Method and system for obtaining 3D profile of sample

Abstract

A method for obtaining a 3D profile of a sample is provided. The method includes receiving a first data set and a second data set related to a first set of structural parameters and a second set of structural parameters from a first metrology tool and a second metrology tool, wherein the first set of structural parameters and the second set of structural parameters collectively have at least one structural parameter; analyzing the first data set to obtain values of the first set of structural parameters; analyzing the second data set to obtain values of a second set of structural parameters, wherein values obtained from the analysis of the first data set for at least some of the common structural features are used to constrain the analysis of the second data set; and generating a 3D profile of the sample by combining values obtained in the analysis of the first data set and the second data set.

Description

Technical Field

[0001] The present disclosure generally relates to metrology systems and methods, and in particular, to using a combination of different scanning techniques for 3D profiling of structures and samples. Background Art

[0002] Scanning electron microscopy (SEM) is a widely used technique for characterizing and in particular for imaging various types of samples such as, but not limited to, metals, ceramics, biological samples, etc. One of the distinguishing features of SEM includes the high spatial resolution of SEM, enabling the characterization of the topography, microstructural features, and surface defects of samples at the nanoscale level. In addition to surface characterization, SEM can be used in combination with energy-dispersive X-ray spectroscopy (EDS) to obtain semi-quantitative elemental composition information. While analyzing the top few nanometers of the sample with high resolution, the depth information provided by SEM is very limited and poses significant challenges in the analysis of structures and materials. The lack of reliable depth information hinders the accurate analysis of materials, especially in cases where the 3D profile of the sample is critical, such as thin film characterization, quality control, and failure analysis. Therefore, characterization techniques with high depth sensitivity such as extreme ultraviolet (EUV) measurements and / or soft X-ray techniques such as X-ray reflectivity are employed to characterize the depth properties of the profile. However, these techniques are typically characterized by poor lateral resolution. Specifically, the resolution of conventional EUV or soft X-ray microscopy is limited by the numerical aperture of the X-ray lens, restricting the resolution of conventional EUV or soft X-ray microscopy to a range of approximately several tens of nanometers. This range may be too large to directly measure samples such as modern semiconductor devices. Summary of the Invention

[0003] According to one aspect of the presently disclosed subject matter, there is provided a method for obtaining a 3D profile of a sample, the method comprising:

[0004] Receiving a first data set provided by a first metrology tool and related to a first set of structural parameters of the sample, the first data set including a first set of measured signals and a first set of operating parameters;

[0005] Receiving a second data set provided by a second metrology tool and related to a second set of structural parameters of the sample, the second data set including a second set of measured signals and a second set of operating parameters, wherein the first set of structural parameters and the second set of structural parameters jointly include at least one structural parameter;

[0006] Performing a first analysis, the first analysis including analyzing the first data set to obtain values of the first set of structural parameters;

[0007] Perform a second analysis, the second analysis including analyzing a second data set to obtain values of a second set of structural parameters, wherein values obtained in the first analysis for at least some of the common structural parameters are used to constrain the second analysis; and

[0008] Generate a substantially complete 3D structural profile of the sample based on the values combined from the first and second analyses.

[0009] One of the first set of structural parameters and the second set of structural parameters may include surface structural parameters, wherein the other of the first set of structural parameters and the second set of structural parameters includes surface structural parameters and subsurface (i.e., internal) structural parameters.

[0010] The first metrology tool may be configured to measure the common structural parameters with higher accuracy than the second metrology tool.

[0011] The values obtained for the common structural parameters in the first analysis may have a very high accuracy.

[0012] According to some examples, if any further improvement in the accuracy of the value of the common structural parameter results in an improvement in the accuracy of the values obtained for other structural parameters such that the resulting improvement is not greater in measurement (e.g., within an acceptable measurement tolerance) than the improvement in the accuracy of the value of the common structural parameter, then the value of the common structural parameter may be considered to have "very high accuracy".

[0013] According to some examples, if the value of the common structural parameter is within about ±10% of the reference value, then the value of the common structural parameter may be considered to have "very high accuracy".

[0014] According to some examples, if, given the requirements of the application, the value of the common structural parameter is within the acceptable tolerance of the reference value, then the value of the common structural parameter may be considered to have "very high accuracy". The reference value may be a known value of the structural parameter, e.g., where the method is used to facilitate the detection of defects, structural anomalies, etc. in the sample.

[0015] Among other things, according to any of the above examples, determining when the value of the common structural parameter has a very high accuracy may include repeating some or all of the steps of the method. A person of ordinary skill in the relevant metrology art will have the required level of proficiency to perform the repetition of the steps and to determine, for example, based on the requirements of the task for which the method is being performed, when a very high accuracy has been obtained.

[0016] The method may further include iteratively repeating the first and second analyses, wherein the values obtained for at least some of the common structural parameters in each analysis are used to constrain subsequent analyses.

[0017] The method may further include applying a unified algorithm, wherein the first analysis and the second analysis are performed simultaneously.

[0018] The unified algorithm may further include using values obtained in the second analysis for at least some of the common structural parameters to constrain the first analysis.

[0019] The complete 3D structure profile may include structural parameters related to the width, depth, and / or shape of the elements of the surface layer; the thickness of the subsurface layer and / or structure; the sample thickness; the sample composition; the topography; and / or the phase.

[0020] One of the first metrology tool and the second metrology tool may include a scanning electron microscope (SEM), and / or the other of the first metrology tool and the second metrology tool may include an X-ray metrology tool and / or an extreme ultraviolet (EUV) metrology tool.

[0021] The SEM may be a critical dimension SEM.

[0022] The SEM may be configured to provide measured signals obtained from scans at different energies, measured signals obtained from scans at different angles, measured signals configured to facilitate SEM tomography, measured signals indicating voltage contrast, and / or measured signals including energy dispersive spectroscopy signals.

[0023] The first set of measured signals may be obtained from backscattered electrons, secondary electrons, and / or low-loss electrons.

[0024] The X-ray metrology tool and / or the EUV metrology tool may be configured to perform soft X-ray metrology, reflectometry, X-ray diffraction, coherent diffraction imaging, ptychography, EUV tomography, and / or X-ray tomography.

