A load layer preparation method and preparation device of a radio frequency filter
Patent Information
- Application Number
- CN202611136823.6
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- Filing Date
- 2026-07-29
- Publication Date
- 2026-10-30
AI Technical Summary
根据各个所述晶圆分区对应的射频滤波器带宽数据与各个所述晶圆分区对应的初始刻蚀深度量,结合所述射频滤波器对应的目标带宽数据,确定所述目标刻蚀量;
Smart Images

Figure CN123019645A_ABST
Abstract
Description
Technical Field
[0001] This application relates to the field of filter technology, and more specifically, to a method and apparatus for fabricating a load layer of a radio frequency filter. Background Technology
[0002] FBAR filters are radio frequency (RF) filters based on film bulk acoustic resonators (FBARs). In the fabrication of RF devices, the preparation of the top electrode and load layer of the RF filter typically employs a step-by-step process. Taking molybdenum (Mo) as both the electrode and load layer material as an example, the basic traditional process flow is as follows: First, a top electrode Mo layer is deposited on a piezoelectric layer. After negative photoresist patterning, a low-temperature (e.g., 10°-20°C) Mo load layer is deposited. Finally, the photoresist and excess metal layers are removed using a lift-off process to complete the patterning. This step-by-step fabrication method results in a cumbersome process with numerous steps and a long cycle time. Furthermore, the negative photoresist lift-off process is prone to problems such as rough pattern edges and incomplete removal.
[0003] There is currently no effective technical solution to the above problems. Summary of the Invention
[0004] The purpose of this application is to provide a method and apparatus for fabricating a load layer for a radio frequency (RF) filter. This simplifies the fabrication process, improves the accuracy of load layer thickness control, enhances the bandwidth consistency of the RF filter, and increases product yield and device performance stability.
[0005] In a first aspect, this application provides a method for fabricating a load layer of a radio frequency filter, comprising: A composite metal layer is deposited on the piezoelectric layer of the radio frequency filter; A mask layer is prepared on the composite metal layer; According to the preset target etching amount corresponding to the radio frequency filter, the area in the composite metal layer that is not covered by the mask layer is etched; Remove the mask layer to obtain the top electrode layer and load layer of the radio frequency filter.
[0006] The above approach simplifies the fabrication process, improves the accuracy of load layer thickness control, enhances the bandwidth consistency of RF filters, and increases product yield and device performance stability.
[0007] Optionally, etching the area of the composite metal layer not covered by the mask layer according to the preset target etching amount corresponding to the RF filter includes: The wafer in which the composite metal layer is disposed in the radio frequency filter is divided to obtain at least two wafer partitions; Based on the initial etching depth corresponding to each wafer partition, the RF filter bandwidth data corresponding to each wafer partition is determined. The target etching amount is determined based on the RF filter bandwidth data corresponding to each wafer partition and the initial etching depth corresponding to each wafer partition, combined with the target bandwidth data corresponding to the RF filter. According to the target etching amount, the area in the composite metal layer not covered by the mask layer is etched.
[0008] Through the above technical solution, this application can effectively solve the problem that traditional etching methods cannot guarantee the accuracy of the load layer thickness, significantly improve the accuracy of the load layer thickness, and thus ensure that the RF filter can meet the preset target bandwidth requirements.
[0009] Optionally, determining the RF filter bandwidth data corresponding to each wafer partition based on the initial etching depth of each wafer partition includes: The composite metal layer on each wafer partition is initially etched according to the initial etching depth corresponding to each wafer partition; After the initial etching is completed, the RF filter bandwidth data corresponding to each wafer partition is detected.
[0010] By clearly defining the specific implementation process for obtaining the bandwidth data corresponding to the initial etching depth of each wafer partition, we can ensure the accurate matching between the initial etching depth and the corresponding bandwidth data, providing a reliable and accurate data foundation for determining the target etching amount and ensuring that the thickness of the load layer obtained in the final processing meets the design requirements.
[0011] Optionally, determining the target etching amount based on the RF filter bandwidth data corresponding to each of the wafer partitions and the initial etching depth corresponding to each of the wafer partitions, combined with the target bandwidth data corresponding to the RF filter, includes: A straight line is obtained by fitting the RF filter bandwidth data corresponding to each wafer partition with the initial etching depth corresponding to each wafer partition; Based on the fitted straight line, the etching depth corresponding to the target bandwidth data is taken as the target etching amount.
[0012] Optionally, depositing a composite metal layer on the piezoelectric layer of the RF filter includes: The composite metal layer is deposited in a single step on the piezoelectric layer of the radio frequency filter using a physical vapor deposition process.
[0013] Optionally, the step of fabricating a mask layer on the composite metal layer includes: A positive photoresist is coated on the surface of the composite metal layer; The positive photoresist is exposed and developed to form a patterned mask layer on the composite metal layer.
