Multi-step process inspection method

Abstract

An image analysis method for identifying features in an image of a portion of an array of features formed by a multi-step process, the method comprising: analyzing changes in features visible in the image; and associating the features of the image with steps of the multi-step process based at least in part on the results of the analysis.

Description

[0001] CROSS-REFERENCE TO RELATED APPLICATIONS

[0002] This application claims priority to EP application 20156290.7 filed on February 10, 2020, the entire contents of which are incorporated herein by reference. Technical Field

[0003] The present disclosure relates to an inspection method, and in particular to an inspection method for manufacturing a device using a lithography apparatus. Background Art

[0004] A photolithographic apparatus is a machine that applies a desired pattern to a substrate, typically to a target portion of the substrate. A photolithographic apparatus can be used, for example, to manufacture an integrated circuit (IC). In this example, a pattern forming device (which may alternatively be referred to as a mask or reticle) can be used to generate a circuit pattern to be formed on a separate layer of the IC. The pattern can be transferred to a target portion (e.g., including a portion of a die, a die, or multiple dies) on a substrate (e.g., a silicon wafer). Typically, the pattern is transferred via imaging onto a radiation-sensitive material (resist) layer disposed on the substrate. Generally speaking, a single substrate will contain a network of adjacent target portions that have been continuously patterned.

[0005] To meet the ongoing demand for reduced feature sizes in photolithography, various processes have been proposed that include multiple steps for producing individual layers with smaller dimensions or pitches than can be formed in a single optical patterning step. Examples of such processes include lithography-etch-lithography-etch (LELE), self-aligned double patterning (SADP), and self-aligned quadruple patterning (SAQP). These processes present difficulties for inspection and metrology processes. Summary of the Invention

[0006] The present disclosure aims to provide improved metrology methods, for example, for use in lithographic equipment manufacturing processes.

[0007] According to one embodiment, there is provided an image analysis method for identifying features in an image of a portion of an array of features formed by a multi-step process, the method comprising:

[0008] Analyzing changes in features visible in the image; and

[0009] Based at least in part on the results of the analysis, features of the image are associated with steps of the multi-step process.

[0010] According to one embodiment, a device manufacturing method is provided, comprising:

[0011] forming an array of features on a substrate using a multi-step process;

[0012] obtaining an image of a portion of the array;

[0013] analyzing images as described above to associate features with steps of a multi-step process;

[0014] detecting defects in features of the array; and

[0015] Based on the association of features with steps and the defects detected, remedial actions are performed.

[0016] According to one embodiment, there is provided an image analysis apparatus for identifying features in an image of a portion of an array of features formed by a multi-step process, the apparatus comprising:

[0017] an image analysis module configured to analyze changes in features visible in the image; and

[0018] An association module is configured to associate features of the image with steps of the multi-step process based at least in part on the analysis results.

[0019] According to one embodiment, there is provided a method of analyzing an image of a portion of an array of features formed by a self-aligned quadruple patterning process, the method comprising:

[0020] Identify multiple features in an image;

[0021] Assigning a feature value to each identified feature, the feature value representing a change in position or shape of the feature;

[0022] grouping the features into a first group, a second group, a third group, and a fourth group, each group comprising a set of aligned features, the first group, the second group, the third group, and the fourth group being adjacent to each other in that order;

[0023] determining a first correlation value between a change in a characteristic value of the first group and a change in a characteristic value of the second group;

[0024] determining a second correlation value between changes in the characteristic value of the second group and changes in the characteristic value of the third group;

[0025] If the first correlation value is higher than the second correlation value, the first and second groups are associated with the first spacer of the self-aligned quadruple patterning process, otherwise the second and third groups are associated with the first spacer of the self-aligned quadruple patterning process. BRIEF DESCRIPTION OF THE DRAWINGS

[0026] Embodiments will now be described by way of example with reference to the accompanying drawings, in which

[0027] Figure 1Depicts a photolithography apparatus that, along with other apparatus, forms a production facility for semiconductor devices;

[0028] Figures 2A to 2F The steps in the self-aligned quadruple patterning process are depicted;

[0029] Figures 3A to 3D The effects of errors in the self-aligned quadruple patterning process are depicted;

[0030] Figure 4 An example of an image depicting a portion of an array of features;

[0031] Figure 5 An example of an image depicting a portion of an array of features including a defect;

[0032] Figure 6 depicts an example of the results of a process for contouring and determining the centroid of a feature;

[0033] Figure 7 Describes grouping features into columns; and

[0034] Figure 8 is a flow chart of a device manufacturing method. DETAILED DESCRIPTION

[0035] Electronic devices consist of circuits formed on a piece of silicon called a substrate. Many circuits can be formed together on the same piece of silicon and are called integrated circuits, or ICs. The size of these circuits has decreased dramatically, allowing more to fit onto a substrate. For example, the IC chip in a smartphone can be as small as a thumbnail but may contain over 2 billion transistors, each less than 1 / 1000 the size of a human hair.

