Overlay compensation method and device, storage medium, and electronic device

Abstract

A method and device for overlay compensation, a storage medium, and an electronic device. The method includes: obtaining initial feedforward compensation data and feedback compensation data corresponding to a first initial layer; obtaining alignment information of the first initial layer, and determining an alignment method based on the alignment information of the first initial layer; the alignment information of the first initial layer includes: identification information of the alignment layer; when the alignment method is a direct alignment method, the alignment compensation data of the first initial layer is obtained based on the alignment information of the first initial layer, and the difference between the initial feedforward compensation data and the alignment compensation data is used as the feedforward compensation data of the first initial layer; using the feedforward compensation data and feedback compensation data of the first initial layer, overlay compensation is performed on the first initial layer, and a corresponding photolithography pattern is formed on the first initial layer. The above scheme can improve overlay accuracy.

Description

Technical Field

[0001] The present invention relates to the field of semiconductor technology, and in particular to an overlay compensation method and device, a storage medium, and an electronic device. Background Art

[0002] The fabrication process of semiconductor devices typically involves forming multiple film layers stacked one above the other, with various components formed within each layer. During this process, the current layer must be aligned with the previous layer so that a component formed in the current layer corresponds to or connects to a component in the previous layer. Therefore, overlay between the layers is a significant factor affecting device performance.

[0003] In practical applications, systematic and accidental errors can cause deviations in the alignment of the current layer pattern with the previous layer pattern during the photolithography process, resulting in overlay deviation. Since integrated circuit chips are manufactured by adding multiple circuit layers, if the current layer and the previous layer are misaligned, the chip will not function properly. Therefore, during the formation of the current layer, it is extremely important to minimize overlay deviation and ensure that it is within the tolerance range.

[0004] Currently, a Correction Per Exposure (CPE) model is usually used to correct the overlay deviation, but the overlay accuracy is still low. Summary of the Invention

[0005] The problem to be solved by the present invention is to improve the overlay accuracy.

[0006] To solve the above problem, an embodiment of the present invention provides an overlay compensation method, which includes:

[0007] Providing a wafer to be compensated; the wafer to be compensated has an overlay reference layer and a first initial layer located on the overlay reference layer;

[0008] Acquire initial feedforward compensation data and feedback compensation data corresponding to the first initial layer; the initial feedforward compensation data is actual compensation data of the overlay reference layer, and the feedback compensation data is feedback compensation data obtained when performing an overlay test on the first initial layer;

[0009] Acquire alignment information of the first initial layer, and determine an alignment method based on the alignment information of the first initial layer; the alignment information of the first initial layer includes: identification information of the alignment layer;

[0010] When the alignment mode is a direct alignment mode, obtaining alignment compensation data of the first initial layer based on the alignment information of the first initial layer, and using a difference between the initial feedforward compensation data and the alignment compensation data as the feedforward compensation data of the first initial layer;

[0011] The feedforward compensation data and the feedback compensation data of the first initial layer are used to perform overlay compensation on the first initial layer, and a corresponding photolithography pattern is formed on the first initial layer.

[0012] Optionally, the alignment information of the first initial layer further includes: alignment model information and alignment mark position information on the alignment layer.

[0013] Optionally, obtaining alignment compensation data of the first initial layer based on the alignment information of the first initial layer includes:

[0014] Taking the registration test point position closest to each first alignment mark position as the corresponding alignment detection position, and obtaining the alignment detection position information of each alignment detection position in the alignment layer;

[0015] Based on the alignment detection position information, alignment compensation data of the first initial layer is calculated using an alignment model for performing photolithography alignment on the first initial layer.

[0016] Optionally, the feedback compensation data is obtained by weighting the feedback compensation data obtained when performing an overlay test on the first initial layer of multiple batches of wafers according to the number of samples and the sampling time and then performing weighted averaging.

[0017] Optionally, the method further includes:

[0018] When the alignment method is an indirect alignment method, the initial feedforward compensation data is used as the feedforward compensation data of the first initial layer.

[0019] Optionally, performing overlay compensation on the first initial layer by using the feedforward compensation data and the feedback compensation data of the first initial layer includes:

[0020] The sum of the feedforward compensation data and the feedback compensation data of the first initial layer is used as the final overlay compensation data of the first initial layer to perform overlay compensation on the first initial layer.

[0021] An embodiment of the present invention further provides an overlay compensation device, comprising:

[0022] A providing unit provides a wafer to be compensated; the wafer to be compensated has an overlay reference layer and a first initial layer located on the overlay reference layer;

[0023] A first acquisition unit is adapted to acquire initial feedforward compensation data and feedback compensation data corresponding to the first initial layer; the initial feedforward compensation data is actual compensation data of the overlay reference layer, and the feedback compensation data is feedback compensation data obtained when performing an overlay test on the first initial layer;

[0024] A second acquiring unit is configured to acquire alignment information of the first initial layer and determine an alignment method based on the alignment information of the first initial layer; the alignment information of the first initial layer includes identification information of the alignment layer;

[0025] a feedforward compensation data determining unit, adapted to, when the alignment mode is a direct alignment mode, obtain alignment compensation data of the first initial layer based on the alignment information of the first initial layer, and use a difference between the initial feedforward compensation data and the alignment compensation data as the feedforward compensation data of the first initial layer;

[0026] The overlay unit is adapted to perform overlay compensation on the first initial layer by utilizing the feedforward compensation data and the feedback compensation data of the first initial layer, so that the first initial layer forms a corresponding lithography pattern.

