A skeleton-diaphragm coupled piezoelectric MEMS loudspeaker and its fabrication method

By employing a skeleton-diaphragm coupling structure, along with a skeleton support structure and an organic sealing layer, the air leakage and acoustic short circuit problems in the low-frequency range of piezoelectric MEMS loudspeakers are solved, achieving efficient low-frequency response and large displacement output, making it suitable for miniature loudspeakers and wearable devices.

CN123123794APending Publication Date: 2026-11-17SICHUAN UNIV
View PDF 0 Cites 0 Cited by

Patent Information

Application Number
CN202611321583.7
Authority / Receiving Office
CN · China
Patent Type
Applications(China)
Current Assignee / Owner
Filing Date
2026-08-28
Publication Date
2026-11-17

AI Technical Summary

Technical Problem

Existing piezoelectric MEMS loudspeakers suffer from severe air leakage, insufficient low-frequency sound pressure, small displacement of the vibration unit, and acoustic short circuits in the low-frequency range, making it difficult to achieve large amplitude output and low distortion characteristics.

Method used

It adopts a skeleton-diaphragm coupling structure, including a skeleton support structure, a piezoelectric drive structure and an organic sealing layer. The skeleton support structure improves the diaphragm stiffness, and the organic sealing layer reduces air leakage, achieving effective acoustic sealing and large displacement output.

Benefits of technology

It improves low-frequency acoustic sealing capability, reduces low-frequency acoustic short-circuit phenomenon, achieves improved low-frequency response performance and large displacement output, reduces harmonic distortion, and is suitable for high-performance miniature loudspeakers and wearable acoustic systems.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure CN123123794A_ABST
    Figure CN123123794A_ABST
Patent Text Reader

Abstract

This invention relates to the field of microelectromechanical systems (MEMS) technology, specifically a skeleton-diaphragm coupled piezoelectric MEMS loudspeaker and its fabrication method. The loudspeaker includes a substrate layer, a skeleton support structure, a piezoelectric driving structure, an organic sealing layer, and a acoustic cavity structure. The piezoelectric driving structure comprises a lower electrode layer, a piezoelectric functional layer, and an upper electrode layer, which generates an inverse piezoelectric effect by applying an alternating driving voltage to drive the diaphragm to bend and vibrate. The skeleton support structure is longitudinally connected to the piezoelectric driving structure to improve the overall out-of-plane stiffness of the diaphragm and achieve uniform transmission of driving force. The organic sealing layer covers the skeleton and the piezoelectric driving structure and is sealed to the surrounding fixed area, forming a front and rear acoustic cavity isolation structure. This invention, through the coupling effect of the skeleton and the organic sealing layer, reduces the system's equivalent stiffness while ensuring airtightness, achieving low-frequency large displacement, high sound pressure output, and low distortion operation, effectively improving the low-frequency response performance of the MEMS loudspeaker.
Need to check novelty before this filing date? Find Prior Art

Description

Technical Field

[0001] This invention relates to the field of microelectromechanical systems (MEMS) technology, specifically to a skeleton diaphragm coupled piezoelectric MEMS loudspeaker and its fabrication method. Background Technology

[0002] MEMS loudspeakers are miniature acoustic devices manufactured using microelectromechanical systems (MEMS) technology. They achieve electroacoustic energy conversion through coupling of microscale mechanical structures with electrical drive units. Compared to traditional dynamic loudspeakers, MEMS loudspeakers offer advantages such as small size, light weight, ease of array integration, low power consumption, and suitability for semiconductor mass production. Therefore, they have broad application prospects in TWS earphones, smartphones, AR / VR devices, hearing aids, wearable electronics, and ultra-thin terminal devices. Currently, MEMS loudspeakers mainly utilize electrostatic, electromagnetic, thermoacoustic, and piezoelectric technologies. Among these, piezoelectric MEMS loudspeakers are gradually becoming a key research and industrialization focus due to their high driving efficiency, low power consumption, good structural stability, and ease of integration with CMOS-compatible processes.

[0003] In recent years, MEMS loudspeakers based on piezoelectric thin films such as AlN, AlScN, and PZT have gradually become a research hotspot. High sound pressure output can be achieved by fabricating a piezoelectric driving layer and an elastic diaphragm on an SOI substrate. Traditional piezoelectric MEMS loudspeakers typically employ cantilever beam structures, nested ring structures, or spring-mass structures. Sound is generated by the piezoelectric layer undergoing bending deformation under electric field excitation, thereby driving the vibrating unit to vibrate. However, due to the size limitations of MEMS structures and the limited strain capacity of the piezoelectric materials themselves, the displacement of the vibrating unit is small, making it difficult to push a sufficient volume of air to achieve high sound pressure output. This is especially problematic in the low-frequency region where severe sound pressure attenuation is more likely to occur.

