JP2013143231A - High-frequency microswitch - Google Patents

Abstract

PROBLEM TO BE SOLVED: To increase temperature fluctuation resistance of a high-frequency microswitch.SOLUTION: A first movable beam portion is formed in a silicon layer adhered partially to a silicon substrate through an insulating layer, and has a first signal wiring and a first electrical drive portion which are arranged in the longitudinal direction. A second movable beam portion is formed in the silicon layer, and has a second signal wiring and the first electric drive portion which are arranged in the longitudinal direction. The first movable beam portion and the second movable beam portion are disposed side by side at positions symmetrical to each other with either an axial symmetry or a rotational symmetry.

Description

  The present invention relates to a high-frequency microswitch, and relates to an improvement in temperature fluctuation resistance of a high-frequency microswitch such as an RF-MEMS (MicroElectro Mechanical Systems) switch.

  In recent years, in various technical fields, devices having a micro structure formed by a micromachining technique have been applied. Among them, RF switch devices that are expected to be used for RF (high frequency) devices due to their good linearity have been designed and prototyped.

  In such an RF switch device, a Si wafer or the like is used as a substrate for manufacturing the device, and a metal material such as Au or Cu is often used as the wiring in order to pass an RF signal with low loss.

JP 2000-188049 A

  In such an RF switch device, since different materials are used, there is a possibility that the shape of the movable part is deformed when a temperature change of the device occurs due to a difference in thermal expansion coefficient. In particular, when different materials exist in the movable region of the RF switch device, the contact gap amount may vary.

  In an extreme case, there is a possibility that the contact portion is turned on even when the drive voltage is not applied, or may not be turned on even when the drive voltage is applied. To explain.

FIG. 22 is a schematic plan view of a prototype RF-MEMS switch, and FIG. 23 is a schematic cross-sectional view of the prototype RF-MEMS switch. As shown in the figure, a device region separated by a slit 54 is formed on a bonded substrate 50 composed of a silicon substrate 51 / SiO ₂ film 52 / silicon layer 53, and a portion from the center of the device region is defined as a movable beam portion 60. To do. The movable beam portion 60 is formed of a silicon layer 53 that is lifted from the silicon substrate 51, and a piezoelectric driving portion 61 and a signal wiring 66 are disposed along the longitudinal direction.

The piezoelectric driver 61 is a laminated structure of a lower electrode 62 / PZT63 / upper electrode 64, lower electrode 62 is connected to a driving-electrode pad 65 _1, the upper electrode 64 is connected to the driving electrode pads 65 _2. The lower electrode 62 is made of Pt, and the upper electrode 64 is made of Pt or Au. The drive electrode pads 65 ₁ and 65 ₂ are formed by Au plating.

One end of the signal line 66 is connected to the signal electrode pads 67 _2, the other end of the signal line 66 is opposed to maintaining the connection wiring 68 and the contact gap with the contact portion 70. The connection wiring 68 is formed by Au electroless plating integrally with the signal electrode pad 67 ₁ and the support portion 69. Note that the signal line 66 is formed by Au sputtering film, the signal electrode pad 67 ₂ is formed by Au plating film.

FIG. 24 is an explanatory diagram of dimensions of a prototype RF-MEMS switch. Here, an example of the dimensions will be described. The size of the piezoelectric drive unit 61 is 320 μm × 160 μm, and the width of the movable beam unit 60 where the piezoelectric drive unit 61 is provided is 170 μm. The size of the drive electrode pads 65 ₁ and 65 ₂ connected to the piezoelectric drive unit 61 is 150 μm × 150 μm. On the other hand, the width of the signal wiring 66 is 30 μm, and the size of the movable beam portion 60 where the signal wiring 66 is provided is 400 μm × 35 μm.

  FIG. 25 is an explanatory diagram of the switching operation of the prototype RF-MEMS switch. As shown in FIG. 25A, the signal wiring 66 and the contact portion 70 have, for example, a gap of 0.3 μm in the voltage OFF state. It is in a state of being opposed to each other.

As shown in FIG. 25 (b), when a voltage of 5V to 20V is applied to the voltage driving unit 61, the movable beam unit 60 is bent and lifted, and the signal wiring 66 and the contact unit 70 come into contact with each other to be in an ON state. The life of such an RF-MEMS switch is about 10 ⁸ to ¹⁰ 10 times as the number of switching times.

  However, when a temperature change occurs, as shown in FIG. 25C, the shape of the movable beam portion 60 is deformed, causing a change in the contact gap amount, and the switch is not applied with the drive voltage. There is a problem of being turned on. This is an extreme example, and in reality, the ON state is reached at a voltage lower than the normal ON voltage. In this case, as a phenomenon that actually becomes a problem, it is conceivable that the return force as the spring of the movable beam portion when the applied voltage is turned off becomes small, and the contact is difficult to separate.

