Proportional dynamic etching compensation algorithm

Abstract

The invention discloses a proportional dynamic etching compensation algorithm, which comprises the following steps of: acquiring a graph to be processed, and extracting graph characteristics; merging the intersected graphic features in the graphic features so as to process the to-be-processed graphic into at least two mutually independent and closed polygons; discretizing each edge of the polygon into a plurality of line segments, and obtaining the line width of the line segment and the line distance relative to the adjacent polygon; calculating compensation values corresponding to the line width and the line distance of each line segment based on a preset compensation rule; and reconstructing the polygon based on the calculated compensation value. Compared with the prior art, the method has the advantage that the defect of overcompensation or undercompensation generated in the layout with sparse and dense patterns coexisting in the traditional one-cutting compensation method is overcome.

Description

Technical Field

[0001] This invention relates to a proportional dynamic etching compensation algorithm, belonging to the field of process compensation technology. Background Technology

[0002] In the PCB semiconductor field, due to physical effects and process parameters, the pattern etched and developed by the exposure machine often deviates from the designed pattern.

[0003] Traditional compensation methods apply a fixed compensation value to all line edges. However, with the continuous shrinking of integrated circuit feature sizes and the increasing wiring density, complex situations arise in the layout, including dense lines, isolated lines, and combinations of different line widths and spacings. A single fixed compensation value cannot adapt to all scenarios: for dense areas, it may lead to overcompensation, causing pattern bridging (short circuits); for isolated lines, it may lead to undercompensation, causing pattern breaks (open circuits). In other words, static compensation strategies based on a single rule can no longer meet the stringent requirements for pattern accuracy and consistency in advanced manufacturing processes.

[0004] In view of this, it is indeed necessary to propose a proportional dynamic etching compensation algorithm to solve the above problems. Summary of the Invention

[0005] To address the aforementioned technical problems, this invention provides a proportional dynamic etching compensation algorithm. This algorithm can accurately compensate for each line segment, overcoming the overcompensation or undercompensation defects caused by the traditional "one-size-fits-all" compensation in layouts where sparse and dense graphics coexist.

[0006] The technical solution of this invention is:

[0007] A proportional dynamic etching compensation algorithm, comprising:

[0008] Acquire the image to be processed and extract its features;

[0009] The intersecting graphic features in the graphic features are merged to process the graphic to be processed into at least two independent and closed polygons;

[0010] Each edge of the polygon is discretized into several line segments, and the line width and line spacing of the line segment relative to the adjacent polygon are obtained.

[0011] The compensation value corresponding to the line width and line spacing of each line segment is calculated based on the preset compensation rules.

[0012] The polygon is reconstructed based on the calculated compensation value.

[0013] As a further improvement of the present invention, obtaining the line width and line distance of the line segment relative to the adjacent polygon includes: emitting a normal from the line segment, the intersection point of the normal with the polygon containing the line segment being the first intersection point, the intersection point with the adjacent polygon closest to the line segment being the second intersection point, the length of the normal between the line segment and the first intersection point being the line width, and the length of the normal between the line segment and the second intersection point being the line distance.

[0014] As a further improvement of the present invention, the line segment emits two normals, and the two normals emit in opposite directions.

[0015] As a further improvement of the present invention, the line segment is one of horizontal lines, vertical lines, and diagonal lines.

[0016] As a further improvement of the present invention, the preset compensation rule includes: calculating the line-to-space ratio and resolution accuracy based on the line width and line spacing of each line segment, and calculating the corresponding compensation value through the line-to-space ratio and the resolution accuracy, wherein the line-to-space ratio is the ratio of the line width and the line spacing, and the resolution accuracy is the smaller value between the line width and the line spacing.

[0017] As a further improvement of the present invention, the preset rules include a line-space ratio compensation table. When the line-space ratio and the resolution are calculated, the corresponding compensation value is obtained based on the line-space ratio compensation table.

[0018] As a further improvement of the present invention, when the line-to-space ratio and the analytical accuracy are within the range of the line-to-space ratio compensation table, but there is no corresponding compensation value, bilinear interpolation is used based on the line-to-space ratio compensation table to obtain the corresponding compensation value.

[0019] As a further improvement of the present invention, when the line-to-space ratio and the analytical accuracy are outside the range of the line-to-space ratio compensation table, a fixed compensation value is directly selected based on the division of the line-to-space ratio and the analytical accuracy in the line-to-space ratio compensation table.

