Frequency mixing MEMS piezoelectric loudspeaker and manufacturing method thereof

Abstract

The invention provides a frequency mixing MEMS piezoelectric loudspeaker and a manufacturing method thereof, the frequency mixing MEMS piezoelectric loudspeaker comprises a plurality of MEMS piezoelectric sound production units, the structure of a vibrating diaphragm of each unit is a cantilever beam structure, a bridge type structure or a periphery reinforcing structure, if the sizes of cavities of all the MEMS piezoelectric sound production units are the same, the structures of the vibrating diaphragms of at least two MEMS piezoelectric sound production units are different, and if the sizes of the cavities of all the MEMS piezoelectric sound production units are the same, the structures of the vibrating diaphragms of all the MEMS piezoelectric sound production units are different. If the size of the cavity of at least one MEMS piezoelectric sound production unit is different from the size of the cavity of the other MEMS piezoelectric sound production unit, the structures of the vibrating diaphragms of the plurality of MEMS piezoelectric sound production units are the same or different, and the sizes of the cavities and the structures of the vibrating diaphragms are related to the resonant frequency of the MEMS piezoelectric sound production units. According to the loudspeaker, the plurality of MEMS piezoelectric sounding units have different diaphragm structures and / or have different cavity sizes, so that the output sound pressure level SPL of the loudspeaker can be improved, and the audio frequency response range is widened.

Description

Technical Field

[0001] The present invention relates to a loudspeaker manufactured based on a MEMS piezoelectric process, and in particular to a high-performance mixed-frequency MEMS piezoelectric loudspeaker capable of achieving full-range coverage and even ultrasound, and a manufacturing method thereof. Background Art

[0002] For a single-frequency sound system, it is generally necessary to compromise between high-frequency and low-frequency reproduction, usually in the bass and sacrifice high-frequency performance. Traditional two-way or even multi-way speakers use multiple smaller BAs (Balanced Armatures) assembled together, or use electric speakers for the bass area, and hybrid solutions as tweeter BAs to achieve applications in a wide frequency range. In recent years, micro speakers based on piezoelectric MEMS technology have small size and light weight, while the speed and accuracy of linear actuators such as MEMS chips are higher than voice coils, thereby achieving higher bandwidth and more accurate high-frequency reproduction. As a result, they are used in hybrid dual-frequency or audio speaker modules, integrating MEMS speakers as tweeters into ring-shaped electric woofers to create a 2-way system. These speaker modules integrate woofers and tweeter MEMS speakers to build TWS (True Wireless Stereo) system-on-chip (SoC) compatible "drop-in" speaker solutions. The superior high-frequency response of the MEMS piezoelectric speaker enables a wider sound field, providing a higher level of clarity and presence for speech, vocals and instruments, allowing the woofer to focus on the low-frequency energy required for deep bass response and active noise reduction. The woofer and tweeter are perfectly matched in the module design, and this solution can improve the sound quality and also use MEMS drivers for effective active noise reduction (ANC) applications.

[0003] However, these assembled high and low frequency two-way audio speaker systems will take up more space in the headphones, and the audio signal driving circuit of the tweeter MEMS speaker is different from that of the traditional ring electric woofer, which cannot ensure seamless integration with the existing system. It needs to configure an additional special driving circuit, and there are problems such as complex system control. In summary, the relevant technology lacks effective solutions to the problems of MEMS-based piezoelectric speakers in providing low-frequency sound pressure level SPL output and expanding the audio frequency response range. Summary of the invention

[0004] The purpose of the present invention is to provide a mixing MEMS piezoelectric speaker and a manufacturing method thereof to improve the low-frequency output sound pressure level SPL of the piezoelectric MEMS speaker, so that the micro speaker has excellent treble and bass performance and a wide dynamic range, and achieves a wider audio frequency response range.

[0005] A mixed-frequency MEMS piezoelectric loudspeaker provided by the present invention comprises: a plurality of MEMS piezoelectric sound generating units arranged in an array structure; each MEMS piezoelectric sound generating unit comprises: a substrate and a composite component located on the upper surface of the substrate; the composite component comprises at least one piezoelectric composite unit and a dry film material; the middle area on the back of the substrate is a cavity; in the composite component, a diaphragm is formed in an area opposite to the cavity; the structure of the diaphragm is a cantilever beam structure, a bridge structure or a structure reinforced all around; wherein, if the size of the cavity of each MEMS piezoelectric sound generating unit is the same, then there are at least two MEMS piezoelectric sound generating units whose diaphragm structures are different; if the size of the cavity of at least one MEMS piezoelectric sound generating unit is different from the size of the cavity of other MEMS piezoelectric sound generating units, then the structures of the diaphragms of the plurality of MEMS piezoelectric sound generating units are the same or different.

[0006] Furthermore, if the size of the cavity of each MEMS piezoelectric sound unit is the same, the resonant frequency of the MEMS piezoelectric sound unit whose diaphragm is a bridge-type structure is lower than the resonant frequency of the MEMS piezoelectric sound unit whose diaphragm is a structure reinforced all around, and is higher than the resonant frequency of the MEMS piezoelectric sound unit whose diaphragm is a cantilever beam structure.

[0007] Furthermore, for MEMS piezoelectric sound generating units having the same diaphragm structure, when the lengths of the cavities are consistent, as the aspect ratio increases, the resonant frequency of the MEMS piezoelectric sound generating unit increases.

[0008] Further, the dry film material in the combined component is used to fill the gap of each piezoelectric composite unit; or, the dry film material in the combined component is used to support each piezoelectric composite unit.

[0009] Further, in each diaphragm, a set of opposite designated edges are anchored at an edge region of the substrate.

[0010] Furthermore, for each piezoelectric composite unit in each MEMS piezoelectric sound unit, at least a portion of the piezoelectric composite unit belongs to the diaphragm; the position of the piezoelectric composite unit in the diaphragm is: the middle area of ​​the diaphragm, or, at least a portion of the piezoelectric composite unit is anchored in the edge area of ​​the substrate.

[0011] Furthermore, there is a gap between the diaphragms of every two adjacent MEMS piezoelectric sound generating units.

[0012] The present invention provides a method for manufacturing a mixed-frequency MEMS piezoelectric speaker, the method comprising: thermally growing an insulating layer on the upper surface and the lower surface of a preset first substrate respectively to obtain a first substrate with an insulating layer; on the insulating layer on the upper surface of the first substrate with the insulating layer, depositing a seed layer, a bottom electrode layer, an intermediate layer and a top electrode layer in sequence from bottom to top to obtain a first device; wherein the intermediate layer is a piezoelectric layer, or the intermediate layer is formed by overlapping and stacking a plurality of piezoelectric layers and a plurality of electrode layers; wherein the upper layer of the bottom electrode layer and the lower layer of the top electrode layer are both piezoelectric layers; processing the first device to obtain a MEMS piezoelectric sound-generating unit; wherein the MEMS piezoelectric sound-generating unit has a cavity, and the MEMS piezoelectric The structure of the diaphragm of the sound unit is a cantilever beam structure, a bridge structure or a structure reinforced all around; the steps of thermally growing an insulating layer on the upper surface and the lower surface of a preset first substrate are repeated to obtain a plurality of MEMS piezoelectric sound units; the plurality of MEMS piezoelectric sound units are arranged in an array structure to obtain a mixed-frequency MEMS piezoelectric speaker; wherein, if the size of the cavity of each MEMS piezoelectric sound unit is the same, then there are at least two MEMS piezoelectric sound units whose diaphragms have different structures; if the size of the cavity of at least one MEMS piezoelectric sound unit is different from the size of the cavity of other MEMS piezoelectric sound units, then the structures of the diaphragms of the plurality of MEMS piezoelectric sound units are the same or different.

[0013] Furthermore, the first device is processed to obtain a MEMS piezoelectric sound unit, including the steps of: patterning the bottom electrode layer, the middle layer and the top electrode layer in the first device to form at least one piezoelectric composite unit to obtain a patterned second device; depositing a passivation layer on the second device to isolate the electrical connection of each top electrode layer, and patterning the passivation layer to form a pad area to lead out the bottom electrode layer and / or the top electrode layer of each piezoelectric composite unit to obtain a third device; depositing a metal layer on each pad area in the third device to form a pad, and bonding a metal wire to obtain a fourth device; filling the fourth device with a covering dry film material, and patterning the dry film material to obtain a fifth device; thinning the bottom surface of the first substrate of the fifth device and etching a cavity to obtain a MEMS piezoelectric sound unit.

[0014] Furthermore, the first device is processed to obtain the MEMS piezoelectric sound generating unit, including the steps of: thermally growing insulating layers on the upper and lower surfaces of the preset second substrate, and covering the insulating layer on the upper surface of the second substrate with a dry film material to obtain a sixth device; bonding the first device to the upper surface of the dry film material of the sixth device so that the top electrode layer of the first device is combined with the dry film material, and removing the first substrate with the insulating layer of the first device to obtain a seventh device; and patterning the bottom electrode layer, the middle layer and the top electrode layer in the seventh device to form at least one piezoelectric composite. A unit is formed to obtain an eighth device after patterning; a passivation layer is deposited on the eighth device to isolate the electrical connection of each top electrode layer, and the passivation layer is patterned to form a pad area to lead out the bottom electrode layer and / or the top electrode layer of each piezoelectric composite unit to obtain a ninth device; a metal layer is deposited on each pad area in the ninth device to form a pad, and a metal wire is bonded to obtain a tenth device; the dry film material is patterned to obtain an eleventh device; the bottom surface of the second substrate of the eleventh device is thinned and a cavity is etched to obtain a MEMS piezoelectric sound unit.

[0015] The mixed frequency MEMS piezoelectric loudspeaker and the manufacturing method thereof provided by the present invention comprise: a plurality of MEMS piezoelectric sound generating units arranged in an array structure; each MEMS piezoelectric sound generating unit comprises: a substrate and a combination component located on the upper surface of the substrate; the combination component comprises at least one piezoelectric composite unit and a dry film material; the middle area of ​​the back of the substrate is a cavity; in the combination component, a diaphragm is formed in the area opposite to the cavity; the structure of the diaphragm is a cantilever beam structure, a bridge structure or a surrounding reinforcement structure; wherein, if the size of the cavity of each MEMS piezoelectric sound generating unit is the same, then there are at least two MEMS piezoelectric sound generating units with different diaphragm structures; if the size of the cavity of at least one MEMS piezoelectric sound generating unit is different from the size of the cavity of other MEMS piezoelectric sound generating units, then the structures of the diaphragms of the plurality of MEMS piezoelectric sound generating units are the same or different; the loudspeaker can improve the output sound pressure level SPL of the device and widen the audio frequency response range by arranging the plurality of MEMS piezoelectric sound generating units with different diaphragm structures and / or different cavity sizes. BRIEF DESCRIPTION OF THE DRAWINGS

[0016] In order to more clearly illustrate the specific implementation methods of the present invention or the technical solutions in the prior art, the drawings required for use in the specific implementation methods or the description of the prior art will be briefly introduced below. Obviously, the drawings described below are some implementation methods of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying creative work.

[0017] Figure 1A structural schematic diagram of a frequency mixing MEMS piezoelectric speaker provided by an embodiment of the present invention;

[0018] Figure 2 A schematic structural diagram of another frequency mixing MEMS piezoelectric speaker provided by an embodiment of the present invention;

[0019] Figure 3(a) , 3(b) , 3(c), 3(d), 3(e), and 3(f) are schematic diagrams of top views of a MEMS piezoelectric sound generating unit provided in an embodiment of the present invention;

[0020] FIG4( a) is a schematic diagram of a relationship between a cavity width and a resonant frequency provided by an embodiment of the present invention;

[0021] FIG4( b ) is a schematic diagram of a relationship between a cavity length and a resonant frequency provided by an embodiment of the present invention;

[0022] FIG4( c ) is a schematic diagram of the relationship between the width of a composite layer and the resonant frequency provided by an embodiment of the present invention;

[0023] FIG4( d ) is a schematic diagram of the relationship between the thickness of a piezoelectric layer and the resonant frequency provided by an embodiment of the present invention;

[0024] FIG4(e) is a schematic diagram of the relationship between the thickness of a support layer and the resonant frequency provided by an embodiment of the present invention;

[0025] FIG5( a ) is a schematic diagram of a cross-sectional view of a mixing frequency MEMS piezoelectric speaker provided by an embodiment of the present invention;

[0026] Figure 5(b) , 5(c) 5(d) is a schematic diagram of a top view of a mixing frequency MEMS piezoelectric speaker provided by an embodiment of the present invention;

[0027] FIG6( a ) is a schematic diagram of a cross-sectional view of another mixing frequency MEMS piezoelectric speaker provided by an embodiment of the present invention;

[0028] Figure 6(b) , 6(c) 6(d) is a schematic diagram of a top view of another mixing frequency MEMS piezoelectric speaker provided by an embodiment of the present invention;

[0029] FIG. 7( a ) is a schematic diagram of a cross-sectional view of another mixing frequency MEMS piezoelectric speaker provided in an embodiment of the present invention;

[0030] Figure 7(b) , 7(c) 7(d) is a schematic diagram of a top view of another mixing frequency MEMS piezoelectric speaker provided by an embodiment of the present invention;

[0031] FIG8( a ) is a schematic diagram of a cross-sectional view of another mixing frequency MEMS piezoelectric speaker provided by an embodiment of the present invention;

[0032] Figure 8(b) , 8(c) 8(d) is a schematic diagram of a top view of another mixing frequency MEMS piezoelectric speaker provided by an embodiment of the present invention;

[0033] FIG9( a ) is a schematic diagram of a cross-sectional view of another mixing frequency MEMS piezoelectric speaker provided in an embodiment of the present invention;

[0034] Figure 9(b) , 9(c) 9(d) is a schematic diagram of a top view of another mixing frequency MEMS piezoelectric speaker provided by an embodiment of the present invention;

[0035] FIG10( a ) is a schematic diagram of a cross-sectional view of another mixing frequency MEMS piezoelectric speaker provided in an embodiment of the present invention;

[0036] Figure 10(b) , 10(c) 10(d) is a schematic diagram of a top view of another mixing frequency MEMS piezoelectric speaker provided by an embodiment of the present invention;

[0037] FIG11( a ) is a schematic diagram of a cross-sectional view of another mixing frequency MEMS piezoelectric speaker provided in an embodiment of the present invention;