  Therefore, an object of the present invention is to improve the temperature fluctuation tolerance of a high-frequency microswitch.

  From one aspect to be disclosed, a silicon substrate, a silicon layer partially adhered to the silicon substrate via an insulating layer, a first signal wiring and a first layer formed on the silicon layer are provided. A first movable beam portion arranged in the longitudinal direction of the first movable beam portion, and the silicon layer provided in a symmetrical position of either line symmetry or rotational symmetry with respect to the first movable beam portion. And a second movable beam portion in which the second signal wiring and the first electric driving portion are arranged in the longitudinal direction, and the first signal wiring on the first electric driving portion side of the first signal wiring. At least a wiring connection portion provided with a contact portion provided at an end, wherein one of the first movable beam portion and the second movable beam portion is a driving portion in a switching operation, The other of the movable beam portion and the second movable beam portion is a switch. RF microswitch is provided, which is a non-driving unit in grayed operation.

  According to the disclosed high-frequency microswitch, it is possible to improve temperature fluctuation tolerance.

1 is a configuration explanatory diagram of a high-frequency microswitch according to an embodiment of the present invention. 1 is a schematic plan view of an RF-MEMS switch according to Embodiment 1 of the present invention. It is AA 'sectional drawing of RF-MEMS switch of Example 1 of this invention. It is explanatory drawing of the switching operation of RF-MEMS switch of Example 1 of this invention. It is a schematic top view of RF-MEMS switch of Example 2 of this invention. It is a schematic top view of RF-MEMS switch of Example 3 of this invention. It is explanatory drawing of the manufacturing process to the middle of the RF-MEMS switch of Example 3 of this invention. It is explanatory drawing of the manufacturing process until the middle of FIG. 7 or subsequent of the RF-MEMS switch of Example 3 of this invention. It is explanatory drawing of the manufacturing process after FIG. 8 of RF-MEMS switch of Example 3 of this invention. It is explanatory drawing of the switching operation of RF-MEMS switch of Example 3 of this invention. It is a schematic top view of RF-MEMS switch of Example 4 of this invention. It is explanatory drawing of the switching operation of RF-MEMS switch of Example 4 of this invention. It is a schematic plan view of the RF-MEMS switch of Example 5 of the present invention. It is explanatory drawing of switching operation of RF-MEMS switch of Example 5 of this invention. It is a schematic top view of RF-MEMS switch of Example 6 of this invention. It is a schematic sectional drawing of RF-MEMS switch of Example 6 of this invention. It is explanatory drawing of the manufacturing process to the middle of the RF-MEMS switch of Example 6 of this invention. It is explanatory drawing of the manufacturing process after FIG. 17 of RF-MEMS switch of Example 6 of this invention. It is a schematic top view of RF-MEMS switch of Example 7 of this invention. It is a schematic sectional drawing of RF-MEMS switch of Example 7 of this invention. It is a schematic plan view of the RF-MEMS switch of Example 8 of the present invention. It is a schematic plan view of a prototype RF-MEMS switch. It is a schematic sectional drawing of a prototype RF-MEMS switch. It is explanatory drawing of the dimension of the prototype RF-MEMS switch. It is explanatory drawing of the switching operation | movement of the prototype RF-MEMS switch.

  Here, with reference to FIG. 1, the high frequency microswitch of embodiment of this invention is demonstrated. FIG. 1 is a configuration explanatory diagram of a high-frequency microswitch according to an embodiment of the present invention, and is shown here as a plan view, but the basic cross-sectional structure is the same as the structure shown in FIG.

As shown in the figure, two movable beam portions 2 ₁ and 2 _{2 in} which signal wirings 4 ₁ and 4 ₂ and electric drive portions 3 ₁ and 3 ₂ are arranged in the longitudinal direction are symmetrically arranged, and one signal the end portion of the wiring 4 ₁ is provided with a wiring connection portion 5 having a contact portion 6. That is, conventionally regard to contact fixed to have side, is obtained by a movable beam part 2 ₁ of the same structure as the contact of the movable side. Here, an example is shown in which the pair of movable beam portions 2 ₁ and 2 ₂ are arranged line-symmetrically, but they may be arranged rotationally symmetrically. Note that the symmetry does not require strict symmetry in a mathematical sense.

Electrical drive unit ₃ 1, _{3 2} are connected to a driving electrode pad _{9 11,} 9 21, _{9 3,} respectively. Further, the signal wirings 4 _1, 4 ₂ One end is connected to the signal electrode pads _81, 82, the other end of the signal wiring 4 ₁ is provided with a wire connection portion 5, the wire connecting portion 5 contact portion 6 provided on the end faces the signal lines 4 _2.

