 is movable in the Y direction by elastically deforming the elastic spring 37.

  In the electrostatic relay 31B having the above structure, a DC voltage source is connected between the fixed electrode portion 35 and the movable electrode portion 36, and the DC voltage is turned on and off by a control circuit or the like. In the fixed electrode portion 35, one terminal of the DC voltage source is connected to the electrode pad layer 65. The other terminal of the DC voltage source is connected to the support portion 38. Since the support portion 38 and the elastic spring 37 have electrical conductivity, and the support portion 38, the elastic spring 37, and the movable electrode portion 36 are electrically connected, the voltage applied to the support portion 38 is the movable electrode portion 36. Will join.

  When a DC voltage is applied between the fixed electrode portion 35 and the movable electrode portion 36 by a DC voltage source, the static electricity is generated between the branch portion 68 of the branch electrode portion 67 and the comb tooth portion 75 of the comb tooth electrode portion 74. Electric attraction is generated. However, since the structures of the fixed electrode portion 35 and the movable electrode portion 36 are formed symmetrically with respect to the center line of each fixed electrode portion 35, the electrostatic attractive force in the X direction acting on the movable electrode portion 36 is balanced, and the movable electrode The part 36 does not move in the X direction. On the other hand, the distance between each comb tooth portion 75 and the branch portion 68 located on the side close to the movable contact portion 34 is the branch portion located on the side far from the movable contact portion 34 adjacent to the comb tooth portion 75. Since each comb tooth 75 is attracted to the movable contact portion side and the elastic spring 37 is bent, the movable electrode portion 36 moves in the Y direction. As a result, the movable contact portion 34 moves to the fixed contact portion 33 side, the movable contact 56 contacts the fixed contacts 46a and 46b, and the space between the fixed contact 46a and the fixed contact 46b (main circuit) is electrically closed.

  Further, when the DC voltage applied between the fixed electrode part 35 and the movable electrode part 36 is released, the electrostatic attractive force between the branch part 68 and the comb tooth part 75 disappears. As a result, the movable electrode part 36 moves backward in the Y direction, the movable contact 56 is separated from the fixed contacts 46a and 46b, and the space between the fixed contact 46a and the fixed contact 46b (main circuit) is opened.

  Such an electrostatic relay 31B is manufactured by the following process. First, a Si substrate (another Si wafer having conductivity) is bonded to the upper surface of a base substrate 32 (Si wafer, SOI wafer, etc.) whose entire surface is covered with an insulating film, and a metal material is bonded to the upper surface of the Si substrate. An electrode film is formed by vapor deposition. Next, this electrode film is patterned by a photolithography technique, and a fixed electrode 63 is formed on the upper surface of the fixed electrode substrate 61 in the pad portion 66 by the electrode film.

  Subsequently, an adhesion layer is formed on the upper surface of the Si substrate from above the electrode film, and a wiring layer and a contact layer are laminated thereon. Next, the contact layer, the wiring layer, and the adhesion layer are patterned to form the wiring pattern portion 48 of the fixed contact portion 33 and the wiring pattern portion 58 of the movable contact portion 34. In addition, an electrode pad layer 65 is formed on the fixed electrode 63 in the pad portion 66.

  Thereafter, the Si substrate is etched using a photoresist or the like as an etching mask, and the fixed contact substrate 41 of the fixed contact portion 33, the movable contact substrate 51 of the movable contact portion 34, and the fixed electrode portion 35 of the Si substrate remaining in each region are etched. The fixed electrode substrate 61, the movable electrode substrate 71 of the movable electrode portion 36, the elastic spring 37, and the support portion 38 are produced.

  Finally, the insulating film in the region exposed from the Si substrate and the insulating film on the lower surface of the movable contact portion 34 and the movable electrode portion 36 are removed by etching and cut into individual electrostatic relays 31B.

  In the manufacturing process of the electrostatic relay 31B, the movable contact part 34 and the fixed electrode part 35 are manufactured by the same process as that described in relation to the switch 31 of the first embodiment, for example. The fixed contacts 46a and 46b and the movable contact 56 of the movable contact portion 34 are side surfaces parallel to the growth direction of the contact layer, and a contact having good smoothness and parallelism can be obtained without polishing. Therefore, also in this electrostatic relay 31B, the same effect as the switch 31 of Embodiment 1 can be obtained.