[0014] Optionally, removing the mask layer includes: The mask layer is removed using a wet stripping process.
[0015] Optionally, etching the area of the composite metal layer not covered by the mask layer according to the preset target etching amount corresponding to the RF filter includes: According to the target etching amount, the area of the composite metal layer not covered by the mask layer is thinned and etched using a preset physical vapor deposition machine until the thickness of the area reaches the preset load layer thickness corresponding to the load layer.
[0016] Optionally, the fabrication method further includes, before depositing a composite metal layer on the piezoelectric layer of the radio frequency filter: A support layer is fabricated on the substrate of the radio frequency filter; A bottom electrode is fabricated on the support layer; A piezoelectric layer of the radio frequency filter is fabricated on the bottom electrode.
[0017] Secondly, this application provides an apparatus for fabricating a load layer of a radio frequency filter, comprising: A composite metal layer deposition module is used to deposit a composite metal layer on the piezoelectric layer of an RF filter; A mask fabrication module is used to fabricate a mask layer on the composite metal layer; The etching module is used to etch the area in the composite metal layer that is not covered by the mask layer according to the preset target etching amount corresponding to the radio frequency filter; A mask removal module is used to remove the mask layer to obtain the top electrode layer and load layer of the radio frequency filter.
[0018] As can be seen from the above, the method for fabricating the load layer of the RF filter provided in this application deposits a composite metal layer of the top electrode and the load layer in one step, and then etches the top electrode layer and the load layer of corresponding thickness according to the preset target etching amount. This replaces the traditional step-by-step deposition process, simplifies the overall fabrication process, and solves the problem of low filter bandwidth consistency and low yield caused by poor load layer thickness accuracy in traditional processes by precisely controlling the thickness of the load layer through quantitative etching. This improves the load layer thickness control accuracy, enhances the bandwidth consistency of the RF filter, and thus improves the product yield and the stability of device performance.
[0019] Other features and advantages of this application will be set forth in the following description and will be apparent in part from the description or may be learned by practicing embodiments of this application. The objectives and other advantages of this application may be realized and obtained by means of the structures particularly pointed out in the written description and the accompanying drawings. Attached Figure Description
[0020] Figure 1 This is a first flowchart of a method for fabricating a load layer of a radio frequency filter according to an embodiment of this application.
[0021] Figure 2 This is a second flowchart of a method for fabricating a load layer of a radio frequency filter provided in an embodiment of this application.
[0022] Figure 3 This is a schematic diagram showing the location of wafer partitions provided in an embodiment of this application.
[0023] Figure 4 A schematic diagram illustrating the effect of the fitted straight line provided in the embodiments of this application.
[0024] Figure 5 This is a schematic diagram illustrating the effect of bandwidth range data provided in the embodiments of this application compared to bandwidth range data of conventional processes.
[0025] Figure 6 This is a schematic diagram illustrating the yield performance of the radio frequency filter provided in the embodiments of this application.
[0026] Figure 7 This is a schematic diagram of a load layer fabrication apparatus for a radio frequency filter provided in an embodiment of this application.
[0027] Labeling explanation: 100, composite metal layer; 110, mask layer; 120, top electrode layer; 130, load layer; 21, composite metal layer deposition module; 22, mask fabrication module; 23, etching module; 24, mask removal module. Detailed Implementation
[0028] The technical solutions of the embodiments of this application will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only a part of the embodiments of this application, and not all of the embodiments. The components of the embodiments of this application described and shown in the accompanying drawings can generally be arranged and designed in various different configurations. Therefore, the following detailed description of the embodiments of this application provided in the accompanying drawings is not intended to limit the scope of the claimed application, but merely represents selected embodiments of this application. All other embodiments obtained by those skilled in the art based on the embodiments of this application without inventive effort are within the scope of protection of this application.
[0029] It should be noted that similar reference numerals and letters in the following figures indicate similar items; therefore, once an item is defined in one figure, it does not need to be further defined and explained in subsequent figures. Furthermore, in the description of this application, terms such as "first," "second," etc., are used only to distinguish descriptions and should not be construed as indicating or implying relative importance.
[0030] Existing RF filter fabrication technologies suffer from problems such as poor film thickness uniformity, inconsistent bandwidth, low yield (the yield of traditional processes is only around 85%), and a wide bandwidth range. For example, the precision of low-temperature Mo long film deposition can typically only be controlled within ±(20Å-30Å), resulting in significant out-of-band rejection losses and affecting device performance stability. Specifically, existing traditional processes, due to the use of adhesive-coated peeling methods, can only use low-temperature Mo to deposit the load layer of the RF filter. However, the film thickness precision of low-temperature Mo deposition is typically only ±30Å, and the deviation increases with the thickness of the film. For example, the deviation can reach ±55Å when depositing a 1000Å load layer, making it difficult to achieve high product precision and uniformity requirements.
