Reference image grouping in superposition metrology

Abstract

A superposition metrology system may include a controller for: receiving metrology data associated with a plurality of superposition targets on one or more samples; generating reference metrics for at least some of the plurality of superposition targets based on the metrology data, wherein the reference metrics are associated with one or more properties of the respective superposition targets that contribute to superposition errors; classifying the plurality of superposition targets into one or more groups based on the reference metrics calculated for the plurality of superposition targets; generating reference images for at least some of the one or more groups; generating corrected metrology data using the associated reference images for at least some of the one or more groups; and generating superposition measurements for the plurality of superposition targets based on the corrected metrology data.

Description

[0001] CROSS-REFERENCE TO RELATED APPLICATIONS

[0002] This application claims the benefit of U.S. Provisional Application No. 63 / 306,103, filed on February 3, 2022, with Einat Peled, Naama Cohen, and Yuval Lamhot as inventors, under 35 U.S.C. §119(e), the entire text of which is incorporated herein by reference. Technical Field

[0003] The present disclosure relates generally to overlay metrology and, more particularly, to mitigating tool-induced errors in overlay metrology. Background Art

[0004] Overlay metrology provides a measure of the relative registration (or offset) between features manufactured using different patterning processes. Some overlay metrology techniques determine overlay measurements by illuminating a portion of a sample containing overlapping features associated with different patterning processes of interest and collecting data associated with the symmetry of the overlapping features. For example, some scatterometry overlay metrology (SCOL) utilizes an overlay target containing periodic features associated with different patterning processes on different layers in an overlapping region of the sample. In such a configuration, opposite diffraction orders (e.g., + / -1 diffraction order, + / -2 diffraction orders, or the like) may have a common amplitude when the features of the overlay target are symmetrical and may differ in amplitude when the features of the overlay target are offset. Thus, overlay measurements associated with the patterning process may be determined based on this symmetry data. However, such techniques are sensitive to measurement non-uniformity, which may manifest as errors in the overlay measurement and are often referred to as tool induced shift (TIS) errors. Therefore, there is a need to develop systems and methods for addressing the above-mentioned deficiencies. Summary of the invention

[0005] According to one or more illustrative embodiments, a superposition metrology system is disclosed. In one illustrative embodiment, the system includes a controller for implementing a metrology recipe. In another illustrative embodiment, the controller receives metrology data associated with a plurality of superposition targets on one or more samples. In another illustrative embodiment, the controller generates reference metrics for at least some of the plurality of superposition targets based on the metrology data, wherein the reference metrics are associated with one or more properties of the respective superposition targets that contribute to superposition errors. In another illustrative embodiment, the controller classifies the plurality of superposition targets into one or more groups based on the reference metrics calculated for the plurality of superposition targets. In another illustrative embodiment, the controller generates reference images for at least some of the one or more groups. In another illustrative embodiment, the controller generates corrected metrology data using the associated reference images for at least some of the one or more groups. In another illustrative embodiment, the controller generates superposition measurements for the plurality of superposition targets based on the corrected metrology data.

[0006] According to one or more illustrative embodiments, a superposition metrology system is disclosed. In one illustrative embodiment, the system includes an optical subsystem. In another illustrative embodiment, the optical subsystem includes one or more optical elements configured to illuminate a sample using light from an illumination source according to a metrology recipe. In another illustrative embodiment, the optical subsystem includes one or more optical elements configured to direct at least a portion of the light from the sample to a detector located at a pupil plane. In another illustrative embodiment, the system includes a controller for implementing the metrology recipe. In another illustrative embodiment, the controller receives metrology data associated with a plurality of superposition targets on one or more samples. In another illustrative embodiment, the controller generates reference metrics for at least some of the plurality of superposition targets based on the metrology data, wherein the reference metrics are associated with one or more properties of the corresponding superposition targets that contribute to superposition errors. In another illustrative embodiment, the controller classifies the plurality of superposition targets into one or more groups based on the reference metrics calculated for the plurality of superposition targets. In another illustrative embodiment, the controller generates reference images for at least some of the one or more groups. In another illustrative embodiment, the controller generates corrected metrology data using the associated reference images for at least some of the one or more groups. In another illustrative embodiment, the controller generates overlay measurements of the plurality of overlay targets based on the corrected metrology data.

[0007] According to one or more illustrative embodiments, a superposition method is disclosed. In one illustrative embodiment, the method includes receiving metrology data associated with a plurality of superposition targets on one or more samples. In another illustrative embodiment, the method includes generating reference metrics for at least some of the plurality of superposition targets based on the metrology data, wherein the reference metrics are associated with one or more properties of the respective superposition targets that contribute to superposition errors. In another illustrative embodiment, the method includes classifying the plurality of superposition targets into one or more groups based on the reference metrics calculated for the plurality of superposition targets. In another illustrative embodiment, the method includes generating reference images for at least some of the one or more groups. In another illustrative embodiment, the method includes generating corrected metrology data using the associated reference images for at least some of the one or more groups. In another illustrative embodiment, the method includes generating superposition measurements for the plurality of superposition targets based on the corrected metrology data.

[0008] It should be understood that both the foregoing summary and the following detailed description are exemplary and explanatory only and do not necessarily limit the invention as claimed. The accompanying drawings, which are incorporated in and constitute a part of this specification, illustrate embodiments of the invention and together with the summary, serve to explain the principles of the invention. BRIEF DESCRIPTION OF THE DRAWINGS

[0009] Those skilled in the art may better appreciate the numerous advantages of the present disclosure by referring to the accompanying drawings.

[0010] Figure 1A is a block diagram of a superimposed metering system according to one or more embodiments of the present disclosure.

[0011] Figure 1B is a simplified schematic diagram of an optical subsystem according to one or more embodiments of the present disclosure.

[0012] Figure 2A is a top view of a single unit of a stacked target having periodic features according to one or more embodiments of the present disclosure.

[0013] Figure 2B According to one or more embodiments of the present disclosure Figure 2A Side view of a single unit in.

[0014] Figure 3 is a conceptual diagram of a pupil image illustrating generation of a differential signal according to one or more embodiments of the present disclosure.

[0015] Figure 4 is a flow chart illustrating steps performed in a method for generating different reference images for metrology data associated with different groups of overlay targets in accordance with one or more embodiments of the present disclosure.

[0016] Figure 5 The present invention is to illustrate the use and Figure 3 Conceptual diagram of a pupil image in which differential signals are generated at different pupil center positions.

[0017] Figure 6 is a graph of TIS as a function of pupil center shift for several overlay targets, illustrating the relationship between pupil center position, TIS, and pupil center slope for different overlay targets, according to one or more embodiments of the present disclosure.

[0018] Fig. 7A is a graph of TIS as a function of pupil center shift based on metrology data from multiple overlay targets without reference image correction in accordance with one or more embodiments of the present disclosure.

[0019] Figure 7B is a graph of TIS as a function of pupil center shift based on metrology data from a plurality of overlay targets corrected using a reference image generated based on metrology data from particular ones of the overlay targets in accordance with one or more embodiments of the present disclosure.

