Measurement system and method for determining characteristics of one or more structures on a substrate

Abstract

A metrology system and associated method for determining a characteristic of interest associated with at least one structure on a substrate is described. The metrology system includes a processor configured to computationally determine phase and amplitude information based on characteristics of detected scattered radiation, the scattered radiation having been reflected or scattered by the at least one structure as a result of illuminating the at least one structure with illumination radiation in a measurement acquisition, the processor configured to use the determined phase and amplitude to determine the characteristic of interest.

Description

[0001] This application is a divisional application of the international application PCT / EP2018 / 075239, which entered the Chinese national phase on June 5, 2020 and has application number 201880079102.3.

[0002] CROSS-REFERENCE TO RELATED APPLICATIONS

[0003] This application claims priority from European application 17194905.0 filed on October 5, 2017, European application 17199764.6 filed on November 2, 2017, and European application 17206967.6 filed on December 13, 2017, which are incorporated herein by reference in their entirety. Technical Field

[0004] The present invention relates to a metrology system or an inspection system for determining properties of a plurality of structures on a substrate. The present invention also relates to a method for determining properties of a plurality of structures on a substrate. Background Art

[0005] A lithographic apparatus is a machine configured to apply a desired pattern to a substrate. A lithographic apparatus can be used, for example, in the manufacture of integrated circuits (ICs). A lithographic apparatus can project a pattern (often referred to as a "design layout" or "design") from a patterning device (e.g., a mask) onto a layer of radiation-sensitive material (resist) disposed on a substrate (e.g., a wafer).

[0006] To project a pattern onto a substrate, a lithographic apparatus can use electromagnetic radiation. The wavelength of this radiation determines the minimum size of features that can be formed on the substrate. Typical wavelengths currently used are 365 nm (i-line), 248 nm, 193 nm, and 13.5 nm. A lithographic apparatus using extreme ultraviolet (EUV) radiation (with a wavelength in the range of 4-20 nm, such as 6.7 nm or 13.5 nm) can be used to form smaller features on a substrate than a lithographic apparatus using radiation having a wavelength of, for example, 193 nm.

[0007] Low-k1 lithography can be used to process features with dimensions smaller than the classical resolution limit of the lithographic apparatus. In such processes, the resolution formula can be expressed as CD = k1 × λ / NA, where λ is the wavelength of the radiation used, NA is the numerical aperture of the projection optics in the lithographic apparatus, CD is the "critical dimension" (usually the minimum feature size printed, but in this case, half the pitch), and k1 is the empirical resolution factor. In general, the smaller k1 is, the more difficult it becomes to reproduce a pattern on the substrate that resembles the shape and dimensions planned by the circuit designer to achieve specific electrical functionality and performance. To overcome these difficulties, complex fine-tuning steps can be applied to the lithographic projection apparatus and / or the design layout. These steps include, for example, but are not limited to: optimization of the NA, customized illumination schemes, use of phase-shifting patterning devices, various optimizations in the design layout (such as optical proximity correction (OPC, sometimes also referred to as "optical and process correction"), or other methods generally defined as "resolution enhancement techniques" (RET). Alternatively, a tight control loop for controlling the stability of the lithographic apparatus can be used to improve the reproduction of patterns at low k1.

[0008] During photolithography, it is desirable to frequently measure the resulting structures, for example, for process control and verification. Various tools are known for performing such measurements, including scanning electron microscopes and various forms of metrology equipment, such as scatterometers. A general term for such tools may be metrology equipment or inspection equipment. In certain embodiments, a relatively small wavelength range of the visible spectrum is used to inspect structures fabricated on a substrate, necessitating shifting to higher and lower wavelengths and utilizing a wider wavelength range during a single measurement. In certain embodiments, the optical system of the inspection or metrology equipment has a relatively high optical density (NA). In certain embodiments of the optical system, stray light and / or ghosting may be a problem. In certain embodiments, darkfield imaging is used, and in some cases, the amount of recorded noise may be too high. It may be possible to address these requirements and / or problems by improving the quality of the optical components used in the inspection or metrology equipment. For example, aberration requirements over a wide wavelength range can become very stringent. This can render the optical components unmanufacturable or prohibitively expensive for the metrology or inspection equipment. Summary of the Invention

[0009] It is an object to provide an effective and efficient solution for an inspection or metrology system or device that addresses one or more of the problems or limitations discussed above.

[0010] Embodiments of the present invention are discussed in the claims and detailed description sections.

[0011] In a first aspect of the invention, a metrology system is provided for determining a characteristic of interest associated with at least one structure on a substrate, the metrology device comprising: a processor configured to computationally determine phase and amplitude information based on characteristics of detected scattered radiation, the scattered radiation having been reflected or scattered by the at least one structure as a result of illuminating the at least one structure with illumination radiation in a measurement acquisition, the processor configured to use the determined phase and amplitude to determine the characteristic of interest.

[0012] In a second aspect of the invention, a method for determining a characteristic of interest associated with at least one structure on a substrate is provided, the method comprising: computationally determining phase and amplitude information based on characteristics of detected scattered radiation, which has been reflected or scattered by the at least one structure as a result of illuminating the at least one structure with illuminating radiation in a measurement acquisition; and using the determined phase and amplitude to determine the characteristic of interest. BRIEF DESCRIPTION OF THE DRAWINGS

[0013] Embodiments of the present invention will now be described, by way of example only, with reference to the accompanying schematic drawings, in which:

[0014] - Figure 1 depicts a schematic overview of a lithographic apparatus;

[0015] - Figure 2 A schematic overview of a lithography cell is depicted;

[0016] - Figure 3 Depicted is a schematic representation of overall photolithography that shows the collaboration between three key technologies used to optimize semiconductor manufacturing;

[0017] - Figure 4 An inspection device according to an embodiment of the present invention is illustrated;

[0018] - Figure 5 A metrology method using EUV radiation is schematically depicted;

[0019] - Figure 6 schematically depicts an EUV metrology apparatus according to an embodiment of the present invention;

[0020] - Figure 7 The known form of the multi-grating target and the outline of the measurement spot on the substrate are depicted;

[0021] Figure 8 Describes the measurements obtained by methods such as those discussed in this paper. Figure 7 an image of the target; and

[0022] Figure 9A measurement method according to an embodiment of the present invention is schematically depicted. DETAILED DESCRIPTION

[0023] In this document, the terms "radiation" and "beam" are used to cover all types of electromagnetic radiation, including ultraviolet radiation (e.g., having a wavelength of 365 nm, 248 nm, 193 nm, 157 nm or 126 nm) and extreme ultraviolet radiation (EUV, e.g., having a wavelength in the range of about 5-100 nm).

[0024] As used herein, the terms "reticle," "mask," or "patterning device" should be broadly interpreted to refer to a general patterning device that can be used to impart a patterned cross-section to an incident radiation beam that corresponds to the pattern to be produced in a target portion of the substrate. The term "light valve" can also be used in this context. In addition to classical masks (transmissive or reflective; binary, phase-shift, hybrid, etc.), examples of other such patterning devices include programmable mirror arrays and programmable LCD arrays.

[0025] Figure 1 A lithographic apparatus LA is schematically depicted. The lithographic apparatus LA comprises an illumination system (also called illuminator) IL configured to condition a radiation beam B (e.g., UV radiation, DUV radiation, or EUV radiation), a mask support (e.g., mask table) MT configured to support a patterning device (e.g., mask) MA and connected to a first positioner PM configured to accurately position the patterning device MA according to specific parameters, a substrate support (e.g., wafer stage) WT configured to hold a substrate (e.g., a resist-coated wafer) W and connected to a second positioner PW configured to accurately position the substrate support according to specific parameters, and a projection system (e.g., a refractive projection lens system) PS configured to project a pattern imparted to the radiation beam B by the patterning device MA onto a target portion C (e.g., comprising one or more dies) of the substrate W.

[0026] In operation, the illumination system IL receives a radiation beam from a radiation source SO, for example, via a beam delivery system BD. The illumination system IL may include various types of optical components, such as refractive, reflective, magnetic, electromagnetic, electrostatic, and / or other types of optical components, or any combination thereof, to direct, shape, and / or control the radiation. The illuminator IL may be used to condition the radiation beam B so as to have a desired spatial and angular intensity distribution in its cross-section at the plane of the patterning device MA.

[0027] The term "projection system" PS as used herein should be broadly interpreted as covering various types of projection systems, including refractive, reflective, catadioptric, anamorphic, magnetic, electromagnetic and / or electrostatic optical systems, or any combination thereof, as appropriate with regard to the exposure radiation used and / or with regard to other factors such as the use of an immersion liquid or the use of a vacuum. Any use of the term "projection lens" herein may be considered synonymous with the more general term "projection system" PS.

[0028] The lithographic apparatus LA may be of a type in which at least a portion of the substrate may be covered by a liquid having a relatively high refractive index (e.g. water) so as to fill the space between the projection system PS and the substrate W - this is also known as immersion lithography. More information on immersion technology is given in US6952253, which is incorporated herein by reference.

[0029] The lithographic apparatus LA may also be of a type having two or more substrate supports WT (also referred to as a "dual stage"). In such a "multi-stage" machine, the substrate supports WT may be used in parallel, and / or steps to prepare a substrate W for subsequent exposure may be performed on a substrate W on one substrate support WT while another substrate W on another substrate support WT is used to expose a pattern on the other substrate W.

[0030] In addition to the substrate support WT, the lithographic apparatus LA can include a measurement platform. The measurement platform is arranged to hold sensors and / or cleaning devices. The sensors can be arranged to measure properties of the projection system PS or properties of the radiation beam B. The measurement platform can hold multiple sensors. The cleaning devices can be arranged to clean a portion of the lithographic apparatus, such as a portion of the projection system PS or a portion of a system for providing immersion liquid. The measurement platform can be moved beneath the projection system PS when the substrate support WT is away from the projection system PS.

[0031] In operation, the radiation beam B is incident on the patterning device (e.g. mask) MA held on the mask support MT and is patterned by the pattern (design layout) present on the patterning device MA. Having traversed the mask MA, the radiation beam B passes through a projection system PS which focuses the beam onto a target portion C of the substrate W. With the aid of the second positioner PW and the position measurement system IF, the substrate support WT can be accurately moved, for example, in order to position a different target portion C in the path of the radiation beam B at a focused and aligned position. Similarly, the first positioner PM and possibly another position sensor ( Figure 1The patterning device MA is positioned relative to the path of the radiation beam B using mask alignment marks M1, M2 and substrate alignment marks P1, P2. Although the substrate alignment marks P1, P2 are shown as occupying dedicated target portions, they can be located in spaces between multiple target portions. When the substrate alignment marks are located between multiple target portions C, the substrate alignment marks P1, P2 are referred to as scribe lane alignment marks.

[0032] like Figure 2 As shown, the lithography apparatus LA may form part of a lithocell LC, sometimes also referred to as a litho cell or (lithography) cluster, which often also includes equipment for performing pre- and post-exposure processes on a substrate W. Conventionally, this equipment includes a spin coater SC for depositing a resist layer, a developer DE for developing the exposed resist, and, for example, a chill plate CH and bake plate BK for regulating the temperature of the substrate W (e.g., for regulating the solvent in the resist layer). A substrate handling device or robot RO picks up substrates W from input / output ports I / O1 and I / O2, moves the substrates between the various process equipment, and then transfers the substrates W to a loading station LB of the lithography apparatus LA. These devices in the lithocell are often collectively referred to as a coating and developing system and are typically controlled by a coating and developing system control unit TCU, which itself may be controlled by a supervisory control system SCS, which may also control the lithography apparatus LA, for example, via a litho control unit LACU.

[0033] In order to ensure that the substrates W exposed by the lithographic apparatus LA are correctly and consistently exposed, it is desirable to inspect the substrates to measure properties of the patterned structures, such as overlay errors between subsequent layers, line thickness, critical dimensions (CD), etc. To this end, an inspection tool (not shown) may be included in the lithography cell LC. If an error is detected, for example, adjustments may be made to subsequent exposures of the substrate or other processing steps to be performed on the substrate W, particularly if the inspection is performed before other substrates W from the same batch or lot are yet to be exposed or processed.