[0020] As a further improvement of the present invention, the line-to-space ratio compensation table includes a horizontal line compensation table and a vertical line compensation table. When the line segment is a horizontal line or a vertical line, the corresponding compensation value is obtained based on the horizontal line compensation table or the vertical line compensation table. When the line segment is a diagonal line, the corresponding compensation value is obtained based on the horizontal line compensation table and the vertical line compensation table.

[0021] As a further improvement of the present invention, the angle between the oblique line and the horizontal line is defined as θ, and a first compensation value is obtained based on the horizontal line compensation table using the line width and line spacing of the oblique line. The second compensation value is obtained based on the vertical line compensation table. Through the formula: The compensation value val corresponding to the oblique line is calculated.

[0022] The beneficial technical effects of this invention are as follows: The proportional dynamic etching compensation algorithm of this invention merges the extracted graphic features to form at least two independent and closed polygons; each edge of the polygon is divided into several line segments, and the line width and line spacing of the line segment relative to the adjacent polygon are obtained; the compensation value corresponding to the line width and line spacing of each line segment is calculated based on the preset compensation rules; the polygon is reconstructed based on the calculated compensation value. In this way, by calculating the line width and line spacing of each line segment, accurate compensation is achieved for each line segment, overcoming the overcompensation or undercompensation defects caused by the traditional "one-size-fits-all" compensation in the layout where sparse and dense graphics coexist. Attached Figure Description

[0023] Figure 1 This is a flowchart of a proportional dynamic etching compensation algorithm conforming to a preferred embodiment of the present invention.

[0024] Figure 2 yes Figure 1 The image to be processed.

[0025] Figure 3 It is a graphic formed by combining graphic features.

[0026] Figure 4 yes Figure 1 The display shows the line width of the line segment and its distance from the adjacent polygon.

[0027] Figure 5 yes Figure 1 The line-space ratio compensation table is a preset compensation rule.

[0028] Figure 6 This is the area division table for the line-space ratio compensation table in the diagram.

[0029] Figure 7 yes Figure 1 The positive and negative images are obtained by adding compensation values ​​after obtaining compensation values ​​in the middle.

[0030] Figure 8 Yes Figure 7 The compensation image obtained after merging positive and negative images. Detailed Implementation

[0031] In order to better understand the technical means of the present invention and to implement it in accordance with the contents of the specification, the specific embodiments of the present invention will be further described in detail below with reference to the accompanying drawings and examples. The following examples are used to illustrate the present invention, but are not intended to limit the scope of the present invention.

[0032] Please see Figure 1 As shown, the present invention provides a proportional dynamic etching compensation algorithm, including:

[0033] Acquire the image to be processed and extract its features;

[0034] The intersecting graphic features in the graphic features are merged to process the graphic to be processed into at least two independent and closed polygons;

[0035] Each edge of the polygon is discretized into several line segments, and the line width and line spacing of the line segment relative to the adjacent polygon are obtained.

[0036] The compensation value corresponding to the line width and line spacing of each line segment is calculated based on the preset compensation rules.

[0037] The polygon is reconstructed based on the calculated compensation value.

[0038] Please see Figure 2 and Figure 3 As shown, in the PCB semiconductor field, the obtained circuit diagram typically contains several intersecting lines. Direct calculations based on these lines can lead to inaccurate compensation results, affecting the final product. Therefore, in this embodiment, the intersecting shapes in the circuit diagram are first merged to obtain independent and closed polygons.

[0039] Then, each edge of each polygon is divided into several line segments, which are horizontal, vertical, or diagonal lines. For example, when a certain outline of the polygon is an arc, the arc outline can be divided into several fine diagonal lines. Therefore, in this embodiment, all line segment compensation is based on the fact that the line segments themselves are horizontal, vertical, or diagonal lines.

[0040] Obtaining the line width and line spacing relative to adjacent polygons of a line segment includes: emitting a normal from the line segment; the intersection of the normal with the polygon containing the line segment is the first intersection point; the intersection with the nearest point of the line segment among adjacent polygons is the second intersection point; the length of the normal between the line segment and the first intersection point is the line width; and the length of the normal between the line segment and the second intersection point is the line spacing. The line segment emits two normals, and the two normals are emitted in opposite directions.

