Etch endpoint monitoring method and apparatus

Abstract

The embodiment of the application provides an etching endpoint monitoring method and device, the method comprises the following steps: obtaining the spectrum data collected by a spectrometer in an etching process, and performing filtering processing on the spectrum data; performing data dimension reduction on the filtered spectrum data to obtain at least one characteristic wave band of the spectrum data; performing numerical analysis on the at least one characteristic wave band of the spectrum data to determine a data model of the spectrum data; monitoring the etching state of the real-time collected spectrum data through the data model of the spectrum data, and controlling the etching process according to the monitoring result; the application can effectively reduce the data dimension in the etching process, reduce data redundancy, reduce the misjudgment rate of the interference signal on the plasma etching result, greatly improve the robustness of the plasma etching process state monitoring model, and realize the automatic management of the etching monitoring process, and improve the etching precision.

Description

Technical Field

[0001] This application relates to the field of etching processes, specifically to a method and apparatus for monitoring the etching endpoint. Background Technology

[0002] Currently, inspection technology for single-atom-level processes in large-scale integrated circuits is a core element of integrated circuit manufacturing processes. Miniaturization and three-dimensional structure are the main characteristics of manufacturing process evolution. Correspondingly, the continuous shrinking of feature sizes places increasingly higher demands on the precision of process equipment, especially the etching process control, which needs to reach the atomic level. Meanwhile, the three-dimensional structure leads to a gradual increase in the complexity of integrated circuit architecture, and the proportion of high-precision etching processes is increasing, thus making the requirements for uniformity, low damage, and yield of each process step increasingly stringent.

[0003] Currently, OES (Optical Emission Spectroscopy) is one of the most commonly used non-invasive plasma diagnostic techniques, which can help identify chemical substances and monitor changes in their ratio to the overall plasma in the chamber. However, since a chemical element may emit light in more than one way, OES for etching state monitoring faces problems such as high data dimensionality and redundancy. In high-dimensional space, useful information is very sparse, which weakens the statistical significance of spectral data.

[0004] Furthermore, current etching process control is usually based on manual operation, which has a negative impact on etching efficiency and accuracy. Summary of the Invention

[0005] To address the problems in the prior art, this application provides an etching endpoint monitoring method and apparatus, which can effectively reduce the data dimensionality in the etching process, reduce data redundancy, reduce the misjudgment rate of interference signals on plasma etching results, significantly improve the robustness of the plasma etching process status monitoring model, and realize automated management of the etching monitoring process to improve etching accuracy.

[0006] To solve at least one of the above problems, this application provides the following technical solution:

[0007] According to a first aspect of the embodiments of this application, this application provides an etching endpoint monitoring method, including:

[0008] Acquire spectral data collected by the spectrometer during the etching process, and filter the spectral data;

[0009] The filtered spectral data is subjected to dimensionality reduction to obtain at least one characteristic band of the spectral data.

[0010] Numerical analysis is performed on at least one characteristic band of the spectral data to determine the data model of the spectral data;

[0011] The etching status is monitored by using the data model of the spectral data in real time, and the etching process is controlled based on the monitoring results.

[0012] According to any embodiment of this application, the filtering process for the spectral data includes:

[0013] The spectral data is decomposed using wavelet decomposition to obtain wavelet decomposition coefficients and high-frequency coefficients of each layer of the spectral data.

[0014] Based on the preset coefficient threshold, the wavelet decomposition coefficients, and the high-frequency coefficients of each layer of the spectral data, the spectral data is subjected to inverse transformation to obtain the denoised spectral data.

[0015] The denoised spectral data is sorted based on the signal sequence within a window of a preset width, and the signal located at the median position is determined as the filtering result of the spectral data.

[0016] According to any embodiment of this application, after filtering the spectral data, the method further includes:

[0017] The spectral data is standardized based on the mean and standard deviation of the individual spectral data.

[0018] According to any embodiment of this application, the step of performing data dimensionality reduction on the filtered spectral data to obtain at least one characteristic band of the spectral data includes:

[0019] The redundant wavelengths in the spectral data are filtered based on the preset absolute peak value to obtain the filtered spectral data.

[0020] Determine the average contribution ratio of the filtered spectral data, and determine the optimal peak value of the spectral data based on the average contribution ratio;

[0021] The band corresponding to the optimal peak value is determined as at least one characteristic band of the spectral data.

[0022] According to any embodiment of this application, the step of monitoring the etching status of the real-time acquired spectral data through the data model of the spectral data, and controlling the etching process based on the monitoring results, includes:

[0023] The spectral data collected in real time by the spectrometer is input into the data model of the spectral data to obtain the upper and lower limit ranges of the current spectrum;

[0024] In response to the fact that the spectral intensity of the current spectrum is within the constraint range of the upper and lower limits, the current etching process is maintained;

[0025] In response to the current spectral intensity exceeding the constraint range of the upper and lower limits, the current etching process is terminated.

[0026] According to a second aspect of the embodiments of this application, this application provides an etching endpoint monitoring device, comprising:

[0027] The data acquisition module is used to: acquire spectral data collected by the spectrometer during the etching process, and to filter the spectral data;

[0028] The data dimensionality reduction module is used to: perform data dimensionality reduction on the filtered spectral data to obtain at least one characteristic band of the spectral data;

[0029] The data mining module is used to: perform numerical analysis on at least one characteristic band of the spectral data to determine the data model of the spectral data;

[0030] The status monitoring module is used to: monitor the etching status of the real-time acquired spectral data through the data model of the spectral data, and control the etching process based on the monitoring results.

[0031] According to any embodiment of this application, when the data acquisition module performs filtering processing on the spectral data, it is specifically used for:

[0032] The spectral data is decomposed using wavelet decomposition to obtain wavelet decomposition coefficients and high-frequency coefficients of each layer of the spectral data.

