Method for manufacturing radio frequency MEMS switch and radio frequency MEMS switch

Abstract

The application discloses a manufacturing method of a radio frequency MEMS switch and a structure thereof, and the manufacturing method comprises the following steps: providing a first substrate, forming a co-planar waveguide and a first electrode on the first substrate, and connecting the first electrode with the co-planar waveguide; providing a second substrate, forming a contact structure on the second substrate; anodically bonding one side of the first substrate provided with the co-planar waveguide with one side of the second substrate formed with the contact structure, the contact structure is arranged correspondingly with the co-planar waveguide, and a gap is formed between the contact structure and the co-planar waveguide; forming a second electrode corresponding with the first electrode on the side of the second substrate away from the contact structure; and after the anodic bonding is completed, etching the second substrate to form a movable beam structure. The manufacturing method of the radio frequency MEMS switch can avoid the influence of an external electric field on a movable electrode assembly of the radio frequency MEMS switch during the anodic bonding, and improve the reliability of the switch.

Description

Technical Field

[0001] This invention relates to the field of semiconductor technology, and in particular to a method for fabricating a radio frequency (RF) MEMS switch and the RF MEMS switch itself. Background Technology

[0002] Micro-Electro-Mechanical Systems (MEMS) refer to systems designed and manufactured on a micrometer scale that integrate multiple components and are suitable for low-cost mass production. Radio Frequency Micro-Electro-Mechanical Systems (RF MEMS) are micro-electro-mechanical systems that integrate radio frequency signals and micromechanical systems; this is a technology that will have a significant impact on the radio frequency structures in existing radar and communications systems.

[0003] Among RF MEMS devices, the RF MEMS switch is currently the most attention-grabbing, serving as its core component. An RF MEMS switch is a micromechanical switch operating at frequencies from radio frequency to millimeter wave, relying on mechanical movement to control the switching on and off of transmission lines. Compared to traditional FETs and PIN diode switches, RF MEMS switches offer advantages such as low power consumption, high isolation, low insertion loss, and low intermodulation components. Switches have a wide range of applications in communication systems, serving as crucial control components in radar, electronic countermeasures, and wireless communication.

[0004] In existing technologies, the mainstream process for fabricating floating structures of RF MEMS switches is the sacrificial layer technology. However, traditional sacrificial layer release processes can lead to incomplete sacrificial layer release, failing to guarantee the surface cleanliness and good contact resistance of the RF MEMS switch. Furthermore, traditional sacrificial layer release processes typically use metal as the upper movable structure, but metal has a relatively low elastic modulus, resulting in reduced switch reliability.

[0005] To address the issue of incomplete sacrificial layer release, anodic bonding technology has been developed as an alternative to sacrificial layer technology, avoiding the defects caused by sacrificial layer release. However, the mainstream process for fabricating devices using anodic bonding currently involves completing the fabrication of the upper and lower structures of the dual substrates before bonding. However, anodic bonding uses an external electric field to bond the two substrates, and this external electric field can affect the movable structure (suspended structure) of the RFMEMS switch. Furthermore, the problems caused by the micro-load effect are currently being overlooked, often resulting in etching of the lower side of the suspended structure, leading to a reduction in device thickness and affecting device stability and performance.

[0006] The information disclosed in this background section is intended only to enhance the understanding of the overall background of the invention and should not be construed as an admission or in any way implying that the information constitutes prior art known to those skilled in the art. Summary of the Invention

[0007] The purpose of this invention is to provide a method for fabricating an RF MEMS switch and an RF MEMS switch, which can avoid the influence of the external electric field applied during anodic bonding on the movable electrode components of the RF MEMS switch, thereby improving the reliability of the switch.

[0008] To achieve the above objectives, embodiments of the present invention provide a method for fabricating a radio frequency MEMS switch, comprising: providing a first substrate, forming a coplanar waveguide and a first electrode on the first substrate, the first electrode being connected to the coplanar waveguide; providing a second substrate, forming a contact structure on the second substrate; anoly bonding one side of the first substrate having the coplanar waveguide to one side of the second substrate having the contact structure, the contact structure being correspondingly disposed to the coplanar waveguide and a gap being formed between the contact structure and the coplanar waveguide; forming a second electrode corresponding to the first electrode on the side of the second substrate opposite to the contact structure; and after completing the bonding, etching the second substrate to form a movable beam structure.