[0038] Figure 11(b) , 11(c) 11(d) is a schematic diagram of a top view of another mixing frequency MEMS piezoelectric speaker provided by an embodiment of the present invention;

[0039] FIG. 12( a ) is a schematic diagram of a cross-sectional view of another mixing frequency MEMS piezoelectric speaker provided in an embodiment of the present invention;

[0040] Figure 12(b) , 12(c) 12(d) is a schematic diagram of a top view of another mixing frequency MEMS piezoelectric speaker provided by an embodiment of the present invention;

[0041] FIG. 13( a ) is a schematic diagram of a cross-sectional view of another mixing frequency MEMS piezoelectric speaker provided in an embodiment of the present invention;

[0042] Figure 13(b) , 13(c) 13(d) is a schematic diagram of a top view of another mixing frequency MEMS piezoelectric speaker provided by an embodiment of the present invention;

[0043] FIG. 14( a ) is a schematic diagram of a cross-sectional view of another mixing frequency MEMS piezoelectric speaker provided in an embodiment of the present invention;

[0044] Figure 14(b) , 14(c) 14(d) is a schematic diagram of a top view of another mixing frequency MEMS piezoelectric speaker provided by an embodiment of the present invention;

[0045] FIG. 15( a ) is a schematic diagram of a cross-sectional view of another mixing frequency MEMS piezoelectric speaker provided in an embodiment of the present invention;

[0046] Figure 15(b) , 15(c) 15(d) is a schematic diagram of a top view of another mixing frequency MEMS piezoelectric speaker provided by an embodiment of the present invention;

[0047] Fig.16 A schematic cross-sectional view of a substrate having an insulating layer grown thereon provided by an embodiment of the present invention;

[0048] Fig.17 A schematic diagram of a first device provided by an embodiment of the present invention;

[0049] Fig.18 A schematic diagram of a second device provided by an embodiment of the present invention;

[0050] Fig.19 A schematic diagram of a device covered with a dry film material provided by an embodiment of the present invention;

[0051] Fig. 20 A schematic diagram of a fifth device provided by an embodiment of the present invention;

[0052] Fig.21 A schematic diagram of a device after thinning provided by an embodiment of the present invention;

[0053] Fig. 22 A schematic diagram of a MEMS piezoelectric sound generating unit provided in an embodiment of the present invention;

[0054] Fig.23 A schematic cross-sectional view of a substrate having an insulating layer grown thereon provided by an embodiment of the present invention;

[0055] Fig.24 A schematic diagram of a first device provided by an embodiment of the present invention;

[0056] Fig.25 A schematic diagram of a sixth device provided by an embodiment of the present invention;

[0057] Fig.26 A schematic diagram of a seventh device provided by an embodiment of the present invention;

[0058] Fig. 27 A schematic diagram of an eighth device provided by an embodiment of the present invention;

[0059] Fig.28 A schematic diagram of an eleventh device provided by an embodiment of the present invention;

[0060] Fig.29 A schematic diagram of a device after thinning provided by an embodiment of the present invention;

[0061] Fig.30 A schematic diagram of a device after cavity etching provided by an embodiment of the present invention;

[0062] Fig.31 A schematic diagram of a top view of a frequency mixing MEMS piezoelectric speaker provided in an embodiment of the present invention;

[0063] Fig.32 A schematic diagram of a top view of a frequency mixing MEMS piezoelectric speaker provided in an embodiment of the present invention;

[0064] Fig.33 A schematic diagram of a sound pressure level-frequency curve provided by an embodiment of the present invention. DETAILED DESCRIPTION

[0065] The technical solution of the present invention will be clearly and completely described below in conjunction with the embodiments. Obviously, the described embodiments are part of the embodiments of the present invention, rather than all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0066] As micro speakers become increasingly miniaturized and lightweight, they are also required to have excellent treble and bass performance and a wide dynamic range. In order to meet the ever-increasing sound quality requirements, multi-dynamic coil units, multi-balanced iron units, or dynamic coil / balanced iron units are usually used together in mid-to-high-end headphones to cover different sound ranges. Micro speakers based on piezoelectric MEMS technology have small size and light weight. The speed and accuracy of linear actuators such as MEMS chips are higher than voice coils, thus achieving higher bandwidth and more accurate high-frequency reproduction. Hybrid audio modules are now available that integrate MEMS tweeters into ring-shaped electric woofers, creating a 2-way system. The excellent high-frequency response of the tweeter MEMS speaker enables a wider sound field, providing a higher level of clarity and presence for speech, vocals, and instruments, allowing the woofer to focus on the low-frequency energy required for bass response and active noise reduction. The woofer and tweeter are perfectly matched in the module design, eliminating the need for analog or digital crossovers, improving sound quality, and using MEMS drivers for effective active noise cancellation (ANC) applications.

[0067] However, this two-way audio hybrid audio module will take up more space in the headphones. At the same time, the audio signal driving circuit of the tweeter MEMS speaker driver is different from that of the traditional ring electric woofer, which cannot ensure seamless integration with the existing system. It requires customization of special circuits, and there are also problems such as complex control of the two-way frequency system.

[0068] The related art provides a micromechanical sound transducer, which can realize different bending actuators responsible for different frequency ranges by connecting several bending elements of the bending actuator in series. Therefore, for example, the inner bending actuator can be configured for the high frequency range, while the other actuator on the outer side can be operated for the low frequency range. However, this solution has the problems of large size, easy membrane breakage and low reliability.

[0069] Related technologies provide a two-way speaker module that integrates a dynamically driven woofer, a tweeter MEMS speaker, and a MEMS speaker amplifier to build a "drop-in" speaker solution compatible with today's leading TWS system-on-chip (SoC). However, this solution has the following problems: 1. The two-way speaker module takes up more space in the headset and is complicated to install; 2. The audio signal driving circuit of the MEMS tweeter is different from that of the traditional ring electric woofer, and additional special circuits need to be configured; 3. The dual-driver module and the 2-way system control drive are more complicated.

[0070] The related art provides a piezoelectric MEMS speaker and a design method thereof, and an electronic device, wherein the piezoelectric MEMS speaker includes N adjacent diaphragms, the structures of the N adjacent diaphragms are asymmetric, N is a positive integer greater than 1, and there is a first gap between the adjacent diaphragms, and / or the piezoelectric MEMS speaker includes at least one diaphragm and at least one boundary element adjacent to the diaphragm, the height of the boundary element is not higher than the height of the diaphragm, and there is a second gap between the diaphragm and the boundary element, wherein the size of the first gap and / or the second gap meets the preset conditions, so that when the diaphragm vibrates at the mechanical resonance point, an acoustic short circuit phenomenon occurs, and when the diaphragm is stationary, an acoustic short circuit phenomenon does not occur. However, the solution has the following problems: the low-frequency output sound pressure level SPL is low; multiple cantilever beam units have inconsistent postures due to the manufacturing process, and low reliability.

[0071] The related technology provides an innovative hybrid audio module that combines MEMS and coaxially aligned dynamic drivers through its 2-way audio technology. The MEMS tweeter is integrated into the ring electric woofer to create a 2-way system. However, this design takes up more space in the headphones. The MEMS tweeter driver requires a custom design, and unlike the audio signal driving circuit of the traditional ring electric woofer, it cannot ensure seamless integration with the existing system and requires additional special circuits. In addition, there are problems such as complex installation, complex control of the 2-way system, affected battery life, and poor usability.

[0072] For a single-frequency sound system, when it is necessary to reproduce high and low frequencies, it is usually prioritized to ensure the bass, while sacrificing high-frequency performance. Traditional two-way or even multi-way speakers will use multiple smaller BAs, assembled together, or use electric speakers for the bass area, and a hybrid solution as a tweeter BA to achieve applications in a wide frequency range. At present, there are also electric speakers for the bass area, and hybrid applications as tweeter MEMS speakers. The tweeter MEMS piezoelectric speaker produces inherently more precise movement and starts quickly, which is particularly important for accurately reproducing high frequencies. However, these assembled high and low frequency speaker systems will take up more space in the headphones, and the audio signal driving circuits of the tweeter MEMS piezoelectric speaker and the electric bass speaker are different, and additional special driving circuits need to be configured, and there are problems such as complex system control. In summary, the relevant technology lacks an effective solution to the problem that the output sound pressure level SPL of the MEMS piezoelectric speaker is low and the audio frequency response range is narrow. Based on this, the embodiment of the present invention provides a mixed frequency MEMS piezoelectric speaker and a manufacturing method thereof, which can be applied to applications that need to improve the output sound pressure level and audio frequency response range of the MEMS piezoelectric speaker.

[0073] To facilitate understanding of the present embodiment, a mixed-frequency MEMS piezoelectric speaker disclosed in an embodiment of the present invention is first introduced, and the speaker includes: a plurality of MEMS piezoelectric sound units arranged in an array structure; each MEMS piezoelectric sound unit includes: a substrate and a composite component located on the upper surface of the substrate; the composite component includes at least one piezoelectric composite unit and a dry film material; the middle area on the back of the substrate is a cavity; in the composite component, a diaphragm is formed in the area opposite to the cavity; the structure of the diaphragm is a cantilever beam structure, a bridge structure or a surrounding reinforcement structure; wherein, if the size of the cavity of each MEMS piezoelectric sound unit is the same, there are at least two MEMS piezoelectric sound units whose diaphragms have different structures; if the size of the cavity of at least one MEMS piezoelectric sound unit is different from the size of the cavity of other MEMS piezoelectric sound units, the structures of the diaphragms of multiple MEMS piezoelectric sound units are the same or different.

[0074] The material of the substrate can be silicon, glass, sapphire, etc., which can be selected according to actual needs; the upper surface of the substrate is usually covered with an insulating layer, and the composite component is located on the insulating layer. The composite component can include a dry film material and one or more piezoelectric composite units; wherein the dry film material can serve as the sound film of the MEMS piezoelectric sound generating unit, and can be a flexible material. The use of a flowable and curable polymer flexible material can cover and fill the slot gap and the slot area, and form a covering flexible film after curing; the size and shape of the cavity can be set according to actual needs, and the back of the substrate can be etched to form the cavity; under the same process conditions, the substrate material of each MEMS piezoelectric sound generating unit is consistent, and the substrate is etched to form an independent cavity structure, and the diaphragm of the MEMS piezoelectric generating unit is released. The structural shape, size, etc. of the cavity of each MEMS piezoelectric sound generating unit can be the same or different, and can be set according to actual needs; by changing the size of the cavity, the mechanical resonance point of each MEMS piezoelectric sound generating unit is designed to be within the audio range, or beyond its range of 20KHz, or even in the ultrasonic frequency band. By designing cavity structures of different sizes, MEMS piezoelectric generating units with different resonance frequencies can be obtained, thereby expanding the bandwidth of the mixed-frequency MEMS piezoelectric speaker.

[0075] like Figure 1 The schematic diagram of the structure of a mixed frequency MEMS piezoelectric speaker is shown, the device is composed of three MEMS piezoelectric sound generating units, the bottom center area of ​​the device substrate has three cavity structures, and the sizes of the three cavities are the same. The upper surface of the substrate has an oxide layer with an insulating effect, and at least one piezoelectric composite unit is arranged in the middle of the upper surface of the cavity opposite to the substrate or / and in the edge area of ​​the anchoring substrate, and each piezoelectric composite unit includes: a bottom electrode layer, a piezoelectric layer, and an upper electrode layer structure; the patterned piezoelectric composite unit forms a plurality of groove gap structure areas, and the preset area in each groove is covered with a filling dry film material, so as to form a diaphragm of the MEMS piezoelectric sound generating unit in the upper area of ​​the cavity of the substrate. Using the same MEMS process technology, MEMS tweeters, MEMS mid- and woofers are made on the same wafer and seamlessly integrated into a hybrid MEMS piezoelectric speaker with two-way and multi-way divisions, which can achieve full-range coverage and extend to the ultra-high frequency band to achieve a wide acoustic bandwidth. Compared with traditional dynamic coils, a combination of multiple iron speakers, and 2-way audio speakers that stack a MEMS tweeter and a dynamic woofer, the size of the hybrid MEMS piezoelectric speaker is greatly reduced and can be used in thin headphones or wearable devices with limited space.

[0076] like Figure 1As shown, the hybrid MEMS piezoelectric speaker is a whole speaker system composed of three closely connected MEMS piezoelectric sound generating units, including: substrate 60, cavity 61, cavity 62 and cavity 63, groove 71, groove 72 and groove 73, insulating layer 10, opening 11, piezoelectric composite unit 40a, piezoelectric composite unit 40b, piezoelectric composite unit 40c and piezoelectric composite unit 40d, and dry film material 50. The back of substrate 60 is etched with cavity 61, cavity 62 and cavity 63, and the sizes of these three cavities are the same. Grooves 71, groove 72 and groove 73 are etched on substrate 60, and the shapes and sizes of these three grooves can be the same or / and different. The dry film material 50 is made of flexible material, covering and / or filling the grooves of each unit, and an opening 11 is opened on the diaphragm 30c, and the cross-sectional size of the opening 11 is smaller than the cross-sectional size of the groove 73. The insulating layer 10 is arranged on the substrate 60 , and the piezoelectric composite unit and the filled dry film material constitute each cavity. The diaphragms 30a , 30b , and 30c are arranged on the cavities, and the edges of the piezoelectric composite unit 40a , 40b , and 40c are connected to the substrate 60 .

[0077] In the case of piezoelectric drive, the opening 11 helps to release the residual stress of the diaphragm 30c during production. Covering and / or filling the grooves 71, 72 and 73 with the dry film material 50 can prevent sound leakage. At the same time, the dry film material 50 has a low density, and the material covering and / or filling the groove area increases the deflection displacement of the diaphragm vibration and the output sound pressure level. When the force of the motor is equal, the lower the mass, the higher the diaphragm acceleration, and the high sound pressure level SPL is achieved at high frequencies. At the same time, the system damping can be increased to reduce its resonance Q value and reduce the total harmonic interference THD (Total Harmonic Distortion). There are gaps between the diaphragms 30a, 30b and 30c on the three MEMS piezoelectric generating units, and they are not connected to each other, which can reduce crosstalk between the units.

[0078] like Figure 2 A schematic structural diagram of another hybrid MEMS piezoelectric speaker is shown, which comprises three MEMS piezoelectric sound-generating units interconnected with each other, wherein the multiple adjacent MEMS piezoelectric sound-generating units have a separate single cavity, a support layer, and a diaphragm composed of one or more piezoelectric composite units distributed on the support layer, wherein each piezoelectric composite unit comprises: a bottom electrode layer located on the upper side of the diaphragm, a piezoelectric layer, and an upper electrode layer structure.

