Self-adaptive air pressure control method for semiconductor polishing

Abstract

The invention provides a self-adaptive air pressure control method for semiconductor polishing, and relates to the technical field of polishing control, the method comprises the following steps: obtaining a cascade air pressure control module of chemical mechanical polishing equipment; constructing an original three-dimensional model and a preset polishing target three-dimensional model of the semiconductor wafer to be processed; performing space alignment fitting to generate a to-be-polished model; neighborhood clustering is carried out according to multi-point polishing requirements; based on the multiple clustering areas and the polishing demand identification information, in combination with preset polishing solution information, polishing air pressure optimization is carried out, and multiple optimal polishing air pressure parameters of the multiple clustering areas are obtained; and based on the first air pressure regulator, air pressure control signals are generated according to the multiple optimal polishing air pressure parameters for polishing control, and feedback optimization control is conducted through the second air pressure regulator. According to the invention, the technical problem that the polishing parameter setting is not accurate enough in the prior art can be solved, and the technical effect of improving the semiconductor polishing accuracy and efficiency is achieved.

Description

Technical Field

[0001] The present application relates to the technical field of polishing control, and in particular to an adaptive gas pressure control method for semiconductor polishing. Background Art

[0002] Adaptive air pressure control is designed to automatically adjust the air pressure parameters during the polishing process according to polishing needs and conditions to achieve the best polishing effect.

[0003] Currently, existing semiconductor polishing processes fail to group multiple locations with similar polishing requirements. This results in inaccurate polishing parameter settings for different areas, which in turn affects polishing consistency across the entire surface. Furthermore, polishing air pressure settings may not be optimized for specific areas, leading to over- or under-polishing in certain areas. This increases polishing time and cost, reduces overall polishing efficiency, and impacts product quality and performance. Therefore, a solution to these problems is needed.

[0004] In summary, the existing technology has a technical problem that most of the points with similar polishing requirements are not grouped, resulting in inaccurate settings of polishing parameters in different areas, causing low polishing efficiency, and further affecting product quality and performance. Summary of the Invention

[0005] The purpose of this application is to provide an adaptive air pressure control method for semiconductor polishing, so as to solve the technical problem in the prior art that most of the points with similar polishing requirements are not grouped, resulting in inaccurate settings of polishing parameters in different areas, causing low polishing efficiency and further affecting product quality and performance.

[0006] In view of the above problems, the present application provides an adaptive gas pressure control method for semiconductor polishing.

[0007] The present application provides an adaptive air pressure control method for semiconductor polishing, wherein the method includes: obtaining a cascade air pressure control module of a chemical mechanical polishing device, wherein the cascade air pressure control module includes a first air pressure regulator and a second air pressure regulator; constructing an original three-dimensional model of a semiconductor wafer to be processed and a preset polishing target three-dimensional model; spatially aligning and fitting the original three-dimensional model and the preset polishing target three-dimensional model to generate a model to be polished; performing neighborhood clustering on the polishing requirements of multiple points of the model to be polished to obtain multiple clustering areas, and the multiple clustering areas carry polishing requirement identification information; based on the multiple clustering areas and the polishing requirement identification information, combined with preset polishing liquid information, performing polishing air pressure optimization to obtain multiple optimal polishing air pressure parameters for the multiple clustering areas; based on the first air pressure regulator, generating an air pressure control signal according to the multiple optimal polishing air pressure parameters to perform polishing control, and performing feedback optimization control through the second air pressure regulator.

