Grid line width measuring method and measuring device

Abstract

The invention provides a grid line width measuring method and a measuring device, which can be widely applied to the technical field of photovoltaic measurement, and the grid line width measuring method comprises the following steps: obtaining a morphology parameter of a to-be-measured grid line, if the morphology parameter is in a preset parameter interval, adjusting a measuring parameter of a first measuring unit to a preset measuring parameter, and if the morphology parameter is in the preset parameter interval, adjusting the measuring parameter of a second measuring unit to the preset measuring parameter; measuring the line width of the to-be-measured grid line under a preset measurement parameter through a first measurement unit to obtain a measured line width; the preset parameter interval comprises one or more of a preset depth-to-width ratio interval, a preset geometric feature interval and a preset side wall angle interval of the cross section of the grid line; calibrating the measured line width through a calibration model to obtain a calibrated line width of the to-be-measured grid line; wherein the calibration model is a mapping relation between the first measurement unit and the second measurement unit, and the measurement accuracy of the grid line width of the to-be-measured grid line by the second measurement unit is higher than that of the first measurement unit. The accuracy of grid line width measurement is improved, and the measurement efficiency is improved.

Description

Technical Field

[0001] This application relates to the field of photovoltaic measurement technology, and more specifically, to a method and apparatus for measuring grid linewidth. Background Technology

[0002] In the current field of crystalline silicon solar cell manufacturing, the fineness and uniformity of the electrode grid lines are crucial for improving cell conversion efficiency and reducing costs. While traditional screen printing technology is mature and reliable, its high cost and limited linewidth of silver paste make it difficult to meet the demands of high-efficiency solar cells. In recent years, copper electroplating electrode technology has attracted widespread attention as a low-cost, high-efficiency alternative. It can achieve finer, higher aspect ratio grid line morphologies, significantly reducing shading loss and series resistance, thereby improving cell efficiency.

[0003] However, the application of this cutting-edge technology has encountered bottlenecks in process monitoring. In particular, for metal grid lines with steep sidewalls (sidewall angles greater than 80°) and approximately rectangular cross-sections (such as electroplated copper grid lines), accurate measurement of their linewidth becomes complex and difficult. Scanning electron microscopy (SEM), as a high-precision measurement tool, can provide nanometer-level resolution and is the gold standard for measuring the linewidth of steep-topography grid lines. However, due to its cumbersome operation, time-consuming nature, and somewhat destructive effect on samples, it is unsuitable for batch testing and online monitoring. In contrast, three-dimensional laser microscopy (3D laser microscopy) has become an ideal measurement choice due to its high speed, non-destructive, and non-contact characteristics. However, in practical applications, for grid lines with steep structures, 3D laser microscopy measurement results exhibit significant systematic errors, ultimately leading to overestimation of the measured values. Summary of the Invention

[0004] The main objective of this application is to provide a method and apparatus for measuring grid line width, so as to at least solve the problems of cumbersome measurement or low measurement accuracy in the prior art.

[0005] To achieve the above objectives, according to one aspect of this application, a method for measuring the linewidth of a grating is provided. The method includes: acquiring morphological parameters of a grating to be measured; if the morphological parameters are within a preset parameter range, adjusting the measurement parameters of a first measurement unit to the preset measurement parameters, and measuring the linewidth of the grating to be measured using the first measurement unit under the preset measurement parameters to obtain the measured linewidth; the preset parameter range includes one or more of a preset aspect ratio range, a preset geometric feature range, and a preset sidewall angle range for the cross-section of the grating; calibrating the measured linewidth using a calibration model to obtain the calibrated linewidth of the grating to be measured; wherein the calibration model is a mapping relationship between the first measurement unit and a second measurement unit, and the second measurement unit has a higher accuracy in measuring the linewidth of the grating to be measured than the first measurement unit. When measuring the grid lines with a preset shape, this application uses a second measurement unit with higher measurement accuracy to calibrate the measurement linewidth of the first measurement unit to obtain a calibrated linewidth, which helps to improve the accuracy of grid linewidth measurement. At the same time, the grid linewidth measurement is performed by a first measurement unit with lower measurement complexity, which improves measurement efficiency.

[0006] Optionally, before calibrating the measurement linewidth using a calibration model to obtain the calibration linewidth of the gate line to be measured, the method further includes:

[0007] The second measurement unit measures the linewidth of each first grid line in the grid line sample set to obtain the sample reference value.

[0008] The first measurement unit measures the linewidth of each first grid line under multiple measurement parameters to obtain sample measurement values ​​under each measurement parameter.

[0009] A correlation analysis was performed between the sample measurement values ​​and the corresponding sample benchmark values ​​for each measurement parameter to establish a correlation model for each measurement parameter.

[0010] Multiple measurement parameters and multiple correlation models are screened and optimized. The selected measurement parameters are used as preset measurement parameters, and the selected correlation models are used as calibration models.

[0011] Optionally, the plurality of measurement parameters includes a first edge detection threshold and a second edge detection threshold, and the association model includes an association formula and an association chart; the step of performing association analysis between the sample measurement value and the corresponding sample benchmark value under each measurement parameter to establish an association model under each measurement parameter includes:

[0012] A linear regression analysis is performed on the sample measurement values ​​and corresponding sample baseline values ​​under the first edge detection threshold to obtain the first correlation formula;

[0013] A second correlation formula is obtained by performing linear regression analysis on the sample measurement values ​​and corresponding sample baseline values ​​under the second edge detection threshold.

[0014] A fitting analysis is performed on the sample measurement values ​​and corresponding sample baseline values ​​under the first edge detection threshold to obtain a first correlation chart;

[0015] The sample measurement values ​​and corresponding sample baseline values ​​under the second edge detection threshold are fitted and analyzed to obtain the second correlation chart.

[0016] Optionally, multiple correlation models may be filtered and optimized, including:

[0017] Under the same measurement parameters, each second grid line in the grid line verification set is measured to obtain the measured line width of each second grid line;

[0018] The measured linewidth of each second gate line is calibrated using multiple correlation models to obtain the calibrated linewidth of the second gate line for each correlation model.

[0019] The true linewidth of each second gate line is obtained, and the calibration linewidth of each associated model is compared with the true linewidth to obtain a comparison result for each associated model; wherein, the comparison result is used to characterize the calibration accuracy of the associated model;

[0020] Based on the comparison results, the calibration accuracy of the multiple associated models is sorted in the first order, and the associated model ranked first in the first order is selected as the calibration model.

[0021] Optionally, multiple measurement parameters may be screened and optimized, including:

[0022] The sample measurement values ​​under multiple measurement parameters are calibrated using the calibration model to obtain sample predicted values ​​under multiple measurement parameters. Based on the sample predicted values ​​of multiple measurement parameters and the corresponding sample reference values, the evaluation values ​​of multiple measurement parameters are determined; wherein, the evaluation values ​​are used to characterize the calibration accuracy of the calibration model under different measurement parameters.

