Optimization method of photoetching alignment mark
By comparing and analyzing the signal and pattern of the lithographic alignment mark, optimizing the lithographic alignment mark of the aluminum-silicon copper metal layer, inserting precompensation marks to improve the alignment accuracy, solving the problem of insufficient lithographic alignment accuracy in the prior art, and significantly improving the product yield and performance stability.
Patent Information
- Application Number
- CN202311719801.9
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- Filing Date
- 2023-12-13
- Publication Date
- 2025-06-13
AI Technical Summary
Conventional methods for improving the photolithography alignment accuracy in the prior art cannot effectively improve the photolithography alignment accuracy of the aluminum-silicon copper metal layer.
By obtaining the alignment signal of the original alignment mark and its pattern, analyzing the source and location of the noise, and optimizing the original alignment mark based on this, inserting the precompensation mark for reverse precompensation until the optimized alignment mark is obtained.
It effectively solves the problem of the impact of silicon defects on alignment stability in aluminum-silicon copper metal layer, significantly improves the photolithographic alignment accuracy, and improves product yield and performance stability.
Smart Images
Figure CN120143555A_ABST
Abstract
Description
Technical Field
[0001] The invention belongs to the technical field of semiconductor manufacturing and relates to an optimization method for photolithography alignment marks. Background Art
[0002] The production process of semiconductor chips includes many steps, among which the photolithography process is the core of chip production and a key step in IC manufacturing. Its precision determines the chip process and device performance. In actual production, the production of a chip may go through dozens of photolithography, and some structural layers even require multiple photolithography to form. The main function of photolithography is to transfer the chip design graphics on the mask to the silicon wafer. It usually requires a whole set of complex processes such as silicon wafer surface cleaning and drying, primer coating, spin coating of photoresist, soft baking, alignment exposure, post-baking, development, hard baking, laser etching, etc. Among them, photolithography alignment technology is an important step before exposure and is one of the core technologies of photolithography.
[0003] Photolithography alignment refers to aligning the pattern on the current mask with the existing pattern on the wafer. Figure 1 , showing a general alignment mark pattern and a waveform diagram of an ideal alignment signal based on the alignment mark output. Usually, the photolithography alignment mark 100 is composed of lines 101 periodically arranged on a horizontal plane (including alignment marks for vertical alignment and alignment marks for horizontal alignment). The alignment principle of the photolithography machine alignment is to irradiate the alignment light onto the photolithography alignment mark 100, use a detector to identify the intensity of the reflected light after the alignment light is reflected by the photolithography alignment mark 100 and convert it into an alignment signal 200 output. The ideal alignment signal 200 is a standard waveform, and the position fitting is performed through the peak and trough positions of the alignment signal 200, thereby realizing the alignment of the pattern on the mask and the pattern on the wafer. With the improvement of the technical level of the photolithography machine, when the characteristic size of the line width is reduced to the nanometer level, the performance requirements of the photolithography machine alignment system are put forward for higher alignment accuracy.
[0004] The current method to improve alignment accuracy is to change the size of the lines (such as widening, i.e. increasing the size) while keeping the spacing between the lines in the standard alignment mark unchanged. Figure 1 The value of d in the figure is used to obtain an alignment mark with higher alignment accuracy. After process verification, it is found that this method can indeed greatly improve the alignment accuracy. However, after using this method to improve the alignment mark in the aluminum-silicon-copper metal layer, the alignment accuracy cannot meet the alignment requirements. Of course, the same situation may also exist in the structural layers of other materials.
[0005] Therefore, how to solve the problem that conventional methods for improving alignment accuracy in the prior art cannot effectively improve the photolithography alignment accuracy of the aluminum-silicon-copper metal layer has become an important technical problem that needs to be urgently solved by those skilled in the art.
[0006] It should be noted that the above introduction of the technical background is only for the convenience of clearly and completely explaining the technical solution of the present application and facilitating the understanding of those skilled in the art. It cannot be considered that the above technical solutions are well-known to those skilled in the art just because these solutions are described in the background art part of the present application. Summary of the Invention
[0007] In view of the above-mentioned disadvantages of the prior art, the purpose of the present invention is to provide an optimization method for lithography alignment marks, which is used to solve the problem that the conventional methods for improving alignment accuracy in the prior art cannot effectively improve the lithography alignment accuracy of the aluminum-silicon-copper metal layer.
