Method, system, device and medium for adjusting semiconductor process recipe parameters

Abstract

The embodiments of the present application relate to the field of semiconductor technology, and disclose a method, system, device and medium for adjusting semiconductor process recipe parameters, the method comprising: training a multivariate regression model using a data set containing one or more batches of semiconductor process recipe parameters and corresponding product yields; substituting fixed parameters in the recipe to be adjusted into the multivariate regression model to obtain the numerical value or numerical range of the parameter to be adjusted in the recipe to be adjusted based on the target yield. Through the scheme of the present application, at least the technical problems that the prior art parameter adjustment depends on the experience of engineers, is time-consuming and labor-intensive, and the decision-making is subjective and inconsistent can be solved. The present application can predict the relationship between key process parameters and yield, make intelligent recommendations to engineers, and improve production efficiency.

Description

Technical Field

[0001] The present application relates to the field of semiconductor technology, and in particular to a method, system, device and medium for adjusting semiconductor process recipe parameters. Background Art

[0002] Adjustment of semiconductor process recipe parameters has become a key challenge in the modern integrated circuit manufacturing industry, which is closely related to chip performance, yield and reliability. As chip manufacturing processes continue to shrink, precise control of process parameters becomes increasingly important, and not all parameter adjustments can bring the expected results, which leads to the concepts of process window and critical parameters. The process window defines a set of parameter ranges within which chips that meet specifications can be produced; critical parameters are those that have the greatest impact on the quality of the final product. Identifying and optimizing these parameters is crucial to improving chip yield, reducing costs and ensuring product quality. Adjustment of semiconductor process recipe parameters involves precise control and optimization of these critical parameters. This adjustment is of great significance for maintaining production stability, improving product performance and yield, and reducing costs. Therefore, effective parameter adjustment methods are particularly important for early detection of potential problems, preventing defects and managing complex manufacturing processes, which helps reduce material waste, improve production efficiency and enhance corporate competitiveness.

[0003] At present, the adjustment methods of semiconductor process recipe parameters are mainly based on statistical process control (SPC), design of experiments (DOE) and engineers' experience and judgment. These methods determine the optimal parameter settings by collecting and analyzing a large amount of production data and combining statistical principles. For example, SPC identifies abnormal conditions by monitoring the changing trends of key parameters; DOE finds the best combination by systematically changing multiple parameters. These methods are effective to a certain extent, but they also have obvious limitations. First, they often require a lot of experiments and data collection, which is time-consuming and labor-intensive; second, when faced with highly complex and nonlinear semiconductor manufacturing processes, traditional methods may find it difficult to capture the complex interactions between parameters; finally, these methods rely heavily on engineers' experience and intuition, which may lead to subjectivity and inconsistency in decision-making.

[0004] Therefore, there is an urgent need for a technical solution that can predict the relationship between key process parameters and yield, make intelligent recommendations to engineers, and improve production efficiency. Summary of the invention

[0005] One purpose of the present application is to provide a method, system, device and medium for adjusting semiconductor process recipe parameters, at least to solve the technical problems that the existing technical parameter adjustment relies on the experience of engineers, is time-consuming and labor-intensive, and the decision-making is subjective and inconsistent.

[0006] To achieve the above objectives, some embodiments of the present application provide the following aspects:

[0007] In a first aspect, some embodiments of the present application provide a method for adjusting semiconductor process recipe parameters, comprising: training a multivariate regression model using a data set containing one or more batches of semiconductor process recipe parameters and corresponding product yields; substituting fixed parameters in the recipe to be adjusted into the multivariate regression model to obtain a numerical value or numerical range of the parameter to be adjusted in the recipe to be adjusted based on a target yield.

[0008] Furthermore, a multivariate regression model is trained using a data set containing one or more batches of semiconductor process recipe parameters and corresponding product yields, including: recording the process recipe parameters used for each batch of wafers during the manufacturing process; performing sampling inspection on some wafers in each batch, wherein the inspection includes CD measurement; associating the inspection results with the corresponding process recipe parameters to obtain historical data, wherein the historical data includes the process recipe name, the product number to which the wafer belongs, the value of the process recipe parameter and / or the qualified status of the wafer CD measurement; and constructing a data set based on the historical data to train the multivariate regression model.

[0009] Furthermore, a data set is constructed based on historical data to train a multivariate regression model, including: calculating the product yield corresponding to the process recipe parameters based on the historical data to obtain a data set containing the process recipe parameters and the yield; determining the number of independent variables based on the process recipe parameters in the data set to construct a multivariate regression model; using the numerical values ​​of the process recipe parameters as independent variables and the corresponding yield as the dependent variable to perform fitting training on the multivariate regression model.

