MEMS loudspeaker, loudspeaker module, preparation method for MEMS loudspeaker, and electronic device

Abstract

The present application relates to the technical field of loudspeakers, and provides a loudspeaker module, an electronic device, an MEMS loudspeaker and a preparation method therefor. The MEMS loudspeaker comprises a substrate and a suspended elastic body, wherein the substrate is internally provided with a first chamber, and the first chamber passes through the substrate in the thickness direction of the substrate; the suspended elastic body is located on the substrate, a part of the suspended elastic body is located on the first chamber, the suspended elastic body defines a second chamber, and the first chamber is communicated with the second chamber. The MEMS loudspeaker further comprises a vibrating diaphragm and a piezoelectric unit, wherein the piezoelectric unit is located on the suspended elastic body, the piezoelectric unit is arranged on the periphery of the second chamber, and the vibrating diaphragm at least covers the second chamber. The suspended elastic body of the present application is a suspended structure. When the structure is prepared, a sacrificial layer can be used to planarize the suspended elastic body provided with a recess. In this way, the prepared vibrating diaphragm has a uniform thickness and is relatively thin, and the suspended elastic body is relatively thick, so that the MEMS loudspeaker not only has a high sound pressure level output, but also has improved reliability.

Description

MEMS loudspeakers, loudspeaker modules, MEMS loudspeaker fabrication methods, electronic devices

[0001] This application claims priority to Chinese Patent Application No. 202510201849.3, filed on February 21, 2025, entitled "MEMS loudspeaker, loudspeaker module, method for fabricating MEMS loudspeaker, electronic device", the entire contents of which are incorporated herein by reference. Technical Field

[0002] This application relates to the field of loudspeaker technology, and in particular to a loudspeaker module, electronic device, MEMS loudspeaker, and method for fabricating the same. Background Technology

[0003] Piezoelectric MEMS loudspeakers, manufactured using Micro-Electro-Mechanical System (MEMS) technology, produce sound by vibrating the air in the acoustic cavity through a vibrating diaphragm. Furthermore, due to their miniaturization and integration characteristics, MEMS loudspeakers have become a new trend in loudspeaker development.

[0004] Some existing MEMS loudspeakers struggle to simultaneously meet performance requirements such as high sound pressure level output, power consumption, and reliability within their operating frequency band. For example, it is difficult to ensure high reliability of the device while maintaining high sound pressure level output. Summary of the Invention

[0005] This application provides a loudspeaker module, an electronic device, a MEMS loudspeaker, and a method for fabricating the same. The main objective is to provide a MEMS loudspeaker with a thin and uniformly thick diaphragm and a thick suspended elastomer, enabling the MEMS loudspeaker to not only have a high sound pressure level output but also improve the reliability of the device.

[0006] To achieve the above objectives, the embodiments of this application adopt the following technical solutions:

[0007] In one aspect, this application provides a MEMS speaker, which can be used in electronic devices such as headphones, mobile phones, virtual reality devices, or augmented reality devices.

[0008] The MEMS loudspeaker includes a substrate and a suspended elastomer. The substrate has a first chamber that extends through the substrate along its thickness. The suspended elastomer is located on the substrate, with a portion of it situated on the first chamber. The suspended elastomer also surrounds and forms a second chamber. The first chamber and the second chamber are connected, thus forming the acoustic cavity of the MEMS loudspeaker. The MEMS loudspeaker also includes a diaphragm and a piezoelectric unit. The piezoelectric unit is located on the suspended elastomer and is disposed on the outer periphery of the second chamber. The diaphragm covers at least the second chamber, meaning that the opening of the second chamber is closed by the diaphragm, i.e., the diaphragm is a closed-type diaphragm.

[0009] In the MEMS loudspeaker provided in this application, the piezoelectric unit can drive the suspended elastomer and the diaphragm to vibrate, and the vibration causes the air in the first chamber and the second chamber to produce sound, thus becoming a loudspeaker.

[0010] When fabricating the MEMS loudspeaker provided in this application, a suspended elastomer can be fabricated on a substrate, a groove can be formed in the suspended elastomer, a sacrificial layer can be filled in the groove, a diaphragm can be fabricated, the substrate can be etched to form a first chamber, and the sacrificial layer in the groove can be removed to form a second chamber. Since the groove of the suspended elastomer is filled by the sacrificial layer, the diaphragm can be covered on the filled suspended elastomer using processes such as spin coating. The thickness of the diaphragm fabricated in this way is relatively uniform, and there will be no phenomenon where the diaphragm is thicker inside the groove and thinner outside the groove. The reliability of the device is significantly improved, and the consistency of the device diaphragm and the wafer-level manufacturing process can be improved. In addition, the phenomenon that the diaphragm is thinner at the groove step than inside and outside the groove, resulting in a weak point for vibration fracture, is also avoided, and the reliability of the device is improved. Furthermore, the diaphragm fabricated in this application is also relatively thin. When the diaphragm is thinner, it is beneficial to increase the amplitude, increase the output sound pressure level of the device, and also improve the acoustic and electrical performance of the device and reduce response distortion.

[0011] In addition, when making the diaphragm, since the groove is filled by the sacrificial layer, there is no need to consider filling the groove with the diaphragm. This not only allows for the production of a thinner and more uniform diaphragm, but also enables the suspended elastomer to be thicker. A thicker suspended elastomer has better rigidity and support, and the span between the suspended elastomer and the diaphragm can be made larger, resulting in greater vibration displacement of the suspended elastomer and the diaphragm when the device is working.

[0012] Therefore, the MEMS loudspeaker provided in this application has a thinner and more uniform diaphragm, as well as a thicker suspended elastomer, which not only enables the MEMS loudspeaker to have a high sound pressure level output, but also improves the reliability and consistency of the device and optimizes the device performance.

[0013] In one feasible approach, the thickness of the vibrating diaphragm is less than the thickness of the suspended elastomer.

[0014] Thicker suspended elastomers offer better rigidity and support, allowing for a larger span between the suspended elastomer and the diaphragm, resulting in greater vibration displacement of the suspended elastomer and diaphragm during device operation. Thinner diaphragm thickness, on the other hand, helps to increase amplitude and improve the device's output sound pressure level.

[0015] In one feasible approach, the piezoelectric unit has a vibrating diaphragm, and the space enclosed by the piezoelectric unit is filled with the vibrating diaphragm.

[0016] That is, the vibrating membrane not only covers the piezoelectric unit, but also fills the space enclosed by the piezoelectric unit.

[0017] In one feasible process, after etching to obtain the piezoelectric unit, a spin coating process can be used to obtain a vibrating film. The vibrating film covers the piezoelectric unit and fills the space enclosed by the etched piezoelectric unit without the need for additional complex processes. In this way, not only can a thin and uniform vibrating film be obtained, but the preparation process is also simple, which can reduce manufacturing costs.

[0018] In one possible implementation, the piezoelectric unit has a vibrating diaphragm, and the space enclosed by the piezoelectric unit is covered with the vibrating diaphragm, the space enclosed by the piezoelectric unit being in communication with a second chamber.

[0019] In this structure, the diaphragm not only covers the piezoelectric unit, but the space enclosed by the piezoelectric unit is not filled by the diaphragm; rather, the diaphragm covers the space. This allows for further reduction in the thickness of the diaphragm and improves its thickness uniformity, thereby increasing the amplitude, enhancing the device's sound pressure level output, ensuring a flat diaphragm, high process consistency, and improving manufacturability and product yield.