[0009] In one or more embodiments of the present invention, prior to the step of forming a contact structure on the second substrate, the method further includes: forming an oxide isolation layer on the second substrate; the contact structure is formed on the oxide isolation layer.

[0010] In one or more embodiments of the present invention, prior to the step of forming an oxide isolation layer on the second substrate, the method further includes the step of forming a groove on the second substrate; the oxide isolation layer is formed on the bottom surface of the groove.

[0011] In one or more embodiments of the present invention, forming a coplanar waveguide and a first electrode on the first substrate, wherein the first electrode is connected to the coplanar waveguide, includes: sequentially magnetron sputtering an adhesion layer and a metal layer on the first substrate, and photolithography / IBE etching to form the coplanar waveguide and the first electrode; wherein the coplanar waveguide includes a signal line and a ground line spaced apart from each other, a break is formed in the middle of the signal line, and the first electrode is connected to the ground line.

[0012] In one or more embodiments of the present invention, the contact structure is correspondingly disposed with the coplanar waveguide and a gap is formed between the contact structure and the coplanar waveguide, including: the contact structure is correspondingly disposed with the signal line and the contact structure is suspended above the break of the signal line, and the contact structure is configured such that when it undergoes elastic deformation and contacts the signal line, it can connect the signal line.

[0013] In one or more embodiments of the present invention, before the step of forming a second electrode corresponding to the first electrode on the side of the second substrate opposite to the contact structure, the method further includes: thinning and polishing the second substrate.

[0014] In one or more embodiments of the present invention, the second substrate has a certain elastic modulus, and the second substrate includes SOI.

[0015] An embodiment of the present invention also provides a radio frequency MEMS switch, comprising: a first substrate, a coplanar waveguide, a first electrode, and a movable electrode assembly; the coplanar waveguide is formed on the first substrate and includes a signal line and a ground line spaced apart from each other, the signal line having a break in the middle; the first electrode is disposed on the ground line; the movable electrode assembly includes a movable beam structure and a second electrode, the movable beam structure is mounted on the first substrate, the movable beam structure having a contact structure corresponding to the break position of the signal line, the contact structure being spaced apart from the signal line, and the second electrode being disposed on the movable beam structure and corresponding to the first electrode.

[0016] In the above technical solution, when the RF MEMS switch is in its initial state, the RF signal cannot be transmitted through the signal line because it is disconnected. When a driving voltage is applied to the second electrode, the second electrode is subjected to electrostatic force and moves downward together with the movable beam structure. The contact structure then contacts the disconnected signal line, allowing the RF signal to propagate via the signal line-contact structure-signal line route. This solution changes the vertical position of the movable electrode assembly by applying a driving voltage, thereby changing the on / off state of the signal line and controlling the on / off state of the RF MEMS switch, thus achieving control of the RF signal.

[0017] In one or more embodiments of the present invention, an oxide isolation layer is provided between the contact structure and the movable beam structure.

[0018] In one or more embodiments of the present invention, the ground wire includes a first ground wire and a second ground wire, the first ground wire and the second ground wire are spaced apart from each other, and the signal wire is disposed between the first ground wire and the second ground wire and is parallel to both the first ground wire and the second ground wire.

[0019] In one or more embodiments of the present invention, the movable beam structure is configured as a beam structure made of the top layer silicon of SOI, and the movable beam structure has a certain elastic modulus.

[0020] In one or more embodiments of the present invention, the material of the contact structure includes Au, TiW, AuPt alloy, and AuRh alloy.

[0021] Compared with the prior art, the method for fabricating the radio frequency MEMS switch according to the present invention uses anodic bonding technology instead of sacrificial layer technology, which fundamentally avoids the defects caused by sacrificial layer release, while improving the cleanliness of the radio frequency MEMS switch surface and reducing the contact resistance between the metal contact structure and the signal line.