[0036] Manufacturing these extremely small ICs is a complex, time-consuming, and expensive process, typically involving hundreds of individual steps. An error in even a single step can result in a defect in the finished IC, rendering it useless. Therefore, one goal of the manufacturing process is to avoid such defects in order to maximize the number of functional ICs produced during the process—in other words, to improve the overall yield of the process.

[0037] One component of increasing yield is monitoring the chipmaking process to ensure that it produces a sufficient number of functional integrated circuits. One way to monitor the process is to inspect the chip circuit structures at various stages of their formation. Inspections can be performed using scanning electron microscopes (SEMs), optical inspection systems, and the like. These systems can be used to image these structures, essentially taking "photos" of the structures of the wafer, with the SEM being able to image the smallest of these structures. The images can be used to determine whether the structures were formed in the proper manner. If the structure has defects, the process can be adjusted so that the defects are less likely to recur.

[0038] A key step in the photolithographic manufacturing process for chips is the optical step where an image of a feature is projected onto a substrate (wafer), but there is a limit to how small features can be formed in this way. To make smaller features, there are processes that use chemical steps to build several (for example, four) smaller features based on the one larger feature produced by the optical step. In one example, one step produces every other feature, while a second step produces features in between. An example is the wiring used in large memory chips. These ICs typically have large areas filled with a regular array of wiring, and if one of the features has a defect it can be difficult to determine which step of the process formed the defective feature. Since all the features should be identical, it may be necessary to count the features from the edge of the array in order to determine in which step a particular feature was formed.

[0039] This disclosure proposes a technique for identifying the step in a multi-step process in which a given feature was formed by examining small variations in the shape and / or position of features in an array. The inventors have determined that the nature of these variations, which are not significant enough to affect device functionality or be considered defects, depends on the process step in which they were formed. In one example, a process step produces similar variations in all features produced in that step, allowing the identification of the feature produced in that step by examining correlations between the variations in the features.

[0040] Before describing the embodiments in detail, it is instructive to present an example environment in which the techniques disclosed herein may be implemented.

[0041] Figure 1The diagram illustrates a typical layout of a semiconductor production facility. A lithographic apparatus 100 applies a desired pattern to a substrate. A lithographic apparatus is used, for example, to manufacture integrated circuits (ICs). In this example, a pattern forming device MA, which may alternatively be referred to as a mask or reticle, includes a circuit pattern of features (often referred to as "product features") to be formed on a separate layer of the IC. Via exposure 104 of the pattern forming device on a layer of radiation-sensitive material (resist) provided on the substrate, the pattern is transferred to a target portion (e.g., including a portion of a die, a die, or several dies) on a substrate 'W' (e.g., a silicon wafer). Generally speaking, a single substrate will contain a network of adjacent target portions that are patterned continuously.

[0042] A known lithographic apparatus irradiates each target portion by illuminating a patterning device while simultaneously positioning the target portion of the substrate at an image position on the patterning device. The irradiated target portion of the substrate is referred to as an "exposure field," or simply a "field." The layout of the fields on the substrate is typically a network of adjacent rectangles or other shapes aligned according to a two-dimensional Cartesian coordinate system (e.g., aligned along the X-axis and the Y-axis, with the two axes being orthogonal to each other).

[0043] The requirement for the lithography apparatus is to accurately reproduce the desired pattern onto the substrate. The position and size of the applied product features need to be within a certain tolerance range. Position errors may cause overlay errors (commonly referred to as "overlap"). Overlap is the error in placing a first product feature in a first layer relative to a second product feature in a second layer. The lithography apparatus reduces overlay errors by accurately aligning each wafer with a reference before pattern formation. This is done by measuring the position of alignment marks applied to the substrate. Based on the alignment measurement, the substrate position is controlled during the pattern formation process to prevent overlay errors that are out of tolerance. The alignment mark is usually generated as part of the product image, forming a reference for measuring overlay. Alternatively, the alignment mark of a previously formed layer can be used.

[0044] When the applied dose associated with exposure 104 is not within specifications, errors in the critical dimension (CD) of the product feature may occur. To this end, the lithographic apparatus 100 must be able to accurately control the radiation dose applied to the substrate. CD errors may also occur when the substrate is not correctly positioned relative to the focal plane associated with the pattern image. Focus position errors are generally associated with non-planarity of the substrate surface. The lithographic apparatus reduces these focus position errors by measuring the substrate surface topography using a height sensor prior to patterning. Substrate height correction is applied during subsequent patterning to ensure that the patterning device is correctly imaged (focused) onto the substrate.