[0023] The evaluation values ​​of the plurality of measurement parameters are sorted in a second order, and the measurement parameter corresponding to the first evaluation value in the second order is selected as the preset measurement parameter.

[0024] Optionally, the calibration model includes a first correlation formula, and the step of determining the evaluation values ​​of the multiple measurement parameters based on the sample predicted values ​​of the multiple measurement parameters and the corresponding sample benchmark values ​​includes:

[0025] Based on the sample predicted values ​​of the plurality of measurement parameters and the corresponding sample benchmark values, the first determination coefficient of the first association formula is determined, and the first determination coefficient is used as the evaluation value.

[0026] Alternatively, based on the sample predicted values ​​of multiple measurement parameters and the corresponding sample benchmark values, the first residual of the first association formula is determined, and the first residual is used as the evaluation value.

[0027] Optionally, before measuring the linewidth of each first gate line under multiple measurement parameters using the first measurement unit, the method further includes:

[0028] A measurement position is marked on each of the first grid lines so that the first measurement unit can perform linewidth measurement on each measurement position of the first grid line under multiple measurement parameters; wherein, the measurement position is the measurement position where the first measurement unit performs linewidth measurement.

[0029] Optionally, the method further includes:

[0030] Obtain the design linewidth of the grid line to be measured;

[0031] The calibration model is determined based on the preset linewidth range to which the designed linewidth belongs.

[0032] Optionally, the method further includes:

[0033] If the morphology parameters are not the preset morphology, the linewidth of the grid line to be measured is measured by the first measurement unit to obtain the measurement result.

[0034] To achieve the above objectives, according to another aspect of this application, a grid line width measuring device is provided, which measures the line width of a grid line using the aforementioned grid line width measuring method.

[0035] By applying the technical solution of this application, when measuring the grid lines with a preset morphology, the measurement linewidth of the first measurement unit is calibrated by the second measurement unit with higher measurement accuracy to obtain the calibrated linewidth, which is beneficial to improving the accuracy of grid linewidth measurement; at the same time, the grid linewidth is measured by the first measurement unit with lower measurement complexity, thereby improving measurement efficiency. Attached Figure Description

[0036] The accompanying drawings, which form part of this application, are used to provide a further understanding of this application. The illustrative embodiments and descriptions of this application are used to explain this application and do not constitute an undue limitation of this application. In the drawings:

[0037] Figure 1 A schematic diagram of a grid line cross-section for measuring grid linewidth is shown in an embodiment of this application.

[0038] Figure 2 A schematic diagram of a grid line cross-section for another grid line width measurement provided in an embodiment of this application is shown;

[0039] Figure 3 A schematic flowchart of a gate linewidth measurement method provided in an embodiment of this application is shown;

[0040] Figure 4 A schematic flowchart of a measurement process provided according to an embodiment of this application is shown;

[0041] Figure 5 A flowchart illustrating the screening and optimization process provided according to an embodiment of this application is shown. Detailed Implementation

[0042] It should be noted that, unless otherwise specified, the embodiments and features described in this application can be combined with each other. This application will now be described in detail with reference to the accompanying drawings and embodiments.

[0043] 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. Obviously, the described embodiments are only some embodiments of the present application, and not all embodiments. Based on the embodiments in the present application, all other embodiments obtained by those of ordinary skill in the art without creative effort should fall within the scope of protection of the present application.

[0044] It should be noted that the terms "first," "second," etc., in the specification, claims, and accompanying drawings of this application are used to distinguish similar objects and are not necessarily used to describe a specific order or sequence. It should be understood that such data can be interchanged where appropriate for the embodiments of this application described herein. 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 comprises a series of steps or units is not necessarily limited to those steps or units explicitly listed, but may include other steps or units not explicitly listed or inherent to such processes, methods, products, or apparatus.

[0045] As described in the background section, in related technologies:

[0046] Crystalline silicon solar cell technology is rapidly developing towards higher conversion efficiency and lower production costs. Traditional screen printing technology uses expensive silver paste to prepare electrode grid lines, which accounts for a large proportion of the non-silicon cost of the cell. In addition, the line width is relatively wide (typically >25μm), resulting in significant light loss due to shading, which limits further improvement in cell efficiency.

[0047] To overcome this bottleneck, copper electroplating electrode technology is considered a crucial direction for next-generation industrialization. This technology forms electrode grid lines through patterning, seed layer deposition, and copper electroplating, offering the following significant advantages: Extreme cost reduction: Replacing expensive silver with inexpensive copper significantly reduces metallization costs. Improved efficiency: Achieving finer (<30μm or even <10μm) and high aspect ratio grid line morphologies significantly reduces shading of the cell's light-receiving surface, lowers series resistance, and thus improves the cell's photoelectric conversion efficiency. Superior morphology: The electroplated copper grid lines possess an ideal conductive structure with steep edges and an approximately rectangular cross-section.

[0048] However, this advanced electrode technology presents unprecedented challenges to process monitoring. The width of the grid lines (linewidth) is one of its most critical dimensions. The consistency of the linewidth directly affects: the conductivity of the solar cell (too thin a linewidth increases series resistance and affects the fill factor); the shading loss of the solar cell (too thick a linewidth increases the shading area and reduces short-circuit current); and the process stability of solar cell fabrication (linewidth is an important indicator of the uniformity and stability of the electroplating process). Therefore, achieving rapid, accurate, and non-destructive measurement of the copper-plated grid linewidth is a key step in advancing this technology.

[0049] Scanning electron microscopy (SEM) and three-dimensional laser scanning microscopy (3D laser microscopy) are two widely used tools for measuring critical dimensions (such as linewidth) of microstructures. SEM offers nanometer-level high resolution, and its measurements are often considered benchmarks, especially for structures with steep sidewalls, enabling precise measurements. However, SEM measurements typically require a vacuum environment, are relatively cumbersome and time-consuming, and require sample destruction. 3D laser microscopy, with its advantages of high measurement speed, non-destructive nature, non-contact operation, and ability to provide three-dimensional morphological information, has become a highly efficient measurement method. However, when using 3D laser microscopy to measure specific morphological structures—namely, metal grids with steep sidewalls (sidewall angles typically greater than 80°) and approximately rectangular cross-sectional shapes (such as copper grids formed through electroplating processes)—it becomes challenging to achieve the desired results. Figure 1 When measuring the cross-section of electroplated copper grid lines using SEM, a significant systematic measurement error occurs: the measured linewidth value is more constant than the SEM measurement value.

[0050] This error stems from the limitations of the physical principles of optical microscopes. Due to the diffraction effect of light, the laser spot diffuses when scanning steep edges, causing the edge signal received by the detector to become blurred and broadened, resulting in a systematic outward deviation of the edge position determined by the instrument software. For structures with gently sloping sidewalls (such as traditional screen-printed grid lines, etc.), this error is more pronounced. Figure 2The error is not significant when measuring the cross-section of screen-printed grid lines using SEM; however, for steep structures produced by modern advanced manufacturing processes, this error has seriously affected the accuracy and reliability of 3D laser microscope measurement results, limiting its application in high-precision metrology. Table 1 shows the linewidth data of two different types of grid lines measured using 3D and SEM.

