Filter device

Abstract

Provided is a filter device comprising a plurality of resonators including an elastic wave resonator 1. The elastic wave resonator 1 includes: a piezoelectric layer 5 that has a first main surface 5a and a second main surface 5b on opposite sides from each other; an IDT electrode 8 that is provided on the first main surface 5a of the piezoelectric layer 5 and has a plurality of electrode fingers; a first dielectric film 6 that is provided on the first main surface 5a of the piezoelectric layer 5; and a second dielectric film 7 that is provided on the second main surface 5b of the piezoelectric layer 5. If the specific bandwidth is 8.5% or greater, the thickness of the crossing region of the piezoelectric layer 5 is denoted by Tp, the thickness of the portion of the first dielectric film 6 that overlaps the crossing region when viewed in a plan view is denoted by Td_f, and the thickness of the portion of the second dielectric film 7 that overlaps the crossing region when viewed in a plan view is denoted by Td_b, then 0.322 < (Td_f / Tp) + (Td_b / Tp) < 0.786 and |(Td_f / Tp) - (Td_b / Tp)| < 0.196 are satisfied.

Description

Filter Device

[0001] The present invention relates to a filter device having an acoustic wave resonator.

[0002] Conventionally, filter devices having acoustic wave resonators have been widely used as filters for mobile phones, etc. Recently, an acoustic wave resonator using bulk waves in thickness shear mode has been proposed, as described in Patent Document 1 below. In this acoustic wave resonator, a piezoelectric layer is provided on a support. An IDT (Interdigital Transducer) is provided on the piezoelectric layer. Electrode fingers are arranged alternately in the IDT. By applying an AC voltage between the electrode fingers, bulk waves in thickness shear mode are excited.

[0003] In Patent Document 1, the surface of the piezoelectric layer on which the IDT is provided is referred to as the front surface. The surface of the piezoelectric layer opposite to the front surface is referred to as the back surface. Frequency setting layers, which are dielectric layers, are provided on the front and back surfaces. Patent Document 1 describes that when the ratio of the thickness of the frequency setting layer provided on the back surface to the total thickness of the frequency setting layers is at least 30%, the A2 mode occurring on the higher frequency side than the passband is reduced by 20% or more. The A2 mode refers to a second-order mode in the thickness direction.

[0004] US Patent Application Publication No. 2021 / 044272

[0005] A ladder-type circuit configuration, for example, may be adopted for a filter device having a wide passband for use in 5G, etc. The series arm resonators and parallel arm resonators in the circuit configuration may be acoustic wave resonators that utilize thickness-shear mode bulk waves.

[0006] Here, a typical RF (Radio Frequency) filter requires high attenuation characteristics in a frequency band near an integer multiple of the passband frequency. However, an elastic wave resonator using thickness-shear bulk waves generates not only a second-order mode but also higher-order modes, such as a third-order, fourth-order, or fifth-order mode. These higher-order modes occur at frequencies higher than the second-order mode. When a third-order mode occurs in a parallel-arm resonator in a ladder-type circuit configuration, the response due to the third-order mode is located at a frequency near twice the passband frequency. Therefore, the required attenuation characteristics may not be satisfied at a frequency near twice the passband frequency.

[0007] An object of the present invention is to provide a filter device that can widen the passband and suppress the second and third modes in an acoustic wave resonator.

[0008] A filter device according to the present invention is a filter device having a passband, and includes a plurality of resonators including an elastic wave resonator, wherein the elastic wave resonator includes a piezoelectric layer having a first main surface and a second main surface facing each other, an IDT electrode provided on the first main surface of the piezoelectric layer and having a plurality of electrode fingers, a first dielectric film provided on the first main surface of the piezoelectric layer, and a second dielectric film provided on the second main surface of the piezoelectric layer, wherein a direction orthogonal to a direction in which the plurality of electrode fingers extend is defined as an electrode finger orthogonal direction, and a region where adjacent electrode fingers overlap when viewed from the electrode finger orthogonal direction is a crossing region, and the first dielectric film and the second dielectric film are configured such that, in a plan view, The passband overlaps with a crossover region, and when the highest frequency in the passband is FH, the lowest frequency in the passband is FL, and the center frequency of the passband is FC, a relative bandwidth expressed by ((FH-FL) / FC) x 100[%] is 8.5% or more, and when the thickness of the piezoelectric layer at the crossover region is Tp, the thickness of the portion of the first dielectric film that overlaps with the crossover region in a planar view is Td_f, and the thickness of the portion of the second dielectric film that overlaps with the crossover region in a planar view is Td_b, the following relationships are satisfied: 0.322<(Td_f / Tp)+(Td_b / Tp)<0.786, and |(Td_f / Tp)-(Td_b / Tp)|<0.196.

[0009] According to the filter device of the present invention, the passband can be widened and the second and third modes in the acoustic wave resonator can be suppressed.

[0010] FIG. 1 is a circuit diagram of a filter device according to a first embodiment of the present invention. FIG. 2 is a schematic front cross-sectional view of an elastic wave resonator according to the first embodiment of the present invention. FIG. 3 is a schematic plan view of the elastic wave resonator according to the first embodiment of the present invention. FIG. 4 is a schematic front cross-sectional view showing the vicinity of a pair of electrode fingers of the elastic wave resonator according to the first embodiment of the present invention. FIG. 5 is a diagram showing the impedance-frequency characteristics of an elastic wave resonator according to a comparative example. FIG. 6 is a diagram showing the impedance-frequency characteristics of the elastic wave resonator according to the first embodiment of the present invention. FIG. 7 is a diagram showing the thickness direction distribution of displacement due to first-order mode at a center portion between electrode fingers of a laminated section according to the first embodiment of the present invention. FIG. 8 is a diagram showing the thickness direction distribution of displacement due to second-order mode at a center portion between electrode fingers of a laminated section according to the first embodiment of the present invention. FIG. 9 is a diagram showing the thickness direction distribution of displacement due to third-order mode at a center portion between electrode fingers of a laminated section according to the first embodiment of the present invention. Fig. 10 is a diagram schematically showing the thickness direction distribution of stress due to first-order mode at the center between the electrode fingers of the laminated portion in the first embodiment of the present invention. Fig. 11 is a diagram schematically showing the thickness direction distribution of stress due to second-order mode at the center between the electrode fingers of the laminated portion in the first embodiment of the present invention. Fig. 12 is a diagram schematically showing the thickness direction distribution of stress due to third-order mode at the center between the electrode fingers of the laminated portion in the first embodiment of the present invention. Fig. 13 is a diagram schematically showing the thickness direction distribution of stress due to third-order mode at the center between the electrode fingers of the laminated portion in the case where the third-order mode is suppressed. Fig. 14 is a diagram schematically showing the thickness direction distribution of stress due to third-order mode at the center between the electrode fingers of the laminated portion in the case where the third-order mode is suppressed in a reference example. Fig. 15 is a diagram showing the relationship between the ratio Td_f / Tp and the electromechanical coupling coefficient K of the third mode when Td_f = Td_b. ２ ｅｆｆ 16 is a graph showing the relationship between the ratio Td_f / Tp and the electromechanical coupling coefficient K of the second mode when (Td_f / Tp+Td_b / Tp)=0.54. ２ ｅｆｆFIG. 17 is a diagram showing the relationship between the ratio Td_f / Tp and the ratio Fr3 / Fr1. FIG. 18 is a schematic front cross-sectional view showing a portion of a filter device according to a first modified example of the first embodiment of the present invention. FIG. 19 is a schematic front cross-sectional view showing a portion of a filter device according to a second modified example of the first embodiment of the present invention. FIG. 20 is a schematic front cross-sectional view showing a portion of a filter device according to a third modified example of the first embodiment of the present invention. FIG. 21 is a schematic front cross-sectional view showing a pair of electrode fingers and their vicinity in an elastic wave resonator according to a fourth modified example of the first embodiment of the present invention. FIG. 22 is a schematic front cross-sectional view showing a pair of electrode fingers and their vicinity in an elastic wave resonator according to a second preferred embodiment of the present invention. FIG. 23 is a schematic front cross-sectional view showing a pair of electrode fingers and their vicinity in an elastic wave resonator according to a third preferred embodiment of the present invention. FIG. 24 is a schematic plan view of an elastic wave resonator according to a fourth preferred embodiment of the present invention. FIG. 25 is a schematic front cross-sectional view showing a pair of electrode fingers and their vicinity in an elastic wave resonator according to a fifth preferred embodiment of the present invention. Fig. 26 is a diagram showing the relationship between Tp / p and the fractional bandwidth of an elastic wave resonator. Fig. 27 is a diagram showing the relationship between the fractional bandwidth of an elastic wave resonator and the magnitude of normalized spurious. Fig. 28 is a diagram showing the relationship between Tp / p, the metallization ratio MR, and the fractional bandwidth. Fig. 29 is a diagram showing the relationship between Tp / p and the fractional bandwidth of a LiNbO resonator when Tp / p approaches 0. ３FIG. 30 is a diagram showing a map of the fractional bandwidth with respect to the Euler angles (0°, θ, ψ). FIG. 30 is a schematic diagram of a coordinate system in which the crystal axes of a piezoelectric crystal are transformed by the Euler angles. FIG. 31 is a schematic diagram showing a coordinate system of a piezoelectric layer in which the angle θ at the Euler angles (0°, θ, 0°) of lithium niobate constituting the piezoelectric layer is 0° or more and 60° or less. FIG. 32 is a schematic diagram showing a coordinate system of a piezoelectric layer in which the angle θ at the Euler angles (0°, θ, 90°) of lithium niobate constituting the piezoelectric layer is -15° or more and less than 0°. FIG. 33 is a schematic diagram showing a coordinate system of a piezoelectric layer in which the angle θ at the Euler angles (0°, θ, 0°) of lithium niobate constituting the piezoelectric layer is 180° or more and 240° or less. Fig. 34 is a schematic diagram showing a coordinate system of a piezoelectric layer when the angle θ in the Euler angles (0°, θ, 90°) of lithium niobate constituting the piezoelectric layer is 165° or more and 180° or less. Fig. 35 is a piezoelectric response microscope image showing an example in which polarization inversion has occurred in the piezoelectric layer. Fig. 36 is a diagram showing the admittance frequency characteristics of an elastic wave resonator when polarization inversion has occurred and when it has not occurred in the piezoelectric layer. Fig. 37 is a diagram showing an example of the pass characteristic of a filter device.

[0011] The present invention will be clarified below by describing specific embodiments of the present invention with reference to the drawings.

[0012] It should be noted that the embodiments described in this specification are merely examples, and partial substitution or combination of configurations is possible between different embodiments.

[0013] FIG. 1 is a circuit diagram of a filter device according to a first embodiment of the present invention.

[0014] The filter device 10 is a bandpass filter having a passband. In this specification, the passband refers to the passband guaranteed by the filter device. A specific method for specifying the passband of a filter device will be described later. The filter device 10 shown in FIG. 1 is widely used in communication devices and the like. The passband of the filter device 10 is the n79 passband. The n79 passband is 4400 MHz to 5000 MHz. However, the passband of the filter device 10 is not limited to the above.

[0015] Specifically, the filter device 10 is a ladder-type filter. The filter device 10 has a plurality of series arm resonators and a plurality of parallel arm resonators as a plurality of resonators. The circuit configuration of the filter device 10 is not limited to the circuit configuration shown in FIG. 1 . The filter device 10 may include any suitable ladder-type circuit unit. The ladder-type circuit unit may include at least one series arm resonator and at least one parallel arm resonator.

[0016] In this specification, the highest frequency in a passband is referred to as the upper end, and the lowest frequency in the passband is referred to as the lower end. In this specification, the upper and lower ends of the passband refer to the upper and lower ends of the passband guaranteed by the filter device. When the upper end of the passband is FH, the lower end of the passband is FL, and the center frequency of the passband is FC, FC = (FH + FL) / 2. In the filter device 10 of this embodiment, the fractional bandwidth, expressed by ((FH - FL) / FC) x 100 [%], is 8.5% or greater. Thus, the passband of the filter device 10 is wide. Meanwhile, the upper limit of the fractional bandwidth is not particularly limited, but the fractional bandwidth may be, for example, 40% or less. In this case, the filter device 10 can be easily manufactured.

[0017] Each of the multiple resonators in the filter device 10 is an elastic wave resonator. However, this is not a limitation. The multiple resonators need only include at least one elastic wave resonator. As shown in FIG. 1 , the multiple resonators include an elastic wave resonator 1. In this embodiment, the elastic wave resonator 1 is a parallel arm resonator. A specific configuration of the elastic wave resonator 1 will be described below.

[0018] FIG. 2 is a schematic front cross-sectional view of the acoustic wave resonator according to the first preferred embodiment.

[0019] The acoustic wave resonator 1 has a piezoelectric layer 5, a first dielectric film 6, a second dielectric film 7, and an IDT electrode 8. The piezoelectric layer 5 has a first principal surface 5a and a second principal surface 5b. The first principal surface 5a and the second principal surface 5b face each other. The first dielectric film 6 is provided on the first principal surface 5a. More specifically, the IDT electrode 8 is provided on the first principal surface 5a. The first dielectric film 6 is provided on the first principal surface 5a so as to cover the IDT electrode 8. Meanwhile, the second dielectric film 7 is provided on the second principal surface 5b.

