A photolithography method and system for manufacturing a display panel

Abstract

This invention provides a photolithography method and system for manufacturing display panels, relating to the field of photolithography technology. The invention involves cleaning, applying a base coat, and dividing the display panel substrate into zones to form imageable micro-markers covering all exposure zones. A lower support adhesive and an upper imaging adhesive are then sequentially formed, and their thickness and uniformity are measured. In-situ data acquisition and modeling are performed using the micro-markers as imaging objects, generating exposure correction instructions and an overlap energy table. Digital micromirrors are used for zoned exposure and real-time correction, followed by development to obtain etch-resistant windows with steep sidewalls. Based on exposure records and window quality data, etching windows are set, and low-damage etching and defect restoration are performed to generate stability evaluation data. This method achieves a comprehensive improvement in high-precision exposure, pattern consistency, and electrical stability.

Description

Technical Field

[0001] This invention relates to the field of photolithography technology, and in particular to a photolithography method and system for manufacturing display panels. Background Technology

[0002] Currently, mass production of display panels still primarily relies on multi-step photolithography processes, including photoresist coating, exposure, development, and etching. For large-size glass substrates, mainstream factories typically use stepper or scanning projection lithography equipment to complete pattern transfer, while pre-setting alignment marks on the substrate for position correction and distortion estimation. As process nodes advance towards higher resolutions, some processes are beginning to introduce partitioned exposure methods using digital micromirror devices to improve pattern flexibility and local correction capabilities. After pattern transfer, dry anisotropic etching is often used to form the thin film structure, supplemented by annealing and passivation steps to improve device stability. For oxide semiconductor backplanes, color filters, and color conversion layers, the industry has accumulated relatively mature material and process combinations, but challenges remain in areas such as large-area consistency, stacking accuracy, and long-term reliability.

[0003] On the one hand, both industry and academia are strengthening online metrology and high-order distortion compensation capabilities. Specifically, this involves increasing the density and frequency of in-situ measurements, employing high-order models incorporating amplification, rotation, orthogonality bias, and radial and tangential nonlinearities, and writing these models back to the stage step size and projection mapping in real time. Simultaneously, energy gradation and weight allocation are implemented in the overlapping seam regions of adjacent partitions to reduce seam errors and improve the continuity of partition splicing. On the other hand, aiming for both board size reduction and reliability, process integration focuses more on using double-layer photoresist to form steep sidewalls and etch-resistant supports, coupled with low-damage anisotropic etching conditions and post-etching defect recovery sequences. This allows for the reduction of conductive overlap areas and suppression of parasitic capacitance while maintaining stable mobility and threshold. These two approaches are gradually evolving towards synergy, that is, coupling online compensation, pattern transfer, and defect repair within the same photolithography-etching process chain to achieve higher yields and consistency.

[0004] Despite existing methods for partitioned exposure and distortion compensation, many solutions still rely on offline analysis or low-order models, making it difficult to promptly address the dynamic deformation of large-size glass during handling and heating. Overlapping errors in the seam area often become a key factor limiting yield. Self-aligned and double-layer photoresist solutions adopted to reduce batches are prone to plasma damage and interface traps if they lack low-damage etching windows for oxide semiconductors and systematic subsequent recovery steps. This leads to decreased mobility and significant threshold drift under positive and negative bias temperature stress conditions. Furthermore, current processes often involve segmented optimization of online metrology and pattern transfer, lacking closed-loop control based on measurable indicators. This makes it difficult to continuously correct errors within the same batch and quickly transfer experience to the initial parameter settings of the next batch, leaving room for improvement in process stability and cross-batch consistency. Summary of the Invention

[0005] To overcome the shortcomings of existing technologies, the purpose of this invention is to provide a photolithography method and system for manufacturing display panels. By establishing a closed-loop data process encompassing partitioned exposure, in-situ modeling, and low-damage etching, the invention achieves a comprehensive effect of high exposure accuracy, good pattern consistency, and strong electrical stability during the manufacturing process of display panels.

[0006] To achieve the above objectives, the present invention provides the following solution:

[0007] A photolithography method for manufacturing a display panel includes:

[0008] The surface of the display panel substrate to be coated is cleaned and primed. The effective area of ​​the display panel substrate is divided into several exposure zones according to the projection field of view and the travel of the motion platform. An overlapping zone with a preset width is set between adjacent exposure zones as a seam area, and a partition definition table containing partition boundary numbers and seam area widths is generated.

[0009] Imageable micromarks covering all exposure zones are formed on the display panel substrate. A lower layer of support adhesive and an upper layer of imaging adhesive are coated sequentially, and the thickness and uniformity are measured to obtain the substrate and adhesive layer parameters.

[0010] Using the imageable micro-marker as the imaging object, in-situ optical acquisition is performed partition by partition according to the partition definition table. The spatial coordinates of the imageable micro-marker are extracted and a micro-marker coordinate map and a substrate warping map are established. The parameters of the higher-order distortion model are fitted online in combination with the parameters of the substrate and the adhesive layer. Based on the higher-order distortion model parameters, exposure correction instructions for motion and projection control and an overlap energy table for the seam area are generated.

[0011] A digital micromirror is used to perform exposure partition by partition according to the partition definition table. Mid-section self-correction and stitching overlap control are performed according to the exposure correction command and the overlap energy table. Before, during and after exposure in each exposure partition, in-situ measurement is repeated and the micro-marker coordinate map and the higher-order distortion model parameters are updated to form an exposure record of dose, focal length and alignment residual. Post-baking and development conditions are set according to the substrate and resist layer parameters and development is completed to obtain an etch-resistant window with steep sidewalls. The critical size, sidewall angle and morphological defects of the etch-resistant window are measured to form window quality data.

[0012] Based on the exposure record and the window quality data, the etching window setting is given. Under low-damage anisotropic conditions, the pattern is transferred to the target thin film layer. Etching process monitoring data is collected to fine-tune the power, atmosphere ratio, working pressure and substrate temperature. Based on the etching process monitoring data, the defect restoration sequence is selected and executed. Mobility, threshold and leakage current are measured under bias and temperature stress conditions to form stability evaluation data.

[0013] Preferably, it further includes:

[0014] The higher-order distortion model parameters, the exposure correction instructions, the overlap energy table, the exposure record, the etching window settings, and the stability evaluation data are correlated and analyzed, and the starting input parameters for the next batch are generated by combining the partition definition table with the substrate and adhesive layer parameters.

[0015] Preferably, the surface of the display panel substrate to be coated is cleaned and primed. The effective area of ​​the display panel substrate is divided into several exposure zones according to the projection field of view and the travel of the motion platform. An overlapping band with a preset width is set between adjacent exposure zones as a seam area. A partition definition table containing partition boundary numbers and seam area widths is generated, including:

[0016] The display panel substrate is positioned to determine the substrate reference point and boundary, obtain the outline and size of the effective area, and output the effective area data and coordinate reference data.

[0017] Complete the cleaning and primer treatment of the surface to be coated, and record the surface condition data; the surface condition data includes at least cleanliness, primer thickness and uniformity;

[0018] Obtain the projection field of view size parameters and motion platform travel parameters, and calculate the minimum coverage partitioning scheme by combining the effective area data, outputting a partitioned grid draft; the expression of the partitioned grid draft is: ;

[0019] in, The horizontal dimension of the effective area; The effective area is the dimension in the vertical direction; This represents the horizontal dimension of the field of view in a single projection. This represents the vertical dimension of the field of view in a single projection. The preset overlap width in the horizontal direction for adjacent exposure zones; The preset overlap width of adjacent exposure zones in the vertical direction; The step size of the partitioned grid in the horizontal direction is equal to the field size minus the overlap width; The vertical step size of the partitioned grid is equal to the field size minus the overlap width. This represents the number of horizontal partitions, and coverage is ensured by rounding up. The number of vertical partitions is determined by rounding up to ensure coverage. and The coordinates of the reference points for the effective area are used to determine the grid start point; For the first Liede The coordinates of the pivot point of the row partition; The coordinate set of the draft partitioned grid; The actual projected field of view rectangular area corresponding to the pin point; the coverage condition is used to verify that the grid completely covers the effective area and is usually satisfied. and ;

[0020] The preset width of the overlap band is determined based on the target alignment accuracy and the superposition error tolerance, and the seam width rules are output.

[0021] Based on the proposed partition grid, the boundaries of adjacent partitions are corrected according to the stitching width rules to ensure full coverage without gaps or excessive overlap. Each partition is assigned a unique number according to the exposure path, and the partition boundary and number data are output.

[0022] The partition definition table is generated by integrating the partition boundaries, numbering data, and seam width rules. It is then verified by calculating the coverage ratio and seam consistency to form a verified partition definition table.

[0023] Preferably, imageable micromarks covering all exposure zones are formed on the display panel substrate, and a lower support adhesive and an upper imaging adhesive are sequentially coated, with thickness and uniformity measured to obtain substrate and adhesive layer parameters, including:

[0024] Based on the partition definition table, the shape, size and spacing of the imageable micro-markers are determined, so that each exposure partition and the stitching area of ​​the exposure partition contains a marker array for in-situ identification, and the micro-marker layout and deployment data are output.

[0025] Imageable micromarks are formed on the surface of the display panel substrate according to the micromark pattern. The micromark contrast and edge contour are acquired by calibration imaging to obtain micromark imaging quality data.

[0026] A lower support adhesive material is uniformly coated on the display panel substrate and cured to form a lower support adhesive film layer. The film thickness and in-plane uniformity of the lower support adhesive film layer are measured, and the lower support adhesive film thickness data and the lower support adhesive in-plane uniformity data are output.