[0008] To achieve the above purpose and other related purposes, the present invention provides an optimization method for lithography alignment marks, including the following steps:
[0009] Provide an original alignment mark;
[0010] Obtain the original alignment signal of the original alignment mark and compare the original alignment signal with the pattern of the original alignment mark to obtain a comparison result, where the comparison result includes the noise source affecting the accuracy of the original alignment signal and the position of the noise source;
[0011] Optimize the original alignment mark based on the comparison result according to the first optimization rule to obtain a first alignment mark;
[0012] Verify the alignment accuracy of the first alignment mark. If the alignment accuracy of the first alignment mark meets the alignment requirements, use the first alignment mark as the optimized alignment mark; if the alignment accuracy of the first alignment mark does not meet the alignment requirements, optimize the first alignment mark according to the second optimization rule until an optimized alignment mark is obtained.
[0013] Optionally, providing an original alignment mark and obtaining the original alignment signal of the original alignment mark and comparing the original alignment signal with the pattern of the original alignment mark include the following steps:
[0014] Provide a first wafer;
[0015] Form the original alignment mark in a preset area of the first wafer, and place the wafer with the original alignment mark formed therein in a lithography machine to collect the original alignment signal;
[0016] Compare the original alignment signal with the pattern of the original alignment mark to analyze the noise source and record the position of the noise source.
[0017] Optionally, the first wafer includes an aluminum-silicon-copper metal layer, and the original alignment marks are formed in the aluminum-silicon-copper metal layer.
[0018] Optionally, the noise source includes at least one of silicon precipitation in the original alignment marks and silicon residue in the gaps of the original alignment marks.
[0019] Optionally, the first optimization rule is to insert pre-compensation marks at the positions of the noise sources for reverse pre-compensation.
[0020] Optionally, the method of inserting pre-compensation marks at the positions of the noise sources includes roughening etching at the positions of the noise sources to form pre-compensation marks.
[0021] Optionally, the pattern of the original alignment marks includes a plurality of first strip marks arranged at intervals in the horizontal direction, and the pre-compensation marks include a plurality of second strip marks arranged at intervals in the horizontal direction.
[0022] Optionally, the first optimization rule further includes increasing the width of at least a part of the first strip marks adjacent to the positions of the noise sources.
[0023] Optionally, the second optimization rule includes adjusting at least one of the number of the second strip marks in the pre-compensation marks, the width of the second strip marks, and the spacing between two adjacent second strip marks.
[0024] Optionally, verifying the alignment accuracy of the first alignment marks includes the following steps:
[0025] Provide a second wafer and form first alignment marks in a preset area of the second wafer;
[0026] Place the second wafer with the first alignment marks formed thereon in a lithography machine for multiple lithography alignments and record the position coordinates of the first alignment marks each time during lithography alignment;
[0027] Calculate the range of position coordinate deviations of the first alignment marks during multiple lithography alignments. If the range of position coordinate deviations is within a preset range, the alignment accuracy of the first alignment marks meets the alignment requirements; if the range of position coordinate deviations exceeds the preset range, the alignment accuracy of the first alignment marks does not meet the alignment requirements.
[0028] Optionally, the preset range is less than 20 nm.
[0029] As described above, the optimization method of the lithography alignment mark of the present invention obtains the original alignment signal of the original alignment mark, compares the original alignment signal with the pattern of the original alignment mark to obtain the noise source and the position of the noise source that affect the accuracy of the original alignment signal, and optimizes the original alignment mark based on the comparison result to obtain the first alignment mark, thereby effectively solving the problem that the silicon defects in the aluminum-silicon-copper metal layer affect the alignment stability, effectively improving the alignment accuracy, and greatly improving the product yield and performance stability. BRIEF DESCRIPTION OF THE DRAWINGS
[0030] Figure 1 Shown is a general alignment mark pattern and a waveform diagram of an ideal alignment signal output based on the alignment mark.
[0031] Figure 2 Shown is a waveform diagram of the alignment signal collected during the lithography alignment of the lithography alignment mark in the aluminum-silicon-copper metal layer and a picture of the lithography alignment mark.
[0032] Figure 3 Shown is a flowchart of the steps of the optimization method of the lithography alignment mark of the present invention.
[0033] Figure 4 Shown is a schematic structural diagram of the original alignment mark in the optimization method of the lithography alignment mark of the present invention.
