Semiconductor etching monitoring method, device, equipment and storage medium

Abstract

The embodiments of the present invention relate to the field of semiconductor etching and disclose a semiconductor etching monitoring method, apparatus, device and storage medium, including: determining the spectral vector of each pixel point in the spectral image data of a wafer; selecting a pixel point of the etching layer to be determined as a first reference pixel point, and determining the spectral vector corresponding to the first reference pixel point as a first reference vector; performing a similarity comparison between the spectral vector of each pixel point in the spectral image data and the first reference vector to determine the distribution area of ​​the etching layer and the termination layer in the spectral image data; for the distribution area of ​​the termination layer, selecting a spectral band range that meets preset conditions from multiple spectral bands to be determined as the characteristic spectrum segment of the termination layer; based on the characteristic spectrum segment of the termination layer, monitoring the etching endpoint of the wafer to be etched. This application solves the problem of inaccurate monitoring of the etching endpoint in actual processes, monitors the etching endpoint according to the characteristic spectrum segment of the termination layer, and improves the accuracy of monitoring the etching endpoint.

Description

Technical Field

[0001] Embodiments of the present invention relate to the field of semiconductor technology, and in particular to a semiconductor etching monitoring method, device, equipment, and storage medium. Background Art

[0002] In the semiconductor manufacturing process, a common method for forming patterns on the wafer surface is to perform a photolithography + etching process, in which the photolithography process "prints" the required pattern from the mask plate to the photoresist layer, and then the etching process transfers it to the wafer surface.

[0003] For semiconductor etching processes, "aspect ratio" is a critical process parameter. There are two common approaches to controlling etch depth: one is to calculate the film thickness and etch rate, convert them into theoretical etch time, and then adjust them multiple times to achieve the optimal main process time. The other approach is to set a film endpoint and monitor it during the actual process. When the product is etched to that position, the process stops.

[0004] However, in actual processes, due to problems such as chip complexity, changes in film material or machine contamination, it is easy to cause inaccurate monitoring of the etching endpoint, and even "over-etching" may occur, resulting in product scrapping. Summary of the Invention

[0005] The purpose of the present invention is to at least provide a semiconductor etching monitoring method, device, equipment and storage medium, which can at least solve the problems in actual processes such as chip complexity, film material changes or machine contamination, which can easily cause inaccurate etching endpoint monitoring and even "over-etching" leading to product scrapping. At least it can achieve the monitoring of the etching endpoint based on the characteristic spectrum of the termination layer, thereby improving the accuracy of etching endpoint monitoring.

[0006] To solve the above technical problems, at least one embodiment of the present application provides a semiconductor etching monitoring method, comprising: acquiring spectral image data of a wafer after etching, wherein the spectral image data includes image data of an etched layer and a stop layer in the wafer under multiple spectral bands;

[0007] Determining a spectral vector of each pixel in the spectral image data of the wafer;

[0008] Selecting a pixel point of the etching layer as a first reference pixel point, and determining a spectral vector corresponding to the first reference pixel point as a first reference vector;

[0009] performing a similarity comparison between the spectral vector of each pixel point in the spectral image data and the first reference vector to determine the distribution areas of the etching layer and the termination layer in the spectral image data;

[0010] With respect to the distribution area of ​​the termination layer, a spectral band range that meets a preset condition is selected from a plurality of spectral bands, and the spectral band range that meets the preset condition is determined as a characteristic spectral band of the termination layer;

[0011] Based on the characteristic spectrum of the stop layer, the etching endpoint of the wafer to be etched is monitored.

[0012] At least one embodiment of the present application further provides a semiconductor etching monitoring device, comprising: an image acquisition component, an imaging component, a filtering component, and a control unit;

[0013] The control unit is used to obtain spectral image data of the etched wafer through an image acquisition component, an imaging component and a filtering component, wherein the spectral image data includes image data of the etching layer and the termination layer in the wafer under multiple spectral bands; determine the spectral vector of each pixel point in the spectral image data of the wafer; select a pixel point of the etching layer as a first reference pixel point, and determine the spectral vector corresponding to the first reference pixel point as a first reference vector; compare the spectral vector of each pixel point in the spectral image data with the first reference vector for similarity, and determine the distribution area of ​​the etching layer and the termination layer in the spectral image data; for the distribution area of ​​the termination layer, select a spectral band range that meets preset conditions from multiple spectral bands, and determine the spectral band range that meets the preset conditions as the characteristic spectral band of the termination layer; based on the characteristic spectral band of the termination layer, monitor the etching end point of the wafer to be etched.

[0014] At least one embodiment of the present application further provides a semiconductor etching monitoring device, comprising an acquisition module for acquiring spectral image data of a wafer after etching, wherein the spectral image data comprises image data of an etched layer and a stop layer in the wafer under multiple spectral bands;

[0015] a vector determination module, configured to determine a spectral vector of each pixel in the spectral image data of the wafer;

[0016] a reference selection module, configured to select a pixel point of the etching layer as a first reference pixel point, and determine a spectral vector corresponding to the first reference pixel point as a first reference vector;

[0017] a comparison module, configured to compare the spectral vector of each pixel point in the spectral image data with the first reference vector for similarity, and determine the distribution areas of the etching layer and the termination layer in the spectral image data;

[0018] a spectrum band determination module, configured to select a spectrum band range that meets preset conditions from a plurality of spectrum bands with respect to a distribution area of ​​the termination layer, and determine the spectrum band range that meets the preset conditions as a characteristic spectrum band of the termination layer;

[0019] The monitoring module is used to monitor the etching endpoint of the wafer to be etched based on the characteristic spectrum of the stop layer.

[0020] At least one embodiment of the present application also provides an electronic device, comprising: at least one processor; and a memory communicatively connected to the at least one processor; wherein the memory stores instructions executable by the at least one processor, and the instructions are executed by the at least one processor to enable the at least one processor to execute the above-mentioned semiconductor etching monitoring method.

[0021] At least one embodiment of the present application further provides a computer-readable storage medium storing a computer program, wherein the computer program implements the above-mentioned semiconductor etching monitoring method when executed by a processor.

[0022] The semiconductor etching monitoring method, device, equipment and storage medium provided by the embodiments of the present application, after obtaining the spectral image data of the etching layer and the termination layer in the wafer after etching, determines the spectral vector of each pixel point in the spectral image data of the wafer, selects a pixel point of the etching layer as the first reference pixel point, and determines the spectral vector corresponding to the first reference pixel point as the first reference vector, and then compares the spectral vector of each pixel point in the spectral image data with the first reference vector for similarity to determine the distribution area of ​​the etching layer and the termination layer in the spectral image data. Then, for the distribution area of ​​the termination layer, a spectral band range that meets the preset conditions is selected from multiple spectral bands, and the spectral band range that meets the preset conditions is determined as the characteristic spectral band of the termination layer. Then, based on the characteristic spectral band of the termination layer, the etching end point of the wafer to be etched is monitored. In this way, by acquiring the characteristic spectrum of the stop layer, the etching endpoint of the wafer is monitored in a targeted manner based on the characteristic spectrum of the stop layer, effectively reducing the probability of inaccurate etching endpoint monitoring or even "over-etching" leading to product scrapping due to problems such as chip complexity, film material changes or machine contamination in actual processes.