[0020] In one feasible approach, the space enclosed by the piezoelectric unit is filled with a vibrating diaphragm, but no vibrating diaphragm is provided on the piezoelectric unit.

[0021] Since a diaphragm is not placed on the piezoelectric unit, but instead filled in the space, the flatness and thickness uniformity of the diaphragm can be further improved, thus enhancing the reliability of the device.

[0022] In one possible implementation, the MEMS loudspeaker also includes a dielectric layer; the dielectric layer is located between the piezoelectric element and the suspended elastomer, for example, at least a portion of the dielectric layer is located between the piezoelectric element and the suspended elastomer.

[0023] In some feasible fabrication processes, after the groove in the suspended elastic body is filled with a sacrificial layer, when the sacrificial layer is thinned, a depression will appear on the surface of the sacrificial layer. As a result, the thickness uniformity of the fabricated vibrating membrane will be affected by the presence of the depression. However, this application uses a dielectric layer grown on the sacrificial layer to reduce the depression depth and improve the thickness uniformity of the vibrating membrane.

[0024] In one possible implementation, the piezoelectric unit encloses a space whose projection onto the suspended elastomer is located outside the boundary of the second chamber, and the space is connected to the second chamber via a dielectric layer; the wall of the suspended elastomer used to enclose the second chamber is a first wall; the wall of the dielectric layer used to enclose the space is a second wall; and the wall of the piezoelectric unit used to enclose the space is a third wall, with the second and third walls being flush.

[0025] This can be understood as follows: since the projection of the space onto the suspended elastomer is located outside the boundary of the second chamber, that is, the opening of the space is larger than the opening of the second chamber, and because the second wall of the dielectric layer is flush with the third wall of the piezoelectric unit, some parts of the suspended elastomer have stacked piezoelectric units and dielectric layers, while others do not. From a fabrication process perspective, the method for obtaining this example structure is relatively simple and easy to implement. For example, after fabricating the dielectric layer and piezoelectric unit film on the suspended elastomer and sacrificial layer, the dielectric layer can be etched away when etching the piezoelectric unit film.

[0026] Therefore, this structure, by utilizing the dielectric layer, can not only reduce the depth of the sacrificial layer and improve the thickness uniformity of the diaphragm, but also has a simple fabrication process, which can reduce manufacturing costs.

[0027] In one feasible embodiment, the piezoelectric units enclose a space, the projection of which onto the suspended elastomer lies outside the boundary of the second chamber, and the space is connected to the second chamber via a permeable dielectric layer; the wall of the suspended elastomer used to enclose the second chamber is a first wall; the wall of the dielectric layer used to enclose the space is a second wall; and the walls of the multiple piezoelectric units used to enclose the space are a third wall, with the first and second walls being flush.

[0028] Because the second wall of the dielectric layer is flush with the first wall of the suspended elastomer, that is, some parts of the suspended elastomer have stacked piezoelectric units and dielectric layers, while others have dielectric layers but no piezoelectric units. From a fabrication process perspective, for example, when etching the sacrificial layer, the dielectric layer located on the sacrificial layer can be etched away without posing a challenge to the fabrication process.

[0029] In this example, the dielectric layer can not only reduce the depth of the sacrificial layer and improve the thickness uniformity of the vibrating membrane, but also has a relatively simple and easy-to-implement fabrication process.

[0030] In one possible implementation, the second wall surface is flush with the third wall surface; the vibrating diaphragm covers the piezoelectric unit and fills the space enclosed by the piezoelectric unit and the space enclosed by the dielectric layer.

[0031] In this example, a vibrating membrane is not only placed on the stacked dielectric layer and piezoelectric units, but also fills the space within it. In one feasible process, after etching the piezoelectric units and dielectric layer, a spin-coating process can be used to fabricate the vibrating membrane, which covers the piezoelectric units and fills the space enclosed by multiple piezoelectric units and the space within the dielectric layer. No additional complex processes are required. This not only allows for the fabrication of a thin and uniformly thick vibrating membrane, but also simplifies the fabrication process and reduces manufacturing costs.

[0032] In one feasible approach, with the second and third walls flush, the vibrating diaphragm fills the space enclosed by multiple piezoelectric units and the space enclosed by the dielectric layer.

[0033] In this example, the vibrating membrane can be left unplaced on the stacked dielectric layer and piezoelectric unit, instead filling the space, which further makes the thickness of the vibrating membrane uniform.

[0034] In one feasible manner, with the second and third walls flush, the vibrating diaphragm covers multiple piezoelectric units and the space enclosed by the multiple piezoelectric units.

[0035] In this structure, the diaphragm covers multiple piezoelectric units, but the space enclosed by these units is not filled by the diaphragm; rather, it covers the space. This allows for further reduction in the thickness of the diaphragm and improves its thickness uniformity, thereby increasing the amplitude, enhancing the device's sound pressure level output, ensuring a flat diaphragm, and improving process consistency, thus enhancing manufacturability and product yield.

[0036] In one feasible manner, with the first wall surface flush with the second wall surface, the vibrating membrane covers multiple piezoelectric units, fills the space enclosed by the multiple piezoelectric units, and covers the space enclosed by the dielectric layer.

[0037] Because the second wall of the dielectric layer is flush with the first wall of the suspended elastomer, stacked piezoelectric units and a dielectric layer are present on a portion of the suspended elastomer, while a dielectric layer is provided on another portion without piezoelectric units. In this example, the thickness of the diaphragm can be reduced, further improving the uniformity of the diaphragm thickness, increasing the amplitude, enhancing the sound pressure level output of the device, resulting in a flat diaphragm, high process consistency, and improved manufacturability and product yield.

[0038] In one possible implementation, the projection of the space onto the substrate lies within the boundary of the first chamber, and the projection of the second chamber onto the substrate lies within the boundary of the first chamber.

[0039] Since the projection of the second chamber onto the substrate is within the boundary of the first chamber, the suspended elastic body can act as an elastic vibrating plate. The suspended elastic body and the vibrating diaphragm can cause the air in the first and second chambers to vibrate. Furthermore, since the projection of the space onto the substrate is within the boundary of the first chamber, the vibrating diaphragm can have a larger vibrating area while ensuring its stability and reliability, thereby increasing the vibration amplitude and outputting a high sound pressure level.

[0040] In one possible implementation, the substrate includes a silicon layer and a silicon oxide layer disposed on the silicon layer, the silicon oxide layer being stacked between the silicon layer and a suspended elastomer, the suspended elastomer being the silicon layer.

[0041] For example, the MEMS speaker can be fabricated using silicon-on-insulator (SOI), wherein the top silicon layer of the silicon-on-insulator (SOI) can be etched to obtain the suspended elastomer of the example in this application.

[0042] In one feasible approach, the thickness d1 of the vibrating diaphragm is: 1μm≤d1≤3μm.

[0043] In one feasible approach, the thickness of the vibrating diaphragm is 2 μm to 3 μm.

[0044] In one feasible approach, the thickness d2 of the suspended elastomer is: 8μm≤d1≤15μm.

[0045] In one feasible approach, the thickness of the suspended elastomer is 8 μm to 10 μm.

[0046] In one possible implementation, the diaphragm is a flexible diaphragm, the material of which includes at least one of PI, Flexfiner SA, polyurethane, Parylene C, or PVI-3.