[0034] Figure 5 Shown is a schematic structural diagram of the first alignment mark in the optimization method of the lithography alignment mark of the present invention.
[0035] Figure 6 Shown is a picture of the original alignment mark without optimization and its corresponding waveform diagram of the alignment signal.
[0036] Figure 7 Shown is a picture of the first alignment mark optimized by using the optimization method of the present invention and its corresponding waveform diagram of the alignment signal.
[0037] Figure 8 Shown is a physical optical microscopic picture after alignment based on the original alignment mark.
[0038] Figure 9 Shown as Figure 8 a simplified effect diagram of region C in
[0039] Figure 10 Shown is a physical optical microscopic picture after alignment based on the first alignment mark.
[0040] Figure 11 Shown as Figure 10 a simplified effect diagram of region D in
[0041] DESCRIPTION OF REFERENCE NUMERALS
[0042] 100 Lithography alignment mark
[0043] 101 Strip area
[0044] 200 Alignment signal
[0045] 10 Original alignment mark
[0046] 11 First strip mark
[0047] 20 First alignment mark
[0048] 21 Pre-compensation mark
[0049] 211 Second strip mark
[0050] 30a, 30b First pattern
[0051] 40a, 40b Second pattern
[0052] Regions A - D
[0053] Steps S1 - S4 Detailed implementation manners
[0054] The following illustrates the implementation manners of the present invention through specific examples. Those skilled in the art can easily understand other advantages and effects of the present invention from the content disclosed in this specification. The present invention can also be implemented or applied through other different specific implementation manners. Various details in this specification can also be modified or changed based on different viewpoints and applications without departing from the spirit of the present invention.
[0055] Please refer to Figures 2 to 11 . It should be noted that the diagrams provided in this embodiment only illustrate the basic concept of the present invention in a schematic manner. Therefore, only the components related to the present invention are shown in the diagrams, rather than being drawn according to the number, shape, and size of the components in actual implementation. The types, quantities, and proportions of the components in actual implementation can be arbitrarily changed, and the component layout type may also be more complex.
[0056] After analyzing the problem that the method of changing the line size while keeping the spacing between lines in the existing standard alignment marks unchanged to obtain higher alignment accuracy cannot effectively improve the lithography alignment accuracy of the aluminum-silicon-copper metal layer, please refer to Figure 2 , which shows the alignment signal waveform diagram and the lithography alignment mark picture (collected by the lithography machine stage) obtained when performing lithography alignment on the lithography alignment marks in the aluminum-silicon-copper metal layer. Figure 2 Part A in shows the morphology presented when defects occur in the lithography alignment mark. Figure 2In part B, the waveform corresponding to the position of the alignment signal waveform in part A fluctuates. For Figure 2 After collecting and analyzing the defects in the middle, it is found that the defect is silicon particles. After further analysis and verification, it is obtained that the silicon particles are generated during the process of etching the lithography alignment marks and precipitated from the surface of the lithography alignment marks after the lithography alignment marks are made. That is, the reason for affecting the alignment accuracy is that due to the material characteristics of the aluminum-silicon-copper metal layer, after etching in the aluminum-silicon-copper metal layer to form alignment marks, there will be silicon residue and / or silicon precipitation (at the gaps between the alignment marks or above the alignment marks) in the alignment marks. Whether it is silicon residue or silicon precipitation in the lithography alignment marks will affect the waveform of the alignment signal based on the lithography alignment marks and cause signal interference, thereby affecting the alignment accuracy.
