A design method of bulk acoustic wave resonator, filter, duplexer and multiplexer

Abstract

This invention provides a design method for bulk acoustic wave resonators, filters, duplexers, and multiplexers. The method includes: constructing an original Mason model; extracting a new model from the Mason model that accurately maps the longitudinal wave velocity and material density of piezoelectric materials to parallel resonance; and using the new model to determine the longitudinal wave velocity and material density parameters for design. This design method simplifies redundant parameters, discards irrelevant parameters, sets finite parameters for relatively fixed parameters, and establishes a mapping model for these finite parameters. Therefore, compared to existing technology models, it can more accurately determine material parameters, such as longitudinal wave velocity and material density parameters. Furthermore, this invention also provides a design method for doped bulk acoustic wave resonators. Compared to existing doping design methods, this doping design method can more stably obtain the design parameters of the resonator at the target frequency, and significantly reduces the design cycle and design cost.

Description

Technical Field

[0001] This invention relates to the field of bulk acoustic wave resonator technology, and in particular to a design method for bulk acoustic wave resonators, filters, duplexers, and multiplexers. Background Technology

[0002] The core component of a bulk acoustic wave (BAW) filter is the BAW resonator, which can also be configured as a duplexer or multiplexer. The BAW resonator comprises a stacked structure, consisting of a top electrode, a piezoelectric layer, and a bottom electrode, arranged from top to bottom. Due to its small size and low insertion loss, the BAW resonator has become one of the most widely used filters in the mobile communication field.

[0003] The design process of bulk acoustic wave (BAW) resonators typically involves more than one engineering fabrication run before the final design parameters are determined. The current design flow involves first using a Mason or MBVD model, then adjusting the Mason or MBVD model parameters based on the measured results from the engineering fabrication run. The final product design parameters are obtained when the model values ​​and measured values ​​are within a certain error range. However, this design flow is fraught with challenges due to the large number of model parameters, making parameter adjustments difficult. Furthermore, the quality of the model and the determination of parameter values ​​directly affect the design accuracy. After the first fabrication run, the center frequency of the BAW filter often differs from the design value by tens to hundreds of megahertz. This is because the design of the BAW filter involves the use of BAW resonator material parameters, which refer to the longitudinal wave velocity and material density of each layer in the stacked structure. Because the design values ​​of the material parameters for bulk acoustic wave (BAW) resonators are often reference values, there is a deviation between these values ​​and the actual values ​​after fabrication. This deviation will also lead to a certain discrepancy between the center frequency of the BAW filter formed after fabrication and its design value. The magnitude of this deviation directly affects the difficulty and timeline of subsequent design processes, and may even require multiple fabrication verifications.

[0004] Therefore, there is an urgent need in this field for a design method for bulk acoustic resonators that can reduce design deviations, save design costs, and shorten the design cycle. Furthermore, with the increasing popularity of doping processes, how to design a doping process flow based on existing undoped processes is also a problem that urgently needs to be solved in this field. Summary of the Invention

[0005] To overcome the aforementioned deficiencies in the prior art, the present invention provides a design method for a bulk acoustic resonator, a filter, a duplexer, and a multiplexer, the design method comprising:

[0006] Step S101: Construct a Mason model of a bulk acoustic resonator; wherein the bulk acoustic resonator includes a stacked structure, which includes at least a top electrode, a piezoelectric layer and a bottom electrode;

[0007] Step S102: Adjust the parameters in the Mason model to observe the changes in the resonator frequency;

[0008] The parameters that are adjusted include, but are not limited to, the thickness of each layer of the resonator stack, the static capacitance C0, and the loss resistance R. S R0, R m Longitudinal wave velocity, material density, etc.;

[0009] Step S103: Discard parameters that are unrelated to the resonant frequency, establish the correspondence between the parallel resonant frequency and the longitudinal wave velocity and material density, and form a new equivalent model;

[0010] The acoustic equivalent part of the new equivalent model is exactly the same as the initial Mason model. Therefore, the material parameters in the new model, such as longitudinal wave velocity and material density, can be applied to the initial Mason model without affecting the accuracy of the model.

[0011] Step S104: Design using the new equivalent model, verify the design using engineering fabrication, and fit and adjust the longitudinal wave velocity and material density parameters in the new equivalent model.

[0012] Step S105: Substitute the fitted and adjusted longitudinal wave velocity and material density into the initial Mason model, and fit and adjust other design parameters based on the resonator parameters measured after the engineering fabrication in step S104.

[0013] Step S106: After fitting and adjusting, the final design parameters of the bulk acoustic resonator are obtained. These parameters are then used to design resonators, filters, duplexers, or multiplexers.

[0014] According to one aspect of the invention, the Mason model includes an acoustic equivalent model and an electrical equivalent circuit;

[0015] The acoustic equivalent model includes at least a cascaded top electrode equivalent circuit, a piezoelectric layer equivalent circuit, and a bottom electrode equivalent circuit.

[0016] The electrical equivalent circuit includes a static capacitor, a loss resistor, and an electrical port.

[0017] The acoustic equivalent model and the electrical equivalent circuit are coupled through a floating ground and an ideal transformer.