[0031] A filter is a frequency filter that allows signals within a specific frequency range to pass through. This frequency range is called the passband, and the frequency range outside the passband is called the out-of-band. The filter must block out-of-band signals from passing through, and this ability of the filter to block out-of-band signals is called out-of-band rejection.
[0032] Based on the above, this application provides a method and apparatus for fabricating a load layer of an RF filter. By depositing a composite metal layer on the piezoelectric layer of the RF filter in one step and fabricating a mask layer on the composite metal layer, the area of the composite metal layer not covered by the mask layer is etched according to the preset target etching amount corresponding to the RF filter, thereby obtaining the top electrode layer and load layer of the corresponding thickness of the RF filter. This simplifies the overall fabrication process, improves the thickness control accuracy of the load layer, and thus improves the bandwidth consistency of the RF filter. It solves the problem of low filter bandwidth consistency and low yield caused by poor load layer thickness accuracy in traditional processes, and improves product yield and device performance stability.
[0033] refer to Figures 1-6 This application provides a method for fabricating a load layer of an RF filter, comprising: Step S1: Deposit a composite metal layer 100 on the piezoelectric layer of the radio frequency filter; Step S2: Prepare a mask layer 110 on the composite metal layer 100; Step S3: Etch the area of the composite metal layer 100 that is not covered by the mask layer 110 according to the preset target etching amount corresponding to the radio frequency filter. Step S4: Remove the mask layer 110 to obtain the top electrode layer 120 and the load layer 130 of the RF filter.
[0034] Composite metal layer 100: In this embodiment, composite metal layer 100 refers to a composite metal layer 100 on the piezoelectric layer of the RF filter, in which a top electrode layer 120 and a load layer 130 are deposited in a single step. This composite structure aims to simplify the fabrication process and improve production efficiency. For example, molybdenum (Mo) can be used as the material of composite metal layer 100, and its total thickness is the sum of the thickness of the top electrode and the thickness of the load layer 130. Taking a WIFI 2.4G filter as an example, the thickness of the top electrode is 1770 Å, and the thickness of the load layer 130 is 300 Å, then the total thickness of composite metal layer 100 is 1770 Å + 300 Å = 2070 Å.
[0035] In traditional processes, the top electrode layer 120 and the load layer 130 are typically deposited in steps, leading to a complex process flow requiring multiple photolithography, deposition, and lift-off operations. This embodiment integrates the functions of the top electrode layer 120 and the load layer 130 into a single deposition step by depositing a composite metal layer 100 at a high temperature range of 200°C-230°C. For example, a physical vapor deposition (PVD) process can be used to deposit a molybdenum (Mo) layer on the piezoelectric layer in a single step as the composite metal layer 100, the total thickness of which is the sum of the thickness of the top electrode and the load layer 130. This single-step deposition method significantly simplifies the process flow, reduces the number of steps, and thus shortens the fabrication cycle.
[0036] Subsequently, a mask layer 110 is fabricated on the composite metal layer 100. The function of the mask layer 110 is to protect the area of the composite metal layer 100 where the load layer 130 needs to be formed, preventing it from being removed during subsequent etching, while exposing the area where the top electrode layer 120 needs to be formed. Specifically, a patterned mask layer 110 can be formed on the composite metal layer 100 by coating a positive photoresist onto the surface of the composite metal layer 100 and then exposing and developing the positive photoresist. This method of fabricating the patterned mask layer 110 can precisely define the areas of the top electrode and the load layer 130, providing an accurate processing range for subsequent etching operations.
[0037] Next, the areas of the composite metal layer 100 not covered by the mask layer 110 are etched according to the preset target etching amount corresponding to the RF filter. By precisely controlling the etching amount, the composite metal layer 100 in the exposed areas is thinned to the thickness required for the load layer 130. For example, ion beam etching (IBE) technology can be used to quantitatively etch the non-patterned areas using a physical vapor deposition machine (i.e., a Trim machine) to obtain the required load layer thickness. This quantitative etching method can precisely control the remaining thickness of the uncovered areas, ensuring that the load layer thickness meets the design requirements, thereby improving the film thickness accuracy and uniformity of the load layer 130.
[0038] Finally, the mask layer 110 is removed to obtain the top electrode layer 120 and the load layer 130 of the RF filter. After etching, the mask layer 110 has completed its protection and patterning tasks and needs to be removed. The mask layer 110 can be removed using a wet stripping process. After removing the mask layer 110, the area of the composite metal layer 100 covered by the mask layer 110 forms the load layer 130, while the uncovered and etched area forms the top electrode layer 120. The required top electrode layer 120 and load layer 130 can be obtained simultaneously in a single process. High-temperature (200°-230°) deposition replaces low-temperature (10°-20°) deposition. Combining positive photolithography and physical vapor deposition processes simplifies the process, improves etching accuracy, and enhances the thickness accuracy and uniformity of the load layer 130.