[0020] In this embodiment, the piezoelectric layer 5 is made of lithium niobate. More specifically, the piezoelectric layer 5 is made of LiNbO ３ The first dielectric film 6 and the second dielectric film 7 are made of silicon oxide. More specifically, the first dielectric film 6 and the second dielectric film 7 are made of SiO ２ In this specification, the expression "a certain component is made of a certain material" includes cases where a small amount of impurities are contained to the extent that the electrical characteristics of the elastic wave resonator are not significantly deteriorated. Note that the materials of the piezoelectric layer 5, the first dielectric film 6, and the second dielectric film 7 are not limited to those mentioned above.

[0021] The elastic wave resonator 1 has a support member 2. The support member 2 includes a support substrate 3 and an insulating layer 4. The insulating layer 4 is provided on the support substrate 3. A second dielectric film 7 is provided on the insulating layer 4. Thus, the piezoelectric layer 5 is indirectly provided on the insulating layer 4 via the second dielectric film 7.

[0022] The material of the support substrate 3 may be, for example, a semiconductor such as silicon or a ceramic such as aluminum oxide. The material of the insulating layer 4 may be an appropriate dielectric such as silicon oxide or tantalum oxide. The support member 2 may be composed of only the support substrate 3. Alternatively, the support member 2 may not necessarily be provided.

[0023] Fig. 3 is a schematic plan view of an elastic wave resonator according to the first embodiment. Fig. 2 above is a schematic cross-sectional view taken along line II in Fig. 3. Note that wiring connected to elastic wave resonator 1 and first dielectric film 6 are omitted in Fig. 3. The same applies to the other schematic plan views.

[0024] The IDT electrode 8 has a pair of bus bars and a plurality of electrode fingers. The pair of bus bars is specifically a first bus bar 16 and a second bus bar 17. The first bus bar 16 and the second bus bar 17 face each other. The plurality of electrode fingers is specifically a plurality of first electrode fingers 18 and a plurality of second electrode fingers 19. One ends of the plurality of first electrode fingers 18 are connected to the first bus bar 16, respectively. One ends of the plurality of second electrode fingers 19 are connected to the second bus bar 17, respectively. The plurality of first electrode fingers 18 and the plurality of second electrode fingers 19 are interdigitated with each other.

[0025] The IDT electrode 8 is made of a laminated metal film. Specifically, in the IDT electrode 8, a Ti layer and an Al layer are laminated in this order from the piezoelectric layer 5 side. Note that the material of the IDT electrode 8 is not limited to the above. Alternatively, the IDT electrode 8 may be made of a single-layer metal film.

[0026] Hereinafter, the first electrode fingers 18 and the second electrode fingers 19 may be collectively referred to simply as electrode fingers. The direction in which the multiple electrode fingers extend is referred to as the electrode finger extension direction, and the direction perpendicular to the electrode finger extension direction is referred to as the electrode finger orthogonal direction. When viewed from the electrode finger orthogonal direction, the region where adjacent first electrode fingers 18 and second electrode fingers 19 overlap is the intersection region A. The first dielectric film 6 and the second dielectric film 7 only need to overlap the intersection region A in a planar view.

[0027] In this specification, a plan view refers to a view along the stacking direction of the support member 2 and the piezoelectric layer 5 from a direction corresponding to the top in Fig. 2. Note that in Fig. 2, for example, of the support substrate 3 side and the piezoelectric layer 5 side, the piezoelectric layer 5 side is the top. Furthermore, in this specification, a plan view is synonymous with a view from the principal surface opposing direction. The principal surface opposing direction is the direction in which the first principal surface 5a and the second principal surface 5b of the piezoelectric layer 5 face each other. More specifically, the principal surface opposing direction is, for example, the normal direction of the first principal surface 5a.

[0028] The elastic wave resonator 1 is configured to be able to utilize thickness-shear mode bulk waves. Specifically, the elastic wave resonator 1 of this embodiment is configured to be able to utilize thickness-shear mode primary bulk waves as the principal mode. More specifically, in the elastic wave resonator 1, when the thickness of the piezoelectric layer 5 in the crossing region A is Tp and the center-to-center distance between the adjacent first electrode fingers 18 and second electrode fingers 19 is p, Tp / p is 0.5 or less. This allows the thickness-shear mode bulk waves to be suitably excited.

[0029] The excitation region C is a region where adjacent first electrode fingers 18 and second electrode fingers 19 overlap in the orthogonal direction and is a region between the centers of the adjacent first electrode fingers 18 and second electrode fingers 19. In other words, the intersection region A includes multiple excitation regions C. The intersection region A and the excitation region C are regions of the piezoelectric layer 5 defined based on the configuration of the IDT electrode 8. In each excitation region C, a thickness-shear mode bulk wave is excited. However, the acoustic wave resonator 1 may be configured to be able to use a plate wave. In this case, the excitation region is the intersection region A.

[0030] In this embodiment, the cut angle of the piezoelectric layer 5 is ZY. ZY means that the Z axis of the crystal constituting the piezoelectric layer 5 is perpendicular to the first main surface 5a of the piezoelectric layer 5, and the Y axis of the crystal is parallel to the direction perpendicular to the electrode fingers. In this embodiment, the crystal constituting the piezoelectric layer 5 is LiNbO ３ However, the material and cut angle of the piezoelectric layer 5 are not limited to those described above.

[0031] FIG. 4 is a schematic front cross-sectional view showing a pair of electrode fingers and their vicinity in the acoustic wave resonator according to the first preferred embodiment.

[0032] In the following, the thickness of the portion of the first dielectric film 6 that overlaps with the crossing region A in a planar view is defined as Td_f. Specifically, the thickness Td_f is the thickness of the portion of the first dielectric film 6 that is located between the electrode fingers. The thickness of the portion of the second dielectric film 7 that overlaps with the crossing region A in a planar view is defined as Td_b.

[0033] This embodiment is characterized by a fractional bandwidth of 8.5% or more, 0.322 < (Td_f / Tp) + (Td_b / Tp) < 0.786, and |(Td_f / Tp) - (Td_b / Tp)| < 0.196. When a filter device is used in a communication device, if the filter device has a wide passband, higher-order modes such as second-order and third-order modes tend to degrade communication quality. In contrast, in this embodiment, the filter device 10 has the above-described configuration, which enables the passband to be widened and the second-order and third-order modes to be suppressed in the acoustic wave resonator 1. The effect of suppressing second-order and third-order modes will be described in detail by comparing this embodiment with a comparative example. Then, it will be shown that problems due to higher-order modes are likely to occur when the fractional bandwidth is 8.5% or more.

[0034] The comparative example differs from the first embodiment in that 0.786<(Td_f / Tp)+(Td_b / Tp). The impedance frequency characteristics of the first embodiment and the comparative example were derived by FEM (Finite Element Method) simulation. The design parameters of the elastic wave resonator 1 having the configuration of the first embodiment are as follows. The electrode finger pitch in the design parameters is the center-to-center distance between adjacent first electrode fingers 18 and second electrode fingers 19 in the direction perpendicular to the electrode fingers. The electrode finger width is the dimension of the electrode fingers in the direction perpendicular to the electrode fingers.

[0035] Piezoelectric layer: Material... LiNbO ３ , Cut angle: ZY, Thickness Tp: 356 nm First dielectric film: Material: SiO２ , thickness Td_f...89 nm Second dielectric film: material...SiO ２ , thickness Td_b...89 nm IDT electrode: layer structure...Ti layer / Al layer from the piezoelectric layer side, thickness...10 nm / 100 nm from the piezoelectric layer side Electrode finger pitch: 4 μm Electrode finger width: 0.8 μm Td_f / Tp: 0.25 Td_b / Tp: 0.25 (Td_f / Tp)+(Td_b / Tp): 0.5 |(Td_f / Tp)-(Td_b / Tp)|: 0

[0036] The design parameters of the acoustic wave resonator of the comparative example are as follows:

[0037] Piezoelectric layer: Material... LiNbO ３ , Cut angle: ZY, Thickness Tp: 294 nm First dielectric film: Material: SiO ２ , thickness Td_f... 132 nm Second dielectric film: material... SiO ２ , thickness Td_b...132 nm IDT electrode: layer structure...Ti layer / Al layer from the piezoelectric layer side, thickness...10 nm / 100 nm from the piezoelectric layer side Electrode finger pitch: 4 μm Electrode finger width: 0.8 μm Td_f / Tp: 0.45 Td_b / Tp: 0.45 (Td_f / Tp)+(Td_b / Tp): 0.9 |(Td_f / Tp)-(Td_b / Tp)|: 0

[0038] The resonant frequency of the elastic wave resonator 1 according to the first embodiment for comparison is 4200 MHz. Therefore, the elastic wave resonator 1 can be suitably used as a parallel arm resonator in a filter device 10 having a pass band of n79. The resonant frequency of the elastic wave resonator according to the comparative example is also 4200 MHz. The pass band of n79 is 4400 MHz to 5000 MHz.

[0039] Fig. 5 is a diagram showing the impedance-frequency characteristics of an elastic wave resonator according to a comparative example. Fig. 6 is a diagram showing the impedance-frequency characteristics of an elastic wave resonator according to the first preferred embodiment. Figs. 5 and 6 show the pass band n79. Figs. 5 and 6 also show frequency band B of 8800 MHz to 10000 MHz, which is twice the frequency of the pass band. Arrow D2 in Figs. 5 and 6 indicates the vicinity of the frequency at which the second-order mode occurs. Arrow D3 indicates the vicinity of the frequency at which the third-order mode occurs.

[0040] As shown in Figure 5, in the comparative example, the second-order mode occurring near 7000 MHz is suppressed. On the other hand, a large spurious response due to the third-order mode occurs near 9500 MHz. The spurious response occurs near the center frequency of frequency band B, which is twice the frequency in the pass band of n79. Therefore, when the acoustic wave resonator in the comparative example is used as a parallel arm resonator in a filter device having a pass band of n79, the attenuation characteristics deteriorate near frequency band B. This causes a deterioration in the quality of communication in the filter device.

[0041] In contrast, as shown in Fig. 6, in the first embodiment, not only the second-order mode but also the third-order mode can be suppressed. The acoustic wave resonator 1 of the first embodiment is used as a parallel arm resonator in the filter device 10 having a passband of n79. This improves the attenuation characteristics near frequency band B. Therefore, when the filter device 10 is used in a communication device, the quality of communication is less likely to deteriorate.

[0042] As described above, the acoustic wave resonator 1 can suppress the second and third modes. The reason for this is explained below. As shown in FIG. 2 , the acoustic wave resonator 1 has a laminated portion 9. Specifically, the laminated portion 9 is a portion where the piezoelectric layer 5, the first dielectric film 6, and the second dielectric film 7 are laminated. In the first embodiment, the laminated portion 9 is located in a portion that does not overlap with the IDT electrode 8 in a plan view. The laminated portion 9 includes a plurality of inter-electrode finger central portions 9 a. Specifically, the inter-electrode finger central portions 9 a are portions of the laminated portion 9 that are located in the center between the electrode fingers. FIGS. 7 to 12 show instantaneous displacement and stress distributions for each mode in the inter-electrode finger central portions 9 a of the laminated portion 9.

[0043] Fig. 7 is a diagram schematically showing the thickness direction distribution of displacement due to first-order mode at the center between the electrode fingers of the laminated section in the first embodiment. Fig. 8 is a diagram schematically showing the thickness direction distribution of displacement due to second-order mode at the center between the electrode fingers of the laminated section in the first embodiment. Fig. 9 is a diagram schematically showing the thickness direction distribution of displacement due to third-order mode at the center between the electrode fingers of the laminated section in the first embodiment.

[0044] 7 to 9, the horizontal axis indicates the magnitude of displacement, and the vertical axis indicates the position in the thickness direction of the laminated portion 9. For convenience, the sign of displacement in one of two opposing directions perpendicular to the thickness direction of the laminated portion 9 is represented as +, and the sign of displacement in the other direction is represented as -. Specifically, the right side in FIGS. 7 to 9 is represented as +, and the left side is represented as -. The upper end of the vertical axis in FIGS. 7 to 9 indicates the surface of the first dielectric film 6. The lower end of the vertical axis in FIGS. 7 to 9 indicates the surface of the second dielectric film 7. The upper and lower ends of the vertical axis are the same in FIGS. 10 to 14, which will be described later.

[0045] As shown in FIG. 7 , the first mode is a wave whose displacement direction is perpendicular to the thickness direction of the laminated portion 9. Note that the first mode in the first embodiment is a thickness-shear first mode. Similarly, as shown in FIGS. 8 and 9 , the second and third modes are also waves whose displacement direction is perpendicular to the thickness direction of the laminated portion 9. In other words, the first, second, and third modes in the first embodiment are thickness-direction modes. The higher the mode order, the shorter the wavelength. Specifically, in the first embodiment, half the wavelength of the first mode corresponds to the entire thickness of the laminated portion 9. One wavelength of the second mode corresponds to the entire thickness of the laminated portion 9. 1.5 wavelengths of the third mode corresponds to the entire thickness of the laminated portion 9.

[0046] FIG. 10 is a diagram schematically illustrating the thickness direction distribution of stress due to first-order mode at the center between the electrode fingers of the laminated portion in the first embodiment. FIG. 11 is a diagram schematically illustrating the thickness direction distribution of stress due to second-order mode at the center between the electrode fingers of the laminated portion in the first embodiment. FIG. 12 is a diagram schematically illustrating the thickness direction distribution of stress due to third-order mode at the center between the electrode fingers of the laminated portion in the first embodiment. In FIGS. 10 to 12 , the horizontal axis represents the magnitude of stress, and the vertical axis represents the position in the thickness direction of the laminated portion 9. For convenience, among the stresses that generate thickness shear mode within the laminated portion 9, the sign of the stress that causes one deformation is represented as +, and the sign of the stress that causes the other deformation is represented as −. Specifically, the right side in FIGS. 10 to 12 is represented as +, and the left side is represented as −. The same applies to FIGS. 13 and 14 described below.