[0018] According to another aspect of the present invention, the parameters adjusted in step S201 include: the thickness of each layer of the resonator, the static capacitance C0, and the loss resistance R. S R0, Rm Longitudinal wave velocity, material density.

[0019] According to another aspect of the present invention, other design parameters in step S105 include: static capacitance C0, loss resistance R. S R0, R m Effective electromechanical coupling coefficient kt 2 .

[0020] According to another aspect of the present invention, the piezoelectric layer is an undoped piezoelectric layer.

[0021] Based on the above design steps, this invention also provides a design method for doped resonators, filters, duplexers, and multiplexers, including:

[0022] Step S201: Select resonators that meet the design frequency requirements from the previous engineering wafer fabrication and extract the measured data of the resonator during the fabrication process.

[0023] Step S202: Calculate the longitudinal wave velocity and material density after doping, and substitute them into the initial Mason model to obtain the theoretical frequency shift.

[0024] Step S203: Adjust the stack thickness so that its frequency offset is equal to the theoretical frequency offset in the above steps, and the offset direction is opposite, thereby obtaining the stack thickness of the doped resonator.

[0025] Step S204: Substitute the parameters obtained in step S203 into the initial Mason model to set the doped acoustic resonator, filter, or duplexer; perform doping engineering fabrication; and fit and adjust the doping design parameters based on the measured parameters of the doped resonator after the doping engineering fabrication.

[0026] Step S205: Obtain the doping design parameters of the final doped bulk acoustic wave resonator, and use these doping design parameters to design doped bulk acoustic wave resonators, filters, duplexers or multiplexers.

[0027] According to one aspect of the present invention, the longitudinal wave velocity and material density values ​​after doping are calculated by using the atomic mass ratio of the material instead of the material density ratio to calculate the relationship between the longitudinal wave velocity and the material density of the doped material and the undoped material.

[0028] According to another aspect of the present invention, the doping design parameters fitted and adjusted in step S204 include: the thickness of each layer of the resonator, the static capacitance C0, and the loss resistance R. S R0, R m .

[0029] According to another aspect of the present invention, the doping design parameters in step S205 include: longitudinal wave velocity, material density, static capacitance C0, and loss resistance R. S R0, R m Effective electromechanical coupling coefficient kt 2 .

[0030] The method for setting up a bulk acoustic wave resonator provided by this invention first constructs a Mason model of the bulk acoustic wave resonator. Then, based on the Mason model, two parameters related to the parallel resonant frequency, namely the longitudinal wave velocity and the material density, are extracted. Based on this, a new model is re-established that can accurately map the longitudinal wave velocity and material density of the piezoelectric material in the longitudinal wave resonance state to the parallel resonance in its electrical characteristics. The longitudinal wave velocity and material density parameters for design are then determined using the new model. The determined longitudinal wave velocity and material density parameters are then input into the original Mason model for circuit design. Finally, the parameters are fitted and adjusted based on measured data from the engineering fabrication resonator to obtain the final design parameters. This invention simplifies redundant parameters, discards irrelevant parameters, and sets a finite number of relatively fixed parameters, establishing a mapping model for these finite parameters. Therefore, compared to existing technology models, it can more accurately determine material parameters, such as the longitudinal wave velocity and material density parameters.

[0031] The above design is an undoped design approach. Based on this, doped resonators can also be designed. Using samples from undoped engineering wafers, resonators in suitable frequency bands are selected. Design is based on measured parameters such as stack thickness. Assuming the stack thickness remains unchanged after doping, the relationship between the longitudinal wave velocity and density of the doped material and those of the undoped material is calculated using formulas. This is then substituted into the initial Mason model to obtain the theoretical frequency shift. The stack thickness is then directly adjusted based on the above to produce the theoretical frequency shift obtained above, but in the opposite direction, thus obtaining the original stack thickness for engineering wafer testing. The reason for using this method to determine the stack thickness instead of directly calculating it is to obtain a more stable engineering wafer of the resonator at the target frequency. Material density and sound velocity are more accurately determined through subsequent measured analysis and adjustment. Furthermore, the design methods for doped resonators, filters, duplexers, and multiplexers of this invention can be directly improved upon the undoped design, significantly reducing design costs and time compared to existing design methods. Attached Figure Description

[0032] Other features, objects, and advantages of the invention will become more apparent from the following detailed description of non-limiting embodiments with reference to the accompanying drawings:

[0033] Figure 1(a) is a design flowchart of a bulk acoustic resonator, filter, duplexer and multiplexer according to the present invention;

[0034] Figure 1(b) is a design flowchart of a doped acoustic resonator, filter, duplexer and multiplexer according to the present invention;

[0035] Figure 2 It is the Mason equivalent model of a bulk acoustic resonator;

[0036] Figure 3(a) is Figure 2 The equivalent circuit structure of the Mason model shown;

[0037] Figure 3(b) shows the equivalent circuit structure of the new model of the present invention;

[0038] Figure 4 This is a new equivalent model of the present invention;

[0039] The same or similar reference numerals in the accompanying drawings represent the same or similar parts. Detailed Implementation

[0040] To enable those skilled in the art to better understand the present application, the technical solutions in the embodiments of the present application will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present application, and not all embodiments. Based on the embodiments in the present application, all other embodiments obtained by those of ordinary skill in the art without creative effort should fall within the scope of protection of the present application.

