JP2005235719A - Electromechanical switch and its manufacturing method - Google Patents

Abstract

<P>PROBLEM TO BE SOLVED: To provide a small sized, thin, and light weighted electromechanical switch in which a response speed of switching is made greater, and integration with other circuit constituting components on the substrate is realized. <P>SOLUTION: In the electromechanical switch M1, a recessed part 2 of which opposed side walls 3, 3 are respectively inclined faces is formed on a substrate 1, a lower electrode 5 is installed on the bottom face 4 of the recessed part 2, an upper movable electrode 7 deformable so that its center part may contact the lower electrode 5 is bridged over the opposed side walls 3, 3 of the recessed part 2, and drive electrodes 8 which are respectively opposed to each other on both sides of the center part of the upper movable electrode 7 are installed on the side wall 3 of the recessed part 2. The electromechanical switch in which response speed is made greater, small sized, thin, and light weighted, and which copes with a high speed transmission system and communication equipment, is provided. <P>COPYRIGHT: (C)2005,JPO&NCIPI

Description

  The present invention relates to an electromechanical switch, and more particularly to an electromechanical switch that uses electrostatic force and is used as an electrostatic microswitch or electrostatic microrelay as a wireless device, a portable information terminal, or the like.

  In recent years, rapid advances in the field of telecommunications have been supported by improvements in electronic devices and systems that enable high-speed transfer of large amounts of information. Among such electronic devices and systems, electronic signal path switching switches are important in all communication systems.

  At present, solid-state switches including MESFETs and PIN diodes are mainly used as the electronic signal path switching switches, and switches having sufficient characteristics at high frequencies are particularly useful.

  Among these solid state switches, PIN diodes are commonly used RF switches. However, PIN diodes must always be forward-biased to obtain the carriers necessary for signal transmission in a low impedance state, and the switching characteristics are non-linear and the breakdown voltage is low. Or a large insertion loss at a high frequency.

  Further, a micro mechanical switch may be used instead of the solid switch. Conventional micromechanical switches function like normal mechanical switches, have low insertion loss, good isolation, can handle large power, and require less switching and static power to drive There are many advantages.

  However, the micromechanical switch has a problem that it is difficult to respond at high speed when the size is reduced and the volume occupied by the micromechanical switch is reduced. An example of the configuration of a microelectromechanical (MEM: microelectromechanical) switch which is one of such conventional micromechanical switches is shown in Patent Document 1 or Patent Document 2, for example.

  The MEM switch disclosed in Patent Document 1 includes an anchor structure 112, a lower electrode 113, an open circuit formed on a substrate 111, as shown in a schematic cross-sectional view of the MEM switch MM1 in FIG. A signal line 114 having a gap forming a gap, a cantilever arm 115 attached to the anchor structure 112 and extending over the lower electrode 113 and the signal line gap, and spaced from the anchor structure 112 on the cantilever arm 115 And a cantilever arm 112 positioned above the lower electrode 113, and a contact portion 116 positioned facing the gap of the signal line 114 and an upper electrode 117 formed on the cantilever arm 115. And a part of the upper electrode 117 form a capacitor structure that can be electrostatically attracted toward the lower electrode 113 when a voltage is selectively applied to the upper electrode 117.

  Patent Document 2 also discloses an anchor structure formed on a substrate, a bottom electrode, and two independent signal lines, each of which is a signal line having a gap forming an open circuit; A spring suspension attached to the anchor structure and extending in a direction substantially perpendicular to the bottom electrode direction; a shorting piece attached to the other end of the spring suspension and formed at a portion remote from the spring suspension A micro-platform structure having a short-circuit piece disposed in a manner facing a gap in the signal line; and a short-circuit formed on the signal line and disposed to face the short-circuit piece. An electrical contact post that forms a capacitor structure that is attracted towards the bottom electrode by static electricity when a voltage is selectively applied on the piece; Switch is disclosed.

  MEM devices such as MEM switches such as these are also integrated with other control circuits in order to operate sufficiently well in the microwave region. For example, passive components (resistors, capacitors and inductors) and at least one for operating as a single pole double throw switch (SPDT) to drive the flow of power signals to and from other components in the microwave system Along with one other switch, a MEM switch is placed in the integrated circuit.

  One problem arises when trying to realize the integration of circuits of these types with silicon integrated circuit elements. That is, the conditions of the temperature treatment process vary widely between MEM devices (having members like electrodes as components) and passive components (like bias resistors not having members like electrodes). As a result, the electrical resistance of the signal line increases due to the diffusion of the electrode material, and the electrode peels off due to the deterioration of the adhesion of the electrode.

Therefore, it is required to efficiently manufacture a device in which a MEM switch is integrated by simultaneously manufacturing a passive component and a MEM device. Among them, the development of a smaller and faster electromechanical switch is important, and its development is energetically advanced, and improvement in performance is desired in the last few years.
Japanese Patent Laid-Open No. 9-17300 JP 2001-143595 A

  However, in the conventional electromechanical switch, the distance between the upper electrode and the lower electrode is formed constant, and it is necessary to increase the area of the upper electrode or the lower electrode to ensure electrostatic force. There is a limit to downsizing, and it is difficult to increase the speed. For example, in the electromechanical switch having the conventional structure shown in Patent Document 1, Patent Document 2, and the like, the electrode interval for generating the driving electrostatic force is constant, so the switching time is small. Since the time is sometimes slow, there is a problem that a large voltage is required to speed up the response time.

  In general, the electrostatic force (Coulomb attractive force) generated between the upper electrode and the lower electrode is proportional to the square of the applied voltage, is proportional to the area of the electrodes facing each other, and is 2 of the distance between the electrodes facing each other. Inversely proportional to the power. Therefore, in order to simultaneously realize a lower drive voltage and a higher switching speed, it is conceivable to increase the area of the opposing electrodes or reduce the interval between the opposing electrodes. However, these methods are not actually effective methods. Increasing the area of the opposing electrodes increases the mass of the movable electrode if the electrode thickness is the same, which is disadvantageous for speeding up. On the other hand, reducing the thickness of the electrode causes a problem in terms of mechanical strength. In addition, reducing the interval between the electrodes facing each other generates a large parasitic capacitance between the electrodes. Therefore, if the electrical signal to be turned on / off is a high frequency, the electrical signal can be satisfactorily transmitted even in the off state. It becomes impossible to turn off, resulting in a problem that the isolation is lowered.

  Therefore, there is a limit to speeding up the response of an electromechanical switch having a conventional structure, and it is difficult in principle to achieve both low driving voltage and high-speed switching. Therefore, development of higher-speed communication equipment is advanced. Therefore, the realization of an electromechanical switch corresponding to a high-speed response is desired.

  The present invention has been devised to solve the above-described problems in the prior art, and its purpose is to increase the response speed of switching, and to reduce the size, thickness, An object of the present invention is to provide an electromechanical switch which can be reduced in weight and can be integrated with other circuit components on a substrate.