[0008] One or more technical solutions provided in this application have at least the following technical effects or advantages:

[0009] By obtaining a cascade air pressure control module of a chemical mechanical polishing device, wherein the cascade air pressure control module includes a first air pressure regulator and a second air pressure regulator; constructing an original three-dimensional model of a semiconductor wafer to be processed and a preset polishing target three-dimensional model; spatially aligning and fitting the original three-dimensional model and the preset polishing target three-dimensional model to generate a model to be polished; performing neighborhood clustering of polishing requirements at multiple points of the model to be polished to obtain multiple clustering areas, wherein the multiple clustering areas carry polishing requirement identification information; based on the multiple clustering areas and the polishing requirement identification information, combined with preset polishing liquid information, performing polishing air pressure optimization to obtain multiple optimal polishing air pressure parameters for the multiple clustering areas; based on the first air pressure regulator, generating an air pressure control signal according to the multiple optimal polishing air pressure parameters for polishing control, and performing feedback optimization control through the second air pressure regulator, that is, by performing neighborhood clustering of polishing requirements at multiple points of the model to be polished and performing polishing air pressure optimization based on the clustering results, the technical goal of improving the accuracy and efficiency of semiconductor polishing is ultimately achieved, and the technical effect of improving product quality is achieved.

[0010] The above description is only an overview of the technical solution of the present application. In order to more clearly understand the technical means of the present application, which can be implemented in accordance with the contents of the description, and to make the above and other purposes, features and advantages of the present application more obvious and easy to understand, the specific implementation methods of the present application are specifically listed below. It should be understood that the content described in this section is not intended to identify the key or important features of the embodiments of the present application, nor is it intended to limit the scope of the present application. Other features of the present application will become easy to understand through the following description. BRIEF DESCRIPTION OF THE DRAWINGS

[0011] In order to more clearly illustrate the technical solutions in this application or the prior art, the following briefly introduces the drawings required for use in the embodiments or the description of the prior art. Obviously, the drawings described below are merely exemplary, and a person of ordinary skill in the art can obtain other drawings based on the provided drawings without creative work.

[0012] Figure 1 This is a flow chart of the adaptive gas pressure control method for semiconductor polishing in this application;

[0013] Figure 2 This is a schematic flow chart of obtaining multiple clustering regions in the adaptive gas pressure control method for semiconductor polishing in this application. DETAILED DESCRIPTION

[0014] This application provides an adaptive air pressure control method for semiconductor polishing, resolving the technical problem in existing technologies where, due to the lack of grouping of multiple points with similar polishing requirements, polishing parameters for different areas are not accurately set, resulting in low polishing efficiency and further impacting product quality and performance. This method achieves the technical goal of improving the accuracy and efficiency of semiconductor polishing and achieves the technical effect of improving product quality.

[0015] Below, the technical solutions in this application will be clearly and completely described with reference to the accompanying drawings. Obviously, the described embodiments are only part of the embodiments of this application, rather than all the embodiments of this application. It should be understood that this application is not limited to the example embodiments described herein. Based on the embodiments of this application, all other embodiments obtained by ordinary technicians in this field without making creative work are within the scope of protection of this application. It should also be noted that, for the convenience of description, only the parts related to this application, rather than all of them, are shown in the accompanying drawings.

[0016] Example

[0017] Please see the attached Figure 1 The present application provides an adaptive gas pressure control method for semiconductor polishing, wherein the method specifically comprises the following steps:

[0018] Step 1: Obtain a cascade air pressure control module of a chemical mechanical polishing device, wherein the cascade air pressure control module includes a first air pressure regulator and a second air pressure regulator;

[0019] Specifically, chemical mechanical polishing equipment is used in semiconductor manufacturing to remove tiny bumps and unevenness from the surface of silicon wafers, achieving a highly smooth surface. During the manufacturing process, a cascaded air pressure control module precisely regulates the air pressure during the polishing process, ensuring polishing quality and efficiency. The cascaded air pressure control module includes a first air pressure regulator and a second air pressure regulator. The first air pressure regulator is an electric proportional valve. The second air pressure regulator contains an air pressure sensor and a PID controller.