[0027] An upper imaging adhesive material is uniformly coated onto the lower support adhesive film layer and cured to form an upper imaging adhesive film layer. The film thickness and in-plane uniformity of the upper imaging adhesive film layer are measured, and the upper imaging adhesive film thickness data and the upper imaging adhesive in-plane uniformity data are output. These data are then combined with the lower support adhesive film thickness data and the lower support adhesive in-plane uniformity data to generate composite data of adhesive layer thickness and uniformity.

[0028] The substrate and adhesive layer parameters are generated by summarizing the micromarker layout and deployment data, the micromarker imaging quality data, and the adhesive layer thickness and uniformity composite data.

[0029] Preferably, using the imageable micro-marker as the imaging object, in-situ optical acquisition is performed partition by partition according to the partition definition table. The spatial coordinates of the imageable micro-marker are extracted, and a micro-marker coordinate map and a substrate warpage map are established. High-order distortion model parameters are fitted online based on the substrate and adhesive layer parameters. Exposure correction instructions for motion and projection control and an overlap energy table for the seam area are generated based on the high-order distortion model parameters, including:

[0030] Based on the partition definition table and the micro-marker layout, the location and order of the acquisition points in each exposure partition are determined, and acquisition point plan data is output to guide in-situ optical acquisition partition by partition;

[0031] Image acquisition is performed on each of the exposure zones according to the acquisition point plan data, and an in-situ image set is output.

[0032] Micromarks are identified in the in-situ image set and their spatial coordinates in the substrate coordinate system are obtained. A micromark coordinate map organized by partition is formed by combining the coordinate reference data, and the difference field between the nominal position and the measured position is calculated.

[0033] Based on the difference field fitting, the deformation distribution in the plane and the micro-out plane is obtained, and the substrate warping diagram is output.

[0034] The micro-marker coordinate map and the substrate warpage map are combined with the substrate and adhesive layer parameters for robust fitting to obtain high-order distortion model parameters including magnification, rotation, orthogonality deviation, radial and tangential terms, and to generate fitting residual statistics.

[0035] Based on the higher-order distortion model parameters and the fitting residual statistics, exposure correction instructions for motion and projection control are calculated, and an overlapping energy table is generated in the stitching area according to the local residual and the stitching width rule, so that the exposure dose in the overlapping band changes monotonically along the stitching normal, and the sum of the doses of the adjacent partitions in the overlapping band is within the tolerance range of the preset target dose.

[0036] Preferably, a digital micromirror is used to perform exposure zone by zone according to the zone definition table. Mid-section self-correction and stitching overlap control are performed based on the exposure correction command and the overlap energy table. Before, during, and after exposure in each zone, in-situ measurement is repeated, and the micro-marker coordinate map and the higher-order distortion model parameters are updated to form an exposure record containing the dose, focal length, and alignment residuals. Post-baking and development conditions are set according to the substrate and adhesive layer parameters, and development is completed to obtain an etch-resistant window with steep sidewalls. The critical dimensions, sidewall angles, and morphological defects of the etch-resistant window are measured to form window quality data, including:

[0037] Before the current exposure zone enters the exposure phase, an in-situ image is acquired, the micro-marker coordinate map of the current exposure zone is updated, the alignment deviation and in-field distortion residual are calculated, the zone correction increment is obtained, and the step size, projection magnification and projection mapping parameters are corrected according to the zone correction increment to generate a pre-exposure correction record.

[0038] Within the seam area, the exposure dose is monotonically gradually varied along the seam normal according to the overlapping energy table. Exposure is completed in the main area of ​​the current exposure zone according to the target dose. During the exposure process, the step size and scanning speed are finely adjusted according to the zone correction increment, and the zone exposure data is output.

[0039] When half of the exposure in a zone has been completed, an in-situ image is acquired again, the parameters of the higher-order distortion model are updated, and the secondary correction increment is calculated. The secondary correction increment is written back to the remaining trajectory of the current exposure zone and the starting compensation of the next zone to form a mid-section self-correction record.

[0040] After the zonal exposure is completed, in-situ metrology is performed and the micro-marker coordinate map and the higher-order distortion model parameters are updated. The zonal exposure data, the pre-exposure correction record and the mid-section self-correction record are summarized to generate an exposure record entry containing dose, focal length and alignment residual, and archived as the exposure record according to the zonal number.

[0041] The substrate and adhesive layer parameters and the exposure record entries are called to determine the post-baking temperature and time, as well as the developer ratio and development time. Post-baking and development are performed to obtain the etching resistance window of the current exposure zone.

[0042] The critical size, sidewall angle and surface morphology of the etching-resistant window are measured to obtain initial quality data. The initial quality data of each exposure zone are associated and merged with the exposure record according to the zone number to form the window quality data.

[0043] Preferably, the etching window is set based on the exposure record and the window quality data. The pattern is transferred to the target thin film layer under low-damage anisotropic conditions. Etching process monitoring data is collected to fine-tune power, atmosphere ratio, working pressure, and substrate temperature. A defect recovery sequence is selected and executed based on the etching process monitoring data. Mobility, threshold current, and leakage current are measured under bias and temperature stress conditions to form stability evaluation data, including:

[0044] a) Call the exposure record and the window quality data, and combine them with the tolerance of the target sidewall angle and critical size to calculate the power setting value and tolerance range, the atmosphere ratio setting value and tolerance range, the working pressure setting value and tolerance range, and the substrate temperature setting value and tolerance range, and output the etching window setting table.

[0045] b) Based on the etching window setting table, perform pattern transfer, collect endpoint signals, removal rate, sidewall angle and substrate temperature, generate etching process monitoring data, and when any monitored quantity exceeds the corresponding tolerance range, calculate the power fine-tuning increment, atmosphere ratio fine-tuning increment, working pressure fine-tuning increment and substrate temperature fine-tuning increment respectively and implement fine-tuning to form a parameter adjustment record.

[0046] c) Calculate the proxy amount of ion energy and the proxy amount of temperature rise based on the etching process monitoring data, compare them with the preset low damage threshold, generate a low damage criterion report, and return to step b) to perform fine-tuning and continue etching;

[0047] d) Select the type and order of the defect restoration sequence based on the etching process monitoring data and low damage criterion report, and provide the set values ​​and tolerance ranges for restoration atmosphere, restoration temperature and restoration time, and output the defect restoration selection table.

[0048] e) Execute the defect restoration sequence, record the temperature curve and time curve, and generate a restoration execution record;

[0049] f) Based on the etching window setting table, formulate a bias and temperature stress test plan, perform mobility, threshold and leakage current measurements on the sample or array, and generate stability evaluation data;

[0050] g) The etching window setting table, the etching process monitoring data, the parameter adjustment record, the low damage criterion report, the defect restoration selection table, the restoration execution record, and the stability evaluation data are merged into a graphic transfer data package; the graphic transfer data package is used as the input for batch-level closed-loop update and as the starting input parameter for the next batch etching window setting.

[0051] A photolithography system for manufacturing display panels, comprising:

[0052] The partition definition and substrate pretreatment module is used to clean and apply a base coat to the surface of the display panel substrate to be coated. The effective area of ​​the display panel substrate is divided into several exposure partitions according to the projection field of view and the travel of the motion platform. An overlapping band with a preset width is set between adjacent exposure partitions as a seam area, and a partition definition table containing partition boundary numbers and seam area widths is generated.

[0053] The micro-marker placement and double-layer photoresist film formation module is used to form imageable micro-markers covering all the exposure zones on the display panel substrate, and sequentially coat the lower layer of support adhesive and the upper layer of imaging adhesive and measure the thickness and uniformity to obtain the substrate and adhesive layer parameters.

[0054] The in-situ metrology and high-order distortion modeling module is used to perform in-situ optical acquisition partition by partition according to the partition definition table, using the imageable micro-marker as the imaging object, extracting the spatial coordinates of the imageable micro-marker and establishing a micro-marker coordinate map and a substrate warping map, combining the substrate and adhesive layer parameters to fit the high-order distortion model parameters online, and generating exposure correction instructions for motion and projection control and an overlap energy table for the seam area according to the high-order distortion model parameters.

[0055] The partitioned exposure and development control module is used to perform exposure partition by partition using a digital micromirror according to the partition definition table, perform mid-section self-correction and stitching overlap control according to the exposure correction command and the overlap energy table, repeat in-situ measurement and update the micro-marker coordinate map and the higher-order distortion model parameters before, during and after exposure in each exposure partition, form an exposure record recording dose, focal length and alignment residual, set post-baking and development conditions according to the substrate and adhesive layer parameters and complete development to obtain an etch-resistant window with steep sidewalls, and measure the critical size, sidewall angle and morphological defects of the etch-resistant window to form window quality data;

[0056] The low-damage etching and defect recovery module is used to set the etching window based on the exposure record and the window quality data, transfer the pattern to the target thin film layer under low-damage anisotropic conditions, collect etching process monitoring data to fine-tune the power, atmosphere ratio, working pressure and substrate temperature, select and execute the defect recovery sequence based on the etching process monitoring data, and measure the mobility, threshold and leakage current under bias and temperature stress conditions to form stability evaluation data.

[0057] The present invention discloses the following technical effects:

[0058] (1) This invention divides the exposure zones on the display panel substrate according to the projection field of view and the travel of the motion platform, and sets an overlap zone with a preset width between adjacent exposure zones as a seam area. This invention can establish a clear spatial partition structure on a large-size substrate. This structure enables subsequent exposure operations to have independent control boundaries within each zone, avoids the accumulation of cross-zone superposition errors, and achieves traceable management of zone boundary and seam area data through the generation of a partition definition table, thereby significantly improving the consistency of exposure path planning and seam.