[0027] Optionally, the alignment information of the first initial layer acquired by the second acquisition unit further includes: alignment model information and alignment mark position information on the alignment layer.

[0028] Optionally, the second acquiring unit includes:

[0029] an alignment position information determining subunit, adapted to use the position of the registration test point closest to each first alignment mark position as the corresponding alignment detection position, and obtain alignment detection position information for each alignment detection position in the alignment layer;

[0030] The alignment compensation data calculation subunit is adapted to calculate the alignment compensation data of the first initial layer based on the alignment detection position information and using an alignment model for performing photolithography alignment on the first initial layer.

[0031] Optionally, the feedback compensation data acquired by the first acquisition unit is obtained by weighting the feedback compensation data when performing registration tests on the first initial layers of multiple batches of wafers according to the number of samples and the sampling time, and then performing weighted averaging.

[0032] Optionally, the feedforward compensation data determining unit is further adapted to use the initial feedforward compensation data as the feedforward compensation data of the first initial layer when the alignment method is an indirect alignment method.

[0033] Optionally, the overlay unit is adapted to use the sum of the feedforward compensation data and the feedback compensation data of the first initial layer as the final overlay compensation data of the first initial layer to perform overlay compensation on the first initial layer.

[0034] An embodiment of the present invention further provides a computer-readable storage medium having a computer program stored thereon, wherein the computer program is executed by a processor to implement the steps of any of the above methods.

[0035] An embodiment of the present invention further provides an electronic device, comprising a memory and a processor, wherein the memory stores a computer program that can be run on the processor, and the processor executes the steps of any of the above methods when running the computer program.

[0036] Compared with the prior art, the technical solution of the embodiment of the present invention has the following advantages:

[0037] Using the solution of the present invention, when the alignment method is direct alignment, that is, when the film layer used for lithography alignment is an overlay reference layer, the difference between the initial feedforward compensation data and the alignment compensation data is used as the feedforward compensation data for the first initial layer. Overlay compensation is then performed on the first initial layer using the feedforward compensation data and feedback compensation data of the first initial layer. Thus, in the case of direct alignment, by stripping the alignment compensation data from the initial feedforward compensation data, repeated high-order exposure compensation of the exposure parameters of the first initial layer can be avoided, thereby improving the accuracy of overlay compensation. BRIEF DESCRIPTION OF THE DRAWINGS

[0038] Figure 1 It is a schematic diagram of the alignment method;

[0039] Figure 2 is a flow chart of an overlay compensation method according to an embodiment of the present invention;

[0040] Figure 3 is a schematic diagram of the cross-sectional structure of a wafer to be compensated in an embodiment of the present invention;

[0041] Figure 4 is a schematic diagram of the distribution of registration test points on a first initial layer in an embodiment of the present invention;

[0042] Figure 5 1 is a schematic diagram of the distribution of first alignment detection positions on an alignment layer in an embodiment of the present invention;

[0043] Figure 6 1 is a schematic diagram showing the distribution of first alignment detection positions and registration test points in an exposure unit in an embodiment of the present invention;

[0044] Figure 7 is a schematic diagram of an overlay compensation process in an embodiment of the present invention;

[0045] Figure 8 is the difference between the first overlay error and the second overlay error in the X direction for different numbers of wafers;

[0046] Figure 9 is the difference between the first overlay error and the second overlay error in the Y direction for different numbers of wafers;

[0047] Figure 10 1 is a schematic structural diagram of an overlay compensation device according to an embodiment of the present invention;

[0048] Figure 11 Schematic diagram of the structure of an exposure system in an embodiment of the present invention. DETAILED DESCRIPTION

[0049] In the overlay compensation of semiconductor lithography engineering, the first step is to find the specific position of the alignment layer, align the current layer and the alignment layer, and then use the overlay reference layer as a reference to finely compensate the exposure parameters of the current layer. Finally, use the compensated exposure parameters to expose and form a graphic current layer.

[0050] When aligning the current layer and the alignment layer, two alignment methods can be used: direct alignment and indirect alignment, based on different alignment models. Direct alignment uses the overlay reference layer as the alignment layer. Indirect alignment uses a layer other than the overlay reference layer as the alignment layer. Whether direct or indirect alignment is used, the current layer will ultimately be able to complete photolithography alignment.

