Etching end point determination method and related equipment

Abstract

The invention provides an etching end point determination method and related equipment, and the method comprises the steps: obtaining first spectrum data used for interference measurement and second spectrum data used for optical emission spectrum measurement in an etching processing process; respectively carrying out feature extraction on the first spectral data and the second spectral data to obtain a first feature of the first spectral data and a second feature of the second spectral data; performing feature fusion on the first feature and the second feature based on an etching rate to obtain a fusion feature of the etching treatment; and determining an end point of the etching processing based on the fusion feature. By using the etching end point determination method provided by the invention, the reliability and accuracy of etching end point determination can be effectively improved.

Description

Technical Field

[0001] This application relates to the field of etching technology, and in particular to a method and related equipment for determining the etching endpoint. Background Technology

[0002] In semiconductor manufacturing processes, etching is one of the key steps in chip pattern transfer. To ensure device dimensional accuracy and interlayer structural consistency, it is crucial to accurately determine the etching endpoint during the etching process. Determining the endpoint too early may result in incomplete etching of the film layer, leading to incomplete pattern transfer due to residual film. Conversely, determining the endpoint too late may damage the underlying material. Therefore, the accuracy of etching endpoint determination directly impacts product yield and process stability.

[0003] In existing technologies, common methods for detecting etching endpoints mainly include optical emission spectroscopy (OES) and interferometry endpoint (IEP). OES reflects the state of the reaction process by monitoring changes in the intensity of the emission spectrum in the plasma during etching. It has the advantages of fast response speed and applicability to various materials, but its measurement results are prone to fluctuations under conditions of low emission intensity or strong signal interference. The IEP method estimates the film thickness by monitoring changes in the interference period of the reflected light signal, and can accurately reflect the film removal process. However, when the surface roughness is large or the reflectivity is reduced, the interference fringes are prone to distortion or disappearance. Summary of the Invention

[0004] In view of this, this application provides a method and related equipment for determining the etching endpoint, which can effectively improve the reliability and accuracy of etching endpoint determination.

[0005] In a first aspect, embodiments of this application provide a method for determining the etching endpoint, the method comprising: Acquire the first spectral data used for IEP and the second spectral data used for OES during the etching process; Feature extraction is performed on the first spectral data and the second spectral data respectively to obtain a first feature of the first spectral data and a second feature of the second spectral data. The first feature includes the observed value of the film thickness, and the second feature includes one or more of the observed value of the plasma concentration and the observed value of the intensity of the characteristic peak in the plasma emission spectrum. Based on the etching rate, the first feature and the second feature are fused to obtain the fused feature of the etching process. The fused feature includes one or more of the following: an estimated value of the etching rate, an estimated value of the film thickness, an estimated value of the intensity of the characteristic peak in the plasma emission spectrum, and an estimated value of the plasma concentration. The endpoint of the etching process is determined based on the fusion characteristics.

[0006] Therefore, in the embodiments of this application, first spectral data for IEP and second spectral data for OES are acquired during the etching process. Features are extracted from both types of spectral data to obtain first and second features, respectively. These features are then fused based on the etching rate, and the etching endpoint is determined based on the fused features. Specifically, the first feature is, for example, an observed value of film thickness, and the second feature is, for example, one or more of an observed value of plasma concentration and an observed value of the intensity of characteristic peaks in the plasma emission spectrum. Since the etching rate can reflect the changes in the observed film thickness, the observed values ​​of the intensity of characteristic peaks in the plasma emission spectrum, and the observed values ​​of plasma concentration, the etching rate can correlate the first and second features, allowing for mutual monitoring and verification, and enabling the correction of their respective values. This results in more accurate estimates of the etching rate, film thickness, plasma concentration, and the intensity of characteristic peaks in the plasma emission spectrum. In summary, feature fusion allows for the simultaneous utilization of the IEP signal's sensitive response to film thickness changes and the OES signal's characterization of plasma chemical reactions. Furthermore, the two are correlated through etching rate, enabling more efficient and precise fusion. The resulting fused features can adaptively reflect etching conditions under different materials, process formulations, and equipment, achieving a comprehensive assessment of the etching process status and effectively improving the reliability and accuracy of etching endpoint determination.

[0007] In conjunction with the first aspect, in one possible implementation, the feature fusion based on the etching rate of the first feature and the second feature to obtain the fused feature of the etching process includes: Establish a first relationship between film thickness and etching rate, and a second relationship between plasma concentration or the intensity of characteristic peaks in plasma emission spectrum and etching rate; The first feature and the second feature are used as observations for Kalman filtering, and the fused feature is used as the state variable for Kalman filtering. The fused feature is obtained by fusing the first feature and the second feature based on the first relationship and the second relationship using the Kalman filtering algorithm.

[0008] Therefore, in the embodiments of this application, the etching rate is used as an intermediate state variable to correlate film thickness with plasma concentration or the intensity of characteristic peaks in the plasma emission spectrum, forming a two-way closed-loop feedback path. On the one hand, by utilizing the optimal estimation characteristics of Kalman filtering, recursive fusion and mutual verification of IEP and OES features are achieved, significantly improving the fusion accuracy. On the other hand, by explicitly establishing physical relationship models (first and second relationships), the interpretability of feature fusion is enhanced, enabling the fused features to adaptively reflect the actual etching state under different material, process formulation, and equipment conditions, thereby further improving the reliability and accuracy of etching endpoint determination.

[0009] In conjunction with the first aspect, in one possible implementation, the step of using the Kalman filter algorithm to fuse the first feature and the second feature based on the first relation and the second relation to obtain the fused feature includes: The fused feature is obtained by fusing the first feature and the second feature based on the Kalman gain matrix, the first relation, and the second relation using the Kalman filtering algorithm. The Kalman gain matrix reflects the signal-to-noise ratio of the interferometric measurement and the optical emission spectrum measurement.

[0010] Therefore, in the embodiments of this application, adaptive weight allocation is achieved by dynamically reflecting the signal-to-noise ratio levels of IEP and OES measurements through the Kalman gain matrix. This dynamic adjustment mechanism significantly enhances the robustness of the system in complex process environments, ensuring that the fusion results always tend towards the high-confidence measurement source, thereby further improving the stability and anti-interference capability of etching endpoint determination.

[0011] In conjunction with the first aspect, in one possible implementation, as the noise of the interferometric measurement increases, the weights in the Kalman gain matrix corresponding to the first feature decrease; and / or When the noise in the optical emission spectrum measurement increases, the weight corresponding to the second feature in the Kalman gain matrix decreases.