[0025] The EUV metrology tool may use light with a wavelength between 1 nm and 50 nm. The X-ray metrology tool may use light with a wavelength of 1 nm or less.

[0026] The sampling rate of the first metrology tool (e.g., SEM, X-ray metrology tool, EUV metrology tool, etc.) may be lower than the sampling rate of the second metrology tool, and the first metrology tool is configured to obtain a limited number of measurements, wherein the second analysis includes extrapolating a portion of the 3D structure profile sampled by the second metrology tool.

[0027] The sampling rate of the first metrology tool (e.g., SEM, X-ray metrology tool, EUV metrology tool, etc.) may be lower than the sampling rate of the second metrology tool, wherein the second analysis includes evaluating the similarity between the measurements obtained by the second metrology tool to detect structural anomalies in the complete 3D structure profile.

[0028] The first analysis and / or the second analysis may implement one or more algorithms based on physical simulation of a sample, phase retrieval, library search, gradient descent optimization, and / or machine learning based on measurements of previous samples.

[0029] The sample may be a semiconductor structure.

[0030] According to another aspect of the presently disclosed subject matter, there is provided a system for obtaining a 3D profile of a sample, the system being configured to execute code which is configured to perform the above method.

[0031] It should be understood that the above method, as well as the present specification and the appended claims herein, should not be construed as limited to being performed in the order presented. Ordinal numbers (first, second, etc.) are used for convenience only and are not intended to indicate sequence. For example, the first analysis and the second analysis may be performed in any suitable order.

[0032] Unless otherwise defined, all technical and / or scientific terms used herein have the same meaning as commonly understood by one of ordinary skill in the art to which this disclosure pertains. In case of conflict, the patent specification, including definitions, will prevail. As used herein, the indefinite article is understood to mean "at least one" or "one or more" unless clearly otherwise understood from the context. BRIEF DESCRIPTION OF THE DRAWINGS

[0033] Some embodiments of the present disclosure are described herein with reference to the drawings. The specification, together with the drawings, enables those of ordinary skill in the art to understand how to practice some embodiments. The drawings are for illustrative purposes and are not intended to show the structural details of the embodiments in more detail than is necessary for a basic understanding of the present disclosure. For clarity, some of the objects shown in the figures are not drawn to scale. Additionally, two different objects in the same figure may be drawn to different scales. In particular, some objects may be greatly enlarged in scale compared to other objects in the same figure.

[0034] Figure 1 A flowchart showing an example of an ordered joint processing method for obtaining a 3D profile of a sample according to some embodiments is shown;

[0035] Figure 2 A flowchart showing an example of an iterative joint processing method for obtaining a 3D profile of a sample according to some embodiments is shown; and

[0036] Figure 3 A flowchart showing an example of a unified joint processing method for obtaining a 3D profile of a sample according to some embodiments is shown. DETAILED DESCRIPTION

[0037] The principles, uses, and implementations taught herein can be better understood with reference to the accompanying description and drawings. After carefully reading the description and drawings herein, those skilled in the art will be able to implement the teachings herein without undue effort or experimentation.

[0038] In the following description, various aspects of the invention will be described. For purposes of explanation, specific details are set forth in order to provide a thorough understanding of the invention. However, it will also be apparent to those skilled in the art that the invention may be practiced without the specific details presented herein. Additionally, well-known features may be omitted or simplified in order not to obscure the invention.

[0039] As used herein, the term "about" may be used to specify a value of a quantity or parameter (e.g., the length of an element) as being within a continuous range of values that are close to (and including) a given (prescribed) value. According to some embodiments, "about" may specify that the value of a parameter is between 80% and 120% of a given value. For example, the statement "the length of the element is about equal to 1 m" is equivalent to the statement "the length of the element is between 0.8 m and 1.2 m". According to some embodiments, "about" may specify that the value of a parameter is between 90% and 110% of a given value. According to some embodiments, "about" may specify that the value of a parameter is between 95% and 105% of a given value.

[0040] As used herein, the terms "substantially" and "about" may be used interchangeably.

[0041] As used herein, according to some embodiments, the term "measured signal" may refer to any signal / data obtained from a first metrology tool and a second metrology tool. According to some embodiments, the measured signal may refer to raw data (i.e., unprocessed data) and / or processed data obtained from the first metrology tool and / or the second metrology tool.

[0042] According to some embodiments, one of the first metrology tool and the second metrology tool may include an SEM, and the other of the first metrology tool and the second metrology tool may include an EUV and / or X-ray metrology tool. According to some embodiments, the measured signal from the SEM may include an SEM image obtained from a sample / structure, etc. According to some embodiments, the SEM image may include a top view, a side view, a cross-sectional side view, or any other type of view of the sample / structure. According to some embodiments, the measured signal from the SEM may include an inclined SEM image (i.e., an image obtained at a non-normal angle), etc. According to some embodiments, an SEM image may be obtained based on using any available SEM-based detection method (such as but not limited to backscattered electrons, low-loss electrons, secondary electrons, etc. or any combination thereof). According to some embodiments, the measured signal from the SEM may include images obtained at several different energies, landing energy sweeps, etc. According to some embodiments, the measured signal from the SEM may include SEM tomography, etc. According to some embodiments, the measured signal from the SEM may include voltage contrast images, etc. According to some embodiments, the measured signal from the SEM may include energy-dispersive spectroscopy (EDS) data, etc.