[0057] Based on the above findings, three methods have been tried to improve the alignment accuracy: 1. Change the film material, change the aluminum-silicon-copper thin film to aluminum-copper or pure aluminum. Although this method can eliminate the influence of the above defects in the alignment marks, in addition to possible leakage and other effects on the electrical characteristics of the product, the aluminum grain boundaries will also interfere with the alignment signal; 2. Improve the growth parameters of the aluminum-silicon-copper metal thin film (that is, change the manufacturing process of the metal layer) to improve the influence of the above defects on the alignment stability. However, this method not only cannot completely improve the interference caused by the above defects to the alignment signal, but also affects the mass production stability; 3. Use a more advanced lithography machine to upgrade the optical path and light intensity to reduce the interference of silicon precipitation, etc. on the alignment signal. However, this method requires additional equipment purchase costs, resulting in higher product costs, and also cannot completely eliminate the influence of the above defects on the alignment. Therefore, after further analyzing the specific reasons for the interference of the above defects on the alignment signal, it is considered that usually the lithography alignment marks include periodically spaced line regions and gap regions located between the line regions. Since the height of the line regions is higher than that of the gap regions, after being irradiated by the alignment light, the line regions appear as bright regions and the gap regions appear as dark regions. The peaks of the alignment signal output correspond to the positions of the bright regions and the valleys correspond to the dark regions. At this time, the difference between the dark regions and the bright regions is not significant. However, the presence of silicon precipitation or silicon residue will make the corresponding regions appear darker after being irradiated by the alignment light, thus being reflected as anomalies in the alignment signal waveform. Based on the above analysis, therefore, after multiple verifications and summaries, a technical solution to deal with the influence of silicon precipitation and silicon residue on the alignment signal accuracy is provided. The general solution idea is to process the regions where silicon residue and silicon precipitation occur and the regions where the above defects may occur to cover these regions, so that the entire region appears as a darker region, which can avoid the interference of the alignment signal caused by the locally darker region. The specific solution is as follows.
[0058] The present invention provides an optimization method for lithography alignment marks. Please refer to Figure 3, which is shown as the step flow chart of the optimization method, includes the following steps:
[0059] S1: Provide an original alignment mark;
[0060] S2: Obtain the original alignment signal of the original alignment mark and compare the original alignment signal with the pattern of the original alignment mark to obtain a comparison result, where the comparison result includes the noise source affecting the accuracy of the original alignment signal and the position of the noise source;
[0061] S3: Optimize the original alignment mark according to a first optimization rule to obtain a first alignment mark;
[0062] S4: Verify the alignment accuracy of the first alignment mark. If the alignment accuracy of the first alignment mark meets the alignment requirements, use the first alignment mark as the optimized alignment mark; if the alignment accuracy of the first alignment mark does not meet the alignment requirements, optimize the first alignment mark according to a second optimization rule until an optimized alignment mark is obtained.
[0063] First, please refer to Figure 2 and Figure 4 , and execute steps S1 and S2. As Figure 4 shown, provide an original alignment mark 10; as Figure 2 shown, obtain the original alignment signal of the original alignment mark 10 and compare the original alignment signal with the pattern of the original alignment mark 10 to obtain a comparison result, where the comparison result includes the noise source affecting the accuracy of the original alignment signal and the position of the noise source. The original alignment mark is the alignment mark to be improved by the technical solution of the present invention. Ideally, the original comparison signal obtained based on the original alignment mark is a standard ideal waveform diagram (see Figure 1 ). In this waveform diagram, the peaks and valleys change periodically. However, as Figure 2 shown, due to defects in the local area of the original alignment mark, the waveform diagram of the original alignment signal is locally shifted or fluctuated compared with the ideal waveform diagram. Therefore, after obtaining the original alignment signal, comparing the original alignment signal with the pattern of the original alignment mark can know the noise source and the position of the noise source that cause the abnormal waveform of the original alignment signal, and then optimize the original alignment mark based on the noise source and the position of the noise source.
[0064] As an example, providing an original alignment mark 10 and obtaining the original alignment signal of the original alignment mark 10 and comparing the original alignment signal with the pattern of the original alignment mark 10 include the following steps:
[0065] Provide a first wafer (not shown in the drawings);
[0066] Form the original alignment marks in a preset area of the first wafer, and place the wafer with the original alignment marks formed therein into a lithography machine to collect original alignment signals;
[0067] Compare the original alignment signals with the patterns of the original alignment marks to analyze the sources of noise and record the positions of the sources of noise.
[0068] As an example, the preset area includes at least one of the scribing lanes of the first wafer and the edge area of the first wafer (the "edge area" refers to the area on the first wafer where no devices are fabricated), and the above positions of the original alignment marks are set to avoid wasting the effective area of the first wafer due to the fabrication of alignment marks.
[0069] As an example, the first wafer includes an aluminum-silicon-copper metal layer, and the original alignment marks 10 are formed in the aluminum-silicon-copper metal layer.
[0070] Further, the sources of noise include at least one of silicon precipitation in the original alignment marks 10 and silicon residue in the gaps of the original alignment marks 10. Of course, when the formation position of the original alignment marks is not the aluminum-silicon-copper metal layer, the sources of noise may be other situations.
