WO2011148698A1

WO2011148698A1 - Mems switch

Info

Publication number: WO2011148698A1
Application number: PCT/JP2011/055743
Authority: WO
Inventors: 知徳藤井; 健太郎中村; 泰成入枝; 貴之高野
Original assignee: 太陽誘電株式会社
Priority date: 2010-05-28
Filing date: 2011-03-11
Publication date: 2011-12-01
Also published as: TW201203301A; JP2011249192A; JP5298072B2

Abstract

Disclosed is a MEMS switch that minimizes closed-switch stiction and allows stable signal transmission when the switch is in a closed state. A first fixed terminal (FT11) and second fixed terminal (FT12) in the disclosed MEMS switch (10-1) have a cross-sectional shape in which the ends of said fixed terminals approach the closest to a movable terminal (MT) and the parts of said fixed terminals other than said ends are farther from the movable terminal (MT) than said ends are. When the switch is closed, the movable terminal (MT) contacts the ends of the first fixed terminal (FT11) and the second fixed terminal (FT12), with each contact area constituting a line.

Description

MEMS switch

The present invention relates to a MEMS (Micro Electro Mechanical Systems) switch.

Semiconductor switches using diodes, FETs, etc., deteriorate in characteristics such as transmission loss increase and isolation decrease as the frequency increases. However, since the MEMS switch is unlikely to cause such deterioration of characteristics, it is attracting attention as a switching device suitable for use in a high-frequency circuit, for example, a circuit through which a high-frequency signal of several GHz to several hundred GHz flows.

Generally, a MEMS switch has a multilayer structure created by using a known thin film forming method. This MEMS switch has a first fixed terminal composed of the end of the first signal line, a second fixed terminal composed of the end of the second signal line, and a movable terminal composed of a conductive layer formed on the movable lever. Both signal lines are made conductive by bringing the movable terminals into contact with both fixed terminals, while the two signal lines are made non-conductive by releasing the contact.

First, a known example of a MEMS switch will be described with reference to FIG. In this description, for convenience of explanation, front, back, top, bottom, left, and right are referred to as top, bottom, front, back, left, and right, respectively, toward the paper surface of FIG. The orientations corresponding to these in FIGS. 1B and 1C are also referred to.

As shown in FIGS. 1 (A) and 1 (B), the first signal line 101 and the second signal line 102 have a rectangular shape when viewed from above, and the tips of the first signal line 101 and the second signal line 102 have a predetermined interval (no symbol). It is arranged to face each other in parallel. The end of the first signal line 101 is used as the first fixed terminal 101a, and the end of the second signal line 102 is used as the second fixed terminal 102a.

The movable lever (no symbol) is formed on the lever body 103 having a rectangular shape when viewed from above, an insulating layer 104 formed on the lever body 103, and formed on the insulating layer 104 so that the shape when viewed from above is rectangular. And a movable terminal 105 made of a conductive layer. The movable terminal 105 is arranged so that the upper surface thereof faces the lower surfaces of both the

fixed terminals

101a and 102a in parallel with a predetermined distance CL101.

As shown in FIG. 1C, a drive actuator (not shown) is operated to displace the movable terminal 105 of the movable lever upward, and the displacement causes the upper surface of the movable terminal 105 to be moved to the lower surfaces of both the

fixed terminals

101a and 102a. By making surface contact, both

signal lines

101 and 102 can be brought into conduction. On the other hand, as shown in FIG. 1B, the operation of the drive actuator is stopped to return the movable terminal 105 of the movable lever to the initial position, and the upper surface of the movable terminal 105 and the lower surfaces of both the

fixed terminals

101a and 102a By releasing the surface contact, both

signal lines

101 and 102 in the conductive state can be brought into a non-conductive state.

This MEMS switch is configured such that both the

signal lines

101 and 102 are brought into conduction by bringing the upper surface of the movable terminal 105 into surface contact with the lower surfaces of both the

fixed terminals

101a and 102a. Therefore, the van der Waals force and meniscus are provided. Due to force or the like, there is a possibility that the phenomenon that the movable terminal 105 sticks to both the

fixed terminals

101a and 102a in a conductive state, so-called stiction, may occur. When stiction occurs, it is difficult to release the contact between the upper surface of the movable terminal 105 and the lower surfaces of both the

fixed terminals

101a and 102a even if the operation of the drive actuator is stopped.

Next, another known example of the MEMS switch (see Patent Document 1) will be described with reference to FIG. In this description, for convenience of explanation, the front, back, top, bottom, left, and right are referred to as the top, bottom, front, back, left, and right, respectively, as viewed in FIG. 2A. The directions corresponding to these in FIG. 2C are also referred to.