[0051] For example, refer to Figure 1 , the alignment layer in alignment process AL is layer L1, the overlay reference layer in the X direction in overlay process OVL is layer L3, and the overlay reference layer in the Y direction is layer L1. Since the front layer in alignment process AL and the overlay reference layer in the X direction in overlay process OVL are different film layers, alignment process AL is indirect alignment with respect to the overlay process in the X direction. Since the alignment layer in alignment process AL and the overlay reference layer in the Y direction in overlay process OVL are the same film layer, alignment process AL is direct alignment with respect to the overlay process in the Y direction.

[0052] During the alignment process, the alignment method can be selected based on the alignment mask placement strategy or process limitations. Traditional alignment models are low-order linear. With the continuous development of semiconductor processes, higher-order alignment models (i.e., alignment models of order 2 or higher) are widely used.

[0053] Using a high-level alignment model for alignment results in minimal alignment residue during exposure, facilitating subsequent overlay compensation. Furthermore, using a high-level alignment model for alignment can mitigate subtle wafer deformations caused by thermal effects or stress during polishing processes, preventing these deformations from affecting overlay measurements by impacting alignment.

[0054] In order to make the overlay as good as possible, after alignment using the high-order alignment model, high-order CPE will also be used to complete the overlay fine compensation.

[0055] At present, when using CPE to correct the overlay error of the current layer, the feedforward compensation data and the feedback compensation data of the current layer are usually calculated separately first, and the sum of the feedforward compensation data of the current layer and the feedback compensation data of the current layer is used as the final compensation data of the current layer. Based on the final compensation value of the current layer, the exposure parameters of the current layer are adjusted, so that exposure can be performed according to the adjusted exposure parameters.

[0056] The so-called feedforward compensation data refers to the actual compensation data of the overlay reference layer. Each wafer batch (lot) has its own independent feedforward compensation data, which is independent of the feedforward compensation data of other wafer batches.

[0057] Feedback compensation data refers to the feedback compensation data obtained during the layer overlay test. After adjusting exposure parameters using the layer's feedforward compensation data, if the overlay test results reveal that some high-order CPEs still require compensation to keep overlay errors within acceptable specifications, the high-order feedback compensation data from the overlay test (excluding any CPE feedforward compensation data) serves as the feedback compensation data. This feedback compensation data is the average of multiple lots and is relatively constant.

[0058] Research has found that when the alignment process is direct, high-order exposure compensation is directly applied to the exposure parameters of the current layer when the high-order alignment model is used to align the current layer. In this case, the CPE feedforward compensation data of the current layer includes compensation data supplemented by the high-order alignment model. This is equivalent to repeatedly applying high-order exposure compensation to the exposure parameters of the current layer during the same overlay compensation process, which is obviously unreasonable. The repeated high-order exposure compensation of the current layer's exposure parameters using high-order alignment and high-order CPE is called crosstalk.

[0059] Possible improvements to the crosstalk problem between high-order alignment and high-order CPE include the following two methods:

[0060] The first method involves updating the alignment method, switching from direct alignment to indirect alignment. This eliminates crosstalk between high-order alignment and high-order CPEs. However, switching from direct alignment to indirect alignment weakens the alignment signal to a certain extent (alignment marks are further apart). This can lead to more alignment residue, hindering subsequent CPE compensation. Furthermore, this weakened alignment signal increases uncertainty in the alignment pattern, potentially leading to miscompensation.

[0061] The second method is to allow crosstalk to exist and partially compensate for it in the feedback compensation data. However, since alignment compensation data varies between wafer batches and even between wafers in the same batch, and feedforward compensation data is also related to the previous layer, the feedforward compensation data of different wafer batches is likely to be different. Both feedforward compensation data and alignment compensation are dynamic. Therefore, the crosstalk between feedforward compensation data and alignment compensation is also dynamic. However, feedback compensation data is the average of feedback compensation data when multiple batches of wafers form the current film layer, and is usually relatively unchanged. If the dynamic crosstalk is partially compensated in the feedback compensation data, the dynamic difference becomes relatively unchanged.

[0062] For example, if the crosstalk values ​​for three wafer batches during layer formation are 3, 5, and 7, respectively, then the feedback compensation data will be detuned to 5, 5, and 5. This means that not all wafer batches can be overlaid to optimal compensation, and two of them will degrade, which is clearly unacceptable. In actual mass production, the differences in feedback compensation data between wafer batches can be significant.

[0063] To address this problem, the present invention provides an overlay compensation method. When the alignment method is a direct alignment method, the difference between the initial feedforward compensation data and the alignment compensation data is used as the feedforward compensation data of the first initial layer. The feedforward compensation data and feedback compensation data of the first initial layer are then used to compensate the exposure parameters of the first initial layer, so that the first initial layer is exposed using the compensated exposure parameters. In this way, in the case of direct alignment, by stripping the alignment compensation data from the initial feedforward compensation data, it is possible to avoid repeated high-order exposure compensation for the exposure parameters of the current layer, thereby improving the accuracy of overlay compensation. Moreover, with this solution, there is no need to change the alignment method, thereby avoiding erroneous compensation caused by changes in the alignment method and improving compensation accuracy. In addition, overlay compensation can be performed separately for different batches of wafers, so that all batches of wafers are overlaid to the optimal level.