[0039] In some embodiments, etching is performed on the area of the composite metal layer 100 not covered by the mask layer 110 according to a preset target etching amount corresponding to the RF filter. This includes: dividing the wafer on which the composite metal layer 100 is disposed in the RF filter to obtain at least two wafer partitions; determining the RF filter bandwidth data corresponding to each wafer partition based on the initial etching depth corresponding to each wafer partition; determining the target etching amount based on the RF filter bandwidth data corresponding to each wafer partition and the initial etching depth corresponding to each wafer partition, combined with the target bandwidth data corresponding to the RF filter; and etching the area of the composite metal layer 100 not covered by the mask layer 110 according to the target etching amount.
[0040] This application's solution uses a regional testing and calibration method to determine the target etching amount for each wafer partition, adapting it to its specific conditions. This solves the problem of poor bandwidth consistency caused by using theoretical values for etching the entire wafer. In the above-mentioned method for fabricating the load layer of the RF filter, a composite metal layer 100 is first deposited on the piezoelectric layer of the RF filter in a single step. This composite metal layer 100 contains the materials required for the top electrode layer 120 and the load layer 130. Subsequently, a mask layer 110 is fabricated on the composite metal layer 100, and the areas not covered by the mask layer 110 are etched to form the load layer 130. To ensure the accuracy of the load layer thickness, this solution further precisely designs the preset target etching amount. First, the wafer with the composite metal layer 100 is divided into at least two wafer partitions, and the initial etching depth and RF filter bandwidth data corresponding to each wafer partition are determined. This allows for the subsequent execution of different initial etching depths for each region, thereby forming a load layer 130 and a top electrode layer 120 of corresponding thicknesses in each region. Then, by applying different initial etching depths to these wafer partitions and measuring the corresponding RF filter bandwidth data for each partition, a practical correspondence between etching depth and filter bandwidth can be established. Based on these measured data, combined with the preset target bandwidth data of the RF filter, the actual target etching amount required to achieve the target bandwidth can be accurately determined through data fitting. Finally, the composite metal layer 100 is etched according to this calibrated target etching amount to obtain a precisely thick load layer 130. This method allows the etching process to fully consider possible process differences within the wafer, avoids the inaccuracies caused by etching with a single theoretical value, ensures high-precision control of the load layer thickness, and thus guarantees that the RF filter can meet the expected target bandwidth requirements.
[0041] The following is a specific example. As a concrete implementation, taking a WIFI 2.4G filter as an example, the thickness of the top electrode is 1770 Å, and the thickness of the load layer 130 is 300 Å. Therefore, the total thickness of the composite metal layer 100 is 1770 Å + 300 Å = 2070 Å. Specifically, the wafer can be divided into four quadrants using ion beam etching (IBE) coordinates, forming four wafer partitions. To quickly locate the optimal process window, different initial etching depths can be applied to these four wafer partitions. This initial etching depth is set around the theoretical thickness of the load layer 130 (i.e., 300 Å). For example, the first wafer partition (e.g., ...) Figure 3 In section B), a 260 Å etching is performed on the second wafer partition (e.g., ...). Figure 3 C) performs 285Å etching, the third wafer partition (e.g. Figure 3 D) performs 315Å etching, the fourth wafer partition (e.g. Figure 3E) performs an etching at 340 Å, such as Figure 3 As shown. After these initial etchings are completed, the RF filter bandwidth data corresponding to each wafer partition is obtained through testing with a network analyzer (WAT). For example, S-parameter data (i.e., S2P data) can be obtained. Subsequently, data parsing tools such as MATLAB are used to process the S2P data to obtain parameters such as the left-hand frequency of -20dB, the right-hand frequency of -20dB, and the -20dB bandwidth, thereby determining the bandwidth data based on the initial etching depth conditions corresponding to the wafer partition. Next, using these different etching conditions and their corresponding bandwidth data, a straight line is fitted to obtain a fitted straight line, as shown. Figure 4 As shown. If the target bandwidth data corresponding to the RF filter is 113MHz, the etching depth corresponding to the 113MHz target bandwidth data can be found based on the fitted line. For example, the fitting result may show that the optimal bandwidth of 113MHz corresponds to an ion beam etching depth of 220Å, at which point the step measurement value may be 237Å. Finally, according to this determined target etching depth, i.e., 220Å, the area of the composite metal layer 100 not covered by the mask layer 110 is thinned and etched until the thickness of this area reaches the preset load layer thickness corresponding to the load layer 130, for