  In the electromechanical switch of the present invention, concave portions each having an inclined side wall on the substrate are formed, a lower electrode is disposed on the bottom surface of the concave portion, and a central portion is disposed between the opposing side walls of the concave portion. An upper movable electrode that can be deformed so as to move to the bottom surface side and be electrically connected to the lower electrode is spanned, and the side wall of the recess is opposed to both sides of the central portion of the upper movable electrode. A drive electrode is provided.

  In the electromechanical switch of the present invention, in the above configuration, the upper movable electrode is disposed on the lower surface of the second substrate that is spanned between the opposing side walls of the recess, and the second substrate is It is spanned between the side walls of the recess and joined to the substrate.

  In the method of manufacturing the electromechanical switch of the present invention, the recesses whose inclined side walls are inclined on the substrate are formed, the lower electrode is disposed on the bottom surface of the recess, and the driving electrodes are disposed on the sidewalls. A step of filling the concave portion with a sacrificial layer, spanning the sacrificial layer between the opposing side walls, a central portion facing the lower electrode, and both sides of the central portion being the drive electrodes A step of forming the upper movable electrode so as to face the substrate, and then a step of removing the sacrificial layer.

  Further, the electromechanical switch manufacturing method of the present invention is the step of forming the concave portion in the above configuration, wherein polysilane is applied on the substrate, and the portion of the applied polysilane that becomes the concave portion is irradiated with light. The method further comprises the step of forming the concave portion by removing the portion irradiated with light by immersing in an alkaline solution and then heating the portion of polysilane applied by heating in an oxygen atmosphere to silicon oxide as the portion not irradiated with light. It is a feature.

  Furthermore, the electromechanical switch manufacturing method of the present invention is characterized in that, in the above-described configuration, in the light irradiation, the light amount is changed stepwise or continuously with respect to the portion of the recess that becomes the side wall. It is.

  According to the electromechanical switch of the present invention, the recesses whose inclined side walls are inclined are formed on the substrate, the lower electrode is disposed on the bottom surface of the recess, and the central portion is between the opposing side walls of the recess. The upper movable electrode, which can be deformed so as to move to the bottom surface side and be electrically connected to the lower electrode, is stretched over, and driving electrodes are arranged on the side walls of the recesses so as to face both sides of the central portion of the upper movable electrode. Since the distance between the upper movable electrode and the driving electrode is not constant but gradually increases, the electrostatic force generated between the upper movable electrode and the driving electrode is large. Therefore, the upper movable electrode starts to move downward even at a small applied voltage in the portion near the base of the upper movable electrode, and the adjacent parts sequentially move downward accordingly. It made. As a result, the electrode interval becomes shorter than the initial interval, so that it can be driven with a small applied voltage. In addition, the concave part with a sloped side wall provided on the substrate can generate the necessary electrostatic force with a small applied voltage, so that the switching response speed can be increased without significantly changing the manufacturing process. Become. In addition, by forming the recess without removing the surface of the substrate by etching or the like, a part of the components can be integrated on the substrate, and the size, thickness, and weight can be reduced. . Therefore, according to the electromechanical switch of the present invention, the electrostatic force required for driving the upper movable electrode can be obtained with a small applied voltage, and the response speed can be increased, so that it is compatible with higher-speed transmission systems and communication devices. An electromechanical switch can be provided.

  According to the electromechanical switch of the present invention, the upper movable electrode is disposed on the lower surface of the second substrate that is spanned between the opposing side walls of the recess, and the second substrate has the recess formed therein. When the substrate is formed to be bridged between the sidewalls of the recess and joined, the upper movable electrode is disposed on the lower surface of the substrate on which the recess is formed. Since each substrate is manufactured through a separate process and finally combined, the manufacturing process of each substrate can be individually optimized and designed, which simplifies the process and allows the desired upper movable parts of various shapes Since the electrodes can be stably formed and disposed on the lower surface of the second substrate, the yield can be improved.

  In addition, according to the method for manufacturing an electromechanical switch of the present invention, a concave portion having inclined side walls on the substrate is formed, a lower electrode is provided on the bottom surface of the concave portion, and a driving electrode is provided on the side wall. A step of filling the concave portion with a sacrificial layer, spanning between the side walls facing the sacrificial layer, the central portion facing the lower electrode, and both sides of the central portion facing the driving electrode, respectively. Since the method includes a step of forming the upper movable electrode and a step of removing the sacrificial layer after that, a space between the upper movable electrode and the driving electrode is formed on the substrate by forming a recess having a sloped side wall. Is not constant but gradually increases from the base of the upper movable electrode to the central portion, and then the concave portion is filled with the sacrificial layer. Less In addition, since the wall surface of the recess is an inclined surface, no bubbles are formed at the corner where the bottom surface and the wall surface of the recess intersect, so it is not necessary to remove the bubbles generated in the sacrificial layer in the step of filling the sacrificial layer. As a result, the process can be simplified.

  In addition, since such a manufacturing method can be performed by a wafer process, a small and high performance electromechanical switch excellent in integration can be provided at low cost.

  In addition, according to the method for manufacturing an electromechanical switch of the present invention, the step of forming concave portions each having a sloped side wall facing the substrate, the polysilane being applied on the substrate, the concave portion of the applied polysilane and A step of forming a concave portion using silicon oxide as a non-light-irradiated portion of polysilane applied by heating in an oxygen atmosphere after removing the light-irradiated portion by irradiating the portion with light and immersing in an alkaline solution. Further, when it is provided, a special vacuum device such as a dry etching device is not required, and a concave portion having an arbitrary shape is formed by arbitrarily selecting a light irradiation portion that irradiates light onto, for example, polysilane coated on a substrate. Therefore, it is possible to easily form a silicon oxide film having a desired shape, so that the step of forming a recess on the substrate can be simplified.

  Furthermore, according to the method for manufacturing an electromechanical switch of the present invention, in the light irradiation to the concave portion of the polysilane coated on the substrate, the amount of light is stepwise or continuously applied to the portion that becomes the side wall of the concave portion. Is not required, a special vacuum apparatus such as a dry etching apparatus is not required, and a silicon oxide film forming an arbitrary three-dimensional shape can be easily formed. A silicon oxide film having a desired shape for forming a concave portion can be formed, and the electromechanical switch can be easily reduced in size and speed.

  And by using the electromechanical switch of the present invention, an arbitrary interval can be set unlike the case where the interval between the upper movable electrode and the driving electrode is constant, so that the lower electrode, the upper movable electrode and The degree of freedom of the electrode shape for the driving electrode is increased, and the degree of freedom for the arrangement of the driving electrode in the concave portion whose side wall provided on the substrate is an inclined surface is increased. It can easily cope with speeding up. Therefore, even when the frequency is increased, it is possible to design an optimum signal line for removing high-frequency noise for the lower electrode and the upper movable electrode, and integration with other circuit components on the substrate is also possible. It is possible to provide a low-cost electromechanical switch that is small and has a very simple structure.

