Method for optimizing line width uniformity of maskless photoetching machine and photoetching system

Abstract

The invention provides a method for optimizing the line width uniformity of a maskless photoetching machine and a photoetching system.The method comprises the steps that firstly, through exposure and measurement of a specific test pattern, line width distribution data of all positions in an exposure area are obtained; calculating a photoetching data adjustment amount required for compensating uneven energy distribution on the basis of the data; and finally, digital adjustment of local exposure energy is realized by adjusting gray values of corresponding pixels in photoetching graphic data of a target product, so that line width errors caused by characteristics or drifting of an optical system are compensated. Under the condition that hardware optical setting does not need to be adjusted, the uniformity of the pattern line width can be effectively improved, and the process stability and the equipment adaptability are improved.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of photoetching machine, in particular, to a method for optimizing line width uniformity of maskless photoetching machine and a photoetching system. BACKGROUND

[0002] As a new emerging semiconductor patterning process, maskless photoetching technology abandons the physical mask used in traditional photoetching, and instead directly controls the irradiation light beam through a spatial light modulator (such as a digital micro-mirror device, DMD), so as to generate the required pattern on the substrate surface. This technology has the advantages of high flexibility, short cycle, relatively low cost, etc., and is particularly suitable for small batch, multi-variety or research and development stage device manufacturing.

[0003] In the actual operation of the maskless photoetching system, the stability and uniformity of the optical performance are crucial. Among them, two key optical indicators directly affect the quality of the final photoetching pattern: one is the perpendicularity of the optical axis relative to the surface of the workpiece (usually referred to as the declination or perpendicularity), and the other is the uniformity of the in-plane light intensity distribution of the DMD modulation. The ideal perpendicularity can ensure that the exposure light beam is perpendicular to the incident, so that the photoresist forms a steep and clear profile after development. The ideal in-plane energy uniformity can ensure that the light dose received at each position in the exposure area is consistent, thereby obtaining a line width uniform and size accurate pattern.

[0004] However, in the actual equipment debugging and long-term use process, the above two indicators often have a mutually restrictive relationship. When the optical axis is perpendicular to the processing surface, due to the inconsistent reflection ability of each small mirror of the DMD, the energy of the processing surface is inconsistent, causing the processed lines to be uneven; if the optical lens is adjusted to improve the uniformity of the energy on the processing surface, the optical axis will be inclined, causing the side of the pattern to be inclined. This contradiction makes it difficult to achieve long-term optimal state of both indicators simply by adjusting the hardware.

[0005] Therefore, the industry urgently needs a method that can effectively compensate for the pattern line width error caused by changes in optical properties under the condition that the hardware state is relatively fixed. SUMMARY

[0006] In view of the defects in the prior art, the purpose of the present application is to provide a method for optimizing the line width uniformity of a maskless photoetching machine and a photoetching system.

[0007] According to the method for optimizing the line width uniformity of a maskless photoetching machine provided by the present application, the following steps are included: Step S1: Expose a test pattern on the substrate using a maskless lithography machine. The test pattern includes multiple measurement marks distributed within the scanning exposure area. Step S2: Develop the exposed substrate to form a solid pattern corresponding to the test pattern; Step S3: Measure the line width of each of the measurement marks in the entity graphic in at least one direction to obtain a line width measurement dataset; Step S4: Based on the linewidth measurement dataset, calculate the deviation between the linewidth dimensions at different locations within the scanned exposure area and the target linewidth dimensions; Step S5: Based on the deviation value, calculate the amount of exposure energy adjustment required to compensate for the deviation at the corresponding position; Step S6: Based on the exposure energy adjustment amount, process the original lithographic pattern data of the target product to generate compensated lithographic data; wherein, the processing includes adjusting the data value in the pattern boundary area of ​​the original lithographic pattern data according to the exposure energy adjustment amount to simulate the enhancement or reduction of exposure energy; Step S7: Use the compensated lithography data to control the maskless lithography machine to expose the product.