[0033] Based on the preset coefficient threshold, the wavelet decomposition coefficients, and the high-frequency coefficients of each layer of the spectral data, the spectral data is subjected to inverse transformation to obtain the denoised spectral data.

[0034] The denoised spectral data is sorted based on the signal sequence within a window of a preset width, and the signal located at the median position is determined as the filtering result of the spectral data.

[0035] According to any embodiment of this application, after filtering the spectral data, a normalization processing module is further included, used for:

[0036] The spectral data is standardized based on the mean and standard deviation of the individual spectral data.

[0037] According to any embodiment of this application, when the data dimensionality reduction module performs data dimensionality reduction on the filtered spectral data to obtain at least one characteristic band of the spectral data, it is specifically used for:

[0038] The redundant wavelengths in the spectral data are filtered based on the preset absolute peak value to obtain the filtered spectral data.

[0039] Determine the average contribution ratio of the filtered spectral data, and determine the optimal peak value of the spectral data based on the average contribution ratio;

[0040] The band corresponding to the optimal peak value is determined as at least one characteristic band of the spectral data.

[0041] According to any embodiment of this application, when the state monitoring module monitors the etching state of the real-time acquired spectral data through the data model of the spectral data and controls the etching process based on the monitoring results, it is specifically used for:

[0042] The spectral data collected in real time by the spectrometer is input into the data model of the spectral data to obtain the upper and lower limit ranges of the current spectrum;

[0043] In response to the fact that the spectral intensity of the current spectrum is within the constraint range of the upper and lower limits, the current etching process is maintained;

[0044] In response to the current spectral intensity exceeding the constraint range of the upper and lower limits, the current etching process is terminated.

[0045] According to a third aspect of the embodiments of this application, this application provides an electronic device, including a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the processor executes the program to implement the steps of the etching endpoint monitoring method.

[0046] According to a fourth aspect of the embodiments of this application, this application provides a computer-readable storage medium having a computer program stored thereon, which, when executed by a processor, implements the steps of the etching endpoint monitoring method described above.

[0047] According to a fifth aspect of the embodiments of this application, this application provides a computer program product, including a computer program / instructions that, when executed by a processor, implement the steps of the etching endpoint monitoring method.

[0048] As can be seen from the above technical solution, this application provides an etching endpoint monitoring method and apparatus. It acquires spectral data collected by a spectrometer during the etching process and filters the spectral data; performs data dimensionality reduction on the filtered spectral data to obtain at least one characteristic band of the spectral data; performs numerical analysis on the at least one characteristic band of the spectral data to determine the data model of the spectral data; and monitors the etching status of the real-time acquired spectral data using the data model of the spectral data, and controls the etching process based on the monitoring results. This effectively reduces the data dimensionality during the etching process, reduces data redundancy, lowers the misjudgment rate of interference signals on plasma etching results, significantly improves the robustness of the plasma etching process status monitoring model, and simultaneously achieves automated management of the etching monitoring process, improving etching accuracy. Attached Figure Description

[0049] To more clearly illustrate the technical solutions in the embodiments of this application or the prior art, the drawings used in the description of the embodiments or the prior art will be briefly introduced below. Obviously, the drawings described below are some embodiments of this application. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0050] Figure 1 This is one of the flowcharts illustrating the etching endpoint monitoring method in the embodiments of this application;

[0051] Figure 2 This is a second schematic flowchart of the etching endpoint monitoring method in the embodiments of this application;

[0052] Figure 3 The original spectral data Mg_285nm of the etching endpoint monitoring method in the embodiments of this application;

[0053] Figure 4 The image shows the spectral intensity variation curve of Mg element after wavelet-median processing of the etching endpoint monitoring method in this application embodiment;

[0054] Figure 5 The original spectral data Co_345nm of the etching endpoint monitoring method in the embodiments of this application;

[0055] Figure 6 The image shows the Co element spectral intensity change curve after wavelet-median processing of the etching endpoint monitoring method in this application embodiment;

[0056] Figure 7 This is the third flowchart illustrating the etching endpoint monitoring method in the embodiments of this application;

[0057] Figure 8These are partial original spectral change curves of the etching endpoint monitoring method in the embodiments of this application;

[0058] Figure 9 The spectral data are the dimensionality-reduced values ​​of the etching endpoint monitoring method in the embodiments of this application.

[0059] Figure 10 This is the fourth flowchart illustrating the etching endpoint monitoring method in the embodiments of this application;

[0060] Figure 11 The image of the Co_345nm fitting function obtained by cubic spline interpolation for the etching endpoint monitoring method in the embodiments of this application;

[0061] Figure 12 The image of the Mg_285nm fitting function obtained by cubic spline interpolation in the etching endpoint monitoring method of this application embodiment;

[0062] Figure 13 This is an image of the etching endpoint detection method in the embodiments of this application;

[0063] Figure 14 This is the encapsulated monitoring system interface for the etching endpoint monitoring method in the embodiments of this application;

[0064] Figure 15 This is the fifth flowchart illustrating the etching endpoint monitoring method in the embodiments of this application;

[0065] Figure 16 This is one of the structural diagrams of the etching endpoint monitoring device in the embodiments of this application;

[0066] Figure 17 This is a schematic diagram of the structure of the electronic device in the embodiments of this application. Detailed Implementation

[0067] To make the objectives, technical solutions, and advantages of the embodiments of this application clearer, the technical solutions of the embodiments of this application will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of this application, not all embodiments. Based on the embodiments of this application, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of this application.

[0068] The acquisition, storage, use, and processing of data in this application all comply with the relevant provisions of national laws and regulations.

[0069] Considering the problems of high data dimensionality and large redundancy in etching status monitoring via OES, this application provides an etching endpoint monitoring method and device, which effectively reduces the data dimensionality in the etching process, reduces data redundancy, reduces the misjudgment rate of interference signals on plasma etching results, greatly improves the robustness of the plasma etching process status monitoring model, and realizes automated management of the etching monitoring process, thereby improving etching accuracy.