[0023] In some optional embodiments, the step of determining the spectral vector of each pixel in the spectral image data of the wafer includes: determining the grayscale value of the pixel in the etching layer under each spectral band as the spectral intensity of the pixel in the etching layer; determining the spectral intensity of the pixel in the etching layer under multiple spectral bands as the spectral vector of each pixel in the etching layer; determining the grayscale value of the pixel in the termination layer under each spectral band as the spectral intensity of the pixel in the termination layer; and determining the spectral intensity of the pixel in the termination layer under multiple spectral bands as the spectral vector of each pixel in the termination layer. By determining the spectral vectors of the etching layer and the termination layer, it is convenient to monitor the etching endpoint of the wafer to be etched after subsequent data processing.

[0024] In some optional embodiments, the step of performing a similarity comparison between the spectral vector of each pixel point in the spectral image data and the first reference vector to determine the distribution area of ​​the etching layer and the termination layer in the spectral image data includes: performing a similarity comparison between the spectral vector of each pixel point in the spectral image data and the first reference vector to determine a similarity comparison value; detecting whether the similarity comparison value is within a preset matching range; when the similarity comparison value is within the preset matching range, determining that the pixel points with the similarity comparison value within the preset matching range are pixel points of the etching layer, and the remaining pixel points are pixel points of the termination layer. By performing a similarity comparison between the spectral vector of each pixel point in the spectral image data and the first reference vector, the film layer where the pixel point is located is determined, thereby distinguishing the distribution area of ​​the etching layer and the termination layer.

[0025] In some optional embodiments, the step of selecting a spectral band range that meets preset conditions from a plurality of spectral bands for the distribution area of ​​the termination layer, and determining the spectral band range that meets the preset conditions as the characteristic spectral band of the termination layer includes: calculating the pixel points in the distribution area of ​​the termination layer in each spectral band to determine the average value of the termination spectrum intensity of each spectral band; detecting whether the average value of the termination spectrum intensity is greater than a preset light intensity threshold; when the average value of the termination spectrum intensity is greater than the preset light intensity threshold, selecting a spectral band whose average value of the termination spectrum intensity is greater than the preset light intensity threshold from the plurality of spectral bands, thereby obtaining the spectral band range of the termination layer; and determining the spectral band range of the termination layer as the characteristic spectral band of the termination layer. By obtaining the characteristic spectral band of the termination layer, the etching endpoint of the wafer is monitored in a targeted manner according to the characteristic spectral band of the termination layer, effectively reducing the probability of inaccurate etching endpoint monitoring or even "over-etching" leading to product scrapping due to problems such as chip complexity, film material changes or machine contamination in actual processes.

[0026] In some optional embodiments, the method further includes: selecting a spectral band range that meets preset conditions from a plurality of spectral bands for the distribution area of ​​the etching layer, and determining the spectral band range that meets the preset conditions as the characteristic spectral band of the etching layer; the step of monitoring the etching endpoint of the wafer to be etched based on the characteristic spectral band of the termination layer includes: monitoring the etching endpoint of the wafer to be etched based on the characteristic spectral band of the termination layer and the characteristic spectral band of the etching layer. After determining the characteristic spectral band of the etching layer and the characteristic spectral band of the termination layer, the characteristic spectral band of the etching layer and the characteristic spectral band of the termination layer can be monitored in a targeted manner, which can further improve the accuracy of monitoring the etching endpoint of the wafer to be etched.

[0027] In some optional embodiments, the etching layer includes an etched area and a reserved area; the method further includes: selecting a pixel point in the etched area in the etched layer as a second reference pixel point, and determining the spectral vector corresponding to the second reference pixel point as a second reference vector; performing a similarity comparison between the spectral vector of each pixel point in the spectral image data and the second reference vector to determine the distribution area of ​​the etched area, the reserved area, and the termination layer in the spectral image data. After determining the characteristic spectrum of the etched area, the characteristic spectrum of the reserved area, and the characteristic spectrum of the termination layer, the characteristic spectrum of the etched area, the characteristic spectrum of the reserved area, and the characteristic spectrum of the termination layer can be monitored in a targeted manner, which can not only improve the etching accuracy when etching the etched area, but also further improve the accuracy of monitoring the etching endpoint of the wafer to be etched. BRIEF DESCRIPTION OF THE DRAWINGS

[0028] One or more embodiments are exemplarily described by the figures in the corresponding drawings, and these exemplified descriptions do not constitute limitations on the embodiments.

[0029] Figure 1 This is a flow chart of a semiconductor etching monitoring method provided by an embodiment of the present application;

[0030] Figure 2 is a structural block diagram of a semiconductor etching monitoring device in one embodiment of the present application;

[0031] Figure 3 is a structural diagram of an electronic device provided by another embodiment of the present application;

[0032] Figure 4 is a flow chart of a semiconductor etching monitoring method in another embodiment of the present application;

[0033] Figure 5 is a schematic diagram of an etching layer and a stop layer provided in one embodiment of the present application;

[0034] Figure 6is a structural diagram of a semiconductor etching monitoring device provided by another embodiment of the present application;

[0035] Figure 7 This is a schematic diagram of a filter component in a semiconductor etching monitoring device provided in another embodiment of the present application. DETAILED DESCRIPTION

[0036] In order to make the purpose, technical solutions and advantages of the embodiments of the present application clearer, each embodiment of the present application will be described in detail below with reference to the accompanying drawings. However, it will be understood by those skilled in the art that in each embodiment of the present application, many technical details are proposed to enable the reader to better understand the present application. However, even without these technical details and various changes and modifications based on the following embodiments, the technical solutions claimed in the present application can be implemented. The division of the following embodiments is for convenience of description and should not constitute any limitation on the specific implementation of the present application. The various embodiments can be combined and referenced with each other under the premise of no contradiction.

[0037] To facilitate understanding of the embodiments of the present application, relevant content regarding the semiconductor etching monitoring method is first introduced here.

[0038] In the semiconductor manufacturing process, a common method for forming patterns on the wafer surface is to perform a photolithography + etching process, in which the photolithography process "prints" the required pattern from the mask plate to the photoresist layer, and then the etching process transfers it to the wafer surface.

[0039] For semiconductor etching processes, "aspect ratio" is a critical process parameter. There are two common approaches to controlling etch depth: one is to calculate the film thickness and etch rate, convert them into theoretical etch time, and then adjust them multiple times to achieve the optimal main process time. The other approach is to set a film endpoint and monitor it during the actual process. When the product is etched to that position, the process stops.

[0040] However, in actual processes, due to problems such as chip complexity, changes in film material or machine contamination, it is easy to cause inaccurate monitoring of the etching endpoint, and even "over-etching" may occur, resulting in product scrapping.

[0041] In order to solve the above-mentioned technical problems in actual processes, inaccurate etching endpoint monitoring may be easily caused due to problems such as chip complexity, film material changes or machine contamination, and even "over-etching" may occur, resulting in product scrapping. The present invention proposes a semiconductor etching monitoring method. The implementation details of the semiconductor etching monitoring method of this embodiment are specifically described below. The following content is only the implementation details provided for easy understanding and is not necessary for the implementation of this solution.