[0022] The method for fabricating a radio frequency MEMS switch according to the present invention adopts the steps of anodic bonding followed by photolithography to fabricate the radio frequency MEMS switch, thereby avoiding the influence of the external electric field applied during anodic bonding on the movable electrode components.

[0023] The method for fabricating an RF MEMS switch according to an embodiment of the present invention deposits an oxide isolation layer on the surface of a second substrate, which avoids the adverse effects of micro-load on the RF MEMS switch and protects the integrity of the overall structure during the deep silicon etching process of the second substrate.

[0024] The radio frequency MEMS switch of the present invention uses SOI top-layer silicon (single crystal silicon) as the main material of the movable beam structure, which avoids the fatigue problem of traditional metal structures over a long period of time, enhances the reliability of the radio frequency MEMS switch, and ensures the working life of the radio frequency MEMS switch. Attached Figure Description

[0025] Figure 1 This is a flowchart illustrating a method for fabricating an RF MEMS switch according to an embodiment of the present invention.

[0026] Figures 2A to 2I This is a process diagram of an embodiment of the radio frequency MEMS switch of the present invention.

[0027] Figure 3 The figure shows the simulation results of the insertion loss of an RF MEMS switch according to an embodiment of the present invention using HFSS finite element simulation software.

[0028] Figure 4 The figure shows the simulation results of the isolation of the radio frequency MEMS switch according to one embodiment of the present invention using HFSS finite element simulation software.

[0029] Figure 5 This is a schematic diagram of the structure of a radio frequency MEMS switch according to an embodiment of the present invention. Detailed Implementation

[0030] The specific embodiments of the present invention will now be described in detail with reference to the accompanying drawings, but it should be understood that the scope of protection of the present invention is not limited to the specific embodiments.

[0031] Unless otherwise expressly stated, throughout the specification and claims, the term "comprising" or its variations such as "including" or "comprises" shall be understood to include the stated elements or components without excluding other elements or other components.

[0032] First, a terminology explanation is provided for the proper nouns used in this application.

[0033] Isolation: The ratio of the power transmitted when the switch is open to the power transmitted when the circuit is on, indicating the degree of isolation of the signal when the switch is off.

[0034] Insertion loss: The ratio of the power transmitted to the load when the switch is on to the power transmitted to the load by an ideal switch; it refers to the extent to which the internal resistance attenuates the signal.

[0035] Drive voltage: The minimum DC bias voltage required to move the upper electrode to the contact position with the lower electrode plate.

[0036] Switch response time: The time required for the upper electrode to move from the on position to the off position.

[0037] Switch life: The minimum number of normal operating cycles a switch can complete before it fails.

[0038] Micro-load effect: During plasma etching, the etching rate becomes interdependent with the aspect ratio. In other words, wide trenches are etched faster than narrow trenches.

[0039] As mentioned in the background section, radio frequency (RF) MEMS switches have always been a research hotspot both domestically and internationally. Many research institutes and universities in China are conducting RF MEMS research, such as the 13th and 55th Research Institutes of China Electronics Technology Group Corporation, Tsinghua University, the Institute of Microelectronics of Peking University, Southeast University, Nanjing University of Science and Technology, Shandong University, and North University of China. Some of these institutions have already achieved switches with high reliability. However, most research is still in the stages of material selection, modeling and analysis, simulation, structural optimization, and process improvement. They have not yet manufactured RF MEMS switch products applicable to real-world situations, nor have they formed an industrial chain for switches and related products. There is still a significant gap between China and foreign MEMS research and manufacturing capabilities.

[0040] Meanwhile, significant challenges remain in the fabrication of RF MEMS switches in China. Firstly, sacrificial layer technology is often used to create floating structures; however, incomplete release of the sacrificial layer can affect the cleanliness of the contact surface and contact resistance. Secondly, bonding processes are used to create floating structures, often involving building the upper and lower substrates before bonding. The applied electric field during bonding can affect the upper movable electrode assembly, for example, by introducing stress. Furthermore, a metal top electrode is typically used to reduce the switch's drive voltage; however, the low elastic modulus of the metal structure layer sacrifices the switch's reliability.