  Hereinafter, an electromechanical switch and a manufacturing method thereof according to the present invention will be described in detail with reference to the drawings.

  FIG. 1 is a schematic plan view showing an electromechanical switch M1 which is an example of an embodiment of an electromechanical switch of the present invention, and FIG. 2 is a cross-sectional view taken along line AA ′ in FIG. Is a cross-sectional view taken along the line AA ′ at the time of ON. In these drawings, 1 is a substrate, 2 is a recess, 3 is a side wall of the recess 2, 4 is a bottom surface of the recess 2, 5 is a lower electrode, 6 is a signal line connected to the lower electrode 5, 7 is an upper movable electrode, 8 is a driving electrode, and 9 is a material for forming a recess.

  As shown in FIGS. 1 and 2, the electromechanical switch M1 of the present invention has a concave portion 2 in which the side walls 3 and 3 facing each other are inclined on the substrate 1, and a lower portion 4 is formed on the bottom surface 4 of the concave portion 2. An electrode 5 is disposed, and a deformable upper movable electrode 7 is stretched between the opposing side walls 3, 3 of the recess 2 so that the central portion contacts the lower electrode 5, and the upper movable electrode is placed on the side wall 3 of the recess 2. The drive electrodes 8 are disposed on both sides of the central portion 7 so as to face each other. When the electromechanical switch M1 of the present invention is turned off, as shown in FIG. 2, the upper movable electrode 7 is, for example, horizontally stretched between the upper ends of the side walls 3 and 3 of the recess 2, and the recess 2 It is disposed on the bottom surface 4 and is disposed integrally with the lower electrode 5 so as to be electrically insulated from the signal line 6 connected to the lower electrode 5, facing the lower electrode 5 and the driving electrode 8. Yes.

  When the switch is turned on, as shown in FIG. 3, by applying a voltage between the upper movable electrode 7 and the lower electrode 5, the central portion of the upper movable electrode 7 is electrostatically attracted to the lower electrode 5, and the recess 2 The upper movable electrode 7 comes into contact with the signal line 6 that is integrally disposed with the lower electrode 5 and connected thereto, and is electrically connected to the lower electrode 5. At this time, according to the electromechanical switch M1 of the present invention, the driving electrodes are disposed on the side walls 3 and 3 that are the inclined surfaces of the recess 2 over which the upper movable electrode 7 is stretched so as to face the upper movable electrode 7. 8 is arranged, both sides of the central portion of the upper movable electrode 7 are electrostatically applied to the drive electrode 8 by applying a drive voltage between the drive electrode 8 and the upper movable electrode 7. Thus, the central portion of the upper movable electrode 7 is deformed toward the bottom surface 4 of the recess 2 and can be assisted by applying a driving force, and the static electricity toward the lower electrode 5 of the upper movable electrode 7 can be supported. It is possible to speed up the deformation operation by increasing the force.

  The electromechanical switch M1 of the present invention forms the recess 2 on the substrate 1 by using the recess forming material 9 or by processing the upper surface of the substrate 1, and the lower electrode 5 on the bottom surface 4 of the recess 2. And the signal line 6 is configured by disposing the driving electrode 8 on the side wall 3 and spanning the plate-like upper movable electrode 7 between the upper ends of the side walls 3 of the recess 2. There is no significant change or difficulty, and it is advantageous for downsizing, thinning, and weight reduction. Therefore, according to the electromechanical switch M1 of the present invention, the electromechanical switch M1 is suitable for a MEM switch that is required to operate at high speed, and can be used for high-speed and large-capacity information transmission system equipment.

  One of the greatest advantages of MEM switches is good isolation characteristics. In order to obtain this good isolation characteristic, it is necessary to ensure a sufficient distance between the lower electrode 5 and the signal line 6 and the upper movable electrode 7. On the other hand, it is desirable that the distance between the upper movable electrode 7 and the driving electrode 8 be as small as possible. On the other hand, in the electromechanical switch M1 of the present invention, the lower electrode 5 and the signal line 6 are disposed on the bottom surface 4 of the recess 2, and the upper movable electrode 7 is stretched between the opposing side walls 3 and 3 of the recess 3, By disposing the driving electrode 8 with the opposite side walls 3 and 3 of the concave portion 3 as inclined surfaces, the upper movable electrode 7 is provided with sufficient space between the lower electrode 5 and the signal line 6 and the upper movable electrode 7. On the other hand, by disposing the driving electrode 8 at an angle and close to each other, good isolation characteristics, a low driving voltage and a high switching speed are realized.

For example, as shows the principle of parallel electrodes model the inclined electrode model respectively in FIGS. 4 (a) and (b), the movable electrode (a) the width length in W is l ₂ -l ₁ (upper) and driving electrode (lower side) are arranged in parallel at intervals g (e.g. 2 mm), when the electrostatic attractive force acting on the movable electrode upon application of a driving voltage V in this case as F _1,
F ₁ = ε ₀ · W · V ² · (l ₁ −l ₂ ) / (2 g)
It becomes. On the other hand, (b) the width length in W the maximum interval is _l 2 -l ₁ of the movable electrode (upper side) and the driving electrode (lower side) and g (e.g. 2 mm), the _{l 1} on the opposite side extended length minute position are arranged obliquely so that the fulcrum, when the electrostatic attractive force acting on the movable electrode upon application of a driving voltage V in this case and F _2,
F ₂ = ε ₀ · W · V ² · (1 / l ₁ −1 / l ₂ ) / ( ² tan ² θ)
It becomes. Here, when l ₁ = 20 μm, l ₂ = 200 μm, and tan θ = 1/100, a value of F ₂ / F ₁ = 10 is obtained. That is, in this case, by arranging the movable electrode and the driving electrode obliquely, it is possible to obtain a driving force 10 times that of the parallel arrangement. It can be seen that switching can be performed.

  In the electromechanical switch M1 of the present invention, the substrate 1 is preferably an insulating substrate made of an insulating material such as silicon, glass, alumina, or zirconia because it is preferable that the loss when transmitting an electric signal is small. In addition, when the MEM switch is configured, the substrate 1 has a length of 1.5 mm or less, a width of 1 mm or less, and a thickness of 1 mm or less so that a region for forming the switch can be obtained. Is preferred.

  The recess 2 formed on the upper surface of the substrate 1 may be formed by bonding a recess forming material 9 made of an insulating material similar to the substrate 1 on the substrate 1 as shown in FIG. You may form by processing the upper surface of 1. The side wall 3 of the recess 2 has inclined side walls 3 and 3 facing the upper movable electrode 7 between upper ends thereof, and a driving electrode 8 is disposed on the inclined surface.