[0042] Example 1:

[0043] The semiconductor etching monitoring method of this embodiment can be applied to electronic devices with communication, computing and data storage capabilities. The specific process can be as follows: Figure 1 As shown, including:

[0044] Step 101 : Acquire spectral image data of a wafer after etching, wherein the spectral image data includes image data of an etched layer and a stop layer in the wafer under multiple spectral bands.

[0045] Specifically, a microscopic hyperspectral camera captures spectral image data of the etched wafer as a sample, obtaining spectral image data of the etched layer and the stop layer within the wafer. A spectral image is a set of images acquired by capturing the wafer at different spectral bands. Spectral image data not only reveals the spatial relationships between objects within a plane, but also shows how object details vary with light wavelength, facilitating material differentiation.

[0046] Step 102 : determining a spectral vector of each pixel in the spectral image data of the wafer.

[0047] Specifically, during semiconductor manufacturing, the materials of the etching layer and mask are typically fixed. The etching layer is made of materials such as metal, SiO2, Si3N4, etc., the mask is typically made of photoresist, and the stop layer is made of a different material than the etching layer. These objects will form an emission spectrum after being illuminated by a light source or after etching glow discharge. In this embodiment, for an object of a certain material, any pixel of that material in the spectral image data is selected, and the spectral vector of that pixel is obtained in multiple spectral bands. The above operation is repeated to obtain the spectral vector of each pixel in the spectral image data of the wafer.

[0048] Step 103 : Select a pixel point of the etching layer as a first reference pixel point, and determine a spectral vector corresponding to the first reference pixel point as a first reference vector.

[0049] Specifically, a pixel of the etching layer is selected from the spectral image data and determined as the first reference pixel. Furthermore, pixels with clear details and features are preferentially selected in the selection of the first reference pixel. After the first reference pixel is determined, the spectral vector corresponding to the first reference pixel is determined as the first reference vector. It should be noted that since the spectral vector of each pixel in the spectral image data of the wafer has been obtained in step 102, in this embodiment, the spectral vector corresponding to the first reference pixel can be directly obtained and determined as the first reference vector.

[0050] Step 104 : performing a similarity comparison between the spectral vector of each pixel point in the spectral image data and the first reference vector to determine the distribution areas of the etching layer and the stop layer in the spectral image data.

[0051] Specifically, the first reference vector is the reference vector of the first reference pixel point of the etching layer. When the spectral vector of each pixel point in the spectral image data is compared with the first reference vector for similarity, that is, when comparing which pixels in the spectral image data belong to the pixels of the etching layer, after determining the pixels belonging to the etching layer, the pixels other than the pixels of the etching layer are the pixels of the termination layer, thereby distinguishing the distribution areas of the etching layer and the termination layer.

[0052] In some examples, the first reference vector is denoted as X. The first reference vectors in different spectral bands form a vector subset. When the total number of spectral bands is n, the vector subset has n elements. The similarity between the spectral vector of each pixel in the spectral image data and the first reference vector is compared using the following relationship:

[0053]

[0054] Where X is the first reference vector, Y is the spectral vector of a pixel in the spectral image data, X i is the vector subset of the first reference vector, Y i is a vector subset of the spectral vector of a certain pixel point in the spectral image data, and D is a similarity comparison value obtained by comparing the spectral vector of each pixel point in the spectral image data with the first reference vector.

[0055] In some examples, when the material of the etching layer is SiO2, a pixel point of the etching layer is selected as the first reference pixel point. The first reference pixel point is a reference pixel point selected in the etching layer whose material is SiO2. By obtaining the spectral vector of the first reference pixel point under multiple spectral bands and comparing the spectral vector of each pixel point in the spectral image data with the first reference vector for similarity, it is possible to determine which pixel points in the spectral image data belong to SiO2, that is, which pixel points belong to the pixel points of the etching layer.

[0056] Step 105 : With respect to the distribution area of ​​the termination layer, a spectral band range that meets preset conditions is selected from a plurality of spectral bands, and the spectral band range that meets the preset conditions is determined as a characteristic spectrum band of the termination layer.

[0057] Specifically, after determining the distribution areas of the etching layer and the termination layer, preset conditions are set. Based on the distribution area of ​​the termination layer, a spectral band that meets the preset conditions is selected from multiple spectral bands, and the spectral band that meets the preset conditions is used as the spectral band range, and then the spectral band range is used as the characteristic spectral band of the termination layer. It should be noted that the spectral band range includes one spectral band and may also include multiple spectral bands. When the spectral band range includes one spectral band, the spectral band is the spectral band range of the termination layer. When the spectral band range includes multiple spectral bands, the multiple spectral bands are integrated, and the integrated spectral band is used as the spectral band range of the termination layer.

[0058] Step 106 : Monitoring the etching endpoint of the wafer to be etched based on the characteristic spectrum of the stop layer.

[0059] Specifically, after determining the characteristic spectrum of the stop layer, the characteristic spectrum of the stop layer can be monitored in a targeted manner. Thus, when etching the wafer to be etched, the etching endpoint of the wafer to be etched can be monitored based on the characteristic spectrum of the stop layer. Compared with existing etching endpoint identification methods, this method not only reduces the amount of computational data required for etching endpoint monitoring, but also effectively reduces the probability of inaccurate etching endpoint monitoring, which can easily occur in actual processes due to problems such as chip complexity, changes in film material, or machine contamination, and even "over-etching" leading to product scrapping.

[0060] In some examples, the photolithography pattern can be used as a reference image and compared with the spectral image under the characteristic spectrum of the termination layer to further determine the location of the etching residue and improve the accuracy of etching endpoint monitoring.

[0061] In some examples, after determining the characteristic spectrum of a stop layer of a certain material, the characteristic spectrum of the stop layer of that material can be organized and summarized for direct reuse in subsequent etching endpoint monitoring processes of the same type of wafer, thereby improving data utilization. If the material of the stop layer of the wafer changes, steps 101 to 106 need to be repeated to obtain the characteristic spectrum of the stop layer after the material change.

[0062] In this embodiment, after obtaining the spectral image data of the etching layer and the stop layer in the wafer after etching, the spectral vector of each pixel in the spectral image data of the wafer is determined, a pixel of the etching layer is selected and determined as a first reference pixel, and the spectral vector corresponding to the first reference pixel is determined as a first reference vector. Then, the spectral vector of each pixel in the spectral image data is compared with the first reference vector for similarity to determine the distribution area of ​​the etching layer and the stop layer in the spectral image data. Then, for the distribution area of ​​the stop layer, a spectral band range that meets the preset conditions is selected from multiple spectral bands, and the spectral band range that meets the preset conditions is determined as the characteristic spectrum of the stop layer. Then, based on the characteristic spectrum of the stop layer, the etching endpoint of the wafer to be etched is monitored. In this way, by acquiring the characteristic spectrum of the stop layer, the etching endpoint of the wafer is monitored according to the characteristic spectrum of the stop layer in a targeted manner, effectively reducing the probability of inaccurate etching endpoint monitoring caused by problems such as chip complexity, film material changes, or machine contamination in actual processes, or even "over-etching" leading to product scrapping.