[0047] In one feasible approach, the width of the diaphragm is between 1 mm and 5 mm. This increases the sound pressure level output while reducing total harmonic distortion.

[0048] Secondly, this application also provides a speaker module, which includes a housing, a circuit board and a MEMS speaker as described in any of the above implementations. The circuit board and the MEMS speaker are disposed inside the housing, and the MEMS speaker is electrically connected to the circuit board. The housing forms a sound cavity, and a sound hole communicating with the sound cavity is opened on the housing. The MEMS speaker is disposed inside the sound cavity.

[0049] In the speaker module of this application example, since it includes the MEMS speaker in the above implementation, the diaphragm in the MEMS speaker is not only thinner but also has a more uniform thickness, and the suspended elastomer is also thicker. This makes the MEMS speaker not only have a high sound pressure level output, but also improve the reliability and consistency of the device and optimize the device performance. Therefore, the performance of the speaker module can be improved.

[0050] Thirdly, this application also provides an electronic device, which includes a housing and a speaker module as described above. The housing has a sound outlet and a mounting cavity communicating with the sound outlet is formed inside the housing. The speaker module is housed in the mounting cavity and the sound hole of the speaker module communicates with the mounting cavity.

[0051] The electronic device provided in this application includes a speaker module in any of the above implementations. The diaphragm in the MEMS speaker is not only thin but also has a relatively uniform thickness, and the suspended elastomer is also relatively thick. This makes the MEMS speaker not only have a high sound pressure level output, but also improves the reliability and consistency of the device and optimizes the device performance. Therefore, the performance of the electronic device can be improved.

[0052] Fourthly, this application also provides a method for fabricating a MEMS loudspeaker, the method comprising:

[0053] A suspended elastomer with grooves is fabricated on a substrate;

[0054] Fill the grooves with a sacrificial layer;

[0055] Piezoelectric elements are fabricated on a suspended elastomer filled with a sacrificial layer, and the piezoelectric elements are disposed on the outer periphery of the groove.

[0056] A vibrating membrane is prepared, which at least covers the sacrificial layer;

[0057] The substrate and sacrificial layer are etched to form a first chamber within the substrate, and a second chamber is formed surrounded by a suspended elastomer. The first and second chambers are connected.

[0058] When fabricating a MEMS loudspeaker using the method provided in this application, a sacrificial layer is filled into the groove of the suspended elastomer before fabricating the diaphragm, making the surface of the suspended elastomer flat. This allows for a more uniform and thinner diaphragm thickness during fabrication. Since the groove is filled with a sacrificial layer, there is no need to consider filling the groove with the diaphragm. This not only allows for the fabrication of a thinner and more uniform diaphragm but also enables the suspended elastomer to be thicker. A thicker suspended elastomer has better rigidity and support, and the acoustic cavity of the MEMS loudspeaker (the connecting first and second chambers) can be made larger, thereby enabling the device to have a high sound pressure level output.

[0059] In one feasible approach, after filling the groove with a sacrificial layer and before fabricating multiple piezoelectric units, the fabrication method further includes: fabricating a dielectric layer on a suspended elastomer filled with a sacrificial layer.

[0060] When the sacrificial layer is filled into the groove of the suspended elastic body and then the sacrificial layer is thinned, a depression will appear on the surface of the sacrificial layer. As a result, the thickness uniformity of the vibrating membrane will be affected by the presence of the depression. However, this application uses a dielectric layer grown on the sacrificial layer to reduce the depth of the depression and improve the thickness uniformity of the vibrating membrane.

[0061] In one feasible approach, piezoelectric cells are fabricated on a suspended elastomer filled with a sacrificial layer, the piezoelectric cells enclosing a space.

[0062] The vibrating diaphragm includes:

[0063] A vibrating membrane is covered over multiple piezoelectric units and filled in a space; or, a vibrating membrane is covered over multiple piezoelectric units and filled in a space; or, a vibrating membrane is filled in a space.

[0064] In one feasible manner, the material of the sacrificial layer includes at least one of germanium silicon, silicon, or germanium.

[0065] In the fabrication process, when the sacrificial layer is made of at least one of germanium silicon, silicon, or germanium, the growth efficiency and release efficiency are relatively high. Attached Figure Description

[0066] Figure 1 is a schematic diagram of the structure of an electronic device according to an example of this application;

[0067] Figure 2 is a structural schematic diagram of a speaker module provided in an embodiment of this application;

[0068] Figure 3 is a schematic diagram of the structure of a MEMS loudspeaker provided in an embodiment of this application;

[0069] Figure 4 is a schematic diagram of the structure of a MEMS loudspeaker provided in an embodiment of this application;

[0070] Figure 5 is a schematic diagram of the structure of a piezoelectric unit provided in an embodiment of this application;

[0071] Figure 6 is a top view of a MEMS loudspeaker provided in an embodiment of this application;

[0072] Figures 7 to 13 are schematic diagrams of the process structure after each step is completed in the fabrication method of a MEMS loudspeaker provided in the embodiments of this application;

[0073] Figure 14 is a schematic diagram of the structure of a MEMS loudspeaker provided in an embodiment of this application;

[0074] Figure 15 is a schematic diagram of the structure of a MEMS loudspeaker provided in an embodiment of this application;

[0075] Figure 16 is a schematic diagram of the structure of a MEMS loudspeaker provided in an embodiment of this application;

[0076] Figure 17 is a schematic diagram of the structure of a MEMS loudspeaker provided in an embodiment of this application;

[0077] Figure 18 is a schematic diagram of the structure of a MEMS loudspeaker provided in an embodiment of this application;

[0078] Figure 19 is a schematic diagram of the structure of a MEMS loudspeaker provided in an embodiment of this application;

[0079] Figure 20 is a schematic diagram of the structure of a MEMS loudspeaker provided in an embodiment of this application;

[0080] Figure 21 is a schematic diagram of the structure of a MEMS loudspeaker provided in an embodiment of this application;

[0081] Figures 22 to 38 are schematic diagrams of the process structure after each step is completed in the fabrication method of a MEMS loudspeaker provided in the embodiments of this application;

[0082] Figure 39 is a flowchart of a method for fabricating a chip stacking structure according to an embodiment of this application.

[0083] Reference numerals: 100-Electronic device; 100a-Earphone shell; 100b-Speaker module; 101-Sound hole; 102-Sound cavity; 200-Cantilever film layer; 300-Sacrificial layer; 400-SOI wafer; 500-First dielectric layer; 10-Shell; 20-MEMS speaker; 30-Circuit board; 1-Substrate; 2-Suspended elastomer; 3-Piezoelectric unit; 4-Vibrating diaphragm; 5-Dielectric layer; 6-Passivation layer; 31-First electrode; 32-Piezoelectric layer; 33-Second electrode; 51-First dielectric layer; 52-Second dielectric layer; 401-Bottom silicon; 402-Silicon oxide layer; 403-Top silicon. Detailed Implementation

[0084] The following embodiments of this application will be described in conjunction with the accompanying drawings.

[0085] The technical solutions of this application can be applied to various electronic devices that use semiconductor devices. For example, the electronic devices in the embodiments of this application can be mobile phones, tablets, laptops, smart home devices, smart wearable devices (e.g., smartwatches, smart bracelets, smart glasses, smart helmets), virtual reality (VR) glasses, VR helmets, augmented reality (AR) glasses, AR helmets, landline handsets (pickups), medical assistive devices (such as hearing aids), and various headphones (such as wireless or wired headphones) and other devices with speakers.