[0070] To effectively reduce data dimensionality and redundancy during the etching process, and to automate the management of etching monitoring to improve etching accuracy, this application provides an embodiment of an etching endpoint monitoring method. See [link to embodiment]. Figure 1 The etching endpoint monitoring method specifically includes the following:

[0071] Step S101: Acquire the spectral data collected by the spectrometer during the etching process, and perform filtering processing on the spectral data.

[0072] This application utilizes a spectrometer as the device, employing software to acquire and receive data from the spectrometer in real time. This data serves as a human-machine communication interface for subsequent model building and real-time status monitoring, allowing the acquisition of spectral information from the plasma etching chamber into the computer. This spectral information can include emission spectral data of the plasma within a specific wavelength range. The acquired data is raw spectral data without any processing, i.e., the initial data directly obtained from the spectrometer.

[0073] Next, the acquired spectral data is filtered. Because the etching process may be affected by the chamber environment or other external factors, interference information may exist in the spectral data. Filtering, through specific processing methods, can remove or reduce noise, interference, and redundant information in the data, thereby improving data quality and accuracy.

[0074] The filtering process can employ various methods and algorithms, such as wavelet transform, sliding window, and median filtering. Wavelet transform can decompose spectral data into sub-signals of different frequency ranges, and high-frequency noise can be removed through thresholding. Median filtering utilizes the median value of the data within a window as the filtered result, and is suitable for removing impulse noise and other outliers. By selecting appropriate filtering methods and parameters, noise in the data can be effectively reduced, and relevant information related to the etching process can be extracted.

[0075] Step S102: Perform dimensionality reduction on the filtered spectral data to obtain at least one characteristic band of the spectral data.

[0076] The filtered spectral data will then undergo dimensionality reduction. This dimensionality reduction aims to reduce the dimensionality of the dataset while retaining the most representative and crucial information. Dimensionality reduction reduces data redundancy, improves computational efficiency, and reduces dataset storage requirements while preserving data characteristics.

[0077] The purpose of dimensionality reduction on the filtered spectral data is to extract at least one characteristic band. This characteristic band represents a representative spectral wavelength or frequency band selected during the dimensionality reduction process, and is used for subsequent analysis and modeling. Dimensionality reduction removes a large amount of redundant information from the original spectral data, retaining important features related to the etching process state.

[0078] Step S103: Perform numerical analysis on at least one characteristic band of the spectral data to determine the data model of the spectral data.

[0079] By performing numerical analysis on at least one characteristic band of the spectral data, a data model can be established to describe and represent the relationship between the characteristics of the spectral data and the etching process state. This data model can be a mathematical model or a statistical model. Through numerical analysis and modeling of the characteristic bands, the patterns and potential etching state information in the spectral data are revealed.

[0080] The established data model can be selected and optimized according to actual needs to improve its accuracy and robustness. This model can be a linear model, a nonlinear model, or a supervised or unsupervised learning model. By performing numerical analysis on characteristic bands, the model can learn and capture key features in the spectral data and correlate them with the etching process state.

[0081] For example, cubic spline interpolation can be used to model the characteristic bands obtained from the dimensionality-reduced spectral data, constructing an interpolation function for each wavelength with respect to the running time.

[0082] Step S104: Monitor the etching status of the real-time acquired spectral data using the data model of the spectral data, and control the etching process based on the monitoring results.

[0083] By using the established data model of spectral data, the etching status can be monitored based on the real-time acquired spectral data. That is, the real-time acquired spectral data is input into the data model, which then analyzes and calculates the data to determine the current etching status.

[0084] The data model can perform calculations and predictions based on the characteristic band values ​​of spectral data and existing training data. By comparing these values ​​with known etching state labels, the model can determine the current state of the etching process, such as whether the etching is nearing or has reached its endpoint. The judgment result can be determined based on pre-set thresholds or rules.

[0085] The system can control the etching process by having the hardware be driven by software to end the current etching process in real time, thereby avoiding over-etching or under-etching, effectively improving the accuracy and stability of the etching process, reducing errors and variations in the etching process, and further optimizing the efficiency and product quality of integrated circuit manufacturing.

[0086] As can be seen from the above description, the etching endpoint monitoring method provided in this application can effectively reduce the data dimension in the etching process, reduce data redundancy, reduce the misjudgment rate of interference signals on plasma etching results, greatly improve the robustness of the plasma etching process status monitoring model, and realize the automated management of the etching monitoring process to improve etching accuracy.

[0087] To remove as much Gaussian white noise and impulse noise as possible from the raw data during plasma etching, this application preferably employs a combination of wavelet filtering followed by median filtering. In one embodiment of the etching endpoint monitoring method of this application, see... Figure 2 The filtering process for the spectral data may further include the following:

[0088] Step S101A: Perform wavelet decomposition on the spectral data to obtain wavelet decomposition coefficients and high-frequency coefficients of each layer of the spectral data;

[0089] First, wavelet decomposition is performed on the spectral data. Wavelet decomposition is a signal decomposition technique that breaks down a signal into sub-signals of different frequency ranges, yielding wavelet decomposition coefficients and high-frequency coefficients for each layer of the spectral data. These coefficients contain information at different frequencies, with the high-frequency coefficients primarily containing details and noise information from the signal.

[0090] Specifically, wavelet decomposition can be performed on the spectral data to extract high-frequency coefficients. Wavelet coefficients greater than a threshold are considered valid data, while those less than the threshold are considered noise. Finally, the obtained low-frequency and high-frequency coefficients are recombined to obtain denoised data.

[0091] Step S101B: Perform inverse transform processing on the spectral data based on the preset coefficient threshold, the wavelet decomposition coefficients, and the high-frequency coefficients of each layer of the spectral data to obtain the denoised spectral data;

[0092] Inverse transformation processing is performed on the spectral data based on the preset coefficient threshold, wavelet decomposition coefficients and high-frequency coefficients of each layer of the spectral data. An appropriate threshold r is selected to perform threshold processing on the high-frequency coefficients obtained from wavelet decomposition. Wavelet inverse transformation is performed through the wavelet low-frequency coefficients and the high-frequency coefficients of each layer, and finally a signal estimation value, that is, the signal after denoising is obtained, which can effectively remove noise components in the spectral data and improve the quality and accuracy of the data.