An ON state in contact when voltage is applied to the signal wiring 4 ₂ in electrical drive unit 3 _second side provided with the contact portion 6 and the signal line 4 ₂ raised is bent movable beam section 2 _2. When stopping the application of the driving voltage in order to switch OFF, the OFF state contact section 6 has deviated force of the spring with the movable beam section 2 ₂ acts away the contact.

Incidentally, the movable beam section 2 ₁ as a non-driving unit when the switching operation is preferably formed to have the same dimensions as the movable beam section 2 ₂ serving as a driving unit at the time of switching operation. In addition, it is desirable that the metal lengths of the signal wirings 4 ₁ and 4 ₂ , which are most likely to cause an influence at the time of temperature change due to the difference in expansion coefficient between the base silicon layer and the silicon layer, be made uniform.

With such a configuration, the movable beam section 2 ₁ but not to actively operate during normal ON / OFF operation, the movable beam section 2 for driving when there is a temperature change of the device during ON / OFF operation _In synchronism with the position variation of ₂ , the same amount of position variation occurs in the same direction. Thereby, since the gap between the contact portion 6 and the signal line 4 ₂ is to be kept constant irrespective of the temperature variation, or the ON state to not apply a driving voltage or a driving voltage The problem of not turning on even when applied is eliminated.

It is not necessary to provide a driving electrode pad 9 ₁₁ The electrical drive unit 3 ₁ provided in the movable beam section 2 ₁ as a non-driving portion during a switching operation. However, the driving electrode pads 9 ₁₁ by leaving to allow application of a voltage from the outside, can be expected new effects which deviate the OFF sometimes actively contacts.

That is, normally, when the contact portion is in an ON state, an adhesive force due to van der Waals force or the like is generated. Spring force with the Stopping movable beam section 2 ₂ application of the driving voltage in order to turn OFF the contact, the contact portion deviates acts in a direction in which the contacts are separated. However, by repeating ON / OFF of the contact portion, the contact area changes due to deformation of a minute area of the contact portion, and the contact force does not turn off beyond the strength of the spring to be separated from the adhesive force. A phenomenon called sticking may occur.

At this time, by applying a voltage in a direction contacts electrical drive unit 3 ₁ of the movable beam section 2 ₁ as a non-driving unit when the switching operation leaves, apparently, the same effects as strong divergence force Can be obtained. As a result, it is possible to realize a high-frequency microswitch whose contacts are less sticky than in the past.

In addition, the wiring connection portion 5 may extend from both signal wirings 4 ₁ , 4 ₂ so as to face each other, and the contact portion 6 may be formed from the center, thereby making the shape more symmetrical. the gap between the contact portion 6 and the signal line 4 ₂ is kept constant irrespective of the temperature variation since.

The electrical drive units 3 ₁ and 3 ₂ may be piezoelectric drive units or electrostatic drive units. When the piezoelectric drive unit is used, a laminated structure including a lower electrode / piezoelectric layer / upper electrode is provided. Note that PZT is typically used as the piezoelectric layer.

  In the case of an electrostatic drive unit, a lower electrode and an upper electrode facing the lower electrode through a gap may be provided. In this case, a metal layer such as Pt may be used for the lower electrode. Alternatively, an ion implantation layer may be used.

  Next, the RF-MEMS switch according to the first embodiment of the present invention will be described with reference to FIGS. 2 is a schematic plan view of the RF-MEMS switch according to the first embodiment of the present invention, and FIG. 3 is a cross-sectional view taken along the alternate long and short dash line connecting AA ′ in FIG. 2. 3A is a cross-sectional view in the OFF state, and FIG. 3B is a cross-sectional view in the ON state.

As shown in FIGS. 2 and 3, a bonded substrate 10 made of a silicon layer 13 having a thickness of 5 μm to 25 μm is prepared by polishing the silicon substrate bonded to the silicon substrate 11 via the SiO ₂ film 12. . Slits 14 reaching the silicon substrate 11 are formed in the bonded substrate 10 to form movable beam portions 20 ₁ and 20 ₂ arranged in line symmetry. The SiO ₂ film 12 in contact with the bottom surfaces of the movable beam portions 20 ₁ and 20 ₂ is removed after the step of forming the slits 14. Alternatively, a large number of micro holes may be formed in the silicon layer 13 in the initial stage of device formation, and the SiO ₂ film 12 may be removed in advance. Further, before the silicon layer 13 is bonded to the silicon substrate 11, the SiO ₂ film 12 at the corresponding location may be removed in advance.

In the movable beam portions 20 ₁ and 20 ₂ , signal wirings 26 ₁ and 26 _{2 made of} an Au sputtered film and piezoelectric driving units 21 ₁ and 21 ₂ are arranged in the longitudinal direction, respectively, and an end portion of _one signal wiring 26 ₁ The connection wiring 28 provided with the contact portion 30 is integrally formed with the support portion 29 by gold plating. At this time, the contact gap between the contact portion 30 and the signal line 26 ₂ facing portion in the OFF state is designed to be 0.3 [mu] m. One end of the signal lines ₂₆ 1, 26 ₂ are respectively connected to the signal electrode pads ₂₇ 1, 27 _2.