As shown in FIGS. 2 (A) and 2 (B), the first signal line 201 and the second signal line 202 have a rectangular shape when viewed from above, and the tips of the first signal line 201 and the second signal line 202 have a predetermined interval (no symbol). It is arranged to face each other in parallel. The end of the first signal line 201 is used as the first fixed terminal 201a, and the end of the second signal line 202 is used as the second fixed terminal 202a. A plurality of

hemispherical protrusions

201b and 202b are provided on the lower surfaces of the

fixed terminals

201a and 202a so as to face the upper surface of the movable terminal 205 described later.

The movable lever (without reference numeral) is formed on the lever body 203 having a rectangular shape when viewed from above, an insulating layer 204 formed on the lever body 203, and on the insulating layer 204 so that the shape when viewed from above is rectangular. And a movable terminal 205 made of a conductive layer. The movable terminal 205 is arranged so that the upper surface thereof faces the lower ends of the plurality of

hemispherical protrusions

201b and 202b of the

fixed terminals

201a and 202a in parallel with a predetermined distance CL201.

As shown in FIG. 2C, a drive actuator (not shown) is actuated to displace the movable terminal 205 of the movable lever upward, and the displacement causes the upper surface of the movable terminal 205 to move to the plurality of

fixed terminals

201a and 202a. The

signal lines

201 and 202 can be brought into conduction by bringing the lower ends of the

hemispherical protrusions

201b and 202b into multipoint contact. On the other hand, as shown in FIG. 2 (B), the operation of the drive actuator is stopped and the movable terminal 205 of the movable lever is returned to the initial position, and the upper surface of the movable terminal 205 and a plurality of

fixed terminals

201a and 202a. By releasing the multipoint contact with the lower ends of the

hemispherical protrusions

201b and 202b, both

signal lines

201 and 202 in the conductive state can be brought into a non-conductive state.

This MEMS switch is configured to bring both

signal lines

201 and 202 into a conductive state by bringing the upper surface of the movable terminal 205 into multipoint contact with the lower ends of the plurality of

hemispherical protrusions

201b and 202b of both

fixed terminals

201a and 202a. Therefore, compared with the surface contact type MEMS switch of FIG. 1, the contact area between the movable terminal 205 and the

fixed terminals

201a and 202a in the conductive state can be reduced. Can be suppressed.

In order to realize the multipoint contact in the MEMS switch shown in FIG. 2, at least the heights of the plurality of hemispherical protrusions 201b on the first fixed terminal 201a side are equal to each other, and the second fixed terminal 202a side The heights of the plurality of hemispherical protrusions 202b need to be the same.

However, since the height of the

hemispherical protrusions

201b and 202b is on the order of μm or nm, and the

hemispherical protrusions

201b and 202b are formed using a known thin film forming method, The height of the plurality of hemispherical protrusions 201b on the first fixed terminal 201a side is likely to vary, and the height of the plurality of hemispherical protrusions 202b on the second fixed terminal 202a side is also likely to vary.

This variation in height causes variations in the number of contact points in the multi-point contact, so that the contact area between the movable terminal 205 and both

signal lines

201 and 202 in the conductive state is likely to vary, and the contact There is a high possibility that transmission loss will vary due to variations in area. That is, although the multipoint contact type MEMS switch shown in FIG. 2 can suppress the occurrence of the stiction, it is difficult to perform stable signal transmission in a conductive state.

JP 2007-012558 A

An object of the present invention includes providing a MEMS switch that can suppress the occurrence of stiction in a conductive state and can perform stable signal transmission in the conductive state.

In order to achieve the above object, an embodiment of the present invention is provided in a first fixed terminal composed of an end of a first signal line, a second fixed terminal composed of an end of a second signal line, and a movable lever. The first fixed terminal and the second fixed terminal, and a movable terminal facing the first fixed terminal and the second fixed terminal at an interval. The first signal is obtained by bringing the movable terminal into contact with the first fixed terminal and the second fixed terminal. A MEMS switch for bringing a line and a second signal line into a conductive state and releasing the contact to bring the first signal line and the second signal line into a non-conductive state, wherein the first fixed terminal and the second fixed terminal are respectively The end edge of the first terminal is closest to the movable terminal, and the portion excluding the end edge has a cross-sectional shape farther from the movable terminal than the end edge. Make line contact with the edge of the terminal.

According to this MEMS switch, since both signal lines are made conductive by bringing the movable terminal into line contact with the edges of both fixed terminals, compared to the conventional surface contact type MEMS switch, The contact area between the movable terminal and both fixed terminals can be reduced. That is, the reduction of the contact area can surely suppress the occurrence of stiction (a phenomenon in which the movable terminal sticks to both fixed terminals in the conductive state).