[0063] In some embodiments, the step of determining the spectral vector of each pixel in the spectral image data of the wafer includes:

[0064] Determine the grayscale value of the pixel point in the etching layer under each spectral band as the spectral intensity of the pixel point in the etching layer;

[0065] Determine the spectral intensity of the pixel points in the etching layer under multiple spectral bands as the spectral vector of each pixel point in the etching layer;

[0066] The gray value of the pixel point in the termination layer under each spectral band is determined as the spectral intensity of the pixel point in the termination layer;

[0067] The spectral intensities of the pixels in the termination layer under multiple spectral bands are determined as the spectral vectors of each pixel in the termination layer.

[0068] Specifically, for a certain material object, any pixel of the material in the spectral image data is selected, and the grayscale value of the pixel in a certain spectral band is read as the spectral intensity of the pixel in the spectral band. Then, the grayscale value of the pixel in the next spectral band is read again as the spectral intensity of the pixel in the next spectral band. The above steps are repeated in the order of wavelengths to obtain the spectral vector of the pixel in multiple spectral bands. In this embodiment, for the etching layer, a pixel of the etching layer is selected in the spectral image data, and the grayscale value of the pixel in a certain spectral band is read as the spectral intensity of the pixel in the spectral band. Then, the grayscale value of the pixel in the next spectral band is read again as the spectral intensity of the pixel in the next spectral band. The above steps are repeated in the order of wavelengths to obtain the spectral vector of the pixel in the etching layer in multiple spectral bands. The spectral intensities of the pixels in the etching layer in the multiple spectral bands are then determined as the spectral vector of each pixel in the etching layer. For the termination layer, select the pixel point of the etching layer in the spectral image data, read the grayscale value of the pixel point in a certain spectral band as the spectral intensity of the pixel point in the spectral band, then switch to the next spectral band, and read the grayscale value of the pixel point in the next spectral band again as the spectral intensity of the pixel point in the next spectral band. Repeat the above steps in the order of wavelength to obtain the spectral vector of the pixel point in the termination layer in multiple spectral bands, and then determine the spectral intensity of the pixel point in the termination layer in multiple spectral bands as the spectral vector of each pixel point in the termination layer. By determining the spectral vectors of the etching layer and the termination layer, it is convenient to monitor the etching endpoint of the wafer to be etched after subsequent data processing. It should be noted that the determination of the spectral vectors of the etching layer and the termination layer can be carried out simultaneously.

[0069] In some embodiments, the step of comparing the spectral vector of each pixel point in the spectral image data with the first reference vector for similarity to determine the distribution area of ​​the etching layer and the termination layer in the spectral image data includes:

[0070] Comparing the spectral vector of each pixel in the spectral image data with the first reference vector for similarity, and determining a similarity comparison value;

[0071] Detect whether the similarity comparison value is within the preset matching range;

[0072] When the similarity comparison value is within the preset matching range, the pixels whose similarity comparison value is within the preset matching range are determined to be pixels of the etching layer, and the remaining pixels are pixels of the termination layer.

[0073] Specifically, the spectral vector of each pixel in the spectral image data is compared with the first reference vector for similarity through the relationship (1) to obtain a similarity comparison value. When the similarity comparison value corresponding to a certain pixel is high, it means that the pixel is similar to the first reference pixel. In this embodiment, it is detected whether the similarity comparison value is within a preset matching range. When the similarity comparison value is within the preset matching range, the pixel points whose similarity comparison values ​​are within the preset matching range are determined to be pixel points of the etching layer, and the remaining pixel points are pixel points of the termination layer, thereby distinguishing the distribution areas of the etching layer and the termination layer.

[0074] In some examples, the preset matching range is set to (0.85, 1). When the similarity comparison value obtained by comparing the spectral vector of a pixel point in the spectral image data with the first reference vector falls within (0.85, 1), it means that the pixel point is similar to the first reference pixel point and the pixel point belongs to the etched layer.

[0075] In some examples, when the material of the etching layer is SiO2, a pixel point of the etching layer is selected as the first reference pixel point. The first reference pixel point is a reference pixel point selected in the etching layer made of SiO2. The SiO2 curve is formed by obtaining the spectral vector of the first reference pixel point under multiple spectral bands, and the spectral vector of a pixel point in the spectral image data under multiple spectral bands forms a comparison curve. The SiO2 curve and the comparison curve are matched. When the similarity comparison value of the two falls into (0.85, 1), it indicates that both belong to SiO2, that is, the pixel point belongs to the pixel point of the etching layer.

[0076] In some embodiments, based on the distribution area of ​​the termination layer, the step of selecting a spectral band range that meets a preset condition from a plurality of spectral bands and determining the spectral band range that meets the preset condition as a characteristic spectral band of the termination layer includes:

[0077] Calculate the pixel points in the distribution area of ​​the termination layer of each spectral band to determine the average termination spectrum intensity of each spectral band;

[0078] Detect whether the average value of the termination spectrum intensity is greater than a preset light intensity threshold;

[0079] When the average value of the termination spectrum intensity is greater than the preset light intensity threshold, a spectrum band whose average value of the termination spectrum intensity is greater than the preset light intensity threshold is selected from the multiple spectrum bands, thereby obtaining the spectrum band range of the termination layer;

[0080] The spectral band range of the termination layer is determined as the characteristic spectrum band of the termination layer.

[0081] Specifically, after determining the distribution area of ​​the termination layer, the spectral intensity of each pixel point in the distribution area of ​​the termination layer under each spectral band is summed to obtain the sum of the regional spectral intensity values ​​under each spectral band. When calculating the average value of the termination spectrum intensity of each spectral band, it is calculated using the following relationship: average value of termination spectrum intensity = sum of regional spectral intensity values / number of regional pixels.

[0082] In this embodiment, after determining the average termination spectrum intensity of each spectral band, the average termination spectrum intensity is tested to see if it is greater than a preset light intensity threshold. When the average termination spectrum intensity is greater than the preset light intensity threshold, a spectral band whose average termination spectrum intensity is greater than the preset light intensity threshold is selected from the multiple spectral bands, thereby obtaining the spectral band range of the termination layer, and determining the spectral band range of the termination layer as the characteristic spectrum range of the termination layer. In this way, the etching endpoint of the wafer is monitored in a targeted manner based on the characteristic spectrum range of the termination layer, effectively reducing the probability of inaccurate etching endpoint monitoring, which can be caused by problems such as chip complexity, film material changes, or machine contamination in actual processes, and even the probability of "over-etching" leading to product scrapping.

[0083] In some examples, when the average intensity value of a region in a spectral image is greater than 0.75, the brightness and darkness transitions are significantly greater than in other regions. Therefore, the preset light intensity threshold is set to 0.75. When the average intensity of the termination spectrum is greater than 0.75, the spectral band with an average intensity greater than 0.75 is selected from the multiple spectral bands to obtain the spectral band range of the termination layer, and the spectral band range of the termination layer is determined as the characteristic spectral band of the termination layer.

[0084] In some embodiments, the semiconductor etching monitoring method further includes:

[0085] For the distribution area of ​​the etching layer, a spectral band range that meets the preset conditions is selected from multiple spectral bands, and the spectral band range that meets the preset conditions is determined as a characteristic spectral band of the etching layer;

[0086] The steps of monitoring the etching endpoint of the wafer to be etched based on the characteristic spectrum of the stop layer include:

[0087] Based on the characteristic spectrum of the stop layer and the characteristic spectrum of the etching layer, the etching endpoint of the wafer to be etched is monitored.