[0086] This application does not impose any special limitations on the specific form of the above-described electronic device. For ease of explanation, the following description uses the earphone shown in Figure 1 as an example.

[0087] Figure 1 is a perspective view of an electronic device 100 provided in some embodiments of this application. Figure 2 is a structural schematic diagram of the speaker module 100b in the electronic device 100 provided in some embodiments of this application.

[0088] Referring to Figures 1 and 2, the earphone 100 can be a true wireless stereo (TWS) earphone. The earphone 100 includes an earphone shell 100a and a speaker module 100b. The earphone shell 100a has a mounting cavity, and a sound outlet is formed on the earphone shell 100a, communicating with the mounting cavity. The speaker module 100b is housed within the mounting cavity. The sound emitted by the speaker module 100b can be transmitted to the outside through the sound outlet. Some earphones may also include components such as a microphone, gyroscope, voice sensor, Bluetooth, and power supply.

[0089] In some examples, referring to FIG2, the speaker module 100b includes a housing 10, a circuit board 30, and a MEMS speaker 20. The circuit board 30 and the MEMS speaker 20 are disposed within the housing 10, and the MEMS speaker 20 is disposed on the circuit board 30.

[0090] The housing 10 forms a sound cavity 102, and a sound hole 101 is provided on the housing 10. The sound hole 101 communicates with the sound cavity 102 and communicates with the mounting cavity of the earphone housing 100a, thereby communicating with the sound outlet on the earphone housing 100a.

[0091] Circuit board 30 is electrically connected to MEMS speaker 20. Circuit board 30 can send acoustic signals to MEMS speaker 20 to excite MEMS speaker 20 to vibrate and produce sound.

[0092] The speaker module 100b shown in Figure 2 also includes a driver chip, which is electrically connected to the MEMS speaker 20. The driver chip sends acoustic-electric signals to the MEMS speaker 20. The driver chip can be an application-specific integrated circuit (ASIC) chip, which excites the MEMS speaker 20 to output sound energy by sending acoustic-electric signals to the MEMS speaker 20.

[0093] In an electronic device, a housing may be included, on which a sound outlet is provided, and a mounting cavity communicating with the sound outlet is formed inside the housing; the speaker module in the example of Figure 2 is housed in the mounting cavity, and the sound hole of the speaker module is communicating with the mounting cavity.

[0094] To improve the performance of MEMS loudspeakers, such as significantly improving the sound pressure level (SPL) performance while maintaining high reliability and consistency and achieving low levels of total harmonic distortion (THD), this application provides some MEMS loudspeaker structures, as shown below.

[0095] As shown in Figure 3, Figure 3 is a partial structural diagram of a MEMS loudspeaker 20 provided in an embodiment of this application.

[0096] In this example, the MEMS loudspeaker 20 includes a substrate 1 and a suspended elastomer 2; the substrate 1 has a first chamber that extends through the substrate 1 along its thickness direction. For example, in Figure 3, the thickness direction of the substrate 1 is the Z direction, and the first chamber extends through two opposite surfaces of the substrate 1 in the Z direction.

[0097] The suspended elastomer 2 is located on the substrate 1. In some examples, the suspended elastomer 2 can be directly attached to the surface of the substrate 1; or, in other examples, other film structures can be stacked between the suspended elastomer 2 and the substrate 1.

[0098] Continuing with Figure 3, part of the suspended elastomer 2 is located on the first chamber. This can be understood as follows: the suspended elastomer 2 includes a fixed portion Q1 and a cantilever portion Q2 connected together. The fixed portion Q1 is disposed on and connected to the substrate 1, while the cantilever portion Q2 is suspended on the first chamber. In other words, the suspended elastomer 2 is a suspended elastomer structure that functions as an elastomer vibrating plate.

[0099] In addition, the suspended elastic body 2 surrounds and forms a second chamber, and the first chamber is connected to the second chamber. In this way, the connected first chamber and the second chamber form the acoustic cavity of the MEMS loudspeaker 20.

[0100] The suspended elastic body 2 surrounds and forms the second chamber. This can be understood as follows: the second chamber is formed by the suspension elastic body 2, and the upper top surface (or upper curved surface) of the second chamber is basically consistent with the upper surface of the suspension elastic body 2, or the upper top surface of the second chamber is closer to the upper surface of the suspension elastic body 2 than the lower surface of the suspension elastic body 2. As shown in Figure 3, the MEMS loudspeaker 20 also includes a piezoelectric unit 3 and a diaphragm 4. The piezoelectric unit 3 is located on the suspended elastic body 2, and at least part of the diaphragm 4 covers the second chamber. That is, the opening of the second chamber is closed by the diaphragm 4, i.e., the diaphragm 4 is a closed diaphragm.

[0101] The projection of the second chamber onto the substrate 1 lies within the boundary of the first chamber, which allows the suspended elastic body to act as an elastic vibrating plate. The suspended elastic body and the vibrating membrane can cause the air in the first and second chambers to vibrate.

[0102] The working principle of the MEMS loudspeaker 20 shown in Figure 3 is as follows: through the piezoelectric effect, the piezoelectric unit 3 converts electrical energy into mechanical energy. The piezoelectric unit 3 drives the cantilever part Q2 of the suspended elastic body 2 and the diaphragm 4 to vibrate. This vibration causes the air in the first chamber and the second chamber to produce sound.

[0103] In some examples, substrate 1 may be a silicon substrate; in other examples, as shown in FIG4, which is a partial structural diagram of another MEMS loudspeaker 20 provided in an embodiment of this application, substrate 1 may include a silicon layer 11 and a silicon oxide layer 12, with the silicon oxide layer 12 stacked on the silicon layer 11 and the suspended elastomer 2 disposed on the silicon oxide layer 12. In still other examples, substrate 1 may be other substrate structures.

[0104] The material of the suspended elastomer 2 can be at least one of silicon, polycrystalline silicon, germanium, etc., which have elasticity. For example, when the suspended elastomer 2 is a silicon layer, the suspended elastomer 2, silicon oxide layer 12 and silicon layer 11 shown in Figure 4 can be made of silicon-on-insulator (SOI). The following describes how to make this MEMS loudspeaker using silicon-on-insulator (SOI).

[0105] In some feasible structures, as illustrated in Figure 5, a piezoelectric unit 3 may be included. The piezoelectric unit 3 may comprise a stacked first electrode 31, a piezoelectric layer 32, and a second electrode 33, with the piezoelectric layer 32 stacked between the first electrode 31 and the second electrode 33. In other structures, additional film layers may be added to the piezoelectric unit structure shown in Figure 5, such as adding a third electrode and a second piezoelectric layer, with the second piezoelectric layer stacked between the first electrode 33 and the third electrode.

[0106] The diaphragm 4 in this application example can be made of a flexible material, such as at least one of PI, Flexfiner SA, polyurethane, Parylene C, or PVI-3. The diaphragm 4 made of a flexible material has a larger amplitude, which can improve the sound pressure level of the MEMS loudspeaker. Furthermore, the flexibility of the diaphragm 4 is beneficial for broadening the operating frequency bandwidth.