[0079] like Figure 2As shown, the mixed frequency MEMS piezoelectric speaker is a whole speaker system composed of three closely connected MEMS piezoelectric sound generating units, including: substrate 60, cavity 61, cavity 62 and cavity 63, insulating layer 10, opening 11, piezoelectric composite unit 40a, piezoelectric composite unit 40b, piezoelectric composite unit 40c and piezoelectric composite unit 40d, and dry film material 50. Cavity 61, cavity 62 and cavity 63 are etched on the back of substrate 60, and the sizes of these three cavities are inconsistent. Dry film material 50 is made of flexible material and is the supporting layer of each MEMS piezoelectric generating unit. Opening 11 is opened on diaphragm 30a and diaphragm 30c. The insulating layer 10 is disposed on the substrate 60 , and the piezoelectric composite unit and the filled dry film material constitute the diaphragms 30a, 30b, and 30c on each cavity, and the edges of the piezoelectric composite unit 40a, 40b, and 40c are connected to the substrate 60 .

[0080] In the case of piezoelectric drive, the opening 11 helps to release the residual stress of the diaphragm 30a and the diaphragm 30c during manufacturing. The dry film material 50 has a low density. As a supporting layer material of the diaphragm, it can increase the deflection displacement of the diaphragm vibration and the output sound pressure level. When the force of the motor is equal, the lower the mass, the higher the diaphragm acceleration, and the high sound pressure level SPL is achieved at high frequencies. At the same time, the system damping can be increased to reduce its resonance Q value and reduce the total harmonic interference THD.

[0081] Figure 1 The dry film material 50 serves as a filling material for gaps and slots, and forms a sound-generating vibration film structure with the piezoelectric unit. Figure 2 The dry film material 50 serves as a support layer. Figure 1 and Figure 2 In the process, the dry film material 50 can prevent sound leakage, and the gas will not flow out from the slotted gap, thereby improving the output sound pressure level of the MEMS piezoelectric sound generating unit in the low frequency band. At the same time, when the diaphragm is divided into multiple unconnected cantilever beam structures, the various parts can be connected by flexible materials to form a vibrating whole. After processing, the postures of the various parts remain consistent, and they move up and down together as a whole under drive; the slotted parts are covered and filled with flexible materials, which reduces the moving mass of the vibration system and is easy to produce larger vibrations, thereby obtaining a larger displacement. When the force of the motor is equal, the lower the mass and the greater the acceleration of the diaphragm, the higher the sound pressure level SPL achieved at high frequencies.

[0082] Taking the use of flexible materials as an example of dry film materials, when the flexible material is used to cover the slot of the rigid material, its width is usually greater than the slot width. The flexible material should have inherent high mechanical properties, excellent thermal stability, good chemical resistance and low dielectric constant, excellent adhesion performance, and can be used as an insulating shielding layer in the MEMS process. The selection of flexible materials needs to meet the following conditions: 1) Compatible with MEMS processes, can be used for patch, exposure, etching and other processes; 2) Reliability issues, dust and water resistance, and at the same time meet the rigid material plate fit, under different humidity conditions, and under force or excitation, there will be no film tearing, delamination, peeling and other phenomena. Excellent electrical insulation can effectively prevent current leakage and electrical breakdown; 3) Wide temperature range: maintain its physical properties at extreme temperatures (between -200℃ and +300℃), flexible materials can withstand high temperatures and withstand the temperature of the patch packaging process without causing reliability problems in the device, such as causing the rigid material to break; 4) Excellent mechanical properties: high strength, high toughness, strong impact resistance, a certain elongation rate, and the flexible material will not be damaged when the device is deflected. 5) The difference in the coefficient of thermal expansion (CTE) between the material and the substrate is small: the CTE is about 20 to 50 ppm / °C. In other words, the volume changes very little when the temperature changes, minimizing the bending of the chip. The flexible material can be selected from but not limited to one of the following materials: PI (polyimide), PSPI (Photosensitive Polyimide), polyurethane, Parylene C (C-type polyparaxylene), PVI-3 (PVI English full name Poly Vinyl Isobutyl ether, vinyl isobutyl ether), polydimethylsiloxane or other organic films. The thickness of the flexible material is between 0.5μm and 30μm.

[0083] With the development of wearable electronic devices, mixed-frequency MEMS piezoelectric speakers with smaller size, lower power consumption, better performance, full frequency coverage, and higher sound pressure level have become an inevitable trend. Through the monolithic integration of three frequency units or multiple frequency units, the bandwidth of the mixed-frequency MEMS piezoelectric speaker is improved to achieve full frequency range and even ultrasonic range applications.

[0084] Generally speaking, the mode merging method can be adopted. By using a rectangular structure of a hybrid MEMS piezoelectric speaker with a large aspect ratio, multiple modes can be excited in a narrow frequency range. The excited modes are all odd modes. The design uses high-damping dry film materials to form a rigid-flexible vibration sound system, increase the damping of the sound system, and reduce the quality factor Q of the device resonance. When the device is working, the peak-to-peak value of the excited different modes will become less narrow and merge in a wide frequency region with a large response to form an ultra-wide bandwidth. Due to the cancellation of opposite propagation motions, there is a lack of even modes in the response spectrum. There is a lack of even modes in the device response spectrum. Taking the second mode as an example, the membrane can be divided into two regions, which are symmetrical and move in opposite directions. However, the mechanical force generated by the inverse piezoelectric effect is unidirectional, and it is difficult for the unidirectional force to couple to this opposite motion and excite the second mode. Other even modes are also difficult to be excited for the same reason, so only odd modes are electrically excited. Multiple modes are excited in a narrow frequency range, and the excited modes are all odd modes. Multiple modes are merged together, so the frequency bandwidth is significantly widened. By observing the modal frequency ratio of the rectangular device, it can be found that the speed of frequency increase depends on the aspect ratio K of the device cavity. The higher K is, the slower the increase rate (referring to the interval size of the first-order, third-order, fifth-order and other modal frequencies), that is, the higher the modal frequency is, the closer it is to the fundamental frequency. Within a certain frequency range, the number of vibration modes of the device is controlled by the parameter K, and the aspect ratio K is generally greater than 3.

[0085] In addition to mode merging, an array structure of multiple transducer units with closely spaced units of different sizes and different natural frequencies can be designed, and the peaks corresponding to all modes are merged so that the composite bandwidth is the sum of the bandwidths of the individual transducers to form a larger bandwidth, which is achieved through the entire array. Using a rectangular membrane design and by properly selecting the aspect ratio, multiple resonant modes with closely spaced resonant frequencies can be achieved in a single device, increasing the frequency bandwidth of the mixed-frequency MEMS piezoelectric speaker, not only outputting a higher sound pressure level in the full-band (20Hz-20kHz) range, but also having better acoustic performance in the ultrasonic range beyond 20kHz to achieve a wide acoustic bandwidth.

[0086] like Figure 1 and Figure 2 As shown, an orthographic projection is made from the cavity to the upper surface of the substrate, and the area corresponding to the projection is the diaphragm. The diaphragm is parallel to the upper surface of the substrate. The diaphragm can be a hard substrate or a flexible substrate. The material can be single crystal silicon, polycrystalline silicon, glass, polymer (for example, polyimide (PI), polyethylene terephthalate (PET)), organic or inorganic compound, any suitable material or combination thereof. The thickness of the diaphragm is 0.1 to 50um, which can be set according to actual needs.

[0087] In this embodiment, the structure of the diaphragm can be a cantilever beam structure, a bridge structure or a surrounding reinforcement structure; as shown in FIG3(a), a schematic diagram of a top view of a MEMS piezoelectric sound unit shows a cantilever beam structure, and the dotted lines 10a, 10b, 10c, and 10d respectively represent the edges of the etching size of the cavity on the back of the MEMS piezoelectric sound unit. The four dotted lines form a rectangular structure, which is also the boundary line of the device diaphragm. The deep silicon etching process etches away the middle layer of silicon in the back cavity to form a hollow structure cavity, which is released into the diaphragm structure of the MEMS piezoelectric sound unit. 40a indicates that the piezoelectric composite unit of the MEMS piezoelectric sound unit is located on the left side of the device, with the left edge anchored on the substrate 60 and the right side located on the diaphragm. Similarly, there is a piezoelectric composite unit on the right side of the device that is symmetrically distributed with 40a along the longitudinal axis. 20a, 20b, and 20c are gaps, gaps 20a and 20b are located below the diaphragm, gap 20c is located in the middle of the diaphragm and divides the diaphragm into a left-right symmetrical cantilever beam structure, gaps 20e, 20f, 20g, and 20h are evenly distributed on the right side of the device, one end of the gap extends beyond the edge area of ​​the cavity, while the other end does not exceed the edge area of ​​the piezoelectric composite unit on the diaphragm. Similarly, there is such a gap on the left side of the device and is distributed in a longitudinally symmetrical structure with 20e, 20f, 20g, and 20h, which is not marked in the figure. It is designed to release the residual stress of the diaphragm and the piezoelectric composite unit 40a in the MEMS piezoelectric sound unit, thereby reducing the risk of rupture due to excessive deformation of the diaphragm caused by residual stress. The device structure is a typical cantilever beam structure, and the diaphragm is a rectangular structure with a symmetrical longitudinal axis. The cantilever beam structure may have convex or concave phenomena after MEMS production, and the two cantilever beams are not in the same horizontal position, which will cause the posture of the diaphragm to be inconsistent. Under piezoelectric drive, the vibration phase and amplitude of the two diaphragms may be inconsistent, pushing the air to cancel each other out. Therefore, a dry film flexible connection structure can be designed to connect the left and right cantilever beam diaphragms to make it an overall structure to avoid the diaphragm from canceling each other out.

[0088] As shown in Figure 3(b), another schematic diagram of a top view of a MEMS piezoelectric sound unit shows another cantilever beam structure. The dotted lines in areas 10a, 10b, 10c, and 10d respectively represent the edges of the etching size of the cavity on the back of the MEMS piezoelectric sound unit. The four dotted lines form a rectangular structure, which is also the boundary line of the device film. The deep silicon etching process etches away the middle layer of silicon in the back cavity to form a hollow structure cavity, which is released into the diaphragm structure of the MEMS piezoelectric sound unit. The piezoelectric composite unit 40a of the MEMS piezoelectric sound unit is located on the left side of the device, with the left edge anchored on the substrate 60 and the right side located on the diaphragm. Similarly, there is a piezoelectric composite unit on the right side of the device that is symmetrically distributed with 40a along the longitudinal axis. 20a, 20b, 20c, 20d, 20m, and 20n are gaps, gaps 20a and 20b are located below the diaphragm, gaps 20c, 20d, 20m, and 20n are located in the middle of the diaphragm to divide the diaphragm into left-right symmetrical cantilever beam structures, gaps 20e, 20f, 20g, and 20h are evenly distributed on the right side of the device, one end of the gap extends beyond the edge area of ​​the cavity, while the other end does not exceed the edge area of ​​the piezoelectric composite unit on the diaphragm. Similarly, there is such a gap on the left side of the device in a longitudinally symmetrical structure with 20e, 20f, 20g, and 20h, which is not marked in the figure and is designed to release the residual stress of the diaphragm and the piezoelectric composite unit 40a in the MEMS piezoelectric sound unit, thereby reducing the risk of rupture due to excessive deformation of the diaphragm caused by residual stress.

[0089] As shown in Figure 3(c), another schematic diagram of a top view of a MEMS piezoelectric sound unit shows a bridge structure. The device is symmetrical about the longitudinal axis and the transverse axis. 20a and 20b are gaps located below the diaphragm, and the left and right ends of the diaphragm are constraint sections. In this MEMS piezoelectric sound unit, the middle layer of silicon on the back of the substrate 60 is etched away by a deep silicon etching process to form a hollow cavity. The edges of the cavity are 10a, 10b, 10c, and 10d, which are also the boundary lines of the diaphragm. In this embodiment, 40a is a piezoelectric composite unit located on the left side of the device, with the left edge anchored on the substrate 60 and the right side located on the diaphragm. Similarly, there is a piezoelectric composite unit on the right side of the device that is symmetrically distributed with 40a along the longitudinal axis.

[0090] As shown in Figure 3(d), another schematic diagram of a top view of a MEMS piezoelectric sound unit shows another bridge structure, 20a and 20b are gaps located below the diaphragm, and the left and right ends of the diaphragm are constraint sections. In this MEMS piezoelectric sound unit, the middle layer of silicon on the back of the substrate 60 is etched away by a deep silicon etching process to form a hollow cavity. The edges of the cavity are 10a, 10b, 10c, and 10d, which are also the boundaries of the diaphragm. In this embodiment, 40a is a piezoelectric composite unit, which is located on the left side of the device, with the left edge anchored on the substrate 60 and the right side located on the diaphragm. There is a piezoelectric composite unit 40b in the middle of the device, and a piezoelectric composite unit on the right side of the device is distributed in a longitudinally symmetrical structure with 40a.

[0091] FIG3(e) is a schematic diagram of a top view of another MEMS piezoelectric sound unit, which shows a surrounding reinforcement structure. In the MEMS piezoelectric sound unit, the middle layer of silicon on the back of the substrate 60 is etched away by a deep silicon etching process to form a hollow cavity. The edges of the cavity are 10a, 10b, 10c, and 10d, which are also the boundary lines of the diaphragm. In this embodiment, 40a is a piezoelectric composite unit, which is located on the left side of the device, with the left edge anchored on the substrate 60 and the right side located on the diaphragm. Similarly, there is a piezoelectric composite unit on the right side of the device that is symmetrically distributed with 40a along the longitudinal axis. One end of the gaps 20a, 20b, 20c, and 20d extends to the edge area of ​​the substrate 60 that is not hollow, and the other end of the gaps 20a, 20b, 20c, and 20d exceeds the middle area of ​​the piezoelectric composite unit 40a. In this embodiment, the gaps 20a, 20b, 20c and 20d divide the piezoelectric composite unit 40a near the edge of the cavity, but not completely, that is, the piezoelectric composite unit 40a is still an integral structure. The gaps 20e, 20f, 20g, and 20h are evenly distributed on the right side of the device, and one end of the gap extends beyond the edge area of ​​the cavity, while the other end does not exceed the edge area of ​​the piezoelectric composite unit on the diaphragm. Similarly, there are such gaps on the left side of the device that are distributed in a longitudinally symmetrical structure with 20e, 20f, 20g, and 20h, which are not marked in the figure and are designed to release the residual stress of the diaphragm and the piezoelectric composite unit 40a in the MEMS piezoelectric sound unit, reducing the risk of rupture due to excessive deformation of the diaphragm caused by residual stress.