[0004] Existing open-type cantilever piezoelectric MEMS loudspeakers typically employ multiple piezoelectric cantilever beams to drive the central vibrating unit. This type of structure offers advantages such as simple manufacturing and fast response. However, due to its reliance on air viscosity to create a limited acoustic seal, it suffers from significant air leakage in the low-frequency range, resulting in insufficient low-frequency sound pressure level output. Furthermore, the displacement capability of this type of structure is limited by the strain limit of the piezoelectric material, making it difficult to achieve large amplitude output. Another common structure is the spring-mass type MEMS loudspeaker, which improves low-frequency response by lowering the system's resonant frequency. While this structure can achieve higher amplitudes in specific frequency bands, the lack of an effective sealing structure between the front and rear cavities of the vibrating unit easily leads to severe acoustic short circuits at low frequencies, causing rapid attenuation of the low-frequency sound pressure level. Furthermore, some existing technologies combine flexible sealing films with MEMS vibration structures, such as using materials like Parylene-C or PDMS to create flexible sealing layers to enhance airtightness. While these solutions can suppress acoustic short circuits to some extent, the flexible film layer typically covers the vibration area directly, increasing the overall stiffness of the vibration system and consequently reducing the effective displacement and acoustic conversion efficiency of the vibration unit. Simultaneously, the flexible film is prone to generating additional tensile stress under large displacement operating conditions, causing nonlinear deformation and harmonic distortion in the vibration unit. Therefore, how to ensure large displacement output while simultaneously maintaining structural stability, low-frequency acoustic sealing capability, and low distortion characteristics has become a crucial technical problem that urgently needs to be solved in the current field of piezoelectric MEMS loudspeakers. Therefore, in response to the above situation, there is an urgent need to develop a skeleton-diaphragm coupled piezoelectric MEMS loudspeaker and its fabrication method to overcome the shortcomings in current practical applications. Summary of the Invention

[0005] The purpose of this invention is to provide a skeleton diaphragm coupled piezoelectric MEMS loudspeaker and its fabrication method, so as to solve the problems mentioned in the background art.

[0006] To achieve the above objectives, the present invention provides the following technical solution:

[0007] A skeleton diaphragm coupled piezoelectric MEMS loudspeaker includes a substrate layer and a skeleton support structure disposed on the substrate layer, and also includes a piezoelectric driving structure and an organic sealing layer;

[0008] The substrate layer has a back cavity structure inside, which extends upward to the buried oxide insulating layer and communicates with the space below the skeleton support structure.

[0009] The skeleton support structure includes one or more cantilever support units, one end of each cantilever support unit is fixed to the buried oxide insulating layer above the substrate, and the other end extends to other areas;

[0010] The piezoelectric drive structure is located in the central region of the skeleton support structure and is coupled to the inner end of the cantilever support unit.

[0011] The organic sealing layer covers the skeleton support structure and the piezoelectric drive structure, and its periphery is sealed to the outer fixing structure above the substrate layer, thus isolating the front acoustic cavity and the rear acoustic cavity of the speaker.

[0012] As a further aspect of the present invention: the piezoelectric driving structure includes a lower electrode layer, a piezoelectric functional layer, and an upper electrode layer;

[0013] The lower electrode layer, the piezoelectric functional layer, and the upper electrode layer are sequentially stacked on the skeleton support structure;

[0014] As a further embodiment of the present invention: the skeleton support structure is one or more of the following: a folded beam structure, a ring-shaped reinforcing structure, or a multi-level bifurcated structure;

[0015] The skeleton support structure is made of monocrystalline silicon, polycrystalline silicon, SiNx or metal materials.

[0016] As a further aspect of the present invention: the organic sealing layer is prepared using Parylene-C, PI, PDMS, SU-8 or a composite flexible polymer material;

[0017] The Young's modulus of the organic sealing layer is lower than that of the Young's modulus of the material of the skeleton support structure;

[0018] The organic sealing layer is connected to the skeleton support structure through an adhesive layer, surface bonding, or integrated deposition process.

[0019] As a further aspect of the present invention: the skeleton support structure forms physical gaps through etching, so that the effective unfolded length of the drive beam is greater than its straight span.

[0020] As a further aspect of the present invention: the organic sealing layer is connected to the top of the skeleton support structure and the piezoelectric drive structure. The organic sealing layer is suspended above the physical gap between adjacent structures or fills the physical slit, and is connected to the organic membrane body above the slit.

[0021] A method for fabricating a skeleton diaphragm coupled piezoelectric MEMS loudspeaker includes the following steps:

[0022] S1. Composite substrate preparation: Select a composite wafer with a substrate layer, a buried oxide insulating layer and a top silicon device layer, and sequentially prepare a lower electrode layer, a piezoelectric functional layer and an upper electrode layer on the surface of the top silicon device layer.