[0034] An inspection apparatus (which may also be referred to as a metrology apparatus) is used to determine properties of a substrate W, in particular to determine how properties vary between different substrates W or how properties associated with different layers of the same substrate W vary between different layers. The inspection apparatus may alternatively be configured to identify defects on the substrate W and may, for example, be part of the lithography cell LC, or may be integrated into the lithography apparatus LA, or may even be a stand-alone device. The inspection apparatus may measure properties of a latent image (the image in the resist layer after exposure), a semi-latent image (the image in the resist layer after a post-exposure bake step PEB), or a developed resist image (where exposed or unexposed portions of the resist have been removed), or even an etched image (after a pattern transfer step such as etching).

[0035] Typically, the patterning process in the lithographic apparatus LA is one of the most critical steps in the process, which requires that the structures on the substrate W be dimensioned and placed with high accuracy. In order to ensure this high accuracy, three systems can be combined into a so-called "holistic" control environment, such as Figure 3 Schematically depicted. One of these systems is the lithography apparatus LA, which is (substantially) connected to a metrology tool MT (a second system) and a computer system CL (a third system). The key to this "holistic" environment is optimizing the collaboration between these three systems to enhance the overall process window and provide a tight control loop to ensure that the patterning performed by the lithography apparatus LA remains within the process window. The process window defines the range of process parameters (e.g., dose, focus, overlay) within which a particular manufacturing process will produce a defined result (e.g., a functional semiconductor device), and typically allows for variations in process parameters during the lithography process or patterning process within this range.

[0036] The computer system CL can use (a portion of) the design layout to be patterned to predict which resolution enhancement technology to use and perform computational lithography simulations and calculations to determine which mask layouts and lithographic equipment settings achieve the largest overall process window for the patterning process (in Figure 3 Typically, the resolution enhancement technique is arranged to match the patterning possibilities of the lithographic apparatus LA. The computer system CL may also be used to detect where within the process window the lithographic apparatus LA is currently operating (e.g. using input from the metrology tool MT) to predict whether there are defects due to, for example, suboptimal processing (e.g., in the process window). Figure 3 (depicted by the arrow pointing to “0” in the second scale SC2).

[0037] The metrology tool MT may provide input to the computer system CL to enable accurate simulations and predictions, and may provide feedback to the lithographic apparatus LA to identify possible drifts, e.g., in the calibration state of the lithographic apparatus LA (in Figure 3 ) depicted by multiple arrows in the third scale SC3).

[0038] During photolithography, it is desirable to frequently measure the resulting structures, for example, for process control and verification. Various tools are known for performing such measurements, including scanning electron microscopes or various forms of metrology equipment, such as scatterometers. Examples of known scatterometers often rely on the provision of specialized metrology targets, such as underfilled targets (targets in the form of simple gratings or overlapping gratings in different layers, which are large enough so that the measurement beam produces a spot smaller than the grating) or overfilled targets (whereby the illumination spot partially or completely encompasses the target). In addition, the use of metrology tools (e.g., angle-resolved scatterometers that illuminate underfilled targets such as gratings) allows for the use of so-called reconstruction methods, in which the properties of the grating can be calculated by simulating the interaction of scattered radiation with a mathematical model of the target structure and comparing the simulated and measured results. The parameters of the model are adjusted until the simulated interaction produces a diffraction pattern similar to the diffraction pattern observed from an actual target.

[0039] A scatterometer is a versatile instrument that allows measurement of parameters of a lithographic process by placing a sensor in the pupil of the scatterometer objective or in a plane conjugated to the pupil (this measurement is generally referred to as pupil-based measurement), or by placing a sensor in the image plane or in a plane conjugated to the image plane (in which case the measurement is generally referred to as image-based or field-based measurement). Such scatterometers and associated measurement techniques are further described in patent applications US20100328655, US2011102753A1, US20120044470A, US20110249244, US20110026032, or EP1,628,164A, which are incorporated herein by reference in their entirety. The aforementioned scatterometers are capable of measuring multiple targets from multiple gratings in one image using light ranging from soft x-rays and visible light to near-IR wavelengths.

[0040] exist Figure 4A measurement device, such as a scatterometer, is depicted in FIG. The measurement device comprises a broadband (white light) radiation projector 2 that projects radiation 5 onto a substrate W. The reflected or scattered radiation 10 is passed to a spectrometer detector 4, which measures the spectrum 6 of the specularly reflected radiation 10 (i.e., a measurement of the intensity I as a function of wavelength λ). From this data, a processing unit PU can reconstruct the structure or profile 8 that gave rise to the detected spectrum, for example by rigorous coupled wave analysis and nonlinear regression or by comparison with a library of simulated spectra. In general, for the reconstruction, the general form of the structure is known, and some parameters are assumed based on knowledge of the process for manufacturing the structure, leaving only a few parameters of the structure to be determined from the scatterometry data. Such a scatterometer can be configured as a normal incidence scatterometer or an oblique incidence scatterometer.

[0041] EUV spectral reflectometry

[0042] Figure 5 The diagram shows the EUV measurement method, and Figure 6 The EUV metrology equipment 300 is shown. The equipment can be used to measure Figure 1

[0046] An example of an EUV metrology device 244 for measuring parameters of a substrate W processed in a fabrication system of FIG. The illumination radiation used by the EUV metrology device may include radiation in a wavelength range from 0.1 nm to 100 nm, or alternatively in a wavelength range from 1 nm to 100 nm, or alternatively in a wavelength range from 1 nm to 50 nm, or alternatively in a wavelength range from 10 nm to 20 nm.

[0043] exist Figure 5 In FIG, a target T is schematically represented as a one-dimensional grating structure at the origin comprising a spherical reference coordinate system. The X-axis, Y-axis and Z-axis are defined relative to the target. (Of course, in principle any arbitrary coordinate system can be defined, and each component can have its own local reference coordinate system that can be defined relative to the reference coordinate system shown). The direction of the periodicity D of the target structure is aligned with the X-axis. The figure is not a true perspective view, but a schematic illustration only. The XY plane is the plane of the target and substrate, and is shown tilted towards the viewer for clarity, as indicated by the oblique view of circle 302. The Z direction defines a direction N that is orthogonal to the substrate. In FIG. Figure 5 In FIG, one of the incident rays is labeled 304 and has a grazing angle of incidence α. In this example, incident ray 304 (and all incident rays forming radiation spot S) lies substantially in a plane parallel to the XZ plane (i.e., the plane defining directions D and N and represented by circle 306). Reflected ray 308, which is not scattered by the periodic structure of target T, emerges toward the right of the target in the figure and has an elevation angle α.

[0044] To perform spectroscopic reflectometry, the ray 308 and other reflected rays are separated into a spectrum 310 comprising rays of different wavelengths. The spectrum can be generated, for example, using a grazing incidence diffraction grating 312. The spectrum is detected by a spectrum detector 313. The spectrum detector 313, which can be, for example, a CCD image detector having an array of pixels, is used to convert the spectrum into an electrical signal and ultimately into digital data for analysis.

[0045] In addition to spectrum 310, a diffraction order detector 350 can be used to detect higher (non-zero) diffraction orders 352 (e.g., at least the +1 and -1 orders, and possibly other higher orders). Although one diffraction order detector 350 is shown here, more than one higher-order detector can be used; for example, a first higher-order detector for the +1 order and a second higher-order detector for the -1 order. Diffraction order detector 350 can be, for example, a CCD image detector having an array of pixels.

[0046] In a practical system, the spectrum of radiation 304 may experience temporal variations that can interfere with analysis. To normalize the detected spectrum 310 and / or higher diffraction orders 352 relative to these variations, a reference spectrum is captured by reference spectrum detector 314. To generate the reference spectrum, source radiation 316 is diffracted by another diffraction grating 318. The zeroth-order reflected rays from grating 318 form the incident rays 304, while the first-order diffracted rays 320 from grating 318 form the reference spectrum detected by reference spectrum detector 314. An electrical signal and data representing the reference spectrum are obtained for analysis.

[0047] From the measured spectra 310 and / or the higher diffraction orders 352 obtained for one or more values of the angle of incidence α, measurements of properties of the target structure T may be calculated in a manner further described below.

[0048] Go to Figure 6 , providing EUV measurement equipment 300 for Figure 5 The method measures the properties of a measurement target T formed on a substrate W. The various hardware components are represented schematically. The actual implementation of these components can be performed by a person skilled in the art applying a mixture of existing components and specially designed components according to well-known design principles. A support (not shown in detail) is provided for holding the substrate in a desired position and orientation relative to the other components to be described. The radiation source 330 provides radiation to the irradiation system 332. The irradiation system 332 provides an EUV irradiation radiation beam represented by ray 304, which forms a focused radiation spot on the target T. The irradiation system 332 also provides a reference spectrum 320 to the reference spectrum detector 314. Components 312, 313, etc. can be conveniently regarded as a spectrum detection system 333.

[0049] In this example, the substrate W is mounted on a movable support with a positioning system 334, which enables the angle of incidence α of the ray 304 to be adjusted and / or the x, y and z position of the substrate W to be adjusted. In this example, for convenience, it is chosen to tilt the substrate W to change the angle of incidence, while the source 330 and the illumination system 332 remain stationary. In order to capture the reflected ray 308, the detection system 333 is provided with an additional movable support 336, which enables it to be moved by an angle 2α relative to the stationary illumination system, or by an angle α relative to the substrate. In a reflecting grazing incidence system, it is convenient to define the angle of incidence α by reference to the plane of the substrate, as shown in the figure. Of course, it can also be defined as the angle between the direction of incidence of the incident ray I and the direction N normal to the substrate.

[0050] Additional actuators (not shown) are provided for bringing each target T to the location of the focused spot S of the radiation. (From another perspective, the spot is brought to the location of the target.) In practical applications, a series of individual targets or target locations may be measured on a single substrate, as well as on a series of substrates. In principle, it is unimportant whether the substrate and target are moved and reoriented while the illumination system and detectors 313, 350 remain stationary, or whether the substrate remains stationary while the illumination system and detectors 313, 350 move, or whether the relative motion of different components is achieved by a combination of these techniques. The present disclosure encompasses all such variations.

[0051] If you have already referred Figure 5 As depicted, radiation reflected by the target T and substrate W is separated into a spectrum 310 of radiation of different wavelengths before striking a spectral detector 313. Spectral detector 313 and / or diffraction order detector 350 may comprise, for example, a position-sensitive EUV detector, typically an array of detector elements. In either case, the array may be a linear array, but in practice, a two-dimensional array of elements (pixels) may be provided. Spectral detector 313 and / or diffraction order detector 350 may be, for example, a CCD (charge-coupled device) image sensor.

[0052] Processor 340 receives signals from detectors 350, 313, and 314. Specifically, signal ST from spectral detector 313 represents the target spectrum, signal SF from diffraction order detector 350 represents the higher-order diffraction pattern, and signal SR from detector 314 represents the reference spectrum. Processor 340 is capable of subtracting the reference spectrum from the target spectrum to obtain a reflectance spectrum of the target, which is normalized for variations in the reflectance spectrum of the target relative to the source spectrum. The resulting reflectance spectra for one or more incident angles are used in the processor to calculate a measurement of a property of the target (e.g., CD or overlap). Similarly, processor 340 is capable of subtracting the reference spectrum from higher-order diffraction patterns (spectra) 352 to obtain higher-order diffraction patterns, which are normalized for variations in the source spectrum. These higher-order diffraction patterns 352 can be compared in an intensity asymmetry measurement to calculate a measurement of a property of the target (e.g., overlap or focus).