[0020] Step 2: constructing an original 3D model of the semiconductor wafer to be processed and a preset polishing target 3D model;

[0021] Specifically, a sample of the finished product of the semiconductor wafer to be processed is obtained, representing the final state of the wafer after the complete polishing process. The finished product sample is scanned using laser scanning technology. The laser scanning technology emits a laser beam and measures the time it takes to reflect back to obtain three-dimensional coordinate data of the sample surface. The obtained sample point cloud data is processed to generate a three-dimensional model, and the point cloud data is converted into a continuous surface in three-dimensional space, thereby constructing a three-dimensional model of the sample to accurately reflect the shape and details of the sample surface. Based on the processed three-dimensional model, a preset polishing target three-dimensional model is generated as a reference during the polishing process to guide the movement and parameter adjustment of the polishing equipment. By comparing the original three-dimensional model of the semiconductor wafer to be processed with the preset polishing target three-dimensional model, the amount of material to be removed and the polishing path are determined, thereby achieving precise polishing control.

[0022] Step 3: performing spatial alignment and fitting on the original three-dimensional model and the preset polishing target three-dimensional model to generate a model to be polished;

[0023] Specifically, an alignment benchmark is selected and the point cloud data of the original 3D model is aligned with the point cloud data of the preset polishing target 3D model. The differences between the original 3D model and the preset polishing target 3D model are compared. The differences are subtracted from the original 3D model to obtain the model to be polished.

[0024] Step 4: performing neighborhood clustering on the polishing requirements of multiple points of the model to be polished to obtain multiple clustering areas, wherein the multiple clustering areas carry polishing requirement identification information;

[0025] Specifically, multiple points are selected on the model to be polished, and polishing thickness information is extracted at these points. The deviation between the polishing thickness information at these points, known as the neighborhood thickness deviation, is calculated. The calculated neighborhood thickness deviation is compared with a preset neighborhood deviation threshold. If the neighborhood thickness deviation is less than or equal to the preset threshold, it indicates that the polishing requirements of adjacent points are similar, and they are merged into clusters. Multiple clusters are then obtained, each with points having similar polishing requirements.

[0026] Step 5: Based on the multiple clustered areas and the polishing requirement identification information, combined with the preset polishing liquid information, perform polishing gas pressure optimization to obtain multiple optimal polishing gas pressure parameters for the multiple clustered areas;

[0027] Specifically, in the historical polishing control record database, the polishing requirement identification information of multiple cluster areas is searched to find historical records similar to the current polishing requirements. For each cluster area, the historical records similar to the current polishing requirements are used to construct the polishing consistency fitness optimization. The candidate parameter set is generated by random combination and cross-expansion of the historical polishing control record set to find the optimal polishing air pressure parameter.

[0028] Step six: Based on the first air pressure regulator, generate an air pressure control signal according to the multiple optimal polishing air pressure parameters to perform polishing control, and perform feedback optimization control through the second air pressure regulator.

[0029] Specifically, based on multiple optimal polishing air pressure parameters, an air pressure control signal is generated by the first air pressure regulator to perform polishing control, and the air pressure data of the cylinder is collected in real time through the second air pressure regulator to reflect the actual air pressure conditions during the polishing process. The collected real-time air pressure data of the cylinder is sent to the first air pressure regulator to compensate for pressure control losses.

[0030] The adaptive gas pressure control method for semiconductor polishing can achieve the technical goal of improving the accuracy and efficiency of semiconductor polishing and achieve the technical effect of improving product quality.

[0031] Furthermore, the present application further comprises the following steps:

[0032] The first air pressure regulator is an electric proportional valve, and the second air pressure regulator includes an air pressure sensor and a PID controller arranged at the cylinder position in the chemical mechanical polishing equipment, the air pressure sensor is connected to one end of the PID controller, and the other end of the PID controller is connected to the first air pressure regulator;

[0033] The second air pressure regulator is used to perform feedback optimization of the air pressure control signal of the first air pressure regulator.