[0041] To address the aforementioned technical problems, this invention creatively proposes a method for fabricating a radio frequency (RF) MEMS switch. This method employs a sequence of anodic bonding followed by photolithography to fabricate the RF MEMS switch, thereby avoiding the impact of the external electric field applied during anodic bonding on the movable electrode assembly.

[0042] The following will describe in detail, with reference to the accompanying drawings, the fabrication method and structure of the radio frequency MEMS switch according to an embodiment of the present invention.

[0043] like Figure 1 As shown, an embodiment of the present invention provides a method for fabricating a radio frequency MEMS switch, comprising: s1: providing a first substrate, forming a coplanar waveguide and a first electrode on the first substrate, the first electrode being connected to the coplanar waveguide; s2: providing a second substrate, forming an oxide isolation layer on the second substrate; s3: forming a contact structure on the oxide isolation layer; s4: anoly bonding the side of the first substrate with the coplanar waveguide to the side of the second substrate with the contact structure, the contact structure being correspondingly disposed to the coplanar waveguide and a gap being formed between the contact structure and the coplanar waveguide; s5: thinning and polishing the second substrate; s6: forming a second electrode corresponding to the first electrode on the side of the second substrate away from the contact structure; s7: after completing the anoly bonding, etching the second substrate to form a movable beam structure.

[0044] Figures 2A to 2I This is a process diagram of the radio frequency MEMS switch described above in this invention. See below for reference. Figures 2A to 2I The fabrication method of the radio frequency MEMS switch of the present invention is described in detail.

[0045] like Figure 2AAs shown, a first substrate 10 is prepared: BF33 glass is selected as the first substrate 10 and it is cleaned; a coplanar waveguide (111 / 112 / 120) and a first electrode (130 / 140) are fabricated on the surface of the first substrate 10. The coplanar waveguide (111 / 112 / 120) includes signal lines 120 and ground lines (111 / 112) arranged at intervals. A break is formed in the middle of the signal line 120, and the first electrode (130 / 140) is connected to the ground line (111 / 112). Specifically, a 20 nm thick Ti and an 800 nm thick Au are sequentially magnetron sputtered on the first substrate 10, with Ti serving as the adhesion layer and Au serving as the metal layer. Then, photolithography is used to achieve the structure of the coplanar waveguide (111 / 112 / 120) and the first electrode (130 / 140), and IBE etching is used to form the coplanar waveguide (111 / 112 / 120) and the first electrode (130 / 140).

[0046] like Figure 2B As shown, the second substrate 20 is prepared: a high-resistivity SOI with a top silicon thickness of 5um is selected as the second substrate 20 and it is cleaned.

[0047] like Figure 2C As shown, a 3µm groove is formed by photolithography and deep silicon etching on the top silicon surface of the second substrate 20.

[0048] like Figure 2D As shown, an oxide isolation layer 30 is formed on the bottom surface of the groove. Specifically, a 200nm silicon dioxide layer is deposited in the groove by PECVD, and the silicon dioxide isolation layer structure is formed by photolithography and IBE etching.

[0049] like Figure 2E As shown, a contact structure 40 is fabricated on the oxide isolation layer 30. Specifically, a 20 nm thick Ti and a 500 nm thick Au are magnetron sputtered onto the oxide isolation layer 30, with Ti serving as the adhesion layer. Then, photolithography is used to achieve the contact structure, and IBE etching is used to form the contact structure 40. In other embodiments, the contact structure 40 can also be made of materials with high resistivity and hardness, such as TiW, AuPt alloys, and AuRh alloys, which can significantly extend the device lifespan.

[0050] like Figure 2F As shown, the first substrate 10 and the second substrate 20 are anoly bonded. By alignment, the SOI and the glass are combined using a standard anoly bonding process.

[0051] like Figure 2G As shown, the second substrate 20 is thinned and polished. The SOI bottom silicon is thinned and polished, the remaining bottom silicon is removed by deep silicon etching, and the SOI buried oxide layer is removed by NLD etching.