[0059] (2) This invention forms imageable micro-markers covering the entire exposure zone on the display panel substrate, and sequentially coats a lower support adhesive and an upper imaging adhesive, measuring their thickness and uniformity. This invention establishes a quantitative characterization of the adhesive layer thickness and in-plane uniformity before exposure. This step ensures the consistency of the depth of focus and the stability of the imaging signal during subsequent in-situ optical acquisition, making the identification coordinates of the micro-markers accurate and reliable, thereby reducing optical focal length drift and alignment errors caused by film thickness fluctuations, and providing a stable film formation basis for high-precision exposure.

[0060] (3) This invention uses imageable micro-markers as imaging objects, performs in-situ optical acquisition partition by partition, extracts spatial coordinates and establishes micro-marker coordinate maps and substrate warping maps, and combines substrate and adhesive layer parameters to fit high-order distortion model parameters online. This invention can obtain a complete model including magnification, rotation, orthogonality deviation and polynomial distortion. Based on the exposure correction instructions and overlapping energy table generated by this model, partition-level projection motion correction and stitching energy distribution control are realized, so that fine compensation can be performed in the exposure stage based on the real-time fitting results, thereby improving the pattern alignment accuracy and the energy continuity of partition stitching.

[0061] (4) This invention achieves dynamic self-correction during the exposure stage by using a digital micromirror to perform exposure in sections and updating the micro-marker coordinate map and high-order distortion model parameters multiple times in-situ before, during, and after exposure. The etching window is set based on the exposure record and window quality data, and pattern transfer is completed under low-damage anisotropy conditions. Combined with etching process monitoring and defect restoration operations, stability evaluation data for mobility, threshold, and leakage current are finally generated. This process chain ensures parameter transfer and quality feedback throughout the entire process from exposure and development to etching, resulting in more stable pattern feature size control, more complete etching boundaries, and higher device consistency. Attached Figure Description

[0062] To more clearly illustrate the technical solutions in the embodiments of the present invention or the prior art, the drawings used in the embodiments will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0063] Figure 1 A flowchart of the method provided in an embodiment of the present invention;

[0064] Figure 2 This is a schematic diagram of the system structure provided in an embodiment of the present invention. Detailed Implementation

[0065] The technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0066] The purpose of this invention is to provide a photolithography method and system for manufacturing display panels, which realizes precise zoning control and dynamic process correction throughout the entire photolithography manufacturing process of display panels, making exposure splicing more accurate, pattern forming more stable, and etching damage less, thereby improving the overall process consistency and reliability of display panels.

[0067] To make the above-mentioned objects, features and advantages of the present invention more apparent and understandable, the present invention will be further described in detail below with reference to the accompanying drawings and specific embodiments.

[0068] Figure 1 The method flowchart provided in the embodiments of the present invention is as follows: Figure 1 As shown, the present invention provides a photolithography method for manufacturing a display panel, comprising:

[0069] Step 100: Clean and apply a base coat to the surface of the display panel substrate to be coated. Divide the effective area of ​​the display panel substrate into several exposure zones according to the projection field of view and the travel of the motion platform. Set an overlapping zone with a preset width between adjacent exposure zones as a seam area. Generate a partition definition table containing partition boundary numbers and seam area widths.

[0070] Step 200: Form imageable micro-markers covering all the exposure zones on the display panel substrate, sequentially coat the lower support adhesive and the upper imaging adhesive, and measure the thickness and uniformity to obtain the substrate and adhesive layer parameters.

[0071] Step 300: Using the imageable micro-marker as the imaging object, perform in-situ optical acquisition partition by partition according to the partition definition table, extract the spatial coordinates of the imageable micro-marker and establish a micro-marker coordinate map and a substrate warping map, combine the substrate and adhesive layer parameters to fit the high-order distortion model parameters online, and generate exposure correction instructions for motion and projection control and an overlap energy table for the seam area according to the high-order distortion model parameters.

[0072] Step 400: Using a digital micromirror, exposure is performed partition by partition according to the partition definition table. Mid-section self-correction and stitching overlap control are performed according to the exposure correction command and the overlap energy table. Before, during and after exposure in each exposure partition, in-situ measurement is repeated and the micro-marker coordinate map and the higher-order distortion model parameters are updated to form an exposure record of recorded dose, focal length and alignment residual. Post-baking and development conditions are set according to the substrate and adhesive layer parameters and development is completed to obtain an etch-resistant window with steep sidewalls. The critical size, sidewall angle and morphological defects of the etch-resistant window are measured to form window quality data.

[0073] Step 500: Based on the exposure record and the window quality data, set the etching window, transfer the pattern to the target thin film layer under low-damage anisotropic conditions, collect etching process monitoring data to fine-tune the power, atmosphere ratio, working pressure and substrate temperature, select and execute the defect restoration sequence based on the etching process monitoring data, and measure the mobility, threshold and leakage current under bias and temperature stress conditions to form stability evaluation data.

[0074] Specifically, in step 100 of this embodiment, the display panel substrate is fixed on the support platform. A reference origin, two orthogonal reference axes, and the four-sided contour of the substrate are determined by a combination of edge positioning pins and visual recognition. The identified outline polygon and its horizontal and vertical dimensions are recorded as valid area data, and the coordinates of the reference origin, the directions of the two reference axes, and the measurement resolution are recorded as coordinate reference data. The above two types of data serve as inputs for subsequent partitioning and in-situ acquisition, ensuring that the partition boundaries and coordinate solutions share the same coordinate system and do not cause ambiguity in reference.

[0075] In this embodiment, the surface to be coated is cleaned and primed according to a predetermined process, and rapid characterization is performed after the steps are completed. The residual particle density, the determination results of surface watermarks or organic residues, the thickness of the prime coating and its in-plane uniformity are statistically analyzed as surface condition data. Among them, the thickness of the prime coating is measured using non-contact thickness measurement, and the in-plane uniformity is expressed as the percentage of the deviation between the maximum and minimum values ​​of the full-width measurement points relative to the average value. This surface condition data is used for the preliminary verification of subsequent development conditions and in-situ imaging quality, and also serves as a traceable record when solving the partitioned mesh draft and seam width rules.

[0076] This embodiment reads the single field of view size of the projection system and the single travel limit of the stage, and performs layout within the coordinate system of the effective area data and coordinate reference data: First, the partition step distances in the horizontal and vertical directions are determined, and the partition step distance is defined as "single field of view size minus preset overlap width"; then, a set of partition anchor point coordinates is generated in the order of "starting from the reference origin, translating one partition step distance horizontally until the effective area is covered, then translating one partition step distance vertically to cover the next row", and each anchor point is associated with a rectangular area of ​​the same size as the single field of view; when the last column or the last row has an edge residue, an upward rounding strategy is used to add a column or row so that the rectangular area and the overlap band jointly cover the boundary of the effective area without leaving any blanks. The set of anchor point coordinates and the set of rectangular areas formed above are called the partition grid draft, which includes the anchor point coordinates of each partition, the corresponding rectangular coverage area, and the row and column indexes, for subsequent boundary correction and numbering calls.

[0077] This embodiment determines the overlap band according to the following seam width rules: the overlap width shall not be less than the sum of six times the composite standard deviation of the superposition error obtained from in-situ measurement of the same batch and the edge forming margin; the composite standard deviation is obtained by taking the square root of the sum of the variances of in-situ alignment error, repeatability error, thermal deformation residual, and projection distortion residual; the edge forming margin is determined by the upper limit of the allowable deviation of the critical size of the anti-etching window after development; the upper limit of the overlap width shall not exceed one-fifth of the single field of view size in the corresponding direction, and shall not be greater than the partition step distance in that direction; the overlap width shall be rounded up to an integer multiple of the common integer multiple of the projected pixel size and the minimum step distance of the stage, and shall not be less than sixteen times the projected pixel size, to ensure sufficient subsequent energy gradient sampling points. Based on these rules, the adjacent partition boundaries of the partition grid draft are corrected to ensure "full coverage, no blanks, and no excessive overlap," and unique numbers are assigned row by row and column by column according to the actual exposure path starting from the reference origin, outputting the partition boundary and number data. The partition boundary and number data are integrated with the seam width rules to generate a partition definition table (as shown in Table 1). Then, the coverage ratio and seam consistency are used for trial calculation and verification. After the verification is passed, the partition definition table is used as the direct input for subsequent in-situ acquisition and partition exposure, and the verification results are recorded for traceability.