[0093] Step S101C: sorting the denoised spectral data based on the signal sequence within a window with a preset width, and determining the signal at the median position as the filtering result of the spectral data.

[0094] Next, within the window of preset width, the denoised spectral data is sorted based on the signal sequence. According to the position of the signal within the sorted window, the signal at the median position is determined as the filtering result. The median is the intermediate value in the signal sequence sorted from small to large, which can effectively filter out outliers and noise, and extract effective information of the spectral data.

[0095] Specifically, when the signal sequence X_j(-∞<j<∞) after wavelet processing is subjected to filtering processing again, a window with an odd length L is firstly defined, L=2n+1, where n is a positive integer. Assuming that at a certain moment, the signal samples in the window are X(i-n), …, X(i), …, X(i+n), where X(i) is the signal sample value at the center of the window. After sorting the L signal sample values in ascending order, the sample value at position i is taken as the output value of median filtering: y(i)=Med[X(i-n), …, X(i), …, X(i+n)]. The changes of spectral intensity curves of Mg and C elements before and after filtering are shown in Figure 3 , Figure 4 , Figure 5 , Figure 6 . It can be seen from Figure 3 , Figure 4 comparison that before processing, there are many data burrs which are not conducive to subsequent data analysis and judgment, and the change curve after wavelet-median processing is smoother.

[0096] Through the above embodiments, the present application realizes filtering processing of spectral data by performing wavelet decomposition, inverse transformation processing, sorting and median determination on spectral data, which can effectively remove noise and extract effective information of spectral data, providing a more accurate data basis for subsequent data analysis and modeling.

[0097] In an optional embodiment, after filtering the spectral data, the method further comprises:

[0098] standardizing the spectral data based on the mean and standard deviation of the single-column spectral data.

[0099] Specifically, after filtering the spectral data, the process also includes a standardization step. Standardization is a data preprocessing method used to transform data from different scales into data with the same scale, facilitating better comparison and analysis.

[0100] In this embodiment, the spectral data can be standardized based on the mean and standard deviation of a single spectral data set. The purpose of standardization is to transform the data into a Gaussian distribution, resulting in a mean of 0 and a standard deviation of 1. This eliminates scale differences between different spectral data sets, improving data comparability and consistency.

[0101] The specific steps of standardization are as follows:

[0102] First, the average value of each wavelength is calculated for the filtered spectral data column by column to obtain the mean value of the spectral data for a single column.

[0103] Then, the standard deviation of each wavelength is calculated column by column for the filtered spectral data to obtain the standard deviation of a single column of spectral data.

[0104] The standardized formula can be expressed as:

[0105] X-μ / σ (1)

[0106] Where μ is the mean of the single-column spectral data and σ is the standard deviation of the single-column spectral data. After standardization, the dataset is transformed from X(1), X(2), ..., X(n) into X^'(1), X^'(2), ..., X^'(n), which satisfies a Gaussian distribution.

[0107] By standardizing the filtered spectral data based on the mean and standard deviation of the individual spectral data, the scale differences between spectral data can be eliminated, the comparability and consistency of the data can be improved, and a more accurate basis can be provided for subsequent data analysis and modeling.

[0108] To help extract key information from spectral data, reduce data dimensionality, and provide a more accurate and efficient data foundation for subsequent data analysis and modeling, in one embodiment of the etching endpoint monitoring method of this application, see [reference needed]. Figure 7 The step of performing dimensionality reduction on the filtered spectral data to obtain at least one characteristic band of the spectral data may further include the following:

[0109] Step S102A: Filter redundant wavelengths in the spectral data based on preset absolute peak values ​​to obtain filtered spectral data.

[0110] The absolute peak characterization selects wavelength variables with significant information based on peak wavelength characteristics. These wavelength variables are used for subsequent data analysis and modeling, while redundant wavelengths have little informational value. The filtered spectral data is processed according to a preset absolute peak filtering method to obtain filtered spectral data. This data contains the selected wavelength variables with high information content, reducing redundant information in the data.

[0111] Step S102B: Determine the average contribution ratio of the filtered spectral data, and determine the optimal peak value of the spectral data based on the average contribution ratio.

[0112] The filtered spectral data is further analyzed to calculate the average contribution ratio for each wavelength. This average contribution ratio is a statistic used to quantify the degree of information loss in the dimensionality-reduced data. By calculating the average contribution ratio for each wavelength, it is possible to determine how many wavelength variables to retain during dimensionality reduction, thus balancing the relationship between data dimensionality and information loss.

[0113] Based on the calculated average contribution ratio, the optimal peak value of the spectral data is determined. The optimal peak value is a wavelength variable with a high contribution ratio, plays an important role in the spectral data, and can be used as a characteristic band for subsequent data analysis and modeling.

[0114] Step S102C: Determine the band corresponding to the optimal peak as at least one characteristic band of the spectral data.

[0115] Based on the band corresponding to the optimal peak, at least one characteristic band of the spectral data is determined. Characteristic bands are wavelength variables with a high contribution ratio selected during data dimensionality reduction; they contain important information and can represent the characteristics of the original spectral data.

[0116] To further illustrate this solution, this application also provides a specific application example, which includes the following:

[0117] The obtained spectral data is further dimensionality reduced to filter out effective feature wavelength data suitable for model construction:

[0118] First, select the absolute peak value, that is, filter out redundant wavelength variables based on the peak wavelength characteristics.

[0119] Secondly, iterative sorting is performed, which further reduces the number of wavelengths in the data based on the linear correlation between variables. During this process, a new statistic, the average contribution ratio, is constructed to quantify the degree of information loss after dimensionality reduction.