[0092] As shown in FIG3(f), another schematic diagram of a top view of a MEMS piezoelectric sound unit shows another surrounding reinforcement structure. In the MEMS piezoelectric sound unit, the middle layer of silicon on the back of the substrate 60 is etched away by a deep silicon etching process to form a hollow cavity. The edges of the cavity are 10a, 10b, 10c, and 10d, which are also the boundary lines of the diaphragm. In this embodiment, 40a is a piezoelectric composite unit, which is located on the left side of the device, with the left edge anchored on the substrate 60 and the right side located on the diaphragm. There is a piezoelectric composite unit 40b in the middle of the device, and there is a piezoelectric composite unit on the right side of the device that is symmetrically distributed with 40a along the longitudinal axis. The linear gaps 20a, 20b, 20c, and 20d are mainly to release the residual stress of the device diaphragm, reduce the risk of deformation or even rupture of the diaphragm caused by stress, and at the same time, the device is structurally decoupled so that a larger drive displacement and sound pressure level output can be obtained. The audio driving signal applied to the piezoelectric composite unit 40a corresponds to the output sound wave signal. The audio driving signal is generally a sinusoidal signal of a DC voltage plus an AC voltage, such as DC 10V+AC 10VP, or driven by a bipolar signal.

[0093] For PZT (lead zirconate titanate film) piezoelectric film, the nonlinear relationship of its voltage is unique. To drive the PZT piezoelectric film, an AC voltage + a DC bias voltage DC is applied. The DC bias voltage DC is mainly used to stabilize the polarization direction of the PZT film to prevent deflection. The excitation or bias voltage should be consistent with the direction of the residual polarization. The magnitude of the depolarization pressure is not only related to the magnitude of the external bias electric field, but also to the direction of the electric field, and the residual polarization intensity is also related to the external bias electric field. The DC bias voltage applied to the PZT piezoelectric film controls its polarization and inherent stress, which can tune the frequency of the two closely spaced resonant modes of the MEMS piezoelectric sound unit, and can also expand the linear working range of the MEMS piezoelectric sound unit, thereby reducing the total harmonic distortion of the MEMS piezoelectric sound unit and improving the output sound pressure level.

[0094] When the diaphragms of the MEMS piezoelectric sound-generating units in the mixed-frequency MEMS piezoelectric speaker vibrate on both sides of their corresponding resonant frequencies, the phase and amplitude of the diaphragm vibration may be inconsistent, and the vibration postures of the diaphragms may be inconsistent. When the vibration phases of the diaphragms are offset or opposite, the sound of each diaphragm will be weakened or even offset, so that the output sound pressure level of the mixed-frequency MEMS piezoelectric speaker cannot reach the maximum value. Therefore, each MEMS piezoelectric sound-generating unit can be independently controlled and driven, and driving voltages of different sizes and phases can be applied to each MEMS piezoelectric sound-generating unit to ensure that the diaphragms of each MEMS piezoelectric sound-generating unit always vibrate in the same direction to push the air, thereby obtaining the maximum sound pressure level output. For example: In a two-way array micro speaker, the first resonance frequencies of the two low-frequency and high-frequency MEMS piezoelectric sound-generating units are designed to be fA and f B , when the speaker applies a driving signal with a frequency f X Less than f A When the same phase signal is used to drive the array unit, it is called in-phase drive. X Greater than f A Less than f B If the same-phase drive is still used, the vibration directions of the two MEMS piezoelectric sound units will be opposite, resulting in the cancellation of the output pressure, which will cause a deep valley in the SPL curve. Therefore, it is necessary to implement another "out-of-phase drive" driving method in this frequency range. Between the first resonance frequency of the two MEMS piezoelectric sound units, the input signal of this low-frequency unit is reversed by 180 degrees, so that the movement of all diaphragms will remain unchanged. With this driving method, the sound pressure level drop will not occur between the two resonances, so that the sound pressure level frequency curve will be relatively flat, and high SPL and wide bandwidth will be obtained.

[0095] An array micro-speaker composed of MEMS piezoelectric sound units with different resonant frequencies, multiple MEMS piezoelectric sound units with different resonances driven by the same signal, and a phase difference of 180° between two adjacent MEMS piezoelectric sound units, which means that the vibration directions of the two adjacent MEMS piezoelectric sound units are opposite, resulting in the cancellation of the output pressure, which leads to a deep valley in the SPL curve. Therefore, in addition to controlling the phase of each unit separately, it is also a good measure to reduce the Q value to reduce the phase difference. When the maximum phase difference is kept within 90°, the mutually canceling sound pressure will be greatly suppressed. In the MEMS array speaker, each MEMS piezoelectric sound unit has an independent cavity structure, which is then achieved by applying appropriate acoustic damping. This method avoids the deep SPL valley. Even in a large array, the driving electrical signal only needs unidirectional electrical input control, and after appropriate acoustic damping tuning, a flat output sound pressure level curve can be obtained.

[0096] It can be understood that in this embodiment, the shape of the piezoelectric composite unit can be a square, rectangle, circle, polygon or other irregular graphic structures. At the same time, the linear gaps 20a, 20b, 20c, 20d can be further divided into two or more line segments, and the shape of the gap can also be other shapes such as an S-shaped curve, a semi-ellipse, an arc, etc. In addition, the position of the etched area on the diaphragm and the size of the etched area can have many variations, and are not limited to the examples provided in this solution.

[0097] Two-way and multi-way mixing MEMS piezoelectric speakers require reasonable design of the resonant frequency of each MEMS piezoelectric sound unit and the amount of air pushed. While ensuring the emission sound pressure level SPL of each frequency band, the mixing MEMS piezoelectric speaker can be higher than 20kHz and the bandwidth can be expanded to 80kHz high-resolution area. There are many ways to change the resonant frequency of the MEMS piezoelectric sound unit, such as the device structure design, the material of each film layer in the diaphragm, the thickness of the film layer, the cavity size of the device diaphragm, and the mass distribution on the diaphragm. For MEMS piezoelectric chips made from the same wafer using the MEMS process, the material and thickness of each film layer in the device diaphragm are consistent, so the conditions that can be changed are as follows:

[0098] 1) Device structure, such as cantilever beam structure, bridge structure, surrounding reinforcement structure, etc. Due to the different constraints of the device diaphragm, the resonant frequency of the device is different.

[0099] 2) The cavity size of the device diaphragm. For a rectangular diaphragm structure, the cavity length has little effect on the resonant frequency of the device, but the width of the cavity has a greater impact on the resonant frequency of the device.

[0100] 3) Mass distribution on the diaphragm. For devices with the same structural form (cantilever beam structure, bridge structure, and surrounding reinforcement structure), the diaphragm can be divided into an effective piezoelectric drive area and an unprepared drive area. The ineffective piezoelectric drive area increases the mass on the diaphragm and thus reduces the resonant frequency of the device.

[0101] In actual implementation, the size of the cavity and the structure of the diaphragm are closely related to the resonant frequency of the MEMS piezoelectric sound unit. In one of the present embodiments, when the size of the cavity of each MEMS piezoelectric sound unit in the mixed frequency MEMS piezoelectric speaker is the same, different diaphragm structures can be used to make the diaphragms of at least two MEMS piezoelectric sound units have different structures, so as to improve the output sound pressure level SPL of the mixed frequency MEMS piezoelectric speaker and widen the audio frequency response range. For example, if the mixed frequency MEMS piezoelectric speaker includes three MEMS piezoelectric sound units, the diaphragm structure of the first MEMS piezoelectric sound unit can adopt a cantilever beam structure, and the diaphragms of the other two MEMS piezoelectric sound units can both adopt a bridge structure, or both adopt a surrounding reinforcement structure, or the diaphragm structure of the second MEMS piezoelectric sound unit adopts a bridge structure, and the diaphragm structure of the third MEMS piezoelectric sound unit adopts a surrounding reinforcement structure, etc. The specific configuration can be based on actual needs. No limitation is made here.

[0102] In another embodiment of the present invention, if the size of the cavity of at least one MEMS piezoelectric sound unit is different from the size of the cavity of other MEMS piezoelectric sound units, the structures of the diaphragms of multiple MEMS piezoelectric sound units may be completely identical or may be completely different. If the structure of the diaphragm of each MEMS piezoelectric sound unit is the same, the output sound pressure level SPL of the mixed-frequency MEMS piezoelectric speaker is improved and the audio frequency response range is widened mainly by setting different cavity sizes. If the structures of the diaphragms of multiple MEMS piezoelectric sound units are not completely identical, the output sound pressure level SPL of the mixed-frequency MEMS piezoelectric speaker is improved and the audio frequency response range is widened by setting different cavity sizes and adopting different diaphragm structures, and the two influencing factors are combined to further improve the flexibility of application.

[0103] The above-mentioned mixed frequency MEMS piezoelectric speaker includes multiple MEMS piezoelectric sound units arranged in an array structure. The structure of the diaphragm of each MEMS piezoelectric sound unit can be a cantilever beam structure, a bridge structure or a surrounding reinforcement structure. If the size of the cavity of each MEMS piezoelectric sound unit is the same, then there are at least two MEMS piezoelectric sound units with different diaphragm structures; if the size of the cavity of at least one MEMS piezoelectric sound unit is different from the size of the cavity of other MEMS piezoelectric sound units, then the structures of the diaphragms of multiple MEMS piezoelectric sound units are the same or different. Since the size of the cavity and the structure of the diaphragm are related to the resonant frequency of the MEMS piezoelectric sound unit, the speaker can improve the output sound pressure level SPL of the mixed frequency MEMS piezoelectric speaker and widen the audio frequency response range by arranging multiple MEMS piezoelectric sound units with different diaphragm structures and / or different cavity sizes.

[0104] By manufacturing two, three or even multiple cantilever beam structures, bridge structures and surrounding reinforcement structures with different resonant frequencies under the same process conditions, a hybrid MEMS piezoelectric speaker with two-way and multi-way division is formed. Each MEMS piezoelectric sound unit in the hybrid MEMS piezoelectric speaker can be optimized for a specific frequency band, thereby improving the sound pressure level within the design frequency band of the hybrid MEMS piezoelectric speaker, reducing the sound pressure level peak at the resonance, and extending the SPL response to extend the high frequency to 80kHz to achieve a wide acoustic bandwidth.

[0105] Furthermore, if the size of the cavity of each MEMS piezoelectric sound unit is the same, the resonant frequency of the MEMS piezoelectric sound unit whose diaphragm is a bridge-type structure is lower than the resonant frequency of the MEMS piezoelectric sound unit whose diaphragm is a structure reinforced all around, and is higher than the resonant frequency of the MEMS piezoelectric sound unit whose diaphragm is a cantilever beam structure.

[0106] The length and width of the cavity size of the MEMS piezoelectric sound unit, or the material and thickness of the diaphragm, are closely related to the resonant frequency of the device. Through the simulation software, the finite element simulation of the MEMS piezoelectric sound unit with cantilever beam structure, bridge structure, and surrounding reinforcement structure is carried out, and the relationship between the length and width of the cavity of each unit, the width of the piezoelectric composite unit, the thickness of the piezoelectric layer, the thickness of the dry film material and other parameters and the corresponding resonant frequency of each unit can be obtained. A schematic diagram of the relationship between the cavity width and the resonant frequency is shown in Figure 4 (a), and a schematic diagram of the relationship between the cavity length and the resonant frequency is shown in Figure 4 (b). From the relationship between the cavity width, cavity length and resonant frequency of the piezoelectric sound unit with cantilever beam structure, bridge structure, and surrounding reinforcement structure in the two figures, it can be seen that the cavity width of each unit of the device has a greater influence on the resonant frequency of the device. The larger the width of the cavity, the lower the resonant frequency can be obtained. The resonant frequency of the three curves in Figure 4 (a) shows a decreasing trend as the cavity width increases. Among them, the frequency decrease trend of the MEMS piezoelectric sound unit with surrounding reinforcement structure is higher than that of the other two unit structures. It must be fast; from the relationship between the cavity length and resonant frequency of the MEMS piezoelectric sound units of the cantilever beam structure, bridge structure, and surrounding reinforced structure in Figure 4(b), it can be seen that the cavity length of each unit of the device has little effect on the resonant frequency of the device. The resonant frequencies of the three curves in the figure are linear as the cavity length increases, and do not change significantly. It can also be seen from the figure that the MEMS piezoelectric sound units of the cantilever beam structure, bridge structure, and surrounding reinforced structure with the same size structure made by the same MEMS process have the largest resonant frequency of the surrounding reinforced structure, followed by the bridge structure, and the resonant frequencies of these two rectangular structures are much greater than the cantilever beam structure.

[0107] Figure 4(c) shows a schematic diagram of the relationship between the composite layer width and the resonant frequency, Figure 4(d) shows a schematic diagram of the relationship between the piezoelectric layer thickness and the resonant frequency, and Figure 4(e) shows a schematic diagram of the relationship between the support layer thickness and the resonant frequency. From these three figures, the relationship between the composite layer width (i.e., the piezoelectric composite unit width), piezoelectric layer thickness, support layer thickness (i.e., dry film material thickness) and the resonant frequency of the MEMS piezoelectric sound unit with cantilever beam structure, bridge structure, and surrounding reinforcement structure is shown. It can be seen that the composite layer width of each unit of the device has a great influence on the resonant frequency of the device. The graph curve presents a parabolic form. The resonant frequency of each unit first increases with the increase of the composite layer width. When the composite layer width of each unit reaches a certain value, the resonant frequency of each unit reaches the highest point, and then decreases with the increase of the composite layer width. It can be seen from Figure 4(c) that the resonant frequency of the unit of the cantilever beam structure is the largest when the composite layer width is 1000um; the magnitude of the resonant frequency of the three curves in Figure 4(d) increases with the increase of the piezoelectric layer thickness of the piezoelectric sound unit of the cantilever beam structure, bridge structure, and surrounding reinforcement structure. The greater the piezoelectric layer thickness, the greater the resonant frequency of each unit; from Figure 4(e), the relationship between the support layer thickness and the resonant frequency of the piezoelectric sound unit of the cantilever beam structure, bridge structure, and surrounding reinforcement structure can be seen. The greater the support layer thickness of each unit of the device, the greater the resonant frequency of each unit. The magnitude of the resonant frequency of the three curves in the figure increases with the increase of the support layer thickness.

