WO2023241034A1

WO2023241034A1 - Simulation model optimization method for bulk acoustic resonator

Info

Publication number: WO2023241034A1
Application number: PCT/CN2023/071378
Authority: WO
Inventors: 李国强
Original assignee: 河源市艾佛光通科技有限公司
Priority date: 2022-06-16
Filing date: 2023-01-09
Publication date: 2023-12-21
Also published as: CN115169272A

Abstract

The present application belongs to the field of third-generation semiconductor materials and devices. Disclosed is a simulation model optimization method for a bulk acoustic resonator. On the basis of an established physical simulation model of a film bulk acoustic resonator, a resistance parameter in a behavior simulation model is extracted from a test curve and is connected to the physical simulation model in series, so as to perfect the physical simulation model by means of correcting and optimizing same, and fine data tuning is then performed on the resistance parameter, so as to improve the fitting degree of the optimized physical simulation model at a series-parallel resonance point. By means of the method, simulation data is accurate, the accuracy of a Q value at the series-parallel resonance point is improved, and a subsequent simulation design executed on this basis is more real and effective.

Description

A simulation model optimization method for bulk acoustic wave resonators

This application claims priority to the Chinese patent application filed with the China Patent Office on June 16, 2022, with application number 202210686739.7 and the invention title "A simulation model optimization method for bulk acoustic wave resonators", the entire content of which is incorporated by reference. in this application.

Technical field

This application belongs to the field of third-generation semiconductor materials and devices, and particularly relates to a simulation model optimization method applied to ADS software bulk acoustic wave resonators.

Background technique

The mainstream frequency bands of global mobile communications are mainly concentrated in the range of 0.9-5.5GHz. Using filters can select the required frequencies to pass specific frequency components in the signal while greatly attenuating other frequency components. At present, with the arrival of the 5G network era based on 4G LTE technology, the operating frequencies required by smartphones and various mobile communication devices are constantly increasing, and the supported radio frequency bands are also increasing. At the same time, the size requirements for antennas and transceivers are becoming more stringent. In order to avoid mutual interference and call drops between signals of different frequencies, sufficient filters must be configured in each communication device to isolate the radio frequency band to ensure accurate transmission and reception of signals. With the large-scale use of available spectrum, newly allocated frequency bands are getting closer and closer to existing frequency bands, and the protected frequency bands are getting narrower, making it increasingly difficult to isolate frequency bands. In recent years, Film Bulk Acoustic Resonator (FBAR) has the advantages of high frequency, miniaturization, high performance, low power consumption, high power capacity, etc., and its manufacturing process is compatible with IC process and can be integrated. , which is conducive to reducing device power consumption and reducing device size. It is currently the only RF front-end filter that can be integrated. Therefore, FBAR filters will become the core components of future 5G high-frequency communications.

The thin film bulk acoustic resonator mainly consists of three parts: the substrate, the acoustic reflection layer, and the sandwich piezoelectric oscillation stack composed of upper and lower electrodes and a piezoelectric film sandwiched between the upper and lower electrodes. When a radio frequency RF voltage is applied between the two electrodes, an alternating electric field is generated in the piezoelectric oscillation stack. Through the inverse piezoelectric effect of the piezoelectric film, part of the electrical energy is converted into bulk acoustic waves that propagate along the thickness direction of the film and are transmitted between the two electrodes. Reflected back and forth, resonance will occur when the propagation of bulk acoustic waves in the piezoelectric oscillation stack is exactly half a wavelength or an odd multiple of a half wavelength, that is, the fundamental frequency wavelength of the resonance is approximately equal to twice the thickness of the piezoelectric oscillation stack.

The commonly used physical simulation model studies the film thickness, basic physical properties, and electrode facing area of each layer of material under ideal conditions. Since the piezoelectric effect transduction of the piezoelectric layer and the electrode layer, support layer, and lining are ignored, The mechanical loss caused by sound wave reflection in the bottom ordinary acoustic layer, as well as the dielectric loss and electrode impedance loss of the non-electrode film layer, cause the physical simulation model to deviate from the real value.