[0008] Preferably, the test pattern is a cross-shaped mark distributed in an array, and in step S3, the line width of each cross-shaped mark is measured in the horizontal and vertical directions respectively.

[0009] Preferably, step S4 specifically includes: arranging the linewidth measurement data according to their positions within the scanning exposure area, calculating the average value of the measurement data in the same row or column in at least one direction, comparing the actual measurement value at each position with the average value in the corresponding direction, and obtaining the deviation value.

[0010] Preferably, in step S5, the calculation of the exposure energy adjustment is based on the energy-linewidth response characteristics of the photoresist, and a linear model is used to convert the linewidth deviation value into the corresponding energy adjustment value.

[0011] Preferably, in step S6, the original photolithographic pattern data is in bitmap format, and the adjustment data value specifically refers to adjusting the grayscale value of the corresponding pixel in the bitmap.

[0012] Preferably, in step S6, when it is necessary to increase the line width, the grayscale value of the pixels in the adjacent non-graphic area outside the graphic is increased; when it is necessary to decrease the line width, the grayscale value of the pixels in the edge area of ​​the graphic is decreased.

[0013] Preferably, when increasing the grayscale value of pixels in non-graphic areas, the initial grayscale value is not lower than the data value corresponding to the photosensitive threshold of the photoresist.

[0014] A maskless photolithography system according to the present invention includes: Maskless lithography machine, which includes digital micromirror devices for patterned exposure; Line width measuring device, used to measure the line width of patterns on a substrate after development; The data processing unit is configured to execute the method for optimizing the linewidth uniformity of a maskless lithography machine to generate compensated lithography data and send it to the maskless lithography machine.

[0015] Compared with the prior art, the present invention has the following beneficial effects: 1. This invention preprocesses lithography data using software algorithms to directly compensate for uneven energy distribution in the DMD plane caused by inherent characteristics of the optical system or optical axis misalignment. This allows for prioritizing optimal optical perpendicularity in hardware without compromising on the optical path for uniformity, thus resolving the technical contradiction in the prior art where two key indicators are difficult to balance.

[0016] 2. Compared to hardware calibration methods that require downtime and precise optical adjustments, the software compensation scheme of this invention offers faster adjustment speeds and greater operational flexibility. New compensation data can be generated simply through test exposure, measurement, and data processing, without requiring changes to physical components, significantly reducing maintenance costs, time costs, and technical barriers.

[0017] 3. This invention establishes a quantitative compensation mechanism based on measured linewidth data and a photoresist energy response model, and considers actual process parameters such as the photosensitivity threshold, making the calculation of energy adjustment more accurate. Attached Figure Description

[0018] Other features, objects, and advantages of the present invention will become more apparent from the following detailed description of non-limiting embodiments with reference to the accompanying drawings: Figure 1 This is a schematic diagram of the test pattern array distribution in an embodiment of the present invention; Figure 2 This is a schematic diagram of the measurement of the cross mark line width in an embodiment of the present invention; Figure 3 This is a schematic diagram of line width measurement data statistics in an embodiment of the present invention; Figure 4 This is a schematic diagram of a partial area of ​​the original bitmap in an embodiment of the present invention; Figure 5 This is a partial schematic diagram of the bitmap after energy compensation in an embodiment of the present invention. Detailed Implementation

[0019] The present invention will now be described in detail with reference to specific embodiments. These embodiments will help those skilled in the art to further understand the present invention, but do not limit the invention in any way. It should be noted that those skilled in the art can make several changes and improvements without departing from the concept of the present invention. These all fall within the protection scope of the present invention.