[0020] Fig. 8A is a chart illustrating a graph of pupil center slope for a stacking target across a sample in accordance with one or more embodiments of the present disclosure.

[0021] Figure 8B is a diagram illustrating a diagram of groups into which overlay targets are classified according to one or more embodiments of the present disclosure. DETAILED DESCRIPTION

[0022] Reference will now be made in detail to the disclosed subject matter illustrated in the accompanying drawings. The present disclosure has been particularly shown and described with respect to certain embodiments and specific features thereof. The embodiments set forth herein should be regarded as illustrative rather than restrictive. It should be readily apparent to one of ordinary skill in the art that various changes and modifications may be made to form and details without departing from the spirit and scope of the present disclosure.

[0023] Embodiments of the present disclosure relate to systems and methods for generating reference images for scatterometry overlay (SCOL) metrology measurements to mitigate tool induced shift (TIS) errors associated with measurement non-uniformity. The reference image may characterize the measurement non-uniformity and may be used to at least partially remove the effects of such measurement non-uniformity and associated TIS errors during measurement. In embodiments, multiple reference images are generated that may be suitable for different groups of metrology data (e.g., metrology data from different groups of overlay targets). For example, metrology data from various overlay targets across samples may be grouped based on a metric associated with the measurement non-uniformity and a different reference image may be generated for each group. Metrology data associated with subsequent measurements of overlay targets on the same or different samples may then be grouped based on the metric used to select an appropriate reference image. As an illustration, one source of measurement non-uniformity may include non-uniformity of printed features that may vary across samples. Therefore, a single reference image may not mitigate the associated TIS errors with the same effectiveness. In some embodiments, multiple reference images are generated and applied to different groups of features.

[0024] SCOL measurement is based on metrology data generated by illuminating a portion of a sample having features associated with two patterning processes (e.g., photolithography exposure, etching process, or the like) and collecting the resulting light from the sample, wherein the features associated with different patterning processes may be on the same or different layers of the sample. SCOL technology can generally utilize an overlay target comprising one or more units, but this is not required. As used herein, an overlay target may generally refer to any portion of a sample suitable for overlay measurement and may include, but is not limited to, special features designed for the purpose of overlay measurement or device features associated with a manufactured device. In addition, the features of the overlay target (referred to herein as target features) may be on any layer or combination of layers of the sample. For example, the target feature may be located on a photoresist layer. In this case, the feature may be characterized by a refractive index change caused by photolithography exposure in the photoresist layer. As another example, the target feature may be located on a process layer. In this case, the feature may be characterized by a change in the material composition caused by etching the sample after patterning with a photoresist or any other patterning process.

[0025] It is considered herein that TIS errors in SCOL metrology can be attributed to a variety of sources, including illumination non-uniformity in the overlay metrology system and specific physical and optical properties of the sample features being measured. Since process variations can result in slight deviations between the properties of the overlay targets across samples, it may be the case that a single reference image may not be sufficient to mitigate TIS errors in the metrology targets across samples.

[0026] Embodiments of the present disclosure relate to systems and methods for classifying metrology data from different metrology targets across a sample into groups (referred to herein as reference groups) and providing different reference images for different reference groups.

[0027] Any suitable technique may be used to group metrology data from different stacking targets into different reference groups. In some embodiments, the reference groups are based on metrics derived from measured or expected process variations across samples. In some embodiments, the reference groups are based on metrics derived from the metrology data itself. For example, in SCOL measurements based on pupil plane images, the metric may be a pupil center slope associated with a variation of a determined stacking based on a selection of a pupil center position. As an illustration, some pupil plane SCOL techniques may determine stacking based on difference signals associated with the difference in amplitude between diffraction orders of opposite signs (e.g., +1 and -1 diffraction orders) or more generally opposite pupil coordinates. Defining these difference signals according to pupil center positions allows the determined stacking measurements to vary for different selected pupil center positions. It is contemplated herein that pupil center slope may be a suitable metric for grouping metrology data from stacking targets across samples for the purpose of reference image generation. In other words, the reference image may effectively mitigate TIS errors for stacking targets with similar pupil center slope values. Therefore, pupil center slope may be an indirect metric of sample variation that may contribute to TIS errors.

[0028] In some embodiments, the grouping derived for one sample may be applied to additional samples. Continuing with the previous example, the pupil center slope may be determined for metrology data associated with an overlay target on one or more additional samples. This data may then be grouped based on previously defined bins representing a range of pupil slopes and the corresponding reference image may be used for overlay determination.

[0029] Additional embodiments of the present disclosure are directed to monitoring the value of pupil center slope (or any other selected metric) over time across multiple samples as a function of sample position. It is contemplated herein that variations in pupil center slope (or any other selected metric) over time across samples at a particular sample position may be indicative of process variations. This information may then be used directly for process control (e.g., for generating correctable terms for a process tool to compensate for such process variations) and / or for adjusting the definition of a reference group. For example, it may be the case that a reference group may generally be associated with a spatial region of samples used to define the reference group (e.g., a radial spatial region associated with radial process variations). However, if metrology data associated with a particular sample position begins to classify into different reference groups over time, it may be the case that the definition of the reference group (e.g., the definition of bins associated with a range of pupil center slope) may need to be updated.

[0030] Reference Figure 1A8 , a system and method for generating grouped reference images for overlay metrology are described in more detail according to one or more embodiments of the present disclosure.

[0031] Figure 1A is a block diagram of an overlay metrology system 100 in accordance with one or more embodiments of the present disclosure.

[0032] In some embodiments, the overlay metrology system 100 includes an optical subsystem 102 for generating metrology data from an overlay target 104 on a sample 106. In some embodiments, the overlay metrology system 100 further includes a controller 108 having one or more processors 110 configured to execute program instructions maintained on a memory 112 (e.g., a storage medium). The controller 108 can be communicatively coupled with any component of the overlay metrology system 100, such as, but not limited to, the detector 124. In this manner, the controller 108 can generate overlay measurements based on the metrology data.

[0033] The overlay metrology system 100 may generally be configured according to one or more metrology recipes. A metrology recipe may include a set of parameters for controlling various aspects of an overlay measurement, such as, but not limited to, illumination of the sample, collection of light from the sample in the form of metrology data, position of the sample during a measurement, or processing steps for generating a measurement based on collected metrology data. In this manner, the optical subsystem 102 may be configured to provide a selected type of measurement for a selected overlay target design.

[0034] For example, the metrology recipe may include parameters of the illumination beam 114, such as, but not limited to, the illumination wavelength, the illumination pupil distribution (e.g., the distribution of illumination angles and the associated intensities of the illumination at those angles), the polarization of the incident illumination, or the spatial distribution of the illumination. As another example, the metrology recipe may include collection parameters, such as, but not limited to, the collection pupil distribution (e.g., the desired distribution of angular light from the overlay target 104 to be used for measurement and the associated filtered intensities at those angles), the collection field stop settings for selecting portions of the overlay sample 104 of interest, the polarization of the collected light, wavelength filters, or parameters for controlling one or more detectors. As another example, the metrology recipe may include various parameters associated with the design of the overlay target 104, such as, but not limited to, the location and orientation of sample features (e.g., the pitch of grating features along a particular direction). As another example, the metrology recipe may include various parameters associated with the location of the sample 106 during measurement, such as, but not limited to, the sample height, the sample orientation, whether the sample is static during measurement, or whether the sample is in motion during measurement (along with associated parameters describing speed, scanning pattern, or the like). As another example, a metrology recipe may include various processing steps for producing a determined overlay measurement (eg, a specific value of an overlay to be output) based on metrology data.