Contents of the invention

In order to solve the above technical problems, this application provides the following solutions.

This application provides a simulation model optimization method for a bulk acoustic wave resonator, including:

Select the set of frequency points in the smooth part of the bulk acoustic wave resonator test curve, and calculate the first capacitor, second capacitor, first inductor, first resistor, and second resistor based on the series resonant frequency and parallel resonant frequency and their respective quality factors. and a third resistor;

According to the first capacitor, the second capacitor, the first inductor, the first resistor, the second resistor and the third resistor, a behavioral simulation model circuit is drawn in the ADS simulation software, and the Compare the above test curves;

Substitute the behavioral simulation model circuit into the original physical simulation model, and set geometric parameters and physical parameters for the original physical simulation model;

Fine-tune the first resistor, the second resistor and the third resistor, determine an optimized physical simulation model based on the fine-tuned resistance values, and use the optimized physical simulation model to perform subsequent simulation work.

Compared with the existing technology, the beneficial effects of this application are:

In order to obtain a physical simulation model that is closer to the real value and correct the gap between the mechanical loss, dielectric loss and electrode impedance loss and the real value, we extracted the parameters in the behavioral simulation model from the test raw data and performed curve fitting, and then These values are used for data fine-tuning to ensure accurate results, and then the fine-tuning parameters are used to correct and optimize the physical simulation model. This method makes the simulation data accurate, improves the accuracy of the Q value at the series-parallel resonance point, and makes the subsequent simulation design more realistic and effective.

Description of the drawings

Figure 1 is the original physical simulation model;

Figure 2 is the behavioral simulation model;

Figure 3 is the optimized physical simulation model;

Figure 4 is a comparison of the simulation results of the optimized physical simulation model, behavioral simulation model and test data;

Figure 5 is a comparison of the simulation results of the optimized physical simulation model and the original physical simulation model.

Detailed ways

In order to make the purpose, technical solutions and advantages of the embodiments of the present application clearer, the technical solutions in the embodiments of the present application will be clearly and completely described below in conjunction with the drawings in the embodiments of the present application. Obviously, the described embodiments These are only some of the embodiments of this application, not all of them. Based on the embodiments in this application, all other embodiments obtained by those of ordinary skill in the art without creative efforts fall within the scope of protection of this application.

The terms "first", "second", "third", "fourth", etc. (if present) in the description and claims of this application and the above-mentioned drawings are used to distinguish similar objects and are not necessarily used for Describe a specific order or sequence. It is to be understood that the data so used are interchangeable under appropriate circumstances so that the embodiments of the application described herein can be practiced in sequences other than those illustrated or described herein.

It should be understood that in the various embodiments of the present application, the size of the sequence numbers of each process does not mean the order of execution. The execution order of each process should be determined by its functions and internal logic, and should not be determined by the execution order of the embodiments of the present application. The implementation process constitutes no limitation.

It should be understood that in this application, "comprising" and "having" and any variations thereof are intended to cover non-exclusive inclusion. For example, a process, method, system, product or equipment that includes a series of steps or units is not necessarily limited to Those steps or elements that are expressly listed may instead include other steps or elements that are not expressly listed or that are inherent to the process, method, product or apparatus.

It should be understood that in this application, "plurality" means two or more. "And/or" is just an association relationship that describes related objects, indicating that three relationships can exist. For example, and/or B can mean: A alone exists, A and B exist simultaneously, and B alone exists. . The character "/" generally indicates that the related objects are in an "or" relationship. "Includes A, B and C" and "includes A, B, C" means that it includes all three of A, B and C, and "includes A, B or C" means that it includes one of A, B and C. "Including A, B and/or C" means including any one, any two or three of A, B and C.

It should be understood that in this application, "B corresponding to A", "B corresponding to A", "A corresponding to B" or "B corresponding to A" means that B is associated with A, according to A B can be determined. Determining B based on A does not mean determining B only based on A, but can also determine B based on A and/or other information. The matching between A and B means that the similarity between A and B is greater than or equal to the preset threshold.