[0107] Figure 6 illustrates one arrangement of the piezoelectric units 3 and the suspended elastomer 2 according to an example of this application. Figure 6 is a top view of multiple piezoelectric units 3 and the suspended elastomer 2. In this example, a second chamber is formed in the suspended elastomer 2. Multiple piezoelectric units 3 are provided, and these units can be spaced apart along the outer periphery of the second chamber. For example, in Figure 6, four piezoelectric units 3 can be spaced apart along the outer periphery of the second chamber. Through the piezoelectric effect, the piezoelectric units 3 can drive the suspended elastomer 2 to vibrate.

[0108] Continuing with Figure 6, the second chamber includes a main chamber S1 and an extension chamber S2, with the extension chamber S2 extending between two adjacent piezoelectric units 3. Alternatively, it can be understood that the extension chamber S2 extends to the corner of the suspended elastic body 2, that is, a chamber is formed at the corner of the suspended elastic body 2. This can further increase the vibration amplitude of the suspended elastic body 2 and improve the sound pressure level of the MEMS loudspeaker.

[0109] In other examples, a piezoelectric unit may be included, for example, multiple piezoelectric units in Figure 6 are connected to form a piezoelectric unit.

[0110] In some structures, the width of the diaphragm 4 (as shown in the Y direction of Figure 6) can be between 1 mm and 5 mm. This can increase the sound pressure level output, reduce total harmonic distortion, and also increase the amplitude. The width of the diaphragm 4 can be understood as the width of the diaphragm 4 above the second chamber.

[0111] Figures 7 to 13 are process structure diagrams of each step in a MEMS loudspeaker fabrication method according to an embodiment of this application.

[0112] As shown in Figure 7, a cantilever film 200 is formed on substrate 1.

[0113] As shown in Figure 8, a groove is formed by etching the cantilever film 200. This groove is used to form the second chamber in the example described above. After etching the cantilever film 200, the suspended elastomer 2 can be formed.

[0114] As shown in Figure 9, a sacrificial layer 300 is formed. For example, the sacrificial layer 300 can be formed by epitaxial growth technology, and the sacrificial layer 300 can at least fill the groove in Figure 8.

[0115] As shown in Figure 10, the structure shown in Figure 9 is flattened to expose the suspended elastomer 2.

[0116] As shown in Figure 11, multiple piezoelectric units 3 are fabricated on the suspended elastomer 2.

[0117] As shown in Figure 12, a vibrating membrane 4 is fabricated, which at least covers the sacrificial layer 300. For example, in the example of Figure 12, the vibrating membrane 4 covers the piezoelectric element 3, the exposed suspended elastomer 2, and the sacrificial layer 300.

[0118] As shown in Figure 13, etching is performed from the back side of the substrate 1 to etch a first chamber within the substrate 1 and to etch a sacrificial layer 300 to form a second chamber within the suspended elastic body, such that the vibrating diaphragm 4 covers the opening of the second chamber, and the first chamber and the second chamber are connected.

[0119] Based on the fabrication methods shown in Figures 7 to 13, it can be seen that before the diaphragm 4 is fabricated as shown in Figure 12, the grooves in the suspended elastomer 2 are filled with the sacrificial layer 300, which flattens the surface of the suspended elastomer 2. In some related fabrication processes, the grooves in the suspended elastomer 2 are not filled, and the diaphragm 4 is directly fabricated. Compared with the above-mentioned related technologies, the diaphragm 4 fabricated in this application has a more uniform thickness and does not exhibit the phenomenon of thicker thickness inside the groove and thinner thickness outside the groove. When the thickness of the diaphragm 4 is more uniform, the vibration mode of the MEMS speaker is less prone to distortion or distortion. Therefore, the MEMS speaker with uniform thickness provided by this application can optimize and improve device performance and enhance the sound quality of the device.

[0120] Furthermore, a more uniform thickness of the diaphragm 4 can improve the reliability of the device. For example, compared to coating the diaphragm in a deeper groove, directly coating the diaphragm without filling the groove with a sacrificial layer results in a thinner diaphragm at the groove steps compared to the inner and outer surfaces of the groove. This makes the groove steps prone to vibration-induced weak points, leading to device failure. However, the diaphragm in this example is essentially on the same plane, eliminating corner breakage weak points and thus improving the device's reliability.

[0121] In some processes, multiple MEMS loudspeakers can be fabricated on a single substrate. Using the examples shown in Figures 7 to 13 above, multiple MEMS loudspeakers can not only improve the reliability of each device, but also improve the consistency of multiple devices, thereby increasing the product yield.

[0122] As shown in Figure 12, the diaphragm 4 fabricated using the method described in this application has a relatively small thickness. In some feasible processes, for example, a flexible diaphragm can be formed by spin-coating a polymer-based flexible material. The thickness of the diaphragm 4 can be from 1 micrometer (μm) to 3 micrometers (μm), for example, from 2 μm to 3 μm. When the diaphragm 4 is thinner, it is beneficial to increase the amplitude of the diaphragm, thereby increasing the output sound pressure level of the MEMS loudspeaker. In planar spin-coating of the diaphragm, there are no limitations such as viscosity, the selection range of diaphragm materials is wider, the diaphragm is flat, and the process consistency is high.

[0123] Continuing with Figure 12, because the groove is filled by the sacrificial layer, while ensuring a small and uniform thickness of the diaphragm 4, the thickness of the suspended elastomer 2 can be larger. For example, the thickness of the suspended elastomer 2 can be 8μm to 15μm, or even 8μm to 10μm. This is because a larger thickness of the suspended elastomer 2 will not affect the thickness and uniformity of the subsequently fabricated diaphragm 4. When the thickness of the suspended elastomer 2 is larger, its rigidity and support are better, the span between the suspended elastomer and the diaphragm can be larger, and the vibration displacement during device operation is greater.

[0124] The thick suspended elastomer 2 can reduce warping caused by process stress, which helps multiple piezoelectric units maintain the same horizontal plane, resulting in high consistency and improved device mode stability, especially for low-frequency response, reducing total harmonic distortion (THD). It also improves the structural mechanical strength of the device and enhances its reliability.

[0125] The acoustic cavity in this application example includes not only a first chamber in the substrate, but also a second chamber in the suspended elastomer. The two interconnected chambers can further enhance the sound pressure level output.

[0126] In some examples, the thickness of the diaphragm 4 is less than the thickness of the suspended elastomer 2. A thicker suspended elastomer provides better rigidity and support, allowing for a larger span between the suspended elastomer and the diaphragm, resulting in greater vibration displacement of the suspended elastomer and the diaphragm during device operation. A thinner diaphragm thickness is beneficial for increasing the amplitude and improving the device's output sound pressure level.

[0127] Figures 14 to 16 are schematic diagrams of the structures of some MEMS loudspeakers in this application.

[0128] As shown in Figure 14, the piezoelectric unit 3 encloses a space, which is connected to the second chamber. The vibrating membrane 4 covers the piezoelectric unit 3 and the space, meaning that the space is not filled with the vibrating membrane 4. This results in a more uniform thickness and a thinner vibrating membrane 4.

[0129] In the example of Figure 14, in addition to the first and second chambers, the space enclosed by the piezoelectric unit is also included, which can increase the volume of the acoustic cavity of the MEMS loudspeaker, thereby enabling a higher sound pressure level output.

[0130] As shown in Figure 15, the piezoelectric unit 3 forms a space that is connected to the second chamber. The vibrating membrane 4 covers the piezoelectric unit 3 and fills the space.