[0088] Specifically, after determining the distribution areas of the etching layer and the termination layer, the average etching spectrum intensity in the distribution area of ​​the termination layer under each spectral band is determined according to the distribution area of ​​the etching layer, and then it is detected whether the average etching spectrum intensity is greater than the preset light intensity threshold. When the average etching spectrum intensity is greater than the preset light intensity threshold, the spectral band whose average etching spectrum intensity is greater than the preset light intensity threshold is selected from multiple spectral bands, thereby obtaining the spectral band range of the etching layer, and the spectral band range of the etching layer is determined as the characteristic spectral band of the etching layer.

[0089] In this embodiment, after determining the characteristic spectrum of the etching layer and the characteristic spectrum of the stop layer, the characteristic spectrum of the etching layer and the characteristic spectrum of the stop layer can be monitored in a targeted manner, which can further improve the accuracy of etching endpoint monitoring of the wafer to be etched.

[0090] In some embodiments, the etching layer includes an etched area and a reserved area;

[0091] The method also includes:

[0092] A pixel point in the etched area of ​​the etched layer is selected as a second reference pixel point, and a spectral vector corresponding to the second reference pixel point is determined as a second reference vector;

[0093] The spectral vector of each pixel point in the spectral image data is compared with the second reference vector for similarity, and the distribution areas of the carved area, the reserved area and the termination layer in the spectral image data are determined.

[0094] Specifically, the etched layer consists of an engraved area and a reserved area. The engraved area is etched to a specified depth according to a specified shape. The surface of the reserved area is covered by a mask (mostly photoresist) to prevent the etched layer covered by the mask from being etched. Because the surface of the reserved area is covered by the mask, the spectral image shows that the reserved area and the engraved area are made of different materials. The reserved area appears to be made of the same material as the mask, such as photoresist, while the engraved area appears to be made of the same material as the etched layer itself, such as metal, SiO2, or Si3N4.

[0095] In this embodiment, a pixel in the cutout region of the etched layer is selected from the spectral image data and determined as a second reference pixel. After determining the second reference pixel, the spectral vector corresponding to the second reference pixel is determined as the second reference vector. It should be noted that since the spectral vector of each pixel in the spectral image data of the wafer has been obtained in step 102, in this embodiment, the spectral vector corresponding to the second reference pixel can be directly obtained and determined as the second reference vector. The second reference vector is the reference vector of the second reference pixel in the cutout region of the etched layer. When the spectral vector of each pixel in the spectral image data is compared with the second reference vector for similarity, that is, which pixels in the spectral image data belong to the cutout region of the etched layer are compared. After determining the pixels belonging to the cutout region of the etched layer, the pixels of the etched layer obtained by the similarity comparison based on the first reference vector are combined. The pixels in the etched layer other than the pixels in the cutout region are pixels in the reserved region, and the pixels other than the pixels in the etched layer are pixels in the stop layer, thereby distinguishing the distribution areas of the cutout region, the reserved region, and the stop layer.

[0096] In some examples, the similarity comparison value of the spectral vector of each pixel in the spectral image data and the second reference vector can also be calculated by using the relationship (1). At this time, in this embodiment, X in the relationship (1) is the second reference vector, Y is the spectral vector of a pixel in the spectral image data, and X i is the vector subset of the second reference vector, Y i is a vector subset of the spectral vector of a certain pixel point in the spectral image data, and D is a similarity comparison value obtained by comparing the spectral vector of each pixel point in the spectral image data with the second reference vector.

[0097] In some examples, when the material of the etching layer is SiO2, a pixel point in the engraved area of ​​the etching layer is selected as the second reference pixel point. The second reference pixel point is the reference pixel point selected in the engraved area whose material is SiO2. By obtaining the spectral vector of the second reference pixel point under multiple spectral bands and comparing the spectral vector of each pixel point in the spectral image data with the second reference vector for similarity, it is possible to determine which pixel points in the spectral image data belong to SiO2, that is, which pixel points belong to the pixel points in the engraved area.

[0098] In some examples, after determining the distribution areas of the engraved area, the reserved area, and the termination layer, the average etching spectrum intensity in the distribution area of ​​the engraved area under each spectral band is determined based on the distribution area of ​​the engraved area, and then the average etching spectrum intensity is detected to see if it is greater than a preset light intensity threshold. When the average etching spectrum intensity is greater than the preset light intensity threshold, a spectral band whose average etching spectrum intensity is greater than the preset light intensity threshold is selected from multiple spectral bands, thereby obtaining a spectral band range of the engraved area, and determining the spectral band range of the engraved area as a characteristic spectral band of the engraved area. Based on the distribution area of ​​the reserved area, the average etching spectrum intensity in the distribution area of ​​the reserved area under each spectral band is determined, and then the average etching spectrum intensity is detected to see if it is greater than a preset light intensity threshold. When the average etching spectrum intensity is greater than the preset light intensity threshold, a spectral band whose average etching spectrum intensity is greater than the preset light intensity threshold is selected from multiple spectral bands, thereby obtaining a spectral band range of the reserved area, and determining the spectral band range of the reserved area as a characteristic spectral band of the reserved area.

[0099] In this embodiment, after determining the characteristic spectrum of the etched area, the characteristic spectrum of the retained area, and the characteristic spectrum of the termination layer, the characteristic spectrum of the etched area, the characteristic spectrum of the retained area, and the characteristic spectrum of the termination layer can be monitored in a targeted manner, which can not only improve the etching accuracy when etching the etched area, but also further improve the accuracy of monitoring the etching endpoint of the wafer to be etched.

[0100] Example 2:

[0101] Based on the first embodiment, this embodiment further provides a semiconductor etching monitoring device, which is used in the semiconductor etching monitoring method of the first embodiment. The semiconductor etching monitoring device of this embodiment includes: an image acquisition component, an imaging component, a filtering component and a control unit;

[0102] The control unit is used to obtain spectral image data of the etched wafer through an image acquisition component, an imaging component and a filtering component, wherein the spectral image data includes image data of the etching layer and the termination layer in the wafer under multiple spectral bands; determine the spectral vector of each pixel point in the spectral image data of the wafer; select a pixel point of the etching layer as a first reference pixel point, and determine the spectral vector corresponding to the first reference pixel point as a first reference vector; compare the spectral vector of each pixel point in the spectral image data with the first reference vector for similarity, and determine the distribution area of ​​the etching layer and the termination layer in the spectral image data; for the distribution area of ​​the termination layer, select a spectral band range that meets preset conditions from multiple spectral bands, and determine the spectral band range that meets the preset conditions as the characteristic spectral band of the termination layer; based on the characteristic spectral band of the termination layer, monitor the etching end point of the wafer to be etched.