[0035] In some embodiments, the overlay metrology system 100 operates as a SCOL metrology tool such that a metrology recipe is associated with a SCOL measurement technique.

[0036] In a general sense, a stacking target 104 suitable for SCOL measurement techniques (e.g., SCOL stacking target 104) may include target features that are 180 degree rotationally symmetric for each patterning process of interest and are located in an overlapping region of the sample 106. In some embodiments, the target features associated with any of the patterning processes are periodic such that they produce discrete diffraction orders from the incident illumination. However, this is not required and the target features associated with any of the patterning processes need not be periodic.

[0037] Reference Figure 2A to 2B , describing a non-limiting example of a SCOL overlay target 104. In a general sense, the overlay target 104 may include one or more cells 202, each cell 202 including overlapping target features associated with two or more patterning processes of interest. In some embodiments, the overlay target 104 includes target features dedicated to overlay measurements. In some embodiments, the overlay target 104 includes an area of ​​the sample 106 that includes device features (e.g., features associated with a device fabricated on the sample 106).

[0038] Figure 2A is a top view of a single unit 202 of a superimposed target 104 having periodic features according to one or more embodiments of the present disclosure. Figure 2B According to one or more embodiments of the present disclosure Figure 2A A side view of a single unit 202 in FIG.

[0039] In some embodiments, cell 202 includes a first layer of printing elements 204 located on a first layer 206 of sample 106 and a second layer of printing elements 208 located on a second layer 210 of sample 106 , the first layer of printing elements 204 and the second layer of printing elements 208 being oriented such that regions including the first layer of printing elements 204 and the second layer of printing elements 208 overlap to form a grating-on-grating structure. Figure 2B Further depicted is a substrate 212 underlying the first layer of printing elements 204 and the second layer of printing elements 208 .

[0040] The first layer printing elements 204 and the second layer printing elements 208 in any particular unit 202 may be designed to have a plurality of printing elements along any direction (e.g., Figure 2BThe expected offset (f0) of the first layer 204 and the second layer 208 is preferably any expected offset (f0) of the first layer 204 and the second layer 208. For example, an expected offset of zero (f0=0) may specify that when the physical overlay is also zero (e.g., no overlay error), the first layer 204 and the second layer 208 completely overlap. In this configuration, the relative shift between the first layer 204 and the second layer 208 indicates an overlay error during manufacturing. As another example, a non-zero expected offset (f0≠0) may specify that when the physical overlay is zero (e.g., no overlay error), the first layer 204 and the second layer 208 exhibit this expected offset. The overlay target 104 may generally be formed of any number of cells 202 that may have any combination of expected offsets (e.g., values ​​of f0).

[0041] Although not explicitly illustrated, the overlay target 104 may be adapted for overlay measurements along multiple directions (e.g., orthogonal directions). In general, the measurement directions may correspond to the periodic directions of the first layer of printing elements 204 and the second layer of printing elements 208 (e.g., the periodic directions of the grating-on-grating structure). As an illustration, Figure 2A and 2B The cell 202 depicted in shows periodicity along the X direction and is suitable for superposition measurements along the X direction.

[0042] In some embodiments, the superposition target 104 includes a plurality of Figure 2A and 2B The one or more units 202 have a periodicity along a second direction (e.g., the X direction depicted in FIG. 1 ) and a periodicity along a second direction (e.g., Figure 2A and 2B In some embodiments, overlay target 104 includes one or more cells 202 having periodicity along both directions. For example, first layer printing elements 204 and second layer printing elements 208 may include structures that are periodic in both the X and Y directions (e.g., a hatch pattern, a grid of square or rectangular structures, or the like).

[0043] Generally speaking Figure 2A and 2B , it should be understood that Figure 2A and 2BThe descriptions together with the associations are merely illustrative and should not be interpreted as limitations of the present disclosure. For example, the first layer 206 and the second layer 210 may generally have any thickness. As another example, the sample 106 may include any number of layers between the first layer printing element 204 and the second layer printing element 208 and / or between the first layer printing element 204 and the substrate 212. As another example, the first layer printing element 204 and the second layer printing element 208 may generally include any 180 degree rotationally symmetric features and need not include grating-on-grating or even periodic features.

[0044] Various SCOL techniques have been developed to generate overlay measurements based on metrology data from the overlay target 104. Metrology data for the SCOL techniques can generally be generated in a pupil plane and / or a field plane. For example, a pupil plane image can capture various diffraction orders from target features associated with each patterning process. In such a configuration, the intensity at any location of the pupil plane can be associated with the interference of diffraction from target features associated with different patterning processes. As another example, more generally, various field plane images can be generated based on selected combinations of diffraction orders or light in different portions of the pupil plane. For example, the overlay metrology system 100 can include various components for isolating desired diffraction orders for a particular field plane image.

[0045] As an illustration, a first order SCOL technique may be based on metrology data associated with first order diffraction from target features associated with different patterning processes. Such techniques may typically require metrology data from two units 202 with different expected offsets (e.g., ±f0). In such a configuration, a superposition measurement may be determined based on a difference signal associated with the difference between the positive and negative first order diffraction from each unit (e.g., more generally, opposite pupil coordinates). These difference images may be generated directly in the pupil plane images or based on separate field plane images for each unit (e.g., one image based on positive first order diffraction and one image based on negative first order diffraction).

[0046] As another illustration, zero-order SCOL technology may be based on metrology data associated with zero-order light from superimposed targets, such as, but not limited to, zero-order diffraction from overlapping target features or opposite-order diffraction (e.g., positive first-order diffraction from target features in one layer and negative first-order diffraction from target features in another layer).

[0047] Various non-limiting examples of SCOL techniques are generally described in Adel et al., “Diffraction order control in overlay metrology—a review of the roadmap options,” Proc. Of SPIE, Vol. 6922, 692202-1 (2008); U.S. Pat. No. 7,317,531, entitled “Apparatus and methods for detecting overlay errors using scatterometry,” issued on January 8, 2008; U.S. Pat. No. 10,197,389, entitled “Approaches in first order scatterometry overlay based on introduction of auxiliary electromagnetic fields,” issued on February 5, 2019; and U.S. Pat. No. WO 2017 / 031587, entitled “Apparatus and methods for detecting overlay errors using scatterometry,” issued on March 16, 2017. International Publication No. 2017 / 044283, the entire text of all of the aforementioned documents are incorporated herein by reference.

[0048] With reference now to equations (1) to (8), the overlay determination and use of reference images to mitigate TIS errors are described in more detail according to one or more embodiments of the present disclosure. For purposes of illustration, a one-stage SCOL technique based on pupil plane images is described. However, it should be understood that this is non-limiting and that the concepts described herein can be extended by one of ordinary skill in the art to any SCOL technique using any combination of pupil plane metrology data or field plane metrology data.