Depending on the context, "if" as used herein may be interpreted as "when" or "when" or "in response to determination" or "in response to detection."

The technical solution of the present application will be described in detail below with specific examples. The following specific embodiments can be combined with each other, and the same or similar concepts or processes may not be described again in some embodiments.

Embodiment 1

This embodiment provides a simulation model optimization method for a bulk acoustic wave resonator, which specifically includes:

According to the first capacitor, the second capacitor, the first inductor, the first resistor, the second resistor and the third resistor, draw the behavioral simulation model circuit in the ADS simulation software and compare it with the test curve;

Substituting the behavioral simulation model circuit into the original physical simulation model, and setting geometric parameters and physical parameters for the original physical simulation model;

On the one hand, select the set of frequency points in the smooth part of the bulk acoustic wave resonator test curve. In the test curve shown in Figure 4, select the flatter and smoother part of the test, such as within the frequency range of 2.2-2.3 or frequency 2.4-2.5. Select more than 4, preferably 4 or 5 frequency points. The series resonance frequency is f _s , the parallel resonance frequency is f _p , and the quality factor represents the Q value at the series resonance frequency or the parallel resonance frequency. The calculated first capacitance is the average static capacitance C ₀ , the second capacitance is the dynamic capacitance C _m , the first inductance is the dynamic inductance L _m , the first resistance is the dielectric loss R ₀ of the piezoelectric film, and the second The resistance is the ohmic loss of the motor and the lead loss _Rs , and the third resistance is the mechanical loss _RM of the piezoelectric film.

On the one hand, after calculating the six values of R ₀ , R _s , R _M , C ₀ , C _m , and L _m , the behavioral simulation model circuit is drawn in the ADS simulation software. The specific behavioral simulation model is the MBVD model. Draw the The MBVD simulation circuit is shown in Figure 2. The MBVD curve drawn in Figure 4 is the behavioral simulation model curve. Compare the MBVD curve with the test curve.

On the one hand, the original physical simulation model is specifically the original MASON model as shown in Figure 1. After substituting the MBVD model in the previous step into the MASON model as shown in Figure 3, the corresponding material and geometric parameters are set for the MASON model.

On the one hand, R ₀ , R _s , and R _M are fine-tuned, and the fine-tuned resistor values are substituted into the MASON model to obtain the final optimized MASON model. The curve comparison between the optimized MASON model and the original MASON model is shown in Figure 5 As shown in the figure, it can be seen that the deviation of the optimized MASON model is very small. This method is simple to operate, ensuring the true accuracy of subsequent filter design and other processes, and reducing the deviation between design and manufacturing device performance.

On the one hand, the test curve is measured by an RF probe or a vector network analyzer, and the data used to extract the model can be read directly on the instrument or subsequently in an export file.

Through the above-mentioned simulation model optimization method of the bulk acoustic wave resonator, the simulation data is accurate, the accuracy of the Q value at the series and parallel resonance points is improved, and the subsequent simulation design is more realistic and effective.

Specifically, setting the geometric parameters of the original MASON model includes setting the thickness of the material layer and the area facing the electrode. The preferred setting parameters of this plan are: the set material layer thickness includes the bottom electrode thickness of 267nm, the piezoelectric layer thickness is set to 1100nm, the top electrode thickness is 281nm, the protective layer thickness is 200nm, and the electrode facing area is set to 7000μm ² . It can be understood that the set material layer does not necessarily include the above four layers, and the protective layer can sometimes be omitted; it is also possible that the set material layer will include other material layers in addition to the bottom electrode, piezoelectric layer, and top electrode, which will not be discussed here. limited.

Specifically, the bulk acoustic wave resonator includes a bottom electrode, a top electrode and a piezoelectric layer. The material of the bottom electrode and the top electrode is Mo; the piezoelectric layer material is a single crystalline aluminum nitride layer, which is a thin film material with a piezoelectric effect. for AIN, ZnO and PZT.