[0047] As shown in Figures 10 to 12, the distribution in the thickness direction due to each mode is expressed as a waveform. The stress waveform due to the first mode corresponds to a waveform of a wave that is out of phase with the wave of the first mode shown in Figure 7. In other words, the wavelength of the stress waveform due to the first mode is the same as the wavelength of the wave in the first mode. The same is true for the second and third modes.

[0048] As shown in Figure 11, the stress due to the secondary mode is distributed in both the positive and negative directions. In the distribution of the stress due to the secondary mode, the integral value in the positive direction is the same as the integral value in the negative direction. Therefore, the integral value of the stress due to the secondary mode throughout the thickness direction of the laminate 9 is 0.

[0049] 12, the stress due to the third mode is also distributed in both the positive and negative directions. However, the integral value of the stress due to the third mode over the entire thickness direction of the laminated portion 9 is not zero.

[0050] The integral value of the stress distribution in the thickness direction located in the piezoelectric layer is proportional to the electromechanical coupling coefficient, as shown in J. Kaitila, "Review of Wave Propagation in BAW Thin Film Devices Progress and Prospects," Proc. IEEE Ultrasonics Symposium Proceedings, 2007, pp. 120-129.

[0051] The integral of the stress due to each mode located on the piezoelectric layer 5 varies depending on the relationship between the thickness Tp of the piezoelectric layer 5, the thickness Td_f of the first dielectric film 6, and the thickness Td_b of the second dielectric film 7. In the above design parameters of the first embodiment, Td_f = Td_b. In this case, the integral of the stress due to the second-order mode located on the piezoelectric layer 5 is 0. Therefore, the electromechanical coupling coefficient of the second-order mode is small. Therefore, as shown in FIG. 6 , the second-order mode can be suppressed in the first embodiment. In other words, in theory, the second-order mode is not excited in the first embodiment.

[0052] In the above design parameters of the first embodiment, Td_f = Td_b = 0.25Tp. Therefore, the thickness Tp of the piezoelectric layer 5 is 2 / 3 of the total thickness of the laminated portion 9. In this case, (Td_f / Tp) + (Td_b / Tp) = 0.5. The suppression of the third mode by this will be explained below.

[0053] Fig. 13 is a diagram schematically showing the distribution of stress in the thickness direction due to the third-order mode at the center between the electrode fingers of the laminated section when the third-order mode is suppressed. The solid lines extending left and right on the upper side of Fig. 13 indicate the boundary between the piezoelectric layer 5 and the first dielectric film 6. This is also true in Fig. 14, which will be described later. The solid lines extending left and right on the lower side of Fig. 13 indicate the boundary between the piezoelectric layer 5 and the second dielectric film 7.

[0054] The total thickness of the laminated portion 9 corresponds to 1.5 wavelengths of the waveform of the stress due to the third-order mode. FIG. 13 shows a case where one wavelength, which is 2 / 3 of this, corresponds to the thickness of the piezoelectric layer 5. In this case, the integral value of the portion of the stress due to the third-order mode located in the piezoelectric layer 5 is 0. Therefore, the electromechanical coupling coefficient of the third-order mode is small. Therefore, as shown in FIG. 6, the third-order mode can be suppressed in the first embodiment.

[0055] However, the third-order mode is not only suppressed when (Td_f / Tp) + (Td_b / Tp) = 0.5. For example, when (Td_f / Tp) + (Td_b / Tp) is near 0.5, the thickness of the piezoelectric layer 5 corresponds to approximately one wavelength of the waveform of the stress due to the third-order mode. In this case, the absolute value of the integral of the portion of the stress due to the third-order mode located at the piezoelectric layer 5 is small. Therefore, the electromechanical coupling coefficient of the third-order mode can be reduced, and the third-order mode can be suppressed.

[0056] FIG. 14 shows, as a reference example, the distribution of stress due to the third-order mode in a configuration in which only the first dielectric film 6 is provided out of the first dielectric film 6 and the second dielectric film 7. In the reference example, the laminated portion is the portion where the first dielectric film 6 and the piezoelectric layer 5 are laminated. Td_f = 0.5Tp, Td_b = 0, and (Td_f / Tp) + (Td_b / Tp) = 0.5. In the reference example as well, the thickness Tp of the piezoelectric layer 5 is 2 / 3 of the total thickness of the laminated portion. This suppresses the third-order mode. However, in the reference example, the second-order mode is difficult to suppress.

[0057] From the above, when (Td_f / Tp) + (Td_b / Tp) is around 0.5, the third-order mode is suppressed regardless of the relationship between the thickness Td_f of the first dielectric film 6 and the thickness Td_b of the second dielectric film 7.

[0058] In reality, differences in physical properties such as density and elastic constant among the piezoelectric layer 5, the first dielectric film 6, and the second dielectric film 7 affect the integral value of the third mode stress at the portion located at the piezoelectric layer 5. Therefore, even if (Td_f / Tp) + (Td_b / Tp) is not equal to 0.5, the integral value may become 0.

[0059] As with the third mode, other higher modes can also be suppressed if the absolute value of the integral of the stress due to the other higher modes located in the piezoelectric layer 5 is small. Below, the conditions under which the integral becomes 0 in the fourth to eighth modes are shown.

[0060] The integral value of the stress due to the fourth mode is 0 if Td_f = Td_b or (Td_f / Tp) + (Td_b / Tp) is 1. The integral value of the stress due to the fifth mode is 0 if (Td_f / Tp) + (Td_b / Tp) is 0.25 or 1.5. The integral value of the stress due to the sixth mode is 0 if Td_f = Td_b or (Td_f / Tp) + (Td_b / Tp) is 0.5 or 2. The integral value of the stress due to the seventh mode is 0 if (Td_f / Tp) + (Td_b / Tp) is 0.167, 0.75, or 2.5. The integral value of the stress due to the eighth mode is 0 if Td_f=Td_b or (Td_f / Tp)+(Td_b / Tp) is 0.333, 1, or 3.

[0061] In the above design parameters of the first embodiment, Td_f=Td_b. Therefore, in the first embodiment, not only the second and third modes but also the fourth, sixth, and eighth modes can be suppressed.

[0062] 10, the stress due to the first mode is distributed only in the positive direction. Therefore, the greater the ratio of the thickness Tp of the piezoelectric layer 5 to the total thickness of the laminate 9, the greater the integral value of the stress due to the first mode at the portion located on the piezoelectric layer 5. This makes it possible to increase the electromechanical coupling coefficient in the first mode.

[0063] 10 , the closer to the surface side of the first dielectric film 6 in the laminated portion 9, the smaller the absolute value of the stress due to the first mode. Similarly, the closer to the surface side of the second dielectric film 7 in the laminated portion 9, the smaller the absolute value of the stress due to the first mode. Therefore, the closer the thickness Td_f of the first dielectric film 6 is to the thickness Td_b of the second dielectric film 7, the larger the integral value of the stress due to the first mode in the portion located at the piezoelectric layer 5. This makes it possible to increase the electromechanical coupling coefficient in the first mode.

[0064] In the above design parameters of the first embodiment, Td_f=Td_b, so that the second and third modes can be suppressed and the first mode can be excited strongly.

[0065] Furthermore, the relationship between the thickness Tp of the piezoelectric layer 5, the thickness Td_f of the first dielectric film 6, and the thickness Td_b of the second dielectric film 7, which can suppress the second-order mode and the third-order mode, was derived. More specifically, the admittance-frequency characteristics of the elastic wave resonator 1 were derived using the equation shown in "V. Plessky, et al. "A formula for the admittance of laterally excited bulk wave resonators (XBARs)." Electronics Letters 57.20 (2021): 773-775." By using this equation, it is possible to derive the admittance-frequency characteristics including the first-order mode and higher-order modes in the thickness direction such as the second-order mode and the third-order mode.

[0066] In deriving the admittance frequency characteristics, the resonance frequency of the first mode was set to coincide with the resonance frequency in the impedance frequency characteristics shown in Fig. 6. Specifically, the thickness Tp of the piezoelectric layer 5, the thickness Td_f of the first dielectric film 6, and the thickness Td_b of the second dielectric film 7 were adjusted.

[0067] The resonant frequency Fr and the anti-resonant frequency Fa of each of the first, second and third modes in the admittance frequency characteristics were determined. Then, the electromechanical coupling coefficient K of each mode was calculated using the resonant frequency Fr and the anti-resonant frequency Fa of each mode according to the following formula: ２ ｅｆｆ was calculated.

[0068]

[0069] The ratio Td_f / Tp of the thickness Td_f of the first dielectric film 6 to the thickness Tp of the piezoelectric layer 5, the ratio Td_b / Tp of the thickness Td_b of the second dielectric film 7 to the thickness Tp of the piezoelectric layer 5, and the electromechanical coupling coefficient K ２ ｅｆｆ The relationship between

[0070] First, under the condition that the second-order mode is not excited, the ratio Td_f / Tp and the electromechanical coupling coefficient K ２ ｅｆｆ Specifically, in a state where Td_f / Tp=Td_b / Tp, the electromechanical coupling coefficient K ２ ｅｆｆ was calculated.

[0071] FIG. 15 shows the relationship between the ratio Td_f / Tp and the electromechanical coupling coefficient K of the third mode when Td_f=Td_b. ２ ｅｆｆ Since Td_f=Td_b, the relationship shown in FIG. 15 is a function of the ratio Td_b / Tp and the electromechanical coupling coefficient K ２ ｅｆｆ It is the same as the relationship between

[0072] As shown in FIG. 15, when the ratio Td_f / Tp is around 0.27, the electromechanical coupling coefficient K ２ ｅｆｆ Since Td_f=Td_b, when the ratio Td_f / Tp and the ratio Td_b / Tp are around 0.27 and (Td_f / Tp)+(Td_b / Tp) is around 0.54, the electromechanical coupling coefficient K ２ ｅｆｆ Specifically, the electromechanical coupling coefficient K ２ ｅｆｆ is approximately 0. In this case, the third-order mode is hardly excited. Note that since Td_f=Td_b, the second-order mode is not excited.

[0073] However, even when (Td_f / Tp) + (Td_b / Tp) is other than 0.54, the electromechanical coupling coefficient K ２ ｅｆｆ In the following, the electromechanical coupling coefficient K ２ ｅｆｆ 10 shows the range of (Td_f / Tp)+(Td_b / Tp) in which the third mode can be suppressed.

[0074] When 0.322<(Td_f / Tp)+(Td_b / Tp)<0.786, the electromechanical coupling coefficient K ２ ｅｆｆ When 0.354<(Td_f / Tp)+(Td_b / Tp)<0.746, the electromechanical coupling coefficient K ２ ｅｆｆ When 0.39<(Td_f / Tp)+(Td_b / Tp)<0.704, the electromechanical coupling coefficient K ２ ｅｆｆ When 0.432<(Td_f / Tp)+(Td_b / Tp)<0.65, the electromechanical coupling coefficient K ２ ｅｆｆWhen 0.488<(Td_f / Tp)+(Td_b / Tp)<0.586, the electromechanical coupling coefficient K ２ ｅｆｆ can be made smaller than 0.1%.

[0075] Under the condition that the third mode is hardly excited, the ratio Td_f / Tp and the electromechanical coupling coefficient K ２ ｅｆｆ Specifically, in a state where (Td_f / Tp) + (Td_b / Tp) = 0.54, the electromechanical coupling coefficient K ２ ｅｆｆ was calculated.

[0076] FIG. 16 shows the relationship between the ratio Td_f / Tp and the electromechanical coupling coefficient K of the second mode when (Td_f / Tp+Td_b / Tp)=0.54. ２ ｅｆｆ FIG.

[0077] As shown in FIG. 16, when 0.172<Td_f / Tp<0.368, the electromechanical coupling coefficient K ２ ｅｆｆ is smaller than 2%. On the other hand, the range of the ratio Td_b / Tp of the thickness Td_b of the second dielectric film 7 to the thickness Tp of the piezoelectric layer 5 is (Td_f / Tp) + (Td_b / Tp) = 0.54, and therefore 0.172 < Td_b / Tp < 0.368. In these ranges of the ratio Td_f / Tp and the ratio Td_b / Tp, |(Td_f / Tp) - (Td_b / Tp)| < 0.196. In this range, the electromechanical coupling coefficient K ２ ｅｆｆ can be made smaller than 2%, and the second mode can be suppressed.

[0078] In the following, the electromechanical coupling coefficient K ２ ｅｆｆ The figure shows the range of |(Td_f / Tp)-(Td_b / Tp)| in which the second mode can be effectively suppressed.

[0079] If 0.186<Td_f / Tp<0.354, then |(Td_f / Tp)−(Td_b / Tp)|<0.168, and the electromechanical coupling coefficient K ２ ｅｆｆ When 0.203<Td_f / Tp<0.337, |(Td_f / Tp)-(Td_b / Tp)|<0.134, and the electromechanical coupling coefficient K ２ ｅｆｆ When 0.223<Td_f / Tp<0.317, |(Td_f / Tp)−(Td_b / Tp)|<0.094, and the electromechanical coupling coefficient K ２ ｅｆｆ When 0.249<Td_f / Tp<0.291, |(Td_f / Tp)−(Td_b / Tp)|<0.042, and the electromechanical coupling coefficient K ２ ｅｆｆ can be made smaller than 0.1%.