[0049] In some embodiments, a differential signal such as the following may be defined based on the metrology data:

[0050]

[0051]

[0052] where I is the intensity at a particular location in the pupil plane, the subscript (+ / -1) refers to the diffraction order, and the superscript (±f0) refers to the expected offset of the measured cell 202. In this way, corresponds to the differential signal generated from cell 202 with an expected offset of +f0, while corresponds to the differential signal generated from cell 202 having an expected offset of -f0.

[0053] It should be noted that the superposition target 104 need not include periodic features and therefore need not provide discrete diffraction orders. Thus, equations (1) and (2) can be generalized for all pupil coordinates.

[0054] Figure 3 is a conceptual diagram of a pupil image (eg, an image in a collection pupil plane) illustrating generation of a differential signal according to one or more embodiments of the present disclosure.

[0055] Specifically, Figure 3 Illustrated are 0th order diffraction 302, positive first order diffraction 304, and negative first order diffraction 306. This distribution may be produced by illuminating cell 202 comprising periodic first layer 204 and second layer 208 printing elements using a single illumination beam at a normal angle of incidence.

[0056] like Figure 3 , the differential signal D(k) may correspond to the difference between the intensity at pupil coordinate k and the intensity at pupil coordinate −k, where coordinates k and −k are 180 degrees rotationally symmetric about pupil center location 308. Furthermore, the differential signal may be formed as an average (e.g., a weighted average) of all pupil coordinates.

[0057] These differential signals can then be combined into so-called K and G signals:

[0058]

[0059]

[0060] The overlay measurement (OVL) can then be generated as

[0061]

[0062] As long as G≠0.

[0063] Assuming OVL<<P and f0<<P, equations (3) and (4) indicate that the K signal depends on the actual value of the superimposed OVL measured at the location of the superimposed target 104, while the G signal does not depend on the actual value. Instead, the G signal depends on which is negligible under these conditions. In this case, Equation 5 simplifies to

[0064]

[0065] As previously described herein, reference images may be used to at least partially mitigate TIS errors in overlay measurements. TIS errors may be due to a number of sources, including, but not limited to, non-uniformity of measurements performed by the optical subsystem 102 (e.g., non-uniform illumination beams, non-uniformity in illumination paths, non-uniformity in light collection paths, or the like) and / or non-uniformity in the overlay target 104. Such TIS errors may typically be observed from overlay measurements made using the overlay target 104 at two different orientations relative to the optical subsystem 102. In particular, TIS may be characterized as:

[0066]

[0067] where OVL0 corresponds to a superposition measurement with the superposition target 104 having the sample in a first orientation relative to the optical subsystem 102 and where OVL 180 Corresponds to an overlay measurement of the same overlay target 104 rotated 180 degrees relative to the optical subsystem 102.

[0068] In this characterization, the TIS-independent TIS-corrected overlay measurement (OVL) can be determined as:

[0069]

[0070] It should be noted that a TIS-corrected overlay measurement as defined by equation (8) may typically require two measurements of the same overlay target 104 at a 180 degree rotation. However, performing such dual measurements for each overlay target 104 of interest across a sample may be time consuming and result in relatively low measurement throughput.

[0071] To alleviate the need to measure each overlay target 104 twice at different rotations, a reference image may be generated that may capture relevant non-uniformities. For example, a reference image may be generated based on metrology data captured for the overlay target 104 at two orientations (e.g., at 0 and 180 degrees). As an example, where the overlay metrology system 100 is configured to determine overlay based on a pupil image (e.g., an image generated by the detector 124 at the pupil plane 138), the reference image may be generated by summing the pupil images (e.g., metrology data) generated at a 180 degree rotation. Furthermore, the reference image may be made symmetric to prevent noise amplification. For example, the reference image may be made symmetric by dividing the reference image by the average of the reference image and a 180 degree rotated version of itself. The effects of asymmetry on the calibration of SCOL measurements are generally described in the following patents: U.S. Patent No. 9,909,982 issued on March 6, 2018, U.S. Patent No. 9,164,397 issued on October 20, 2015, and U.S. Patent No. 9,390,492 issued on July 12, 2016, all of which are incorporated herein by reference in their entirety.

[0072] Regardless of how the reference image is generated, the reference image may be used to correct metrology data associated with a single measurement at a single orientation of the overlay target 104. For example, the pupil image (eg, metrology data) may be divided by the reference image to mitigate the effects of TIS errors.

[0073] However, the effectiveness of a reference image for mitigating TIS error is related to how well the reference image captures the measurement non-uniformity that contributes to TIS error. It should be recognized herein that many previous SCOL techniques have utilized a single reference image for all overlay targets 104 on the sample 106 based on the assumption that the measurement non-uniformity that contributes to TIS error is primarily due to non-uniformity in illumination and / or light collection through the optical subsystem 102. However, it is carefully contemplated herein that sample variations between overlay targets 104 may also contribute to TIS error such that a single reference image may not accurately mitigate TIS error for all overlay targets 104.

[0074] Figure 4 is a flow chart illustrating steps performed in a method 400 for generating different reference images for metrology data associated with different groups of overlay targets 104 in accordance with one or more embodiments of the present disclosure. Applicants should note that the embodiments and enabling techniques previously described herein in the context of an overlay metrology system 100 should be interpreted as extending to the method 400. However, it should be further noted that the method 400 is not limited to the architecture of the overlay metrology system 100.

[0075] In some embodiments, method 400 includes a step 402 of receiving metrology data associated with superposition targets 104 on one or more samples 106. For example, the metrology data may include any combination of pupil plane or field plane images suitable for superposition measurement using the SCOL technique (e.g., based on a selected metrology recipe). The metrology data generated in step 402 may be associated with all or a subset of the superposition targets 104 on the one or more samples 106. In addition, the metrology data may be associated with superposition measurements along one or more measurement directions.

[0076] In some embodiments, the method 400 includes a step 404 of generating a reference metric for each of the overlay targets 104 based on the metrology data, wherein the reference metric is associated with one or more properties of the respective overlay targets 104 that contribute to the overlay error. In some embodiments, the method 400 includes a step 406 of classifying the overlay targets 104 into one or more groups based on the reference metric. In some embodiments, the method 400 includes a step 408 of generating a reference image for at least some of the one or more groups.

[0077] In a general sense, steps 406 and 408 may provide reference metrics based on any number of groups. It is contemplated herein that the number of groups may depend on the specific characteristics of each particular sample 106 or samples 106 within a particular batch and the desired number of groups may not be known a priori. Thus, method 400 may facilitate analyzing the overlay target 104 on each sample 106 based on the reference metrics (e.g., step 404), determining a suitable number of groups that would benefit from separate reference images (e.g., step 406), and generating associated reference images (e.g., step 408).

[0078] As previously described herein, the reference image may be generated using any suitable technique known in the art. In some embodiments, the reference image for a particular group is generated based on metrology data from a single representative overlay target 104 in the group (e.g., metrology data from multiple orientations of overlay targets 104). In some embodiments, the reference image for a particular group is generated based on metrology data from multiple representative overlay targets 104.