  For the slopes of the side walls 3 and 3 of the recess 2, the maximum height of the slope defines the distance between the upper movable electrode 7 and the lower electrode 5. 3 μm or less is preferable. The interval between the inclined surfaces where the bottom surface 4 and the side walls 3 and 3 intersect is preferably 150 μm or less, for example, so that a region for forming the lower electrode 5 and the signal line 6 can be obtained. In addition, in the case of configuring a MEM switch, it is preferable that the slope has a length of 220 μm or less and a width of 170 μm or less so that a region for forming the driving electrode 8 can be obtained on the slope.

  The bottom surface 4 of the recess 2 is provided with a lower electrode 5 and a signal line 6 connected thereto on this surface, so that the portion where these are disposed is a central portion of the opposed upper movable electrode 7. The plane is preferably parallel to the plane. Furthermore, an insulating material such as a thermal oxide film may be formed on the surface of the bottom surface 4 of the recess 2 in order to suppress leakage from the lower electrode 5 and the signal line 6 to the substrate 1 side during signal transmission.

  The lower electrode 5 disposed on the bottom surface 4 of the recess 2 acts to electrostatically attract the upper movable electrode 7 to the lower electrode 5 by applying a voltage between the lower movable electrode 7 and the signal line 6. It is intended to improve the adhesion with the upper movable electrode 7 and to reduce leakage from the signal line 6 during signal transmission. It adheres well to the substrate 1 and is electrostatically connected to the upper movable electrode 7. A material capable of generating a large force is preferable, and it is desirable to use Au, Al, Cu or the like. Further, if the MEM switch is configured, the configuration of the lower electrode 5 is a stacked configuration of Ti / Pt / Au, Cr / Au, etc. from the lower side (substrate 1 side). 300 μm or less and a width of 100 μm or less, and the thickness of each layer is preferably 0.1 μm / 0.2 μm / 0.5 μm in the former case, and 0.5 μm / 1 μm in the latter case.

  A signal line 6 disposed on the bottom surface 4 of the recess 2 and connected to the lower electrode 5 is provided below the position facing the central portion of the upper movable electrode 7 with a certain interval, and transmits a signal when turned on. Are electrically connected to the upper movable electrode 7 and are made of a material having a small electric resistance, and Au, Cu, Al, or the like is preferable. In the case of configuring a MEM switch, for example, the length is preferably 300 μm or less, the width is 70 μm or less, and the film thickness is preferably 1 μm if it is Au.

  Here, an example is shown in which the signal line 6 is disposed on the lower electrode 5, but this arrangement reduces the loss during signal transmission, and the lower electrode 5 is connected to the substrate 1 and the signal line 6. It plays a role in improving adhesion. The signal line 6 does not necessarily have to be disposed on the lower electrode 5. If the signal line 6 is connected to the lower electrode 5 and is electrically connected to the upper movable electrode 7 when turned on, the lower electrode 5 And the signal line 6 may be arranged so as to be adjacent to each other at a predetermined interval in the same surface of the bottom surface 4 of the central portion 2. Further, the signal line 6 may be disposed on the bottom surface 4 of the central portion 2, and the lower electrode 5 may also be disposed on the side wall 3 of the central portion 2 serving as the driving electrode 8.

The upper movable electrode 7, which is stretched between the opposing side walls 3, 3 of the recess 2 and can be deformed so that the central portion contacts the lower electrode 5, moves downward, with the central portion being on the substrate 1 side. The electrode 5 or the signal line 6 is in contact with the signal line 6 and is electrically connected to transmit the signal, or the central portion moves upward to leave the lower electrode 5 or the signal line 6 to be in the off state. It is desirable that the drive response to electrostatic force be fast. Au, Cr, Al, Ni or the like may be used for the upper movable electrode 7 as described above. The upper movable electrode 7 preferably has a multilayer structure in order to reduce warping due to internal stress during formation or use. For example, if the MEM switch is to be configured, the upper movable electrode 7 is formed from the lower side, for example, SiO 2 ₂ / Cr / Au laminated structure, preferably having a length of 750 μm or less, a width of 300 μm or less, and a thickness of each layer of 0.2 μm / 0.2 μm / 0.3 μm.

  The driving electrode 8 disposed on the side wall 3 of the recess 2 so as to face both sides of the central portion of the upper movable electrode 7 applies a voltage between the upper movable electrode 7 and electrostatically. Force is generated, and thereby the upper movable electrode 7 is driven, and an electrostatic force can be generated between the upper movable electrode 7 and the adhesion to the side wall 3 that is a slope is improved. High is desirable. For such a driving electrode 8, Au, Cr, Al or the like may be used. In addition, as a configuration of the driving electrode 8, in the case of configuring a MEM switch, for example, it is a laminated structure such as Ti / Pt / Au or Cr / Au from the lower side, and the dimension is 300 μm or less in length. The width is preferably 100 μm or less, and the thickness of each layer is preferably 0.1 μm / 0.2 μm / 0.5 μm in the former case and 0.5 μm / 1 μm in the latter case.

  In the electromechanical switch M1 of the present invention configured as described above, an electrostatic force directed to the lower electrode 5 of the upper movable electrode 7 is strengthened by the driving electrode 8. That is, since the distance between the upper movable electrode 7 and the driving electrode 8 is not constant but gradually increases, the magnitude of the electrostatic force generated between the upper movable electrode 7 and the driving electrode 8 is increased. In the portion close to the base of the upper movable electrode 7, the upper movable electrode 7 starts to move downward even at a small applied voltage, and the adjacent portions sequentially move downward accordingly. As a result, since the electrode interval between the upper movable electrode 7 and the driving electrode 8 is shorter than the initial interval, the deformation operation of the upper movable electrode 7 can be speeded up with a small applied voltage. In addition, it is advantageous for miniaturization, thinning, and weight reduction without significant changes or difficulties in the manufacturing process, and is suitable for MEM switches that need to be operated at high speed. It is compatible with high-speed and large-capacity information transmission system equipment.

  Such an electromechanical switch M1 of the present invention is used in an electric signal transmission system in a measuring device such as a communication device such as a mobile phone or a semiconductor tester, and can reduce attenuation of a transmission signal when turned on. Since signal leakage at the time of off can be suppressed, a device capable of high-speed response and stable operation is obtained.

  FIG. 6 is a schematic perspective view showing an electromechanical switch M2 which is another example of the embodiment of the electromechanical switch of the present invention, and FIG. 7 is a cross-sectional view taken along the line AA ′ in the off state. FIG. 8 is a cross-sectional view taken along line AA ′ at the time of turning on. In these drawings, 11 is a substrate, 12 is a recess, 13 is a side wall of the recess 12, 14 is a bottom surface of the recess 12, 15 is a lower electrode, 17 is an upper movable electrode, 18 is a drive electrode, and 19 is a recess forming material. These are the same as those in the electromechanical switch M1 in the example shown in FIGS.