[0064] In order to make the above-mentioned objects, features and advantages of the present invention more obvious and easy to understand, specific embodiments of the present invention are described in detail below with reference to the accompanying drawings.

[0065] Reference Figure 2 An embodiment of the present invention provides an overlay compensation method, which may include the following steps:

[0066] Step 21 : providing a wafer to be compensated, wherein the wafer to be compensated has an overlay reference layer and a first initial layer located on the overlay reference layer.

[0067] Reference Figure 3 , providing a wafer 10 to be compensated, wherein the wafer 10 to be compensated includes a substrate 101 , an overlay reference layer 102 arranged on the substrate 101 , and a first initial layer 103 located above the overlay reference layer 102 .

[0068] In a specific implementation, the overlay reference layer 102 is the front layer in the overlay process, and the first initial layer 103 is the current layer in the overlay process.

[0069] In a specific implementation, the wafer 10 to be compensated may further include an alignment layer. During the photolithographic alignment process of the first initial layer, a plurality of alignment mark position information may be formed on the alignment layer. Based on the alignment mark position information, alignment detection position information of the alignment layer may be obtained, where the alignment detection position information represents the detection position information of each alignment mark. Subsequently, the preset alignment model may be used to align the first initial layer based on the alignment detection position information.

[0070] In a specific implementation, the preset alignment model may be a radial basis function alignment model or other alignment models, which is not limited here.

[0071] In a specific implementation, the alignment layer can be the same film layer as the overlay reference layer 102, or a different film layer. When the alignment layer can be the same film layer as the overlay reference layer 102, the alignment method is a direct alignment method. When the alignment layer and the overlay reference layer 102 are different film layers, the alignment method is an indirect alignment method. When the alignment layer and the overlay reference layer 102 are different film layers, the alignment layer is formed below the overlay reference layer 102.

[0072] Step 22, obtaining initial feedforward compensation data and feedback compensation data corresponding to the first initial layer; the initial feedforward compensation data is the actual compensation data of the overlay reference layer, and the feedback compensation data is the feedback compensation data obtained when the first initial layer is subjected to an overlay test.

[0073] In practice, obtaining initial feedforward compensation data is very easy because the feedforward registration compensation data can directly use the overlay compensation data of the previous film layer. The overlay compensation data of the previous film layer also includes the feedforward compensation data and feedback compensation data of the previous film layer. When the current film layer is the first layer, the overlay compensation data of the previous film layer is 0.

[0074] In practice, feedback compensation data for the first initial layer relies on the calculation of measured data, requiring a feedback mechanism and a complex process. Specifically, after the photolithographic pattern is formed on the first initial layer, an overlay test is performed to measure the overlay of each film layer on the wafer after overlay. The required compensation data is calculated based on the actual overlay conditions and used as the feedback compensation data for the first initial layer. During the overlay test, the overlay conditions on the wafer must be measured so that feedback compensation data can be calculated based on the wafer's overlay conditions.

[0075] In one embodiment of the present invention, a full sampling measurement method can be used to perform an alignment test on the film layer. Full sampling measurement means that all sampling points in all exposure units on the wafer surface are measured. This method has high measurement accuracy and wide coverage, but low efficiency. At this time, the feedback alignment compensation data of the current film layer during the alignment test is the weighted average of the feedback alignment compensation data of the current film layer of wafers from multiple other batches during the alignment test, weighted according to the number of samples and the sampling time. The more samples there are, the higher the weight; the closer the sampling time is to the current exposure, the higher the weight.

[0076] In another embodiment of the present invention, a dynamic SSO measurement (also called dynamic partial sampling measurement) method can be used to perform registration testing on the film layer. Dynamic SSO measurement refers to dividing all sampling points in all exposure units on the wafer into several non-overlapping groups, measuring only one group on each wafer, and finally combining the measurement results of several groups as the measurement results of a batch of wafers. At this time, the feedback compensation data is the sum of the wafer-level compensation data and the exposure unit-level compensation data. Among them, the wafer-level compensation data is the weighted average of the feedback compensation data of wafers from multiple other batches after weighting according to the number of samples and the sampling time. The exposure unit-level compensation data is the feedback compensation data after the measurement data of wafers from multiple other batches are superimposed. The more samples there are, the higher the weight they occupy; the closer the sampling time is to the current exposure, the higher the weight they occupy.

