Acoustic element

Abstract

The present invention is provided with: a voltage generation unit which has a vibrating body 22a1 and a fixed potential electrode 24 that is in contact with the receiving surface of the vibrating body 22a1, and which is capable of generating a voltage on the output surface of the vibrating body 22a1; and an impedance conversion element that sets, as a potential generation unit, a local region of the output surface located at a stress concentration site of the vibrating body 22a1, and that sets the potential of the potential generation unit as a control voltage. The impedance conversion element has a base region 14 of a first conductivity type, and first and second main electrode regions 15a and 15b of a second conductivity type. The potential generation unit is connected to the base region 14 between the first main electrode region 15a and the second main electrode region 15b, and controls a potential barrier generated in a channel between the first main electrode region 15a and the second main electrode region 15b for controlling carrier movement by means of the potential of the potential generation unit.

Description

Technical Field

[0001] The present invention relates to an acoustic element utilizing piezoelectric effect, and in particular to an acoustic element suitable for receiving purposes. Background Art

[0002] As disclosed in non-patent document 1 and patent document 1, a technology for forming a piezoelectric micromachined ultrasonic transducer (PMUT) array in a single chip is known. However, non-patent document 1 discloses only the PMUT array. In addition, in patent document 1, a first chip forming the PMUT array and a second chip forming the integrated circuit are combined together by a bonding component called a bonding layer. Therefore, when the PMUT is used as a receiving element, the distance between the PMUT and the amplifier must be long, which causes a deterioration in the signal-to-noise ratio (S / N ratio) due to the braking impedance of the wiring connecting them.

[0003] Non-patent document 2 points out that by connecting the PMUT to the gate of a field effect transistor (FET) via a wire, it is possible to achieve high sensitivity and broadband by improving the S / N ratio. However, the conclusion of non-patent document 2 points out that since the bandwidth has not been improved, further improving the sensitivity and broadband is still a problem to be solved. In order to realize a highly sensitive and broadband ultrasonic receiver, it is not the macroscopic structure that is important, but the detailed layout inside the acoustic element, such as from which part of the PMUT and how the voltage is extracted. The invention described in non-patent document 2 only studies the separate structure in which the PMUT as a separate element and the FET as a separate element are connected by external wiring. Therefore, in the invention described in non-patent document 2, the microscopic stress distribution in the internal structure of the PMUT and the output voltage distribution in the internal structure related to the stress distribution are not considered. In addition, the invention described in non-patent document 2 does not study the structure that makes the PMUT itself broadband, the structure that most effectively utilizes the PMUT output, and the integrated structure required for broadband including reducing the braking impedance.

[0004] Patent Document 2 and Patent Document 3 show a structure in which a FET as a preamplifier is formed on a substrate, and a PMUT directly connected to the gate of the FET is arranged directly above the gate of the FET. However, in both structures, there is a void or cavity between the PMUT and the gate of the FET, and the two are connected by a long wiring (through hole). This is because in Patent Document 2 and Patent Document 3, the PMUT and the FET are different components, and the PMUT must eventually be combined directly above the FET. In this case, the high sensitivity and broadband of the ultrasonic receiver cannot be fully realized, especially due to the large braking impedance caused by the through hole. In any case, when a system using these PMUTs performs ultrasonic reception, satisfying characteristics such as high sensitivity, broadband, and low output impedance becomes a problem faced by practical application. In addition, receiving sensors that use piezoelectric effects such as PMUTs have very high impedance, and there is a matching problem with amplification circuits such as preamplifiers.

[0005] Prior art literature

[0006] Patent Literature

[0007] Patent Document 1: U.S. Patent No. 10562069 Patent Document 2: Japanese Patent Publication No. 2019-503095 Patent Document 3: Japanese Patent Publication No. 2020-535776

[0008] Non-patent literature Non-patent literature 1: D. Jiao et al., "High Fill Factor Array of Piezoelectric Micromachined Ultrasonic Transducers with Large Quality Factor", Sensors and Materials, Vol. 32, No. 5, 2020, pp. 1785-1795.

[0009] Non-patent document 2: Hiroki Makino et al., "Development of high-sensitive and wideband FET-based ultrasound receiver directly driven by piezoelectric effect", 2015 IEEE International Ultrasonics Symposium Proceedings. Summary of the invention

[0010] Problems to be Solved by the Invention The present invention has been conceived in view of the above-mentioned problems, and an object of the present invention is to provide an acoustic element which can achieve high sensitivity and wide bandwidth, reduce output impedance, and is suitable for a receiving sensor.

[0011] Means for Solving the Problems In order to achieve the above-mentioned object, the acoustic element of the first mode of the present invention includes: (a) a voltage generating unit having a vibrating body and a fixed potential electrode in contact with the receiving surface of the vibrating body, and generating a voltage by piezoelectric effect on the output surface of the vibrating body opposite to the receiving surface (i.e., the surface on the other side of the receiving surface); (b) an impedance conversion element that uses a specific part of the output surface of the stress concentration part of the vibrating body as a potential transfer area and operates with the potential of the potential transfer area as a control voltage. The vibrating body can be composed of a piezoelectric layer (piezoelectric body) in its entirety, or only a part of the stress concentration part can be composed of a piezoelectric layer (piezoelectric body). That is, the voltage generating unit only needs to have a structure in which at least a piezoelectric layer (piezoelectric body) is provided at the stress concentration part of the vibrating body in the vibrating body having a continuous structure as a stress field. In addition, even if the vibrating body is composed of a piezoelectric layer as a whole, the piezoelectric constant of the piezoelectric layer can be distributed, thereby having a vibrating body structure with a non-uniform piezoelectric effect that enhances the piezoelectric effect at the stress concentration part. Therefore, the impedance conversion element uses a specific part of the piezoelectric layer located at the stress concentration point as a potential transfer region and operates using the potential of the potential transfer region as a control voltage. In the acoustic element of the first aspect, the ultrasonic pressure applied to the vibrator is used as an output signal of the impedance conversion element.

[0012] Effects of the Invention According to the present invention, it is possible to provide an acoustic element that achieves high sensitivity and wide bandwidth, reduces output impedance, and is suitable for reception. BRIEF DESCRIPTION OF THE DRAWINGS

[0013] Figure 1A 1 is a schematic plan view showing the overall structure of the acoustic element according to the first embodiment of the present invention, which is a unit cell (unit cell) that can constitute the acoustic element integrated circuit.

[0014] Figure 1B This is a plan view that enlarges a portion of the acoustic element of FIG. 1 and shows the planar layout of the impedance conversion element through the voltage generating portion.

[0015] Figure 2 It means from Figure 1B Schematic cross-sectional view of the structure viewed along the II-II direction.

[0016] Figure 3A This is an equivalent circuit that illustrates a model in which the gate terminal of an MIS transistor is connected to the output of a piezoelectric element.

[0017] Figure 3BThis is an equivalent circuit of the acoustic element of the first embodiment.

[0018] Figure 4A Schematic cross-sectional view showing a parallel toroidal stress distribution enhancement structure of a simulation object.

[0019] Figure 4B It is explained in Figure 4A A cross-sectional view of five potential transfer regions arranged on the output surface of the vibrating body of the stress distribution enhancement structure.

[0020] Figure 4C As Figure 4A Simulation results when aluminum nitride (AlN) is used in the stress distribution enhancement structure, with graphs showing transient response characteristics in five potential transfer regions.

[0021] Figure 4D This is to explain the relationship between the diameter and thickness of the convex portion of the piezoelectric layer composed of AlN and Figure 4C A graph showing transient response characteristics in five potential transfer regions for different conditions.

[0022] Figure 4E This is a graph showing the frequency dependence of voltage sensitivity when lead zirconate titanate (PZT) is used as the piezoelectric layer, with the aspect ratio of the central protrusion being used as a parameter.

[0023] Figure 4F This is a graph showing the frequency dependency of voltage sensitivity when PZT is used as the piezoelectric layer, the aspect ratio of the central protrusion is fixed to 1, and the diameter of the central protrusion is used as a parameter.

[0024] Figure 5A is a schematic cross-sectional view showing another parallel toroidal type stress distribution enhancing structure.

[0025] Figure 5B It is explained in Figure 5A A cross-sectional view of five potential transfer regions arranged on the output surface of the vibrating body of the stress distribution enhancement structure.

[0026] Figure 5C It means in Figure 5A Figure 2. Stress distribution enhancement diagram of the simulation results when PZT is used in the piezoelectric layer of the structure.

[0027] Figure 5D This means that when using PZT, Figure 5A The graph shows the frequency dependence of the voltage sensitivity with the diameter of the protrusion of the piezoelectric layer of the stress distribution enhancement structure as a parameter.

[0028] Figure 5E This is a model cross-sectional view for explaining the asymmetric vibration when the piezoelectric layer of the single piezoelectric crystal structure is bent upward and downward.

[0029] Fig. 5F It means in Figure 5A The stress distribution is enhanced when PZT is used in the piezoelectric layer of the structure. Fig. 5F (a) represents the stress in the XX direction, Fig. 5F (b) represents the stress in the YY direction, Fig. 5F (c) Graph showing stress in the XY directions.

[0030] Fig. 6A Schematic cross-sectional view showing the arrangement of ideal electrodes for minute potential detection as a simulation object in order to verify the voltage distribution of the detailed stress distribution enhancement structure.

[0031] Figure 6B It means that in Fig. 6A Figure 2. Simulated transient response of an ideal electrode for detecting tiny potentials when using PZT in the piezoelectric layer of the enhanced structure.

[0032] Figure 6C is to have Fig. 6A A diagram illustrating the voltage distribution obtained in a simulation using the diameter of the convex portion in the central part of the vibration cavity as a parameter when PZT is used in the piezoelectric layer of the stress distribution enhancement structure of an ideal electrode for detecting minute potentials.

[0033] Fig. 7A is a cross-sectional view schematically illustrating a V-shaped stress distribution enhancement structure.

[0034] Figure 7B It is explained in Fig. 7A Figure 2. Stress distribution in the piezoelectric layer of the enhanced structure when using PZT and simulated transient response.

[0035] Figure 7C is with Figure 7B A graph showing the transient response simulation results when PZT is used in the piezoelectric layer of the stress distribution enhancement structure for different thicknesses, etc.

[0036] Fig. 8A It is a cross-sectional view schematically illustrating the Λ-shaped stress distribution enhancement structure.

[0037] Figure 8B It means in Fig. 8A Figure 2. Stress distribution in the piezoelectric layer of the enhanced structure when using PZT in the transient response simulation results.

[0038] Fig. 9A is a cross-sectional view schematically illustrating an M-shaped stress distribution enhancement structure.

[0039] Fig. 9B It means in Fig. 9A Figure 2. Stress distribution in the piezoelectric layer of the enhanced structure when using PZT in the transient response simulation results.

[0040] Fig. 10A is a cross-sectional view schematically illustrating a W-shaped stress distribution enhancement structure.

[0041] Fig. 10B It means in Fig. 10A Figure 2. Stress distribution in the piezoelectric layer of the enhanced structure when using PZT in the transient response simulation results.

[0042] Fig.11A is a cross-sectional view schematically illustrating a flat disk-type general structure without enhanced stress distribution.

[0043] Fig. 11B It means in Fig.11A Figure 2 is a diagram of transient response simulation results when PZT is used in the piezoelectric layer of a flat disk-type general structure.

[0044] Fig. 11C It means in Fig.11A Figure 2 is a diagram of transient response simulation results when AlN is used in the piezoelectric layer of a flat disk-type general structure.

[0045] Fig.11D is Fig.11A When AlN is used in the piezoelectric layer of a flat disk-shaped ordinary structure, Fig. 11C This is a diagram illustrating the transient response when the film thickness etc. are different.

[0046] Fig.12 This is a graph showing the frequency dependency of voltage sensitivity when PZT is used as the piezoelectric layer with stress distribution enhancement structures of different shapes as parameters (Part 1).

[0047] Fig.13 This is a graph showing the frequency dependency of voltage sensitivity when PZT is used as the piezoelectric layer with stress distribution enhancement structures of different shapes as parameters (part 2).

[0048] Fig.14 corresponds to Figure 2 A is a schematic cross-sectional view showing a main part of an acoustic element according to a first modified example of the first embodiment of the present invention, showing a structure viewed in the II-II direction.

[0049] Fig.15 corresponds to Figure 2 A is a schematic cross-sectional view showing a main part of an acoustic element according to a second modified example of the first embodiment of the present invention, as viewed in the II-II direction.

[0050] Fig.16A Yes Description Fig.15 A cross-sectional view showing a process of manufacturing an acoustic element according to a second modified example of the first embodiment is shown.

[0051] Fig. 16B Yes Description Fig.16A The following processes are cross-sectional views of the method for manufacturing the acoustic element according to the second modification example of the first embodiment.

[0052] Fig. 16C Yes Description Fig. 16B The following processes are cross-sectional views of the method for manufacturing the acoustic element according to the second modification example of the first embodiment.

[0053] Fig.16D Yes Description Fig. 16C The following processes are cross-sectional views of the method for manufacturing the acoustic element according to the second modification example of the first embodiment.

[0054] Fig.16E Yes Description Fig.16D The following processes are cross-sectional views of the method for manufacturing the acoustic element according to the second modification example of the first embodiment.

[0055] Fig.16F Yes Description Fig.16E The following processes are cross-sectional views of the method for manufacturing the acoustic element according to the second modification example of the first embodiment.

[0056] Figure 16G Yes Description Fig.16F The following processes are cross-sectional views of the method for manufacturing the acoustic element according to the second modification example of the first embodiment.

[0057] Fig.16H Yes Description Figure 16G The following processes are cross-sectional views of the method for manufacturing the acoustic element according to the second modification example of the first embodiment.

[0058] Fig.16I Yes Description Fig.16H The following processes are cross-sectional views of the method for manufacturing the acoustic element according to the second modification example of the first embodiment.

[0059] Fig.16J Yes Description Fig.16I The following processes are cross-sectional views of the method for manufacturing the acoustic element according to the second modification example of the first embodiment.

[0060] Figure 16K Yes Description Fig.16J The following processes are cross-sectional views of the method for manufacturing the acoustic element according to the second modification example of the first embodiment.

[0061] Figure 16L Yes Description Figure 16K The following processes are cross-sectional views of the method for manufacturing the acoustic element according to the second modification example of the first embodiment.

[0062] Fig.17 This is an equivalent circuit diagram for explaining in a model manner the technical concept of the acoustic element according to the second embodiment of the present invention.

[0063] Fig.18A This is a plan view showing a portion of the acoustic element according to the second embodiment by enlarging the voltage generating portion, with attention paid to the impedance conversion element. Fig.18B It means from Fig.18A A cross-sectional view showing an overview of the structure viewed along the XVIIIB-XVIIIB direction.

[0064] Fig.18C is a cross-sectional view showing the general structure of an acoustic element according to a modified example of the second embodiment of the present invention (corresponding to the Fig.18A Cross-sectional structure viewed in the XVIIIB-XVIIIB direction).

[0065] Fig.18D is the material used for the center column as a parameter, indicating Fig.18C FIG. 1 is a graph showing the frequency dependency of the voltage sensitivity of the acoustic element according to the modified example of the second embodiment.

[0066] Fig.19A The diagram is a process cross-sectional view for explaining a process of manufacturing the impedance conversion element side chip as a method of manufacturing the acoustic element according to the second embodiment. Fig.19B is a process cross-sectional view of the acoustic element manufacturing method of the second embodiment, for explaining the following Fig.19A Next, the manufacturing process of the impedance conversion element side chip is carried out.

[0067] Fig.19C is a process cross-sectional view of the acoustic element manufacturing method of the second embodiment, for explaining the following Fig.19B Next, the manufacturing process of the impedance conversion element side chip is carried out.

[0068] Fig.19D is a process cross-sectional view of the acoustic element manufacturing method of the second embodiment, for explaining the following Fig.19C Next, the manufacturing process of the impedance conversion element side chip is carried out.

[0069] Fig.19E is a process cross-sectional view of the acoustic element manufacturing method of the second embodiment, for explaining the following Fig.19D Next, the manufacturing process of the impedance conversion element side chip is carried out.

[0070] Fig.19F It is used to explain and Fig.19E This is a cross-sectional view of a process of manufacturing an acoustic element according to a second embodiment, in which a process flow of manufacturing a new voltage generating unit side chip is started, different from the previous process of manufacturing an impedance conversion element side chip.

[0071] Figure 19Gis a process cross-sectional view of the acoustic element manufacturing method of the second embodiment, for illustrating Fig.19F Next is the manufacturing process of the voltage generating unit chip.

[0072] Fig.19H is a process cross-sectional view of the acoustic element manufacturing method of the second embodiment, for illustrating Figure 19G Next is the manufacturing process of the voltage generating unit chip.

[0073] Fig.19I is a process cross-sectional view of the acoustic element manufacturing method of the second embodiment, for illustrating Fig.19H Next is the manufacturing process of the voltage generating unit chip.

[0074] Fig.19J is a process cross-sectional view of a method for manufacturing an acoustic element according to a second embodiment, for explaining Fig.19I Next is the manufacturing process of the voltage generating unit chip.

[0075] Figure 19K is a process cross-sectional view of a method for manufacturing an acoustic element according to a second embodiment, for explaining the steps of Figures 19A to 19E The impedance conversion element side chip manufactured in the process of Figures 19F to 19J The voltage generating unit side chip manufactured in the process is directly bonded.

[0076] Fig. 20 This is an equivalent circuit diagram for explaining the technical concept of the acoustic element according to the third embodiment of the present invention in a model manner.

[0077] Fig.21A This is a plan view showing a portion of the acoustic element according to the third embodiment by enlarging the voltage generating section with attention paid to the impedance conversion element. Fig. 21B is from Fig.21A Schematic cross-sectional view of the structure viewed in the XXIB-XXIB direction.

[0078] Fig. 22 This is an equivalent circuit diagram for explaining in a model manner the technical concept of the acoustic element according to the modified example of the third embodiment of the present invention.

[0079] Fig.23 It is a schematic cross-sectional view showing the structure of an acoustic element according to a modification of the third embodiment.

[0080] Fig.24 This is an equivalent circuit diagram for explaining in a model manner the technical concept of the acoustic element according to the fourth embodiment of the present invention.

[0081] Fig.25 This is an equivalent circuit diagram for explaining in a model manner the technical concept of the acoustic element according to the fifth embodiment of the present invention.

[0082] Fig.26 yes Fig.25 A schematic cross-sectional view showing the structure of an acoustic element according to a fifth embodiment is shown.

[0083] Fig. 27 FIG. 1 is a plan view for explaining a layout in which an acoustic element integrated circuit is divided into a transmitting element and a receiving element and arranged two-dimensionally as an example of another embodiment.

[0084] Fig.28A This is a plan view for illustrating an example of a planar layout of an impedance conversion element constituting an acoustic element of another embodiment (the structure of the main part of the orthogonal complementary element circuit constituting the impedance conversion element is illustrated through a voltage generating portion located above it).

[0085] Fig.28B It is used to illustrate Fig.28A The diagram illustrates the operation of the orthogonal complementary element circuit shown in FIG. 1 , with the potential barrier of electrons in one main current being changed to a saddle point of the potential distribution.

[0086] Fig.28C It is used to illustrate Fig.28A FIG. 1 is a diagram showing changes in the potential barrier of holes of the other main current in the orthogonal complementary element circuit shown.

[0087] Fig.28D It is used to illustrate in a model way Fig.28A The illustrated diagram is an equivalent circuit diagram of a circuit configuration example in the case where an orthogonal complementary element circuit is used as an impedance conversion element.

[0088] Fig.29 Yes means Fig.28A A plan layout diagram of a modified example of the acoustic element shown.

[0089] Fig.30 This is a graph comparing the frequency dependency of voltage sensitivity when AlN and PZT are used for the central protrusion and when the material of the remaining flat plate-shaped portion is different from that of the central protrusion and when the material is the same as that of the central protrusion.

[0090] Fig.31 This is a model diagram used to explain the stress asymmetry generated when a single piezoelectric chip type structure with a disk-shaped vibration cavity is bent up and down to form a convex surface. DETAILED DESCRIPTION

[0091] Hereinafter, with reference to the accompanying drawings, the acoustic elements of the first to fifth embodiments are described as specific examples of the present invention. In the following description of the acoustic elements of the first to fifth embodiments and the description of the drawings, the same or similar parts are marked with the same or similar figure marks and described uniformly. However, it should be noted that the drawings are schematic, and the relationship between the thickness and the plane size, the size ratio of each component, etc. are different from the actual situation. Therefore, the specific thickness, size, size, etc. should be judged in a more diverse manner based on the main purpose of the technical ideas that can be understood in the following description. In addition, the drawings of course also include parts with different size relationships and ratios.

[0092] In addition, the acoustic elements of the first to fifth embodiments shown below are examples to illustrate the device structure, manufacturing method, etc. used for the technical idea of ​​the present invention, and the technical idea of ​​the present invention does not limit the materials, shapes, structures or configurations of the components and the steps of the manufacturing method to the contents shown in the first to fifth embodiments. That is, the technical idea of ​​the present invention is not limited to the following description of the acoustic elements of the first to fifth embodiments, and various changes can be made within the technical scope specified by the claims recorded in the patent claims. In conventional PMUTs, a dual piezoelectric chip type structure in which a piezoelectric layer is formed on both sides of a fixed potential electrode is generally adopted. In addition to the structure of two piezoelectric layers, a dual piezoelectric chip type structure is also known, in which a piezoelectric layer is formed on one side of a fixed potential electrode, and an elastic layer composed of a material having a different rigidity from the piezoelectric layer is formed on the other side. However, in the acoustic elements of the first to fifth embodiments shown below, for example Fig.11A As shown, the description is based on a voltage generating unit of a unimorph type (single-morph type) structure consisting of a vibrating body 22f consisting of a single piezoelectric layer and a fixed potential electrode 24a in contact with the receiving surface of the vibrating body 22f. That is, in the description of the acoustic element of the first to fifth embodiments, the description is mainly centered around the voltage generating unit of the unimorph type structure and the receiving element consisting of an impedance conversion element that works with the potential of a specific part (local) of the voltage generating unit as a control voltage. However, the description of the unimorph type structure is only for convenience and is not intended to limit the scope of application of the present invention to the unimorph type structure, and a bimorph type structure is also possible.

[0093] like Fig.11A The structure of the flat disk type vibration cavity shown is a conventionally known structure from the perspective of focusing on the vibration cavity. However, as far as the inventors know, for a single piezoelectric chip type structure, Fig.11A Examples of flat disk-shaped vibration cavities are not common. Fig.11A The thin flat disk-shaped vibration cavity 21f shown in FIG. Fig.31 As shown, consider the normal direction of the disk ( Fig.31The vibrating body 22f composed of the flat piezoelectric layer is bent and vibrated on the upper and lower sides. Fig.31 As shown, when the flat plate-like vibrating body 22f is bent on the upper side to form an upward convex surface and when it is bent on the lower side to form a downward convex surface, mutually asymmetric vibrations with different intervals are generated. Fig.31 is a model diagram, assuming that the vibrating body 22f is in the horizontal direction ( Fig.31 The X direction in the image is flat.

[0094] Fig.31 (a) is a diagram showing the displacement of a vibrating body 22f made of PZT when a disk-shaped vibration cavity 21f with a diameter of 66 μm is formed in the X direction. Fig.31 (a) Assuming that there is no load, 19 circles are arranged at equal intervals of about 3.7 μm in the X direction, indicating the displacement degree of the circles in the Y direction at different distances from the center along the X direction. Fig.31 In case (a), when the vibrating body 22f is bent convexly on the upper side, the interval of ○ in the center of the vibrating body 22f tends to expand in the X direction, and when the vibrating body 22f is bent convexly on the lower side, the interval of ○ in the center of the vibrating body 22f tends to shrink in the X direction, indicating that the strain of the vibrating body 22f is uneven, and therefore the stress inside the vibrating body 22f is also uneven. Fig.31 In (a), the shape of the circle when the vibrator 22f is bent downward indicates the displacement position of each circle 0.13 μs after the ultrasonic pulse arrives, and the shape of the circle when the vibrator 22f is bent upward indicates the displacement position of each circle 0.18 μs after the ultrasonic pulse arrives.

[0095] Fig.31 (b) is a case where a disk-shaped vibration cavity 21f with a diameter of 40 μm is formed in the X direction. x Ti 1-x O 3 The unloaded state of the vibrating body 22f made of PZT is modeled with 19 circles arranged at equal intervals of about 2.2 μm in the X direction. Fig.31 In the model case (b), when the flat vibrator 22f is bent on the upper side to form a curved surface, the interval of the circles in the center of the vibrator 22f is reduced in the X direction, and when the flat vibrator 22f is bent on the lower side to form a curved surface, the interval of the circles in the center of the vibrator 22f is expanded in the X direction. Fig.31 In (b), the shape of the circle arranged when the vibrator 22f is bent at the bottom side represents the displacement position of each circle 0.14μs after the ultrasonic pulse arrives, and the shape of the circle arranged when the vibrator 22f is bent at the top side represents the displacement position of each circle 0.19μs after the ultrasonic pulse arrives. Fig.31In the case of (b), it is also shown that the strain of the vibrating body 22f is not uniform, and therefore the stress inside the vibrating body 22f is not uniform.

[0096] That is, in the unimorph type structure, since the strain and stress of the vibrating body 22f constituting the voltage generating portion are not uniform, the output surface (surface on the other side of the receiving surface) of the vibrating body 22f opposite to the receiving surface (surface on the other side of the receiving surface) is Fig.11A A specific portion of the output surface of the vibrating body 22f (the lower main surface) generates a higher voltage than other portions, and is characterized by the in-plane non-uniformity of the generated voltage. The "specific portion" of the output surface of the vibrating body 22f is an area corresponding to the stress concentration portion of the vibrating body 22f, and the specific portion within the surface defines the potential transfer region of the acoustic element of the first to fifth embodiments. The impedance conversion element operates with the potential of the potential transfer region as a control voltage, and uses the ultrasonic pressure applied to the piezoelectric layer as the output signal of the impedance conversion element, and uses the output signal as the output signal of the acoustic element of the first to fifth embodiments.

[0097] When the impedance conversion element is composed of a single active element, the "first main electrode region" of the active element refers to a semiconductor region that becomes either the source region or the drain region in a field effect transistor (FET), a static induction transistor (SIT), a high electron mobility transistor (HEMT) or a new transistor structure equivalent thereto. In a bipolar junction transistor (BJT), an insulated gate bipolar transistor (IGBT) or a semiconductor element of a new structure equivalent thereto, it refers to a semiconductor region that becomes either the emitter region or the collector region. In addition, in a thyristor such as an MIS-controlled static induction thyristor (SI thyristor) or a semiconductor element of a new structure equivalent thereto, it refers to a semiconductor region that becomes either the anode region or the cathode region. In FET, SIT, HEMT, etc., the "second main electrode region" refers to a semiconductor region that becomes either the source region or the drain region outside the above-mentioned first main electrode region. In BJT, IGBT, etc., it refers to a region that becomes either the emitter region or the collector region outside the above-mentioned first main electrode region. In an SI thyristor or the like, it refers to a semiconductor region that serves as either an anode region or a cathode region other than the first main electrode region.

[0098] In this way, when the impedance conversion element is composed of a single active element, if the "first main electrode region" is a source region, the "second main electrode region" is a drain region, and the "main current" flows between the first main electrode region 15a and the second main electrode region 15b. If the "first main electrode region" is an emitter region, the "second main electrode region" is a collector region. If the "first main electrode region" is an anode region, the "second main electrode region" is a cathode region. In the case of MISFET, if the bias relationship is exchanged, the function of the "first main electrode region" and the function of the "second main electrode region" may be exchanged. When the impedance conversion element is composed of a plurality of active elements, such as a CMOS type circuit configuration or a differential amplifier, for each of the plurality of active elements, a pair of the first main electrode region 15a and the second main electrode region 15b, a pair of the third main electrode region 15d and the fourth main electrode region 15c, and so on are defined in sequence. For example, in the CMOS type circuit configuration described in the third embodiment, the "third main electrode region" in FET, SIT, HEMT, etc. refers to a semiconductor region that serves as either a source region or a drain region, and the "fourth main electrode region" in FET, SIT, HEMT, etc. refers to a semiconductor region that serves as either a drain region or a source region that is paired with the third main electrode region.

[0099] In addition, the words "upper", "lower", "up", "lower" and the like in the following description, as well as the directional definitions using "up" and "down", are defined only for the convenience of description and do not limit the technical idea of ​​the present invention. For example, if the object is rotated 90° to observe, the upper and lower parts will be converted to left and right readings, and if it is rotated 180° to observe, the upper and lower parts will be reversed to read. In addition, in the following description, the case where the first conductivity type is p-type and the second conductivity type is n-type is used as an example for explanation. However, the opposite conductivity type relationship can also be selected, that is, the first conductivity type is set to n-type and the second conductivity type is set to p-type. The first conductivity type and the second conductivity type do not have to be defined together in the following first to fifth embodiments. For example, in the following description of the first to fourth embodiments of the MIS transistor (hereinafter referred to as "MIS transistor"), the first conductivity type is defined as p-type and the second conductivity type is defined as n-type, but in the description of the fifth embodiment of the junction transistor, the first conductivity type is set to n-type and the second conductivity type is set to p-type. In the description of each embodiment, as long as the definitions of the first conductivity type and the second conductivity type are common, it is sufficient. Furthermore, in the description of each embodiment, an opposite conductivity type relationship may be selected.

[0100] Similarly, in the description of the first to fourth embodiments, the first main electrode region is described as a drain region and the second main electrode region is described as a source region, but in the description of the second to fifth embodiments, the definition of the first main electrode region as a source region and the second main electrode region as a drain region is easier to understand. Within the description of each embodiment, as long as the definition of the first main electrode region and the second main electrode region is the same. In addition, within the description of each embodiment, the opposite definition relationship between the first main electrode region and the second main electrode region can also be selected. In addition, the "+" and "-" attached after "n" and "p" respectively represent semiconductor regions with relatively high or low impurity densities compared to semiconductor regions without "+" and "-". In addition, the "--" attached after "n" and "p" represents a high resistivity semiconductor region that is very close to the intrinsic semiconductor (i layer). However, even if the semiconductor regions are attached with the same "+" or "-", or the semiconductor regions are not attached with "+" and "-", it does not mean that their respective impurity densities are strictly the same.

[0101] First embodiment

[0102] Figure 1A The regular hexagonal unit cell X i,j (Although not shown in the figure) with other adjacent regular hexagonal unit cells X (i-1),(j+2) , …, X (i-1),(j+1) , X i,(j+1) , X (i+1),(j+1) , …, X (i-1),j , X (i+1),j , …, X i,(j-1) , X (i+1),(j-1) , ..., arranged together into a two-dimensional matrix, is a unit element that can form an acoustic element integrated circuit pattern. That is, Figure 1A , Figure 1B and Figure 2 As shown, the acoustic element of the first embodiment of the present invention includes: a voltage generating part (22a1, 24a) that generates a non-uniform voltage through a piezoelectric effect, and an impedance conversion element (14, 15a, 15b) that operates with the potential of a specific part (local part) of the voltage generating part (22a1, 24a) as a control voltage. As can be seen from the following description, the acoustic element of the first embodiment can improve the receiving sensitivity and is therefore suitable for a receiving element. Figure 1A The unit cell X shown i,j They can be arranged on the same plane (the same curved surface) to form a two-dimensional matrix.