[0051]

[0052] Table 1

[0053] To address the aforementioned issues, embodiments of this application provide a method for measuring the linewidth of a gate line. When measuring a gate line with a preset shape, the linewidth measured by the first measuring unit is calibrated using a second measuring unit with higher measurement accuracy to obtain a calibrated linewidth, which helps improve the accuracy of the gate linewidth measurement. At the same time, the measurement efficiency is improved by using a first measuring unit with lower measurement complexity for the gate linewidth measurement.

[0054] The technical solutions of the present invention will be clearly and completely described below with reference to the accompanying drawings in the embodiments of the present invention.

[0055] Figure 3 This is a flowchart illustrating a gate linewidth measurement method according to an embodiment of this application. The gate linewidth measurement method includes:

[0056] Step S100: Obtain the morphological parameters of the grid line to be measured. If the morphological parameters are within a preset parameter range, adjust the measurement parameters of the first measurement unit to the preset measurement parameters, and measure the line width of the grid line to be measured through the first measurement unit to obtain the measured line width. The preset parameter range includes one or more of the following: a preset depth-to-width ratio range of the grid line cross-section, a preset geometric feature range, and a preset sidewall angle range.

[0057] Step S200: The measurement linewidth is calibrated using a calibration model to obtain the calibration linewidth of the grating to be measured; wherein, the calibration model is a mapping relationship between the first measurement unit and the second measurement unit, and the second measurement unit has a higher accuracy in measuring the linewidth of the grating with a preset shape than the first measurement unit.

[0058] The preset measurement parameters of this application correspond to the calibration model. The measurement linewidth of the first measurement unit under the preset measurement parameters is calibrated through the corresponding calibration model. The preset measurement parameters are the parameters set by the first measurement unit during measurement. The morphological parameters are parameters such as the aspect ratio, geometric features (i.e., shape), and sidewall angle of the cross-section of the grid line. The preset parameter range can be one or more of the preset aspect ratio, preset shape, and preset sidewall angle of the cross-section of the grid line. Specifically, the preset aspect ratio can be a given range of aspect ratios, and the preset sidewall angle can also be a given range of sidewall angles. For example, the preset morphology can be a metal grid line with a sidewall angle typically greater than 80°, a cross-sectional shape approximately rectangular, and a cross-sectional aspect ratio greater than the preset aspect ratio. Here, the sidewall angle is the angle formed by the sidewall connected to the bottom edge and the bottom edge in the cross-section of the grid line. The preset geometric feature range can be set based on the preset shape of the grid line cross-section. For example, if it is necessary to measure a grid line with an approximately rectangular cross-section, the preset geometric features can be set as the bounding box and area ratio range of the cross-section of the grid line. By obtaining the minimum bounding rectangle (i.e., the bounding box) of the cross-section of the grating line to be measured, the area of ​​the bounding box is determined. Simultaneously, the shape area of ​​the cross-section of the grating line to be measured is obtained, and the ratio of the bounding box to the area of ​​the cross-section of the grating line to be measured (i.e., the morphological parameter in this application) is calculated. Then, it is determined whether the calculated ratio of the bounding box to the area is within a preset range (i.e., a preset parameter range), and whether the grating line width measurement should be performed using the scheme described in steps S100 to S400. Alternatively, for grating lines whose cross-section is approximately rectangular, the preset geometric features can be set as the angles of the four vertex angles of the grating line's cross-section. In this case, the vertex angles are the morphological parameters in this application, and the angle range is the preset parameter range in this application. Of course, the above embodiments regarding morphological parameters and preset parameter ranges are illustrative examples. This application does not limit the specific physical meaning of the morphological parameters or the specific values ​​of the preset parameter ranges; those skilled in the art can set them according to their needs.

[0059] It should be noted that the calibration model in this application can be a calibration formula, a calibration chart, or an artificial intelligence model. This application does not specifically limit the form of the calibration model. In some embodiments, the first measurement unit is a 3D laser microscope, and the second measurement unit can be a SEM scanning electron microscope. In this application, the measurement complexity of the first measurement unit is lower than the measurement load of the second measurement unit. However, the measurement accuracy of the first measurement unit for the linewidth of the grating line with morphological parameters within a preset parameter range is lower than the measurement accuracy of the second measurement unit for the linewidth of the grating line with morphological parameters within a preset parameter range; that is, for the linewidth of the grating line with morphological parameters within a preset parameter range, the measurement accuracy of the second measurement unit is higher than that of the first measurement unit. Therefore, this application reflects the mapping relationship between the first and second measurement units through a pre-set calibration model, and then calibrates the measurement linewidth obtained by the first measurement unit through the calibration model, which improves the measurement accuracy, reduces the measurement difficulty, and improves the measurement efficiency.

[0060] The technical solution of this embodiment can solve the problem of inaccurate measurement results caused by the greater consistency of the measured values ​​of 3D laser microscopes compared to scanning electron microscopes when measuring the linewidth of steep-topped gratings. This solution obtains the morphological parameters of the grating to be measured and confirms that it falls within a preset parameter range (i.e., a steep topography). Then, using a first measurement unit (3D laser microscope) under preset measurement parameters, the linewidth is measured to obtain an initial measurement value. Next, a calibration model based on the mapping relationship between the first and second measurement units is used to calibrate the measured linewidth, resulting in a more accurate calibrated linewidth. The calibration model is established by comparing and analyzing data from a standard sample set under the two measurement units. By adjusting different measurement parameters, such as the laser focus and scanning speed of the 3D laser microscope or the threshold in the edge detection algorithm, the measurement parameter-model combination with the smallest error compared to the SEM measurement result is found. This parameter combination constitutes the preset measurement parameters and the calibration model. After the calibration model is corrected, this method can match the measurement results of the 3D laser microscope with the high-precision measurement results of the SEM, thereby eliminating systematic measurement errors and improving the accuracy and reliability of the measurement. The technical solution of this invention not only retains the advantages of 3D laser microscopy—fast, non-destructive, and three-dimensional measurement—but also corrects its systematic errors when measuring the linewidth of steeply shaped grid lines, ensuring the accuracy and consistency of measurement results. This provides crucial technical support for process monitoring of copper electroplating electrode technology in the manufacture of crystalline silicon solar cells. This enables 3D laser microscopy to be widely used as a high-efficiency, high-precision measurement tool in quality control within large-scale production environments, reducing reliance on expensive and time-consuming SEM measurements and improving production efficiency and cost-effectiveness.

[0061] Optionally, before calibrating the measurement linewidth using a calibration model to obtain the calibration linewidth of the gate line to be measured, the method further includes:

[0062] The linewidth of each first grid line in the grid line sample set is measured by the second measurement unit to obtain the sample reference value;

[0063] The first measurement unit measures the linewidth of each first grid line under multiple measurement parameters to obtain sample measurement values ​​for each measurement parameter.