[0071] As an example, in order to verify the sources of noise, a first picture of the original alignment marks 10 is obtained after the original alignment marks 10 are formed and a second picture of the original alignment marks 10 is obtained after the original alignment signals are collected. By comparing the first picture and the second picture, it can be known whether the specific source of noise is silicon residue or silicon precipitation (silicon residue is generated after the original alignment marks are etched, and both the first picture and the second picture exist; silicon precipitation is formed after the original alignment marks are etched and before the original alignment signals are collected, and only the second picture exists while the first picture does not exist).
[0072] Next, perform step S3, please refer to Figure 5 , optimize the original alignment marks 10 according to a first optimization rule based on the comparison result to obtain first alignment marks 20.
[0073] As an example, the first optimization rule is to insert pre-compensation marks 21 at the positions of the sources of noise for reverse pre-compensation.
[0074] Further, the method of inserting pre-compensation marks 21 at the positions of the sources of noise includes roughening etching at the positions of the sources of noise to form pre-compensation marks 21.
[0075] As an example, the pattern of the original alignment mark 10 includes a plurality of first strip marks 11 arranged at intervals in the horizontal direction, and the pre-compensation mark 21 includes a plurality of second strip marks 211 arranged at intervals in the horizontal direction. The purpose of performing the roughening etching is that the reflected light after the alignment light irradiates this area is weaker than the reflected light when no roughening etching is performed, so that the difference in the reflection intensity between the second strip marks 211 in the pre-compensation mark 21 and the first strip marks 11 in the original alignment mark 10 can be widened.
[0076] As an example, the number range of the second strip members 211 in the compensation mark 21 is 3 to 5, and the width of the second strip members 211 and the distance between two adjacent second strip members 211 can be adjusted based on actual needs on the premise of meeting the reverse pre-compensation.
[0077] As an example, the first optimization rule further includes increasing the width of at least a part of the first strip mark 11 adjacent to the position of the noise source. The specific width increase degree and width increase position only need to achieve the effect that the first strip mark 11 can reach a strong signal (correspondingly, the second strip mark 211 corresponds to a weak signal).
[0078] Next, step S4 is executed to verify the alignment accuracy of the first alignment mark 20. If the alignment accuracy of the first alignment mark 20 meets the alignment requirements, the first alignment mark 20 is used as the optimized alignment mark; if the alignment accuracy of the first alignment mark 20 does not meet the alignment requirements, the first alignment mark 20 is optimized according to the second optimization rule until an optimized alignment mark is obtained.
[0079] As an example, verifying the alignment accuracy of the first alignment mark 20 includes the following steps:
[0080] A second wafer (not shown in the drawings) is provided, and the first alignment mark 20 is formed in a preset area of the second wafer;
[0081] The second wafer formed with the first alignment mark 20 is placed in a lithography machine for multiple lithography alignments, and the position coordinates of the first alignment mark 20 are recorded each time lithography alignment is performed;
[0082] Calculate the position coordinate deviation range of the first alignment mark 20 during multiple lithography alignments. If the position coordinate deviation range is within the preset range, the alignment accuracy of the first alignment mark 20 meets the alignment requirements; if the position coordinate deviation range exceeds the preset range, the alignment accuracy of the first alignment mark 20 does not meet the alignment requirements.
[0083] It should be noted that in this embodiment, the method for verifying whether the first alignment mark can be used as an optimized alignment mark is quantitative measurement, that is, using the optimized first alignment mark as the actual alignment mark to perform a conventional lithography alignment process to collect whether the repeatability of its specific coordinate positions meets the alignment requirements. In fact, it is also possible to collect the alignment signal of the optimized first alignment mark and compare the waveform diagram of the alignment signal with the waveform diagram of the standard alignment signal (such as Figure 1 shown). If the waveform diagrams of the two are basically consistent or the difference is extremely small (for example, the overlap degree of the waveform diagrams of the two is more than 99%), it can also be regarded that the first alignment mark does not need to be further optimized and can be used as an optimized alignment mark. Of course, in order to verify the accuracy and ensure the product yield, it is preferably to use the above verification method in this embodiment to verify the alignment accuracy of the first alignment mark.
[0084] As an example, the number of times of performing lithography alignment on the second wafer formed with the first alignment mark 20 in a lithography machine includes 10 to 100 times. In actual application, the verification times can be selected based on actual needs, preferably 20 to 30 times, which can effectively verify the alignment accuracy without affecting the verification efficiency due to a large number of verification times.
