Method and system for scatterometry modeling

Abstract

The present application relates to a method and system for scatterometry modeling. A metrology system may receive a model for measuring one or more selected properties of a target including features distributed in a selected pattern based on regression of spectral scatterometry data from a scatterometry tool over a range of wavelengths. The metrology system may further generate a weighting function for the model to de-emphasize portions of the spectral scatterometry data associated with wavelengths where light captured by the scatterometry tool when measuring the target is predicted to include undesired diffraction orders. The metrology system may further direct the spectral scatterometry tool to generate scatterometry data of one or more measurement targets including manufacturing features distributed in the selected pattern. The metrology system may further measure the selected properties of the one or more measurement targets based on regression of the scatterometry data of the one or more measurement targets to the model weighted by the weighting function.

Description

[0001] CROSS REFERENCE TO RELATED APPLICATIONS

[0002] This application claims the benefit of U.S. Provisional Application Serial No. 62 / 794,510, filed on January 18, 2019, entitled CD MODELING IN THE PRESENCE OF HIGHER DIFFRACTION ORDERS, by inventors named Phillip Atkins, Liequan Lee, Krishnan Shankar, David Wu, and Emily Chiu, under 35 U.S.C. §119(e), the entire contents of which are incorporated herein by reference. Technical Field

[0003] The present invention relates generally to scatterometry metrology, and more particularly, the present invention relates to spectroscopic scatterometry metrology. Background Art

[0004] Scatterometry provides metrological measurements based on the angle at which analytical light is emitted from a sample rather than analyzing an image of the sample. Light incident on a sample is reflected, scattered, and / or diffracted at various angles based on the structure and composition of features on the sample. In addition, the angle and / or intensity of the reflected, scattered, and / or diffracted light typically varies with different wavelengths of the incident light. Therefore, spectral scatterometry captures scatterometry signals at multiple wavelengths to further characterize features on a sample.

[0005] Scatterometry metrology typically utilizes models to relate scatterometry data (e.g., reflected, scattered, and / or diffracted light) to one or more properties of the sample, including but not limited to physical dimensions of features (e.g., critical dimensions (CD)), feature heights, separation distances between features on one or more sample layers, material properties of one or more sample layers, or optical properties of one or more sample layers.

[0006] It is often desirable to limit the space of the sample dedicated to metrology to leave space for manufacturing features of interest. A technique for reducing the size of the metrology target is to utilize a large numerical aperture when collecting light. However, it may be the case that a large numerical aperture will capture light associated with unwanted diffraction orders from the sample, which can introduce noise into the fit of the scatterometry data to the model. It is therefore desirable to provide systems and methods that eliminate the above-mentioned drawbacks. Summary of the invention

[0007] According to one or more illustrative embodiments of the invention, a metrology system is disclosed. In an illustrative embodiment, the system includes a controller. In another illustrative embodiment, the controller receives a model for measuring one or more selected properties of a target including one or more features distributed in a selected pattern based on regression of spectral scatterometry data from a scatterometry tool within a selected wavelength range. In another illustrative embodiment, the controller generates a weighting function of the model to de-emphasize a portion of the spectral scatterometry data associated with one or more wavelengths within the selected wavelength range where light captured by the scatterometry tool when measuring the target is predicted to include one or more undesired diffraction orders. In another illustrative embodiment, the controller directs the spectral scatterometry tool to generate scatterometry data of one or more measurement targets within the selected wavelength range, wherein the one or more measurement targets include manufacturing features distributed in the selected pattern. In another illustrative embodiment, the controller measures the one or more selected properties of the one or more measurement targets based on regression of the scatterometry data of the one or more measurement targets to the model weighted by the weighting function.

[0008] According to one or more illustrative embodiments of the invention, a metrology system is disclosed. In an illustrative embodiment, the system includes a scatterometry tool for generating spectral scatterometry data of a target within a selected wavelength range. In another illustrative embodiment, the system includes a controller. In another illustrative embodiment, the controller receives a model for measuring one or more selected properties of a target including one or more features distributed in a selected pattern based on regression of the spectral scatterometry data from the scatterometry tool within the selected wavelength range. In another illustrative embodiment, the controller generates a weighting function of the model to de-emphasize a portion of the spectral scatterometry data associated with one or more wavelengths within the selected wavelength range where light captured by the scatterometry tool when measuring the target is predicted to include one or more undesired diffraction orders. In another illustrative embodiment, the controller directs the spectral scatterometry tool to generate scatterometry data of one or more measurement targets within the selected wavelength range, wherein the one or more measurement targets include manufacturing features distributed in the selected pattern. In another illustrative embodiment, the controller measures the one or more selected properties of the one or more measurement targets based on regressing the scatterometry data of the one or more metrology targets to the model weighted by the weighting function.

[0009] According to one or more illustrative embodiments of the invention, a metrology method is disclosed. In one illustrative embodiment, the method includes receiving a model for measuring one or more selected properties of a target including one or more features distributed in a selected pattern based on regression of spectral scatterometry data from a scatterometry tool within a selected wavelength range. In another illustrative embodiment, the method includes generating a weighting function of the model to de-emphasize a portion of the spectral scatterometry data associated with one or more wavelengths within the selected wavelength range where light captured by the scatterometry tool when measuring the target is predicted to include one or more undesired diffraction orders. In another illustrative embodiment, the method includes directing the spectral scatterometry tool to generate scatterometry data of one or more measurement targets within the selected wavelength range, wherein the one or more measurement targets include manufacturing features distributed in the selected pattern. In another illustrative embodiment, the method includes measuring the one or more selected properties of the one or more measurement targets based on regression of the scatterometry data of the one or more measurement targets to the model weighted by the weighting function.

[0010] It should be understood that both the above general description and the following detailed description are for illustration and explanation only and are not necessarily limiting of the present invention.The accompanying drawings, which are incorporated in and constitute a part of this specification, illustrate embodiments of the present invention and together with the detailed description, serve to explain the principles of the present invention. BRIEF DESCRIPTION OF THE DRAWINGS

[0011] Those skilled in the art may better understand the numerous advantages of the present invention by referring to the accompanying drawings, in which:

[0012] Figure 1A is a block diagram of a scatterometry metrology system according to one or more embodiments of the present invention;

[0013] Figure 1B is a conceptual diagram of a scatterometry metrology tool according to one or more embodiments of the present invention;

[0014] Figure 1C is a conceptual diagram of a scatterometry metrology tool configured with a common objective lens according to one or more embodiments of the present invention;

[0015] Figure 2 is a flow chart illustrating steps performed in a method for scatterometry metrology according to one or more embodiments of the present invention;

[0016] Figure 3 is a flow chart illustrating sub-steps associated with generating a weighting function based on measurements of a reference target according to one or more embodiments of the present invention;

[0017] Figure 4Ais a perspective view of a test pitch target according to one or more embodiments of the present invention;

[0018] Figure 4B is a perspective view of a modified pitch target according to one or more embodiments of the present invention;

[0019] Figure 5 The invention is based on the comparison and Figure 4B The modified pitch residual associated with the modified pitch target described in Figure 4A A plot of the weights generated by the test pitch residuals associated with the test pitch targets described in;

[0020] Figure 6 is a flow chart illustrating sub-steps for generating a weighting function based on estimating a total contamination intensity associated with a portion of undesired diffraction orders captured by a scatterometry metrology tool in accordance with one or more embodiments of the present invention;

[0021] Figure 7 The invention is based on one or more embodiments of the present invention. Figure 4A A plot of weights generated by the estimated total contamination intensity associated with the measurements of the test pitch target 402 illustrated in FIG. 4 ; and

[0022] Figure 8 is a plot illustrating weights generated based on a linear combination of residual weighting and ray tracing methods according to one or more embodiments of the present invention. DETAILED DESCRIPTION

[0023] Reference will now be made in detail to the disclosed subject matter illustrated in the accompanying drawings. The present invention has been particularly shown and described with respect to specific embodiments and specific features thereof. The embodiments described herein should be considered to be intended to be illustrative and not limiting. It should be readily apparent to those skilled in the art that various changes and modifications in form and details may be made without departing from the spirit and scope of the present invention.

[0024] Embodiments of the present invention are directed to systems and methods for spectral scatterometry that incorporate a model for measuring one or more properties of a sample based on scatterometry data that includes one or more desired diffraction orders and further incorporate weights to de-emphasize scatterometry data associated with undesired diffraction orders.

[0025] Scatterometry metrology tools typically illuminate a sample and capture (e.g., using a detector) associated light emitted from the sample in a pupil plane. In this regard, the angle and intensity of light from the sample may be captured. This scatterometry data may then be related to any of a variety of properties of the sample through a model. For example, features on a sample may reflect, scatter, and / or diffract incident light based on various physical or optical sample properties, such as, but not limited to, a number of sample layers, a pattern (e.g., distribution) of features on one or more layers, a composition of a layer, a thickness of a layer, or an optical property of a layer. Thus, a model may incorporate various simulations and / or estimates of the interaction of incident light with the sample to provide a relationship between the physical or optical properties of the sample and the corresponding scatterometry data. Furthermore, because the scattering, reflection, and diffraction of light typically vary as a function of wavelength, a spectroscopic scatterometry tool may capture scatterometry signals at multiple wavelengths simultaneously or sequentially.