[0045] To verify the overlay error and CD error associated with the lithography process, the patterned substrate is inspected by a metrology device 140. Common examples of metrology devices are scatterometers and scanning electron microscopes. Scatterometers typically measure the characteristics of dedicated metrology targets. These metrology targets are representative of the product features, except that their size is usually larger to allow for precise measurement. Scatterometers measure overlay by detecting the asymmetry of the diffraction pattern associated with the overlay metrology target. Critical dimensions are measured by analyzing the diffraction pattern associated with the CD metrology target. The CD metrology target is used to measure the result of the most recently exposed layer. The overlay target is used to measure the difference between the position of the previous layer and the position of the most recent layer. Electron beam (e-beam) based inspection tools such as scanning electron microscopes (SEMs) can generally provide excellent results when measuring small overlay and CD values.

[0046] Within a semiconductor production facility, the lithography apparatus 100 and metrology equipment 140 form part of a "lithography cell" or "lithography cluster." The lithography cluster also includes a coating apparatus 108 for applying photoresist to a substrate W, a bake apparatus 110, a developer 112 for developing the exposed pattern into a physical resist pattern, an etching station 122, an apparatus 124 for performing a post-etch annealing step, and may also include processing equipment 126. The metrology equipment is configured to inspect the substrate after development 112 or after further processing (e.g., etching). The various devices within the lithography cell are controlled by a supervisory control system (SCS), which issues control signals 166 to control the lithography apparatus via the lithography apparatus control unit (LACU) 106 to execute conditional measures R. The SCS allows the operation of different devices to maximize throughput and product yield. An important control mechanism is feedback 146 (via the SCS) from the metrology equipment 140 to the various devices, particularly the lithography apparatus 100. Based on the characteristics of the measurement feedback, corrective actions are determined to improve the processing quality of subsequent substrates. The SCS can be a single computer or multiple computers, which may or may not communicate. The conditional means R can be implemented as a single conditional means or multiple independent conditional means. For example, the conditional means for a process step such as etching may be completely independent of the conditional means for checking the results of the process step (e.g., etching). For example, two or more conditional means for various steps can be interrelated so that one conditional means is adjusted to take into account the results of another conditional means on the same or different substrates.

[0047] The performance of lithographic equipment is typically controlled and corrected through methods such as Advanced Process Control (APC), as described in US2012008127A1. Advanced process control techniques use measurements of metrology targets applied to the substrate. A manufacturing execution system (MES) schedules APC measurements and transmits the measurement results to a data processing unit. The data processing unit transforms the characteristics of the measurement data into conditional means including instructions for the lithographic equipment. This method is very effective in suppressing drift phenomena associated with lithographic equipment.

[0048] Processing metrology data into corrective actions performed by processing devices is important for semiconductor manufacturing. In addition to metrology data, characteristics of individual patterning devices, substrates, processing devices, and other contextual data may be required to further optimize the manufacturing process. A framework in which available metrology and contextual data are used to optimize the lithography process as a whole is often referred to as part of integrated lithography. For example, contextual data related to CD errors on a reticle can be used to control various devices (lithography equipment, etching stations) so that these CD errors do not affect the yield of the manufacturing process. Subsequent metrology data can be used to verify the effectiveness of the control strategy and determine additional corrective measures.

[0049] Self-aligned quadruple patterning (SAQP), which may also be referred to as sidewall-assisted quadruple patterning, is a technique that produces features at a quarter the pitch of features produced by an optical exposure (photolithography) step. SAQP has been developed to print lines or regular holes at a denser pitch without reducing the exposure wavelength. The process will refer to Figures 2A to 2F Provide explanation.

[0050] exist Figures 2A to 2F In each of the figures, the left portion is a top view of the result of the process steps, while the right portion is a cross-sectional view. In a first step, a resist feature 200 (e.g., a line) is formed on the wafer. The width of the line is equal to 3P and is part of an array of similar lines with a pitch of 8P. In a second step, a conformal coating process is performed to apply a first spacer 201 on each of the sidewalls of the resist feature 200, as shown in FIG. Figure 2B The coating process is carefully controlled so that each of the first spacers 201 has a width P.

[0051] The original resist features 200 are removed so as to leave two first spacers 201, which are now separated by a gap of width 3P, as shown in FIG. Figure 2C Then, a second conformal coating process is performed to form a second spacer layer 202 on each side of each of the two first spacers 201, as shown in FIG. Figure 2DAgain, the coating process is carefully controlled so that the second spacers 202 each have a width P. This leaves a gap between the two spacer clusters of width P.

[0052] Then, the first spacer 201 is chemically removed so that only the second spacer 202 remains, as shown in FIG. Figure 2E As shown. Each of the second spacers 202 has a width P, and the gaps between the second spacers 202 also have a width P. Therefore, the pitch of the new pattern is 2P, while the pitch of the original pattern is 8P. Finally, a pattern transfer step, such as an etching step, is performed, and the spacers are removed (if they were not removed during the etching process) to produce a set of trenches 203, as shown. Figure 2F shown.