[0012] Therefore, in the embodiments of this application, a noise-adaptive weight adjustment mechanism is achieved through the Kalman gain matrix. When the noise of interferometric or optical emission spectroscopy measurements increases due to process fluctuations, equipment aging, or environmental interference, the system automatically reduces the fusion weight of the corresponding feature, thereby effectively isolating the influence of unreliable signals on the fusion result and ensuring that the etching endpoint determination is always based on the most reliable measurement information, thus significantly improving the accuracy of endpoint detection.

[0013] In conjunction with the first aspect, in one possible implementation, determining the endpoint of the etching process based on the fusion features includes: The endpoint of the etching process is determined based on one or more of the following: The estimated etching rate is less than a first threshold; The estimated thickness of the film layer is less than the second threshold; The change in the estimated thickness of the film layer is less than the third threshold. The estimated intensity of the characteristic peak in the plasma emission spectrum is greater than the fourth threshold or less than the fifth threshold. The estimated change in the intensity of the characteristic peaks in the plasma emission spectrum is greater than the sixth threshold. The estimated value of the plasma concentration is greater than the seventh threshold or less than the eighth threshold; The change in the estimated plasma concentration is greater than the ninth threshold.

[0014] Therefore, the embodiments of this application provide a multi-dimensional and configurable etching endpoint determination strategy. Since the estimated values ​​of each feature are optimized using Kalman filtering, their accuracy is significantly higher than the direct observation values ​​of traditional single-measurement methods. Therefore, higher detection accuracy can be achieved by determining the endpoint based on any of the aforementioned indicators. Simultaneously, the multi-indicator comprehensive determination mechanism allows for flexible configuration of weights or logic according to specific process requirements, effectively reducing the risk of misjudgment based on a single indicator and significantly improving the method's versatility and adaptability to different etching processes.

[0015] Secondly, embodiments of this application provide an etching endpoint determination apparatus, the apparatus comprising: Acquisition circuitry is used to acquire first spectral data for IEP and second spectral data for OES during the etching process. An extraction circuit is used to extract features from the first spectral data and the second spectral data respectively to obtain a first feature of the first spectral data and a second feature of the second spectral data. The first feature includes an observed value of the film thickness, and the second feature includes one or more of an observed value of the plasma concentration and an observed value of the intensity of the characteristic peak in the plasma emission spectrum. A fusion circuit is used to fuse the first feature and the second feature based on the etching rate to obtain the fused feature of the etching process. The fused feature includes one or more of the following: an estimated value of the etching rate, an estimated value of the film thickness, an estimated value of the intensity of the characteristic peak in the plasma emission spectrum, and an estimated value of the plasma concentration. A determining circuit is used to determine the endpoint of the etching process based on the fusion features.

[0016] In conjunction with the second aspect, in one possible implementation, any number of the acquisition circuit, the extraction circuit, the fusion circuit, and the determination circuit are heterogeneously deployed.

[0017] Thirdly, embodiments of this application provide an etching endpoint determination apparatus, including a memory and a processor. The memory is used to store a computer program, the computer program includes program instructions, and the processor is configured to invoke the program instructions to execute the steps in the method designed in the first aspect above.

[0018] Fourthly, embodiments of this application provide a spectrometer that includes the apparatus described in the second or third aspect.

[0019] Fifthly, embodiments of this application provide a semiconductor device that includes the apparatus described in the second or third aspect.

[0020] In a sixth aspect, embodiments of this application provide a computer-readable storage medium storing a computer program for electronic data interchange, wherein the computer program causes a computer to perform some or all of the steps described in the method of the first aspect of embodiments of this application.

[0021] In a seventh aspect, embodiments of this application provide a computer program product, wherein the computer program product includes a non-transitory computer-readable storage medium storing a computer program operable to cause a computer to perform some or all of the steps described in the method of the first aspect of embodiments of this application. The computer program product may be a software installation package.

[0022] The beneficial effects of the technical solutions in the second to sixth aspects can be found in the technical effects of the technical solution in the first aspect, and will not be repeated here. Attached Figure Description

[0023] To more clearly illustrate the technical solutions in the embodiments of this application, the drawings used in the description of the embodiments 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 from these drawings without creative effort.

[0024] Figure 1 A schematic diagram of an etching system provided for an embodiment of this application; Figure 2 A schematic diagram of a heterogeneous etching system provided for an embodiment of this application; Figure 3 A schematic diagram of another heterogeneous etching system provided for an embodiment of this application; Figure 4A schematic diagram of an integrated etching system provided for an embodiment of this application; Figure 5 A flowchart illustrating a method for determining the etching endpoint provided in this application. Figure 6 A functional unit block diagram of an etching endpoint determination device provided for embodiments of this application; Figure 7 This is a schematic diagram of an etching endpoint determination device provided for an embodiment of this application.

[0025] Explanation of reference numerals in the attached figures: 600 - Etching endpoint determination device; 601 - Acquisition circuit; 602 - Extraction circuit; 603 - Fusion circuit; 604 - Determination circuit; 700 - Etching endpoint determination device; 710 - Processor; 720 - Memory; 721 - Computer program or instructions. Detailed Implementation

[0026] To enable those skilled in the art to better understand the present application, the technical solutions in the embodiments of the present application will be clearly and completely described below with reference to the accompanying drawings.

[0027] The terms "first," "second," etc., in the specification, claims, and accompanying drawings of this application are used to distinguish different objects, not to describe a specific order. Furthermore, the terms "comprising" and "having," and any variations thereof, are intended to cover non-exclusive inclusion. For example, a process, method, system, product, or apparatus that includes a series of steps or units is not limited to the listed steps or units, but may optionally include steps or units not listed, or may optionally include other steps or units inherent to these processes, methods, products, or apparatuses.

[0028] In the embodiments of this application, "at least one item" or similar expressions refer to any combination of these items, including any combination of a single item or a plurality of items. "One or more" refers to one or more items, while "multiple" refers to two or more items. For example, "at least one item" of a, b, or c can represent the following seven cases: a, b, c; a and b; a and c; b and c; a, b, and c. Each of a, b, and c can be an element or a set containing one or more elements.

[0029] In the embodiments of this application, "connection" refers to various connection methods, such as direct connection or indirect connection, to achieve communication between devices. The embodiments of this application do not impose any limitations on this. In the description of this application, it should be noted that, unless otherwise explicitly specified and limited, the terms "installation," "connection," and "connection" should be interpreted broadly. In one example, it can be a fixed connection, a detachable connection, or an integral connection; it can be a mechanical connection, an electrical connection, or a connection that allows mutual communication; it can be a direct connection or an indirect connection through an intermediate medium; it can be the internal communication of two components or the interaction between two components.

[0030] In this document, the term "implementation" means that a specific feature, structure, or characteristic described in connection with an implementation may be included in at least one implementation of this application. The appearance of this phrase in various places in the specification does not necessarily refer to the same implementation, nor is it a separate or alternative implementation mutually exclusive with other implementations. It will be explicitly and implicitly understood by those skilled in the art that the implementations described herein can be combined with other implementations.