example, 237Å. Therefore, this application does not directly use the theoretical thickness of the load layer 130 of 300Å for etching, but obtains the target etching depth (i.e., 220Å) through the above technical solution. The load layer thickness obtained by etching with 220Å is more consistent with the target bandwidth data of 113MHz corresponding to the RF filter. That is, this application can effectively solve the problem that traditional etching methods cannot guarantee the accuracy of the load layer thickness, significantly improve the accuracy of the load layer thickness, and thus ensure that the RF filter can meet the preset target bandwidth requirements. Furthermore, by comparing the bandwidth range data of the RF filter in this application with the bandwidth range data of conventional processes, such as... Figure 5 As shown, the bandwidth of RF filters manufactured using traditional processes is approximately 11MHz, while the bandwidth of the RF filter of this application is reduced to approximately 4MHz. This demonstrates that the RF filters prepared using the method described in this application exhibit highly consistent frequency response, effectively improving the bandwidth consistency problem of RF filters. Consequently, this significantly reduces defective products due to insertion loss / out-of-band rejection, resulting in a substantial decrease in yield loss and thus improving product yield and device performance stability. Experiments have shown that the RF filter of this application achieves a yield rate of over 95%. Figure 6 As shown, the yield of RF filters prepared by traditional processes is significantly improved (only about 85%). By combining this method with the one-time deposition of the composite metal layer 100, this application can achieve high-precision control of the load layer thickness while simplifying the process flow, overcoming the inherent defect of poor film thickness accuracy in traditional low-temperature Mo deposition.
[0042] In some implementations, the RF filter bandwidth data corresponding to each wafer partition is determined based on the initial etching depth corresponding to each wafer partition, including: performing initial etching on the composite metal layer 100 on each wafer partition according to the initial etching depth corresponding to each wafer partition; and detecting the RF filter bandwidth data corresponding to each wafer partition after the initial etching is completed.
[0043] The step of "performing initial etching of the composite metal layer 100 on each wafer partition according to the initial etching depth corresponding to each wafer partition" aims to create a composite metal layer 100 with a specific initial thickness for each wafer partition to simulate the impact of different etching depths on the performance of the RF filter. By applying different initial etching depths to different partitions, the relationship between etching depth and filter bandwidth can be systematically explored. This initial etching depth can be precisely thinned by using ion beam etching (IBE) technology, through precise control of the ion beam energy, angle, and etching time, to achieve the preset initial etching depth.
[0044] The purpose of the step "detecting the RF filter bandwidth data for each wafer partition after initial etching" is to obtain the actual bandwidth performance data of the RF filters on each wafer partition after different initial etching depths. Direct measurement provides the most accurate and realistic correlation between etching depth and bandwidth. Detection methods may include, but are not limited to: using a network analyzer to measure the S-parameters of the RF filters and then extracting key bandwidth parameters such as -20dB bandwidth and -3dB bandwidth from the S-parameter data; or, using specialized automated test equipment (ATE) for batch testing to quickly obtain the bandwidth data for each partition.
[0045] Specifically, for each wafer partition, the composite metal layer 100 on the wafer partition can be initially etched according to the initial etching depth corresponding to the wafer partition, so that the load layer 130 and the top electrode layer 120 can be formed on the wafer partition in the future. After the initial etching is completed, the RF filter bandwidth data corresponding to the wafer partition can be detected.
[0046] The solution proposed in this application, by clearly defining the specific implementation process for obtaining the bandwidth data corresponding to the initial etching depth of each wafer partition, can ensure the accurate matching of the initial etching depth and the corresponding bandwidth data, providing a reliable and accurate data basis for subsequently determining the target etching amount, and ensuring that the thickness of the load layer obtained in the final processing meets the design requirements.
[0047] In some implementations, the target etching amount is determined based on the RF filter bandwidth data corresponding to each wafer partition and the initial etching depth corresponding to each wafer partition, combined with the target bandwidth data corresponding to the RF filter. This includes: performing a linear fitting between the RF filter bandwidth data corresponding to each wafer partition and the initial etching depth corresponding to each wafer partition to obtain a fitted line; and using the etching depth corresponding to the target bandwidth data as the target etching amount based on the fitted line.