[0053] It should be appreciated that the above process represents an idealized version of the SAQP process. In reality, there will be some variation in the ideal (or nominal) pattern in terms of feature location (e.g., due to overlay) and / or width (CD error). Thus, overlay error, CD error, and CD uniformity (CDu) error, as well as random fluctuations and defects, can cause the lithographically defined features 200 to have non-straight edges. The effects of this non-straight edge are as follows: Figures 3A to 3D shown, where variations from nominal feature sizes are exaggerated.

[0054] Figure 3A Two adjacent lithographically defined features 300a, 300b are shown, with variations in their edges in this example line. When a conformal coating process is performed to produce the first spacers 301a to 301d, as shown in FIG. Figure 3B As shown, the first spacers take on the shape of the respective edges of the lithographically defined features 300a, b and are therefore not straight. The width of the first spacers may also vary at this stage. Figure 3C As shown, second spacers 302a to 302g are formed during the second conformal coating process, and then the first spacers 301a to 301d are removed to reach Figure 3D The final stage is shown. In Figure 3, the isotropic lines are labeled 0, 1, 2, and 3.

[0055] In the event of variations, knowing which line in the final pattern corresponds to which line in the intermediate process step is valuable for the etch recorder, as it allows the recorder to see which process step is most prone to errors and should therefore be optimized. One way to find out what is in the final pattern is to take an addressed SEM image, i.e., an SEM image where the position in the pattern is known. However, in many cases, the accuracy of SEM addressing is not sufficient to determine which line is which based on its position in the image. Figure 4is a sketch indicating what such an image would look like. Since the line width is likely to be around 20nm to 30nm, it can be seen that positioning errors in SEM images of this order will result in incorrect attribution of the lines in the image. Figure 5 is similar to Figure 4 , but shows where some features merge to form defects.

[0056] It should be pointed out that Figure 4 and Figure 5 Depicted is an array of features after a cutting process step has been performed such that each of the original line features has been divided into a plurality of shorter, but still elongated, line segments. The present technique can be applied to images of the array of lines before the cutting step, or to images of the array of features after the cutting step. In the latter case, line segments from columns or regions of the same line feature can be grouped together and treated as one feature. It should be noted that the pattern of elongated features can have features aligned with either the x-axis or the y-axis of the lithographic apparatus, or aligned at intermediate angles. The coordinate system of the SEM is ideally aligned with the coordinate system of the lithographic apparatus, but this need not be the case.

[0057] One possibility for providing greater certainty about which line is which is to image the border of the array and count the lines from that border to determine which line is which. However, this approach has the disadvantage that only a small portion of the array can be imaged, and the array border itself can be more prone to error. Additionally, imaging only the sides or corners of the array may not be representative.

[0058] Thus, the present disclosure provides a method for determining which lines in an array are associated with which process steps in a multi-step process, for example, which gaps correspond to the first spacer in a SAQP process, by correlating fluctuations in the placement of adjacent lines.

[0059] The disclosed techniques are applicable to a variety of multi-step processes, i.e., processes where a single lithographically defined feature is transformed into multiple features in the final device through additional process steps. The techniques are particularly applicable when the multi-step process includes a conformal coating process, i.e., a process that forms a layer of constant width or thickness.

[0060] according to Figure 3D , it can be seen that the lines indicated using 0 and 1 have similar profiles and have the same fluctuations. The reason for this is that the deposition step is usually very conformal, as it is done using ALD (Atomic Layer Deposition). Therefore, the variation in placement that occurs is the same as the original line edge roughness (LER) to the left of the printed line on the left in the first lithography step ( Figure 3A ), and the gaps between these lines correspond to the first spacers 301a. Therefore, it is expected that the correlation of position fluctuations between line 0 and line 1 is high.

[0061] For the same reason, the placement correlation between line 2 and line 3 is expected to be high, while the placement correlation between line 1 and line 2 or between line 3 and line 0 is expected to be low or insignificant. Therefore, by determining the placement correlation between adjacent lines, the correlation between two of the four adjacent line pairs (0-1, 1-2, 2-3, 3-4) is expected to be large, while the correlation between the other two pairs is expected to be small. The large correlation corresponds to the gap caused by the first spacer.

[0062] Now, reference Figure 8 A more detailed description of the example process is given.

[0063] A pattern comprising an array of features in a single layer is formed S1 on a substrate and imaged S2 using a scanning electron microscope or similar tool. The pattern is formed in a multi-step process such as SAQP, where different features, or different portions of a feature, can be defined by different steps of the multi-step process. The imaging step can be performed after the pattern has been transferred to the substrate (e.g., by etching features), or while the pattern is still defined by sacrificial features (such as spacers).