[0010] Furthermore, the fixed parameters in the recipe to be adjusted are substituted into the multivariate regression model to obtain the numerical value or numerical range of the parameters to be adjusted in the recipe to be adjusted based on the target yield, including: obtaining parameter adjustment instructions; dividing the parameters in the recipe to be adjusted in response to the parameter adjustment instructions to obtain fixed parameters and parameters to be adjusted; substituting the numerical values ​​of the fixed parameters as independent variables into the multivariate regression model to obtain a univariate regression model; substituting the set target yield as the dependent variable into the univariate regression model to solve the independent variable of the parameter to be adjusted; and obtaining the numerical value or numerical range of the parameter to be adjusted based on the solution result.

[0011] Furthermore, the method also includes: presenting a curve graph of the univariate regression model; and presenting a numerical table of the curve graph based on a set step size.

[0012] Furthermore, the multivariate regression model includes a multivariate quadratic regression model.

[0013] In the second aspect, some embodiments of the present application also provide a semiconductor process recipe parameter adjustment system that applies the adjustment method described in the above embodiments, including: a parameter unit, a measurement unit and a calculation unit; wherein the parameter unit is used to obtain process recipe parameters; the measurement unit is used to obtain CD measurement results; the calculation unit is used to train a multivariate regression model using a data set containing one or more batches of semiconductor process recipe parameters and corresponding product yields; and the fixed parameters in the recipe to be adjusted are substituted into the multivariate regression model to obtain the numerical value or numerical range of the parameter to be adjusted in the recipe to be adjusted based on the target yield.

[0014] In a third aspect, some embodiments of the present application further provide an electronic device, comprising: one or more processors; and a memory storing computer program instructions, wherein the computer program instructions, when executed, cause the processor to perform the steps of the method described above.

[0015] In a fourth aspect, some embodiments of the present application further provide a computer-readable medium having computer program instructions stored thereon, wherein the computer program instructions can be executed by a processor to implement the method as described above.

[0016] In a fifth aspect, some embodiments of the present application further provide a computer program product, comprising a computer program / instruction, which implements the steps of the method described above when executed by a processor.

[0017] Compared with the related art, the solution provided in the embodiment of the present application can automatically set a reasonable control range for key process parameters by using historical data and machine learning algorithms through intelligent parameter setting, reducing the reliance on the personal experience of engineers. Furthermore, through data analysis and intelligent algorithms, the system can identify and recommend more accurate parameter control values, thereby improving the stability of the production process. Furthermore, through automated and intelligent parameter control settings, the time for manual analysis is reduced, the production preparation and adjustment process is accelerated, and the overall production efficiency is improved. BRIEF DESCRIPTION OF THE DRAWINGS

[0018] One or more embodiments are exemplarily described by pictures in the corresponding drawings, and these exemplified descriptions do not constitute limitations on the embodiments. Elements with the same reference numerals in the drawings represent similar elements, and unless otherwise stated, the figures in the drawings do not constitute proportional limitations.

[0019] Figure 1 A schematic flow chart of a method for adjusting semiconductor process recipe parameters provided in an embodiment of the present application;

[0020] Figure 2 A schematic diagram of a flow chart of a method for training a multivariate regression model provided in an embodiment of the present application;

[0021] Figure 3 A flow chart of a method for adjusting parameters provided in an embodiment of the present application;

[0022] Figure 4 A schematic diagram showing a regression model provided in an embodiment of the present application;

[0023] Figure 5 A schematic diagram of a recommended parameter range provided in an embodiment of the present application;

[0024] Figure 6 A schematic diagram of the structure of a semiconductor process recipe parameter adjustment system provided in an embodiment of the present application;

[0025] Figure 7 An exemplary structural diagram of an electronic device provided in an embodiment of the present application. DETAILED DESCRIPTION

[0026] In order to make the purpose, technical solution and advantages of the embodiments of the present application clearer, the technical solution in the embodiments of the present application will be clearly and completely described below in conjunction with the drawings in the embodiments of the present application. Obviously, the described embodiments are part of the embodiments of the present application, not all of the embodiments. Based on the embodiments in the present application, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of this application.

[0027] The embodiments of the present application disclose a method, system, device and medium for adjusting semiconductor process recipe parameters. The method predicts the relationship between key parameters and yield based on historical measurement data and yield, thereby finding the key parameter range where the yield is higher than a threshold, making intelligent recommendations to engineers, reducing reliance on engineers' experience, and improving production efficiency.

[0028] First embodiment

[0029] Figure 1 A schematic diagram of a method for adjusting semiconductor process recipe parameters provided in an embodiment of the present application. Figure 1 As shown, the first embodiment of the present application relates to a method for adjusting semiconductor process recipe parameters, comprising the following steps:

[0030] Step S101 , training a multivariate regression model using a data set including one or more batches of semiconductor process recipe parameters and corresponding product yields.