[0026] It is often the case that a physical or optical property of a manufactured sample differs from a target value. Thus, a model may often include one or more floating parameters such that scatterometry data associated with a manufactured sample may be fit to the model. Accordingly, the resulting values ​​of the floating parameters may be used to determine measured values ​​of one or more properties of the sample. For example, fitting the scatterometry data to the model may provide measurements of one or more features, such as, but not limited to, feature size (e.g., critical dimension (CD)), feature height, sidewall angle, spacing between features on a selected sample layer (e.g., pitch of distributed features), or spacing between features on different layers (e.g., overlay measurement). By way of another example, fitting the scatterometry data to the model may provide measurements of one or more sample layers, such as, but not limited to, layer thickness or refractive index.

[0027] It may be the case that scatterometry models for different applications (e.g., measurements of targets incorporating different features) may require different scatterometry data as input. For example, optical critical dimension (OCD) measurements may (but not necessarily) utilize a model that incorporates scatterometry data associated with a single diffraction order (e.g., the 0th order) of the scatterometry. Accordingly, scatterometry data associated with undesired diffraction orders (e.g., higher order diffraction) may introduce noise when fitting to the model. Accordingly, undesired diffraction orders may contaminate the measurement and may cause measurement inaccuracies.

[0028] It should be recognized here that spectral scatterometry models may generally benefit from scatterometry data across a wide bandwidth range to provide information sufficient for accurate and precise determination of the values ​​of the floating parameters. However, in practice, it may be desirable to utilize a particular scatterometry tool (e.g., a particular spectral scatterometry tool) to characterize a target that results in the collection of at least some undesirable scatterometry data (e.g., associated with desired diffraction orders). Continuing with the above example, it may be desirable to perform OCD measurements on a target having features with a large pitch distribution sufficient to cause a particular spectral scatterometry tool to collect at least some higher order diffraction for at least some wavelengths within the bandwidth of the tool. For example, the angle at which light is diffracted from periodic features on a sample may generally decrease as the pitch increases and the wavelength of the light decreases. Thus, a target suitable for characterizing some state-of-the-art logic devices or static random access memory (SRAM) devices may have features with a large pitch sufficient to result in the collection of at least some higher order diffraction, particularly when using visible light or ultraviolet (UV) light.

[0029] However, it should be recognized herein that the impact of undesired scatterometry data may vary with different wavelengths. Embodiments of the present invention are directed to generating a weighting function for a scatterometry model to de-emphasize undesired scatterometry data, such as scatterometry data associated with undesired diffraction orders. Thus, undesired scatterometry data may be utilized to the extent that it does not unduly affect the accuracy of the measurement, which may be determined based on any selected goodness of fit metric. Accordingly, the degree of contamination introduced at each wavelength may be balanced with the benefit of data across a wide spectral range to provide an accurate fit. For example, it should be noted that data may be excluded only for wavelengths that introduce (or are expected to introduce) contamination. However, this approach may provide a close fit of the model based on a selected goodness of fit metric, but the limited amount of collected data may actually reduce the accuracy of the measurement. In contrast, weighting scatterometry data based on the degree of contamination according to embodiments of the present invention may result in highly accurate measurements regardless of the contamination.

[0030] Embodiments of the present invention include systems and methods for generating weighting functions for a spectral scatterometry model designed to relate spectral scatterometry data from a selected spectral scatterometry tool (e.g., a spectral scatterometry tool with a selected measurement bandwidth and numerical aperture) to one or more properties of interest (e.g., CD or the like) of a test target (e.g., a measurement target). For example, the test target may include a known distribution of features in one or more sample layers.

[0031] In some embodiments, a weighting function is generated by measuring spectral scatterometry data of a test pitch target and at least one modified pitch target. For example, the test pitch target may be identical to the test target of interest, while the modified pitch target may include features distributed according to the same pattern as the test target of interest but with sufficiently small pitch values ​​to avoid capturing undesired diffraction orders by the selected spectral scatterometry tool. Furthermore, the scatterometry data may be fit to a model and wavelength-dependent residuals may be determined. Then, a wavelength-dependent weighting function may be generated based on the ratio of the residuals of the modified pitch target and the test pitch target.

[0032] In some embodiments, the weighting function is generated based on an estimate of contamination at each wavelength, which may be generated using various techniques including, but not limited to, ray tracing techniques or rigorous coupled wave analysis (RCWA) techniques.

[0033] In some embodiments, a weighting function is generated based on a combination of multiple techniques. For example, a weighting function may be generated as, but not limited to, a linear combination of multiple weighting techniques.

[0034] Now generally refer to Figures 1A to 8 Systems and methods for scatterometry metrology using a weighted model are described in more detail below.

[0035] Figure 1A is a block diagram of a scatterometry metrology system 100 in accordance with one or more embodiments of the present invention.

[0036] In one embodiment, the scatterometry metrology system 100 includes a scatterometry metrology tool 102 for generating scatterometry data associated with one or more samples. The scatterometry metrology tool 102 may include any type of scatterometry metrology system known in the art suitable for providing scatterometry metrology signals at one or more wavelengths. For example, the scatterometry metrology tool 102 may include, but is not limited to, a spectrometer, a spectroscopic ellipsometer with one or more illumination angles, a spectroscopic ellipsometer for measuring Mueller matrix elements (e.g., using a rotating compensator), a spectroscopic reflectometer, or a scatterometer.

[0037] In addition, the scatterometry metrology tool 102 may include a single metrology tool or multiple metrology tools. For example, a metrology tool including multiple hardware configurations is generally described in U.S. Pat. No. 7,933,026, the entire contents of which are incorporated herein by reference. A metrology system incorporating multiple metrology tools is generally described in U.S. Pat. No. 7,478,019, the entire contents of which are incorporated herein by reference. Focused beam ellipsometry based on primary reflective optics is generally described in U.S. Pat. No. 5,608,526, the entire contents of which are incorporated herein by reference. The use of apodizers to mitigate the effects of optical diffraction that causes the spread of illumination spots beyond the size defined by the geometric optics is generally described in U.S. Pat. No. 5,859,424, the entire contents of which are incorporated herein by reference. The use of a high numerical aperture tool with simultaneous multi-incident angle illumination is generally described in U.S. Pat. No. 6,429,943, the entire contents of which are incorporated herein by reference.

[0038] The scatterometry metrology tool 102 may generate scatterometry data associated with any location on the sample. In one embodiment, the scatterometry metrology tool 102 generates scatterometry data of a device feature on the sample. Accordingly, the scatterometry metrology tool 102 may directly characterize the feature of interest. In another embodiment, the scatterometry metrology tool 102 generates scatterometry data of one or more metrology targets (e.g., targets) that include manufacturing features designed to represent device features on the sample. Accordingly, measurements of one or more metrology targets distributed across the sample may be attributed to the device features. For example, it may be the case that the size, shape, or distribution of sample features may not be suitable for accurate metrology measurement. In contrast, the metrology targets may include features on one or more sample layers that have sizes, shapes, and distributions that are adapted such that the scatterometry data of the targets are highly sensitive to one or more selected physical or optical properties of the features. The scatterometry data of the targets may then be related to specific values ​​of the selected properties (e.g., through a model).

[0039] Metrology targets may be designed to be sensitive to and thus facilitate the measurement of various physical or optical properties, including, but not limited to, CD, overlay, sidewall angle, film thickness, film composition, or process-related parameters such as focus, dose, and the like. To this end, metrology targets may include any combination of periodic structures, such as one-, two-, or three-dimensional periodic structures, or isolated non-periodic features. Characterization of non-periodic structures using metrology tools is generally described in U.S. Patent No. 9,291,554, issued on March 22, 2016, the entire contents of which are incorporated herein by reference. In addition, metrology targets may be generally characterized as having one or more spatial frequencies, such as one or more pitches, attributable to a pattern or distribution of features. The use of symmetric target designs in scatterometry overlay metrology is generally described in U.S. Patent Publication No. 2015 / 0204664, published on July 23, 2015, the entire contents of which are incorporated herein by reference.

[0040] Metrology targets may be located at multiple sites on the sample. For example, targets may be located within scribe lines (e.g., between dies) and / or within the die themselves. Multiple targets may be measured simultaneously or sequentially by the same or multiple metrology tools as described in U.S. Pat. No. 7,478,019, the entire contents of which are incorporated herein by reference.