[0064] The image from the SEM is processed by first outlining the structures S3 and then determining the center of each structure. Any suitable algorithm can be used to determine the outline of the object in the image. In particular, it is desirable to select an algorithm that is particularly suitable for the shape of the structure to be outlined. The center of a feature can be defined as its centroid, but other definitions of the center of a feature are also possible. Examples include the center of the area, the geometric midpoint (i.e., the midpoint between the extreme points of the outline in two orthogonal directions (e.g., x and y)), and the center line. The present invention can also make use of other characteristics of the feature, such as the position of an edge, or the magnitude of a dimension, such as a line width. The center point or center line can be defined relative to a portion of the feature. Weighting can be applied to points (e.g., boundary points) to calculate the average position.

[0065] In some cases, it may be desirable to remove S4 distortion, as well as the true distortion on the wafer, from the SEM image (particularly by decomposing the grid distortion of the determined centroids into static and time-varying SEM contributions). A suitable method is described in European Patent Application 18210026.3, which is incorporated herein by reference. Distortion removal can also be performed before centroids are contoured and determined, but the amount of processing required is reduced if performed on the centroids. Not removing distortion may reduce, but not eliminate, the differences in correlation values ​​obtained and used later in the process.

[0066] The next step is to label (eg, number) all the lines of the S5 structure. Suitable labeling schemes include 0, 1, 2, 3, 0, 1, 2, etc. Which line is called 0 is arbitrary.

[0067] Then, the coefficients of multiple correlations between the structure in a given line of S6 and a number of adjacent structures in the lines next to it are determined. Figure 7 As shown in Figure 7 Labeled feature columns (or groups) are shown, with their centers indicated by crosses. To determine the coefficient of the multiple correlation, we define the vector quantity of the correlation for the portion of the variance of the structure placement in a given line that is explained by the position fluctuations of the holes in the adjacent lines:

[0068]

[0069] where P_left is the position of the left hole, P_(right,i) is the placement of the i-th neighbor on the right, and r is the correlation coefficient.

[0070] Furthermore, we define the correlation matrix Q of the neighborhood structure in the right line, where

[0071]

[0072] Then, the coefficient of the multiple correlation can be calculated by the following equation:

[0073]

[0074] In experiments conducted using 15 different images of examples of SAQP-formed patterns, a large R 2 , but smaller correlations are observed between the other two pairs of lines. The pair of lines with the larger correlation is associated with the first spacer S7, and the pair of lines with the smaller correlation is associated with the second spacer.

[0075] Defects in the pattern imaged by the SEM are detected S8. Defect detection can be performed before, after, or in parallel with the above steps to associate the line with a specific process step. By combining knowledge about which feature is associated with which process step and the location of the defect, remedial action to be taken S9 can be determined. Possible forms of remedial action may include adjusting the process applied to subsequent substrates, reprocessing substrates that have already been processed. In some cases, defects can be resolved by adjusting subsequent steps performed on the same substrate. In some cases, remedial action may include scrapping a substrate that does not meet specifications before further work is performed on it. Any action aimed at improving yield or throughput or otherwise addressing the detected defects can be considered a remedial action.

[0076] Other features that can be used with the techniques disclosed herein include cut lines for SADP engineering and SAQP processes. The techniques described herein can be used to analyze the dependencies when placing such cut features.

[0077] Another possibility is to measure the line before cutting and determine the power spectral density (PSD) of the correlation. This may provide a stronger signal to identify the different features and therefore be more correlated for 'better print' cases where there is less LER.

[0078] Furthermore, the high-frequency content of the PSD of the LER can be investigated for all edges. Specifically, it is expected that in some cases the high-frequency content inside and outside the original lithographically defined line will be different, so the location of the original lithographically defined line (Gap 1-2 or Gap 3-0) can be determined.

[0079] Thus, the techniques described herein are able to distinguish the first spacers, lines, and spaces that separate gaps from the original lithographically defined pattern without a well-addressed SEM.

[0080] While specific techniques have been described above, it will be appreciated that the disclosure may be practiced otherwise than as described.

[0081] One embodiment may include a computer program comprising one or more sequences of machine-readable instructions configured to instruct Figure 1 The various devices shown perform the measurement and optimization steps and control the subsequent exposure process as described above. For example, the computer program can be used in Figure 1 The computer program is executed in the control unit LACU or the supervisory control system SCS or a combination of the two. A data storage medium (eg semiconductor memory, magnetic disk or optical disk) in which such a computer program is stored may also be provided.

[0082] While specific reference may have been made above to optical lithography, it should be appreciated that the techniques disclosed herein can be used in other applications, such as imprint lithography. In the case of imprint lithography, the topography in the patterning device defines the pattern produced on the substrate. The topography of the patterning device can be pressed into a resist layer supplied to the substrate, and the resist is then cured by applying electromagnetic radiation, heat, pressure, or a combination thereof. The patterning device removes the resist, leaving a pattern in the resist after it has cured.