[0103] exist Figure 1AIn the example, the case where the plane pattern is a regular hexagon is illustrated, but the plane pattern of the voltage generating part (22a1, 24a) is not limited to a regular hexagon, and various plane patterns such as a regular octagon, a regular dodecagon, and a regular icosagon can also be used. Taking into account the corner rounding of the photolithography process in the manufacturing process, the regular icosagon on the mask layer will be closer to the real shape in the finished product shape after the manufacturing process. However, from the perspective of suppressing the harmonics generated when the voltage generating part (22a1, 24a) vibrates in a drum-like manner, regular polygons of regular hexagons and above are preferred. In addition, wurtzite-type compound semiconductors such as aluminum nitride (AlN), gallium nitride (GaN), zinc oxide (ZnO) and cadmium sulfide (CdS) used as the vibrating body 22a1 belong to the hexagonal crystal system and have good compatibility with the regular hexagonal shape (perovskite structures such as PZT belong to the cubic crystal system).

[0104] from Figure 2 As can be seen from the cross-sectional view, the voltage generating portion (22a1, 24a) has a vibrating body 22a1 and a receiving surface ( Figure 2 The fixed potential electrode 24a in contact with the upper main surface of the vibrator 22a1 is a single piezoelectric chip type structure. The voltage generating portion (22a1, 24a) of the single piezoelectric chip type structure is provided on the output surface (surface on the other side of the receiving surface) opposite to the receiving surface of the vibrator 22a1. Figure 2 A specific part of the lower main surface of the inner wall generates a higher voltage than other parts through the piezoelectric effect. Figure 6C The uneven voltage distribution shown in the figure produces a voltage. Figure 6C In the figure, the horizontal axis represents the distance from the center, but the "specific part" of the output surface of the vibrator 22a1 refers to, for example, the part corresponding to Figure 4B Marked as P 0 The potential transfer area in the inner area of ​​the part corresponding to the stress concentration part of the vibrating body 22a1. Figure 2 As shown, the vibrating body 22a1 may be entirely composed of a piezoelectric layer, or only a portion of the stress concentration portion may be composed of a piezoelectric layer, and the remaining portion may be a non-piezoelectric layer. For example, only the columnar portion constituting the central convex portion may be composed of a piezoelectric layer, and the remaining plate-shaped portion and the peripheral portion may be non-piezoelectric layers. That is, the voltage generating portion only needs to have a structure in which at least a piezoelectric layer is provided at the stress concentration portion of the vibrating body 22a1. Therefore, the impedance conversion element (14, 15a, 15b) uses a specific portion of the piezoelectric layer at least located at the stress concentration portion as a potential transfer region, and operates with the potential of the potential transfer region as a control voltage.

[0105] The impedance transformation elements (14, 15a, 15b) are Figure 2The central part of the vibrator 22a1 is operated with the potential of the potential transfer area located at the downward convex part of the output surface side of the vibrator 22a1 as the control voltage. That is, the output voltage of the voltage generating unit (22a1, 24a) generated by the ultrasonic pressure applied to the vibrator 22a1 is used as the input signal of the impedance conversion element (14, 15a, 15b). Then, the output signal converted by the impedance conversion element (14, 15a, 15b) becomes the receiving element X. i,j The output signal of the vibrator 22a1 is Figure 1A Indicated by dotted lines, Figure 1B The double-dotted line in the figure represents the regular hexagonal shape, which is the convex part of the regular hexagonal prism. Figure 2 As shown, around the convex portion provided in the form of a step structure at the central portion of the output surface side of the vibrating body 22a1, the parallel toroidal vibration cavity 21a1 is provided in a centrally symmetrical manner around the convex portion. Figure 1A It can be understood from the plan view that the ring surface constituting the vibration cavity 21a1 has a topological structure whose plane pattern is a regular hexagon. Figure 2 In the cross-sectional view of FIG. 2 , the vibration cavities 21a1 are located on both sides of the central downward convex portion. However, in reality, the two vibration cavities 21a1 located on both sides of the central convex portion are an integral structure that is continuous with each other on the front side and the back side of the paper.

[0106] like Figure 1A As shown, on the outside of the outer wall (indicated by the dotted line) of the regular hexagonal vibration cavity 21a1, the dotted line defining the outline of the regular hexagonal fixed potential electrode 24a pattern is set as a plane pattern surrounding the vibration cavity 21a1. Figure 1A Correspondingly, in Figure 1B The positions of the inner wall of the outer side of the vibration cavity 21a1 and the outer peripheral position of the fixed potential electrode 24a are indicated by double-dotted lines on the left and right sides. The plane pattern of the ring surface constituting the vibration cavity 21a1 and the plane pattern of the convex part located in the center of the ring surface are regular hexagons. This is because Figure 1A The unit cell X shown i,j The geometric similarity requirement is due to the shape of a regular hexagon. For example, if the unit cell X i,j If the shape of the ring is a regular octagon, a regular dodecagon, a regular icosagon or a circle, it is ideal for the outermost periphery of the ring surface and the planar pattern of the convex portion located in the center of the ring surface to adopt a similar shape of a regular octagon, a regular dodecagon, a regular icosagon or a circle.

[0107] The impedance conversion element (14, 15a, 15b) of the acoustic element according to the first embodiment is as follows: Figure 1B and Figure 2As shown, the substrate region 14 is composed of a first conductivity type (p-type) semiconductor region, and a second conductivity type (n-type) semiconductor region is disposed separately from each other on the substrate region 14. + The active element of the first main electrode region 15a and the second main electrode region 15b is composed of a semiconductor region of the type. Figure 2 As shown, the upper surface of the base region 14 between the first main electrode region 15a and the second main electrode region 15b is connected to the potential transfer region ( Figure 2 For example, the base region 14 is set to GaN with a bandgap of 3.4 to 3.5 eV, and the vibrator 22a1 is set to AlN with a bandgap of 5 to 6.2 eV to form an AlN / GaN heterojunction, thereby forming a heterojunction field effect transistor (heterojunction FET). Both GaN and AlN have a wurtzite crystal structure, and the lattice constant of GaN in the a-axis direction is 0.318 nm, while the lattice constant of AlN in the a-axis direction is 0.311 nm, so the lattice mismatch is about 2.4%, which is good for heterojunction compatibility.

[0108] When AlN (with a longer AlN bond length) is epitaxially grown on the Ga plane of GaN with a plane orientation in the c-axis direction, AlN will experience tensile strain due to the difference in lattice constants between the two, and piezoelectric polarization will be applied inside AlN. GaN is also a polar crystal, so it also has piezoelectric polarization. Due to the effects of these two piezoelectric polarizations, positive fixed charges are generated at the AlN / GaN interface, and a two-dimensional electron gas is generated at the interface. In the case of using a base region 14 composed of a GaN substrate, the first main electrode region 15a and the second main electrode region 15b of the impedance conversion element (14, 15a, 15b) can also be composed of GaN. Even in the case of using a base region 14 composed of a Si substrate, as Fig.14 As shown in FIG. 1 , by using the buffer layer 13h, AlN can also be epitaxially grown on the buffer layer 13h. In addition, if defects at the heterojunction interface are not considered, AlN can also be epitaxially grown directly on the Si substrate.

[0109] Around the base region 14, an element isolation insulating film 16 is provided in a frame shape to define the range of the active region. Figure 2In the cross-sectional view, the element isolation insulating film 16 is located on both sides of the active region, but the element isolation insulating films 16 located on both sides of the active region are an integral structure that is continuous with each other on the front and back sides of the paper. Inside the active region pattern surrounded by the element isolation insulating film 16, the range (region) of the upper surface of the substrate region 14 is defined, and the first main electrode region 15a and the second main electrode region 15b are arranged in the upper part near the upper surface (surface) of the active region surrounded by the element isolation insulating film 16. Inside the rectangular substrate region 14 sandwiched between the first main electrode region 15a and the second main electrode region 15b, a channel for carrier flow is formed near the heterojunction interface of the substrate region 14. Figure 1B In the center, the range of the regular hexagonal potential transfer area is represented by a double-dotted line. The regular hexagonal potential transfer area, which is defined as the convex part of the output surface side of the vibrator 22a1, is arranged above the channel near the heterojunction interface in order to be able to impart a selective electrostatic induction effect to the channel locally. The channel is formed near the heterojunction interface due to the change in the surface potential distribution in the depth direction caused by the difference in the work function of the materials constituting the heterojunction, which causes the energy band of carrier accumulation to bend. In the case of AlN / GaN heterojunction, when it is desired to accumulate a two-dimensional electron cloud at the bend of the energy band, it is ideal to set GaN to i-type and AlN to n-type, so that the energy band at the conduction band end of GaN is bent to below the Fermi level.

[0110] Then, a potential barrier that controls the movement of carriers is formed in the channel through the npn structure between the first main electrode region 15a and the second main electrode region 15b. In the regular hexagonal region shown by the double dotted line, a selective electrostatic induction effect occurs within the depth from the heterojunction to the Debye length range. As a result, the height of the potential barrier generated between the first main electrode region 15a and the second main electrode region 15b is controlled by the potential of the potential transfer region of the vibrating body 22a1. The potential of the substrate region 14 can be set to a fixed potential such as the ground potential. This operation is similar to that of an ordinary heterojunction FET, but unlike an ordinary heterojunction FET, there is no gate electrode. Figure 2 In the structure shown, if the fixed potential electrode 24a is grounded, the Fermi level of the energy band diagram becomes 0V at the position of the fixed potential electrode 24a. When ultrasonic waves are input, the Fermi level inside the vibrator 22a1, especially near the heterojunction interface, vibrates. When the Fermi level at the heterojunction interface vibrates, the potential barrier height of the npn structure changes. In the case of the impedance conversion element (14, 15a, 15b) of the acoustic element involved in the first embodiment, as shown in FIG. Figure 3B As shown in the equivalent circuit of FIG. 1 , the vibrating body 22a1 itself functions as the gate dielectric layer of a normal MIS transistor. Furthermore, although the gate electrode is not present, Figure 3B The first brake capacitor C shownd1 (=C gs ) and the second brake capacitor C d2 (=C gd ) determines the gate voltage for the local area, i.e. the channel.

[0111] Since the acoustic element according to the first embodiment does not have a gate electrode, it is not necessary to connect the PMUT as a single element and the FET as a single element through external wiring as described in Non-Patent Document 2. Therefore, the acoustic element according to the first embodiment does not have a floating impedance problem. In addition, since the voltage generating unit (22a1, 24a) adopts the Figures 4A to 6C The stress enhancement structure shown achieves high sensitivity and broadband, so the acoustic element involved in the first embodiment can achieve high sensitivity and broadband. In particular, a specific position where the output surface of the vibrating body 22a1 generates the highest voltage is selected as the potential transfer area, and the active element of the impedance conversion element (14, 15a, 15b) is driven, so it is easy to achieve high sensitivity.

[0112] like Figure 2 As shown, an interlayer insulating film 17 is provided on the first main electrode region 15a and the second main electrode region 15b. Through contact holes opened in the interlayer insulating film 17, the first main electrode region 15a is connected to the first main electrode wiring 25a, and the second main electrode region 15b is connected to the second main electrode wiring 25b. Figure 1B In the figure, the rectangular region indicated by the dotted line below the first main electrode wiring 25a and the second main electrode wiring 25b is partially hidden, and the first main electrode region 15a and the second main electrode region 15b of the active element are disposed apart from each other with their short sides facing each other.

[0113] A third contact plug 41c connected to the first power source is disposed on the upper right oblique side of the hexagon constituting the outer periphery of the fixed potential electrode 24a. The third contact plug 41c is connected to the fixed potential electrode 24a. Figure 1A In the above description, the first power source is described as the ground potential (GND) as an example, but it is not limited to the ground potential (GND), as long as it is a fixed potential. In addition, two rectangular grooves (cutouts) are provided as a plane pattern on the outer periphery of the hexagonal fixed potential electrode 24a, and the two grooves are respectively provided with a second power source V DD The first contact plug 41a and the signal wiring R iThat is, the first contact plug 41a is arranged inside the groove provided on the left longitudinal side of the hexagonal planar pattern constituting the outer periphery of the fixed potential electrode 24a, and the second contact plug 41b is arranged inside the groove provided on the right longitudinal side. The first contact plug 41a is connected to the first main electrode wiring 25a, and the second contact plug 41b is connected to the second main electrode wiring 25b. The signal wiring R i The output resistor Rs is connected to the amplifier 81. If the first main electrode region 15a is used as the drain region and the second main electrode region 15b is used as the source region, then Figure 1A The circuit shown is an example of a source grounded circuit form.

[0114] Figure 3A The above paragraph shows that Figure 2 The equivalent circuit of the piezoelectric element in which the voltage generating section (22a1, 24a) acts is shown. The "A" enclosed in the figure represents an ideal transformer and is called the "force coefficient" and corresponds to the transformer's magnification. The force coefficient A is the force F of the mechanical system. o The voltage V of the electrical system o The conversion factor,

[0115] F o =AV o ……(1)

[0116] Connect the mechanical system circuit on the right side of the upper section and the electrical system circuit on the left side of the upper section. The magnitude of the force coefficient A depends on the magnitude of the d constant and the e constant of the piezoelectric material. In other words, the force coefficient A depends on the structure, the electromechanical coupling coefficient of the piezoelectric material, etc., but only acts as a conversion ratio. As shown in formula (1), the larger the force coefficient A, the higher the conversion efficiency of electrical and mechanical energy.

[0117] Figure 3A The mechanical system equivalent circuit shown on the right side of the upper paragraph is a circuit composed of an ideal transformer with a force coefficient A connected in series with a mechanical loss resistor (equivalent mechanical resistor) r, an equivalent mass (m+M), and an equivalent flexibility s. The equivalent flexibility s is the reciprocal of the equivalent stiffness c (s=1 / c). The equivalent mass M corresponds to the part that generates the mechanical system force and is a mass component that contributes to the input of ultrasonic waves. The equivalent mass m is the inherent mass component of the piezoelectric element other than the equivalent mass M. In addition, an internal resistor Z is connected to the mechanical loss resistor r and the equivalent flexibility s that constitute the series circuit. 0 The current source is almost infinite, and the vibration source vibrates due to ultrasonic input. The electrical system equivalent circuit shown on the left side of the upper paragraph consists of a brake capacitor C connected in parallel to an ideal transformer with a force coefficient A. d and brake resistor (dielectric loss resistor) R d express.

[0118] Figure 3A The lower section shows the small signal equivalent circuit of a MIS transistor (insulated gate transistor) having a gate insulating film (gate oxide film) structure. Heterojunction FET and HEMT are also types of MIS transistors. Figure 3A The connection node between the upper circuit and the lower circuit is shown as the gate terminal G of the MIS transistor. The gate terminal G corresponds to the gate electrode provided on the gate insulating film. The terminal corresponding to the first main electrode region 15a of the impedance conversion element (14, 15a, 15b) on the left side of the lower section is shown as the drain terminal D of the MIS transistor, and the terminal corresponding to the second main electrode region 15b on the right side of the lower section is shown as the source terminal S of the MIS transistor. Figure 3A The bottom side shows the substrate terminal B of the MIS transistor. The gate-source capacitance C is shown between the gate terminal G and the source terminal S. gs , showing a gate-drain capacitance C between the gate terminal G and the drain terminal D gd , showing a gate-substrate capacitance C between the gate terminal G and the substrate terminal B gb .

[0119] The substrate source capacitance C is shown between the substrate terminal B and the source terminal S. bs , showing a substrate drain capacitance C between the substrate terminal B and the drain terminal D bd , showing a source-drain conductance g between the source terminal S and the drain terminal D sd =1 / r sd (=ΔI DS / ΔV DS ). In addition, between the source terminal S and the drain terminal D, a current source g m v gs and current source g mb v bs Connect in parallel. m is the gate mutual conductance ΔI DS / ΔV GS , g mb is the substrate mutual conductance ΔI DS / ΔV BS . Figure 3A The gate-source capacitance C of the MIS transistor small signal equivalent circuit shown in the following paragraph is gs and gate-drain capacitance C gd Corresponds to the gate insulation film capacitance of the MIS transistor. Gate substrate capacitance C gb A portion of also corresponds to a portion of the gate insulating film capacitance.

[0120] However, Figure 2 The impedance conversion element (14, 15a, 15b) shown in the figure does not have a structure in which a gate insulating film exists directly below the voltage generating portion (22a1, 24a). Figure 3A This does not represent an equivalent circuit of the acoustic element according to the first embodiment. In the acoustic element according to the first embodiment, the vibrating body 22a1 is used as a gate insulating film of the MIS transistor. Figure 2 The equivalent circuit of the acoustic element according to the first embodiment in which the vibration body 22a1 itself functions as a gate insulating film corresponds to the structure of the impedance conversion element (14, 15a, 15b) connected to the voltage generating portion (22a1, 24a) as shown in FIG. Figure 3B shown.

[0121] exist Figure 3B On the left side, the gate-source capacitance C gs and gate-drain capacitance C gd In the electrical system equivalent circuit of the voltage generating unit (22a1, 24a), an ideal transformer with a force coefficient A and a braking resistor R d Ladder circuit connected in parallel. That is, the gate-source capacitance C gs As the first brake capacitor C d1 (C gs =C d1 ), gate-drain capacitance C gd As the second brake capacitor C d2 (C gd =C d2 ) play a role respectively. Gate substrate capacitance C gb A portion of the first main electrode region 15a and the second main electrode region 15b also functions as a brake capacitor of the voltage generating section (22a1, 24a). In addition, the potential barrier height for controlling the movement of carriers generated in the channel defined between the first main electrode region 15a and the second main electrode region 15b is increased by the first brake capacitor C d1 (=C gs ), the second brake capacitor C d2 (=C gd ) and the gate-substrate capacitance C gb A part of it is controlled by the potential of the vibrating body 22a1.

[0122] The piezoelectric constant has d form (strain·charge form), e form (stress·charge form), g form (strain·electric field form), and h form (stress·electric field form). The linear piezoelectric basic equation is given for each form and is used according to the purpose. The linear piezoelectric basic equation expressed in d form is:

[0123] D i =d iα T α +ε ij T E j ……(2a)

[0124] S α =sαβ E T β +d iα E i ……(2b)

[0125] α, β=xx, yy, zz,..., xy...(2c)

[0126] i, j = x, y, z...(2d)

[0127] Among them, D i is the electric flux density (C / m 2 ), T α is the stress (N / m 2 ), S α is the strain (unitless), E i is the electric field (V / m). In addition, d iα is the piezoelectric d constant (C / N), s αβ E is the elastic compliance under constant electric field (m 2 / N), ε ij T is the dielectric constant (pF / m) under constant stress. The subscripts defined in equations (2c) and (2d) indicate directions. Fig.31 The Y direction (longitudinal direction) in corresponds to the z direction used in equations (2a) and (2d), Fig.31 The X direction (horizontal direction) in corresponds to the x direction or y direction used in equations (2a) and (2d). The superscripts of the second term on the right side of the coefficient in equation (2a) and the first term on the right side of (2b) indicate the conditions. The coefficient with the superscript E indicates the value measured under the condition of constant electric field, and the coefficient with the superscript T indicates the value measured under the condition of constant stress.

[0128] On the other hand, the basic piezoelectric equation can be expressed in e-form as:

[0129] D i =e iα S α +ε ij S E j ……(3a)

[0130] T α =c αβ E S β -e iα E i ……(3b)

[0131] Among them, e iα is the piezoelectric e constant (C / m 2 ), cαβ E is the elastic stiffness (N / m 2 ), ε ij S is the dielectric constant (pF / m) under constant strain. The subscript indicates the direction defined in equations (2c) and (2d). The superscript of the second term on the right side of equation (3a) and the first term on the right side of equation (3b) indicates the condition. The coefficient with the superscript E indicates the value measured under constant electric field conditions, and the coefficient with the superscript S indicates the value measured under constant strain conditions. Equations (2a) and (3a) represent the direct piezoelectric effect, and equations (2b) and (3b) represent the inverse piezoelectric effect. In an ideal single crystal piezoelectric material whose polarization direction is consistent with the z direction, the number of independent components from the elastic compliance tensor, etc. can be reduced due to the isotropy of the material.

[0132] For example, the e-form linear piezoelectric basic equation (3a) of AlN can be expressed as:

[0133] Formula 1

[0134]

[0135] Assuming the piezoelectric constant e iα Very small, the second term on the right side of the linear piezoelectric basic equation (3b) is omitted. The electric flux density D at the end surface of the piezoelectric film x =D y =D z = 0 (open circuit), the linear piezoelectric basic equation (3b) when AlN is used as a piezoelectric film can be simplified to:

[0136] Formula 2

[0137]

[0138] When AlN is a piezoelectric film, let the electric flux density D in equations (4a) and (4b) be i = 0, the corresponding stress T α The electric field E x , E y , E z Formula 3

[0139]

[0140] From the electric field E x , E y , E z Determine the piezoelectric emf.

[0141] PZT is divided into a variety of commercial materials according to its composition, such as PZT-4, PZT-5, PZT-5A, PZT-5H, PZT-8, etc. Among them, as a soft PZT with a large piezoelectric constant, we focus on PZT-5H, which is generally used for receiving probes, etc. The e-form linear piezoelectric basic equation (3a) of PZT-5H can be expressed as:

[0142] Formula 4

[0143]

[0144] The second term on the right side of equation (3b) is omitted. Also, the electric flux density D at the end surface of the piezoelectric film is x =D y =D z = 0 (open circuit), the linear piezoelectric basic equation (3b) when PZT-5H is used as a piezoelectric film can be simplified to:

[0145] Formula 5

[0146]

[0147] When PZT-5H is the piezoelectric film, let the electric flux density D in equations (4a) and (4b) be i = 0, the corresponding stress T α The electric field E x , E y , E z for:

[0148] Formula 6

[0149]

[0150] The electric field E in equation (5c) can be x , E y , E z The piezoelectric electromotive force is obtained, but by comparing equation (4c) and equation (5c), it can be judged that when the electric flux density D i When =0, there is almost no significant difference in the piezoelectric potential of pyroelectric AlN and ferroelectric PZT-5H in response to stress.

[0151] As shown in the following formula (6), the dielectric includes a piezoelectric, the piezoelectric includes a pyroelectric (polar crystal), and the pyroelectric includes a ferroelectric. As long as the vibrating body 22a1 of the voltage generating portion (22a1, 24a) of the acoustic element involved in the first embodiment has a continuous structure as a stress field, the material of the stress concentration portion constituting at least a part of the vibrating body 22a1 is a piezoelectric, and the remaining part may be a non-piezoelectric.

[0152]

[0153] Crystals are divided into 32 families based on the symmetry of their crystal structure, 21 of which have no symmetry. One of the crystal point groups (cubic point group O-432) is not piezoelectric due to another symmetry, but the other 20 crystal point groups are piezoelectric. Among the piezoelectrics that are a subset of dielectrics, crystal (SiO 2 ), lanthanum gallium silicon oxide (La 3 Ga 5 SiO 14 ), gallium phosphate (GaPO 4 ) are well-known as piezoelectric materials. In addition, although it also depends on the crystallinity, aluminum oxide (Al2O3) of corundum type oxide 2 O 3 ), rutile type titanium dioxide (TiO 2 ), rock salt type oxides such as magnesium oxide (MgO), fluorite type oxides such as hafnium oxide (HfO 2 ) and zirconium oxide (ZrO 2 ) etc. can be used as the material constituting at least a part of the vibration body 22a1, that is, the material of the stress concentration part of the vibration body 22a1.

[0154] That is, the vibrating body 22a1 can be Figure 2 As shown, the entire structure is composed of piezoelectric bodies of 20 crystal point groups, or only the central downward convex portion (stress concentration site) can be composed of piezoelectric bodies of 20 crystal point groups, and the remaining plate-like parts can be composed of non-piezoelectric layers belonging to the remaining 12 crystal point groups among the 32 types. In other words, as long as it has a continuous structure as a stress field, the voltage generating part (22a1, 24a) can be a structure in which at least a piezoelectric layer is provided at the stress concentration site of the stress field formed by the vibrating body 22a1. Therefore, the impedance conversion element (14, 15a, 15b) of the acoustic element involved in the first embodiment can use at least a specific part of the piezoelectric layer provided in the downward convex portion (stress concentration site) as a potential transfer area, and operate with the potential of the potential transfer area as a control voltage.

[0155] Like hafnium aluminate (HfAlO x ) and hafnium silicate (HfSi x O y) can also be used as at least a material for the stress concentration portion of the vibrating body 22a1. That is, an oxide containing at least one of strontium (Sr), aluminum (Al), magnesium (Mg), yttrium (Y), hafnium (Hf), zirconium (Zr), tantalum (Ta), and bismuth (Bi), or a silicon nitride containing these elements can be used as at least a material for the stress concentration portion of the vibrating body 22a1. Zinc blende compound semiconductors such as gallium arsenide (GaAs), gallium phosphide (GaP), zinc sulfide (ZnS), zinc telluride (ZnTe), zinc selenide (ZnSe), and cadmium telluride (CdTe), and wurtzite compound semiconductors such as AlN, GaN, CdS, ZnO, and cadmium selenide (CdSe) also have piezoelectricity, and therefore can be used as at least a material for the stress concentration portion of the vibrating body 22a1. On the other hand, a semiconductor having a diamond structure (point symmetry group O h =m3m) single element semiconductors such as silicon (Si), germanium (Ge), diamond (C), etc. cannot exhibit piezoelectricity and therefore cannot be used as materials for the stress concentration part of the vibrating body 22a1, but can be used as materials for the remaining plate-like parts other than the stress concentration part.

[0156] As shown in formula (6), pyroelectrics are a subset of piezoelectrics. Among piezoelectrics, pyroelectrics are usually polarized and are prone to charge when the temperature changes. Among the 20 crystal families with piezoelectricity, 10 crystal point groups show pyroelectricity, also known as "polar crystals". That is, according to the Schönflis notation, the triclinic system C 1 , monoclinic C S , C 2 , orthorhombic system C 2V 、Tetragonal system C 4 , C 4V , trigonal system C 3 , C 3V , hexagonal system C 6 , C 6V The crystal point group of wurtzite structure is hexagonal C 6V Wurtzite compound semiconductors such as AlN, GaN, ZnO, and CdS are pyroelectrics. Hexagonal silicon carbide (SiC) is also a pyroelectric. AlN and GaN show piezoelectricity in a specific direction of the crystal, but when Sc is added, the crystal structure changes slightly, and ions move more easily in that direction, improving the piezoelectricity. For example, Sc x Al 1-x The piezoelectricity of N increases with the increase of Sc concentration x, and when x is ~ 43%, it shows the same piezoelectricity as LiTaO 3 Very large piezoelectricity. Tourmaline ((Ca, Na)X 3 Al 6 (BO 3 )3 Si 6 O 18 (OH, F)4; X = Mg, Fe, Mn, Li, Al) also has pyroelectric properties.

[0157] Ferroelectrics are a subset of pyroelectrics as shown in equation (6). Among pyroelectrics, the spontaneous polarization direction changes when an external electric field is applied. Piezoelectrics and pyroelectrics are supported by the symmetry of the crystal structure, but the definition of the symmetry of the crystal structure of ferroelectrics is ambiguous. In addition to PZT, ferroelectrics also include HfO 2 、Hf zirconium oxide (Hf 1-x Zr x O 2 ), barium titanate (BaTiO 3 ), SrTiO 3 ), magnesium titanate (MgTiO 3 ), calcium titanate (CaTiO 3 ), bismuth titanate (Bi 3 Ti 3 O 12 ), lead titanate (PbTiO 3 ), lead zirconate (PbZrO 3 ), barium strontium titanate (BaSrTiO 3 ), strontium ruthenate (SrRuO 3 ), SrBiTiO 2 Ta 2 O 9 ) and other perovskite oxides, bismuth layered ferroelectrics (BLSF), polyvinylidene fluoride (PVDF), Rochelle salt (NaKC 4 H 3 O 6 ·4H 2 O) etc.

[0158] <Study on stress distribution strengthening structure>

[0159] In the existing PMUT technology such as the invention described in non-patent document 2, no new internal structure and geometric shape for realizing high sensitivity and broadband of the PMUT itself have been studied. The acoustic element involved in the first embodiment is designed by considering the structure of the voltage generating part (22a1, 24a) with non-uniform potential distribution presenting different potentials at the output surface side of the vibrating body 22a1, so as to achieve high sensitivity and broadband. For example, considering the non-uniform potential distribution, it is allowed to adopt a composite structure in which a piezoelectric layer is arranged only at the stress concentration point to achieve high sensitivity and broadband. As shown in equations (2a) to (5c), in order to produce a position-dependent potential distribution, it is preferable to adopt a "stress distribution reinforcement structure" that produces a non-uniform potential distribution inside the vibrating body 22a1.

[0160] Although it has been used Fig.31 It has been explained that even in the case of using a flat disk-shaped vibration cavity known in the past, when the vibrating body 22f is bent upward to form an upward convex surface, and when it is bent downward to form a downward convex surface, the intervals between ○ and ○ are different, and mutually asymmetric vibrations will be performed. As a stress distribution strengthening structure for generating a potential distribution that depends on a position, what kind of piezoelectric layer and the shape of the vibration cavity are more ideal, for which the "piezoelectric fluctuation analysis software PZFlex" originally developed by Weidlinger Associates in the United States was used for simulation. By simulating the transient response of the generated voltage in various stress distribution strengthening structures, it is explored what kind of structure can be used as a new structure that achieves high sensitivity and broadband through the piezoelectric effect. Then, a specific candidate position is selected as the potential transfer area, and a specific position where the generated voltage is higher is microscopically explored in various stress distribution strengthening structures so that a piezoelectric layer can be configured at least in this potential transfer area.

[0161] - Parallel Truss Type -

[0162] First, for example Figure 4A The voltage generated at each position in the stress distribution reinforcement structure with parallel truss type vibration cavity 21a1 around the central convex part is simulated to study the voltage distribution on the output surface, which is used as the basis for high sensitivity and broadband. "Parallel truss type" refers to a three-dimensional structure in which the main surfaces (top and bottom) corresponding to the space constituting the truss are parallel to each other. The parallel truss type vibration cavity 21a1 is based on such a structure: in a space with a diameter d 11 (=2r 11 ) of the silicon substrate 11 with a cylindrical recess on the left and right sides sa1 A vibrating body 22a1 having a columnar downward convex portion is inserted in the center of the output surface (bottom). In cross section, the rectangular space surrounding the downward convex portion forms a parallel truss that is continuous in the depth direction of the paper. Figure 1A and Figure 1B As shown in the plan view of , the acoustic element according to the first embodiment is a parallel truss type vibration cavity 21a1 having a hexagonal planar pattern. Figure 1A The structure shown in the figure is a structure in which a vibrating body 22a1 having a regular hexagonal prism convex portion is inserted into the central part of the vibration cavity 21a1, which is different from the axisymmetric three-dimensional shape simplified in the simulation in terms of geometric plane pattern. However, whether the plane pattern is circular or hexagonal has no significant influence on the simulation results of the voltage distribution generated by the piezoelectric effect in principle.