[0034] Specifically, if Figure 2As shown, the first air pressure regulator in the cascaded air pressure control module utilizes an electric proportional valve. This valve uses an electrical signal to control the position of the valve core, thereby regulating flow or pressure. In chemical mechanical polishing equipment, the electric proportional valve serves as the main regulator, connected to the high-pressure air source. It regulates the input air pressure to a preset initial value and, based on signals from the control circuit, opens and closes the air supply and exhaust solenoid valves, thereby controlling the air pressure within the pilot chamber. A pressure sensor detects the signal and feeds it back to the control circuit. The electromagnetic force moves the valve core, ultimately outputting the desired air pressure. The accuracy and stability of the electric proportional valve ensures basic air pressure control during the polishing process. Furthermore, the second air pressure regulator consists of an air pressure sensor and a PID controller, located at the cylinder position within the chemical mechanical polishing equipment. The air pressure sensor measures air pressure changes within the cylinder. Using physical carriers such as resistors, capacitors, thermoelectric devices, and semiconductors, the air pressure changes are converted into an electrical signal output, which is then converted into voltage, current, or frequency for transmission, processing, and control. By monitoring the air pressure at the cylinder position in real time, the air pressure sensor provides real-time feedback data to the PID controller. The PID controller receives the feedback signal from the air pressure sensor and, based on the deviation between the setpoint and the measured value, uses proportional, integral, and differential calculations to generate a control signal, which is then transmitted to the electrical proportional valve in the first air pressure regulator for fine-tuning. The PID controller's goal is to keep the actual air pressure as close as possible to the setpoint, thereby achieving precise control of the air pressure during the polishing process.

[0035] Through the coordinated work of the electric proportional valve, air pressure sensor and PID controller, precise and stable control of the air pressure in the chemical mechanical polishing equipment is achieved, providing a strong guarantee for high-quality polishing operations.

[0036] Furthermore, the present application further comprises the following steps:

[0037] Extracting multiple polishing thickness information of multiple points based on the model to be polished;

[0038] extracting first polishing thickness information and second polishing thickness information of adjacent first and second points from the plurality of polishing thickness information;

[0039] Calculating a neighborhood thickness deviation between the first polishing thickness information and the second polishing thickness information;

[0040] Determine whether the neighborhood thickness deviation meets a preset neighborhood deviation threshold, and if so, merge the first point and the second point;

[0041] Repeating the neighborhood point merging analysis until the plurality of points are traversed, and obtaining the plurality of clustering areas based on the merging results;

[0042] The average values of the polishing thickness information within the plurality of clustering regions are calculated to generate the polishing requirement identification information.

[0043] Specifically, multiple points are selected on the model to be polished, and the polishing thickness information at these points is extracted. The multiple points can be evenly distributed across the entire model, or they can be selected specifically based on the complexity of the model surface and the polishing requirements.

[0044] Then, a first adjacent point and a second adjacent point are randomly selected from the extracted points, and first polishing thickness information and second polishing thickness information are obtained to analyze the polishing thickness difference between the adjacent points.

[0045] Next, the deviation between the polishing thickness information of the first point and the second point, ie, the neighborhood thickness deviation, is calculated to reflect the degree of difference in polishing requirements between adjacent points.

[0046] Next, the calculated neighborhood thickness deviation is compared with the preset neighborhood deviation threshold. If the neighborhood thickness deviation is less than or equal to the preset threshold, it means that the polishing requirements of adjacent points are similar and can be merged into one cluster area.

[0047] Then, the polishing thickness information of adjacent points is repeatedly extracted, the neighborhood thickness deviation is calculated, and the points are judged and merged until all points are traversed, thereby gradually forming multiple clustering areas, and the points in each area have similar polishing requirements.

[0048] Next, for each formed cluster area, the mean value of the polishing thickness information of all points in the cluster area is calculated to represent the overall polishing requirement of the cluster area, and corresponding polishing requirement identification information is generated.

[0049] By clustering multiple points on the model to be polished according to polishing requirements and generating clear polishing requirement identification information for each cluster area, it helps to optimize the polishing path, reduce errors in the polishing process, and improve the overall polishing efficiency and quality.

[0050] Furthermore, the present application further comprises the following steps:

[0051] Performing three-dimensional structural alignment on the original three-dimensional model and the preset polishing target three-dimensional model;

[0052] The original three-dimensional model is subjected to removal position extraction based on the alignment result, and the model to be polished is generated using the removal position extraction result.