[0023] S2. Piezoelectric drive stack construction: The upper electrode layer, piezoelectric functional layer and lower electrode layer are patterned through multiple photolithography and etching processes to form multiple independent piezoelectric drive units.

[0024] S3. Skeleton structure definition: The top silicon device layer is patterned by etching process to define the skeleton support structure and its physical gaps. The skeleton support structure is set in the central region and coupled to multiple piezoelectric drive units.

[0025] S4. Organic sealing layer integration: A low-modulus organic sealing layer is integrated above the skeleton support structure, allowing the organic sealing layer to cross physical gaps and be sealed to the surrounding fixed area, forming a front and rear acoustic cavity isolation structure.

[0026] S5. Structure Release and Acoustic Cavity Construction: A back cavity structure is formed in the substrate layer using a substrate back etching process, and the skeleton support structure is released, making the skeleton support structure movable.

[0027] As a further aspect of the present invention: in step S4, the integration of the organic sealing layer can employ an ex-situ transfer process, specifically including:

[0028] Pre-fabricated organic sealing films were prepared on an auxiliary substrate;

[0029] Surface activation treatment was performed on the front side of the MEMS structure and the surface of the pre-fabricated organic sealing film.

[0030] The prefabricated organic sealing film is aligned and covered over the skeleton support structure, so that the prefabricated organic sealing film spans the physical gap and is bonded to the top surface of the skeleton support structure and the outer fixing structure.

[0031] Remove the auxiliary substrate to form a skeleton diaphragm coupling structure.

[0032] As a further aspect of the present invention: after step S2, a passivation layer and interconnect pad preparation step is included, specifically including:

[0033] Deposit a passivation layer on the device surface;

[0034] Contact holes are formed in the passivation layer using photolithography and etching processes to expose the surfaces of the upper and lower electrode layers, respectively.

[0035] A metal layer is deposited and patterned to form metal interconnect pads, which are electrically connected to the upper and lower electrode layers through contact holes, respectively.

[0036] As a further aspect of the present invention: in step S5, the structural release and acoustic cavity construction specifically include:

[0037] Deep reactive ion etching is used to remove the substrate silicon material from the back side of the substrate layer to form a back cavity structure.

[0038] Hydrofluoric acid solution is used to remove the exposed buried oxygen insulation layer, allowing the skeleton support structure to be completely released.

[0039] Compared with the prior art, the beneficial effects of the present invention are:

[0040] The skeleton support structure is set in the central area of ​​the piezoelectric drive structure and coupled to multiple cantilever support units to form in-plane support for the diaphragm, thereby improving the overall out-of-plane stiffness of the diaphragm and effectively suppressing local collapse and nonlinear deformation.

[0041] An organic sealing layer covers the skeleton structure and is sealed to the outer fixed area, isolating the front and rear acoustic cavities, thereby significantly reducing air leakage, improving low-frequency acoustic sealing capability, and effectively suppressing low-frequency acoustic short circuits.

[0042] The organic sealing layer seals physical gaps to suppress acoustic short circuits, and combined with the skeleton structure to uniformly transmit driving force, it achieves low-frequency sound pressure level flattening and effectively improves low-frequency response performance.

[0043] The skeleton support structure improves the overall stiffness of the diaphragm and suppresses local nonlinear deformation, thereby reducing harmonic distortion under large amplitude operating conditions and improving acoustic output quality.

[0044] The high compliance structure of the skeleton support structure enables large displacement output, while the organic sealing layer maintains low out-of-plane equivalent stiffness while ensuring air sealing capability. The skeleton structure further improves the overall stability, thus balancing large displacement output capability and structural reliability.

[0045] The device is fabricated using microelectromechanical systems (MEMS) technology and features small size and easy array integration, making it suitable for high-performance miniature loudspeakers and wearable acoustic systems. Attached Figure Description

[0046] Figure 1 This is a schematic diagram of the cross-sectional structure of the device in Embodiment 1 of the present invention.

[0047] Figure 2 This is a schematic diagram of the exploded structure of the device assembly in Embodiment 1 of the present invention.

[0048] Figure 3 This is a schematic diagram of the device vibration modes in Embodiment 1 of the present invention.

[0049] Figure 4 This is a simulation curve of the mid-to-low frequency sound pressure level in Embodiment 1 of the present invention.

[0050] Figure 5 This is a schematic diagram of the preparation method according to an embodiment of the present invention.

[0051] Figure 6This is a schematic diagram of the front structure of the annular reinforced frame in Embodiment 2 of the present invention.

[0052] Figure 7 This is a schematic diagram of the front structure of the multi-level branched skeleton in Embodiment 3 of the present invention.