[0043] According to some embodiments, the measured signal obtained from the EUV and / or X-ray metrology tool may include a reflectometer, such as an X-ray reflectometer (XRR), etc. Those skilled in the art will understand that reflection can be measured at different angles and / or different wavelengths. According to some embodiments, the measured signal obtained from the EUV and / or X-ray metrology tool may include diffraction measurements, such as X-ray diffraction (XRD), i.e., a measured signal including diffraction patterns at different angles or wavelengths, etc. According to some embodiments, the measured signal obtained from the EUV and / or X-ray metrology tool may include coherent diffraction imaging (CDI), etc. In other words, in some embodiments, the measured signal may include a signal obtained from an imaging method based on a diffraction pattern of a sample. According to some embodiments, the measured signal obtained from the EUV and / or X-ray metrology tool may include ptychography (i.e., an imaging method based on diffraction patterns of a sample obtained with several lateral displacements of the sample), etc. According to some embodiments, the measured signal obtained from the EUV and / or X-ray metrology tool may include EUV and / or X-ray tomography, etc. According to some embodiments, the measured signal obtained from the EUV and / or X-ray metrology tool may include a signal obtained from a soft X-ray tool / method, etc.

[0044] In some embodiments, the term "EUV" may refer to an approximate wavelength range of 1 nm to 50 nm. In some embodiments, the term "EUV" may refer to an approximate wavelength range of 1 nm to 40 nm, 1 nm to 30 nm, 10 nm to 50 nm, or 10 nm to 40 nm.

[0045] In some embodiments, the term "X-ray" may refer to an approximate wavelength range of about 1 nm or less. In some embodiments, the term "X-ray" may refer to an approximate wavelength range of about 0.1 nm to 1 nm, 0.5 nm to 1 nm, 0.8 nm to 1 nm, etc.

[0046] According to some embodiments, the disclosed method for obtaining a 3D profile of a sample can be used to characterize various types of samples, such as but not limited to semiconductor devices, semiconductor device patterns, chip architectures, electronic devices, etc. Those skilled in the art understand that the continuous reduction in the size of semiconductor devices requires improving the resolution and accuracy of metrology tools to characterize the parameters of interest of the sample (e.g., structural parameters / properties, such as film thickness, size, die alignment, presence of defects and / or anomalies, etc.). Therefore, it is usually necessary to use multiple metrology tools.

[0047] SEM- and EUV-based metrology tools are widely used to characterize samples. Those skilled in the art can understand that the measured signal obtained from an SEM (e.g., an SEM image) typically includes a top view or the topmost image of the structure of the sample with high resolution and high accuracy (e.g., about 1 nm or less), while providing poor depth information. For example, the sample under inspection may include a structure with multiple layers, where each of the multiple layers includes a different lateral pattern. Therefore, in such a case, the obtained SEM image may display different patterns as their superimposed pattern. Thus, it hinders the extraction of a complete, reliable, and accurate 3D profile of the sample.

[0048] Those skilled in the art can further understand that EUV and X-ray reflection techniques are characterized by poor lateral resolution (e.g., about 10 nm or higher) and are generally based on indirect measurements, where the light is not tightly focused and the diffraction pattern reflected from the sample is collected.

[0049] Currently, data analysis is generally based on inverse problem algorithms, which aim to determine the structure based on the received input light and the resulting scattering pattern. However, due to the ambiguity of the solution, it is usually difficult to solve such inverse problems. Therefore, the inverse problem algorithm may converge to an incorrect solution. In addition, in cases where multiple solutions lead to the same result, the inverse problem algorithm may not be able to distinguish between them. Additionally, the presence of measurement noise may exacerbate and increase the error probability of the inverse problem algorithm.

[0050] At least some of the methods according to the presently disclosed subject matter can facilitate combining the advantages of each of the metrology tools to obtain a substantially complete 3D profile of a sample, at least with respect to some physical parameters, and the obtained 3D profile has greater accuracy than any one of the metrology tools would provide on its own. At least some of the methods according to the presently disclosed subject matter can facilitate combining the direct measurement of individual structures (e.g., obtained from SEM) with the average properties of the regions of the sample being tested (e.g., obtained from X-ray / EUV), such that their combination results in obtaining improved, more precise, and complete structural parameters of the sample being inspected, such as a substantially complete 3D profile of the sample being inspected.

[0051] According to some embodiments, the substantially complete 3D profile can include structural information of the sample, such as the surface and subsurface properties of the sample. According to some embodiments, the substantially complete 3D profile can include a combination of the structural information of the sample and the semi-quantitative and / or quantitative elemental analysis of the sample, etc. According to some embodiments, the substantially complete 3D structural profile can include at least one of the following: the width, depth, and / or shape of the elements of the surface layer; the thickness of the subsurface layer and / or structure; the sample thickness; the sample composition; the topography; and the phase. As a non-limiting example, the 3D profile of the sample can include the thickness of the sample and / or the thickness of each of the layers or components of the sample, the sample / layer composition, and the topography of the sample. In some embodiments, the substantially complete 3D profile can include determining / identifying the presence of structural anomalies in the sample / structure. According to some embodiments, the anomalies can include any type of discontinuity; defects (e.g., point defects, 2D defects, 3D defects, etc.); the presence of contaminants; misalignment; improper variations of design values such as width, thickness, material composition, density, etc.; or any combination thereof, etc.

[0052] According to some embodiments, the substantially complete 3D profile can include comparing the sample / structure being inspected with a reference structure, such as but not limited to comparing with a desired sample / structure (e.g., without defects, anomalies, etc.).

[0053] Reference Figure 1 , Figure 1 FIG. 100 is a flow chart showing an example of an ordered joint processing method for obtaining a 3D profile of a sample. According to some embodiments, the joint processing method can be a computer-implemented method.

[0054] According to some embodiments, in step 102, the method can include receiving a first data set from a first metrology tool. According to some embodiments, the first data set can include a first set of measured signals from the first metrology tool. According to some embodiments, the first data set can include a first set of measured signals from the first metrology tool and a first set of operating parameters.

[0055] According to some embodiments, the first data set may include a set of raw (i.e., unprocessed) measured signals obtained from a first metrology tool. Additionally or alternatively, in some embodiments, the first data set may include a set of processed first measured signals.

[0056] According to some embodiments, the first data set may include single or multiple measurements obtained from the same sample. According to some embodiments in which the first metrology tool is a SEM, the first set of measured signals may include multiple top view images and / or multiple side view images and / or multiple tilted images of the same sample. According to some embodiments, the first set of measured signals obtained from a SEM may include multiple images obtained at different energies and / or different detection modes, etc. According to some embodiments, the first set of measured signals obtained from a SEM may include one or more EDS measurements, etc.