[0103] Specifically, the image acquisition component can be a variable-focus high-definition industrial camera that can achieve grayscale imaging with a maximum resolution of 1920*1080 and 16-bit, while ensuring an imaging rate of 3 frames per second. The imaging component includes an objective lens and a lens group. The objective lens combined with the focusing function of the industrial camera can achieve a 5-10x magnification function. The lens group is used for optical path collimation to facilitate the control of light focusing and transmission. The filter component consists of a replaceable filter chip group combined with a carrier mechanism, such as Figure 7 As shown, the filter chipset is designed in a square, 3x3 configuration and placed on the surface of the carrier. The chips are designed as bandpass filters, with each chip corresponding to a specific spectral band of the substance. In actual applications, the filter assembly is placed before the image acquisition component. Before entering the camera, light passes through the filter assembly. The light is then separated by the filter chips within the filter assembly and imaged on the surface of the industrial camera. The resulting image is the spectral image of the target being measured.

[0104] In some cases, a single defect often corresponds to multiple filter chips. Therefore, semiconductor etch monitoring equipment also includes a driver assembly. This assembly utilizes a miniature motor connected to the filter assembly, driving the filter assembly's movement and controlling channel switching. Combined with an industrial camera, this assembly automatically captures data from all spectral bands. The filter assembly and driver assembly work together: in the initial position, the industrial camera lens focuses on the first filter chip. After capturing a spectral image, the driver assembly advances the filter assembly to the next chip. This cycle repeats until data from all filter chips is collected.

[0105] Specifically, the control unit uses a conventional MCU or programmable logic chip as the main controller. It can be programmed to control image acquisition, channel switching, and the coordination of drive components, achieving automated acquisition at a rate of three frames per second. Furthermore, the control unit can perform image consistency calculations to compare the results obtained by the industrial camera with the actual chip layout.

[0106] When there is an error in identifying the film layer (etching layer and termination layer), the "pseudo-defect" is removed by the local contrast method, and the final position of the defect is determined. Specifically, for the defective image with identification errors, the edge extraction algorithm in the field of image processing is used for edge extraction. The edge extraction algorithm can be a convolution algorithm or a Sobel algorithm to obtain the edge position data of each defect inside the scene. Then, the local contrast in each area is calculated, and the calculation is performed using the following relationship: local contrast = sum of the grayscale values ​​of the pixels in the area / total number of pixels. Then, the area with a local contrast higher than a certain threshold is selected as the defect area. In some examples, it is more appropriate to set the local contrast threshold to 0.75-0.85. In this way, the probability of the occurrence of errors in identifying the film layer (etching layer and termination layer) is effectively reduced.

[0107] This application is aimed at the etching process of semiconductor manufacturing, and can realize real-time and rapid monitoring of the end point of the etching layer. The present invention is mainly based on the principle of spectral imaging, and uses the differences in components within the substance to distinguish substances, which can effectively identify the film layers containing O, N or obtained by unconventional processes in the semiconductor manufacturing process. In addition, unlike conventional light reflection spectrum and other monitoring schemes, this scheme uses the reflection characteristics of the substance itself. As long as there are differences in components inside the substance, its reflection spectrum must be different, and it is only necessary to obtain the position of the different spectral segments through calculation. At the same time, a major feature of this scheme is: single acquisition, long-term use. That is, the obtained material spectrum segment data is summarized and can be directly called for later use, avoiding repeated operations.

[0108] Example 3:

[0109] This embodiment provides exemplary contents of the first and second embodiments, that is, provides an exemplary process of a semiconductor etching monitoring method and a semiconductor etching monitoring device, and the specific contents include:

[0110] like Figure 5 As shown, the mask used in semiconductor etching is typically a photoresist. Its structure consists of an open area (i.e., the immediate open area) and a reserved area to create the desired pattern. The non-open area (i.e., the reserved area) serves to block etching, while the open area is etched to the desired depth and topography. When etching a layer of a wafer, with a properly configured etch stop layer, the film corresponding to the open area is completely etched, revealing the next layer beneath it. During semiconductor manufacturing, the compositions of the etched film and the mask are typically fixed. The etched film (the open area) can be metal, SiO2, Si3N4, etc., while the mask is typically photoresist. When these materials are illuminated by a light source or by glow discharge during etching, an emission spectrum is generated. Theoretically, substances with different compositions must produce different emission spectra. Therefore, compositional differentiation can be achieved by simply finding the spectral locations that distinguish between different substances. This is also the most commonly used solution for endpoint monitoring of etching processes. Objects are distinguished by monitoring the spectral information of gas by-products in the process under glow discharge, which is also called optical emission spectroscopy.

[0111] The monitoring method described in this proposal is also based on the reflectance spectra of materials, but it does not rely solely on spectral differences between materials. Instead, it analyzes all materials in the scene and, leveraging spectral differences, identifies characteristic spectral segments that distinguish each object at the image level and presents them as images. This approach is more intuitive, efficient, and less prone to error. Furthermore, when sufficient data is collected, this method can achieve real-time, high-precision monitoring of the stop layer in the etching process.

[0112] like Figure 4 As shown in FIG, the overall process of the method described in this scheme is divided into the following steps:

[0113] S1. Data Collection

[0114] Data collection needs to be completed using a microscopic hyperspectral camera. The sample collected is an etched wafer, and finally a series of spectral images of the etched area and the retained area on the wafer surface are obtained. The so-called spectral image refers to a set of images obtained by collecting the same scene under different spectral bands. Such data can not only show the spatial relationship of objects in the plane, but also show the relationship between the details of the object and the wavelength of light, which is convenient for material differentiation. The data is arranged in the order of light wavelength to obtain a set of two-dimensional planes combined with spectral dimensions. As shown in the attached figure Figure 2 As shown, the data to be collected includes the film in the etched area, the film in the retained area, and other locations outside the current film layer. The spectral camera should have sufficient spatial and spectral resolution. Taking silicon wafer etching with a minimum line width of 1 μm as an example, the optimal camera parameters are a spatial resolution of 2048*2048, an objective lens magnification of 5-50 times, and a spectral resolution of 2.8 nm.

[0115] The spectrum described in this solution is a visible light reflection spectrum with a wavelength of 400 to 720 nm.

[0116] S2. Target Identification and Labeling

[0117] The data needs to be further analyzed to calculate the corresponding position relationship of different types of film layers in the image, so as to obtain the position information of the edges of objects of different materials. The purpose of this is to be able to match the spectral information with the object material, which facilitates the subsequent characteristic spectrum calculation.

[0118] Spectral curves are used to distinguish objects of different materials. The data collection step generates spectral images at different wavelengths at the same location. This data can then be classified and labeled using appropriate algorithms. This invention employs a vector matching approach for data classification. The specific process involves spectral vector reading, target matching calculation, and target labeling.

[0119] 1) Spectral vector reading. For a specific material object, select any pixel in the spectral image and read the grayscale value at that point, which is the spectral intensity of the material in that band. Then switch to the same position in the next band and read the grayscale point again. Repeat these steps in order of wavelength to obtain a set of spectral vectors.

[0120] 2) Target matching calculation. Select any pixel point in the inscribed area in the spectral image (preferably a pixel with clear details and features), and record it as the reference vector X, X iis a vector subset of the reference vector. When the total number of bands is n, the subset has n elements. The similarity comparison between the reference vector and other vectors is performed through the relationship (1) to obtain a similarity comparison value. In this embodiment, in the relationship (1), X is the reference vector, Y is the spectral vector of a certain pixel point, and X i is a vector subset of the base vector, Y i is a vector subset of the spectral vector at a particular pixel. Similarity comparison values ​​closer to 1 indicate higher similarity. For example, if a spectral image contains materials such as SiO2, Si, and photoresist, and the SiO2 curve is used for target matching, if the matching degree between the two curves is between 0.85 and 1, both are considered SiO2.