[0053] Specifically, an alignment reference is selected. For example, the alignment reference can be a known point, line, or surface that exists in both the original 3D model and the predetermined polishing target 3D model and has identical geometric properties. A point cloud registration algorithm, such as the iterative closest point algorithm, is used to align the point cloud data of the original 3D model and the point cloud data of the predetermined polishing target 3D model. During the alignment process, the optimal rotation and translation between the two point clouds is iteratively calculated to minimize the distance between corresponding points.

[0054] After the 3D structure is aligned, the original 3D model is compared with the pre-set polishing target 3D model. By calculating the difference between the two models, the areas requiring removal are extracted. Boolean operations are performed on the point cloud data to generate point cloud data representing the removal areas. Based on the removal location map, one or more removal layers are constructed on the original 3D model, representing the portions of material to be progressively removed. The removal layers are subtracted from the original 3D model to produce the model to be polished, which reflects the desired intermediate state during the polishing process and provides precise guidance for the polishing equipment.

[0055] By ensuring the accuracy of polishing operations, optimizing polishing paths, and reducing unnecessary material removal, polishing efficiency and quality are improved.

[0056] Furthermore, the present application further comprises the following steps:

[0057] Based on the multiple clustered areas and the polishing requirement identification information, searching historical polishing control records to obtain multiple historical polishing control record sets for the multiple clustered areas, wherein any historical polishing control record includes polishing control parameters and surface roughness and material removal rate of the semiconductor wafer;

[0058] Constructing a polishing consistency fitness function for the multiple clustering regions:

[0059] The plurality of historical polishing control record sets are subjected to consistency optimization through the polishing consistency fitness function to generate the plurality of optimal polishing gas pressure parameters.

[0060] Specifically, the polishing control record database is searched based on the polishing requirement identification information of multiple clustered areas to find historical records similar to the current polishing requirements. The historical records contain polishing control parameters and corresponding information such as semiconductor wafer surface roughness and material removal rate. The material removal rate is the amount of waste material that needs to be polished away. Waste material is material that has been over-polished or damaged.

[0061] Then, for each cluster region, the consistency of polishing results under different polishing control parameters is evaluated by the polishing consistency fitness function.

[0062] Next, the polishing consistency fitness function is used to perform consistency optimization on multiple historical polishing control record sets. The candidate parameter set is generated by random combination and cross-expansion of the historical polishing control record sets, and the polishing consistency fitness function is used to find the optimal polishing air pressure parameter to meet the polishing consistency and quality requirements. The polishing control parameters are continuously adjusted to find the optimal parameter combination that can make the polishing results most consistent.

[0063] By using the information in the historical polishing control records, constructing a polishing consistency fitness function and performing consistency optimization, the optimal polishing gas pressure parameters for multiple clustering areas are generated, which will help improve the stability and efficiency of the polishing process while ensuring that the surface quality of semiconductor wafers meets the requirements.

[0064] Furthermore, the present application further comprises the following steps:

[0065] The polishing consistency fitness function is:

[0066]

[0067] FIT=σ 2 [|FQ i -FQ i+1 |];

[0068] Among them, FQ i is the independent fitness of the i-th cluster area; SR i is the surface roughness of the i-th cluster area; MRR i is the material removal rate of the i-th cluster area; S MRR is the standard material removal rate threshold; i∈N, N is the number of regions in the multiple clustering regions; ω1 and ω2 are weight coefficients; FIT is the polishing consistency fitness of multiple clustering regions; σ 2 [|FQ i -FQ i+1 |] is the variance of the independent fitness difference between the i-th cluster region and the i+1-th cluster region.