[0053] In practice, the radiation from source 330 can be provided as a series of short pulses, and signals SR and ST can be captured together for each pulse. Before being gathered into the overall reflectance spectrum of the target for the incident angle, the difference signal of each individual pulse is calculated. In this way, the instability of the source spectrum between the correction pulses is obtained. The pulse rate can be thousands or even tens of thousands per second (Hz). For example, the number of pulses gathered to measure a reflectance spectrum can be tens or hundreds. Even with so many pulses, the physical measurement only takes a fraction of a second.

[0054] When applying this EUV spectroscopic reflectometry technique for metrology in semiconductor manufacturing, it is possible to measure targets using small gratings. Multiple diffraction spectra are captured using detectors 350, 313, and 314, while setting the grazing incidence angle α to various values. Using the spectrum detected by spectral detector 313 and a mathematical model of the target structure, reconstruction calculations can be performed to obtain measurements of CD and / or other parameters of interest. Alternatively or additionally, complementary higher diffraction orders detected by diffraction order detector 350 can be compared to determine asymmetries in the target structure, thereby determining one or more relevant parameters of interest, such as overlay, focus, or dose, depending on the target properties.

[0055] In one embodiment of the scatterometer MT, the scatterometer MT is adapted to measure the overlay of two misaligned gratings or periodic structures by measuring the reflection spectrum and / or an asymmetry in the detection configuration, the asymmetry being related to the degree of overlap. Similar methods can be used to measure focus on specialized targets formed with a focus-dependent asymmetry. In the case of overlap, the two (typically overlapping) grating structures can be applied in two different layers (not necessarily continuous layers) and can be formed at substantially the same location on the wafer. The scatterometer can have a symmetrical detection configuration, such as described in commonly owned patent application EP1,628,164A, so that any asymmetry can be clearly distinguished. This provides a simple method for measuring misalignment in gratings. Further examples of measuring overlay error between two layers containing periodic structures by measuring the asymmetry of the target periodic structure can be found in PCT Patent Application Publication No. WO 2011 / 012624 or U.S. Patent Application US 20160161863, which are incorporated herein by reference in their entirety.

[0056] Figure 7An exemplary metrology target T is shown on a substrate W, such as that which can be used to measure overlay. Target T can include the entirety of a composite grating or sub-target 32, 33, 34, 35 formed primarily in resist by a photolithography process, but can also include the entirety of a composite grating or sub-target 32, 33, 34, 35 formed after, for example, an etching process. For overlay applications, the sub-targets 32, 33, 34, 35 can be a pair of similar gratings (same pitch, CD, SWA, etc.) formed in the same location on the wafer, but not necessarily in consecutive layers. The metrology equipment will measure the misalignment between the two overlapping gratings, which is called an overlay measurement. In an embodiment, target T can be designed for dark field measurements using a suitable scatterometer. The dark field target will typically be made smaller than the available illumination spot 31 (a typical target is 5x5 square microns, while the illumination spot diameter is 35 microns). This way, there is enough space to use multiple overlapping sub-targets 32, 33, 34, 35 that can be measured simultaneously, allowing measurement of multiple functions. As shown, the directions of sub-targets 32, 33, 34, 35 can be different so as to diffract the incident radiation in the X and Y directions. In the specific example shown, sub-targets 32 and 34 are X-direction sub-targets biased to +d and -d respectively, and sub-targets 33 and 35 are Y-direction sub-targets biased to +d and -d respectively. Alternatively, measurement in only one direction may require only half of the sub-targets, that is, only the sub-targets corresponding to that direction. Although four sub-targets are shown, another embodiment may include a larger matrix to obtain the desired accuracy. For example, a 3×3 array of 9 composite sub-targets can have biases of -4d, -3d, -2d, -d, 0, +d, +2d, +3d, +4d. Separate images of these sub-targets can be identified in the image captured by the detection system.

[0057] In an embodiment, the asymmetry of the target can then be determined, and therefore, for example, the overlap. This can be done using an image processor and controller by comparing the intensity values obtained for the +1 order and -1 order (or other complementary higher order) of each periodic structure 32-35 to identify any differences in their intensities (i.e., intensity asymmetry). The term "difference" is not intended to refer only to subtraction. The difference can be calculated in the form of a ratio or a sum. The measured intensity asymmetry of multiple periodic structures, together with (if applicable) knowledge of the overlap bias of these periodic structures, is used to calculate one or more performance parameters of the lithography process near the target T. The performance parameter of interest is overlap. Other performance parameters of the lithography process, such as focus and / or dose, can be calculated. One or more performance parameters can be fed back to improve the lithography process, for example, to improve the measurement and calculation processes of the scatterometer itself and to improve the design of the target T.

[0058] More specifically, using the method described in, for example, PCT Patent Application Publication No. WO2011 / 012624 or U.S. Patent Application No. US 20160161863, the entire contents of which are incorporated herein by reference, the overlap between two layers containing sub-targets 32 to 35 can be measured by a method comprising the following steps. In an initial step, by Figure 2 A lithography unit processes a substrate (e.g., a semiconductor wafer) one or more times to create a structure including a target comprising a periodic structure 32-35. In a next step, a first diffraction pattern is obtained from the periodic structure 32-35 using one of the first-order diffraction beams (e.g., a -1 diffraction beam). In an embodiment, a first illumination mode is used. Then, whether by, for example, changing the illumination mode, changing the imaging mode, or by rotating the substrate W 180° in the field of view of the measurement device, another first-order diffraction beam (+1 diffraction beam) can be used to obtain a second diffraction pattern from the periodic structure. Thus, +1 diffracted radiation is captured in the second image. In an embodiment, the illumination mode is changed and a second illumination mode is used. In an embodiment, tool-induced artifacts such as TIS (tool-induced shift) can be removed by measuring in the 0° and 180° substrate directions. The first diffraction pattern and the second diffraction pattern are then compared, for example, by calculating the difference in intensity levels within the diffraction pattern of each sub-target.

[0059] Figure 8 As part of the above method (for example, by using Figure 6 equipment) use Figure 7 An example of an image that can be formed on and detected by a sensor for a target. The black rectangle represents the image field on the sensor, within which the illumination spot 31 on the substrate is imaged into the corresponding circular area 41. Within this image, rectangular areas 42-45 represent images of the small target gratings 32 to 35. If the target is located in the product area, product features may also be visible at the periphery of this image field. The image processor and controller PU use pattern recognition to process these images to identify separate images 42 to 45 of the gratings 32 to 35. This eliminates the need for precise alignment of the images at specific locations within the sensor frame, significantly improving the throughput of the entire measurement device. Once the separate images of the gratings have been identified, the intensities of those individual images can be measured, for example by averaging or summing the intensity values of selected pixels within the identified areas. The intensities and / or other properties of the images can be compared with each other. These results can be combined to measure different parameters of the lithography process. Overlap performance is an important example of such a parameter.

[0060] Embodiments of the present invention discussed below may be implemented in one of the metrology or inspection apparatuses discussed above.Embodiments of the present invention relate to methods and / or metrology or inspection apparatuses for determining a characteristic of interest of at least one structure (e.g., a target) on a substrate.

[0061] In current metrology equipment, spatially incoherent light sources are typically used. To increase photon flux (and thus reduce measurement time), it is desirable to (at least in part) use a coherent light source (such as a white light laser) with an AOTF (acousto-optical tunable filter) to select the measurement wavelength and / or bandwidth. Coherent illumination sources can also allow for smaller illumination spots, which is also beneficial (e.g., to support smaller target sizes or to prevent illumination of structures near the target). Furthermore, the supported wavelength range in current metrology equipment is typically limited to approximately 400 nm to 900 nm.

[0062] In general, suitable, high-quality intensity image measurements require that the optics have low aberration specifications over a wide wavelength range λ, allowing flexibility in selecting the optimal measurement wavelength and / or bandwidth. A wide wavelength range will enable measurements covering a wide range of different materials, stacks, and applications. At the same time, the optics should also have a large numerical aperture (NA) (e.g., NA>0.8) and a large field of view to minimize crosstalk between adjacent structures. Other considerations are a large dynamic range (low ghosting / reflections) and compatibility with the dark-field measurement principle, which suppresses the zeroth order.

[0063] Implementing all of these requirements and desired features in a single metrology device is difficult or impossible, as many of these features impose competing demands on the sensor optics to maintain sufficiently low aberration performance. Specifically, increasing the wavelength range of the illumination radiation significantly beyond the current 400nm to 900nm while meeting other requirements will degrade the aberration performance of the sensor optics. This will result in increased aberrations, which in turn will degrade the detector intensity image quality.

[0064] In particular, it is desirable to facilitate a larger wavelength range, for example, 200 nm to 2000 nm, in combination with a larger field of view (>50 μm). Rather than attempting to achieve this objective optically while maintaining aberration performance, it is proposed to do so by allowing the sensor optics to have larger aberrations. Of course, simply allowing larger aberrations within the sensor optics will have an unacceptable impact on image quality unless some measures are taken to compensate for the effects of these optical aberrations. Therefore, it is proposed to use computational imaging techniques to compensate for the negative effects of relaxing the aberration performance requirements within the sensor optics.

[0065] Thus, it is proposed to use computational imaging in a metrology apparatus to measure structures on a substrate formed using a photolithographic process.

[0066] It should be noted that instead of a metrology device, a metrology system can also be used. For example, measurement and image processing do not have to be performed in the same device. A separate image processing device can be coupled to the metrology device to form a metrology system. It should also be noted that instead of a metrology device or a metrology system, an inspection device or an inspection system can be used. For example, an inspection device including embodiments of the present invention can be used to detect defects in a substrate or defects in a structure on a substrate. In such embodiments, a characteristic of a structure on a substrate can be related to a defect in the structure, the absence of a specific portion of the structure, or the presence of an unwanted structure on the substrate.

[0067] In an alternative embodiment, a metrology device is also proposed, comprising an optical sensor arrangement in which the optics for illumination and detection branches are separated to reduce ghosting, stray light and / or reflections.

[0068] More specifically, a direct measurement based on (e.g., relatively low-quality) intensity measurements, replacing the target detector intensity image (i.e., an image of the illuminated target including the amplitude / intensity of the electric field at the detector) with phase acquisition, is proposed. This describes the interaction between the target and the illuminating radiation in terms of the electric field amplitude and phase. This description can include, for example, a representation of the electric and / or magnetic fields at a plane immediately above the target. In such an embodiment, the illuminated target electric and / or magnetic field images are modeled as equivalent source descriptions using extremely small current dipoles and / or magnetic current dipoles on a surface in a plane parallel to the target (e.g., two-dimensional). Such a plane can, for example, be a plane immediately above the target, such as a plane focused according to the Rayleigh criterion. It should be noted that current metrology equipment detects intensity images at conjugate planes, either directly above or within the target. However, the location of the model plane is unimportant: once the amplitude and phase in one plane are known, they can be propagated forward or backward in time to any other plane (e.g., focused, defocused, or even the pupil plane). The selected plane may be located before the (e.g., non-ideal, aberrated) optics such that, in a re-imaging step following a phase acquisition algorithm, the target can be computationally re-imaged under preferred circumstances (e.g., assuming ideal sensor optics and / or (almost) infinite numerical aperture, and / or a specific (partial) coherence, and / or an optimized pupil (complex) transfer mask / function). Alternatively, the description may include a complex transmission of the target or its two-dimensional equivalent.

[0069] Typically, the measured target has unity magnetic permeability and non-unity permittivity.Thus, in an embodiment it is proposed to represent the combination of target and illumination using only current dipoles and not magnetic dipoles.

[0070] The proposed phase acquisition can be used to obtain additional target information, for example, not only intensity / amplitude, but also phase information and / or an extended wavelength range. Moreover, the proposed phase acquisition can be used to obtain better quality target information, which can be used, for example, to calculate overlap or focus (for example, using existing overlap or focus algorithms). The better quality target information may be related to intensity / amplitude alone, or to both intensity / amplitude and phase. For example, such better quality target information can be generated by aberration correction in the sensor optics in the phase acquisition algorithm and / or by a priori knowledge of the target in the phase acquisition algorithm.