  In this example, the recess forming material 19 is formed on the upper surface of the substrate 11 as a film made of an insulating material, and a part of the film is removed to form the recess 12 in which the opposing side walls 13 and 13 are inclined surfaces. Yes. The lower electrode 15 is disposed so as to serve also as a signal line from the bottom surface 14 of the recess 12 to the upper surface of the recess forming material 19 as a substrate, and a part of the portion located directly below the upper movable electrode 17 is disconnected. It has a structure. This disconnected portion is a portion that is electrically connected when the upper movable electrode 17 moves and contacts, and the lower portion of the lower electrode 15 is disconnected on the lower surface of the upper movable electrode 17 that contacts the disconnected portion. In order that the conductive portion for connecting the electrode has good conductivity and can be connected to the upper movable electrode 17 while ensuring good insulation, it is opposed to the lower electrode 15. On the bottom surface, for example, a contact electrode 20 having a conductive portion made of an Au layer having a thickness of several thousand angstroms, which is chemically stable and has good conductivity, is connected to the conductive portion of the contact electrode 20 and the upper movable electrode 17. For example, an insulating layer made of silicon oxide having a thickness of about several hundred to several thousand angstroms is interposed between the main body and the main body. In addition, when silicon oxide is used for this insulating layer, a thin film such as Cr having a film thickness of about several hundred angstroms, for example, is oxidized in order to improve the adhesion with the Au layer which is the conductive part of the contact electrode 20. It may be provided between the silicon layer and the Au layer.

  Driving electrodes 17 are arranged on a part of the side walls 13 and 13 of the recess 12 so as to face both sides of the central portion of the upper movable electrode 17, respectively. The upper movable electrode 17 is disposed so as to span between the opposite side walls 13 and 13 of the recess 12, and is disposed so as to face the lower electrode 15 and the driving electrode 17 with a predetermined gap therebetween.

  When the electromechanical switch M2 of the present invention is also turned off, the upper movable electrode 17 is, for example, horizontally stretched between the upper ends of the side walls 13 and 13 of the recess 12 as shown in FIG. It is disposed on the bottom surface 14 and is opposed to the lower electrode 15 and the driving electrode 18 while being electrically insulated from the lower electrode 15.

  When the switch is turned on, as shown in FIG. 8, by applying a voltage between the upper movable electrode 17 and the lower electrode 15, the central portion of the upper movable electrode 17 is electrostatically attracted to the lower electrode 15, and the recess 12 The upper movable electrode 17 contacts the disconnected portion of the lower electrode 15 with the contact electrode 20 and is electrically connected to the lower electrode 15, and the lower electrode 15 is disconnected. The parts are also electrically connected. Thus, by electrically connecting the disconnected portion of the lower electrode 15 with the contact electrode 20, it is possible to control on / off of an electric signal propagating through the lower electrode 15 that also serves as a signal line. At this time, according to the electromechanical switch M2 of the present invention, the driving electrodes are arranged on the side walls 13 and 13 which are the inclined surfaces of the recess 12 over which the upper movable electrode 17 is stretched so as to face the upper movable electrode 17. 18 is arranged, both sides of the central portion of the upper movable electrode 17 are electrostatically driven by applying a driving voltage between the driving electrode 18 and the upper movable electrode 17. Thus, the central portion of the upper movable electrode 17 is deformed toward the bottom surface 14 of the recess 12 and can be assisted to give a driving force, and static electricity toward the lower electrode 15 of the upper movable electrode 17 It is possible to speed up the deformation operation by increasing the force.

  According to the electromechanical switch M2 of the present invention, low voltage driving is essential and high-speed on / off switching is necessary. For example, the electromechanical switch M2 is suitable for an antenna changeover switch for a mobile phone or the like. It will be compatible with capacity information transmission system equipment.

  FIG. 9 is a schematic cross-sectional view (when off) showing an electromechanical switch M3 which is still another example of the embodiment of the electromechanical switch of the present invention. In FIG. 9, 31 is a substrate, 32 is a recess, 33 is a side wall of the recess 32, 34 is a bottom surface of the recess 32, 35 is a lower electrode, 37 is an upper movable electrode, 38 is a drive electrode, 39 is a recess forming material, Reference numeral 40 denotes a contact electrode, which is the same as that in the electromechanical switches M1 and M2 in the examples shown in FIGS. 1 to 3 and FIGS.

  This electromechanical switch M3 is an example in which the slopes of the opposing side walls 33 and 33 of the recess 32 are formed in multiple stages. The manufacturing process is the same as that of the electromechanical switch M2 described later. The inclination angle of the slope of the side wall 33 of the recess 32 is, for example, using polysilane for the recess forming material 39, based on the good linearity of the photosensitive light irradiation amount and the reaction in the depth direction, It can be easily set by setting the opening amount of the photomask in multiple stages. In the electromechanical switch of the present invention, the switching speed is limited by an operation at an extremely early stage in which the upper movable electrode is attracted to the driving electrode. In the electromechanical switch M3 in this example, the slopes of the side walls 33 and 33 of the recess 32 are multi-staged to reduce the angle with respect to the upper movable electrode 36 on the upper side of the driving electrode 38, so that the upper movable electrode 36 and the driving electrode 38 are reduced. , The driving force generated in the upper movable electrode 36 is abruptly increased because it is inversely proportional to the square of the distance between the two electrodes, thereby increasing the initial driving force applied to the upper movable electrode 36. It can be set, and higher speed driving becomes possible. In order to increase the speed of the electromechanical switch, it is extremely important how the upper movable electrode in a stationary state can reach a predetermined speed in a short time, and the side walls 33 and 33 of the recess 32 are formed as in the electromechanical switch M3. An electrode structure in which the driving electrode 38 is disposed as a multi-stage slope is an extremely effective means. It is also useful to change the inclined structure of the side walls 33 and 33 of the recess 32 continuously instead of multiple stages as in this example, and the switching speed can be increased by the same principle. It is feasible.

  FIG. 10 is a schematic exploded sectional view showing an electromechanical switch M4 which is still another example of the embodiment of the electromechanical switch of the present invention. In FIG. 10, 51 is a substrate, 52 is a recess, 53 is a side wall of the recess 52, 54 is a bottom surface of the recess 52, 55 is a lower electrode, 57 is an upper movable electrode, 58 is a drive electrode, 59 is a recess forming material, 60 is a contact electrode, and 61 is a second substrate. These are the same as the electromechanical switches M1, M2, and M3 in the examples shown in FIGS. 1 to 3, 6 to 8, and 9.