[0078] Table 1 Partition Definition Table

[0079]

[0080] Furthermore, in this embodiment, after obtaining the field of view size of a single projection and the movable travel of the motion platform, the minimum partitioning scheme that can achieve complete coverage is calculated in conjunction with the effective area data of the display panel substrate. Specifically, the spacing between each exposure partition in the horizontal and vertical directions is first determined. This spacing is equal to the size of the single projection field of view minus the preset overlap width of adjacent exposure areas in that direction. Then, the required number of partitions is obtained by dividing the horizontal and vertical dimensions of the effective area by the aforementioned spacing and rounding up, ensuring that all partitions can completely cover the effective area after splicing. Next, using the reference origin of the effective area as the starting point, the anchor point coordinates of each partition are calculated sequentially in row and column order, ensuring that each partition is continuously arranged in the horizontal and vertical directions. When edge margins appear, compensation is made by adding a column or row to ensure that no blank areas appear in the effective area. The actual exposure rectangle range of each partition is determined based on the coordinates of each anchor point and the size of the single projection field of view, forming a draft partition grid consisting of a set of partition coordinates and corresponding rectangular areas, used for subsequent boundary correction and numbering. After completing the draft partition grid, it is necessary to verify the coverage of the effective area by the partition layout. The coverage condition requires that the horizontal stitching length plus the corresponding overlap width of all partitions be no less than the horizontal dimension of the effective area, and the vertical stitching length plus the corresponding overlap width be no less than the vertical dimension of the effective area. If this condition is not met, the number of partitions is automatically increased until complete coverage is achieved. The horizontal and vertical dimensions of the effective area are obtained by substrate positioning measurements; the field of view of a single projection is given by the optical parameters of the projection system; the overlap width is determined by process requirements and is usually smaller than the field of view of a single projection; the anchor point coordinates are calculated sequentially from the reference origin and the partition spacing. This calculation method ensures that all partitions are continuously connected in the horizontal and vertical directions, without omissions or unnecessary overlaps, thus forming a standardized spatial partitioning dataset for exposure.

[0081] In this embodiment, the effective area of ​​the display panel substrate can have a horizontal dimension of 1500 and a vertical dimension of 1850; the horizontal and vertical dimensions of a single field of view can be 50 and 50 respectively; the upper limit of the single stroke of the support stage can be 520 and 520; the reference origin is set at the lower left corner of the effective area; the lower limit of the overlap width is determined by six times the composite standard deviation plus the edge forming margin. If the composite standard deviation of the in-place alignment error, repeat positioning error, thermal deformation residual, and projection distortion residual is 0.08, and the edge forming margin is 0.12, then the lower limit of the overlap width is 0.60; the upper limit of the overlap width is limited to 20% of the single field of view size, corresponding to 10; the overlap width also needs to be rounded up to the projection distance. The projection pixel size and the minimum step size of the support stage are both integer multiples of each other. If the projection pixel size is 0.01 and the minimum step size of the support stage is 0.1, then the common integer multiple is 0.1. The final overlap width can be adjusted upwards to the nearest value of 0.6 or 0.7. The partition step size is calculated by subtracting the overlap width from the single field of view size. If the overlap width is 0.6, then the partition step size is 49.4. The calculated number of columns and rows is rounded up to ensure coverage. For example, 31 columns are needed horizontally and 38 rows are needed vertically. The representative target range for the base coating thickness is 50 to 200 mm, and the in-plane uniformity target is better than ±5%. The particle residual density control target is less than 0.5 mm per square millimeter. The above values ​​are only representative settings for this embodiment. Applicants can make equivalent adjustments under different equipment and size conditions without departing from the principle of this embodiment.

[0082] Specifically, in step 200 of this embodiment, based on the preceding partition definition table, an array of imageable micro-markers for in-situ identification is deployed in the central region and seam area of ​​each exposure partition. The micro-markers employ a high-contrast structure combining cross, swastika, or ring shapes. Their size and spacing are determined according to the partition size and subsequent imaging resolution, ensuring that each partition and its seam contain at least three independently identifiable marker points. The markers are formed by patterned etching after depositing a thin metal film on the substrate surface, or by local refractive index differences in a transparent phase structure; both aim to provide stable edge contours and sufficient imaging contrast. After marker formation, a calibration imaging is performed, and the resulting contrast, edge sharpness, and position repeatability are extracted as micro-marker imaging quality data. This data, along with the marker's shape, size, coordinates, and their correspondence with the partition number, is compiled into a micro-marker layout and deployment data for input in subsequent in-situ metrology and modeling steps.

[0083] In this embodiment, a lower support adhesive material is uniformly applied to a substrate with micromarks using a slot coating method. After pre-baking and curing, a lower support adhesive film layer providing sidewall support and etching resistance is obtained. To quantify the contribution of this film layer to the stability of subsequent pattern transfer, non-contact film thickness scanning is performed across the entire area using a regular grid, and at least one data point near the seam is collected in each exposure zone. The in-plane uniformity is calculated using the maximum, minimum, and average values ​​of the full-area measurement points, and output as lower support adhesive film thickness data and lower support adhesive in-plane uniformity data. A one-to-one correspondence is established between these two types of data and the zone number, which is used both for setting the target film thickness of the upper imaging adhesive and for pre-verification of the time and temperature settings during development.

[0084] Above the lower support film layer, the upper imaging adhesive material is applied using the same slit coating method. After pre-baking and curing, the upper imaging adhesive film layer for exposure imaging is obtained. A full-width film thickness scan is performed on this film layer, and data is collected at the center of each exposure zone and near the seam. The upper imaging adhesive film thickness data and the uniformity data within the upper imaging adhesive surface are obtained using the same statistical method. Subsequently, the film thickness and uniformity of the upper and lower layers are merged according to the zone number to obtain composite data describing the adhesive layer thickness and uniformity of the entire photolithography stack. This composite data is then combined with the micromarker layout and placement data, and the micromarker imaging quality data to form the substrate and adhesive layer parameters used for subsequent in-situ metrological modeling, exposure focal length setting, and development condition setting. These parameters are stored in the form of a data table and are called as the initial input at the beginning of each batch.

[0085] In a typical implementation, the micro-markers use a combination of cross and ring markers. Each marker has an outer diameter of 30 micrometers and a line width of 3 micrometers. The marker spacing within each zone is 10 millimeters, and the marking density within the seam area is reduced to 2 millimeters. The imaging contrast threshold is not less than 0.6, and the edge transition width is not greater than 1 micrometer. The target film thickness of the lower support adhesive is 1.5 micrometers, with in-plane uniformity better than ±3%, and a pre-baking temperature of 110 degrees Celsius for 120 seconds. The target film thickness of the upper imaging adhesive is 1.2 micrometers, with in-plane uniformity better than ±3%, and a pre-baking temperature of 95 degrees Celsius for 90 seconds. The measurement point spacing for full-width film thickness scanning is 25 millimeters, with no fewer than 1,000 sampling points per image. At least one additional measurement point is collected near the seam in each exposure zone. The above values ​​are representative settings for this embodiment, and equivalent adjustments can be made in actual applications without departing from the technical principles of this embodiment.

[0086] Further, in step 300 of this embodiment, the partition definition table and micro-marker layout are first read. Following the row and column order of each exposure partition, the location and order of acquisition points for in-situ identification within that partition are determined, generating acquisition point plan data. Acquisition points cover the main body of the partition and the seam area, with appropriate densification within the seam area to ensure resolution for subsequent seam control. Subsequently, image acquisition is performed according to the acquisition point plan to obtain the in-situ image set for that batch. This embodiment uses the previously established coordinate reference data as the sole reference system to identify imageable micro-markers appearing in the in-situ image set, outputting their actual spatial coordinates in the substrate reference system, and organizing them according to partition and row / column indices to form a micro-marker coordinate map. Simultaneously with forming the micro-marker coordinate map, the difference field between the nominal and actual positions of each micro-marker is calculated. This difference field serves only as input data for subsequent deformation estimation and distortion fitting, without altering the original coordinate information.

[0087] This embodiment uses the difference field obtained in the previous section to estimate the overall deformation of the substrate and outputs a substrate warpage map. The warpage map includes in-plane components and negligible out-of-plane components, and is organized using partitions and row / column indexes to ensure consistency with micro-marker coordinates. Figure 1 A one-to-one correspondence is established. Based on this, robust modeling is performed by combining substrate and adhesive layer parameters: using magnification deviation, rotation deviation, orthogonality deviation, and polynomial radial and tangential components as modeling elements, and through an anomaly-resistant fitting process, high-order distortion model parameters for this batch and this partition set are obtained. To facilitate quality assessment, this embodiment also outputs fitting residual statistics, including the mean, extreme values, and standard deviation of the residual distribution, and records them separately for partitions and seam areas, so that they can be directly referenced in subsequent energy distribution calculations and mid-section self-correction.

[0088] This embodiment calculates exposure correction commands for motion and projection control based on high-order distortion model parameters and fitting residual statistics. These commands include at least step size fine-tuning, scan speed fine-tuning, projection magnification correction, and local correction for projection mapping. For the seam area of ​​each exposure zone, an overlap energy table (as shown in Table 2) is generated based on the local residuals within that area and predetermined seam width rules. This ensures that the exposure dose within the overlap band monotonically varies along the seam normal, and that the sum of the doses of adjacent zones within the overlap band falls within the tolerance range of a preset target dose. This embodiment summarizes the micro-marker coordinate map, substrate warpage map, high-order distortion model parameters, fitting residual statistics, exposure correction commands, and overlap energy table into an in-situ metering data package. This data package serves as the direct input for zoned exposure and mid-section self-calibration, and also as the basis for batch-level closed-loop updates.

[0089] Table 2. Examples of overlapping energy representation

[0090]

[0091] In a typical implementation, each exposure zone has no fewer than 9 acquisition points, and the acquisition point density within the seam area is twice that of the main body of the zone; the minimum resolution of a single in-situ image is no less than 0.5 micrometers per pixel; the contrast threshold for micro-marker recognition is 0.6, and the uncertainty of edge localization is no greater than 0.2 micrometers; the in-plane component of the warp map is output in units of the zone grid, with a spatial sampling interval of no more than 5 millimeters; the batch determination threshold for the fitting residual is a mean of no more than 0.05 micrometers and a standard deviation of no more than 0.2 micrometers; the update cycle for the step size adjustment and projection magnification correction of the exposure correction command is no more than once per zone; the overlapping energy table uses a monotonically gradual change of no less than 16 dose levels, and the target dose tolerance is set to ±3%. The above values ​​are representative settings of this embodiment and can be equivalently adjusted according to equipment and product requirements without departing from the principle of this embodiment.