[0163] The thickness of the flat plate-shaped portion of the vibrating body 22a1 excluding the central convex portion is t 22a1 , the diameter of the central convex part is set to d22a1 A fixed potential electrode 24a at a ground potential is disposed on the upper surface of the vibrating body 22a1. A fixed potential electrode protective film 23 made of epoxy resin, which functions as a matching layer, is formed with a thickness of t 23 The fixed potential electrode 24a and the vibrating body 22a1 without the fixed potential electrode 24a are configured. In the simulation, in order to verify the voltage distribution generated by the piezoelectric effect under the vibrating body 22a1, as shown in FIG. Figure 4B As shown, five ideal electrodes for potential detection are configured (at Figure 4B Then, according to the arrangement of the five ideal potential detection electrodes shown in Table 1, the reference radius r is divided from the inside. 11 =(d 11 -d 22a1 ) / 2 direction (outward) is numbered as position P 0 , P 1 ,……,P 4 , the individual name of the potential transfer area is defined as the central area P 0 , the first peripheral area P 1 , ..., the fourth peripheral area P 4 .

[0164] Table 1

[0165] Table 1 Figure 4B Definition of case

[0166]

[0167] Central Area 0 The line width and spacing along the radial direction of the outer side are the same, and the four first peripheral areas P surrounded by the same spacing 1 , the second peripheral area P 2 , the third peripheral area P 4 , the fourth peripheral area P 4 Since the line width and spacing are the same, the dividing reference radius r of the vibration cavity 21a1 is 11 =(d 11 -d 22a1 ) / 2 is divided into nine equal parts, thus defining the central area P 0 , the first peripheral area P 1 , ..., the fourth peripheral area P 4 The position along the radial direction. That is, Figure 4B As shown, the central area P 0 The ideal electrode for potential detection is a ring area surrounding the central downward convex part. Therefore, the radius of the central downward convex part is d. 22a1 / 2, Central Area P 0 Defined as the radius from position d22a1 / 2 to radius position (1 / 9)r 11 +d 22a1 / 2 annular area (region) with a certain line width.

[0168] Similarly, as defined in Table 1, the first peripheral area P 1 Defined as the radius from the position (2 / 9)r 11 +d 22a1 / 2 to radius position (3 / 9)r 11 +d 22a1 / 2. The second peripheral area P 2 Defined as the radius from the position (4 / 9)r 11 +d 22a1 / 2 to radius position (5 / 9)r 11 +d 22a1 / 2 a certain line width of the annular area, the third peripheral area P 3 Defined as the radius from the position (6 / 9)r 11 +d 22a1 / 2 to radius position (7 / 9)r 11 +d 22a1 / 2 is a ring area with a certain line width. The fourth peripheral area P 4 Defined as the radius from position (8 / 9)r 11 +d 22a1 / 2 to the outermost set (radius position r 11 ) area.

[0169] In the simulation, they are respectively arranged in the central area P 0 , the first peripheral area P 1 , ..., the fourth peripheral area P 4 The five ideal potential detection electrodes are virtual electrodes that are electrically open (open), have zero thickness, and have zero contact resistance with the vibrating body 22a1. 0 , the first peripheral area P 1 , ..., the fourth peripheral area P 4 In this paper, the transient response of the generated voltage generated by the piezoelectric effect is verified, and the voltage distribution on the output surface of the vibrator 22a1 is studied. 0 , the first peripheral area P 1 , ..., the fourth peripheral area P 4 The radial electric field inside the ideal electrode is short-circuited. Figure 4B In the 4th peripheral area P 4 The outer periphery of the ideal electrode for potential detection and the silicon substrate 11 sa1 contact, but the silicon substrate 11sa1 It is semi-insulating (high resistivity), and in the simulation it is assumed that the fourth peripheral region P 4 The ideal electrode for potential detection is in an electrically floating state.

[0170] Figure 4C The entire vibrating body 22a1 is made of AlN, and the diameter d of the central convex portion of the parallel truss type vibration cavity 21a1 of the voltage generating part (22a1, 24a) is 22 a 1 = 1.2μm, the potential transfer area is in the central area P 0 , the first peripheral area P 1 , ..., the fourth peripheral area P 4 The figure shows the transient response of the voltage generated when ultrasonic waves are received, with the position of the vibration body 22a1 as a parameter. 22 a 1 = 0.2 μm, fixed potential electrode protection film 23 thickness t 23 =3μm. Figure 4C The line type legend in the upper right corner shows that the thick solid line represents the central area P 0 The central voltage Vp of the generated voltage 0 The transient response of Figure 4C In the figure, the thick dotted line indicates the first peripheral area P 1 The first peripheral voltage V that generates the voltage P1 , the thin dotted line represents the second peripheral area P 2 The second peripheral voltage V P2 , the thin solid line represents the third peripheral area P 3 The third peripheral voltage V P3 , the single-dot chain line represents the fourth peripheral area P 4 The fourth peripheral voltage V4 of the generated voltage P When a single-shot sinusoidal ultrasonic wave with a pressure vibration of a maximum vibration amplitude of ±1Pa is input to the voltage generating section (22a1, 24a), the central area P 0 , the first peripheral area P 1 , ..., the fourth peripheral area P 4 The pulse response voltage signals generated by the piezoelectric effect show that the second vibration amplitude is the largest around 0.13μs, and the third and subsequent vibrations are transient responses of attenuated vibrations.

[0171] Specifically, in Figure 4A and Figure 4B When the fixed potential electrode 24a is connected to the ground potential (GND), as shown in FIG. Figure 4CAs shown, with the ground potential (GND=0V) as the center, it presents transient responses of positive and negative vibrations. Figure 4C The central voltage V P0 The first vibration amplitude, which peaked at about 0.06μs after the ultrasonic pulse input, was about +0.34μV, and the second vibration amplitude, which peaked at about 0.13μs, was about -0.52μV, reaching the maximum amplitude value. Thereafter, the third vibration amplitude, which peaked at about 0.19μs, was about +0.25μV, and the fourth vibration amplitude, which peaked at about 0.26μs, decayed to about -0.13μV.

[0172] and Figure 4B Different, if the central area P 0 The ideal electrode for potential detection is arranged in an island shape at the lower end of the central downward protrusion (only for the position definition, corresponding to keeping the central protrusion diameter unchanged in Table 1 and making d 22a1 / 2=0), then the central voltage V P0 Increase to Figure 4C This is considered to be the result of the piezoelectric electromotive force generated by the transverse compression and stretching of the film near the central downward convex portion of the vibrating body 22a1 and the piezoelectric electromotive force generated by the longitudinal compression and stretching along the central downward convex portion reinforcing each other.

[0173] exist Figure 4C The second peripheral voltage V indicated by the thin dotted line P2 The first vibration amplitude, which reaches a peak value at about 0.06μs after the ultrasonic pulse input, is about -0.26μV, while the second vibration amplitude, which reaches a peak value at about 0.13μs, is about +0.38μV, reaching the maximum amplitude value. However, the second peripheral voltage V P2 Absolute value of amplitude |V P2 |Less than the central voltage V P0 Absolute value of amplitude |V P0 |, and the central voltage V P0 and the second peripheral voltage V P2 After that, the second peripheral region P reaches a peak value at about 0.19 μs. 2 The second peripheral voltage V P2 The 3rd vibration amplitude is about -0.18 μV, and the 4th vibration amplitude, which reaches a peak value around the time of about 0.26 μs, decays to about +0.09 μV. Since the peak values ​​of the vibration amplitudes after the 5th vibration amplitude become smaller, their description is omitted.

[0174] exist Figure 4CThe fourth peripheral voltage V indicated by the single-dot chain line P4 The first vibration amplitude reaches a peak value of about +0.25μV at about 0.06μs, and the second vibration amplitude reaches a peak value of about -0.32μV at about 0.13μs, reaching the maximum amplitude value. P4 Absolute value of amplitude |V p4 |, than the central voltage V P0 and the second peripheral voltage V P2 The absolute value of |V P0 | and |V P2 | is the third largest value. 4th peripheral voltage V P4 and the second peripheral voltage V P2 Inversely proportional to the central voltage V P0 After that, the 4th peripheral voltage V P4 The amplitude of the third vibration, which reaches a peak value at approximately 0.19 μs, is approximately +0.17 μV, and the amplitude of the fourth vibration, which reaches a peak value at approximately 0.26 μs, decays to approximately -0.08 μV.

[0175] exist Figure 4C The first peripheral voltage V represented by the thick dashed line P1 The absolute value of the amplitude of the transient response vibration |V P1 |, than the 4th peripheral voltage V P4 The absolute value of |V P4 | is smaller and is the fourth largest value. The first peripheral voltage V P1 and the second peripheral voltage V P2 Vibrate in phase with each other, the first peripheral voltage V P1 With the central voltage V P0 and the 4th peripheral voltage V P4 Vibrate in opposite phases. Figure 4C The third peripheral voltage V represented by the thin solid line P3 The absolute value of the amplitude of the transient response vibration |V P3 |, than the first peripheral voltage V P1 The absolute value of |V P1 | is the fifth largest value and also the smallest value. P3 and the first peripheral voltage V P1 and the second peripheral voltage V P2 Vibrate in phase with each other, the third peripheral voltage V P3 With the central voltage V P0 and the 4th peripheral voltage V P4 Vibrate in opposite phases to each other.

[0176] That is, in Figure 4A and Figure 4B The potential relationship generated by the piezoelectric effect on the output surface of the vibrating body 22a1 of the voltage generating unit (22a1, 24a) is: P0 |>|V P2 |>|V P4 |>|V P1 |>|V P3 |……(7)

[0177] It presents a non-uniform potential distribution with different potentials at different positions. Figure 4C is to Figure 1A The voltage generated when the first power supply is connected to the ground potential (GND), that is, the center voltage V P0 , the first peripheral voltage V P1 , the second peripheral voltage V P2 , the third peripheral voltage V P3 and the 4th peripheral voltage V P4 . Figure 1A The circuit connections shown are different. If the first power supply is connected to the bias voltage V bias >0, the central voltage V generated by the voltage generating unit (22a1, 24a) P0 , the first peripheral voltage V P1 , ..., the fourth peripheral voltage V P4 , will become a bias voltage V different from 0V bias Centered at a bias voltage V bias That is, select the V that makes the impedance conversion element (14, 15a, 15b) have the highest amplification factor. bias , and connect the 1st power supply to V bias , the sensitivity of the acoustic element involved in the first embodiment can be improved. Figure 4D and Figure 4C Similarly, in the composition Figure 4A and Figure 4B When the vibrating body 22a1 of the voltage generating part (22a1, 24a) is made of AlN, and the diameter d of the central convex part of the parallel truss type vibration cavity 21a1 of the voltage generating part (22a1, 24a) is 22a1 =2.0μm, Figure 4C When the situation is more serious, Figure 4B The central area P of the defined position 0 , the first peripheral area P 1 , ..., the fourth peripheral area P 4 The position of the potential transfer area is used as a parameter to simulate the transient response of the voltage generated when receiving ultrasonic waves. 0 , the first peripheral area P 1 , ..., the fourth peripheral area P 4The positions are shown in Table 1. The thickness t of the vibrating body 22a1 outside the central convex portion 22a1 =1μm, Figure 4C The fixed potential electrode protection film 23 has a thickness t 23 =3μm, and Figure 4C Same situation.

[0178] exist Figure 4D The central voltage V at the central portion of the vibrating body 22a1 indicated by the thick solid line is P0 The first vibration amplitude, which reaches a peak value at about +0.14μV after about 0.03μs after the ultrasonic pulse input, and the second vibration amplitude, which reaches a peak value at about -0.24μV after about 0.13μs, reaches the maximum amplitude value. Figure 4C In the case of the first central voltage V around the time of about 0.06 μs after the ultrasonic pulse input, P0 The vibration amplitude is about +0.34μV, and the second vibration amplitude is about 0.52μV after about 0.13μs. Figure 4D The central voltage V P0 Vibration amplitude ratio Figure 4C Moreover, the first vibration, which reaches its peak at about 0.03μs after the pulse input, shows a waveform distortion that gradually decays at about 0.06μs, and is not a single sine wave. Figure 4C In this case, the sine wave vibration has a peak value at 0.06μs. The central voltage V P0 The amplitude of the third vibration, which peaked at about 0.19 μs, was about +0.15 μV, and the amplitude of the fourth vibration, which peaked at about 0.26 μs, decayed to about -0.06 μV. Waveform distortion was also observed in the third and fourth vibrations.

[0179] Although not shown in the figure, if the central area P 0 The ideal electrode for potential detection is arranged in an island shape at the lower end of the central downward convex portion (only focusing on the position definition, corresponding to keeping the convex portion diameter unchanged in Table 1 and making d 22a1 / 2=0), the central voltage V P0 Increase to Figure 4D This is about twice as much as the case shown. Figure 4C It is considered that the piezoelectric electromotive force generated by the transverse compression and stretching of the film near the central downward convex portion of the vibrating body 22a1 and the piezoelectric electromotive force generated by the longitudinal compression and stretching along the central downward convex portion are mutually enhanced (the increase in the longitudinal piezoelectric effect of the central convex portion will be referred to later). Figure 4E and Figure 4F for explanation).

[0180] exist Figure 4D The second peripheral voltage V indicated by the thin dotted line P2 The first vibration amplitude, which reaches a peak value at about 0.06μs after the ultrasonic pulse input, is about -0.14μV, while the second vibration amplitude, which reaches a peak value at about 0.13μs, is about +0.24μV, reaching the maximum amplitude value. Figure 4D As shown, the central voltage V P0 and the second peripheral voltage V P2 Inverse phase relationship, but the second peripheral voltage V p2 The absolute value of |V P2 |With the central voltage V P0 The absolute value of |V P0 | is roughly the same. After that, the second peripheral voltage V reaches a peak value around the time of about 0.19 μs. P2 The amplitude of the third vibration is about -0.14μV, and the amplitude of the fourth vibration, which reaches a peak around the time of about 0.26μs, decays to about +0.06μV.

[0181] exist Figure 4D The fourth peripheral voltage V indicated by the single-dot chain line in the vicinity of the outer wall (side wall) of the vibration cavity 21a1 and the position farthest from the central protrusion is P4 The first vibration amplitude reaches a peak value of about +0.15μV at about 0.06μs, which is higher than the central voltage V P0 The first vibration amplitude is slightly larger than +0.14μV. The second vibration amplitude, which reaches a peak value at about -0.24μV after about 0.13μs relative to the first vibration amplitude, reaches the maximum amplitude value, which is similar to the central voltage V P0 The amplitude of the second vibration is roughly the same. Therefore, the fourth peripheral voltage V P4 Absolute value of amplitude |V P4 |With the central voltage V P0 The absolute value of |V P0 | and the second peripheral voltage V P2 The absolute value of |V P2 |About the same. 4th peripheral voltage V P4 and the second peripheral voltage V P2 Inversely proportional to the central voltage V P0 After that, the 4th peripheral voltage V P4 The third vibration amplitude, which reaches a peak value at about 0.19 μs, is about +0.13 μV, which is larger than the center voltage V P0 The amplitude of the fourth vibration, which is slightly smaller and reaches a peak around the time of about 0.26 μs, decays to about -0.06 μV.

[0182] exist Figure 4D The first peripheral voltage V P1 The transient response is that the first vibration peak positively vibrates to about +0.05μV at about 0.03μs after the ultrasonic pulse input. Thereafter, the second vibration peak negatively vibrates to about -0.03μV at about 0.06μs. The first peripheral voltage V P1 presents a central voltage V P0 , the second peripheral voltage V P2 , 4th peripheral voltage V P4 Different unique vibration waveforms. And, with Figure 4C The first peripheral voltage V P1 The transient response behavior is also different. Figure 4D The first peripheral voltage V P1 The transient response behavior increases to about +0.07μV around 0.06μs. Figure 4C The first peripheral voltage V P1 The transient response behavior is the same as that of Figure 4C is the second vibration peak, and Figure 4D The difference is that it is the third vibration peak. The first peripheral voltage V P1 The absolute value of the amplitude of the transient response vibration |V P1 | is the central voltage V P0 , the second peripheral voltage V P2 , 4th peripheral voltage V P4 The absolute value of |V P0 |、|V P2 |、|V P4 |The fourth smaller value in. Except for the first vibration peak around 0.03μs, the first peripheral voltage V P1 and the second peripheral voltage V P2 In addition to the first vibration peak, the first peripheral voltage V P1 With the central voltage V P0 and the 4th peripheral voltage V P4 Vibrate in opposite phases to each other.

[0183] In addition, Figure 4D The third peripheral voltage V represented by the thin solid line P3 The absolute value of the amplitude of the transient response vibration is greater than the first peripheral voltage V P1 The third peripheral voltage V P3 The first surrounding area P 1 , the first peripheral voltage V P1 and the second peripheral voltage V P2Vibrate in opposite phases, which is similar to Figure 4C The transient response shown is different. Similarly, the third peripheral voltage V P3 With the central voltage V P0 and the 4th peripheral voltage V P4 They vibrate in phase with each other, which is also consistent with Figure 4C That is, when the diameter d of the convex portion at the center of the vibration cavity 21a1 is 22a1 Compare Figure 4C The case is thicker, and the thickness t of the vibrating body 22a1 is 22a1 Compare Figure 4C When the thickness is thicker, the potential relationship generated by the piezoelectric effect on the output surface of the vibrating body 22a1 is:

[0184] |V P0 |≈|V P2 |≈|V P4 |……(8a)

[0185] |V P0 |,|V P2 |,|V P4 |>|V P1 |>|V P3 |……(8b)

[0186] It shows a non-uniform potential distribution with position dependence.

[0187] Figure 4E When the entire vibrator 22a1 is made of PZT, the central area P detected by the ideal electrode for potential detection provided just below the central convex portion is 0 The central voltage V P0 The thickness t of the flat plate portion of the vibrating body 22a1 other than the central convex portion is 22a1 Set to 2.4μm. Figure 4B The diameter d of the central convex portion shown 22a1 Redefine as diameter D, and set diameter D = 2.4 μm = t 22a1 , represents the frequency dependence of voltage sensitivity when the aspect ratio (=H / D) of the height H of the central convex portion corresponding to the diameter D is changed. Figure 4E As shown in FIG. 1 , when the aspect ratio H / D=1, a sensitivity curve with a peak frequency of 7.3MHz is presented in the 5-10MHz frequency band of receiving ultrasound. When the aspect ratio H / D=1, the central voltage V P0 The maximum voltage sensitivity is about 26.4μV / Pa. Compared with the aspect ratio H / D=1, when the aspect ratio H / D=2, that is, the height increases, the peak frequency of the sensitivity curve is 6.8MHz, slightly shifted to the low frequency side, and the relative bandwidth becomes narrower. The central voltage V p0The maximum voltage sensitivity is about 47.7μV / Pa, and the maximum voltage sensitivity increases by about 1.8 times. It can be observed that the longitudinal piezoelectric effect increases with the increase of the aspect ratio H / D.

[0188] When the aspect ratio H / D=3 and the height is further increased, the peak frequency of the sensitivity curve is 6.5MHz, slightly shifted to the lower frequency side than when H / D=2, and the bandwidth is further narrowed slightly. P0 The maximum voltage sensitivity is about 68.2μV / Pa. When the aspect ratio H / D=3, the maximum voltage sensitivity increases to about 2.6 times that when the aspect ratio H / D=1. It can be observed that the longitudinal piezoelectric effect increases with the increase of the aspect ratio H / D. In addition, when the aspect ratio H / D=4, the height is further increased, and the peak frequency of the sensitivity curve is 6.3MHz, which is slightly shifted to the lower frequency side than when H / D=3, and the bandwidth is further narrowed slightly. The central voltage V P0 The maximum voltage sensitivity is about 85.5 / Pa. When the aspect ratio H / D=4, the maximum voltage sensitivity increases to about 3.2 times that when the aspect ratio H / D=1. It can be observed that the longitudinal piezoelectric effect increases with the increase of the aspect ratio H / D. Further, when the aspect ratio H / D=8, the sensitivity curve shifts to the lower frequency side than when H / D=4, and the bandwidth further narrows slightly. The central voltage V P0 The maximum voltage sensitivity is about 139.5 / Pa. When the aspect ratio H / D=8, the maximum voltage sensitivity increases to about 5.3 times that when the aspect ratio H / D=1. It can be observed that the longitudinal piezoelectric effect increases with the increase of the aspect ratio H / D.

[0189] Figure 4F and Figure 4E Similarly, when the entire vibrator 22a1 is made of PZT, the central area P detected just below the central convex portion is shown. 0 The central voltage V P0 The sensitivity of Figure 4E In the figure, the frequency dependence of the voltage sensitivity is shown when the aspect ratio (=H / D) of the central convex height H corresponding to the diameter D is changed when the diameter D = 2.4 μm (constant), and Figure 4F In the figure, it is shown that when the aspect ratio H / D=1 (constant), the diameter D=d 22a1 The frequency dependence of the voltage sensitivity at Figure 4F The thickness t of the flat plate portion of the vibrating body 22a1 other than the central convex portion is 22a1 Also 2.4 μm (constant). Figure 4F Middle diameter d 22a1 =2.4μm sensitivity curve and Figure 4E The sensitivity curve is the same when the aspect ratio H / D = 1. At the peak frequency of 7.3MHz, the central voltage VP0 The maximum voltage sensitivity is about 26.4μV / Pa.

[0190] With diameter d 22a1 =2.4μm, when the diameter d 22a1 = 4.8μm, when the central convex part is made thicker, the sensitivity curve shifts slightly to the high frequency side to a peak frequency of 7.8MHz, and the bandwidth becomes wider. 22a1 =4.8μm, the center voltage V P0 The maximum voltage sensitivity is about 12.8μV / Pa, and the maximum voltage sensitivity decreases to the diameter d 22a1 = about 0.48 times of that at 2.4 μm, it can be observed that as the diameter d 22a1 The longitudinal piezoelectric effect decreases relatively as the diameter d increases. 22a1 =7.2μm, when the central convex part is made thicker, the peak frequency becomes 8.6MHz, and the sensitivity curve is proportional to the diameter d 22a1 = 4.8μm, the higher frequency side is further shifted, and the bandwidth is slightly wider. 22a1 =7.2μm when the center voltage V P0 The maximum voltage sensitivity is reduced to about 8.3μV / Pa. Therefore, at a diameter of d 22 a 1 =7.2μm, the maximum voltage sensitivity is the diameter d 22 a 1 = about 0.31 times of that at 2.4 μm, it can be observed that as the central convex portion diameter d 22a1 As the longitudinal piezoelectric effect increases, it decreases relatively.

[0191] The parallel toroidal stress distribution reinforcement structure can be formed with the downward protrusion on the output surface of the vibrator as the center, or with the upward protrusion on the silicon substrate as the center. Figure 4A Different, such as Figure 5A As shown, the protrusion in the central part of the vibration cavity is the silicon substrate 11 sa2 The numerical simulation results when a part protrudes upward are as follows Figure 5C and Figure 5D As shown. Since the silicon substrate 11 having a cylindrical recess sa2 The central portion of the silicon substrate 11 has an upward protrusion as a step structure. sa2 The vibrating body 22a2 is inserted into the vibrating body 22a2, thereby forming a parallel toroidal vibration cavity 21a2 around the upward protrusion of the central part made of silicon. A fixed potential electrode 24a connected to the ground potential is arranged on the upper surface of the vibrating body 22a2. A fixed potential electrode protective film 23 made of epoxy resin is arranged on the fixed potential electrode 24a and on the vibrating body 22a2 where the fixed potential electrode 24a does not exist.

[0192] exist Figure 5B In the case of , as shown in Table 2, for the arrangement of 5 ideal potential detection electrodes from the center along the radius r 11 (=d 11 / 2) Direction (outward) Positions numbered sequentially P 0 , P 1 ,……,P 4 , the individual name of the potential transfer area is defined as the central area P 0 , the first peripheral area P 1 , ..., the fourth peripheral area P 4 . Compared with the Figure 4B The definition of the situation is different. The central potential transfer area (central area) P 0 The ideal electrode for potential detection is a circular island area.

[0193] Table 2

[0194] Table 2 General definitions

[0195]

[0196] Central Area 0 The line width and spacing width along the radial direction of the outer side are the same, and the four first peripheral areas P surrounded by the same spacing 1 , the second peripheral area P 2 , the third peripheral area P 4 , the fourth peripheral area P 4 Since the line width and the spacing width are the same, the radius r of the vibration cavity 21a1 is 11 (=d 11 / 2) is divided into 9 parts, defining the central area P 0 , the first peripheral area P 1 , ..., the fourth peripheral area P 4 That is, as defined in Table 2, the radius r of the vibration cavity 21a1 11 =d 11 / 2, Central Area P 0 Defined as the distance from the center to the radius (1 / 9) r 11 Also as defined in Table 2, the first peripheral area P 1 Defined as the radius from the position (2 / 9)r 11 To the radius position (3 / 9)r 11 The second peripheral area P 2 Defined as the radius from the position (4 / 9)r 11 To the radius position (5 / 9)r 11The third peripheral area P 3 Defined as the radius from the position (6 / 9)r 11 To the radius position (7 / 9)r 11 The fourth peripheral area P 4 Defined as the radius from position (8 / 9)r 11 To the outermost set (radius position r 11 ) area.

[0197] For convenience, in Figure 5B In the figure, five potential detection ideal electrodes are embedded in the lower surface of the vibrating body 22a2. However, in the simulation, the thickness of the potential detection ideal electrodes is zero. Figure 5B The surface of the embedded vibrating body 22a2 is also as shown in Figure 4B The plate-like structure shown protruding from the surface of the vibrating body 22a1 does not bring about any essential difference. 0 The line width and spacing width along the radial direction of the outer side are the same, and the four first peripheral areas P surrounded by the same spacing 1 , the second peripheral area P 2 , the third peripheral area P 4 , the fourth peripheral area P 4 are concentric rings. Figure 4B The same situation is mentioned in Figure 5B The fourth peripheral area P 4 The outer periphery of the potential detection ideal electrode and the silicon substrate 11 sa2 In the simulation, it is assumed that the silicon substrate 11 sa2 is semi-insulating, the fourth peripheral area P 4 The potential detection ideal electrode is in an electrically suspended state. Similarly, the central area P 0 The entire lower surface of the potential detection electrode is in contact with the silicon substrate 11 sa2 The upper end of the upward projection contacts the silicon substrate 11. sa2 is semi-insulated, the central area P 0 For potential detection, the ideal electrode is in an electrically suspended state.

[0198] Since the line width and space width are the same, Figure 5B In the arrangement shown, the radius r of the vibration cavity 21a1 is 11 (=d 11 / 2) by the first peripheral area P 1 , ..., the fourth peripheral area P 4 Therefore, the first peripheral area P 1 , ..., the fourth peripheral area P 4 The line width ΔrL =r 11 / 9 is constant. The diameter d of the upward protrusion in the central part of the vibration cavity 21a2 is 22a2 =2Δr L That is, the central area P is determined as in Table 2. 0 , the first peripheral area P 1 , ..., the fourth peripheral area P 4 Location, central area P 0 , the first peripheral area P 1 , ..., the fourth peripheral area P 4 The central voltage V p0 , the first peripheral voltage V p1 , ..., the fourth peripheral voltage V4 p The transient responses of Figure 5C shown.

[0199] Figure 5C This is the simulation result when PZT is used for the entire vibrating body 22a2. The protrusion diameter d at the center of the parallel toroidal vibration cavity 21a2 of the voltage generating part (22a2, 24a) is 22a2 =2Δr L = The voltage generated when receiving ultrasound at 6.7μm, showing the central voltage Vp 0 , the first peripheral voltage Vp 1 , ..., the fourth peripheral voltage V4 P The transient response of each. This is the thickness t of the vibrating body 22a2 22a2 =2.4 μm, the thickness t of the fixed potential electrode protection film 23 23 =6μm. Figure 5C As shown in the legend in the upper right corner, the thick solid line represents the central voltage V P0 The transient response of Figure 5C In the figure, the thick dotted line represents the first peripheral voltage V P1 , the thin dotted line represents the second peripheral voltage V P2 , the thin solid line represents the third peripheral voltage V P3 , the single-dot dash line represents the fourth peripheral voltage V P4 Their respective transient responses. Figure 5C When the ultrasonic wave consisting of a single pulse sine wave of pressure vibration with a maximum vibration amplitude of ±1Pa shown on the left is input to the voltage generating unit (22a2, 24a), the central area P 0 , the second peripheral area P 1 , ..., the fourth peripheral area P 4 The pulse response voltage signals generated by the piezoelectric effect show a transient response. The second vibration amplitude is the largest at about 0.13μs, and the third and subsequent vibrations are attenuated vibrations.

[0200] Specifically, Figure 5C The central voltage V P0 The first vibration amplitude reached a peak value at about -0.62μV, which was about 0.06μs after the ultrasonic pulse was input. In contrast, the second vibration amplitude reached a peak value at about +0.92μV, which was about the maximum amplitude value, at about 0.13μs. Thereafter, the third vibration amplitude reached a peak value at about -0.55μV, which was about 0.19μs after the ultrasonic pulse was input, and the fourth vibration amplitude reached a peak value at about 0.26μs after the ultrasonic pulse was input, which was about -0.33μV. Figure 5C The first peripheral voltage V shown in P1 The first vibration amplitude is about +0.13μV, reaching its peak value at about 0.06μs after the ultrasonic pulse input. In contrast, the second vibration amplitude reaches its peak value at about -0.18μV, reaching its maximum amplitude value, at about 0.13μs. However, the first peripheral voltage V P1 The absolute value of the amplitude |V P1 |Less than the central voltage V P0 The absolute value of the amplitude |V P0 |, central voltage V P0 and the first peripheral voltage V P1 After that, the first peripheral voltage V reaches a peak value at about 0.19 μs. P1 The third vibration amplitude is about +0.12μV, and the fourth vibration amplitude, which reaches a peak value at about 0.26μs, decays to about -0.03μV. Since the peak value of the vibration amplitude becomes smaller, the description of the fifth and subsequent vibrations is omitted. Figure 5C The fourth peripheral voltage V represented by a single-dot chain line P4 The first vibration amplitude reaches a peak value at about 0.06μs. In contrast, the second vibration amplitude reaches a peak value at about -0.09μV at about 0.13μs, reaching the maximum amplitude value. The fourth peripheral voltage V P4 The absolute value of the amplitude |V P4 |Less than the central voltage V P0 The absolute value of the amplitude |V P0 | and the first peripheral voltage V P1 The absolute value of the amplitude |V P1 |, is the third smallest value. The first peripheral voltage V P1 , the third peripheral voltage V P3 , the fourth peripheral voltage V P4 With the central voltage V P0 , the second peripheral voltage V P2 In anti-phase relationship, the central voltage VP0 and the second peripheral voltage V P2 Thereafter, the fourth peripheral voltage V P4 The amplitude of the third vibration, which reaches a peak value at about 0.19 μs, is about +0.03 μV, and the amplitude of the fourth vibration, which reaches a peak value at about 0.26 μs, decays to about -0.01 μV.