[0069] Specifically, the independent fitness FQ i is the fitness value calculated separately for the i-th cluster area, based on the surface roughness SR of the i-th cluster area i , material removal rate MRR of the i-th cluster area i and the standard material removal rate threshold S MRR For comprehensive evaluation. i∈N, N is the number of cluster regions. ω1 and ω2 are weight coefficients used for weighted calculation. The surface roughness SR of the i-th cluster region i The smaller it is, the independent fitness FQ of the i-th cluster area iThe larger the value, the smaller the value. Material removal rate MRR of the i-th cluster area i and the standard material removal rate threshold S MRR The smaller the difference is, the higher the independent fitness FQ of the i-th cluster area i Furthermore, the polishing consistency fitness FIT of multiple cluster regions is based on the variance σ of the independent fitness difference between the i-th cluster region and the i+1-th cluster region. 2 [|FQ i -FQ i+1 |] is obtained. The variance of the independent fitness differences is used to measure the consistency of polishing results between different clusters. A smaller variance indicates closer polishing results across clusters, higher consistency, and a smoother polished surface. Conversely, a smaller variance indicates a smoother polishing surface.

[0070] The polishing results of multiple clustered areas are evaluated by the polishing consistency fitness function, and the polishing control parameters that make the polishing consistency optimal are found for adjustment and optimization.

[0071] Furthermore, the present application further comprises the following steps:

[0072] Based on the multiple historical polishing control record sets of the multiple clustered regions, randomly combining the control parameters of the multiple regions to obtain multiple regional parameter combination results;

[0073] Cross-expanding the regional parameter combination results to obtain a regional candidate parameter set;

[0074] Obtaining multiple polishing consistency fitnesses of multiple regional candidate parameters in the regional candidate parameter set by calculating the polishing consistency fitness function;

[0075] According to the multiple polishing consistency fitnesses, the region candidate parameters corresponding to the minimum polishing consistency fitness are extracted to obtain the multiple optimal polishing pressure parameters.

[0076] Specifically, control parameters are randomly selected and combined from multiple historical polishing control records for multiple clustered regions. Each clustered region has a set of historical polishing control records, including different polishing air pressure parameters and other control parameters. By randomly combining parameters, multiple regional parameter combination results are generated.

[0077] Then, the regional parameter combination results are cross-expanded, and new regional candidate parameter sets are generated by exchanging or mixing parameters in different regional parameter combinations to expand the search range and increase the possibility of finding the optimal parameters.

[0078] Next, for each candidate parameter in each region, the polishing consistency fitness function is used to calculate its polishing consistency fitness. By calculating the polishing consistency fitness, the performance of each candidate parameter in achieving the polishing consistency and quality requirements is evaluated.

[0079] Next, the polishing consistency fitness of candidate parameters in all regions is compared, and the parameter with the smallest fitness is selected as the optimal polishing pressure parameter, indicating that the parameter with the smallest fitness meets the polishing quality requirements while maintaining polishing consistency.

[0080] A candidate parameter set is generated by random combination and cross-expansion of historical polishing control record sets, and the optimal polishing gas pressure parameters are found using the polishing consistency fitness function to meet the polishing consistency and quality requirements.

[0081] Furthermore, the present application further comprises the following steps:

[0082] Polishing the first cluster area is controlled by using a first optimal polishing pressure parameter among the multiple optimal polishing pressure parameters, and real-time cylinder pressure is collected by the pressure sensor in the second pressure regulator;

[0083] Sending the real-time air pressure of the cylinder to the PID controller to compensate for pressure control loss, generating the feedback control signal and sending it to the first air pressure regulator;

[0084] The first air pressure regulator optimizes and controls the electric proportional valve according to the feedback control signal.

[0085] Specifically, a first optimal polishing pressure parameter from among the multiple optimal polishing pressure parameters is randomly selected and used to control polishing of the first clustered area. During the polishing process, the air pressure data of the cylinder is collected in real time by the air pressure sensor in the second air pressure regulator to reflect the actual air pressure during the polishing process.

[0086] The collected real-time cylinder pressure data is then sent to a PID controller to compensate for pressure control losses. By calculating and processing the deviation between the real-time pressure and the set pressure, a corresponding feedback control signal is generated to compensate for any pressure losses that may occur during the polishing process. The PID controller calculates the feedback control signal based on the real-time pressure data and the set optimal polishing pressure parameters. The feedback control signal contains instructions for adjusting the pressure and is sent to the first pressure regulator to bring the actual pressure closer to the set optimal value.