Specifically, the material and geometric parameters are set, and the component parameters of the original physical simulation model are updated. The component parameters are specifically the sound velocity, acoustic impedance, electromechanical coupling coefficient, clamping dielectric constant, attenuation factor, and electrode of the piezoelectric layer. The sound velocity, acoustic impedance and attenuation factor of the layer.

As shown in Table 1, they are the acoustic property values of each material in commonly used FBAR. For example, the parameters of each component selected for the resonator in this solution are: VAR1 uses a clamping dielectric constant of 9.5*10-11F/m, The acoustic impedance is 3.7*107kg/m ² s, the sound velocity of the electrode layer is 11350m/s, the electromechanical coupling coefficient is 6%, and the attenuation factor is 800dB/m AIN material. VAR2 uses an acoustic impedance of 6.39*107kg/m ² s. The electrode layer is made of Mo material with a sound speed of 6213m/s and an attenuation factor of 500dB/m. The type of the resonator may be an air cavity type thin film bulk acoustic resonator, a Bragg reflection type thin film bulk acoustic resonator or a reverse etching type thin film bulk acoustic resonator. Define the above physical parameters for each component in the simulation software, and update the physical parameters of the component to the MASON model in Figures 1 and 3.

Table 1 Acoustic property values of various materials in commonly used FBAR

Specifically, select a frequency point set of 4 or 5 frequency points, find the imaginary part of the impedance characteristic curve at each frequency point, calculate the static capacitance at the corresponding frequency point, and then average the static capacitance at all frequency points. to the first capacitor. The first capacitor is the average static capacitance C ₀ . Find the imaginary part X _i of the impedance characteristic curve Z(1,1) at each frequency point _fi , calculate C _i at the corresponding frequency point, and calculate the C _i , the average value is the static capacitance C ₀ , and the calculation formula is as follows:

According to the first capacitance C ₀ and the series resonance frequency f _s and parallel resonance frequency f _p , the second capacitance is calculated. The second capacitance is the dynamic capacitance C _m . The calculation formula is as follows:

According to the second capacitance C _m and the series resonance frequency f _s , the first inductance is calculated. The first inductance is the dynamic inductance L _m . The calculation formula is as follows:

Through the above formula, it can be calculated that the first capacitor is the average static capacitance C ₀ , the second capacitor is the dynamic capacitance C _m , and the first inductor is the dynamic inductance L _m .

Specifically, the quality factor Q is divided into the quality factor Q _s at the series resonance frequency and the quality factor Q _p at the parallel resonance frequency. The quality factor Q can be calculated in two ways:

Method 1: Calculate the quality factor Q _s at the series resonance frequency and the quality factor Q _p at the parallel resonance frequency through the following formulas, where

is the impedance phase:

Method 2: Obtained by calculation through simulation software. Specifically, enter the following formula in the simulation software:

q1＝abs(diff(phaserad(50*(1+S(2,2))/(1-S(2,2)))))

Through the above two methods, the quality factor Q _s at the series resonance frequency and the quality factor Q _p at the parallel resonance frequency can be calculated respectively.

Specifically, select the frequency point set to be 4 or 5 frequency points, find the real part R _i of the impedance characteristic curve at each frequency point, and take the average value as R, then R _s + R ₀ =R, and then according to The Q _s and Q _p , using the formula:

Calculate the first resistance R ₀ , the second resistance R _S , and the third resistance R _M .

After calculation, a set of better numerical combinations were obtained: C ₀ =0.643pF, C _m =0.033pF, L _m =144.976nH, R ₀ =1.0359Ω, R _S =0.0268Ω, R _M =0.8102Ω, the coordination is specific Parameter selection of the resonator size: the bottom electrode thickness is set to 267nm, the piezoelectric layer thickness is set to 1100nm, the top electrode thickness is 281nm, the protective layer thickness is 200nm, and the area is 7000μm ^2. The drawn MBVD model is shown in Figure 2 .