[0052] like Figure 2H As shown, a second electrode (51 / 52) is fabricated on the second substrate 20. A 20 nm thick Ti layer and a 200 nm thick Au layer are sequentially magnetron sputtered, with Ti serving as the adhesion layer. Then, photolithography and IBE etching are performed to form the second electrode structure.

[0053] like Figure 2I As shown, the final movable beam structure 60 is fabricated. A single-crystal silicon movable beam structure is formed through photolithography and deep silicon etching, followed by laser scribing.

[0054] In the above technical solution, the present invention uses a step of first anodic bonding and then photolithography to fabricate the radio frequency MEMS switch, which avoids the influence of the external electric field applied by anodic bonding on the movable electrode components, and further improves the stability and performance of the radio frequency MEMS switch.

[0055] Figure 3 This is a simulation result of the insertion loss of an RF MEMS switch fabricated using HFSS finite element simulation software according to one embodiment of the present invention. (Refer to...) Figure 3 As the operating frequency of the RF MEMS switch increases, the performance of the RF MEMS switch deteriorates. At a frequency of 5 GHz, its insertion loss is better than -0.10 dB.

[0056] Figure 4 This is a simulation result of the isolation of an RF MEMS switch fabricated using HFSS finite element simulation software according to one embodiment of the present invention. (Refer to...) Figure 4 The RF MEMS switch exhibits the best switching performance at an operating frequency of 5 GHz, with an isolation of -36 dB.

[0057] In the above embodiments, the movable beam structure is designed to reduce the driving voltage of the switch. The theoretical calculation formula for the driving voltage is as follows:

[0058]

[0059]

[0060] K total =nK m

[0061]

[0062] Where K is the elastic modulus of the structure, E is the elastic modulus of the second electrode material, w is the width of the beam, t is the thickness of the beam, l is the length of the beam, and K m V is the elastic coefficient of a single hinge, n is the number of hinges, and V pε is the switching drive voltage, g0 is the air gap, ε0 is the vacuum dielectric constant, and A is the overlapping area of ​​the first and second electrodes.

[0063] Based on the theoretical formulas above, it can be concluded that reducing the thickness and width of the beam, increasing the beam length, reducing the number of folded beams, or increasing the degree of bending of the folded beams can all result in a smaller elastic coefficient for the second electrode, thereby reducing the driving voltage of the switch. However, excessively reducing the elastic coefficient to obtain a low driving voltage makes the movable beam structure prone to downward pull-in and collapse, which will affect the reliability of the switch.

[0064] Based on the above considerations, one embodiment of the present invention provides a radio frequency MEMS switch, such as... Figure 5 As shown, this design allows the switch to have a relatively low drive voltage while also ensuring high reliability. The theoretical drive voltage of the switch, calculated using the formula, is 10.75V.

[0065] like Figure 5 As shown, an embodiment of the present invention provides an RF MEMS switch manufactured by the above-described method for fabricating an RF MEMS switch, comprising a first substrate 10, a coplanar waveguide (111 / 112 / 120), a first electrode (130 / 140), a second electrode (51 / 52), a movable beam structure 60, an oxide isolation layer 30, and a contact structure 40.

[0066] The first substrate 10 serves as a base, providing a supporting foundation for the overall switch structure. In one specific embodiment, the first substrate 10 is made of BF33 glass because BF33 glass has a smaller loss angle, which can reduce the loss of the first substrate 10.

[0067] A coplanar waveguide is formed on a first substrate 10. The coplanar waveguide includes signal lines 120 and ground lines spaced apart from each other. The ground lines include a first ground line 111 and a second ground line 112, which are spaced apart and parallel to each other. The signal lines 120 are disposed between the first ground lines 111 and the second ground lines 112 and are parallel to and equidistant from both the first ground lines 111 and the second ground lines 112. A break is provided in the middle of the signal lines 120. In a specific embodiment, Au is selected as the transmission line of the coplanar waveguide because gold has high electrical conductivity and stable chemical properties.

[0068] The first electrodes are respectively disposed on both sides of the signal line 120 and connected to the first ground line 111 and the second ground line 112. Preferably, the first electrodes are respectively disposed on the first ground line 111 and the second ground line 112.