[0057] According to some embodiments, in step 104, the method may include receiving a second data set from a second metrology tool. According to some embodiments, the second data set may include a second set of measured signals and a second set of operation parameters from the second metrology tool.

[0058] According to some embodiments, the second data set may include a set of raw (i.e., unprocessed) measured signals obtained from the second metrology tool. Additionally or alternatively, in some embodiments, the second data set may include a set of processed second measured signals.

[0059] According to some embodiments, the second metrology tool is different from the first metrology tool. In some embodiments, one of the first metrology tool or the second metrology tool is configured to collect the structural properties of the subsurface layer and / or structure of the sample, and the other of the first metrology tool or the second metrology tool is configured to collect the structural properties of the surface layer of the sample.

[0060] Alternatively, in some embodiments, the method may include the steps of: receiving both a first data set from a first metrology tool and a second data set from a second metrology tool.

[0061] According to some embodiments, one of the first metrology tool or the second metrology tool may include a SEM, and the other of the first metrology tool or the second metrology tool may include at least one of an X-ray and an EUV tool.

[0062] According to some embodiments, the measured signals from a SEM tool (i.e., the first metrology tool or the second metrology tool) may include at least one or more of the following: SEM tomography; SEM voltage contrast images; landing energy sweeps; tilted SEM images; energy dispersive spectroscopy (EDS) data; and SEM images obtained by one or more of the following: backscattered electrons, secondary electrons, and low-loss electrons, etc.

[0063] According to some embodiments, the X-ray and EUV tools may include one or more of the following: soft X-rays, reflectometers, X-ray diffraction, coherent diffraction imaging, ptychography, EUV and / or X-ray tomography, etc.

[0064] According to some embodiments, the second data set may include a single measurement or multiple measurements obtained from the same sample. According to some embodiments, the second set of measured signals may include multiple EUV scattering patterns obtained from several incident angles and / or wavelengths.

[0065] According to some embodiments, in a case where the sampling rate of the first metrology tool may be lower than the sampling rate of the second metrology tool, the first metrology tool is configured to obtain a limited number of measurements to reduce the sampling time.

[0066] According to some embodiments, in step 106, the method may include analyzing the first data set to obtain a first portion of the structural parameters of the sample. According to some embodiments, the method may include analyzing the first data set to obtain a first portion of the structural parameters of the sample with high accuracy. In some embodiments, the first portion of the structural parameters of the sample with high accuracy may then be used to facilitate the analysis performed in step 108. According to some embodiments, high accuracy may include a case where the structural parameter to be measured is known to be within about 10% of the actual value of the structural parameter, about 5% of the actual value of the structural parameter, or about 1% of the actual value of the structural parameter, etc. According to some embodiments, high accuracy may include a case where the structural parameter to be measured is known to be within about 10% or less of the actual value of the structural parameter, about 5% or less of the actual value of the structural parameter, or about 1% or less of the actual value of the structural parameter, etc. As a non-limiting example, if the width or thickness of a certain feature of a semiconductor device is about 10 nm, estimating the width or thickness with high accuracy may include an estimation error of less than about 1 nm, less than about 0.5 nm, or less than about 0.1 nm. Those skilled in the art should understand that the required "sufficiently high" accuracy depends on the specific parameters and accuracy requirements.

[0067] According to some embodiments, the method may include analyzing a first data set to obtain a first portion of the structural parameters of a sample with a predefined accuracy or an initial accuracy, such that the first portion of the analyzed structural parameters is configured to facilitate the analysis performed in step 108, thereby obtaining a more accurate result. According to some embodiments, the predefined / initial accuracy may include an allowed estimation error, etc., and the allowed estimation error is one or more of the following: about 50%, about 40%, about 30%, about 20%, about 10%, about 5%, or about 1% of the actual value of the parameter to be measured. According to some embodiments, the predefined / initial accuracy may be within one or more ranges of the allowed estimation error: about 40% to 50%, about 30% to 40%, about 10% to 30%, about 20% to 30%, about 10% to 20%, etc.

[0068] As a non-limiting example, in the case where the first metrology tool includes an SEM, the first portion of the structural parameters may include the width and shape / morphology of the top layer of the sample with a high accuracy (such as, for example, about 10% or less of the actual value of the structural parameter), the width and shape of the buried layer with a lower accuracy (such as, for example, within the range of about 10% to 20%, 20% to 30% of the actual value of the structural parameter, etc.), and there may be no thickness data for each of the buried layers or there may be thickness data for each of the buried layers with a low accuracy (such as, for example, within the range of about 40% to 50% of the actual thickness).

[0069] As another non-limiting example, in the case where the first metrology tool includes an XRR tool, the first portion of the structural parameters may include the thickness of the layer with a high accuracy (such as, for example, within about 10% or less of the actual thickness), while the shape and / or width of the line at each of the layers may be obtained with a low accuracy (such as, for example, an estimation error within the range of about 10% to 30%, within the range of about 20% to 40%, etc.).

[0070] According to some embodiments, in step 108, the method may include analyzing a second data set to obtain a second portion of the structural parameters of the sample while locking the first portion of the structural parameters of the sample (obtained in step 106). Thus, in some embodiments, the accuracy of the structural parameters of the sample is improved.

[0071] According to some embodiments, the analysis performed in steps 106 and 108 may include or be based on physical simulation, library search-based algorithms, gradient descent optimization, machine learning methods based on measurements of previous samples (such as, for example, standard samples, previously tested samples, etc.), or any combination thereof. According to some embodiments, the machine learning-based methods may include linear or non-linear regression methods, estimators based on the correlation between structural parameters and detection signals or features in the detection signals, artificial neural networks, etc.