[0079] The reference metric may include any metric associated with a property of the overlay target 104 that contributes to an overlay error (e.g., a TIS error). Such an overlay error may be associated with a non-uniformity of the overlay target 104 and may vary based on a location on the sample 106. For example, it may be the case that a sample may exhibit process variations that vary with radial distribution. As an illustration, a property such as, but not limited to, film thickness, a sidewall angle of a fabricated feature, or asymmetry of a fabricated feature may vary with radial distribution across the sample. In addition, one or more such process variations may affect the overlay measurement such that a single reference image may not be suitable for mitigating TIS errors across the entire sample. Therefore, measurements of one or more such process variations may be used as a basis for dividing metrology data into reference groups each having a different reference image. Process variation detection is generally described in U.S. Patent No. 10,699,969, issued on June 30, 2020, the entirety of which is incorporated herein by reference.

[0080] In some embodiments, the reference metric is based on at least one of a physical measurement, an expected measurement, or a simulation of one or more properties of superimposed target 104. Such measurements may be performed at the location of superimposed target 104 or at different locations on sample 106 and extrapolated to the location of superimposed target 104. For example, the reference metric may be based on, for example, but not limited to, target features (e.g., Figure 2A to 2B The asymmetry of the first layer printing element 204 and / or the second layer printing element 208 depicted in FIG. 1 ), the sidewall angle along any direction, the critical dimension (CD), or the layer thickness (e.g., as shown in FIG. 1 ). Figure 2A to 2B Measurement of the thickness of the first layer 206 and / or the second layer 210 depicted in FIG.

[0081] In some embodiments, the reference metric is derived from the metrology data itself. In this way, no additional measurements, simulations, or assumptions may be required.

[0082] Now, the generation of reference metrics based on pupil plane SCOL technology is described according to one or more embodiments of the present disclosure. However, it should be understood that this description is provided purely for illustrative purposes and should not be interpreted as limiting. Instead, the concepts disclosed herein can be extended to field plane SCOL technology by a person of ordinary skill in the art.

[0083] In some embodiments, the reference metric includes or is based on a pupil center slope measurement generated based on metrology data prior to correction using the reference image, where the pupil center slope may correspond to a variation of the overlay measurement determined based on the selection of the pupil center position 308 .

[0084] like Figure 3 , an overlay measurement may be determined based on a difference image generated from a pupil plane image of light emanating from the overlay target 104. However, this approach may be sensitive to the choice of pupil center position 308.

[0085] Figure 5 The present invention is to illustrate the use and Figure 3 Conceptual diagram of pupil images (e.g., images in the collection pupil plane) that produce differential signals at different pupil center positions 308. Figure 5 , the opposite pupil coordinates are defined by the pupil center position 308. Thus, the differential signal and resulting overlay determination may be affected by the selection of the pupil center position 308. Additionally, the selection of the pupil center position 308 may affect the TIS error associated with the measurement, as described in more detail below.

[0086] Pupil center slope (PC slope ), the relationship between the resulting stacking measurement and TIS can be described by equations (9) to (11), which share the same variables and definitions as equations (7) and (8):

[0087] OVL0=OVL+Δp·PC slope (9)

[0088] OVL 180 =OVL-Δp·PC slope (10)

[0089] TIS=Δp·PC slope (11)

[0090] where Δp corresponds to a pupil center displacement from a reference position (eg, an arbitrary position) and may have units of pixels as measured by a pupil plane image (eg, metrology data).

[0091] Figure 6 is a graph of TIS as a function of pupil center displacement (Δp) of several superimposed targets 104 according to one or more embodiments of the present disclosure, illustrating the relationship between pupil center position 308, TIS and pupil center slope (PC) for different superimposed targets 104. slope ). For example, the TIS may generally vary approximately linearly with the selection of the pupil center position 308 (eg, with the pupil center shift), where the pupil center slope (PC slope ) corresponds to the slope of this linear relationship. However, the metrology data for different superimposed targets 104 may show a pupil center slope (PC slope ) and may further have different linear constants that affect the value (e.g., zero) to which TIS is minimized.

[0092] Although not explicitly shown, a determined overlay value (e.g., OVL0, OVL 180 or similar) and pupil center position 308 (or Figure 6 Similar curves are shown for the pupil center shift shown in .

[0093] It should be noted that a typical pupil plane-based SCOL technique may select the pupil center position 308 in various ways. For example, the pupil center position 308 may be selected to minimize the TIS of all measured samples. Figure 6 6 is illustrated by dashed line 602 indicating the selection of pupil center shift (and associated pupil center position 308) that attempts to optimize both minimum average TIS and 3σ variation across the measured stack target 104. Pupil center optimization techniques are generally described in International Patent Publication No. 2014 / 138741, published on September 12, 2014, which is incorporated herein by reference in its entirety.

[0094] In this paper, after careful consideration, the pupil center slope (PC slope ) is a metric that characterizes a property of the sample 106 at the location of the superimposed target 104 and is independent of the particular optical subsystem 102 used to generate the associated metrology data. In particular, simulations and SCOL theory indicate that the pupil center slope is related to the thickness of the measured layer (e.g., as Fig. 7A ). Therefore, the pupil center slope is well suited as a reference metric for grouping overlay objects with similar physical properties based solely on metrology data (e.g., pupil plane images of overlay objects 104) without requiring additional measurements.

[0095] Furthermore, when a reference image is generated based on metrology data from a particular overlay target 104 (e.g., based on metrology data from two orientations of the particular overlay target 104 rotated 180 degrees from each other), the pupil center slope (PC) calculated based on the corrected metrology data may be used to generate the reference image. slope ) can be approximately zero (at least within a suitable range of positions of pupil center position 308). Therefore, the TIS at this position is also close to zero or constant (e.g., as given by equations (9) to (11) and Figure 3 , 5 and 6 instructions).

[0096] Fig. 7A and 7B A simulation is shown that illustrates the effect of the selection of a reference image on the TIS of a plurality of overlay objects 104 according to one or more embodiments of the present disclosure. Specifically, Fig. 7A and 7B is based on simulations using a superimposed target 104 in locations with different layer thicknesses. Fig. 7A is a graph of TIS as a function of pupil center shift based on simulated metrology data from multiple superimposed targets 104 without reference image correction in accordance with one or more embodiments of the present disclosure. Fig. 7A In FIG. 1 , each superimposed target 104 (represented by a location on the sample 106) has a different pupil center slope (PC) indicating a different physical property of the associated superimposed target 104. slope ). Figure 7B FIG. 1 is a graph of TIS as a function of pupil center shift based on simulated metrology data from multiple overlay targets 104 corrected using a reference image generated based on simulated metrology data from a particular one of the overlay targets 104 in accordance with one or more embodiments of the present disclosure. Specifically, FIG. 1 is a graph of TIS as a function of pupil center shift based on simulated metrology data from a plurality of overlay targets 104 corrected using a reference image generated based on simulated metrology data from a particular one of the overlay targets 104 at location 3. Figure 7B Reference image. Figure 7B In the example, the corrected pupil center slope (and TIS) is approximately zero for the range of pupil center shifts considered. Fig. 7A The superimposed target 104 of metrology data having similar values ​​of pupil center slope of the image (as depicted in FIG. 1 ) is also well corrected by the same reference image as indicated by similar low values ​​of corrected pupil center slope and TIS. Fig. 7A The superposition target 104 of the metrology data of substantially different values ​​of the pupil center slope (as depicted in FIG. Figure 7B The relatively high corrected pupil center slope and TIS values ​​in indicate that it is well corrected by the reference image.