[0020] This invention aims to provide a method for improving the linewidth uniformity of maskless lithography machines by optimizing lithography data. The core idea is as follows: First, the linewidth distribution error of the current lithography system within the exposure area is measured and quantified using a standard test pattern containing distribution characteristics. Then, based on this error data, the required local exposure energy adjustment to compensate for this error is calculated in reverse. Finally, this adjustment is applied to the lithography pattern data of an actual product, and pre-distortion compensation is performed on the data through digital image processing. This achieves the elimination of systematic energy distribution unevenness during physical exposure without changing the hardware optical settings, resulting in a more uniform linewidth.

[0021] Phase 1: Testing and Calibration Phase. The goal of this phase is to obtain the energy-linewidth distribution characteristics of the current lithography system under specific process conditions. This includes the following steps: Step S101: Prepare and expose test pattern.

[0022] First, a standard test pattern file is designed. This test pattern preferably contains a series of regularly arranged measurement markers to systematically evaluate the entire exposure field of view. In a preferred embodiment, as shown in the attached... Figure 1 As shown, an array of 64 cross-shaped markers in 8 rows (AH) and 8 columns (1-8) is used. Each cross-shaped marker consists of clearly defined horizontal and vertical lines. This test pattern is converted into exposure instructions executable by the lithography machine through the standard data processing flow of a maskless lithography machine.

[0023] Step S102: Perform exposure and post-processing.

[0024] Using the data generated above, a maskless lithography machine is controlled to expose a substrate coated with a specific photoresist. After exposure, standard post-baking, developing, and etching processes are performed on the substrate to form a solid structure on the substrate, such as trenches or lines, based on the exposed pattern.

[0025] Step S103: Measure the line width data.

[0026] The 64 cross-shaped markings formed on the substrate were measured one by one using critical dimension measuring equipment (such as CD-SEM). (Refer to...) Figure 2As shown, for each cross mark, the width of its horizontal line in the vertical direction (denoted as the Y value) and the width of its vertical line in the horizontal direction (denoted as the X value) are measured respectively. Finally, two sets (X group and Y group) of 64 line width measurement data are obtained, each data corresponding to a specific physical position (A1 to H8) in the exposure field of view.

[0027] Step S104: Data processing and deviation analysis.

[0028] The measured data will be organized and analyzed to reveal the distribution pattern of linewidth within the field of view.

[0029] As attached Figure 3 As shown, the 64 measurements in the X direction are filled into an 8x8 two-dimensional matrix according to their corresponding physical locations (A1-H8). Similarly, the measurements in the Y direction are filled into another matrix.

[0030] To analyze the overall trend of line width in a certain direction, statistical analysis can be performed on the data. For example, to analyze the distribution of line width in the horizontal direction (X direction), the average X value of each column (e.g., A1-H1, A2-H2, ..., A8-H8) can be calculated. These eight column averages constitute a profile curve describing the distribution of line width along the horizontal axis in the X direction (e.g., ...). Figure 3 (As shown by the gray broken line). This profile curve reflects the inherent uniformity of the energy distribution in the X direction of the system. The analysis in the Y direction is similar.

[0031] Define the target linewidth. Typically, the ideal linewidth for a given direction can be the average of all measurements taken within the entire field of view (or the target value set by the process).

[0032] Calculate the deviation. For each measurement location (e.g., location A2), subtract the measured value in the X direction from the target line width in the X direction (e.g., the aforementioned total average value in the X direction) to obtain the line width deviation value ΔX in the X direction at that location (e.g., if the target is 1.29 μm and the measured value is 1.266 μm, then ΔX = 0.024 μm, indicating that the line width is too narrow and the vertical line needs to be thickened by 0.028). Similarly, calculate the line width deviation value ΔY in the Y direction (e.g., if the target is 1.32 μm and the measured value is 1.309 μm, then ΔY = 0.011 μm, indicating that the line width is too narrow and the horizontal line needs to be thickened by 0.011). These deviation values ​​ΔX and ΔY quantify the energy error of the system at various local locations.