[0041] In this embodiment, we assume that the line width and line spacing from any point on each small edge segment to other polygons are the same, such as... Figure 4As shown, the line width can be obtained by emitting an inward normal ray from the line segment, and the line spacing can be obtained by emitting an outward normal ray from the line segment. Thus, the line width and line spacing of the line segment are obtained. If the outward normal ray does not find an intersection with an adjacent polygon, the line spacing is set to infinity. Then, the infinite line spacing and the obtained line width are substituted into the preset compensation rule to obtain the corresponding compensation value. Of course, this is based on the fact that the line spacing and line width are calculated using a normal ray emitted from the line segment in the opposite direction. In other embodiments, a normal ray can be emitted in the same direction as the normal ray used to calculate the line width at the beginning to calculate the line spacing, or when the line spacing is determined to be infinite, a normal ray can be emitted again in the same direction as the normal ray used to calculate the line width, and then the line spacing can be calculated. These can all be explicitly set according to the preset calculation rule, and there are no restrictions on this.

[0042] Preferably, the preset compensation rule includes: calculating the line-to-space ratio and resolution accuracy based on the line width and line spacing of each line segment, and calculating the corresponding compensation value using the line-to-space ratio and the resolution accuracy, wherein the line-to-space ratio is the ratio of the line width to the line spacing, and the resolution accuracy is the smaller value between the line width and the line spacing. In this embodiment, the compensation value is calculated using the line-to-space ratio and the resolution accuracy.

[0043] Please see Figure 5 As shown, preferably, the preset rules include a line-to-space ratio compensation table. When the line-to-space ratio and the resolution are calculated, the corresponding compensation value is obtained based on the line-to-space ratio compensation table. The line-to-space ratio compensation table is as follows: Figure 5 As shown, "line" represents the presence of a graphic (line width), and "space" represents the absence of a graphic (line spacing). The line-to-space ratio is the ratio of line width to line spacing. The resolution of 4µm, 6µm, 8µm…20µm represents the smaller of the line width and line spacing of the segment to be compensated. When the line-to-space ratio is greater than 1, it represents the line spacing; when it is less than 1, it represents the line width. "Horizontal" indicates the compensation value corresponding to a horizontal line, and "vertical" indicates the compensation value corresponding to a vertical line. This line-to-space ratio compensation table can be pre-set based on previous experiments, thus providing greater convenience for subsequent operations.

[0044] In other words, a high line-to-space ratio (e.g., 5:1) means "the lines are wide and very close to each other (very small spacing)." During exposure, light diffracts severely in the narrow gaps, resulting in insufficient energy and making it difficult for the photoresist at the gaps to develop completely. During etching, the etchant does not exchange smoothly in the narrow gaps, slowing down the etching rate. The lines are prone to expansion, and the gaps tend to narrow, potentially leading to short circuits. Therefore, they need to be pre-thinned to counteract the expansion trend during manufacturing. A low line-to-space ratio (e.g., 1:5) means "the lines are very thin and far from other lines (very large spacing)." Isolated thin lines are not surrounded by adjacent patterns for obstruction or interference, and they are subjected to more thorough lateral attack or etching during exposure and etching. The lines are prone to shrinkage, becoming thinner and potentially leading to open circuits. Therefore, they need to be pre-thickened to counteract the shrinkage trend during manufacturing. A medium line-to-space ratio (e.g., 1:1) is in a relatively balanced critical state, with compensation values ​​falling between the two. Of course, the specific compensation values ​​can be adjusted according to different situations, and there are no restrictions on this.

[0045] When the line-to-space ratio and the analytical accuracy are within the range of the line-to-space ratio compensation table, but there is no corresponding compensation value, bilinear interpolation is used based on the line-to-space ratio compensation table to obtain the corresponding compensation value.

[0046] Given the following four adjacent sets of analytical precision values ​​and line-to-space ratios in the table:

[0047]

[0048] The corresponding compensation accuracy is:

[0049]

[0050] Then the resolution Linear space ratio Data ,exist Linear interpolation is performed on the above, for 1. The interpolation formula is:

[0051]

[0052] for 2. The interpolation formula is:

[0053]

[0054] Then Interpolation is performed in the direction, and the final interpolation result is:

[0055]

[0056] That is, the final compensation value is obtained through bilinear interpolation.