[0201] Figure 5C The third peripheral voltage V represented by the thin solid line P3 The absolute value of the amplitude of the transient response vibration |V P3 |With the fourth peripheral voltage V P4 The absolute value of V P4 | Roughly the same, but overall higher than the fourth peripheral voltage V P4 The absolute value of V P4 | is slightly smaller, and is the fourth smallest value. The third peripheral voltage V P3 and the first peripheral voltage V P1 , the fourth peripheral voltage V P4 The third peripheral voltage V P3 With the central voltage V P0 , the second peripheral voltage V P2 They vibrate in anti-phase relation to each other. Figure 5C The second peripheral voltage V represented by the thin dotted line P2 The absolute value of the amplitude of the transient response vibration |V P2 | is less than the third peripheral voltage V P3 The absolute value of |V P3 | and the fourth peripheral voltage V P4 The absolute value of V P4 |, is the smallest fifth value. The second peripheral voltage V P2 With the central voltage V P0 The second peripheral voltage V P2 and the first peripheral voltage V P1 , the third peripheral voltage V P3 , the fourth peripheral voltage V P4 They vibrate in anti-phase relation to each other. Figure 5A and Figure 5B The potential relationship of the output surface of the vibrating body 22a2 of the voltage generating unit (22a2, 24a) generated by the piezoelectric effect is a position-dependent potential distribution with a magnitude relationship as follows:

[0202] |V P0 |>|V P1 |>|V P4 |>|V P3 |>|V P2 |……(9)

[0203] Figure 5D and Figure 5C The same means that in the composition Figure 5A and Figure 5B The simulation results of the frequency dependence of the voltage sensitivity (μV / Pa) when the vibrating body 22a2 of the voltage generating unit (22a2, 24a) uses PZT. In the simulation, the thickness t of the vibrating body 22a2 is 22b = 2.4 μm, the thickness t of the fixed potential electrode protection film 23 on the vibrating body 22a2 23 =6μm, the protrusion diameter d22b of the central part of the parallel toroidal vibration cavity of the voltage generating part is the parameter.

[0204] like Figure 5D As shown, in the receiving frequency range of 5MHz to 10MHz, when the protrusion diameter d 22b = 1.2μm (reference value) when the maximum voltage sensitivity is about 10.6μV / Pa, which is the maximum value of the parameters used in the simulation. In the receiving frequency range of 5MHz to 10MHz, when the protrusion diameter d 22b = 2.4μm is twice the reference value, the maximum voltage sensitivity is about 5.4 (μV / Pa), which is reduced to about half. 22b = 3.6μm is 3 times the reference value, the maximum voltage sensitivity is about 3.3μV / Pa, which is reduced to about 1 / 3. 22b = 4.8μm is 4 times the reference value, the maximum voltage sensitivity is about 2.5μV / Pa, which is reduced to about 1 / 4. 22b =6.0μm is 5 times the reference value, the maximum voltage sensitivity is about 2.0μV / Pa, which is reduced to about 1 / 5. 22b =7.2μm is 6 times the reference value, the maximum voltage sensitivity is about 1.7μV / Pa, which is reduced to about 1 / 6. Figure 5D It can be seen that the convex diameter d of the central part of the parallel toroidal vibration cavity 21a2 of the voltage generating part (22a2, 24a) is 22b The larger it is, the smaller the maximum voltage sensitivity is.

[0205] Figure 5A and Figure 5B The voltage generating portion (22a2, 24a) shown in the figure is at the protrusion of the central part, and the vibrating body 22a2 is in the up-down direction ( Figure 5E The vibration of the silicon substrate 11 sa2 In the case where the central part is fixed, the vibrating body 22a2 located above the toroidal vibration cavity 21a2 is fixed. Figure 5EIt is shown in a wing shape, vibrating as an upward convex curved surface and a downward convex curved surface on both sides of the raised central portion. Figure 5E This indicates that the displacement inside the upwardly convex curved surface and the displacement inside the downwardly convex curved surface are asymmetric to each other. Figure 5E This is a model diagram for explaining the displacement and stress inside the vibrating body 22a2. In the no-load state without bending, it is assumed that the vibrating body 22a2 is in the horizontal direction ( Figure 5E The figure shows a model in which 18 zeros are arranged evenly at intervals of about 3.7 μm in the X direction when arranged flatly in an unloaded state, and these zeros are connected and arranged in the lateral direction by elastic bodies (springs).

[0206] Please note that Figure 5E The Y direction used in corresponds to the z direction in equations (2a) to (5c), Figure 5E The x direction corresponds to the x direction or the y direction in equations (2a) to (5c). Figure 5E In the stress model shown, Figure 5A and Figure 5B The voltage generating part (22a2, 24a) shown in the figure has a structure in which the vibrating body 22a2 is fixed at the protrusion in the central part, forming a diameter d 11 =Toroidal vibration cavity 21a2 with a diameter of 66 μm. Figure 5E is Figure 5A and Figure 5B In the structure shown, when PZT is used in the vibrating body 22a2, it is a diagram for explaining the displacement inside the vibrating body 22a2 in a vibrating state. Figure 5E The lower part of the diagram shows the arrangement of the zeros when the vibrating body 22a2 is bent downward as two downward convex curves, and the displacement position of each zero when 0.13μS has passed after the ultrasonic pulse arrives. On the other hand, the upper part shows the arrangement of the zeros when the vibrating body 22a2 is bent upward as two upward convex curves, and the displacement position of each zero when 0.18μS has passed after the ultrasonic pulse arrives. It shows the degree of displacement of zeros arranged at equal intervals in the X direction in the unloaded state by the vibration of the vibrating body 22a2 in the X direction.

[0207] exist Figure 5EIn the case of the above, when the vibrating body 22a2 is bent upward convexly on both sides of the fixed position of the central part, the interval between 0 and 0 in the vicinity of the fixed position of the central part tends to be reduced in the X direction. On the other hand, when the vibrating body 22a2 is bent downward convexly on both sides of the fixed position of the central part, the interval between 0 and 0 in the vicinity of the fixed position of the central part tends to be expanded in the X direction, so the change in the interval between 0 and 0 of the vibrating body 22a2 is asymmetric when bending upward and bending downward. Therefore, it can be seen that the stress inside the vibrating body 22a2 is also uneven.

[0208] exist Figure 5A and 5B When the vibrator 22a2 uses PZT, Fig. 5F (a) represents the stress change in the XX direction, Fig. 5F (b) represents the stress change in the YY direction, Fig. 5F (c) shows the stress variation in the XY direction. Fig. 5F (a)~ Fig. 5F The X and Y directions used in (c) are the same as Figure 5E The X direction and Y direction defined in are the same. Fig. 5F (a)~ Fig. 5F (c) The line types in the upper right corner are shown in the legend. The thick solid line represents the central area P 0 The central stress T p0 The transient response of Fig. 5F (a)~ Fig. 5F In (c), the thick dotted line indicates the first peripheral area P 1 The first peripheral stress T generated in p1 , the thin dotted line indicates the second peripheral area P 2 The second peripheral stress T generated in p2 , the thin solid line represents the third peripheral area P 3 The third peripheral stress T generated in p3 , the single-dot chain line represents the fourth peripheral area P 4 The fourth peripheral stress T4 generated in p Their respective transient responses. Figure 5C As shown on the left, when an ultrasonic wave consisting of a single pulse sine wave of pressure vibration with a maximum vibration amplitude of ±1Pa is input to the voltage generating unit (22a2, 24a), in the central area P 0 , the first peripheral area P 1 , ..., the fourth peripheral area P 4 The corresponding central stress T in the XX direction generated in the vibrating body 22a2 p0 , the first peripheral stress T p1 , ..., the fourth peripheral stress T4 pIt exhibits a transient response, with the second vibration amplitude being the largest at about 0.13 μs, and the third and subsequent vibrations being attenuated vibrations.

[0209] by Fig. 5F The central stress T in the XX direction is represented by the thick solid line in (a). p0 For example, the first vibration amplitude reaching the peak value at about 0.06μs after the ultrasonic pulse input is about -50N / m 2 In contrast, the second vibration amplitude, which reaches a peak value around 0.13 μs, is about +120 N / m. 2 , reaching the maximum amplitude value. After that, the third vibration amplitude reached the peak value at about 0.19μs, which was about -110N / m 2 The fourth vibration amplitude reaches a peak value of about +60N / m at about 0.26μs. 2 The fifth vibration amplitude, which reaches a peak value at about 0.31μs, is about -40N / m 2 After reaching the peak value at about 0.37μs, the amplitude of the sixth vibration decays to about +20N / m 2 Other primary peripheral stresses T p1 , ..., the fourth peripheral stress T4 p The same trend is also shown. Fig. 5F (a) The central stress T in the XX direction generated in the vibrating body 22a2 p0 , the first peripheral stress T p1 , ..., the fourth peripheral stress T4 p The central stress T p0 maximum.

[0210] like Fig. 5F As shown in (b), the stress (N / m 2 ) also shows that the second vibration amplitude is the largest around 0.13μs, and the third and subsequent vibrations are transient responses of attenuated vibrations. The central stress T in the YY direction represented by the thick solid line p0 For example, the first vibration amplitude reaching the peak value at about 0.06μs after the ultrasonic pulse input is about +15N / m 2 In contrast, the second vibration amplitude, which reaches a peak value around 0.13 μs, is about -42 N / m. 2 , reaching the maximum amplitude value. After that, the third vibration amplitude reached the peak value at about 0.19μs, which was about +39N / m 2 The fourth vibration amplitude, which reaches a peak value at about 0.26μs, is about -22N / m. 2The fifth vibration amplitude, which reaches a peak value at about 0.31μs, is about +12N / m. 2 After reaching the peak value at about 0.37μs, the amplitude of the sixth vibration decays to about -8N / m 2 Other primary peripheral stresses T p1 , ..., the fourth peripheral stress T4 p The same trend is also shown. Fig. 5F (b) The central stress T in the YY direction generated in the vibrating body 22a2 p0 , the first peripheral stress T p1 , ..., the fourth peripheral stress T4 p In the XX direction, the central stress T p0 maximum.

[0211] Similarly, when an ultrasonic pulse composed of a sine wave is input to the voltage generating section (22a2, 24a), in the central area P 0 , the first peripheral area P 1 , ..., the fourth peripheral area P 4 The central stress T in the XY direction generated in the corresponding vibrating body 22a2 p0 , the first peripheral stress T p1 , ..., the fourth peripheral stress T4 p Also like Fig. 5F As shown in (c), the second vibration amplitude is the largest around 0.13μs, and the third and subsequent vibrations are transient responses of attenuated vibrations. Fig. 5F The central stress T in the XY direction is represented by the thick solid line in (c). p0 For example, the first vibration amplitude reaching the peak value at about 0.06μs after the ultrasonic pulse input is about +12N / m 2 In contrast, the second vibration amplitude, which reaches a peak value around 0.13 μs, is about -35 N / m. 2 , reaching the maximum amplitude value. After that, the third vibration amplitude reached the peak value at about 0.19μs, which was about +32N / m 2 The fourth vibration amplitude, which reaches a peak value at about 0.26μs, is about -20N / m. 2 The fifth vibration amplitude, which reaches a peak value at about 0.31μs, is about +11N / m. 2 After reaching the peak value at about 0.37μs, the amplitude of the sixth vibration decays to about -7N / m 2 Other primary peripheral stresses T p1 , ..., the fourth peripheral stress T4 p The same trend is also shown. Fig. 5F(c) Among the stresses in the XY direction generated in the vibrating body 22a2, the central stress T p0 maximum.

[0212] Next, Figure 5B The central part of the Fig. 6A As shown, Figure 5B The central area P 0 Subdivided into subdivided central area P 00 , the first subdivided surrounding area P 01 , the second subdivided surrounding area P 02 , the third subdivision surrounding area P 03 , and verify Figure 5B The central bulge and the small potential distribution of the voltage generated near the central bulge. 00 Located in the center of the central bulge. The subdivided central area P around the center 00 , the first subdivided surrounding area P 01 , the second subdivided surrounding area P 02 , the third subdivision surrounding area P 03 Located around it. Fig. 6A In the subdivision central area P 00 , the first subdivided surrounding area P 01 , the second subdivided surrounding area P 02 , the third subdivision surrounding area P 03 The positions are configured with Figure 5B Ideal electrode for potential detection in more detailed situations. Silicon substrate 11 sa The vibrating body 22a2 is inserted into the vibrating body 22a2, and a parallel toroidal vibrating cavity 21a2 is provided around the protrusion of the central part. The vibrating body 22a2 is provided with a fixed potential electrode 24a and a fixed potential electrode protective film 23. Figure 5B same.

[0213] The first subdivision surrounding area P outside the ideal electrode for potential detection 01 , the second subdivided surrounding area P 02 , the third subdivision surrounding area P 03 The position is surrounded by three potential detection ideal electrodes as concentric rings. The first subdivision surrounding area P 01 , the second subdivided surrounding area P 02 , the third subdivision surrounding area P 03 The potential detection ideal electrode is in the radial direction, so that the line width Δr L and the spacing width Δr S Same (Δr L =Δr S) and are arranged at the same spacing. Since the line width and spacing width of the potential detection ideal electrode are the same, the potential detection ideal electrode will have a central protrusion radius r 22a2 (=d 22a2 / 2) is divided into 7 parts, defining the subdivision central area P 00 , the first subdivided surrounding area P 01 , the second subdivided surrounding area P 02 , the third subdivision surrounding area P 03 The position along the radius. Therefore, the convex diameter d of the central part of the vibration cavity 22a2 Roughly configured in the third subdivision surrounding area P 03 The potential detection ideal electrode P 03 The outer diameter is 14Δr S Equal (14Δr S =d 22a2 ). And, let the radius of the central protrusion be r 22a2 =d 22a2 / 2, subdivided into central area P 00 Defined as: from the center to the radius position (1 / 7) r 22a2 , an area of ​​an ideal electrode for detecting a ring potential with a certain line width.

[0214] Similarly, the first subdivided surrounding area P 01 Defined as: from the radius position (2 / 7)r 22a2 To the radius position (3 / 7)r 22a2 , an area where an ideal potential detection electrode with a certain line width is configured. The second subdivision peripheral area P 02 Defined as: from the radius position (4 / 7)r 22a2 To the radius position (5 / 7)r 22a2 , an area where an ideal potential detection electrode with a certain line width is configured, and the third subdivision peripheral area P 03 Defined as: From the radius position (6 / 7)r 22a2 To the radius position (7 / 7)r 22a2 , an area where an ideal potential detection electrode with a certain line width is configured. In addition, the fourth subdivided peripheral area P 04 Defined as: From the radius position (8 / 7)r 22a2 To the radius position (9 / 7)r 22a2 , an area of ​​an ideal potential detection ring electrode with a certain line width is arranged on the vibration cavity 21a2 outside the central protrusion. 03 Defined as: From the radius position (10 / 7)r 22a2 To the radius position (11 / 7)r 22a2On the vibration cavity 21a2 outside the central protrusion, an area of ​​an ideal ring-shaped potential detection electrode with a certain line width is arranged.

[0215] In the simulation, it is assumed that the subdivided central area P 00 , the first subdivided surrounding area P 01 , ..., the fifth subdivision surrounding area P 05 The six potential detection ideal electrodes are all in an electrically open state (open circuit). On the output surface of the vibrating body 22a2, the central area P of each subdivision is 00 , the first subdivided surrounding area P 01 , ..., the fifth subdivision surrounding area P 05 The radial electric field inside the potential detection ideal electrode is short-circuited by the potential detection ideal electrode.

[0216] Figure 6B When the vibrating body 22a2 is entirely made of PZT, the convex diameter d of the central portion of the parallel toroidal vibrating cavity 21a2 of the voltage generating portion (22a2, 24a) is 22a2 = 2.4μm, the transient response of voltage is generated when ultrasonic wave is received, and the potential transfer area is subdivided into the central area P 00 , the first subdivided surrounding area P 01 , ..., the fifth subdivision surrounding area P 05 The position of is a diagram of the parameters. This is the thickness t of the vibrating body 22a2 excluding the central protrusion. 22a2 =2.4 μm, the thickness t of the fixed potential electrode protection film 23 0203 =6.0μm. Figure 6B The line type legend in the upper right corner shows that the thick solid line represents the subdivision of the central area P 00 The voltage generated in the subdivision center voltage V P00 The transient response of Figure 6B In the figure, the thick dotted line indicates the first subdivision surrounding area P 01 The voltage generated in the first subdivision peripheral voltage V P01 , the dotted line represents the second subdivision surrounding area P 02 The voltage generated in the second subdivision peripheral voltage V P02 , the single-point dash line represents the third subdivision surrounding area P 03 The voltage generated in the third subdivision peripheral voltage V P03 , the thin solid line represents the fourth subdivision surrounding area P 04 The voltage generated in the fourth subdivision is the peripheral voltage V P04 , the thin dotted line represents the fifth subdivision surrounding area P 05 The voltage generated in the fifth subdivision is the peripheral voltage V P05 respective transient responses.

[0217] like Figure 6B As shown on the left, when an ultrasonic wave consisting of a 1.5-cycle sine wave that vibrates asymmetrically at 0.3 Pa on the positive side and 1 Pa on the negative side is input to the voltage generating unit (22a2, 24a), the subdivided central area P 00 , the first subdivided surrounding area P 01 , ..., the fifth subdivision surrounding area P 05 The voltage signal generated in the experiment shows that the amplitude of the second vibration reaches the maximum value at about 0.13μs, and the third and subsequent vibrations are transient responses of attenuated vibrations.

[0218] Specifically, Figure 6B The first subdivided central region P of the central portion of the vibrating body 22a2 is indicated by a thick solid line. 00 The subdivision center voltage V P00 The amplitude of the first vibration, which peaked at about -1.3μV after about 0.06μs after the ultrasonic pulse was input, was about +3.5μV, reaching the maximum amplitude value. After that, the amplitude of the third vibration, which peaked at about 0.19μs, was about -3.2μV, and the amplitude of the fourth vibration, which peaked at about 0.26μs, decayed to about +1.8μV. Figure 6B The first subdivided peripheral area P indicated by the thick dashed line 01 The first subdivision peripheral voltage V P01 The first vibration amplitude, which reaches a peak value at about 0.06μs after the ultrasonic pulse input, is about -1.2μV. In contrast, the second vibration amplitude, which reaches a peak value at about 0.13μs, is about +3.4μV, reaching the maximum amplitude value. However, the first subdivision peripheral voltage V P01 The absolute value of the amplitude |V P01 | Less than the subdivision center voltage V P00 The absolute value of the amplitude |V P00 |. After that, the first subdivision peripheral voltage V reaches a peak value at about 0.19 μs. P01 The third vibration amplitude is about -3.1 μV, and the fourth vibration amplitude, which reaches a peak value at about 0.26 μs, decays to about +1.7 μV. Since the peak value of the vibration amplitude becomes smaller, the description of the fifth and subsequent vibrations is omitted.

[0219] Figure 6B The second subdivision surrounding area P shown by the dotted line 02 The second subdivision peripheral voltage V P02 The absolute value of the amplitude of the transient response vibration |V P02 |, overall, it is better than the first subdivision surrounding area P01 The first subdivision peripheral voltage V P01 The absolute value of V P01 | is slightly smaller, which is the third value. In addition, Figure 6B The third subdivision surrounding area P shown by the single-dot chain line 03 The third subdivision peripheral voltage V P03 The absolute value of the amplitude of the transient response vibration |V P03 |, overall, it is higher than the 2nd subdivision peripheral voltage V P02 The absolute value of V P02 | is slightly smaller, which is the 4th value. In addition, Figure 6B The fourth subdivision surrounding area P shown by a thin solid line 04 The 4th subdivision peripheral voltage V P04 The absolute value of the amplitude of the transient response vibration |V P04 |, overall, the peripheral voltage V P03 The absolute value of V P03 |Slightly smaller, the 5th value.

[0220] and, Figure 6B The fifth subdivision surrounding area P shown by the thin dotted line 05 The 5th subdivision peripheral voltage V P05 The first vibration amplitude reaches a peak value of about -0.2μV at about 0.06μs, and the second vibration amplitude reaches a peak value of about +0.6μV at about 0.13μs, reaching the maximum amplitude value. The 5th subdivision peripheral voltage V P05 The absolute value of the amplitude |V P05 | Ratio subdivision central voltage V P00 To the 4th subdivision peripheral voltage V P04 The absolute value of V P00 |To|V P04 | is the smallest value. It should be noted that the subdivided central voltage V P00 To the 5th subdivision peripheral voltage V P05 are in phase relationship. After that, the 5th subdivision peripheral voltage V P05 The amplitude of the third vibration, which reaches a peak value near the time of about 0.19 μs, is about -0.4 μV, and the amplitude of the fourth vibration, which reaches a peak value near the time of about 0.26 μs, decays to about +0.2 μV. That is, the potential relationship generated by the piezoelectric effect on the output surface of the vibration body 22a2 of the voltage generating unit (22a2, 24a) is:

[0221] |V P00 |>|V P01 |>|V P02 |>|V P03 |>|VP04 |>|V P05 |……(10)

[0222] The potential distribution exhibits position dependence with a magnitude relationship.

[0223] Figure 6C The diameter d of the central convex portion of the vibration cavity 21a2 is shown in the case where the vibration body 22a2 uses PZT. 22a2 is a parameter showing the relationship between the voltage (μV) generated on the vibrating body 22a2 and the generated position (μm) measured from the center. 22a2 =2.4 μm, the thickness t of the fixed potential electrode protection film 23 23 =6μm. Figure 6C As shown, the curve representing the voltage position dependence is symmetrical about the center of the voltage generating portion (22a2, 24a), and it can be seen that the diameter d of the convex portion in the central part of the vibration cavity 21a2 is 22a2 The smaller the convex portion is, the greater the voltage generated is. Specifically, when the convex portion diameter d 22a2 = 2.4 μm (reference value), the maximum generated voltage is about 3.2 μV, which is the largest among the parameters used in the simulation. 22a2 =4.8μm, which is twice the reference value, the maximum generated voltage is about 1.5μV, which is about half the value. 22a2 =7.2μm, which is 3 times the reference value, the maximum generated voltage is about 1.0μV, which drops to about 1 / 3; when the convex diameter d 22a2 = 9.6μm, which is 4 times the reference value, the maximum generated voltage is about 0.8μV, which is reduced to about 1 / 4. Figure 6C It can be seen that the diameter d of the convex portion of the central part of the vibration cavity 21a2 is 22a2 The larger it is, the smaller the maximum generated voltage is.

[0224] - V-shaped (inclined torus) -

[0225] Next, the vibrating body 22b constituting the voltage generating unit (22b, 24b) is subjected to the following conditions: Fig. 7A The stress distribution of the V-shaped cross-sectional structure shown in the figure is strengthened. The voltage distribution on the output surface of the structure is studied at different positions as a basis for high sensitivity and wide bandwidth. The vibrating body 22b is arranged on the silicon substrate 11 sb A grounded fixed potential electrode 24b is disposed on the vibrating body 22b, and a fixed potential electrode protective film 23 made of resin is disposed on the fixed potential electrode 24b. The vibrating body 22b is disposed on the silicon substrate 11 sb The concave part of the Fig. 7AAs shown, since the cross-sectional shape is V-shaped, the lower tip of the V-shaped is in contact with the silicon substrate 11 sb Therefore, in the space formed by the concave portion, on both sides of the V-shaped cut-in area, there are inclined toroidal vibration cavities 21b with two triangular cross-sections. "Inclined toroidal" refers to the cross-sectional structure of the space forming the toroid, such as Fig. 7A As shown in FIG. 1 , a three-dimensional structure is shown in which one main surface is inclined relative to the other main surface. In the simulation, according to the same definition as in Table 2, the central area P at the lower tip of the V-shape 0 Configure the ideal electrode for potential detection and place it in the central area P 0 Around, along the V-shaped shape, configure the central area P 0 The first peripheral area P of the four concentric rings 1 , ..., the fourth peripheral area P 4 . Figure 7B This is a simulation result when PZT is used for the entire vibrating body 22b. sb Point contact with the vibrating body 22b (assuming contact width d 22b =0μm), the voltage transient response is generated in the potential transfer area, that is, the central area P 0 , the first peripheral area P 1 , ..., the fourth peripheral area P 4 The position of the vibration body 22b is shown as a parameter. 22b =4.8 μm, the thickness t of the fixed potential electrode protection film 23 23 =6μm. Figure 7B When a single-frequency sine wave with a maximum vibration amplitude of ±1 Pa shown on the left is input to the voltage generating section (22b, 24b), the potential transfer region, i.e., the central region P 0 , the first peripheral area P 1 , ..., the fourth peripheral area P 4 The impulse responses generated in the two experiments show that the second vibration amplitude reaches the maximum at about 0.13μs, and the third and subsequent vibrations are transient responses with attenuated vibrations. Specifically, Figure 7B The central area P indicated by the thick solid line 0 The central voltage V p0 The transient response of the ultrasonic pulse input is about -0.8μV at the first vibration amplitude after about 0.06μs. In contrast, the second vibration amplitude, which reaches a peak after about 0.13μs, is about +1.5μV, reaching the maximum amplitude value. Thereafter, the third vibration amplitude, which reaches a peak after about 0.19μs, is about -1.2μV, and the fourth vibration amplitude, which reaches a peak after about 0.26μs, decays to about +0.7μV. Figure 7BThe first peripheral voltage V p1 The transient response of the first vibration is about -0.6μV after about 0.06μs of pulse input. In contrast, the second vibration reaches a peak value of about +1.2μV after about 0.13μs, reaching the maximum amplitude value. However, the first peripheral voltage V p1 The absolute value of the amplitude |V p1 |Ratio to central voltage V p0 The absolute value of the amplitude |V p0 | Small. After that, the first peripheral voltage V reaches a peak value after about 0.19 μs. p1 The 3rd vibration amplitude is about -0.7 μV, and the 4th vibration amplitude, which reaches a peak value after about 0.26 μs, decays to about +0.6 μV. The peak values ​​of the vibration amplitudes after the 5th vibration amplitude become smaller, so the description is omitted.

[0226] Figure 7B The second peripheral voltage V p2 The absolute value of the amplitude |V p2 | Overall, the first peripheral voltage V p1 The absolute value of the amplitude |V p1 | is slightly smaller, which is the third value. In addition, Figure 7B The third peripheral voltage V represented by the thin solid line p3 The absolute value of the amplitude |V p3 | Overall, the second peripheral voltage V p2 The absolute value of the amplitude |V p2 |Slightly smaller, the 4th value.

[0227] and, Figure 7B The fourth peripheral voltage V represented by a single-dot chain line p4 The transient response of the 1st vibration amplitude is about 0.2μV after about 0.06μs. In contrast, the 2nd vibration amplitude reaches the peak value of about +0.4μV after about 0.13μs, reaching the maximum amplitude value. However, the 4th peripheral voltage V p4 The absolute value of the amplitude |V p4 |Ratio to central voltage V p0 ~3rd peripheral voltage V p3 The absolute value of the amplitude |V p0 |~|V p2 | is the smallest value. It should be noted that the central voltage V p0 To the 4th peripheral voltage V p4 Therefore, the absolute values ​​of the potentials generated on the output surface of the vibrator 22b are related to each other as follows:

[0228] |V p0 |>|V p1|>|V p2 |>|V p3 |>|V p4 |……(11)

[0229] The potential distribution exhibits position dependence with a magnitude relationship.

[0230] Figure 7C and Figure 7B Similarly, it shows Fig. 7A The simulation results are shown in the case where the vibrating body 22b uses PZT. When it is assumed that the central portion of the vibration cavity 21b is in the silicon substrate 11 sb The contact width d with the vibrating body 22b 22b = 6.7μm, the voltage transient response is generated, and the potential transfer area, that is, the central area P 0 , the first peripheral area P 1 , ..., the fourth peripheral area P 4 The position of the vibration body 22b is shown as a parameter. 22b =2.4 μm, the thickness t of the fixed potential electrode protection film 23 23 =4μm. Figure 7C When a single-frequency sine wave with a maximum vibration amplitude of ±1 Pa shown on the left is input to the voltage generating section (22b, 24b), the potential transfer region, i.e., the central region P 0 , the first peripheral area P 1 , ..., the fourth peripheral area P 4 The impulse responses generated in Figure 7B Similar to the example of , it shows that the amplitude of the second vibration reaches the maximum around 0.13 μs, and the third and subsequent vibrations are transient responses of attenuated vibrations.

[0231] Specifically, Figure 7C The central area P indicated by the thick solid line 0 The central voltage V p0 The transient response is that the first vibration amplitude is about 0.6μV after about 0.06μs after the pulse input. In contrast, the second vibration amplitude peak is about +0.9μV after about 0.13μs, reaching the maximum amplitude value. Thereafter, the third vibration amplitude peak is about -0.5μV after about 0.19μs, and the fourth vibration amplitude peak decays to about +0.3μV after about 0.26μs. The vibration amplitude peaks after the fifth time become smaller, so the description is omitted.

[0232] Figure 7C The first peripheral voltage V p1The transient response of the first vibration amplitude peak is about 0.16μV after about 0.06μs of pulse input. In contrast, the second vibration amplitude peak is about +0.18μV after about 0.13μs, reaching the maximum amplitude value. However, the first peripheral voltage V p1 The absolute value of the amplitude |V p1 |Ratio to central voltage V p0 The absolute value of the amplitude |V p0 | Small. After that, the first peripheral voltage V reaches a peak value after about 0.19 μs. p1 The amplitude of the third vibration is about -0.12μV, and the amplitude of the fourth vibration, which reaches a peak after about 0.26μs, decays to about +0.6μV.

[0233] Figure 7C The second peripheral voltage V p2 The absolute value of the amplitude |V p2 |, overall, the first peripheral voltage V p1 The absolute value of the amplitude |V p1 | slightly larger. In addition, Figure 7C The third peripheral voltage V represented by the thin solid line p3 The absolute value of the amplitude |V p3 |, overall, the first peripheral voltage V p1 and the second peripheral voltage V p2 The absolute value of the amplitude |V p1 | and |V p2 |Slightly smaller, the 4th value.

[0234] and, Figure 7C The fourth peripheral voltage V represented by a single-dot chain line p4 The second vibration amplitude peak value after about 0.13 μs also reaches the maximum amplitude value. However, the fourth peripheral voltage V p4 The absolute value of the amplitude |V p4 |Ratio to central voltage V p0 ~3rd peripheral voltage V p3 The absolute value of the amplitude |V p0 |~|V p3 | is the smallest value. It should be noted that the central voltage V p0 ~4th peripheral voltage V p4 Therefore, the absolute values ​​of the potentials generated on the output surface of the vibrator 22b are related to each other as follows:

[0235] |V p0 |>|V p2 |>|V p1 |>|V p3 |>|V p4 |……(12)

[0236] The potential distribution exhibits position dependence with a magnitude relationship.

[0237] -Λ shape-

[0238] Then, if Fig. 8A As shown in FIG. 1 , the voltage distribution on the output surface of the stress distribution reinforcement structure having a cross-sectional structure of an Λ-shaped structure is studied at each position as a basis for high sensitivity and wide bandwidth. The vibrating body 22c is arranged on the silicon substrate 11 sc The grounded fixed potential electrode 24c is disposed on the vibrating body 22c, and the fixed potential electrode protection film 23 made of resin is disposed on the fixed potential electrode 24c. The vibrating body 22c is on the silicon substrate 11 sc The top is a cone that tapers upward, Fig. 8A As shown, the cross-sectional shape is a Λ-shaped shape, and the bottom two sides of the Λ-shaped shape are adjacent to the silicon substrate 11 sc Therefore, on the silicon substrate 11 sc A conical vibration cavity 21c with an isosceles triangle cross section is provided on the lower surface of the vibration body 22c. In addition, on the lower surface of the vibration body 22c, in order to verify the generated voltage distribution, a circular potential detection ideal electrode is arranged at the upper tip of the Λ shape, and a potential detection ideal electrode P is arranged around the central potential detection ideal electrode P 0 Four concentric ring-shaped ideal potential detection electrodes are configured. Starting from the inner side, the potential transfer area is defined as the central area P 0 , the first peripheral area P 1 , ..., the fourth peripheral area P 4 (See Table 2). Then, by verifying the central area P 0 , the first peripheral area P 1 , ..., the fourth peripheral area P 4 The transient response of the voltages generated in each of the two circuits is simulated to simulate the voltage distribution on the output surface (lower surface) of the vibrating body 22c.