[0053] Figure 8 This is a schematic diagram of the front structure of another multi-level bifurcated skeleton in Embodiment 3 of the present invention.

[0054] In the diagram: 100 - substrate layer, 200 - buried oxide insulating layer, 300 - skeleton support structure, 400 - lower electrode layer, 500 - piezoelectric functional layer, 600 - upper electrode layer, 700 - passivation layer, 800 - metal interconnect pad, 900 - organic sealing layer. Detailed Implementation

[0055] The technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0056] The specific implementation of the present invention will be described in detail below with reference to specific embodiments.

[0057] Please see Figures 1-8 The present invention provides a skeleton diaphragm coupled piezoelectric MEMS loudspeaker, including a substrate layer 100, a buried oxygen insulating layer 200, a piezoelectric driving structure, a skeleton support structure 300, an organic sealing layer 900, and a acoustic cavity structure.

[0058] A skeleton support structure 300 is disposed on the substrate layer 100 and the buried oxide insulating layer 200, and is coupled to multiple cantilever support units. The skeleton support structure 300 is used to improve the overall out-of-plane stiffness of the vibrating part, suppress local collapse and nonlinear deformation, and achieve uniform transmission of driving force within the vibration region. The skeleton support structure 300 can be one or more combinations of a folded beam structure, a ring-reinforced structure, or a multi-level bifurcated structure, and is made of monocrystalline silicon, polycrystalline silicon, SiNx, or metallic materials, with a thickness of 5μm to 80μm. Since this invention does not require the use of air viscosity in a solid structure for sealing, the shape of the skeleton support structure 300 can be freely selected and designed.

[0059] A piezoelectric driving structure is disposed on the skeleton support structure 300 and includes a lower electrode layer 400, a piezoelectric functional layer 500, and an upper electrode layer 600. By applying an alternating driving voltage, the piezoelectric functional layer 500 undergoes the inverse piezoelectric effect, thereby driving the diaphragm to produce bending vibration. The piezoelectric functional layer 500 is preferably made of AlN, AlScN, or PZT material. The shapes of each layer of the piezoelectric driving structure are similar to the moving parts in the skeleton support structure 300, and are used to drive the moving parts in the skeleton support structure 300.

[0060] An organic sealing layer 900 covers the skeleton support structure 300 and the piezoelectric drive structure, and is sealed to the surrounding fixed area to form a front and rear acoustic cavity isolation structure, thereby reducing air leakage and suppressing low-frequency acoustic short circuits. The organic sealing layer 900 is made of Parylene-C, PI, PDMS, SU-8, or a composite flexible polymer material, with a thickness ranging from 0.5 μm to 20 μm. The organic sealing layer 900 is connected to the skeleton support structure 300 and the piezoelectric drive structure through adhesive bonding, surface bonding, or integrated deposition processes, forming a coordinated vibration structure with the skeleton support structure 300 and the piezoelectric drive structure. The skeleton support structure 300 only supports a localized area of ​​the organic sealing layer 900, allowing the organic sealing layer 900 to maintain low out-of-plane equivalent stiffness while ensuring airtightness, thus balancing low-frequency large displacement output capability.

[0061] A back cavity structure is provided inside the substrate layer 100 and the buried oxide insulating layer 200. The back cavity is connected to the lower part of the skeleton support structure 300 to provide the air buffer space required for diaphragm vibration. The back cavity is formed by deep reactive ion etching process.

[0062] This invention achieves low-frequency large displacement, high sound pressure output, and low distortion operation by coupling the skeleton support structure 300, the piezoelectric drive structure, and the organic sealing layer 900, while ensuring air sealing capability.

[0063] Example 1:

[0064] Please see Figure 1 and Figure 2 This embodiment provides a skeleton diaphragm coupled piezoelectric MEMS loudspeaker, which includes, from bottom to top, a substrate layer 100, a buried oxygen insulating layer 200, a skeleton support structure 300, a lower electrode layer 400, a piezoelectric functional layer 500, an upper electrode layer 600, a passivation layer 700, a metal interconnect pad 800, and an organic sealing layer 900.

[0065] The substrate layer 100 is fabricated using the bottom silicon layer of an SOI substrate, with a thickness of 450 μm. A back cavity structure is provided within the substrate layer 100, which is connected to the lower part of the framework support structure 300, providing the air buffer space required for the vibration of the framework and diaphragm. The back cavity is formed using a deep reactive ion etching process, which achieves a highly vertical sidewall morphology, thereby maximizing the effective volume of the acoustic cavity and ensuring that the diaphragm has sufficient vibration space.