[0069] The movable beam structure 60 is mounted on the first substrate 10 and has a certain elastic modulus. In one specific embodiment, the movable beam structure 60 is made of Si, the top layer of an SOI wafer, which has a large elastic modulus and good mechanical properties.

[0070] A second electrode is disposed on the surface of the movable beam structure 60 facing away from the first substrate 10, and the position of the second electrode corresponds vertically to that of the first electrode. The movable beam structure 60 separates the first electrode and the second electrode, which effectively avoids failure caused by contact wear between the first electrode and the second electrode. In a specific embodiment, the first electrode and the second electrode are made of Au, which has a low DC resistance.

[0071] The movable beam structure 60 has an oxide isolation layer 30 of the same or different shape as the first substrate 10 on the side close to it. The oxide isolation layer 30 is made of the traditional dielectric material Si3N4, which is not easily etched during plasma etching and plays a good protective role, avoiding the adverse effects of micro-load effects on the RF MEMS switch, and protecting the integrity of the overall structure during the deep silicon etching of the second substrate.

[0072] A contact structure 40 is provided on the movable beam structure 60 corresponding to the break position of the signal line 120. The contact structure 40 and the signal line 120 maintain a predetermined distance apart when no driving voltage is applied. The contact structure 40 and the movable beam structure 60 are isolated by an oxide insulating layer 30. In one specific embodiment, the contact structure 40 is made of Au, as gold has high electrical conductivity and stable chemical properties.

[0073] For example, a recessed structure may be formed on the first substrate 10, and a coplanar waveguide and a first electrode may be formed in the recess of the first substrate 10. Alternatively, the movable beam structure 60 may be configured as an arch bridge structure, mounted on the first substrate 10, with the oxide isolation layer 30 and the contact structure 40 disposed on its surface facing the first substrate 10.

[0074] The aforementioned radio frequency MEMS switch also includes leads and electrode pads, etc. The arrangement of the leads and the size and position of the electrode pads should be slightly adjusted according to the specific test conditions, which will not be further elaborated in this invention.

[0075] The following uses the above embodiments as examples to briefly explain the usage or principle of the radio frequency MEMS switch of the present invention.

[0076] When the RF MEMS switch is in its initial state, the RF signal cannot be transmitted through the signal line 120 because it is disconnected. When a driving voltage is applied to the second electrode, the second electrode is subjected to electrostatic force and moves downward together with the movable beam structure 60, and the contact structure 40 comes into contact with the disconnected signal line 120. When the contact structure 40 is in contact with the signal line 120, the RF signal can propagate through the signal line-contact-signal line route. The RF MEMS switch of the present invention changes the vertical position of the movable electrode assembly by applying a driving voltage, thereby changing the on / off state of the signal line 120, and thus controlling the on / off state of the RF MEMS switch, thereby achieving control of the RF signal.

[0077] Compared with the prior art, the method for fabricating the radio frequency MEMS switch according to the present invention uses anodic bonding technology instead of sacrificial layer technology, which fundamentally avoids the defects caused by sacrificial layer release, while improving the cleanliness of the radio frequency MEMS switch surface and reducing the contact resistance between the metal contact structure and the signal line.

[0078] The method for fabricating a radio frequency MEMS switch according to the present invention adopts the steps of anodic bonding followed by photolithography to fabricate the radio frequency MEMS switch, thereby avoiding the influence of the external electric field applied during anodic bonding on the movable electrode components.

[0079] The method for fabricating an RF MEMS switch according to an embodiment of the present invention deposits an oxide isolation layer on the surface of a second substrate, which avoids the adverse effects of micro-load on the RF MEMS switch and protects the integrity of the overall structure during the deep silicon etching process of the second substrate.

[0080] The radio frequency MEMS switch of the present invention uses SOI top-layer silicon (single crystal silicon) as the main material of the movable beam structure, which avoids the fatigue problem of traditional metal structures over a long period of time, enhances the reliability of the radio frequency MEMS switch, and ensures the working life of the radio frequency MEMS switch.