In addition, since the line contact can be reliably performed over the entire predetermined contact area, the contact area between the movable terminal and both fixed terminals in the conductive state varies as compared with the conventional multi-point contact type MEMS switch. It can be suppressed. That is, it is possible to suppress a variation in transmission loss by suppressing the variation in the contact area, and to perform stable signal transmission in a conductive state.

According to each embodiment of the present invention, it is possible to provide a MEMS switch that can suppress the occurrence of stiction in a conductive state and can perform stable signal transmission in the conductive state.

The above object and other objects, structural features, and operational effects of the present invention will be further clarified by the following description and attached drawings.

1A is a partial top view of a MEMS switch showing a known example, FIG. 1B is a cross-sectional view taken along line S101-S101 in FIG. 1A, and FIG. It is sectional drawing which shows the state which carried out. 2A is a partial top view of a MEMS switch showing another known example, FIG. 2B is a cross-sectional view taken along line S201-S201 in FIG. 2A, and FIG. It is sectional drawing which shows the state which conduct | electrically_connected. FIG. 3 is a top view of the MEMS switch showing the first embodiment of the present invention. 4A is a sectional view taken along line S11-S11 in FIG. 3, FIG. 4B is a sectional view taken along line S12-S12 in FIG. 3, and FIG. 4C is taken along line S13-S13 in FIG. It is sectional drawing which follows. 5 (A) and 5 (B) are diagrams for explaining the operation of the MEMS switch shown in FIG. 1, and FIG. 5 (C) is a diagram showing a contact area of both fixed terminals with respect to the movable terminal. FIG. 6 is a view showing a modification of the shape of both fixed terminals shown in FIG. 7A and 7B are diagrams for explaining an example of a method for obtaining both the fixed terminals shown in FIG. 4A and the both fixed terminals shown in FIG. FIG. 8A is a partial top view of a MEMS switch showing a second embodiment of the present invention, and FIG. 8B is a diagram showing a contact area of both fixed terminals with respect to a movable terminal. FIG. 9A is a partial top view of a MEMS switch showing a third embodiment of the present invention, and FIG. 9B is a diagram showing a contact area of both fixed terminals with respect to the movable terminal. FIG. 10A is a partial top view of a MEMS switch showing a fourth embodiment of the present invention, and FIG. 10B is a diagram showing a contact area of both fixed terminals with respect to a movable terminal. FIG. 11 is a top view of a MEMS switch showing a fifth embodiment of the present invention.

First Embodiment First, a MEMS switch 10-1 to which the present invention is applied will be described with reference to FIGS. In this description, for convenience of explanation, the front, back, top, bottom, left, and right are referred to as top, bottom, front, back, left, and right, respectively, and correspond to these in other drawings. The direction is also referred to. In FIG. 8A, FIG. 9A, FIG. 10A, FIG. 11 and the drawings corresponding thereto, the reference regarding the orientation follows the same rules as in FIG.

The MEMS switch 10-1 has a multi-layer structure created by using a known thin film forming method, and the front-rear dimension can be about 3.0 mm and the left-right dimension can be about 1.5 mm.

As shown in FIGS. 3 and 4 (A) to 4 (C), the base layer 11 is formed using Si or the like as a material so that the top view shape is rectangular. An insulating layer 12 made of SiO ₂ or the like is formed on the entire upper surface of the base layer 11. The base layer 11 and the insulating layer 12 are formed with through

holes

11a and 12a having a U-shape when viewed from above. Inside the through

holes

11a and 12a, there is formed a movable lever ML having a rectangular shape when viewed from above, which is composed of a part 11b of the base layer 11 and a part 12b of the insulating layer 12.

A plurality (three in the figure) of through-holes 11b1 and 12b1 are formed in front of the center of the movable lever ML in the front-rear direction. The rear part of the movable lever ML from the through holes 11b1 and 12b1 corresponds to the main body part MLa, the front part of the through holes 11b1 and 12b1 corresponds to the movable part MLb, and the part in which the through holes 11b1 and 12b1 are formed. Corresponds to the hinge part MLc.

The first electrode layer 13 is formed on the upper surface (the upper surface of the portion 12b of the insulating layer 12) of the movable lever ML so as to extend from the upper surface to the rear side. The first electrode layer 13 is formed so that its top view shape forms a rectangle, and has an overhang portion 13a whose top view shape forms a rectangle on the right side of the rear part. The first electrode layer 13 has, for example, a multilayer structure including a Ti layer having a thickness of 5 nm and a Pt layer having a thickness of 200 nm formed thereon. In addition, a piezoelectric layer 14 made of PZT or the like is formed on the upper surface of the first electrode layer 13 (excluding the overhanging portion 13 a) in the same top view shape as the first electrode layer 13. Further, the second electrode layer 15 is formed on the upper surface of the piezoelectric layer 14 in the same top view shape as the piezoelectric layer 14. The second electrode layer 15 has, for example, a multilayer structure including a Ti layer having a thickness of 5 nm and a Pt layer having a thickness of 200 nm formed thereon. The first electrode layer 13, the piezoelectric layer 14, and the second electrode layer 15 constitute a drive actuator (no symbol) for deforming the body portion MLa of the movable lever ML so as to warp.