  The electromechanical switch M4 is provided on the lower surface of the second substrate 61 in which the upper movable electrode 57 is spanned between the opposing side walls 53 and 53 of the recess 52, and the second substrate 61 is attached to the substrate 51, This is an example in which the side walls 53 and 53 of the recess 52 are stretched and joined. According to such an electromechanical switch M4, the upper movable electrode 57 is disposed on the lower surface of the second substrate 61, and the upper movable electrode 57 is recessed by bonding the second substrate 61 to the substrate 51. Since the substrate 51 and the second substrate 61 formed with the concave portion 52 are manufactured through different processes and finally combined with each other, an arbitrary shape is obtained. The upper movable electrode 57 having a shape can be stably produced and spanned between the side walls 53 of the recess 52, and the same effect as the electromechanical switch M2 of the present invention described above can be realized.

  In the electromechanical switch M4, the upper movable electrode 57 is spanned between the side walls 53 and 53 of the recess 52 by bonding the second substrate 61 to the substrate 51. In this case, it is not always necessary to use a sacrificial layer as will be described later. In addition, since the upper movable electrode 57 is disposed on the second substrate 61 in advance, the formation of the lower electrode 55 and the driving electrode 58 on the substrate 51 and the formation of the upper movable electrode 57 on the second substrate 61 are performed. Each can be performed in an independent process, for example, it is possible to simplify the manufacturing process, such as omitting the formation process and the processing process of the sacrificial layer, and the substrate 51 and the upper movable electrode 57 of an arbitrary shape can be independently processed. Since it can be manufactured, it can be manufactured stably, which is advantageous for improving the yield.

  In order to bridge and bond the second substrate 61 between the side walls 53 and 53 of the recess 52 of the substrate 51, for example, a material 59 for forming the recess and the second substrate 61 provided with the upper movable electrode 57 are provided. A technique such as so-called anodic bonding in which high voltage is applied between the electrodes and bonding may be used. In addition to this, bonding may be performed using, for example, solder or resin.

  Next, the electromechanical switch manufacturing method of the present invention will be described by taking the electromechanical switch M1 shown in FIGS. 1 to 3 as an example.

  In the manufacturing method of the electromechanical switch M1 according to the present invention, the concave portions 2 whose side walls 3 and 3 facing each other are inclined are formed on the substrate 1, and the lower electrode 5 is formed on the bottom surface 4 of the concave portion 2 and the side walls 3 and 3 are formed. The step of disposing the driving electrodes 8 and 8 on the substrate, the step of filling the recess 2 with a sacrificial layer (not shown), and spanning the sacrificial layer between the opposing side walls 3 and 3, A step of forming the upper movable electrode 7 so that the portion faces the lower electrode 5 and both sides of the central portion face the driving electrodes 8 and 8, respectively, and then a step of removing the sacrificial layer. It is characterized by this.

  An electromechanical switch M1 of the present invention as shown in FIG. 1 uses a silicon wafer as a substrate 1 and spins an insulating material of a recess forming material 9 for forming a recess 2 having a hollow structure on the substrate 1. Apply by the coating method.

  Next, in order to form the concave portion 2 in the concave portion forming material 9 formed on the substrate 1, a resist film is applied on the concave portion forming material 9 by a spin coat method or the like. Using a photomask in which a pattern is formed using a thin film, photolithography is performed according to the pattern to form an opening for forming the recess 2 in the resist film.

Next, the recess forming material 9 in the opening of the resist film is wet-etched using an organic solvent such as acetone, IPA, butanol, or a semiconductor manufacturing gas such as CF ₄ , C ₂ F ₆ , Ar, or O _2. The desired recesses 2 are formed in the recess forming material 9 by removing to a predetermined depth, usually up to the upper surface of the substrate 1, by dry etching using. Regarding the shape, size, etc. of the recess 2, if the MEM switch is configured, the maximum depth of the recess 2 is preferably 3 μm or less, which is the distance between the upper movable electrode 7 and the lower electrode 5, The length of the opening at the upper end is 710 μm or less, the width is 260 μm or less, the length of the bottom surface 4 is 150 μm or less and the width is 170 μm or less, and the side walls 3 and 3, the lower electrode 5 and the signal line 6 are formed in the recess 2. The dimensions are preferably such that the region is obtained.

In addition, in order to make the opposite side walls 3 and 3 of the concave portion 2 have a slope, if a slope is formed using a resist mask on the SiO ₂ film, a pattern is formed with a Cr thin film on the glass surface. A pattern opening is formed by photolithography after applying a resist using a photomask, and an etching rate difference between the sacrificial layer material and the resist during dry etching using CF ₄ and Ar may be used.

  As an example of the recess 2, the driving electrode 8 formed on the side walls 3 and 3 that are the slopes of the recess 2 can be formed with a uniform thickness toward the bottom surface 4 of the recess 2. In order to stably obtain an electrostatic force, the dimensions for obtaining the region for forming the driving electrode 8 on the side wall 3 are 220 μm in length, 170 μm in width, and the height of the side wall 3. And the distance between the side walls 3 and 3 may be 150 μm (the distance between the corners where the bottom surface 4 and the side walls 3 and 3 intersect).

  A metal thin film such as Cu, Ti, Cr, Au or the like is deposited on the side walls 3 and 3 and the bottom surface 4 of the recess 2 having the inclined side walls 3 and 3 formed as described above by an electron beam evaporation method and a resistance heating evaporation method. Then, vacuum film formation is performed using a sputtering method or the like, processed into a predetermined pattern, and the driving electrodes 8 and 8 and the lower electrode 3 are disposed, respectively.

  Next, a resist film is applied on the driving electrodes 8 and 8 and the lower electrode 3 by a spin coat method or the like, and photolithography is performed according to the pattern using a photomask in which a pattern is made of a Cr thin film on a glass surface, for example. In line, for example, an opening for the signal line 6 is formed so as to be positioned on the lower electrode 5. Next, the signal line 6 is formed by forming a metal thin film such as Cu, Ti, Cr, Au, or Pt by using a vacuum film forming method or the like. As an example of the signal line 6, in order to improve the adhesion to the substrate 1 and suppress the diffusion of Ti to the upper Au layer to reduce the loss during signal transmission, the Ti / A stacked structure of Pt / Au may be used, and the thickness of each layer may be 0.1 μm / 0.2 μm / 0.5 μm.

Next, in order to form the upper movable electrode 7 spanned between the opposing side walls 3 and 3 of the recess 2, the driving electrodes 8 and 8 are provided on the side walls 3 and 3, and the lower electrode 5 and the signal line are provided on the bottom surface 4. The recesses 2 each provided with 6 are formed by depositing SiO ₂ , polysilicon or the like as a sacrificial layer material by a sputtering method or the like, and filled with the sacrificial layer. This sacrificial layer is for obtaining a region for forming the upper movable electrode 7 spanned between the opposing side walls 3 and 3 of the concave portion 2. For example, the sacrificial layer is formed by CVD using TEOS (tetramethoxy germanium). The SiO ₂ film formed by chemical vapor deposition is filled so that the height of the upper surface is flush with the upper end portion of the recess 2 without generating a gap between the side walls 3.