[0131] This can be understood as: the space is filled with a vibrating membrane 4. In some manufacturing processes, after the piezoelectric unit is etched to obtain the space, the vibrating membrane can be prepared by spin coating. This allows the vibrating membrane to not only cover multiple piezoelectric units but also fill the space. This process is simple and easy to implement.

[0132] As shown in Figure 16, the piezoelectric unit 3 encloses a space that communicates with the second chamber, and the vibrating membrane 4 fills the space. In this example, the vibrating membrane 4 is not covered on the piezoelectric unit 3; instead, it fills the space. Since the vibrating membrane is not formed on the piezoelectric unit 3, this further ensures that the thickness of the vibrating membrane 4 is uniform.

[0133] Figure 17 is a schematic diagram of another MEMS loudspeaker provided in an embodiment of this application.

[0134] This example of a MEMS loudspeaker includes a substrate 1, a suspended elastomer 2, a piezoelectric element 3, a diaphragm 4, and a first chamber and a second chamber. The example MEMS loudspeaker also includes a dielectric layer 5 located between the piezoelectric element 3 and the suspended elastomer 2. The dielectric layer 5 can electrically isolate the piezoelectric element 3 and the suspended elastomer 2; for example, the dielectric layer 5 may be made of silicon dioxide.

[0135] In the example of Figure 17, the space enclosed by the piezoelectric unit 3 penetrates the dielectric layer 5. The projection of this space onto the suspended elastic body 2 lies outside the boundary of the second chamber. In this example,

[0136] The walls of the space enclosing the piezoelectric unit 3 (wall M in Figure 17) and the walls of the space enclosing the dielectric layer 5 (wall N in Figure 17) are basically flush.

[0137] Thus, the vibrating diaphragm 4 partially covers the piezoelectric unit 3, and the vibrating diaphragm 4 fills the space enclosed by the piezoelectric unit 3 and the space enclosed by the dielectric layer 5. That is, the vibrating diaphragm 4 partially covers the piezoelectric unit 3, partially covers the suspended elastomer 2, and partially covers the second chamber.

[0138] Figure 18 is a schematic diagram of another MEMS loudspeaker provided in an embodiment of this application.

[0139] In this example, the dielectric layer 5 can be multiple layers, such as the first dielectric layer 51 and the second dielectric layer 52 stacked in Figure 18. The material of the first dielectric layer 51 can be the same as or different from the material of the second dielectric layer 52. For example, the material of the first dielectric layer 51 can be SiO2 or SixNy, and the material of the second dielectric layer 52 can be SiO2 or SixNy.

[0140] By employing multi-layered stacked dielectric layers, the stress of the dielectric layers can be adjusted, reducing phenomena such as warping in the device and optimizing device reliability.

[0141] In the example of Figure 18, the vibrating membrane 4 partially covers the piezoelectric unit 3, and the vibrating membrane 4 fills the space enclosed by the piezoelectric unit 3 and the space enclosed by the dielectric layer 5.

[0142] Figure 19 is a schematic diagram of the structure of another MEMS loudspeaker provided in an embodiment of this application.

[0143] In this example, the wall of the space enclosing the medium layer 5 (wall N in Figure 19) is substantially flush with the wall of the second chamber enclosing the suspended elastomer 2 (end face R in Figure 19).

[0144] The vibrating membrane 4 covers the piezoelectric unit 3 and the space enclosed by the piezoelectric unit 3. The space enclosed by the dielectric layer 5 is not filled with the vibrating membrane 4.

[0145] In this way, the vibrating diaphragm 4 partially covers the piezoelectric unit 3, partially covers the dielectric layer 5, and partially covers the space enclosed by the dielectric layer.

[0146] Figure 20 is a schematic diagram of the structure of another MEMS loudspeaker provided in an embodiment of this application.

[0147] In this embodiment, a portion of the suspended elastomer 2 is covered by the stacked dielectric layer 5 and piezoelectric unit 3, while the area near the second chamber is not covered by the dielectric layer 5 and piezoelectric unit 3.

[0148] The vibrating diaphragm 4 is located on the portion of the suspended elastomer 2 that is not covered by the dielectric layer 5 and the piezoelectric unit 3, and on the second chamber.

[0149] This can be understood as follows: the walls of the space enclosed by the piezoelectric unit 3 (wall M in Figure 20) and the walls of the space enclosed by the dielectric layer 5 (wall N in Figure 20) are basically flush. The vibrating diaphragm 4 fills the space enclosed by the piezoelectric unit 3 and the space enclosed by the dielectric layer 5.

[0150] The diaphragm 4 in this example has a more uniform thickness, which further reduces device vibration distortion or distortion and improves the sound quality of the device.

[0151] Figure 21 is a schematic diagram of the structure of another MEMS loudspeaker provided in an embodiment of this application.

[0152] In this example, the piezoelectric units 3 enclose a space that communicates with the second chamber, which in turn communicates with the first chamber. The walls of the space enclosing the piezoelectric units 3 (wall M in Figure 21) and the walls of the space enclosing the dielectric layer 5 (wall N in Figure 21) are substantially flush. The vibrating diaphragm 4 is located on the piezoelectric units 3 and also in the space. That is, the vibrating diaphragm 4 is not placed in the gaps between the piezoelectric units 3.

[0153] The MEMS loudspeaker illustrated in Figure 21 has a thinner and more uniform diaphragm 4, which increases the vibration displacement of the suspended elastomer and the diaphragm, further reducing device shape distortion or distortion and improving the sound quality of the device.

[0154] Figures 22 to 34 are process structure diagrams of each step in a method for fabricating a MEMS loudspeaker according to an embodiment of this application.

[0155] As shown in Figure 22, a silicon-on-insulator (SOI) wafer 400 is obtained. The silicon-on-insulator (SOI) wafer 400 includes a bottom silicon 401 and a top silicon 403, and a silicon oxide layer 402 stacked between the top silicon 403 and the bottom silicon 401.

[0156] As shown in Figure 23, a first dielectric layer 500 is formed on the silicon SOI wafer 400 on the insulating layer. For example, the first dielectric layer 500 is located on the surface of the top silicon 403.

[0157] Alternatively, in some processes, the top silicon 403 can be thermally oxidized to obtain the first dielectric layer 500.

[0158] As shown in Figure 24, the first dielectric layer 500 is etched to create a groove. This step is used to form a mask layer for subsequent patterning of the suspended elastomer.

[0159] As shown in Figure 25, the top layer silicon 403 is etched.

[0160] For example, dry etching can be used to etch the top silicon 403 to form a groove, with the etching stopping at the silicon oxide layer 402. Alternatively, in other embodiments, the etching can stop on the surface of the bottom silicon 401.

[0161] After completing the steps shown in Figure 25, the patterned top silicon 403 can be the suspended elastomer 2 of the MEMS speaker.

[0162] As shown in Figure 26, a sacrificial layer of 300 is grown.

[0163] For example, an epitaxial growth technique can be used to epitaxially grow a sacrificial layer 300 inside and outside the groove.

[0164] The material of the sacrificial layer 300 may include at least one of Si, Ge, GeSi, etc. Of course, it may also include other materials.

[0165] Using at least one of Si, Ge, GeSi, or other materials as the sacrificial layer material can result in higher growth and release efficiency of the sacrificial layer.