[0041] In another embodiment, the scatterometry metrology system 100 includes a controller 104. In another embodiment, the controller 104 includes one or more processors 106 configured to execute program instructions stored on a storage medium 108 (e.g., a memory). In this regard, the one or more processors 106 of the controller 104 may perform any of the various process steps described in this disclosure. Furthermore, the storage medium 108 may store any type of data for use by any component of the scatterometry metrology system 100. For example, the storage medium 108 may store a protocol for the scatterometry metrology tool 102, scatterometry data generated by the scatterometry metrology tool 102, or the like.

[0042] The one or more processors 106 of the controller 104 may include any processing element known in the art. In this regard, the one or more processors 106 may include any microprocessor-type device configured to execute algorithms and / or instructions. In one embodiment, the one or more processors 106 may be comprised of a desktop computer, a mainframe computer system, a workstation, an image computer, a parallel processor, or any other computer system (e.g., a network computer) configured to execute a program (which is configured to operate the scatterometry metrology system 100) as described in this disclosure. It should be further recognized that the term "processor" may be broadly defined to encompass any device having one or more processing elements that execute program instructions from the non-transitory memory medium 108.

[0043] The memory medium 108 may include any storage medium known in the art that is suitable for storing program instructions that can be executed by the associated one or more processors 106. For example, the memory medium 108 may include a non-transitory memory medium. By way of another example, the memory medium 108 may include, but is not limited to, a read-only memory, a random access memory, a magnetic or optical storage device (such as a disk), a tape, a solid-state drive, and the like. It should be further noted that the memory medium 108 may be housed in a common controller housing with the one or more processors 106. In one embodiment, the memory medium 108 may be remotely located relative to the physical location of the one or more processors 106 and the controller 104. For example, the one or more processors 106 of the controller 104 may access a remote memory (such as a server) that is accessible via a network (such as the Internet, an intranet, and the like). Therefore, the above description should not be interpreted as a limitation of the present invention, but is merely illustrative.

[0044] Additionally, the controller 104 and any associated components (e.g., the processor 106, the memory medium 108, or the like) may include one or more controllers housed in a common housing or multiple housings. Furthermore, the controller 104 may be integrated with and / or perform functions of any component in the scatterometry metrology system 100.

[0045] The controller 104 may perform any number of the processing or analysis steps disclosed herein, including but not limited to receiving, generating, or applying a model to relate scatterometry data to selected attributes of sample features, which may involve a number of algorithms. For example, the metrology target may be modeled (parameterized) using any technique known in the art, including but not limited to a geometry engine, a process modeling engine, or a combination thereof. The use of process modeling is generally described in U.S. Patent Publication No. 2014 / 0172394, the entire contents of which are incorporated herein by reference. The geometry engine may be, but not necessarily, implemented by AcuShape software (a product provided by KLA-TENCOR). By way of another example, an electromagnetic (EM) solver may be used, but not limited to, to model the optical interaction of an illumination beam with a metrology target on a sample. Furthermore, the EM solver may utilize any method known in the art, including but not limited to rigorous coupled wave analysis (RCWA), finite element method analysis, moment analysis, surface integral techniques, volume integral techniques, or finite difference time domain analysis.

[0046] The controller 104 may further analyze the data collected from the scatterometry metrology tool 102 using any data fitting to model and optimization techniques known in the art to apply the collected data to a model, including but not limited to a library, a fast degradation model, regression, a machine learning algorithm such as a neural network, a support vector machine (SVM), a dimensionality reduction algorithm such as principal component analysis (PCA), independent component analysis (ICA), local linear embedding (LLE), and the like, a sparse representation of the data such as Fourier or wavelet transforms, Kalman filters, algorithms for facilitating matching with the same or different tool types, and the like. For example, the data collection and / or fitting may be, but not necessarily, performed by a signal response metrology (SRM), a product provided by KLA-Tencor Corporation.

[0047] In another embodiment, the controller 104 uses an algorithm that does not include modeling, optimization, and / or fitting (e.g., phase characterization or the like) to analyze the raw data generated by the scatterometry metrology tool 102. It should be noted here that the computational algorithms executed by the controller may (but not necessarily) be adapted to the metrology application by using parallelization, distributed computing, load balancing, multi-service support, design and implementation of computational hardware, or dynamic load optimization. Furthermore, various implementations of the algorithms may (but not necessarily) be executed by the controller 104 (e.g., by firmware, software, or field programmable gate arrays (FPGAs) and the like) or one or more programmable optical elements associated with the scatterometry metrology tool 102.

[0048] Figure 1B is a conceptual diagram of a scatterometry metrology tool 102 in accordance with one or more embodiments of the present invention.

[0049] In one embodiment, scatterometry metrology tool 102 includes an illumination source 110 to generate an illumination beam 112. Illumination beam 112 may include light of one or more selected wavelengths including, but not limited to, ultraviolet (UV) radiation, visible radiation, or infrared (IR) radiation.

[0050] In one embodiment, the illumination source 110 includes a laser source. For example, the illumination source 110 may include, but is not limited to, one or more narrowband laser sources, one or more broadband laser sources, one or more supercontinuum laser sources, one or more white light laser sources, or the like. Furthermore, the illumination source 110 may include any type of laser source known in the art, including, but not limited to, a diode laser source or a diode pumped laser source.

[0051] In another embodiment, the illumination source 110 includes a lamp source. By way of another example, the illumination source 110 may include, but is not limited to, an arc lamp, a discharge lamp, an electrodeless lamp, and the like. Accordingly, the illumination source 110 may provide an illumination beam 112 having low coherence (e.g., low spatial coherence and / or temporal coherence).

[0052] In another embodiment, the illumination source 110 comprises a broadband plasma (BBP) illumination source. Accordingly, the illumination beam 112 may comprise radiation emitted by a plasma. For example, the BBP illumination source 110 may (but need not) comprise one or more pump sources (e.g., one or more lasers) configured to focus into a volume of a gas to cause energy to be absorbed by the gas to create or maintain a plasma suitable for emitting radiation. Furthermore, at least a portion of the plasma radiation may be used as the illumination beam 112.

[0053] In another embodiment, the illumination source 110 provides a tunable illumination beam 112. For example, the illumination source 110 may include a tunable illumination source such as one or more tunable lasers and the like. By way of another example, the illumination source 110 may include a broadband illumination source coupled to a tunable filter.

[0054] The illumination source 110 can further provide an illumination beam 112 having any temporal profile. For example, the illumination beam 112 can have a continuous temporal profile, a modulated temporal profile, a pulsed temporal profile, or the like.

[0055] In another embodiment, the illumination source 110 directs the illumination beam 112 to the sample 114 via the illumination path 116 and collects light emitted from the sample (e.g., collected light 118) via the collection path 120. The collected light 118 may include any combination of light from the sample 114 generated in response to the incident illumination beam 112, such as, but not limited to, reflected light, scattered light, diffracted light, or luminescent light of the sample 114.

[0056] In another embodiment, the sample 114 is located on a sample stage 122. The sample stage 122 may include any type of translation stage known in the art suitable for aligning one or more features (e.g., a target) on the sample 114 within the scatterometry metrology tool 102 to perform scatterometry measurements. For example, the sample stage 122 may include any combination of a linear translation stage, a rotation stage, and / or a flip / tilt stage.

[0057] In one embodiment, the illumination path 116 may include an illumination focusing element 124 to focus the illumination beam 112 onto the sample 114. The illumination path 116 may include one or more illumination beam conditioning components 126 suitable for modifying and / or conditioning the illumination beam 112. For example, the one or more illumination beam conditioning components 126 may include, but are not limited to, one or more polarizers, one or more filters, one or more beam splitters, one or more apodizers or one or more beam shapers, one or more diffusers, one or more homogenizers, or one or more lenses.

[0058] In another embodiment, the collection path 120 may include a collection focusing element 128 to capture the collection light 118 from the sample 114. In another embodiment, the scatterometry metrology system 100 includes a detector 130 configured to detect at least a portion of the collection light 118 emitted from the sample 114 through the collection path 120. The detector 130 may include any type of optical detector known in the art suitable for measuring illumination received from the sample 114. For example, the detector 130 may include, but is not limited to, a CCD detector, a CMOS detector, a TDI detector, a photomultiplier tube (PMT), an avalanche photodiode (APD), and the like. In another embodiment, the detector 130 may include a spectral detector suitable for identifying the wavelength of radiation emitted from the sample 114.

[0059] The collection path 120 may further include any number of collection beam conditioning elements 132 to direct and / or modify the illumination collected by the collection focusing element 128, including but not limited to one or more lenses, one or more filters, one or more polarizers, or one or more phase plates.

[0060] In another embodiment, Figure 1B The scatterometry metrology tool 102 depicted in FIG. 1 may facilitate multi-angle illumination of a sample 114 and / or one or more illumination sources 110 (eg, coupled to one or more additional detectors 130). Accordingly, Figure 1B The scatterometry metrology tool 102 depicted in FIG. 1 may perform multiple metrology measurements. In another embodiment, the scatterometry metrology tool 102 may include multiple detectors 130 to facilitate multiple metrology measurements performed by the scatterometry metrology tool 102 (eg, multiple metrology tools).