[0031] It should be understood that in the semiconductor manufacturing process, process recipe parameters refer to various variables that control the manufacturing process, such as temperature, pressure, gas flow, exposure time, etc. These parameters have a direct impact on the quality of the final product. The embodiment of the present application requires that the process recipe parameters used for each batch of wafers be recorded in detail during the manufacturing process. These parameters may include, but are not limited to, photolithography process parameters: exposure energy, focal length, mask offset, etc.; etching process parameters: gas flow, RF power, pressure, etc.; thin film deposition parameters: temperature, pressure, gas ratio, etc.; chemical mechanical planarization (CMP) parameters: pressure, rotation speed, polishing liquid composition, etc. These parameters should be recorded in a structured form to facilitate subsequent data processing and analysis.

[0032] In one embodiment, the process recipe parameters used for each batch of wafers are recorded during the manufacturing process; a portion of the wafers in each batch are sampled and tested, and the test includes CD measurement; the test results are associated with the corresponding process recipe parameters and stored to obtain historical data, and the historical data includes the process recipe name, the product number of the wafer, the value of the process recipe parameter and / or the qualified status of the CD measurement of the wafer; a data set is constructed based on the historical data to train a multivariate regression model (the training method is as follows Figure 2 As shown in Figure 2 for expanded instructions).

[0033] Quality inspection is a key step to ensure that the product meets the specifications. In the present invention, a portion of the wafers in each batch are sampled for inspection. CD measurement is particularly mentioned here, which is an important indicator in semiconductor manufacturing and is used to measure the size of key structures on the wafer. CD measurement can use a variety of techniques, such as: scanning electron microscopy (SEM), atomic force microscopy (AFM), optical CD measurement (OCD). In addition to CD measurement, quality inspection can also include other parameters, such as: film thickness, surface roughness, electrical performance testing, defect detection, and these test results will be used to evaluate the yield and quality of the product.

[0034] The multivariate regression model is a statistical method used to analyze the relationship between multiple independent variables (i.e., recipe parameters in the embodiment of the present application) and dependent variables (i.e., yield rate in the embodiment of the present application). The model establishes a mathematical relationship between parameters and yield rate through the analysis of historical data, which can be used for prediction and optimization. In addition, the multivariate regression model can be used in various forms, such as linear regression, polynomial regression, ridge regression, Lasso regression, elastic network regression, etc.

[0035] Step S102 , substituting the fixed parameters in the recipe to be adjusted into the multivariate regression model to obtain the value or value range of the parameter to be adjusted in the recipe to be adjusted based on the target yield.

[0036] A recipe to be adjusted is a recipe that needs to have its parameters adjusted to achieve a specific goal. The recipe usually contains multiple parameters, which can be divided into two categories: fixed parameters and parameters to be adjusted. Fixed parameters are parameters that remain unchanged during the adjustment process, while parameters to be adjusted are parameters that need to be calculated to determine their optimal value or appropriate range.

[0037] Traditional recipe adjustment methods usually rely on empirical rules or simple linear models, which often do not work well when dealing with much more complex parameter systems. They may ignore the interactions between parameters, resulting in adjustment results that are not accurate or reliable.

[0038] In addition, traditional methods usually require a lot of experiments and adjustments, which is time-consuming, labor-intensive and inefficient. The applicant found that by establishing an accurate multivariate regression model, the complex relationship between recipe parameters and yield can be better captured. This method is based on the principle of data-driven and can take into account the nonlinear relationship and interaction between parameters, thereby providing more accurate parameter adjustment suggestions.

[0039] Specifically, a parameter adjustment instruction is obtained; in response to the parameter adjustment instruction, the parameters in the recipe to be adjusted are divided to obtain fixed parameters and parameters to be adjusted; the values ​​of the fixed parameters are substituted into the multivariate regression model as independent variables to obtain a univariate regression model; the set target yield is substituted into the univariate regression model as the dependent variable to solve the independent variable of the parameter to be adjusted; and the value or value range of the parameter to be adjusted is obtained based on the solution result.

[0040] Figure 2 A flowchart of a method for training a multivariate regression model provided in an embodiment of the present application is shown in FIG. Figure 2 As shown, the method for training a multivariate regression model includes:

[0041] Step S201 , calculating the product yield corresponding to the process recipe parameters based on historical data to obtain a data set including the process recipe parameters and the yield.

[0042] Historical data refers to a large amount of production data that has been accumulated in the semiconductor manufacturing process in the above embodiments, including but not limited to the setting values ​​of various process parameters and the corresponding product qualification rate data. Semiconductor process recipe parameters refer to various controllable factors that affect product quality and performance in the semiconductor manufacturing process in the above embodiments, such as temperature, pressure, gas flow, reaction time, etc. These parameters are usually set based on experience or process requirements and have a direct impact on the yield of the final product. Product yield refers to the ratio of the number of products that meet quality standards to the total number of products within a certain period of time or a certain number of products, usually expressed as a percentage. Yield is an important indicator for measuring the efficiency of the semiconductor manufacturing process and product quality.