[0239] Figure 8B This is the simulation result when the entire vibrator 22c is made of PZT. 22c =3.6 μm, the thickness t of the fixed potential electrode protection film 23 on the vibrating body 22c 23 = 6μm, a voltage transient response occurs in the central area P as the potential transfer area. 0 ~4th surrounding area P 4 The position of the parameter is displayed. Figure 8B When a single-frequency sine wave with a maximum vibration amplitude of ±1 Pa shown on the left is input to the voltage generating section (22c, 24c), the central region P as the potential transmission region is 0, the first peripheral area P 1 , ..., the fourth peripheral area P 4 The transient response of the impulse response generated in each case is that the amplitude of the second vibration reaches the maximum at about 0.13μs, and the transient response after the third vibration is a decaying vibration.

[0240] Specifically, Figure 8B The central voltage V of the central part of the vibrating body 22c indicated by the thick solid line is p0 The transient response is that the first vibration amplitude peak is about +0.61μV after about 0.06μs after the pulse input. Compared with the first vibration amplitude peak, the second vibration amplitude peak is about -1.0μV after about 0.13μs, reaching the maximum amplitude value. Thereafter, the third vibration amplitude peak is about +0.63μV after about 0.19μs, and the fourth vibration amplitude peak decays to about -0.5μV after about 0.26μs. Figure 8B The first peripheral voltage V p1 The transient response of the first vibration amplitude peak is about +0.42μV after about 0.06μs after the pulse input. Compared with the first vibration amplitude peak, the second vibration amplitude peak is about -0.74μV after about 0.13μs, reaching the maximum amplitude value. However, the first peripheral voltage V p1 The absolute value of the amplitude |V p1 |Ratio to central voltage V p0 The absolute value of the amplitude |V p0 | Small. After that, the first peripheral voltage V reaches a peak value after about 0.19 μs. p1 The peak value of the 3rd vibration amplitude is about +0.45 μV, and the 4th vibration amplitude, which reaches the peak value after about 0.26 μs, decays to about -0.38 μV. The peak values ​​of the vibration amplitudes after the 5th vibration amplitude become smaller, so the description is omitted.

[0241] Figure 8B The second peripheral voltage V p2 The absolute value of the amplitude |V p2 | Overall, the first peripheral voltage V p1 The absolute value of the amplitude |V p1 | is slightly smaller, which is the third value. In addition, Figure 8B The third peripheral voltage V represented by the thin solid line p3 The absolute value of the amplitude |V p3 | Overall, the second peripheral voltage V p2 The absolute value of the amplitude |V p2 |Slightly smaller, the 4th value.

[0242] and, Figure 8B The fourth peripheral voltage V represented by a single-dot chain linep4 The transient response of the 1st vibration amplitude peak is about +0.28μV after about 0.06μs. Compared with the 1st vibration amplitude peak, the 2nd vibration amplitude peak is about -0.5μV after about 0.13μs, reaching the maximum amplitude value. However, the 4th peripheral voltage V p4 The absolute value of the amplitude |V p4 |Ratio to central voltage V p0 ~3rd peripheral voltage V p3 The absolute value of the amplitude |V p0 |~|V p3 | is the smallest value. In addition, the central voltage V p0 ~4th peripheral voltage V p4 Therefore, the mutual relationship between the absolute values ​​of the potentials generated on the output surface of the vibrator 22c is the same as that of equation (11), showing a position-dependent potential distribution with a magnitude relationship.

[0243] -M shape-

[0244] Next, if Fig. 9A As shown in FIG. 1 , the voltage distribution on the output surface of the vibrating body 22d having an M-shaped cross-sectional structure is studied at each position as a basis for high sensitivity and wide bandwidth. The vibrating body 22d is arranged on the silicon substrate 11 sd The grounded fixed potential electrode 24d is disposed on the vibrating body 22d, and the fixed potential electrode protective film 23 made of resin is disposed on the fixed potential electrode 24d. Since the cross-sectional structure is an M-shaped structure, the vibrating body 22d has a tapered shape that tapers downward in the center, and has a tapered portion that surrounds the tapered shape and tapers upward. In addition, the lower end of the tapered shape in the center of the M-shaped vibrating body 22d is close to the silicon substrate 11. sd The two sides of the bottom are respectively connected to the silicon substrate 11 sd Therefore, on the silicon substrate 11 sd The inclined toroidal vibration cavity 21d is formed by two spaces with roughly isosceles triangle cross-sections in a mirror-symmetrical manner. In addition, in order to verify the generated voltage distribution, a circular potential detection ideal electrode and four concentric ring-shaped potential detection ideal electrodes surrounding the potential detection ideal electrode in the central part are arranged at the lower end of the cone in the central part of the M-shaped vibration body 22d, and the potential transfer area is defined as the central area P 0 , the first peripheral area P 1 , ..., the fourth peripheral area P 4 By verifying the central area P 0 , the first peripheral area P 1 , ..., the fourth peripheral area P 4The transient response of the voltages generated in each of the two vibrations is simulated to simulate the voltage distribution on the output surface (lower surface) of the vibration body 22d.

[0245] Fig. 9B This is a simulation result when the entire vibrator 22d is made of PZT. 22d =3.6 μm, the thickness t of the fixed potential electrode protection film 23 on the vibrating body 22d 23 =6μm, the voltage transient response is generated, with the potential transfer area, i.e. the central area P 0 , the first peripheral area P 1 , ..., the fourth peripheral area P 4 The position of the parameter is displayed. Fig. 9B When a single-frequency sine wave with a maximum vibration amplitude of ±1 Pa shown on the left is input to the voltage generating section (22d, 24d), the potential transfer region, i.e., the central region P 0 , the first peripheral area P 1 , ..., the fourth peripheral area P 4 The transient responses of the impulse responses generated in the figure show that the amplitude of the second vibration reaches a maximum around 0.13 μs, and the third and subsequent vibrations are attenuated transient responses.

[0246] Specifically, Fig. 9B The central voltage V of the central part of the vibrating body 22d indicated by the thick solid line is p0 The transient response is that the first vibration amplitude peak is about +0.38μV after about 0.06μs after the pulse input. Compared with the first vibration amplitude peak, the second vibration amplitude peak is about -0.65μV after about 0.13μs, reaching the maximum amplitude value. Thereafter, the third vibration amplitude peak is about +0.4μV after about 0.19μs, and the fourth vibration amplitude peak decays to about -0.35μV after about 0.26μs. The vibration amplitude peaks after the fifth time become smaller, so the description is omitted. Fig. 9B The first peripheral voltage V p1 The transient response of the pulse input changes irregularly and reaches a peak value after about 0.13μs. p1 Other than the central voltage V p0 , the second peripheral voltage V p2 ~4th peripheral voltage V p4 The absolute value of the amplitude |V p0 |、|V p2 |~|V p4 Any one of | is smaller, which is the minimum value.

[0247] Fig. 9BThe second peripheral voltage V p2 The transient response is that the first vibration amplitude peak is about 0.42μV after about 0.06μs after the pulse input. Compared with the first vibration amplitude peak, the second vibration amplitude peak is about +0.65μV after about 0.13μs, reaching the maximum amplitude value. Thereafter, the third vibration amplitude peak is about -0.44μV after about 0.19μs, and the fourth vibration amplitude peak decays to about +0.35μV after about 0.26μs. The vibration amplitude peaks after the fifth time become smaller, so the description is omitted. The second peripheral voltage V p2 The absolute value of the amplitude |V p2 |With the central voltage V p0 The absolute value of the amplitude |V p0 |About the same, it is the maximum value.

[0248] in addition, Fig. 9B The third peripheral voltage V represented by the thin solid line p3 The transient response is that the first vibration amplitude peak is about 0.3μV after about 0.06μs after the pulse input. Compared with the first vibration amplitude peak, the second vibration amplitude peak is about +0.5μV after about 0.13μs, reaching the maximum amplitude value. Thereafter, the third vibration amplitude peak is about -0.4μV after about 0.19μs, and the fourth vibration amplitude peak decays to about +0.24μV after about 0.26μs. The vibration amplitude peaks after the fifth time become smaller, so the description is omitted.

[0249] and, Fig. 9B The fourth peripheral voltage V represented by a single-dot chain line p4 The transient response is that the first vibration amplitude peak is about 0.3μV after about 0.06μs. Compared with the first vibration amplitude peak, the second vibration amplitude peak is about +0.5μV after about 0.13μs, reaching the maximum amplitude value. The fourth peripheral voltage V p4 The absolute value of the amplitude |V p4 |With the third peripheral voltage V p3 The absolute value of the amplitude |V p3 | is roughly the same. However, the 4th peripheral voltage V p4 The absolute value of the amplitude |V p4 |Ratio to central voltage V p0 The absolute value of the amplitude |V p0 |, the second peripheral voltage V p2 The absolute value of the amplitude |V p2 | Small. In addition, the second peripheral voltage V p2 ~4th peripheral voltage V p4 vibrate in phase with each other, and they are related to the central voltage Vp0 Therefore, the absolute value of the potential generated on the output surface of the vibrator 22d is related to each other as follows:

[0250] |V p0 |≈|V p2 |>|V p3 |≈|V p4 |>|V p1 |……(13)

[0251] The potential distribution exhibits position dependence with a magnitude relationship.

[0252] -W shape-

[0253] Next, if Fig. 10A As shown in FIG. 1 , the voltage distribution on the output surface of the vibrator 22e having a W-shaped cross-sectional structure is studied at each position as a basis for high sensitivity and wide bandwidth. The vibrator 22e is arranged on a silicon substrate 11 having a cylindrical recess. se On the vibrating body 22e, a grounded fixed potential electrode 24e is arranged, and a fixed potential electrode protective film 23 made of resin is arranged on the fixed potential electrode 24e. Since the cross-sectional shape is a W-shape, the vibrating body 22e has a cone that tapers upward in the central part of the W. In addition, around the conical space in the central part of the W, there is a downward-pointing protrusion that surrounds the cone in a circular ring shape. In the cross-sectional structure, an inclined toroidal structure represented by two downward-pointing inverted triangles on both sides of the triangular space in the central part of the W is formed. In addition, the lower end of the downward-pointing protrusion at approximately half of the radial direction from the central part to the vibration cavity 21e of the W-shaped vibrating body 22e is close to the silicon substrate 11. se Therefore, on the silicon substrate 11 se In the concave portion, a part of the conical vibration cavity 21e is provided in the central part, and the other part of the vibration cavity 21e with a right triangle cross section is provided at the end. In addition, in order to verify the generated voltage distribution, a circular potential detection ideal electrode is arranged at the upper tip of the cone in the central part of the W-shaped vibrating body 22e, and a concentric ring potential detection ideal electrode is arranged around the central potential detection ideal electrode, and the potential transfer area is defined as the central area P and the potential transfer area P, respectively. 0 and the first peripheral area P 1 (See Table 2). In addition, a concentric annular potential detection ideal electrode is arranged at the lower end of the downwardly tapered bulge around the central portion of the W-shaped vibrating body 22e, which is defined as the second peripheral area P 2 .

[0254] In addition, in the area surrounding the second peripheral region P 2The second peripheral area P is arranged on the inclined surface of the vibrating body 22e exposed in the vibration cavity 21e. 2 The concentric ring-shaped potential detection ideal electrode is defined as the third peripheral area P 3 Then, on the lower surface of the vibrating body 22e at the outermost periphery of the vibration cavity 21e, a third peripheral region P is arranged. 3 The concentric ring-shaped potential detection ideal electrode is defined as the fourth peripheral area P 4 Then, by verifying the potential transfer area, that is, the central area P 0 , the first peripheral area P 1 , ..., the fourth peripheral area P 4 The transient response of the voltages generated in each of the two circuits is simulated to simulate the voltage distribution on the output surface (lower surface) of the vibrating body 22e.

[0255] Fig. 10B This is the simulation result when the entire vibrator 22e is made of PZT. 22e =4.5 μm, the thickness t of the fixed potential electrode protection film 23 on the vibrating body 22e 23 =6μm, the voltage transient response is generated, with the potential transfer area, i.e. the central area P 0 , the first peripheral area P 1 , ..., the fourth peripheral area P 4 The position of the parameter is displayed. Fig. 10B When a single-frequency sine wave with a maximum vibration amplitude of ±1 Pa shown on the left is input to the voltage generating section (22e, 24e), the potential transfer region, i.e., the central region P 0 , the first peripheral area P 1 , ..., the fourth peripheral area P 4 The transient responses of the impulse responses generated in the figure show that the amplitude of the second vibration reaches a maximum around 0.13 μs, and the third and subsequent vibrations are attenuated transient responses.

[0256] Specifically, Fig. 10B The second peripheral area P indicated by the dotted line 2 The second peripheral voltage V p2 The transient response is that the first vibration amplitude peak is about 0.6μV after about 0.06μs after the pulse input. Compared with the first vibration amplitude peak, the second vibration amplitude peak is about +0.98μV after about 0.13μs, reaching the maximum amplitude value. Thereafter, the third vibration amplitude peak is about -0.68μV after about 0.19μs, and the fourth vibration amplitude peak decays to about +0.45μV after about 0.26μs. The vibration amplitude peaks after the fifth time become smaller, so the description is omitted. The second peripheral voltage V p2The absolute value of the amplitude |V p2 |Ratio to central voltage V p0 , the first peripheral voltage V p1 , the third peripheral voltage V p3 and the 4th peripheral voltage V p4 The absolute value of the amplitude |V p0 |、|V p1 |、|V p3 | and |V p4 Any one of | is greater, which is the maximum value.

[0257] Fig. 10B The central area P indicated by the thick solid line 0 The central voltage V p0 The transient response is that the first vibration amplitude peak is about +0.28μV after about 0.06μs after the pulse input. Compared with the first vibration amplitude peak, the second vibration amplitude peak is about -5.0μV after about 0.13μs, reaching the maximum amplitude value. Thereafter, the third vibration amplitude peak is about +0.28μV after about 0.19μs, and the fourth vibration amplitude peak decays to about -0.22μV after about 0.26μs. The vibration amplitude peaks after the 5th time become smaller, so the description is omitted. Center voltage V p0 and the second peripheral voltage V p2 In anti-phase relationship, the absolute value of its amplitude |V p0 |Second only to the 2nd peripheral voltage V p2 The absolute value of the amplitude |V p2 |, is the second largest.

[0258] Fig. 10B The third peripheral voltage V represented by the thin solid line p3 The transient response is that the first vibration amplitude peak is about 0.26μV after about 0.06μs after the pulse input. Compared with the first vibration amplitude peak, the second vibration amplitude peak is about +0.4μV after about 0.13μs, reaching the maximum amplitude value. Thereafter, the third vibration amplitude peak is about -0.32μV after about 0.19μs, and the fourth vibration amplitude peak decays to about +0.18μV after about 0.26μs. The vibration amplitude peaks after the 5th time become smaller, so the description is omitted. The third peripheral voltage V p3 is the third largest value.

[0259] Fig. 10B The fourth peripheral voltage V represented by a single-dot chain line p4 The absolute value of the amplitude |V p4 | Overall, the third peripheral voltage V p3 is slightly smaller and is the fourth largest value. Fig. 10BThe first peripheral voltage V p1 The absolute value of the amplitude |V p1 | Overall, the 4th peripheral voltage V p4 The absolute value of the amplitude |V p4 | is slightly smaller and is the minimum value. In addition, in the potential transfer region, that is, the first peripheral region P 1 ~4th surrounding area P 4 The first peripheral voltage V p1 ~4th peripheral voltage V p4 vibrate in phase with each other, and they are in phase with the central region P 0 The central voltage V p0 Therefore, in the case of a W-shaped stress distribution reinforcement structure, the absolute value of the potential generated on the output surface of the vibrating body 22e is related to each other as follows:

[0260] |V p2 |>|V p0 |>|V p3 |>|V p4 |>|V p1 |……(14)

[0261] The potential distribution exhibits position dependence with a magnitude relationship.

[0262] - Flat disc type conventional structure (traditional structure) -

[0263] Finally, for comparison, Fig.11A As shown, in the case where the cross section of the vibration cavity formed by the vibration body 22f is a flat disk-type conventional structure with parallel plates (stress distribution non-reinforced structure), the voltage distribution generated on the output surface of the vibration body 22f is simulated at each position. Fig. 11B The simulation results when PZT is used as the vibrator 22f are shown. Fig. 11C and Fig.11D shows the simulation results when using AlN. A flat disk-shaped conventional structure, such as Fig.11A As shown in FIG. 1 , the voltage generating portion (22f, 24f) is composed of a vibrating body 22f forming a uniform flat surface, a grounded fixed potential electrode 24f above the vibrating body 22f, and a fixed potential electrode protective film 23 made of epoxy resin above the fixed potential electrode 24f. Furthermore, the voltage generating portion (22f, 24f) is arranged on a silicon substrate 11 having a concave portion. sf The receiving surface and the output surface of the vibrating body 22f facing each other are on the silicon substrate 11 sf The concave portion of the silicon substrate 11 forms a Euclidean plane where the Gaussian curvature and the mean curvature are always zero. sfA thin cylindrical (disc-shaped) vibration cavity 21f is formed on the concave portion, and the cross-sectional structure of the vibration cavity 21f is a parallel plate structure with a certain thickness and a receiving surface and an output surface parallel to each other.

[0264] In the stress distribution non-strengthened structure, in order to verify the generated voltage distribution, such as Fig.11A As shown, a disk-shaped potential detection ideal electrode is arranged at the central part of the lower surface of the vibrating body 22f, and four concentric ring-shaped potential detection ideal electrodes are arranged around the central potential detection ideal electrode. In the stress distribution non-strengthening structure, the same as defined in Table 2, for one disk and four concentric circles, numbered P in sequence from the center 0 , P 1 ,……,P 4 , define the potential transfer area as the central area P 0 , the first peripheral area P 1 , ..., the fourth peripheral area P 4 .

[0265] Fig. 11B This is the simulation result of the stress distribution of the non-reinforced structure using PZT as the entire vibrator 22f. 22f =5 μm, the thickness of the fixed potential electrode protection film 23 is t 23 =6μm, the voltage transient response is generated, with the potential transfer area, i.e. the central area P 0 , the first peripheral area P 1 , ..., the fourth peripheral area P 4 The position of the parameter is displayed. Fig. 11B When the single-frequency sinusoidal ultrasonic wave vibrating at a maximum of ±1 Pa shown on the left is input to the voltage generating section (22f, 24f), the potential transfer area, i.e., the central area P 0 , the first peripheral area P 1 , ..., the fourth peripheral area P 4 The voltage signals generated in the experiment show that the amplitude of the second vibration reaches the maximum at about 0.13 μs, and the third and subsequent vibrations are transient responses of attenuated vibrations.

[0266] Fig. 11B The first peripheral voltage V p1, the first vibration amplitude reaching a peak value of about +0.12μV about 0.06μs after the ultrasonic pulse input. After the first vibration amplitude peak, the second vibration amplitude peak value about 0.13μs later is about -0.22μV, reaching the maximum amplitude value. Thereafter, the third vibration amplitude peak value about 0.19μs later is about +0.15μV, and the fourth vibration amplitude peak value about 0.26μs later decays to about -0.12μV. The vibration amplitude becomes smaller after the fifth time, so the description is omitted. The first peripheral voltage V p1 Other than the central voltage V p0 , the second peripheral voltage V p2 ~4th peripheral voltage V p4 They are all big, which is the maximum value.

[0267] Fig. 11B The second peripheral voltage V p2 , the first vibration amplitude, which reaches a peak value about 0.06μs after the ultrasonic pulse input, is about +0.1μV. After the first vibration amplitude peak, the second vibration amplitude peak is about -0.16μV about 0.13μs after the first vibration amplitude peak, reaching the maximum amplitude value. Thereafter, the third vibration amplitude peak is about +0.11μV about 0.19μs after, and the fourth vibration amplitude peak decays to about -0.08μV about 0.26μs after. The vibration amplitude becomes smaller after the fifth time, so the description is omitted. The second peripheral voltage V p2 and the first peripheral voltage V p1 In phase relationship, the second peripheral voltage V p2 The absolute value of the amplitude |V p2 |At the 1st peripheral voltage V p1 The absolute value of the amplitude |V p1 |Then the second largest.

[0268] Fig. 11B The third peripheral voltage V represented by the thin solid line p3 , the first vibration amplitude reaching a peak value of about +0.06μV about 0.06μs after the ultrasonic pulse input. After the first vibration amplitude peak, the second vibration amplitude peak value about 0.13μs later is about -0.1μV, reaching the maximum amplitude value. Thereafter, the third vibration amplitude peak value about 0.19μs later is about +0.06μV, and the fourth vibration amplitude peak value about 0.26μs later decays to about -0.05μV. The vibration amplitude becomes smaller after the fifth time, so the description is omitted. The third peripheral voltage V p3 is the third largest value.

[0269] Fig. 11B The fourth peripheral voltage V represented by a single-dot chain line p4 The absolute value of the amplitude |V p4 | Overall, the first peripheral voltage Vp1 ~3rd peripheral voltage V p3 The absolute value of the amplitude |V p1 |~|V p3 | is the fourth largest value. Fig. 11B The central voltage V p0 The absolute value of the amplitude |V p0 | Overall, the 4th peripheral voltage V p4 The absolute value of the amplitude |V p4 | is slightly smaller and is the smallest. In addition, these central voltages V p0 ~4th peripheral voltage V p4 Therefore, in the case of a conventional flat disk structure using PZT as the vibrator 22f and not actively strengthening the stress distribution, the absolute value of the potential generated on the output surface of the vibrator 22f is:

[0270] |V p1 |>|V p2 |>|V p3 |>|V p4 |>|V p0 |……(15)

[0271] It can be seen from this that even if Fig.11A The conventional structure shown does not actively enhance the stress distribution and also has a position-dependent potential distribution.

[0272] Fig. 11C This is a simulation result of a conventional flat disk structure using AlN as the entire vibrator 22f. 22f =0.2 μm, the thickness t of the fixed potential electrode protection film 23 23 =3μm, the voltage transient response is generated, with the potential transfer area, i.e. the central area P 0 , the first peripheral area P 1 , ..., the fourth peripheral area P 4 The position is displayed as a parameter. Fig. 11C When the ultrasonic wave vibrating at a maximum of ±1 Pa shown on the left is input to the voltage generating section (22f, 24f), the central area P 0 , the first peripheral area P 1 , ..., the fourth peripheral area P 4 The voltage signals generated in Fig. 11B The situation is the same as that of , showing a transient response in which the amplitude of the second vibration reaches a maximum at around 0.13μs, and the third and subsequent vibrations are attenuated vibrations.

[0273] Fig. 11C The central voltage V p0The first vibration amplitude, which reaches a peak value about 0.06μs after the ultrasonic pulse input, is about -0.57μV. After the first vibration amplitude peak, the second vibration amplitude peak is about +0.9μV about 0.13μs after the first vibration amplitude peak, reaching the maximum amplitude value. Thereafter, the third vibration amplitude peak is about -0.48μV about 0.19μs after the third vibration amplitude peak, and the fourth vibration amplitude peak decays to about +0.28μV about 0.26μs after the fourth vibration amplitude peak. The vibration amplitude becomes smaller after the fifth time, so the description is omitted. Central voltage V p0 The absolute value of the amplitude |V p0 | than the first peripheral voltage V p1 ~4th peripheral voltage V p4 The absolute value of the amplitude |V p1 |~|V p3 |All are large, which is the maximum value.

[0274] Fig. 11C The first peripheral voltage V p1 , the first vibration amplitude, which reaches a peak value about 0.06μs after the ultrasonic pulse input, is about -0.42μV. After the first vibration amplitude peak, the second vibration amplitude peak is about +0.7μV about 0.13μs after the first vibration amplitude peak, reaching the maximum amplitude value. Thereafter, the third vibration amplitude peak is about -0.4μV about 0.19μs after the third vibration amplitude peak, and the fourth vibration amplitude peak decays to about +0.22μV about 0.26μs after the fourth vibration amplitude peak. The vibration amplitude becomes smaller after the fifth time, so the description is omitted. The first peripheral voltage V p1 With the central voltage V p0 In phase relationship, the first peripheral voltage V p1 The absolute value of the amplitude |V p1 |At the center voltage V p0 The absolute value of the amplitude |V p0 |Then the second largest.

[0275] Fig. 11C The fourth peripheral voltage V represented by a single-dot chain line p4 , the first vibration amplitude reaching a peak value of about +0.32μV about 0.06μs after the ultrasonic pulse input. After the first vibration amplitude peak, the second vibration amplitude peak value about 0.13μs later is about -0.46μV, reaching the maximum amplitude value. Thereafter, the third vibration amplitude peak value about 0.19μs later is about +0.25μV, and the fourth vibration amplitude peak value about 0.26μs later decays to about -0.16μV. The vibration amplitude becomes smaller after the 5th time, so the description is omitted. The 4th peripheral voltage V p4 With the central voltage V p0 , the first peripheral voltage V p1 In anti-phase relationship, the absolute value of the amplitude |V p4| is the third largest value.

[0276] Fig. 11C The first peripheral voltage V p2 The first vibration amplitude, which reaches a peak value about 0.06μs after the ultrasonic pulse input, is about -0.2μV. After the first vibration amplitude peak, the second vibration amplitude peak is about +0.34μV about 0.13μs after the first vibration amplitude peak, reaching the maximum amplitude value. Thereafter, the third vibration amplitude peak is about -0.18μV about 0.19μs after the third vibration amplitude peak, and the fourth vibration amplitude peak decays to about +0.12μV about 0.26μs after the fourth vibration amplitude peak. The vibration amplitude becomes smaller after the fifth time, so the description is omitted. The second peripheral voltage V p2 With the central voltage V p0 , the first peripheral voltage V p1 In phase with the 4th peripheral voltage V p4 In anti-phase relationship, the absolute value of the amplitude |V p2 | is the fourth largest value.

[0277] Fig. 11C The third peripheral voltage V represented by the thin solid line p3 The absolute value of the amplitude |V p3 |With the 4th peripheral voltage V p4 In phase relationship, but overall higher than the 4th peripheral voltage V p4 The absolute value of the amplitude |V p4 | is small, which is the smallest. Therefore, in the case of a flat disk-type conventional structure using AlN as the vibrating body 22f and not actively strengthening the stress distribution, the relationship between the absolute value of the potential generated on the output surface of the vibrating body 22f is:

[0278] |V p0 |>|V p1 |>|V p4 |>|V p2 |>|V p3 |……(16)

[0279] It can be seen from this that even for conventional structures that do not actively enhance stress distribution, there is a position-dependent potential distribution.

[0280] Fig.11D and Fig. 11C Similarly, this is the simulation result of a conventional flat disk structure in which AlN is used throughout the vibrating body 22f. 22f =0.2μm and Fig. 11C The same as the case of the fixed potential electrode protection film 23 thickness t 23 =6μm, Fig. 11C The case is thick when the voltage transient response is generated to the central area P 0, the first peripheral area P 1 , ..., the fourth peripheral area P 4 The position is displayed as a parameter. Fig.11D When the ultrasonic wave vibrating at a maximum of ±1 Pa shown on the left is input to the voltage generating section (22f, 24f), the central area P 0 , the first peripheral area P 1 , ..., the fourth peripheral area P 4 The voltage signals generated in Fig. 11C The situation is the same as that of , showing a transient response in which the amplitude of the second vibration reaches a maximum at around 0.13μs, and the third and subsequent vibrations are attenuated vibrations.

[0281] Fig.11D The central voltage V p0 The first vibration amplitude, which reaches a peak value about 0.06μs after the ultrasonic pulse input, is about -0.29μV. After the first vibration amplitude peak, the second vibration amplitude peak is about +0.34μV about 0.13μs after the first vibration amplitude peak, reaching the maximum amplitude value. Thereafter, the third vibration amplitude peak is about -0.22μV about 0.19μs after the third vibration amplitude peak, and the fourth vibration amplitude peak decays to about +0.12μV about 0.26μs after the fourth vibration amplitude peak. The vibration amplitude becomes smaller after the fifth time, so the description is omitted. Central voltage V p0 The absolute value of the amplitude |V p0 | than the first peripheral voltage V p1 ~4th peripheral voltage V p4 The absolute value of the amplitude |V p1 |~|V p4 |All are large, which is the maximum value.

[0282] Fig.11D The first peripheral voltage V p1 , the first vibration amplitude reaching a peak value about 0.06μs after the ultrasonic pulse input is about -0.24μV. After the first vibration amplitude peak value, the second vibration amplitude peak value about 0.13μs after is about +0.26μV, reaching the maximum amplitude value. Thereafter, the third vibration amplitude peak value about 0.19μs after is about -0.18μV, and the fourth vibration amplitude peak value about 0.26μs after decays to about +0.08μV. The vibration amplitude becomes smaller after the fifth time, so the description is omitted. The first peripheral voltage V p1 With the central voltage V p0 In phase relationship, the first peripheral voltage V p1 The absolute value of the amplitude |V p1 |At the center voltage V p0 The absolute value of the amplitude |V p0 |Then the second largest.

[0283] Fig.11D The fourth peripheral voltage V represented by a single-dot chain line p4 , the first vibration amplitude reaching a peak value of about +0.17μV about 0.06μs after the ultrasonic pulse input. After the first vibration amplitude peak, the second vibration amplitude peak value about 0.13μs later is about -0.21μV, reaching the maximum amplitude value. Thereafter, the third vibration amplitude peak value about 0.19μs later is about +0.12μV, and the fourth vibration amplitude peak value about 0.26μs later decays to about -0.06μV. The vibration amplitude becomes smaller after the 5th time, so the description is omitted. The 4th peripheral voltage V p4 With the central voltage V p0 , the first peripheral voltage V p1 In anti-phase relationship, the absolute value of the amplitude |V p4 | is the third largest value.

[0284] Fig.11D The second peripheral voltage V p2 The first vibration amplitude, which reaches a peak value about 0.06μs after the ultrasonic pulse input, is about -0.12μV. After the first vibration amplitude peak, the second vibration amplitude peak is about +0.15μV about 0.13μs after the first vibration amplitude peak, reaching the maximum amplitude value. Thereafter, the third vibration amplitude peak is about -0.09μV about 0.19μs after the third vibration amplitude peak, and the fourth vibration amplitude peak decays to about +0.05μV about 0.26μs after the fourth vibration amplitude peak. The vibration amplitude becomes smaller after the fifth time, so the description is omitted. The second peripheral voltage V p2 With the central voltage V p0 , the first peripheral voltage V p1 In phase with the 4th peripheral voltage V p4 In anti-phase relationship, the absolute value of the amplitude |V p2 | is the fourth largest value.

[0285] exist Fig.11D The third peripheral voltage V represented by the thin solid line p3 The absolute value of the amplitude |V p3 |With the fourth peripheral voltage V p4 In phase, but overall higher than the fourth peripheral voltage V p4 The absolute value of V p4 | is small, which is the smallest. Therefore, in the case of using AlN for a flat disk-shaped general structure of the vibrating body 22f, even if the thickness of the fixed potential electrode protection film 23 is t 23 Increase, the absolute value of the potential generated on the output surface of the vibrator 22f is related to Fig. 11C The same is true for , as shown in equation (16). Therefore, even if Fig.11AEven in a conventional structure that is not actively intended to enhance stress distribution as shown, it can be seen that there is a position-dependent potential distribution.