[0061] Furthermore, the scatterometry metrology tool 102 can facilitate multi-angle illumination of the sample 114 and / or one or more illumination sources 110. Accordingly, the scatterometry metrology tool 102 can perform multiple metrology measurements. In another embodiment, one or more optical components can be mounted to a rotatable arm (not shown) that pivots about the sample 114, such that the angle of incidence of the illumination beam 112 on the sample 114 can be controlled by the position of the rotatable arm.

[0062] In some embodiments, one or more components are common to both the illumination path 116 and the collection path 120 . Figure 1C 1 is a conceptual diagram of a scatterometry metrology tool 102 configured with a common objective lens according to one or more embodiments of the present invention. In one embodiment, the scatterometry metrology tool 102 includes a beam splitter 134 oriented so that an objective lens 136 can simultaneously direct an illumination beam 112 to a sample 114 and capture a collection light 118 emitted from the sample 114. In this regard, the objective lens 136 can replace or be combined with Figure 1B The illumination focusing element 124 and / or collection focusing element 128 operate.

[0063] Generally speaking Figures 1A to 1C , the controller 104 may be communicatively coupled to any component of the scatterometry metrology system 100. In one embodiment, the controller 104 is communicatively coupled to the illumination source 110 to provide one or more selected wavelengths of illumination for scatterometry measurements. In another embodiment, the controller 104 is configured to one or more elements of the illumination path 116 to direct the adjustment of the angle of incidence between the illumination beam 112 and the sample 114.

[0064] Figure 2 2 is a flow chart illustrating steps performed in a method 200 for scatterometry metrology according to one or more embodiments of the present invention. Applicants should note that the embodiments and enabling techniques previously described herein in the context of the scatterometry metrology system 100 should be interpreted as extending to the method 200. However, it should be further noted that the method 200 is not limited to the architecture of the scatterometry metrology system 100.

[0065] In one embodiment, the method comprises step 202 of receiving a model for measuring a selected property of a target comprising features distributed in a selected pattern based on spectral scatterometry data within a selected wavelength range.

[0066] As previously described herein in the context of the scatterometry metrology system 100, but not limited to the architecture of the scatterometry metrology system 100, a scatterometry model may relate scatterometry data to one or more properties of a target (e.g., a metrology target). For example, the target may be modeled using any technique known in the art, including, but not limited to, a geometry engine, a procedural modeling engine, or a combination thereof.

[0067] In one embodiment, the model is adapted to measure one or more selected properties of a particular measurement target of interest (e.g., a measurement target) that includes features distributed in a selected pattern. For example, the measurement target may correspond to a metrology target designed to provide measurements related to and representative of device features of a particular type fabricated on the same sample. In this regard, the model may incorporate various physical properties of the measurement target of interest, such as, but not limited to, the number of sample layers, the CD of features in any sample layer, the height of features in any sample layer, the separation distance between features (e.g., the pitch associated with periodically distributed features), or the sidewall angle. The model may further incorporate various optical properties of the measurement target of interest, such as, but not limited to, the thickness or composition of one or more sample layers.

[0068] The model may determine the scattering, reflection, and / or diffraction of the illumination beam 112 based on any number of simulations and / or estimates of the interaction of the illumination beam 112 with the target of interest. For example, the optical interaction of the illumination beam with the metrology target on the sample may be modeled using, but not limited to, an electromagnetic (EM) solver. Furthermore, the EM solver may utilize any method known in the art, including, but not limited to, rigorous coupled wave analysis (RCWA), finite element method analysis, moment analysis, surface integral techniques, volume integral techniques, or finite difference time domain analysis.

[0069] In one embodiment, the model received in step 202 is adapted or otherwise tailored for use with a particular spectral scatterometry tool. For example, the model may be adapted for use with the spectral range of the illumination source 110. By way of another example, the model may incorporate the numerical aperture of the illumination path 116 and / or the collection path 120. In this regard, the model may take into account the characteristics of the scatterometry metrology tool 102 to provide an accurate relationship between properties of target expected scatterometry data (e.g., the signal generated by the detector 130).

[0070] In another embodiment, the model includes one or more floating parameters, so that the scatterometry data can be fit to the model to determine specific values ​​of the floating parameters, which can then be related to specific values ​​of selected properties of the target. Thus, fitting the scatterometry data to the model can provide measurements of any number of physical or optical properties of the target.

[0071] The scatterometry data may be fitted to a model using any technique known in the art, such as, but not limited to, libraries, fast degradation models, regression, machine learning algorithms such as neural networks, support vector machines (SVMs), dimensionality reduction algorithms such as principal component analysis (PCA), independent component analysis (ICA), local linear embedding (LLE), and the like, sparse representations of the data such as Fourier or wavelet transforms, Kalman filters, algorithms for facilitating matching with the same or different tool types, and the like. In one embodiment, the scatterometry data may be fitted to a model using a maximum likelihood estimation (MLE) regression technique. Accordingly, the values ​​of the floating parameters may be determined based on the input data set by maximizing the likelihood function.

[0072] The model of step 202 may be received by any source. In one embodiment, the model is received from an external source, such as an external server, an external controller, or the like. In another embodiment, the model is generated at least in part by the controller 104. In this regard, one or more aspects of the model may be stored in and received from the storage medium 108.

[0073] In another embodiment, the method includes step 204 of generating a weighting function of the model to de-emphasize portions of the spectral scatterometry data associated with one or more undesired diffraction orders. For example, it may be the case that a measurement target of interest may have characteristics that result in a distribution of pitches (e.g., spatial frequencies) of undesired diffraction orders being collected when measured using the scatterometry metrology tool 102.

[0074] It should be recognized here that the difference between desired diffraction orders and undesired diffraction orders may vary based on the particular model and / or the particular property of interest being measured. Thus, the examples herein should be interpreted as illustrative and not limiting. In one embodiment, the one or more desired diffraction orders include a single diffraction order, such that the undesired diffraction orders include the remaining diffraction orders produced under the measurement conditions provided by the scatterometry metrology tool 102. For example, as previously described herein in the context of an OCD scatterometry tool, but not limited to an OCD scatterometry tool, the one or more desired diffraction orders may include a single diffraction order (e.g., the 0th order), such that the undesired diffraction orders are associated with higher orders of diffraction. In another embodiment, the one or more desired diffraction orders include two or more diffraction orders.

[0075] The portion of the spectral scatterometry data associated with one or more undesirable diffraction orders may be determined using any technique known in the art. For example, step 204 may include predicting which wavelengths of the illumination beam 112 will result in undesirable diffraction orders of light captured by the scatterometry metrology tool 102 using any combination of simulations, estimates, or measurements of reference targets.

[0076] Furthermore, step 204 may include a prediction of the extent to which undesired diffraction orders introduce noise into the fit and introduce measurement inaccuracies. Accordingly, a weighting function may variably de-emphasize scatterometry data associated with a wavelength based on the predicted impact. Thus, scatterometry data associated with wavelengths that induce undesired diffraction orders with relatively high impact may be suppressed to a greater extent than scatterometry data associated with wavelengths that induce undesired diffraction orders with relatively low impact.

[0077] In another embodiment, the method includes step 206 of directing the spectroscopic scatterometry tool to generate scatterometry data for one or more measurement targets (e.g., measurement targets of interest) within a selected wavelength range. For example, the one or more measurement targets may be located on a sample of interest. In another embodiment, the method includes step 208 of measuring selected properties of the one or more measurement targets based on the scatterometry data of the one or more test targets regressed to a model weighted by a weighting function. Thus, the model received in step 202 and modified by the weighting function of step 204 may be applied to measure any selected property (e.g., physical and / or optical property) on any number of measurement targets of interest on any number of samples of interest.

[0078] In another embodiment, although not shown in the figure, the measurement of the selected property can be provided as feedback and / or feedforward data to one or more additional process tools. For example, the measurement of the selected property on the measurement target in a batch can be provided as feedback data to one or more process tools (such as semiconductor manufacturing tools) to mitigate any process deviations of samples in the same batch or different batches. By way of another example, the measurement of the selected property on the measurement target in a batch can be provided as feedforward data to one or more process tools to compensate for any process deviations identified for the current batch.

[0079] Referring again to step 204, it should be noted here that the weighting function used to de-emphasize the scatterometry data based on the predicted impact may be generated in a variety of ways in accordance with the present invention. Figures 3 to 5 Weighting functions based on measurements of reference targets, including at least one reference target having a feature spacing modified to be free of undesirable diffraction orders, are generally described. Figure 6 and 7 In general, a weighting function based on a theoretical estimate of the effect of higher order diffraction is described. Figure 8 In more detail, the weighting function may comprise a combination (eg, a linear combination) of two or more weighting schemes.