[0033] The second stage: establishing the energy compensation map. The goal of this stage is to convert linewidth deviation into operable data compensation parameters. Specifically, it includes the following steps: Step S201: Establish the deviation-energy regulation model.

[0034] There is a correlation between the linewidth of a photoresist and the received exposure dose (energy). For positive photoresists, within a certain process window, the linewidth increases with increasing exposure dose (the opposite is true for negative photoresists). Based on this principle, a simple linear transformation model can be established.

[0035] For example, if the size of a single pixel is known to be 1 micrometer, and the 256 gray levels of an 8-bit bitmap correspond to an energy coverage range from just underexposed to saturated exposure, then the linewidth transformation amount K can be calculated as: pixel size / total number of gray levels = 1 / 256. Therefore, for ΔX = 0.024 μm, the calculated exposure compensation amount ΔG = (0.024) / (1 / 256) ≈ 6.14, rounded down to 6. A positive value indicates that more energy is needed to widen the linewidth.

[0036] Step S202: Consider the photosensitivity threshold.

[0037] When implementing energy compensation, special attention must be paid to the photoresist's photosensitivity threshold. For areas requiring increased energy (corresponding to wider linewidths), compensation is typically achieved by appropriately increasing the grayscale value in the unexposed areas at the edge of the pattern (originally black / low grayscale areas in the bitmap). If the initial grayscale value of this area is 0 (representing no exposure), directly adding a calculated ΔG value may still result in a new grayscale value below the critical value (photoresist sensitivity threshold) for the photoresist to react, rendering the compensation ineffective. Therefore, when increasing light intensity in black areas, the photosensitivity threshold needs to be pre-assigned. Assuming the photoresist currently in use has a photosensitivity threshold of 5, which translates to a grayscale value of 8, meaning that to increase brightness by 8 in black areas, the actual calculation requires an increase of 8 + 8 = 16 brightness.

[0038] The third stage: Compensation processing of product graphic data. This stage applies the established compensation rules to the lithography data of the actual product. Specifically, it includes the following steps: Reference Figure 4 As shown, for directions where the graphic needs to be widened (e.g., the Y direction needs to be widened): Find the adjacent background (black) pixels along the outer edge of the graphic (white) in that direction. Increase the grayscale values ​​of these background pixels based on the calculation results of steps S201 and S202 (e.g., to increase the brightness in the Y direction by 2, plus a photosensitivity threshold of 8, i.e., to increase the brightness of the black area by 10). This is equivalent to slightly expanding outwards from the edge of the graphic, as shown in the attached diagram. Figure 5 As shown by the light gray line appearing above the horizontal line.

[0039] For directions where the image needs to be narrowed: the grayscale value of the inner edge pixels of the image (white) in that direction is appropriately reduced based on the calculation results. This is equivalent to reducing the exposure intensity of the image edges, causing the image edges to slightly shrink inward after development.

[0040] Since a single pixel location may involve compensation in both the X and Y directions, a comprehensive consideration is required during the operation. After the compensation operations at all locations are completed, the final "compensated lithography data" with uniformity optimization is generated.

[0041] The generated compensated lithography data is loaded into a maskless lithography machine for exposing the product substrate. Since the data already includes pre-compensation information for the uneven energy distribution of the current system, the inherent energy inhomogeneity of the system will be effectively canceled after actual exposure, thereby achieving significantly improved linewidth uniformity in the developed product pattern.

[0042] Based on the same inventive concept, this invention also provides a maskless lithography system, comprising: a maskless lithography machine (including a DMD), a linewidth measurement device (such as a CD-SEM), and a data processing unit. The data processing unit integrates or connects a memory and a processor, wherein the memory stores a computer program. When executed by the processor, the program can automatically or semi-automatically implement the method steps described in the "testing and calibration stage," "establishment of energy compensation mapping," and "compensation processing of product graphic data," ultimately outputting optimized lithography data to the lithography machine.