[0031] First, the etching system to which the etching endpoint determination method proposed in the embodiments of this application is applicable will be described. The etching system typically includes etching equipment and endpoint detection equipment.

[0032] In this embodiment, see Figure 1 , Figure 1 This is a schematic diagram of an etching system applicable to the embodiments of this application.

[0033] See Figure 1 ,exist Figure 1 In the etching system shown, the endpoint detection equipment includes an IEP light source, an IEP spectrometer, an OES spectrometer, a spectral feature extraction unit, a feature fusion unit, and an endpoint determination unit.

[0034] The IEP light source provides the white / monochrome light source required for IEP endpoint detection; the IEP spectrometer acquires the first spectral data for IEP and sends it to the spectral feature extraction unit; the OES spectrometer acquires the second spectral data for OES and sends it to the spectral feature extraction unit; the spectral feature extraction unit preprocesses the first and second spectral data and extracts the first feature of the first spectral data and the second feature of the second spectral data; the feature fusion unit fuses the first and second features and sends the fused feature to the endpoint determination unit; the endpoint determination unit performs high-precision real-time determination of the etching endpoint based on the fused feature.

[0035] In this embodiment, one or more of the above-mentioned spectral feature extraction unit, feature fusion unit, and endpoint determination unit can be heterogeneously deployed in other structures. For example, the spectral feature extraction unit can be heterogeneously deployed on architectures such as Field-Programmable Gate Array (FPGA), Application-Specific Integrated Circuit (ASIC) chips, and Advanced RISC Machine (ARM) + FPGA; the feature fusion unit can be heterogeneously deployed on architectures such as FPGA, ASIC, and Central Processing Unit (CPU); and the endpoint determination unit can be heterogeneously deployed on architectures such as Personal Computer (PC) host computer.

[0036] For example, consider a scenario where the spectral feature extraction unit and the feature fusion unit are heterogeneously deployed on an FPGA, and the endpoint determination unit is heterogeneously deployed on a PC host computer. Figure 2 As shown, the spectrometers (IEP spectrometer, OES spectrometer) send raw spectral data to the FPGA carrying the spectral feature extraction unit and the feature fusion unit via an Ethernet interface. After receiving the spectral data, the FPGA performs preprocessing on the raw spectral data through its spectral data preprocessing unit, including filtering, baseline correction, and background subtraction, to remove noise interference and enhance the separability of spectral features. Then, the preprocessed spectral data is sent to the spectral feature extraction unit for feature extraction. Depending on the process requirements, different algorithm paths can be selected, such as Principal Component Analysis (PCA) implemented through hardening, stripe counting, peak finding algorithms, and model fitting, to extract key spectral features representing plasma concentration and film thickness. Next, the key spectral features are sent to the feature fusion unit to fuse the features obtained from IEP and OES, resulting in fused features. Finally, the fused features are sent to the PC host computer, where the endpoint determination unit determines the etching endpoint.

[0037] In this embodiment, such as Figure 3 As shown, the spectral feature extraction unit, feature fusion unit, and endpoint determination unit can also be uniformly and heterogeneously deployed on a PC host computer. For example, a system-on-chip (SoC) architecture integrating hardware acceleration units and embedded processors can be adopted. The SoC chip is set up in the PC host computer, and this SoC chip uses spectral data fusion and etching endpoint determination based on the collaborative processing of hardware IP cores and embedded CPUs. Or, as... Figure 4As shown, the spectral feature extraction unit and the feature fusion unit can also be integrated into the spectrometer. For example, spectral data fusion can be achieved by integrating a dedicated custom chip inside the spectrometer.

[0038] It should be noted that, in Figure 3 and Figure 4 In the structure, the operation mode of each unit can be referred to the above. Figure 2 The operation of each unit will not be elaborated here.

[0039] In this embodiment, the heterogeneous deployment approach improves system real-time performance and enhances parallel processing capabilities. Each module can be flexibly configured according to the actual usage environment, exhibiting high compatibility with older equipment and effectively improving the real-time performance and stability of the endpoint detection system. Simultaneously, the integrated deployment approach allows the data processing flow—from spectral acquisition, data preprocessing, feature extraction to feature fusion—to be completed within the spectrometer. This significantly reduces the data transmission load and processing latency of the endpoint detection system, resulting in higher integration and response speed, further enhancing its real-time performance and stability. Users can choose a suitable system architecture for deployment based on their actual operating environment, greatly expanding the applicability of this solution.

[0040] Understandable Figure 1 In the etching system shown, the form and number of the IEP light source, IEP spectrometer, OES spectrometer, spectral feature extraction unit, feature fusion unit and endpoint determination unit are only examples and do not constitute a limitation on the implementation of this application.

[0041] For example, the etching system may also include servers or other devices.

[0042] For example, the etching system may include other light source devices in addition to the IEP light source shown.

[0043] For example, the etching system may include other spectrometer devices in addition to the IEP spectrometer and OES spectrometer shown.

[0044] For example, in addition to the spectral feature extraction unit, feature fusion unit, and endpoint determination unit shown, the etching system may include other functional units.

[0045] The technical solutions in the embodiments of this application will be clearly and completely described below with reference to the above content and the accompanying drawings. Obviously, the described embodiments are only a part of the embodiments of this application, and not all of them. Based on the embodiments in this application, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of this application. At the same time, these embodiments can be related to each other or independent of each other, and the same content between different embodiments can be referenced by each other, which will not be elaborated here.

[0046] The following will combine Figure 5 This application describes a method for determining the etching endpoint in one of its embodiments. Figure 5 A flowchart illustrating an etching endpoint determination method provided in this application is shown. This method can be applied to... Figure 1 The etching system shown specifically includes the following steps: S501: Acquire the first spectral data for IEP and the second spectral data for OES during the etching process.

[0047] In this embodiment, the first spectral data can be acquired by an IEP spectrometer in the etching system, and the second spectral data can be acquired by an OES spectrometer in the etching system.

[0048] It should be noted that other methods and apparatuses in the art that can acquire first spectral data for IEP and second spectral data for OES during the etching process can also be applied to this application, and this application does not limit them.

[0049] S502: Perform feature extraction on the first spectral data and the second spectral data respectively to obtain the first feature of the first spectral data and the second feature of the second spectral data.

[0050] In this embodiment, the first feature is typically the thickness information of the film layer, such as the observed value of the film layer thickness, and the second feature is typically the concentration information of the plasma, such as one or more of the observed value of the plasma concentration and the observed value of the intensity of the characteristic peak at a given wavelength in the plasma emission spectrum.

[0051] It should be noted that the feature extraction method for the first spectral data can refer to the feature extraction method in the existing IEP technology, and the feature extraction method for the second spectral data can also refer to the feature extraction method in the existing OES technology, which will not be elaborated here.