Specifically, fine-tune the first resistor R ₀ , the second resistor R _S , and the third resistor R _M calculated above. The specific adjustment method is to use the tune tuning tool, optimize tool, or in the ads simulation software, set R ₀ =1.0359 Ω, R _S =0.0268Ω, R _M =0.8102Ω are input, and the three parameters are continuously adjusted one by one, so that the difference between the two curves in Figure 4 becomes smaller and smaller. Finally, the fine-tuning operation is completed when they approximately overlap. The results are R ₀ =1.1359Ω, R _S =0.0268Ω, and R _M =1.0402Ω. Afterwards, the fine-tuned first resistor R ₀ , the second resistor R _S , and the third resistor R _M can be substituted into the optimized resonator MASON model to perform subsequent filter simulation work.

In this embodiment, in order to obtain a MASON model that is closer to the real value and correct the gap between the mechanical loss R _M , dielectric loss R ₀ and electrode impedance loss R _S and the real value, the parameters in the MBVD model are extracted from the original test data. Curve fitting, and then perform data fine-tuning on these values. After ensuring that the results are accurate, use the fine-tuned R ₀ , R _S , and R _M to correct and optimize the MASON model. This method makes the simulation data accurate, improves the accuracy of the Q value at the series-parallel resonance point, and makes the subsequent simulation design more realistic and effective.

Embodiment 2

This implementation also provides another simulation model optimization method of the bulk acoustic wave resonator, which specifically includes: setting the first resistor R ₀ , the second resistor R _S , and the third resistor R _M to all have resistance values of 1Ω. For the first resistor R ₀ , the second resistor R _S and the third resistor R _M are fine-tuned, the optimized physical simulation model is determined based on the fine-tuned resistance value, and the optimized physical simulation model is used for subsequent simulation work.

Specifically, the calculation steps of R ₀ , R _S , and _RM in the embodiment can be skipped, and the resistance values of the first resistor R ₀ , the second resistor R _S , and the third resistor _RM are directly preset to be 1Ω. According to the implementation The first capacitance calculated in Example 1 is the average static capacitance C ₀ , the second capacitance is the dynamic capacitance C _m , the first inductance is the dynamic inductance L _m , and the parameter selection to match the specific resonator size: bottom electrode thickness setting is 267nm, the piezoelectric layer thickness is set to 1100nm, the top electrode thickness is 281nm, the protective layer thickness is 200nm, and the area is 7000μm2. The MBVD model is established as shown in Figure 2, and substituted into the original MASON model as shown in Figure 3. Then R ₀ , R _S and R _M are fine-tuned. The specific adjustment method is to use the tune tuning tool, optimize tool or in the ads simulation software, input R ₀ = 1Ω, R _S = 1Ω, R _M = 1Ω, and let R ₀ and R _S and R _M are adjusted around 1Ω, so that the difference between the two curves in Figure 4 becomes smaller and smaller, and finally they approximately overlap to complete the fine-tuning operation.

Compared with Embodiment 1, this embodiment can more quickly obtain R ₀ , R _S , The adjustment result of R _M is convenient for operation.

Finally, it should be noted that the above embodiments are only used to illustrate the technical solution of the present application, but not to limit it; although the present application has been described in detail with reference to the foregoing embodiments, those of ordinary skill in the art should understand that: The technical solutions described in the foregoing embodiments can still be modified, or some or all of the technical features can be equivalently replaced; and these modifications or substitutions do not deviate from the essence of the corresponding technical solutions from the technical solutions of the embodiments of the present application. scope.

Claims

A simulation model optimization method for bulk acoustic wave resonators, including:

Select the set of frequency points in the smooth part of the bulk acoustic wave resonator test curve, and calculate the first capacitor, second capacitor, first inductor, first resistor, and second resistor based on the series resonant frequency and parallel resonant frequency and their respective quality factors. and a third resistor;

According to the first capacitor, the second capacitor, the first inductor, the first resistor, the second resistor and the third resistor, a behavioral simulation model circuit is drawn in the ADS simulation software, and the Compare the above test curves;

Substitute the behavioral simulation model circuit into the original physical simulation model, and set materials, geometric parameters and physical parameters for the original physical simulation model;