[0071] A (candidate) phase acquisition algorithm has been designed that can be combined with any optical sensor layout since it only requires the evaluation of the forward optical model and the calculation of its derivatives. More details of the design of this phase acquisition algorithm will be described later.

[0072] Alternatively, or in combination with prior art methods, a variety of measurements can be performed. To achieve diversity, the imaging system can be (slightly) modified between multiple measurements. An example of a diverse measurement is full focus stepping, i.e., by obtaining measurements at different focus positions. Alternative methods for introducing diversity include, for example, using different illumination wavelengths or different wavelength ranges, modulating the illumination, or changing the angle of incidence of the illumination on the target between multiple measurements.

[0073] In embodiments, the methods described herein may be performed on a processor forming part of a metrology device, for example, more particularly by executing suitable algorithms for phase acquisition and determination of the characteristic of interest. In this way, the existing imaging detection optics of the metrology device may be replaced by new / other detection optics, a phase acquisition algorithm, and optionally computationally re-imaging the reconstructed electric field (both amplitude and phase) into a detector intensity and / or phase image.

[0074] Figure 9 is a flow chart schematically depicting the method disclosed herein. Figure 9A metrology device 900 is shown that includes sensor optics and a detector 910. As already described, the sensor optics can have increased capabilities (e.g., large wavelength range / field of view / NA) at the expense of aberration performance. During measurement acquisition, an image 920 (e.g., of a target) is captured on the detector 910. However, due to aberrations in the sensor optics, this image 920 may be of insufficient quality. A phase acquisition algorithm 930 is used to determine the amplitude A and phase φ of the electric field at a plane parallel to the target (e.g., directly above the target). Using a forward model of an ideal sensor (e.g., aberrations and coherence), the target is re-imaged 940 to obtain the intensity I of the field at the plane of the detector 910 (in addition to the phase φ). No target model is required. The characteristic of interest 960 is then calculated in a conventional manner using a suitable algorithm 950. For example, the overlap can be calculated based on the intensity asymmetry (intensity difference) in the normal and complementary (positive and negative) higher diffraction order pairs.

[0075] The phase acquisition algorithm can be combined with any of the following three optical sensor / detector layouts:

[0076] • Measurements of targets with the detector located in / near the image plane (also called the field plane).

[0077] o This will likely require the use of an objective lens, since without an objective lens (ie using free-space propagation) the detector would need to be impractically close to the target (eg target to detector distance less than 100 μm for large fields of view).

[0078] • Measurements of targets with the detector located in / near the pupil plane (also called the Fourier plane).

[0079] This can be implemented in two ways:

[0080] o Use a lens between the target and the detector to act as the detection optics.

[0081] o Using free space propagation between target and detector (i.e., in Figure 5 and Figure 6 ), whereby the detector is located in / near the far field.

[0082] • Measurements of targets where the detector is located between the image plane and the pupil plane.

[0083] Detectors may be placed in more than one of these locations, so that any combination of two or more of these layouts is possible by using multiple detectors each located in a different plane.

[0084] It can be shown that having the detector located in the image plane or in the pupil plane will result in comparable photon shot noise performance on the measured target quadrant intensity for high numerical aperture conditions when using the phase acquisition method as described herein.

[0085] An advantage of locating the detector in / near the image plane and / or the pupil plane and / or between the image plane and the pupil plane is that multiple target defocussings (i.e. multiple focus settings) can be used to obtain more information about the same target (diversity measurement). This is (in principle) not possible when the detector is located in / near the far field (i.e. when only free-space propagation is used between the target and the detector).

[0086] In order to obtain a variety of measurement results at different focus settings (all-focus measurements), it is obvious that the distance between the sample (target) and the sensor can be changed between image captures. This can be achieved by shifting the sample. However, this approach leads to a large amount of calculations, because the light is propagated digitally through the entire optical system to obtain the complete electric field at the sample plane. In order to reduce the amount of calculations, the sensor can be shifted instead of the sample. In this way, in order to acquire the field at the detector plane, only free space propagation between the measurement planes is required. For example, this phase acquisition can be done by an iterative Fresnel propagation algorithm, or by the transfer intensity equation (TIE). Now, the detector plane field only needs to be propagated digitally once through the setting to acquire the light field on the object plane.

[0087] However, the required displacement of the sensor is much longer than the displacement of the target. This is proportional to the square of the magnification of the optical system. This results in a significant speed limitation: the sensor has to be translated over a long distance. To address this problem, another all-focus measurement setup has been proposed, which enables all-focus measurements without mechanical displacement of the sample or sensor. In this setup, the focal length of the imaging system is tuned electronically. In this way, the distance between the final optical element and the image plane is fixed. One possible implementation uses an electrically tunable lens (ETL), which is located on the low NA side of the imaging system. Such an ETL has a fast tuning response, which allows for fast all-focus measurements. ETLs are commercially available and, in addition to fast tuning, also offer high transmission bandwidth, low distortion and low cost.

[0088] For the lensless example, an advantage of positioning the detector in / near the pupil plane is that if the distance between the target and the detector is large enough (e.g., ≥50 mm), the aberrations of the detection optics can be negligible, or if the distance is small (e.g., <50 mm), the aberrations are well known and do not require calibration.

[0089] For example, the detection optics aberration calibration can be based on measuring the same target for multiple different rotation angles (Rz) and / or position offsets (in the x and y directions). Similar methods can also be used to calibrate / characterize the illumination spot amplitude, phase and / or coherence distribution (e.g., illumination spot).

[0090] The target can be illuminated using spatially incoherent illuminating radiation (e.g., from a spatially incoherent source), spatially coherent illuminating radiation (e.g., from a spatially coherent source), or spatially partially coherent illuminating radiation (e.g., from a spatially partially coherent source). In the latter case of using partially coherent illuminating radiation, and when the detector is located in / near the image plane, its effects can be modeled using the coherent system summation method. If the detector is located in / near the pupil plane, then Scherr's theorem can provide an alternative modeling method that is less computationally demanding than the coherent system summation method.

[0091] It is typically desirable to maximize the ratio of wavelength to target pitch (hereinafter referred to as the λ / p ratio). However, as the numerical aperture (NA) of the optics increases, the likelihood of capturing diffraction orders higher than the first (e.g., the second) increases accordingly. In current scatterometry systems, this typically corrupts the captured image. The proposed method described herein provides additional opportunities to eliminate the effects of second-order diffraction using computational techniques.

[0092] A first proposal for minimizing the effects of second-order diffraction involves using multiple low-NA lens and detector pairs, for example adjacent to each other, so that the same detector does not capture both diffraction orders. This can be implemented practically via computational imaging systems such as those described herein, as the proposed optics for such systems are relatively simple (e.g., detection optics comprising a double aspheric lens between the target and the detector).

[0093] Another proposal for minimizing the effects of second-order diffraction involves utilizing a phase acquisition algorithm as described herein. Instead of acquiring only one amplitude and phase image of the target in the manner described, it is proposed to simultaneously acquire multiple images of the target—one for each diffraction order. More specifically, it is proposed to simultaneously acquire an amplitude and phase image for each diffraction order capable of propagating to the sensor. In addition to being applicable to scalar conditions, this approach is also applicable to full vector conditions.

[0094] As will be described in further detail below (in the section entitled "Example Phase Acquisition Algorithm"), a regularization of the total variation excitation (i.e., a priori knowledge) can be applied to the amplitude and phase images corresponding to multiple illumination wavelengths and / or illumination polarizations. It is proposed that such regularization can also be applied to the amplitude and phase images generated by multiple diffraction orders propagating to the sensor. It should be noted that propagating one phase and amplitude image per diffraction order is equivalent to obtaining the electromagnetic light field radiated by the sample and the geometric parameters of the target, thereby subsequently splitting the field into multiple diffraction orders by (for example) Fourier decomposition or the light propagation itself.

[0095] It will be appreciated that electromagnetic light fields (intensity and phase) can also be obtained using a holographic setup.

[0096] Whether holographically or using computational phase acquisition, the target can be computationally re-imaged to introduce arbitrary (e.g., selected / optimized) pupil filtering / masking. In this way, it is possible to acquire a (computationally) re-imaged sensed image of the target that includes a selected / optimized combination of diffraction order information derived from the (acquired) target.

[0097] Further extending this, it is possible to use the phase acquisition algorithm to obtain information about the zeroth and higher diffraction orders. This can be done by: a) starting with a high-resolution image and performing an appropriate Fourier transform (Fourier decomposition), thereby filtering the high-resolution image; or b) starting with separate images for each relevant diffraction order and applying additional regularization and / or constraints to the phase acquisition algorithm (e.g., image similarity of different diffraction orders).

[0098] In embodiments, any of the methods described herein may include computationally re-imaging a target electric field (e.g., electric and / or magnetic fields at the target level, including complex target images) under different conditions, such as for different detection optics, aperture obscuration conditions, and / or illumination conditions. This applies whether the complex target image is measured using, for example, computational phase acquisition or directly using holography.

[0099] For example, computationally optimized illumination conditions can include computationally re-imaging with completely incoherent illumination. This typically results in smoother images with fewer ringing artifacts. This can be considered a form of coherence filtering, which can be beneficial in reducing ringing artifacts in the detector intensity image that would otherwise affect the measured average intensity.

[0100] This computational re-imaging technique can also be used to apply apodization digitally. Currently, to reduce crosstalk between multiple structures being measured by engineering the (complex) point spread function of an imaging system, apodization can be applied by means of physical apodizers in (or near) the pupil plane of the imaging system. Such physical apodizers discard light, resulting in additional reflections and losses in the beam path. Digital apodization algorithms also have the advantage that they allow dynamic tuning of the apodization, enabling it to be tailored to a specific target / wafer / device structure combination.

[0101] Additionally, digital propagation and / or tuning of illumination conditions may include optimization of:

[0102] (Digital) illumination coherence: The electric field at the target level is assumed to be fully spatially coherent, partially spatially coherent, or spatially incoherent. This can be achieved, for example, by introducing an illuminator in the re-imaging model. Alternatively, it is possible to modify the spatial coherence of the target directly (e.g., assuming that points in the target are mutually incoherent beyond a certain distance). This is possible when the field at the target level is acquired computationally, so in that sense, no explicit illuminator in the model is needed when acquiring the target.

[0103] (Digital) illumination obscuration; i.e., areas where the target illumination is blocked;

[0104] (Digital) spectral shape of the illumination;

[0105] (Digital) illumination polarization; this requires performing separate measurements for at least two different illumination polarizations, or alternatively, presenting a very good model of the illumination and the target;

[0106] (Digital) polarizers in (digital) optics, which, in the case of a fully vectorial implementation (in which case there is access to the polarization component of the electric field), suppress the above-mentioned crosstalk effects;

[0107] Selecting / optimizing a combination of diffraction orders (as already described); for example, by optimizing pupil filtering / masking to acquire a derived (re-imaged) sensed image of the target that includes the selected / optimized combination of diffraction order information originating from the (acquired) target;

[0108] Optimize or change one or more conditions in the optical path from the target to the sensor. Essentially, anything that can be physically changed in the imaging system can be changed digitally in its digital model; this can include, for example: changing optical details of the detection optics (e.g., changing the NA of the detection optics), changing any other lenses / optical components (e.g., changing the focal length and / or introducing / modifying / removing aberrations), introducing (digitally) filters in the detection branch;

[0109] • Selecting / optimizing different focus settings, e.g., re-imaging at a different focus value than the focus corresponding to the actual measurement. Re-imaging to focus on different layers in an image (e.g., obtained at a single focus setting).

[0110] It is also possible to average multiple images (i.e., average multiple images calculated by means of re-imaging as described above). In the rest of this section, for the sake of brevity, images that are re-imaged in a computationally efficient manner are described as "re-images". Such averaging can be useful, for example, when measuring thick stacks, where the distance between the top and bottom gratings of the overlapping target can be large (e.g., with respect to the depth of focus). In that case, at least one of the edges of the two gratings (i.e., the top and bottom gratings) is always blurred. This leads to process variations that affect the overlay performance. To address this issue, it is proposed, for example, to calculate a re-image at best focus for the top grating and another re-image at best focus for the bottom grating. These two re-images can then be averaged, and the overlay inferred from the averaged re-image (e.g., using the asymmetry strength), since the averaged re-image can be more process-robust.