[0041] It should be noted that the terms "first," "second," etc., in the specification, claims, and accompanying drawings of this application are used to distinguish similar objects and are not necessarily used to describe a specific order or sequence. It should be understood that such data can be interchanged where appropriate so that the embodiments of this application described herein can be implemented in orders other than those illustrated or described herein. Furthermore, the terms "comprising" and "having," and any variations thereof, are intended to cover a non-exclusive inclusion; for example, a process, method, system, product, or apparatus that includes a series of steps or apparatuses is not necessarily limited to those explicitly listed, but may include other steps or apparatuses not explicitly listed or inherent to such processes, methods, products, or apparatuses.

[0042] This invention provides a design method for a bulk acoustic wave resonator. Referring to Figure 1(a), Figure 1(a) is a flowchart of a design method for a bulk acoustic wave resonator according to a specific embodiment of the present invention. As shown in Figure 1(a), the method includes the following steps:

[0043] Step S101: Construct the Mason model of the bulk acoustic resonator;

[0044] The bulk acoustic resonator includes a stacked structure, which includes at least a top electrode, a piezoelectric layer, and a bottom electrode.

[0045] Step S102: Adjust the parameters in the Mason model to observe the changes in the resonator frequency;

[0046] The parameters that are adjusted include, but are not limited to, the thickness of each layer of the resonator stack, the static capacitance C0, and the loss resistance R. S R0, R m Longitudinal wave velocity, material density, etc.;

[0047] In step S103, parameters unrelated to the resonant frequency are discarded, and the correspondence between the parallel resonant frequency, longitudinal wave velocity, and material density is established to form a new equivalent model.

[0048] In step S104, engineering fabrication verification is performed, and the longitudinal wave velocity and material density parameters are fitted and adjusted in the new equivalent model.

[0049] In step S105, the fitted and adjusted longitudinal wave velocity and material density are substituted into the initial Mason model, and other design parameters are fitted and adjusted according to the resonator parameters measured after the engineering fabrication in step S104.

[0050] The other design parameters include, but are not limited to, static capacitance C0 and loss resistance R. S R0, R m Effective electromechanical coupling coefficient kt 2 .

[0051] In step S106, after fitting and adjustment, the final design parameters of the bulk acoustic resonator are obtained, and these parameters are used to design the resonator, filter, duplexer and multiplexer.

[0052] Specifically, in step S101, a Mason model of the bulk acoustic wave resonator is constructed using simulation software (such as ADS). The bulk acoustic wave resonator includes a stacked structure, which at least includes a top electrode, a piezoelectric layer, and a bottom electrode. The bulk acoustic wave resonator can be an air-cavity type thin-film bulk acoustic wave resonator, a Bragg reflection type thin-film bulk acoustic wave resonator, or a reverse-etched type thin-film bulk acoustic wave resonator. These three types of thin-film bulk resonators will be collectively referred to as bulk acoustic wave resonators in the following text.

[0053] The Mason model of a bulk acoustic wave (SAW) resonator includes an acoustic equivalent model and an electrical equivalent circuit. Regardless of whether the SAW resonator is an air-cavity type, a Bragg-reflector type, or a reverse-etched type, its electrical equivalent circuit is the same, consisting of electrical ports, static capacitance, and loss resistance. The acoustic equivalent model varies depending on the SAW resonator's stack-up structure. Specifically, the acoustic equivalent model is formed by cascading the equivalent circuits of each layer in the SAW resonator's stack-up structure. If the SAW resonator's stack-up structure, from top to bottom, consists of a top electrode, a piezoelectric layer, and a bottom electrode, then the acoustic equivalent model in the Mason model of this SAW resonator includes the cascaded equivalent circuits of the top electrode, the piezoelectric layer, and the bottom electrode. If the stacked structure of a bulk acoustic wave resonator, from top to bottom, consists of a passivation layer, a top electrode, a piezoelectric layer, a bottom electrode, and a seed layer, then the acoustic equivalent model in the Mason model of this bulk acoustic wave resonator includes cascaded equivalent circuits for the passivation layer, top electrode, piezoelectric layer, bottom electrode, and seed layer. Similarly, if the stacked structure of the bulk acoustic wave resonator, from top to bottom, consists of a top electrode, a piezoelectric layer, a bottom electrode, and a Bragg reflector layer, then the acoustic equivalent model in the Mason model of this bulk acoustic wave resonator includes cascaded equivalent circuits for the top electrode, piezoelectric layer, bottom electrode, and Bragg reflector layer. The acoustic equivalent model and the electrical equivalent circuit are coupled through a floating ground and an ideal transformer.

[0054] In this embodiment, the bulk acoustic wave resonator is an air cavity type thin-film bulk acoustic wave resonator. Specifically, the bulk acoustic wave resonator includes a substrate and a stacked structure located on the substrate. The stacked structure consists of a top electrode, a piezoelectric layer, and a bottom electrode from top to bottom, wherein a cavity is formed between the bottom electrode and the substrate.