[0064] A correlation analysis was performed between the sample measurement values ​​and the corresponding sample benchmark values ​​for each measurement parameter to establish a correlation model for each measurement parameter.

[0065] Multiple measurement parameters and multiple correlation models are screened and optimized. The selected measurement parameters are used as preset measurement parameters, and the selected correlation models are used as calibration models.

[0066] This application establishes a grid line sample set. The morphological parameters of each first grid line in the sample set are within a preset parameter range. Since the second measurement unit is more accurate in measuring the linewidth of the grid lines within the preset parameter range, this application measures each first grid line using the second measurement unit to obtain a sample reference value. Then, the first measurement unit measures the linewidth of each first grid line under multiple measurement parameters, and performs correlation analysis between the sample measurement value under each measurement parameter and the corresponding sample reference value of the first grid line to establish a correlation model. This yields multiple measurement parameter-correlation model combinations (including the calibration model of this application). The multiple measurement parameter-correlation model combinations are then screened and optimized to select the optimal combination, which is the preset measurement parameters and calibration model. It should be noted that the measurement parameters in this application are the measurement setting parameters of the first measurement unit. Specifically, these can be parameters set before measurement, such as objective magnification, laser wavelength, scanning mode, pinhole diameter, laser power, and edge detection thresholds (height threshold / high threshold / low threshold). It is understood that, in one embodiment, by setting an edge detection threshold for the measurement unit, variations in reflectivity or height of different materials can be detected, and the detection accuracy can be controlled. Of course, the above description of the measurement parameters is illustrative and not intended to limit the specific types of measurement parameters.

[0067] It is understood that this application requires two quantities determined through screening and optimization: preset measurement parameters and a calibration model, corresponding to multiple measurement parameters collected before calibration and multiple types of correlation models. Specifically, multiple measurement parameters can be screened and optimized first, using a unified correlation model to determine the sample measurement value and sample reference value corresponding to each measurement parameter among the multiple measurement parameters, and then performing difference evaluation. Based on the difference evaluation results, the optimal measurement parameter is determined, i.e., the preset measurement parameter is obtained. Then, based on the preset measurement parameters, the sample measurement values ​​and sample reference values ​​of multiple types of correlation models are determined, and the difference evaluation is also performed to determine the optimal type of correlation model, i.e., the calibration model is obtained. In another embodiment, multiple types of correlation models can be screened and optimized first, using a unified measurement parameter to screen the correlation models under multiple types, determining the optimal correlation model as the calibration model, and then, based on the calibration model, the optimal measurement parameter is selected by screening the measurement data under multiple measurement parameters. In another embodiment, multiple measurement parameters are set, and the measurement data under each parameter are fitted using multiple types of correlation models to obtain multiple measurement parameter-correlation model combinations. Then, the measurement results of these multiple measurement parameter-correlation model combinations are evaluated, and the optimal combination is selected to obtain the preset measurement parameters and calibration model. Those skilled in the art can set specific processing logic for the selection and optimization process according to actual needs; this application does not impose specific limitations. The multiple types of correlation models in this application can be correlation models in formula form, correlation models in chart form, or correlation models in intelligent model form.

[0068] In this embodiment, a second measurement unit is introduced to measure the linewidth of each first grid line in the grid line sample set, obtaining a sample reference value, which forms the basis for establishing the calibration model. Subsequently, the first measurement unit is used to measure the linewidth of each first grid line under multiple measurement parameters, obtaining sample measurement values ​​under each measurement parameter. This step constructs the original dataset of the calibration model. Correlation analysis is performed between the sample measurement values ​​under each measurement parameter and the corresponding sample reference value to establish a correlation model for each measurement parameter. This step is the core of the calibration model construction; through data analysis and model fitting, the relationship between the measured values ​​and the reference values ​​under different test parameter settings is quantitatively described. Next, the measurement parameters and correlation models are screened and optimized. The screened measurement parameters are used as preset measurement parameters, and the screened correlation models are used as calibration models. This step ensures the accuracy of the calibration model and the optimality of the measurement parameters, providing crucial technical support for subsequent accurate measurements. The results show that the above technical solution can significantly improve the accuracy and reliability of 3D laser microscope measurement of grid line width in steep morphology, eliminate systematic errors caused by differences in measurement parameter settings, and make the measurement results highly consistent with high-precision SEM measurement values. This meets the requirements for precise control of grid line width in the cell manufacturing process and improves production efficiency and quality control.

[0069] Optionally, multiple measurement parameters include a first edge detection threshold and a second edge detection threshold, and the correlation model includes a correlation formula and a correlation chart; correlation analysis is performed on the sample measurement values ​​and corresponding sample benchmark values ​​under each measurement parameter to establish a correlation model under each measurement parameter, including:

[0070] A linear regression analysis is performed on the sample measurements and corresponding sample baseline values ​​based on the first edge detection threshold to obtain the first correlation formula;

[0071] A linear regression analysis was performed on the sample measurements and corresponding sample baseline values ​​based on the second edge detection threshold to obtain the second correlation formula.

[0072] A fitting analysis is performed on the sample measurements and corresponding sample baseline values ​​based on the first edge detection threshold to obtain the first correlation chart;

[0073] A fitting analysis is performed on the sample measurements and corresponding sample baseline values ​​based on the second edge detection threshold to obtain the second correlation chart.

[0074] It is understood that this application can also perform other types of correlation analysis on the sample measurement value and the corresponding sample benchmark value under the first edge detection threshold to obtain the correlation model, that is, the calibration model in this application. The calibration model can be a formula, a chart, or a structured database, etc. This application does not limit the specific form of the calibration model.

[0075] In some embodiments, the association model in this application may be an association formula or an association table, and the measurement parameter may be a first edge detection threshold or a second edge detection threshold. Therefore, this application includes:

[0076] A linear regression analysis is performed on the sample measurement values ​​under the first edge detection threshold and the sample baseline values ​​corresponding to the first grid line to obtain the first correlation formula;

[0077] A second correlation formula is obtained by performing linear regression analysis on the sample measurement values ​​under the second edge detection threshold and the corresponding sample reference values ​​of the first grid line.

[0078] A fitting analysis is performed on the sample measurement values ​​under the first edge detection threshold and the sample reference values ​​corresponding to the first grid line to obtain the first correlation chart;

[0079] The sample measurement values ​​under the second edge detection threshold and the corresponding sample reference values ​​of the first grid line are fitted and analyzed to obtain the second correlation chart.

[0080] Of course, the measurement parameters in this application can also be objective lens magnification, pinhole size, etc. For different measurement parameters or different combinations of measurement parameters, and combined with different calibration methods (i.e., calibration models or correlation models), different measurement parameter-correlation model combinations are obtained. Then, through a unified evaluation method, multiple measurement parameter-correlation model combinations are screened and optimized.