[0108] The quality factor of the hybrid MEMS piezoelectric speaker resonance is usually relatively high, with a sharp and narrow peak frequency response. Flexible dry film materials with greater damping can be used to increase system damping to reduce its resonant Q value. At the same time, the aspect ratio of a single rectangular structure MEMS piezoelectric sound unit is optimized. When the aspect ratio is greater than 3, it helps to have more modes in each frequency range. When the low Q value MEMS piezoelectric sound unit is working, it excites its base membrane and harmonic modes, so that multiple relatively wide-band smooth peaks are mixed together and form a frequency response that is relatively flat to the resonant frequency. In addition, an array structure consisting of multiple rectangular structure MEMS piezoelectric sound units with inconsistent resonant frequencies is designed, and the design of resonant mixing further optimizes the speaker bandwidth.

[0109] According to the mixed frequency MEMS piezoelectric speaker of this embodiment, the mechanical resonance point of the low frequency unit of the mixed frequency MEMS piezoelectric speaker can be designed to be within the audible sound range of 20Hz-20kHz. When the first-order resonance frequency of the mixed frequency MEMS piezoelectric speaker is 2-5KHz, under certain driving conditions, the diaphragm driving displacement is large, and the output sound pressure level of the mixed frequency MEMS piezoelectric speaker in the low frequency range is large, and it can be used in low frequency speakers. The mechanical resonance point of the high frequency unit or ultra-high frequency unit of the mixed frequency MEMS piezoelectric speaker can be designed to be greater than 20kHz outside the audible sound. The first and third odd harmonics of the mixed frequency MEMS piezoelectric speaker are outside the audible sound frequency range. The nonlinearity and THD of the mixed frequency MEMS piezoelectric speaker are low, and it can be used in high frequency speakers. The speaker system composed of the low frequency unit, the high frequency unit, and the ultra-high frequency unit will extend the bandwidth of the frequency far higher than 20kHz to the high-resolution field of 80kHz, optimize the speaker bandwidth, and realize high-resolution and ultrasonic applications.

[0110] In order to produce high sound pressure level output in a small-sized structure, it is required that the piezoelectric multilayer membrane structure of the diaphragm can undergo large bending deformation under the action of voltage drive. Under the large deformation of the diaphragm structure, the hybrid MEMS piezoelectric speaker will produce a relatively obvious geometric nonlinear effect. For the cantilever beam structure, bridge structure, and surrounding reinforcement structure with one end constraint of the same size, the maximum displacement produced by the diaphragm drive of the cantilever beam structure is larger, and its geometric nonlinear effect is most obvious. This geometric nonlinear effect causes the hybrid MEMS piezoelectric speaker to produce a higher total harmonic distortion THD, and at the same time limits the further deformation of the diaphragm of the hybrid MEMS piezoelectric speaker and the sound pressure level SPL output. Through the MEMS manufacturing process, the residual stress level of the diaphragm of the hybrid MEMS piezoelectric speaker is adjusted to make the device diaphragm plane flat or have a small convexity, so that the entire diaphragm stress will present a certain compressive stress state, and then apply a suitable DC bias voltage to the hybrid MEMS piezoelectric speaker to reduce the stiffness of the diaphragm and enhance the low-frequency sound pressure level SPL output of the device.

[0111] When the mixed frequency MEMS piezoelectric speaker is used as a speaker or sound generator, in order to produce sound, the larger the effective radiation area of ​​the device, the higher the sound pressure generated. The larger the displacement of the diaphragm radiation area, the higher the sound pressure generated. When a single MEMS piezoelectric sound unit has a limited output sound pressure level SPL under excitation, a mixed frequency MEMS piezoelectric speaker composed of two or more MEMS piezoelectric sound units arranged in cascade with each other can be used to increase the sound pressure level output of the device by increasing the effective radiation area of ​​the mixed frequency MEMS piezoelectric speaker. In a two-way or even multi-way sound system, the diaphragm of the mid- and low-frequency MEMS piezoelectric sound unit has a low stiffness, and a more flexible diaphragm can be designed to improve the sound pressure level (SPL) in the mid- and low-frequency conditions as a woofer, and provide the low-frequency energy required for bass response and active noise reduction. The high-frequency MEMS piezoelectric sound unit is designed with greater stiffness and an inherently high resonant frequency. The high-frequency response can achieve a wider sound field, provide a higher level of clarity and presence for speech, vocals and musical instruments, and achieve ideal performance at high frequencies. Two-way or even multi-way audio technology uses a single driver module to optimize each speaker for a specific frequency range. No additional special circuits are required, reducing the complexity of the crossover system control drive.

[0112] Furthermore, for a MEMS piezoelectric sound generating unit having the same diaphragm structure, as the aspect ratio of the cavity increases, the resonance frequency of the MEMS piezoelectric sound generating unit increases.

[0113] The length and width of the cavity of the MEMS piezoelectric sound unit with cantilever beam structure, bridge structure, and surrounding reinforcement structure affect the output sound pressure level of the device. The MEMS piezoelectric sound unit with a large aspect ratio can achieve a higher resonance frequency and a larger sound pressure level than the MEMS piezoelectric sound unit with a small aspect ratio. For example, the aspect ratio of the diaphragm of the MEMS piezoelectric sound unit can be greater than 3 to improve the performance of the device. Under the same process conditions (the same preparation process and compact structural layout can be achieved), piezoelectric MEMS array speakers with two-way and multi-way division are made. The MEMS tweeter and MEMS mid-bass speaker are seamlessly integrated into a package. Compared with the traditional method of stacking two speakers together, the thickness of the module is greatly reduced. When in thin headphones or wearable devices with limited space, smaller MEMS array speaker modules increase the space allowed for other components to be used.

[0114] Furthermore, the dry film material in the combined component is used to fill the gap of each piezoelectric composite unit; or, the dry film material in the combined component is used to support each piezoelectric composite unit. In actual implementation, the dry film material can fill the gap of the piezoelectric composite unit, and the piezoelectric composite unit can also be arranged on the upper surface of the dry film material to support the piezoelectric composite unit through the dry film material.

[0115] Furthermore, in each diaphragm, a set of designated edges are anchored to the edge area of ​​the substrate. The diaphragm of each MEMS piezoelectric sound unit is located above the cavity of the MEMS piezoelectric sound unit, and the left and right sides can be fully anchored to the edge of the substrate, or partially anchored to the edge of the substrate. If it is a four-sided reinforced structure, the four sides of the diaphragm, the top, bottom, left, and right sides, are all anchored to the edge of the substrate.

[0116] Furthermore, for each piezoelectric composite unit in each MEMS piezoelectric sound unit, at least a portion of the piezoelectric composite unit belongs to the diaphragm; the position of the piezoelectric composite unit in the diaphragm is: the middle area of ​​the diaphragm, or, at least a portion of the piezoelectric composite unit is anchored in the edge area of ​​the substrate.

[0117] The diaphragm is located above the substrate, and at least one piezoelectric composite unit is arranged on the rigid-flexible film. The piezoelectric composite unit can deform the diaphragm under the drive of an electrical signal, thereby generating sound waves. The piezoelectric composite unit can be located in the middle area of ​​the diaphragm, or a part of it can be located on the diaphragm, and the other part can be anchored in the edge area of ​​the substrate. Each piezoelectric composite unit includes: a single piezoelectric layer structure: electrode layer / piezoelectric layer / electrode layer, a double piezoelectric layer structure: electrode layer / piezoelectric layer / electrode layer / piezoelectric layer / electrode layer, and a multi-piezoelectric layer structure: electrode layer / piezoelectric layer / electrode layer... / piezoelectric layer / electrode layer; the material of the piezoelectric layer can be a single crystal piezoelectric material or a polycrystalline material, for example: single crystal AlN (aluminum nitride), doped AlN, single crystal PZT, doped PZT, ZnO (zinc oxide) and other piezoelectric materials. The thickness of the piezoelectric layer is usually between 0.1 and 25um, and the thickness of each piezoelectric layer can be the same or different. The material of the electrode layer can be a single crystal metal material such as Mo, Pt, Al, Cu, or an alloy material. The thickness of the electrode layer is usually between 0.01 and 5 um, the thickness of each electrode layer can be the same, and the thickness of different electrode layers can be different.

[0118] Furthermore, there is a gap between the diaphragms of every two adjacent MEMS piezoelectric sound generating units.

[0119] Each MEMS piezoelectric sound unit can make sound independently. Two or more MEMS piezoelectric sound units form an array structure to increase the sound area of ​​the device and improve the output sound pressure level. Under the same process conditions, the MEMS piezoelectric sound unit has the same membrane structure, and each unit is closely connected to achieve a compact structural layout. And each unit structure can be a cantilever beam structure, a bridge structure, and a four-sided reinforcement structure, and then form a variety of array structures with two-way and multi-way MEMS piezoelectric sound units. The structure of the mixing MEMS piezoelectric speaker is an array arrangement structure of more than two rectangular units, and the length-to-width ratio of the effective diaphragm of the rectangular structure is greater than 3. The mechanical resonance point of each MEMS piezoelectric sound unit can be the same or different. The device improves the sound pressure level within the design frequency band of a single mixing MEMS piezoelectric speaker through mixing, and expands the SPL response, extending the high frequency to 80kHz.

[0120] The neutral layer of the entire film of the hybrid MEMS piezoelectric speaker is designed at the center of the diaphragm, which is conducive to the opposite forces (tension and pressure) of the upper and lower piezoelectric composite layers of the diaphragm. Through reasonable electrical connections and applying appropriate drives, the output sound pressure level of the device can be improved. In specific implementation, according to the size of the emission sound pressure level, a single MEMS piezoelectric sound unit can be used, or multiple MEMS piezoelectric sound units arranged in a preset array can be driven by electrical parallel to increase the output sound pressure level, wherein the diaphragms of multiple MEMS piezoelectric sound units are coplanar.

[0121] An electric field is applied to the piezoelectric layer of each MEMS piezoelectric sound unit, wherein the first piezoelectric layer and the second piezoelectric layer belonging to the same MEMS piezoelectric sound unit have an electric field in the same direction, and the deflection directions of the silicon films on both sides of the electric field applied to the piezoelectric layer of each MEMS piezoelectric sound unit are opposite. The polarization directions of the piezoelectric layers in all MEMS piezoelectric sound units on the upper side of the diaphragm structure are the same, and the piezoelectric layer electric field in the composite layer of multiple MEMS piezoelectric sound units is electrically divided by the electrodes, and the electrode layers are connected in parallel.

[0122] In the array of mixed frequency MEMS piezoelectric speakers, the cavity size of each MEMS piezoelectric sound unit is consistent. It is only by designing different forms of MEMS piezoelectric sound units, such as cantilever beam structure, bridge structure and four-sided fixed structure, and by adjusting the width parameters of the piezoelectric composite unit on the diaphragm to set the resonant frequency of each MEMS piezoelectric sound unit, that the mixed frequency MEMS piezoelectric speakers with two-way and multi-way division can be made under the same process conditions (same preparation process, compact structure layout can be achieved). In this way, the cavity size of each MEMS piezoelectric sound unit in the mixed frequency MEMS piezoelectric speaker is consistent, and there will be no phenomenon of over-etching or incomplete etching of the top of the cavity structure due to inconsistent etching cavity size, which reduces the problem of inaccurate and poor consistency of cavity etching, which has a great impact on device performance, and there will be no problem of over-etching or incomplete etching affecting the vibration frequency and sensitivity of the device, and in severe cases, the device structure may be damaged and fail. Therefore, this solution can not only obtain a mixed-frequency MEMS piezoelectric speaker with good performance consistency, but also greatly improve the yield of the entire on-chip device.

[0123] FIG5(a) is a schematic diagram of a cross-sectional view of a mixing frequency MEMS piezoelectric speaker, FIG5(b) is a schematic diagram of a top view of a mixing frequency MEMS piezoelectric speaker, FIG5(c) is a schematic diagram of a top view of another mixing frequency MEMS piezoelectric speaker, and FIG5(d) is a schematic diagram of a top view of another mixing frequency MEMS piezoelectric speaker, wherein FIG5(a) can adopt any one of the structures in FIG5(b), FIG5(c), and FIG5(d); the left and right MEMS piezoelectric sound generating units can adopt the same diaphragm structure, and the middle MEMS piezoelectric sound generating unit can adopt the same diaphragm structure. The EMS piezoelectric sound unit adopts a different diaphragm structure. In Figure 5(b), the diaphragms of the three MEMS piezoelectric sound units are all reinforced structures on all sides. There is a gap in the middle of the diaphragms of the left and right MEMS piezoelectric sound units, and there is no gap in the middle of the diaphragm of the middle MEMS piezoelectric sound unit; in Figure 5(c), the diaphragms of the left and right MEMS piezoelectric sound units have a bridge structure, and the diaphragm of the middle MEMS piezoelectric sound unit has a reinforced structure on all sides; in Figure 5(d), the diaphragms of the three MEMS piezoelectric sound units all have a bridge structure.

[0124] 6(a) is a schematic diagram of a cross-sectional view of another mixing MEMS piezoelectric speaker, FIG6(b) is a schematic diagram of a top view of another mixing MEMS piezoelectric speaker, FIG6(c) is a schematic diagram of a top view of another mixing MEMS piezoelectric speaker, and FIG6(d) is a schematic diagram of a top view of another mixing MEMS piezoelectric speaker, wherein FIG6(a) can adopt any one of the structures in FIG6(b), FIG6(c), and FIG6(d); the left and right MEMS piezoelectric sound units can adopt the same diaphragm structure, and the middle MEMS piezoelectric sound unit can adopt a different diaphragm structure. The structure of the diaphragm, in Figure 6(b), the structure of the diaphragm of the left and right MEMS piezoelectric sound units is a structure reinforced all around, and the structure of the diaphragm of the middle MEMS piezoelectric sound unit is a cantilever beam structure; in Figure 6(c), the diaphragms of the three MEMS piezoelectric sound units are all bridge structures, wherein there is no gap in the middle of the diaphragms of the left and right MEMS piezoelectric sound units, and there is a gap in the middle of the diaphragm of the middle MEMS piezoelectric sound unit; in Figure 6(d), the structure of the diaphragm of the left and right MEMS piezoelectric sound units is a structure reinforced all around, and the structure of the diaphragm of the middle MEMS piezoelectric sound unit is a cantilever beam structure.