[0072] According to some embodiments, the analysis performed in steps 106 and 108 may include extrapolating a portion of the 3D structural profile sampled by the first metrology tool and / or the second metrology tool, etc. As a non-limiting example, in a case where the sampling rate of the first metrology tool is lower than that of the second metrology tool, the first metrology tool may be configured to obtain a limited number of measurements to reduce the sampling time. Thus, in some embodiments, the analysis may include extrapolating a portion of the 3D profile obtained / sampled by the second metrology tool. In some embodiments, the analysis may include evaluating the similarity between the measurements obtained by the second metrology tool to detect structural anomalies in the complete 3D profile of the sample. According to some embodiments, the anomalies may include any type of discontinuity; defects (e.g., point defects, 2D defects, etc.); the presence of contaminants; misalignment; improper variations in design values such as width, thickness, material composition, density, etc.; or any combination thereof.

[0073] According to some embodiments, in step 110, the method may include generating a substantially complete 3D profile of the sample. According to some embodiments, generating a substantially complete 3D profile of the sample may include combining a first portion and a second portion of the structural parameters.

[0074] According to some embodiments, in a case where the first metrology tool includes an SEM and the second metrology tool includes a diffraction-based EUV imaging tool (such as, for example, tomography or any other method), the analysis of the first data set from the first metrology tool can be used as an initial estimate for a phase retrieval algorithm applied to the first metrology tool. Thus, the accuracy of the generated 3D profile of the sample is improved.

[0075] Reference Figure 2 , Figure 2 FIG. 200 shows a flowchart of an example of an iterative joint processing method for obtaining a 3D profile of a sample. According to some embodiments, the iterative joint processing method may be a computer-implemented method.

[0076] According to some embodiments, in step 202, the method may include receiving a first data set from a first metrology tool. According to some embodiments, the first data set may include a first set of measured signals and a first set of operating parameters from the first metrology tool.

[0077] According to some embodiments, in step 204, the method may include receiving a second data set from a second metrology tool. According to some embodiments, the second data set may include a second set of measured signals and a second set of operating parameters from the first metrology tool.

[0078] In some embodiments, one of the first metrology tool or the second metrology tool is configured to collect structural properties of a subsurface layer and / or structure of a sample, and the other of the first metrology tool and the second metrology tool is configured to collect structural properties of a surface layer of the sample.

[0079] Alternatively or additionally, in some embodiments, the method may include receiving both a first data set and a second data set respectively from each of the first metrology tool and the second metrology tool.

[0080] According to some embodiments, one of the first metrology tool or the second metrology tool may include an SEM, and the other of the first metrology tool or the second metrology tool may include at least one of an X-ray and an EUV tool.

[0081] According to some embodiments, the measured signal from the SEM tool (i.e., the first metrology tool or the second metrology tool) may include at least one or more of the following: SEM tomography; SEM voltage contrast image; landing energy sweep; tilted SEM image; energy dispersive spectroscopy (EDS) data; and SEM images obtained by one or more of the following: backscattered electrons, secondary electrons, and low-loss electrons, etc.

[0082] According to some embodiments, the X-ray and EUV tools may include one or more of the following: soft X-ray, reflectometer, X-ray diffraction, coherent diffraction imaging, ptychography, EUV and / or X-ray tomography, etc. or any combination thereof.

[0083] According to some embodiments, and as depicted in flowchart 200, the method may include iteratively analyzing the first data and the second data to simultaneously match structural parameters from a first set of measured signals and a second set of measured signals. According to some embodiments, as elaborated in more detail in steps 206 to 208, the matching includes repeatedly feeding the parameters obtained from the first data set into the analysis of the second data, and vice versa.

[0084] According to some embodiments, in step 206, the method may include analyzing the first data set to obtain a first portion of the estimated structural parameters of the sample, thereby obtaining an updated first set of structural parameters. According to some embodiments, and as Figure 2 depicted, step 206 further includes feeding the updated first set of estimated structural parameters into step 208.

[0085] According to some embodiments, in step 208, the method may include analyzing a second data set to obtain a second portion of the structural parameters of the sample, thereby obtaining an updated set of second estimated structural parameters. According to some embodiments, the updated set of second estimated structural parameters may be refined / optimized at least in part based on the updated set of first estimated structural parameters. According to some embodiments, and as Figure 2 depicted, step 208 further includes feeding the updated set of second estimated structural parameters back to step 206. Then, in step 206, the analysis includes using the updated set of second estimated structural parameters and the first data set and / or (previous) updated set of first estimated structural parameters to refine the structural parameter estimation, thereby obtaining an (new) updated set of first estimated structural parameters. Then, in some embodiments, the (new / refined) updated set of first estimated structural parameters is fed back into step 208 to refine the updated set of second estimated structural parameters.

[0086] In some embodiments, steps 206 to 208 are performed iteratively until a sufficiently accurate parameter estimation is achieved. According to some embodiments, a sufficiently accurate parameter estimation may be achieved when a match is found between the estimated structural parameters of each of the metrology tools. In some embodiments, a sufficiently accurate parameter estimation may be achieved when the change in parameter estimation between consecutive iterative steps becomes small enough. As a non-limiting example, when the estimation change is less than about 1% of the estimated value.

[0087] According to some embodiments, the analysis performed in steps 206 to 208 may include or be based on physical simulations, library search-based algorithms, gradient descent optimization, machine learning methods based on measurements of previous samples (e.g., standard samples, previously tested samples, etc.), or any combination thereof. According to some embodiments, machine learning-based methods may include linear or nonlinear regression methods, estimators based on the correlation between structural parameters and detection signals or features in the detection signals, artificial neural networks, etc.

[0088] According to some embodiments, the analysis performed in steps 206 to 208 may include extrapolating a part of the 3D structure profile, etc. As a non-limiting example, in a case where the sampling rate of the first metrology tool is lower than that of the second metrology tool, the first metrology tool may be configured to obtain a limited number of measurements to reduce the sampling time. Thus, in some embodiments, the analysis may include extrapolating a part of the 3D profile obtained / sampled by the second metrology tool. In some embodiments, the analysis may include evaluating the similarity between the measurements obtained by the second metrology tool to detect structural anomalies in the complete 3D profile of the sample. According to some embodiments, the anomalies may include any type of discontinuity; defects (e.g., point defects, 2D defects, etc.); the presence of contaminants; misalignment; improper variations in design values such as width, thickness, material composition, density, etc.; or any combination thereof.