[0080] 15 and 16, it can be seen that the second-order mode and the third-order mode can be suppressed in the configuration of the first embodiment described above. That is, in the first embodiment, 0.322<(Td_f / Tp)+(Td_b / Tp)<0.786 and |(Td_f / Tp)-(Td_b / Tp)|<0.196. As a result, the electromechanical coupling coefficients K ２ ｅｆｆ can be made smaller than 2%, and the second and third modes can be suppressed.

[0081] It is preferable that 0.354<(Td_f / Tp)+(Td_b / Tp)<0.746 and |(Td_f / Tp)-(Td_b / Tp)|<0.168. Thereby, the electromechanical coupling coefficients K ２ ｅｆｆ can be made smaller than 1.5%.

[0082] It is more preferable that 0.39<(Td_f / Tp)+(Td_b / Tp)<0.704 and |(Td_f / Tp)-(Td_b / Tp)|<0.134. Thereby, the electromechanical coupling coefficients K ２ ｅｆｆ can be made smaller than 1%.

[0083] It is more preferable that 0.432<(Td_f / Tp)+(Td_b / Tp)<0.65 and |(Td_f / Tp)-(Td_b / Tp)|<0.094. Thereby, the electromechanical coupling coefficients K ２ ｅｆｆ can be made smaller than 0.5%.

[0084] It is even more preferable that 0.488<(Td_f / Tp)+(Td_b / Tp)<0.586 and |(Td_f / Tp)-(Td_b / Tp)|<0.042. Thereby, the electromechanical coupling coefficients K ２ ｅｆｆ can be made smaller than 0.1%.

[0085] In addition, within the above-mentioned preferable range, the electromechanical coupling coefficient K ２ ｅｆｆ Therefore, the acoustic wave device according to the present invention can be suitably used as an acoustic wave resonator in a filter device with a wide passband such as 5G.

[0086] High-frequency filters for general communications require high attenuation characteristics in frequency bands near integer multiples of the passband frequency. For example, frequency band B shown in FIG. 5 is a frequency band whose frequency is twice that of the passband frequency. For example, if a third-order mode occurs in frequency band B, the attenuation characteristics in frequency band B deteriorate. This can cause a deterioration in the quality of communication in the high-frequency filter. However, whether or not the third-order mode is likely to occur in frequency band B depends on the fractional bandwidth. In addition, the ratio of the third-order mode resonant frequency to the first-order mode resonant frequency depends on the ratio Td_f / Tp, etc. Note that when the first-order mode resonant frequency is Fr1 and the third-order mode resonant frequency is Fr3, the ratio is Fr3 / Fr1.

[0087] FIG. 17 is a diagram showing the relationship between the ratio Td_f / Tp and the ratio Fr3 / Fr1.

[0088] First, the anti-resonance frequency of the first-order mode of the parallel arm resonator is located within the passband of the filter device. The resonant frequency Fr1 of the first-order mode of the parallel arm resonator is slightly lower than the lower end FL of the passband of the filter device. As shown in FIG. 17 , the resonant frequency Fr3 of the third-order mode is approximately 2.2 to 2.9 times the resonant frequency Fr1 of the first-order mode. Therefore, the third-order mode occurs at a frequency that is at least twice the lower end FL of the passband of the filter device. When the passband is relatively narrow, the third-order mode occurs at a frequency that is at least twice the upper end FH of the passband. In other words, the third-order mode does not occur in frequency band B. Therefore, when the filter device is used in a communication device, degradation of communication quality is unlikely to occur.

[0089] On the other hand, when the passband is relatively wide, the third-order mode occurs at a frequency lower than twice the upper limit FH of the passband. For example, a wide passband is required for a 5G high-frequency filter device. Furthermore, when the fractional bandwidth is 8.5% or more, the third-order mode is likely to occur in frequency band B. As described above, deterioration of the attenuation characteristics in frequency band B causes deterioration in communication quality. Therefore, when the fractional bandwidth of the filter device is 8.5% or more, it is particularly necessary to suppress the third-order mode. This is described in detail below.

[0090] The lower end FL of the passband of the filter device was set to 4400 MHz, and the upper end FH of the passband was changed to change the bandwidth of the passband. When the bandwidth of the passband is R, R = FH - FL. The difference between the lower end FL of the passband and the resonant frequency of the first mode of the parallel arm resonator was set to 0.33 times the bandwidth R. In other words, |FL - Fr1| / R = 0.33.

[0091] 5, the parallel arm resonator has a first-order mode resonant frequency Fr1 of 4200 MHz and a third-order mode resonant frequency Fr3 of 9500 MHz. In this case, the third-order mode resonant frequency Fr3 of the parallel arm resonator is 2.26 times the first-order mode resonant frequency Fr1. Similarly, in this study, Fr3 / Fr1 = 2.26. However, in this study, the first-order mode resonant frequency Fr1 is not limited to 4200 MHz.

[0092] Under the conditions of FL = 4400 MHz, |FL - Fr1| / R = 0.33, and Fr3 / Fr1 = 2.26, when the resonant frequency Fr3 of the third-order mode of the parallel arm resonator coincides with a frequency twice the upper end FH of the pass band, the upper end FH is 4817 MHz. In this case, the fractional bandwidth is 9%. If the fractional bandwidth is wider, the resonant frequency Fr3 of the third-order mode is lower than the frequency twice the upper end FH of the pass band. The frequency twice the upper end FH of the pass band is the highest frequency in frequency band B. Therefore, when the fractional bandwidth value is 9% or more, the third-order mode occurs in frequency band B. However, in reality, variations occur in each frequency due to temperature changes and manufacturing variations. Therefore, it is particularly necessary to suppress the third-order mode when the fractional bandwidth is 8.5% or more.

[0093] In the first embodiment, the fractional bandwidth of the filter device 10 is 8.5% or more, and the second-order mode and the third-order mode can be suppressed in the acoustic wave resonator 1. Therefore, the filter device 10 can be suitably used as a filter device for 5G or the like, and the communication performance is unlikely to deteriorate.

[0094] The configuration of the first embodiment will be described in more detail below.

[0095] 1 , the filter device 10 has a first signal terminal 13, a second signal terminal 14, a plurality of series arm resonators, and a plurality of parallel arm resonators. The first signal terminal 13 is connected to an input potential. The second signal terminal 14 is connected to an output potential. Note that the first signal terminal 13 may be connected to an output terminal, and the second signal terminal 14 may be connected to the input potential. The first signal terminal 13 and the second signal terminal 14 may be configured as, for example, electrode pads or wiring.

[0096] Specifically, the plurality of series arm resonators are series arm resonator S1, series arm resonator S2, series arm resonator S3, series arm resonator S4, and series arm resonator S5. The plurality of series arm resonators are connected in series with each other between the first signal terminal 13 and the second signal terminal 14. Note that, from the first signal terminal 13 side, the series arm resonator S1, series arm resonator S2, series arm resonator S3, series arm resonator S4, and series arm resonator S5 are arranged in this order.

[0097] Specifically, the multiple parallel arm resonators are an elastic wave resonator 1, a parallel arm resonator P2, a parallel arm resonator P3, and a parallel arm resonator P4. The elastic wave resonator 1 is connected between the connection point between the series arm resonators S1 and S2 and ground potential. The parallel arm resonator P2 is connected between the connection point between the series arm resonators S2 and S3 and ground potential. The parallel arm resonator P3 is connected between the connection point between the series arm resonators S3 and S4 and ground potential. The parallel arm resonator P4 is connected between the connection point between the series arm resonators S4 and S5 and ground potential.

[0098] In designing a ladder filter, it is necessary to make the resonant frequencies of the series arm resonators and the parallel arm resonators different from each other. This allows for a desired passband, etc. In addition, by making the resonant frequencies of multiple series arm resonators different from each other, or by making the resonant frequencies of multiple parallel arm resonators different from each other, it is possible to improve the filter characteristics.

[0099] 2, an acoustic wave resonator is formed by providing an IDT electrode 8 as a functional electrode on a piezoelectric layer 5. Such a filter device 10 is sometimes called a chip. Within one chip, it is sufficient that at least the resonant frequencies of the series arm resonators and the parallel arm resonators are different from each other.

[0100] In the elastic wave resonator using thickness-shear mode bulk waves according to the present invention, the resonant frequency can be adjusted by adjusting the thickness of at least one of the piezoelectric layer 5, the first dielectric film 6, and the second dielectric film 7. A single chip may include a plurality of elastic wave resonators having different thicknesses of the piezoelectric layer 5, the first dielectric film 6, and the second dielectric film 7. This allows the resonant frequencies of at least the series arm resonators and the parallel arm resonators to be different from one another.

[0101] First to third modified examples of the first embodiment show examples in which a plurality of elastic wave resonators are configured in which the thicknesses of the piezoelectric layer 5, the first dielectric film 6, or the second dielectric film 7 are different from one another. Note that in this specification, taking the first embodiment shown in FIG. 2 as an example, the thickness of the piezoelectric layer 5 in each elastic wave resonator is the thickness of a portion of the piezoelectric layer 5 located in a central portion 9 a between the electrode fingers of the laminated portion 9. Similarly, the thicknesses of the first dielectric film 6 and the second dielectric film 7 in each elastic wave resonator are the thicknesses of the portions of the first dielectric film 6 and the second dielectric film 7 located in a central portion 9 a between the electrode fingers of the laminated portion 9.

[0102] The circuit configurations of the first to third modified examples are the same as that of the first embodiment. In the first to third modified examples, the fractional bandwidth is 8.5% or greater. Furthermore, in all elastic wave resonators, 0.322<(Td_f / Tp)+(Td_b / Tp)<0.786 and |(Td_f / Tp)-(Td_b / Tp)|<0.196 are satisfied. As a result, in the first to third modified examples, as in the first embodiment, the passband can be widened and the second- and third-order modes can be suppressed in the elastic wave resonators.

[0103] In addition, in at least one elastic wave resonator, 0.322<(Td_f / Tp)+(Td_b / Tp)<0.786 and |(Td_f / Tp)−(Td_b / Tp)|<0.196 may be satisfied.

[0104] In a first modified example shown in Fig. 18 , the piezoelectric layer 5A has a thickness change region 5X. The thickness change region 5X is a region in a layer where the thickness changes. In this modified example, a plurality of acoustic wave resonators share the same piezoelectric layer 5A. In the portion shown in Fig. 18 , a thickness change region 5X is provided between adjacent acoustic wave resonators. As a result, in the portion shown in Fig. 18 , acoustic wave resonators having piezoelectric layers 5A with different thicknesses are adjacent to each other.

[0105] In a second modified example shown in Fig. 19 , a first dielectric film 6A has a thickness-varying region 6X. In the portion shown in Fig. 19 , acoustic wave resonators having first dielectric films 6A with different thicknesses are adjacent to each other. On the other hand, in a third modified example shown in Fig. 20 , a second dielectric film 7A has a thickness-varying region 7X. In the portion shown in Fig. 20 , acoustic wave resonators having second dielectric films 7A with different thicknesses are adjacent to each other.

[0106] In the first to third modified examples, each thickness change region is configured as a step portion. In the piezoelectric layer, the first dielectric film, or the second dielectric film, portions of different thickness are provided with the thickness change region as a boundary. Note that the thickness change region may be configured as an inclined surface connecting the portions of different thickness.

[0107] The portions of the piezoelectric layer 5A shown in Fig. 18 that are different in thickness may be formed by, for example, film formation or etching. The same applies to the first dielectric film 6A and the second dielectric film 7A shown in Figs. 19 and 20. However, dielectrics such as silicon oxide are easier to form or etch than piezoelectrics such as lithium niobate. For this reason, it is desirable to provide portions of the first dielectric film 6A that are different in thickness and / or to provide portions of the second dielectric film 7A that are different in thickness.

[0108] Hereinafter, the number indicating the number of thicknesses of the piezoelectric layer, the first dielectric film, or the second dielectric film in a filter device will be referred to as the number of thickness levels of the piezoelectric layer, the first dielectric film, or the second dielectric film. The greater the number of thickness levels of the piezoelectric layer in a filter device, the more reliably the filter characteristics of the filter device can be improved. The same applies when the number of thickness levels of the first dielectric film or the second dielectric film is large.

[0109] On the other hand, if the piezoelectric layer, the first dielectric film, or the second dielectric film have a large number of thickness levels, the manufacturing process for the filter device becomes complicated, which may lead to increased costs. For this reason, it is desirable that the number of thickness levels per filter device, i.e., per chip, be six or less.

[0110] As shown in FIG. 18 , the piezoelectric layer 5A of the first modified example has a membrane portion 5Y. Specifically, the membrane portion 5Y is a portion of the piezoelectric layer 5A that overlaps with the cavity 2a. For example, if the thickness change region 5X of the piezoelectric layer 5A were located in the membrane portion 5Y, the membrane portion 5Y could be damaged. This is because, when the thickness change region 5X overlaps with the cavity 2a in a planar view, stress is likely to concentrate in the thickness change region 5X. More specifically, stress is likely to concentrate in the thickness change region 5X when a temperature change occurs or when mechanical vibration or impact is applied. Because such an area of ​​stress concentration is located in the membrane portion 5Y, the membrane portion 5Y could be damaged.

[0111] As in the first modification, it is desirable that the thickness changing region 5X be located within the region where the laminated portion 9A, which is made up of the piezoelectric layer 5A, the first dielectric film 6, and the second dielectric film 7, is in contact with the insulating layer 4. This makes it difficult for the membrane portion 5Y to be damaged.