The piezoelectric drive units 21 ₁ and 21 ₂ have a laminated structure of a lower electrode 22 ₁ / PZT 23 ₁ / upper electrode 24 ₁ and a lower electrode 22 ₂ / PZT 23 ₂ / upper electrode 24 ₂ , respectively. The upper electrodes 24 ₁ and 24 ₂ are connected to the drive electrode pads 25 ₁₁ and 25 ₂₁ , respectively, and the lower electrodes 22 ₁ and 22 ₂ are connected to the common drive electrode pad 25 ₃ . The lower electrodes 22 ₁ and 22 ₂ and the upper electrodes 24 ₁ and 24 ₂ are formed of a Pt sputtering film.

FIG. 4 is an explanatory diagram of the switching operation of the RF-MEMS switch according to the first embodiment of the present invention, and is shown as a cross-sectional view along the alternate long and short dash line connecting BB ′ in FIG. As shown in FIG. 4 (a), in the OFF state, the gap between the contact portion 30 and the signal line 26 ₂ is maintained to be a gap of 0.3 [mu] m.

As shown in FIG. 4 (b), when a voltage is applied between the driving electrode pads 25 ₂₁ and the driving electrode pad 25 _3, the contact portion raised is bent movable beam section 20 ₂ as shown in FIG. 3 (b) an ON state in contact with the 30 and the signal line 26 _2.

When stopping the application of the driving voltage in order to switch OFF, spring force possessed by the movable beam section 20 _2, and the contact portion 30 acts in a direction in which the contact leaves are deviated, again, shown in FIG. 4 (a) It becomes OFF state. In this case, by bending the movable beam portion 20 ₁ by applying a voltage to the piezoelectric driving unit 21 _1, it may be a force for causing the divergence between the contacts becomes greater.

The voltage applied to the piezoelectric driving unit 21 ₁ need not be the same as the voltage to be applied at the ON to the piezoelectric driving unit 21 _2, may be half the voltage if it is sufficient to deviate the contacts. Moreover, it is not always necessary to apply the voltage in the OFF state as to the time for applying the voltage, and it is sufficient to apply the voltage for a very short time until the contact where the adhesive force between the contacts becomes a problem.

FIG. 4C is an explanatory diagram of the OFF state when a temperature change occurs. Even when caused temperature changes, since the movable beam section 20 ₁ for supporting the movable beam section 20 ₂ also signal lines 26 ₁ connection wiring 28 via the supporting signal line 26 ₂ changes in the same way, the contact gap As a result, there is almost no change and sticking does not occur.

In the first embodiment of the present invention, since the form of the movable beam sections 20 _1, 20 ₂ for supporting the signal line 26 _1, 26 ₂ symmetrically, act to offset the amount of variation due to a temperature change each other , The change of the contact gap can be suppressed.

Next, the RF-MEMS switch according to the second embodiment of the present invention will be described with reference to FIG. Figure 5 is a schematic plan view of the RF-MEMS switch according to the embodiment 2 of the present invention, in which lost connection with the drive electrode pads to the piezoelectric driving unit 21 ₁ provided in the movable beam section 20 _1.

That is, the voltage applied to the piezoelectric driving unit 21 ₁ provided in the movable beam section 20 ₁ serving as a non-driving unit at the time of switching is obtained by omitted since there is no need in principle. This eliminates the need for a wiring and drive electrode pad forming step for the external power supply, and eliminates the need for a space for providing the wiring and the drive electrode pad. Other configurations are exactly the same as those of the RF-MEMS switch of the first embodiment, and the cross-sectional structure along the alternate long and short dash line connecting BB ′ is the same as the structure shown in FIG.

Next, with reference to FIG. 6 thru | or FIG. 10, the RF-MEMS switch of Example 3 of this invention is demonstrated. FIG. 6 is a schematic plan view of an RF-MEMS switch according to the third embodiment of the present invention. The basic configuration is the same as that of the first embodiment, but the contact portions are symmetrical. That is, to shorten the length of the connecting wires 28 ₁ to be connected to the signal lines 26 _1, is provided with a connection wiring 28 ₂ to the signal lines 26 ₂ via the supporting portion 29 _2.

Next, although the manufacturing process of the contact portion will be briefly described, the formation process of the piezoelectric drive unit and the electrode pad is omitted here. First, as shown in FIG. 7A, a resist mask 15 having openings corresponding to slits is provided on a bonded substrate 10 composed of a silicon substrate 11 / SiO ₂ film 12 / silicon layer 13.