[0286] - Comparison of characteristics of stress distribution reinforcement structures -

[0287] Focusing on the case of using PZT as the piezoelectric layer, Figure 5A and 5B The parallel annular stress distribution strengthening structure (T) shown in Fig. 7A The V-shaped stress distribution strengthening structure (V 1 , V 2 ), Fig. 8A The Λ-type stress distribution strengthening structure (Λ) shown in Fig. 9A The M-shaped stress distribution strengthening structure (M) shown in the figure Fig. 10A The voltage sensitivity and frequency bandwidth of each of the W-shaped stress distribution reinforcement structure (W) shown are compared with the characteristics of the flat disk-shaped ordinary structure (P) as the basis for achieving high sensitivity and wide bandwidth.

[0288] like Fig.12 As shown, in the frequency bandwidth of receiving ultrasonic waves within the range of 5 to 10 MHz, V 1 The maximum voltage sensitivity shown by the sensitivity curve of the V-shaped stress distribution strengthening structure represented by the thick solid line is about 2.9μV / Pa, which is the highest. Next, the maximum voltage sensitivity shown by the sensitivity curve of the W-shaped stress distribution strengthening structure represented by the dotted line marked by W is about 2.2μV / Pa, which is the second highest. The maximum voltage sensitivity shown by the sensitivity curve of the Λ-shaped stress distribution strengthening structure represented by the thin solid line marked by Λ is about 1.9μV / Pa, which is the third highest, and the maximum voltage sensitivity shown by the sensitivity curve of the M-shaped stress distribution strengthening structure represented by the thin solid line marked by M is about 1.4μV / Pa, which is the fourth highest. Compared with these stress strengthening structures, the maximum receiving sensitivity of the traditional structure (flat disk-type ordinary structure) represented by the single-point dashed line marked by P is only about 0.4μV / Pa, which is the lowest.

[0289] like Fig.13 As shown in the figure, the maximum voltage sensitivity of the parallel toroidal stress distribution strengthening structure represented by the solid line marked with T and the maximum voltage sensitivity represented by V 2The maximum voltage sensitivity of the V-shaped (inclined toroidal) stress distribution strengthening structure represented by the dotted line is slightly more than 1.2μV / Pa, which is the highest. In contrast, the maximum voltage sensitivity of the conventional structure (flat disk-type ordinary structure) represented by the single-dot chain line marked with P is only slightly more than 0.4μV / Pa, which is the lowest. The frequency characteristics of the sensitivity of the parallel toroidal stress distribution strengthening structure show a wide band exceeding 0.6μV / Pa in the frequency bandwidth range of 5 to 15MHz. Although the frequency characteristics of the sensitivity of the V-shaped stress distribution strengthening structure are slightly narrower than those of the parallel toroidal stress distribution strengthening structure, they also show a frequency band similar to that of the parallel toroidal stress distribution strengthening structure. On the other hand, it can be seen that the frequency characteristics of the sensitivity of the conventional structure (flat disk-type ordinary structure) represented by the single-dot chain line are very narrow compared with the parallel toroidal and V-shaped stress distribution strengthening structures.

[0290] - First Modification of the First Embodiment -

[0291] exist Figure 2 In the example of a single heterojunction FET structure, a hexagonal potential transfer region defined as a convex portion on the output surface side of the vibrator 22a1 is formed in a channel on the surface of a rectangular region sandwiched between the first main electrode region 15a and the second main electrode region 15b in a parallel toroidal stress distribution reinforcement structure. However, as Fig.14 As shown in FIG. 1 , in the acoustic element according to the first modification of the first embodiment, the buffer layer 13h is sandwiched between the vibrating body 22a1 and the base region 14 to form a double heterojunction. As described above, since AlN and GaN have good compatibility as heterojunctions, even if AlN is heteroepitaxially grown on the base region 14 composed of a GaN substrate as the vibrating body 22a1, the interface energy level and lattice defects between GaN and AlN are relatively small. However, if GaN is inserted between GaN and AlN, the double heterojunction is formed. 1-x Al x The N layer as a buffer layer 13h can further reduce the interface energy level and lattice defects between GaN and AlN. The Ga with a longer bond length is epitaxially grown on the Ga surface of GaN with a c-axis orientation. 1-x Al x N, due to the difference in lattice constants between the two, Ga 1-x Al x Tensile strain is generated in N, and Ga 1- x Al x N has piezoelectric polarization inside. GaN, as a polar crystal, also has piezoelectric polarization. 1-x Al xAt the N / GaN heterojunction interface, due to the difference in work function, a potential distribution will be generated in the direction perpendicular to the interface, just like ordinary heterojunction FETs. Therefore, the end band structures of the conduction band and valence band will bend as they approach the interface.

[0292] If GaN is set as i-type, Ga 1-x Al x N is set as n-type, then in Ga 1-x Al x Two-dimensional electron gas is generated at the N / GaN heterojunction interface. In addition, the two piezoelectric polarization effects can be simultaneously present. Unlike ordinary heterojunction FETs, the acoustic element according to the first variant of the first embodiment does not have a gate electrode. Although there is no gate electrode, Ga 1-x Al x N internal, especially Ga 1-x Al x The Fermi level of the N / GaN heterojunction interface will vibrate due to ultrasonic input. Since there is no gate electrode in the acoustic element according to the first variant of the first embodiment, there is no need to connect the PMUT and FET through external wiring as described in Non-Patent Document 2, so there is no floating impedance problem. In addition, since the use of Figures 4A to 6C The voltage generating portion (22a1, 24a) of the stress strengthening structure shown in the figure achieves high sensitivity and broadband. The voltage generating portion (22a1, 24a) itself achieves high sensitivity and broadband. The acoustic element according to the first modified example of the first embodiment can achieve high sensitivity and broadband. In particular, by microscopically studying the internal structure of the stress strengthening structure, a specific position on the output surface of the vibrating body 22a1 where the highest voltage is generated is selected as a potential transfer area, and the active element of the impedance conversion element (14, 15a, 15b) is driven, so that high sensitivity can be easily achieved.

[0293] On the other hand, when GaN or AlN is heteroepitaxially grown as the vibrating body 22a1 on the base region 14 composed of the Si substrate having a diamond-type crystal structure to constitute the acoustic element according to the first modified example of the first embodiment, since the lattice constant of Si is 0.543 nm, the lattice mismatch is large. When GaN or AlN is heteroepitaxially grown as the vibrating body 22a1 using the base region 14 composed of the Si substrate, as shown in FIG. Fig.14 As shown, it is necessary to set up 1-x Al x N and other buffer layers 13h to form a double heterojunction structure. When a polar crystal layer such as GaN or AlN is epitaxially grown on a non-polar substrate such as Si, a phase reversal of the atomic arrangement may occur. This is because the atomic arrangement of the polar crystal is different depending on whether the step of the substrate is composed of 1 atom or 2 atoms, which is called an "anti-phase domain". In order to prevent the anti-phase domain, it is preferred to use an off-angle substrate.

[0294] When SiC, which is a polar crystal, is heteroepitaxially grown on the base region 14 formed of the Si substrate as the vibration body 22a1 to form the acoustic element according to the first modification of the first embodiment, C 2 H 4 The surface of the Si substrate is carbonized by hydrocarbon gases such as , and SiC is grown using it as a buffer layer 13h. When the S1 surface is carbonized, β-SiC (cubic crystal) grows epitaxially. Alternatively, sphalerite-type crystal boron phosphide (BP) can be formed on the base region 14 composed of the Si substrate as a buffer layer 13h, and SiC is heteroepitaxially grown through the buffer layer 13h to form a double heterojunction. Fig.14 The other structures other than the buffer layer 13h shown in FIG. Figure 2 The contents described in are repeated and therefore omitted.

[0295] - Second Modification of the First Embodiment -

[0296] and Figure 2 The structure shown is similar to Fig.15 The acoustic element according to the second variant of the first embodiment shown in FIG. 1 also adopts a parallel annular stress distribution reinforcement structure. Fig.15 As shown, the first main electrode wiring line 25a is connected to the first main electrode region 15a, and the second main electrode wiring line 25b is connected to the second main electrode region 15b through the contact holes opened in the interlayer insulating film 17 provided on the first main electrode region 15a and the second main electrode region 15b. Fig.15 In the embodiment, the first main electrode wiring line 25a is led to the left through the element separation insulating film 16, and the second main electrode wiring line 25b is led to the right through the element separation insulating film 16. In the acoustic element according to the second modified example of the first embodiment, the cavity inner wall protection film 19 covers the structure above the first main electrode wiring line 25a and the second main electrode wiring line 25b. Figure 2 The convex portion of the output surface side of the vibrator 22a1 reaches the upper surface of the base region 14 through the cavity inner wall protection film 19, and the potential transfer region of the vibrator 22a1 and the upper surface of the base region 14 form a heterojunction. Fig.15 Other structures other than the cavity inner wall protective film 19 shown in Figure 2 The contents described in are repeated and therefore omitted.

[0297] Since the acoustic element according to the second modification of the first embodiment does not have a gate electrode, the floating impedance problem described in Non-Patent Document 2 does not exist. Figures 4A to 6CThe voltage generating portion (22a1, 24a) of the stress strengthening structure shown in the figure achieves high sensitivity and wide bandwidth. As a result, the acoustic element according to the second modified example of the first embodiment can achieve high sensitivity and wide bandwidth. In particular, by microscopically studying the internal structure of the stress strengthening structure, a specific position on the output surface of the vibrating body 22a1 where the highest voltage is generated is selected as a potential transfer area, and the active element of the impedance conversion element (14, 15a, 15b) is driven, so that high sensitivity can be easily achieved.

[0298] =Method for manufacturing acoustic element according to second variation of first embodiment=

[0299] use Fig.16A ~Figure 16R Description Fig.15 The method for manufacturing an acoustic element according to the second variant of the first embodiment is exemplified in the following. It should be noted that the method for manufacturing an acoustic element described below is only an example, and as long as the structure described in the claims can be realized, it can of course be realized by various other manufacturing methods. In addition, the names with ordinal numbers such as "first photoresist film" used in the following description are only used for the rhetorical requirement to distinguish it from other photoresist films, and do not mean the first photoresist film used in the actual process. In fact, in Fig.16A The stage before the structure shown has generally been subjected to processing using a photoresist film.

[0300] (a) First, a base region 14 is prepared, which is composed of a p-type silicon substrate with a (100) plane as the main surface and a resistivity of 0.1 to 10 Ωcm. Then, a device isolation insulating film 16 such as a silicon oxide film is formed to a thickness of about 0.5 to 1.2 μm using the known LOCOS or STI technology. Inside the pattern of the active region surrounded by the device isolation insulating film 16, a device isolation insulating film 16 is formed as follows: Fig.16A The n shown + The first main electrode region 15a and the second main electrode region 15b are sectional structures opposite to each other. Examples of the planar patterns of the first main electrode region 15a and the second main electrode region 15b may be, for example, Figure 2 Afterwards, ion implantation for MOSFET threshold control can be added as needed.

[0301] (b) Then, on the first main electrode region 15a, the second main electrode region 15b, and the base region 14 exposed between the first main electrode region 15a and the second main electrode region 15b, as shown in FIG. Fig. 16BAs shown in FIG. 1 , a first insulating film 17p such as a silicon oxide film having a thickness of about 10 to 250 nm is deposited by a deposition method such as CVD. When the substrate region 14 is Si, thermal oxidation can also be used. After the first insulating film 17p is deposited by a method such as CVD, it is flattened by a method such as chemical mechanical polishing (CMP) until the first insulating film 17p is flattened. Fig. 16B The upper surface of the element isolation insulating film 16 is exposed. Thereafter, a first photoresist film is coated on the first insulating film 17p, and the first photoresist film is exposed and developed using a photolithography technique to form a first etching mask for opening a contact hole. Using the first etching mask, a dry etching technique such as reactive ion etching (RIE) is used, such as Fig.16F As shown, the first insulating film 17p is selectively etched to form contact hole patterns on the upper portions of the first main electrode region 15a and the second main electrode region 15b. Hereinafter, the first insulating film 17p having the contact holes formed therein is referred to as an "interlayer insulating film 17".

[0302] (c) Further, on the first insulating film 17p, a first conductor film composed of a conductor layer such as a doped polysilicon (DOPOS) film or a high melting point metal is deposited on the entire surface with a thickness of about 80 to 200 nm by a deposition method such as CVD. A second photoresist film is applied on the first conductor film, and the second photoresist film is exposed and developed using a photolithography technique to form a second etching mask for patterning the first main electrode wiring line 25a and the second main electrode wiring line 25b. Using the second etching mask, a dry etching technique such as RIE is used, such as Fig. 16C As shown, the first conductor film is selectively etched to form a pattern of a first main electrode wiring line 25a connected to the first main electrode region 15a and a pattern of a second main electrode wiring line 25b connected to the second main electrode region 15b. Fig. 16C In the figure, the interlayer insulating film 17 exposed between the first main electrode wiring line 25a and the second main electrode wiring line 25b is defined as a “channel protection insulating film 17c”.

[0303] (d) Then, a third photoresist film is coated on the channel protection insulating film 17c, the first main electrode wiring line 25a and the second main electrode wiring line 25b, and the third photoresist film is exposed and developed using a photolithography technique to form a third etching mask for exposing the upper portion of the channel. Using the third etching mask, the channel protection insulating film 17c is selectively etched by a dry etching technique such as RIE, such as Fig.16D As shown, a contact hole pattern is opened between the first main electrode region 15 a and the second main electrode region 15 b to expose the upper surface of the substrate region 14 .

[0304] (e) Then, on the first main electrode wiring line 25a and the second main electrode wiring line 25b and the base region 14 exposed between the first main electrode region 15a and the second main electrode region 15b, as shown in FIG. Fig.16E As shown, a cavity inner wall protective film 19 such as a silicon oxide film with a thickness of about 180 to 250 nm is deposited by a deposition method such as CVD. Then, a sacrificial layer material layer 27p such as tungsten (W) is deposited on the entire surface by a deposition method such as sputtering, vacuum evaporation or CVD. Then, planarization is performed by a CMP method or the like.

[0305] (f) Then, a fourth photoresist film is coated on the sacrificial layer material layer 27p, and the fourth photoresist film is exposed and developed using a photolithography technique to form a fourth etching mask for forming a sacrificial layer. Using the fourth etching mask, a dry etching technique such as RIE is used, such as Figure 16G As shown, the sacrificial layer material layer 27p is selectively etched to form a pattern of the sacrificial layer 27. Figure 16G In the cross-sectional view of FIG. , the patterns of the two sacrificial layers 27 are symmetrically located on both sides. However, from the appearance, in Figure 16G The two sacrificial layers 27 are symmetrically located on both sides of the Figure 16G The front side of the paper and the back side of the paper are continuous with each other in an annular structure.

[0306] (g) Further, a fifth photoresist film is coated on the sacrificial layer 27 forming the annular structure and the cavity inner wall protection film 19 exposed in the center of the sacrificial layer 27 of the annular structure, and the fifth photoresist film is exposed and developed using a photolithography technique to form a fifth etching mask for exposing the upper part of the channel. Using the fifth etching mask, the cavity inner wall protection film 19c is selectively etched by a dry etching technique such as RIE to expose the upper surface of the substrate region 14 in the center of the sacrificial layer 27 of the annular structure.

[0307] (h) A contact hole pattern on the upper surface of the substrate region 14 exposed in the center of the sacrificial layer 27 of the ring structure, such as Fig.16H As shown, the vibrator 22a1 is heteroepitaxially grown on the upper surface of the base region 14. For example, when the base region 14 is GaN, an AlN layer can be grown as the vibrator 22a1 by a light-excited molecular layer epitaxy (MLE) method. The MLE of the AlN layer is, for example, in a vacuum, trimethylaluminum (TMAI) as a group III element gas and ammonia (NH4) as a group V element gas. 3) and other raw material gases are introduced alternately in a time-division manner in beam form to grow AlN layers in units of one molecular layer. Photoexcitation MLE is an epitaxial growth method that further irradiates the surface of the growth layer with ultraviolet light such as an excimer laser, and uses ultraviolet energy to assist the growth, based on this MLE method. Photoexcitation MLE utilizes the surface migration caused by light energy to heteroepitaxially grow AlN layers in units of one molecular layer at low temperatures below 1000°C. The AlN layer can be grown at about 350 to 1200°C, but by utilizing ultraviolet energy, good crystallinity can be obtained even at low temperatures below 1000°C. Alternatively, when the substrate region 14 is Si, HfO can be grown by CVD at about 400°C using tetraethylmethylaminohafnium (TEMAH) or the like. 2 The layer serves as the vibrating body 22a1. The surface of the vibrating body 22a1 may be planarized by a method such as CMP as needed.

[0308] (i) Then, on the vibrating body 22a1, a second conductor film of aluminum (Al) or aluminum alloy is deposited on the entire surface by a deposition method such as sputtering, vacuum evaporation, or CVD. Further, a sixth photoresist film is applied on the second conductor film, and a pattern for forming a fixed potential electrode is formed by photolithography. The sixth photoresist film is used as a sixth etching mask to selectively etch the second conductor film to form a fixed potential electrode. Fig.16I The pattern of the fixed potential electrode 24 is shown.

[0309] (j) Next, a fixed electrode protection base film (not shown) such as epoxy resin is deposited on the vibrating body 22a1 by spin coating or the like to cover the pattern of the fixed potential electrode 24. Further, a seventh photoresist film is coated on the fixed electrode protection base film, and a pattern for introducing a removal liquid is formed by photolithography. The seventh photoresist film is used as a seventh etching mask, as shown in FIG. Fig.16J As shown, a liquid introduction hole 28 is formed which penetrates the fixed electrode protection underlying film and the vibrating body 22a1 and reaches the sacrificial layer 27. Fig.16J The cross-sectional shape after the seventh photoresist film serving as the seventh etching mask is removed is shown, but as described above, illustration of the fixed electrode protection underlying film is omitted.

[0310] (k) Then, for example, heated hydrogen peroxide solution (H 2 O 2 ) is introduced into the sacrificial layer 27 from the liquid introduction hole 28. When the heated hydrogen peroxide water is introduced from the liquid introduction hole 28, the sacrificial layer 27 is selectively dissolved by wet etching. When the sacrificial layer 27 is dissolved, Figure 16KAs shown, a parallel toroidal vibration cavity 21a1 is formed on the cavity inner wall protection film 19. Thereafter, a fixed potential electrode protection film 23 such as epoxy resin with a thickness of about 400 to 600 nm is deposited on the fixed electrode protection base film by spin coating or the like. Figure 16L As shown, the liquid introduction hole 28 is closed. In this process of closing the liquid introduction hole 28, by treating with an inert gas such as helium (He) gas as a main component in a reduced pressure atmosphere of about 1 kPa, the interior of the vibration cavity 21a1 becomes a reduced pressure state that can be regarded as almost a vacuum, wherein the inert gas is a main component, and the acoustic element according to the second variant of the first embodiment is completed.

[0311] Second Embodiment In the acoustic element according to the first embodiment, attention is paid to the fact that the vibrating body 22a1 functions as a gate insulating layer. Figure 3B In the acoustic element of the second embodiment of the present invention, attention is paid to the structure modeled by the equivalent circuit of Figure 3A The equivalent circuit structure shown in Figure 3A yes Figure 3B That is, Fig.17 , Fig.18A and Fig.18B As shown, the acoustic element according to the second embodiment is a receiving element X i,j , which includes a voltage generating unit 1a that generates a non-uniform voltage by piezoelectric effect, and an impedance conversion element 2b that operates the potential of a specific position (locality) of the voltage generating unit 1a as a control voltage. Fig.18B As can be seen from the cross-sectional view, the voltage generating unit 1a has a monopole structure, including a vibrating body 22a1 similar to the acoustic element of the first embodiment and a receiving surface ( Fig.18B The fixed potential electrode 24a is in contact with the upper main surface of the vibrating body 22a1. A fixed potential electrode protective film 23 such as epoxy resin is provided on the fixed potential electrode 24a. Fig.18B As shown, the entirety is composed of a piezoelectric layer, or only the central downward convex portion (stress concentration portion) is composed of a piezoelectric layer, and the remaining portion is a non-piezoelectric layer (not shown). That is, the voltage generating portion 1a only needs to be a portion of the vibrating body 22a1, that is, a structure in which at least a piezoelectric layer is provided at the stress concentration portion of the vibrating body 22a1. Therefore, the impedance conversion element 2b can use at least a specific part of the piezoelectric layer provided at the downward convex portion (stress concentration portion) as a potential transfer region, and operate the potential of the potential transfer region as a control voltage. As the material to be used for the central downward convex portion of the vibrating body 22a1 as the stress concentration portion, as mentioned above, materials belonging to 20 crystal point groups with piezoelectricity among the 32 crystal families can be used. Since it only needs to be used at least in the stress concentration portion, it can of course also be used as Fig.18BAs shown in FIG. 1 , the entire vibrating body 22a1 is made of a material having piezoelectricity. Among them, materials such as HfO2, AlN, and PZT, which belong to the 10 crystal point groups showing pyroelectricity, are suitable. The voltage generating portion 1a of the unipolar structure is on the output surface ( Fig.18B A specific position of the lower main surface of the vibrator 22a1 has a voltage distribution that generates a higher voltage than other positions by the piezoelectric effect. The "specific position" of the output surface of the vibrator 22a1 is, for example, Fig.17 The central region P is schematically shown in the potential transfer region model. 0 , the first peripheral area P 1 , ..., the fourth peripheral area P 4 . Fig.17 The central region P in the model representation 0 , the first peripheral area P 1 , ..., the fourth peripheral area P 4 For example, the central area P 0 As the center, to surround the central area P 0 The pattern of the first peripheral area P 1 , the second peripheral area P 2 , the third peripheral area P 3 , the fourth peripheral area P 4 It is a pattern surrounded by concentric hexagonal rings. It should be noted that Fig.17 In the model representation of the voltage generating unit 1a, the P 0 , P 1 ,……,P 4 It only indicates the location of the potential transfer area, and does not mean that in reality P 0 , P 1 ,……,P 4 There are electrodes at the rear. Fig. 20 , Fig. 22 , Fig.24 , Fig.25 , Fig.28D The same is true for etc.

[0312] like Fig.17 As shown, the impedance conversion element 2b includes an active element Q as a circuit element. b , which is an insulated gate transistor (MIS transistor), in which the first main electrode region 15a is connected to the first power supply (GND or negative power supply Vss), and the second main electrode region 15b is connected to the first power supply (GND or negative power supply Vss) through the output resistor R d Connected to a second power supply V with a higher potential than the first power supply DD As an active element Q b, which can correspond to MIS transistors such as MOSFET, MOSSIT, MISFET, MISIT, HEMT, etc. The active element Q constituting the impedance conversion element 2b b As Fig.18B The first peripheral region P shown in the center portion is a convex portion of the output surface of the vibrating body 22a1. 0 That is, the impedance conversion element 2b of the acoustic element according to the second embodiment includes an active element Q b , which uses the pressure of the ultrasonic wave Φ applied to the vibrating body 22a1 of the voltage generating unit 1a as the first peripheral area P 0 Then, the acoustic element according to the second embodiment converts the active element Q into a potential input and outputs it as a signal after impedance conversion. b The output signal of the acoustic element is used as the output signal of the acoustic element.

[0313] The output surface side convex portion of the vibrating body 22a1 of the voltage generating portion 1a of the acoustic element according to the second embodiment is as follows Fig.18A The double-dotted line shows the regular hexagonal prism convex part. Fig.18B As shown, around the convex portion provided at the center of the output surface side of the vibrating body 22a1, the parallel toroidal vibration cavity 21a1 is provided around the convex portion in a centrally symmetrical manner. The parallel toroids constituting the vibration cavity 21a1 have a topological structure whose plane pattern is a hexagon. Fig.18B In the cross-sectional view, from the appearance, the two vibration cavities 21a1 are located on both sides of the central convex portion, but they are continuous with each other on the front and back sides of the paper and form an integrated structure.

[0314] Active element Q of the impedance conversion element 2b of the acoustic element according to the second embodiment b ,like Fig.18B As shown, the substrate 14 is composed of a first conductive type (p-type) semiconductor region, and a second conductive type (n-type) semiconductor region is disposed separately from each other on the substrate 14. + The active element Q comprises a first main electrode region 15a and a second main electrode region 15b formed of a semiconductor region of the type. b like Fig.18B As shown, the active element Q having the gate dielectric film 13 contacts the upper surface of the base region 14 between the first main electrode region 15a and the second main electrode region 15b and includes a gate dielectric film 13 having a wider bandgap than the base region 14 and a control electrode 37 disposed on the gate dielectric film 13. b In the case of , the insulation breakdown of the gate dielectric film 13 may become a problem due to the excessive gate voltage. Fig.17 As shown, in the active component Q bThe control electrode 37 and the second power supply V DD Connect a positive voltage clamping diode D between k1 A negative voltage clamping diode D is connected between the control electrode 37 and the first power supply (GND). k2 By clamping the positive voltage through diode D k1 and negative voltage clamping diode D k2 It is connected to the control electrode 37 and is configured so that an excessive gate voltage is not applied to the gate dielectric film 13 .

[0315] Fig.17 exemplifies the circuit connection of connecting the fixed potential electrode 24a to the ground potential (GND), but is not limited to this circuit connection. Figure 4C , 4D , 5C, 7B, 7C, 8B, 9B, 10B, etc., when the fixed potential electrode 24a is connected to the ground potential (GND), the central voltage V of the voltage generating unit 1a p0 , the first peripheral voltage V p1 , the second peripheral voltage V p2 , the third peripheral voltage V p3 and the fourth peripheral voltage V p4 It will become a vibration waveform that vibrates in the positive and negative directions with 0V as the center. Fig.17 The circuit connection of the example is different. If the fixed potential electrode 24a is connected to the bias voltage V bias >0, the central voltage V generated by the voltage generating unit 1a p0 , the first peripheral voltage V p1 , the second peripheral voltage V p2 , the third peripheral voltage V p3 and the fourth peripheral voltage V p4 will become a bias voltage V different from 0V bias Centered at a bias voltage V bias That is, if the operating point V with the highest amplification factor on the IV characteristic of the impedance conversion element 2b is selected, bias and connect the fixed potential electrode 24a to V bias , the sensitivity of the acoustic element according to the second embodiment can be improved. That is, as long as the fixed potential electrode 24a is at a fixed potential, any bias voltage V can be selected. bias It is also possible to adjust any desired bias voltage V from a variable voltage source. bias Circuit structure.

[0316] Depending on the film formation conditions, the silicon oxide film (SiO 2 The bandgap of aluminum oxide (Al 2 O3 The band gap of silicon nitride film is about 8.3eV. 3 N 4 The band gap of the active element Q is about 4.9 eV. b It can be considered to have a heterojunction gate structure similar to that of a heterojunction FET. Fig.17 As shown, in the acoustic element according to the second embodiment, the active element Q b The control electrode 37 is electrically connected to the potential generating position P 0 , and the potential generates position P 0 The potential control in the active element Q b The height of the potential barrier generated in the channel defined between the first main electrode region 15a and the second main electrode region 15b and controlling the movement of carriers. Since the impedance conversion element 2b of the acoustic element according to the second embodiment includes the active element Q having the gate dielectric film 13 and the control electrode 37 b , as an equivalent circuit for explaining the main components, can correspond in form to the circuit already explained in the first embodiment. Figure 3A .

[0317] Around the base region 14, an element isolation insulating film 16 is provided in a frame shape to define the active element Q b The range of the active area. Fig.18B In the cross-sectional representation, the element separation insulating film 16 is located on both sides of the active area, but similar to the structure of the acoustic element according to the first embodiment, the element separation insulating films 16 located on both sides of the active area are continuous with each other on the front and back sides of the paper, forming an integral structure. Inside the pattern of the active area surrounded by the element separation insulating film 16, the range (region) of the upper surface of the substrate area 14 is defined, and the first main electrode area 15a and the second main electrode area 15b are provided in the upper part near the upper surface (surface) of the active area surrounded by the element separation insulating film 16. Fig.18A In the center, the range of the regular hexagonal potential transmission area is indicated by a two-dot chain line. For the hexagonal potential transmission area defined as the output surface side convex portion of the vibrating body 22a1, the active element Q b The channel is formed on the surface of the rectangular region sandwiched between the first main electrode region 15a and the second main electrode region 15b. Within the active region, in the active element Q b The height of the potential barrier to carrier movement generated in the channel between the first main electrode region 15 a and the second main electrode region 15 b is electrostatically controlled by the potential of the potential transfer region of the vibrating body 22 a 1 .

[0318] like Fig.18BAs shown, an interlayer insulating film 17 is provided on the first main electrode region 15a and the second main electrode region 15b. Through the contact holes opened in the interlayer insulating film 17, the first main electrode wiring line 25a is connected to the first main electrode region 15a, and the second main electrode wiring line 25b is connected to the second main electrode region 15b. Fig.18A and Fig.18B As shown, an input signal distribution line 38 composed of a conductor is provided on the control electrode 37, and they are connected by metal bonding. Fig.18A The input signal distribution line 38 is a distribution line in a length direction orthogonal to the length direction of the first main electrode distribution line 25a and the second main electrode distribution line 25b, and connects the control electrodes 37 to the p + The n-type anode region 42 and the n-type cathode region 43 correspond to Fig.17 The equivalent circuit representation of the positive voltage clamping diode D k1 , by p + The pn junction diode formed by the n-type anode region 42 and the n-type cathode region 41 is Fig.18A The area of ​​the paper located above the pattern of the first main electrode wiring line 25a and the second main electrode wiring line 25b. + The length direction of the pn junction diode formed by the n-type anode region 42 and the n-type cathode region 41 is parallel to the length direction of the patterns of the first main electrode wiring line 25 a and the second main electrode wiring line 25 b .

[0319] Similarly, corresponding to Fig.17 The negative voltage clamping diode D k2 ,exist Fig.18A The lower area of ​​the paper is provided with a + The pn junction diode formed by the n-type anode region 44 and the n-type cathode region 43 has its length direction parallel to the pattern of the first main electrode wiring line 25a and the second main electrode wiring line 25b. k2 The anode region 44 is connected to the first power supply side diode distribution line 36, and the positive voltage clamping diode D k1 The cathode region 41 is connected to the second power supply side diode distribution line 35. Fig.18BAs shown, a columnar (block-shaped) connection plug 29 is embedded as a contact portion (contact area) from the tip side of the central convex portion on the output surface side of the vibrating body 22a1, and the connection plug 29 is in ohmic contact with the vibrating body 22a1. The surface layer of the lower end portion of the connection plug 29, that is, the lower surface of the connection plug 29, can use a metal that is softer than the upper portion, and the connection plug 29 is formed by a stacked structure of hard metal and soft metal. The lower end portion of the connection plug 29 that functions as the electrical contact portion of the voltage generating part 1a is processed into a flat surface and becomes an end surface that is connected to the upper surface of the input signal wiring 38. In addition, by making physical contact between the lower surface of the connection plug 29 (that is, the connection end surface) and the upper surface of the input signal wiring 38, the central area P of the voltage generating part 1a is 0 The potential of the active element Q constituting the impedance conversion element 2b b The control voltage is input.

[0320] In order to reduce the ohmic contact resistance between the vibrating body 22a1 and the connecting plug 29, a contact resistance improving layer such as an alloy layer may be sandwiched between them. The connecting plug 29 does not necessarily have to be embedded in the tip of the central protrusion, and may also be plate-shaped to form a simple plane connection between the tip of the protrusion and the plate-shaped surface. If the connecting plug 29 is regarded as a component, the voltage generating unit 1a includes the vibrating body 22a1, the fixed potential electrode 24a, the fixed potential electrode protective film 23 and the connecting plug 29. In addition, a new connection means such as a bump may be further sandwiched between the connecting plug 29 and the input signal distribution line 38.