A movable terminal MT made of a conductive layer is formed on the upper surface of the movable portion MLb of the movable lever ML (the upper surface of the part 12b of the insulating layer 12) so that the shape of the upper surface when viewed from above is rectangular. The movable terminal MT (conductive layer) has, for example, a multilayer structure including a Ti layer having a thickness of 5 nm and an Au layer having a thickness of 200 nm formed thereon.

A power input terminal 16 made of a conductive layer is formed on the upper surface of the rear portion of the second electrode layer 15 so that the shape of the upper surface view is rectangular. The power input terminal 16 (conductive layer) has, for example, a multilayer structure including a Ti layer having a thickness of 5 nm and an Au layer having a thickness of 200 nm formed thereon.

The ground terminal 17 made of a conductive layer is formed on the upper surface of the overhanging portion 13a of the first electrode layer 13 so that the shape of the top view is rectangular. The ground terminal 17 (conductive layer) has, for example, a multilayer structure composed of a Ti layer having a thickness of 5 nm and an Au layer having a thickness of 200 nm formed thereon.

On the upper surface of the front portion of the insulating layer 12, a first signal line SL11 having a predetermined width made of a conductive layer and a second signal line SL12 made of a conductive layer and having the same width as the first signal line SL11 are arranged on the left and right sides of the movable terminal MT. It is formed so as to be symmetrical with respect to the center in the direction. The first signal line SL11 (conductive layer) and the second signal line SL12 (conductive layer) are, for example, (1) a Ti layer having a thickness of 5 nm, an Au layer having a thickness of 200 nm formed thereon, and an upper layer thereof. A multi-layer structure (four-layer structure) composed of a 3 μm thick Au layer formed thereon and a 200 nm thick SiO ₂ layer formed thereon, and (2) a 5 nm thick Ti layer and formed thereon. And a multilayer structure (three-layer structure) composed of an Au layer having a thickness of 200 nm and an Au layer having a thickness of 3 μm formed thereon.

As can be seen from FIG. 4A, the first signal line SL11 has a portion joined to the insulating layer 12 and a non-joined portion continuous with the portion. An end portion (right end portion in the figure) of the non-joined portion has a cross-sectional shape inclined toward the movable terminal MT, and the end portion is used as the first fixed terminal FT11. Similarly, the second signal line SL12 has a portion joined to the insulating layer 12 and a non-joined portion continuous with the portion, and an end portion (left end portion in the figure) of the non-joined portion is connected to the movable terminal MT. The end portion is used as the second fixed terminal FT12. Further, the distal ends of both fixed terminals FT11 and FT12 face each other in parallel in the left-right direction with a predetermined interval (no symbol). The end edges of both fixed terminals FT11 and FT12 face the upper surface of the movable terminal MT in parallel with a predetermined distance CL11. In FIG. 4A, θ represents the tilt angle of both fixed terminals FT11 and FT12. The inclination angle θ is 1 to 45 degrees, preferably 3 to 5 degrees.

In the MEMS switch 10-1, when both the signal lines SL11 and SL12 are brought into conduction, the power input terminal 16 and the ground terminal 17 are connected to a variable DC power supply (not shown), and the power input terminal 16 is connected to the variable DC power supply. A drive voltage is applied to. By applying this driving voltage, the piezoelectric layer 14 of the driving actuator is contracted by the piezoelectric effect, and the main body MLa of the movable lever ML is warped up by the contraction. Along with the deformation, the movable terminal MT on the movable portion MLc is displaced upward, and the displacement causes the upper surface of the movable terminal MT to come into contact with the edges of both the fixed terminals FT11 and FT12 so as to slightly push the movable terminal MTc. The upper surface of the MT is in line contact with the edges of the fixed terminals FT11 and FT12 under a predetermined contact pressure (see FIGS. 5A and 5B).

The movable portion MLb warps together with the main end portion MLa, but since there is a hinge portion MLc which is more flexible than the movable portion MLb, the movable terminal MT contacts the edges of the fixed terminals FT11 and FT12. When it starts, the movable part MLb tilts by utilizing the flexibility of the hinge part MLc, and the tilting causes the upper surface of the movable terminal MT to be in line contact with the edges of the fixed terminals FT11 and FT12. Further, since the upper surface of the movable terminal MT is in contact so as to slightly push up the edges of the fixed terminals FT11 and FT12, even if the edges of the fixed terminals FT11 and FT12 are displaced in the vertical direction, this push-up action is performed. This offsets the displacement, and the upper surface of the movable terminal MT is surely in line contact with the end edges of the fixed terminals FT11 and FT12. That is, the line contact between the movable terminal MT and the fixed terminals FT11 and FT12 is, as shown in FIG. 5C, a predetermined contact area CR11 and CR12 according to the width of the edge of the fixed terminals FT11 and FT12. It is reliably performed over substantially the entire area.