[0081] The foregoing description of specific exemplary embodiments of the invention is for illustrative and explanatory purposes. These descriptions are not intended to limit the invention to the precise forms disclosed, and it will be apparent that many changes and variations can be made in accordance with the foregoing teachings. The exemplary embodiments were chosen and described in order to explain the specific principles of the invention and its practical application, thereby enabling those skilled in the art to implement and utilize various different exemplary embodiments of the invention, as well as various different choices and variations. The scope of the invention is intended to be defined by the claims and their equivalents.

Claims

1. A method for fabricating a radio frequency MEMS switch, characterized in that, include: A first substrate is provided, and a coplanar waveguide and a first electrode are formed on the first substrate, wherein the first electrode is connected to the coplanar waveguide; A second substrate is provided, an oxide isolation layer is formed on the second substrate, and a contact structure is formed on the oxide isolation layer; The first substrate with a coplanar waveguide is anoly bonded to the second substrate with a contact structure, wherein the contact structure is correspondingly provided with the coplanar waveguide and a gap is formed between the contact structure and the coplanar waveguide; After the bonding is completed, a second electrode corresponding to the first electrode is formed on the side of the second substrate opposite to the contact structure; After completing the step of forming the second electrode, the second substrate is etched to form a movable beam structure.

2. The method for fabricating an RF MEMS switch as described in claim 1, characterized in that, Prior to the step of forming an oxide isolation layer on the second substrate, the method further includes: The step of forming a groove on the second substrate; The oxide isolation layer is formed on the bottom surface of the groove.

3. The method for fabricating an RF MEMS switch as described in claim 1, characterized in that, The method of forming a coplanar waveguide and a first electrode on the first substrate, wherein the first electrode is connected to the coplanar waveguide, includes: A magnetron sputtering adhesive layer and a metal layer are sequentially deposited on the first substrate, and a coplanar waveguide and a first electrode are formed by photolithography / IBE etching; wherein, the coplanar waveguide includes a signal line and a ground line arranged at intervals, a break is formed in the middle of the signal line, and the first electrode is connected to the ground line.

4. The method for fabricating an RF MEMS switch as described in claim 3, characterized in that, The contact structure is correspondingly disposed to the coplanar waveguide, and a gap is formed between the contact structure and the coplanar waveguide, including: The contact structure is configured to correspond to the signal line, and the contact structure is suspended above the break in the signal line. The contact structure is configured to connect the signal line when it undergoes elastic deformation and contacts the signal line.

5. The method for fabricating an RF MEMS switch as described in claim 1, characterized in that, Before the step of forming a second electrode corresponding to the first electrode on the side of the second substrate opposite to the contact structure, the method further includes: The second substrate is then thinned and polished.

6. The method for fabricating an RF MEMS switch as described in claim 1, characterized in that, The second substrate has a certain elastic modulus, and the second substrate includes SOI.

7. An RF MEMS switch fabricated using the method for fabricating an RF MEMS switch as described in any one of claims 1-6, characterized in that, include: First substrate; A coplanar waveguide is formed on the first substrate and includes signal lines and ground lines spaced apart from each other, with a break in the middle of the signal lines. The first electrode is disposed on the ground wire; as well as A movable electrode assembly includes a movable beam structure and a second electrode. The movable beam structure is mounted on a first substrate. A contact structure is provided on the movable beam structure corresponding to the break position of the signal line. The contact structure is spaced apart from the signal line. The second electrode is disposed on the movable beam structure and corresponds to the first electrode. An oxide isolation layer is provided between the contact structure and the movable beam structure.

8. The method for fabricating an RF MEMS switch as described in claim 7, characterized in that, The ground wire includes a first ground wire and a second ground wire, which are spaced apart from each other. The signal line is disposed between the first ground wire and the second ground wire and is parallel to both the first ground wire and the second ground wire.

9. The method for fabricating an RF MEMS switch as described in claim 7, characterized in that, The movable beam structure is configured as a beam structure made of the top layer silicon of SOI, and the movable beam structure has a certain elastic modulus.

Images