[0080] Figure 3 3 is a flowchart 300 illustrating sub-steps associated with generating a weighting function based on measurements of a reference target according to one or more embodiments of the present invention. For example, flowchart 300 may represent a method associated with Figure 2Sub-steps associated with step 204.

[0081] In one embodiment, step 204 includes sub-step 302 of measuring selected properties of one or more modified pitch reference targets. For example, the modified pitch reference target may include a pattern of features that are the same or substantially the same as the measurement target of interest, but where the spacing distances are modified and / or scaled to limit diffraction to desired diffraction orders. In this regard, the modified pitch reference target may simulate the material and geometry of the measurement target of interest to provide an accurate representation of the measurement target as closely or substantially as possible, while also limiting diffraction to desired orders.

[0082] In another embodiment, step 204 includes a sub-step 304 of determining a modified pitch residual (eg, as a function of wavelength) between the scatterometry data and the model for one or more modified pitch reference targets.

[0083] Any number of modified pitch reference targets may be measured in sub-step 302. In one embodiment, a single modified pitch reference target is measured in sub-step 302. Thus, the modified pitch residual may be represented as the residual between the scatterometry data and the model for a single modified pitch reference target at each wavelength. In another embodiment, multiple modified pitch reference targets are measured in sub-step 302. For example, multiple modified pitch reference targets may be manufactured with variations in various physical or optical properties that may be expected to occur through deviations in the manufacturing process. Thus, the modified pitch residual may be represented by a statistical combination (e.g., a mean, a median, or the like) of the residuals between the scatterometry data and the model for one or more modified pitch reference targets at each wavelength.

[0084] In another embodiment, step 204 includes sub-step 306 of measuring selected properties of one or more test pitch reference targets corresponding to one or more measurement targets. For example, the test pitch reference target may have features distributed in the same pattern as the measurement target of interest and at a scale relative to the measurement target of interest. In this regard, the test pitch reference target may be identical to the measurement target of interest or otherwise fully simulate the measurement target of interest.

[0085] In another embodiment, step 204 includes a sub-step 308 of determining a test pitch residual (eg, as a function of wavelength) between the scatterometry tool and the model for one or more modified pitch reference targets.

[0086] Any number of test pitch reference targets may be measured in sub-step 306. In one embodiment, a single test pitch reference target is measured in sub-step 306. Thus, the test pitch residual may be represented as the residual between the scatterometry data and the model for a single test pitch reference target at each wavelength. In another embodiment, multiple test pitch reference targets are measured in sub-step 306. For example, multiple test pitch reference targets may be manufactured with variations in various physical or optical properties that may be expected to occur through deviations in the manufacturing process. Thus, the test pitch residual may be represented by a statistical combination (e.g., a mean, a median, or the like) of the residuals between the scatterometry data and the model for one or more test pitch reference targets at each wavelength.

[0087] It should be noted here that any weighting scheme may be used to measure and regress the modified pitch reference target (e.g., in sub-step 302) and the test pitch reference target (e.g., in sub-step 306) to determine the associated modified pitch and test pitch residuals. Furthermore, the weighting scheme used to generate the modified pitch residuals may be the same or different than the weighting scheme used to generate the test pitch residuals. In this regard, it should be recognized here that a weighting scheme suitable for determining the modified pitch residuals based on measurements of the modified pitch reference target may not be equally suitable for determining the test pitch residuals based on measurements of the test pitch reference target.

[0088] In one embodiment, a uniform weighting scheme is used to generate the modified pitch residuals and / or the test pitch residuals. In another embodiment, a wavelength and / or system dependent weighting scheme is used to generate the modified pitch residuals and / or the test pitch residuals. For example, the weighting scheme may be adapted to compensate for any known signal biases of a particular scatterometry metrology tool 102. By way of another example, the weighting scheme may (but is not necessarily) be limited only to wavelengths where contamination of undesired diffraction orders is not expected to occur. Since the modified pitch and test pitch residuals are provided as inputs to a contamination sensitive weighting function (e.g., see sub-step 310 below), limiting the information from wavelengths associated with contamination (or predicted contamination) does not affect the performance of the resulting weighting function.

[0089] In another embodiment, step 204 includes sub-step 310 of generating a weighting function based on a ratio of the modified pitch residual to the test pitch residual.

[0090] It should be recognized here that there are three major sources of error that contribute to the calculation of residuals associated with fitting scatterometry data to a model: modeling error, contamination associated with undesired diffraction orders, and various non-modeling errors (e.g., calibration errors and the like). If the modeling error and various non-modeling errors can be estimated or otherwise accounted for, the difference between the residuals (e.g., modified pitch residuals and test pitch residuals) can account for errors associated with contamination caused by undesired diffraction orders. Thus, comparing (e.g., via a ratio) the modified pitch residuals to the test pitch residuals at each wavelength can provide an estimate of the effect of contamination at each wavelength, which can then be used to weight the fit of the scatterometry data to the model.

[0091] In one embodiment, the weighting function (W res (λ)) is expressed as the absolute value of the ratio between the modified pitch residual and the test pitch residual, as described by Equation 1:

[0092]

[0093] Where R MP (λ) represents the modified pitch residual, R TP (λ) represents the test pitch residual, and λ represents the wavelength within the spectrum of the scatterometry metrology tool 102. In addition, the weighting function (W res (λ)) can be bounded between values ​​0 and 1.

[0094] Reference Figures 4A to 5 , a simulation example illustrating the generation of a weighting function based on measurements of a reference target is described. Figure 4A is a perspective view of a test pitch target 402 in accordance with one or more embodiments of the present invention. Figure 4B 4 is a perspective view of a modified pitch target 404 according to one or more embodiments of the present invention. In this example, the model is designed to provide at least CD measurements of the top and bottom portions of the fin structure. Furthermore, for this example, the model is based on the 0th order diffraction and higher order diffraction represents undesirable diffraction orders that may contribute to noise in the measurement.

[0095] Figure 4AThe test pitch target 402 described in the figures includes a three-dimensional grating structure including overlapping features in two layers. The test pitch target 402 may be (but not necessarily) suitable as a large pitch SRAM target for generating metrology measurements representing characteristics of an SRAM device. For example, the test pitch target 402 includes: a first layer 406 including a first layer feature 408; a second layer 410 including a second layer feature 412; and a third layer 414 (e.g., a substrate layer) without any features. The first layer features 408 and the second layer features 412 each include two groups of fin structures (e.g., a first group 416 and a second group 418), wherein the fin structures are periodically distributed with a pitch 420 of 1.25 microns (μm) along the Y direction and a pitch 422 of 2.5 microns along the X direction. For example, the two groups of fin structures may be (but not necessarily) manufactured by cutting the fin structures along the Y direction. The first layer features 408 and the second layer features 412 may overlap (e.g., Figure 4A ), but they do not necessarily need to overlap.

[0096] In addition to the first layer feature 408 and the second layer feature 412 each including a single set of fin structures, Figure 4B 4. The modified pitch target 404 illustrated in FIG. 4 includes a two-dimensional grating structure that is substantially the same as the test pitch target 402. For example, the modified pitch target 404 corresponds to the test pitch target 402 without the cut lines along the Y direction. Accordingly, the modified pitch target 404 may have substantially the same properties as the test pitch target 402, such as, but not limited to, the CD of the fin structure (e.g., along the Y direction), the fin height, the pitch along the Y direction, the layer thickness, the layer composition, or the layer refractive index.

[0097] Figure 5 The invention is based on the comparison and Figure 4B The modified pitch residual associated with the modified pitch target 404 described in Figure 4A 500 is a plot of the weights generated by the test pitch residuals associated with the test pitch target 402 described in FIG. Figure 3 The weighting function is generated by the steps described in . For example, the weights may correspond to a weighting function suitable for (but not limited to) a maximum likelihood estimator.

[0098] The modified pitch residual and the test pitch residual are each generated as the average of the residuals from 13 measurements, where the 13 measurements are generated by perturbing various properties of the modified pitch target 404 and the test pitch target 402, such as, but not limited to, the pitch of the fin structures (e.g., the first layer features 408 and the second layer features 412 along the Y direction), the top and bottom CDs of the fin structures, the material properties of the sample layers, and the spacing between the fin structure groups along the X direction of the test pitch target 402. In this way, the perturbations can represent deviations in the manufacturing process.

[0099] Figure 5 Also illustrated is a contamination cutoff wavelength 502, below which at least some undesirable higher order diffraction is captured (e.g., by the scatterometry metrology tool 102). Figure 5 , the weights produced using this technique have non-zero values ​​throughout the spectral range, including wavelengths below the cutoff wavelength 502. Thus, this extension of the usable wavelengths enables the utilization of a significant portion of the bandwidth of the scatterometry metrology tool 102, including wavelengths where undesirable higher order diffraction is captured, but with limited impact on the accuracy of the measurement.