[0121] By performing this calculation on the thin film in the engraved area, the thin film in the reserved area, and other positions other than the current film layer (i.e., the termination layer), the distribution range of all substances in the image can be obtained.

[0122] S3. Calculation of characteristic spectrum segments

[0123] Select a band with a strong average spectral intensity as the characteristic spectral band. Generally, when the average intensity value of a region within the image is greater than 0.75, the transition between bright and dark areas will be significantly higher than other areas. Therefore, set the threshold range above 0.75. The average spectral intensity value is calculated as: sum of the regional spectral intensity values ​​ / number of pixels in the region.

[0124] At this point, the characteristic spectral band positions of the etched area and the retained area on the surface of a wafer can be obtained through calculation. Then, when identifying the etch stop layer, it is only necessary to monitor the corresponding spectral band in a targeted manner, which can greatly reduce the amount of calculated data.

[0125] In addition, the lithography pattern can be used as a reference image and compared with the image taken under the characteristic spectrum band. The accuracy can be further improved by comparing the consistency.

[0126] Furthermore, after a single data collection and calculation, the results can be consolidated and summarized, allowing for direct reuse in subsequent monitoring processes. If the wafer surface coating changes, such as SiO2 changing to Si3N4, only a new collection and calculation is required to record the data, improving data utilization.

[0127] In some embodiments, as Figure 6 As shown, the semiconductor etching monitoring equipment should include: an image acquisition component, an imaging component, a filtering component, a driving component, and a computing component.

[0128] The image acquisition component is designed as a variable-focus high-definition industrial camera, which can achieve grayscale imaging with a maximum resolution of 1920*1080 16-bit, while ensuring an imaging rate of 3 frames per second. The imaging component includes an objective lens and a lens group. The objective lens combined with the focusing function of the industrial camera can achieve a 5-10x magnification function. The lens group is used for optical path collimation to facilitate the control of light focusing and transmission. The filter component is the core part of the device, consisting of a replaceable filter chip group combined with a carrier mechanism. Figure 7 As shown, the filter chipset is designed in a square, 3x3 configuration and placed on the surface of the carrier. The chips are designed as bandpass filters, with each chip corresponding to a characteristic wavelength band of the substance. In actual use, the filter assembly is placed before the image acquisition component. Before entering the camera, light passes through the filter assembly, where it is split by the filter chips and imaged on the camera surface. The resulting image is an image of the target's characteristic wavelength band.

[0129] In practice, a single defect often corresponds to multiple filter chips. Therefore, the filter assembly typically works in conjunction with a driver module: In the initial position, the camera lens focuses on the first filter chip. After capturing a photo, the driver module immediately drives the filter module to the next chip. This cycle repeats until data from all filter chips is collected.

[0130] The driving component uses a micro-drive motor connected to the filter component to drive the filter component to move, control channel switching, and combine with camera shooting to realize automatic collection of all characteristic band data.

[0131] The computing component (also known as the control unit) uses a conventional MCU or programmable logic chip as the main controller. It can be programmed to control image acquisition, channel switching, and the coordination of drive components, achieving automated acquisition of three frames per second. Furthermore, the computing component performs image consistency calculations to compare camera results with the actual chip layout.

[0132] The local contrast method removes "pseudo-defects" and determines the final location of the defect. The specific calculation method is as follows: 1. For the defect image, edge extraction is performed using an image processing edge extraction algorithm, such as a convolution algorithm or a Sobel algorithm, to obtain the location data of each defect edge within the scene. 2. The local contrast within each area is calculated using the following formula: local contrast = sum of the grayscale values ​​of the pixels in the area / total number of pixels. 3. Defect areas are identified as having a local contrast above a certain threshold. Experiments have shown that a local contrast threshold of 0.75 to 0.85 is suitable.

[0133] Example 4:

[0134] Another embodiment of the present application relates to a semiconductor etching monitoring device. The implementation details of the semiconductor etching monitoring device of this embodiment are specifically described below. The following content is only for the convenience of understanding the implementation details, and is not necessary for the implementation of this solution. The schematic diagram of the semiconductor etching monitoring device of this embodiment can be as follows Figure 2 As shown, a semiconductor etching monitoring device includes:

[0135] An acquisition module 201 is configured to acquire spectral image data of the etched wafer, wherein the spectral image data includes image data of the etched layer and the stop layer in the wafer under multiple spectral bands;

[0136] A vector determination module 202 is used to determine the spectrum vector of each pixel in the spectrum image data of the wafer;

[0137] The reference selection module 203 is configured to select a pixel point of the etching layer as a first reference pixel point, and determine a spectral vector corresponding to the first reference pixel point as a first reference vector;

[0138] A comparison module 204 is configured to compare the spectral vector of each pixel in the spectral image data with the first reference vector for similarity, and determine the distribution areas of the etching layer and the termination layer in the spectral image data;

[0139] The spectrum band determination module 205 is configured to select a spectrum band range that meets a preset condition from a plurality of spectrum bands with respect to the distribution area of ​​the termination layer, and determine the spectrum band range that meets the preset condition as a characteristic spectrum band of the termination layer;

[0140] The monitoring module 206 is configured to monitor the etching endpoint of the wafer to be etched based on the characteristic spectrum of the stop layer.

[0141] It is worth mentioning that all modules involved in this embodiment are logical modules. In actual applications, a logical unit can be a physical unit, a part of a physical unit, or a combination of multiple physical units. In addition, to highlight the innovation of this application, this embodiment does not include units that are not closely related to solving the technical problem proposed by this application. However, this does not mean that other units do not exist in this embodiment.

[0142] Embodiment 5:

[0143] Another embodiment of the present application relates to an electronic device, such as Figure 3As shown, it includes: at least one processor 901; and a memory 902 that is communicatively connected to the at least one processor 901; wherein the memory 902 stores instructions that can be executed by the at least one processor 901, and the instructions are executed by the at least one processor 901 to enable the at least one processor 901 to execute the semiconductor etching monitoring method in the above-mentioned embodiments.

[0144] The memory and processor are connected using a bus, which can include any number of interconnected buses and bridges. The bus connects various circuits of one or more processors and memories. The bus can also connect various other circuits such as peripheral devices, voltage regulators, and power management circuits. These are all well known in the art and are therefore not described further herein. The bus interface provides an interface between the bus and the transceiver. The transceiver can be a single component or multiple components, such as multiple receivers and transmitters, providing a unit for communicating with various other devices over a transmission medium. Data processed by the processor is transmitted over a wireless medium via an antenna. Furthermore, the antenna receives data and transmits it to the processor.

[0145] The processor is responsible for managing the bus and general processing, and can also provide various functions, including timing, peripheral interfaces, voltage regulation, power management, and other control functions. Memory can be used to store data used by the processor when performing operations.

[0146] Example 6:

[0147] Another embodiment of the present application relates to a computer-readable storage medium storing a computer program, which implements the above method embodiment when executed by a processor.