[0081] In this embodiment, to address the need for precise measurement of the linewidth of steep-topped grating lines, this invention proposes a systematic calibration method. This method first defines two key measurement parameters: a first edge detection threshold and a second edge detection threshold, used to set the standards for identifying grating line edges in the 3D laser microscope edge detection algorithm. Then, the precise linewidth values ​​of the standard sample set are obtained using SEM as a benchmark. Next, the same standard sample set (grating sample set) is scanned multiple times at different height thresholds using a 3D laser microscope, collecting the corresponding original linewidth measurements. Based on the collected data, linear regression analysis is performed between the measurements at the first edge detection threshold and the SEM benchmark values ​​to derive a first correlation formula, and simultaneously, a first correlation chart is obtained. The same processing is applied to the measurements at the second edge detection threshold to obtain a second correlation formula and a second correlation chart. These formulas and charts constitute the core of the calibration model, enabling the rapid acquisition of accurate calibrated linewidth values ​​by looking up tables or calculating based on the original measurements at a specific height threshold when measuring unknown samples. This technical solution not only quantifies the distribution characteristics of errors under different threshold settings, but also allows users to flexibly select the most suitable calibration model according to actual needs, thereby significantly improving the accuracy and reliability of the measurement of steep-topography grid line width while ensuring measurement speed.

[0082] Optionally, multiple correlation models may be filtered and optimized, including:

[0083] Under the same measurement parameters, each second grid line in the grid line verification set is measured to obtain the measured line width of each second grid line;

[0084] The measured linewidth of each second gate line is calibrated using multiple correlation models to obtain the calibrated linewidth of the second gate line for each correlation model.

[0085] The true linewidth of each second gate line is obtained, and the calibration linewidth of each associated model is compared with the true linewidth to obtain a comparison result for each associated model; wherein, the comparison result is used to characterize the calibration accuracy of the associated model;

[0086] Based on the comparison results, the calibration accuracy of the multiple associated models is sorted in the first order, and the associated model ranked first in the first order is selected as the calibration model.

[0087] This application can verify the calibration effect of the measurement parameter-correlation model combination using the second grid line in the grid line verification set, in order to screen and optimize the measurement parameters and correlation models. Alternatively, this application can also verify the calibration effect of the measurement parameter-correlation model combination using the first grid line in the grid line sample, in order to screen and optimize the measurement parameters and correlation models. Specifically, a measurement parameter is first fixed, and each second grid line in the grid line verification is measured using this parameter. The measurement results are then calibrated using multiple correlation models to obtain multiple calibration linewidths. By comparing the calibration linewidths with the actual linewidths, the optimal correlation model is selected as the calibration model.

[0088] In this embodiment, the process of selecting and optimizing measurement parameters and correlation models was further optimized for the accurate measurement of the linewidth of steep-topography grating lines. Each second grating line was measured under a 3D laser microscope using a set of grating line validation parameters to obtain its linewidth measurement results. This validation set serves to comprehensively evaluate the accuracy and consistency of the data under the measurement parameters. Next, based on the measurement results of each second grating line, the correlation model was systematically adjusted and optimized to find the optimal correlation model that minimizes errors and improves model fit.

[0089] Optionally, multiple measurement parameters may be screened and optimized, including:

[0090] The sample measurement values ​​under multiple measurement parameters are calibrated using the calibration model to obtain sample predicted values ​​under multiple measurement parameters. Based on the sample predicted values ​​of multiple measurement parameters and the corresponding sample reference values, the evaluation values ​​of multiple measurement parameters are determined; wherein, the evaluation values ​​are used to characterize the calibration accuracy of the calibration model under different measurement parameters.

[0091] The evaluation values ​​of the plurality of measurement parameters are sorted in a second order, and the measurement parameter corresponding to the first evaluation value in the second order is selected as the preset measurement parameter.

[0092] After determining the calibration model, this application calibrates sample measurements obtained under multiple measurement parameters using the calibration model to obtain sample predicted values ​​(the calibrated linewidth after calibration). Then, by comparing the sample predicted values ​​with the sample reference values ​​(i.e., the true values), evaluation values ​​for multiple measurement parameters are obtained. The optimal measurement parameters, i.e., the preset measurement parameters, are determined through a second ranking of the evaluation values. This application calibrates sample measurements using different combinations of measurement parameters and correlation models to obtain sample predicted values. Based on the sample predicted values ​​and the corresponding reference values ​​for the grid lines, an evaluation system is established to obtain various evaluation values. Based on the ranking of the evaluation values, the evaluation values ​​are selected, thereby obtaining the preset measurement parameters and the calibration model.

[0093] Optionally, the calibration model includes a first correlation formula, and the step of determining the evaluation values ​​of the multiple measurement parameters based on the sample predicted values ​​of the multiple measurement parameters and the corresponding sample benchmark values ​​includes:

[0094] Based on the sample predicted values ​​of the plurality of measurement parameters and the corresponding sample benchmark values, the first determination coefficient of the first association formula is determined, and the first determination coefficient is used as the evaluation value.

[0095] Alternatively, based on the sample predicted values ​​of multiple measurement parameters and the corresponding sample benchmark values, the first residual of the first association formula is determined, and the first residual is used as the evaluation value.

[0096] In this application, the evaluation system can screen and optimize the combination of measurement parameters and correlation models using the coefficient of determination or residuals of the sample predicted values ​​and the sample benchmark values. In this embodiment, based on the SEM baseline linewidth values ​​and multiple sets of raw linewidth data from the 3D laser microscope, an optimal calibration formula or lookup table is determined through mathematical modeling analysis. Simultaneously, a set of "optimal measurement parameters" that maximizes measurement accuracy and correlation is identified. Finally, in the actual measurement process, a 3D laser microscope is used to scan the sample under test according to the pre-determined preset measurement parameters. The obtained raw measurement values ​​are input into the established calibration model to calculate the calibration linewidth value that matches the SEM accuracy. This method not only overcomes the influence of optical diffraction on the measurement accuracy of steep grating lines but also maintains the fast and non-destructive characteristics of the 3D laser microscope, achieving high-precision, efficient, and standardized measurement of grating linewidths. Specifically, in determining the first evaluation value, the quality of the calibration model can be evaluated by calculating the first coefficient of determination or the first residual. The first coefficient of determination reflects the goodness of fit of the linewidth calibration model, while the first residual directly shows the degree of deviation between the model's predicted value and the actual SEM measurement value. Together, these two help us select the most suitable combination of parameters and thus establish the most accurate calibration model.

[0097] Optionally, before measuring the linewidth of each first gate line under multiple measurement parameters using the first measurement unit, the method further includes:

[0098] A measurement position is marked on each first grid line so that the first measurement unit can perform linewidth measurement on each first grid line measurement position under multiple measurement parameters; wherein, the measurement position is the measurement position where the first measurement unit performs linewidth measurement.