[0043] By calculating the product yield corresponding to the process recipe parameters, a data set containing process recipe parameters and yield can be established. This data set is the basis for the subsequent establishment of a multivariate regression model. The quality of the data set directly affects the accuracy and reliability of the model. Therefore, in this step, preprocessing operations such as data cleaning and outlier processing may be required to ensure the validity and representativeness of the data.

[0044] Step S202, determining the number of independent variables based on the process recipe parameters in the data set to construct a multiple regression model. The independent variables in this application refer to the various process recipe parameters that affect the product yield, and the dependent variable is the product yield. The multiple regression model is a statistical analysis method used to study the relationship between a dependent variable and multiple independent variables. In this application, the purpose of the multiple regression model is to establish a mathematical relationship between the process recipe parameters (independent variables) and the product yield (dependent variable).

[0045] Determining the number of independent variables is a key step in building a multivariate regression model. This step may involve the process of variable selection, that is, selecting the most explanatory and predictive parameters from the many process parameters that may affect the yield as the independent variables of the model. The methods of variable selection may include but are not limited to stepwise regression, principal component analysis, Lasso regression, etc. Selecting an appropriate number of independent variables can both ensure the explanatory power of the model and avoid overfitting problems.

[0046] Building a multiple regression model usually involves determining the mathematical form of the model. The most common form is the linear multiple regression model, which has the general expression: ,in is the dependent variable (product yield), is the independent variable (process recipe parameter), is the regression coefficient to be estimated, is a random error term. Depending on the actual situation, nonlinear relationships may also need to be considered, such as quadratic terms, interaction terms, etc.

[0047] Step S203, using the numerical value of the process recipe parameter as an independent variable and the corresponding yield as a dependent variable to perform fitting training on the multivariate regression model. The purpose of this step is to estimate the regression coefficient in the model through statistical methods so that the model can best fit the existing data.

[0048] Fitting training usually uses the least squares method or other optimization algorithms, such as gradient descent. The goal of these methods is to minimize the error between the predicted value and the actual value. During the training process, multiple iterations may be required to continuously adjust the regression coefficients until the preset convergence condition is reached or the maximum number of iterations is reached.

[0049] The training process may also involve model verification and evaluation. Common methods include cross-validation and holdout methods. These methods can help evaluate the generalization ability of the model, that is, the performance of the model on unseen data. At the same time, the goodness of fit and prediction accuracy of the model can also be evaluated by calculating indicators such as the coefficient of determination (R²), mean square error (MSE), and mean absolute error (MAE).

[0050] The advantage of the method of this application is that it provides a systematic method to analyze the relationship between semiconductor process recipe parameters and product yield. By establishing a multivariate regression model, the influence of each process parameter on the yield can be quantified, thereby providing data support for process optimization. This method can help engineers better understand the process, identify key parameters, and make targeted improvements on this basis.

[0051] In addition, this method also has predictive capabilities. By inputting new process recipe parameters, the possible product yield can be predicted, which is of great significance for production planning and risk assessment. For example, when introducing new process parameters or adjusting existing parameters, the model can be used to pre-evaluate the possible impact, thereby reducing trial and error costs and improving the scientific nature of decision-making.

[0052] Figure 3 The following is a flow chart of a parameter adjustment method provided in an embodiment of the present application. After the model is trained, the model can be used to adjust the semiconductor process recipe parameters. Figure 3 As shown, the parameter adjustment method includes:

[0053] Step S301, obtaining parameter adjustment instructions. Parameter adjustment instructions can come from multiple sources, such as instructions input by operators through a human-computer interaction interface, or instructions automatically generated by an automation system based on real-time monitoring data. The purpose of this step is to determine the parameters that need to be adjusted or the timing and range of parameter adjustment.

[0054] Step S302, in response to the parameter adjustment instruction, divide the parameters in the recipe to be adjusted to obtain fixed parameters and parameters to be adjusted. First, identify the specific values ​​of each parameter in the recipe currently being used, and these values ​​will be used as a benchmark when calculating. Then determine which parameters are the parameters to be adjusted that need to be optimized this time. For example, suppose there are four key parameters X1, X2, X3 and X4 in the current recipe, and their values ​​are 20, 30, 40 and 50 respectively. If X1 needs to be optimized this time, then X1 is the parameter to be adjusted, and the current values ​​of X2, X3 and X4 will be used as known conditions in subsequent calculations. It should be noted that although X2, X3 and X4 remain unchanged temporarily in this optimization, they are also adjustable parameters in essence, but adjustment is not considered in the current optimization task. This analysis method enables the embodiment of the present application to focus on the optimization space of a single parameter while keeping other parameters unchanged, thereby simplifying the complexity of the problem. At the same time, this method also retains the flexibility of optimizing other parameters in the future.