[0148] That is, those skilled in the art will understand that all or part of the steps in the above-mentioned embodiments can be implemented by instructing the relevant hardware through a program, which is stored in a storage medium and includes a number of instructions for causing a device (which may be a single-chip microcomputer, chip, etc.) or a processor to execute all or part of the steps of the methods described in each embodiment of the present application. The aforementioned storage medium includes: a USB flash drive, a mobile hard disk, a read-only memory (ROM), a random access memory (RAM), a magnetic disk, or an optical disk, etc., various media that can store program code.

[0149] Those skilled in the art will appreciate that the above embodiments are specific embodiments for implementing the present application, and that in actual applications, various changes may be made thereto in form and detail without departing from the spirit and scope of the present application.

Claims

1. A semiconductor etching monitoring method, characterized in that: include: Acquiring spectral image data of the etched wafer, wherein the spectral image data includes image data of the etched layer and the stop layer in the wafer under multiple spectral bands; Determining a spectral vector of each pixel in the spectral image data of the wafer; Selecting a pixel point of the etching layer as a first reference pixel point, and determining a spectral vector corresponding to the first reference pixel point as a first reference vector; performing a similarity comparison between the spectral vector of each pixel point in the spectral image data and the first reference vector to determine the distribution areas of the etching layer and the termination layer in the spectral image data; With respect to the distribution area of ​​the termination layer, a spectral band range that meets a preset condition is selected from a plurality of spectral bands, and the spectral band range that meets the preset condition is determined as a characteristic spectral band of the termination layer; Based on the characteristic spectrum of the stop layer, the etching endpoint of the wafer to be etched is monitored.

2. The semiconductor etching monitoring method according to claim 1, wherein: The step of determining the spectral vector of each pixel point in the spectral image data of the wafer includes: Determine the grayscale value of the pixel point in the etching layer under each spectral band as the spectral intensity of the pixel point in the etching layer; Determining the spectral intensity of the pixel points in the etching layer under multiple spectral bands as the spectral vector of each pixel point in the etching layer; Determine the grayscale value of the pixel point in the termination layer under each spectral band as the spectral intensity of the pixel point in the termination layer; The spectral intensities of the pixels in the termination layer under a plurality of spectral bands are determined as the spectral vector of each pixel in the termination layer.

3. The semiconductor etching monitoring method according to claim 1, wherein: The step of comparing the similarity of the spectral vector of each pixel point in the spectral image data with the first reference vector to determine the distribution areas of the etching layer and the termination layer in the spectral image data includes: performing a similarity comparison between the spectral vector of each pixel point in the spectral image data and the first reference vector to determine a similarity comparison value; Detecting whether the similarity comparison value is within a preset matching range; When the similarity comparison value is within a preset matching range, the pixels whose similarity comparison value is within the preset matching range are determined to be pixels of the etching layer, and the remaining pixels are pixels of the termination layer.

4. The semiconductor etching monitoring method according to claim 1, wherein: The step of selecting a spectral band range that meets preset conditions from a plurality of spectral bands for the distribution area of ​​the termination layer, and determining the spectral band range that meets the preset conditions as a characteristic spectral band of the termination layer includes: Calculating the pixel points in the distribution area of ​​the termination layer in each spectral band to determine the average termination spectrum intensity of each spectral band; Detecting whether the average value of the termination spectrum intensity is greater than a preset light intensity threshold; When the average value of the termination spectrum intensity is greater than the preset light intensity threshold, a spectrum band whose average value of the termination spectrum intensity is greater than the preset light intensity threshold is selected from the plurality of spectrum bands, thereby obtaining the spectrum band range of the termination layer; The spectral band range of the termination layer is determined as the characteristic spectrum band of the termination layer.

5. The semiconductor etching monitoring method according to claim 1, wherein: The method further comprises: With respect to the distribution area of ​​the etching layer, a spectral band range that meets a preset condition is selected from a plurality of spectral bands, and the spectral band range that meets the preset condition is determined as a characteristic spectral band of the etching layer; The step of monitoring the etching endpoint of the wafer to be etched based on the characteristic spectrum of the stop layer includes: Based on the characteristic spectrum of the stop layer and the characteristic spectrum of the etching layer, the etching endpoint of the wafer to be etched is monitored.

6. The semiconductor etching monitoring method according to any one of claims 1 to 5, characterized in that: The etching layer includes an etched area and a reserved area; The method further comprises: Selecting a pixel point in the etched area of ​​the etched layer as a second reference pixel point, and determining a spectral vector corresponding to the second reference pixel point as a second reference vector; A similarity comparison is performed between the spectral vector of each pixel point in the spectral image data and the second reference vector to determine the distribution areas of the carved area, the reserved area and the termination layer in the spectral image data.

7. A semiconductor etching monitoring device, characterized in that: include: Image acquisition component, imaging component, filtering component and control unit; The control unit is configured to acquire spectral image data of the etched wafer through an image acquisition component, an imaging component, and a filtering component, wherein the spectral image data includes image data of the etched layer and the stop layer in the wafer under multiple spectral bands; determine a spectral vector of each pixel point in the spectral image data of the wafer; select a pixel point of the etched layer as a first reference pixel point, and determine the spectral vector corresponding to the first reference pixel point as a first reference vector; The spectral vector of each pixel point in the spectral image data is compared with the first reference vector for similarity to determine the distribution area of ​​the etching layer and the termination layer in the spectral image data; for the distribution area of ​​the termination layer, a spectral band range that meets preset conditions is selected from multiple spectral bands, and the spectral band range that meets the preset conditions is determined as the characteristic spectral band of the termination layer; based on the characteristic spectral band of the termination layer, the etching end point of the wafer to be etched is monitored.

8. A semiconductor etching monitoring device, characterized in that: include: An acquisition module, configured to acquire spectral image data of the etched wafer, wherein the spectral image data includes image data of the etched layer and the stop layer in the wafer under multiple spectral bands; a vector determination module, configured to determine a spectral vector of each pixel in the spectral image data of the wafer; a reference selection module, configured to select a pixel point of the etching layer as a first reference pixel point, and determine a spectral vector corresponding to the first reference pixel point as a first reference vector; a comparison module, configured to compare the spectral vector of each pixel point in the spectral image data with the first reference vector for similarity, and determine the distribution areas of the etching layer and the termination layer in the spectral image data; a spectrum band determination module, configured to select a spectrum band range that meets preset conditions from a plurality of spectrum bands with respect to a distribution area of ​​the termination layer, and determine the spectrum band range that meets the preset conditions as a characteristic spectrum band of the termination layer; The monitoring module is used to monitor the etching endpoint of the wafer to be etched based on the characteristic spectrum of the stop layer.

9. An electronic device, characterized in that: include: at least one processor; as well as, a memory communicatively connected to the at least one processor; wherein, The memory stores instructions that can be executed by the at least one processor, and the instructions are executed by the at least one processor to enable the at least one processor to perform the semiconductor etching monitoring method according to any one of claims 1 to 7.

10. A computer-readable storage medium storing a computer program, characterized in that: When the computer program is executed by a processor, the semiconductor etching monitoring method according to any one of claims 1 to 7 is implemented.

Images