[0077] In a specific implementation, no matter which method is used to perform the registration test on the patterned first initial layer, the CPE compensation data of the patterned first initial layer at each registration test point can be obtained. Figure 4 Schematic diagram of the distribution of registration test points on the first initial layer 103, wherein each point is a registration test point. Figure 4As can be seen, the first initial layer 103 has numerous and densely packed alignment test points, enabling support for higher-order data compensation. The CPE compensation data for each alignment test point includes the position information for that alignment test point and the CPE compensation data at each alignment test point location. Each CPE compensation data includes an initial feedforward compensation value and a feedback compensation value.

[0078] Step 23: Acquire alignment information of the first initial layer, and determine an alignment method based on the alignment information of the first initial layer; the alignment information of the first initial layer includes: identification information of the alignment layer.

[0079] In a specific implementation, the alignment information of the first initial layer can be obtained from the alignment report. Based on the alignment information of the first initial layer, the identification information of the alignment layer can be determined, thereby determining the alignment method. Specifically, referring to Figure 3 When the film layer used for photolithography alignment is the overlay reference layer 102, the alignment method is determined to be direct alignment, and step 24 can be performed. When the film layer used for photolithography alignment is not the overlay reference layer 102, the alignment method is determined to be indirect alignment, and step 25 can be performed.

[0080] Step 24: obtaining alignment compensation data of the first initial layer based on the alignment information of the first initial layer, and using the difference between the initial feedforward compensation data and the alignment compensation data as the feedforward compensation data of the first initial layer.

[0081] In practical applications, the actual alignment compensation data for the first initial layer can only be obtained from the exposure information after a single exposure of the first initial layer. The first initial layer must then be reworked to perform overlay compensation and achieve a second exposure. After the first exposure before rework, the alignment compensation value at each alignment detection position on the alignment layer can be obtained. The alignment compensation values ​​at each alignment detection position on the alignment layer constitute the actual alignment compensation data for the first initial layer. The alignment detection positions are the detection positions corresponding to the alignment marks. Each alignment detection position corresponds to an alignment mark.

[0082] To this end, in one embodiment of the present invention, obtaining alignment compensation data of the first initial layer based on the alignment information of the first initial layer may include:

[0083] The position of the registration test point closest to each alignment mark position is used as the corresponding alignment detection position to obtain alignment detection position information for each alignment layer. Based on the alignment detection position information, the alignment compensation data of the first initial layer is calculated using the alignment model for performing lithography alignment on the first initial layer.

[0084] In a specific implementation, the alignment information of the first initial layer also includes: alignment model information and alignment mark position information on the alignment layer. Therefore, the alignment model used when aligning the first initial layer and the positions of the alignment marks on the alignment layer can be obtained from the alignment report.

[0085] Research has found that alignment compensation data derived from the same alignment model using registration test points approximating the actual alignment mark positions closely matches the actual alignment compensation data for the first initial layer, exhibiting a highly linear relationship. Therefore, in embodiments of the present invention, the registration test point closest to each alignment mark position is used as the corresponding alignment detection position. This eliminates the need for initial exposure and rework, simplifies the process, improves compensation efficiency, and reduces costs.

[0086] Taking the radial basis function alignment model to align the first initial layer as an example, we can get Figure 5 A distribution diagram of the alignment detection positions is shown in FIG, where each point represents an alignment detection position.

[0087] Figure 6 This is a schematic diagram showing the distribution of alignment detection positions and registration test points in an exposure unit. Figure 6 , the registration test point closest to the alignment mark can be found among the registration test points and used as the alignment detection position. The registration test point that can be used as the alignment detection position should be within a preset range corresponding to the alignment mark. This preset range can be adjusted based on actual conditions. For example, the preset range can be [2mm, -2mm].

[0088] Taking the alignment mark 61 as an example, the closest alignment test point to the alignment mark 61, that is, the alignment test point 62, can be found within the range of each control mark [2mm, -2mm], and the alignment test point 62 can be used as the alignment detection position corresponding to the alignment mark 61.

[0089] After determining each alignment detection position on the alignment layer, the same alignment model can be used to obtain alignment compensation information corresponding to each alignment detection position.

[0090] In an embodiment of the present invention, when direct alignment is used for the first initial layer, the difference between the initial feedforward compensation data and the alignment compensation data is used as the feedforward compensation data for the first initial layer. Specifically, the alignment compensation data is stripped from the initial feedforward compensation data to obtain the feedforward compensation data for the first initial layer. The feedforward compensation data for the first initial layer includes the final overlay compensation value for each alignment test point on the overlay reference layer. The alignment compensation data includes alignment compensation information for each first alignment detection position on the overlay reference layer.

[0091] When the alignment test point closest to the alignment mark is used as the corresponding alignment detection position, the alignment test points on the overlay reference layer can be divided into two categories: first alignment test points and second alignment test points. The first alignment test points are alignment test points on the overlay reference layer that have both alignment compensation information and a final overlay compensation value, while the second alignment test points are alignment test points on the overlay reference layer that only have the final overlay compensation value. The difference between the initial feedforward compensation data and the alignment compensation data, i.e., the difference between the final overlay compensation value and the alignment compensation information of the first alignment test point is used as the feedforward compensation data for the first alignment test point; the final overlay compensation value of the second alignment test point is used as the feedforward compensation data for the second alignment test point. The feedforward compensation data of the first alignment test point and the feedforward compensation data of the second alignment test point constitute the feedforward compensation data of the first initial layer.