[0112] In the second and third modified examples shown in FIGS. 19 and 20 , the piezoelectric layer 5 also includes a membrane portion 5Y. Stress is also likely to concentrate in the thickness change region 6X shown in FIG. 19 and the thickness change region 7X shown in FIG. 20 . Therefore, if the thickness change region 6X or the thickness change region 7X is located in a portion of the first dielectric film 6A or the second dielectric film 7A that is laminated with the membrane portion 5Y of the piezoelectric layer 5, the membrane portion 5Y may be damaged. As in the second and third modified examples, it is desirable that the thickness change region 6X or the thickness change region 7X be located within the region where the laminate portion 9B or the laminate portion 9C is in contact with the insulating layer 4. This makes the membrane portion 5Y less likely to be damaged.

[0113] The circuit configuration of the filter device 10 is not limited to the above. As described above, the filter device 10 may include a ladder-type circuit unit. In the first embodiment, the elastic wave resonator 1 is the parallel arm resonator closest to the first signal terminal 13 among the multiple parallel arm resonators. However, the arrangement of the elastic wave resonators 1 is not limited to the above.

[0114] As shown in FIG. 2 , a recess is provided in the insulating layer 4. A second dielectric film 7 and a piezoelectric layer 5 are provided on the insulating layer 4 so as to close the recess. This forms a hollow portion. This hollow portion is the cavity 2a. In the first embodiment, the support member 2 and the piezoelectric layer 5 are arranged so that a portion of the support member 2 and a portion of the piezoelectric layer 5 face each other with the cavity 2a in between. However, the recess in the support member 2 may be provided across the insulating layer 4 and the support substrate 3. Alternatively, a recess provided only in the support substrate 3 may be closed by the insulating layer 4. The recess may be provided in the piezoelectric layer 5, for example. The cavity 2a may be a through-hole provided in the support member 2.

[0115] The cavity 2a is the acoustic reflecting portion of the present invention. At least a portion of the IDT electrode 8 needs to overlap with the acoustic reflecting portion in a planar view. Specifically, at least a portion of the multiple electrode fingers needs to overlap with the acoustic reflecting portion in a planar view. The acoustic reflecting portion can confine the energy of the elastic wave on the piezoelectric layer 5 side. Note that an acoustic reflecting film, which will be described later, may be provided as the acoustic reflecting portion. However, when the acoustic reflecting portion is the cavity 2a, the energy of the elastic wave can be further confined on the piezoelectric layer 5 side.

[0116] It is preferable that the excitation regions C overlap with the acoustic reflecting portions in a plan view, so that the energy of the elastic waves can be more reliably confined on the piezoelectric layer 5 side.

[0117] In the first embodiment, the piezoelectric layer 5 is made of lithium niobate. However, the material of the piezoelectric layer 5 is not limited to the above, and may be, for example, LiTaO ３ Lithium tantalate such as AlN, aluminum nitride such as ScAlN, or PZT (lead zirconate titanate) can also be used.

[0118] However, it is preferable that the piezoelectric layer 5 be made of lithium niobate. Lithium niobate is a single crystal with low viscous loss, which can reduce the energy loss of the main mode. In addition, the high piezoelectricity of lithium niobate makes it easy to widen the passband of the filter device 10.

[0119] Among lithium niobate, it is more preferable to use Z-cut, 80° to 160° rotated Y-cut X-propagation, or 70° to 100° rotated Y-cut X-90° propagation lithium niobate as the material for the piezoelectric layer 5. This increases the electromechanical coupling coefficient of the first-order mode. When the piezoelectric layer 5 is made of Z-cut lithium niobate, the electromechanical coupling coefficient of the first-order mode can be increased regardless of the propagation direction. The propagation direction here refers to a direction that is parallel to the first main surface 5a of the piezoelectric layer 5 and parallel to the direction perpendicular to the electrode fingers. In this specification, when "to" is used in a cut angle range, it indicates a range that is equal to or greater than the number before "to" and equal to or less than the number after "to".

[0120] In the first embodiment, the first dielectric film 6 and the second dielectric film 7 are made of silicon oxide. The materials for the first dielectric film 6 and the second dielectric film 7 are not limited to those described above, and may be, for example, silicon nitride, aluminum oxide, silicon oxynitride, or silicon oxycarbide such as SiOC. However, it is preferable that the first dielectric film 6 and the second dielectric film 7 are made of silicon oxide. In this case, the absolute value of the temperature coefficient of frequency (TCF) of the acoustic wave resonator 1 can be reduced, and the frequency-temperature characteristics of the acoustic wave resonator 1 can be improved. Additionally, the acoustic wave resonator 1 can be easily processed.

[0121] The materials of the first dielectric film 6 and the second dielectric film 7 may be different from each other. The first dielectric film 6 and the second dielectric film 7 are not limited to single-layer dielectric films. At least one of the first dielectric film 6 and the second dielectric film 7 may be a laminated film. In the first embodiment, the first dielectric film 6 and the second dielectric film 7 have the same thickness, but they do not necessarily have to have the same thickness and may be different from each other. The thickness of these films is selected appropriately based on conditions such as filter characteristics, reliability, and ease of manufacture.

[0122] 2 may be formed integrally from the same material. In this case, the thickness Td_b of the second dielectric film 7 is the thickness of the portion of the second dielectric film 7 that overlaps the intersection region A in plan view. Alternatively, the insulating layer 4 and the second dielectric film 7 may be formed individually from different materials.

[0123] In the first embodiment, of the first main surface 5a and the second main surface 5b of the piezoelectric layer 5, the second main surface 5b is the main surface on the support member 2 side. However, of the first main surface 5a and the second main surface 5b, the first main surface 5a may be the main surface on the support member 2 side. In this case, the first dielectric film 6 is provided on the insulating layer 4. The insulating layer 4 and the first dielectric film 6 may be provided integrally using the same material. In this case, the thickness Td_f of the first dielectric film 6 is the thickness of the portion of the first dielectric film 6 that overlaps with the intersection region A in plan view. Alternatively, the insulating layer 4 and the first dielectric film 6 may be provided separately using different materials.

[0124] 4 , in the first embodiment, the thickness Td_f of the first dielectric film 6 is thinner than the thickness of each electrode finger of the IDT electrode 8. Therefore, the surface of the first dielectric film 6 has an uneven shape. Specifically, the distance between the surface of the portion of the first dielectric film 6 covering the electrode fingers and the first main surface 5 a of the piezoelectric layer 5 is different from the distance between the surface of the portion of the first dielectric film 6 located between the electrode fingers and the first main surface 5 a. However, the present invention is not limited to this.

[0125] For example, in a fourth modification of the first embodiment shown in FIG. 21 , the thickness of the first dielectric film 6B is greater than the thickness of each electrode finger of the IDT electrode 8. The first dielectric film 6B covers the IDT electrode 8. The surface of the first dielectric film 6B is flat. Even in this case, it is sufficient that the fractional bandwidth is 8.5% or greater, 0.322<(Td_f / Tp)+(Td_b / Tp)<0.786, and |(Td_f / Tp)-(Td_b / Tp)|<0.196. This allows the passband of the filter device to be widened and the second- and third-order modes to be suppressed in the acoustic wave resonator, as in the first embodiment.

[0126] Second to fifth embodiments show examples of laminated structures or electrode configurations in elastic wave resonators that differ from those of the first embodiment. In the second to fifth embodiments, as in the first embodiment, the fractional bandwidth is 8.5% or more, 0.322<(Td_f / Tp)+(Td_b / Tp)<0.786, and |(Td_f / Tp)-(Td_b / Tp)|<0.196. This allows the second to fifth embodiments to widen the passband and suppress the second- and third-order modes in the elastic wave resonator.

[0127] FIG. 22 is a schematic front cross-sectional view showing the vicinity of a pair of electrode fingers of an elastic wave resonator according to the second preferred embodiment.

[0128] This embodiment differs from the first embodiment in that the first dielectric film 6 in the acoustic wave resonator 21 is provided between the first main surface 5 a of the piezoelectric layer 5 and the IDT electrode 8. In all other respects, the filter device of this embodiment has the same configuration as the filter device 10 of the first embodiment.

[0129] The configuration in which the first dielectric film 6 is provided between the piezoelectric layer 5 and the IDT electrode 8 in the acoustic wave resonator 21 can be adopted to configurations other than that of the second embodiment of the present invention.

[0130] FIG. 23 is a schematic front cross-sectional view showing the vicinity of a pair of electrode fingers of an elastic wave resonator according to the third preferred embodiment.

[0131] This embodiment differs from the first embodiment in that the acoustic reflection portion of the acoustic wave resonator 31 is an acoustic reflection film 38. This embodiment also differs from the first embodiment in that the support member 32 is made of only a support substrate. Except for the above points, the filter device of this embodiment has the same configuration as the filter device 10 of the first embodiment.

[0132] An acoustic reflection film 38 is provided on the surface of the support member 32. A second dielectric film 7 is provided on the acoustic reflection film 38. Therefore, the piezoelectric layer 5 is indirectly provided on the acoustic reflection film 38 via the second dielectric film 7. It is sufficient that the support member 32 and the piezoelectric layer 5 are arranged so that at least a portion of the support member 32 and at least a portion of the piezoelectric layer 5 face each other with the acoustic reflection film 38 sandwiched between them.

[0133] The acoustic reflecting film 38 is a laminate of multiple acoustic impedance layers. Specifically, the acoustic reflecting film 38 has multiple low acoustic impedance layers and multiple high acoustic impedance layers. The low acoustic impedance layers are layers with relatively low acoustic impedance. More specifically, the multiple low acoustic impedance layers of the acoustic reflecting film 38 are the low acoustic impedance layer 33a, the low acoustic impedance layer 33b, and the low acoustic impedance layer 33c.

[0134] On the other hand, the high acoustic impedance layer is a layer with a relatively high acoustic impedance. More specifically, the multiple high acoustic impedance layers of the acoustic reflecting film 38 are the high acoustic impedance layer 34a and the high acoustic impedance layer 34b. The low acoustic impedance layers and the high acoustic impedance layers are alternately stacked. The low acoustic impedance layer 33a is the layer of the acoustic reflecting film 38 that is located closest to the piezoelectric layer 5. A second dielectric film 7 is provided directly on the low acoustic impedance layer 33a.

[0135] The acoustic reflection film 38 has three low acoustic impedance layers and two high acoustic impedance layers, although it is sufficient that the acoustic reflection film 38 has at least one low acoustic impedance layer and one high acoustic impedance layer.

[0136] The low acoustic impedance layer may be made of, for example, silicon oxide or aluminum, while the high acoustic impedance layer may be made of, for example, a metal such as platinum or tungsten, or a dielectric such as aluminum nitride, silicon nitride, or hafnium oxide.

[0137] For example, the second dielectric film 7 may function as a low acoustic impedance layer. In this case, too, it is sufficient that 0.322<(Td_f / Tp)+(Td_b / Tp)<0.786 and |(Td_f / Tp)-(Td_b / Tp)|<0.196 are satisfied.

[0138] When elastic waves are excited in the acoustic wave resonator 31, heat is generated in the multiple excitation regions C. However, in this embodiment, the portions of the piezoelectric layer 5 where the multiple excitation regions C are located are supported by the acoustic reflection film 38. This improves the heat dissipation properties of the acoustic wave resonator 31. Therefore, even when a large amount of power is applied to the acoustic wave resonator 31, the acoustic wave resonator 31 is less likely to be damaged. This improves the power durability of the acoustic wave resonator 31 and the filter device. In addition, because the piezoelectric layer 5 is supported by the acoustic reflection film 38, the piezoelectric layer 5 is less likely to be damaged by thermal stress or mechanical shock.

[0139] The acoustic wave resonator 31 having the acoustic reflection film 38 as the acoustic reflection portion may be adopted in configurations other than the third embodiment of the present invention.

[0140] FIG. 24 is a schematic plan view of an elastic wave resonator according to the fourth preferred embodiment.

[0141] This embodiment differs from the first embodiment in that the acoustic wave resonator 41 has a pair of reflectors 46 and 47. This embodiment also differs from the first embodiment in that the acoustic wave resonator 41 is configured to utilize plate waves such as Lamb waves. Except for the above points, the filter device of this embodiment has the same configuration as the filter device 10 of the first embodiment.

[0142] The reflectors 46 and 47 are provided on the first main surface 5a of the piezoelectric layer 5. Specifically, the reflectors 46 and 47 face each other across the IDT electrode 8 in the direction perpendicular to the electrode fingers. When the IDT electrode 8 is provided on the first dielectric film 6, it is sufficient that each reflector is also provided on the first dielectric film 6.

[0143] The reflector 46 has a pair of reflector bus bars and a plurality of reflector electrode fingers 45. Specifically, the pair of reflector bus bars is a first reflector bus bar 43 and a second reflector bus bar 44. The first reflector bus bar 43 and the second reflector bus bar 44 face each other. One end of each of the plurality of reflector electrode fingers 45 is connected to the first reflector bus bar 43. The other end of each of the plurality of reflector electrode fingers 45 is connected to the second reflector bus bar 44. The reflector 47 has a similar configuration to the reflector 46.

[0144] When the elastic wave resonator 41 is configured to be able to use plate waves, it is preferable to have a pair of reflectors, which can improve the resonance characteristics of the elastic wave resonator 41.

[0145] FIG. 25 is a schematic plan view illustrating a pair of electrode fingers and their vicinity in an elastic wave resonator according to a fifth preferred embodiment of the present invention.