[0085] As an example, the preset range is less than 20 nm. It should be noted that usually, in order to achieve the alignment of the mask pattern and the original pattern on the wafer, multiple alignment marks are usually set, and a part of them is used to achieve the alignment in the first horizontal direction, and another part is used to achieve the alignment in the second horizontal direction (the first horizontal direction is perpendicular to the second horizontal direction), so as to enable precise positioning. The technical solution of the present invention only takes one of the alignment marks as an example for optimization. If the lithography alignment mark for achieving the alignment in the first horizontal direction is optimized, the above preset range refers to the allowable range of position deviation in the first horizontal direction. If the lithography alignment mark for achieving the alignment in the second horizontal direction is optimized, the above preset range refers to the allowable range of position deviation in the second horizontal direction. Of course, the above preset range is a numerical range summarized after multiple experimental verifications, that is, when verifying the alignment accuracy of the first alignment mark, the position coordinates are collected while performing the lithography alignment process and the actual alignment situation is statistically analyzed, and then the numerical range is summarized. When the deviation between the position coordinates of a certain time and any other position coordinates is more than 20 nm, the actual lithography alignment picture also shows a large alignment deviation and does not meet the requirements of the overlay accuracy in the process; correspondingly, when the deviation between the position coordinates of a certain time and all other position coordinates is within 20 nm, even if there is a slight deviation in the actual lithography alignment picture, it is within the allowable range of the overlay accuracy deviation in the process.
[0086] As an example, the second optimization rule includes adjusting at least one of the number of the second strip-shaped marks 211 in the pre-compensation mark 21, the width of the second strip-shaped mark 211, and the spacing between two adjacent second strip-shaped marks 211. In fact, after multiple verifications and practices, the alignment accuracy and stability of the first alignment mark 20 obtained by optimizing the original alignment mark 10 according to the first optimization rule meet the alignment requirements in most cases. However, there are still individual cases where the first alignment mark 20 cannot achieve good alignment accuracy. At this time, appropriate adjustment of the specific structure of the pre-compensation mark 21 can achieve good alignment accuracy.
[0087] Specifically, please refer to Figure 6 and Figure 7 , where Figure 6 shows the picture of the original alignment mark without optimization and its corresponding alignment signal waveform diagram, Figure 7 shows the picture of the first alignment mark optimized by using the optimization method of the present invention and its corresponding alignment signal waveform diagram. By comparing Figure 6 and Figure 7 , it can be seen that when not optimized, the alignment signal of the original alignment mark is more or less affected by defects such as silicon precipitation or silicon residue, resulting in abnormal points in the waveform, thus affecting the alignment accuracy. After optimization, whether there is still silicon precipitation or silicon residue in the first alignment mark, it will not affect the alignment signal (since the pre-compensation mark is set in the area where silicon precipitation or silicon residue is located, the waveform of the alignment signal corresponding to this area will not be abnormal due to sudden brightness and darkness), and the waveform diagram of the alignment signal is close to the standard alignment signal waveform. Please refer to Figures 8 to 11 , where Figure 8 shows the physical optical microscopic picture after alignment based on the original alignment mark, Figure 9 shows Figure 8 the simplified effect picture of area C in Figure 10 shows the physical optical microscopic picture after alignment based on the first alignment mark, Figure 11 shows Figure 10 the simplified effect picture of area D in Figure 8 and Figure 9 show that after lithographic alignment based on the original alignment mark, the alignment accuracy between the first pattern 30a and the second pattern 40a is poor. It can be visibly seen from the physical optical microscopic picture that the second pattern 40a is significantly shifted left and downward relative to the first pattern 30a. And as Figure 10 and Figure 11As shown, after lithographic alignment based on the first alignment mark, the center of the first pattern 30b coincides basically with the center of the second pattern 40b, or only has a slight offset. That is, after practical verification, the problem that the silicon defects in the aluminum-silicon-copper metal layer affect the alignment stability can be effectively solved by optimizing the original alignment mark according to the optimization scheme of the present invention, thereby effectively improving the alignment accuracy. According to the data statistically obtained after a series of verifications, the alignment accuracy of the optimized alignment mark is improved by more than 50%, and the overlay error of the wafer can be effectively improved.
[0088] It should be noted that although the optimization of the lithographic alignment mark in the present invention is carried out around the lithographic alignment mark in the aluminum-silicon-copper metal layer, in fact, the technical solution of the present invention is also applied to the optimization of the lithographic alignment mark of other structural layers that may have similar problems.