[0099] In this embodiment, before measuring the linewidth of each first grid line, the method further includes marking the measurement position on the first grid line to ensure that the first measurement unit is aligned to the same position for measurement under various measurement parameters. This design not only improves the comparability and accuracy of the measurement results but also eliminates data deviations caused by inconsistent positions, laying the foundation for establishing a reliable calibration model. Through precise positioning, whether in high-resolution SEM scanning or under different parameter settings of a 3D laser microscope, it can be ensured that the linewidth is measured at the same position of the first grid line. This effectively avoids the influence of positional variables on the measured values ​​during data analysis, ensuring a direct comparison between the SEM reference value and the original measurement value from the 3D laser microscope, and improving the efficiency and accuracy of establishing the calibration model. Of course, precise control of the measurement position can also be achieved through other methods, such as using an autofocus system or preset image recognition markers. These methods can also ensure the consistency of measurements and the reliability of data, thereby improving the standardization level of the entire measurement process and the accuracy of the measurement results.

[0100] Optionally, the method further includes:

[0101] Obtain the design linewidth of the grid line to be measured;

[0102] The calibration model is determined based on the preset linewidth range to which the designed linewidth belongs.

[0103] In some embodiments, for the grid lines to be measured with different designed linewidths, a calibration model corresponding to the preset linewidth range to which the designed linewidth belongs is determined. Different preset calibration models are set for different preset linewidth ranges to increase calibration accuracy. In this embodiment, the designed linewidth of the grid line to be measured is obtained, and the calibration model is determined according to the preset linewidth range to which the designed linewidth belongs. This step embodies an intelligent and personalized measurement strategy, making the selection of calibration models more accurate and fully adaptable to the measurement needs of grid lines of different specifications. The designed linewidth, as a process parameter, is predetermined in the cell manufacturing process. For each specific grid line, its designed linewidth belongs to a preset linewidth range. The designed linewidth of the sample to be measured is compared with the linewidth range of the standard sample set to select the calibration model that best matches it. This range-based model selection mechanism avoids measurement deviations that may be caused by universal models, ensuring high accuracy of measurement results. In practical applications, when the designed linewidth of the grid line to be measured is 16μm, if it belongs to the linewidth range of 15μm to 20μm, then a model calibrated for that range is used for measurement. This approach not only improves measurement accuracy but also simplifies the operation process, avoiding the tedious step of predicting the SEM value of each sample before measurement, thus greatly improving measurement efficiency. Through the above embodiments, this invention can more flexibly adapt to the measurement needs of different production lines, providing a solid technical guarantee for large-scale mass production. In other embodiments not shown in the figures, the design linewidth can be divided into more refined intervals, and the calibration model will be established based on these refined intervals, further improving measurement accuracy and applicability.

[0104] Optionally, the method further includes:

[0105] If the morphology parameters are not within the preset parameter range, the line width of the grid line to be measured is measured by the first measurement unit to obtain the measurement result.

[0106] In this embodiment, if the basic morphology of the grating line to be measured is not a preset steep morphology, the linewidth of the grating line will be measured by the first measurement unit to obtain the corresponding measurement result. This supplementary solution ensures that even when facing grating lines with non-standard morphologies, reliable data can be obtained using appropriate measurement methods, thereby expanding the scope of application of the invention beyond the precise measurement of steep morphology grating lines. With this design, regardless of the sidewall angle of the grating line, the most suitable measurement method can be flexibly selected to meet the precise measurement needs of different morphological structures. In practical applications, the first measurement unit can be other optical or non-optical measurement devices adapted to non-steep morphology measurements, such as white light interferometers or atomic force microscopes. These devices can provide high-precision linewidth measurements without damaging the sample, compensating for the limitations of 3D laser microscopes in non-steep morphology measurements, thus ensuring the comprehensiveness and efficiency of the entire measurement system.

[0107] To achieve the above objectives, according to another aspect of this application, a grid line width measuring device is provided, which measures the line width of a grid line using the aforementioned grid line width measuring method.

[0108] The grid linewidth measurement method and apparatus provided in this application, through a systematic calibration process, can accurately correct the systematic errors present in 3D laser microscopy when measuring grid lines with steep sidewalls, greatly improving the accuracy and reliability of the measurement. Compared with traditional methods, this application's solution not only maintains the high efficiency and non-destructive measurement characteristics of 3D laser microscopy, but also achieves alignment between the measurement results and SEM accuracy by introducing a high-precision SEM benchmark, significantly improving measurement efficiency and data quality. Furthermore, this solution optimizes the selection of measurement parameters, ensuring optimal results across different linewidth ranges. This has significant practical value and economic implications for promoting the development of crystalline silicon solar cell technology, especially the industrialization of copper electroplating electrode technology.

[0109] To enable those skilled in the art to better understand the technical solution of this application, the implementation process of the gate line width measurement method of this application will be described in detail below with reference to specific embodiments.

[0110] This invention aims to solve the problem of systematically overestimating linewidth measurements when measuring grating lines on steep sidewalls using 3D laser microscopes. Its core idea is not to attempt to change the physical principles of 3D laser microscopes, but rather to use a systematic calibration method to accurately map its measurements to recognized SEM benchmark values. This achieves SEM-level measurement accuracy while retaining the advantages of high efficiency, non-destructive testing, and three-dimensional measurement capabilities of 3D laser microscopes.

[0111] This application embodiment constructs a complete closed-loop system from "measurement" to "calibration", which mainly includes the following three core components:

[0112] 1. Establishing a Standard Sample Set and Acquiring Reference Data: Prepare or select a set of grating line samples covering the expected measurement range (i.e., the preset linewidth interval) and possessing steep sidewall morphology as a standard sample set (i.e., the grating line sample set). Use a high-precision SEM to measure each grating line (i.e., the first grating line) in this sample set to obtain its reference value for linewidth (i.e., the sample reference value Ws). This step establishes a "true" reference system for the measurement, facilitating the establishment of a calibration model.

[0113] 2.3D Laser Microscope Measurement Parameter Systematic Optimization and Data Acquisition: The above-mentioned standard sample set was scanned using a 3D laser microscope. A series of controllable measurement parameters (i.e., measurement parameters, such as different objective magnifications, pinhole size, and height threshold in the edge detection algorithm) were systematically adjusted to obtain multiple sets of original measurement values ​​(i.e., sample measurement values ​​Wl in this application) for the same grating line under different parameters.

[0114] 3. Constructing a precise mathematical calibration model: Data analysis is performed on the Wl sequences and corresponding Ws sequences obtained under each set of measurement parameters to establish a quantitative relationship model between them. This model can be a calibration formula (Ws=a×Wl+b), a calibration curve, or a lookup table. By comparing the goodness of fit (e.g., R² value) of the model under different parameters, a set of optimal measurement parameters (i.e., the preset measurement parameters in this application) is determined, ensuring that under these parameters, the correlation between the raw data from the 3D laser microscope and the SEM reference values ​​is highest and the deviation is lowest.