[0089] According to some embodiments, in step 210, the method may include obtaining a substantially complete 3D profile of the sample by combining the estimated structural parameters. According to some embodiments, the iterative refinement of the estimated structural parameters in steps 206 to 208 improves the accuracy of the estimated structural parameters, thereby improving the generated 3D profile of the sample.

[0090] Reference Figure 3 , Figure 3 FIG. 300 is a flow chart showing an example of a unified joint processing method for obtaining a 3D profile of a sample. According to some embodiments, the unified joint processing method may be a computer-implemented method.

[0091] According to some embodiments, in step 302, the method may include receiving a first data set from a first metrology tool. According to some embodiments, the first data set may include a first set of measured signals and a first set of operating parameters from the first metrology tool.

[0092] According to some embodiments, in step 304, the method may include receiving a second data set from a second metrology tool. According to some embodiments, the second data set may include a second set of measured signals and a second set of operating parameters from the first metrology tool.

[0093] In some embodiments, one of the first metrology tool or the second metrology tool is configured to collect the structural properties of the subsurface layer and / or structure of the sample, and the other of the first metrology tool and the second metrology tool is configured to collect the structural properties of the surface layer of the sample.

[0094] According to some embodiments, one of the first metrology tool or the second metrology tool may include an SEM, and the other of the first metrology tool or the second metrology tool may include at least one of an X-ray and an EUV tool.

[0095] According to some embodiments, the X-ray and EUV tools may include one or more of the following: soft X-rays, X-ray reflectometers, X-ray diffraction, coherent diffraction imaging, ptychography, EUV and / or X-ray tomography, etc. or any combination thereof.

[0096] Alternatively or additionally, in some embodiments, the method may include receiving both a first data set and a second data set from each of a first metrology tool and a second metrology tool, respectively.

[0097] According to some embodiments, in step 306, the method may include feeding the first data set and the second data set into a unified algorithm.

[0098] According to some embodiments, the unified algorithm may include or be based on physical simulations, library search-based algorithms, gradient descent optimization, machine learning methods based on measurements of previous samples (e.g., standard samples, previously tested samples, etc.), etc. or any combination thereof. According to some embodiments, the machine learning-based methods may include linear or non-linear regression methods, estimators based on the correlation between structural parameters and detected signals or features in the detected signals, artificial neural networks, etc.

[0099] According to some embodiments, in step 308, the method may include analyzing the first data set and the second data set as a unified data set by applying the unified algorithm to obtain the estimated structural parameters of the sample. According to some embodiments, the estimated structural parameters are configured to match both the first data set and the second data set simultaneously. According to some embodiments, analyzing the first data set and the second data set as a unified data set may not separate the data obtained from the first metrology tool and / or the second metrology tool.

[0100] According to some embodiments, the analysis performed in step 308 may include extrapolating a portion of the 3D structure profile, etc. As a non-limiting example, in a case where the sampling rate of the first metrology tool is lower than the sampling rate of the second metrology tool, the first metrology tool may be configured to obtain a limited number of measurements to reduce the sampling time. Thus, in some embodiments, the analysis may include extrapolating a portion of the 3D profile obtained / sampled by the second metrology tool. In some embodiments, the analysis may include evaluating the similarity between the measurements obtained by the second metrology tool to detect structural anomalies in the complete 3D profile of the sample. According to some embodiments, the anomalies may include any type of discontinuity; defects (e.g., point defects, 2D defects, 3D defects, etc.); the presence of contaminants; misalignment; improper variations in design values such as width, thickness, material composition, density, etc.; or any combination thereof.

[0101] According to some embodiments, in step 310, the method may include outputting the 3D profile of the sample.

[0102] In the description and claims of the present application, the words "comprising" and "having" and their forms are not limited to the members in the lists that may be associated with these words.

[0103] According to one aspect of some embodiments, a system for 3D profile analysis of a sample / structure is disclosed herein. According to some embodiments, the system is configured to execute code that is configured to perform an ordered joint processing method for obtaining a 3D profile of the sample / structure. According to some embodiments, the system is configured to execute code that is configured to receive a first data set, the first data set including a first set of measured signals and a first set of operating parameters from a first metrology tool. According to some embodiments, the system is configured to execute code that is further configured to receive a second data set, the second data set including a second set of measured signals and a second set of operating parameters from a second metrology tool, the second metrology tool being different from the first metrology tool, and wherein one of the first metrology tool or the second metrology tool is configured to collect the structural properties of a subsurface layer of the sample and / or the structure, and the other of the first metrology tool and the second metrology tool is configured to collect the structural properties of the surface layer of the sample.

[0104] According to some embodiments of the system, one of the first metrology tool and the second metrology tool may include an SEM, and the other of the first metrology tool and the second metrology tool may include EUV and / or soft X-rays. According to some embodiments, the first set of operating parameters or the second set of operating parameters may include SEM operating parameters such as, but not limited to, acceleration voltage, beam current, vacuum level, focal length, electron gun performance parameters, etc. or any combination thereof. According to some embodiments, the first set of measured signals or the second set of measured signals may include secondary electron signals and / or backscattered electron signals obtained from the SEM, etc. As a non-limiting example, the first metrology tool may include an SEM, wherein the first set of measured signals may include SEM images such as top view images and cross-sectional images, EDS measurements, etc. or combinations thereof.

[0105] According to some embodiments, the wavelength range of the EUV tool may be from about 1 nm to about 50 nm, from about 10 nm to about 50 nm, from about 1 nm to about 40 nm, from about 1 nm to about 30 nm, etc.

[0106] According to some embodiments, the wavelength range of the X-ray tool may be about 1 nm or less. According to some embodiments, the wavelength range of the X-ray tool may be from about 0.1 nm to about 1 nm.

[0107] According to some embodiments, the first set of measured signals and / or the second set of measured signals may include unprocessed (i.e., raw) data obtained from the first metrology tool and / or the second metrology tool. Alternatively or additionally, the first set of measured signals and / or the second set of measured signals may include processed data obtained from the first metrology tool and / or the second metrology tool. According to some embodiments, the processed data may include analyzed results and / or processed signals obtained from the corresponding metrology tool, and the processed signals enable additional features to be extracted therefrom.