[0097] Therefore, it is contemplated herein that a reference image may characterize the interaction of the optical subsystem 102 with the measured overlay target 104 (or a location of the measured overlay target 104), in addition to the non-uniformity of the optical subsystem 102 alone. Because the characteristics of the overlay target 104 may vary across different locations of the sample 106 due to factors such as, but not limited to, reflectivity variations (which may require different illumination intensities for each point) or other imperfections in the measurement or associated calculations, the effectiveness of a particular reference image in mitigating TIS may vary across samples 106 or more generally with process variations.

[0098] However, the TIS of multiple overlay targets 104 having different characteristics may be reduced by first (e.g., in step 406) grouping metrology data from the overlay targets 104 into one or more groups based on the value of the pupil center slope (e.g., as measured prior to correction by any reference image), and then (e.g., in step 408) generating a reference image for each group based on the metrology data of the overlay targets 104 within the respective group. As described by equations (9) to (11) and Figure 3 , 5 6, the TIS value may be determined based on a determined superposition measurement (eg, an OVL0 measurement) or by a calculated TIS value (eg, a combination of OVL0 and OVL 180 The measurement of the two is correlated) to produce a value for the pupil center slope. However, using a superimposed measurement from a single set of metrology data (e.g., an OVLO measurement) may be faster and therefore desirable in some applications.

[0099] Again, more general reference Figure 4 , step 406 of sorting metrology data from overlay target 104 into groups based on a reference metric (eg, pupil center slope or any other suitable metric) may be performed in a variety of ways within the spirit and scope of the present disclosure.

[0100] In some embodiments, the range of values ​​of the reference metric observed in the metrology data from the overlay target 104 can be divided into any number of bins representing one or more groups. In this way, the metrology data associated with a particular overlay target 104 can be classified based on the corresponding bin containing the value of the calculated reference metric for this overlay target 104.

[0101] In a general sense, any number of groups and associated reference images may be generated. However, since the generation of reference images requires metrology data at two sample orientations (e.g., 0 and 180 degree orientations), a large number of groups and their associated reference images will require more data collection at multiple sample orientations. In this case, there is a tradeoff between achieving minimum TIS (by having a large number of groups) and faster measurement of the entire sample (by having a smaller number of groups).

[0102] Fig. 8A and 8B The depiction categorizes metrology data from the overlay target 104 across samples into three groups. Fig. 8A is a diagram illustrating a graph of pupil center slope (eg, a selected reference metric) for a superimposed target 104 across a sample 106 in accordance with one or more embodiments of the present disclosure. Fig. 8A , the reference metric exhibits a clear radial pattern. As previously described herein, this pattern can be associated with a radial layer thickness pattern across the sample 106. However, it should be understood that Fig. 8A This is illustrative only and the reference metric may have any distribution across samples 106 .

[0103] Figure 8B 4 is a diagram illustrating a diagram of groups into which overlay objects 104 are classified (eg, based on step 406) according to one or more embodiments of the present disclosure. Figure 8B , groups may encompass spatial regions of the sample 106 having similar values ​​of a reference metric. In this particular example, a first group 802a includes a central region of the sample, a second group 802b includes a mid-radial region, and a third group 802c includes an edge region of the sample 106. However, it should be understood that the groupings do not necessarily need to be contiguous across the sample 106. For example, several distinct regions of the sample 106 may fall into a common group.

[0104] Reference again Figure 4 , additional steps of method 400 are now described according to one or more embodiments of the present disclosure.

[0105] In some embodiments, the method 400 includes a step 410 of generating corrected metrology data for the overlay targets 104 using associated reference images of one or more groups (e.g., at least the overlay targets 104 in the groups for which the reference images are generated). In some cases, all measured overlay targets 104 are classified into groups (e.g., in step 406) and reference images are generated for each of the groups (e.g., in step 408). In some embodiments, the method 400 includes a step 412 of generating overlay measurements for a plurality of overlay targets based on the corrected set of metrology data. In this manner, the provided overlay measurements associated with each overlay target 104 can be well corrected for errors such as, but not limited to, TIS based on the appropriate selection of reference images.

[0106] It is further contemplated herein that the reference group and associated reference images are applicable to additional overlay targets 104 on the same or different samples 106 .

[0107] In some embodiments, the set of overlay targets 104 used to classify the reference group and generate the associated reference image is only a subset of the overlay targets 104 on the sample 106 (e.g., as defined by the metrology recipe). In this case, metrology data generated for overlay targets 104 that are not in this original set may simply be classified based on the existing group. As an illustration, in the case of classification based on bins or ranges of the reference metric, values ​​of the reference metric may be generated for additional overlay targets 104 (e.g., as a corollary of step 404) such that the additional overlay targets 104 may be classified into groups based on associated bins or ranges of the reference metric (e.g., as a corollary of step 406). Subsequently, the associated reference image for the group may be used to generate a corrected overlay measurement for each of the additional overlay targets 104 (e.g., as a corollary of steps 410 and 412).

[0108] Similarly, the reference group and associated reference image are applicable to the additional overlay targets 104 on the additional sample 106. In this case, the value of the reference metric may be determined for the additional overlay targets 104 on the additional sample 106 to be classified based on the existing group. Subsequently, the associated reference image for the group may be used to generate corrected overlay measurements for each of the additional overlay targets 104. It should be noted that the metrology data associated with the additional overlay targets 104 may be classified based on the reference metric and does not necessarily need to be classified into the same group as the adjacent overlay targets 104 or the overlay targets 104 in the same position on the previous sample 106. In this way, the most suitable reference image may be used for each overlay target 104.

[0109] It is further contemplated herein that a reference group as defined at one point in time may need to be updated. For example, it may be the case that one or more manufacturing processes drift sufficiently over time that the effectiveness of one or more reference images for the associated group may decrease.

[0110] In a general sense, method 400 or portions thereof (eg, step 406 associated with defining a range of reference metrics corresponding to any of the groups and / or step 408 associated with generating an associated reference image for any of the groups) may be repeated or updated using any technique.

[0111] In some embodiments, the method 400 or a portion thereof may be performed at periodic time intervals. In some embodiments, the method 400 or a portion thereof may be performed after a trigger condition. Any suitable trigger condition may be used. For example, a non-limiting example of a trigger condition may be the number of times (in an absolute number or in a selected time window) that the superimposed target 104 at a particular location on the sample 106 is classified into a group different from the original sample 106. For example, it may be expected that the superimposed target 104 at a particular location on the sample 106 should have relatively consistent physical parameters and therefore relatively consistent reference metrics. Under this assumption, changes in the reference metrics associated with a particular location over time may indicate sample-to-sample process variations. At some point, such sample-to-sample variations may cause the reference image for one or more groups to be invalid (e.g., there may be a TIS error above a selected threshold). However, updating the classification of the group (e.g., the range of the reference image associated with any one of the groups) and / or the reference image for any one of the groups may again provide valid TIS error correction for all superimposed targets 104.