[0120] Finally, the optimal peak value is selected, identifying the peak wavelength variable in the input absorption spectrum dataset. During any N processing runs, at a certain time point T... j The output is a set of wavelength indices, which represent the wavelengths at time point T. j The peak wavelength intensity at that time.

[0121] Figure 8 , Figure 9 The figures show the spectral diagrams before and after dimensionality reduction. As shown, while significantly reducing the data dimensionality, the key core information in the process engineering was also preserved.

[0122] The formula for calculating the mean contribution ratio (MCR) of a statistic is shown below:

[0123]

[0124] Among them, R r,k The sorting coefficients are obtained by using a subset of wavelengths generated from the absolute peak selection and iterative sorting process to represent each column of original spectral data.

[0125] The absolute peak value is defined as SBW, and the calculation formula is shown below:

[0126] SBW = {W i-1 +F <W i >W i+1 +F} (3)

[0127] Where F is the interference factor of the spectrometer itself, and W i This represents the actual spectral data.

[0128] The sorting coefficient is defined as R:

[0129]

[0130] Among them, Y i Indicates the actual value. Indicates the predicted value. This represents the mean.

[0131] This application performs dimensionality reduction on filtered spectral data by filtering based on preset absolute peak values, calculating the average contribution ratio, and determining the optimal peak value. This results in at least one characteristic band of the spectral data, which helps extract key information from the spectral data, reduces the dimensionality of the data, and provides a more accurate and efficient data foundation for subsequent data analysis and modeling.

[0132] In order to promptly detect abnormalities during the etching process and take corresponding control measures to ensure the stability of the etching process and good etching results, in one embodiment of the etching endpoint monitoring method of this application, see [reference needed]. Figure 10The step of monitoring the etching status of the real-time acquired spectral data using the data model of the spectral data, and controlling the etching process based on the monitoring results, may specifically include the following:

[0133] Step S104A: Input the spectral data collected in real time by the spectrometer into the data model of the spectral data to obtain the upper and lower limit ranges of the current spectrum.

[0134] The real-time acquired spectral data is input into a pre-built data model. Through data model analysis, the upper and lower limits of the current spectrum can be obtained, which reflects the constraint range of spectral intensity under normal etching conditions.

[0135] During a stable etching process, the system reads data from the spectrometer in real time as input to the model, using a pre-constructed function model. After the model calculation is complete (ms-level response), the upper and lower limits [lower(t), upper(t)] of the spectral intensity of the corresponding spectral band at the current running time t are obtained, such as... Figure 11 , Figure 12 As shown, upper and lower bound functions for Mg_285nm and Co_345nm were constructed.

[0136] Step S104B: In response to the current spectrum intensity being within the constraint range of the upper and lower limits, maintain the current etching process.

[0137] The system compares the real-time acquired spectral data with the upper and lower limit intervals to determine whether the current spectral intensity is within the constraint range of the upper and lower limit intervals. If the spectral intensity is within the constraint range, it indicates that the etching process is proceeding normally, and the system will maintain the current etching process.

[0138] Step S104C: In response to the current spectrum intensity exceeding the constraint range of the upper and lower limit intervals, terminate the current etching process.

[0139] In response to the current spectrum exceeding the constraint range: If the spectral intensity of the current spectrum exceeds the upper and lower limit ranges, it indicates an abnormality in the etching process. In this case, the system will terminate the current etching process and take appropriate control measures to avoid adverse etching results or other problems.

[0140] Figure 13 The diagram shows that during one etching process, the spectral intensity of Mg decreases while that of Co increases, indicating that Mg is nearing complete etching, and the underlying Co layer is gradually exposed. The intersection of the upper and lower bounds of the Mg and Co spectral intensities marks the endpoint of this etching operation. The packaged system is shown below. Figure 14 As shown.

[0141] This application can monitor the etching status of the acquired spectral data in real time and control the etching process based on the monitoring results. Through data model analysis and judgment, abnormalities in the etching process can be detected in a timely manner, and corresponding control measures can be taken to ensure the stability of the etching process and good etching results.

[0142] Figure 15 A system flowchart illustrating this application is shown, such as Figure 15 As shown, in one embodiment, this application may include the following six basic steps:

[0143] 1. Data Acquisition: The spectral information of the plasma etching chamber is acquired by a spectrometer and transferred to a computer to obtain the raw spectral data without any processing.

[0144] 2. Data selection: The obtained spectral data is filtered according to the etching process characteristics;

[0145] 3. Preprocessing: Standardize the processed data;

[0146] 4. Data Transformation: Through dimensionality reduction methods, characteristic bands that can be used for subsequent analysis are obtained from highly redundant spectral data;

[0147] 5. Data Mining: Based on the obtained characteristic values ​​of the bands, numerical analysis methods are used to model each characteristic band;

[0148] 6. Monitoring Focus / Faults: For the acquired spectral data, the etching status is monitored through the pre-built model. If the endpoint signal is detected, the software can drive the hardware in real time to end the current process, thereby realizing real-time monitoring and control of the etching endpoint signal.

[0149] To effectively reduce data dimensionality and redundancy during the etching process, and to automate the etching monitoring process and improve etching accuracy, this application provides an embodiment of an etching endpoint monitoring device for implementing all or part of the etching endpoint monitoring method. See [link to embodiment]. Figure 16 The etching endpoint monitoring device specifically includes the following components:

[0150] The data acquisition module 1101 is used to: acquire spectral data collected by the spectrometer during the etching process, and perform filtering processing on the spectral data;

[0151] Data dimensionality reduction module 1102 is used to: perform data dimensionality reduction on the filtered spectral data to obtain at least one characteristic band of the spectral data;

[0152] Data mining module 1103 is used to: perform numerical analysis on at least one characteristic band of the spectral data to determine the data model of the spectral data;

[0153] The status monitoring module 1104 is used to: monitor the etching status of the real-time acquired spectral data through the data model of the spectral data, and control the etching process according to the monitoring results.