[0092] Furthermore, in this embodiment, when generating exposure correction instructions, the deviation between the actual position of the micromarks in each zone and their designed nominal position is taken into account, along with the in-plane warping trend of the substrate, the difference in adhesive layer thickness, and the imaging characteristics of the equipment projection system. The calculation process obtains the correction amounts for overall magnification, rotation angle, and orthogonality error through multi-point registration, and then establishes a high-order deformation distribution model locally in each zone to correct the spatial offset of each point. The main parameters used for calculation include the substrate reference coordinates, the center position of the micromarks in each zone, the actual measurement data of the pattern boundary after development, and the distortion model results of the previous batch. To ensure the stability of the modeling results, outlier removal and weighted fitting are used to ensure that the root mean square value of the fitting residual is less than a preset threshold. The calculated correction amounts are converted into motion control instructions and projection optical compensation instructions for subsequent exposure step adjustment, scanning speed adjustment, and projection mapping correction.

[0093] When forming the overlapping energy table, this embodiment uses the local residual distribution of adjacent zones within the seam area as input to calculate the gradual distribution of exposure energy along the normal direction of the seam strip. The starting and ending energies of the gradual distribution are determined by the target dose setting and the tolerance range. The target dose is determined comprehensively based on the photosensitive curve of the material, layer thickness, and the absorption rate of the imaging adhesive. The tolerance range is generally set to within three percent of the target dose. The monotonicity of the energy distribution is achieved through the digital micromirror grayscale allocation of the light source, with at least sixteen grayscale levels to ensure smooth energy changes without abrupt changes. The system verifies whether the total energy of adjacent zones within the overlapping zone falls within the allowable range through trial calculations. If overexposure or underexposure trends occur, the slope of the grayscale allocation curve is automatically adjusted to achieve dose balance. Through the above control, the exposure dose of each zone can be continuous and there can be no obvious brightness difference at the seam when splicing large areas, thereby ensuring the consistency of the microstructure of the display panel.

[0094] Furthermore, step 400 of this embodiment includes:

[0095] Before starting the current exposure zone, in-situ imaging of the zone is completed according to the zone definition table and acquisition point plan. The image is identified and denoised based on the micro-marker map to generate the latest micro-marker coordinate map of the zone. The latest coordinates are registered with the design nominal coordinates at multiple points to obtain the spatial distribution of alignment deviation and in-field distortion residuals. Anomalies with insufficient brightness, insufficient contrast, or deviation exceeding the threshold are eliminated or replaced to ensure the stability and reliability of the data involved in the calculation. Based on this, the step size adjustment, projection magnification correction, and projection mapping local correction are calculated. These three are collectively referred to as the zone correction increment and are written into the pre-exposure correction record of the current zone in the order of "step size first, then magnification, then mapping". The pre-exposure correction record includes at least the effective time, row and column index, correction value, anomaly point processing result, and data version number, which serve as the basis for subsequent zone exposure and the generation of exposure record entries.

[0096] Within the seam area, a grayscale distribution is set according to the overlapping energy table, ensuring a monotonically gradual change in the overlapping band along the seam normal. Exposure is performed in the main area of ​​the zone according to the target dose. During exposure, the step size and scan speed are fine-tuned in real-time based on the zone correction increment, with update cycles and maximum changes set to prevent over-adjustment and trajectory jitter. Simultaneously, dose and focal length are monitored online and recorded promptly. If dose drift or focal length deviation exceeds a preset threshold, a local retest is triggered, and the step size is corrected first, followed by the scan speed, in a predetermined order. After completing the exposure of the zone, the zone exposure data is exported, including at least the dose distribution, scan trajectory, focal length setting, real-time monitoring values, and timestamp. This data is then linked to the zone correction increment for subsequent aggregation into exposure record entries and use in setting development conditions.

[0097] When the exposure process of a given zone is approximately halfway complete, in-situ imaging is performed again to update the higher-order distortion model parameters, calculate the secondary correction increment, and immediately apply it to the remaining trajectory of the zone after this calculation. Simultaneously, it is preset as the starting compensation for the next zone, forming a mid-section self-correction record. The order of application of the secondary correction increment is consistent with the pre-exposure correction, and the effective range and duration are indicated in the record. After the zone exposure is completed, in-situ metrology is performed again to update the micro-marker coordinate map and higher-order distortion model parameters of that zone. Subsequently, the zone exposure data, pre-exposure correction record, and mid-section self-correction record are merged to generate an exposure record entry for that zone. Each entry includes at least statistics on dose, focal length, alignment residuals, and a list of abnormal events, and is accumulated by zone number as the exposure record for this batch. The exposure record entry is directly used for setting the development conditions for this zone and solving the subsequent etching window; the batch-level exposure record is used for subsequent closed-loop updates.

[0098] By calling the substrate and adhesive layer parameters, as well as the exposure record entries for that zone, the temperature and time for post-baking, the developer ratio, and the development time are determined. After post-baking and development, the etching resistance window for that zone is obtained. Subsequently, three measurements are performed: critical size, sidewall angle, and surface morphology defects. A minimum number of sampling positions are set in the seam zone and the main body area to ensure representativeness. After noise reduction and consistency verification, the measurement data are summarized into initial quality data and associated and merged with the exposure records according to the zone number and spatial location to form window quality data. The window quality data is used as input for etching window setting and defect restoration sequence selection, and is also incorporated into batch-level closed-loop updates to optimize the starting input and sampling point plan for the next batch.

[0099] Furthermore, in the acquisition and recognition process, each zone has no fewer than 9 acquisition points, and the density of acquisition points in the seam area is twice that of the main area; the spatial resolution of a single in-situ image is no less than 0.5 micrometers per pixel; acquisition is performed once each at "before exposure, during exposure, and after exposure"; the total acquisition and processing time for a single acquisition does not exceed 0.5 seconds; the recognition success rate is no less than 99.5%; the contrast threshold is no less than 0.6; the edge positioning uncertainty is no greater than 0.2 micrometers; the threshold for judging abnormal points is that points with a contrast lower than 0.6 or a positional deviation higher than 0.5 micrometers are discarded or replaced by interpolation of nearby qualified points. In distortion and correction, the absolute value of the step size adjustment amplitude does not exceed 0.2 micrometers; the absolute value of the projection magnification correction does not exceed 0.005%; the local correction of the projection mapping is output according to the grid, and the grid center spacing is no greater than 5 millimeters; the write-back delay of the mid-section self-correction does not exceed 100 milliseconds; the zone mean of the alignment residual does not exceed 0.05 micrometers, the standard deviation does not exceed 0.2 micrometers, and the maximum residual of the seam area does not exceed 0.3 micrometers. In overlapping energy and dose control, the representative value of the overlap band width is 600 micrometers, and it is allowed to be adjusted according to rules within the range of 300 to 1000 micrometers; the number of gray levels is not less than 16; the total dose of adjacent zones within the overlap band is allowed to deviate from the target dose by no more than 3; the smooth segments at both ends of the gradient curve each occupy 10 to 20 of the overlap band width; the energy change of a single level does not exceed 0.5 of the target dose; and the dose difference between adjacent sampling points does not exceed 1.

[0100] Furthermore, the dose stability of the main area is better than 2; the allowable deviation of focal length is no greater than 0.5 micrometers; the dose monitoring sampling cycle is no greater than 0.1 seconds; the online fine-tuning range of scanning speed does not exceed 2, and the upper limit of a single change is set to 1 of the initial speed. The post-baking temperature is 95 to 110 degrees Celsius, and the time is 90 to 120 seconds; the development time is 50 to 90 seconds, the rinsing time is 30 to 60 seconds, and the drying temperature is 100 to 130 degrees Celsius, and the time is 60 to 120 seconds; the critical size deviation does not exceed 0.2 micrometers; the sidewall angle is controlled between 85 and 90 degrees; the three-times standard deviation of edge roughness does not exceed 0.1 micrometers; and the surface morphology defect density is no higher than 0.5 defects per square millimeter. The above values ​​are representative ranges and thresholds that can be directly implemented. Applicants can make equivalent adjustments based on differences in equipment and materials without changing the technical principles.

[0101] In step 500 of this embodiment, the exposure record and window quality data generated in the previous process are first retrieved. The dose distribution, focal length setting, alignment residual, critical size, sidewall angle, and surface morphology measurement results of the partition are read. Based on the allowable deviation of the sidewall angle and critical size according to the process target, the required range of etching removal rate and sidewall morphology for each thin film layer is deduced, and the set values ​​and tolerances of power supply, atmosphere ratio, working pressure, and substrate temperature are given accordingly. The set values ​​are used to start etching, and the tolerances are used for online monitoring qualification judgment and fine-tuning trigger. The four parameters are solidified into an etching window setting table with the structure of "target value, upper limit, lower limit, allowable change step size, and verification index", which serves as the direct input for subsequent pattern transfer start-up and on-the-fly etching adjustment. At the same time, the source of the setting (the corresponding exposure record entry and window quality data entry number) is recorded to ensure traceability.