Next, as shown in FIG. 7B, the slit 14 is formed by performing dry etching such as a polishing process using the resist mask 15 as a mask. Thereafter, the movable beam section ₂₀ 1, 20 ₂ of the lower portion of the SiO ₂ film 12 surrounded by the slit 14, which is a hollow be wet etching. In this case, in the drawing, the remaining SiO ₂ film 12 is highlighted, but actually it is further retracted. As another method, the SiO ₂ film 12 under the movable beam portions 20 ₁ and 20 ₂ is previously removed by arranging many finer holes in the same process as in FIGS. 7A and 7B. May be. In this case, the remaining SiO ₂ film 12 does not retreat excessively.

Next, as shown in FIG. 7C, after the resist mask 15 is removed, signal wirings 26 ₁ and 26 ₂ are formed on the movable beam portions 20 ₁ and 20 ₂ by mask sputtering of the Au film, respectively.

Next, as shown in FIG. 7D, a first sacrificial layer 16 is formed on the entire surface using TEOS (tetraethoxysilane). Next, as shown in FIG. 8E, the openings 17 for the ends of the signal wirings 26 ₁ and 26 ₂ are formed.

Then, as shown in FIG. 8 (f), by electrolytic plating with Au using a plating frame (not shown), the supporting portion 29 ₁ to the signal lines 26 _1, also to the signal line 26 ₂ the supporting portion 29 ₂ and the connection wiring 28 ₂ are integrally formed.

Next, as shown in FIG. 8G, the second sacrificial layer 18 is again formed on the entire surface using TEOS. Then, as shown in FIG. 9 (h), forming a recess 19 ₂ for forming the concave portion 19 ₁ and the contact portion with respect to the supporting portion 29 _1. Incidentally, the concave portion 19 ₂ for forming the contact portion is formed so as to face the end of the connecting wiring 28 _2.

Then, as shown in FIG. 9 (i), by electrolytic plating with Au using a plating frame (not shown), with a support portion 29 ₃ and the contact portion 30 to be connected to the support portion 29 _first connecting wire 28 ₁ are integrally formed.

Next, as shown in FIG. 9J, the second sacrificial layer 18 and the first sacrificial layer 16 are sequentially removed by etching using hydrofluoric acid vapor. Since the TEOS film is easier to etch than the SiO ₂ film formed by the normal silicon thermal oxidation process, the SiO ₂ film 12 is not undesirably removed in this process.

FIG. 10 is an explanatory diagram of the switching operation of the RF-MEMS switch according to the third embodiment of the present invention. As shown in FIG. 10 (a), in the OFF state, the gap between the contact portion 30 and the connecting wiring 28 ₂ is maintained to be a gap of 0.3 [mu] m.

As shown in FIG. 10 (b), when a voltage is applied between the driving electrode pads 25 ₂₁ and the driving electrode pad 25 _3, the contact portion raised is bent movable beam section 20 ₂ as shown in FIG. 3 (b) an ON state in contact with the 30 and the connection wiring 28 _2.

When stopping the application of the driving voltage in order to switch OFF, spring force possessed by the movable beam section 20 _2, and the contact portion 30 acts in a direction in which the contact leaves are deviated, again, shown in Fig. 10 (a) It becomes OFF state. In this case, by bending the movable beam portion 20 ₁ by applying a voltage to the piezoelectric driving unit 21 _1, it may be a force for causing the divergence between the contacts becomes greater.

FIG. 10C is an explanatory diagram of an OFF state when a temperature change occurs. Even when it caused temperature changes, also the movable beam section 20 ₁ for supporting the connection wiring 28 ₁ movable beam section 20 ₂ also via the signal lines 26 ₁ to support the connection wiring 28 ₂ via the signal line 26 ₂ same Thus, the contact gap hardly changes, so that sticking does not occur.

In Example 3 of the present invention, the movable beam section 20 _1, 20 ₂ for supporting the signal line 26 _1, 26 ₂ with symmetrically formed, since the contact portion is also symmetrically formed, due to the temperature change The amount of variation can be canceled out more effectively.

Next, with reference to FIG.11 and FIG.12, the RF-MEMS switch of Example 4 of this invention is demonstrated. Figure 11 is a schematic plan view of the RF-MEMS switch according to the embodiment 4 of the present invention, the basic configuration is rotated a pair of the movable beam section ₂₀ 1, 20 ₂ of the RF-MEMS switch of Examples 1 It corresponds to a symmetrical arrangement. By arranging them rotationally symmetrical in this way, they are divided into two independent regions by the slit 14, and each of the piezoelectric drive units 21 ₁ and 21 ₂ has independent drive electrode pads 25 ₁₁ and 25 ₁₂ and drive electrodes, respectively. Connected to pads 25 ₂₁ and 25 ₂₂ .