[0146] This embodiment differs from the first embodiment in that an IDT electrode is provided on both the first principal surface 5 a and the second principal surface 5 b of the piezoelectric layer 5 in the acoustic wave resonator 51. Except for the above, the filter device of this embodiment has the same configuration as the filter device 10 of the first embodiment.

[0147] As in the first embodiment, an IDT electrode 8 is provided on the first main surface 5a of the piezoelectric layer 5. On the other hand, an IDT electrode 58 is provided on the second main surface 5b. The IDT electrode 58 has the same configuration as the IDT electrode 8. That is, the IDT electrode 58 has a pair of bus bars and a plurality of electrode fingers. In this embodiment, the crossing region defined based on the IDT electrode 8 and the crossing region defined based on the IDT electrode 58 overlap in a planar view.

[0148] In this embodiment, the thickness Td_b of the second dielectric film 7 is the thickness of the portion of the second dielectric film 7 located between the electrode fingers.

[0149] The acoustic wave resonator 51 uses a thickness-shear mode as its main mode. However, the acoustic wave resonator 51 may be configured to use a plate wave. In this case, a pair of reflectors is preferably provided on each of the first principal surface 5 a and the second principal surface 5 b of the piezoelectric layer 5. Specifically, a pair of reflectors is preferably provided on the first principal surface 5 a so as to sandwich the IDT electrode 8 in the direction perpendicular to the electrode fingers, and a pair of reflectors is preferably provided on the second principal surface 5 b so as to sandwich the IDT electrode 58 in the direction perpendicular to the electrode fingers, and to face each other.

[0150] A preferred configuration of the present invention will be described below.

[0151] When using thickness-shear bulk waves as the primary mode, Tp / p≦0.5 is preferable, and Tp / p≦0.24 is even more preferable. This allows the thickness-shear bulk waves to be suitably excited and the fractional bandwidth of the elastic wave resonator to be sufficiently large. Note that the fractional bandwidth referred to here does not refer to the fractional bandwidth of the filter device, but to the fractional bandwidth of a single elastic wave resonator. The fractional bandwidth is expressed as (|Fa-Fr| / Fr)×100[%].

[0152] FIG. 26 is a graph showing the relationship between Tp / p and the bandwidth ratio of an elastic wave resonator.

[0153] As is clear from Figure 26, when Tp / p > 0.5, the fractional bandwidth is less than 5%. In contrast, when Tp / p ≤ 0.5, the fractional bandwidth can be made 5% or more. Therefore, the electromechanical coupling coefficient of the bulk wave in thickness shear mode can be increased. When Tp / p ≤ 0.24 or less, the fractional bandwidth can be made 7% or more. Therefore, the electromechanical coupling coefficient of the bulk wave in thickness shear mode can be effectively increased. However, while the electromechanical coupling coefficient of the first-order thickness shear mode can be increased, the second-order and third-order modes are suppressed in the present invention.

[0154] When the metallization ratio of the electrode fingers to the excitation region C is MR, it is preferable to satisfy MR≦1.75(Tp / p)+0.075. In this case, the value of the fractional bandwidth of the acoustic wave resonator does not become too large, and spurious emissions between the resonant frequency and the antiresonant frequency can be suppressed. Details of this are described below.

[0155] In this specification, the metallization ratio MR of the electrode fingers to the excitation region C is the proportion of the portion of the piezoelectric layer 5 that is covered with the metal that constitutes the electrode fingers in the excitation region C when viewed in a plan view. Specifically, the metallization ratio MR is the ratio of the area of ​​the first electrode fingers 18 and the second electrode fingers 19 in the excitation region C to the area of ​​the excitation region C when viewed in a plan view. When the width of the electrode fingers located in the excitation region C is constant, the metallization ratio MR can also be calculated by dividing the sum of the widths of the electrode fingers located in the excitation region C by the dimension of the excitation region C in the direction perpendicular to the electrode fingers.

[0156] Fig. 27 is a diagram showing the relationship between the relative bandwidth of an elastic wave resonator and the magnitude of normalized spurious. Fig. 27 shows the results of measuring the amount of phase rotation of spurious every time the relative bandwidth is changed by changing the thickness of the piezoelectric layer and the dimensions of the electrode fingers. Note that the normalized magnitude of spurious in Fig. 27 is specifically a value in which the amount of phase rotation of the spurious impedance is normalized by 180°. The results shown in Fig. 27 are for a Z-cut LiNbO ３ Although this is the result when a piezoelectric layer made of this material was used, the same tendency is observed when a piezoelectric layer having another cut angle is used.

[0157] In the region surrounded by ellipse E in Figure 27, the normalized magnitude of the spurious response between the resonant frequency and the anti-resonant frequency is 1.0. If the bandwidth fraction of the elastic wave resonator exceeds 17%, the normalized magnitude of the spurious response may be 1.0 or more. For this reason, it is preferable that the bandwidth fraction be 17% or less. This makes it possible to suppress the spurious response between the resonant frequency and the anti-resonant frequency.

[0158] 28 is a diagram showing the relationship between Tp / p, metallization ratio MR, and fractional bandwidth, in which the results of calculating fractional bandwidth for different Tp / p and metallization ratio MR are shown.

[0159] In Figure 28, the hatched area is the area where the fractional bandwidth is 17% or less. The boundary between this hatched area and the non-hatched area is roughly represented by dashed line G. Dashed line G is represented by MR = 1.75(Tp / p) + 0.075. It is preferable that MR ≤ 1.75(Tp / p) + 0.075. In this case, it is easy to keep the fractional bandwidth at 17% or less.

[0160] On the other hand, the dashed-dotted line G1 in Figure 28 indicates the boundary where the slope of the change in metallization ratio MR with respect to changes in Tp / p is the same as that of the dashed line G, and where the fractional bandwidth is 17% or less over the entire range. The dashed-dotted line G1 is represented by MR = 1.75(Tp / p) + 0.05. It is more preferable that MR ≤ 1.75(Tp / p) + 0.05. In this case, the fractional bandwidth can be more reliably kept at 17% or less.

[0161] FIG. 29 shows the results of LiNbO when Tp / p approaches 0. ３ 29 is a diagram showing a map of fractional bandwidths with respect to Euler angles (0°, θ, ψ) of the frequency band of the optical fiber 10. The hatched area in FIG. 29 is a region where a fractional bandwidth of at least 5% or more can be obtained, and the range of this region can be approximated to the ranges expressed by the following formulas (1), (2), and (3).

[0162] (0°±10°, 0° to 20°, any ψ) ... Equation (1) (0°±10°, 20° to 80°, 0° to 60° (1-(θ-50) ２ / 900) １／２ ) or (0°±10°, 20° to 80°, [180°-60° (1-(θ-50) ２ / 900) １／２ ]~180°) ...Formula (2) (0°±10°, [180°-30°(1-(ψ-90) ２ / 8100) １／２ ] to 180°, any ψ) ...Equation (3)

[0163] It is preferable that the Euler angles (φ, θ, ψ) of the lithium niobate constituting the piezoelectric layer are within the range of the above formula (1), formula (2), or formula (3). This allows the relative bandwidth of the elastic wave resonator to be sufficiently wide. The same applies when the piezoelectric layer is made of lithium tantalate.

[0164] The Euler angles of the piezoelectric crystals constituting the piezoelectric layer affect the material constants of the piezoelectric crystals, such as the piezoelectric constant, elastic constant, and dielectric constant. As a result, when the Euler angles are in the above range, the relative bandwidth of the elastic wave resonator becomes wider.

[0165] In this specification, when any angle range in Euler angles is expressed using "to", this indicates that the angle range is equal to or greater than the value before "to" and equal to or less than the value after "to".

[0166] The Euler angles are one of the indices that express the in-plane orientation of a piezoelectric crystal used in an elastic wave device such as an elastic wave resonator. The angles φ, θ, and ψ in the Euler angles are rotation angles with respect to the crystal axis of the piezoelectric crystal. Specifically, for example, the Euler angles are the rotation angles when the crystal axis of the piezoelectric crystal is transformed into the coordinate system when the piezoelectric crystal is used in the elastic wave device.

[0167] Euler angles will be described in more detail below. In this description, four right-handed coordinate systems will be used. The right-handed coordinate system is a coordinate system with three mutually orthogonal axes, and the three axes are assigned to the thumb, index finger, and middle finger of the right hand in that order. For example, in the case of (α, β, γ), the α axis is assigned to the thumb, the β axis to the index finger, and the γ axis to the middle finger.

[0168] When the Euler angles are (φ, θ, ψ), the coordinate system is transformed by the following rotation operations 1) to 3). 1) Rotate (α, β, γ) by "φ" around the γ axis to obtain (α1, β1, γ1). Next, 2) rotate (α1, β1, γ1) by "θ" around the α1 axis to obtain (α2, β2, γ2). Next, 3) rotate (α2, β2, γ2) by "ψ" around the γ2 axis to obtain the orientation (α3, β3, γ3). Note that the direction of a right-handed screw is the positive direction of rotation.

[0169] By performing the rotation operations 1) to 3) above, (α, β, γ) becomes (α3, β3, γ3). The coordinate systems of (α, β, γ) and (α3, β3, γ3) share the same origin. Note that Euler angles and coordinate transformation methods are also described in the "Elastic Wave Device Technology Handbook, page 549."

[0170] Unless otherwise specified, the following examples are examples in which the material and Euler angles of piezoelectric layer 5 shown in FIG. 3 are changed in the configuration of elastic wave resonator 1 in filter device 10 shown in FIG.

[0171] When manufacturing an elastic wave device such as the elastic wave resonator 1, for example, a wafer made of a piezoelectric crystal is divided into pieces to obtain a plurality of piezoelectric layers 5. For example, a configuration in which IDT electrodes 8 are provided on the main surface of the wafer or the piezoelectric layer 5 has a coordinate system including an axis extending in the normal direction to the main surface and an axis extending in the direction perpendicular to the electrode fingers. Euler angles are indicators that indicate the relationship between the crystal axes of the wafer or the piezoelectric layer 5 and the above coordinate system.

[0172] Specifically, when the crystal axes of the wafer and the piezoelectric layer 5 are (X, Y, Z) and the coordinate system of the wafer and the piezoelectric layer 5 is (x1, x2, x3), (X, Y, Z) is transformed into (x1, x2, x3) using the Euler angles (φ, θ, ψ).

[0173] FIG. 30 is a schematic diagram of a coordinate system in which the crystal axes of a piezoelectric crystal are transformed by Euler angles.

[0174] 30 schematically shows the wafer 105 and the IDT electrodes 8 provided on the wafer 105. The x1 axis in the coordinate system of the wafer 105 extends parallel to the electrode finger orthogonal direction. The x3 axis extends in the normal direction to the main surface of the wafer 105. The x2 axis is an axis perpendicular to both the x1 axis and the x3 axis. This is also true for the piezoelectric layer 5 shown in FIG. 3 and other figures after the wafer 105 is divided. Specifically, the x3 axis extends in the normal direction to the first main surface 5a of the piezoelectric layer 5, and the x1 axis extends parallel to the electrode finger orthogonal direction.

[0175] The piezoelectric layer 5 has a negative surface and a positive surface in the polarization direction. In this specification, a main surface in which approximately 95% or more is a positive surface in the polarization direction is defined as a positive surface. A main surface in which approximately 95% or more is a negative surface in the polarization direction is defined as a negative surface. Furthermore, in this specification, the direction from "-" to "+" in the polarized state is defined as the +Z direction. The +Z direction is the polarization direction of the piezoelectric crystals that make up the piezoelectric layer 5.

[0176] FIG. 31 is a schematic diagram showing the coordinate system of the piezoelectric layer when the angle θ in the Euler angles (0°, θ, 0°) of the lithium niobate constituting the piezoelectric layer is greater than or equal to 0° and less than or equal to 60°.

[0177] The angle at which the x3 axis, which extends in the normal direction to the main surface of the piezoelectric layer 5, is inclined with respect to the Z axis of the piezoelectric layer 5 is angle θ in the Euler angles (φ, θ, ψ). In the example shown in Fig. 31, the angles φ and ψ in the Euler angles (φ, θ, ψ) are 0°. Therefore, the x1 axis, which extends parallel to the direction orthogonal to the electrode fingers, coincides with the X axis. The +Z direction, which is the polarization direction of the piezoelectric layer 5, is inclined by angle θ toward the x2 axis with respect to the x3 axis.

[0178] In this example, the first main surface 5a of the piezoelectric layer 5 is the positive surface, while the second main surface 5b of the piezoelectric layer 5 is the negative surface. An IDT electrode 8 is provided on the positive surface of the piezoelectric layer 5.

[0179] U.S. Pat. No. 10,790,802 describes that when the angle β in the Euler angles (0°, β, 0°) of the piezoelectric substrate is between 0° and 60°, thickness-shear mode bulk waves are suitably excited and unwanted waves are suppressed. In U.S. Pat. No. 10,790,802, the piezoelectric substrate is made of lithium niobate. Therefore, when the angle θ in the Euler angles (0°, θ, 0°) of the lithium niobate constituting the piezoelectric layer 5 is between 0° and 60°, thickness-shear mode bulk waves are suitably excited.

[0180] Furthermore, FIG. 32 below shows another example in which bulk waves in thickness shear mode are preferably excited.

[0181] FIG. 32 is a schematic diagram showing the coordinate system of the piezoelectric layer when the angle θ in the Euler angles (0°, θ, 90°) of the lithium niobate constituting the piezoelectric layer is greater than or equal to −15° and less than 0°.