In the contact process, the upper surface of the movable terminal MT comes into contact with the edges of the fixed terminals FT11 and FT12 slightly pushed up. If the fixed terminals FT11 and FT12 have appropriate rigidity and the inclination angle θ is appropriately set based on the rigidity and the pushing force, the line contact is shown in FIG. This can be ensured over substantially the entire contact area CR11 and CR12. In other words, even if the upper surface of the movable terminal MT contacts with the upper edges of the fixed terminals FT11 and FT12 pushed up, the line contact can be ensured without any problem, and the line contact does not become a surface contact.

On the other hand, when the signal lines SL11 and SL12 in the conductive state are brought into the non-conductive state, the application of the drive voltage to the power input terminal 16 may be canceled. By releasing the application of the drive voltage, the operation of the drive actuator is stopped, the movable lever ML is restored, the movable terminal MT is returned to the initial position, and the upper surface of the movable terminal MT and the edges of the fixed terminals FT11 and FT12 are connected. The line contact is released (see FIG. 4A).

In FIG. 4A, the two fixed terminals FT11 and FT12 have cross-sectional shapes inclined toward the movable terminal MT. However, as shown in FIG. 6, the fixed terminals FT11 ′ and FT12 ′ are movable terminals. Even when a cross-sectional shape curved toward the MT is employed, the line contact and the release thereof can be similarly performed. The cross-sectional shapes of the two fixed terminals are not limited to those shown in FIGS. 4A and 6. In short, the edges of the two fixed terminals are closest to the movable terminal MT and the edges are excluded. Even if a portion whose cross-sectional shape is further away from the movable terminal MT than the end edge is used as both the fixed terminals, the line contact and the release can be similarly performed.

Both signal lines SL11 and SL12 including both fixed terminals FT11 and FT12 shown in FIG. 4A and both signal lines SL11 and SL12 including both fixed terminals FT11 ′ and FT12 ′ shown in FIG. Since it is extremely thin, it is practically difficult to bend and form the cross-sectional shape shown in FIG. 4A or the cross-sectional shape shown in FIG. .

Therefore, in order to make both the fixed terminals (FT11 and FT12, FT11 ′ and FT12 ′) have the cross-sectional shape shown in FIG. 4A and the cross-sectional shape shown in FIG. 6, both signal lines SL11 and SL12 are formed. In the process, both fixed terminals (FT11 and FT12, FT11 ′ and FT12 ′) need to be inclined or curved. Below, the suitable example of a method clarified through various experiments is demonstrated.

(First method example)
In the first method example, in the process of forming both signal lines SL11 and SL12 including both fixed terminals (FT11 and FT12, FT11 ′ and FT12 ′), internal stress (compressive stress or The basic idea is to generate a tensile stress or the like, and to incline or curve the portion by the internal stress.

Here, both signal lines SL11 and SL12 are “formed on the 5 nm thick Ti layer, the 200 nm thick Au layer formed thereon, and the 3 μm thick Au layer formed thereon. A case of having a “multilayer structure (four-layer structure) composed of a SiO ₂ layer having a thickness of 200 nm” will be described as an example, and a first method example will be specifically described with reference to FIG. 7A corresponds to a Ti layer having a thickness of 5 nm, LA12 corresponds to an Au layer having a thickness of 200 nm, LA13 corresponds to an Au layer having a thickness of 3 μm, and LA14 corresponds to an SiO ₂ layer having a thickness of 200 nm.

In this case, both signal lines SL11 and SL12 are formed by forming a Ti layer LA11 by DC sputtering on a patterned resist (not shown) and forming an Au layer LA12 by DC sputtering on the Ti layer LA11. A step of forming a plating seed layer (no symbol) comprising the layer LA11 and the Au layer L12, a step of forming an Au layer LA13 by electrolytic plating on the plating seed layer, and a SiO ₂ layer by plasma CVD or RF sputtering on the Au layer LA12 It is created through the step of forming LA14 and the step of removing the resist used in forming the plating seed layer. In the step of forming the SiO ₂ layer LA14, the substrate temperature is set to 300 ° C., which is higher than the normal temperature, and the temperature is lowered to 20 ° C. after the formation of the SiO ₂ layer LA14.