[0055] Step S303, substitute the numerical value of the fixed parameter as an independent variable into the multivariate regression model to obtain a univariate regression model. This step simplifies the complex multivariate problem into a univariate problem, thereby reducing the complexity of the solution. A univariate regression model refers to a regression model with only one independent variable, which can more intuitively describe the relationship between a single parameter and a target variable. Specifically, in combination with the embodiment described above, if the parameter X1 is to be optimized, the current values ​​(30, 40, 50) of X2, X3 and X4 are substituted into the multivariate regression model. In this way, a univariate quadratic regression model containing only X1 as a variable is obtained. This univariate model more intuitively describes the relationship between X1 and the target yield. In this way, the effect of the change of X1 on the yield can be studied while keeping other conditions unchanged. For example, it is assumed that the obtained univariate quadratic model is: Y = aX1^2+ bX1 + c, where Y represents the yield, and a, b, and c are constants. By solving this univariate quadratic equation, the value range of X1 that makes Y greater than a certain target threshold can be found. This range is the recommended adjustment range of X1 when other parameters remain unchanged. Similarly, X1, X3, and X4 can be fixed to study the optimization range of X2; X1, X2, and X4 can be fixed to study the optimization range of X3; and X1, X2, and X3 can be fixed to study the optimization range of X4. Through this parameter-by-parameter analysis method, the potential optimization space of each parameter can be fully understood. By substituting fixed parameters into the multivariate regression model, the relationship between the parameters to be adjusted and the target yield is actually studied while other conditions are fixed.

[0056] Step S304, substitute the set target yield as the dependent variable into the univariate regression model to solve the independent variable of the parameter to be adjusted. The target yield refers to the expected product qualification rate, which is an important indicator for measuring the quality of the semiconductor manufacturing process. By substituting the target yield into the model, the parameter values ​​required to achieve the target can be reversely derived. This solution process may involve numerical analysis methods, such as Newton's method, dichotomy method, etc., and the specific method selected may depend on the complexity of the model and the accuracy requirements of the solution.

[0057] Step S305, based on the solution result, obtain the value or value range of the parameter to be adjusted. The output of this step may be a specific value, indicating the optimal value of the parameter to be adjusted; or it may be a value range, indicating the acceptable range of the parameter to be adjusted. The provision of the value range can provide greater flexibility for the operator, while also taking into account the errors and fluctuations that may exist in actual production (such as Figure 4 or Figure 5 as shown).

[0058] Figure 4 A schematic diagram showing a regression model provided in an embodiment of the present application. The multivariate regression model shown in the figure includes a multivariate quadratic regression model. After obtaining the trained model and determining the parameters to be adjusted, a curve graph of the univariate regression model can be presented, and / or a numerical table of the curve graph can be presented based on a set step size. Taking a lithography device as an example, Figure 4 The effect of exposure energy on CD measurement yield of a certain product is demonstrated.

[0059] Figure 5 A schematic diagram of a recommended parameter range provided in an embodiment of the present application. Specifically, the model predicts the yield of the next product measurement after the key parameters are set, and reflects the relationship between the key parameters predicted by the model and the yield in the curve chart. The user selects the appropriate yield and automatically and intelligently recommends the appropriate range. According to the yield threshold, the appropriate key parameter control range is automatically recommended. For example, if the engineer selects the yield threshold of 99%, RMS automatically and intelligently recommends the exposure energy control range of 20.25~21.5.

[0060] Second embodiment

[0061] Figure 6A schematic diagram of a semiconductor process recipe parameter adjustment system provided in an embodiment of the present application. It should be understood that the system shown in the figure is exemplary and not restrictive. This means that the system architecture involved is not limited to a specific form or design, but is presented as an example. In other words, the architecture shown in the figure can be regarded as a way of expression to clearly describe related concepts and relationships, and does not exclude other forms of architecture. Therefore, when interpreting the architecture in the picture, it should be understood that the model is flexible and diverse, and its purpose is to provide an exemplary description rather than a restrictive provision on a specific form.

[0062] Specifically, if Figure 6 As shown, a semiconductor process recipe parameter adjustment system is used to execute the above adjustment method, the system comprising: a parameter unit 601, a measurement unit 602 and a calculation unit 603; wherein the parameter unit 601 is used to obtain process recipe parameters; the measurement unit 602 is used to obtain CD measurement results; the calculation unit 603 is used to train a multivariate regression model using a data set containing one or more batches of semiconductor process recipe parameters and corresponding product yields; the fixed parameters in the recipe to be adjusted are substituted into the multivariate regression model to obtain the value or value range of the parameter to be adjusted in the recipe to be adjusted based on the target yield.