[0125] 7(a) is a schematic diagram of a cross-sectional view of another mixing frequency MEMS piezoelectric speaker, FIG7(b) is a schematic diagram of a top view of another mixing frequency MEMS piezoelectric speaker, FIG7(c) is a schematic diagram of a top view of another mixing frequency MEMS piezoelectric speaker, and FIG7(d) is a schematic diagram of a top view of another mixing frequency MEMS piezoelectric speaker, wherein FIG7(a) can adopt any one of the structures in FIG7(b), FIG7(c), and FIG7(d); the left and right MEMS piezoelectric sound units can adopt the same diaphragm structure, and the middle MEMS piezoelectric sound unit adopts a different diaphragm structure. In FIG7(b), the diaphragm structure of the left and right MEMS piezoelectric sound units is a surrounding reinforcement structure, and the diaphragm structure of the middle MEMS piezoelectric sound unit is a bridge structure; in FIG7(c), the diaphragms of the three MEMS piezoelectric sound units are all bridge structures; in FIG7(d), the diaphragms of the three MEMS piezoelectric sound units are surrounding reinforcement structures.

[0126] 8(a) is a schematic diagram of a cross-sectional view of another mixing MEMS piezoelectric speaker, FIG8(b) is a schematic diagram of a top view of another mixing MEMS piezoelectric speaker, FIG8(c) is a schematic diagram of a top view of another mixing MEMS piezoelectric speaker, and FIG8(d) is a schematic diagram of a top view of another mixing MEMS piezoelectric speaker; as shown in FIG8(a), any one of the structures in FIG8(b), FIG8(c), and FIG8(d) can be adopted; in FIG8(b), the structure of the diaphragm of the left and right MEMS piezoelectric sound units is a bridge structure, and the structure of the diaphragm of the middle MEMS piezoelectric sound unit is a surrounding reinforcement structure; in FIG8(c), the structure of the diaphragm of the left and right MEMS piezoelectric sound units is a surrounding reinforcement structure, and the structure of the diaphragm of the middle MEMS piezoelectric sound unit is a cantilever beam structure; in FIG8(d), the structure of the diaphragms of the three MEMS piezoelectric sound units are all cantilever beam structures.

[0127] 9(a) is a schematic diagram of a cross-sectional view of another mixing MEMS piezoelectric speaker, FIG9(b) is a schematic diagram of a top view of another mixing MEMS piezoelectric speaker, FIG9(c) is a schematic diagram of a top view of another mixing MEMS piezoelectric speaker, and FIG9(d) is a schematic diagram of a top view of another mixing MEMS piezoelectric speaker; as shown in FIG9(a), any one of the structures shown in FIG9(b), FIG9(c), and FIG9(d) can be adopted; in FIG9(b), the structure of the diaphragm of the left and right MEMS piezoelectric sound units is a cantilever beam structure, and the structure of the diaphragm of the middle MEMS piezoelectric sound unit is a bridge structure; in FIG9(c), the structure of the diaphragm of the left and right MEMS piezoelectric sound units is a cantilever beam structure, and the structure of the diaphragm of the middle MEMS piezoelectric sound unit is a surrounding reinforcement structure; in FIG9(d), the structure of the diaphragm of the left and right MEMS piezoelectric sound units is a cantilever beam structure, and the structure of the diaphragm of the middle MEMS piezoelectric sound unit is a surrounding reinforcement structure.

[0128] 10(a) is a schematic diagram of a cross-sectional view of another mixing MEMS piezoelectric speaker, FIG10(b) is a schematic diagram of a top view of another mixing MEMS piezoelectric speaker, FIG10(c) is a schematic diagram of a top view of another mixing MEMS piezoelectric speaker, and FIG10(d) is a schematic diagram of a top view of another mixing MEMS piezoelectric speaker; as shown in FIG10(a), any one of the structures in FIG10(b), FIG10(c) and FIG10(d) may be adopted; FIG10(b) ), the structure of the diaphragm of the left and right MEMS piezoelectric sound units is a structure reinforced all around, and the structure of the diaphragm of the middle MEMS piezoelectric sound unit is a cantilever beam structure; in Figure 10(c), the structure of the diaphragm of the left and right MEMS piezoelectric sound units is a bridge structure, and the structure of the diaphragm of the middle MEMS piezoelectric sound unit is a cantilever beam structure; in Figure 10(d), the structure of the diaphragm of the left and right MEMS piezoelectric sound units is a structure reinforced all around, and the structure of the diaphragm of the middle MEMS piezoelectric sound unit is a cantilever beam structure.

[0129] FIG11(a) is a schematic diagram of a cross-sectional view of another mixing MEMS piezoelectric speaker, FIG11(b) is a schematic diagram of a top view of another mixing MEMS piezoelectric speaker, FIG11(c) is a schematic diagram of a top view of another mixing MEMS piezoelectric speaker, and FIG11(d) is a schematic diagram of a top view of another mixing MEMS piezoelectric speaker; FIG11(a) can adopt any one of the structural forms of FIG11(b), FIG11(c), and FIG11(d); in FIG11(b), three MEMS piezoelectric sound generators are shown. The structure of the diaphragm of the unit is a structure reinforced on all sides, wherein the diaphragms of the left and right MEMS piezoelectric sound units have a gap in the middle, while the diaphragm of the middle MEMS piezoelectric sound unit has no gap in the middle; in Figure 11(c), the structure of the diaphragms of the left and right MEMS piezoelectric sound units are both cantilever beam structures, while the structure of the diaphragm of the middle MEMS piezoelectric sound unit is a bridge structure; in Figure 11(d), the structure of the diaphragms of the left and right MEMS piezoelectric sound units are cantilever beam structures, while the structure of the diaphragm of the middle MEMS piezoelectric sound unit is a structure reinforced on all sides.

[0130] 12(a) is a schematic diagram of a cross-sectional view of another mixing MEMS piezoelectric speaker, FIG12(b) is a schematic diagram of a top view of another mixing MEMS piezoelectric speaker, FIG12(c) is a schematic diagram of a top view of another mixing MEMS piezoelectric speaker, and FIG12(d) is a schematic diagram of a top view of another mixing MEMS piezoelectric speaker; FIG12(a) can adopt any one of the structural forms of FIG12(b), FIG12(c), and FIG12(d); in FIG12(b), the structures of the diaphragms of the three MEMS piezoelectric sound units are all bridge structures; in FIG12(c), the structures of the diaphragms of the three MEMS piezoelectric sound units are all surrounded by reinforcement structures; in FIG12(d), the structures of the diaphragms of the left and right MEMS piezoelectric sound units are bridge structures, and the structure of the diaphragm of the middle MEMS piezoelectric sound unit is a surrounded by reinforcement structure.

[0131] 13(a) is a schematic diagram of a cross-sectional view of another mixing MEMS piezoelectric speaker, FIG13(b) is a schematic diagram of a top view of another mixing MEMS piezoelectric speaker, FIG13(c) is a schematic diagram of a top view of another mixing MEMS piezoelectric speaker, and FIG13(d) is a schematic diagram of a top view of another mixing MEMS piezoelectric speaker; FIG13(a) can adopt any one of the structural forms of FIG13(b), FIG13(c), and FIG13(d); in FIG13(b), the structures of the diaphragms of the left and right two MEMS piezoelectric sound units are all four-sided reinforced structures, and the structure of the diaphragm of the middle MEMS piezoelectric sound unit is a bridge structure; in FIG13(c), the structures of the diaphragms of the three MEMS piezoelectric sound units are all four-sided reinforced structures; in FIG13(d), the structure of the diaphragm of the MEMS piezoelectric sound unit on the left is a four-sided reinforced structure, and the structure of the diaphragms of the two MEMS piezoelectric sound units on the right is a bridge structure.

[0132] 14(a) is a schematic diagram of a cross-sectional view of another mixing frequency MEMS piezoelectric speaker, FIG14(b) is a schematic diagram of a top view of another mixing frequency MEMS piezoelectric speaker, FIG14(c) is a schematic diagram of a top view of another mixing frequency MEMS piezoelectric speaker, and FIG14(d) is a schematic diagram of a top view of another mixing frequency MEMS piezoelectric speaker; FIG14(a) can adopt any one of the structural forms of FIG14(b), FIG14(c), and FIG14(d); in FIG14(b), the structures of the diaphragms of the three MEMS piezoelectric sound units are all reinforced structures on all sides; in FIG14(c), the structures of the diaphragms of the three MEMS piezoelectric sound units are all bridge structures; in FIG14(d), the structure of the diaphragm of the MEMS piezoelectric sound unit on the left is a cantilever beam structure, and the structures of the diaphragms of the two MEMS piezoelectric sound units on the right are reinforced structures on all sides.

[0133] 15(a) is a schematic diagram of a cross-sectional view of another mixing frequency MEMS piezoelectric speaker, FIG15(b) is a schematic diagram of a top view of another mixing frequency MEMS piezoelectric speaker, FIG15(c) is a schematic diagram of a top view of another mixing frequency MEMS piezoelectric speaker, and FIG15(d) is a schematic diagram of a top view of another mixing frequency MEMS piezoelectric speaker; FIG15(a) can adopt any one of the structural forms of FIG15(b), FIG15(c), and FIG15(d); in FIG15(b), the structures of the diaphragms of the three MEMS piezoelectric sound units are all reinforced structures on all sides; in FIG15(c), the structures of the diaphragms of the left and right MEMS piezoelectric sound units are all reinforced structures on all sides, and the structure of the diaphragm of the middle MEMS piezoelectric sound unit is a bridge structure; in FIG15(d), the structures of the diaphragms of the three MEMS piezoelectric sound units are all bridge structures.

[0134] The hybrid MEMS piezoelectric speaker improves the traditional speaker architecture, replacing the centralized coil / magnet drive with a monolithic piezoelectric drive. The speed and accuracy of the MEMS piezoelectric linear actuator are higher than the voice coil, thus achieving higher bandwidth and more accurate high-frequency reproduction. Under the same process conditions (same preparation process, compact structure layout), piezoelectric MEMS array speakers with two-way and multi-way division are made for the entire audible frequency range; seamlessly integrating the tweeter and mid-bass unit of the MEMS piezoelectric speaker with an array structure into a package helps to achieve a wider acoustic bandwidth, and the thickness of the device is greatly reduced compared to the traditional method of stacking two speakers together. When in thin headphones or wearable devices with limited space, smaller hybrid MEMS piezoelectric speakers increase the space allowed for other components.

[0135] This solution can produce two, three or even multiple MEMS piezoelectric speakers with different resonant frequencies under the same process conditions (same preparation process, compact structure layout can be achieved). A cantilever beam structure, bridge structure and surrounding reinforcement structure are mixed to form a MEMS piezoelectric speaker with two-way and multi-way division. Each MEMS piezoelectric sound unit in the MEMS piezoelectric speaker can be optimized for a specific frequency band. The low-frequency unit provides a high sound pressure level (SPL) in the low-frequency range, and the mid-high frequency unit presents perfect mid-high frequency performance at mid-high frequencies. The sound pressure level in the design frequency band of a single MEMS piezoelectric speaker is improved by mixing, and the SPL response is extended to extend the high frequency to 100kHz. The problem of low output sound pressure level SPL in the low-frequency band, narrow audio frequency response and insufficient flatness of the sound pressure level frequency curve in the audio range of existing MEMS piezoelectric speakers is solved.

[0136] This solution uses the same MEMS process technology to produce a single wide-band, high sound pressure level, low THD hybrid MEMS piezoelectric speaker on the same wafer, which can achieve full-range coverage and extend to ultra-high frequency bands, while providing near-ideal audio characteristics such as high fidelity and low distortion, replacing traditional multiple ring-iron speaker combinations and this 2-way audio hybrid audio module speaker. Provide TWS manufacturers with a variety of configuration options to best suit their personal headphone designs.

[0137] The above-mentioned mixing frequency MEMS piezoelectric speaker can bring the following technical benefits:

[0138] 1) This scheme adopts a MEMS piezoelectric sound unit with a rectangular structure. Compared with the MEMS piezoelectric sound unit with a circular or square structure, there is a lack of even modes in the response spectrum. Multiple modes are excited in a narrow frequency range, and the excited modes are all odd modes. In the structural design, the larger the aspect ratio, the smaller the difference between the resonant frequencies, and the multiple modes are merged together, so the frequency bandwidth is significantly widened.

[0139] 2) Multiple cantilever beam structures, bridge structures or surrounding reinforcement structures of the same size and different resonant frequencies can be designed, and through reasonable layout, two-way and multi-way mixing MEMS piezoelectric speaker array devices can be formed, so that multiple relatively wide-band smooth peaks can be mixed together to form a relatively flat frequency response relative to the resonant frequency, thereby improving the bandwidth of the device while improving the sound pressure level of the speaker.

[0140] 3) Similar structure, compact layout, consistent device film material and thickness parameters, consistent preparation process, easy control, and good design flexibility. Piezoelectric MEMS actuators respond faster and provide a larger range of motion, achieving rich and loud audio over the full bandwidth (20Hz-20kHz) and even in the ultrasonic frequency range.

[0141] 4) Flexible materials are used to fill the rigid area, reducing the moving mass of the diaphragm. When the force of the motor is equal, the lower the mass, the higher the diaphragm acceleration, achieving high sound pressure level SPL at high frequencies and achieving excellent high-frequency response.

[0142] 5) When the cavity size of each MEMS piezoelectric sound unit in the mixed frequency MEMS piezoelectric speaker is consistent, the accuracy, consistency and uniformity of the cavity etching of the entire wafer can be improved, thereby ensuring the consistency of device performance and greatly improving the yield of the entire on-chip device.

[0143] 6) The preparation process is consistent, and under the same size and area, the rectangular array structure has higher performance. Using multiple units to form the array structure can increase the sound area, thereby further improving the sound pressure level.

[0144] 7) Grooving releases the stress of the piezoelectric composite unit and the silicon vibration membrane, and covering them with flexible materials can prevent sound leakage and improve the low-frequency SPL of the speaker. At the same time, a rigid-flexible vibration membrane can be formed by flexible materials, which can increase the system damping to reduce its resonance Q value, thereby making the sound pressure level curve smoother and reducing the total harmonic interference THD.