[0154] According to any embodiment of this application, when the data acquisition module performs filtering processing on the spectral data, it is specifically used for:

[0155] The spectral data is decomposed using wavelet decomposition to obtain wavelet decomposition coefficients and high-frequency coefficients of each layer of the spectral data.

[0156] Based on the preset coefficient threshold, the wavelet decomposition coefficients, and the high-frequency coefficients of each layer of the spectral data, the spectral data is subjected to inverse transformation to obtain the denoised spectral data.

[0157] The denoised spectral data is sorted based on the signal sequence within a window of a preset width, and the signal located at the median position is determined as the filtering result of the spectral data.

[0158] According to any embodiment of this application, after filtering the spectral data, a normalization processing module is further included, used for:

[0159] The spectral data is standardized based on the mean and standard deviation of the individual spectral data.

[0160] According to any embodiment of this application, when the data dimensionality reduction module performs data dimensionality reduction on the filtered spectral data to obtain at least one characteristic band of the spectral data, it is specifically used for:

[0161] The redundant wavelengths in the spectral data are filtered based on the preset absolute peak value to obtain the filtered spectral data.

[0162] Determine the average contribution ratio of the filtered spectral data, and determine the optimal peak value of the spectral data based on the average contribution ratio;

[0163] The band corresponding to the optimal peak value is determined as at least one characteristic band of the spectral data.

[0164] According to any embodiment of this application, when the state monitoring module monitors the etching state of the real-time acquired spectral data through the data model of the spectral data and controls the etching process based on the monitoring results, it is specifically used for:

[0165] The spectral data collected in real time by the spectrometer is input into the data model of the spectral data to obtain the upper and lower limit ranges of the current spectrum;

[0166] In response to the fact that the spectral intensity of the current spectrum is within the constraint range of the upper and lower limits, the current etching process is maintained;

[0167] In response to the current spectral intensity exceeding the constraint range of the upper and lower limits, the current etching process is terminated.

[0168] As can be seen from the above description, the etching endpoint monitoring device provided in this application embodiment can effectively reduce the data dimension in the etching process, reduce data redundancy, reduce the misjudgment rate of interference signals on plasma etching results, greatly improve the robustness of the plasma etching process status monitoring model, and realize the automated management of the etching monitoring process to improve etching accuracy.

[0169] From a hardware perspective, in order to effectively reduce the data dimensionality during the etching process, reduce data redundancy, and achieve automated management of the etching monitoring process to improve etching accuracy, this application provides an embodiment of an electronic device for implementing all or part of the etching endpoint monitoring method. The electronic device specifically includes the following components:

[0170] The system comprises a processor, memory, a communications interface, and a bus; wherein the processor, memory, and communications interface communicate with each other via the bus; the communications interface is used to realize information transmission between the etching endpoint monitoring device and core business systems, user terminals, and related databases and other related devices; the logic controller can be a desktop computer, tablet computer, or mobile terminal, etc., and this embodiment is not limited to these. In this embodiment, the logic controller can be implemented with reference to the embodiments of the etching endpoint monitoring method and the etching endpoint monitoring device in the embodiments, the content of which is incorporated herein, and repeated details will not be described again.

[0171] It is understood that the user terminal may include smartphones, tablet computers, network set-top boxes, portable computers, desktop computers, personal digital assistants (PDAs), in-vehicle devices, smart wearable devices, etc. Among these, the smart wearable devices may include smart glasses, smartwatches, smart bracelets, etc.

[0172] In practical applications, the etching endpoint monitoring method can be partially executed on the electronic device side as described above, or all operations can be completed in the client device. The choice can be made based on the processing power of the client device and the limitations of the user's usage scenario. This application does not impose any limitations on this. If all operations are completed in the client device, the client device may further include a processor.

[0173] The aforementioned client device may have a communication module (i.e., a communication unit) that can communicate with a remote server to achieve data transmission. The server may include a server on the task scheduling center side; in other implementation scenarios, it may also include a server on an intermediate platform, such as a server on a third-party server platform that has a communication link with the task scheduling center server. The server may include a single computer device, a server cluster consisting of multiple servers, or a distributed server structure.

[0174] Figure 17 This is a schematic block diagram illustrating the system configuration of the electronic device 9600 according to an embodiment of this application. Figure 17 As shown, the electronic device 9600 may include a central processing unit 9100 and a memory 9140; the memory 9140 is coupled to the central processing unit 9100. It is worth noting that... Figure 17 This is an example; other types of structures can also be used to supplement or replace this structure to achieve telecommunications functions or other functions.

[0175] In one embodiment, the etching endpoint monitoring method function can be integrated into the central processing unit 9100. The central processing unit 9100 can be configured to perform the following control:

[0176] Step S101: Acquire the spectral data collected by the spectrometer during the etching process, and perform filtering processing on the spectral data.

[0177] Step S102: Perform dimensionality reduction on the filtered spectral data to obtain at least one characteristic band of the spectral data.

[0178] Step S103: Perform numerical analysis on at least one characteristic band of the spectral data to determine the data model of the spectral data.

[0179] Step S104: Monitor the etching status of the real-time acquired spectral data using the data model of the spectral data, and control the etching process based on the monitoring results.

[0180] As can be seen from the above description, the electronic device provided in this application embodiment effectively reduces the data dimension in the etching process, reduces data redundancy, reduces the misjudgment rate of interference signals on plasma etching results, greatly improves the robustness of the plasma etching process status monitoring model, and realizes automated management of the etching monitoring process, thereby improving etching accuracy.

[0181] In another embodiment, the etching endpoint monitoring device can be configured separately from the central processing unit 9100. For example, the etching endpoint monitoring device can be configured as a chip connected to the central processing unit 9100, and the etching endpoint monitoring method function can be implemented through the control of the central processing unit.