[0052] S503: Based on the etching rate, the first feature and the second feature are fused to obtain the fused feature of the etching process.

[0053] In this embodiment, the fusion feature is used to identify the physical characteristics of the etched area during the etching process. It may include one or more of the following: the estimated value of the film thickness, the estimated value of the plasma concentration, the estimated value of the intensity of the characteristic peak in the plasma emission spectrum, and the estimated value of the etching rate. It can accurately reflect the current etching status and then make a high-precision determination of the etching endpoint.

[0054] In existing technologies, film thickness, the intensity of characteristic peaks in plasma emission spectra, and plasma concentration can be obtained through IEP and OES, i.e., by analyzing first and second spectral data. These methods are susceptible to various types of noise interference, consequently compromising the accuracy of the obtained film thickness, characteristic peak intensity, and plasma concentration data. Therefore, in this embodiment, etching rate is incorporated as an intermediate parameter to fuse the first and second characteristics, resulting in more accurate film thickness, characteristic peak intensity, and plasma concentration data.

[0055] Specifically, the etching rate can reflect the changes in observed film thickness, the intensity of characteristic peaks in the plasma emission spectrum, and the plasma concentration during the etching process. The etching rate can then be derived from both the first and second characteristics, and since the etching rate should be unique at any given moment, the first and second characteristics can mutually monitor and verify each other, allowing for the correction of their respective values. This leads to more precise measurements of various parameters, including film thickness, plasma concentration, and the intensity of characteristic peaks in the plasma emission spectrum.

[0056] It should be noted that, in this embodiment, any form of statistical filtering or probabilistic fusion mechanism can be used to fuse the first feature and the second feature based on the etching rate. Examples include, but are not limited to, Kalman filtering, Bayesian fusion, particle filtering, or neural network methods.

[0057] The following section will take Kalman filtering as an example, and will explain in detail the method of feature fusion based on etching rate for the first and second features, including the estimated values ​​of film thickness, plasma concentration and etching rate.

[0058] First, a feature vector is constructed based on the first and second features. This feature vector is used to identify the changes in the etched region observed by IEP and OES techniques during time processing. This feature vector can also be called the observation vector. Specifically, the feature vector can be as follows:

[0059] Among them, z k iepThis represents the first feature obtained by extracting features from the spectral data at time k in the first spectral data using IEP technology, z. k oes This represents the second feature obtained by extracting features from the spectral data at time k in the second spectral data using OES technology, z k This represents the eigenvector corresponding to time k.

[0060] Then, a vector of fused features is constructed. As mentioned earlier, this vector of fused features is used to identify the physical characteristics of the etched region during the etching process, that is, the actual physical state of the etched region. Consequently, this vector can also be called a state vector. Therefore, the vector of fused features can be constructed using film thickness and plasma concentration. At the same time, in order to make the components in the constructed vector interrelated, the etching rate can be added as a new component to the vector of fused features to identify the dynamic coupling relationship between film thickness and plasma concentration.

[0061] Specifically, a first relationship can be established between the film thickness and the etching rate, which can be expressed by the kinetic equation shown in the following formula ①: ………① Where, d k s represents the film thickness of the etched region at time k. k Δt represents the etching rate, and Δt represents the time interval.

[0062] It can be seen that the film thickness d k With etching rate s k There is a certain relationship between them, through the film thickness d k The etching rate s can be derived by reverse engineering. k .

[0063] Simultaneously, a second relationship can be established between plasma concentration or the intensity of characteristic peaks in the plasma emission spectrum and the etching rate. Taking plasma concentration as an example, plasma concentration c k The chemical reactivity of the etching products in the reaction plasma is considered, while the etching rate sk directly reflects the physical removal rate of the etched area. Based on this, the second relationship can be expressed by the empirical process model shown in the following formula ②: ………② Where 'a' represents the process-related index.

[0064] It can be seen that the plasma concentration c k With etching rate s k There is also a certain connection between them; similarly, it can be determined by plasma concentration c. k Reverse derive the etching rate s k .

[0065] Therefore, in this embodiment, the etching rate s is... k As an intermediate variable, the film thickness d k With plasma concentration c k This is related to the thickness d of the film. k With plasma concentration c k The characteristics during processing are due to the etching rate s k The participation of these elements allows for mutual integration and calibration, which in turn makes the determination of the etching endpoint more accurate in subsequent use.

[0066] Based on this, the vector of fused features can be represented as follows:

[0067] Where, x k This represents the vector of fused features at time k.

[0068] Finally, in this embodiment, a first model can be constructed based on Kalman filtering. The first and second features are used as observations for Kalman filtering, and the fused features are used as state variables. The Kalman filtering algorithm is then used to fuse the first and second features based on the first and second relations to obtain the fused features. Specifically, the feature vector z constructed above... k As an observable of Kalman filtering, the vector x of fused features k As the state variables of the Kalman filter, the feature vector x is processed through the Kalman filter, the first relation, and the second relation. k The values ​​of the components included in the etching process are estimated, and then the physical characteristics of the etched area are determined, so as to accurately determine the etching endpoint.

[0069] Specifically, based on the etching process, the feature vector x is fused. k The state prediction equation in Kalman filtering needs to be satisfied, specifically as shown in equation ③: ………③ Where, x' k+1 Let x represent the prior state estimate of the fused features at time k+1. k Let A represent the posterior state estimate of the fused features at time k, A represent the state transition matrix, B represent the control matrix, reflecting the influence of the control input on the prediction, and w represent the posterior state estimate of the fused features at time k. k This indicates a control input.

[0070] It should be noted that, in this embodiment, the prior state estimate refers to the predicted value obtained directly through Kalman filtering without correction using measured values, while the posterior state estimate refers to the estimate obtained after correction using measured values. The vector of fused features includes the posterior state estimate.

[0071] In this embodiment, Bw k It can be represented as noise in the state transition process, reflecting random disturbances and modeling errors.

[0072] At the same time, the state transition matrix A needs to satisfy the covariance prediction equation in Kalman filtering, specifically as shown in equation ④: ………④ Among them, P' k+1 Let P be the prior error covariance matrix at time k+1. k Let represent the posterior error covariance matrix at time k, and Q represent the random error. In this embodiment, to achieve joint correction of OES and IEP data, a Kalman gain matrix of the following form is also constructed in the first model:

[0073] Where, n 11 For z k iep Regarding the film thickness d k The corrected weights, n 12 For z k oes Regarding the film thickness d k Indirectly corrected weights, n 21 For z k iep For etching rate s k The corrected weights, n 22 For z k oes For etching rate s k The corrected weights, n 31 For z k iep For plasma concentration c k Indirectly corrected weights, n 32 For z k oes For plasma concentration c k Corrected weights.