[0055] The Mason model of the bulk acoustic resonator is as follows: Figure 2 As shown, it includes an acoustic equivalent model 10 and an electrical equivalent circuit 20, wherein the acoustic equivalent model 10 and the electrical equivalent circuit 20 are coupled through a floating ground and an ideal transformer (represented by N in the figure).

[0056] The acoustic equivalent model 10 includes a cascaded top electrode equivalent circuit 101, a piezoelectric layer equivalent circuit 102, and a bottom electrode equivalent circuit 103. One end of the top electrode equivalent circuit 101 is grounded, and the other end is connected to one end of the piezoelectric layer equivalent circuit 102. The other end of the piezoelectric layer equivalent circuit 102 is connected to one end of the bottom electrode equivalent circuit 103, and the other end of the bottom electrode equivalent circuit 103 is grounded. (See figure.) , , The acoustic impedances of the top electrode equivalent circuit 101, the piezoelectric layer equivalent circuit 102, and the bottom electrode equivalent circuit 103 are respectively represented. , , The electrical lengths of the top electrode equivalent circuit 101, the piezoelectric layer equivalent circuit 102, and the bottom electrode equivalent circuit 103 are respectively represented.

[0057] The electrical equivalent circuit 20 includes electrical ports (including a first electrical port P1 and a second electrical port P2), static capacitors (including a first static capacitor C0 and a second static capacitor -C0), and loss resistors (including a first loss resistor R). m The second loss resistor R0 and the third loss resistor R s ), where R m R0 and R s This reflects the different types of losses in the bulk acoustic resonator. Specifically, the first loss resistance R... m The first loss resistor R0 reflects the mechanical loss of the bulk acoustic wave resonator, and the second loss resistor R0 reflects the transverse leakage loss of the bulk acoustic wave resonator. The third loss resistor R s This reflects the electrode loss of the bulk acoustic resonator.

[0058] The specific structure of the electrical equivalent circuit 20 is as follows: the first electrical port P1 and the first loss resistor R m One end is connected, the second electrical port P2 is connected to the third loss resistor R s One end is connected to the third loss resistor R. s The other end is connected to one end of the second static capacitor -C0, one end of the second loss resistor R0 is connected to the first electrical port P1, and the other end of the second loss resistor R0 is connected to one end of the first static capacitor C0, and the other end of the first static capacitor C0 is connected to the third loss resistor R s The node between the second static capacitor -C0, and the first loss resistor R m The other end of the capacitor is connected to the other end of the second static capacitor -C0 and to the two ends of the winding on one side of the ideal transformer N, respectively. The two ends of the winding on the other side of the ideal transformer N are grounded and connected to the floating ground, respectively.

[0059] In step S102, the parameters in the Mason model are adjusted to observe the change in the resonant frequency of the resonator. The specific steps are as follows:

[0060] Adjust the control variables in the Mason model individually, such as the stack thickness, static capacitance C0, -C0, and loss resistance R. m R0, R s and electromechanical coupling coefficient Kt 2The parameters such as longitudinal wave velocity (v) and material density (ρ) are checked to see if the resonant frequency changes. The resonant frequencies checked at this stage can include both series and parallel resonant frequencies of the resonator. In normal design processes, the importance of each parameter in the Mason model is not entirely the same; different parameters may affect the same performance of the resonator, and their contributions to performance also differ. Some parameters in the Mason model are mainly determined by the material itself, such as material density and longitudinal wave velocity, while others need to be continuously adjusted during the design process. However, all these parameters, whether essentially fixed or changing with the design, have parameter errors between the design and the actual fabrication. Therefore, the core of this method is to reduce the correlation between the number of variables and the resonant frequency, prioritizing the accurate determination of relatively fixed parameter variables. Longitudinal wave velocity and material density are relatively stable parameters during the design process. Therefore, the main purpose of adjusting variables and checking the resonator's changes in this step is to obtain a precise mapping between longitudinal wave velocity, material density, and a specific resonant frequency.

[0061] In step S103, parameters unrelated to the resonant frequency are discarded, and a new equivalent model is established to represent the relationship between the parallel resonant frequency, the longitudinal wave velocity, and the material density; the specific steps are as follows:

[0062] The resonant frequencies of resonators are generally classified into series resonant frequencies and parallel resonant frequencies. During the modification of the control variables in the Mason model, a close mapping relationship was found between the parallel resonant frequency and the longitudinal wave velocity, material density, and stack thickness. The following theoretical analysis illustrates this mapping relationship between the parallel resonant frequency and these parameters.

[0063] (1-1)

[0064] Wherein, Equation 1-1 is the electrical impedance expression of the Mason model, aluminum carbide (AlN) is the piezoelectric layer material in the bulk acoustic wave resonator, molybdenum (Mo) is the material of the upper and lower electrodes in the bulk acoustic wave resonator; C0 is the static capacitance of the bulk acoustic wave resonator, and Z... t (Z) t =jZ Mo tanθ Mo ) and Z b (Zb=jZ) Mo tanθ' Mo Z is the input impedance viewed from the upper and lower boundaries of aluminum nitride, respectively, looking upwards and downwards. AlN / Z Mo This is the acoustic characteristic impedance of aluminum nitride and molybdenum. θ AlN / θ Mo / θ' MoIt refers to the phase shift that occurs during the propagation of sound waves through the piezoelectric layer, top electrode, and bottom electrode. n represents... Figure 2 The turns ratio of the coil in an ideal transformer N is specifically expressed as: n = 2θ AlN / (kt 2 ωC0Z AlN (where kt) 2 It is the electromechanical coupling coefficient of the piezoelectric layer material.