[0087] Then, after receiving the feedback control signal, the first air pressure regulator optimizes and controls the electric proportional valve according to the instructions in the feedback control signal, and adjusts the opening of the electric proportional valve, thereby achieving precise control of the polishing air pressure.

[0088] By achieving real-time monitoring and precise control of the polishing air pressure, the air pressure is ensured to be stable during the polishing process and the optimal polishing effect is achieved, thereby improving the quality and consistency of the polishing and also improving the efficiency and stability of the polishing process.

[0089] In summary, the adaptive air pressure control method for semiconductor polishing provided by this application has the following technical effects:

[0090] By obtaining a cascade air pressure control module of a chemical mechanical polishing device, wherein the cascade air pressure control module includes a first air pressure regulator and a second air pressure regulator; constructing an original three-dimensional model of a semiconductor wafer to be processed and a preset polishing target three-dimensional model; spatially aligning and fitting the original three-dimensional model and the preset polishing target three-dimensional model to generate a model to be polished; performing neighborhood clustering of polishing requirements at multiple points of the model to be polished to obtain multiple clustering areas, wherein the multiple clustering areas carry polishing requirement identification information; based on the multiple clustering areas and the polishing requirement identification information, combined with preset polishing liquid information, performing polishing air pressure optimization to obtain multiple optimal polishing air pressure parameters for the multiple clustering areas; based on the first air pressure regulator, generating an air pressure control signal according to the multiple optimal polishing air pressure parameters for polishing control, and performing feedback optimization control through the second air pressure regulator, that is, by performing neighborhood clustering of polishing requirements at multiple points of the model to be polished and performing polishing air pressure optimization based on the clustering results, the technical goal of improving the accuracy and efficiency of semiconductor polishing is ultimately achieved, and the technical effect of improving product quality is achieved.

[0091] The above description of the disclosed embodiments is intended to enable one skilled in the art to implement or use the present application. Various modifications to these embodiments will be readily apparent to one skilled in the art, and the general principles defined herein may be implemented in other embodiments without departing from the spirit or scope of the present application. Therefore, the present application is not limited to the embodiments shown herein, but is intended to conform to the widest scope consistent with the principles and novel features disclosed herein.

[0092] Obviously, those skilled in the art may make various changes and modifications to the present application without departing from the spirit and scope of the present application. Thus, if these modifications and variations of the present application fall within the scope of the present application and its equivalents, the present application is intended to include these modifications and variations.

Claims

1. An adaptive gas pressure control method for semiconductor polishing, characterized in that: The method comprises: Obtaining a cascade air pressure control module of a chemical mechanical polishing device, wherein the cascade air pressure control module includes a first air pressure regulator and a second air pressure regulator; Constructing an original three-dimensional model of the semiconductor wafer to be processed and a preset polishing target three-dimensional model; Performing spatial alignment and fitting on the original three-dimensional model and the preset polishing target three-dimensional model to generate a model to be polished; Performing neighborhood clustering on the polishing requirements of multiple points of the model to be polished to obtain multiple clustering areas, wherein the multiple clustering areas carry polishing requirement identification information; Based on the multiple clustered areas and the polishing requirement identification information, combined with preset polishing liquid information, performing polishing gas pressure optimization to obtain multiple optimal polishing gas pressure parameters for the multiple clustered areas; Based on the first air pressure regulator, an air pressure control signal is generated according to the multiple optimal polishing air pressure parameters to perform polishing control, and feedback optimization control is performed through the second air pressure regulator.

2. The method according to claim 1, wherein The first air pressure regulator is an electric proportional valve, and the second air pressure regulator includes an air pressure sensor and a PID controller arranged at the cylinder position in the chemical mechanical polishing equipment, the air pressure sensor is connected to one end of the PID controller, and the other end of the PID controller is connected to the first air pressure regulator; The second air pressure regulator is used to perform feedback optimization of the air pressure control signal of the first air pressure regulator.