[0074] In this embodiment, the Kalman gain matrix G k This reflects the signal-to-noise ratio (SNR) of the IEP and OES. For example, when the IEP is interfered with, its noise increases, and the covariance of the IEP automatically increases, thereby reducing k. 11 k21 and k 31 When an OES is disturbed, its noise increases, the covariance of the OES automatically increases, and k automatically decreases. 12 k 22 and k 32 Subsequently, the risk of misjudgment caused by anomalies in either IEP or OES was effectively suppressed, significantly improving the system robustness.

[0075] Specifically, the Kalman gain matrix G k It can be shown in formula ⑤: ………⑤ Where H represents the observation matrix and R represents the measurement noise covariance matrix, reflecting the sensor error in IEP and OES measurements.

[0076] Therefore, based on the Kalman gain matrix G k and the input feature vector z at time k k The predicted value directly obtained from the first model is the prior state estimate, which is x'. k After correction, the posterior state estimate x is obtained. k That is, the fusion feature.

[0077] Specifically, the prior state estimate, x' k The correction process can be represented by the following formula ⑥: ………⑥ Where, x' k Let z represent the prior state estimate of the fusion feature at time k. k -Hx' k () represents the difference between the measured value and the predicted value at time k.

[0078] Similarly, for the prior covariance matrix P' at time k... k In addition, it is also necessary to use the Kalman gain matrix G. k After correction, the posterior covariance matrix P at time k is obtained. k Then, the prior covariance matrix P' at time k+1 is determined using the above formula ④. k+1 This enables subsequent predictions.

[0079] Specifically, the prior covariance matrix P' k The correction process can be represented by the following formula ⑦: ………⑦ Where I represents the identity matrix.

[0080] Based on this, the prediction process at time k in the first model above is as follows: First, based on the posterior covariance matrix P at time k-1, using formula ④ k-1 Determine the prior covariance matrix P' at time k. k Meanwhile, using formula ③, the posterior state estimate x based on time k-1 is obtained. k-1 Determine the prior state estimate x' at time k. k Then, based on the prior covariance matrix P' using formula ⑤... k Determine the Kalman gain matrix G k Finally, based on Equation ⑥, the Kalman gain matrix G is used... k Prior state estimate x' k and the input feature vector z at time k k Determine the posterior state estimate x at time k. k As the fusion feature output, the posterior covariance matrix P at time k is determined using Equation ⑦. k This is so that it can be used in the prediction process at time k+1.

[0081] Therefore, based on the above description of the first model, the three components of the fused feature vector include the etching rate s. k Film thickness d k and plasma concentration c k This will form a two-way closed-loop feedback path, as follows: (1) Forward path: film thickness d k →Etching rate s k →Plasma concentration c k 1. In the first model, the film thickness d is determined by the first feature in the input feature vector. k The etching rate s can be affected by the state transition matrix A. k ; 2. Etching rate s k The change will indirectly affect the plasma concentration c through the concentration decay model. k .

[0082] (2) Reverse path: plasma concentration c k →Etching rate s k →Film thickness d k 1. In the first model, the plasma concentration c is determined by the second feature in the input feature vector. k It can be obtained through the Kalman gain matrix G k n 22 Corrected etching rate s k ; 2. The corrected etching rate sk The film thickness d at the next moment can be changed using formula ①. k+1 .

[0083] Therefore, the addition of etching rate enables the first and second features extracted by IEP and OES to mutually verify each other, resulting in a higher degree of fusion between the two, thereby improving the accuracy of etching endpoint determination based on the fused feature.

[0084] In this embodiment, by constructing the aforementioned feature vector, the fused feature vector, and the first model, the first and second features extracted through IEP and OES are used to determine the actual physical characteristics of the etched region. Subsequently, the etch endpoint can be determined with high precision using the actual physical characteristics of the etched region. However, the aforementioned first model is only an initial model and requires further training with appropriate training data to become a usable feature fusion model.

[0085] Specifically, historical data from the etching process can be obtained. This historical data may include first historical spectral data used for IEP (In-Process Etching), second historical spectral data used for OES (Out-of-Process Etching), and historical physical characteristics of the etched region, namely, film thickness variation data and plasma concentration variation data. Among these, feature extraction of the first historical spectral data yields the z-value corresponding to the first model. k iep The data; after feature extraction from the second historical spectral data, we can obtain the z corresponding to the first model. k oes The data; the film thickness variation data can correspond to the film thickness d in the first model. k The data; plasma concentration variation data can correspond to the plasma concentration c in the first model. k The data; after taking the first derivative of the film thickness variation data, the etching rate s corresponding to the first model can be obtained. k The data is then used to train the first model, determining the state transition matrix A, control matrix B, observation matrix H, etc., thus obtaining a feature fusion model that can be used in practice.

[0086] In this embodiment, after obtaining the feature fusion model, the feature vector constructed from the first feature and the second feature can be input into the feature fusion model to obtain the fused features.

[0087] S504: Determine the endpoint of the etching process based on the fusion features.

[0088] In some embodiments, determining the endpoint of the etching process based on fusion characteristics includes: determining the endpoint based on one or more of the following: an estimated etching rate less than a first threshold; an estimated film thickness less than a second threshold; a change in the estimated film thickness less than a third threshold; an estimated intensity of a characteristic peak in the plasma emission spectrum greater than a fourth threshold or less than a fifth threshold; a change in the estimated intensity of a characteristic peak in the plasma emission spectrum greater than a sixth threshold; an estimated plasma concentration greater than a seventh threshold or less than an eighth threshold; and a change in the estimated plasma concentration greater than a ninth threshold. Here, the first to ninth thresholds can be determined based on the properties of the etching material.

[0089] In this embodiment, the fusion features include one or more of the following: an estimated value of the film thickness of the etched region, an estimated value of the plasma concentration, an estimated value of the intensity of the characteristic peak in the plasma emission spectrum, and an estimated value of the time rate. Therefore, when the fusion features only include one of the estimated values ​​of the film thickness, the intensity of the characteristic peak in the plasma emission spectrum, or the plasma concentration, the determination of whether the etching endpoint has been reached can be based on existing methods for determining the single etching endpoint using film thickness, the intensity of the characteristic peak in the plasma emission spectrum, or the plasma concentration. For example, the determination of whether the etching endpoint has been reached can be made by determining that: the estimated film thickness is less than a second threshold; the change in the estimated film thickness is less than a third threshold; the estimated intensity of the characteristic peak in the plasma emission spectrum is greater than a fourth threshold or less than a fifth threshold; the change in the estimated intensity of the characteristic peak in the plasma emission spectrum is greater than a sixth threshold; the estimated plasma concentration is greater than a seventh threshold or less than an eighth threshold; or the change in the estimated plasma concentration is greater than a ninth threshold. Since the fusion characteristics include the estimated values ​​of film thickness, the intensity of characteristic peaks in the plasma emission spectrum, or the plasma concentration, which are obtained based on the first model mentioned above, they are more accurate than the film thickness or plasma concentration obtained by direct measurement through IEP or OES in the prior art. Therefore, even if the existing determination method is used to determine the etching endpoint, a more accurate determination result can be obtained.