Fine-tune the first resistor, the second resistor and the third resistor, determine an optimized physical simulation model based on the fine-tuned resistance values, and use the optimized physical simulation model to perform subsequent simulation work.
The simulation model optimization method of a bulk acoustic wave resonator according to claim 1, wherein setting the geometric parameters includes setting the material layer thickness and the electrode facing area.
The simulation model optimization method of the bulk acoustic wave resonator according to claim 2, wherein the bulk acoustic wave resonator includes a bottom electrode, a top electrode and a piezoelectric layer, and the materials of the bottom electrode and the top electrode are Mo, so The piezoelectric layer material is a single crystal aluminum nitride layer.
The simulation model optimization method of the bulk acoustic wave resonator according to claim 2, wherein setting the physical parameters includes updating the component parameters of the original physical simulation model, and the component parameters are specifically the sound velocity and acoustic impedance of the piezoelectric layer. , electromechanical coupling coefficient, clamping dielectric constant, attenuation factor, sound velocity of the electrode layer, acoustic impedance and attenuation factor.
The simulation model optimization method of a bulk acoustic wave resonator according to claim 1, further comprising: selecting the frequency point set to be 4 or 5 frequency points, and finding the imaginary part of the impedance characteristic curve at each frequency point, Calculate the static capacitance at the corresponding frequency point, and then average the static capacitance at all frequency points to obtain the first capacitance;

Calculate the second capacitance according to the first capacitance and the series resonant frequency and the parallel resonant frequency;

The first inductance is calculated based on the second capacitance and the series resonance frequency.
The simulation model optimization method of the bulk acoustic wave resonator according to claim 5, further comprising: the quality factor is calculated by the following formula:

in
is the impedance phase, f p is the parallel resonance frequency, f s is the series resonance frequency;

Or calculated through simulation software, the quality factor Q s at the series resonance frequency and the quality factor Q p at the parallel resonance frequency can be obtained respectively.
The simulation model optimization method of the bulk acoustic wave resonator according to claim 6, further comprising: selecting the frequency point set to be 4 or 5 frequency points, and finding the real part R of the impedance characteristic curve at each frequency point. i , take the average value as R, then R s + R 0 =R, and then according to the quality factor Q S and quality factor Q p , use the formula:

Calculate the first resistance R 0 , the second resistance R S , and the third resistance R M .
The simulation model optimization method of the bulk acoustic wave resonator according to claim 7, further comprising fine-tuning the calculated first resistance R 0 , the second resistance R S and the third resistance RM .
The simulation model optimization method of a bulk acoustic wave resonator as claimed in claim 5, wherein the first capacitance is calculated by the following formula:

Among them, fi is the frequency point; Xi is the imaginary part Xi of the impedance characteristic curve Z(1,1) at each frequency point fi; n is the number of frequency points.
The simulation model optimization method of a bulk acoustic wave resonator as claimed in claim 5, wherein the second capacitance is calculated by the following formula:

Among them, C 0 is the first capacitance; f s is the series resonance frequency; f p is the parallel resonance frequency.
The simulation model optimization method of a bulk acoustic wave resonator as claimed in claim 5, wherein the first inductance is calculated by the following formula:

Among them, C m is the second capacitance; f s is the series resonance frequency.
The simulation model optimization method of the bulk acoustic wave resonator according to claim 6, wherein the calculation obtained by the simulation software is to enter the following formula in the simulation software:

q1=abs(diff(phaserad(50*(1+S(2,2))/(1-S(2,2))))).
The simulation model optimization method of the bulk acoustic wave resonator according to claim 1, further comprising: setting the resistance values of the first resistor, the second resistor and the third resistor to 1Ω, and setting the resistance values of the third resistor to 1Ω. A resistor, the second resistor and the third resistor are fine-tuned, an optimized physical simulation model is determined based on the fine-tuned resistance value, and the optimized physical simulation model is used for subsequent simulation work.
The simulation model optimization method of a bulk acoustic wave resonator according to any one of claims 1 to 13, wherein the test curve is measured by a radio frequency probe or a vector network analyzer.