[0048] The proposed solution establishes a correlation between bandwidth and etching depth by fitting actual acquired process data, accurately determining the etching amount that meets the target bandwidth requirements and solving the problem of insufficient precision in the target etching amount. The method first divides the wafer containing the composite metal layer 100 in the RF filter into at least two wafer partitions. Based on the initial etching depth of each wafer partition, the RF filter bandwidth data corresponding to each wafer partition is determined. These wafer partitions and their corresponding initial etching depths and bandwidth data constitute a set of discrete process data points in actual production. Based on this, a linear fit is performed on these actual acquired process data—that is, a linear fit is performed between the RF filter bandwidth data corresponding to each wafer partition and the initial etching depth corresponding to each wafer partition—to obtain a fitted straight line. This fitted straight line establishes a quantitative relationship model between bandwidth and etching depth, which reflects the actual situation of the current batch process, making this relationship more consistent with actual production. By utilizing measured data from each wafer partition obtained in actual production for fitting, the resulting fitted line more accurately characterizes the intrinsic relationship between bandwidth and etching depth, thus laying the foundation for subsequent precise calculations. Subsequently, based on this fitted line, the target bandwidth data corresponding to the preset RF filter is substituted into the fitted relationship to calculate the etching depth required to achieve the target bandwidth, which is then used as the final target etching depth. This method, relying on the linear correspondence obtained through fitting, can directly, quickly, and accurately calculate the etching depth required for the target bandwidth, overcoming the problem of insufficient accuracy caused by directly using theoretical values for etching in traditional methods. In this way, the final etched load layer thickness accurately meets design requirements, ensuring that the RF filter bandwidth meets design goals and significantly improving the consistency and stability of product performance.
[0049] In some embodiments, depositing a composite metal layer 100 on the piezoelectric layer of the radio frequency filter includes: depositing the composite metal layer 100 on the piezoelectric layer of the radio frequency filter in a single step using a physical vapor deposition process.
[0050] Specifically, a composite metal layer 100 is deposited in a single step on the piezoelectric layer of the RF filter using physical vapor deposition (PVD). This composite metal layer 100 contains all the metal materials for the subsequently formed top electrode layer 120 and load layer 130. This single-step deposition method avoids the multiple process steps required by traditional step-by-step deposition of the top electrode metal layer and load layer 130, thus significantly simplifying the overall fabrication process and shortening the processing cycle. Simultaneously, by reducing process steps and operations, it effectively avoids the accumulation of errors and process defects that may be introduced by multiple depositions, improving production efficiency and yield. As a high-precision thin film fabrication technology, physical vapor deposition enables precise control over film thickness and uniformity. Compared to the low-temperature deposition method used for the load layer 130 in traditional step-by-step fabrication, physical vapor deposition has significant advantages in film density, adhesion, and purity, and is more likely to yield a metal layer with better uniformity. This provides a solid technological foundation for obtaining a precisely thick load layer 130 and top electrode layer 120 through subsequent etching processes, thereby improving the problems of large film thickness deviation and poor uniformity in traditional processes.
[0051] In some embodiments, the preparation of a mask layer 110 on the composite metal layer 100 includes: coating a positive photoresist on the surface of the composite metal layer 100; and exposing and developing the positive photoresist to form a patterned mask layer 110 on the composite metal layer 100.
[0052] After depositing a composite metal layer 100 on the piezoelectric layer of the RF filter in a single step, a positive photoresist is applied to the surface of the composite metal layer 100. The composite metal layer 100 coated with the positive photoresist is then exposed and developed. During exposure, a mask template transmits the preset pattern information to the positive photoresist, exposing the photoresist corresponding to the areas to be etched. During development, a developer selectively dissolves the exposed photoresist, precisely forming a patterned mask layer 110 on the composite metal layer 100. This effectively solves the problems of rough pattern edges and incomplete stripping that are common with traditional negative photoresists during mask fabrication. This high-precision mask layer 110 ensures the accuracy of the etching process, resulting in the final top electrode layer 120 and load layer 130 having precise dimensions and shapes, thereby effectively improving the performance and yield of the RF filter.
[0053] In some embodiments, removing the mask layer 110 includes: removing the mask layer 110 using a wet stripping process.
[0054] In the fabrication of the load layer 130 of the RF filter, a composite metal layer 100 is first deposited on the piezoelectric layer of the RF filter in a single step, followed by the fabrication of a patterned mask layer 110 on the composite metal layer 100. Next, according to the preset target etching amount corresponding to the RF filter, the areas of the composite metal layer 100 not covered by the mask layer 110 are precisely etched to form the top electrode layer 120 and the load layer 130 of the RF filter. After the etching step is completed, to obtain the final device structure, the remaining mask layer 110 needs to be removed. This application uses a wet stripping process to accomplish this step. The wet stripping process utilizes a specific chemical solution to selectively dissolve or expand the mask material, causing it to detach from the etched metal layer surface. This method avoids the physical damage that may be caused by traditional dry stripping, while ensuring that the mask layer 110 is completely removed without leaving any residue, thereby guaranteeing the structural integrity and surface cleanliness of the top electrode layer 120 and the load layer 130.