[0166] In some processes, sacrificial layer materials that can withstand process temperatures of around 450 to 1100 degrees Celsius can be selected, such as Si, Ge, and GeSi.

[0167] As shown in Figure 27, the planarization sacrificial layer is 300.

[0168] In some processes, chemical mechanical polishing (CMP) can be used to planarize and thin the sacrificial layer 300, stopping the etching on the surface of the first dielectric layer 500. That is, the first dielectric layer 500 can serve as an etching stop layer, protecting the suspended elastomer 2 located below the first dielectric layer 500, so that the thickness of the suspended elastomer 2 conforms to the design dimensions.

[0169] In one of the selectable processes, as shown in Figure 27, after planarizing the sacrificial layer 300, the first dielectric layer 500 can be removed. For example, the first dielectric layer 500 can be removed using a wet etching technique.

[0170] As shown in Figure 28, the regrown medium layer 5.

[0171] In some processes, as shown in Figures 26 and 27, when planarizing the sacrificial layer 300, because the material of the sacrificial layer 300 is different from the material of the first dielectric layer 500 which serves as the etch stop layer, the top surface of the sacrificial layer 300 will exhibit a butterfly-shaped depression as shown in Figure 27 during the CMP process. The butterfly-shaped depression that occurs during the CMP process can be reduced by using the regrowth dielectric layer 5 shown in Figure 28.

[0172] In some examples, the dielectric layer 5 can serve as an electrical isolation layer for subsequently fabricated piezoelectric units.

[0173] Figure 28 shows the fabrication of a single dielectric layer 5; in other examples, multiple stacked dielectric layers can be fabricated, such as stacked SiO2 layers and SixNy layers, which can reduce stress.

[0174] Figure 29 shows the coating of the piezoelectric unit.

[0175] For example, the first electrode, piezoelectric layer, and second electrode of the piezoelectric unit can be deposited using a thin-film sputtering deposition process.

[0176] As shown in Figure 30, the piezoelectric unit film layer is etched to obtain multiple piezoelectric units 3.

[0177] In some processes, a portion of the dielectric layer 5 not covered by the piezoelectric unit 3 can be etched to expose the sacrificial layer 300, resulting in the structure shown in Figure 30.

[0178] As shown in Figure 31, in some examples, a passivation layer 6 can be deposited on the piezoelectric unit 3 using a coating technique, and the first electrode and the second electrode in the piezoelectric unit can be interconnected by etching and coating techniques.

[0179] The passivation layer 6 can be selected from at least one of Al2O3, TEOS, or SixNy. This passivation layer 6 can isolate the device from external environmental conditions such as moisture, protect the device, and improve the reliability of the device.

[0180] As shown in Figure 32, the flexible vibrating membrane 4 is coated. For example, the flexible vibrating membrane can be coated onto the device surface using a spin coating process.

[0181] The vibrating diaphragm 4 can be made of at least one of the following materials: PI, Flexfiner SA, polyurethane, Parylene C, or PVI-3.

[0182] As shown in Figure 33, the substrate is etched to form the first chamber. In this example, dry etching can be used to etch the bottom silicon 401 from the back of the substrate, stopping at the silicon oxide layer 402. This silicon oxide layer 402 can be understood as an etching stop layer to avoid over-etching and etching the suspended elastomer 2.

[0183] As shown in Figure 34, the sacrificial layer 300 is removed to form a second chamber, thereby releasing the diaphragm 4. For example, wet etching or dry etching can be used to remove the exposed silicon oxide layer 402 and the sacrificial layer 300, exposing the diaphragm 4 and allowing it to be released. In this way, the MEMS loudspeaker of this application example can be fabricated.

[0184] As shown in Figures 31 and 32 above, before the vibrating membrane 4 is made, the groove in the suspended elastomer 2 is filled with the sacrificial layer 300. That is, the surface of the suspended elastomer 2 is smoothed. The vibrating membrane 4 made in this way is not only thin, but also relatively uniform in thickness. For example, the thickness of the vibrating membrane 4 can be 2μm to 3μm.

[0185] In some other manufacturing processes, after completing Figure 29 above, the processes shown in Figures 35 to 38 can be performed.

[0186] As shown in Figure 35, the piezoelectric unit 3 can be etched without etching the first dielectric layer 5, thus preserving the first dielectric layer 5.

[0187] As shown in Figure 36, the flexible vibrating membrane 4 is coated. For example, the flexible vibrating membrane can be coated onto the device surface using a spin coating process.

[0188] As shown in Figure 37, the substrate is etched to form the first chamber. In this example, dry etching can be used to etch the bottom silicon 401 from the back of the substrate, stopping at the silicon oxide layer 402.

[0189] As shown in Figure 38, the sacrificial layer 300 is removed to form a second chamber, and the first dielectric layer 5 is etched further to release the diaphragm 4. For example, wet etching or dry etching can be used to remove the exposed silicon oxide layer 402, the sacrificial layer 300, and the first dielectric layer 5, exposing the diaphragm 4 and allowing it to be released. Thus, the MEMS loudspeaker of this application example is obtained.

[0190] In the two different preparation methods of the above example, after filling the sacrificial layer 300, a first dielectric layer 5 is prepared, such as a silicon dioxide layer. The first dielectric layer 5 can reduce the butterfly-shaped depression of the sacrificial layer 300 and improve the flatness (or planarity) of the subsequently prepared vibrating membrane 4. For example, the depression on the lower surface of the vibrating membrane 4 can not exceed 20 nm, or not exceed 50 nm, or not exceed 100 nm, or not exceed 300 nm.

[0191] Furthermore, the first dielectric layer 5 in this example can also relieve the stress on the sacrificial layer 300, thereby optimizing device performance.

[0192] MEMS loudspeakers with different structures based on the above examples can be fabricated according to the process flow diagram shown in Figure 39.

[0193] Step S1: A suspended elastomer with grooves, also called a cantilever, is fabricated on a substrate.

[0194] For example, the MEMS speaker can be fabricated using a silicon-on-insulator (SOI) wafer, and the suspended elastomer can be the top silicon layer of the SOI.

[0195] Step S2: Fill the groove with a sacrificial layer.

[0196] For example, at least one of materials such as Si, Ge, and GeSi can be selected to prepare the sacrificial layer.

[0197] Step S3: A piezoelectric element is fabricated on a suspended elastomer (also called a cantilever) filled with a sacrificial layer. The piezoelectric element is disposed on the outer periphery of the groove.

[0198] Step S4: Prepare a vibrating membrane, which at least covers the sacrificial layer.

[0199] Since the groove of the suspended elastomer is filled by the sacrificial layer in step S2, i.e., flattening treatment, the resulting vibrating diaphragm is not only thinner but also has a more uniform thickness.

[0200] Step S5: Etch the substrate and sacrificial layer to form a first chamber within the substrate and a second chamber within the suspended elastomer (also called a cantilever). The first and second chambers are connected. This produces a suspended elastomer structure.

[0201] In some alternative fabrication processes, after filling the sacrificial layer within the groove and before fabricating multiple piezoelectric units, the fabrication method further includes: fabricating a dielectric layer on a suspended elastomer filled with the sacrificial layer. This dielectric layer can reduce the butterfly-shaped indentation of the vibrating membrane and further improve the thickness uniformity of the vibrating membrane. For example, the size of the indentation of the vibrating membrane into the second chamber is less than 0.2 μm, the vibrating membrane is flat, and the thickness uniformity is good.