[0100] In addition, Figure 5 As described in , the weights generated based on comparing the modified pitch residuals to the test pitch residuals can compensate for additional errors (eg, modeled errors and / or non-modeled errors). This can be understood by simplifying the weighting function for a specific value above the cutoff wavelength 502.

[0101] Characterize by comparing the quality of the fit obtained using the weighting function with a full simulation including higher order diffraction Figure 5 The accuracy of the weighting function in the . In this case, regressing the scatterometry data from the test pitch target 402 incorporating a full simulation of higher-order diffraction resulted in a chi-squared fit metric of 282.5, which was used as a benchmark. In contrast, regressing the scatterometry data from the test pitch target 402 incorporating only 0-order diffraction and a uniform weighting scheme (e.g., no weights applied to correction for higher-order contamination) resulted in a chi-squared value of 435.8, which indicates a relatively poor model fit. In addition, application Figure 5 The weighting function of resulted in a chi-square value of 297.4, which indicates a relatively good model fit and comparable results to the full simulation.

[0102] It should be noted here that generating a weighting function based on comparison of residuals from a reference target as described herein may enable an accurate fit to the model without requiring any calculation of undesired diffraction orders or their associated effects.

[0103] Reference Figure 6 and 7 , describes in more detail the generation of weighting functions by estimating the effects of contamination by undesired diffraction orders. This approach may be used instead of or in conjunction with the generation of weighting functions based on residuals from a reference target as previously described herein.

[0104] In one embodiment, the weighting function is generated based on estimating the total contamination intensity associated with the portion of undesired diffraction orders captured by the scatterometry metrology tool 102. For example, the weights may be proportional to the total contamination intensity within a selected range.

[0105] The total contamination intensity associated with undesired diffraction orders may be estimated in a variety of ways in accordance with the present invention. It should be recognized here that the weighting function does not necessarily need to be determined with the same accuracy as the underlying model. Given that the model provides a relationship between scatterometry data (e.g., scattered, reflected, and / or diffracted light captured by the scatterometry metrology tool 102) and the properties of the measured sample, the weighting function, when fitted to the model, merely adjusts the relative influence of the scatterometry data at a particular wavelength. Thus, it may be the case that an approximation or estimate that would not otherwise be suitable for use within a model may be suitable for generating a weighting function to be applied to the model.

[0106] Figure 6 6 is a flowchart 600 illustrating sub-steps for generating a weighting function based on estimating the total contamination intensity associated with a portion of undesired diffraction orders captured by the scatterometry metrology tool 102 in accordance with one or more embodiments of the present invention. For example, the flowchart 600 may represent a method for generating a weighting function based on estimating the total contamination intensity associated with a portion of undesired diffraction orders captured by the scatterometry metrology tool 102. Figure 2 However, it should be understood that the following model is for illustration only and should not be interpreted as limiting.

[0107] In one embodiment, step 204 includes sub-step 602 of estimating the trapped intensities associated with one or more undesired diffraction orders of light at wavelengths of interest, such as wavelengths within the spectral range of the scatterometry metrology tool 102. The trapped intensities associated with the undesired diffraction orders at a particular wavelength may be estimated using any technique known in the art, including, but not limited to, ray tracing techniques, RCWA techniques, direct measurement of one or more undesired diffraction orders coupled with a model that provides sufficient order separation, or the like. Furthermore, any combination of techniques may be used to estimate the trapped intensities associated with the undesired diffraction orders.

[0108] In one embodiment, a ray tracing model is used to estimate the contamination intensity associated with each diffraction order. For example, the ray tracing model may consider the projection of an infinitesimal focus size of light from an illumination aperture onto the sample and the corresponding projection of rays from the sample to the collection apertures associated with the various diffraction orders. Thus, the contamination intensity associated with each diffraction order may be determined based on the extent to which light from each diffraction order passes through the collection aperture.

[0109] The ray tracing model may (but not necessarily) be implemented using the following formula. In one embodiment, the ray tracing model may trace an infinitesimal focused spot size illumination ray (e.g., illumination beam 112) projected from at least a portion of the illumination aperture onto the sample. For example, rays may be traced from points along a two-dimensional curve representing the illumination aperture (e.g., the boundary of the illumination aperture). It should be noted here that tracing rays from the boundary of the illumination aperture may provide an accurate estimate of the contamination intensity.

[0110] For the pitch values ​​d along the X and Y directions respectivelyx and y The goal is to generate a reciprocal lattice:

[0111]

[0112] where θ OA is defined as the X pitch and the Y pitch (e.g. d x With d y ) is the orthogonal angle of the unit cell between and represents the principal vector in real space, and and denotes the corresponding principal vector in wave space.

[0113] Furthermore, the wave vector of the ray emitted from the sample can be written as the incident wave vector and the diffraction orders m in the X and Y directions respectively x and m y Function:

[0114]

[0115] The value Can be constrained to be positive to exclude evanescent waves.

[0116] From this, the rays emitted from the sample can be projected onto the plane containing the collection aperture. Furthermore, the illuminated area on the collection aperture for each of the diffraction orders can be determined by identifying which ray projections fall within the interior of the collection aperture, determining the edge intersection points between the projected rays and the aperture, and finding the polygon determined by the interior points and edge intersection points for each of the diffraction orders.

[0117] Finally, the illuminated area on the collection aperture of all undesired diffraction orders can be used to determine the intensity of the diffraction orders passing through the collection aperture (eg, contamination intensity).

[0118] In one embodiment, step 204 includes a sub-step 604 of estimating a total contamination intensity associated with the combined capture intensities of the undesired diffraction orders. Various techniques may be used to estimate the total contamination intensity associated with the combined capture intensities of the undesired diffraction orders. In one embodiment, the total contamination intensity is generated as the sum of the capture intensities of the undesired diffraction orders that pass through the collection aperture. In another embodiment, the various capture intensities of the undesired diffraction orders are scaled based on the area of ​​the contamination in the collection pupil and then summed. In this regard, both the intensity of the captured light associated with the undesired diffraction orders and the extent to which the captured light covers the collection aperture may be considered to contribute to the effects of the contamination.

[0119] For example, the total pollution intensity (I C,TOT (λ) can (but not necessarily) be expressed as:

[0120]

[0121] Where m x and m y represents the diffraction orders in two orthogonal directions, I C (λ,m x ,m y ) represents the contamination level (m) at the collection numerical aperture of the scatterometry metrology tool 102 x ,m y )’s average intensity, A Coll represents the area of ​​the collecting pore, and A C (λ,m x ,m y ) represents the contamination area on the collection pore.

[0122] In another embodiment, step 204 includes sub-step 606 of generating a weighting function based on the total pollution intensity.

[0123] Then, various techniques may be used to generate a weighting function based on the estimated total contamination intensity. In one embodiment, an upper threshold (n1) is defined relative to a reference level (I0) to describe a condition where the estimated total contamination intensity overwhelms the signal. For wavelengths that meet this condition, the weight may be set to 0 to exclude the associated scattering measurement data from the analysis. In another embodiment, a lower threshold (n0) may be defined relative to a reference level (I0) to describe a condition where the estimated total contamination intensity is negligible. For wavelengths that meet this condition, the weight may be set to 1 to provide full weight to the associated scattering measurement data. In another embodiment, the weight may be transitioned between 1 and 0 for wavelengths between the upper and lower thresholds. For example, the weighting function (W HO (λ)) can follow a linear relationship between 1 and 0, as described in Equation 5:

[0124]

[0125] It should be appreciated that any combination of upper and lower thresholds can be used to generate weights based on the total pollution intensity. For example, n0 or n1 can be set to 0. In addition, the values ​​of the upper and lower thresholds can be determined using any technique known in the art. In one embodiment, the upper and / or lower thresholds are specifically determined. For example, the upper and / or lower thresholds can be determined to provide a target baseline fit metric (e.g., a chi-square value or the like). In another embodiment, the upper and / or lower thresholds are determined based on the expected noise of the signal. For example, signals with a selected noise level (e.g., 3σ or the like) can be reduced or eliminated.

[0126] Any metric may be used to select the reference level (I0). In one embodiment, the reference level (I0) is selected to correspond to a noise variance level (σ(λ)) associated with any selected signal, such as, but not limited to, a collection signal or contamination intensity. In another embodiment, the reference level (I0) is selected based on one or more post-processing signals, such as, but not limited to, reflectivity measurements, harmonic signals (such as associated with spectroscopic ellipsometry or the like), or calculated sample Mueller.

[0127] However, it should be appreciated that the above example describing the weighting function as a linear function between two boundary conditions is for illustration only and should not be interpreted as limiting. In another embodiment, a nonlinear relationship between an upper threshold and a lower threshold may be used to define the weighting function. In another embodiment, the lower threshold is not defined and a non-zero weight may be applied to all wavelengths within the spectral range of the scatterometry metrology tool 102.