12A and 12B are explanatory diagrams of the switching operation of the RF-MEMS switch according to the fourth embodiment of the present invention. FIG. 12A shows the OFF state, FIG. 12B shows the ON state, and FIG. 12C shows the temperature. The OFF state after the change is shown. The distance between the signal line 26 ₁ and the signal line 26 ₂ only differs basic operation and effects are the same as those in Example 1 above.

Next, with reference to FIG.13 and FIG.14, the RF-MEMS switch of Example 5 of this invention is demonstrated. Figure 13 is a schematic plan view of the RF-MEMS switch according to the embodiment 5 of the present invention, the basic configuration is rotated a pair of the movable beam section ₂₀ 1, 20 ₂ of the RF-MEMS switch according to the embodiment 3 of the It corresponds to a symmetrical arrangement. By arranging them rotationally symmetrical in this way, they are divided into two independent regions by the slit 14, and each of the piezoelectric drive units 21 ₁ and 21 ₂ has independent drive electrode pads 25 ₁₁ and 25 ₁₂ and drive electrodes, respectively. Connected to pads 25 ₂₁ and 25 ₂₂ .

14A and 14B are explanatory diagrams of the switching operation of the RF-MEMS switch according to the fifth embodiment of the present invention. FIG. 14A shows the OFF state, FIG. 14B shows the ON state, and FIG. 14C shows the temperature. The OFF state after the change is shown. The distance between the signal line 26 ₁ and the signal line 26 ₂ only differs basic operation and effects are the same as in Example 3 above.

  Next, an RF-MEMS switch according to Example 6 of the present invention will be described with reference to FIGS. FIG. 15 is a schematic plan view of the RF-MEMS switch according to the sixth embodiment of the present invention, which is the same as the RF-MEMS switch according to the first embodiment except that the piezoelectric driving unit is replaced with an electrostatic driving unit. .

  FIG. 16 is a schematic cross-sectional view of the RF-MEMS switch according to the sixth embodiment of the present invention. 16A is a cross-sectional view taken along the alternate long and short dash line connecting AA ′ in FIG. 15, and FIG. 16B is a cross-sectional view taken along the alternate long and short dash line connecting BB ′ and FIG. c) is a cross-sectional view after a temperature change along the alternate long and short dash line connecting BB ′.

As shown, the electrostatic driving unit ₃₁ 1, 31 ₂ is composed of an ion-implanted layer ₃₂ 1, 32 ₂ and the upper electrode ₃₃ 1, 33 ₂ serving as a lower electrode, both of the gap is, for example, a 1μm . The upper electrodes 33 ₁ and 33 ₂ are formed by electroplating Au integrally with the support portions 34 ₁ and 34 _{2 on} both sides. Further, in the step of forming the ion implantation layers 32 ₁ and 32 ₂ , wiring regions 35 ₁ and 35 ₂ connected to the drive electrode pads 25 ₁₁ and 25 ₂₁ are simultaneously formed.

Next, the manufacturing process will be briefly described. First, as shown in FIG. 17A, P is ion-implanted into the silicon layer 13 of the bonded substrate 10 composed of the silicon substrate 11 / SiO ₂ film 12 / silicon layer 13, thereby forming an n ⁺ -type ion implantation layer 32. ₂ is formed. Note that employs the same steps in the side of the movable beam section 20 _1.

Next, as shown in FIG. 17B, the slit (14) is formed by performing dry etching such as a polishing process using a resist mask (not shown) as a mask. Thereafter, the lower portion of the SiO ₂ film 12 of the movable beam section 20 ₂ and hollow by wet etching.

Then, as shown in FIG. 17 (c), after removing the resist mask, to form the signal lines 26 ₂ on the movable beam section 20 ₂ by masking sputtering Au film. Next, as shown in FIG. 17D, a sacrificial layer 36 is formed on the entire surface using TEOS.

Then, as shown in FIG. 18 (e), to expose the driving electrode pad forming region and the signal electrode pad forming region, so as to face the end of the recess 37 of the signal line 26 ₂ for forming the contact portion Form.

Then, as shown in FIG. 18 (f), by electrolytic plating with Au using a plating frame (not shown), to form the driving electrode pads 25 ₃ and the signal electrode pads 27 _2, the upper electrode 33 ₂ And the connection wiring 28 provided with the contact portion 30 are integrally formed. At this time, the supporting portion 34 ₂ to both sides of the upper electrode 33 ₂ are integrally formed. Next, as shown in FIG. 18G, the sacrificial layer 36 is removed by etching using hydrofluoric acid vapor.

The switching operation in the sixth embodiment is the same as that in the case of the piezoelectric drive, except that the driving unit is a piezoelectric drive or an electrostatic drive. That is, when the switch is ON state, is driven by utilizing the attracting force by applying a mutually different polarities of the voltage between the ion implanted layer 32 ₁ and the upper electrode 33 ₁ to be the lower electrode.