[0057] When the line-to-space ratio and the resolution are outside the range of the line-to-space ratio compensation table, a fixed compensation value is directly selected based on the division of the line-to-space ratio and the resolution in the line-to-space ratio compensation table, according to a fixed rule.

[0058] In this embodiment, the data outside the table range is divided into a total of 8 regions, such as... Figure 6 As shown:

[0059] Region 1: Line-to-space ratio less than 1:5, resolution less than 4µm, directly use the compensation value in the upper left corner of the table, i.e. -0.7µm.

[0060] Region 2: The line-to-space ratio is between 1:5 and 5:1, and the resolution is less than 4um. The compensation values ​​interpolated in the first column of the table are used, namely -0.70um, -0.89um, -1.08um, -1.82um, -2.03um, -0.80um, -1.08um, -1.06um, -1.30um, and -1.75um.

[0061] Region 3: Line-to-space ratio greater than 5:1, resolution less than 4µm, compensation value in the lower left corner of the table, i.e. -1.75µm.

[0062] Region 4: Line-to-space ratio greater than 5:1, resolution between 4um and 20um, use the interpolation compensation values ​​from the last row of the table, namely -1.75um, -1.84um, -1.72um, -1.47um, -1.58um, -1.56um.

[0063] Region 5: Line-to-space ratio greater than 5:1, resolution greater than 20µm, use the compensation value in the lower right corner of the table, i.e. -1.56µm.

[0064] Region 6: Line-to-space ratio is between 1:5 and 5:1, resolution is greater than 20um, use the interpolation compensation values ​​in the last column of the table, namely -0.81um, -0.87um, -1.37um, -1.55um, -1.54um, -0.87um, -0.98um, -1.33um, -1.41um, -1.56um.

[0065] Region 7: Line-to-space ratio less than 1:5, resolution greater than 20µm, directly use the compensation value in the upper right corner of the table, i.e. -0.81µm.

[0066] Region 8: Line-to-space ratio less than 1:5, resolution between 4um and 20um, use the interpolation compensation values ​​in the first row of the table, namely -0.70um, -1.00um, -1.04um, -0.94um, -0.83um, -0.81um.

[0067] Of course, the data in this embodiment are all obtained from experiments. In other embodiments, adjustments can be made according to the actual situation, as long as the rules of this embodiment are met.

[0068] like Figure 5 As shown, the line-space ratio compensation table includes a horizontal line compensation table and a vertical line compensation table. When the line segment is a horizontal line or a vertical line, the corresponding compensation value is obtained based on the horizontal line compensation table or the vertical line compensation table. When the line segment is a diagonal line, the corresponding compensation value is obtained based on the horizontal line compensation table and the vertical line compensation table.

[0069] Define the angle between the diagonal line and the horizontal line as θ, and obtain the first compensation value based on the horizontal line compensation table using the line width and spacing of the diagonal line. The second compensation value is obtained based on the vertical line compensation table. Through the formula: The compensation value val corresponding to the oblique line is calculated.

[0070] If there is no corresponding value in the horizontal or vertical line compensation table, you can go back to the previous step and determine the compensation value for the horizontal or vertical line using bilinear interpolation or fixed rules, and then calculate the compensation value for the diagonal line using the formula.

[0071] In this embodiment, for values ​​not outside the table, a bilinear interpolation algorithm is used to compensate using data from surrounding cells. In practice, for more complex situations, mathematical modeling of the interpolation can be performed to obtain compensation values ​​that are more consistent with reality.

[0072] like Figure 7 As shown, after obtaining the compensation value based on the line-to-space ratio analysis, corresponding positive and negative polygons with the compensation value are added. For example... Figure 8 As shown, a polygon merging operation is performed to obtain the graphic after adding compensation values. It can be seen that the original graphic with horizontal lines of 18µm line width and 6µm line spacing has had a compensation value of -1.95µm added. For more complex graphics, a graphic smoothing operation can be added to reduce the amount of data in the graphic.

[0073] The positive and negative polygons include positive polygons (i.e., polygons that will expand outwards under the compensation value) and negative polygons (i.e., polygons that will shrink inwards under the compensation value). All positive polygons are merged to obtain the etched area, and all negative polygons are merged to obtain the retained area. Finally, a complete graphic with added compensation values ​​is obtained.