[0108] According to some embodiments of the system, the system is configured to execute code that is further configured to analyze a first data set to obtain a first portion of the structural parameters of a sample with high accuracy; analyze a second data set while locking the first portion of the structural parameters to obtain a second portion of the structural parameters; and generate a substantially complete 3D structural profile of the sample based on combining the first portion and the second portion of the structural parameters.

[0109] According to some embodiments of the system, the system may include a processing unit. According to some embodiments, the processing unit may include at least one processor. According to some embodiments, the system may include non-volatile memory, etc. According to some embodiments, the system may include computer hardware, software, etc. or a combination thereof.

[0110] According to some embodiments of the system, the system may be configured to execute code that is configured to output an image depicting the 3D profile of the sample / structure. Additionally or alternatively, in some embodiments, the system may be configured to execute code that is configured to output analyzed values of the 3D profile of the sample / structure, such as but not limited to: the width, depth, and / or shape of the elements at the top layer; the thickness of the subsurface layer and / or structure; the sample thickness; the sample composition; the topography; one or more phases present in the structure / sample; etc.; or any combination thereof.

[0111] According to some embodiments, the system may be configured to execute code at least in part based on physical simulation. According to some embodiments, the system may be configured to execute code at least in part based on one or more library search-based algorithms. According to some embodiments, the system may be configured to execute code at least in part based on gradient descent-based methods such as, for example, stochastic gradient descent. According to some embodiments, the system may be configured to execute code at least in part based on signal processing methods. According to some embodiments, the system may be configured to execute code at least in part based on machine learning methods (e.g., based on measurements of previous samples (e.g., samples previously tested by the current system and / or different systems), standard / reference samples, etc.). Non-limiting examples of related machine learning-based methods may include linear or non-linear regression methods, estimators based on the correlation between structural parameters and detected signals or features in detected signals, artificial neural networks, etc.

[0112] According to some embodiments, in a case where the sampling rate of a first metrology tool may be lower than the sampling rate of a second metrology tool, the system may be further configured to execute code configured to output instructions for obtaining a limited number of measurements to the first metrology tool. In some embodiments, the system may be configured to execute code configured to extrapolate a portion of a 3D structure profile sampled by the second metrology tool.

[0113] According to some embodiments, in a case where the sampling rate of a first metrology tool may be lower than the sampling rate of a second metrology tool, the system may be further configured to execute code configured to evaluate the similarity between measurements obtained by the second metrology tool to detect structural anomalies in a substantially complete 3D structure profile.

[0114] According to one aspect of some embodiments, an iterative joint processing system for 3D profile analysis of a sample / structure is disclosed herein. According to some embodiments, the system is configured to execute code configured to perform an iterative joint processing method for obtaining a 3D profile of a sample / structure. According to some embodiments, the system is configured to execute code configured to receive a first data set including a first set of measured signals and a first set of operating parameters from a first metrology tool. According to some embodiments, the system is configured to execute code further configured to receive a second data set including a second set of measured signals and a second set of operating parameters from a second metrology tool different from the first metrology tool.

[0115] According to some embodiments, the system is configured to execute code that is further configured to iteratively analyze a first data set and a second data set to simultaneously match structural parameters from a first set of measured signals and a second set of measured signals. According to some embodiments, the matching includes repeatedly feeding parameters obtained from the first data set into the analysis of the second data, and vice versa, to determine an updated set of estimated structural parameters of the sample until a substantially complete 3D structural profile of the sample is obtained. According to some embodiments, each of the updated sets of estimated structural parameters of the sample (at each iteration) includes an improved / refined value of the estimated structural parameter, thereby improving the accuracy of the substantially complete 3D structural profile.

[0116] According to some embodiments, the system is configured to execute code that is further configured to output a substantially complete 3D profile of the sample / structure.

[0117] According to some embodiments of the system, one of the first metrology tool and the second metrology tool may include an SEM, and the other of the first metrology tool and the second metrology tool may include EUV and / or soft X-ray. According to some embodiments, the first set of operating parameters or the second set of operating parameters may include SEM operating parameters such as, but not limited to, acceleration voltage, beam current, vacuum level, focal length, electron gun performance parameters, etc. or any combination thereof. According to some embodiments, the first set of measured signals or the second set of measured signals may include secondary electron signals and / or backscattered electron signals obtained from an SEM, etc. As a non-limiting example, the first metrology tool may include an SEM, wherein the first set of measured signals may include SEM images such as top view images and cross-sectional images, EDS measurements, etc. or a combination thereof.

[0118] According to some embodiments, the wavelength range of the EUV tool may be from about 1 nm to about 50 nm, from about 10 nm to about 50 nm, from about 1 nm to about 40 nm, from about 1 nm to about 30 nm, etc.

[0119] According to some embodiments, the wavelength range of the X-ray tool may be about 1 nm or less. According to some embodiments, the wavelength range of the X-ray tool may be from about 0.1 nm to about 1 nm.

[0120] According to some embodiments, the first set of measured signals and / or the second set of measured signals may include unprocessed (i.e., raw) data obtained from the first metrology tool and / or the second metrology tool. Alternatively or additionally, the first set of measured signals and / or the second set of measured signals may include processed data obtained from the first metrology tool and / or the second metrology tool. According to some embodiments, the processed data may include analyzed results and / or processed signals obtained from the corresponding metrology tool, etc., and the processed signals enable additional features to be extracted therefrom.

[0121] According to some embodiments of the system, the system may include a processing unit. According to some embodiments, the processing unit may include at least one processor. According to some embodiments, the system may include non-volatile memory, etc. According to some embodiments, the system may include computer hardware, software, etc. or a combination thereof.

[0122] According to some embodiments of the system, the system may be configured to execute code that is configured to output an image depicting a 3D profile of a sample / structure. Additionally or alternatively, in some embodiments, the system may be configured to execute code that is configured to output an analyzed value of the 3D profile of the sample / structure, such as but not limited to: the width, depth, and / or shape of elements at the top layer; the thickness of subsurface layers and / or structures; the sample thickness; the sample composition; the topography; one or more phases present in the structure / sample; or any combination thereof.