[0182] like Figure 17 As shown, the electronic device 9600 may further include: a communication module 9110, an input unit 9120, an audio processor 9130, a display 9160, and a power supply 9170. It is worth noting that the electronic device 9600 does not necessarily need to include these components. Figure 17 All components shown; in addition, the electronic device 9600 may also include Figure 17 For components not shown, please refer to existing technologies.

[0183] like Figure 17 As shown, the central processing unit 9100, sometimes also referred to as a controller or operating control, may include a microprocessor or other processor device and / or logic device, which receives inputs and controls the operation of various components of the electronic device 9600.

[0184] The memory 9140 may be, for example, one or more of a cache, flash memory, hard drive, removable media, volatile memory, non-volatile memory, or other suitable devices. It may store the aforementioned failure-related information, and also store a program for executing that information. The central processing unit 9100 may execute the program stored in the memory 9140 to perform information storage or processing, etc.

[0185] Input unit 9120 provides input to central processing unit 9100. Input unit 9120 may be, for example, a keypad or touch input device. Power supply 9170 provides power to electronic device 9600. Display 9160 displays images and text. Display may be, for example, an LCD display, but is not limited thereto.

[0186] The memory 9140 can be a solid-state memory, such as a read-only memory (ROM), random access memory (RAM), a SIM card, etc. It can also be a memory that retains information even when power is off, can be selectively erased, and contains more data; examples of this type of memory are sometimes referred to as EPROMs. The memory 9140 can also be some other type of device. The memory 9140 includes a buffer memory 9141 (sometimes referred to as a buffer). The memory 9140 may include an application / function storage unit 9142 for storing application programs and function programs or processes for executing the operation of the electronic device 9600 via the central processing unit 9100.

[0187] The memory 9140 may also include a data storage unit 9143 for storing data, such as contacts, digital data, pictures, sounds, and / or any other data used by the electronic device. The driver storage unit 9144 of the memory 9140 may include various drivers for the electronic device's communication functions and / or for performing other functions of the electronic device (such as messaging applications, address book applications, etc.).

[0188] The communication module 9110 is a transmitter / receiver 9110 that transmits and receives signals via the antenna 9111. The communication module (transmitter / receiver) 9110 is coupled to the central processing unit 9100 to provide input signals and receive output signals, which can be the same as in a conventional mobile communication terminal.

[0189] Based on different communication technologies, multiple communication modules 9110 can be configured in the same electronic device, such as cellular network modules, Bluetooth modules, and / or wireless LAN modules. The communication module (transmitter / receiver) 9110 is also coupled to a speaker 9131 and a microphone 9132 via an audio processor 9130 to provide audio output via the speaker 9131 and receive audio input from the microphone 9132, thereby realizing typical telecommunications functions. The audio processor 9130 may include any suitable buffer, decoder, amplifier, etc. Additionally, the audio processor 9130 is coupled to a central processing unit 9100, enabling on-device recording via the microphone 9132 and on-device playback of stored sound via the speaker 9131.

[0190] Embodiments of this application also provide a computer-readable storage medium capable of implementing all steps of the etching endpoint monitoring method with a server or client execution subject in the above embodiments. The computer-readable storage medium stores a computer program that, when executed by a processor, implements all steps of the etching endpoint monitoring method with a server or client execution subject in the above embodiments. For example, when the processor executes the computer program, it implements the following steps:

[0191] Step S101: Acquire the spectral data collected by the spectrometer during the etching process, and perform filtering processing on the spectral data.

[0192] Step S102: Perform dimensionality reduction on the filtered spectral data to obtain at least one characteristic band of the spectral data.

[0193] Step S103: Perform numerical analysis on at least one characteristic band of the spectral data to determine the data model of the spectral data.

[0194] Step S104: Monitor the etching status of the real-time acquired spectral data using the data model of the spectral data, and control the etching process based on the monitoring results.

[0195] As can be seen from the above description, the computer-readable storage medium provided in the embodiments of this application effectively reduces the data dimension in the etching process, reduces data redundancy, reduces the misjudgment rate of interference signals on plasma etching results, greatly improves the robustness of the plasma etching process status monitoring model, and realizes automated management of the etching monitoring process, thereby improving etching accuracy.

[0196] Embodiments of this application also provide a computer program product capable of implementing all steps of the etching endpoint monitoring method in the above embodiments, where the execution subject is a server or a client. When executed by a processor, this computer program / instruction implements the steps of the etching endpoint monitoring method. For example, the computer program / instruction implements the following steps:

[0197] Step S101: Acquire the spectral data collected by the spectrometer during the etching process, and perform filtering processing on the spectral data.

[0198] Step S102: Perform dimensionality reduction on the filtered spectral data to obtain at least one characteristic band of the spectral data.

[0199] Step S103: Perform numerical analysis on at least one characteristic band of the spectral data to determine the data model of the spectral data.

[0200] Step S104: Monitor the etching status of the real-time acquired spectral data using the data model of the spectral data, and control the etching process based on the monitoring results.

[0201] As can be seen from the above description, the computer program product provided in this application embodiment effectively reduces the data dimension in the etching process, reduces data redundancy, reduces the misjudgment rate of interference signals on plasma etching results, greatly improves the robustness of the plasma etching process status monitoring model, and realizes automated management of the etching monitoring process, thereby improving etching accuracy.

[0202] Those skilled in the art will understand that embodiments of the present invention can be provided as methods, apparatus, or computer program products. Therefore, the present invention can take the form of a completely hardware embodiment, a completely software embodiment, or an embodiment combining software and hardware aspects. Furthermore, the present invention can take the form of a computer program product embodied on one or more computer-usable storage media (including, but not limited to, disk storage, CD-ROM, optical storage, etc.) containing computer-usable program code.