Next, a resist film is applied by spin coating or the like onto the sacrificial layer and the concave portion forming material 9 in which the concave portion 2 is filled so that the upper surface is flush with the upper surface of the concave portion forming material 9. Photolithography is performed according to the pattern using a photomask having a pattern made of a Cr thin film on the surface, the central portion faces the lower electrode 5, and both sides of the central portion face the driving electrodes 8 and 8, respectively. As described above, the opening for the upper movable electrode 7 is formed across the opposing side walls 3. Next, a metal thin film such as Al, Cu, Ti, Cr, Au or the like is formed by using a vacuum film forming method or the like to form the upper movable electrode 7. As an example of the upper movable electrode 7, Au and Cr are used as materials and the SiO ₂ / Cr from the lower side is used as a material so as to be lightweight and capable of being driven by a small electrostatic force so that high-speed response is possible. It is a Cr / Au laminated structure, and the dimensions may be formed with a length of 710 μm, a width of 260 μm, and a thickness of each layer of 0.2 μm / 0.2 μm / 0.3 μm.

  Thereafter, the sacrificial layer is etched and removed from the recess 2, whereby the lower electrode 5 and the signal line 6 are arranged on the bottom surface of the recess 2, and the driving electrodes 8 and 8 are arranged on the inclined surfaces of the opposing side walls 3 and 3. A hollow structure recess 2 is formed in which the upper movable electrode 7 is stretched between the side walls 3 and 3 so as to be opposed thereto.

For etching removal of the sacrificial layer, for example, when SiO ₂ formed by CVD using TEOS is used as the sacrificial layer material, etching is performed using an organic solvent such as acetone, IPA, butanol or the like as an etchant. It is good to carry out and dry in a continuous process. At this time, in order to relieve the surface tension so that the upper movable electrode 7 is pulled to the bottom surface 4 side of the concave portion 2 and does not come into contact with the lower electrode 5 by the surface tension of the etching solution during drying, supercritical drying is performed. It is preferable to do so. In addition, a typical resist that can be formed by applying a sacrificial layer by applying the sacrificial layer material to the surface of the substrate 1 by spin coating and filling the recess 2 is used, for example, TSMR-8900 manufactured by Tokyo Ohka Kogyo Co., Ltd. In some cases, the sacrificial layer may be removed by dry etching using a semiconductor manufacturing gas such as CF ₄ , C ₂ F ₆ , Ar, or O ₂ to form the hollow recess 2.

  In the electromechanical switch M1 of the present invention manufactured as described above, a voltage is applied between the upper movable electrode 7 and the driving electrode 8, and the displacement amount of the upper movable electrode 7 is the distance between the lower electrode 5 and the movable electrode 7. As a result, the response time of the upper movable electrode 7 was about 8 μsec at a driving voltage of 10 V, and good response characteristics were obtained.

  FIG. 5 shows the drive voltage application time and the upper movable electrode 7 or the cantilever of the electromechanical switch M1 of the present invention produced as described above and the comparative example using the MEM switch MM1 shown in FIG. The evaluation result of the relationship with the displacement amount of the arm 115 is shown by a diagram. In FIG. 5, the horizontal axis represents the drive voltage application time (unit: sec), the vertical axis represents the displacement (unit: m), the solid characteristic curve represents the relationship with respect to the example, and the dotted characteristic curve represents the comparative example. Shows the relationship. The magnitude of the drive voltage was 10V.

  As can be seen from the results shown in FIG. 5, the time required to move the same amount of displacement as the distance between the electrodes in the comparative example was 10.8 μsec, whereas the same displacement as the distance between the electrodes in the example. The time required to move the quantity was 7.78 μsec, the switching time could be shortened by 3.02 μsec, and the response characteristics could be increased. This is used in the electrical signal transmission system of a communication system to reduce the attenuation of transmission signals when turned on and to suppress signal leakage when turned off, creating a signal transmission system with stable operation with a faster response. It is advantageous in that it can be done.

  Next, a method for manufacturing the electromechanical switch of the present invention will be described using the electromechanical switch M2 shown in FIGS. 6 to 8 as an example. The basic content of the manufacturing method is the same as that described with the electromechanical switch M1 as an example.

  The substrate 11 is usually made of silicon, but glass or the like may be used. In this example, silicon oxide is used for the recess forming material 19.

  For the formation of the recess 12, a method using polysilane is suitable, and the photosensitivity of polysilane may be used. Polysilane is a silicon-based polymer in which Si atoms are connected in a chain. For example, when irradiated with ultraviolet light, silicon bonds (-Si-Si- bonds) are photolyzed, and oxygen atoms are arranged between the silicon bonds. Silanol groups acting as siloxane bonding sites and acid sites are generated. Therefore, when polysilane is applied onto the substrate 11, the portion that becomes the recess 12 is irradiated with light, and this is immersed in an alkaline developer that is an alkaline solution, the portion where silanol groups and the like are generated by irradiation with light is an alkaline solution. It is dissolved and removed. That is, by selecting a portion to be irradiated with light, the portion irradiated with light can be selectively removed, and the recess 12 can be processed with an arbitrary planar shape. The selection of the light irradiation part can be performed by the opening pattern of the photomask using a general photolithography technique. At this time, since the photoresponsiveness of polysilane sufficiently responds to ultraviolet light from a general mercury lamp, a special light source or the like is not required.

  Thereafter, by heating at, for example, several 100 ° C. in an oxygen atmosphere, the portion of the polysilane that has not been irradiated with light becomes silicon oxide, and the recess 12 is formed. At this time, if the entire surface is exposed to ultraviolet light with sufficient strength before heating in an oxygen atmosphere, the portion of polysilane not previously irradiated with light reacts with the ultraviolet light to break the bond between silicon, Oxygen atoms can enter and good silicon oxide can be formed.

  Polysilane has a modifying group in the side chain, but the modifying group is propyl group, hexyl group, phenylmethyl group, trifluoropropyl group, nonafluorohexyl group, tolyl group, biphenyl group, phenyl group, cyclohexyl group. As long as the polysilane has photosensitivity and becomes silicon oxide by heating in an oxygen atmosphere, it can be appropriately selected within the range.

  Here, using the property that the reaction in the depth direction with respect to the irradiation amount of polysilane is linear, for example, the aperture ratio of the photomask that determines the light irradiation site and the irradiation amount so that the side wall 13 of the recess 12 is a slope is stepped. In addition, by controlling the amount of irradiation light by changing it continuously or continuously, it is possible to form the recesses 12 whose opposed side walls 13 and 13 are inclined surfaces very easily.