3. The method according to claim 1, wherein Performing neighborhood clustering on the polishing requirements of multiple points of the model to be polished to obtain multiple clustering areas, wherein the multiple clustering areas carry polishing requirement identification information, including: Extracting multiple polishing thickness information of multiple points based on the model to be polished; extracting first polishing thickness information and second polishing thickness information of adjacent first and second points from the plurality of polishing thickness information; Calculating a neighborhood thickness deviation between the first polishing thickness information and the second polishing thickness information; Determine whether the neighborhood thickness deviation meets a preset neighborhood deviation threshold, and if so, merge the first point and the second point; Repeating the neighborhood point merging analysis until the plurality of points are traversed, and obtaining the plurality of clustering areas based on the merging results; The average values of the polishing thickness information within the plurality of clustering regions are calculated to generate the polishing requirement identification information.

4. The method according to claim 1, wherein Performing spatial alignment and fitting on the original three-dimensional model and the preset polishing target three-dimensional model to generate a model to be polished includes: Performing three-dimensional structural alignment on the original three-dimensional model and the preset polishing target three-dimensional model; The original three-dimensional model is subjected to removal position extraction based on the alignment result, and the model to be polished is generated using the removal position extraction result.

5. The method according to claim 1, wherein Based on the multiple clustered areas and the polishing requirement identification information, combined with preset polishing liquid information, polishing gas pressure optimization is performed to obtain multiple optimal polishing gas pressure parameters for the multiple clustered areas, including: Based on the multiple clustered areas and the polishing requirement identification information, searching historical polishing control records to obtain multiple historical polishing control record sets for the multiple clustered areas, wherein any historical polishing control record includes polishing control parameters and surface roughness and material removal rate of the semiconductor wafer; Constructing a polishing consistency fitness function for the multiple clustering regions: The plurality of historical polishing control record sets are subjected to consistency optimization through the polishing consistency fitness function to generate the plurality of optimal polishing gas pressure parameters.

6. The method according to claim 5, wherein The polishing consistency fitness function is: FIT=σ 2 [|FQ i -FQ i+1 |]; Among them, FQ i is the independent fitness of the i-th cluster area; SR i is the surface roughness of the i-th cluster area; MRR i is the material removal rate of the i-th cluster area; S MRR is the standard material removal rate threshold; i∈N, N is the number of regions in the multiple clustering regions; ω1 and ω2 are weight coefficients; FIT is the polishing consistency fitness of multiple clustering regions; σ 2 [|FQ i -FQ i+1 |] is the variance of the independent fitness difference between the i-th cluster region and the i+1-th cluster region.

7. The method according to claim 5, wherein Performing consistency optimization on multiple historical polishing control record sets using the polishing consistency fitness function to generate the multiple optimal polishing gas pressure parameters includes: Based on the multiple historical polishing control record sets of the multiple clustered regions, randomly combining the control parameters of the multiple regions to obtain multiple regional parameter combination results; Cross-expanding the regional parameter combination results to obtain a regional candidate parameter set; Obtaining multiple polishing consistency fitnesses of multiple regional candidate parameters in the regional candidate parameter set by calculating the polishing consistency fitness function; According to the multiple polishing consistency fitnesses, the region candidate parameters corresponding to the minimum polishing consistency fitness are extracted to obtain the multiple optimal polishing pressure parameters.

8. The method according to claim 2, wherein Based on the first air pressure regulator, an air pressure control signal is generated according to the multiple optimal polishing air pressure parameters to perform polishing control, and feedback optimization control is performed through the second air pressure regulator, including: Polishing the first cluster area is controlled by using a first optimal polishing pressure parameter among the multiple optimal polishing pressure parameters, and real-time cylinder pressure is collected by the pressure sensor in the second pressure regulator; Sending the real-time air pressure of the cylinder to the PID controller to compensate for pressure control loss, generating the feedback control signal and sending it to the first air pressure regulator; The first air pressure regulator optimizes and controls the electric proportional valve according to the feedback control signal.

Images