[0111] Another generalization of this averaging can include applying a technique similar to optical color weighting (OCW) typically used in wafer alignment (e.g., OCW in Menchtchikov et al., "Reduction in overlay error from mark asymmetry using simulation, ORION, and alignment models," Transactions of SPIE, vol. 10587, id. 105870C, p. 10 (2018); incorporated herein by reference). In this proposal, two or more re-images are calculated and combined using weighting. The weighting is such that the sum of the weights is equal to 1, but the individual weights can be negative. Thus, the re-images are summed according to the weights attributed to each re-image, where the sum of the weights is equal to 1, to obtain a single weighted re-image. This makes it possible to train the weights so that the measurement becomes less sensitive to process variations.

[0112] Weighting / averaging can be performed on the following dimensions:

[0113] • Focus (as per the example above).

[0114] ·wavelength.

[0115] ·polarization.

[0116] Spatial coherence.

[0117] Pupil apodization / occlusion.

[0118] The diffraction orders of the target.

[0119] The advantage of weighting intensity reimages over weighting multiple overlap estimates (i.e., one overlap estimate for each individual reimage) (the latter example is most similar to what is done in OCW for wafer alignment) is that it is possible to visually inspect / optimize for favorable features in the averaged reimages. This is advantageous because no (external) absolute overlap reference may be used for training. An example of a favorable feature could be minimal intensity ripple within the overlapping target region of interest.

[0120] Example Phase Acquisition Algorithm

[0121] Loss Function

[0122] In an embodiment, it is proposed to include a priori (target) knowledge in a loss function which forms a starting point to derive / design the phase acquisition algorithm. In such an embodiment, the prior knowledge may be based on certain observations; for example, there are a number of regularities between each of a plurality of images of the target. The plurality of images may be obtained in a single measurement (e.g., a measurement using more than one illumination condition, e.g., a multi-wavelength measurement) or from the described diversity of measurements (different focus levels, etc.). In this context, measuring under different illumination conditions may comprise using illumination radiation, wherein one or more of the wavelength and / or polarization and / or temporal and / or spatial coherence of the illumination radiation is varied between the plurality of measurements. It may be observed that, regardless of the illumination condition / focus setting, each image comprises a flat structure having essentially Figure 8 of the form (assuming the target is Figure 7In this way, it can be observed that each image has the same or very similar position and shape for each region of interest (e.g., 4 rectangular or square ROIs, each quarter of the larger square or rectangular target area), and that each region of interest includes a region of relatively high intensity with a relatively flat intensity distribution. It is proposed in an embodiment to exploit this similarity between multiple images by a generalization of the total variation or vector total variation regularization (i.e., applying an L1 penalty term to the gradient of the target image). The advantage of this vector generalization is that it introduces a coupling between different illumination conditions (this coupling is referred to as wavelength coupling in the following, although it is more generally applicable to the coupling of measurements corresponding to changes in other illumination conditions). This wavelength coupling may be similar to the coupling caused by introducing dispersion models (also called n&k models because they describe the relationship of the refractive index n and / or extinction coefficient k over the entire wavelength), but does not require information such as the thickness of the layer, the complex dielectric constant of the layer, or the geometric parameters of the target to be specified.

[0123] It should be clear that diversity, a priori target knowledge, or both can be used in phase acquisition. With sufficient diversity, it should not be necessary to apply constraints and / or regularization based on a priori target knowledge to ensure convergence. Similarly, by using a priori target knowledge to constrain and / or regularize the phase acquisition algorithm, diversity (e.g., measurements at different defocus levels) will not be necessary. By using diversity and applying a priori target knowledge constraints and / or regularization simultaneously, higher accuracy or better convergence guarantees can be achieved.

[0124] The wavelength coupling resulting from the use of prior knowledge (e.g., as a constraint, regularization, or a hybrid of regularization and constraints) can also be used to suppress coherent imaging artifacts such as image speckle. In conventional imaging systems, a well-known approach to suppress speckle is to increase the bandwidth. This paper proposes to perform phase reconstruction for multiple wavelengths / illumination conditions simultaneously. The resulting wavelength coupling introduces a form of averaging (which is achieved by the proposed prior knowledge regularization). This results in a computationally equivalent speckle reduction in conventional imaging systems by increasing the bandwidth.

[0125] Performing phase reconstruction for multiple wavelengths / illumination conditions also enables the identification and removal of coherent imaging artifacts, which are particularly likely to result from the proposed use of spatially coherent light. Such coherent imaging artifacts can include interactions between adjacent targets such as ringing artifacts and speckle (for the purposes of this disclosure, speckle is classified as a coherent imaging artifact; it is a related effect, albeit arising from path length variations such as surface roughness). Ringing artifacts induced by the optics should be distinguished from true processing effects, which happen to resemble such ringing artifacts in the image. It can be observed that ringing artifacts induced by the optics show large variations in wavelength, while true processing effects do not. Thus, strong wavelength dependence can be used in the image acquisition algorithm to remove most of the effects of optics-induced ringing artifacts and speckle, without suppressing true processing effects, since the ringing-like effects produced by true processing effects have spatial frequencies that are independent of wavelength. One way to achieve this is to use a vector generalization of the proposed total variation or vector total variation regularization. For example, wavelengths may be selected in a recipe creation setting based on optimizing a series of KPIs such as sensitivity, stack sensitivity, process robustness, and grating imbalance.

[0126] In phase acquisition, there are essentially two possible algorithmic approaches at the highest level. The first of these approaches is a "feasibility problem" or projection onto convex and / or non-convex set methods (such as the Gerchberg-Saxton algorithm), and the second approach is an optimization problem approach. Since the proposed forward optical model (see below) does not have a closed form inverse operator, the gradient of the nonlinear forward optical model is used to implement the reverse mapping. Such gradient-based minimization methods may include, for example, gradient descent, Newton's method, Gauss-Newton method, or quasi-Newton method. Therefore, an optimization approach is more appropriate because when the gradient of the nonlinear forward optical model is used to implement the reverse mapping, a feasibility-based or projection-based approach will ultimately lead to an optimization problem. It should also be understood that multiple detector intensity measurements will be used in the proposed phase acquisition algorithm.

[0127] A phase acquisition loss function is proposed having a data fidelity term and a prior information term (i.e., a regularization term). Optionally, one or more constraints may be imposed on the phase acquisition minimization loss function using the prior information. In an embodiment, the data fidelity term comprises the least squares sum of the differences between the modeled intensity and the measured intensity (i.e., the L2 norm). The L2 norm is proposed because the noise in the measured intensity will be close to a Gaussian distribution and least squares minimization will result in a minimum variance unbiased estimate. Additionally, least squares minimization may lead to the possibility of using a Gauss-Newton approximation to the resulting Newton equation in the future (depending on the properties of the regularization term in the phase acquisition loss function).

[0128] For the prior information term (i.e., regularization term) of the phase acquisition loss function, we propose using the L1 norm, more specifically the total variation norm. The total variation norm is related to the contour length of the level set representing the (unknown) target (phase and amplitude). Since the unknown target consists of three separate polarization directions, its measurement may also involve more than one illumination condition. In embodiments, a vector extension of the total variation norm is used. In the specific case described herein, the unknown target is a complex function (i.e., not only intensity, but also amplitude and phase), which means that an additional extension of the vector total variation norm may be appropriate. This proposed additional extension can be based on the fact that the explicit singular values of the Jacobian matrix are equal to the gradient magnitude of the actual vector field. Utilizing the vector extension of the total variation norm has the additional benefit of penalizing the misregistration (i.e., overlap) of the resulting target amplitude and phase between illumination conditions. Note that the prior information term of the phase acquisition loss function operates at the local target level (i.e., pixel neighborhood), which is beneficial from the perspective of algorithm parallelization.

[0129] In an embodiment, the phase acquisition loss function L may be in the following form:

[0130] L=F D +G T (1)

[0131] Fidelity term F D In a specific embodiment, it can be in the following form

[0132]

[0133]

[0134]

[0135]

[0136] In a specific embodiment, the prior information term or regularization term may be in the following form:

[0137]

[0138]

[0139]

[0140]

[0141] in:

[0142] W represents the total number of wavelengths measured;

[0143] S(w) denotes the total number of detector intensity sample positions for wavelength identifier w∈{1,2,…,W};

[0144] U represents the total number of regularized target dipole current sample locations;

[0145] λ w represents the measurement wavelength (or more generally, the illumination condition) for wavelength identifier w∈{1,2,…,W};

[0146] J T，p represents the target dipole current distribution for three polarization directions p∈{x,y,z};

[0147] μ represents the regularization multiplier, whose value (≥0) will be chosen based on the appropriate level of regularization;

[0148] Representation matrix The maximum singular value of

[0149] and represents the partial derivative operators in the target x direction and the target y direction respectively;

[0150] For w∈{1,2,…,W} and s∈{1,…,S(w)} (x w，s ,y w，s , z w，s ) represents the intensity sample location; and

[0151] For u∈{1,…,U}(x u ,y u ,λ w ) represents the target dipole current sample position.

[0152] It should be noted that the above equation (7) only gives one embodiment of the vector total variation norm. Another embodiment may include the nuclear norm, that is, In general, any suitable scalar function of the singular values and / or eigenvalues of the matrix within the square brackets may be used.

[0153] Note also that the most dominant singular value of the real Jacobian matrix (i.e., if it is a real matrix, then it is g T (u)) is equal to the non-negative gradient magnitude. In addition, in the implementation of the phase acquisition algorithm on the processor, the target can be assumed to be periodic and discrete, thus implementing the use of Fast Fourier Transform (FFT) to calculate the Fourier transform in the digital evaluation of the forward optical model optics, and implementing the cyclic difference matrix The use of to calculate the derivative of equation (7) and

[0154] Equations (4), (5), (6) and (7) will now be redefined into their periodic and discrete forms, as described by the following equations (11), (12) and (13). These are more practical forms according to embodiments that can be solved more efficiently and are therefore the forms proposed for use herein;

[0155]

[0156]

[0157]

[0158]

[0159]

[0160]

[0161] in:

[0162] Indicates the phase and amplitude of the acquired target;

[0163] n=3w-2;

[0164] U now represents the total number of Cartesian sample positions of the regularized target dipole current in the x-direction;

[0165] V represents the total number of Cartesian sample positions of the regularized target dipole current in the y direction;

[0166] For w∈{1,…,W} and s∈{1,…,S(w)} (x w，s ,y w，s , z w ) represents (a subset of) the detector intensity Cartesian sample positions, such that all or a subset of said Cartesian sample positions may be selected;

[0167] Extract Matrix n columns;

[0168] Extract Matrix m rows;

[0169] x T Represents a vector x the transpose of ; and

[0170] and Denote a circulant difference matrix which, for each row, has -1s at the diagonal elements and ones at the appropriate off-diagonal elements, with all other elements being zero.

[0171] Example Optimization Algorithm

[0172] The phase acquisition loss function, as described by equation (1), is a convex function with no continuous derivatives due to the L1 regularization term. This simplified example is provided by the equation |x| = σ1([x 0]). Therefore, it is not advantageous to implement equation (10) using a gradient-based optimization algorithm. Instead, it is proposed to implement equation (10) using an approximation-based optimization algorithm, as will be described in this section.

[0173] The optimization algorithm that can be used is the original dual algorithm based on forward-backward. The term forward-backward applies to the separation of the loss function L into a first part with continuous first-order derivatives (e.g., the data fidelity term F D ) and a second part with a non-discontinuous first-order derivative (e.g., the prior information term G T ). Because the prior information term G T The gradient of does not exist, so the gradient is replaced by its approximation operator, which leads to the use of the data fidelity term F D The term “primal-dual” refers to the fact that both the primal and dual optimization problems are solved simultaneously, which is beneficial because G T The approximate operator for is easier to compute in its dual form.