[0092] Reference Figure 7 , CPE_FF_total represents a schematic diagram of the initial feedforward compensation data corresponding to the first initial layer, alignment_data represents a schematic diagram of the alignment compensation data of the first initial layer, and after stripping the alignment compensation data of the first initial layer from the initial feedforward compensation data corresponding to the first initial layer, the data schematic diagram obtained is represented by CPE_FF_Partial.

[0093] Step 25: Use the initial feedforward compensation data as the feedforward compensation data of the first initial layer.

[0094] That is, when the alignment method is indirect alignment, since there is no crosstalk, there is no need to strip the alignment compensation data from the initial feedforward compensation data, and the initial feedforward compensation data can be directly used as the feedforward compensation data of the first initial layer.

[0095] Step 26 : Using the feedforward compensation data and feedback compensation data of the first initial layer, perform overlay compensation on the first initial layer, and form a corresponding photolithography pattern on the first initial layer.

[0096] In a specific implementation, when the alignment method is direct alignment, refer to Figure 7 , the sum of CPE_FF_Partial and the feedback compensation data CPE_FB of the first initial layer can be used as the final overlay compensation data CPE_total.

[0097] When the alignment method is indirect alignment, the sum of CPE_FF_total and the feedback compensation data CPE_FB of the first initial layer may be used as the final overlay compensation data CPE_total.

[0098] The exposure parameters of the first initial layer are compensated using the final overlay compensation data CPE_total, and the compensated exposure parameters are used to control the exposure of the first initial layer to form a pattern on the first initial layer, completing the overlay of the first initial layer and obtaining a patterned first initial layer.

[0099] In practical applications, after overlaying the first initial layer, the first initial layer will deviate from the overlay reference layer in the X and Y directions. The deviation of the first initial layer from the overlay reference layer in the X direction is called the X-direction overlay error. The deviation of the first initial layer from the overlay reference layer in the Y direction is called the Y-direction overlay error.

[0100] The overlay error obtained after overlaying the first initial layer using the overlay compensation method of the embodiment of the present invention is called the first overlay error. The overlay error obtained after overlaying the first initial layer using the existing CPE is called the second overlay error.

[0101] Figure 8 It represents the difference between the first overlay error and the second overlay error in the X direction for different numbers of wafers, where the horizontal axis represents the number of wafers and the vertical axis represents the difference between the first overlay error and the second overlay error in the X direction. Figure 9 It represents the difference between the first overlay error and the second overlay error in the Y direction for different numbers of wafers, where the horizontal axis represents the number of wafers and the vertical axis represents the difference between the first overlay error and the second overlay error in the Y direction.

[0102] from Figure 8 and Figure 9 It can be seen that the difference between the first overlay error and the second overlay error is mostly negative in both the X and Y directions, indicating that the first overlay error is smaller than the second overlay error, that is, the first overlay error obtained by the overlay compensation method in the embodiment of the present invention is smaller, and the lithography effect is better.

[0103] From the above content, it can be seen that the overlay compensation method in the embodiment of the present invention automatically identifies the lithography alignment method and, during direct alignment, strips the alignment compensation data from the initial feedforward compensation data of the current layer. After high-order alignment, the dynamic accuracy of the feedforward compensation data can still be ensured, thereby improving the overlay accuracy.

[0104] In order to enable those skilled in the art to better understand and implement the present invention, the apparatus, computer-readable storage medium, and electronic device corresponding to the above method are described in detail below.

[0105] Reference Figure 10The embodiment of the present invention further provides an overlay compensation device 100, which may include: a providing unit 110, a first acquiring unit 120, a second acquiring unit 130, a feedforward compensation data determining unit 140, and an overlay unit 150.

[0106] The providing unit 110 provides a wafer to be compensated; the wafer to be compensated has an overlay reference layer and a first initial layer located on the overlay reference layer;

[0107] The first acquisition unit 120 is adapted to acquire initial feedforward compensation data and feedback compensation data corresponding to the first initial layer; the initial feedforward compensation data is actual compensation data of the overlay reference layer, and the feedback compensation data is feedback compensation data obtained when performing an overlay test on the first initial layer;

[0108] The second acquiring unit 130 is adapted to acquire the alignment information of the first initial layer and determine an alignment method based on the alignment information of the first initial layer;

[0109] The feedforward compensation data determining unit 140 is adapted to, when the alignment mode is a direct alignment mode, obtain alignment compensation data of the first initial layer based on the alignment information of the first initial layer, and use a difference between the initial feedforward compensation data and the alignment compensation data as the feedforward compensation data of the first initial layer;

[0110] The overlay unit 150 is adapted to perform overlay compensation on the first initial layer using the feedforward compensation data and the feedback compensation data of the first initial layer, and form a corresponding lithography pattern on the first initial layer.