[0063] In summary, the embodiments of the present application transform the traditional process of engineers manually setting the process parameter control range based on experience and offline data into a process algorithm intelligently recommending the key parameter control range based on historical measurement data and yield.

[0064] Intelligent parameter setting can use historical data and machine learning algorithms to automatically set reasonable control ranges for key process parameters, reducing reliance on engineers’ personal experience. Furthermore, through data analysis and intelligent algorithms, the system can identify and recommend more accurate parameter control values, thereby improving the stability of the production process. Furthermore, automated and intelligent parameter control settings reduce manual analysis time, speed up production preparation and adjustment processes, and improve overall production efficiency.

[0065] It is not difficult to find that this embodiment is a system embodiment corresponding to the first embodiment, and this embodiment can be implemented in conjunction with the first embodiment. The relevant technical details mentioned in the first embodiment are still valid in this embodiment, and in order to reduce repetition, they are not repeated here. Accordingly, the relevant technical details mentioned in this embodiment can also be applied in the first embodiment.

[0066] It is worth mentioning that all modules involved in this embodiment are logic modules. In practical applications, a logic unit can be a physical unit, a part of a physical unit, or a combination of multiple physical units. In addition, in order to highlight the innovative part of this application, this embodiment does not introduce units that are not closely related to solving the technical problems proposed by this application, but this does not mean that there are no other units in this embodiment.

[0067] Third embodiment

[0068] In addition, some embodiments of the present application also provide an electronic device. The electronic device may be a digital computer in various forms, such as a laptop computer, a desktop computer, a workbench, a personal digital assistant, a server, a blade server, a mainframe computer, etc. The electronic device may also be a mobile device in various forms, such as a personal digital processing, a cellular phone, a smart phone, a wearable device, and other similar computing devices.

[0069] The electronic device includes: one or more processors; and a memory storing computer program instructions, wherein when the computer program instructions are executed, the processor executes the steps of the method provided in any one or more of the above embodiments. Figure 7 An exemplary structural diagram of the electronic device is disclosed. Figure 7 As shown, the electronic device includes: one or more processors 1101, a memory 1102, and interfaces for connecting various components, including high-speed interfaces and low-speed interfaces. The various components are connected to each other using different buses and can be installed on a common motherboard or installed in other ways as needed. The processor can process instructions executed in the electronic device, including instructions stored in or on the memory to display graphical information of the GUI on an external input / output device (such as a display device coupled to the interface). In some other embodiments, if necessary, multiple processors and / or multiple buses can be used with multiple memories and multiple memories. Similarly, multiple electronic devices can be connected, and each device provides some necessary operations (for example, as a server array, a group of blade servers, or a multi-processor system). Among them, the components shown in this article, their connections and relationships, and their functions are only examples, and are not intended to limit the implementation of the present application described and / or required herein.

[0070] The electronic device may further include: an input device 1103 and an output device 1104. The processor 1101, the memory 1102, the input device 1103 and the output device 1104 may be connected via a bus or other means. Figure 2 The example of connecting through bus is taken in the following.

[0071] The input device 1103 can receive input digital or character information, and generate key signal input related to the user settings and function control of the electronic device, such as a touch screen, a keypad, a mouse, a track pad, a touch pad, an indicator rod, one or more mouse buttons, a trackball, a joystick and other input devices. The output device 1104 may include a display device, an auxiliary lighting device (e.g., an LED) and a tactile feedback device (e.g., a vibration motor), etc. The display device may include, but is not limited to, a liquid crystal display (LCD), a light emitting diode (LED) display and a plasma display. In some embodiments, the display device may be a touch screen.

[0072] To provide interaction with a user, the electronic device may be a computer. The computer has: a display device (e.g., a CRT (cathode ray tube) or LCD (liquid crystal display) monitor) for displaying information to the user; and a keyboard and a pointing device (e.g., a mouse or a trackball), through which the user can provide input to the computer. Other types of devices may also be used to provide interaction with a user; for example, the feedback provided to the user may be any form of sensory feedback (e.g., visual feedback, auditory feedback, or tactile feedback); and input from the user may be received in any form (including acoustic input, voice input, or tactile input).

[0073] Fourth embodiment

[0074] In the embodiments of the present application, a computer program / instruction is stored on a computer-readable medium, and when the computer program / instruction is executed by a processor, the steps of the method provided by any one or more of the above embodiments are implemented. The computer-readable medium may be included in the electronic device described in the above embodiments; or it may exist independently without being assembled into the device. The above computer-readable medium carries one or more computer-readable instructions.