[0202] In some processes, multiple piezoelectric units are fabricated on a suspended elastomer filled with a sacrificial layer, and the multiple piezoelectric units surround a space; wherein, the fabrication of the vibrating membrane includes: covering the multiple piezoelectric units with the vibrating membrane and filling the space with the vibrating membrane; or, covering the multiple piezoelectric units with the vibrating membrane and covering the space with the vibrating membrane; or, filling the space with the vibrating membrane.

[0203] Based on the above-described fabrication method, the MEMS loudspeaker of this application features a closed-loop diaphragm that reduces sound leakage and provides a relatively large amplitude. The flexible polymer diaphragm is soft and lightweight, capable of generating a large amplitude. Furthermore, the suspended elastomer has a large thickness, providing good support and reliability. The second and first chambers can be made larger; for example, the diaphragm size can be increased, with the opening width exceeding 1 mm. Consequently, the MEMS loudspeaker of this application exhibits a significantly improved sound pressure level (SPL) performance, high reliability and consistency, and a low level of total harmonic distortion (THD).

[0204] In the description of this specification, specific features, structures, materials, or characteristics may be combined in any suitable manner in one or more embodiments or examples.

[0205] The above description is merely a specific embodiment of this application, but the scope of protection of this application is not limited thereto. Any variations or substitutions that can be easily conceived by those skilled in the art within the scope of the technology disclosed in this application should be included within the scope of protection of this application. Therefore, the scope of protection of this application should be determined by the scope of the claims.

Claims

1. A MEMS speaker, characterized by, include: A substrate having a first chamber therein, the first chamber penetrating the substrate along its thickness direction; A suspended elastomer is located on the substrate, a portion of the suspended elastomer is located on the first chamber, and the suspended elastomer surrounds and forms a second chamber, with the first chamber and the second chamber communicating with each other; A piezoelectric unit is located on the suspended elastomer and is disposed on the outer periphery of the second chamber; A vibrating diaphragm that at least partially covers the second chamber.

2. The MEMS loudspeaker of claim 1, wherein, The thickness of the vibrating diaphragm is less than the thickness of the suspended elastomer.

3. The MEMS loudspeaker according to claim 1 or 2, characterized in that, The piezoelectric unit has the vibrating diaphragm, and the space enclosed by the piezoelectric unit is covered by the vibrating diaphragm, the space enclosed by the piezoelectric unit communicating with the second chamber; or, The piezoelectric unit has the vibrating membrane, and the space enclosed by the piezoelectric unit is filled with the vibrating membrane; or, The space enclosed by the piezoelectric unit is filled with the vibrating diaphragm.

4. The MEMS loudspeaker of claim 1 or 2, wherein, The MEMS loudspeaker also includes a dielectric layer; The dielectric layer is located between the piezoelectric unit and the suspended elastomer.

5. The MEMS loudspeaker of claim 4, wherein, The piezoelectric unit encloses a space, the projection of the space onto the suspended elastic body is located outside the boundary of the second chamber, and the space penetrates the dielectric layer and communicates with the second chamber; The wall surface of the suspended elastomer used to enclose the second chamber is the first wall surface; The wall of the medium layer that encloses the space is a second wall; The wall surface of the piezoelectric unit that forms the space is a third wall surface; The second wall surface is flush with the third wall surface, or the first wall surface is flush with the second wall surface.

6. The MEMS loudspeaker according to claim 5, characterized in that, When the second wall surface is flush with the third wall surface, The vibrating diaphragm covers the piezoelectric unit and fills the space enclosed by the piezoelectric unit and the space enclosed by the dielectric layer.

7. The MEMS loudspeaker according to claim 5, characterized in that, When the second wall surface is flush with the third wall surface, The vibrating diaphragm fills the space enclosed by the piezoelectric unit and the space enclosed by the dielectric layer.

8. The MEMS loudspeaker according to claim 5, characterized in that, When the second wall surface is flush with the third wall surface, The vibrating diaphragm covers the piezoelectric unit and the space enclosed by the piezoelectric unit.

9. The MEMS loudspeaker according to claim 5, characterized in that, When the first wall surface is flush with the second wall surface, The vibrating diaphragm covers the piezoelectric unit, fills the space enclosed by the piezoelectric unit, and covers the space enclosed by the dielectric layer.

10. The MEMS loudspeaker of any one of claims 5-9, wherein, The projection of the space enclosed by the piezoelectric unit onto the substrate lies within the boundary of the first chamber, and the projection of the second chamber onto the substrate lies within the boundary of the first chamber.

11. The MEMS loudspeaker of any one of claims 1-10, wherein, The substrate includes a silicon layer and a silicon oxide layer disposed on the silicon layer, the silicon oxide layer being stacked between the silicon layer and the suspended elastomer, the suspended elastomer being a silicon layer.

12. The MEMS loudspeaker of any one of claims 1-11, wherein, The thickness d1 of the vibrating diaphragm is: 1μm≤d1≤3μm, and / or the thickness d2 of the suspended elastomer is: 8μm≤d1≤15μm.

13. The MEMS loudspeaker of any one of claims 1-12, wherein, The width of the vibrating diaphragm is 1 mm to 5 mm.

14. A speaker module, characterized by include: case, The circuit board and the MEMS speaker as described in any one of claims 1-13, wherein the circuit board and the MEMS speaker are disposed within the housing, and the MEMS speaker is electrically connected to the circuit board; The housing forms an acoustic cavity, and an acoustic hole communicating with the acoustic cavity is provided on the housing. The MEMS speaker is disposed inside the acoustic cavity.

15. An electronic device, comprising: include: The outer casing has a sound outlet, and an installation cavity communicating with the sound outlet is formed inside the outer casing; The speaker module as claimed in claim 14, wherein the speaker module is housed within the mounting cavity, and the sound hole of the speaker module communicates with the mounting cavity.

16. A method of manufacturing a MEMS speaker, the method comprising: The preparation method includes: A suspended elastomer with grooves is fabricated on a substrate; A sacrificial layer is filled into the groove; A piezoelectric element is fabricated on the suspended elastomer filled with the sacrificial layer, the piezoelectric element being disposed on the outer periphery of the groove; A vibrating diaphragm is prepared, wherein the vibrating diaphragm at least covers the sacrificial layer, and the thickness of the vibrating diaphragm is less than the thickness of the suspended elastomer; The substrate and the sacrificial layer are etched to form a first chamber within the substrate and a second chamber is formed surrounded by the suspended elastomer, wherein the first chamber and the second chamber are connected.

17. The method of claim 16, wherein the MEMS speaker is prepared by: After filling the groove with a sacrificial layer and before fabricating the piezoelectric unit, the fabrication method further includes: A dielectric layer is formed on the suspended elastomer filled with the sacrificial layer.

18. The method of claim 16 or 17, wherein, A piezoelectric element is fabricated on the suspended elastomer filled with the sacrificial layer, the piezoelectric element enclosing a space; The vibrating diaphragm is prepared by: The vibrating membrane is covered on the piezoelectric unit and filled within the space enclosed by the piezoelectric unit; Alternatively, the vibrating membrane may be covered on the piezoelectric unit and the space enclosed by the piezoelectric unit; or the vibrating membrane may be filled within the space enclosed by the piezoelectric unit.

19. The method of claim 16-18, wherein, The material of the sacrificial layer includes at least one of germanium silicon, silicon, or germanium.

Images