[0066] A buried oxide insulating layer 200 is disposed above the substrate layer 100, and is made of silicon dioxide with a thickness of 1 μm. As a buried oxide insulating layer of the SOI substrate, the buried oxide insulating layer 200 serves two purposes: firstly, it achieves electrical isolation between the substrate layer 100 and the device layer above it; secondly, it acts as an etching stop layer during deep reactive ion etching, ensuring precise control over the thickness of the framework support structure 300, thereby improving the consistency and yield of the device's mechanical properties.

[0067] The skeleton support structure 300 is disposed above the buried oxide insulating layer 200 and is fabricated using the top single-crystal silicon device layer on an SOI substrate with a thickness of 5 μm. The skeleton support structure 300 is processed into a serpentine folded structure, with physical gaps of 2 μm to 5 μm width formed by anisotropic etching, increasing the effective unfolded length of a single drive beam to 2.5 to 4 times its linear span. By adopting a folded beam structure, the effective length of the beam can be significantly extended within a limited chip area, thereby greatly improving the mechanical compliance of the structure. This allows the stress generated by piezoelectric actuation to be efficiently converted into a large vertical displacement output, effectively improving the vibration amplitude and acoustic output capability of the diaphragm.

[0068] The lower electrode layer 400 is disposed above the skeleton support structure 300 and includes a 20 nm thick titanium seed layer and a 100 nm thick platinum electrode layer. The titanium seed layer is used to improve the adhesion between the platinum electrode and the silicon substrate and prevent the electrode from falling off or delaminating during long-term vibration. The platinum electrode, as the lower electrode, together with the upper electrode, forms a piezoelectrically driven electric field application circuit, which has excellent chemical stability and conductivity.

[0069] A piezoelectric functional layer 500 is disposed above the lower electrode layer 400 and is fabricated using a PZT piezoelectric ceramic thin film with a thickness of 1 μm. A 150 nm thick SRO (strontium ruthenium ruthenium oxide) / LNO (lanthanum nickel oxide) buffer layer is placed below the piezoelectric functional layer 500 to improve the crystallinity and orientation of the PZT thin film, thereby enhancing the piezoelectric coefficient and electromechanical conversion efficiency of the piezoelectric material. When an alternating driving voltage is applied above and below the piezoelectric functional layer 500, the piezoelectric functional layer 500 generates an inverse piezoelectric effect, undergoing in-plane stretching deformation, which in turn drives the framework support structure 300 to bend and vibrate, realizing the conversion of electrical energy into mechanical energy.

[0070] The upper electrode layer 600 is disposed above the piezoelectric functional layer 500, and is made of platinum material with a thickness of 100 nm. The upper electrode layer 600 is defined as the input terminal of the driving signal. In conjunction with the lower electrode layer 400, it forms a driving electric field at both ends of the piezoelectric functional layer 500, ensuring that the electric field can act uniformly on the piezoelectric material and improve the driving efficiency.

[0071] The passivation layer 700 covers the entire surface of the device structure and is fabricated using PECVD silicon dioxide, with a thickness of 200 nm. The passivation layer 700 is used to protect the underlying piezoelectric functional layer 500, upper electrode layer 600, and lower electrode layer 400 from moisture, ions, and mechanical damage from the external environment, thereby improving the long-term reliability and stability of the device and achieving electrical isolation between the electrodes of each layer.

[0072] Metal interconnect pads 800 are disposed above the passivation layer 700 and include a 20nm thick chromium adhesion layer and a 200nm thick gold conductive layer. The metal interconnect pads 800 are electrically connected to the upper electrode layer 600 and the lower electrode layer 400 respectively through contact holes in the passivation layer 700, for use as input of external drive signals. Using gold as the pad material provides excellent conductivity and oxidation resistance, facilitating subsequent wire bonding and package integration.

[0073] An organic sealing layer 900, fabricated from a PDMS organic thin film with a thickness of 5 μm to 10 μm and a Young's modulus of 1.5 MPa to 3.0 MPa, is disposed on the top layer of the device structure. The organic sealing layer 900 spans and completely seals all physical gaps between the skeleton support structure 300 and the piezoelectric drive structure, forming a sealed connection with the surrounding fixed area. This isolates the air cavities before and after the diaphragm, effectively preventing air leakage through physical gaps, suppressing low-frequency acoustic short circuits, and significantly improving the low-frequency sound pressure output performance of the speaker. Because the Young's modulus of the organic sealing layer 900 is much lower than that of the silicon-based skeleton support structure 300, the mechanical load it generates during servo operation accounts for a lower proportion of the total system stiffness. Therefore, while achieving acoustic sealing, it does not significantly suppress the diaphragm displacement output, balancing low-frequency sealing capability and large displacement output characteristics. The organic sealing layer 900 and the skeleton support structure 300 are surface-bonded through oxygen plasma treatment, ensuring reliable connection and good process compatibility.