[0112] Reference again Figure 1A and 1B , additional aspects of the overlay metrology system 100 are described in more detail according to one or more embodiments of the present disclosure.

[0113] The one or more processors 110 of the controller 108 may include any processor or processing element known in the art. For the purposes of this disclosure, the term "processor" or "processing element" may be broadly defined to encompass any device having one or more processing or logic elements, such as one or more microprocessor devices, one or more application specific integrated circuit (ASIC) devices, one or more field programmable gate arrays (FPGAs), or one or more digital signal processors (DSPs). In this sense, the one or more processors 110 may include any device configured to execute algorithms and / or instructions, such as program instructions stored in a memory. In one embodiment, the one or more processors 110 may be embodied as a desktop computer, a mainframe computer system, a workstation, an image computer, a parallel processor, a networked computer, or any other computer system configured to execute a program configured to operate or operate in conjunction with the overlay metrology system 100, as described throughout this disclosure.

[0114] In addition, the different subsystems of the superposition metering system 100 may include processors or logic elements suitable for implementing at least a portion of the steps described in the present disclosure. Therefore, the above description should not be interpreted as limiting the embodiments of the present disclosure but is merely illustrative. In addition, the steps described throughout the present disclosure may be implemented by a single controller 108 or alternatively multiple controllers. In addition, the controller 108 may include one or more controllers housed in a common housing or multiple housings. In this way, any controller or combination of controllers can be individually packaged as a module suitable for integration into the superposition metering system 100.

[0115] The memory 112 may include any storage media known in the art suitable for storing program instructions executable by the associated one or more processors 132. For example, the memory 112 may include a non-transitory memory medium. As another example, the memory 112 may include, but is not limited to, a read-only memory (ROM), a random access memory (RAM), a magnetic or optical storage device (e.g., a disk), a tape, a solid-state drive, and the like. It should be further noted that the memory 112 may be housed in a common controller housing with the one or more processors 110. In one embodiment, the memory 112 may be remotely located relative to the physical location of the one or more processors 110 and the controller 108. For example, the one or more processors 110 of the controller 108 may access a remote memory (e.g., a server) that is accessible via a network (e.g., the Internet, an intranet, and the like).

[0116] Figure 1B is a simplified schematic diagram of the optical subsystem 102 according to one or more embodiments of the present disclosure.

[0117] In embodiments, the optical subsystem 102 may be configured according to a metrology recipe (e.g., an overlay recipe) to generate overlay measurements associated with an overlay target 104 having a design based on the metrology recipe. For example, the optical subsystem 102 may direct one or more illumination beams 114 from an illumination source 116 to the overlay target 104 on the sample 106 (e.g., via an illumination path 118), collect light or other radiation (referred to herein as measurement light 120) emitted from the overlay target 104 in response to the one or more illumination beams 114 (e.g., via a collection path 122), and generate metrology data from at least one detector 124 based on the measurement light 120. For example, the metrology data may be associated with diffraction of the illumination beam 114 from the overlay target 104. The illumination path 118 and the collection path 122 may further include additional optical elements at any suitable location, such as, but not limited to, a lens 126 and a beam modulation component 128 (e.g., a polarizer, a spectral filter, a spatial filter, a beam stop, an aperture, or the like). In some embodiments, as Figure 1B, the optical subsystem 102 includes one or more beam splitters 130 for allowing simultaneous illumination and collection through a common objective lens 132 or other focusing element. In some embodiments, the optical subsystem 102 includes a translation stage 134 including any number of linear or rotational actuators for fixing and / or positioning the sample 106.

[0118] The optical subsystem 102 may include one or more detectors 124 at any suitable location for collecting metrology data. For example, the optical subsystem 102 may include Figure 1B 136 (e.g., a plane conjugate with the sample 106). As another example, although not illustrated, the optical subsystem 102 may include at least one detector 124 at a pupil plane 138 (e.g., a diffraction plane corresponding to the angular distribution of light from the sample 106). Further, although not illustrated, the optical subsystem 102 may include multiple channels, each with a separate detector 124. In this manner, the optical subsystem 102 may provide multiple simultaneous measurements using multiple detectors 124 at any combination of the field plane 136 or the pupil plane 138. For example, the optical subsystem 102 may include one or more beam splitters (e.g., a non-polarizing beam splitter, a polarizing beam splitter, a dichroic mirror providing spectral selectivity, or the like) for splitting the measurement light 120 into multiple channels for detection. The optical subsystem 102 may further include optical components for modifying the properties of the measurement light 120 within each channel, such as, but not limited to, a polarizer, a polarization rotator, a spectral filter, a spatial filter, or a pupil filter (e.g., a beam stop or aperture in the pupil plane for blocking or passing selected diffraction orders).

[0119] The themes described herein sometimes illustrate different components contained in or connected to other components. It should be understood that such depicted architectures are merely exemplary, and in fact many other architectures that realize the same functionality can be implemented. In a conceptual sense, any arrangement of components to realize the same functionality is effectively "associated" so as to realize the desired functionality. Therefore, any two components combined to realize a specific functionality herein can be considered to be "associated" with each other so as to realize the desired functionality regardless of the architecture or intermediate components. Similarly, any two components so associated can also be considered to be "connected" or "coupled" to each other to realize the desired functionality, and any two components that can be so associated can also be considered to be "coupleable" to each other to realize the desired functionality. Specific examples that can be coupled include but are not limited to physically interactable and / or physically interactive components and / or wirelessly interactive and / or wirelessly interactive components and / or logically interactive and / or logically interactive components.

[0120] It is believed that the present disclosure and many of its attendant advantages will be understood from the foregoing description, and it will be appreciated that various changes may be made to the form, construction, and arrangement of components without departing from the disclosed subject matter or sacrificing all of its substantial advantages. The described form is merely illustrative, and it is intended that such changes be covered and included by the appended claims. Furthermore, it should be understood that the invention is defined by the appended claims.

Claims

1. A superposition metering system, comprising: A controller comprising one or more processors configured to execute program instructions causing the one or more processors to implement a metering recipe by: receiving metrology data associated with a plurality of overlay targets on one or more samples; generating a reference metric for at least some of the plurality of overlay targets based on the metrology data, wherein the reference metric for a particular one of the plurality of overlay targets is associated with one or more properties of the particular one of the plurality of overlay targets that contribute to an overlay error; classifying the plurality of overlay objects into one or more groups based on the reference metric calculated for the plurality of overlay objects; generating a reference image for at least some of the one or more groups, wherein the reference image mitigates tool-induced shift errors of the plurality of overlay targets; generating corrected metrology data using the reference images associated with at least some of the one or more groups; and Overlay measurements of the plurality of overlay targets are generated based on the corrected metrology data.