[0055] In some embodiments, etching is performed on the area of the composite metal layer 100 not covered by the mask layer 110 according to the preset target etching amount corresponding to the radio frequency filter. This includes: according to the target etching amount, thinning etching is performed on the area of the composite metal layer 100 not covered by the mask layer 110 using a preset physical vapor deposition machine until the thickness of the area reaches the preset load layer thickness corresponding to the load layer 130.
[0056] By precisely controlling the etching endpoint, the thickness of the load layer 130 was ensured to meet design requirements, significantly improving the thickness accuracy of the load layer 130. This enables the RF filter to meet the high-precision requirements of high-performance devices for load layer thickness, thereby ensuring the stability of device performance. Simultaneously, utilizing a pre-designed physical vapor deposition (PVD) machine for thinning etching effectively leverages existing equipment, reduces process improvement costs, and ensures compatibility with existing production processes. This precise etching control, combined with the overall solution of one-time deposition of the composite metal layer 100, effectively solves problems such as large load layer thickness deviation, poor film thickness uniformity, poor bandwidth consistency, and significant out-of-band suppression loss in traditional processes, significantly improving the yield and performance stability of the RF filter.
[0057] In some embodiments, the fabrication method further includes, before depositing the composite metal layer 100 on the piezoelectric layer of the radio frequency filter, preparing a support layer on the substrate of the radio frequency filter; preparing a bottom electrode on the support layer; and preparing the piezoelectric layer of the radio frequency filter on the bottom electrode.
[0058] This application constructs a complete and stable device foundation structure by sequentially fabricating a support layer on a substrate, a bottom electrode on the support layer, and a piezoelectric layer on the bottom electrode before depositing the composite metal layer 100 on the piezoelectric layer of the RF filter. First, a support layer is fabricated on the substrate. This support layer effectively isolates the substrate from the influence of acoustic wave resonance and provides a flat and mechanically stable surface, laying the foundation for subsequent thin film growth. Second, a bottom electrode is fabricated on the support layer. As a key component of the resonator, the precise geometry and material properties of the bottom electrode are crucial for achieving the desired electroacoustic conversion efficiency. Finally, a piezoelectric layer is fabricated on the bottom electrode. The piezoelectric layer is the core functional layer for electroacoustic coupling, and its quality directly determines the filter's performance. This progressive fabrication sequence ensures that each layer meets design requirements, providing a high-quality substrate for subsequent deposition of the composite metal layer 100 on the piezoelectric layer and for precise etching. This structured fabrication process enables the piezoelectric layer to support the composite metal layer 100 in an optimal state, thereby ensuring the precise formation of the top electrode layer 120 and the load layer 130, and thus improving the overall performance stability and bandwidth consistency of the RF filter.
[0059] refer to Figure 7 This application provides an apparatus for fabricating a load layer 130 of a radio frequency filter, comprising: Composite metal layer deposition module 21 is used to deposit composite metal layer 100 on the piezoelectric layer of RF filter; The mask fabrication module 22 is used to fabricate a mask layer 110 on the composite metal layer 100; Etching module 23 is used to etch the area of the composite metal layer 100 that is not covered by the mask layer 110 according to the preset target etching amount corresponding to the radio frequency filter. The mask removal module 24 is used to remove the mask layer 110 to obtain the top electrode layer 120 and the load layer 130 of the RF filter.
[0060] This application's solution integrates a composite metal layer deposition module 21, a mask fabrication module 22, an etching module 23, and a mask removal module 24, achieving a fabrication process of quantitative etching after a single deposition of the composite metal layer 100. Specifically, since the composite metal layer deposition module 21 deposits a metal layer containing the top electrode and the load layer 130 material in a single deposition, multiple deposition processes are avoided; the mask layer 110 formed by the mask fabrication module 22 protects the load layer 130 region, thereby ensuring that etching only acts on the load layer 130 region; the etching module 23 precisely controls the target etching amount to achieve the high-precision requirements for the load layer thickness; finally, the mask removal module 24 removes the mask, completing the fabrication. This approach effectively solves the problems of poor film thickness uniformity and poor bandwidth consistency in traditional processes.
[0061] The above technical solution significantly improves the thickness accuracy of the 130mm load layer. At the same time, the process is simplified, the number of steps is reduced, the preparation cycle is shortened, peeling defects are avoided, bandwidth consistency and product yield are improved, out-of-band suppression performance is improved, and device performance stability is enhanced.
[0062] The RF filter load layer fabrication apparatus provided in this application embodiment is used to perform the steps in the RF filter load layer fabrication method provided in the first aspect above. The principle of the RF filter load layer fabrication apparatus provided in this embodiment is the same as that of the RF filter load layer fabrication method provided in the first aspect above, and will not be discussed in detail here.