[0065] Among them, the loss resistance R s R0, R m These represent the ohmic loss caused by the electrodes, the dielectric loss due to the dielectric material, and the mechanical loss due to irregular mechanical vibration boundaries, respectively. These three resistors only affect the quality factor of the resonator and do not affect its resonant frequency. Therefore, Equation 1-1 can be simplified to Equation 1-2.

[0066] (1-2)

[0067] Where θ AlN The specific calculation method is as follows, k AlN Let d be the lossless propagation constant of the piezoelectric layer material. AlN It is the thickness of the piezoelectric layer, V AlN It is the longitudinal wave velocity of aluminum nitride.

[0068] (1-3)

[0069] For ease of derivation, the FBAR impedance expression in formula (1-2) is converted into an ideal FBAR expression. An ideal FBAR is characterized by having only a piezoelectric layer, and the resonator's lateral dimension is much larger than its longitudinal dimension, i.e., Z... t and Z b The values ​​are all approximately zero (both the top and bottom electrodes are removed), resulting in the following equation 1-4:

[0070] (1-4)

[0071] According to the impedance curve of the resonator, when f is the parallel resonant frequency, i.e., f = f p At that time, Z FBAR As it approaches infinity, at this point, θ AlN |f p =(2n+1)π / 2,tanθ AlN / θ AlN Approaching infinity. Combining equations 1-3 and 1-4, fp can be successfully separated. The separation formula is as follows:

[0072] (1-5)

[0073] Where m = 0, 1, 2, 3... From equation (1-5), it can be seen that f p The value of is independent of the electromechanical coupling coefficient, and only depends on V. AlN and d ALN Related, V AlN This refers to the longitudinal wave velocity of aluminum nitride. AlN, a commonly used piezoelectric material in bulk acoustic resonators, has a hexagonal wurtzite structure. Sound waves propagating in a piezoelectric medium can be considered as a linear combination of plane waves propagating in two directions. When the sound wave propagates along the c-axis, if the z-axis of the spatial rectangular coordinate system is chosen to be parallel to the c-axis of the crystal, the longitudinal wave velocity of aluminum nitride can be considered as:

[0074] (1-6)

[0075] Where C 33 e is the elastic stiffness constant of the material. 33 The piezoelectric stress constant of the material is Let V be the dielectric constant of the material, and ρ be the density of the material. Therefore, V... AlN It is closely related to the material density of ALN. Therefore, f p There is an accurate mapping relationship between it and the longitudinal wave velocity, material density, and layer thickness.

[0076] The above theoretically explains the correspondence between the parallel resonant frequency and the longitudinal wave velocity, material density, and layer thickness. The following section uses this theory to establish a new equivalent model.

[0077] Figure 3(a) is Figure 2 The conventional equivalent circuit structure of the Mason model includes a first static capacitor C0, a second static capacitor -C0, and a first loss resistor R. m Second loss resistor R0, third loss resistor R s The system consists of a first capacitor C1, a first inductor L1, an ideal transformer N, a first electrical port P1, and a second electrical port P2. The first electrical port P1 is connected to the first loss resistor R. m One end is connected, the second electrical port P2 is connected to the third loss resistor R s One end is connected to the third loss resistor R. s The other end is connected to one end of the second static capacitor -C0, one end of the second loss resistor R0 is connected to the first electrical port P1, and the other end of the second loss resistor R0 is connected to one end of the first static capacitor C0, and the other end of the first static capacitor C0 is connected to the third loss resistor R s The node between the second static capacitor -C0, and the first loss resistor R mThe other end of the first capacitor is connected to the other end of the second static capacitor -C0, which is connected to the two ends of one winding of the ideal transformer N. The two ends of the other winding of the ideal transformer N are connected to the first capacitor C1 and the first inductor L1, respectively. The first capacitor C1 and the first inductor L1 are connected in series. Based on the above derivation, the conventional equivalent circuit structure in Figure 3(a) can be simplified to the circuit structure shown in Figure 3(b). In the circuit of Figure 3(b), there are a first capacitor C1, a first inductor L1, an ideal transformer N, a first electrical port P1, and a second electrical port P2. One end of the first capacitor C1 is connected to one end of the first inductor L1. The other ends of the first capacitor C1 and the first inductor L1 are connected to the two ends of one winding of the ideal transformer N, respectively. The two ends of the other winding of the ideal transformer N are connected to the first electrical port P1 and the second electrical port P2, respectively. It should be noted that Figures 3(a) and 3(b) have the same resonant frequency.