[0043] In summary, this invention uses software algorithms to intelligently preprocess lithography data, cleverly compensating for energy distribution defects that are difficult for hardware systems to completely eliminate. While maintaining the optimization of hardware indicators such as optical perpendicularity, it significantly improves the uniformity of pattern linewidth, enhances process stability, and improves the overall performance of the equipment.

[0044] Specific embodiments of the present invention have been described above. It should be understood that the present invention is not limited to the specific embodiments described above, and those skilled in the art can make various changes or modifications within the scope of the claims, which do not affect the essence of the present invention. Unless otherwise specified, the embodiments and features described in this application can be arbitrarily combined with each other.

Claims

1. A method for optimizing the linewidth uniformity of a maskless lithography machine, characterized in that, Includes the following steps: Step S1: Expose a test pattern on the substrate using a maskless lithography machine. The test pattern includes multiple measurement marks distributed within the scanning exposure area. Step S2: Develop the exposed substrate to form a solid pattern corresponding to the test pattern; Step S3: Measure the line width of each of the measurement marks in the entity graphic in at least one direction to obtain a line width measurement dataset; Step S4: Based on the linewidth measurement dataset, calculate the deviation between the linewidth dimensions at different locations within the scanned exposure area and the target linewidth dimensions; Step S5: Based on the deviation value, calculate the amount of exposure energy adjustment required to compensate for the deviation at the corresponding position; Step S6: Based on the exposure energy adjustment amount, process the original lithographic pattern data of the target product to generate compensated lithographic data; wherein, the processing includes adjusting the data value in the pattern boundary area of ​​the original lithographic pattern data according to the exposure energy adjustment amount to simulate the enhancement or reduction of exposure energy; Step S7: Use the compensated lithography data to control the maskless lithography machine to expose the product.

2. The method for optimizing linewidth uniformity in a maskless lithography machine according to claim 1, characterized in that, The test pattern is a cross-shaped mark arranged in an array. In step S3, the line width of each cross-shaped mark is measured in both the horizontal and vertical directions.

3. The method for optimizing the linewidth uniformity of a maskless lithography machine according to claim 1 or 2, characterized in that, Step S4 specifically includes: arranging the linewidth measurement data according to their positions within the scanning exposure area, calculating the average value of the measurement data in the same row or column in at least one direction, comparing the actual measurement value at each position with the average value in the corresponding direction, and obtaining the deviation value.

4. The method for optimizing linewidth uniformity in a maskless lithography machine according to claim 1, characterized in that, In step S5, the calculation of the exposure energy adjustment is based on the energy-linewidth response characteristics of the photoresist, and a linear model is used to convert the linewidth deviation value into the corresponding energy adjustment value.

5. The method for optimizing linewidth uniformity in a maskless lithography machine according to claim 1, characterized in that, In step S6, the original photolithographic pattern data is in bitmap format, and the adjustment data value specifically refers to adjusting the grayscale value of the corresponding pixel in the bitmap.

6. The method for optimizing linewidth uniformity in a maskless lithography machine according to claim 5, characterized in that, In step S6, when it is necessary to increase the line width, the grayscale value of the pixels in the adjacent non-graphic area outside the graphic is increased; when it is necessary to decrease the line width, the grayscale value of the pixels in the edge area of ​​the graphic is decreased.

7. The method for optimizing linewidth uniformity in a maskless lithography machine according to claim 6, characterized in that, When increasing the grayscale value of pixels in non-graphic areas, the initial grayscale value set is not lower than the data value corresponding to the photosensitive threshold of the photoresist.

8. A maskless photolithography system, characterized in that, include: Maskless lithography machine, which includes digital micromirror devices for patterned exposure; Line width measuring device, used to measure the line width of patterns on a substrate after development; The data processing unit is configured to perform the method for optimizing the linewidth uniformity of a maskless lithography machine as described in any one of claims 1 to 7, to generate compensated lithography data and send it to the maskless lithography machine.

Images