[0074] Preferably, in the above process, after merging all polygons as a whole, the processing of each polygon is independent in subsequent steps, so parallel operations can be performed, which speeds up the processing of the overall polygons.

[0075] In summary, the proportional dynamic etching compensation algorithm of the present invention merges the extracted graphic features to form at least two independent and closed polygons; divides each edge of the polygon into several line segments and obtains the line width and line spacing of each line segment relative to the adjacent polygons; calculates the compensation value corresponding to the line width and line spacing of each line segment based on a preset compensation rule; and reconstructs the polygon based on the calculated compensation value. In this way, by calculating the line width and line spacing of each line segment, accurate compensation is achieved for each line segment, overcoming the overcompensation or undercompensation defects caused by the traditional "one-size-fits-all" compensation in the layout where sparse and dense graphics coexist.

[0076] The above description is merely a preferred embodiment of the present invention and is not intended to limit the present invention. It should be noted that for those skilled in the art, several improvements and modifications can be made without departing from the technical principles of the present invention, and these improvements and modifications should also be considered within the scope of protection of the present invention.

Claims

1. A proportional dynamic etching compensation algorithm, characterized in that, include: Acquire the image to be processed and extract its features; The intersecting graphic features in the graphic features are merged to process the graphic to be processed into at least two independent and closed polygons; Each edge of the polygon is discretized into several line segments, and the line width and line spacing of the line segment relative to the adjacent polygon are obtained. The compensation value corresponding to the line width and line spacing of each line segment is calculated based on the preset compensation rules. The polygon is reconstructed based on the calculated compensation value.

2. The proportional dynamic etching compensation algorithm according to claim 1, characterized in that, Obtaining the line width and line spacing relative to adjacent polygons of the line segment includes: emitting a normal from the line segment, the intersection point of the normal with the polygon containing the line segment being the first intersection point, the intersection point with the adjacent polygon closest to the line segment being the second intersection point, the length of the normal between the line segment and the first intersection point being the line width, and the length of the normal between the line segment and the second intersection point being the line spacing.

3. The proportional dynamic etching compensation algorithm according to claim 2, characterized in that, The line segment emits two normals, and the two normals emit in opposite directions.

4. The proportional dynamic etching compensation algorithm according to claim 1, characterized in that, The line segment can be one of the following: horizontal, vertical, or diagonal.

5. The proportional dynamic etching compensation algorithm according to claim 4, characterized in that, The preset compensation rules include: calculating the line-to-space ratio and resolution accuracy based on the line width and line spacing of each line segment, and calculating the corresponding compensation value through the line-to-space ratio and the resolution accuracy, wherein the line-to-space ratio is the ratio of the line width and the line spacing, and the resolution accuracy is the smaller value between the line width and the line spacing.

6. The proportional dynamic etching compensation algorithm according to claim 5, characterized in that, The preset rules include a line-to-space ratio compensation table. When the line-to-space ratio and the resolution are calculated, the corresponding compensation value is obtained based on the line-to-space ratio compensation table.

7. The proportional dynamic etching compensation algorithm according to claim 6, characterized in that, When the line-to-space ratio and the analytical accuracy are within the range of the line-to-space ratio compensation table, but there is no corresponding compensation value, bilinear interpolation is used based on the line-to-space ratio compensation table to obtain the corresponding compensation value.

8. The proportional dynamic etching compensation algorithm according to claim 6, characterized in that, When the line-to-space ratio and the resolution are outside the range of the line-to-space ratio compensation table, a fixed compensation value is directly selected based on the division of the line-to-space ratio and the resolution in the line-to-space ratio compensation table.

9. The proportional dynamic etching compensation algorithm according to claim 6, characterized in that, The line-space ratio compensation table includes a horizontal line compensation table and a vertical line compensation table. When the line segment is a horizontal line or a vertical line, the corresponding compensation value is obtained based on the horizontal line compensation table or the vertical line compensation table. When the line segment is a diagonal line, the corresponding compensation value is obtained based on the horizontal line compensation table and the vertical line compensation table.

10. The proportional dynamic etching compensation algorithm according to claim 9, characterized in that, Define the angle between the diagonal line and the horizontal line as θ, and obtain the first compensation value based on the horizontal line compensation table using the line width and spacing of the diagonal line. The second compensation value is obtained based on the vertical line compensation table. Through the formula: The compensation value val corresponding to the oblique line is calculated.

Images