[0090] Alternatively, if the fusion features only include the etching rate, the etching endpoint can be determined by analyzing the characteristics of the etching rate. For example, in some etching processes, as the etching endpoint approaches, the etched material is often also nearing completion, causing the etching rate to drop sharply and stabilize. Then, the etching endpoint can be analyzed by determining if the estimated etching rate is less than a first threshold.

[0091] Alternatively, if the fusion features include multiple estimates, each estimate can be used to make a judgment before a comprehensive analysis is performed, or weights can be set for analysis and judgment. This application does not impose any restrictions on this.

[0092] It should be noted that the above description of determining the etching endpoint is only one possible example and does not limit this application to determining the etching endpoint solely through the above method. Other methods for determining the etching endpoint in the art can also be applied to this application, and this application does not impose any restrictions on them.

[0093] As can be seen, the method of this application, by acquiring first spectral data for IEP and second spectral data for OES during the etching process, extracting features from the two types of spectral data, fusing features based on the etching rate, and determining the etching endpoint based on the fused features, allows for the simultaneous utilization of the sensitive response characteristics of the IEP signal to changes in film thickness and the characterization ability of the OES signal to plasma chemical reactions. By correlating the two through the etching rate, a more efficient and accurate fusion is achieved. The resulting fused features can adaptively reflect the etching conditions under different materials, process formulations, and equipment conditions, enabling a comprehensive judgment of the etching process state and effectively improving the reliability and accuracy of etching endpoint determination.

[0094] The foregoing primarily describes the implementation scheme of this application from a methodological perspective. It is understood that, to achieve the above functions, the apparatus may include hardware structures and / or software circuits corresponding to the execution of each function. Those skilled in the art should readily recognize that, based on the units and algorithm steps of the various examples described in conjunction with the embodiments disclosed herein, this application can be implemented in hardware or a combination of hardware and computer software. Whether a function is executed in a hardware or computer software-driven hardware manner depends on the specific application and design constraints of the technical solution. Those skilled in the art can use different methods to implement the described functions for each specific application, but such implementation should not be considered beyond the scope of this application.

[0095] The embodiments of this application can divide the device into functional units according to the above method examples. For example, each function can be divided into a separate functional unit, or two or more functions can be integrated into one processing unit. It should be noted that the unit division in the embodiments of this application is illustrative and is only a logical functional division, while other division methods may be used in actual implementation.

[0096] When using integrated units, Figure 6 This is a functional unit block diagram of an etching endpoint determination device according to an embodiment of this application. The etching endpoint determination device 600 includes an acquisition circuit 601, an extraction circuit 602, a fusion circuit 603, and a determination circuit 604.

[0097] In this embodiment, the acquisition circuit 601, extraction circuit 602, fusion circuit 603, and determination circuit 604 can be any circuit unit used to receive and process signals, information, etc., or to determine a monitoring mechanism, and no specific limitations are imposed on them.

[0098] In this embodiment, the etching endpoint determination device 600 may further include a storage unit for storing computer program code or instructions executed by the etching endpoint determination device 600. The storage unit may be a memory.

[0099] In this embodiment, the etching endpoint determination device 600 may be a chip or a chip module.

[0100] In this embodiment, the acquisition circuit 601, extraction circuit 602, fusion circuit 603, and determination circuit 604 can be integrated into the communication unit. The communication unit can be a communication interface, transceiver, transceiver circuit, etc.

[0101] In this embodiment, the acquisition circuit 601, extraction circuit 602, fusion circuit 603, and determination circuit 604 can be integrated into the processing unit.

[0102] It should be noted that the processing unit can be a processor or controller, such as a baseband processor, baseband chip, central processing unit (CPU), general-purpose processor, digital signal processor (DSP), application-specific integrated circuit (ASIC), field-programmable gate array (FPGA), or other programmable logic device, transistor logic device, hardware component, or any combination thereof. It can implement or execute the various exemplary logic blocks, modules, and circuits described in conjunction with the disclosure of this application. The processing unit can also be a combination that implements computational functions, such as including one or more microprocessor combinations, a combination of a DSP and a microprocessor, etc.

[0103] In this embodiment, the etching endpoint determination device 600 is used to perform any of the steps performed by network devices / chips / chip modules, etc., as described in the above method embodiments.

[0104] In specific implementation, the acquisition circuit 601, extraction circuit 602, fusion circuit 603, and determination circuit 604 are used to perform any step in the above method implementation, and when performing actions such as sending, other units can be selectively invoked to complete the corresponding operation. A detailed description follows.

[0105] Acquisition circuit 601 is used to acquire first spectral data for interferometric measurement and second spectral data for optical emission spectroscopy measurement during the etching process. Acquisition circuit 601 includes, for example, an IEP spectrometer and an OES spectrometer.

[0106] The extraction circuit 602 is used to extract features from the first spectral data and the second spectral data respectively to obtain a first feature of the first spectral data and a second feature of the second spectral data. The first feature includes an observed value of the film thickness, and the second feature includes one or more of the observed values ​​of plasma concentration and the observed values ​​of the intensity of characteristic peaks in the plasma emission spectrum.

[0107] The fusion circuit 603 is used to perform feature fusion on the first feature and the second feature based on the etching rate to obtain the fused feature of the etching process. The fused feature includes one or more of the following: an estimated value of the etching rate, an estimated value of the film thickness, an estimated value of the intensity of the characteristic peak in the plasma emission spectrum, and an estimated value of the plasma concentration.

[0108] Determining circuit 604 is used to determine the endpoint of the etching process based on the fusion features. Determining circuit 604 includes, for example, a PC host computer.

[0109] In this embodiment, regarding the feature fusion of the first feature and the second feature based on the etching rate to obtain the fused feature of the etching process, the fusion circuit 603 is specifically used for: Establish a first relationship between film thickness and etching rate, and a second relationship between plasma concentration or the intensity of characteristic peaks in plasma emission spectrum and etching rate; The first feature and the second feature are used as observations for Kalman filtering, and the fused feature is used as the state variable for Kalman filtering. The fused feature is obtained by fusing the first feature and the second feature based on the first relationship and the second relationship using the Kalman filtering algorithm.