Also in this case, since the two movable beam portions 20 ₁ and 20 ₂ are formed in line symmetry and the lengths of the signal wirings 26 ₁ and 26 ₂ are the same, as shown in FIG. Even after the change, the two movable beam portions 20 ₁ and 20 ₂ are similarly deformed, so that the gap hardly changes.

  In the sixth embodiment, since the piezoelectric body forming step is not required, the manufacturing process is simplified and the warp of the movable beam portion due to the stress of the piezoelectric body is reduced. In addition, since an ion implantation layer using silicon rather than a metal material is used as the lower electrode, the influence of deformation due to temperature change is less likely to occur.

  Next, with reference to FIG.19 and FIG.20, RF-MEMS switch of Example 7 of this invention is demonstrated. FIG. 19 is a schematic plan view of an RF-MEMS switch according to Example 7 of the present invention. The RF-MEMS switch according to Example 6 except that the ion implantation layer serving as the lower electrode is replaced with an Au film. It is the same.

  FIG. 20 is a schematic cross-sectional view of the RF-MEMS switch according to the seventh embodiment of the present invention. 20A is a cross-sectional view taken along the alternate long and short dash line connecting AA ′ in FIG. 19, and FIG. 20B is a cross-sectional view taken along the alternate long and short dash line connecting BB ′. c) is a cross-sectional view after a temperature change along the alternate long and short dash line connecting BB ′.

As shown in the figure, lower electrodes 38 ₁ and 38 ₂ and wiring layers 39 ₁ and 39 ₂ are simultaneously formed by Au mask sputtering. When the lower electrodes 38 ₁ and 38 ₂ are formed with the same material and the same thickness as the signal wirings 26 ₁ and 26 ₂ , the lower electrodes 38 ₁ and 38 ₂ and the signal wirings 26 ₁ and 26 ₂ are formed simultaneously. . Also in this case, the gap between the lower electrodes 38 ₁ and 38 ₂ and the upper electrodes 33 ₁ and 33 ₂ is, for example, 1 μm.

The switching operation in the seventh embodiment is the same as that in the sixth embodiment except that the lower electrode is different from the ion implantation layer or the metal film. That is, when the switch is ON state, utilizing driving the attracting force by applying a mutually different polarities of the voltage between the lower electrode 38 ₁ and the upper electrode 33 _1.

Also in this case, since the two movable beam portions 20 ₁ and 20 ₂ are formed symmetrically and the lengths of the signal wirings 26 ₁ and 26 ₂ are the same, as shown in FIG. Even after the change, the two movable beam portions 20 ₁ and 20 ₂ are similarly deformed, so that the gap hardly changes.

  In the seventh embodiment, the ion implantation process is not required, and only a general film forming process is required, so that the configuration of the manufacturing apparatus is simplified. In addition, when the lower electrode and the signal wiring are formed at the same time, the number of manufacturing steps can be reduced.

Next, an RF-MEMS switch according to an eighth embodiment of the present invention will be described with reference to FIG. Figure 21 is a schematic plan view of the RF-MEMS switch according to the embodiment 8 of the present invention, the basic configuration is rotated the movable beam sections ₂₀ 1, 20 ₂ a pair of RF-MEMS switch of Examples 6 It corresponds to a symmetrical arrangement.

  As described above, the rotationally symmetrical arrangement eliminates the need for arranging the wide electrostatic drive units in parallel, and thus space saving is possible.

In the description of the above Examples 6 to 8, in order to simplify the description, are formed so as to connect the connection wiring 28 signal lines 26 ₁ only, the above Example 3 or Example As shown in FIG. 5, the contact portions may be symmetrically provided on both signal wirings 26 ₁ and 26 ₂ .
In addition, the drive electrode pad for the electrostatic drive unit on the side serving as the non-drive unit during the switching operation may be omitted.