[0111] In one embodiment of the present invention, the alignment information of the first initial layer acquired by the second acquiring unit 130 further includes alignment model information and alignment mark position information on the alignment layer.

[0112] In one embodiment of the present invention, the second acquiring unit 130 may include: an alignment position information determining subunit 131 and an alignment compensation data calculating subunit 132.

[0113] The alignment position information determining subunit 131 is adapted to use the registration test point position closest to each first alignment mark position as the corresponding alignment detection position to obtain alignment detection position information for each alignment detection position in the alignment layer;

[0114] The alignment compensation data calculation subunit 132 is adapted to calculate the alignment compensation data of the first initial layer based on the alignment detection position information and using an alignment model for performing photolithography alignment on the first initial layer.

[0115] In one embodiment of the present invention, the feedback compensation data acquired by the first acquisition unit 120 is obtained by weighting the feedback compensation data when performing registration tests on the first initial layers of multiple batches of wafers according to the number of samples and the sampling time, and then performing weighted averaging.

[0116] In one embodiment of the present invention, the feedforward compensation data determining unit 140 is further adapted to use the initial feedforward compensation data as the feedforward compensation data of the first initial layer when the alignment method is an indirect alignment method.

[0117] In one embodiment of the present invention, the overlay unit 150 is adapted to use the sum of the feedforward compensation data and the feedback compensation data of the first initial layer as the final overlay compensation data of the first initial layer to perform overlay compensation on the first initial layer.

[0118] Figure 11 FIG. 1 is a schematic structural diagram of an exposure system according to an embodiment of the present invention. Figure 11 In a specific implementation, the first initial layer of the wafer to be supplemented is photolithographically aligned in the alignment scanning device 11, and the alignment information is stored in the exposure machine 13. The overlay compensation device 100 can obtain the alignment information, and obtain the feedforward compensation data and feedback compensation data of the first initial layer, and send the obtained feedforward compensation data and feedback compensation data of the first initial layer to the exposure machine 13. The exposure machine 13 adjusts the exposure parameters of the first initial layer based on the feedforward compensation data and feedback compensation data of the first initial layer, and uses the adjusted exposure parameters to expose the first initial layer to form a pattern on the first initial layer, so that a patterned first initial layer can be obtained. After obtaining the patterned first initial layer, the wafer to be compensated can be sent to the overlay measurement device 14 for overlay measurement. The data obtained from the overlay measurement can be used to update the feedback compensation data.

[0119] An embodiment of the present invention further provides another computer-readable storage medium having computer instructions stored thereon. The computer program is executed by a processor to implement any of the steps of the overlay compensation method in the above embodiments, which will not be described in detail.

[0120] In a specific implementation, the computer-readable storage medium may include: ROM, RAM, magnetic disk or optical disk, etc.

[0121] An embodiment of the present invention also provides an electronic device, wherein the terminal may include a memory and a processor, wherein the memory stores a computer program that can be run on the processor, and when the processor runs the computer program, the steps of any one of the overlay compensation methods in the above embodiments are executed, which will not be repeated here.

[0122] Regarding the various modules / units contained in the various devices and products described in the above embodiments, they can be software modules / units, hardware modules / units, or partly software modules / units and partly hardware modules / units. For example, for various devices and products applied to or integrated in a chip, the various modules / units contained therein can all be implemented in the form of hardware such as circuits, or at least some of the modules / units can be implemented in the form of software programs, which run on the processor integrated inside the chip, and the remaining (if any) modules / units can be implemented in the form of hardware such as circuits; for various devices and products applied to or integrated in a chip module, the various modules / units contained therein can all be implemented in the form of hardware such as circuits, and different modules / units can be located in the same component of the chip module (such as a chip, circuit module, etc.) or in different components, or at least some of the modules / units can be implemented in the form of hardware such as circuits. The element can be implemented in the form of a software program, which runs on the processor integrated inside the chip module, and the remaining (if any) modules / units can be implemented in the form of hardware such as circuits; for various devices and products applied to or integrated in the terminal, the various modules / units contained therein can be implemented in the form of hardware such as circuits, and different modules / units can be located in the same component (for example, chip, circuit module, etc.) or different components in the terminal, or, at least some modules / units can be implemented in the form of a software program, which runs on the processor integrated inside the terminal, and the remaining (if any) modules / units can be implemented in the form of hardware such as circuits.

[0123] Although the present invention is disclosed as above, the present invention is not limited thereto. Any person skilled in the art can make various changes and modifications without departing from the spirit and scope of the present invention. Therefore, the scope of protection of the present invention should be based on the scope defined by the claims.