[0110] In this embodiment, regarding the use of the Kalman filter algorithm to fuse the first feature and the second feature based on the first relationship and the second relationship to obtain the fused feature, the fusion circuit 603 is specifically used for: The fused feature is obtained by fusing the first feature and the second feature based on the Kalman gain matrix, the first relation, and the second relation using the Kalman filtering algorithm. The Kalman gain matrix reflects the signal-to-noise ratio of the interferometric measurement and the optical emission spectrum measurement.

[0111] In this embodiment, when the noise of the interferometric measurement increases, the weight corresponding to the first feature in the Kalman gain matrix decreases; and / or When the noise in the optical emission spectrum measurement increases, the weight corresponding to the second feature in the Kalman gain matrix decreases.

[0112] In this embodiment, regarding determining the endpoint of the etching process based on the fusion features, the determining circuit 604 is specifically configured to: The endpoint of the etching process is determined based on one or more of the following: The estimated etching rate is less than a first threshold; The estimated thickness of the film layer is less than the second threshold; The change in the estimated thickness of the film layer is less than the third threshold. The estimated intensity of the characteristic peak in the plasma emission spectrum is greater than the fourth threshold or less than the fifth threshold. The estimated change in the intensity of the characteristic peaks in the plasma emission spectrum is greater than the sixth threshold. The estimated value of the plasma concentration is greater than the seventh threshold or less than the eighth threshold; The change in the estimated plasma concentration is greater than the ninth threshold.

[0113] In this embodiment, any number of heterogeneous deployments are made in the acquisition circuit 601, extraction circuit 602, fusion circuit 603, and determination circuit 604.

[0114] It should be noted that, Figure 6 The specific implementation of each operation in the implementation method can be found in the description of the method implementation method shown above, and will not be repeated here.

[0115] See Figure 7 , Figure 7 This is a schematic diagram of another etching endpoint determination device proposed in an embodiment of this application. The etching endpoint determination device 700 may include a processor 710, a memory 720, and a communication bus for connecting the processor 710 and the memory 720.

[0116] Optionally, the memory 720 may include, but is not limited to, random access memory (RAM), read-only memory (ROM), erasable programmable read-only memory (EPROM), or compact disc read-only memory (CD-ROM). The memory 720 is used to store the program code executed by the etching endpoint determination device 700 and the data transmitted.

[0117] In this embodiment, the etching endpoint determination device 700 may include an IEP spectrometer and an OES spectrometer to perform IEP measurement and OES measurement.

[0118] In this embodiment, the etching endpoint determination device 700 also includes a communication interface for receiving and sending data.

[0119] In this embodiment, the etching endpoint determination device 700 can be the terminal device, network device, reader, or A-IoT device described above.

[0120] In this embodiment, the processor 710 can be one or more CPUs. If the processor 710 is a CPU, the CPU can be a single-core CPU or a multi-core CPU.

[0121] In this embodiment, the processor 710 can be a baseband chip, a chip, a CPU, a general-purpose processor, a DSP, an ASIC, an FPGA, or other programmable logic devices, transistor logic devices, hardware components, or any combination thereof.

[0122] In a specific implementation, the processor 710 in the etching endpoint determination device 700 executes the computer program or instructions 721 stored in the memory 720 to perform the following operations: Acquire the first spectral data for interferometry and the second spectral data for optical emission spectroscopy during the etching process; Feature extraction is performed on the first spectral data and the second spectral data respectively to obtain a first feature of the first spectral data and a second feature of the second spectral data. The first feature includes the observed value of the film thickness, and the second feature includes one or more of the observed value of the plasma concentration and the observed value of the intensity of the characteristic peak in the plasma emission spectrum. Based on the etching rate, the first feature and the second feature are fused to obtain the fused feature of the etching process. The fused feature includes one or more of the following: an estimated value of the etching rate, an estimated value of the film thickness, an estimated value of the intensity of the characteristic peak in the plasma emission spectrum, and an estimated value of the plasma concentration. The endpoint of the etching process is determined based on the fusion characteristics.

[0123] It should be noted that, Figure 7 The specific implementation of each operation in the above-described method implementation can be found in the description of the method implementation, and will not be repeated here.

[0124] This application also provides a spectrometer, wherein the spectrometer includes any of the etching endpoint determination devices described in the above-described device embodiments.

[0125] As one implementation method, a spectrometer is, for example, Figure 4 The custom spectrometer shown can include IEP and OES spectrometers for interferometric and optical emission spectroscopy measurements, transmitting the measurement data to a processing chip (e.g., a dedicated ASIC processing chip) in the form of analog signals. The processing chip may include, for example, a feature extraction unit and a feature fusion unit.

[0126] This application also provides a semiconductor device, wherein the semiconductor device includes any of the etching endpoint determination devices described in the above-described device embodiments.

[0127] This application also provides a computer storage medium storing a computer program for electronic data interchange, which causes a computer to perform some or all of the steps of any of the methods described in the above method embodiments, wherein the computer includes an electronic device.

[0128] This application also provides a computer program product, which includes a non-transitory computer-readable storage medium storing a computer program operable to cause a computer to perform some or all of the steps of any of the methods described in the above method embodiments. The computer program product may be a software installation package, and the computer may include an electronic device.

[0129] It should be noted that, for the sake of simplicity, the various embodiments described above are all presented as a series of actions. Those skilled in the art should understand that this application is not limited by the described order of actions, as some steps in the embodiments of this application can be performed in other orders or simultaneously. Furthermore, those skilled in the art should also understand that the embodiments described in the specification are preferred embodiments, and the actions, steps, modules, or units involved are not necessarily essential to the embodiments of this application.

[0130] In the above embodiments, the descriptions of each embodiment in this application have different focuses. For parts not described in detail in a certain embodiment, please refer to the relevant descriptions of other embodiments.

[0131] The steps of the methods or algorithms described in this application can be implemented in hardware or by a processor executing software instructions. The software instructions can consist of corresponding software modules, which can be stored in RAM, flash memory, ROM, EPROM, electrically erasable programmable read-only memory (EEPROM), registers, hard disk, portable hard disk, read-only optical disk (CD-ROM), or any other form of storage medium known in the art. An exemplary storage medium is coupled to a processor, enabling the processor to read information from and write information to the storage medium. Of course, the storage medium can also be a component of the processor. The processor and storage medium can reside in an ASIC. Furthermore, the ASIC can reside in a terminal device or management device. Alternatively, the processor and storage medium can exist as discrete components in the terminal device or management device.