Here, the following additional notes are attached to the embodiment of the present invention including Examples 1 to 8.
(Supplementary Note 1) A silicon substrate, a silicon layer partially adhered to the silicon substrate via an insulating layer, a first signal wiring and a first electrical wiring formed on the silicon layer A first movable beam portion in which a drive unit is arranged in the longitudinal direction, and the silicon layer provided in a symmetrical position of either line symmetry or rotational symmetry with respect to the first movable beam portion. And a second movable beam part in which the second signal wiring and the first electric driving unit are arranged in the longitudinal direction, and an end of the first signal wiring on the first electric driving unit side. At least a wiring connection portion provided with a contact portion, wherein one of the first movable beam portion and the second movable beam portion is a driving portion in a switching operation, and the first movable beam portion And the other of the second movable beam portions is in switching operation. RF microswitch, which is a that non-driven unit.
(Supplementary note 2) The high-frequency microswitch according to supplementary note 1, wherein the size of the first signal wiring and the size of the second signal wiring are the same.
(Appendix 3) The high frequency micro wave according to appendix 1 or appendix 2, characterized in that a voltage application terminal is missing in the electric drive section provided in the movable beam section which is a non-drive section in the switching operation. switch.
(Additional remark 4) The 2nd wiring connection part which the said wiring connection part is the length which does not reach the said 2nd signal wiring, and it is connected to the said 2nd signal wiring, and extends just under the said contact part. The high-frequency microswitch according to any one of supplementary notes 1 to 3, characterized by comprising:
(Supplementary note 5) The high-frequency microswitch according to supplementary note 4, wherein the wiring connection portion and the second wiring connection portion are made of thick film plating.
(Supplementary note 6) The high frequency device according to any one of supplementary notes 1 to 5, wherein the electrical drive unit is a piezoelectric drive unit having a laminated structure including a lower electrode / piezoelectric layer / upper electrode. Microswitch.
(Additional remark 7) The said electrical drive part is an electrostatic drive part which consists of a lower electrode and the upper electrode which opposes the said lower electrode through a space | gap, Any one of Additional remark 1 thru | or Additional remark 5 characterized by the above-mentioned. High frequency micro switch.
(Supplementary note 8) The high-frequency microswitch according to supplementary note 7, wherein the lower electrode is made of an ion implantation layer.

DESCRIPTION OF SYMBOLS 1 Silicon substrate 2 ₁ , 2 ₂ Movable beam part 3 ₁ , 3 ₂ Electrical drive part 4 ₁ , 4 ₂ Signal wiring 5 Wiring connection part 6 Contact part 7 Slit 8 ₁ , 8 ₂ Signal electrode pad 9 ₁₁ , 9 ₂₁ , 9 ₃ Driving electrode pads 10, 50 Bonded substrate 11, 51 Silicon substrate 12, 52 SiO ₂ film 13, 53 Silicon layer 14, 54 Slit 15 Resist mask 16 First sacrificial layer 17 Opening 18 Second sacrificial layer 19 ₁ , 19 ₂ concave portions 20 ₁ , 20 ₂ , 60 movable beam portions 21 ₁ , 21 ₂ , 61 piezoelectric drive portions 22 ₁ , 22 ₂ , 62 lower electrodes 23 ₁ , 23 ₂ , 63 PZT
24 ₁ , 24 ₂ , 64 Upper electrodes 25 ₁₁ , 25 ₁₂ , 25 ₂₁ , 25 ₂₂ , 25 ₃ , 65 ₁ , 65 ₂ Drive electrode pads 26 ₁ , 26 ₂ , 66 Signal wirings 27 ₁ , 27 ₂ , 67 ₁ , 67 ₂ signal electrode pads 28 _{1, 28} 2, 68 connection wiring 29 _{1, 29} 2, 29 ₃ supporting portions 30, 70 contact portion ₃₁ 1, 31 ₂ electrostatic driving unit ₃₂ 1, 32 ₂ ion implantation layer 33 _1, 33 ₂ upper electrode ₃₄ 1, 34 ₂ supporting unit ₃₅ 1, 35 ₂ wiring region 36 sacrificial layer 37 recesses ₃₈ 1, 38 ₂ lower electrode ₃₉ 1, 39 ₂ wiring layers 69 supporting portion

Claims (5)

A silicon substrate;
A silicon layer in close contact with the silicon substrate via an insulating layer;
A first movable beam portion formed in the silicon layer and having a first signal wiring and a first electrical drive unit arranged in a longitudinal direction;
The second signal wiring and the first electric drive unit are formed on the silicon layer provided at a symmetrical position of either line symmetry or rotational symmetry with respect to the first movable beam portion. Second movable beam portions arranged in the longitudinal direction;
A wiring connection portion provided with a contact portion provided at an end portion of the first signal wiring on the first electric drive portion side;
One of the first movable beam portion and the second movable beam portion is a drive unit in a switching operation,
The high-frequency microswitch, wherein the other of the first movable beam portion and the second movable beam portion is a non-driving portion in a switching operation.   2. The high-frequency microswitch according to claim 1, wherein a size of the first signal wiring and a size of the second signal wiring are the same. The wiring connection portion has a length that does not reach the second signal wiring;
4. The high-frequency micro according to claim 1, further comprising a second wiring connection portion that is connected to the second signal wiring and extends to a position directly below the contact portion. 5. switch.   4. The high-frequency micro according to claim 1, wherein the electrical driving unit is a piezoelectric driving unit having a laminated structure including a lower electrode / a piezoelectric layer / an upper electrode. switch.   The said electric drive part is an electrostatic drive part which consists of a lower electrode and the upper electrode facing the said lower electrode through a space | gap, The any one of Claim 1 thru | or 3 characterized by the above-mentioned. High frequency micro switch.

Images