[0102] In this embodiment, anisotropic pattern transfer is initiated according to the etching window setting table. Endpoint signals, removal rate, sidewall angle surrogate quantity, and substrate temperature are collected in real time to form etching process monitoring data. The endpoint signals are derived from stable changes in optical emission or interference reflection intensity; the removal rate is derived from the quality difference of the interferometric thickness gauge or calibration sheet; the sidewall angle surrogate quantity is derived from the angular distribution of the scattered signal or the convergence trend of the linewidth over time; and the substrate temperature is derived from the sensor embedded in the vacuum chuck and infrared comparison. The monitoring data is compared with the upper and lower limits of the etching window setting table for each sampling cycle. If any monitored quantity exceeds the corresponding tolerance, fine-tuning is calculated and implemented in the order of "power supply, atmosphere ratio, working pressure, and substrate temperature." The target value and timestamp after fine-tuning are written into the parameter adjustment record. To avoid over-adjustment, this embodiment limits the upper limit of the single change and the cumulative change per unit time. After fine-tuning, the endpoint signals and removal rate are checked to ensure they return to the tolerance range. Etching can only continue if the check passes.

[0103] This embodiment uses two types of surrogate quantities to assess the etching damage level: the surrogate quantity of ion energy is defined as a representative value calculated by combining the effective value of the bias voltage, the RF power density, and the gas rarefaction level; the surrogate quantity of temperature rise is defined as the equivalent value obtained by comparing the increase in the measured temperature of the substrate relative to the initial temperature with the infrared thermometry. Both types of surrogate quantities are compared with preset low-damage thresholds to generate a low-damage criterion report. If either indicator does not meet the low-damage condition, the process returns to the aforementioned fine-tuning procedure, prioritizing a reduction in power supply power, a decrease in the proportion of active components in the atmosphere, a moderate increase in operating pressure, and an increase in the back-side helium cooling flow rate. This process is repeated until the conditions are met or the safety shutdown protection threshold is reached. After etching is completed, this embodiment selects the type and order of defect restoration sequence based on the etching process monitoring data and low damage criterion report. It usually includes oxygen-containing atmosphere annealing and low-power hydrogen-containing or nitrogen-containing passivation treatment. The set values ​​and tolerances of restoration atmosphere, restoration temperature and restoration time are given respectively and solidified into a defect restoration selection table. During restoration, the temperature curve and time curve, atmosphere switching time and pressure relief time are recorded to form a restoration execution record.

[0104] This embodiment calls the etching window setting table to formulate a test plan for combined bias and temperature stress. Representative devices or test structures within the array are selected to perform positive and negative bias stress and high-temperature holding tests, measuring mobility, threshold, and leakage current to generate stability assessment data. Simultaneously, key layout locations (such as the area around seams and high-density interconnect regions) are separately documented to compare electrical drift differences at different locations. Finally, the etching window setting table, etching process monitoring data, parameter adjustment records, low-damage criterion report, defect restoration selection table, restoration execution records, and stability assessment data are merged into a pattern transfer data package. This data package is used for closed-loop verification in this batch and serves as the starting input parameter for etching window setting in the next batch, achieving adaptive optimization between batches.

[0105] Furthermore, in this embodiment, when setting the etching window, the power supply is set to 200-500 ohms, the volume ratio of active to inert components in the atmosphere is 20:80-60:40, the working pressure is 2-20 ohms, the substrate temperature is 20-60 ohms, the target sidewall angle is 85-90 degrees, and the allowable deviation of the critical size is ±0.2 micrometers; the sampling period for online monitoring and fine-tuning is 0.1-0.5 seconds, and the triggering conditions are that the endpoint signal offset exceeds 10%, the removal rate deviates from the target by more than 10%, the sidewall angle change corresponding to the proxy amount exceeds 2 degrees, or the substrate temperature exceeds the tolerance, triggering fine-tuning. The single change in power supply is 5-20 ohms, and the cumulative change within 30 seconds does not exceed 10% of the initial setting; the single change in atmosphere ratio is 2%-5%; the single change in working pressure is 0.2-0.5 ohms; and the single change in substrate temperature is 2-5 ohms. In the low-damage criterion... The upper limit of the effective bias value corresponding to the proxy amount of ion energy is controlled within the range of 30 to 80. The upper limit of the proxy amount of temperature rise is no more than 15 relative to the initial temperature. If either limit is exceeded, the process will return to fine-tuning. The defect restoration sequence can be executed in the order of "annealing followed by passivation". The annealing temperature in the oxygen atmosphere is 200 to 300 degrees Celsius and the duration is 10 to 30 minutes. The passivation power in the hydrogen atmosphere is 50 to 200 and the duration is 30 to 120 seconds. The passivation power in the nitrogen atmosphere is 100 to 300 and the duration is 60 to 180 seconds. The step interval is 2 to 5 minutes of cooling. In the electrical evaluation, the positive bias temperature stress is 60 degrees Celsius with a positive bias of 20 for 1000 seconds. The negative bias temperature stress is the same temperature with a negative bias of 20 for 1000 seconds. The mobility target is not less than 10, the absolute value of the threshold drift does not exceed 0.5, and the leakage current does not exceed 0.000000001.

[0106] Preferably, it further includes:

[0107] Step 600: Perform correlation analysis on the higher-order distortion model parameters, the exposure correction command, the overlap energy table, the exposure record, the etching window setting, and the stability evaluation data, and combine the partition definition table with the substrate and adhesive layer parameters to generate the starting input parameters for the next batch.

[0108] In step 600 of this embodiment, using the data from the completed batch as input, the higher-order distortion model parameters, exposure correction instructions, overlap energy tables, exposure records, etching window settings, and stability evaluation data are aligned and merged in the same coordinate system according to the partition number and seam area location, forming an original data set organized by partition and time sequence. Abnormal entries in the set are first filtered out, with abnormality criteria including low in-situ imaging contrast, alignment residuals exceeding process thresholds, long-term drift in dose monitoring, etching endpoint signal mismatch, and interrupted temperature recording. Subsequently, the key performance of each partition is quantified using a "quality score." The quality score is a comprehensive evaluation of indicators such as alignment residuals, seam consistency, critical size deviation, sidewall angle, and electrical stability according to predetermined weights, used to intuitively reflect the overall quality of the partition from exposure to etching to electrical performance. Simultaneously, a "contribution score" is calculated, which measures the strength of the impact of a setting or correction on quality improvement. Specifically, the changes in parameters in this batch are analyzed in relation to the improvement of corresponding quality indicators, and parameter entries showing stable positive or negative correlations are recorded. To ensure traceability, this embodiment writes the source list and timestamp of each scoring item into the "Association Analysis Report". The report gives the reason for the removal of the data and the source number and sampling location description of the data to be retained.

[0109] After completing the correlation analysis, this embodiment generates an "Initial Input Parameter Table," which is a set of parameters directly called when starting the next batch. It contains four categories: First, initial settings for in-situ metrology and modeling, including the spatial density of acquisition points, the densification ratio within the seam tape, the order selection of high-order models, and the anomaly detection threshold; Second, initial settings for exposure control, including the baseline correction amounts for step size and projection magnification, the initial deformation distribution of the projection mapping, the series of energy gradients in the seam tape and the proportion of smooth sections at both ends, and the target dose tolerance; Third, initial settings for etching... The initial settings for etching and restoration include target ranges for power, atmosphere ratio, operating pressure, and substrate temperature; thresholds for endpoint signals and removal rates; upper limits for low damage threshold control; recommended combinations and priority order of restoration atmosphere, temperature, and duration; fourth, strategy parameters for quality judgment and rollback, including batch thresholds for alignment residuals, critical dimensions, and electrical drift; maximum step size for parameter changes; upper limit for cumulative change per batch; and rollback strategy. The rollback strategy refers to automatically restoring to the steady-state settings of the previous one to three batches when the key indicators of the next batch fall below the threshold. The initial input parameter table undergoes two processes after generation: first, rounding and boundary verification according to the granularity executable by the equipment to ensure that all settings can be actually executed; second, completing source mapping and version registration so that each initial setting can be traced back to specific exposure record entries, window quality data entries, and etching window setting entries. Finally, in this embodiment, the table is distributed to the execution stages of in-situ metrology, motion and projection control, dose management and etching processes, respectively, and loading and verification are completed before the start of the next batch, thereby achieving adaptive optimization and traceability control between batches.

[0110] Corresponding to the above methods, such as Figure 2 As shown, this embodiment also provides a photolithography system for manufacturing and processing a display panel, including:

[0111] The partition definition and substrate pretreatment module is used to clean and apply a base coat to the surface of the display panel substrate to be coated. The effective area of ​​the display panel substrate is divided into several exposure partitions according to the projection field of view and the travel of the motion platform. An overlapping band with a preset width is set between adjacent exposure partitions as a seam area, and a partition definition table containing partition boundary numbers and seam area widths is generated.

[0112] The micro-marker placement and double-layer photoresist film formation module is used to form imageable micro-markers covering all the exposure zones on the display panel substrate, and sequentially coat the lower layer of support adhesive and the upper layer of imaging adhesive and measure the thickness and uniformity to obtain the substrate and adhesive layer parameters.

[0113] The in-situ metrology and high-order distortion modeling module is used to perform in-situ optical acquisition partition by partition according to the partition definition table, using the imageable micro-marker as the imaging object, extracting the spatial coordinates of the imageable micro-marker and establishing a micro-marker coordinate map and a substrate warping map, combining the substrate and adhesive layer parameters to fit the high-order distortion model parameters online, and generating exposure correction instructions for motion and projection control and an overlap energy table for the seam area according to the high-order distortion model parameters.