[0075] The memory 1102 can be used as a non-transient computer-readable storage medium, which can be used to store non-transient software programs, non-transient computer executable programs and modules. The processor 1101 executes various functional applications and data processing of the server by running the non-transient software programs, instructions and modules stored in the memory 1102, so as to implement the program instructions / modules corresponding to the method provided by any one or more embodiments in the embodiments of the present application.

[0076] The memory 1102 may include a program storage area and a data storage area, wherein the program storage area may store an operating system, an application required for at least one function; the data storage area may store data created according to the use of the electronic device, etc. In addition, the memory 1102 may include a high-speed random access memory, and may also include a non-transient memory, such as at least one disk storage device, a flash memory device, or other non-transient solid-state storage device. In some embodiments, the memory 1102 may optionally include a memory remotely arranged relative to the processor 1101, and these remote memories may be connected to the electronic device via a network. Examples of the above-mentioned network include, but are not limited to, the Internet, an intranet, a local area network, a mobile communication network, and combinations thereof.

[0077] It should be noted that the computer-readable medium described in this application may be a computer-readable signal medium or a computer-readable storage medium or any combination of the above two. The computer-readable medium may be, for example, but not limited to, an electrical, magnetic, optical, electromagnetic, infrared or semiconductor system, device or device, or any combination of the above. More specific examples of computer-readable storage media may include, but are not limited to: an electrical connection with one or more wires, a portable computer disk, a hard disk, a random access memory (RAM), a read-only memory (ROM), an erasable programmable read-only memory (EPROM or flash memory), an optical fiber, a portable compact disk read-only memory (CD-ROM), an optical storage device, a magnetic storage device, or any suitable combination of the above. In this application, a computer-readable medium may be any tangible medium containing or storing a program that can be used by or in combination with an instruction execution system, device or device.

[0078] Computer readable media include permanent and non-permanent, removable and non-removable media, and can be implemented by any method or technology to store information. Information can be computer readable instructions, data structures, modules of programs or other data. Examples of computer storage media include, but are not limited to, phase change memory (PRAM), static random access memory (SRAM), dynamic random access memory (DRAM), other types of random access memory (RAM), read-only memory (ROM), electrically erasable programmable read-only memory (EEPROM), flash memory or other memory technology, read-only compact disk (CD-ROM), digital versatile disk (DVD) or other optical storage, magnetic cassettes, magnetic tape disk storage or other magnetic storage devices or any other non-transmission medium that can be used to store information that can be accessed by a computing device.

[0079] Computer program code for performing the operations of the present application may be written in one or more programming languages ​​or a combination thereof, including object-oriented programming languages, such as Java, Smalltalk, C++, and conventional procedural programming languages, such as "C" or similar programming languages. The program code may be executed entirely on the user's computer, partially on the user's computer, as a separate software package, partially on the user's computer and partially on a remote computer, or entirely on a remote computer or server. In the case of a remote computer, the remote computer may be connected to the user's computer through any type of network, including a local area network (LAN) or a wide area network (WAN), or may be connected to an external computer (e.g., via the Internet using an Internet service provider).

[0080] In the above embodiments, all or part of the embodiments may be implemented by software, hardware, firmware or any combination thereof. For example, an application specific integrated circuit (ASIC), a general-purpose computer or any other similar hardware device may be used to implement the embodiments. In some embodiments, the software program of the present application may be executed by a processor to implement the above steps or functions. Similarly, the software program of the present application (including related data structures) may be stored in a computer-readable recording medium, such as a RAM memory, a magnetic or optical drive or a floppy disk and the like. In addition, some steps or functions of the present application may be implemented by hardware, for example, as a circuit that cooperates with a processor to perform various steps or functions.

[0081] Fifth embodiment

[0082] The computer program product provided in the embodiment of the present application includes one or more computer programs / instructions, which, when executed by the processor, generate in whole or in part the process or function described in the embodiment of the present application. The computer may be a general-purpose computer, a special-purpose computer, a computer network, or other programmable device. The computer instructions may be stored in a computer-readable storage medium, or transmitted from one computer-readable storage medium to another computer-readable storage medium. For example, the computer instructions may be transmitted from a website site, a computer, a server, or a data center to another website site, a computer, a server, or a data center by wired (e.g., coaxial cable, optical fiber, digital subscriber line (DSL)) or wireless (e.g., infrared, wireless, microwave, etc.) means. The computer-readable storage medium may be any available medium that a computer can access or a data storage device such as a server or a data center that includes one or more available media integrated. The available medium may be a magnetic medium (e.g., a floppy disk, a hard disk, a tape), an optical medium (e.g., a DVD), or a semiconductor medium (e.g., a solid state disk (SSD)), etc.