[0182] In the example shown in FIG. 32 , the angle φ in the Euler angles (φ, θ, ψ) is 0°, and the angle ψ is 90°. Therefore, the angle of the x1 axis, which extends parallel to the electrode finger orthogonal direction, with respect to the X axis is 90°. On the other hand, the direction of the x2 axis is parallel to the direction of the X axis. However, the direction of the x2 axis and the direction of the X axis are opposite to each other. The +Z direction, which is the polarization direction of the piezoelectric layer 5, is inclined with respect to the x3 axis by an angle equal to the absolute value of the angle θ, toward the opposite side of the x1 axis. This is because the angle θ is a negative value. It can also be said that the +Z direction is inclined with respect to the x3 axis by θ+360° toward the x1 axis.

[0183] In this example, the first main surface 5a of the piezoelectric layer 5 is the positive surface, while the second main surface 5b of the piezoelectric layer 5 is the negative surface. An IDT electrode 8 is provided on the positive surface of the piezoelectric layer 5.

[0184] U.S. Pat. No. 10,797,675 describes that when the angle β in the Euler angles (0°, β, 90°) of the piezoelectric substrate is greater than or equal to -15° and less than 0°, the electromechanical coupling coefficient of the thickness-shear mode bulk wave increases. In U.S. Pat. No. 10,797,675, the piezoelectric substrate is made of lithium niobate. Therefore, when the angle θ in the Euler angles (0°, θ, 90°) of the lithium niobate constituting the piezoelectric layer 5 is greater than or equal to -15° and less than 0°, the thickness-shear mode bulk wave is suitably excited.

[0185] In addition, in "The Journal of the Acoustical Society of Japan, Vol. 36, No. 3 (1980), pp. 141-142," the trigonal (3m) crystal LiTaO ３The document explains the characteristics of surface acoustic waves in a trigonal (3m) crystal. It states that, due to the symmetry of the crystal, it is sufficient to consider the angle θ in the Euler angles (φ, θ, ψ) within the range of 0°≦θ≦180°. In other words, when the angles θ of one and the other trigonal (3m) crystals differ by 180°, the electrical characteristics are considered to be equivalent whether one crystal or the other crystal is used in an acoustic wave device such as an acoustic wave resonator.

[0186] This is because changing the angle θ in the Euler angles (φ, θ, ψ) of a crystal by 180° is equivalent to changing the polarization direction of the crystal by 180°. When the crystal angle θ is changed by 180°, the only electrical characteristic that is said to change is generally the phase of the excited acoustic wave. Therefore, even if the crystal angle θ is changed by 180°, it is believed that there is substantially no change in the electrical characteristics of an acoustic wave device such as an acoustic wave resonator.

[0187] In addition, LiTaO ３ Similarly, LiNbO ３ The same can be said for lithium niobate.

[0188] In the following, an example will be shown in which the range of angle θ in the Euler angles of lithium niobate constituting the piezoelectric layer 5 is different by 180° from the range in the examples shown in FIGS.

[0189] Fig. 33 is a schematic diagram showing a coordinate system of the piezoelectric layer when the angle θ of the lithium niobate constituting the piezoelectric layer at the Euler angles (0°, θ, 0°) is 180° or more and 240° or less. Fig. 34 is a schematic diagram showing a coordinate system of the piezoelectric layer when the angle θ of the lithium niobate constituting the piezoelectric layer at the Euler angles (0°, θ, 90°) is 165° or more and 180° or less.

[0190] 33, similar to the example shown in FIG. 31, the x1 axis extending parallel to the direction perpendicular to the electrode fingers coincides with the X axis. The +Z direction, which is the polarization direction of the piezoelectric layer 5, is inclined at an angle θ toward the x2 axis with respect to the x3 axis. In this example, the first main surface 5a of the piezoelectric layer 5 is the negative surface. Meanwhile, the second main surface 5b of the piezoelectric layer 5 is the positive surface. An IDT electrode 8 is provided on the negative surface of the piezoelectric layer 5.

[0191] In the example shown in Figure 34, similar to the example shown in Figure 32, the x2 axis and the X axis extend in opposite directions. The +Z direction, which is the polarization direction of the piezoelectric layer 5, is inclined toward the x1 axis by an angle θ with respect to the x3 axis. In this example, the first main surface 5a of the piezoelectric layer 5 is the negative surface. Meanwhile, the second main surface 5b of the piezoelectric layer 5 is the positive surface. An IDT electrode 8 is provided on the negative surface of the piezoelectric layer 5.

[0192] In the examples shown in Figures 33 and 34, the range of angle θ in the Euler angles of lithium niobate constituting the piezoelectric layer 5 is different by 180° from the range in the examples shown in Figures 31 and 32. Therefore, in the examples shown in Figures 33 and 34, bulk waves in thickness shear mode are suitably excited, as in the examples shown in Figures 31 and 32.

[0193] In the examples shown in Figures 31 and 32, the IDT electrode 8 is provided on the positive side of the piezoelectric layer 5. On the other hand, in the examples shown in Figures 33 and 34, the IDT electrode 8 is provided on the negative side of the piezoelectric layer 5. The inventors have found that when the IDT electrode is provided on the negative side, polarization inversion is less likely to occur in the piezoelectric layer than when the IDT electrode is provided on the positive side. Note that if polarization inversion occurs in a piezoelectric layer used in an elastic wave resonator, the electrical characteristics of the elastic wave resonator will deteriorate. This is illustrated using a reference example.

[0194] The basic configuration of the reference example is similar to that described in U.S. Pat. No. 1,079,0802. Specifically, a piezoelectric substrate is provided on a support member having a through hole. An IDT electrode is provided on the piezoelectric substrate.

[0195] Fig. 35 is a piezoelectric response microscope image showing an example in which polarization inversion has occurred in the piezoelectric layer. Fig. 36 is a diagram showing the admittance frequency characteristics of an elastic wave resonator in the cases where polarization inversion has occurred in the piezoelectric layer and where it has not. Note that Fig. 35 is based on a piezoelectric response microscope image taken with an atomic force microscope (manufactured by Bruker Japan, product name: Dimension-FastScan (Icon)).

[0196] As shown in Fig. 35, an IDT electrode 118 is provided on the positive surface of the piezoelectric layer 115. The cross-hatched areas in Fig. 35 are areas where polarization inversion occurs on the positive surface. As shown in Fig. 36, when polarization inversion occurs in the piezoelectric layer 115, the characteristics deteriorate both near the resonant frequency and near the antiresonant frequency compared to when polarization inversion does not occur.

[0197] 31 and 32 with the examples shown in Figures 33 and 34, it can be seen that in the examples shown in Figures 33 and 34, polarization inversion is less likely to occur in the piezoelectric layer 5, and the electrical characteristics of the elastic wave resonator 1 are less likely to deteriorate. In addition, the thickness shear mode can be suitably excited.

[0198] When thickness-shear mode bulk waves are preferably excited at Euler angles (0°, θ, 0°), thickness-shear mode bulk waves are also preferably excited when the angle φ in the Euler angles is within a range of 0°±10°. Similarly, thickness-shear mode bulk waves are also preferably excited when the angle ψ in the Euler angles is within a range of 0°±10°.

[0199] For these reasons, it is preferable that the Euler angles (φ, θ, ψ) of the lithium niobate constituting the piezoelectric layer 5 are (0°±10°, 180° to 240°, 0°±10°). This allows thickness-shear mode bulk waves to be suitably excited in the elastic wave resonator 1. In addition, polarization inversion is unlikely to occur in the piezoelectric layer 5, and the electrical characteristics of the elastic wave resonator 1 are unlikely to deteriorate. This also prevents deterioration of the filter characteristics of the filter device 10 including the elastic wave resonator 1.

[0200] The Euler angles (φ, θ, ψ) of the lithium niobate constituting the piezoelectric layer 5 are preferably (0°, 180° to 240°, 0°±3°), and even more preferably (0°, 180° to 240°, 0°). This more reliably and suitably excites thickness-shear mode bulk waves in the elastic wave resonator 1. In addition, polarization inversion is less likely to occur in the piezoelectric layer 5, and the electrical characteristics of the elastic wave resonator 1 are less likely to deteriorate. This also makes it less likely that the filter characteristics of the filter device 10 including the elastic wave resonator 1 will deteriorate.

[0201] Furthermore, when thickness-shear mode bulk waves are preferably excited at Euler angles (0°, θ, 90°), thickness-shear mode bulk waves are also preferably excited when the angle φ in the Euler angles is within the range of 0°±10°.Similarly, thickness-shear mode bulk waves are also preferably excited when the angle ψ in the Euler angles is within the range of 90°±10°.

[0202] The Euler angles (φ, θ, ψ) of the lithium niobate constituting the piezoelectric layer 5 are preferably (0°±10°, 165° to 180°, 90°±10°). This allows thickness-shear mode bulk waves to be suitably excited in the elastic wave resonator 1. In addition, polarization inversion is unlikely to occur in the piezoelectric layer 5, and the electrical characteristics of the elastic wave resonator 1 are unlikely to deteriorate. This also prevents deterioration of the filter characteristics of the filter device 10 including the elastic wave resonator 1.

[0203] The Euler angles (φ, θ, ψ) of the lithium niobate constituting the piezoelectric layer 5 are preferably (0°, 165° to 180°, 90°±3°), and even more preferably (0°, 165° to 180°, 90°). This more reliably and suitably excites thickness-shear mode bulk waves in the elastic wave resonator 1. In addition, polarization inversion is less likely to occur in the piezoelectric layer 5, and the electrical characteristics of the elastic wave resonator 1 are less likely to deteriorate. This also makes it less likely that the filter characteristics of the filter device 10 including the elastic wave resonator 1 will deteriorate.

[0204] However, the elastic wave resonator 1 does not necessarily have to be configured to excite bulk waves in thickness shear mode. For example, the elastic wave resonator 1 may be configured to excite plate waves. Even when the elastic wave resonator 1 utilizes elastic waves other than bulk waves in thickness shear mode, it is preferable that the IDT electrode 8 be provided on the negative surface of the piezoelectric layer 5. This makes it difficult for polarization inversion to occur in the piezoelectric layer 5, and thus prevents the electrical characteristics of the elastic wave resonator 1 from deteriorating.

[0205] Specifically, it is preferable that the Euler angles (φ, θ, ψ) of the piezoelectric crystal of trigonal point group 3m constituting piezoelectric layer 5 are (0°±10°, θ, 0°±10°) and 90°<θ<270°. In this case, IDT electrode 8 is provided on the negative surface of piezoelectric layer 5. Therefore, polarization inversion is unlikely to occur in piezoelectric layer 5, and the electrical characteristics of elastic wave resonator 1 are unlikely to deteriorate. The piezoelectric crystal of trigonal point group 3m is, for example, lithium niobate or lithium tantalate.

[0206] It is more preferable that the Euler angles (φ, θ, ψ) of the piezoelectric crystals of the trigonal point group 3m constituting the piezoelectric layer 5 are (0°, θ, 0°±3°) and 90°<θ<270°. It is even more preferable that the Euler angles (φ, θ, ψ) of the piezoelectric crystals of the trigonal point group 3m constituting the piezoelectric layer 5 are (0°, θ, 0°) and 90°<θ<270°. This makes it difficult for polarization inversion to occur in the piezoelectric layer 5, and the electrical characteristics of the elastic wave resonator 1 to deteriorate.

[0207] On the other hand, it is preferable that the Euler angles (φ, θ, ψ) of the piezoelectric crystal of trigonal point group 3m constituting piezoelectric layer 5 are (0°±10°, θ, 90°±10°) and 90°<θ<270°. In this case, too, IDT electrode 8 is provided on the negative surface of piezoelectric layer 5. Therefore, polarization inversion is unlikely to occur in piezoelectric layer 5, and the electrical characteristics of elastic wave resonator 1 are unlikely to deteriorate.

[0208] The Euler angles (φ, θ, ψ) of the piezoelectric crystals of the trigonal point group 3m constituting the piezoelectric layer 5 are preferably (0°, θ, 90°±3°) and 90°<θ<270°. The Euler angles (φ, θ, ψ) of the piezoelectric crystals of the trigonal point group 3m constituting the piezoelectric layer 5 are preferably (0°, θ, 90°) and 90°<θ<270°. This makes it difficult for polarization inversion to occur in the piezoelectric layer 5, and the electrical characteristics of the elastic wave resonator 1 to deteriorate.

[0209] As described above, when the Euler angles (φ, θ, ψ) of the lithium niobate or lithium tantalate constituting the piezoelectric layer 5 are within the ranges of the above formulas (1), (2), and (3), the fractional bandwidth of the elastic wave resonator can be sufficiently widened. Note that the same effect can be obtained even when the range of the angle θ in formulas (1), (2), and (3) differs by 180°.

[0210] Specifically, for example, it is preferable that the Euler angles (φ, θ, ψ) of the lithium niobate or lithium tantalate constituting the piezoelectric layer 5 be within the range of the following formula (4), formula (5), or formula (6), thereby making it possible to sufficiently widen the fractional bandwidth of the elastic wave resonator 1.