[0083] As used herein, the terms "radiation" and "beam" encompass all types of electromagnetic radiation, including ultraviolet (UV) radiation (e.g., having a wavelength of or about 365 nm, 355 nm, 248 nm, 193 nm, 157 nm, or 126 nm) and extreme ultraviolet (EUV) radiation (e.g., having a wavelength in the range of 1 nm to 100 nm), as well as particle beams (such as ion beams or electron beams). Implementation of scatterometers and other inspection devices can be performed at UV and EUV wavelengths using suitable sources, and the present disclosure is in no way limited to systems using IR and visible radiation.

[0084] Where the context permits, the term "lens" may refer to any one or combination of various types of optical components, including refractive, reflective, magnetic, electromagnetic, and electrostatic optical components. Reflective components are likely to be used in devices operating in the UV range and / or EUV range.

[0085] As used herein, unless expressly stated otherwise, the term "or" encompasses all possible combinations unless not feasible. For example, if it is stated that a component can include A or B, then unless expressly stated otherwise or not feasible, the component can include A, or B, or A and B. As a second example, if it is stated that a component can include A, B, or C, then unless expressly stated otherwise or not feasible, the component can include A, or B, or C, or A and B, or A and C, or B and C, or A, B, and C.

[0086] Various aspects of the disclosure are set out in the following numbered clauses:

[0087] 1. An image analysis method for identifying features in an image of a portion of an array of features formed by a multi-step process, the method comprising:

[0088] Analyzing changes in features visible in the image; and

[0089] Based at least in part on the results of the analysis, features of the image are associated with steps of the multi-step process.

[0090] 2. The method of clause 1, wherein the change is a change in position or a change in shape of the feature.

[0091] 3. The method of clause 2, wherein the change is a change in position of one or more of:

[0092] The centroid of the feature,

[0093] The geometric midpoint of the feature,

[0094] The centerline of the feature,

[0095] The edge of the feature,

[0096] the size of the feature,

[0097] • The centers of multiple parts of a feature.

[0098] 4. The method according to any one of clauses 1 to 3, wherein the analyzing comprises analyzing correlations between changes in different features.

[0099] 5. The method of clause 4, wherein analyzing comprises determining a correlation between a change in the first feature and a change in each of n adjacent features, where n is less than 4.

[0100] 6. The method of clause 4, wherein the analyzing comprises determining correlations between changes in adjacent features.

[0101] 7. A method according to clause 4, 5 or 6, wherein the association associates features having related variations with the same processing step.

[0102] 8. A method according to clause 4, 5, 6 or 7, wherein the associating divides the features into a plurality of sets having similar degrees of relevance and associating each set with a respective one of the processing steps.

[0103] 9. A method according to any of the preceding clauses, wherein the analyzing comprises: selecting a plurality of regions of the image, each region comprising a plurality of features; and determining correlations between features of different regions.

[0104] 10. The method of clause 9, wherein each region comprises a plurality of aligned features.

[0105] 11. The method according to any of the preceding clauses, wherein the multi-step process includes a conformal deposition step.

[0106] 12. A method according to any of the preceding clauses, wherein analyzing comprises determining a profile of each of the features.

[0107] 13. A method according to any of the preceding clauses, wherein the image is an image obtained by scanning electron microscopy.

[0108] 14. A device manufacturing method comprising:

[0109] forming an array of features on a substrate using a multi-step process;

[0110] obtaining an image of a portion of the array;

[0111] The method of any one of clauses 1 to 13, analyzing the image to associate features with steps of a multi-step process;

[0112] detecting defects in features of the array; and

[0113] Based on the association of features with steps and the defects detected, remedial actions are performed.

[0114] 15. An image analysis apparatus for identifying features in an image of a portion of an array of features formed by a multi-step process, the apparatus comprising:

[0115] an image analysis module configured to analyze changes in features visible in the image; and

[0116] An association module is configured to associate features of the image with steps of the multi-step process based at least in part on the results of the analysis.

[0117] 16. The apparatus of clause 15, wherein the change is a change in position or a change in shape of the feature.

[0118] 17. The apparatus of clause 16, wherein the change is a change in position of one or more of:

[0119] The centroid of the feature,

[0120] The geometric midpoint of the feature,

[0121] The centerline of the feature,

[0122] The edge of the feature,

[0123] the size of the feature,

[0124] • The centers of multiple parts of a feature.

[0125] 18. The apparatus according to any of clauses 15 to 17, wherein the image analysis module is configured to analyze correlations between changes in different features.

[0126] 19. The apparatus of any of clauses 15 to 17, wherein the image analysis module is configured to determine a correlation between a change in a first feature and a change in each of n adjacent features, where n is less than 4.

[0127] 20. The apparatus of clause 18, wherein the image analysis module is configured to determine correlations between changes in adjacent features.

[0128] 21. The apparatus of clause 18, 19 or 20, wherein the association module is configured to associate features having related changes with the same processing step.