[0083] The flow chart or block diagram in the accompanying drawings shows the possible architecture, function and operation of the equipment, method and computer program product according to various embodiments of the present application. In this regard, each box in the flow chart or block diagram can represent a module, a program segment or a part of a code, and the module, the program segment or a part of the code contains one or more executable instructions for realizing the specified logical function. It should also be noted that in some alternative implementations, the functions marked in the box can also occur in a different order from the order marked in the accompanying drawings. For example, two boxes represented in succession can actually be executed substantially in parallel, and they can sometimes be executed in the opposite order, depending on the functions involved. It should also be noted that each box in the block diagram and / or flow chart, and the combination of the boxes in the block diagram and / or flow chart can be implemented with a dedicated system for hardware that performs a specified function or operation, or can be implemented with a combination of dedicated hardware and computer instructions.

[0084] The scope of the present application is defined by the appended claims rather than the above description, and it is intended that all changes that fall within the meaning and scope of the equivalent elements of the claims are included in the present application. Any figure mark in the claims should not be regarded as limiting the claims involved. In addition, it is clear that the word "comprising" does not exclude other units or steps, and the singular does not exclude the plural. Multiple units or devices stated in the device claim may also be implemented by one unit or device through software or hardware. The words "first", "second", etc. are only used to distinguish the description, and do not indicate any particular order, nor can they be understood as indicating or implying relative importance.

[0085] The above is only a specific implementation of the present application, but the protection scope of the present application is not limited thereto. Any person skilled in the art can easily mention changes or substitutions within the technical scope disclosed in the present application, which should be included in the protection scope of the present application. Therefore, the protection scope of the present application should be based on the protection scope of the claims, and the above embodiments should be regarded as exemplary and non-restrictive.

Claims

1. A method for adjusting semiconductor process recipe parameters, characterized in that: include: A multivariate regression model is trained using a data set including one or more batches of semiconductor process recipe parameters and corresponding product yields; The method includes: recording the process recipe parameters used for each batch of wafers during the manufacturing process; performing sampling inspection on some wafers in each batch, wherein the inspection includes CD measurement; associating the inspection results with the corresponding process recipe parameters to obtain historical data, wherein the historical data includes the process recipe name, the product number of the wafer, the value of the process recipe parameter and / or the qualified status of the CD measurement of the wafer; constructing a data set based on the historical data to train a multivariate regression model; Substituting the fixed parameters in the recipe to be adjusted into the multivariate regression model to obtain the value or value range of the parameter to be adjusted in the recipe to be adjusted based on the target yield, including: Get parameter adjustment instructions; In response to the parameter adjustment instruction, the parameters in the recipe to be adjusted are divided to obtain fixed parameters and parameters to be adjusted; Substituting the numerical values ​​of the fixed parameters as independent variables into the multivariate regression model to obtain a univariate regression model; Substituting the set target yield rate as the dependent variable into the univariate regression model to solve the independent variable of the parameter to be adjusted; Obtaining a value or a value range of a parameter to be adjusted based on the solution result; presenting a graph of the univariate regression model; A table of values ​​for the graph is presented based on a set step size.

2. The adjustment method according to claim 1, characterized in that: in, Build a dataset based on historical data to train a multivariate regression model, including: Calculate the product yield corresponding to the process recipe parameters based on historical data to obtain a data set including the process recipe parameters and the yield; Determine the number of independent variables based on the process recipe parameters in the data set to build a multivariate regression model; The numerical values ​​of the process recipe parameters are used as independent variables, and the corresponding yield is used as the dependent variable to perform fitting training on the multivariate regression model.

3. The adjustment method according to claim 1, characterized in that: The multivariate regression model comprises a multivariate quadratic regression model.

4. A semiconductor process recipe parameter adjustment system using the adjustment method according to any one of claims 1 to 3, characterized in that: include: parameter unit, measurement unit and calculation unit; among which, The parameter unit is used to obtain process recipe parameters; The measuring unit is used to obtain CD measurement results; The calculation unit is used to train a multivariate regression model using a data set containing one or more batches of semiconductor process recipe parameters and corresponding product yields; substitute the fixed parameters in the recipe to be adjusted into the multivariate regression model to obtain the value or value range of the parameter to be adjusted in the recipe to be adjusted based on the target yield.

5. An electronic device, characterized in that: The electronic device comprises: one or more processors; and A memory storing computer program instructions, which, when executed, cause the processor to perform the steps of the method as claimed in any one of claims 1 to 3.

6. A computer readable medium having a computer program / instructions stored thereon, characterized in that: When the computer program / instructions are executed by a processor, the steps of the method according to any one of claims 1 to 3 are implemented.

7. A computer program product comprising a computer program / instructions, characterized in that When the computer program / instructions are executed by a processor, the steps of the method according to any one of claims 1 to 3 are implemented.

Images