[0063] In this document, relational terms such as first and second are used only to distinguish one entity or operation from another entity or operation, without necessarily requiring or implying any such actual relationship or order between these entities or operations.
[0064] The above description is merely an embodiment of this application and is not intended to limit the scope of protection of this application. Various modifications and variations can be made to this application by those skilled in the art. Any modifications, equivalent substitutions, improvements, etc., made within the spirit and principles of this application should be included within the scope of protection of this application.
Claims
1. A method for fabricating a load layer of an RF filter, characterized in that, include: A composite metal layer (100) is deposited on the piezoelectric layer of the radio frequency filter. A mask layer (110) is prepared on the composite metal layer (100). According to the preset target etching amount corresponding to the radio frequency filter, the area in the composite metal layer (100) not covered by the mask layer (110) is etched; Remove the mask layer (110) to obtain the top electrode layer (120) and load layer (130) of the radio frequency filter.
2. The method for fabricating the load layer of the radio frequency filter according to claim 1, characterized in that, The etching of the area in the composite metal layer (100) not covered by the mask layer (110) according to the preset target etching amount corresponding to the radio frequency filter includes: The wafer in which the composite metal layer (100) is disposed in the radio frequency filter is divided to obtain at least two wafer partitions; Based on the initial etching depth corresponding to each wafer partition, the RF filter bandwidth data corresponding to each wafer partition is determined. The target etching amount is determined based on the RF filter bandwidth data corresponding to each wafer partition and the initial etching depth corresponding to each wafer partition, combined with the target bandwidth data corresponding to the RF filter. According to the target etching amount, the area of the composite metal layer (100) not covered by the mask layer (110) is etched.
3. The method for fabricating the load layer of the radio frequency filter according to claim 2, characterized in that, The step of determining the RF filter bandwidth data corresponding to each wafer partition based on the initial etching depth of each wafer partition includes: The composite metal layer (100) on each wafer partition is initially etched according to the initial etching depth corresponding to each wafer partition; After the initial etching is completed, the RF filter bandwidth data corresponding to each wafer partition is detected.
4. The method for fabricating the load layer of the radio frequency filter according to claim 2, characterized in that, The step of determining the target etching amount based on the RF filter bandwidth data corresponding to each wafer partition and the initial etching depth corresponding to each wafer partition, combined with the target bandwidth data corresponding to the RF filter, includes: A straight line is obtained by fitting the RF filter bandwidth data corresponding to each wafer partition with the initial etching depth corresponding to each wafer partition. Based on the fitted straight line, the etching depth corresponding to the target bandwidth data is taken as the target etching amount.
5. The method for fabricating the load layer of the radio frequency filter according to claim 1, characterized in that, The deposition of a composite metal layer (100) on the piezoelectric layer of the radio frequency filter includes: The composite metal layer (100) is deposited in a single step on the piezoelectric layer of the radio frequency filter using a physical vapor deposition process.
6. The method for fabricating the load layer of the radio frequency filter according to claim 1, characterized in that, The process of fabricating a mask layer (110) on the composite metal layer (100) includes: A positive photoresist is coated on the surface of the composite metal layer (100); The positive photoresist is exposed and developed to form a patterned mask layer (110) on the composite metal layer (100).
7. The method for fabricating the load layer of the radio frequency filter according to claim 1, characterized in that, The removal of the mask layer (110) includes: The mask layer (110) is removed using a wet stripping process.
8. The method for fabricating the load layer of the radio frequency filter according to claim 1, characterized in that, The etching of the area in the composite metal layer (100) not covered by the mask layer (110) according to the preset target etching amount corresponding to the radio frequency filter includes: According to the target etching amount, the area of the composite metal layer (100) not covered by the mask layer (110) is thinned and etched using a preset physical vapor deposition machine until the thickness of the area reaches the preset load layer thickness corresponding to the load layer (130).
9. The method for fabricating the load layer of the radio frequency filter according to claim 1, characterized in that, The fabrication method further includes, before depositing the composite metal layer (100) on the piezoelectric layer of the radio frequency filter: A support layer is fabricated on the substrate of the radio frequency filter; A bottom electrode is fabricated on the support layer; A piezoelectric layer of the radio frequency filter is fabricated on the bottom electrode.
10. An apparatus for fabricating a load layer of a radio frequency filter, characterized in that, include: A composite metal layer deposition module is used to deposit a composite metal layer (100) on the piezoelectric layer of an RF filter. A mask fabrication module is used to fabricate a mask layer (110) on the composite metal layer (100). The etching module is used to etch the area of the composite metal layer (100) that is not covered by the mask layer (110) according to the preset target etching amount corresponding to the radio frequency filter; A mask removal module is used to remove the mask layer (110) to obtain the top electrode layer (120) and load layer (130) of the radio frequency filter.