Claims

1. A method for overlay compensation, characterized in that: include: Providing a wafer to be compensated; the wafer to be compensated has an overlay reference layer and a first initial layer located on the overlay reference layer; Acquire initial feedforward compensation data and feedback compensation data corresponding to the first initial layer; the initial feedforward compensation data is actual compensation data of the overlay reference layer, and the feedback compensation data is feedback compensation data obtained when performing an overlay test on the first initial layer; Acquiring alignment information of the first initial layer, and determining an alignment method based on the alignment information of the first initial layer; The alignment information of the first initial layer includes: identification information of the alignment layer; When the alignment mode is a direct alignment mode, obtaining alignment compensation data of the first initial layer based on the alignment information of the first initial layer, and using a difference between the initial feedforward compensation data and the alignment compensation data as the feedforward compensation data of the first initial layer; Performing overlay compensation on the first initial layer using the feedforward compensation data and the feedback compensation data of the first initial layer, and forming a corresponding photolithography pattern on the first initial layer; The alignment information of the first initial layer further includes: alignment model information and alignment mark position information on the alignment layer.

2. The overlay compensation method according to claim 1, wherein: The obtaining alignment compensation data of the first initial layer based on the alignment information of the first initial layer includes: Taking the registration test point position closest to each alignment mark position as the corresponding alignment detection position, and obtaining the alignment detection position information of each alignment detection position in the alignment layer; Based on the alignment detection position information, alignment compensation data of the first initial layer is calculated using an alignment model for performing photolithography alignment on the first initial layer.

3. The overlay compensation method according to claim 1, wherein: The feedback compensation data is obtained by weighting the feedback compensation data when performing registration tests on the first initial layers of multiple batches of wafers according to the sampling quantity and sampling time and then performing weighted averaging.

4. The overlay compensation method according to claim 1, wherein: Also includes: When the alignment method is an indirect alignment method, the initial feedforward compensation data is used as the feedforward compensation data of the first initial layer.

5. The overlay compensation method according to claim 1, wherein: The method of performing overlay compensation on the first initial layer by using the feedforward compensation data and the feedback compensation data of the first initial layer includes: The sum of the feedforward compensation data and the feedback compensation data of the first initial layer is used as the final overlay compensation data of the first initial layer to perform overlay compensation on the first initial layer.

6. An overprint compensation device, characterized in that: include: A providing unit is provided, which provides a wafer to be compensated; the wafer to be compensated has an overlay reference layer and a first initial layer located on the overlay reference layer; A first acquisition unit is adapted to acquire initial feedforward compensation data and feedback compensation data corresponding to the first initial layer; the initial feedforward compensation data is actual compensation data of the overlay reference layer, and the feedback compensation data is feedback compensation data obtained when performing an overlay test on the first initial layer; a second acquiring unit, adapted to acquire alignment information of the first initial layer and determine an alignment method based on the alignment information of the first initial layer; a feedforward compensation data determining unit, adapted to, when the alignment mode is a direct alignment mode, obtain alignment compensation data of the first initial layer based on the alignment information of the first initial layer, and use a difference between the initial feedforward compensation data and the alignment compensation data as the feedforward compensation data of the first initial layer; an overlay unit adapted to perform overlay compensation on the first initial layer by utilizing the feedforward compensation data and the feedback compensation data of the first initial layer, and form a corresponding lithographic pattern on the first initial layer; The alignment information of the first initial layer further includes: alignment model information and alignment mark position information on the alignment layer.

7. The overprint compensation device according to claim 6, wherein: The second acquiring unit includes: an alignment position information determining subunit, adapted to use the position of the registration test point closest to each first alignment mark position as the corresponding alignment detection position, and obtain alignment detection position information for each alignment detection position in the alignment layer; The alignment compensation data calculation subunit is adapted to calculate the alignment compensation data of the first initial layer based on the alignment detection position information and using an alignment model for performing photolithography alignment on the first initial layer.

8. The overprint compensation device according to claim 6, wherein: The feedback compensation data acquired by the first acquisition unit is obtained by weighting the feedback compensation data when performing an overlay test on the first initial layer of multiple batches of wafers according to the sampling quantity and sampling time and then performing weighted averaging.

9. The overprint compensation device according to claim 6, wherein: The feedforward compensation data determining unit is further adapted to use the initial feedforward compensation data as the feedforward compensation data of the first initial layer when the alignment method is an indirect alignment method.

10. The overprint compensation device according to claim 6, wherein: The overlay unit is adapted to use the sum of the feedforward compensation data and the feedback compensation data of the first initial layer as the final overlay compensation data of the first initial layer to perform overlay compensation on the first initial layer.

11. A computer-readable storage medium having a computer program stored thereon, characterized in that: The computer program is executed by a processor to implement the steps of the method according to any one of claims 1 to 5.

12. An electronic device comprising a memory and a processor, wherein the memory stores a computer program that can be run on the processor, wherein: When the processor runs the computer program, the steps of the method according to any one of claims 1 to 5 are performed.

Images