[0123] According to some embodiments, the system may be configured to execute code at least in part based on physical simulation. According to some embodiments, the system may be configured to execute code at least in part based on one or more library search-based algorithms. According to some embodiments, the system may be configured to execute code at least in part based on gradient descent-based methods such as, for example, stochastic gradient descent. According to some embodiments, the system may be configured to execute code at least in part based on signal processing methods. According to some embodiments, the system may be configured to execute code at least in part based on machine learning methods (e.g., based on measurements of previous samples (e.g., samples previously tested by the current system and / or different systems), standard / reference samples, etc.). According to some embodiments, machine learning-based methods may include linear or nonlinear regression methods, estimators based on the correlation between structural parameters and detected signals or features in detected signals, artificial neural networks, etc.

[0124] According to some embodiments, in a case where the sampling rate of a first metrology tool may be lower than the sampling rate of a second metrology tool, the system may be further configured to execute code that is configured to output instructions for obtaining a limited number of measurements to the first metrology tool. In some embodiments, the system may be configured to execute code that is configured to extrapolate a portion of the 3D structure profile sampled by the second metrology tool.

[0125] According to some embodiments, in a case where the sampling rate of a first metrology tool may be lower than the sampling rate of a second metrology tool, the system may be further configured to execute code that is configured to evaluate the similarity between measurements obtained by the second metrology tool to detect structural anomalies in a substantially complete 3D structure profile.

[0126] In accordance with one aspect of some embodiments, a unified joint processing system for obtaining a 3D profile of a sample is provided. According to some embodiments, the unified processing system is configured to execute code that is configured to: receive a first data set including a first set of measured signals and a first set of operation parameters from a first metrology tool; receive a second data set including a second set of measured signals and a second set of operation parameters from a second metrology tool, wherein one of the first metrology tool or the second metrology tool includes a SEM, and wherein the other of the first metrology tool or the second metrology tool includes at least one of an X-ray and an EUV tool; feed the first data set and the second data set into a unified algorithm; and analyze the first data set and the second data set as a unified data set by applying the unified algorithm to obtain an estimated structural parameter of the sample, the estimated structural parameter being configured to simultaneously match the first data set and the second data set. According to some embodiments, the unified algorithm is at least partially based on physical simulation, library search-based algorithms, gradient descent optimization, machine learning methods based on measurements of previous samples, or any combination thereof.

[0127] According to some embodiments, the machine learning-based method may include linear or non-linear regression methods, estimators based on the correlation between the structural parameter and the detected signal or features in the detected signal, artificial neural networks, and the like.

[0128] Unless otherwise defined, all technical and / or scientific terms used herein have the same meaning as commonly understood by one of ordinary skill in the art to which this disclosure pertains. In case of conflict, the patent specification including definitions will prevail. As used herein, the indefinite articles "a" and "an" mean "at least one" or "one or more components and / or methods" as set forth herein. Other embodiments may be practiced and the embodiments may be implemented in various ways.

[0129] The wording and terminology used herein are for the purpose of description and should not be regarded as restrictive. The citation or identification of any reference in this application should not be construed as an admission that such reference is available as prior art to this disclosure. The section headings used herein are for ease of understanding the specification and should not be construed as necessary limitations.

Claims

1. A method for obtaining a 3D profile of a sample, the method comprising: receiving a first data set provided by a first metrology tool and related to a first set of structural parameters of the sample, the first data set comprising a first measured signal set and a first operating parameter set; receiving a second data set provided by a second metrology tool and relating to a second set of structural parameters of the sample, the second data set comprising a second set of measured signals and a second set of operating parameters, wherein the first set and the second set collectively include at least one structural parameter; performing a first analysis, the first analysis comprising analyzing the first data set to obtain values ​​for the first set of structural parameters; performing a second analysis, the second analysis comprising analyzing the second data set to obtain values ​​for the second set of structural parameters, wherein the values ​​obtained in the first analysis for at least some of the common structural parameters are used to constrain the second analysis; as well as A substantially complete 3D structural profile of the sample is generated based on combining the values ​​obtained in the first analysis and the second analysis.

2. The method of claim 1, wherein one of the first structure parameter group and the second structure parameter group comprises surface structure parameters, and the other of the first structure parameter group and the second structure parameter group comprises surface structure parameters and sub-surface structure parameters. 3 . The method of claim 1 , wherein the first metrology tool is configured to measure the common structural parameter with a higher accuracy than the second metrology tool. The method of claim 1 , wherein the values ​​obtained in the first analysis for the common structural parameters have a high degree of accuracy.

5. The method of claim 1, further comprising iteratively repeating the first analysis and the second analysis, wherein values ​​obtained in each analysis for at least some of the common structural parameters are used to constrain subsequent analyses.

6. The method of claim 1 , further comprising applying a unified algorithm, wherein the first analysis and the second analysis are performed simultaneously, the unified algorithm optionally further comprising constraining the first analysis using values ​​obtained in the second analysis for at least some of the common structural parameters.

7. The method of any one of the preceding claims, wherein one of the first metrology tool and the second metrology tool comprises a scanning electron microscope (SEM) and / or the other of the first metrology tool and the second metrology tool comprises an X-ray metrology tool and / or an extreme ultraviolet (EUV) metrology tool.

8. The method of claim 1 , the first metrology tool having a sampling rate lower than a sampling rate of the second metrology tool, the first metrology tool being configured to obtain a limited number of measurements, wherein the second analysis comprises extrapolating a portion of the 3D structure profile sampled by the second metrology tool.

9. The method of claim 1, the first metrology tool having a sampling rate lower than the sampling rate of the second metrology tool, wherein the second analysis comprises evaluating similarities between measurements obtained by the second metrology tool to detect structural anomalies in the complete 3D structural profile.

10. A system for obtaining a 3D profile of a sample, the system being configured to execute code configured to perform the method according to any one of the preceding claims.

Images