[0129] 22. The apparatus of any of clauses 18 to 21, wherein the association module is configured to: divide the features into a plurality of sets having similar degrees of relevance, and associate each set with a respective one of the processing steps.

[0130] 23. The apparatus of any of clauses 15 to 22, wherein the image analysis module is configured to: select a plurality of regions of the image, each region comprising a plurality of features; and determine correlations between features of different regions.

[0131] 24. The device of clause 22, wherein each region comprises a plurality of aligned features.

[0132] 25. The device of any of clauses 15 to 24, wherein the multi-step process comprises a conformal deposition step.

[0133] 26. The apparatus of any of clauses 15 to 25, wherein analyzing comprises determining a profile of each of the features.

[0134] 27. An inspection apparatus comprising a scanning electron microscope and an image analysis apparatus according to any of clauses 15 to 26, the image analysis apparatus being configured to analyze an image generated by the scanning electron microscope.

[0135] 28. A method of analyzing an image of a portion of an array of features formed by a self-aligned quadruple patterning process, the method comprising:

[0136] Identify multiple features in an image;

[0137] Assigning a feature value to each identified feature, the feature value representing a change in position or shape of the feature;

[0138] grouping the features into a first group, a second group, a third group, and a fourth group, each group comprising a set of aligned features, the first group, the second group, the third group, and the fourth group being adjacent to each other in that order;

[0139] determining a first correlation value between a change in a characteristic value of the first group and a change in a characteristic value of the second group;

[0140] determining a second correlation value between changes in the characteristic value of the second group and changes in the characteristic value of the third group;

[0141] If the first correlation value is higher than the second correlation value, the first and second groups are associated with the first spacer of the self-aligned quadruple patterning process, otherwise the second and third groups are associated with the first spacer of the self-aligned quadruple patterning process.

[0142] 29. The method of clause 17, wherein the characteristic value is one of:

[0143] The location of the feature’s centroid,

[0144] The location of the geometric midpoint of the feature,

[0145] The location of the feature's centerline,

[0146] The location of the feature's edges,

[0147] the size of the feature,

[0148] • The location of the center of a part of a feature.

[0149] 30. The method of clause 28 or 29, wherein the feature is a line feature with a cut, or a line feature without a cut.

[0150] 31. A method of identifying pairs of features generated from an edge of a printed feature using a manufacturing process including a conformal coating step, the method comprising:

[0151] Analyzing the plurality of features to determine correlations between adjacent pairs of features; and

[0152] Based on the correlation, it is determined that a pair of features are both generated based on an edge of a conformally coated printed feature.

[0153] 32. The method according to clause 31, further comprising:

[0154] Based on the correlation, the steps in the manufacturing process that are related to the generation of defects are identified.

[0155] 33. A method according to clause 31 or 32, wherein the analyzing comprises: analyzing the placement of features.

[0156] 34. A computer program comprising computer readable code which, when executed by a computer system, instructs the computer system to perform the method according to any one of clauses 1 to 13 or 28 to 33.

[0157] The breadth and scope of the present technology disclosed herein should not be limited by any of the above-described exemplary embodiments, but should be defined only in accordance with the following claims and their equivalents.

Claims

1. An image analysis method for identifying features in an image of a portion of an array of features formed by a multi-step process, the method comprising: analyzing variations in features contained or indicated in said image and analyzing correlations between said variations in different features of said array; associating features of the image with steps of the multi-step process based at least in part on results of the analyzing; as well as Based on the correlation, a step in the multi-step process that caused the defect is determined. The method of claim 1 , wherein the change is a change in position or a change in shape of the feature.

3. The method of claim 2, wherein the change is a change in position of one or more of: the centroid of the feature, the geometric midpoint of the feature, the centerline of the feature, the edges of the feature, the dimensions of the feature, • The centers of parts of the feature.

4. The method of claim 3, wherein analyzing comprises: A correlation between a variation of a first feature and a variation of each of n adjacent features is determined, where n is less than 4.

5. The method of claim 3, wherein the analyzing comprises: Determine the correlation between changes in adjacent features.

6. A method according to claim 3, 4 or 5, wherein the associating associates features having correlated variations with the same processing step. 7 . The method of claim 6 , wherein the associating divides the features into a plurality of sets having similar degrees of correlation and associates each set with a corresponding one of the processing steps.

8. The method according to any one of the preceding claims, wherein the analyzing comprises: A plurality of regions of the image are selected, each region comprising a plurality of features; and correlations between the features of different regions are determined. The method of claim 8 , wherein each region comprises a plurality of aligned features.

10. The method of any one of the preceding claims, wherein the multi-step process comprises a conformal deposition step.

11. The method according to any one of the preceding claims, wherein the analyzing comprises: A profile of each of the features is determined.

12. The method according to any one of the preceding claims, wherein the image is an image obtained by scanning electron microscopy.