Since the linear expansion coefficient of the SiO ₂ layer LA14 is smaller than the linear expansion coefficient of the Au layer LA13, the lower Au layer LA13 contracts more greatly than the upper SiO ₂ layer LA14 during the temperature drop process (FIG. 7A). ) See the arrow in the middle). The SiO ₂ layer LA14 and the Au layer LA13 are inclined or curved downward due to the compressive stress based on the difference in shrinkage, and accordingly, the thin plating seed layer (Ti layer LA11 + Au layer LA12) is also inclined or curved downward. To do. Thereby, both fixed terminals (FT11 and FT12, FT11 ′ and FT12 ′) have the cross-sectional shape shown in FIG. 4A and the cross-sectional shape shown in FIG.

According to experiments, when the specific method was carried out with the lengths of both fixed terminals (FT11 and FT12, FT11 ′ and FT12 ′) being 90 μm, the edges of the fixed terminals were approximately 1. It could be lowered by 2 μm. In addition, even when the ambient temperature in which both the fixed terminals were created was changed within normal temperature (20 ± 15 ° C.), no substantial change was observed in the numerical value of 1.2 μm.

(Second method example)
In the second method example, in the process of forming both signal lines SL11 and SL12 including both fixed terminals (FT11 and FT12, FT11 ′ and FT12 ′), at least a portion corresponding to both fixed terminals is caused to undergo plastic deformation. The basic idea is to incline or curve the same part by plastic deformation.

Here, both signal lines SL11 and SL12 are “a multilayer structure (3 nm comprising a Ti layer having a thickness of 5 nm, an Au layer having a thickness of 200 nm formed thereon, and an Au layer having a thickness of 3 μm formed thereon. A second method example will be specifically described with reference to FIG. 7B corresponds to a Ti layer having a thickness of 5 nm, LA 22 corresponds to an Au layer having a thickness of 200 nm, and LA 23 corresponds to an Au layer having a thickness of 3 μm.

In this case, both signal lines SL11 and SL12 form a Ti layer LA21 by DC sputtering on the patterned resist RE, form an Au layer LA22 by DC sputtering on the Ti layer LA21, and form Ti layers LA21 and SL21. A step of forming a plating seed layer (no symbol) made of the Au layer L22, a step of forming an Au layer LA23 on the plating seed layer by electrolytic plating, and a step of removing the resist RE used for forming the plating seed layer , Created through. As the resist RE used for forming the plating seed layer, a resist that can be heated and shrunk, for example, a novolac positive resist is used, and as a step before removing the resist RE, the resist RE is heated at 180 ° C. for 1 hour. A step of heating and holding is performed.

Since the resist RE can be heated and shrunk, the resist RE shrinks during the heating process (see the arrow in FIG. 7B). With this contraction, the plating seed layer (Ti layer LA21 + Au layer LA22) and Au layer LA23 thereon are plastically deformed downward and inclined or curved downward. Thereby, both fixed terminals (FT11 and FT12, FT11 'and FT12') have the cross-sectional shape shown in FIG. 4A and the cross-sectional shape shown in FIG.

According to experiments, when the above-mentioned specific method was carried out with the length of both fixed terminals (FT11 and FT12, FT11 ′ and FT12 ′) being 40 μm, the edges of the two fixed terminals were approximately 2. It could be lowered by 0 μm.

In addition, the specific method is described as follows: “Ti layer having a thickness of 5 nm, Au layer having a thickness of 200 nm formed thereon, Au layer having a thickness of 3 μm formed thereon, and a thickness of 200 nm formed thereon. Even when applied to the production of both signal lines SL11 and SL12 having a “multilayer structure (four-layer structure) made of SiO ₂ layers”, both fixed terminals (FT11 and FT12, FT11 ′ and FT12 ′) are used in the same manner as described above. The cross-sectional shape shown in FIG. 4A and the cross-sectional shape shown in FIG. 6 could be obtained.

According to the MEMS switch 10-1 described above, the following effects (1) and (2) can be obtained.

(1) Since the signal lines SL11 and SL12 are brought into conduction by bringing the upper surface of the movable terminal MT into line contact with the end edges of the fixed terminals FT11 and FT12, the surface contact type MEMS shown in FIG. Compared with the switch, the contact area between the movable terminal MT and the fixed terminals FT11 and FT12 in the conductive state can be reduced. That is, the reduction of the contact area can surely suppress the occurrence of stiction (a phenomenon in which the movable terminal MT sticks to both the fixed terminals FT11 and FT12 in the conductive state).

(2) Since the line contact can be surely performed over substantially all the predetermined contact areas CR11 and CR12, the movable terminal MT in the conductive state as compared with the multipoint contact type MEMS switch shown in FIG. And variation in the area of the contact area between the fixed terminals FT11 and FT12 can be suppressed. And it can suppress that a transmission loss varies by suppression of variation in the area of a contact area, and can perform stable signal transmission in a conduction state.