31, 31A Switch 31B Electrostatic relay 32 Base substrate 33 Fixed contact portion 34 Movable contact portion 35 Fixed electrode portion 36 Movable electrode portion 37 Elastic spring 41 Fixed contact substrate 51 Movable contact substrate 43, 53 Adhesion layers 44, 44a, 44b, 54 Wiring layer 45, 45a, 45b, 55 Contact layer 46, 46a, 46b Fixed contact 56 Movable contact 63 Fixed electrode 67 Branched electrode part 74 Comb electrode part A1 Substrate A2 Mold part A3 Adhesion layer A4 Wiring layer A5 Contact layer A7 Recess

Claims (12)

A first contact portion formed with a plurality of layers including a first contact layer above the first substrate and a plurality of layers including a second contact layer formed above the second substrate. Two contact points,
In the first contact layer, an end surface parallel to the growth direction at the time of film formation of the contact layer is used as a contact of the first contact portion.
In the second contact layer, an end surface parallel to the growth direction at the time of film formation of the contact layer is used as a contact of the second contact portion.
In at least one contact portion of the first contact portion and the second contact portion, the contact of the contact portion is more than the layers other than the contact layer in the contact portion and the respective end surfaces of the substrate of the contact portion. Protruding,
The switch according to claim 1, wherein the contact of the first contact portion and the contact of the second contact portion are opposed to each other so that the two contacts are in contact with or separated from each other.   The first and second contact layers are formed of any one of a noble metal, an alloy, a conductive Si-based material, and a conductive oxide. Switch. The first contact portion has a first wiring layer formed above the first substrate, and the first contact layer is formed on an upper surface of the first wiring layer.
The second contact portion has a second wiring layer formed above the second substrate, and the second contact layer is formed on an upper surface of the second wiring layer. The switch according to claim 1.   In at least one of the contact portions where the contact protrudes from the respective end surfaces of the layer other than the contact layer and the substrate, the end surface of the wiring layer of the contact portion extends from the edge of the contact portion on the side in contact with the contact layer. The switch according to claim 3, wherein the switch is an inclined surface that is gradually retracted toward a direction close to the substrate of the part.   The said 1st and 2nd wiring layer is formed with the material in any one of a noble metal, an alloy, Si-type material which has electroconductivity, or an electroconductive oxide, It is characterized by the above-mentioned. Switch. A plurality of layers including a contact layer are grown in the thickness direction of the substrate above the substrate to form a plurality of layers including the contact layer above the substrate, and a mold portion having a predetermined pattern is formed on the uppermost surface thereof. Process,
Etching a plurality of layers including the contact layer using the mold portion as a mask divides the plurality of layers including the contact layer into a plurality of regions, and forms a contact surface by the etched surface of the contact layer. And a process of
Forming a recess in the surface of the substrate by isotropically etching the surface of the substrate between the divided regions of the plurality of layers including the contact layer;
Dividing the substrate into a plurality of regions according to the divided regions of the plurality of layers including the contact layer by anisotropically etching the substrate between the divided regions of the plurality of layers including the contact layer;
In at least one region of the divided regions, the step of etching the layer other than the contact layer to recede the end surface of the layer other than the contact layer from the surface serving as the contact of the contact layer; and
A method for manufacturing a switch, comprising: Form a mold part of a predetermined pattern above the substrate,
A plurality of layers including the contact layer above the substrate by growing a plurality of layers including a contact layer in a thickness direction of the substrate in a plurality of regions excluding the region where the mold part is formed above the substrate. Forming a step;
Removing the mold part, and forming a surface to be a contact by a surface that is in contact with the side surface of the mold part of the contact layer;
Forming a recess in the surface of the substrate by isotropically etching the surface of the substrate between the separated regions of the plurality of layers including the contact layer;
Dividing the substrate into a plurality of regions according to the separated regions of the plurality of layers including the contact layer by anisotropically etching the substrate between the separated regions of the plurality of layers including the contact layer; and
Retreating an end face of a layer other than the contact layer from a face to be a contact of the contact layer by etching a layer other than the contact layer in at least one of the separated areas; and
A method for manufacturing a switch, comprising:   The switch according to claim 6 or 7, wherein the contact layer is formed by a deposition method such as vapor deposition, sputtering, MBE, CVD, plating, spray method, sol-gel method, inkjet method, or screen printing. Production method.   The method for manufacturing a switch according to claim 6 or 7, wherein the plurality of layers including the contact layer are formed by forming contact layers on an upper surface of a wiring layer formed above the substrate. .   In the step of retracting the end face of the layer other than the contact layer from the face to be a contact of the contact layer, the end face of the wiring layer is inclined so as to recede greatly from the contact layer side toward the substrate. The method of manufacturing a switch according to claim 9.   10. The method of manufacturing a switch according to claim 9, wherein the wiring layer is formed by a deposition method such as vapor deposition, sputtering, MBE, CVD, plating, spray method, sol-gel method, ink jet method, or screen printing. .   The switch according to claim 1, and at least one of the first contact portion and the second contact portion is a contact surface between the contacts of the first contact portion and the second contact portion. A relay comprising: an actuator for moving in a vertical direction to contact or separate the contacts.

Images