[0074] In this embodiment, when a drive signal with a peak-to-peak value of 20V and a frequency of 100Hz to 20kHz is input to the metal interconnect pad 800, the piezoelectric functional layer 500 generates an inverse piezoelectric effect, causing the skeleton support structure 300 to bend and vibrate. Due to the high compliance of the skeleton support structure 300, the stress generated by the drive unit is efficiently converted into vertical lifting motion, allowing the 5.557mm² diaphragm to achieve a peak displacement of 12μm on one side under 4Vpp drive. Simultaneously, the organic sealing layer 900 vibrates synchronously with the skeleton support structure 300, completely sealing physical gaps and preventing acoustic short-circuit leakage of air within these gaps. Through the coupling effect of the above structures, this embodiment achieves highly efficient acoustic energy conversion. The 5.557mm² diaphragm, under 4Vpp drive, exhibits a sound pressure level output greater than 95dB at 1kHz and below, demonstrating excellent low-frequency response performance.

[0075] The preparation method of this embodiment is described in detail below, including the following steps:

[0076] Step S1, composite substrate preparation. A PZT-SOI composite wafer is selected, with its structure from bottom to top as follows: a 450μm thick substrate layer (100 silicon), a 1μm thick buried oxide insulating layer (200 silicon dioxide), and a 5μm thick device layer top silicon. On the top silicon surface, a 300nm silicon oxide layer, a 20nm titanium seed layer, a 100nm platinum lower electrode layer (400), a 150nm SRO / LNO buffer layer, a 1μm PZT piezoelectric functional layer (500), and a 100nm platinum upper electrode layer (600) are sequentially prepared by physical vapor deposition or sol-gel method. By using an SOI substrate, the thickness of the device layer can be precisely controlled, ensuring the consistency of the mechanical properties of the framework support structure (300); while the high-quality preparation of the PZT thin film provides a good material foundation for subsequent piezoelectric actuation.

[0077] Step S2, piezoelectric drive stack construction. First, photolithography is performed using a first set of photomasks to pattern the 100nm thick platinum upper electrode layer 600 using ion beam etching, forming the drive region. Next, photolithography is performed using a second set of photomasks to etch and remove the 1μm thick piezoelectric functional layer 500 and the 150nm thick buffer layer, exposing the surface of the lower electrode layer 400. Then, photolithography is performed using a third set of photomasks to etch and remove the 100nm thick platinum lower electrode layer 400 and the 20nm thick titanium seed layer, achieving electrical isolation between each drive branch. Through multiple photolithography and etching processes, the shape and size of the piezoelectric drive unit can be precisely defined, ensuring the electrical independence between each drive unit and providing a foundation for subsequent signal input and drive control.

[0078] Step S3: Graphical definition of the skeleton support structure 300. First, photolithography is performed using a fourth set of photomasks. A 200nm thick passivation layer 700 and a 100nm thick silicon oxide layer are etched using reactive ion etching. Then, deep reactive ion etching is used to etch down to the 5μm thick top silicon of the device layer until the buried oxide insulating layer 200 is reached, thus defining the skeleton support structure 300 and its physical gaps. This step allows for precise control of the shape, width, and gap dimensions of the folded beams, ensuring that the mechanical compliance and vibration characteristics of the skeleton support structure 300 meet design requirements.

[0079] Step S4: Integration of the low-modulus organic sealing layer. The integration of the organic sealing layer 900 includes in-situ film deposition and ex-situ transfer processes, with the appropriate integration method selected based on material characteristics. The in-situ film deposition process uses vacuum phase deposition, solution spin coating, or spray coating to directly form a conformal or quasi-conformal organic sealing film on the topological surface of the framework support structure 300. The ex-situ transfer process prefabricates the organic sealing film on an auxiliary substrate and integrates the prefabricated film across the gap above or below the physical seam through transfer, bonding, and selective bonding processes. This embodiment uses the ex-situ transfer process:

[0080] First, a low Young's modulus PDMS adhesive is spin-coated onto an auxiliary glass substrate and cured at 100°C for 60 minutes to form a uniform organic film with a thickness of 8 μm (i.e., the organic sealing layer 900). Then, the front side of the MEMS structure and the surface of the PDMS film are treated with oxygen plasma to increase surface energy and enhance molecular bonding strength. Next, the PDMS film is precisely aligned and placed over the framework support structure 300, allowing the PDMS film to span the physical gaps of the folded beam and achieve irreversible bonding with the inactive area at the top of the folded beam. Finally, the auxiliary glass substrate is removed, ultimately forming the framework diaphragm coupling structure. The use of an ex-situ transfer process to integrate the organic sealing layer 900 avoids contamination and damage to the MEMS structure caused by direct deposition of organic materials, while also allowing precise control of the thickness and mechanical properties of the organic sealing layer 900, ensuring consistent acoustic performance of the device.