[0115] By implementing the aforementioned key technologies, this invention achieves a perfect balance between high precision and high efficiency, fundamentally correcting systematic errors, ensuring data accuracy and reliability, and improving the repeatability and standardization of the measurement process. Once the calibration model is established, when measuring unknown samples, a rapid scan using a 3D laser microscope under preset measurement parameters is sufficient. Then, the original value Wl is substituted into the calibration model to instantly obtain a high-precision calibration value Ws. Its accuracy is comparable to SEM measurement results and far superior to unprocessed 3D laser microscope data. It also eliminates result fluctuations caused by different operators or arbitrary parameter settings. This invention acknowledges and quantifies the systematic errors caused by optical diffraction. By establishing a mathematical model between Wl and Ws, instead of simply multiplying by a correction coefficient, it precisely describes the laws governing systematic errors (including offset b and slope a). Therefore, calibration is dynamic and precise for different linewidths, thus fundamentally correcting systematic errors.

[0116] Specifically, see Figure 4This is a flowchart of a grid linewidth calibration measurement method provided by an embodiment of the present invention. The method mainly includes two stages: a calibration stage and an application measurement stage.

[0117] I. Calibration Phase: The goal is to establish an accurate mathematical model between 3D laser microscope measurements and SEM baseline values.

[0118] 1. Prepare a standard sample set: Prepare or select a set (usually no fewer than 5) of grid line samples with steep sidewall morphology. These samples should cover the expected range of the linewidth to be tested (e.g., 10 μm to 30 μm) to ensure that the calibration model is effective within this range.

[0119] 2. Obtain SEM baseline values ​​(Ws): Measure each grid line in the standard sample set using a high-precision scanning electron microscope (SEM) to obtain its precise linewidth value, denoted as Ws (e.g., Ws1, Ws2…Wsn). This data will serve as the "true" baseline for calibration. To ensure positional consistency, precise positioning marks should be made on the grid lines.

[0120] 3.3D Laser Microscope Parametric Scanning and Data Acquisition: A 3D laser microscope is used to perform three-dimensional scanning of the same batch of standard samples at identical locations. Measurement parameters need to be systematically adjusted to acquire multiple sets of raw data. Selectable measurement parameters include: height thresholds for edge detection algorithms (e.g., thresholds at 20%, 30%, 40%, and 50% height). For each set of parameters, the corresponding raw linewidth value from the 3D laser microscope is measured and denoted as Wl (e.g., for sample 1, Wl-30% is measured at the 30% threshold).

[0121] 4. Constructing and optimizing the calibration model:

[0122] (1) Data association: Perform association analysis on all Wl values ​​obtained under each set of measurement parameters and the corresponding Ws values.

[0123] (2) Model fitting: Perform linear regression analysis on each data sequence to establish a preliminary calibration model, which can be set as Ws=a×Wl+b (a is the slope and b is the intercept).

[0124] (3) Determine the optimal parameter set: Compare the coefficient of determination (R²) and residuals of the fitted models under different parameters. Select the set of parameters that makes R² closest to 1 and has the smallest residual as the optimal measurement parameters. The calibration model corresponding to these parameters is the final optimal calibration model.

[0125] II. Application Measurement Phase: Utilizing the results obtained in the calibration phase, perform rapid and accurate measurements on unknown samples.

[0126] 1. Standard scan of the sample to be measured: For new samples to be measured, use a 3D laser microscope and call up the preset measurement parameters determined in the calibration stage to scan the grid lines to be measured and obtain the original measurement value W. measure .

[0127] 2. Substitute into the model to calculate the true linewidth (i.e., the calibration linewidth): Substitute W measure Substituting into the optimal calibration model (i.e., the formula Ws=a×Wl+b), the calibrated linewidth value W is calculated. final .

[0128] Output the final result: W final This refers to calibrated linewidth measurement results with accuracy comparable to SEM, which can be directly used for process analysis or quality reports.

[0129] Figure 5 This is a scatter plot and calibration curve diagram of standard sample set measurement data provided in an embodiment of the present invention. The horizontal axis represents the original measurement value Wl from the 3D laser microscope, and the vertical axis represents the SEM baseline value Ws. Each data point in the figure represents the measurement value of a standard sample under specific parameters. Curve L1 represents the calibration curve fitted using a 50% height threshold; its data points are relatively dispersed, with an R² of 0.953. Curve L2 represents the calibration curve fitted using a 40% height threshold; its data points are also relatively dispersed, with an R² of 0.968. Curve L3 represents the optimal calibration curve fitted using the "optimal parameters" (30% height threshold) determined in this invention, with the formula Ws = 1.128 × Wl - 9.21. The data points are closely distributed on both sides of the curve, with an R² value as high as 0.990, indicating an excellent linear relationship. This curve is the calibration model used in subsequent measurements.

[0130] This invention provides a measurement scheme for accurately measuring the linewidth of steep-topography grid lines, comprising the following steps: obtaining SEM baseline linewidth values ​​of a standard sample set; measuring the standard sample set using a 3D laser microscope under various measurement parameters to obtain multiple sets of original linewidth values; establishing a calibration model based on the SEM baseline linewidth values ​​and multiple sets of original linewidth values, and determining a set of optimal measurement parameters; measuring the sample to be tested using a 3D laser microscope and the optimal measurement parameters, and substituting the obtained original measurement values ​​into the calibration model to obtain the calibration linewidth values.

[0131] It should be noted that the above are merely illustrative examples and do not specifically limit the methods provided in this application.

[0132] In the description of this application, it should be understood that if terms such as "center", "longitudinal", "lateral", "length", "width", "thickness", "upper", "lower", "front", "rear", "left", "right", "vertical", "horizontal", "top", "bottom", "inner", "outer", "clockwise", "counterclockwise", "axial", "radial", "circumferential" appear, these terms indicate the orientation or positional relationship based on the orientation or positional relationship shown in the accompanying drawings, and are only for the convenience of describing this application and simplifying the description, and do not indicate or imply that the device or element referred to must have a specific orientation, or be constructed and operated in a specific orientation, and therefore should not be construed as a limitation of this application.

[0133] Furthermore, where the terms "first" and "second" appear, these terms are for descriptive purposes only and should not be construed as indicating or implying relative importance or implicitly specifying the number of technical features indicated. Thus, a feature defined with "first" or "second" may explicitly or implicitly include at least one of that feature. In the description of this application, where the term "multiple" appears, "multiple" means at least two, such as two, three, etc., unless otherwise explicitly specified.

[0134] In this application, unless otherwise expressly specified and limited, the terms "installation," "connection," "joining," and "fixing," etc., should be interpreted broadly. For example, they can refer to a fixed connection, a detachable connection, or an integral part; they can refer to a mechanical connection or an electrical connection; they can refer to a direct connection or an indirect connection through an intermediate medium; they can refer to the internal communication of two components or the interaction between two components, unless otherwise expressly limited. Those skilled in the art can understand the specific meaning of the above terms in this application based on the specific circumstances.