[0145] The embodiment of the present invention further provides a method for manufacturing a mixed frequency MEMS piezoelectric speaker, the method comprising the following steps:

[0146] Step A, thermally growing an insulating layer on the upper surface and the lower surface of a preset first substrate respectively, to obtain a first substrate with an insulating layer;

[0147] The above-mentioned preset first substrate can be an SOI substrate, a silicon substrate, glass, sapphire, etc.; the upper surface and the lower surface of the preset first substrate can be thermally grown with an insulating layer, the thickness of the insulating layer is between 0.1 and 5 um, and can be made of insulating materials such as silicon nitride and aluminum oxide. Fig.16 A schematic cross-sectional view of a substrate with an insulating layer grown thereon is shown.

[0148] Step B, depositing a seed layer, a bottom electrode layer, an intermediate layer and a top electrode layer in sequence from bottom to top on the insulating layer on the upper surface of the first substrate with the insulating layer, to obtain a first device; wherein the intermediate layer is a piezoelectric layer, or the intermediate layer is formed by overlapping and stacking a plurality of piezoelectric layers and a plurality of electrode layers; wherein the upper layer of the bottom electrode layer and the lower layer of the top electrode layer are both piezoelectric layers;

[0149] The seed layer can be made of amorphous silicon or other materials. The seed layer is formed in the early stage of electrodeposition, which is conducive to the high-quality growth of the piezoelectric layer. Specifically, the seed layer can be grown on the insulating layer on the upper surface of the first substrate, the bottom electrode layer can be deposited on the seed layer, and then the middle layer and the top electrode layer can be grown on the bottom electrode layer; for example, the piezoelectric layer can be a single crystal PZT film with a thickness of 2um, and the top electrode can be Pt with a thickness of 0.1um. When forming the bottom electrode layer, the middle layer and the top electrode layer, they can be deposited layer by layer from bottom to top by sputtering; the materials of the bottom electrode and the top electrode layer can be selected from but not limited to one of the following materials or a combination thereof: single crystal or polycrystalline metal films such as aluminum, copper, gold, platinum, molybdenum, chromium, titanium, etc., metal conductive films such as lanthanum nickel oxide (LNO), SRO, or composite metal layers, such as gold / chromium composite layers.

[0150] The material of the piezoelectric layer can be selected from but not limited to one of the following materials or a combination thereof: polycrystalline or single crystal piezoelectric films such as ZnO, AlN, PZT, PVDF, or doped piezoelectric films, such as AlN doped piezoelectric films; the thickness of the piezoelectric layer is between 0.1μm and 15μm. The materials of the piezoelectric layers in the piezoelectric composite unit can be the same material or different materials. The thickness of each piezoelectric layer is consistent, and the thickness of different piezoelectric layers can be different. Fig.17 A schematic diagram of a first device is shown.

[0151] Step C, etching the first device to obtain a MEMS piezoelectric sound unit; wherein the MEMS piezoelectric sound unit has a cavity, and the structure of the diaphragm of the MEMS piezoelectric sound unit is a cantilever beam structure, a bridge structure or a surrounding reinforcement structure.

[0152] Step D, repeatedly performing the steps of thermally growing an insulating layer on the upper surface and the lower surface of a preset first substrate to obtain a plurality of MEMS piezoelectric sound units; arranging the plurality of MEMS piezoelectric sound units in an array structure to obtain a mixed-frequency MEMS piezoelectric speaker; wherein, if the size of the cavity of each MEMS piezoelectric sound unit is the same, then there are at least two MEMS piezoelectric sound units whose diaphragm structures are different; if the size of the cavity of at least one MEMS piezoelectric sound unit is different from the size of the cavity of other MEMS piezoelectric sound units, then the diaphragms of the plurality of MEMS piezoelectric sound units have the same or different structures.

[0153] The first device can be subjected to graphing processing, cavity etching processing, etc. to obtain a MEMS piezoelectric sound unit. The structure of the diaphragm of the MEMS piezoelectric sound unit can be selected as a cantilever beam structure, a bridge structure, or a surrounding reinforcement structure according to actual needs; the aforementioned steps can be repeated to obtain multiple MEMS piezoelectric sound units, and the obtained multiple MEMS piezoelectric sound units can be arranged in an array structure to obtain a mixed-frequency MEMS piezoelectric speaker.

[0154] The manufacturing method of the mixing frequency MEMS piezoelectric speaker can improve the output sound pressure level SPL of the mixing frequency MEMS piezoelectric speaker and widen the audio frequency response range by arranging multiple MEMS piezoelectric sound generating units with different diaphragm structures and / or different cavity sizes.

[0155] The embodiment of the present invention further provides another method for manufacturing a mixed frequency MEMS piezoelectric speaker. The method is implemented on the basis of the method of the above embodiment, and the method comprises the following steps:

[0156] Step 1: thermally grow insulating layers on the upper surface and the lower surface of the preset first substrate respectively to obtain a first substrate with an insulating layer;

[0157] Step 2: on the insulating layer on the upper surface of the first substrate with the insulating layer, a seed layer, a bottom electrode layer, an intermediate layer and a top electrode layer are deposited in sequence from bottom to top to obtain a first device; wherein the intermediate layer is a piezoelectric layer, or the intermediate layer is formed by overlapping and stacking a plurality of piezoelectric layers and a plurality of electrode layers; wherein the upper layer of the bottom electrode layer and the lower layer of the top electrode layer are both piezoelectric layers;

[0158] Step three, patterning the bottom electrode layer, the middle layer and the top electrode layer in the first device to form at least one piezoelectric composite unit, thereby obtaining a patterned second device;

[0159] Specifically, the bottom electrode layer, the middle layer and the top electrode layer in the first device can be etched layer by layer from top to bottom by dry etching and wet etching, so as to pattern the bottom electrode layer, the middle layer and the top electrode layer to form one or more piezoelectric composite units. The PZT piezoelectric layer and the bottom electrode are etched at an angle, which is conducive to the step coverage of the passivation layer. The etching angle of the PZT piezoelectric layer and the bottom electrode can be: 30°-60°, and the overetching amount is <10%; the smaller the roughness of the PZT side wall, the better. The goal is that after the passivation layer is deposited, there is no delamination between the sliced ​​PZT and the passivation layer; such as Fig.18 A schematic diagram of a second device is shown.

[0160] Step 4: depositing a passivation layer on the second device to isolate the electrical connection of each top electrode layer, and patterning the passivation layer to form a pad area to lead out the bottom electrode layer and / or the top electrode layer of each piezoelectric composite unit, thereby obtaining a third device;

[0161] Step 5, depositing a metal layer on each pad region in the third device to form a pad, and bonding a metal wire to obtain a fourth device;

[0162] The deposited passivation layer can isolate the electrical connection of each electrode layer, play an insulating and protective role, and the patterned passivation layer leads to the bottom electrode and the top electrode layer for the electrical connection between different piezoelectric composite units and the connection pad PAD. The material of the passivation layer can be silicon dioxide with a thickness of 0.5um. The liftoff process is used to make the metal PAD and routing, and the material can be aluminum-copper alloy with a thickness of 0.75um.

[0163] The passivation layer needs to have good step coverage and be able to fit tightly on each film layer. The material can be selected from but not limited to one or a combination of the following materials: TEOS (tetraethyl orthosilicate), silicon dioxide, aluminum nitride, PI, aluminum oxide and other oxide insulating materials, or non-conductive organic compounds, etc.

[0164] In this embodiment, the top and bottom electrode layers are led out to the welding pad PAD and the composite piezoelectric units in each region are connected by metal wire deposition and patterning. The material can be selected from but not limited to one or a combination of the following materials: gold, copper, aluminum, and other metal materials, or conductive alloy materials, etc.

[0165] Step six, filling the fourth device with a covering dry film material, and performing a patterning process on the dry film material to obtain a fifth device;

[0166] A layer of flexible film material FleFiner 15, ie, dry film material, with a thickness of 10 um may be coated on the fourth device by a laminator.

[0167] Specifically, the dry film material can be thawed at a temperature of 20°C to 25°C, and the conveyor vacuum laminator is used with a temperature of 65°C (50-80°C), a vacuum time of 20 seconds, a pressurization time of 90 seconds, and a pressurization pressure of 0.4MPa. After vacuum lamination, it is necessary to ensure that the dry film can fill the grooves and gaps of the fourth device and fit the fourth device tightly without generating bubbles. Fig.19 The schematic diagram of a device covered with a dry film material is shown; the flexible dry film material not only needs to be compatible with the MEMS process and be graphed, but also needs to be able to work for a long time, resist aging, and have good mechanical properties and temperature stability. The flexible dry film material also requires a certain elongation, and the (CTE) difference between the material and the substrate minimizes the wafer bending. Organic compounds with low density and large Young's modulus can be selected, such as PVI-3, PI, PET, etc., and the material thickness can be 5um to 50um.

[0168] When patterning dry film materials, you can use parallel light exposure machine exposure, PEB / pre-bake (Post-Exposure Bake, post-exposure baking), stripping of carrier film, development, post-curing and other processes to achieve; during development, the developer is 1wt% Na2CO3aq; the liquid temperature is 25-50°C; the spray pressure is 0.1-0.5MPa; the processing time is 40sec. (30-70sec), and then washed with water. If the development time is less than 30 seconds, the development ability is insufficient, which may cause problems such as poor resolution and incomplete development. If the development time exceeds 100 seconds, the physical properties of the dry film may be affected, which may cause oil loss, side corrosion, poor resistance to gold precipitation and other problems. (Box Oven) When thermally curing, select 120°C / 30min; if Fig. 20 A schematic diagram of a fifth device is shown; the flexible dry film material can not only cover the groove gap, but also be used as an insulating layer material to cover the metal connection, thereby playing the role of electrical insulation and waterproofing.

[0169] Step 7, thinning and etching the bottom surface of the first substrate of the fifth device to obtain a MEMS piezoelectric sound unit; wherein the MEMS piezoelectric sound unit has a cavity, and the structure of the diaphragm of the MEMS piezoelectric sound unit is a cantilever beam structure, a bridge structure or a surrounding reinforcement structure;

[0170] In this embodiment, the fifth device is subjected to wafer thinning CMP (Chemical-Mechanical Planarization) to 300um, which is not only conducive to the thinning of the device thickness, but also can reduce the etching time of the back cavity DRIE. The wafer is temporarily bonded to the carrier substrate using bonding glue to thin the wafer, and then the glue is removed to debond. Fig.21 A schematic diagram of a device after thinning is shown.

[0171] The double-sided exposure technology DRIE is used to etch the back cavity base silicon and the SiO2 layer, thereby releasing the diaphragm of the MEMS piezoelectric sound unit. The DRIE etching back cavity etching accuracy of the device has a great influence on the device performance, especially the width dimension change. Therefore, the accuracy of the alignment mark, whether the device etching is accurate, whether the etching is complete, and whether there is over-etching require strict process control. The back cavity etching angle is required to be: 90°±1.5°, without residue and over-etching. Fig. 22 A schematic diagram of a MEMS piezoelectric sound unit is shown.

[0172] Step eight, repeatedly performing the steps of thermally growing an insulating layer on the upper surface and the lower surface of the preset first substrate to obtain a plurality of MEMS piezoelectric sound units; arranging the plurality of MEMS piezoelectric sound units in an array structure to obtain a mixed-frequency MEMS piezoelectric speaker; wherein, if the size of the cavity of each MEMS piezoelectric sound unit is the same, then there are at least two MEMS piezoelectric sound units whose diaphragm structures are different; if the size of the cavity of at least one MEMS piezoelectric sound unit is different from the size of the cavity of other MEMS piezoelectric sound units, then the diaphragms of the plurality of MEMS piezoelectric sound units have the same or different structures.

[0173] The manufacturing method of the above-mentioned mixing MEMS piezoelectric speaker can improve the output sound pressure level SPL of the mixing MEMS piezoelectric speaker and widen the audio frequency response range by setting multiple MEMS piezoelectric sound units with different diaphragm structures and / or different cavity sizes during the preparation process.

[0174] The embodiment of the present invention further provides another method for manufacturing a mixed frequency MEMS piezoelectric speaker. The method is implemented on the basis of the method of the above embodiment, and the method comprises the following steps:

[0175] Step ten, thermally grow insulating layers on the upper surface and the lower surface of the preset first substrate respectively to obtain a first substrate with an insulating layer;

[0176] The above-mentioned preset first substrate can be an SOI substrate, a silicon substrate, glass, sapphire, etc.; the upper surface and the lower surface of the preset first substrate can be thermally grown with an insulating layer, the thickness of the insulating layer is between 0.1 and 30 um, and can be made of insulating materials such as silicon nitride, silicon carbide, and aluminum oxide. Fig.23 A schematic cross-sectional view of a substrate with an insulating layer grown thereon is shown.

[0177] Step eleven, on the insulating layer on the upper surface of the first substrate with the insulating layer, depositing a seed layer, a bottom electrode layer, an intermediate layer and a top electrode layer in sequence from bottom to top to obtain a first device; wherein the intermediate layer is a piezoelectric layer, or the intermediate layer is formed by overlapping and stacking a plurality of piezoelectric layers and a plurality of electrode layers; wherein the upper layer of the bottom electrode layer and the lower layer of the top electrode layer are both piezoelectric layers;

[0178] The material of the bottom electrode may be selected from but not limited to one of the following materials or a combination thereof: single crystal or polycrystalline metal films such as aluminum, copper, gold, platinum, molybdenum, chromium, titanium, etc., metal conductive films such as lanthanum nickel oxide (LNO), SRO, etc., or composite metal layers, such as: gold / chromium composite layers. Preferably, the thickness of the Pt electrode is between 0.05μm and 3μm. In this case, the piezoelectric layer is a single crystal PZT film with a thickness of 1.5um, and the electrode layer is Pt, each with a thickness of 0.15um.

[0179] The material of the piezoelectric layer may be selected from but is not limited to one of the following materials or a combination thereof: polycrystalline or single crystal piezoelectric films such as ZnO, AlN, PZT, PVDF, or doped piezoelectric films, such as AlN doped piezoelectric films; the thickness of the piezoelectric layer is between 0.1μm and 5μm. The intermediate material may be selected from but is not limited to one of the following materials or a combination thereof: single crystal or polycrystalline metal films such as aluminum, copper, gold, platinum, molybdenum, chromium, titanium, metal conductive films such as lanthanum nickel oxide (LNO), SRO, or composite metal layers, such as gold / chromium composite layers. Preferably, the thickness of the PT electrode is between 0.05μm and 3μm. Fig.24 A schematic diagram of a first device is shown.