[0078] Finally, considering that the series connection of the first inductor L1 and the first capacitor C1 can be equivalent to the acoustic equivalent model 10 in the Mason model of a bulk acoustic resonator, the acoustic equivalent model 10 in the Mason model is used to simplify the first inductor L1 and the first capacitor C1 in the circuit, thereby obtaining... Figure 4 The new equivalent model is shown. For example... Figure 4 As shown, the model includes the same acoustic equivalent model 10 as the Mason model, which is connected to the first electrical port P1 and the second electrical port P2 via a floating ground and an ideal transformer N. Because... Figure 4 The model shown does not include any capacitors, resistors, or inductors. Its resonant frequency is only related to the material and structural parameters of the bulk acoustic resonator. The material parameters of the bulk acoustic resonator refer to the longitudinal wave velocity and material density of each layer in the stacked structure, while the structural parameters refer to the thickness of each layer in the stacked structure. This is consistent with the theoretical analysis above.

[0079] from Figure 2 , Figure 4 It can be seen that the acoustic equivalent parts of the two models are exactly the same, therefore the new model ( Figure 4 Material parameters such as longitudinal wave velocity and material density in the model can be fully applied to the old model. Figure 2 This can be removed from the model without affecting its accuracy.

[0080] In step S104, the design is performed using a new equivalent model. After the engineering fabrication, parameter fitting and adjustment are conducted to obtain the fitted and adjusted material longitudinal wave velocity and material density; specifically:

[0081] A filter was designed using the obtained new equivalent model. During the engineering fabrication process, inline measurements were performed on the resonator to obtain the measured parameters of the filter. The new model was then used to fit and adjust material parameters such as longitudinal wave velocity and material density, thus obtaining the adjusted longitudinal wave velocity and material density parameters. The parameter fitting and adjustment process involves adjusting specific parameter values ​​in the new equivalent model so that the simulated value of the parallel resonant frequency of the new equivalent model and the measured parallel resonant frequency value of the resonator meet the minimum error requirement.

[0082] In step S105, the fitted and adjusted longitudinal wave velocity and material density are substituted into the Mason model, and other design parameters are fitted and adjusted according to the measured resonator parameters after the engineering fabrication in step S104.

[0083] During the fitting process, the longitudinal wave velocity and material density remain constant, while other design parameters are optimized. These other design parameters include, but are not limited to, the static capacitance C0 and the loss resistance R. S R0, R m Effective electromechanical coupling coefficient kt 2 ;

[0084] In step S106, after optimization, the final design parameters of the bulk acoustic wave resonator are obtained, and these parameters are used to design subsequent resonators, filters, duplexers, and multiplexers.

[0085] The above describes a design method for a bulk acoustic wave resonator (BAW) using an undoped piezoelectric layer, such as aluminum nitride (ALN). However, with continuous technological advancements, in pursuit of higher electromechanical coupling coefficients and smaller chip structures, more and more BAW resonators are employing doping processes to fabricate piezoelectric materials. For example, scandium-doped (Sc) aluminum nitride is commonly used as the piezoelectric material. To save costs and shorten the development cycle, developing an accurate doping process design flow based on the existing undoped design process is a problem that urgently needs to be solved by those skilled in the art.

[0086] The following examples use scandium-doped aluminum nitride (AlScN) as a piezoelectric material to illustrate how to improve the design of the doping process using undoped design methods.

[0087] See Figure 1(b), which is a flowchart of the design of a doped acoustic resonator.

[0088] In step S201, resonators that meet the design frequency requirements are selected from the previous engineering wafer fabrication, and the measured data of the resonator during the fabrication process are extracted.

[0089] In the design process of bulk acoustic wave filters, designers will select a variety of different stack thicknesses to make engineering fabrication sheets according to actual needs, so as to facilitate parameter fitting and adjustment and various design projects in the later stage. The number of engineering fabrication sheets can be one or multiple times.

[0090] In the previous engineering wafer fabrication, resonators that meet the frequency requirements of our design are selected. Inline measurement data of the resonator during the fabrication process are extracted. In this embodiment, the required frequency is 4.8-5G. The inline measurement data includes the thickness of each stack structure, such as the thickness of the piezoelectric layer, the thickness of the upper electrode, the thickness of the lower electrode, etc., which are well known to those skilled in the art. The specific values ​​mentioned above are just examples. This method is also applicable to other frequency bands and other measurement data.

[0091] In step S202, the longitudinal wave velocity and material density after doping are calculated and substituted into the initial Mason model to obtain the theoretical frequency offset.

[0092] This embodiment takes 9% scandium (Sc) doped into piezoelectric material aluminum nitride (AlN) as an example, that is, the ratio of Al, Sc and N in AlScN is 0.91:0.09:1. Based on the thickness of the stacked structure measured by Inline in step S201, it is assumed that the thickness remains unchanged after doping. The atomic mass ratio is used to roughly represent the density ratio. The specific density calculation process is shown in formula (1-7).

[0093] (1-7)

[0094] The atomic mass of Al is 26.9, that of N is 14.01, and that of Sc is 44.96.