[0114] The partitioned exposure and development control module is used to perform exposure partition by partition using a digital micromirror according to the partition definition table, perform mid-section self-correction and stitching overlap control according to the exposure correction command and the overlap energy table, repeat in-situ measurement and update the micro-marker coordinate map and the higher-order distortion model parameters before, during and after exposure in each exposure partition, form an exposure record recording dose, focal length and alignment residual, set post-baking and development conditions according to the substrate and adhesive layer parameters and complete development to obtain an etch-resistant window with steep sidewalls, and measure the critical size, sidewall angle and morphological defects of the etch-resistant window to form window quality data;

[0115] The low-damage etching and defect recovery module is used to set the etching window based on the exposure record and the window quality data, transfer the pattern to the target thin film layer under low-damage anisotropic conditions, collect etching process monitoring data to fine-tune the power, atmosphere ratio, working pressure and substrate temperature, select and execute the defect recovery sequence based on the etching process monitoring data, and measure the mobility, threshold and leakage current under bias and temperature stress conditions to form stability evaluation data.

[0116] The beneficial effects of this invention are as follows:

[0117] (i) A partitioning system based on the projection field of view and platform travel is introduced on a large-size substrate, and an overlap band with a preset width is set between adjacent partitions. At the same time, a partition definition table containing partition boundaries, numbers, and overlap band widths is generated, so that the exposure space is clearly discretized and managed digitally. This design decomposes the full-frame error into controllable small errors within the partitions. On the other hand, the overlap band leaves engineering margin for energy gradation and superposition in the seam area, reducing cross-zone cumulative error and splicing stripes from the source, and improving the overall plate superposition accuracy and seam consistency.

[0118] (ii) First, imageable micro-markers covering all zones are formed on the substrate. Then, a lower layer of support adhesive and an upper layer of imaging adhesive are formed sequentially, and their thickness and in-plane uniformity are measured to obtain substrate and adhesive layer parameters that can be directly used for subsequent settings. The micro-markers provide a stable and traceable in-situ identification reference. The double-layer adhesive improves sidewall formation and etching resistance by separating the support and imaging functions. The quantitative input of film thickness and uniformity provides a basis for setting process points such as focal length and development time. As a result, the critical size of the etching resistance window is more stable and the sidewalls are steeper after development, reducing subsequent patterning errors caused by film formation fluctuations.

[0119] (III) Using micro-markers as the imaging target, in-situ optical acquisition is carried out zone by zone to establish micro-marker coordinate maps and substrate warping maps. A high-order distortion model is then fitted online using substrate and adhesive layer parameters. Based on this, exposure correction commands for motion and projection, as well as an overlap energy table for the seam area, are generated. Simultaneously, the model is repeatedly measured and updated before, during, and after exposure in each zone, forming a closed loop of "measurement—modeling—command—verification." This dynamic compensation chain can suppress drift caused by thermal, mechanical, and projection nonlinearities at any time, ensuring that the alignment and dose distribution within the zone and the seam are always controlled, significantly reducing superposition errors and energy mutations.

[0120] (iv) Exposure records and window quality data are used for etching window settings, and pattern transfer is completed under low-damage anisotropy conditions. During etching, endpoints, removal rate, and temperature are recorded and monitored, and power, atmosphere ratio, pressure, and substrate temperature are finely adjusted according to tolerance. After etching, a defect recovery sequence is executed, and finally, mobility, threshold, and leakage current are measured under bias and temperature stress conditions to form stability evaluation data. In this way, "exposure-development-etching-recovery-electrical evaluation" is connected into a measurable quality channel, which enables the sidewall angle, edge roughness, and electrical stability to be optimized simultaneously, reducing the risk of yield in the later stages.

[0121] (v) The entire process generates and calls upon objects such as partition definition tables, substrate and adhesive layer parameters, micro-marker coordinate diagrams, substrate warpage diagrams, high-order distortion model parameters, exposure correction instructions, overlap energy tables, exposure records, window quality data, etching window settings, etching process monitoring and stability assessment data, and manages them in a unified manner under the same coordinate and partition dimension. This data chain provides traceable, comparable, and reusable process assets, providing direct evidence to support continuous calibration within the same batch and parameter settings for subsequent batches, thereby improving process consistency, shortening setup time, and reducing overall manufacturing costs.

[0122] The various embodiments in this specification are described in a progressive manner, with each embodiment focusing on its differences from other embodiments. Similar or identical parts between embodiments can be referred to interchangeably. For the systems disclosed in the embodiments, since they correspond to the methods disclosed in the embodiments, the descriptions are relatively simple; relevant parts can be referred to the method section.

[0123] This document uses specific examples to illustrate the principles and implementation methods of the present invention. The descriptions of the above embodiments are only for the purpose of helping to understand the method and core ideas of the present invention. Furthermore, those skilled in the art will recognize that, based on the ideas of the present invention, there will be changes in the specific implementation methods and application scope. Therefore, the content of this specification should not be construed as a limitation of the present invention.

Claims

1. A photolithography method for manufacturing a display panel, characterized in that, include: The surface of the display panel substrate to be coated is cleaned and primed. The effective area of ​​the display panel substrate is divided into several exposure zones according to the projection field of view and the travel of the motion platform. An overlapping zone with a preset width is set between adjacent exposure zones as a seam area, and a partition definition table containing partition boundary numbers and seam area widths is generated. Imageable micromarks covering all exposure zones are formed on the display panel substrate. A lower layer of support adhesive and an upper layer of imaging adhesive are coated sequentially, and the thickness and uniformity are measured to obtain the substrate and adhesive layer parameters. Using the imageable micro-marker as the imaging object, in-situ optical acquisition is performed partition by partition according to the partition definition table. The spatial coordinates of the imageable micro-marker are extracted and a micro-marker coordinate map and a substrate warping map are established. The parameters of the higher-order distortion model are fitted online in combination with the parameters of the substrate and the adhesive layer. Based on the higher-order distortion model parameters, exposure correction instructions for motion and projection control and an overlap energy table for the seam area are generated. A digital micromirror is used to perform exposure partition by partition according to the partition definition table. Mid-section self-correction and stitching overlap control are performed according to the exposure correction command and the overlap energy table. Before, during and after exposure in each exposure partition, in-situ measurement is repeated and the micro-marker coordinate map and the higher-order distortion model parameters are updated to form an exposure record of dose, focal length and alignment residual. Post-baking and development conditions are set according to the substrate and resist layer parameters and development is completed to obtain an etch-resistant window with steep sidewalls. The critical size, sidewall angle and morphological defects of the etch-resistant window are measured to form window quality data. Based on the exposure record and the window quality data, the etching window setting is given. Under low-damage anisotropic conditions, the pattern is transferred to the target thin film layer. Etching process monitoring data is collected to fine-tune the power, atmosphere ratio, working pressure and substrate temperature. Based on the etching process monitoring data, the defect restoration sequence is selected and executed. Mobility, threshold and leakage current are measured under bias and temperature stress conditions to form stability evaluation data.

2. The photolithography method for manufacturing a display panel according to claim 1, characterized in that, Also includes: The higher-order distortion model parameters, the exposure correction instructions, the overlap energy table, the exposure record, the etching window settings, and the stability evaluation data are correlated and analyzed, and the starting input parameters for the next batch are generated by combining the partition definition table with the substrate and adhesive layer parameters.

3. The photolithography method for manufacturing a display panel according to claim 1, characterized in that, The surface of the display panel substrate to be coated is cleaned and primed. The effective area of ​​the display panel substrate is divided into several exposure zones according to the projection field of view and the travel of the motion platform. An overlap zone with a preset width is set between adjacent exposure zones as a seam area. A partition definition table containing partition boundary numbers and seam area widths is generated, including: The display panel substrate is positioned to determine the substrate reference point and boundary, obtain the outline and size of the effective area, and output the effective area data and coordinate reference data. Complete the cleaning and primer treatment of the surface to be coated, and record the surface condition data; the surface condition data includes at least cleanliness, primer thickness and uniformity; Obtain the projection field of view size parameters and motion platform travel parameters, and calculate the minimum coverage partitioning scheme by combining the effective area data, outputting a partitioned grid draft; the expression of the partitioned grid draft is: ； in, The horizontal dimension of the effective area; The effective area is the dimension in the vertical direction; This represents the horizontal dimension of the field of view in a single projection. This represents the vertical dimension of the field of view in a single projection. The preset overlap width in the horizontal direction for adjacent exposure zones; The preset overlap width of adjacent exposure zones in the vertical direction; The step size of the partitioned grid in the horizontal direction is equal to the field size minus the overlap width; The vertical step size of the partitioned grid is equal to the field size minus the overlap width. This represents the number of horizontal partitions, and coverage is ensured by rounding up. The number of vertical partitions is determined by rounding up to ensure coverage. and The coordinates of the reference points for the effective area are used to determine the grid start point; For the first Liede The coordinates of the pivot point of the row partition; The coordinate set of the draft partitioned grid; The actual projected field of view rectangular area corresponding to the pin point; the coverage condition is used to verify that the grid completely covers the effective area and is usually satisfied. and ; The preset width of the overlap band is determined based on the target alignment accuracy and the superposition error tolerance, and the seam width rules are output. Based on the proposed partition grid, the boundaries of adjacent partitions are corrected according to the stitching width rules to ensure full coverage without gaps or excessive overlap. Each partition is assigned a unique number according to the exposure path, and the partition boundary and number data are output. The partition definition table is generated by integrating the partition boundaries, numbering data, and seam width rules. It is then verified by calculating the coverage ratio and seam consistency to form a verified partition definition table.