[0203] This invention is described with reference to flowchart illustrations and / or block diagrams of methods, apparatus (devices), and computer program products according to embodiments of the invention. It will be understood that each block of the flowchart illustrations and / or block diagrams, and combinations of blocks in the flowchart illustrations and / or block diagrams, can be implemented by computer program instructions. These computer program instructions can be provided to a processor of a general-purpose computer, special-purpose computer, embedded processor, or other programmable data processing apparatus to produce a machine, such that the instructions, which execute via the processor of the computer or other programmable data processing apparatus, generate instructions for implementing the flowchart illustrations and / or block diagrams. Figure 1 One or more processes and / or boxes Figure 1 A device that provides the functions specified in one or more boxes.

[0204] These computer program instructions may also be stored in a computer-readable storage medium that can direct a computer or other programmable data processing device to function in a particular manner, such that the instructions stored in the computer-readable storage medium produce an article of manufacture including instruction means, which are implemented in a process Figure 1 One or more processes and / or boxes Figure 1 The function specified in one or more boxes.

[0205] These computer program instructions may also be loaded onto a computer or other programmable data processing equipment to cause a series of operational steps to be performed on the computer or other programmable equipment to produce a computer-implemented process, thereby providing instructions that execute on the computer or other programmable equipment for implementing the process. Figure 1 One or more processes and / or boxes Figure 1The steps of the function specified in one or more boxes.

[0206] Specific embodiments have been used to illustrate the principles and implementation methods of this invention. The descriptions of the embodiments above are only for the purpose of helping to understand the method and core ideas of this invention. At the same time, for those skilled in the art, there will be changes in the specific implementation methods and application scope based on the ideas of this invention. Therefore, the content of this specification should not be construed as a limitation of this invention.

Claims

1. A method for monitoring the etching endpoint, characterized in that, The method includes: Acquire spectral data collected by the spectrometer during the etching process, and filter the spectral data; The filtered spectral data is subjected to dimensionality reduction to obtain at least one characteristic band of the spectral data. Numerical analysis is performed on at least one characteristic band of the spectral data to determine the data model of the spectral data; The etching status is monitored by using the data model of the spectral data in real time, and the etching process is controlled according to the monitoring results. The step of performing data dimensionality reduction on the filtered spectral data to obtain at least one characteristic band of the spectral data includes: filtering redundant wavelengths in the spectral data based on a preset absolute peak value to obtain filtered spectral data; determining the average contribution ratio of the filtered spectral data, and determining the optimal peak value of the spectral data based on the average contribution ratio; and determining the band corresponding to the optimal peak value as at least one characteristic band of the spectral data.

2. The method according to claim 1, characterized in that, The filtering process for the spectral data includes: The spectral data is decomposed using wavelet decomposition to obtain wavelet decomposition coefficients and high-frequency coefficients of each layer of the spectral data. Based on the preset coefficient threshold, the wavelet decomposition coefficients, and the high-frequency coefficients of each layer of the spectral data, the spectral data is subjected to inverse transformation to obtain the denoised spectral data. The denoised spectral data is sorted based on the signal sequence within a window of a preset width, and the signal located at the median position is determined as the filtering result of the spectral data.

3. The method according to claim 1, characterized in that, After filtering the spectral data, the process further includes: The spectral data is standardized based on the mean and standard deviation of the individual spectral data.

4. The method according to claim 1, characterized in that, The step of monitoring the etching status of the real-time acquired spectral data using the data model of the spectral data, and controlling the etching process based on the monitoring results, includes: The spectral data collected in real time by the spectrometer is input into the data model of the spectral data to obtain the upper and lower limit ranges of the current spectrum; In response to the fact that the spectral intensity of the current spectrum is within the constraint range of the upper and lower limits, the current etching process is maintained; In response to the current spectral intensity exceeding the constraint range of the upper and lower limits, the current etching process is terminated.

5. An etching endpoint monitoring device, characterized in that, The device includes: The data acquisition module is used to: acquire spectral data collected by the spectrometer during the etching process, and to filter the spectral data; A data dimensionality reduction module is used to: perform data dimensionality reduction on the filtered spectral data to obtain at least one characteristic band of the spectral data; specifically, the data dimensionality reduction module is used to: filter redundant wavelengths in the spectral data based on a preset absolute peak value to obtain filtered spectral data; determine the average contribution ratio of the filtered spectral data, and determine the optimal peak value of the spectral data based on the average contribution ratio; and determine the band corresponding to the optimal peak value as at least one characteristic band of the spectral data; a data mining module is used to: perform numerical analysis on at least one characteristic band of the spectral data to determine the data model of the spectral data; The status monitoring module is used to: monitor the etching status of the real-time acquired spectral data through the data model of the spectral data, and control the etching process based on the monitoring results.

6. The apparatus according to claim 5, characterized in that, When the data acquisition module performs filtering processing on the spectral data, it is specifically used for: The spectral data is decomposed using wavelet decomposition to obtain wavelet decomposition coefficients and high-frequency coefficients of each layer of the spectral data. Based on the preset coefficient threshold, the wavelet decomposition coefficients, and the high-frequency coefficients of each layer of the spectral data, the spectral data is subjected to inverse transformation to obtain the denoised spectral data. The denoised spectral data is sorted based on the signal sequence within a window of a preset width, and the signal located at the median position is determined as the filtering result of the spectral data.

7. The apparatus according to claim 5, characterized in that, The status monitoring module, when monitoring the etching status of the real-time acquired spectral data using the data model of the spectral data, and controlling the etching process based on the monitoring results, is specifically used for: The spectral data collected in real time by the spectrometer is input into the data model of the spectral data to obtain the upper and lower limit ranges of the current spectrum; In response to the fact that the spectral intensity of the current spectrum is within the constraint range of the upper and lower limits, the current etching process is maintained; In response to the current spectral intensity exceeding the constraint range of the upper and lower limits, the current etching process is terminated.

8. An electronic device comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, characterized in that, When the processor executes the program, it implements the steps of the etching endpoint monitoring method according to any one of claims 1 to 4.

Images