[0095] Therefore, the density relationship between doped and undoped materials can be obtained:

[0096] (1-8)

[0097] Furthermore, the relationship between the sound velocity of doped materials and the sound velocity of undoped materials can be calculated using formula (1-9):

[0098] (1-9)

[0099] After obtaining the estimated values ​​of the doped material parameters, the new material density parameters and longitudinal wave velocity parameters are substituted into the initial Mason model, along with the thickness of the stacked structure measured in the inline step above. The resonant frequency of the doped model at the desired frequency band will shift to a lower frequency by a certain value (doping affects the material properties, resulting in a certain frequency deviation between the undoped and doped frequencies of the same thickness). This embodiment uses the N79 frequency band as an example, showing that the resonant frequency of the doped model at 4.7 GHz will shift to a lower frequency by 287.94 MHz. Of course, those skilled in the art know that the direction and magnitude of frequency shift differ depending on the doped material, and using other doped materials may also result in a frequency shift to higher frequencies; these will not be illustrated here.

[0100] In step S203, the stack thickness is adjusted so that its frequency offset is equal to the theoretical frequency offset in the above steps, and the offset direction is opposite, thereby obtaining the stack thickness of the doped resonator; the specific steps include:

[0101] After obtaining the theoretical frequency offset, the electrode thickness is reduced in the Mason model (the electrode thickness is reduced when the theoretical frequency offset shifts to lower frequencies and increased when it shifts to higher frequencies; in this embodiment, the offset is to lower frequencies, so the electrode thickness needs to be reduced to increase the frequency) to increase the fundamental frequency of the resonator by the aforementioned frequency offset. The reduced stack thickness is then used as the base stack thickness for the engineering fabrication of the scandium-doped resonator. In this embodiment, by reducing the thickness of the upper electrode, the fundamental frequency of the resonator is increased by 287.94 MHz. Of course, those skilled in the art will know that the fundamental frequency can also be increased by adjusting the thickness of the lower electrode, the thickness of the piezoelectric layer, or different combinations between the upper and lower electrodes and the piezoelectric layer; whether the thickness is reduced or increased depends on the direction of the theoretical frequency offset. These will not be elaborated upon here. It should also be noted that the opposite offset direction here means that if the frequency offset in step S202 is offset towards lower frequencies, then the direction of the resonant frequency offset in step S203 by adjusting the stack thickness is offset towards higher frequencies; similarly, if the frequency offset in step S202 is offset towards higher frequencies, then the direction of the resonant frequency offset in step S203 by adjusting the stack thickness is offset towards lower frequencies.

[0102] In step S204, the above parameters are substituted into the initial Mason model to set the resonator, filter, or duplexer, and the doped engineering fabrication is performed. The design parameters are then fitted and adjusted based on the measured parameters of the doped resonator after the engineering fabrication, including:

[0103] The stack thickness, longitudinal wave velocity, and material density parameters obtained in step S203 are substituted into the initial Mason model for circuit design. The circuit includes resonators, filters, duplexers, multiplexers, etc. Then, based on the measured data of the doped resonator after engineering fabrication, the design parameters are further fitted and adjusted. The fitted and adjusted design parameters include, but are not limited to: C0, R... S R0, R m Kt 2 .

[0104] In step S205, the design parameters of the final doped bulk acoustic wave resonator are obtained, including: using the fitted and adjusted parameters obtained in step S204 as the final design parameters, and using these parameters to design the circuits of the final resonator, filter, duplexer, and multiplexer. The design parameters include, but are not limited to, the static capacitance C0 and the loss resistance R. S R0, R m Effective electromechanical coupling coefficient kt 2 .

[0105] It will be apparent to those skilled in the art that the present invention is not limited to the details of the exemplary embodiments described above, and that the invention can be implemented in other specific forms without departing from the spirit or essential characteristics of the invention. Therefore, the embodiments should be considered illustrative and non-limiting in all respects, and the scope of the invention is defined by the appended claims rather than the foregoing description. Thus, all variations falling within the meaning and scope of equivalents of the claims are intended to be embraced within the present invention. No reference numerals in the claims should be construed as limiting the scope of the claims. Furthermore, it is clear that the word "comprising" does not exclude other components, units, or steps, and the singular does not exclude the plural. Multiple components, units, or devices recited in the system claims may also be implemented by a single component, unit, or device in software or hardware.

[0106] The method for setting up a bulk acoustic wave resonator provided by this invention first constructs a Mason model of the bulk acoustic wave resonator. Then, based on the Mason model, two parameters related to the parallel resonant frequency, namely the longitudinal wave velocity and the material density, are extracted. Based on this, a new model is re-established that can accurately map the longitudinal wave velocity and material density of the piezoelectric material in the longitudinal wave resonance state to the parallel resonance in its electrical characteristics. The longitudinal wave velocity and material density parameters for design are then determined using the new model. The determined longitudinal wave velocity and material density parameters are then input into the original Mason model for circuit design. Finally, the parameters are fitted and adjusted based on measured data from the engineering fabrication resonator to obtain the final design parameters. This invention simplifies redundant parameters, discards irrelevant parameters, and sets a finite number of relatively fixed parameters, establishing a mapping model for these finite parameters. Therefore, compared to existing technology models, it can more accurately determine material parameters, such as the longitudinal wave velocity and material density parameters.