[0135] In this application, unless otherwise expressly specified and limited, the use of descriptions such as "above" or "below" the second feature indicates that the first and second features are in direct contact or indirect contact via an intermediate medium. Furthermore, "above," "on top of," and "over" the second feature can mean that the first feature is directly above or diagonally above the second feature, or simply that the first feature is at a higher horizontal level than the second feature. Similarly, "below," "below," and "under" the second feature can mean that the first feature is directly below or diagonally below the second feature, or simply that the first feature is at a lower horizontal level than the second feature.

[0136] It should be noted that if an element is referred to as being "fixed to" or "set on" another element, it can be directly on the other element or there may be an intervening element. If an element is considered to be "connected to" another element, it can be directly connected to the other element or there may be an intervening element. If so, the terms "vertical," "horizontal," "upper," "lower," "left," "right," and similar expressions used in this application are for illustrative purposes only and do not represent the only possible implementation.

[0137] The technical features of the above embodiments can be combined in any way. For the sake of brevity, not all possible combinations of the technical features in the above embodiments are described. However, as long as there is no contradiction in the combination of these technical features, they should be considered to be within the scope of this specification.

[0138] The embodiments described above are merely illustrative of several implementation methods of this application, and while the descriptions are relatively specific and detailed, they should not be construed as limiting the scope of the patent application. It should be noted that those skilled in the art can make various modifications and improvements without departing from the concept of this application, and these all fall within the protection scope of this application. Therefore, the protection scope of this patent application should be determined by the appended claims.

[0139] The above description is merely a preferred embodiment of this application and is not intended to limit this application. Various modifications and variations can be made to this application by those skilled in the art. Any modifications, equivalent substitutions, improvements, etc., made within the spirit and principles of this application should be included within the protection scope of this application.

Claims

1. A method for measuring grid linewidth, characterized in that, The method for measuring the grid line width includes: The topographic parameters of the grid line to be measured are obtained. If the topographic parameters are within a preset parameter range, the measurement parameters of the first measurement unit are adjusted to the preset measurement parameters, and the line width of the grid line to be measured is measured through the first measurement unit to obtain the measured line width. The preset parameter range includes one or more of the following: a preset aspect ratio range of the grid line cross-section, a preset geometric feature range, and a preset sidewall angle range. The measured linewidth is calibrated using a calibration model to obtain the calibrated linewidth of the gate line to be measured; wherein, the calibration model is a mapping relationship between the first measurement unit and the second measurement unit, and the second measurement unit has a higher accuracy in measuring the gate linewidth of the gate line to be measured than the first measurement unit.

2. The method for measuring grid linewidth according to claim 1, characterized in that, Before calibrating the measured linewidth using a calibration model to obtain the calibrated linewidth of the grid line to be measured, the method further includes: The second measurement unit measures the linewidth of each first grid line in the grid line sample set to obtain the sample reference value. The first measurement unit measures the linewidth of each first grid line under multiple measurement parameters to obtain sample measurement values ​​under each measurement parameter. A correlation analysis was performed between the sample measurement values ​​and the corresponding sample benchmark values ​​for each measurement parameter to establish a correlation model for each measurement parameter. Multiple measurement parameters and multiple correlation models are screened and optimized. The selected measurement parameters are used as preset measurement parameters, and the selected correlation models are used as calibration models.

3. The method for measuring grid linewidth according to claim 2, characterized in that, The multiple measurement parameters include a first edge detection threshold and a second edge detection threshold; the association model includes an association formula and an association chart; the step of performing association analysis between the sample measurement value and the corresponding sample benchmark value under each measurement parameter to establish an association model under each measurement parameter includes: A linear regression analysis is performed on the sample measurement values ​​and corresponding sample baseline values ​​based on the first edge detection threshold to obtain the first correlation formula; A second correlation formula is obtained by performing linear regression analysis on the sample measurement values ​​and corresponding sample baseline values ​​based on the second edge detection threshold. A fitting analysis is performed on the sample measurement values ​​and corresponding sample baseline values ​​based on the first edge detection threshold to obtain a first correlation chart; A fitting analysis is performed on the sample measurement values ​​and corresponding sample baseline values ​​based on the second edge detection threshold to obtain a second correlation chart.

4. The method for measuring grid linewidth according to claim 2, characterized in that, The process involves filtering and optimizing multiple correlation models, including: Under the same measurement parameters, each second grid line in the grid line verification set is measured to obtain the measured line width of each second grid line; The measured linewidth of each second gate line is calibrated using multiple correlation models to obtain the calibrated linewidth of the second gate line for each correlation model. The true linewidth of each second gate line is obtained, and the calibration linewidth of each associated model is compared with the true linewidth to obtain a comparison result for each associated model; wherein, the comparison result is used to characterize the calibration accuracy of the associated model; Based on the comparison results, the calibration accuracy of the multiple associated models is sorted in the first order, and the associated model ranked first in the first order is selected as the calibration model.

5. The method for measuring grid line width according to claim 2, characterized in that, Optimization and selection of multiple measurement parameters are performed, including: The sample measurement values ​​under multiple measurement parameters are calibrated using the calibration model to obtain sample predicted values ​​under multiple measurement parameters. Based on the sample predicted values ​​of multiple measurement parameters and the corresponding sample reference values, the evaluation values ​​of multiple measurement parameters are determined; wherein, the evaluation values ​​are used to characterize the calibration accuracy of the calibration model under different measurement parameters. The evaluation values ​​of the plurality of measurement parameters are sorted in a second order, and the measurement parameter corresponding to the first evaluation value in the second order is selected as the preset measurement parameter.

6. The method for measuring grid linewidth according to claim 5, characterized in that, The calibration model includes a first correlation formula, wherein determining the evaluation values ​​of the multiple measurement parameters based on sample predicted values ​​and corresponding sample benchmark values ​​of the multiple measurement parameters includes: Based on the sample predicted values ​​of the plurality of measurement parameters and the corresponding sample benchmark values, the first determination coefficient of the first association formula is determined, and the first determination coefficient is used as the evaluation value. Alternatively, based on the sample predicted values ​​of multiple measurement parameters and the corresponding sample benchmark values, the first residual of the first association formula is determined, and the first residual is used as the evaluation value.

7. The method for measuring grid line width according to claim 2, characterized in that, Before measuring the linewidth of each first grid line under multiple measurement parameters using the first measurement unit to obtain sample measurement values ​​under each measurement parameter, the method further includes: A measurement position is marked on each of the first grid lines so that the first measurement unit can perform linewidth measurement on each measurement position of the first grid line under multiple measurement parameters; wherein, the measurement position is the measurement position where the first measurement unit performs linewidth measurement.

8. The method for measuring grid line width according to claim 2, characterized in that, The method further includes: Obtain the design linewidth of the grid line to be measured; The calibration model is determined based on the preset linewidth range to which the designed linewidth belongs.

9. The method for measuring grid line width according to claim 2, characterized in that, The method further includes: If the morphology parameters are not within the preset parameter range, the line width of the grid line to be measured is measured by the first measurement unit to obtain the measurement result.

10. A grid linewidth measuring device, characterized in that, The grid line width measuring device measures the grid line width using the grid line width measuring method as described in any one of claims 1 to 9.

Images