4. The photolithography method for manufacturing a display panel according to claim 3, characterized in that, Imageable micromarkers covering all exposure zones are formed on the display panel substrate. A lower support adhesive and an upper imaging adhesive are sequentially coated, and their thickness and uniformity are measured to obtain substrate and adhesive layer parameters, including: Based on the partition definition table, the shape, size and spacing of the imageable micro-markers are determined, so that each exposure partition and the stitching area of ​​the exposure partition contains a marker array for in-situ identification, and the micro-marker layout and deployment data are output. Imageable micromarks are formed on the surface of the display panel substrate according to the micromark pattern. The micromark contrast and edge contour are acquired by calibration imaging to obtain micromark imaging quality data. A lower support adhesive material is uniformly coated on the display panel substrate and cured to form a lower support adhesive film layer. The film thickness and in-plane uniformity of the lower support adhesive film layer are measured, and the lower support adhesive film thickness data and the lower support adhesive in-plane uniformity data are output. An upper imaging adhesive material is uniformly coated onto the lower support adhesive film layer and cured to form an upper imaging adhesive film layer. The film thickness and in-plane uniformity of the upper imaging adhesive film layer are measured, and the upper imaging adhesive film thickness data and the upper imaging adhesive in-plane uniformity data are output. These data are then combined with the lower support adhesive film thickness data and the lower support adhesive in-plane uniformity data to generate composite data of adhesive layer thickness and uniformity. The substrate and adhesive layer parameters are generated by summarizing the micromarker layout and deployment data, the micromarker imaging quality data, and the adhesive layer thickness and uniformity composite data.

5. The photolithography method for manufacturing a display panel according to claim 4, characterized in that, Using the imageable micro-markers as the imaging object, in-situ optical acquisition is performed partition by partition according to the partition definition table. The spatial coordinates of the imageable micro-markers are extracted, and a micro-marker coordinate map and a substrate warpage map are established. High-order distortion model parameters are fitted online based on the substrate and adhesive layer parameters. Exposure correction instructions for motion and projection control and an overlap energy table for the seam area are generated based on the high-order distortion model parameters, including: Based on the partition definition table and the micro-marker layout, the location and order of the acquisition points in each exposure partition are determined, and acquisition point plan data is output to guide in-situ optical acquisition partition by partition; Image acquisition is performed on each of the exposure zones according to the acquisition point plan data, and an in-situ image set is output. Micromarks are identified in the in-situ image set and their spatial coordinates in the substrate coordinate system are obtained. A micromark coordinate map organized by partition is formed by combining the coordinate reference data, and the difference field between the nominal position and the measured position is calculated. Based on the difference field fitting, the deformation distribution in the plane and the micro-out plane is obtained, and the substrate warping diagram is output. The micro-marker coordinate map and the substrate warpage map are combined with the substrate and adhesive layer parameters for robust fitting to obtain high-order distortion model parameters including magnification, rotation, orthogonality deviation, radial and tangential terms, and to generate fitting residual statistics. Based on the higher-order distortion model parameters and the fitting residual statistics, exposure correction instructions for motion and projection control are calculated, and an overlapping energy table is generated in the stitching area according to the local residual and the stitching width rule, so that the exposure dose in the overlapping band changes monotonically along the stitching normal, and the sum of the doses of the adjacent partitions in the overlapping band is within the tolerance range of the preset target dose.

6. The photolithography method for manufacturing a display panel according to claim 1, characterized in that, A digital micromirror is used to perform exposure zone by zone according to the defined zone table. Mid-section self-correction and stitching overlap control are performed based on the exposure correction command and the overlap energy table. Before, during, and after exposure in each zone, in-situ measurement is repeated, and the micro-marker coordinate map and the higher-order distortion model parameters are updated to form an exposure record containing the dose, focal length, and alignment residuals. Post-baking and development conditions are set according to the substrate and resist layer parameters, and development is completed to obtain an etch-resistant window with steep sidewalls. The critical dimensions, sidewall angles, and morphological defects of the etch-resistant window are measured to form window quality data, including: Before the current exposure zone enters the exposure phase, an in-situ image is acquired, the micro-marker coordinate map of the current exposure zone is updated, the alignment deviation and in-field distortion residual are calculated, the zone correction increment is obtained, and the step size, projection magnification and projection mapping parameters are corrected according to the zone correction increment to generate a pre-exposure correction record. Within the seam area, the exposure dose is monotonically gradually varied along the seam normal according to the overlapping energy table. Exposure is completed in the main area of ​​the current exposure zone according to the target dose. During the exposure process, the step size and scanning speed are finely adjusted according to the zone correction increment, and the zone exposure data is output. When half of the exposure in a zone has been completed, an in-situ image is acquired again, the parameters of the higher-order distortion model are updated, and the secondary correction increment is calculated. The secondary correction increment is written back to the remaining trajectory of the current exposure zone and the starting compensation of the next zone to form a mid-section self-correction record. After the zonal exposure is completed, in-situ metrology is performed and the micro-marker coordinate map and the higher-order distortion model parameters are updated. The zonal exposure data, the pre-exposure correction record and the mid-section self-correction record are summarized to generate an exposure record entry containing dose, focal length and alignment residual, and archived as the exposure record according to the zonal number. The substrate and adhesive layer parameters and the exposure record entries are called to determine the post-baking temperature and time, as well as the developer ratio and development time. Post-baking and development are performed to obtain the etching resistance window of the current exposure zone. The critical size, sidewall angle and surface morphology of the etching-resistant window are measured to obtain initial quality data. The initial quality data of each exposure zone are associated and merged with the exposure record according to the zone number to form the window quality data.

7. The photolithography method for manufacturing a display panel according to claim 1, characterized in that, Based on the exposure records and window quality data, an etching window setting is given. Under low-damage anisotropic conditions, the pattern is transferred to the target thin film layer. Etching process monitoring data is collected to fine-tune power, atmosphere ratio, operating pressure, and substrate temperature. A defect recovery sequence is selected and executed based on the etching process monitoring data. Mobility, threshold current, and leakage current are measured under bias and temperature stress conditions to generate stability evaluation data, including: a) Call the exposure record and the window quality data, and combine them with the tolerance of the target sidewall angle and critical size to calculate the power setting value and tolerance range, the atmosphere ratio setting value and tolerance range, the working pressure setting value and tolerance range, and the substrate temperature setting value and tolerance range, and output the etching window setting table. b) Based on the etching window setting table, perform pattern transfer, collect endpoint signals, removal rate, sidewall angle and substrate temperature, generate etching process monitoring data, and when any monitored quantity exceeds the corresponding tolerance range, calculate the power fine-tuning increment, atmosphere ratio fine-tuning increment, working pressure fine-tuning increment and substrate temperature fine-tuning increment respectively and implement fine-tuning to form a parameter adjustment record. c) Calculate the proxy amount of ion energy and the proxy amount of temperature rise based on the etching process monitoring data, compare them with the preset low damage threshold, generate a low damage criterion report, and return to step b) to perform fine-tuning and continue etching; d) Select the type and order of the defect restoration sequence based on the etching process monitoring data and low damage criterion report, and provide the set values ​​and tolerance ranges for restoration atmosphere, restoration temperature and restoration time, and output the defect restoration selection table. e) Execute the defect restoration sequence, record the temperature curve and time curve, and generate a restoration execution record; f) Based on the etching window setting table, formulate a bias and temperature stress test plan, perform mobility, threshold and leakage current measurements on the sample or array, and generate stability evaluation data; g) The etching window setting table, the etching process monitoring data, the parameter adjustment record, the low damage criterion report, the defect restoration selection table, the restoration execution record, and the stability evaluation data are merged into a graphic transfer data package; the graphic transfer data package is used as the input for batch-level closed-loop update and as the starting input parameter for the next batch etching window setting.

8. A photolithography system for manufacturing a display panel, characterized in that, include: The partition definition and substrate pretreatment module is used to clean and apply a base coat to the surface of the display panel substrate to be coated. The effective area of ​​the display panel substrate is divided into several exposure partitions according to the projection field of view and the travel of the motion platform. An overlapping band with a preset width is set between adjacent exposure partitions as a seam area, and a partition definition table containing partition boundary numbers and seam area widths is generated. The micro-marker placement and double-layer photoresist film formation module is used to form imageable micro-markers covering all the exposure zones on the display panel substrate, and sequentially coat the lower support adhesive and the upper imaging adhesive and measure the thickness and uniformity to obtain the substrate and adhesive layer parameters. The in-situ metrology and high-order distortion modeling module is used to perform in-situ optical acquisition partition by partition according to the partition definition table, using the imageable micro-marker as the imaging object, extracting the spatial coordinates of the imageable micro-marker and establishing a micro-marker coordinate map and a substrate warping map, combining the substrate and adhesive layer parameters to fit the high-order distortion model parameters online, and generating exposure correction instructions for motion and projection control and an overlap energy table for the seam area according to the high-order distortion model parameters. The partitioned exposure and development control module is used to perform exposure partition by partition using a digital micromirror according to the partition definition table, perform mid-section self-correction and stitching overlap control according to the exposure correction command and the overlap energy table, repeat in-situ measurement and update the micro-marker coordinate map and the higher-order distortion model parameters before, during and after exposure in each exposure partition, form an exposure record recording dose, focal length and alignment residual, set post-baking and development conditions according to the substrate and adhesive layer parameters and complete development to obtain an etch-resistant window with steep sidewalls, and measure the critical size, sidewall angle and morphological defects of the etch-resistant window to form window quality data; The low-damage etching and defect recovery module is used to set the etching window based on the exposure record and the window quality data, transfer the pattern to the target thin film layer under low-damage anisotropic conditions, collect etching process monitoring data to fine-tune the power, atmosphere ratio, working pressure and substrate temperature, select and execute the defect recovery sequence based on the etching process monitoring data, and measure the mobility, threshold and leakage current under bias and temperature stress conditions to form stability evaluation data.

Images