[0174] An example of an algorithm that minimizes the loss function L and thus solves equation (10) above will now be provided. This algorithm is written in pseudocode using notation that is directly connected to the loss function L method (equation (1)):

[0175] 1.

[0176] 2.

[0177] 3.Setτ∈]0,+∞[

[0178] 4.Setφ∈]0,+∞[

[0179] 5.k←1

[0180] 6.while not converged do

[0181] 7.

[0182] 8.

[0183] 9.

[0184] 10.

[0185] 11.Set η k ∈]0,+∞[

[0186] 12.

[0187] 13.

[0188] 14.k←k+1

[0189] 15. End of loop

[0190] 16. Return

[0191] 17. Return

[0192] in:

[0193]

[0194] yes "Transpose"

[0195] express an approximation operator for the scaled conjugate; and

[0196] express Note that the conjugate function is not the same as the complex conjugate. Note that for clarity, the variables have been omitted in the pseudo-code description above. and Subscript T .

[0197] It should be noted that other unknowns may also be introduced as variables; for example: an uncertain focus position of the target, an uncertain aberration function of the detection optics, or uncertain illumination properties (e.g. illumination amplitude, phase, spatial coherence and / or temporal coherence). This would simply lead to the matrix Lining is filled with additional zeros.

[0198] Note that the matrix Iterative modification of (i.e., the matrix becomes a function of the iteration number k), which can be used to further improve the estimation performance.

[0199] The advantage of this particular forward-backward based original dual algorithm is that it requires a simpler approximation operator The calculation of The left-hand multiplication of is maintained by the optimization algorithm and does not need to be considered in the calculation of the approximation operator itself. Another advantage is that there is no matrix The inverse or decomposition of needs to be calculated.

[0200] It can be shown that:

[0201]

[0202]

[0203] in represents the unit sphere of the matrix nuclear norm.

[0204] To the unit ball of matrix nuclear norm The projection can be easily A suitable method for doing this can be found, for example, in Goldluecke, Bastian, Evgeny Strekalovskiy, and Daniel Cremers, Total Variation of Natural Vectors Arising from Geometric Measurement Theory, SIAM Journal of Imaging Sciences 5, No. 2 (2012): 537-563; (see Corollary 3.6, Theorem 3.7, and Figure 7 ), all of which are incorporated herein by reference. It should be noted that in the above example algorithm, the matrix is a complex matrix, whereas the equivalent matrices in the documents incorporated by reference are real matrices. It will be readily apparent to those skilled in the art that this model can be adapted for complex matrices.

[0205] Other regularization options

[0206] The above algorithm and cost function use a specific regularization (i.e., the prior information term G T Prior target knowledge is incorporated into the phase acquisition problem. However, this is only an example of regularization and other methods that benefit from being able to use prior (target) knowledge. A number of such methods will now be described. It should be noted that any of these a priori forms of regularization and / or "model reduction" methods can be used alone or in any combination as an alternative to the above-mentioned regularization or as one or more other constraints to supplement the regularization. Different regularization methods and combinations thereof may be better for measuring different structures (different stacking and / or target properties) and in different applications.

[0207] A first possible option is to utilize regularization based on "minimum description length" (e.g., Kolmogorov complexity). Example embodiments may compress the target electric field image using a Liv-Zempel and / or run-length-like compression scheme and use the length of the resulting compressed bitstream as an indicator. The compression scheme may exploit the knowledge that the target structure is a simple, repeating structure, so regularization may bias solutions with, for example, high Kolmogorov complexity.

[0208] In a somewhat similar manner, regularization based on matrix / tensor rank, or an approximation of matrix / tensor rank, such as the matrix nuclear norm (i.e., the sum of the singular values of a matrix or tensor), can also be used. For example, such matrix rank or nuclear norm regularization can impose a low-rank constraint on the solution. Again, this will bias solutions that imply a complex target structure.

[0209] Another option is that if measurement results and / or simulation data of the target are available in advance (i.e., before performing the actual measurement), this data can be used to generate (for example) a search library, or to generate and utilize projections to a lower dimensional subspace. For example, the search library can be used to reduce the number of degrees of freedom by restricting the potential solutions to only those that are linear combinations of the library images of the target.

[0210] In another option, where hyperspectral illumination radiation is used (e.g., where measurements are made using 100 or more different illumination wavelength and polarization combinations), a general dispersion model may be used to reduce the total number of degrees of freedom. Determining the general dispersion model may include modeling the variation of wavelength / polarization (or propagation direction in a material) with respect to the refractive index (or similar parameter) for each target using measurement data from each wavelength and polarization combination. The model may then include a model of the target as a function of transmission. The a priori regularization function will in this case operate on the input coefficients of such a dispersion model. Since the acquired phase and amplitude depend on the wavelength / polarization, the model is able to output an image for the wavelength and polarization combination of interest. In an embodiment, the dielectric constant function ε r (ω) can take the following forms:

[0211] ε r (ω)=1+χ(ω) (17)

[0212]

[0213]

[0214] And c re is the real offset constant, a m is the dipole complex amplitude, ω mis the dipole eigenfrequency, is the dipole sampling constant, is the dipole time constant. Note that all poles occur as (conjugate) pole pairs (dipoles), where M represents the total number of pole pairs.

[0215] Example Optics Forward Model

[0216] The optics forward model used to calculate the modeled intensity in the data fidelity term FD of equation (1) can be in the form of a vector forward model. An example coherent vector forward model will now be described. Other suitable models may also be used, for example, the model may be modified to enable or be able to properly handle partially spatially coherent or spatially incoherent illumination radiation. In the following description, the subscripts D, P and T refer to the coordinate systems of the detector, pupil and target, respectively. Note that in this geometry, all lenses between the target and the detector (via the aperture stop) have been replaced by far-field free space propagation. However, in embodiments, the model may additionally model the effect of one or more optical elements (e.g., an objective lens) on the scattered radiation to obtain modeled characteristics of the scattered radiation.

[0217] The vector forward model can start with the following information: the electric field at the entrance pupil is locally a plane wave, so only its tangential electric field E component E P，θ and is non-zero and the normal electric field component E P，ρ′ is zero. This also applies to the electric field at the exit pupil when the electric field at the exit pupil is equal to the electric field at the entrance pupil.

[0218] It is convenient to calculate the propagation of the electric field from the exit pupil to the detector in a Cartesian polarization coordinate system, since in that case the directions of the individual polarization components do not depend on the spatial position at the detector.

[0219]

[0220] in

[0221] Ω(k x ,,k y , z)=D(k x， , k y ,z)K(k x， , k y )A(k x， , k y ) (twenty one)

[0222]

[0223]

[0224]

[0225]

[0226] Detector intensity I for the fully coherent case D，coh (x, y, z) can be calculated by the following equation:

[0227]

[0228] Among them, the sword Represents the combination of the transpose operator and the complex conjugation operator.

[0229] To calculate the electric field at the entrance pupil E P , the combination / product of irradiance and target is modeled as having a value equal to J T The current distribution of the (x, y) is an infinitely dense array of infinitesimal current dipoles. Therefore, the Fourier transform relationship describing the diffraction from the target to the entrance pupil can be of the following form:

[0230]

[0231] Combining the above:

[0232]

[0233]

[0234]

[0235]

[0236] Note that the wavefront aberrations (transmission and / or phase) can additionally be included in this expression. If applied in parallel, and using the vertical polarization coordinate system for example, the aberration matrix function can be located in the matrix and Note that, in contrast to the above assumptions, the aberration function is, in general, the pupil coordinate k x , and k y A four-dimensional function of both the spatial coordinates x and y.

[0237] Note that the matrix is a unitary matrix, and the discrete Fourier transform matrix is also a unitary matrix. These two unitary matrices are commuted. Therefore, in equation (29), the multiplication with the matrix can be exchanged with the inverse discrete Fourier transform operation at the appropriate position. In addition, if the goal is to numerically calculate the detector intensity, see equation (26), the multiplication with the matrix can also be omitted because the detector intensity can be calculated using a spherical polarization coordinate system.

[0238] Summarize

[0239] The use of computational imaging (e.g., phase acquisition) in lithography metrology applications on a target (e.g., measuring a characteristic or parameter of interest such as overlay, focus, dose, or one or more geometric parameter values such as critical dimension, sidewall angle, edge placement error, or local critical dimension uniformity) is presented above. Overlay can include a measure of misalignment of structures in different layers and / or within a layer overlap, i.e., a measure of misalignment of structures in a single layer, such as in a double patterning process.

[0240] In an embodiment, the use of a priori (target) knowledge in phase acquisition is proposed to enable relaxed requirements for optical device specifications (e.g., aberration performance). In most current phase acquisition applications, there is little (a priori) knowledge of the subject being measured (e.g., in biological applications), and therefore no (or only limited) a priori (target) knowledge can be exploited in the phase acquisition algorithm. Furthermore, in most known phase acquisition applications, bright-field measurements are typically used, so extending this technique to dark-field illumination is unknown. Applying a regularization function based on vector total variation to one or more complex fields (i.e., amplitude and phase data, rather than intensity data) is currently unknown, as such functions have so far only been thought to be applied to real fields (e.g., intensity images). The one or more complex fields can include a complex field for each electric dipole current component / direction (i.e., x, y, or z) according to the illumination wavelength and polarization combination.

[0241] The advantages of the proposed method include the following:

[0242] Provides wavelength coupling, i.e., coupling of corresponding measurements of a target under different illumination conditions. In μDBO (micro-diffraction-based overlay) applications, this wavelength enables multi-wavelength (or multi-illumination condition) measurements of the same target (i.e., at coupled wavelengths) without the need for a dispersion model (also known as an n&k model). This dispersion model describes the relationship between the refractive index n and / or the extinction coefficient k over the entire wavelength range. However, as mentioned above, a dispersion model can optionally be used to reduce the number of degrees of freedom.

[0243] • Wavelength coupling can be used to suppress coherent imaging artifacts such as image speckle or ringing artifacts by introducing averaging.

[0244] • Unlike many other known phase acquisition algorithms, the phase acquisition algorithm specifically disclosed above has convergence guarantees (under certain suitable conditions not further discussed here).

[0245] The fact that the target electric field replaces the detector intensity image in the above disclosure allows for additional advantages. This applies whether the target electric field is measured using, for example, computational phase acquisition or directly using holography. These additional advantages include:

[0246] • Digital refocusing of the target is possible after it has been measured.

[0247] • The method described above can be used to computationally re-image the target electric field under different conditions; different conditions including, among others: different detection optics, aperture obscuration conditions and / or illumination conditions.

[0248] • In cases such as thin resist targets, the derived phase may provide a better source of overlay information than the derived amplitude / intensity.

[0249] Currently, it is possible to obtain detector intensity images for different wavelengths, bandwidths, illumination coherence states, and / or different illumination polarization states and / or detector analyzer polarization states. It is now also possible to introduce the following additional variations that allow more / different information to be measured from a single target:

[0250] • Defocusing of the target (note that this is not possible when the detector is located in the pupil plane).

[0251] Aberrations and / or speckle and / or spatial coherence and / or temporal coherence disturbances of the illumination spot.

[0252] Detect aberration disturbances in optical devices.

[0253] In typical metrology device sensors, extensive sensor optics are required / used. Computational propagation of the electric and / or magnetic fields allows computational generation of both pupil and field images without requiring measurement of both. This allows for a reduction in sensor optics, which can reduce size and cost, and / or allow multiple sensors to be provided in parallel to further reduce measurement time and / or increase substrate sampling density.

[0254] The specific phase acquisition algorithm described above is well suited for efficient implementation on, for example, a graphics processing unit (GPU). No linear equations need to be solved. The a priori knowledge portion of the loss function can be evaluated on a pixel-by-pixel basis, leading to cumbersome parallelization of that portion of the loss function. For the data fidelity portion of the loss function, the amount of computation is dominated by computing the FFT (Fast Fourier Transform). Furthermore, the phase acquisition algorithm has a small memory footprint.