[0128] Figure 7 The invention is based on one or more embodiments of the present invention. Figure 4A 700 of the weights generated by the estimated total contamination intensity associated with the measurement of the test pitch target 402 described in FIG. Figure 6 In addition, the upper threshold (n0) 3 and the lower threshold (n0) 0 are used to generate the weighting function. Figure 7 The weight in .

[0129] like Figure 7 , the weights produced using this technique have non-zero values ​​throughout the spectral range, including wavelengths below the cutoff wavelength 502. Thus, this extension of the usable wavelengths enables the utilization of a significant portion of the bandwidth of the scatterometry metrology tool 102, including wavelengths where undesirable higher order diffraction is captured, but with limited impact on the accuracy of the measurement.

[0130] In addition, Figure 7 As described in , the weights generated based on comparing the modified pitch residuals to the test pitch residuals can compensate for additional errors (such as modeled errors and / or non-modeled errors). This can be understood by simplifying the weighting function above a specific value above the cutoff wavelength 502.

[0131] By comparing the quality of the fit obtained using the weighting function with that obtained previously in this paper Figure 5 The full simulation of higher order diffraction described in the background of Figure 7 The accuracy of the weighting function in . Application Figure 7 The weighting function of resulted in a chi-square value of 215.5, which indicates a relatively good model fit and comparable results to the full simulation with a chi-square value of 282.5.

[0132] Reference Figure 8 In some embodiments, the weighting function may be incorporated into a combination of two or more weighting functions. For example, any combination of weighting functions described in accordance with the present invention may be combined into a single weighting function.

[0133] The weighting function can be combined into any linear or nonlinear combination. In one embodiment, the weighting function is the W previously described herein. Res (λ) and W HO (λ). For example, the weighting function can be expressed as:

[0134] W(λ)=αW Res (λ)(1-α)W HO (λ), (6)

[0135] where α is the control combination weighting function W(λ) Res (λ) and W HO (λ). In addition, the value of α may be determined using any technique known in the art. In one embodiment, an upper threshold and / or a lower threshold are specifically determined. For example, the value of α may be determined to provide a target baseline fit metric (e.g., a chi-square value or the like). By way of another example, the value of α may be determined to provide a selected performance relative to a reference measurement from a high precision metrology tool (e.g., a scanning electron microscope, a transmission electron microscope, or the like). For example, a high precision metrology tool may be used to measure one or more target sites with known values ​​of a selected attribute (e.g., CD or the like). Then, the value of α may be determined to provide a selected performance based on one or more metrics for measuring the accuracy of the model-based method relative to the measured values ​​(e.g., R-squared, slope, offset, or the like).

[0136] Figure 8 8 is a plot 800 illustrating weights generated based on a linear combination of a residual weighting method and a ray tracing method according to one or more embodiments of the present invention. In addition, a value of α equal to 0.75 is used to generate Figure 7 By comparing the quality of the fit obtained using the weighting function with that obtained previously in this paper Figure 5 The full simulation of higher order diffraction described in the background of Figure 8 The accuracy of the weighting function in . Application Figure 8 The weighting function of resulted in a chi-square value of 220.0, which indicates a relatively good model fit and comparable results to the full simulation with a chi-square value of 282.5.

[0137] The objectives described herein sometimes illustrate different components contained in or connected to other components. It should be understood that such depicted architectures are for illustration only, and in fact many other architectures that realize the same functionality can be implemented. Conceptually, any arrangement of components for realizing the same functionality is effectively "associated" so that the desired functionality is realized. Therefore, any two components combined to realize a specific functionality herein can be considered to be "associated" with each other so that the desired functionality is realized, regardless of the architecture or intermediate components. Similarly, any two components so associated can also be considered to be "connected" or "coupled" to each other to realize the desired functionality, and any two components that can be so associated can also be considered to be "coupleable" to each other to realize the desired functionality. Specific examples of "coupleable" include, but are not limited to, physically interactable and / or physically interacting components and / or wirelessly interactable and / or wirelessly interacting components and / or logically interactable and / or logically interacting components.

[0138] It is believed that the present invention and its many attendant advantages will be understood from the above description, and it should be understood that various changes may be made to the form, construction and arrangement of components without departing from the disclosed objects or sacrificing all material advantages thereof. The forms described are for illustration only, and the appended claims are intended to cover and encompass such changes. In addition, it should be understood that the present invention is defined by the appended claims.

Claims

1. A metering system, comprising: A controller comprising one or more processors configured to execute program instructions to cause the one or more processors to: receiving a model for measuring one or more selected properties of a target comprising one or more features distributed in a selected pattern based on regression of scatterometry data from a scatterometry tool within a selected wavelength range; generating a weighting function of the model to de-emphasize a portion of the scatterometry data associated with one or more wavelengths within the selected wavelength range in which light captured by the scatterometry tool when measuring the target is predicted to include one or more undesired diffraction orders; directing the scatterometry tool to generate scatterometry data for one or more measurement targets within the selected wavelength range, wherein the one or more measurement targets include fabricated features distributed in the selected pattern; and The one or more selected properties of the one or more measurement targets are measured based on regressing the scatterometry data of the one or more measurement targets to the model weighted by the weighting function.

2. The metering system of claim 1 , wherein generating the weighting function comprises: measuring the one or more selected properties of one or more modified pitch reference targets based on a regression of scatterometry data of the one or more modified pitch reference targets, wherein the one or more modified pitch reference targets include features distributed in the selected pattern, wherein spacing between features is scaled so that light captured by the scatterometry tool when measuring the one or more modified pitch reference targets is predicted to include a single diffraction order within the selected wavelength range; determining a modified pitch residual between the scatterometry data of the one or more modified pitch reference targets and the model; measuring the one or more selected properties of one or more test pitch reference targets based on regression of scatterometry data of the one or more test pitch reference targets, wherein the one or more test pitch reference targets include features distributed in the selected pattern, wherein spacings between features of the one or more test pitch reference targets correspond to spacings between features of the one or more measurement targets; determining a test pitch residual between the scatterometry data of the one or more test pitch reference targets and the model; The weighting function is generated based on an absolute value of a ratio between the modified pitch residual and the test pitch residual.

3. The metering system of claim 2, wherein the weighting function W(λ) comprises: Where R MP (λ) represents the modified pitch residual, R TP (λ) represents the test pitch residual, and λ represents a wavelength within the selected wavelength range. 4 . The metrology system of claim 2 , wherein the test pitch residual is determined using a weighting scheme for determining the modified pitch residual.

5. The metrology system of claim 2, wherein the test pitch residual is determined using a uniform weighting scheme.

6. The metrology system of claim 2, wherein the one or more modified pitch reference targets include: two or more modified pitch reference targets, wherein the values ​​of the modified pitch residuals at specific wavelengths within the wavelength range include: an average of two or more residuals associated with the two or more modified pitch reference targets at the particular wavelength.

7. The metrology system of claim 2, wherein the one or more test pitch reference targets include: two or more test pitch reference targets, wherein the values ​​of the test pitch residuals at specific wavelengths within the wavelength range include: an average of two or more residuals associated with the two or more test pitch reference targets at the particular wavelength.

8. The metering system of claim 1 , wherein generating a weighting function comprises: estimating a captured intensity associated with each of the one or more undesired diffraction orders; estimating a total contamination intensity associated with the captured intensity of each of the undesired diffraction orders; and The weighting function is generated based on the total pollution intensity.

9. The metering system of claim 8, wherein the weighting function W(λ) comprises: Among them I C,TOT (λ) represents the total pollution intensity, I0(λ) represents the reference intensity, λ represents the wavelength within the selected wavelength range, and n0 and n1 represent limit parameters, wherein n0 represents the upper threshold value and n1 represents the lower threshold value.

10. The metering system according to claim 9, wherein the reference intensity comprises: The noise variance of the scatterometry data.

11. The metering system according to claim 9, wherein the total pollution intensity comprises: Where m x and m y represents the diffraction orders in two orthogonal directions, I C (λ,m x ,m y ) represents the contamination level on the collection aperture of the scatterometry tool (m x ,m y )’s average intensity, A Coll represents the area of ​​the collecting pore, and A C (λ,m x ,m y ) represents the contamination area on the collecting pore.

12. The metering system of claim 1 , wherein generating a weighting function comprises: measuring the one or more selected properties of one or more modified pitch reference targets based on a regression of scatterometry data of the one or more modified pitch reference targets, wherein the one or more modified pitch reference targets include features distributed in the selected pattern, wherein spacing between features is scaled so that light captured by the scatterometry tool when measuring the one or more modified pitch reference targets is predicted to include a single diffraction order of the selected wavelength range; determining a modified pitch residual between the scatterometry data of the one or more modified pitch reference targets and the model; measuring the one or more selected properties of one or more test pitch reference targets based on regression of scatterometry data of the one or more test pitch reference targets, wherein the one or more test pitch reference targets include features distributed in the selected pattern, wherein spacings between features of the one or more test pitch reference targets correspond to spacings between features of the one or more measurement targets; determining a test pitch residual between the scatterometry data of the one or more test pitch reference targets and the model; generating a first weighted sub-function based on an absolute value of a ratio between the modified pitch residual and the test pitch residual; estimating a captured intensity associated with each of the one or more undesired diffraction orders; estimating a total contamination intensity associated with the captured intensity of each of the undesired diffraction orders; and A second weighted sub-function is generated based on the total pollution intensity.