13. The method of claim 1 , wherein analyzing the correlation comprises: determining a correlation vector between a given feature and a neighboring feature, the correlation vector representing the portion of the variance of the given feature placement explained by the placement of the neighboring features; Constructing a correlation matrix of the adjacent features; as well as The coefficient of multiple correlation is calculated by using the correlation vector and the correlation matrix to quantify the correlation strength between the given feature and the adjacent features.

14. A device manufacturing method comprising: forming an array of features on a substrate using a multi-step process; obtaining an image of a portion of the array; The method of any one of claims 1 to 13, analysing the image to associate features with steps of the multi-step process; and Defects in features of the array are detected.

15. The method according to claim 14, further comprising: Based on the association of features with steps and the defects detected, remedial actions are performed.

16. The method of claim 14, wherein the multi-step process is a sidewall-assisted quadruple patterning process comprising the following steps: forming resist features on the substrate; performing a conformal coating process to apply a first spacer on each of the sidewalls of the resist feature; removing the resist feature so as to leave two first spacers; performing a second conformal coating process to form a second spacer on each side of each of the two first spacers; chemically removing the first spacer so as to leave only the second spacer; as well as A pattern transfer step is performed and the second spacers are removed to produce a set of trenches.

17. An image analysis apparatus for identifying features in an image of a portion of an array of features formed by a multi-step process, the apparatus comprising: an image analysis module configured to analyze variations in features contained or indicated in the image and to analyze correlations between the variations in different features of the array; as well as an association module configured to associate features of the image with steps of the multi-step process based at least in part on results of the analyzing, The step in the multi-step process that causes the defect is determined based on the correlation.

18. The apparatus of claim 17, wherein the change is a change in position or a change in shape of the feature.

19. The apparatus of claim 18, wherein the change is a change in position of one or more of: the centroid of the feature, the geometric midpoint of the feature, the centerline of the feature, the edges of the feature, the dimensions of the feature, • The centers of parts of the feature.

20. The apparatus of claim 17, wherein the image analysis module is configured to determine the correlation between a change in a first feature and a change in each of n adjacent features, where n is less than 4.

21. The apparatus of claim 17, wherein the image analysis module is configured to determine the correlation between changes in adjacent features.

22. The apparatus of claim 17, wherein the association module is configured to associate features having related changes with the same processing step.

23. The apparatus of claim 22, wherein the associating module is configured to divide the features into a plurality of sets having similar degrees of correlation, and associate each set with a corresponding one of the processing steps.

24. The apparatus of claim 17, wherein the image analysis module is configured to select a plurality of regions of the image and determine correlations between the features of different regions, each region comprising a plurality of features.

25. The apparatus of claim 23, wherein each region comprises a plurality of aligned features.

26. The apparatus of claim 17, wherein the multi-step process includes a conformal deposition step.

27. The apparatus of claim 17, wherein the analyzing comprises: A profile of each of the features is determined.

28. An inspection apparatus comprising a scanning electron microscope and the image analysis apparatus according to claim 17, the image analysis apparatus being configured to analyze an image generated by the scanning electron microscope.

29. A method of analyzing an image of a portion of an array of features formed by a self-aligned quadruple patterning process, the method comprising: identifying a plurality of said features in said image; Assigning a feature value to each identified feature, the feature value representing a change in position or shape of the feature; grouping the features into a first group, a second group, a third group, and a fourth group, each group comprising a set of aligned features, the first group, the second group, the third group, and the fourth group being sequentially adjacent to each other; determining a first correlation value between a change in the characteristic value of the first group and a change in the characteristic value of the second group; determining a second correlation value between a change in the characteristic value of the second group and a change in the characteristic value of the third group; If the first correlation value is higher than the second correlation value, the first group and the second group are associated with the first spacer of the self-aligned quadruple patterning process, otherwise the second group and the third group are associated with the first spacer of the self-aligned quadruple patterning process.

30. The method of claim 29, wherein the characteristic value is one of: the location of the feature’s centroid, the location of the geometric midpoint of the feature, the location of the feature's centerline, the location of the edge of the feature, the dimensions of the feature, • The location of the center of a portion of the feature.

31. The method of claim 29 or 30, wherein the feature is a line feature with a cut or a line feature without a cut.

32. A method of identifying pairs of features generated from an edge of a printed feature using a manufacturing process including a conformal coating step, the method comprising: Analyze multiple features to determine the correlation between adjacent pairs of features; as well as Based on the correlation, it is determined that a pair of features are both generated based on an edge of a printed feature to which a conformal coating is applied.

33. The method of claim 32, further comprising: Based on the correlation, a step in the manufacturing process that causes the defect is determined.

34. The method of claim 32 or 33, wherein analyzing comprises: The placement of the features is analyzed.

35. A computer program comprising computer readable code which, when executed by a computer system, instructs the computer system to perform the method according to any one of claims 1 to 13 or 29 to 34.

Images