2. The stacking metrology system of claim 1, wherein the one or more groups include: Two or more groups.

3. The superimposed metrology system according to claim 1, wherein the metrology data comprises: Pupil plane image.

4. The superimposed metrology system of claim 3, wherein the reference metric comprises: Pupil center slope.

5. The overlay metrology system of claim 3, wherein generating the reference metrics for at least some of the plurality of overlay targets based on the metrology data comprises: calculating preliminary overlay values ​​for at least some of the plurality of overlay targets using different values ​​of selected pupil center positions based on the metrology data; and The reference metric is calculated based on a relationship between the selected pupil center position and the preliminary overlay value.

6. The stacking metrology system according to claim 1, wherein the metrology data comprises: Field plane image.

7. The overlay metrology system of claim 1 , wherein generating the reference metrics for at least some of the plurality of overlay targets based on the metrology data comprises: performing at least one of measuring or simulating one or more characteristics of at least one of the sample or at least some of the plurality of stacking targets at locations of at least some of the plurality of stacking targets; and The reference metric is generated based on the at least one of the measurement or the simulation.

8. The stacking metrology system of claim 7, wherein the one or more characteristics include: Layer thickness.

9. The superimposed metrology system of claim 7, wherein the one or more characteristics include: At least one of a sidewall angle, an asymmetry of one or more fabricated features, or a critical dimension.

10. The overlay metrology system of claim 1, wherein classifying the plurality of overlay targets into one or more groups based on the reference metrics calculated for the plurality of overlay targets comprises: dividing a range of the reference metric calculated for the plurality of superimposed objects into one or more bins; and The plurality of superimposed objects are classified into one or more groups based on the one or more bins.

11. The stacking metrology system of claim 1 , wherein the metrology data comprises first metrology data, wherein generating the reference image for a particular group of the one or more groups comprises: selecting one or more representative overlay objects from the plurality of overlay objects of the specific group; receiving second metrology data associated with the representative overlay target, wherein the second metrology data is generated with the representative overlay target rotated 180 degrees relative to corresponding data from the first metrology data; and The reference image for the particular group is generated based on the second metering data and the corresponding data from the first metering data.

12. The stacking metrology system of claim 11, wherein generating the reference image for the specific group based on the second metrology data and corresponding data from the first metrology data comprises: An image from the second metrology data is summed with a corresponding image from the first metrology data to produce the reference image for the particular group.

13. The stacking metrology system of claim 11, wherein generating the reference image for the specific group based on the second metrology data and a corresponding portion of the first metrology data comprises: summing an image from the second metrology data with a corresponding image from the first metrology data to produce a first reference image for the particular group; and generating a second reference image for the specific group by rotating the first reference image for the specific group by 180 degrees; and The first reference image for the particular group is divided by an average of the first and second reference images for the particular group to generate a corresponding reference image.

14. The superimposed metrology system of claim 1 , wherein generating corrected metrology data using the reference images associated with at least some of the one or more groups comprises: An image in the first metrology data is divided by a corresponding reference image associated with the one or more groups.

15. The overlay metering system of claim 1, wherein the one or more program instructions are further configured to cause the one or more processors to implement the metering recipe by: receiving additional metrology data associated with a plurality of additional overlay targets; generating the reference metrics for at least some of the plurality of additional overlay targets based on the additional metrology data; classifying the plurality of additional overlay targets into one or more groups based on the reference metric calculated for the plurality of additional overlay targets; generating additional corrected metrology data using the reference images associated with at least some of the one or more groups; and Overlay measurements of the plurality of additional overlay targets are generated based on the additional corrected metrology data.

16. The overlay metrology system of claim 15, wherein at least one of the one or more additional overlay targets is located on the one or more additional samples.

17. The overlay metrology system of claim 15, wherein at least one of the one or more additional overlay targets is located on the one or more samples.

18. The overlay metering system of claim 17, wherein the one or more program instructions are further configured to cause the one or more processors to implement the metering recipe by: monitoring a value of the reference metric associated with one or more locations of the one or more additional samples; and When a selected condition based on the monitored value of the reference metric is met: classifying the superimposed objects of the current sample into a new set of one or more groups based on the reference metric; and A reference image for the new set of one or more groups is generated.

19. The overlay metering system of claim 17, wherein the one or more program instructions are further configured to cause the one or more processors to implement the metering recipe by: monitoring a value of the reference metric associated with one or more locations of the one or more additional samples; and When a selected condition based on the monitored value of the reference metric is met, a new reference image for the one or more groups is generated.

20. The overlay metrology system of claim 1, wherein at least some of the plurality of overlay targets include one or more cells having overlapping structures associated with two or more patterning processes.

21. A superimposed metering system, comprising: An optical subsystem comprising: one or more optical elements configured to illuminate the sample with light from an illumination source according to a dosing recipe; and one or more optical elements configured to direct at least a portion of light from the sample to a detector located at a pupil plane; and a controller communicatively coupled to the optical subsystem, wherein the controller includes one or more processors configured to execute program instructions causing the one or more processors to implement the metrology recipe by: receiving metrology data associated with a plurality of superimposed targets on one or more samples from the optical subsystem; generating reference metrics for at least some of the plurality of stacking targets based on the metrology data, wherein the reference metrics are associated with one or more properties of the respective stacking targets that contribute to stacking error; classifying the plurality of overlay objects into one or more groups based on the reference metric calculated for the plurality of overlay objects; generating a reference image for at least some of the one or more groups, wherein the reference image mitigates tool-induced shift errors of the plurality of overlay targets; generating corrected metrology data using the reference images associated with at least some of the one or more groups; and Overlay measurements of the plurality of overlay targets are generated based on the corrected metrology data.

22. A superposition measurement method, comprising: receiving metrology data associated with a plurality of overlay targets on one or more samples; generating reference metrics for at least some of the plurality of stacking targets based on the metrology data, wherein the reference metrics are associated with one or more properties of the respective stacking targets that contribute to stacking error; classifying the plurality of overlay objects into one or more groups based on the reference metric calculated for the plurality of overlay objects; generating a reference image for at least some of the one or more groups, wherein the reference image mitigates tool-induced shift errors of the plurality of overlay targets; generating corrected metrology data using the reference images associated with at least some of the one or more groups; and Overlay measurements of the plurality of overlay targets are generated based on the corrected metrology data.

23. The method of claim 22, further comprising: receiving additional metrology data associated with a plurality of additional overlay targets; generating the reference metrics for at least some of the plurality of additional overlay targets based on the additional metrology data; classifying the plurality of additional overlay targets into one or more groups based on the reference metric calculated for the plurality of additional overlay targets; generating additional corrected metrology data using the reference images associated with at least some of the one or more groups; and Overlay measurements of the plurality of additional overlay targets are generated based on the additional corrected metrology data.

24. The method of claim 23, further comprising: monitoring a value of the reference metric associated with one or more locations of the one or more additional samples; and When a selected condition based on the monitored value of the reference metric is met: classifying the superimposed objects of the current sample into a new set of one or more groups based on the reference metric; and A reference image for the new set of one or more groups is generated.

25. The method of claim 23, further comprising: monitoring a value of the reference metric associated with one or more locations of the one or more additional samples; and When a selected condition based on the monitored value of the reference metric is met, a new reference image for the one or more groups is generated.

Images