[0132] Those skilled in the art will recognize that, in one or more of the examples above, the functions described in the embodiments of this application can be implemented, in whole or in part, by software, hardware, firmware, or any combination thereof. When implemented in software, it can be implemented, in whole or in part, as a computer program product. This computer program product includes one or more computer instructions. When these computer program instructions are loaded and executed on a computer, all or part of the processes or functions described in the embodiments of this application are generated. The computer can be a general-purpose computer, a special-purpose computer, a computer network, or other programmable device. The computer instructions can be stored in a computer-readable storage medium or transmitted from one computer-readable storage medium to another. For example, the computer instructions can be transmitted from one website, computer, server, or data center to another via wired (e.g., coaxial cable, fiber optic, digital subscriber line (DSL)) or wireless (e.g., infrared, wireless, microwave, etc.) means. The computer-readable storage medium can be any available medium accessible to a computer or a data storage device such as a server or data center that integrates one or more available media. The available media can be magnetic media (e.g., floppy disks, hard disks, magnetic tapes), optical media (e.g., digital video discs (DVDs)), or semiconductor media (e.g., solid-state disks (SSDs)).

[0133] The modules / units included in the various devices and products described in the above embodiments can be software modules / units, hardware modules / units, or a combination of both. For example, for devices and products applied to or integrated into a chip, all modules / units can be implemented using hardware methods such as circuits, or at least some modules / units can be implemented using software programs that run on a processor integrated within the chip, while the remaining (if any) modules / units can be implemented using hardware methods such as circuits. For devices and products applied to or integrated into a chip module, all modules / units can be implemented using hardware methods such as circuits. Different modules / units can be located in the same component (e.g., chip, circuit module, etc.) or different components of the chip module, or at least some modules / units can be implemented using hardware methods such as circuits. The implementation is achieved through a software program that runs on the processor integrated within the chip module. The remaining modules / units (if any) can be implemented using hardware methods such as circuits. For various devices and products applied to or integrated into terminal equipment, each of their modules / units can be implemented using hardware methods such as circuits. Different modules / units can be located in the same component (e.g., chip, circuit module, etc.) or different components within the terminal equipment. Alternatively, at least some modules / units can be implemented through a software program that runs on the processor integrated within the terminal equipment, while the remaining modules / units (if any) can be implemented using hardware methods such as circuits.

[0134] The specific embodiments described above further illustrate the purpose, technical solution, and beneficial effects of the embodiments of this application. It should be understood that the above description is only a specific embodiment of the embodiments of this application and is not intended to limit the protection scope of the embodiments of this application. Any modifications, equivalent substitutions, improvements, etc., made on the basis of the technical solution of the embodiments of this application should be included within the protection scope of the embodiments of this application.

Claims

1. A method for etch endpoint determination, the method comprising: The method comprises: acquiring first spectral data for interferometric measurement and second spectral data for optical emission spectral measurement during an etching process; performing feature extraction on the first spectral data and the second spectral data respectively to obtain first features of the first spectral data and second features of the second spectral data, the first features comprising observation values of film thickness, and the second features comprising one or more of observation values of plasma concentration and observation values of intensities of characteristic peaks in plasma emission spectrum; performing feature fusion on the first features and the second features based on etching rate to obtain fusion features of the etching process, the fusion features comprising one or more of estimated values of etching rate, estimated values of film thickness, estimated values of intensities of characteristic peaks in plasma emission spectrum, and estimated values of plasma concentration; determining an endpoint of the etching process based on the fusion features.

2. The method of claim 1, wherein, The feature fusion on the first features and the second features based on etching rate to obtain fusion features of the etching process comprises: establishing a first relationship between film thickness and the etching rate, and a second relationship between plasma concentration or intensities of characteristic peaks in plasma emission spectrum and the etching rate; using the first features and the second features as observation quantities of Kalman filtering, and using the fusion features as state quantities of Kalman filtering, and performing feature fusion on the first features and the second features based on the first relationship and the second relationship by using a Kalman filtering algorithm to obtain the fusion features.

3. The method of claim 2, wherein, The feature fusion on the first features and the second features based on the first relationship and the second relationship by using the Kalman filtering algorithm to obtain the fusion features comprises: performing feature fusion on the first features and the second features based on a Kalman gain matrix, the first relationship and the second relationship by using the Kalman filtering algorithm to obtain the fusion features, the Kalman gain matrix reflecting signal-to-noise ratios of the interferometric measurement and the optical emission spectral measurement.

4. The method of claim 3, wherein, in a case where noise of the interferometric measurement increases, a weight corresponding to the first features in the Kalman gain matrix decreases; and / or in a case where noise of the optical emission spectral measurement increases, a weight corresponding to the second features in the Kalman gain matrix decreases.

5. The method according to any one of claims 1 to 4, characterized in that, The determination of the endpoint of the etching process based on the fusion features comprises: determining the endpoint of the etching process based on one or more of the following: the estimated value of the etching rate is less than a first threshold value; the estimated value of the film thickness is less than a second threshold value; a variation of the estimated value of the film thickness is less than a third threshold value; the estimated value of the intensities of characteristic peaks in plasma emission spectrum is greater than a fourth threshold value, or less than a fifth threshold value; a variation of the estimated value of the intensities of characteristic peaks in plasma emission spectrum is greater than a sixth threshold value; the estimated value of the plasma concentration is greater than a seventh threshold value, or less than an eighth threshold value; a variation of the estimated value of the plasma concentration is greater than a ninth threshold value.

6. An etch endpoint determination apparatus, comprising: The device comprises: acquisition circuitry configured to acquire first spectral data for interferometric measurement and second spectral data for optical emission spectral measurement during an etching process; extraction circuitry configured to extract features from the first spectral data and the second spectral data respectively, to obtain first features of the first spectral data and second features of the second spectral data, the first features comprising an observation value of a film thickness, and the second features comprising one or more of an observation value of a plasma concentration and an observation value of intensity of a characteristic peak in a plasma emission spectrum; fusion circuitry configured to fuse the first features and the second features based on an etching rate, to obtain fused features of the etching process, the fused features comprising one or more of an estimated value of the etching rate, an estimated value of the film thickness, an estimated value of intensity of the characteristic peak in the plasma emission spectrum, and an estimated value of the plasma concentration; determination circuitry configured to determine an endpoint of the etching process based on the fused features.

7. The apparatus of claim 6, wherein any multiple of the acquisition circuitry, the extraction circuitry, the fusion circuitry, and the determination circuitry are heterogeneously deployed.

8. An etch endpoint determination apparatus, comprising: comprising: a processor, and a memory communicatively connected to the processor; the memory stores computer-executable instructions; the processor executes the computer-executable instructions stored in the memory to implement the method of any one of claims 1-5.

9. A spectrometer, characterized by, the apparatus of any one of claims 6-8.

10. A semiconductor device, characterized by comprising: the semiconductor device comprises the apparatus of any one of claims 6-8.

11. A computer readable storage medium characterized by, the computer-readable storage medium stores a computer program, which is executed by the processor to implement the method of any one of claims 1-5. the computer-readable storage medium stores a computer program, which is executed by the processor to implement the method of any one of claims 1-5.

Images