[0211] (0°±10°, 180° to 200°, any ψ) ... Equation (4) (0°±10°, 200° to 260°, 0° to 60° (1-(θ-50) ２ / 900) １／２ ) or (0°±10°, 200° to 260°, [180°-60° (1-(θ-50) ２ / 900) １／２ ]~180°) ...Formula (5) (0°±10°, [360°-30°(1-(ψ-90) ２ / 8100) １／２ ] to 360°, any ψ) ...Equation (6)

[0212] In addition, when the Euler angles (φ, θ, ψ) of the lithium niobate or lithium tantalate constituting the piezoelectric layer 5 are within the range of formula (4), formula (5), or formula (6), the IDT electrode 8 is provided on the negative surface of the piezoelectric layer 5. This makes it difficult for polarization inversion to occur in the piezoelectric layer 5, and the electrical characteristics of the acoustic wave resonator 1 to deteriorate. As a result, the filter characteristics of the filter device 10 including the acoustic wave resonator 1 are unlikely to deteriorate.

[0213] A specific method for identifying the passband of a filter device is described below. First, if a datasheet is available for the filter device product, refer to the datasheet. The passband described in the datasheet is the passband of the filter device. Examples of passbands are shown in Tables 1 to 5 below. Tables 1 and 2 show 4G passbands. Tables 3 to 5 show 5G passbands.

[0214]

[0215]

[0216]

[0217]

[0218]

[0219] If a datasheet for a filter device product does not exist, the passband of the filter device can be identified by measuring the passband characteristics of the filter device. Specifically, the passband characteristics can be measured using, for example, a network analyzer. An example of a network analyzer is "Keysight Technologies, Model No. P5027B."

[0220] When measuring the pass characteristics of a filter device, a jig may be used to connect the filter device to a network analyzer or the like. The jig may have, for example, a pair of connectors and a substrate. One connector is connected to a terminal of the filter device that is connected to an input potential. The other connector is connected to a terminal of the filter device that is connected to an output potential. The substrate grounds the terminal of the filter device that is connected to the ground potential.

[0221] For example, first, the filter device is placed on the substrate of the jig. At this time, the terminal of the filter device connected to the ground potential is brought into contact with the substrate of the jig. As a result, the terminal of the filter device connected to the ground potential is grounded.

[0222] A pair of connectors of the jig is connected to a terminal of the filter device that is connected to the input potential and a terminal that is connected to the output potential, respectively. The characteristic impedance of the path from each connector of the jig to each terminal of the filter device is usually designed to be 50 Ω. Each connector is then connected to an input / output port of a network analyzer, for example, by a cable. As a result, the terminal of the filter device that is connected to the input potential and the terminal that is connected to the output potential are each electrically connected to the input / output port of the network analyzer.

[0223] The conditions for measuring the pass characteristics of the filter device are as follows.

[0224] Temperature: Room temperature Input power: 0 dBm Pass characteristics to be measured: S21

[0225] FIG. 37 is a diagram illustrating an example of the pass characteristic of a filter device.

[0226] The passband characteristics are measured while changing the frequency. The passband of the filter device can be identified by comparing the frequency band where the insertion loss is small, as indicated by the arrows in Figure 37, with the passbands shown in Tables 1 to 5. Note that the passbands shown in Tables 1 to 5 are examples, and the passbands to be compared with the measurement results are not limited to the passbands in Tables 1 to 5.

[0227] Examples of the configuration of the filter device according to the present invention will be summarized below.

[0228] <1> A filter device having a passband, comprising a plurality of resonators including an elastic wave resonator, wherein the elastic wave resonator comprises: a piezoelectric layer having a first main surface and a second main surface facing each other; an IDT electrode provided on the first main surface of the piezoelectric layer and having a plurality of electrode fingers; a first dielectric film provided on the first main surface of the piezoelectric layer; and a second dielectric film provided on the second main surface of the piezoelectric layer, wherein a direction orthogonal to a direction in which the plurality of electrode fingers extend is defined as an electrode finger orthogonal direction, and a region where adjacent electrode fingers overlap each other when viewed from the electrode finger orthogonal direction is a crossing region, and the first dielectric film and the second dielectric film overlap each other in a plan view. a filter device in which, where FH is the highest frequency in the pass band, FL is the lowest frequency in the pass band, and FC is the center frequency of the pass band, a relative bandwidth expressed by ((FH-FL) / FC) x 100[%] is 8.5% or more, and where Tp is the thickness of the piezoelectric layer at the crossing region, Td_f is the thickness of a portion of the first dielectric film that overlaps with the crossing region in a planar view, and Td_b is the thickness of a portion of the second dielectric film that overlaps with the crossing region in a planar view, the following relationships are satisfied: 0.322<(Td_f / Tp)+(Td_b / Tp)<0.786, and |(Td_f / Tp)-(Td_b / Tp)|<0.196.

[0229] <2> The filter device according to <1>, wherein the elastic wave resonator is configured to be able to utilize bulk waves in thickness shear mode.

[0230] <3> The filter device described in <1>, further comprising a support member laminated on the piezoelectric layer, wherein an acoustic reflection portion is formed on the support member at a position that overlaps with the plurality of electrode fingers in a planar view, and wherein Tp / p is 0.5 or less when the center-to-center distance between adjacent electrode fingers is p.

[0231] <4> The filter device according to <3>, wherein Tp / p is 0.24 or less.

[0232] <5> A filter device described in <3> or <4>, in which adjacent electrode fingers overlap each other in a direction perpendicular to the electrode fingers, and the region between the centers of adjacent electrode fingers is an excitation region, and when MR is the metallization ratio of the electrode fingers to the excitation region, MR≦1.75(Tp / p)+0.075 is satisfied.

[0233] <6> A filter device described in any one of <3> to <5>, wherein the acoustic reflection portion is a hollow portion, and the support member and the piezoelectric layer are arranged so that a portion of the support member and a portion of the piezoelectric layer face each other across the hollow portion.

[0234] <7> A filter device described in any one of <3> to <5>, wherein the acoustic reflection portion is an acoustic reflection film including a high acoustic impedance layer having a relatively high acoustic impedance and a low acoustic impedance layer having a relatively low acoustic impedance, and the support member and the piezoelectric layer are arranged so that at least a portion of the support member and at least a portion of the piezoelectric layer face each other with the acoustic reflection film in between.

[0235] <8> The filter device according to <1>, configured to be able to use plate waves.

[0236] <9> The filter device according to any one of <1> to <8>, wherein the piezoelectric layer is made of lithium niobate or lithium tantalate, and the Euler angles (φ, θ, ψ) of the lithium niobate or lithium tantalate constituting the piezoelectric layer are within the range of the following formula (1), formula (2), or formula (3): (0°±10°, 0° to 20°, any ψ) ... formula (1) (0°±10°, 20° to 80°, 0° to 60° (1-(θ-50) ２ / 900) １／２ ) or (0°±10°, 20° to 80°, [180°-60° (1-(θ-50) ２ / 900) １／２ ]~180°) ...Formula (2) (0°±10°, [180°-30°(1-(ψ-90) ２ / 8100) １／２ ] to 180°, any ψ) ...Equation (3)

[0237] <10> The filter device according to any one of claims 1 to 8, wherein the piezoelectric layer is made of lithium niobate or lithium tantalate, and the Euler angles (φ, θ, ψ) of the lithium niobate or lithium tantalate constituting the piezoelectric layer are within the range of the following formula (4), formula (5), or formula (6): (0°±10°, 180° to 200°, any ψ) ... formula (4) (0°±10°, 200° to 260°, 0° to 60° (1-(θ-50) ２ / 900) １／２ ) or (0°±10°, 200° to 260°, [180°-60° (1-(θ-50) ２ / 900) １／２ ]~180°) ...Formula (5) (0°±10°, [360°-30°(1-(ψ-90) ２ / 8100) １／２ ] to 360°, any ψ) ...Equation (6)

[0238] <11> The filter device according to any one of <1> to <10>, wherein the piezoelectric layer is made of lithium niobate.

[0239] <12> The filter device according to any one of <1> to <11>, wherein the first dielectric film and the second dielectric film are made of silicon oxide.

[0240] REFERENCE SIGNS LIST 1... Acoustic wave resonator 2... Support member 2a... Cavity portion 3... Support substrate 4... Insulating layer 5, 5A... Piezoelectric layer 5X... Thickness change region 5Y... Membrane portion 5a, 5b... First and second principal surfaces 6, 6A, 6B... First dielectric film 6X... Thickness change region 7, 7A... Second dielectric film 7X... Thickness change region 8... IDT electrode 9, 9A to 9C... Laminated portion 9a... Center portion between electrode fingers 10... Filter device 13, 14... First and second signal terminals 16, 17... First and second bus bars 18, 19... First and second electrode fingers 21... Acoustic wave resonator 31... Acoustic wave resonator 32... Support member 33a to 33c... Low acoustic impedance layer 34a, 34b... High acoustic impedance layer 38... Acoustic reflection film 41... Acoustic wave resonator 43, 44... First and second reflector bus bars 45... Reflector electrode fingers 46, 47... Reflector 51... Acoustic wave resonator 58... IDT electrode 105... Wafer 115... Piezoelectric layer 118... IDT electrode A... Crossing region C... Excitation region P2 to P4... Parallel arm resonators S1 to S5... Series arm resonators

Claims

1. A filter device having a passband, comprising a plurality of resonators including an elastic wave resonator, the elastic wave resonator comprising: a piezoelectric layer having a first main surface and a second main surface facing each other; an IDT electrode provided on the first main surface of the piezoelectric layer and having a plurality of electrode fingers; a first dielectric film provided on the first main surface of the piezoelectric layer; and a second dielectric film provided on the second main surface of the piezoelectric layer; a direction perpendicular to the direction in which the plurality of electrode fingers extend is defined as an electrode finger perpendicular direction, and a region in which the adjacent electrode fingers overlap when viewed from the electrode finger perpendicular direction is a crossing region; the first dielectric film and the second dielectric film overlap the crossing region in a plan view; a relative bandwidth expressed by ((FH-FL) / FC) x 100[%] is 8.5% or more, where FH is the highest frequency in the pass band, FL is the lowest frequency in the pass band, and FC is the center frequency of the pass band; and where Tp is a thickness of the piezoelectric layer at the crossing region, Td_f is a thickness of a portion of the first dielectric film that overlaps with the crossing region in a planar view, and Td_b is a thickness of a portion of the second dielectric film that overlaps with the crossing region in a planar view, the following relationships are satisfied: 0.322<(Td_f / Tp)+(Td_b / Tp)<0.786, and |(Td_f / Tp)-(Td_b / Tp)|<0.

196.

2. The filter device according to claim 1, wherein the elastic wave resonator is configured to utilize bulk waves in thickness shear mode.

3. The filter device according to claim 1, further comprising a support member laminated on the piezoelectric layer, wherein an acoustic reflection portion is formed on the support member at a position overlapping with the plurality of electrode fingers in a planar view, and wherein Tp / p is 0.5 or less, where p is the center-to-center distance between adjacent electrode fingers.

4. The filter device according to claim 3, wherein Tp / p is 0.24 or less.

5. A filter device as described in claim 3 or 4, wherein adjacent electrode fingers overlap in a direction perpendicular to the electrode fingers, and the region between the centers of adjacent electrode fingers is an excitation region, and when the metallization ratio of the electrode fingers to the excitation region is MR, MR satisfies MR≦1.75(Tp / p)+0.

075.

6. A filter device as described in any one of claims 3 to 5, wherein the acoustic reflection portion is a hollow portion, and the support member and the piezoelectric layer are arranged so that a portion of the support member and a portion of the piezoelectric layer face each other across the hollow portion.

7. A filter device as described in any one of claims 3 to 5, wherein the acoustic reflection portion is an acoustic reflection film including a high acoustic impedance layer having a relatively high acoustic impedance and a low acoustic impedance layer having a relatively low acoustic impedance, and the support member and the piezoelectric layer are arranged so that at least a portion of the support member and at least a portion of the piezoelectric layer face each other with the acoustic reflection film in between.

8. The filter device according to claim 1, which is configured to be able to utilize plate waves.

9. A filter device according to any one of claims 1 to 8, wherein the piezoelectric layer is made of lithium niobate or lithium tantalate, and the Euler angles (φ, θ, ψ) of the lithium niobate or lithium tantalate constituting the piezoelectric layer are within the range of the following formula (1), formula (2), or formula (3): (0°±10°, 0° to 20°, any ψ) ... formula (1) (0°±10°, 20° to 80°, 0° to 60° (1-(θ-50) ２ / 900) １／２ ) or (0°±10°, 20°~80°, [180°-60°(1-(θ-50) ２ / 900) １／２ ]~180°) ...Formula (2) (0°±10°, [180°-30°(1-(ψ-90) ２ / 8100) １／２ ] ~ 180°, any ψ) ... Equation (3) 10. A filter device according to any one of claims 1 to 8, wherein the piezoelectric layer is made of lithium niobate or lithium tantalate, and the Euler angles (φ, θ, ψ) of the lithium niobate or lithium tantalate constituting the piezoelectric layer are within the range of the following formula (4), formula (5), or formula (6): (0°±10°, 180° to 200°, any ψ) ... formula (4) (0°±10°, 200° to 260°, 0° to 60° (1-(θ-50) ２ / 900) １／２ ) or (0°±10°, 200°~260°, [180°-60°(1-(θ-50) ２ / 900) １／２ ]~180°) ...Formula (5) (0°±10°, [360°-30°(1-(ψ-90) ２ / 8100) １／２ ] ~ 360°, any ψ) ... Equation (6) 11. The filter device according to any one of claims 1 to 10, wherein the piezoelectric layer is made of lithium niobate.

12. The filter device according to any one of claims 1 to 11, wherein the first dielectric film and the second dielectric film are made of silicon oxide.

Images