[0081] Step S5, Structure Release and Acoustic Cavity Construction. A back-side etching process is used to remove a 450μm thick silicon substrate layer 100, forming a back cavity structure. The exposed buried oxide insulating layer 200 is then removed using hydrofluoric acid solution, making the skeleton support structure 300 movable. This process, which defines the skeleton structure on the front and releases the cavity on the back, ensures the dimensional accuracy and surface quality of the skeleton support structure 300, while simultaneously achieving complete structure release and ensuring its free vibration.

[0082] Example 2:

[0083] Please see Figure 6The difference between this embodiment and Embodiment 1 is that the skeleton support structure 300 adopts a ring-shaped reinforced structure.

[0084] The annular reinforced frame structure includes a central connecting region and an annular support portion surrounding the central connecting region, with multiple cantilever support units connected to the annular support portion. The annular support portion and the annular reinforced frame are integrally manufactured from the same material and have the same thickness.

[0085] By adopting a ring-shaped reinforced structure, the overall out-of-plane stiffness of the central vibration region can be improved, allowing the organic sealing layer 900 to maintain a more uniform displacement distribution during vibration. This prevents local collapse or warping deformation in the vibration center region and effectively reduces vibration distortion. Simultaneously, the ring-shaped support and the organic sealing layer 900 form a surface contact support, effectively reducing local stress concentration and improving the reliability of the device during long-term cyclic operation.

[0086] In this embodiment, the organic sealing layer 900 and the skeleton structure can be connected by bonding, adhesive bonding, or integral molding. When the device is working, the piezoelectric drive structure drives the annular support to vibrate as a whole, thereby driving the organic sealing layer 900 to move synchronously and realize air vibration to generate sound. Since the annular skeleton structure can improve the overall stiffness of the central area, it can effectively reduce harmonic distortion and improve low-frequency sound pressure output performance under low-frequency, high-amplitude operating conditions.

[0087] The remaining structures and preparation methods of this embodiment are basically the same as those of Embodiment 1, and will not be repeated here.

[0088] Example 3:

[0089] Please see Figure 7 and Figure 8 The difference between this embodiment and Embodiment 1 is that the central frame support structure 300 adopts a multi-level bifurcated frame structure.

[0090] This multi-level branching skeleton structure includes a central support and multiple branch support beams extending outwards, with intervals formed between the branch support beams. The branch support beams radiate outwards from the central support and branch out in multiple levels during their extension, forming a tree-like topology.

[0091] Compared to the skeleton support structure 300 in Embodiment 1, the multi-level branched structure in this embodiment can further reduce the equivalent mass of the central region and improve the stress dispersion capability of the vibration region, allowing the driving force to be transmitted more evenly to the entire vibration region. During device operation, each branch support beam can provide multi-point support for the organic sealing layer 900, thereby reducing local wrinkles and asymmetric vibrations generated by the organic sealing layer 900 under large amplitude conditions, further improving the stability and linearity of vibration. In addition, due to the large opening area between the skeletons, the additional air damping can be reduced, improving the high-frequency response capability of the device and expanding the effective operating frequency band of the loudspeaker.

[0092] The remaining structures and preparation methods of this embodiment are basically the same as those of Embodiment 1, and will not be repeated here.

[0093] It should be noted that, in this invention, although the specification describes the embodiments, not every embodiment contains only one independent technical solution. This way of describing the specification is only for clarity. Those skilled in the art should regard the specification as a whole. The technical solutions in each embodiment can also be appropriately combined to form other embodiments that can be understood by those skilled in the art.

Claims

1. A skeleton-diaphragm coupled piezoelectric MEMS loudspeaker, comprising a substrate layer (100), a buried oxide insulating layer (200), and a piezoelectric driving structure, characterized in that, It also includes a skeleton support structure (300) and an organic sealing layer (900); The substrate layer (100) has a back cavity structure inside, which extends upward to the buried oxide insulating layer (200) and communicates with the space below the skeleton support structure (300). The skeleton support structure (300) includes one or more cantilever support units, one end of each cantilever support unit is fixed to the buried oxide insulating layer (200) above the substrate layer (100), and the other end extends to other areas; The piezoelectric drive structure is located in the central region of the skeleton support structure (300) and is coupled to the inner ends of multiple cantilever support units. The organic sealing layer (900) covers the skeleton support structure (300) and the piezoelectric drive structure, and its periphery is sealed to the outer fixing structure above the substrate layer (100), thus isolating the front acoustic cavity and the rear acoustic cavity of the speaker.