13. The metering system according to claim 12, wherein the first weighted sub-function W Res (λ) includes: Where R MP (λ) represents the modified pitch residual, R TP (λ) represents the test pitch residual, and λ represents the wavelength within the selected wavelength range, wherein the second weighting subfunction W HO (λ) includes: Among them I C,TOT (λ) represents the total pollution intensity, I0(λ) represents the reference intensity, λ represents the wavelength within the selected wavelength range, and n0 and n1 represent limit parameters, wherein n0 represents the upper threshold and n1 represents the lower threshold, wherein the weighting function W(λ) includes: W(λ)=αW HO (λ)+(1-α)W Res (λ), where α is a parameter defining the relative weights of the first weighted sub-function and the second weighted sub-function.

14. The metering system of claim 13, wherein the reference intensity comprises: The noise variance of the scatterometry data.

15. The metering system according to claim 13, wherein the total pollution intensity comprises: Where m x and m y represents the diffraction orders in two orthogonal directions, I C (λ,m x ,m y ) represents the contamination level on the collection aperture of the scatterometry tool (m x ,m y )’s average intensity, A Coll represents the area of ​​the collecting pore, and A C (λ,m x ,m y ) represents the contamination area on the collecting pore.

16. The metrology system of claim 1, wherein the scatterometry tool comprises: an illumination source configured to generate an illumination beam comprising the selected wavelength range; one or more illumination optics that direct the illumination beam to the sample; one or more collection optics that collect light from the sample in response to the illumination beam; and A detector is configured to generate the scatterometry data based on detecting at least a portion of the light collected by the one or more collection optics.

17. A metering system comprising: a scatterometry tool for generating scatterometry data of a target within a selected wavelength range; and a controller communicatively coupled to the scatterometry tool, the controller comprising one or more processors configured to execute program instructions to cause the one or more processors to: receiving a model for measuring one or more selected properties of a target comprising one or more features distributed in a selected pattern based on regression of scatterometry data from the scatterometry tool within the selected wavelength range; generating a weighting function of the model to de-emphasize a portion of the scatterometry data associated with one or more wavelengths within the selected wavelength range in which light captured by the scatterometry tool when measuring the target is predicted to include one or more undesired diffraction orders; directing the scatterometry tool to generate scatterometry data for one or more measurement targets within the selected wavelength range, wherein the one or more measurement targets include fabricated features distributed in the selected pattern; and The one or more selected properties of the one or more measurement targets are measured based on regressing the scatterometry data of the one or more measurement targets to the model weighted by the weighting function.

18. The metrology system of claim 17, wherein the scatterometry tool comprises: an illumination source configured to generate an illumination beam comprising the selected wavelength range; one or more illumination optics that direct the illumination beam to the sample; one or more collection optics that collect light from the sample in response to the illumination beam; and A detector is configured to generate the scatterometry data based on detecting at least a portion of the light collected by the one or more collection optics.

19. The metrology system of claim 17, wherein the scatterometry tool comprises: Spectral reflectance measurement tool.

20. The metrology system of claim 17, wherein the scatterometry tool comprises: Spectroscopic ellipsometry tool.

21. The metering system of claim 17, wherein generating the weighting function comprises: measuring the one or more selected properties of one or more modified pitch reference targets based on a regression of scatterometry data of the one or more modified pitch reference targets, wherein the one or more modified pitch reference targets include features distributed in the selected pattern, wherein spacing between features is scaled so that light captured by the scatterometry tool when measuring the one or more modified pitch reference targets is predicted to include a single diffraction order within the selected wavelength range; determining a modified pitch residual between the scatterometry data of the one or more modified pitch reference targets and the model; measuring the one or more selected properties of one or more test pitch reference targets based on regression of scatterometry data of the one or more test pitch reference targets, wherein the one or more test pitch reference targets include features distributed in the selected pattern, wherein spacings between features of the one or more test pitch reference targets correspond to spacings between features of the one or more measurement targets; determining a test pitch residual between the scatterometry data of the one or more test pitch reference targets and the model; The weighting function is generated based on an absolute value of a ratio between the modified pitch residual and the test pitch residual.

22. The metering system of claim 21, wherein the weighting function W(λ) comprises: Where R MP (λ) represents the modified pitch residual, R TP (λ) represents the test pitch residual, and λ represents a wavelength within the selected wavelength range.

23. The metering system of claim 17, wherein generating the weighting function comprises: estimating a captured intensity associated with each of the one or more undesired diffraction orders; estimating a total contamination intensity associated with the captured intensity of each of the undesired diffraction orders; and The weighting function is generated based on the total pollution intensity.

24. The metering system of claim 23, wherein the weighting function W(λ) comprises: Among them I C,TOT (λ) represents the total pollution intensity, I0(λ) represents the reference intensity, λ represents the wavelength within the selected wavelength range, and n0 and n1 represent limit parameters, wherein n0 represents an upper threshold value and n1 represents a lower threshold value.

25. The metrology system of claim 24, wherein the reference intensity comprises: The noise variance of the scatterometry data.

26. The metering system of claim 24, wherein the total pollution intensity comprises: Where m x and m y represents the diffraction orders in two orthogonal directions, I C (λ,m x ,m y ) represents the contamination level on the collection aperture of the scatterometry tool (m x ,m y )’s average intensity, A Coll represents the area of ​​the collecting pore, and A C (λ,m x ,m y ) represents the contamination area on the collecting pore.

27. The metering system of claim 17, wherein generating the weighting function comprises: measuring the one or more selected properties of one or more modified pitch reference targets based on a regression of scatterometry data of the one or more modified pitch reference targets, wherein the one or more modified pitch reference targets include features distributed in the selected pattern, wherein spacing between features is scaled so that light captured by the scatterometry tool when measuring the one or more modified pitch reference targets is predicted to include a single diffraction order within the selected wavelength range; determining a modified pitch residual between the scatterometry data of the one or more modified pitch reference targets and the model; measuring the one or more selected properties of one or more test pitch reference targets based on regression of scatterometry data of the one or more test pitch reference targets, wherein the one or more test pitch reference targets include features distributed in the selected pattern, wherein spacings between features of the one or more test pitch reference targets correspond to spacings between features of the one or more measurement targets; determining a test pitch residual between the scatterometry data of the one or more test pitch reference targets and the model; generating a first weighted sub-function based on an absolute value of a ratio between the modified pitch residual and the test pitch residual; estimating a captured intensity associated with each of the one or more undesired diffraction orders; estimating a total contamination intensity associated with the captured intensity of each of the undesired diffraction orders; and A second weighted sub-function is generated based on the total pollution intensity.

28. The metering system according to claim 27, wherein the first weighted sub-function W Res (λ) includes: Where R MP (λ) represents the modified pitch residual, R TP (λ) represents the test pitch residual, and λ represents the wavelength within the selected wavelength range, wherein the second weighting subfunction W HO (λ) includes: Among them I C,TOT (λ) represents the total pollution intensity, I0(λ) represents the reference intensity, λ represents the wavelength within the selected wavelength range, and n0 and n1 represent limit parameters, wherein n0 represents an upper threshold and n1 represents a lower threshold, wherein the weighting function W(λ) comprises: W(λ)=αW HO (λ)+(1-α)W Res (λ), where α is a parameter defining the relative weights of the first weighted sub-function and the second weighted sub-function.

29. The metrology system of claim 28, wherein the reference intensity comprises: The noise variance of the scatterometry data.

30. The metering system of claim 28, wherein the total pollution intensity comprises: Where m x and m y represents the diffraction orders in two orthogonal directions, I C (λ,m x ,m y ) represents the contamination level on the collection aperture of the scatterometry tool (m x ,m y )’s average intensity, A Coll represents the area of ​​the collecting pore, and A C (λ,m x ,m y ) represents the contamination area on the collecting pore.

31. A measurement method, comprising: receiving a model for measuring one or more selected properties of a target comprising one or more features distributed in a selected pattern based on regression of scatterometry data from a scatterometry tool within a selected wavelength range; generating a weighting function of the model to de-emphasize a portion of the scatterometry data associated with one or more wavelengths within the selected wavelength range in which light captured by the scatterometry tool when measuring the target is predicted to include one or more undesired diffraction orders; directing the scatterometry tool to generate scatterometry data for one or more measurement targets within the selected wavelength range, wherein the one or more measurement targets include fabricated features distributed in the selected pattern; and The one or more selected properties of the one or more measurement targets are measured based on regressing the scatterometry data of the one or more measurement targets to the model weighted by the weighting function.

Images