Second Embodiment Next, a MEMS switch 10-2 to which the present invention is applied will be described with reference to FIG.

The MEMS switch 10-2 is different from the MEMS switch 10-1 described in the first embodiment, as shown in FIG. 8A, both fixed terminals rather than the width W11 of both signal lines SL21 and SL22. This is because the width W12 of the tip (edge) of FT21 and FT22 is reduced.

In this case, the line contact between the movable terminal MT and both the fixed terminals FT21 and FT22 is, as shown in FIG. 8B, a predetermined contact area CR21 corresponding to the width W12 of the edges of the both fixed terminals FT21 and FT21. And substantially over the entire region of CR22.

According to the MEMS switch 10-2, in addition to the effects (1) and (2) described in the first embodiment, the following effect (3) can be obtained.

(3) The contact area of the fixed terminals FT21 and FT22 with respect to the movable terminal MT can be easily adjusted by changing the width W12 of the edges of the fixed terminals FT21 and FT22.

Third Embodiment Next, a MEMS switch 10-3 to which the present invention is applied will be described with reference to FIG.

The MEMS switch 10-3 is different from the MEMS switch 10-1 described in the first embodiment, as shown in FIG. 9A, the top view shape is substantially the tip of the first fixed terminal FT31. Two convex portions FT31a forming a semicircular shape are integrally formed, and two convex portions 32a having a substantially semicircular shape when viewed from the top are symmetrical to the convex portion FT31a at the tip of the second fixed terminal FT31. In this way, it is integrally formed. In short, the same number of convex portions FT31a and 32a are integrally formed at the tips of both fixed terminals FT31 and FT32.

In this case, the line contact between the movable terminal MT and the fixed terminals FT31 and FT32 is, as shown in FIG. 9B, a predetermined contact area corresponding to the shape (arc) of the edge of each convex portion FT31a and FT32a. This is ensured over substantially the entire region of CR31 and CR32.

According to the MEMS switch 10-3, in addition to the effects (1) and (2) described in the first embodiment, the following effect (4) can be obtained.

(4) The contact area of the fixed terminals FT31 and FT32 with respect to the movable terminal MT can be easily adjusted by changing the size and number of the convex portions FT31a and FT32a. This effect can be similarly obtained even when one convex portion FT31a and FT32a are integrally formed at the tips of both fixed terminals FT31 and FT32, or when three or more convex portions FT31a and FT32a are integrally formed. Even when the top-view shape of each of the convex portions FT31a and FT32a is a U-shape or a rectangle, the same can be obtained.

Fourth Embodiment Next, a MEMS switch 10-4 to which the present invention is applied will be described with reference to FIG.

The MEMS switch 10-4 is different from the MEMS switch 10-1 described in the first embodiment, as shown in FIG. 10A, the top view shape is substantially the tip of the first fixed terminal FT41. Two convex portions FT41a having a semicircular shape are integrally formed, and one convex portion 42a having a substantially semicircular shape when viewed from above is integrally formed at the tip of the second fixed terminal FT41. In short, the number of convex portions FT41a integrally formed at the tips of both fixed terminals FT41 is different from the number of convex portions FT42a integrally formed at the tips of FT42.

In this case, the line contact between the movable terminal MT and the fixed terminals FT41 and FT42 is a predetermined contact corresponding to the shape (arc) of the edge of each convex portion FT41a and FT42a, as shown in FIG. This is ensured over substantially the entire region CR41 and CR42.

According to the MEMS switch 10-4, in addition to the effects (1) and (2) described in the first embodiment, the following effects (5) and (6) can be obtained.

(5) The contact area of the fixed terminals FT41 and FT42 with respect to the movable terminal MT can be easily adjusted by changing the size and number of the convex portions FT41a and FT42a. This effect is obtained when three or more convex portions FT41a are integrally formed at the tip of the first fixed terminal FT41 and two or more convex portions FT42a are integrally formed at the tip of the second fixed terminal FT42. The same can be obtained when the top-view shapes of the parts FT41a and FT42a are U-shaped or rectangular.

(6) As shown in FIG. 10 (A), by arranging the convex portions FT41a and FT42a so that the other convex portion FT42a faces between the two convex portions FT41a, both the fixed terminals FT41 and It is possible to reduce the interval between the tips of the FT42, thereby shortening the signal transmission distance between the two fixed terminals FT41 and FT42 and reducing transmission loss.

Fifth Embodiment Next, a MEMS switch 10-5 to which the present invention is applied will be described with reference to FIG.