[0321] That is, the potential of the connection plug 29 is input to the control electrode 37 through the input signal distribution line 38, and the potential of the control electrode 37 is electrostatically controlled by the gate dielectric film 13 to control the height of the potential barrier to carrier movement generated in the channel between the first main electrode region 15a and the second main electrode region 15b. Fig.18B The structure shown is an example of a structure in which the connection plug 29 of the voltage generating unit side chip provided at the upper part and the input signal distribution line 38 of the impedance conversion element side chip provided at the lower part are in physical contact with each other. Therefore, around the vibration cavity 21a1, the direct bonding insulating film 31 with a mirror surface provided on the voltage generating unit side chip and the cavity inner wall protection film 19 with a mirror surface provided on the impedance conversion element side chip are directly bonded by hydrophilic bonding (hydrogen bonding between silanol groups).

[0322] Compared with the floating impedance when the PMUT and FET are connected by external wiring as described in non-patent document 2, in the acoustic element according to the second embodiment, since a thin plate-shaped input signal distribution line 38 and a control electrode 37 are used, and an electrical connection path is directly connected in the thickness direction with the shortest distance, the floating impedance can be greatly reduced. In addition, compared with the PMUT in non-patent document 2, the voltage generating part 1a adopts a new structure of a stress-enhanced structure to achieve high sensitivity and broadband, so the acoustic element according to the second embodiment can achieve high sensitivity and broadband. In particular, by microscopically studying the internal structure of the voltage generating part 1a, a specific position where the highest voltage is generated is selected as a potential transfer area, and as an input transfer position to the impedance conversion element 2b, so that high sensitivity can be easily achieved.

[0323] - Modification of the Second Embodiment -

[0324] exist Fig.18B In the embodiment, a structure in which the cylindrical connection plug 29 is embedded upward from the tip side of the convex portion downward from the output surface side of the vibrating body 22a1 as the contact portion is exemplified, but the structure is not limited to this. Fig.18C As shown, a connecting plug 29 composed of a columnar conductor may be provided. pil As the central support of the vibration cavity 21a1, and around the connecting plug 29 pil The parallel toroidal vibration cavities 21a1 are arranged in a centrally symmetrical manner. Fig.18B The description of the invention mentions that the connecting plug 29 may be in the form of a plate. Fig.18C The structure shown corresponds to Fig.18B The thickness of the central protrusion is set to zero, and the structure is composed entirely of thick plate-shaped connection plugs 29. Connection plug 29 pil Become Occupy Fig.18A The potential transmission area of ​​the regular hexagonal shape indicated by the double-dashed line in the center is a regular hexagonal prism, and the parallel annular surfaces constituting the vibration cavity 21a1 have a diameter around the connection plug 29. pil The plane pattern is a hexagonal ring topology. pil , the planes are joined at the center of the output surface of the vibrating body 22a1. As a result, the connecting plug 29 pil Corresponding to the contact portion (contact area) whose six sides are exposed to the hexagonal ring surface. Fig.18C In the cross-sectional view, from the appearance, the two vibration cavities 21a1 are located in the center of the connecting plug 29. pil on both sides of the paper, but they are continuous with each other on the front and back of the paper and form an integrated structure.

[0325] like Fig.18C The active components shown are Fig.18B The active components Q shownb Similarly, the gate dielectric film 13 is contacted with the upper surface of the base region 14 between the first main electrode region 15a and the second main electrode region 15b, and includes a gate dielectric film 13 having a wider bandgap than the base region 14 and a control electrode 37 disposed on the gate dielectric film 13. Fig.18C As shown, an input signal distribution line 38 composed of a conductor is connected to the control electrode 37 by metal bonding. pil The flat upper surface of the vibrator 22a1 is in ohmic contact with the vibrator 22a1 at the center of the output surface side of the vibrator 22a1. Although not shown in the figure, the vibrator 22a1 is in ohmic contact with the connecting plug 29 at the center of the output surface side of the vibrator 22a1. pil A contact resistance improving layer such as an alloy layer may be inserted between the flat upper surfaces.

[0326] Then, a soft metal is used as the surface layer (in Fig.18C The connection plug 29 is the lowest layer in the middle pil The flat lower surface of the connector is in physical contact with the upper surface of the input signal distribution line 38. pil electrically connected to the input signal distribution line 38, Fig.17 The central area P of the voltage generating section 1a shown in FIG. 0 The potential of the impedance conversion element 2b is input as a control voltage to the active element Q b .exist Fig.18C In the representation, the connecting plug 29 shown in the cross-sectional view pil The width of the input signal distribution line 38 and the width of the control electrode 37 are shown to be the same, but this is only an example. pi The width of the input signal distribution line 38 may be different from the width of the input signal distribution line 38, and the width of the control electrode 37 may also be different. pil The width of the control electrode 37 may also be different. For example, the width of the control electrode 37 may be greater than Fig.18C Connecting plug 29 shown in pil The width of the control electrode 37 may be larger than that of the connecting plug 29. pil The width is small and T-shaped. Fig.18B The acoustic element according to the second embodiment shown is the same, and therefore repeated description is omitted.

[0327] Compared with the floating impedance when the PMUT and the FET are connected by external wiring as described in Non-Patent Document 2, in the acoustic element according to the modification of the second embodiment, the connection plug 29 pil, the input signal distribution line 38 and the control electrode 37 are directly connected in the thickness direction with the shortest distance to form an electrical connection path, so that the floating impedance can be greatly reduced. In addition, compared with the PMUT in non-patent document 2, the voltage generating part 1a adopts a new structure of stress-enhanced structure to achieve high sensitivity and broadband, so the acoustic element according to the modified example of the second embodiment can achieve high sensitivity and broadband. In particular, a specific position on the output surface of the vibrating body 22a1 where the highest voltage is generated is selected as the potential transfer area, and the short connection plug 29 is used to connect the voltage generating part 1a to the output surface of the vibrating body 22a1. pil As the input transfer position to the active element, it is easy to achieve high sensitivity.

[0328] - Dependence of the hardness of the central pillar material - when using the thickness t 22a1 = 2.4 μm PZT as the vibrator 22a1, change Fig.18C The cylindrical connection plug 29 with a diameter of 2.4 μm acts as a central support pil Voltage sensitivity and half-value bandwidth for material hardness, such as Fig.18D The outer diameter of the ring is d. 11 =60μm(reference Figure 4A Medium 11 ), and assuming that the vibrating body 22a1 is covered with a thickness t 23 = 6μm soft epoxy resin fixed potential electrode protection film 23. Fig.18D As shown in the figure, within the half-value bandwidth of 5 to 11 MHz for receiving ultrasonic waves, the tungsten (W) represented by the thin solid line and the silicon nitride film (Si) represented by the thin dotted line 3 N 4 The maximum voltage sensitivity of the structure with the membrane as the central pillar of the material is about 5.14μV / Pa, which is the highest. Fig.18D In the figure, the thin solid line of W and Si 3 N 4 The thin dashed lines of the films are almost overlapping. W and Si 3 N 4 The film has the highest and second highest Young's modulus among the materials shown in Table 3.

[0329] (Table 3)

[0330] Material Name Poisson's ratio Elastic modulus(Gpa) Cu 0.343 124-129 W 0.28 380-411 Al 0.345 69-70 <![CDATA[Si 3 N 4 ]]> 0.27-0.25 210-310 Si 0.266 107 <![CDATA[SiO 2 ]]> 0.17-0.2 66-94 Polyester 0.05 0.01-0.06 Soft epoxy resin 0.34-0.37 2.6-3 Polyimide 0.30-0.458 0.2-1.3

[0331] Moreover, although it is higher than W and Si 3 N 4 The sensitivity curve of the membrane is slightly lower frequency, but the sensitivity curve of the central pillar structure with copper (Cu) as the material represented by the thick solid line is similar to that of W and Si 3 N 4 The sensitivity curves of the films are almost the same, and when Cu is used as the connecting plug 29pil The maximum voltage sensitivity is also about 5.14μV / Pa. In addition, the silicon oxide film (SiO 2 The sensitivity curve of the central pillar structure of the material is similar to that of W and Si 3 N 4 The sensitivity curves of the film and Cu are almost identical, with the same maximum voltage sensitivity and half-value bandwidth curves.

[0332] In addition, the sensitivity curve of the central pillar structure of aluminum (Al) as the material, represented by the single-dot chain line, is about 0.5MHz lower than the sensitivity curve of Cu, and is lower than that of W and Si. 3 N 4 Film, Cu and SiO 2 The sensitivity curve of the film is slightly lower, with a maximum voltage sensitivity of about 5.0μV / Pa, and the half-value bandwidth curve is almost the same as them. In addition, the sensitivity curve of the central pillar structure of silicon (Si) as the material, represented by the double-dotted line, which is about 0.3MHz lower than the sensitivity curve of Cu, has a maximum voltage sensitivity of about 4.9μV / Pa, which is slightly lower than the sensitivity curve of Al, and the half-value bandwidth curve is similar to that of w and Si. 3 N 4 Film, Cu, SiO 2 The half-value bandwidths of the film and Al are almost the same. 3 N 4 Film, Cu, SiO 2 In the range of films, metals such as Al and Si, and non-organic compounds, it can be seen that the hardness of the material constituting the central pillar structure has no great dependence on the maximum voltage sensitivity and half-value bandwidth.

[0333] In contrast, the two sensitivity curves of polyurethane (PUR) represented by the thick dashed line and polyethylene (PE) represented by the thin solid line as the central pillar structure of the material are located at a lower level than SiO 2 The sensitivity curve of the membrane is on the low-frequency side of about 3.8MHz. The sensitivity curve of the central pillar structure of PUR and PE as the material has a maximum voltage sensitivity of about 3.9μV / Pa, which is lower than that of W, Si 3 N 4A set of sensitivity curves of the central pillar structure composed of metal and non-organic compounds such as membranes are low, and are displayed as two curves overlapping with almost the same trajectory. The sensitivity curves of the central pillar structure of PUR and PE as materials are displayed as half-value bandwidth curves in the 4.1-7.3MHz region (on the lower frequency side than the region of the set of sensitivity curves of metal and non-organic compounds), and the half-value bandwidth is narrowed. In addition, the sensitivity curve of soft epoxy resin (EP resin) represented by the thick solid line has a maximum voltage sensitivity of about 3.5μV / Pa, which is lower than the sensitivity curves of PUR and PE, and shows a half-value bandwidth of 4.1-8.0MHz. Fig.18D It can be seen that, compared with W and Si 3 N 4 Compared with the case where the central pillar structure is composed of metals such as membranes and non-organic compounds, when the central pillar structure is composed of polymer materials such as PUR, the maximum voltage sensitivity is lower, the half-value bandwidth shifts to the low frequency side, and the bandwidth becomes narrower.

[0334] =Method for manufacturing acoustic element according to second embodiment=

[0335] use Figures 19A to 19K A method for directly bonding a voltage generating unit side chip and an impedance conversion element side chip is described. Fig.18B The method for manufacturing the acoustic element according to the second embodiment is exemplified by the following. It should be noted that the method for manufacturing the acoustic element according to the second embodiment described below is only an example, and is not limited to the case where the voltage generating unit side chip and the impedance conversion element side chip are directly bonded by hydrophilic bonding (fusion bonding). If the temperature process accompanying the film formation of the vibrating body 22a1 is not a problem, a heteroepitaxial or CVD method similar to the method for manufacturing the acoustic element of the second variant of the first embodiment described above can also be used. In addition, even in the case where the voltage generating unit side chip and the impedance conversion element side chip are stacked, electrical connection and physical strength can be achieved by bump connection like a normal three-dimensional stacked chip structure, and it is not necessary to be limited to the following manufacturing method. For the bump connection, in addition to the exemplified gold (Au)-Au direct bonding, copper (Cu)-Cu direct bonding, silver (Ag)-tin (Sn) solder, indium (In)-Sn solder, or indirect bonding using eutectic components such as Au-Sn and Cu-Sn are also possible. Therefore, as long as the structure described in the claims can be achieved, the acoustic element according to the second embodiment can of course be realized by various manufacturing methods other than the following examples.

[0336] In addition, as mentioned at the beginning of the description of the method for manufacturing the acoustic element of the second variant of the first embodiment, the names with ordinal numbers such as "the first photoresist film" in the following description are convenient names used for the rhetorical requirement of distinguishing from other photoresist films, and do not mean the first order used in the actual process. Fig.19A The stage before the structure shown in the figure usually has implemented multiple processing stages using photoresist films, and in actual processes it is usually not the "first photoresist film". In addition, the order can be Figures 19A to 19E The impedance conversion element side chip process shown is performed before (before) Figure 19F to Figure 19J The voltage generating unit side chip process is shown, and if the voltage generating unit side chip process is performed first, the ordinal numbers such as "first photoresist film" used for convenience in the following description will of course change.

[0337] (a) First, as the basis of the impedance conversion element side chip, a base region 14 composed of a p-type silicon substrate with a (100) plane as the main surface and a resistivity of 0.1 to 10 Ωcm is prepared. Then, using the known LOCOS or STI technology, an element isolation insulating film 16 such as a silicon oxide film is formed to a thickness of about 0.4 to 1.2 μm. A gate dielectric film 13 and a DPOS film are deposited on the base region 14 where the planar pattern of the active region surrounded by the element isolation insulating film 16 is exposed, and the DPOS film is patterned to form a control electrode 37. By ion implantation, n gate electrodes 37 are formed in a self-aligned manner in the active region using the control electrode 37 as a mask. + The cross-sectional structure of the first main electrode region 15a and the second main electrode region 15b is shown in FIG. An example of a planar pattern of the first main electrode region 15a and the second main electrode region 15b may be, for example Fig.18A When forming the first main electrode region 15a and the second main electrode region 15b, Fig.19A The back of the paper is formed Fig.18A The cathode region 41 shown in the plan view of the drawing is formed on the front side of the drawing. After the cathode regions 41 and 43 are formed, an ion implantation method is used to form a cathode region 41 in the pattern. Fig.18A The p shown in + A p-type anode region 42 is formed inside the pattern of the cathode region 43. + Type anode region 44.

[0338] (b) Then, a first insulating film 17p such as a silicon oxide film having a thickness of about 180 to 250 nm is deposited on the first main electrode region 15a, the second main electrode region 15b, and the exposed substrate region 14 between the first main electrode region 15a and the second main electrode region 15b by a deposition method such as CVD. When the substrate region 14 is Si, thermal oxidation can also be used. After the first insulating film 17p is deposited by a method such as CVD, it is flattened by a method such as chemical mechanical polishing (CMP) until the upper surface of the element separation insulating film 16 is exposed. Thereafter, a first photoresist film is coated on the first insulating film 17p, and the first photoresist film is exposed and developed by photolithography to form a first etching mask for opening contact holes. Using the first etching mask, the first insulating film is selectively etched by a dry etching technique such as reactive ion etching (RIE), and a contact hole pattern is opened in the upper part of the first main electrode region 15a, the second main electrode region 15b, the control electrode 37, the anode region 42, 44, and the cathode region 41, 43.

[0339] (c) Further, a first conductor film composed of a conductor layer such as a DOPOS film or a high melting point metal is deposited on the entire surface of the first insulating film by a deposition method such as CVD to have a thickness of about 80 to 200 nm. A second photoresist film is applied on the first conductor film, and the second photoresist film is exposed and developed using a photolithography technique to form a second etching mask for patterning the first main electrode wiring line 25a and the second main electrode wiring line 25b (refer to Fig.18A ). Using a second etching mask, a dry etching technique such as RIE is used, such as Fig.19A As shown, the first conductor film is selectively etched to form a pattern of first main electrode wiring lines 25a connected to the first main electrode region 15a and a pattern of second main electrode wiring lines 25b connected to the second main electrode region 15b.

[0340] (d) Then, on the first main electrode wiring line 25a, the second main electrode wiring line 25b and the control electrode 37, as shown in FIG. Fig.19B As shown, a cavity inner wall protective film 19 such as a silicon oxide film with a thickness of about 180 to 250 nm is deposited by a deposition method such as CVD. Further, a third photoresist film is coated on the cavity inner wall protective film 19, and the third photoresist film is exposed and developed using a photolithography technique to form a third etching mask for controlling the opening of contact holes on the electrode 37, the anode regions 42, 44, and the cathode regions 41, 43 (refer to Fig.18A ). Using a third etching mask, the cavity inner wall protection film 19 is selectively etched by dry etching techniques such as RIE, and contact holes are opened on the control electrode 37, the anode regions 42, 44, and the cathode regions 41, 43. Fig.19AIn the figure, the contact holes on the anode region 42 and the cathode region 41 are located on the back of the paper, and the contact holes on the anode region 44 and the cathode region 43 are located on the front of the paper (refer to Fig.18A ).

[0341] (e) Next, after removing the third photoresist film, a fourth photoresist film is coated on the cavity inner wall protection film 19, and the fourth photoresist film is exposed and developed using a photolithography technique, thereby forming a fourth etching mask for patterning the Damascus grooves of the input signal wiring 38, the first power side diode wiring 36, and the second power side diode wiring 35 (see FIG. Fig.18A ). Using the fourth etching mask, by dry etching techniques such as RIE, such as Fig.19C As shown, the inner wall protective film 19 of the cavity is selectively etched to form a damascene groove for connecting the control electrode 37 to the anode region 42 and the cathode region 43, the first power supply side diode wiring 36 connected to the anode region 44, and the second power supply side diode wiring 35 connected to the cathode region 41. The damascene groove of the input signal wiring 38 is integrated with the contact hole on the control electrode 37 above the control electrode 37. Fig.19C In the figure, the Damascus groove of the first power supply side diode wiring 36 is located deep in the paper, and the Damascus groove of the second power supply side diode wiring 35 is located near the paper (see Fig.18A ).

[0342] (f) Further, if Fig.19D As shown in FIG. 1 , on the cavity inner wall protection film 19 formed with the Damascus groove, a second conductor film 38p composed of a conductor layer such as a copper (Cu) film with a thickness of about 80-200 nm is deposited by a deposition method such as CVD. Then, a planarization method such as CMP is used to expose the upper surface of the cavity inner wall protection film 19, so that the second conductor film 38p is embedded in each Damascus groove, as shown in FIG. Fig.19E As shown. By embedding the second conductor film 38p into the Damascus groove, an input signal wiring 38 connecting the control electrode 37 to the anode region 42 and the cathode region 43, a first power side diode wiring 36 connected to the anode region 44, and a second power side diode wiring 35 connected to the cathode region 41 are formed as Damascus wiring. The top layer of the Damascus wiring is preferably a soft metal with a Vickers hardness of about 20Hv-30Hv, such as Au. In addition, Au alloys with a Vickers hardness of about 15Hv-120Hv, such as Au-silicon (Si), Au-germanium (Ge), Au-antimony (Sb), Au-tin (Sn), Au-lead (Pb), Au-zinc (Zn), Au-copper (Cu), etc., with an Au content of more than 80%, can also be used in the top layer of the Damascus wiring. Fig.19EIn the figure, the first power supply side diode wiring 36 is located deep in the paper, and the second power supply side diode wiring 35 is located near the paper (see Fig.18A The upper surface of the cavity inner wall protection film 19 which has been flattened by a method such as CMP is finally processed into a mirror surface suitable for the bonding process, thereby completing the "impedance conversion element side chip".

[0343] (g) On the other hand, in order to prepare the "voltage generating portion side chip", a PZT substrate with a thickness of about 100 μm is prepared. On one main surface of the prepared PZT substrate, a second insulating film 31p composed of a silicon oxide film or the like with a thickness of about 180-250 nm is deposited by a deposition method such as CVD. Afterwards, the other main surface of the PZT substrate is adjusted to the thickness required as the vibrator 22a1. In order to maintain the strength of the PZT substrate, the peripheral portion of the PZT substrate can be retained as a frame of the thickness of the base material, and the frame is a recessed portion. Only the thickness of the central portion corresponding to the bottom of the recessed portion is adjusted to the required thickness as the vibrator 22a1. This step structure is also feasible. Then, on the bottom of the recessed portion on the other main surface side of the PZT substrate as the vibrator 22a1, a third conductor film such as Al or Al alloy is deposited on the entire surface using a deposition method such as sputtering, vacuum evaporation or CVD. Next, a fifth photoresist film is coated on the third conductor film, and a pattern for forming a fixed potential electrode is formed using photolithography technology. The third conductor film is selectively etched using the fifth photoresist film as a fifth etching mask to form a Fig.19F The pattern of the fixed potential electrode 24a is shown.

[0344] (h) Next, if Figure 19G As shown in the figure, a fixed potential electrode protection film 23 such as epoxy resin is deposited on the vibrating body 22a1 by spin coating or other methods to cover the pattern of the fixed potential electrode 24a. In order to ensure the strength of the vibrating body 22a1, the thickness of the fixed potential electrode protection film 23 can be thicker than the final design value. Alternatively, a reinforcing substrate (not shown in the figure) can be bonded to the fixed potential electrode protection film 23.

[0345] (i) Then, the vibrator 22a1 is turned over so that one main surface of the PZT substrate faces upward. A sixth photoresist film is coated on the second insulating film 31p on one main surface of the PZT substrate, and a pattern for embedding a connecting plug is formed using a photolithography technique. The sixth photoresist film is used as a sixth etching mask to form a plug insertion groove that penetrates the second insulating film 31p and reaches the vibrator 22a1, as shown in FIG. Fig.19H Next, a fourth conductive film composed of a conductive layer such as a DOPOS film and a high melting point metal is deposited on the second insulating film 31p by a deposition method such as a CVD method, a sputtering method, or a vacuum evaporation method, and its thickness is thicker than the depth of the plug buried groove. After being planarized by a method such as CMP, the connection plug 29 is embedded in the plug insertion groove, as shown in FIG. Fig.19I As shown. At least the uppermost layer of the connection plug 29 is preferably a metal film with a low Vickers hardness, such as Au or Au alloy. In order to reduce the ohmic contact resistance between the vibrating body 22a1 and the connection plug 29, after the connection plug 29 is embedded, heat treatment may be performed to form an alloy layer between the vibrating body 22a1 and the connection plug 29. Alternatively, before the connection plug 29 is embedded in the vibrating body 22a1, an ohmic contact resistance improvement layer composed of other metals or the like is first deposited as a bottom layer, and then the connection plug 29 is embedded, and then heat treatment is performed.

[0346] (j) After that, a seventh photoresist film is coated on the second insulating film 31p, and the seventh photoresist film is exposed and developed by photolithography technology to form a seventh etching mask for forming a sacrificial layer. The second insulating film 31p is selectively etched by dry etching technology such as RIE using the seventh etching mask to form a sacrificial layer. Fig.19J The circular vibration cavity is shown with a pattern of concave portions. Fig.19J In the cross-sectional view of FIG, the two vibration cavities are symmetrically located on both sides with the pattern of the concave portions. However, in reality, Fig.19J The two concave parts for vibration cavities, which seem to be symmetrically located on both sides, are continuous annular structures in front and behind the paper, and are a whole. The second insulating film 31p remaining around the concave part for vibration cavity becomes the insulating film 31 for direct bonding, and the chip on the voltage generating part side is completed. The surface of the insulating film 31 for direct bonding is also processed into a mirror surface.

[0347] (k) Then, the voltage generating unit side chip is turned over so that the direct bonding insulating film 31 side faces downward, and is moved above the impedance conversion element side chip. Figure 19K As shown. Next, the voltage generating unit side chip and the impedance conversion element side chip are directly bonded by hydrophilic bonding through the "glass optical contact" between the direct bonding insulating film 31 and the cavity inner wall protection film 19. The silicon oxide film constituting the ultra-smooth direct bonding insulating film 31 and the silicon oxide film constituting the cavity inner wall protection film 19 can be closely attached only by contact. In hydrophilic bonding, hydrogen bonds between silanol groups formed on the glass surface due to water adsorption make the glass closely attached. In order to achieve strong bonding, post-annealing treatment at about 800°C can be performed. During the post-annealing process, due to thermal expansion, the connection plug 29 extending from the plug buried groove is thermally pressed to the input signal line 38 and bonded to the metal due to thermal expansion during the post-annealing. If the surfaces of the direct bonding insulating film 31 and the cavity inner wall protection film 19 are irradiated with oxygen plasma respectively to form a porous oxide film for plasma activation bonding, the adsorbed excess water and hydrogen generated by the decomposition of the silanol group during the post-annealing process are more easily released, and the post-annealing temperature of the hydrophilic bonding can be reduced to about 300°C. By directly bonding the voltage generating unit side chip and the impedance conversion element side chip, the space formed by the concave portion for the vibration cavity above the cavity inner wall protection film 19 becomes as follows. Fig.18B The parallel annular vibration cavity 21a1 shown completes the acoustic element of the second embodiment.

[0348] Also, as mentioned above, use Figures 19A-19K The acoustic element manufacturing method of the second embodiment described is only an example. Fig.19F The structure shown can also be formed by forming SiO on a Si substrate by a thermal oxidation method or the like. 2 In this case, SiO2 is formed on a Si substrate. 2 On the film, a third conductor film is deposited on the whole by sputtering method or the like, and the third conductor film is selectively etched by photolithography technology to form a Fig.19F The pattern of the fixed potential electrode 24a shown in FIG. Then, on the pattern of the fixed potential electrode 24a, PZT22p is deposited to a desired thickness by a high-frequency magnetron sputtering method, an MOCVD method, a solution coating method (sol-gel method), etc. For example, after high-frequency magnetron sputtering at room temperature, rapid heat treatment at about 500°C or less using a halogen lamp or an excimer laser can be performed to improve the crystallinity of PZT22p.

[0349] Furthermore, if the second insulating film 31p is deposited on the PZT 22p by a CVD method or the like, then Fig.19F In the structure shown, a "substructure" corresponding to a structure without a fixed potential electrode protection film 23 is completed on the Si substrate. However, not only Fig.19F The substructure is due to the SiO 2 The film is set on a thick Si substrate, thus forming a reinforcement structure in which the physical strength of the thin PZT22p is enhanced by the thick Si substrate. The reinforcement structure with the Si substrate is maintained until Figure 19K That is, after the voltage generating unit side chip and the impedance conversion element side chip are directly bonded, the SiO between the Si substrate and the fixed potential electrode 24a is removed. 2 The fixed potential electrode 24a is exposed, and then a fixed potential electrode protective film 23 such as epoxy resin is deposited on the exposed fixed potential electrode 24a, thereby completing the acoustic element of the second embodiment.

[0350] The third embodiment is as already described Figure 4C and 4D As described in the examples in the literature, in a circuit connection in which the fixed potential electrode 24a is connected to the ground potential (GND), the central region P defined in the potential transfer region of the voltage generating section 1a is 0 , the first peripheral area P 1 , ..., the fourth peripheral area P 4The detected potential shows a transient response vibrating in the positive direction and the negative direction. In order to cope with the case where the voltage generating portion 1a generates a voltage waveform vibrating in the positive direction and the negative direction, the acoustic element of the third embodiment of the present invention provides a structure including two types of active elements in the impedance conversion element 2c. That is, Fig. 20 As shown, the acoustic element of the third embodiment is a receiving element X including a voltage generating portion 1a that generates a non-uniform voltage by a piezoelectric effect and an impedance conversion element 2c that operates with a potential at a specific position (local portion) of the voltage generating portion 1a as a control voltage. i,j , and includes a CMOS inverter (Q p , Q n ) as impedance transformation element 2c.

[0351] from Fig. 21B As can be seen from the cross-sectional view of FIG. 1 , the voltage generating portion 1a has a vibrating body 22a1 similar to the acoustic element of the first and second embodiments and a receiving surface (at the Fig. 21B The fixed potential electrode 24a is in contact with the upper main surface (in the middle), forming a single piezoelectric chip type structure. Fig. 21B As shown in FIG. 1 , the entire vibrating body 22a1 may be composed of piezoelectric bodies belonging to 20 crystal point groups, or only the central downward convex portion (stress concentration portion) may be composed of 20 piezoelectric bodies, and the rest may be composed of non-piezoelectric bodies belonging to the other 12 crystal groups. Among the 20 crystal point groups, HfO2, which belongs to the 10 crystal point groups having thermoelectricity, is particularly 2 , AlN, PZT and other materials are suitable as the material of the downward protruding portion (stress concentration portion) in the center of the vibrating body 22a1. That is, the voltage generating part 1a only needs to have a structure in which at least a piezoelectric layer is provided at the stress concentration portion of the vibrating body 22a1. Therefore, the impedance conversion element 2c can use at least a specific part of the piezoelectric layer provided at the downward protruding portion (stress concentration portion) as a potential transfer region, and operate with the potential of the potential transfer region as a control voltage.

[0352] The voltage generating portion 1a of the unimorph structure of the acoustic element of the third embodiment is provided on an output surface (on the opposite side of the receiving surface) of the vibrating body 22a1 opposite to the receiving surface (on the opposite side of the receiving surface) of the vibrating body 22a1. Fig. 21B At least a piezoelectric layer is provided at a specific position (the lower main surface in the figure), and a voltage distribution is provided that generates a higher voltage than other positions by the piezoelectric effect. The "specific position" of the output surface of the vibrating body 22a1 is, for example, Fig. 20 The central region P is schematically shown in the potential transfer region arrangement candidate model. 0 , the first peripheral area P 1 , ..., the fourth peripheral area P 4Any one of them, or a combination of them. Fig. 20 In the model representation, the central area P 0 , the first peripheral area P 1 , ..., the fourth peripheral area P 4 For example, the central area P 0 As the center, to surround the central area P 0 The mode consists of the first peripheral area P 1 , the second peripheral area P 2 , the third peripheral area P 3 , the fourth peripheral area P 4 A pattern of concentric hexagonal rings.

[0353] In the acoustic element of the third embodiment, the CMOS inverter (Q p , Q n ) is connected to the potential transfer region of the voltage generating section 1a, that is, the central region P 0 By connecting the input electrode to the central area P as the potential transfer area 0 , CMOS inverter (Q p , Q n ) of the p-type active element Q p and n-type active element Q n The potential barrier height generated in each channel is determined by the central area P of the voltage generating section 1a. 0 As a p-type active element Q p , corresponding to p-channel MIS transistors such as p-channel MOSFET, p-channel MOSFET, p-channel MISFET, p-channel MISSIT, and p-channel HEMT (hereinafter referred to as "pMIS transistor"). Similarly, as an n-type active element Q n , corresponding to n-channel MIS transistors such as n-channel MOSFET, n-channel MOSIT, n-channel MISFET, n-channel MISSIT, and n-channel HEMT (hereinafter referred to as "nMIS transistor").

[0354] CMOS Inverter (Q p , Q n ) includes connecting the first main electrode region 15a to a first power source V SS (=-V DD ), connecting the second main electrode region 15b to the n-type active element Q of the output signal wiring 25c n As the first active element. Active element Q n The potential barrier in the n-type channel is the active element Q nThe main current (electrons) flowing between the first main electrode region 15a and the second main electrode region 15b is prevented from moving. p , Q n ) includes connecting the third main electrode region 15d to a voltage higher than the first power source V SS A second power supply V with a higher potential DD , connecting the fourth main electrode region 15c to the p-type active element Q of the output signal wiring 25c p As the second active element. Active element Q p The potential barrier in the p-type channel is the active element Q p The CMOS inverter (Q) as the impedance conversion element 2c is used to prevent the movement of the main current (holes) flowing between the third main electrode region 15d and the fourth main electrode region 15c. p , Q n ) is connected to the first power supply V SS (=-V DD ) and the second power supply V DD Between, such as Fig. 20 As shown, it can correspond to the input voltage waveform vibrating in the positive direction and the negative direction, such as in Figure 4C As exemplified in etc.