  In this case, the photomask openings corresponding to the opposing side walls 13 and 13 are formed as a collection of fine openings, and the opening ratio of these fine openings is the same as the shape of the openings. In order not to be reflected in the photosensitive polysilane, for example, a method in which the size of each fine opening is sufficiently small below the optical resolution at the time of light irradiation and arranged at high density, a photomask and the photosensitive polysilane If the aperture ratio is adjusted by adjusting the density of the photomask openings corresponding to the sidewalls 13 and 13 by reducing the optical resolution by separating the distance of several tens to several hundreds of micrometers, the photomask can be transmitted. It is possible to easily change the irradiation light amount per unit area in a stepwise or continuous manner.

  When the recess forming material 19 is formed by applying polysilane, it is appropriate that the thickness is about 10 μm and the depth of the recess 12 is about 2 to 3 μm. The inclination of the side wall 13 is suitably about 1/100 radians.

  After the concave portion 12 is formed by the concave portion forming material 19, the lower electrode 15 and the driving electrode 18 are formed by film formation by vapor deposition or sputtering, and photolithography. For these electrodes, a metal multilayer film such as Au / Cr or a metal single layer film such as Al is suitable, but other materials can be appropriately selected.

  Next, the sacrificial layer is filled in the recess 12. In order to fill the recess 12 with the sacrificial layer, for example, a photosensitive organic material is used as a material, and is applied to the upper surface of the substrate 11 on which the recess 12 is formed by spin coating or the like to fill the recess 12. The portion of the organic material may be removed by photolithography. As the photosensitive organic material used as the sacrificial layer, for example, a photoresist, a photosensitive polyimide, or the like is suitable, but other materials can be appropriately selected.

After filling the recess 12 with the sacrificial layer, about several hundreds of liters of Cr are formed thereon by vapor deposition or sputtering, and a part thereof is removed using photolithography and dry etching techniques to provide an opening. Using Cr as a mask, dry etching is performed again to form a shallow recess of several thousand inches from the opening to the sacrificial layer, and then Cr is removed. Next, an Au / Cr / SiO ₂ laminated film is formed in the shallow recess so that, for example, SiO ₂ is the uppermost layer, and the contact electrode 20 is formed.

After the contact electrode 20 is formed, between the side walls 13 and 13 of the recess 12 on this. The upper movable electrode 17 is formed by vapor deposition, sputtering, photolithography, or the like so as to be stretched. The upper movable electrode 17 is suitably a multilayer film such as an Au / Cr / SiO ₂ laminated film in which, for example, SiO ₂ is the lowermost layer, but other materials can be appropriately selected. Here, in order to prevent a short circuit between the two electrodes when a driving voltage is applied between the driving electrode 18 and the upper movable electrode 17, it is preferable to provide a SiO ₂ film on the lowermost layer of the upper movable electrode 17, If the uppermost layer of the drive electrode 18 is covered with an insulating film, the entire upper movable electrode 17 may be made of only a conductive material such as aluminum or nickel.

  Then, the sacrificial layer is removed by immersing in an organic solvent, and the electromechanical switch M2 of the present invention is completed by drying using a supercritical drying technique or the like.

  In addition, this invention is not limited to the example of the above embodiment, A various change may be added in the range which does not deviate from the summary of this invention. For example, as the insulating material or sacrificial layer material used as the substrate 1 or the recess forming material 9 in the present invention, or the electrode material of the lower electrode 5, the upper movable electrode 7, and the driving electrode 8, the exemplified materials may be used. Although it is desirable, it is not particularly limited thereto.

It is a top view which shows typically an example of embodiment of the electromechanical switch of this invention. FIG. 2 is a cross-sectional view taken along line A-A ′ when the electromechanical switch according to the embodiment of the present invention illustrated in FIG. 1 is off. FIG. 2 is a cross-sectional view taken along the line A-A ′ when the electromechanical switch according to the embodiment of the present invention illustrated in FIG. 1 is on. (A) And (b) is a principle figure of a parallel electrode model and a gradient electrode model, respectively. It is a diagram which shows the relationship between the drive voltage application time and displacement amount in this invention and the conventional electromechanical switch. It is a top view which shows typically the other example of embodiment of the electromechanical switch of this invention. FIG. 7 is a cross-sectional view taken along line A-A ′ in another example of the embodiment of the electromechanical switch of the present invention shown in FIG. 6. FIG. 7 is a cross-sectional view taken along the line A-A ′ when the electromechanical switch according to another embodiment of the present invention shown in FIG. 6 is turned on. It is sectional drawing at the time of OFF of the further another example of embodiment of the electromechanical switch of this invention. FIG. 6 is an exploded cross-sectional view of still another example of the electromechanical switch according to the present invention. It is sectional drawing which shows typically the example (at the time of OFF) of the conventional electromechanical switch.

Explanation of symbols

1, 11, 31, 51: Substrate 2, 12, 32, 52: Recess 3, 13, 33, 53: Side wall 4, 14, 34, 54: Bottom 5, 15, 35, 55: Lower electrode 6: Signal line 7, 17, 37, 57: Upper movable electrode 8, 18, 38, 58: Drive electrode 9, 19, 39, 59: Recess formation material
20, 40, 60: Contact electrode
61: Second substrate M1, M2, M3, M4: Electromechanical switch of the present invention

Claims (5)

Concave portions each having an inclined side wall on the substrate are formed, and a lower electrode is disposed on the bottom surface of the concave portion, and a central portion moves to the bottom surface side between the opposing side walls of the concave portion. An upper movable electrode that can be deformed is electrically connected to the lower electrode, and driving electrodes are disposed on the side walls of the recess so as to face both sides of the central portion of the upper movable electrode. An electromechanical switch characterized by having The upper movable electrode is disposed on a lower surface of a second substrate that is spanned between the opposing side walls of the concave portion, and the second substrate is spanned between the side walls of the concave portion on the substrate. The electromechanical switch according to claim 1, wherein the electromechanical switch is joined. A step of forming concave portions each having an inclined side wall on the substrate, a step of disposing a lower electrode on the bottom surface of the concave portion and a driving electrode on the side wall; and a step of filling the concave portion with a sacrificial layer Forming an upper movable electrode over the sacrificial layer between the opposing sidewalls so that a central portion faces the lower electrode and both sides of the central portion face the driving electrode, respectively. And thereafter, removing the sacrificial layer. A method of manufacturing an electromechanical switch. In the step of forming the recess, a polysilane is applied on the substrate, the portion to be the recess of the applied polysilane is irradiated with light, and the portion irradiated with light is immersed in an alkaline solution to remove oxygen, 4. The method of manufacturing an electromechanical switch according to claim 3, further comprising the step of forming the recess by using a portion of the polysilane that has been heated and applied in an atmosphere that has not been irradiated with light as silicon oxide. 5. The method of manufacturing an electromechanical switch according to claim 4, wherein, in the light irradiation, the light amount is changed stepwise or continuously with respect to a portion that becomes the side wall of the concave portion.

Images