[0180] Step 12, thermally grow insulating layers on the upper surface and the lower surface of the preset second substrate respectively, and cover the insulating layer on the upper surface of the second substrate with a dry film material to obtain a sixth device;

[0181] The process of covering the dry film material can refer to the relevant description in the above embodiments, which will not be repeated here. Fig.25 A schematic diagram of a sixth device is shown.

[0182] Step 13, bonding the first device on the upper surface of the dry film material of the sixth device so that the top electrode layer of the first device is combined with the dry film material, and removing the first substrate with the insulating layer of the first device to obtain the seventh device;

[0183] The first device is combined with the sixth device through a bonding process, specifically the top electrode layer of the first device is combined with the dry film material, and the first substrate with the insulating layer of the first device is removed to obtain the seventh device, such as Fig.26 A schematic diagram of a seventh device is shown.

[0184] Step 14, patterning the bottom electrode layer, the middle layer and the top electrode layer in the seventh device to form at least one piezoelectric composite unit, thereby obtaining an eighth device after patterning;

[0185] First, wet etch the top PZT piezoelectric layer, and then use IBE (Ion Beam Etching) to dry etch the top PZT piezoelectric layer and the middle electrode layer. Wet etching the top PZT piezoelectric layer is to obtain the middle electrode area to facilitate the electrical connection between the subsequent electrode layers and the lead-out metal wiring. Using IBE dry etching to simultaneously pattern the top PZT piezoelectric layer and the middle electrode layer is beneficial to avoid etching the top PZT piezoelectric layer and the middle electrode layer in steps, and avoid over-etching of the PZT film due to different etching selectivities.

[0186] In this embodiment, the etching angle of the PZT piezoelectric layer is 30°-60°, the over-etching amount of the middle electrode and the bottom electrode is less than 30%; the heat critical dimension CD (Critical Dimension) between the top electrode layer, the top piezoelectric layer, the middle electrode layer, the bottom piezoelectric layer and the bottom electrode layer after patterning is less than 3um. Fig. 27 A schematic diagram of an eighth device is shown.

[0187] Step 15, depositing a passivation layer on the eighth device to isolate the electrical connection of each top electrode layer, and patterning the passivation layer to form a pad area to lead out the bottom electrode layer and / or the top electrode layer of each piezoelectric composite unit, thereby obtaining a ninth device;

[0188] In this embodiment, the material of the passivation layer may be selected from but is not limited to one of the following materials or a combination thereof: silicon dioxide, aluminum nitride, PI, aluminum oxide and other oxide insulating materials, or non-conductive organic compounds and the like.

[0189] Step 16, depositing a metal layer on each pad region in the ninth device to form a pad, and bonding a metal wire to obtain a tenth device;

[0190] The liftoff process can be used to deposit and pattern the metal wires. The material is AL&CU alloy with a thickness of 0.5um. In this embodiment, the metal wire deposition and patterning can realize parallel connection between units, series connection between units, etc. The material can be selected from but not limited to one of the following materials or a combination thereof: gold, copper, aluminum, and other metal materials, or conductive alloy materials, etc.

[0191] Step seventeen, performing a patterning process on the dry film material to obtain an eleventh device;

[0192] The processing method of patterning the dry film material can refer to the method in the above embodiment, which will not be repeated here. Fig.28 A schematic diagram of an eleventh device is shown.

[0193] Step 18, thinning and etching the bottom surface of the second substrate of the eleventh device to obtain a MEMS piezoelectric sound unit; wherein the MEMS piezoelectric sound unit has a cavity, and the structure of the diaphragm of the MEMS piezoelectric sound unit is a cantilever beam structure, a bridge structure or a surrounding reinforcement structure;

[0194] The method of thinning the bottom surface of the second substrate of the eleventh device and etching the cavity can refer to the method in the above-mentioned embodiment, which will not be described in detail here; Fig.29 A schematic diagram of a device after thinning is shown; and Fig.30 A schematic diagram of a device after cavity etching is shown.

[0195] Step nineteen, repeatedly performing the steps of thermally growing an insulating layer on the upper surface and the lower surface of the preset first substrate to obtain a plurality of MEMS piezoelectric sound units; arranging the plurality of MEMS piezoelectric sound units in an array structure to obtain a mixed-frequency MEMS piezoelectric speaker; wherein, if the size of the cavity of each MEMS piezoelectric sound unit is the same, then there are at least two MEMS piezoelectric sound units whose diaphragm structures are different; if the size of the cavity of at least one MEMS piezoelectric sound unit is different from the size of the cavity of other MEMS piezoelectric sound units, then the diaphragms of the plurality of MEMS piezoelectric sound units have the same or different structures.

[0196] The beneficial effects of this embodiment are described below. Fig.31 A schematic diagram of a top view of a mixing frequency MEMS piezoelectric speaker is shown, and Fig.32Taking the schematic diagram of a top view of a mixing MEMS piezoelectric speaker as an example, the mixing MEMS piezoelectric speaker has a rectangular structure unit size of 6.5mm*1.5mm, and three units form an array structure with a size of 6.5mm*6.0mm. Single-layer PZT piezoelectric layer structure: SOI substrate silicon wafer, Si / PT / PZT / PT / =2um / 0.15um / 2um / 0.15um, where both structures are rectangular structures reinforced on four sides, the main difference is that the second chip adopts a two-way frequency design, the width of the middle cavity is designed to be relatively wide, the unit resonant frequency is low, the unit cavity width on both sides is relatively narrow, and the unit resonant frequency is high. Each unit transducer in the MEMS piezoelectric array speaker can be optimized for a specific frequency band, where the low-frequency unit provides a high sound pressure level (SPL) in the low-frequency range, and the mid-high frequency unit presents perfect mid-high frequency performance at mid-high frequencies, and the sound pressure level in the design frequency band of a single MEMS piezoelectric speaker is improved by mixing. Two MEMS chips were obtained through MEMS process tape-out. After the samples were packaged, acoustic tests were performed and the following results were obtained: Fig.33 A schematic diagram of a sound pressure level-frequency curve is shown in FIG. 1 , in which the light grey curve is the sample Y4M05-3 (corresponding to Fig.31 The dark grey curve is the curve corresponding to the structure in Fig.32 From the curve corresponding to the structure in the figure, we can see that: Sample Y4M05-3: SPL = 113dB at 1KHz, the low frequency <100Hz has a dive, and the sound pressure level SPL>120dB in the frequency range of 4.0KHz to 20KHz; Sample Y4M06-1: SPL = 116dB at 1KHz, the low frequency <100Hz also has a dive, and the frequency range of 1.5KHz to 20KHz is SPL>120dB. Compared with Y4M05, the sound pressure level SPL of sample Y4M06-1 is 3dB higher at 1KHz, and the frequency range of sound pressure level SPL>120dB is 1.5KHz to 20KHz, which is wider than Y4M05-3. The main reason is that the frequency below the mechanical resonance frequency of the vibration system is the support compliance control area, and the change in stiffness affects the low frequency band; in the frequency range of 20kHz-10kHz, the sound pressure level of the dark gray curve is higher than that of the light gray curve, which shows that the low-frequency unit performance of the sample Y4M06-1 speaker array is better, and the sound pressure level of the mid- and low-frequency is improved. This is mainly because in the mid-frequency area, the system is controlled by the vibration of the mass Mms. The larger the Mms, the lower the sound pressure. In this area, the sound pressure is independent of the frequency. It can be seen that this multi-frequency design can be used to solve the problems of the existing piezoelectric MEMS speakers in the low-frequency band output sound pressure level SPL, the narrow audio frequency response range, and the sound pressure level frequency curve in the audio range is not flat enough.

[0197] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention, rather than to limit it. Although the present invention has been described in detail with reference to the aforementioned embodiments, those skilled in the art should understand that they can still modify the technical solutions described in the aforementioned embodiments, or replace some or all of the technical features therein with equivalents. However, these modifications or replacements do not cause the essence of the corresponding technical solutions to deviate from the scope of the technical solutions of the embodiments of the present invention.

Claims

1. A mixing frequency MEMS piezoelectric speaker, characterized in that: include: Two or more MEMS piezoelectric sound generating units arranged in an array structure; Each of the MEMS piezoelectric sound generating units comprises: a substrate and a composite component located on the upper surface of the substrate; the composite component comprises at least one piezoelectric composite unit and a dry film material; the middle area of ​​the back of the substrate is a cavity; in the composite component, the area opposite to the cavity forms a diaphragm; the structure of the diaphragm is a cantilever beam structure, a bridge structure or a surrounding reinforcement structure; Wherein, if the size of the cavity of each MEMS piezoelectric sound generating unit is the same, then there are at least two MEMS piezoelectric sound generating units with different structures of the diaphragms; If the size of the cavity of at least one of the MEMS piezoelectric sound generating units is different from the size of the cavity of other MEMS piezoelectric sound generating units, the structures of the diaphragms of the plurality of MEMS piezoelectric sound generating units are the same or different.

2. The loudspeaker according to claim 1, characterized in that If the size of the cavity and the thickness of each membrane layer of each MEMS piezoelectric sound unit are the same, the resonant frequency of the MEMS piezoelectric sound unit whose diaphragm is a bridge-type structure is lower than the resonant frequency of the MEMS piezoelectric sound unit whose diaphragm is a structure reinforced all around, and is higher than the resonant frequency of the MEMS piezoelectric sound unit whose diaphragm is a cantilever beam structure.

3. The loudspeaker according to claim 1, characterized in that For MEMS piezoelectric sound generating units with the same diaphragm structure, when the length of the cavity is consistent, the greater the aspect ratio, the higher the resonant frequency of the MEMS piezoelectric sound generating unit.

4. The loudspeaker according to claim 1, characterized in that The dry film material in the combined component is used to fill the gap of each of the piezoelectric composite units; or, the dry film material in the combined component is used to support each of the piezoelectric composite units.

5. The loudspeaker according to claim 1, characterized in that In each of the diaphragms, a set of opposite designated edges are anchored in an edge region of the substrate.

6. The loudspeaker according to claim 1, characterized in that For each piezoelectric composite unit in each MEMS piezoelectric sound unit, at least a portion of the piezoelectric composite unit belongs to the diaphragm; the position of the piezoelectric composite unit in the diaphragm is: the middle area of ​​the diaphragm, or at least a portion of the piezoelectric composite unit is anchored in the edge area of ​​the substrate.

7. The loudspeaker according to claim 1, characterized in that The supporting layer between each two adjacent MEMS piezoelectric sound units is etched away to form a gap, which can isolate the influence of the vibration of adjacent units and help reduce crosstalk.

8. A method for manufacturing a mixed frequency MEMS piezoelectric speaker, characterized in that: The method comprises: Thermally grow insulating layers on the upper surface and the lower surface of the preset first substrate respectively to obtain a first substrate with an insulating layer; On the insulating layer on the upper surface of the first substrate with the insulating layer, a seed layer, a bottom electrode layer, an intermediate layer and a top electrode layer are deposited in sequence from bottom to top to obtain a first device; wherein the intermediate layer is a piezoelectric layer, or the intermediate layer is formed by overlapping and stacking a plurality of piezoelectric layers and a plurality of electrode layers; wherein the upper layer of the bottom electrode layer and the lower layer of the top electrode layer are both piezoelectric layers; Processing the first device to obtain a MEMS piezoelectric sound unit; wherein the MEMS piezoelectric sound unit has a cavity, and the structure of the diaphragm of the MEMS piezoelectric sound unit is a cantilever beam structure, a bridge structure or a surrounding reinforcement structure; Repeat the steps of thermally growing an insulating layer on the upper surface and the lower surface of a preset first substrate to obtain a plurality of MEMS piezoelectric sound units; arrange the plurality of MEMS piezoelectric sound units in an array structure to obtain a mixed-frequency MEMS piezoelectric speaker; wherein, if the size of the cavity of each of the MEMS piezoelectric sound units is the same, then there are at least two MEMS piezoelectric sound units with different diaphragm structures; if the size of the cavity of at least one of the MEMS piezoelectric sound units is different from the size of the cavity of other MEMS piezoelectric sound units, then the diaphragms of the plurality of MEMS piezoelectric sound units have the same or different structures.

9. The method according to claim 8, characterized in that The step of processing the first device to obtain a MEMS piezoelectric sound generating unit comprises: Performing patterning on the bottom electrode layer, the middle layer and the top electrode layer in the first device to form at least one piezoelectric composite unit, thereby obtaining a patterned second device; Depositing a passivation layer on the second device to isolate the electrical connection of each of the top electrode layers, and patterning the passivation layer to form a pad area to lead out the bottom electrode layer and / or the top electrode layer of each of the piezoelectric composite units, thereby obtaining a third device; Depositing a metal layer on each pad region in the third device to form a pad, and bonding a metal wire to obtain a fourth device; Filling the fourth device with a covering dry film material, and performing a patterning process on the dry film material to obtain a fifth device; The bottom surface of the first substrate of the fifth device is thinned and etched into a cavity to obtain a MEMS piezoelectric sound generating unit.

10. The method according to claim 8, characterized in that The step of processing the first device to obtain a MEMS piezoelectric sound generating unit comprises: Thermally grow insulating layers on the upper surface and the lower surface of the preset second substrate respectively, and cover the insulating layer on the upper surface of the second substrate with a dry film material to obtain a sixth device; Bonding the first device on the upper surface of the dry film material of the sixth device so that the top electrode layer of the first device is combined with the dry film material, and removing the first substrate with the insulating layer of the first device to obtain a seventh device; Performing patterning on the bottom electrode layer, the middle layer and the top electrode layer in the seventh device to form at least one piezoelectric composite unit, thereby obtaining an eighth device after patterning; Depositing a passivation layer on the eighth device to isolate the electrical connection of each top electrode layer, and patterning the passivation layer to form a pad area to lead out the bottom electrode layer and / or the top electrode layer of each piezoelectric composite unit, thereby obtaining a ninth device; Depositing a metal layer on each pad region in the ninth device to form a pad, and bonding a metal wire to obtain a tenth device; Performing a patterning process on the dry film material to obtain an eleventh device; The bottom surface of the second substrate of the eleventh device is thinned and cavitated to obtain a MEMS piezoelectric sound unit.

Images