[0355] CMOS inverter (Q p , Q n ) of the n-type active element Q n The first control electrode 37n is connected to the first input signal wiring 39 through the first gate connection plug 38n. On the other hand, the p-type active element Q p The second control electrode 37p is connected to the first input signal wiring 39 through the second gate connection plug 38p. Then, as shown in FIG. Fig. 21B As shown in FIG. 1 , a second input signal wiring 45 is provided on the first input signal wiring 39. The CMOS inverter (Q p , Q n ) The convex portion located on the output side of the vibrating body 22a1 (such as Fig. 21B The central area P 0 The potential is transmitted to the n-type active element Q through the second input signal wiring 45 and the first input signal wiring 39. n The first control electrode 37n and the p-type active element Q p That is, the impedance conversion element 2c of the acoustic element of the third embodiment uses the pressure of the ultrasonic wave Φ applied to the vibrating body 22a1 of the voltage generating unit 1a as the central area P 0 The potential of the CMOS inverter (Q p , Qn ) input signal, and the impedance-converted signal is transmitted from the CMOS inverter (Q p , Q n Then, the acoustic element of the third embodiment outputs the CMOS inverter (Q p , Q n ) is used as the output signal of the acoustic element.

[0356] The convex portion constituting the output surface side of the vibrating body 22a1 of the voltage generating portion 1a of the acoustic element of the third embodiment is as follows. Fig.21A The double-dotted line shows the regular hexagonal columnar protrusion. Fig. 21B As shown, around the protrusion set in the central part of the output surface side of the vibrator 22a1, a parallel annular vibration cavity 21a1 is set in a centrally symmetrical manner to surround the protrusion. The parallel annular structure constituting the vibration cavity 21a1 is a hexagonal topology in the plane pattern. Fig. 21B In the cross-sectional view, it appears that the two vibration cavities 21a1 are located on both sides of the central protrusion, but in fact they are an integrated structure that is continuous with each other on the front and back of the paper.

[0357] CMOS inverter (Q p , Q n ) of the n-type active element Q n like Fig. 21B As shown, there is a well region 52 composed of a first conductive type (p-type) semiconductor region, and a second conductive type (n-type) semiconductor region spaced apart from each other on the well region 52. + The well region 52 is partially disposed on the upper portion of the base region 51 composed of the second conductivity type (n-type) semiconductor region. In addition, the n-type active element Q n like Fig. 21B As shown, the first gate dielectric film 13n, which contacts the upper surface of the well region 52 between the first main electrode region 15a and the second main electrode region 15b and has a wider bandgap than the well region 52, and the first control electrode 37n arranged on the first gate dielectric film 13n is an active element.

[0358] CMOS inverter (Q p , Q n ) of the p-type active element Q p like Fig. 21B As shown, there is a base region 51, and a first conductive type (P + The third main electrode region 15d and the fourth main electrode region 15c are composed of a p-type semiconductor region.p like Fig. 21B As shown, it is an active element having a second gate dielectric film 13p that contacts the upper surface of the substrate region 51 between the third main electrode region 15d and the fourth main electrode region 15c and has a wider bandgap than the substrate region 51, and a second control electrode 37p arranged on the second gate dielectric film 13p.

[0359] like Fig. 20 As shown, in the CMOS inverter (Q p , Q n ) input signal terminal (corresponding to Fig.21A and 21B The first input signal wiring 39 and the second input signal wiring 45) are connected to the second power supply V DD A p-side clamping diode D is connected between kp , between the input signal terminal and the first power supply V SS (=-V DD ) is connected with an n-side clamping diode D kn By clamping the p-side diode D kp and n-side clamping diode D kn Connected to the CMOS inverter (Q p , Q n ) input signal terminal, so that the n-type active element Q n The first gate dielectric film 13n and the p-type active element Q p An excessive gate voltage is not applied to the second gate dielectric film 13p.

[0360] As described in the second embodiment, it can be considered that the CMOS inverter (Q p , Q n ) of the n-type active element Q n and p-type active element Q p Each has a heterojunction gate structure. Fig. 21B As shown in the third embodiment, in the acoustic element, the CMOS inverter (Q p , Q n ) is connected to the potential generating position P 0 Therefore, a CMOS inverter (Q p , Q n ) of the n-type active element Q n The potential barrier height generated in the n-type channel between the first main electrode region 15a and the second main electrode region 15b is determined by the potential generating position P 0 Similarly, the CMOS inverter (Q p , Qn ) of the p-type active element Q p The potential barrier height generated in the p-type channel between the third main electrode region 15d and the fourth main electrode region 15c is also related to the n-type active element Q n Complementarily, the potential generates the position P 0 Potential control.

[0361] Around the well region 52 locally provided on the upper part of the n-type base region 51, an element isolation insulating film 16 is provided in a frame shape for defining the CMOS inverter (Q p , Q n ) of the n-type active element Q n The range of the active region. In defining the n-type active element Q n Adjacent to the frame-shaped bank of the active region pattern, there is also a p-type active element Q defined p That is, the n-type active element Q n The active region and p-type active element Q p The active region of the n-type active element Q is adjacent to each other through the element isolation insulating film 16. n The region on the upper surface of the well region 52 is defined within the active region range, and the first main electrode region 15a and the second main electrode region 15b are provided near the upper surface (surface) of the active region surrounded by the element isolation insulating film 16. p Inside the active region pattern, an area on the upper surface of the substrate region 51 is defined, and a third main electrode region 15d and a fourth main electrode region 15c are provided near the upper surface (surface) of the active region surrounded by the element isolation insulating film 16.

[0362] exist Fig.21A In the center, the range of the potential transfer region of a regular hexagon is indicated by a double-dashed line. The hexagonal potential transfer region defined as the convex portion on the output surface side of the vibrator 22a1 constitutes a CMOS inverter (Q p , Q n ) of the n-type active element Q n The n-type channel is formed near the surface of the rectangular region sandwiched between the first main electrode region 15a and the second main electrode region 15b. p , Q n ) of the p-type active element Q p The p-type channel is formed near the surface of the rectangular region sandwiched between the third main electrode region 15d and the fourth main electrode region 15c. n Inside the active region of the CMOS inverter (Q p , Qn The height of the potential barrier to carrier movement generated in the n-type channel defined between the first main electrode region 15a and the second main electrode region 15b is electrostatically controlled by the potential of the potential transfer region of the vibrating body 22a1. p Inside the active region of the CMOS inverter (Q p , Q n The height of the potential barrier to carrier movement generated in the p-type channel defined between the third main electrode region 15d and the fourth main electrode region 15c is determined by the potential of the potential transfer region of the vibrating body 22a1 and the potential of the n-type active element Q n Complementary control.

[0363] like Fig. 21B As shown, an interlayer insulating film (first interlayer insulating film) 17 is provided on the first main electrode region 15a and the second main electrode region 15b. The first main electrode region 15a is connected to the first main electrode wiring 25a, and the second main electrode region 15b is connected to the output signal wiring 25c through a contact hole opened in the first interlayer insulating film 17. In addition, the third main electrode region 15d is connected to the third main electrode wiring 25d, and the fourth main electrode region 15c is connected to the output signal wiring 25c through a contact hole opened in the first interlayer insulating film 17. The first main electrode wiring 25a, the output signal wiring 25c, the third main electrode wiring 25d, and the n-type active element Q n The first control electrode 37n and the p-type active element Q p A second interlayer insulating film 53 is deposited on each of the second control electrodes 37p. Fig. 21B As shown, the first gate contact hole exposing the top of the first control electrode 37n and the second gate contact hole exposing the top of the second control electrode 37p are respectively opened through the second interlayer insulating film 53 and the first interlayer insulating film 17 thereunder. The first gate connection plug 38n is embedded in the first gate contact hole, and the second gate connection plug 38p is embedded in the second gate contact hole, respectively realizing ohmic contact.

[0364] like Fig. 21B As shown, a first input signal wiring 39 is provided on the second interlayer insulating film 53 so as to electrically connect the first gate connection plug 38n and the second gate connection plug 38p. Fig.21A As shown in the plan view of FIG. 1 , the second input signal wiring 45 is arranged in a direction orthogonal to the longitudinal direction of the first input signal wiring 39. Fig. 21B As can be seen from the cross-sectional view, the second input signal wiring 45 is a wiring provided on and in contact with the first input signal wiring 39. n The first control electrode 37n and the p-type active element Qp The second control electrodes 37p are electrically connected to each other, so that Fig.21A As shown, the second input signal wiring 45 connects the first control electrode 37n and the second control electrode 37p to P + type anode region 42 and an n-type cathode region 43 .

[0365] Corresponds to Fig. 20 The p-side clamping diode D in the equivalent circuit representation kp , by P + The pn junction diode composed of the n-type anode region 42 and the n-type cathode region 41 is arranged at Fig.21A The paper area above the patterns of the first main electrode wiring 25a, the second input signal wiring 45 and the third main electrode wiring 25d. + The longitudinal direction of the pn junction diode formed by the n-type anode region 42 and the n-type cathode region 41 is parallel to the longitudinal direction of the patterns of the first main electrode wiring 25 a and the output signal wiring 25 c .

[0366] Likewise, corresponding to Fig. 20 The n-side clamping diode D kn ,exist Fig.21A The lower area of ​​the paper is arranged with P + The pn junction diode formed by the n-type anode region 44 and the n-type cathode region 43 has its longitudinal direction parallel to the pattern of the first main electrode wiring 25a, the second input signal wiring 45 and the third main electrode wiring 25d. kn The anode region 44 is the first power supply side diode wiring 36, connected to the p-side clamping diode D kp The cathode region 41 is the second power supply side diode wiring 35. Fig. 21B As shown, a connection plug 29 having a multilayer structure conductor with a soft metal surface layer is embedded from the tip side of the central projection on the output surface side of the vibrator 22a1, and its flat surface serves as a connection end surface. In addition, the lower surface (connection end surface) of the connection plug 29 is in physical contact with the upper surface of the second input signal wiring 45, thereby connecting the central area P of the voltage generating unit 1a. 0 The potential of the impedance conversion element 2c is input as a control voltage to the CMOS inverter (Q p , Q n ). If the connection plug 29 is regarded as a component, the voltage generating unit 1 a includes the vibrating body 22 a 1 , the fixed potential electrode 24 a , the fixed potential electrode protection film 23 , and the connection plug 29 .

[0367] That is, the potential of the connection plug 29 is input to the first input signal wiring 39 through the second input signal wiring 45, and the potential of the first input signal wiring 39 is transmitted to the first gate connection plug 38n and the second gate connection plug 38p. The potential of the first gate connection plug 38n is input to the first control electrode 37n, and the potential of the first control electrode 37n electrostatically controls the potential barrier height between the first main electrode region 15a and the second main electrode region 15b through the first gate dielectric film 13n. Similarly, the potential of the second gate connection plug 38p is input to the second control electrode 37p, and the potential of the second control electrode 37p electrostatically controls the potential barrier height between the third main electrode region 15d and the fourth main electrode region 15c through the second gate dielectric film 13p.

[0368] Fig. 21B The structure shown is an example of a structure in which a connection plug 29 provided on the voltage generating unit side chip located on the upper side and a second input signal wiring 45 provided on the impedance conversion element side chip located on the lower side are in physical contact with each other. Therefore, around the vibration cavity 21a1, the direct bonding insulating film 31 with a mirror surface provided on the voltage generating unit side chip and the cavity inner wall protective film 19 with a mirror surface provided on the impedance conversion element side chip are directly bonded by hydrophilic bonding (hydrogen bonding between silanol groups). Compared with the structure connected by external wiring in non-patent document 2, the acoustic element of the third embodiment is directly connected to the first gate connection plug 38n and the second gate connection plug 38p through a thin plate-shaped second input signal wiring 45 and a first input signal wiring 39 to form an electrical path, thereby reducing the floating impedance. In addition, since the voltage generating unit 1a adopts a new structure of a stress-enhanced structure, high sensitivity and broadband are achieved, so according to the acoustic element of the third embodiment, high sensitivity and broadband can be achieved. In particular, by microscopically studying the internal structure of the voltage generating unit 1a, a specific position where the generated voltage is the highest is selected as a potential transfer region and used as an input transmission position to the impedance conversion element 2c, thereby easily achieving high sensitivity.

[0369] - Modification of the third embodiment -

[0370] For convenience, in the acoustic elements of the first to third embodiments, the top (bottom) of the central protrusion of the vibration cavity 21a1 is the central area P 0 However, the potential transmission area for transmitting the potential at a specific position (local) of the voltage generating part 1b is not limited to the center of the vibration cavity 21a1. Fig. 22 As shown, the receiving element X includes a voltage generating unit 1b that generates a non-uniform voltage by piezoelectric effect and an impedance conversion element 2c that operates with the potential at a specific position (local) of the voltage generating unit 1b as a control voltage. i,j.from Fig.23 As can be seen from the cross-sectional view of FIG. 1 , the voltage generating portion 1 b is basically the same as the voltage generating portion 1 a of the acoustic element of the first to third embodiments in that it is a unimorph type structure. Fig. 22 and Fig.23 As shown, the central area P in the center of the vibration cavity 21a1 0 The second peripheral area P is defined by the bottom of the surrounding hexagonal ring edge. 2 The feature of transmitting the potential of the acoustic element 2c to the impedance conversion element 2c is different from the structure of the acoustic element of the first to third embodiments described above.

[0371] The second peripheral area P of the bottom of the hexagonal ring edge is adopted 2 The potential is because the cross-sectional structure of the vibrating body 22e2 that transmits the potential to the impedance conversion element 2c is W-shaped, and the receiving surface of the vibrating body 22e2 (at Fig.23 The fixed potential electrode 24e in contact with the upper main surface (in the middle) also has a W-shaped cross-sectional structure. The vibrating body 22e2 can be made of materials belonging to 20 crystal point groups. Among the 20 materials, HfO belongs to 10 crystal point groups with thermoelectric properties. 2 , AlN, PZT and other materials are suitable for the vibrating body 22e2. Fig.23 The cross-sectional structure shown is W-shaped, but the planar structure of the acoustic element of the third embodiment modification is a hexagonal ring structure having six-fold rotational symmetry about the rotation axis of the connection of the two Vs that appear as a W in the cross-sectional view.

[0372] If used Fig. 10A and Fig. 10B As described above, the voltage generating portion 1b of the unimorph structure having a W-shaped cross section is located in the second peripheral region P at the lower top position of the two V's forming the W shape. 2 The absolute value of the potential |V P2 | Compared to other central areas P 0 , the first peripheral area P 1 , the third peripheral area P 3 , the fourth peripheral area P 4 The absolute value of the potential |V P0 |、|V P1 |、|V P3 |、|V P4 | Big. Although Fig. 10A and Fig. 10B The central area P in the simulation results shown 0 , the first peripheral area P 1 , ..., the fourth peripheral area P 4 is a concentric circle pattern, but even if it is a concentric hexagonal ring shape, there is no essential difference. Fig.23The vibrating body 22e2 shown in the cross section is W-shaped, and the central area P defined by the model at the central convex position 0 It does not occur in the surrounding central area P 0 The second hexagonal peripheral area P 2 than other locations (central area P 0 , the first peripheral area P 1 , the third peripheral area P 3 , the fourth peripheral area P 4 ) a higher voltage. Thus, the output surface (on the opposite side of the receiving surface) of the vibrating body 22e2 is opposite to the receiving surface. Fig.23 The second peripheral area P is a specific position of the lower main surface 2 than other positions, namely the central area P 0 , the first peripheral area P 1 , the third peripheral area P 3 , the fourth peripheral area P 4 The voltage generating portion 1 b having a W-shaped cross-sectional structure exhibits a voltage distribution in which the generated voltage is not uniform for a higher voltage.

[0373] If already in Fig. 10B As shown in the example, the central area P of the voltage generating unit 1b 0 , the first peripheral area P 1 , ..., the fourth peripheral area P 4 The detected potential shows a transient response waveform vibrating in the positive direction and the negative direction. In order to correspond to the input voltage waveform vibrating in the positive direction and the negative direction, the acoustic element of the third embodiment modification is similar to the acoustic element of the third embodiment, such as Fig. 22 As shown, it includes a CMOS inverter (Q p , Q n ) as the impedance conversion element 2c. CMOS inverter (Q p , Q n ) is connected to the second peripheral region P of the voltage generating unit 1b. 2 , so that the CMOS inverter (Q p , Q n ) of the p-type active element Q p and n-type active element Q n The potential barrier height generated in each channel is determined by the second peripheral region P of the voltage generating portion 1b. 2 Potential control.

[0374] CMOS Inverter (Q p , Q n ) includes connecting the first main electrode region 15a to a first power source V SS (=-V DD), connecting the second main electrode region 15b to the n-type active element Q of the output signal wiring 25c n As the first active element. n-type active element Q n The potential barrier in the n-type channel is the active element Q n The main current (electrons) flowing between the first main electrode region 15a and the second main electrode region 15b is prevented from moving. p , Q n ) includes connecting the third main electrode region 15d to a voltage higher than the first power source V SS A second power supply V with a higher potential DD , connecting the fourth main electrode region 15c to the p-type active element Q of the output signal wiring 25c p As the second active element. The p-type active element Q p The potential barrier in the p-type channel is the active element Q p The CMOS inverter (Q) as the impedance conversion element 2c is used to prevent the movement of the main current (holes) flowing between the third main electrode region 15d and the fourth main electrode region 15c. p , Q n ) is connected to the first power supply V SS (=-V DD ) and the second power supply V DD Between, such as Fig. 22 As shown, it can correspond to the input voltage waveform vibrating in the positive direction and the negative direction, such as in Fig. 10B As exemplified in etc.

[0375] CMOS inverter (Q p , Q n ) of the n-type active element Q n The first control electrode 37n is connected to the first input signal wiring 46 through the first gate connection plug 38n. On the other hand, the p-type active element Q p The second control electrode 37p is connected to the first input signal wiring 46 through the second gate connection plug 38p. Then, as shown in FIG. Fig.23 As shown, a hexagonal ring-shaped second input signal wiring 47 is provided on the first input signal wiring 46 .

[0376] The CMOS inverter (Q p , Q n ) will be located at Fig.23 The potential of the second peripheral region P2 of the hexagonal ring-shaped edge portion on the output surface side of the vibration body 22e2, which is located in the middle of the radial direction of the vibration cavity 21e, is transmitted to the n-type active element Q through the hexagonal ring-shaped second input signal wiring 47 and the first input signal wiring 46. nThe first control electrode 37n, the p-type active element Q p That is, the impedance conversion element 2c of the acoustic element of the third embodiment modification uses the pressure of the ultrasonic wave Φ applied to the vibrating body 22e2 of the voltage generating unit 1b as the second peripheral area P 2 The potential of the CMOS inverter (Q p , Q n ) input signal, and the impedance-converted signal is transmitted from the CMOS inverter (Q p , Q n Then, the acoustic element of the third embodiment modification example outputs the CMOS inverter (Q p , Q n ) is used as the output signal of the acoustic element.

[0377] In the acoustic element of the third embodiment modification, as Fig.23 As shown, since a hexagonal annular edge is provided in the radial direction of the output surface side of the vibrator 22e2, in addition to the space of the vibration cavity 21e inside the hexagonal annular edge, an inclined annular vibration cavity 21e space is also provided outside the edge in a centrally symmetrical manner. The inclined annular structure constituting the outer vibration cavity 21e space is close to a right triangle in cross-sectional shape, and its plane pattern is a hexagonal topology. Fig.23 In the cross-sectional view, it appears that the two vibration cavities 21e are located on both sides of the hexagonal annular edge, but in fact they are an integrated structure that is continuous with each other on the front and back of the paper.

[0378] CMOS inverter (Q p , Q n ) of the n-type active element Q n like Fig.23 As shown, there is a well region 52 composed of a p-type semiconductor region, and n-type semiconductor regions are arranged at intervals on the top of the well region 52. + The first main electrode region 15a and the second main electrode region 15b are formed of an n-type semiconductor region. The well region 52 is partially provided on the upper portion of the base region 51 formed of an n-type semiconductor region. In addition, the n-type active element Q n like Fig.23 As shown, the first gate dielectric film 13n, which contacts the upper surface of the well region 52 between the first main electrode region 15a and the second main electrode region 15b and has a wider bandgap than the well region 52, and the first control electrode 37n arranged on the first gate dielectric film 13n is an active element.

[0379] CMOS inverter (Q p , Q n ) of the p-type active element Qp like Fig.23 As shown, it has a base region 51, and P is arranged on the upper part of the base region 51 at intervals. + The third main electrode region 15d and the fourth main electrode region 15c are composed of a p-type semiconductor region. p like Fig.23 As shown, it is an active element having a second gate dielectric film 13p that contacts the upper surface of the substrate region 51 between the third main electrode region 15d and the fourth main electrode region 15c and has a wider bandgap than the substrate region 51, and a second control electrode 37p arranged on the second gate dielectric film 13p.

[0380] like Fig. 22 As shown, in the CMOS inverter (Q p , Q n ) input signal terminal (corresponding to Fig.23 The first input signal wiring 46 and the second input signal wiring 47 in the circuit are connected to the second power supply V DD A p-side clamping diode D is connected between kp , between the input signal terminal and the first power supply V SS (=-V DD ) is connected with an n-side clamping diode D kn By clamping the p-side diode D kp and n-side clamping diode D kn Connected to the CMOS inverter (Q p , Q n ) input signal terminal, so that the n-type active element Q n The first gate dielectric film 13n and the p-type active element Q p An excessive gate voltage is not applied to the second gate dielectric film 13p.

[0381] like Fig. 22 and Fig.23 As shown in FIG. 1 , in the acoustic element of the third embodiment modification, the CMOS inverter (Q p , Q n ) is connected to the potential generating position P 2 Therefore, a CMOS inverter (Q p , Q n ) of the n-type active element Q n The potential barrier height generated in the n-type channel between the first main electrode region 15a and the second main electrode region 15b is determined by the potential generating position P 2 Similarly, the CMOS inverter (Q p , Qn ) of the p-type active element Q p The potential barrier height generated in the p-type channel between the third main electrode region 15d and the fourth main electrode region 15c is also related to the n-type active element Q n Complementarily, the potential generates the position P 2 Potential control.

[0382] Around the well region 52 locally provided on the upper part of the n-type base region 51, an element isolation insulating film 16 is provided in a frame shape for defining the CMOS inverter (Q p , Q n ) of the n-type active element Q n The range of the active region. In defining the n-type active element Q n Adjacent to the frame-shaped bank of the active region pattern, there is also a p-type active element Q defined p That is, the n-type active element Q n The active region and p-type active element Q p The active region of the n-type active element Q is adjacent to each other through the element isolation insulating film 16. n The region on the upper surface of the well region 52 is defined within the active region range, and the first main electrode region 15a and the second main electrode region 15b are provided near the upper surface (surface) of the active region surrounded by the element isolation insulating film 16. p Inside the active region pattern, an area on the upper surface of the substrate region 51 is defined, and a third main electrode region 15d and a fourth main electrode region 15c are provided near the upper surface (surface) of the active region surrounded by the element isolation insulating film 16.

[0383] CMOS inverter (Q p , Q n ) of the n-type active element Q n The n-type channel is formed near the surface of the rectangular region sandwiched between the first main electrode region 15a and the second main electrode region 15b. p , Q n ) of the p-type active element Q p The p-type channel is formed near the surface of the rectangular region sandwiched between the third main electrode region 15d and the fourth main electrode region 15c. nThe height of the potential barrier for carrier movement generated in the n-type channel defined by the first main electrode region 15a and the second main electrode region 15b is electrostatically controlled by the potential of the potential transfer region of the vibrating body 22e2. p The height of the potential barrier for carrier movement generated in the p-type channel defined by the third main electrode region 15d and the fourth main electrode region 15c is the potential of the potential transfer region of the vibrating body 22e2, which is connected to the n-type active element Q n Complementary way to control.

[0384] like Fig.23 As shown, a first interlayer insulating film 17 is provided on the first main electrode region 15a and the second main electrode region 15b. The first main electrode region 15a is connected to the first main electrode wiring 25a, and the second main electrode region 15b is connected to the output signal wiring 25c through a contact hole opened in the first interlayer insulating film 17. In addition, the third main electrode region 15d is connected to the third main electrode wiring 25d, and the fourth main electrode region 15c is connected to the output signal wiring 25c through a contact hole opened in the first interlayer insulating film 17. The first main electrode wiring 25a, the output signal wiring 25c, the third main electrode wiring 25d, and the n-type active element Q n The first control electrode 37n and the p-type active element Q p A second interlayer insulating film 53 is deposited above each of the second control electrodes 37p.

[0385] like Fig.23 As shown, a first gate contact hole exposing the top of the first control electrode 37n and a second gate contact hole exposing the top of the second control electrode 37p are respectively opened through the second interlayer insulating film 53 and the first interlayer insulating film 17 thereunder. The first gate connection plug 38n is embedded in the first gate contact hole, and the second gate connection plug 38p is embedded in the second gate contact hole. The integrated structure connecting the first control electrode 37n and the first gate connection plug 38n may be referred to as a "first gate terminal (37n, 38n)", and the integrated structure connecting the second control electrode 37p and the second gate connection plug 38p may be referred to as a "second gate terminal (37p, 38p)".

[0386] like Fig.23 As shown in FIG. 1 , a first input signal wiring 46 is provided on the second interlayer insulating film 53 so as to electrically connect the first gate connection plug 38n to the second gate connection plug 38p. Although not shown in the plan view, a hexagonal ring-shaped second input signal wiring 47 is provided in a direction intersecting with the longitudinal direction of the first input signal wiring 46. Fig.23As can be seen from the cross-sectional view, the second input signal wiring 47 is a wiring provided on the first input signal wiring 46 and in contact with the first input signal wiring 46 at two locations. The first input signal wiring 46 connects the n-type active element Q n The first control electrode 37n and the p-type active element Q p Therefore, the second input signal wiring 47 connects the first control electrode 37n and the second control electrode 37p to the p-side clamp diode D kp The anode region and the n-side clamping diode D kn It is also possible to refer to the integrated structure connecting the first input signal wiring 46 and the second input signal wiring 47 as "input signal wiring (46, 47)". In this representation, the input signal wiring (46, 47) is connected to the first gate terminal (37n, 38n) and the second gate terminal (37p, 38p).

[0387] Connect to n-side clamping diode D kn The anode region of the first power supply side diode wiring is connected to the p-side clamping diode D kp The cathode region is the second power supply side diode wiring. Fig.23 As shown in FIG. 1 , a connection pad 48 made of a conductor is provided at the bottom of the hexagonal ring-shaped edge portion on the output surface side of the vibrating body 22e2 to achieve ohmic contact. In addition, the lower surface (connection end surface) of the hexagonal ring-shaped connection pad 48 and the upper surface of the hexagonal ring-shaped second input signal wiring 47 are in physical contact with each other, thereby connecting the second peripheral area P of the voltage generating unit 1b. 2 The potential of the impedance conversion element 2c is input as a control voltage to the CMOS inverter (Q p , Q n If the connection pad 48 is regarded as a component, the voltage generating unit 1b includes the vibrating body 22e2, the fixed potential electrode 24e, the fixed potential electrode protection film 23 and the connection pad 48. A connection means such as a bump may be further interposed between the connection pad 48 and the second input signal wiring 47.

[0388] That is, the potential of the hexagonal ring-shaped connection pad 48 is input to the first input signal wiring 46 through the hexagonal ring-shaped second input signal wiring 47, and the potential of the first input signal wiring 46 is transmitted to the first gate connection plug 38n and the second gate connection plug 38p. The potential of the first gate connection plug 38n is input to the first control electrode 37n, and the potential of the first control electrode 37n electrostatically controls the potential barrier height between the first main electrode region 15a and the second main electrode region 15b through the first gate dielectric film 13n. Similarly, the potential of the second gate connection plug 38p is input to the second control electrode 37p, and the potential of the second control electrode 37p electrostatically controls the potential barrier height between the third main electrode region 15d and the fourth main electrode region 15c through the second gate dielectric film 13p.

[0389] Fig.23 The struc...

Claims

1. An acoustic element, characterized in that: include: A voltage generating unit having a vibrating body and a fixed potential electrode in contact with a receiving surface of the vibrating body, wherein at least a piezoelectric layer is provided at a portion of the vibrating body and located at a stress concentration portion of the vibrating body on an output surface side of the vibrating body opposite to the receiving surface; The impedance conversion element uses a specific part corresponding to the stress concentration part of the piezoelectric layer as a potential transfer area and uses the potential of the potential transfer area as a control voltage to operate; The pressure of the ultrasonic wave applied to the vibrating body is used as an output signal of the impedance conversion element.

2. The acoustic element according to claim 1, characterized in that The impedance transformation element comprises: A substrate region, composed of a semiconductor region of a first conductivity type; The first and second main electrode regions are formed by semiconductor regions of the second conductivity type arranged on the upper part of the substrate region and spaced from each other; The potential transfer region is bonded to the upper surface of the base region between the first main electrode region and the second main electrode region, and the potential of the potential transfer region is used to control the height of the potential barrier generated in the channel between the first main electrode region and the second main electrode region.

3. The acoustic element according to claim 1, characterized in that The impedance transformation element comprises: A substrate region, composed of a semiconductor region of a first conductivity type; The first and second main electrode regions are formed by semiconductor regions of the second conductivity type arranged on the upper part of the substrate region and spaced from each other; a gate dielectric film, contacting an upper surface of the substrate region between the first main electrode region and the second main electrode region, and having a wider bandgap than the substrate region; The control electrode is provided on the gate dielectric film and connected to the potential transfer region, and controls the height of the potential barrier generated in the channel between the first main electrode region and the second main electrode region by using the potential of the potential transfer region.

4. The acoustic element according to claim 1, characterized in that The impedance transformation element comprises: A substrate region, composed of a semiconductor region of a first conductivity type; The first and second main electrode regions are formed by semiconductor regions of the first conductivity type which are arranged in the substrate region and spaced apart from each other and have an impurity density higher than that of the substrate region; a surface control electrode region, which is formed by a semiconductor region of the second conductivity type disposed on an upper portion of the substrate region between the first main electrode region and the second main electrode region and forming a pn junction with the substrate region; an auxiliary control electrode region, which is composed of a semiconductor region of a second conductivity type arranged below the surface control electrode region and opposite to the substrate region, and defines a channel region between the first main electrode region and the second main electrode region sandwiched therebetween; The control electrode is metallographically bonded to the surface control electrode region in a manner of being connected to the potential transfer region, and the potential of the potential transfer region is used to control the height of the potential barrier generated in the channel.

5. The acoustic element according to claim 1, characterized in that The impedance conversion element is a CMOS inverter, and the input electrode of the CMOS inverter is connected to the potential transfer region, thereby constituting the potential barrier height generated in each channel of the p-type active element and the n-type active element of the CMOS inverter, which is controlled by the potential of the potential transfer region.

6. The acoustic element according to claim 1, characterized in that The impedance conversion element is a semiconductor integrated circuit, and the potentials of specific multiple potential transfer regions of the voltage generating unit are independently transmitted as independent control voltages of multiple active elements included in the semiconductor integrated circuit, and the multiple active elements are independently controlled.

Images