[0255] Further embodiments are disclosed in subsequent aspects:

[0256] i. A metrology apparatus for determining characteristics of one or more structures on a substrate, the metrology system comprising

[0257] - a sensor detecting a property of electromagnetic radiation being reflected or scattered by said one or more structures,

[0258] -processor, configured as

[0259] - receiving the detected characteristic of said electromagnetic radiation,

[0260] - determining the phase and amplitude of the electric field in the vicinity of one or more structures on said substrate,

[0261] - Using the determined phase and amplitude of the electric field to determine the characteristic.

[0262] ii. The measurement system of aspect i, wherein the processor uses a feasibility problem method, such as, for example, the Gerchberg-Saxton algorithm, to determine the phase and amplitude.

[0263] iii. The measurement system according to aspect i, wherein the processor is configured to determine the phase and amplitude by

[0264] a) defining an optimization problem based on the properties of the electromagnetic radiation being detected, a model of the interaction of the radiation with the one or more structures and knowledge that the radiation impinges on the one or more structures at the moment when the sensor detects the properties of the electromagnetic radiation; and

[0265] b) Finding a solution to the optimization problem.

[0266] iv. The metrology system of aspect iii, wherein the processor is further configured to use knowledge of the one or more structures on the substrate to determine the phase and amplitude.

[0267] v. The metrology system according to aspect iii or iv, wherein regularization is used to incorporate knowledge of the one or more structures into the optimization problem.

[0268] vi. The metrology system of aspect v, wherein the processor is configured to define the optimization problem comprising gradient-based minimization of a loss function.

[0269] vii. The metrology system of aspect vi, wherein the processor is configured to define a loss function comprising the sum of a data fidelity function and a prior knowledge regularization function.

[0270] viii. The metrology system according to any of aspects iii to vii, wherein the processor is configured to use complex numbers in defining the optimization problem and finding the optimum value of the optimization problem.

[0271] ix. The metrology system according to any of the preceding aspects, wherein the sensor is arranged in or near one of: a pupil plane of an optical system, a plane conjugate to the pupil plane of the optical system, and an image plane of the optical system.

[0272] x. A measurement system according to any one of aspects i to ix, wherein the sensor is arranged in or near the far field of the one or more structures on the substrate, and the measurement system is configured to provide free-space propagation for radiation propagating from the one or more structures on the substrate toward the sensor.

[0273] xi. The metrology system according to any of the preceding aspects, comprising an illumination system for illuminating the one or more structures on the substrate, the illumination system comprising a radiation source, the radiation source being one of a spatially coherent light source, a spatially incoherent light source, and a spatially incoherent light source.

[0274] xii. The metrology system according to any of the preceding aspects, wherein the characteristics of one or more structures on the substrate include at least one of the following:

[0275] - overlap values of structures in different layers of said one or more structures,

[0276] - one or more geometric parameter values of the 2D or 3D structure of the one or more structures, such as a critical dimension of the one or more structures, a sidewall angle of the one or more structures, an edge placement error of the one or more structures or a local critical dimension uniformity value of the one or more structures,

[0277] - operating parameters of an apparatus for manufacturing the one or more structures on the substrate, e.g. a focus value relating to the focus of a lithographic apparatus for manufacturing the one or more structures, or a dose value relating to the dose being used by the lithographic apparatus for manufacturing the one or more structures.

[0278] xiii. A method of determining a characteristic of one or more structures on a substrate, the method comprising

[0279] - detecting properties of electromagnetic radiation reflected or scattered by said one or more structures,

[0280] - determining the phase and amplitude of the electric field in the vicinity of said one or more structures on said substrate,

[0281] - Using the determined phase and amplitude of the electric field to determine the characteristic.

[0282] Other embodiments are disclosed in subsequently numbered aspects:

[0283] 1. A metrology system for determining a characteristic of interest about at least one structure on a substrate, the metrology system comprising:

[0284] - A processor configured to:

[0285] Phase and amplitude information is computationally determined from the properties of the detected scattered radiation.

[0286] 2. The metrology system of clause 1, wherein the scattered radiation has been reflected or scattered by the at least one structure as a result of illuminating the at least one structure with illumination radiation in a measurement acquisition.

[0287] 3. The measurement system according to aspect 1 or 2, wherein the processor is further configured to

[0288] The determined phase and amplitude are used to determine the characteristic of interest.

[0289] 4. A measurement system according to any of the preceding aspects, wherein the processor is configured such that computationally determining phase and amplitude information comprises computationally determining the phase and amplitude of an electric field and / or magnetic field and / or source representative of the at least one structure.

[0290] 5. The measurement system according to aspect 4, wherein the processor is further configured to:

[0291] modeling an effect of an interaction between the illuminating radiation and the at least one structure on the scattered radiation to obtain a modeled characteristic of the scattered radiation; and

[0292] The phase and amplitude of the electric field are optimized to minimize the difference between the modeled characteristics of the scattered radiation and the detected characteristics of the scattered radiation.

[0293] 6. A measurement system according to aspect 5, wherein the model is capable of operating to model the effects of the interaction between the irradiating radiation and the at least one structure as a dense array of small current dipoles and / or magnetic current dipoles, wherein optionally, the dense array is an infinitely dense array, and wherein optionally, the small current dipoles and / or magnetic current dipoles are infinitely small.

[0294] 7. The measurement system of clause 6, wherein the dense array of small current dipoles is modeled on a two-dimensional plane.

[0295] 8. A measurement system according to aspects 5, 6 or 7, wherein the measurement system includes one or more optical elements between the at least one structure and the detection plane where the scattered radiation is detected, and the processor is configured to additionally model the effects of the one or more optical elements on the scattered radiation to obtain the modeled characteristics of the scattered radiation.

[0296] 9. The metrology system of any of aspects 5 to 8, wherein the processor is configured to use a priori knowledge of the at least one structure to optimize the phase and amplitude of the electric field.

[0297] 10. The metrology system of aspect 9, wherein the processor is configured to use at least a portion of the a priori knowledge of the at least one structure as regularization and / or constraints for the optimization of the phase and amplitude of the electric field.

[0298] 11. The metrology system of clause 10, wherein the regularization and / or constraints comprise total variation based regularization and / or constraints.

[0299] 12. The metrology system of clause 10, wherein the regularization and / or constraints comprise vector total variation based regularization and / or constraints.

[0300] 13. The metrology system according to any of clauses 10 to 12, wherein the regularization and / or constraints apply coupling between different sets of values for phase and amplitude information of the at least one structure, each set of values relating to a different illumination condition.

[0301] 14. The metrology system of clause 13, wherein the processor is operable to computationally determine the different sets of values for the phase and amplitude information simultaneously, thereby suppressing coherent imaging artifacts such as image speckle and / or ringing artifacts.

[0302] 15. The metrology device according to any one of aspects 10-14, wherein the processor is configured to:

[0303] defining a loss function describing the difference between a characteristic of the modeled scattered radiation and a characteristic of the detected scattered radiation; and

[0304] At least a portion of the prior knowledge of the at least one structure is used as a regularization and / or constraint on the minimization to minimize the loss function.

[0305] 16. A measurement system according to any one of aspects 10 to 15, wherein the processor is configured to define a loss function comprising the sum of a data fidelity function and a priori knowledge regularization function, wherein the data fidelity function describes the difference between the characteristics of the modeled scattered radiation to be minimized and the characteristics of the detected scattered radiation, and the priori knowledge regularization function is capable of operating to perform the regularization and / or constraints on the minimization.

[0306] 17. The metrology system of clause 16, wherein the processor is configured to define the minimization as a gradient-based minimization of the loss function or a Newton-based minimization of the loss function.

[0307] 18. The metrology system of clause 17, wherein the processor is configured to use explicit singular values of a Jacobian matrix associated with the prior knowledge regularization function as gradient magnitudes or approximations of the gradient magnitudes in the gradient-based minimization of the loss function.

[0308] 19. The metrology system of clause 17, wherein the processor is configured to use a suitable function of the singular values and / or eigenvalues of the Jacobian matrix associated with the a priori knowledge regularization function in the gradient-based minimization of the loss function.

[0309] 20. The metrology system of any one of aspects 16 to 19, wherein the processor is configured to:

[0310] modelling the variation of the refractive index of the at least one structure with respect to wavelength and / or polarisation and / or direction of propagation to determine a dispersion model of the structure; and

[0311] The prior knowledge regularization function is applied to input coefficients of the dispersion model.

[0312] 21. The metrology system of any of clauses 10 to 20, wherein the regularization and / or constraints comprise applying a biased minimum description length based regularization or constraint to the optimized complex solution.

[0313] 22. The metrology system of clause 21, wherein the minimum description length based regularization or constraint comprises a Kolmogorov complexity based regularization and / or constraint.

[0314] 23. The metrology system according to any one of clauses 10 to 22, wherein the regularization or constraint comprises applying a low-rank constraint or an approximation thereof to the optimized solution, matrix rank or nuclear norm based regularization and / or constraint.

[0315] 24. A metrology system according to any one of aspects 10 to 23, wherein the processor is configured to use a search library to constrain solutions to the optimization to only those solutions that are linear combinations of library images of the at least one structure included in the search library.

[0316] 25. A measurement system according to any one of aspects 10 to 25, wherein the processor is configured such that the regularization is also based on different sets of values for the phase and amplitude information of the at least one structure, each set of values relating to a different focus setting, the measurement system comprising an electrically tunable lens for changing the focus setting without changing the relative shift between the structure and the sensor.

[0317] 26. The metrology system of any of aspects 5 to 25, wherein the processor is configured to use complex numbers in defining and performing the optimization of the phase and amplitude of the electric field.

[0318] 27. The metrology system of any preceding aspect, wherein the processor is further configured to computationally determine the phase and amplitude information for a plurality of images of the target simultaneously, each image corresponding to a different diffraction order.

[0319] 28. The metrology system according to any of the preceding aspects, wherein the processor is further configured to computationally re-image the measurement acquisition of the at least one structure after measurement to obtain at least one computationally re-imaged image.

[0320] 29. The metrology system of clause 28, wherein computationally re-imaging the measurement acquisition comprises one or more of the following steps:

[0321] digitally altering one or more illumination characteristics, which may include: illumination coherence or coherence at a target level, illumination profile, illumination spectral shape, illumination polarization, and / or digitally applied polarization and / or apodization;

[0322] digitally altering one or more characteristics of the detection optics, which may include: changing the numerical aperture of the detection optics, changing any other characteristic of an optical component of the detection optics, changing aberration characteristics of an optical component of the detection optics, filtering light in the detection optics;

[0323] Digitally change the focus setting of an image.

[0324] 30. The metrology system of aspect 28 or 29, wherein the processor is further configured to:

[0325] computationally re-imaging the measurement acquisition of the at least one structure after measurements for a plurality of different virtual settings of the parameter to obtain a plurality of computationally re-imaged images, and

[0326] The plurality of computationally re-imaged images are averaged to obtain an averaged computationally re-imaged image.

[0327] 31. A measurement system according to aspect 30, wherein the processor is further configured to apply positive, zero or negative weights to each of the computationally reimaged images to obtain a weighted average computationally reimaged image, the sum of the weights for all the computationally reimaged images being 1.

[0328] 32. The metrology system of any preceding aspect, wherein the processor uses a feasibility problem method, such as, for example, the Gerchberg-Saxton algorithm, to determine the phase and amplitude.

[0329] 33. The metrology system of any preceding aspect, further comprising one or more sensors for detecting a characteristic of the scattered radiation after the scattered radiation has been reflected or scattered by the at least one structure.

[0330] 34. The metrology system of aspect 33, wherein the metrology device is configured to prevent a zeroth diffraction order of the scattered radiation from being transmitted towards the one or more sensors.