[0107] The above design is an undoped design approach. Based on this, doped resonators can also be designed. Using samples from undoped engineering wafers, resonators in suitable frequency bands are selected. Design is based on measured parameters such as stack thickness. Assuming the stack thickness remains unchanged after doping, the relationship between the longitudinal wave velocity and density of the doped material and those of the undoped material is calculated using formulas. This is then substituted into the initial Mason model to obtain the theoretical frequency shift. The stack thickness is then directly adjusted based on the above to produce the theoretical frequency shift obtained above, but in the opposite direction, thus obtaining the original stack thickness for engineering wafer testing. The reason for using this method to determine the stack thickness instead of directly calculating it is to obtain a more stable engineering wafer of the resonator at the target frequency. Material density and sound velocity are more accurately determined through subsequent measured analysis and adjustment. Furthermore, the design methods for doped resonators, filters, duplexers, and multiplexers of this invention can be directly improved upon the undoped design, significantly reducing design costs and time compared to existing design methods.

[0108] The above-disclosed embodiments are merely some preferred embodiments of the present invention and should not be construed as limiting the scope of the present invention. Therefore, any equivalent variations made in accordance with the claims of the present invention are still within the scope of the present invention.

Claims

1. A design method for a bulk acoustic resonator, wherein, This design methodology includes: Step S101: Construct the Mason model of the bulk acoustic resonator; The bulk acoustic resonator includes a stacked structure, which includes at least a top electrode, a piezoelectric layer, and a bottom electrode. Step S102: Adjust the parameters in the Mason model to observe the changes in the resonator frequency; Step S103: Discard parameters that are not related to the resonant frequency, and convert the impedance expression of the bulk acoustic resonator into the impedance expression of an ideal bulk acoustic resonator with only a piezoelectric layer. Establish the correspondence between the parallel resonant frequency and the longitudinal wave velocity and material density of the piezoelectric layer to form a new equivalent model. Step S104: Engineering fabrication verification, fitting and adjusting the longitudinal wave velocity and material density parameters in the new equivalent model; Step S105: Substitute the fitted and adjusted longitudinal wave velocity and material density into the initial Mason model, and fit and adjust other design parameters based on the resonator parameters measured after the engineering fabrication in step S104. Step S106: After fitting and adjusting, the final design parameters of the bulk acoustic resonator are obtained. These parameters are then used to design resonators, filters, duplexers, or multiplexers.

2. The method according to claim 1, wherein: The Mason model includes an acoustic equivalent model and an electrical equivalent circuit; The acoustic equivalent model includes at least a cascaded top electrode equivalent circuit, a piezoelectric layer equivalent circuit, and a bottom electrode equivalent circuit. The electrical equivalent circuit includes a static capacitor, a loss resistor, and an electrical port. The acoustic equivalent model and the electrical equivalent circuit are coupled through a floating ground and an ideal transformer.

3. The method according to claim 1, wherein, The parameters adjusted in step S102 include: the thickness of each layer of the resonator, the static capacitance C0, and the loss resistance R. S Loss resistance R0, loss resistance R m Longitudinal wave velocity, material density.

4. The method according to claim 1, wherein, Other design parameters in step S105 include: static capacitance C0, loss resistance R. S Loss resistance R0, loss resistance R m Effective electromechanical coupling coefficient kt 2 .

5. The method according to claim 1, wherein, The parallel resonant frequency is: Where m = 0, 1, 2, 3..., d AlN It is the thickness of the piezoelectric layer, V AlN It is the longitudinal wave velocity of the piezoelectric layer.

6. The method according to claim 1, wherein, The piezoelectric layer is an undoped piezoelectric layer.

7. The method according to any one of claims 1-5, further comprising the following steps: Step S201: Select resonators that meet the design frequency requirements from the previous engineering wafer fabrication and extract the measured data of the resonator during the fabrication process. Step S202: Calculate the longitudinal wave velocity and material density after doping, and substitute them into the initial Mason model to obtain the theoretical frequency shift. Step S203: Adjust the stack thickness so that its frequency offset is equal to the theoretical frequency offset in the above steps, and the offset direction is opposite, thereby obtaining the stack thickness of the doped resonator. Step S204: Substitute the parameters obtained in step S203 into the initial Mason model to set the doped acoustic resonator, filter or duplexer, perform doping engineering fabrication, and fit and adjust the doping design parameters according to the measured parameters of the doped resonator after the doping engineering fabrication. Step S205: Obtain the doping design parameters of the final doped bulk acoustic wave resonator, and use these doping design parameters to design doped bulk acoustic wave resonators, filters, duplexers or multiplexers.

8. The method according to claim 7, wherein, The longitudinal wave velocity and material density values ​​after doping are calculated by using the atomic mass ratio of the material instead of the material density ratio to calculate the relationship between the longitudinal wave velocity and the material density of the doped material and the undoped material.

9. The method according to claim 7, wherein, The doping design parameters fitted and adjusted in step S204 include: the thickness of each layer of the resonator, the static capacitance C0, and the loss resistance R. S Loss resistance R0, loss resistance R m .

10. The method according to claim 7, wherein, The doping design parameters in step S205 include: longitudinal wave velocity, material density, static capacitance C0, and loss resistance R. S Loss resistance R0, loss resistance R m Effective electromechanical coupling coefficient kt 2 .

Images