[0089] In summary, the optimization method of the lithographic alignment mark of the present invention can effectively solve the problem that the silicon defects in the aluminum-silicon-copper metal layer affect the alignment stability, thereby effectively improving the alignment accuracy, and greatly improving the product yield and performance stability. Therefore, the present invention effectively overcomes various shortcomings in the prior art and has high industrial utilization value.
[0090] The above embodiments are only illustrative of the principles and effects of the present invention, and are not used to limit the present invention. Any person familiar with this technology can modify or change the above embodiments without departing from the spirit and scope of the present invention. Therefore, all equivalent modifications or changes completed by those with ordinary knowledge in the technical field without departing from the spirit and technical idea disclosed by the present invention should still be covered by the claims of the present invention.
Claims
1. An optimization method for lithography alignment marks, characterized in that, it includes the following steps: Provide an original alignment mark; Obtain the original alignment signal of the original alignment mark and compare the original alignment signal with the pattern of the original alignment mark to obtain a comparison result, where the comparison result includes the noise source affecting the accuracy of the original alignment signal and the position of the noise source; Optimize the original alignment mark based on the comparison result according to a first optimization rule to obtain a first alignment mark; Verify the alignment accuracy of the first alignment mark. If the alignment accuracy of the first alignment mark meets the alignment requirements, use the first alignment mark as the optimized alignment mark; If the alignment accuracy of the first alignment mark does not meet the alignment requirements, optimize the first alignment mark according to a second optimization rule until an optimized alignment mark is obtained.
2. The optimization method for lithography alignment marks according to claim 1, characterized in that, providing an original alignment mark and obtaining the original alignment signal of the original alignment mark and comparing the original alignment signal with the pattern of the original alignment mark includes the following steps: Provide a first wafer; Form the original alignment mark in a preset area of the first wafer, and place the wafer with the original alignment mark formed thereon in a lithography machine to collect the original alignment signal; Compare the original alignment signal with the pattern of the original alignment mark to analyze the noise source and record the position of the noise source.
3. The optimization method for lithography alignment marks according to claim 2, characterized in that: The first wafer includes an aluminum-silicon-copper metal layer, and the original alignment mark is formed in the aluminum-silicon-copper metal layer.
4. The optimization method for lithography alignment marks according to claim 3, characterized in that: The noise source includes at least one of silicon precipitation in the original alignment mark and silicon residue in the gap of the original alignment mark.
5. The optimization method for lithography alignment marks according to claim 1, characterized in that: The first optimization rule is to insert a pre-compensation mark at the position of the noise source for reverse pre-compensation.
6. The optimization method for lithography alignment marks according to claim 5, characterized in that: The method of inserting a pre-compensation mark at the position of the noise source includes roughening etching at the position of the noise source to form a pre-compensation mark.
7. The optimization method for lithography alignment marks according to claim 5, characterized in that: The pattern of the original alignment mark includes a plurality of first strip marks arranged at intervals in the horizontal direction, and the pre-compensation mark includes a plurality of second strip marks arranged at intervals in the horizontal direction.
8. The optimization method for lithography alignment marks according to claim 7, characterized in that: The first optimization rule further includes increasing the width of at least a part of the first strip mark adjacent to the position of the noise source.
9. The optimization method for lithography alignment marks according to claim 7, characterized in that: The second optimization rule includes adjusting at least one of the number of the second strip marks in the pre-compensation marks, the width of the second strip marks, and the spacing between two adjacent second strip marks.
10. The method for optimizing a lithography alignment mark according to claim 1, wherein: Verifying the alignment accuracy of the first alignment mark includes the following steps: Providing a second wafer, and forming a first alignment mark in a preset area of the second wafer; Placing the second wafer with the first alignment mark formed thereon in a lithography machine for multiple lithography alignments and recording the position coordinates of the first alignment mark each time lithography alignment is performed; Calculating the range of position coordinate deviations of the first alignment mark during multiple lithography alignments. If the range of position coordinate deviations is within a preset range, the alignment accuracy of the first alignment mark meets the alignment requirements. If the range of position coordinate deviations exceeds the preset range, the alignment accuracy of the first alignment mark does not meet the alignment requirements.
11. The method for optimizing a lithography alignment mark according to claim 9, wherein: The preset range is less than 20 nm.