This MEMS switch 10-5 differs from the MEMS switch 10-1 described in the first embodiment in that the longitudinal dimensions of the base layer 11 ′ and the insulating layer 12 ′ are increased as shown in FIG. The movable lever ML composed of a part 11b ′ of the base layer 11 and a part 12b ′ of the insulating layer 12 is lengthened. The movable lever ML ′ includes two main body parts MLa and one movable part MLb. The first electrode layer 13, the piezoelectric layer 14, and the first electrode layer 13 constituting the drive actuator are also formed on the upper surface of the main body MLa on the front side (ie, the point changed from the cantilever structure to the double-sided structure). The second electrode layer 15 is formed, and the power input terminal 16 and the ground terminal 17 corresponding to the drive actuator provided on the upper surface of the front body part MLa are added.

In the MEMS switch 10-5, when both signal lines SL11 and SL12 are brought into conduction, two power input terminals 16 and two ground terminals 17 are connected to a variable DC power supply (not shown), and the variable DC power supply is connected. A drive voltage having the same value is applied to the two power input terminals 16. By applying this drive voltage, the piezoelectric layers 14 of the two drive actuators on the front side and the rear side are contracted due to the piezoelectric effect, and the two main body portions MLa of the movable lever ML ′ are warped up by the contraction. With these deformations, the movable terminal MT on the movable portion MLc is displaced upward, and the displacement causes the upper surface of the movable terminal MT to come into contact with the edges of the fixed terminals FT11 and FT12 so as to slightly lift the movable terminal MTc. The upper surface of the terminal MT is in line contact with the edges of the fixed terminals FT11 and FT12 under a predetermined contact pressure (see FIGS. 5A and 5B).

On the other hand, when both the signal lines SL11 and SL12 in the conductive state are brought into the non-conductive state, the application of the drive voltage to the two power input terminals 16 may be canceled. By releasing the application of the drive voltage, the operation of the two drive actuators is stopped, the movable lever ML ′ is restored, the movable terminal MT is returned to the initial position, and the upper surface of the movable terminal MT and the ends of the fixed terminals FT11 and FT12 are restored. The line contact with the edge is released (see FIG. 4A).

According to the MEMS switch 10-5, the effects (1) and (2) described in the first embodiment can be obtained although the configuration of the movable lever ML 'and the like is different. Further, the features described in the second embodiment, the third embodiment, and the fourth embodiment can be appropriately adopted for the MEMS switch 10-5, and the effect (3) described in the second embodiment by the adoption. ), The effect (4) described in the third embodiment, and the effects (5) and (6) described in the fourth embodiment can be appropriately obtained.

10-1, 10-2, 10-3, 10-4, 10-5 ... MEMS switch, SL11, SL21, SL31, SL41 ... first signal line, FT11, FT11 ', FT21, FT31, FT41 ... first fixed Terminal, FT31a, FT41a ... convex portion, SL12, SL22, SL32, SL42 ... second signal line, FT12, FT12 ', FT22, FT32, FT42 ... second fixed terminal, FT32a, FT42a ... convex portion, CR11, CR12, CR21 , CR22, CR31, CR32, CR41, CR42 ... contact area, ML, ML '... movable lever, MT ... movable terminal.

Claims

The first fixed terminal including the end portion of the first signal line, the second fixed terminal including the end portion of the second signal line, the conductive layer provided on the movable lever, and including the first fixed terminal and the second fixed terminal. A movable terminal facing the fixed terminal at an interval, and bringing the first signal line and the second signal line into a conductive state by bringing the movable terminal into contact with the first fixed terminal and the second fixed terminal; A MEMS switch that uncouples the first signal line and the second signal line by unwinding,
Each of the first fixed terminal and the second fixed terminal has a cross-sectional shape in which each edge is closest to the movable terminal, and a portion excluding the edge is further away from the movable terminal than in the conductive state. The movable terminal makes line contact with the edges of the first fixed terminal and the second fixed terminal.
The MEMS switch according to claim 1, wherein
The first fixed terminal and the second fixed terminal have a cross-sectional shape inclined toward the movable terminal.
The MEMS switch according to claim 1, wherein
The first fixed terminal and the second fixed terminal have a cross-sectional shape curved toward the movable terminal.
The MEMS switch according to any one of claims 1 to 3,
The widths of the edges of the first fixed terminal and the second fixed terminal are smaller than the widths of the first signal line and the second signal line.
The MEMS switch according to any one of claims 1 to 3,
One or more convex portions are integrally formed at the tips of the first fixed terminal and the second fixed terminal, and the movable terminal is connected to the edges of the convex portions of the first fixed terminal and the second fixed terminal in the conductive state. Contact.
The MEMS switch according to claim 5, wherein
The number of convex portions integrally formed at the tip of the first fixed terminal is the same as the number of convex portions integrally formed at the tip of the second fixed terminal.
The MEMS switch according to claim 5, wherein
The number of convex portions integrally formed at the tip of the first fixed terminal is different from the number of convex portions integrally formed at the tip of the second fixed terminal.