2. The skeleton-diaphragm coupled piezoelectric MEMS loudspeaker according to claim 1, characterized in that, The piezoelectric drive structure includes a lower electrode layer (400), a piezoelectric functional layer (500), and an upper electrode layer (600); The lower electrode layer (400), piezoelectric functional layer (500) and upper electrode layer (600) are stacked sequentially on the skeleton support structure (300).

3. The skeleton-diaphragm coupled piezoelectric MEMS loudspeaker according to claim 1, characterized in that, The skeleton support structure (300) is one or more of the following: folded beam structure, ring reinforcement structure or multi-level bifurcated structure; The skeleton support structure (300) is made of monocrystalline silicon, polycrystalline silicon, SiNx or metal materials.

4. The skeleton-diaphragm coupled piezoelectric MEMS loudspeaker according to claim 1, characterized in that, The organic sealing layer (900) is made of Parylene-C, PI, PDMS, SU-8 or a composite flexible polymer material; The Young's modulus of the organic sealing layer (900) is lower than that of the material Young's modulus of the skeleton support structure (300); The organic sealing layer (900) is connected to the skeleton support structure (300) and the piezoelectric drive structure through processes such as bonding, surface bonding or integrated deposition.

5. The skeleton-diaphragm coupled piezoelectric MEMS loudspeaker according to claim 1, characterized in that, The skeleton support structure (300) forms physical gaps through etching, so that the effective unfolding length of the drive beam is greater than its straight span.

6. The skeleton-diaphragm coupled piezoelectric MEMS loudspeaker according to claim 1, characterized in that, The organic sealing layer (900) is connected to the top of the skeleton support structure (300) and the piezoelectric drive structure. The organic sealing layer (900) is suspended above the physical gap between adjacent structures or fills the physical slit, and is connected to the organic membrane body above the slit.

7. A method for fabricating a skeleton-diaphragm coupled piezoelectric MEMS loudspeaker, characterized in that, Includes the following steps: S1. Composite substrate preparation: Select a composite wafer with a substrate layer (100), a buried oxide insulating layer (200) and a top silicon device layer, and sequentially prepare a lower electrode layer (400), a piezoelectric functional layer (500) and an upper electrode layer (600) on the surface of the top silicon device layer. S2. Construction of piezoelectric drive stack: The upper electrode layer (600), piezoelectric functional layer (500) and lower electrode layer (400) are patterned through multiple photolithography and etching processes to form multiple independent piezoelectric drive units. S3. Skeleton structure definition: The top silicon device layer is patterned by etching process to define the skeleton support structure (300) and its physical gaps. The skeleton support structure (300) is set in the central region and coupled to multiple piezoelectric drive units. S4. Organic sealing layer integration: A low-modulus organic sealing layer (900) is integrated above the skeleton support structure (300) and the piezoelectric drive structure, so that the organic sealing layer (900) spans the physical gap and is sealed and connected to the surrounding fixed area to form a front and rear acoustic cavity isolation structure. S5. Structure release and acoustic cavity construction: A back cavity structure is formed in the substrate layer (100) by using a substrate back etching process, and the skeleton support structure (300) is released, so that the skeleton support structure (300) is in an active state.

8. The method for fabricating a skeleton-diaphragm coupled piezoelectric MEMS loudspeaker according to claim 7, characterized in that, In step S4, the integration of the organic sealing layer can be performed using an ex-situ transfer process, specifically including: Pre-fabricated organic sealing films were prepared on an auxiliary substrate; Surface activation treatment was performed on the front side of the MEMS structure and the surface of the pre-fabricated organic sealing film. The prefabricated organic sealing film is aligned and covered over the skeleton support structure (300), so that the prefabricated organic sealing film spans the physical gap and is bonded to the top surface and the outer fixing structure of the skeleton support structure (300). Remove the auxiliary substrate to form a skeleton diaphragm coupling structure.

9. The method for fabricating a skeleton-diaphragm coupled piezoelectric MEMS loudspeaker according to claim 7 or 8, characterized in that, Step S2 is followed by passivation layer and interconnect pad fabrication steps, specifically including: A passivation layer (700) is deposited on the device surface; Contact holes are formed in the passivation layer (700) by photolithography and etching processes, exposing the surfaces of the upper electrode layer (600) and the lower electrode layer (400) respectively; A metal layer is deposited and patterned to form metal interconnect pads (800), which are electrically connected to the upper electrode layer (600) and the lower electrode layer (400) through contact holes, respectively.

10. The method for fabricating a skeleton-diaphragm coupled piezoelectric MEMS loudspeaker according to claim 7 or 8, characterized in that, In step S5, the structural release and acoustic cavity construction specifically include: The substrate silicon material is removed from the back side of the substrate layer (100) by deep reactive ion etching process to form a back cavity structure; The exposed buried oxygen insulation layer (200) is removed by using hydrofluoric acid solution or gas, so that the skeleton support structure (300) is completely released.