[0331] 35. A metrology system according to aspect 33 or 34, wherein at least one of the one or more sensors is arranged in or near one of the following: a pupil plane of the optical system or a plane conjugate to the pupil plane of the optical system.

[0332] 36. The metrology system of clause 33 or 35, wherein at least one of the one or more sensors is arranged in an image plane of the optical system or in a plane conjugate to the image plane.

[0333] 37. A measurement system according to aspects 33, 34, 35 or 36, wherein at least one of the one or more sensors is arranged in or near the far field of the at least one structure, and the measurement system is configured to provide free space propagation for the scattered radiation to propagate from the at least one structure toward the sensor.

[0334] 38. The metrology system of any one of clauses 1 to 35, wherein the metrology system is operable to perform measurement acquisitions on the at least one structure at a plurality of different focus levels; and

[0335] The results of each of these measurement acquisitions are used to determine a characteristic of interest.

[0336] 39. The metrology system according to any of the preceding aspects, wherein the processor is further configured to computationally refocus the measurement acquisition of the at least one structure after measurement.

[0337] 40. The measurement system according to any one of the preceding aspects comprises an illumination system for illuminating the at least one structure on the substrate, the illumination system comprising a radiation source, the radiation source being one of: a spatially coherent light source, a spatially incoherent light source, and a spatially partially coherent light source.

[0338] 41. The metrology system of any preceding aspect, configured to determine aberrations and / or speckle disturbances in an illumination profile of the illumination radiation on the at least one structure; and

[0339] Wherein the processor is configured to use the determined aberrations and / or speckle perturbations in determining the characteristic of interest.

[0340] 42. The metrology system of any preceding aspect, configured to determine aberration disturbances in the detection optics; and

[0341] Wherein the processor is configured to use the determined aberration disturbance in determining the characteristic of interest.

[0342] 43. The metrology system of any preceding aspect, wherein the processor is further configured to computationally determine the phase and amplitude for illumination conditions that differ from those actually used when performing the measurement acquisition.

[0343] 44. The metrology system of any preceding aspect, wherein the characteristic of interest comprises one or more of:

[0344] - an overlap value related to a misalignment of substructures in different layers of said at least one structure,

[0345] a focus value related to the focus of a lithographic apparatus for fabricating the at least one structure, and

[0346] A dose value related to a dose used by the lithographic apparatus to fabricate the at least one structure.

[0347] 45. A measurement system according to any one of aspects 1 to 43, wherein the characteristic of interest includes one or more geometric parameter values of the at least one structure, wherein the one or more geometric parameter values include one or more of the following: a critical dimension associated with the at least one structure, a sidewall angle associated with the at least one structure, an edge placement error associated with the at least one structure, or a local critical dimension uniformity value associated with the at least one structure.

[0348] 46. A measurement system according to any of the preceding aspects, wherein the measurement system is capable of operating to perform dark field measurements such that the zeroth order of the scattered radiation is completely or partially blocked, and the phase information and characteristics of interest are determined based on at least one pair of higher positive and negative diffraction orders of the scattered radiation.

[0349] 47. A method of determining a characteristic of interest associated with at least one structure on a substrate, the method comprising:

[0350] Phase and amplitude information is computationally determined from the properties of the detected scattered radiation.

[0351] 48. A method according to clause 47, wherein the scattered radiation has been reflected or scattered by the at least one structure as a result of illuminating the at least one structure with illumination radiation in the measurement acquisition.

[0352] 49. A method according to clause 47 or 48, further comprising using the determined phase and amplitude to determine a characteristic of interest.

[0353] 50. A method according to clause 47, 48 or 49, wherein computationally determining phase and amplitude information comprises computationally determining the phase and amplitude of an electric and / or magnetic field representative of the at least one structure.

[0354] 51. The method of clause 50, comprising:

[0355] modeling an effect of an interaction between the illuminating radiation and the at least one structure on the scattered radiation to obtain a modeled characteristic of the scattered radiation; and

[0356] The phase and amplitude of the electric field are optimized to minimize the difference between the modeled characteristics of the scattered radiation and the detected characteristics of the scattered radiation.

[0357] 52. The method according to aspect 51 comprises modeling the at least one structure as a dense array of small current dipoles and / or magnetic current dipoles on a two-dimensional plane, wherein, optionally, the dense array is an infinitely dense array, wherein, optionally, the small current dipoles and / or magnetic current dipoles are infinitely small.

[0358] 53. A method according to aspect 51 or 52, wherein one or more optical elements are present between the at least one structure and the detection plane in which the scattered radiation is detected, and the method includes additionally modeling the effects of the one or more optical elements on the scattered radiation to obtain the modeled characteristics of the scattered radiation.

[0359] 54. A method as defined in any one of clauses 51 to 53, comprising using a priori knowledge of the at least one structure to optimize the phase and amplitude of the electric field.

[0360] 55. A method according to clause 54, comprising using the a priori knowledge of the at least one structure as a regularisation and / or constraint for the optimisation of the phase and amplitude of the electric field by:

[0361] defining a loss function describing the difference between a characteristic of the modeled scattered radiation and a characteristic of the detected scattered radiation; and

[0362] The prior knowledge of the at least one structure is used as a regularization and / or constraint on the minimization to minimize the loss function.

[0363] 56. A method according to clause 54 or 55, wherein the regularisation and / or constraints apply coupling between different sets of values for phase and amplitude information of the at least one structure, each set of values relating to a different illumination condition.

[0364] 57. A method according to any one of aspects 54 to 56, comprising defining a loss function comprising the sum of a data fidelity function and a prior knowledge regularization function, wherein the data fidelity function describes the difference between the characteristics of the modeled scattered radiation to be minimized and the characteristics of the detected scattered radiation, and the prior knowledge regularization function is capable of operating to perform the regularization and / or constraint on the minimization.

[0365] 58. A method according to any of clauses 54 to 57, wherein the regularization and / or constraints comprise one or more of:

[0366] applying a biased minimum description length based regularization or constraint to the optimized complex solution; or

[0367] Regularization and / or constraints based on matrix rank or nuclear norm of low-rank constraints are applied to the optimized solution.

[0368] 59. A method according to any one of aspects 51 to 58, comprising using complex numbers in defining and performing the optimization of the phase and amplitude of the electric field.

[0369] 60. The method of any of clauses 47 to 59, comprising sensing the scattered radiation, wherein sensing is performed in or near one or more of:

[0370] a pupil plane of the optical system or a plane conjugate to the pupil plane of the optical system; an image plane of the optical system; and / or

[0371] In or near the far field of the at least one structure so as to provide free space propagation for the scattered radiation to propagate from the at least one structure towards the sensor.

[0372] 61. The method according to any one of aspects 47 to 60, comprising:

[0373] performing measurement acquisition of the at least one structure at a plurality of different focus levels; and

[0374] The results of each of these measurement acquisitions are used to determine a characteristic of interest.

[0375] 62. The method according to any of aspects 47-61, comprising computationally refocusing the measurement acquisition of the at least one structure after measurement.

[0376] 63. The method of any of clauses 47-62, comprising computationally determining the phase and amplitude information for a plurality of images of the target simultaneously, each image corresponding to a different diffraction order.

[0377] 64. A method according to any of clauses 47 to 63, comprising computationally determining the phase and amplitude for illumination conditions different from those actually used when performing the measurement acquisition.

[0378] 65. The method of any of aspects 47-64, comprising computationally re-imaging the measurement acquisition of the at least one structure after measuring to obtain at least one computationally re-imaged image.

[0379] 66. The method of clause 65, wherein computationally re-imaging the measurement acquisition comprises one or more of the following steps:

[0380] digitally altering one or more illumination characteristics, the one or more illumination characteristics including: illumination coherence, target coherence, illumination profile, illumination spectral shape, illumination polarization, and / or digitally applied polarization and / or apodization;

[0381] digitally modifying one or more characteristics of the detection optics, the one or more characteristics comprising: changing the numerical aperture of the detection optics, changing any other characteristics of an optical component of the detection optics, changing aberration characteristics of an optical component of the detection optics, filtering light in the detection optics;

[0382] Digitally change the focus setting of an image.

[0383] 67. A method according to aspect 64 or 65, comprising:

[0384] computationally re-imaging the measurement acquisition of the at least one structure after measurements for a plurality of different virtual settings of the parameter to obtain a plurality of computationally re-imaged images, and

[0385] The plurality of computationally re-imaged images are averaged to obtain an averaged computationally re-imaged image.

[0386] 68. A method according to aspect 67, comprising applying a positive weight, a zero weight or a negative weight to each of the computationally reimaged images to obtain a weighted average computationally reimaged image, the sum of the weights for all the computationally reimaged images being 1.

[0387] 69. The method of any one of aspects 47-68, wherein the characteristic of interest comprises one or more of:

[0388] - an overlap value related to a misalignment of substructures in different layers of said at least one structure,

[0389] an overlay value associated with misalignment of substructures in the same layer of at least one structure in the plurality of patterning processes;

[0390] a focus value related to the focus of a lithographic apparatus for manufacturing the at least one structure,

[0391] a dose value related to a dose used by said lithographic apparatus to manufacture said at least one structure; and / or

[0392] One or more geometric parameter values of the at least one structure, wherein the one or more geometric parameter values include one or more of the following: a critical dimension associated with the at least one structure, a sidewall angle associated with the at least one structure, an edge placement error associated with the at least one structure, or a local critical dimension uniformity value associated with the at least one structure.

[0393] 70. A non-transitory computer program product comprising machine-readable instructions for causing a processor to perform the method according to any of clauses 47-69.

[0394] 71. A metrology apparatus for determining a characteristic of interest associated with at least one structure on a substrate, the metrology apparatus comprising at least one of:

[0395] - one or more radiation sources operable to generate radiation in an emission wavelength range at least partially overlapping with the wavelength range from 200 nm to 2000 nm, or alternatively, said emission wavelength range at least overlapping with half of said wavelength range from 200 nm to 2000 nm,

[0396] - an optical system operable to emit or reflect radiation in said wavelength range from 200 nm to 2000 nm, or alternatively, said optical system operable to emit or reflect radiation in at least half of said wavelength range from 200 nm to 2000 nm, or alternatively, said optical system operable to emit or reflect radiation in at least ¾ of said wavelength range from 200 nm to 2000 nm,

[0397] - said optical system is operable to illuminate said structure with a numerical aperture (NA) greater than 0.4, greater than 0.6, greater than 0.7 or optionally greater than 0.8,

[0398] - the optical system is operable to capture reflected radiation and / or scattered radiation with a detection optics subsystem, wherein the detection optics subsystem has a numerical aperture (NA) greater than 0.4, greater than 0.6, greater than 0.7 or optionally greater than 0.8,

[0399] - the optical system has a wavelength greater than the illumination wavelength used Divide the aberration by 20,

[0400] - the optical system is operable to illuminate the structure with a field of view (FoV) larger than 40 by 40 micrometers, or alternatively larger than 50 by 50 micrometers, or alternatively larger than 75 by 75 micrometers,

[0401] - the optical system has a transmission or reflection from said one or more radiation sources towards said structure of at least 25% or optionally at least 75%,

[0402] The optical system has a transmission or reflectivity of at least 70% or optionally at least 75% for reflected and / or scattered radiation from the structure towards one or more sensors for recording properties of the reflected and / or scattered radiation.

[0403] 72. The metrology apparatus of clause 71, further comprising an optical system configured to transmit only one or more higher diffraction orders towards the sensor.

[0404] 73. The metrology apparatus of clause 72, wherein the optical system further comprises blocking means operable to block a zeroth diffraction order reflected by the structure on the substrate when the structure is illuminated with radiation.

[0405] 74. The metrology apparatus according to any one of aspects 71 to 73, further comprising a processor configured to perform the method according to any one of aspects 47 to 69.

[0406] 75. A metrology system according to any one of aspects 1 to 46, comprising a metrology device according to any one of aspects 71 to 74.

[0407] Although specific reference may be made herein to the use of lithographic apparatus in IC manufacturing, it should be understood that the lithographic apparatus described herein may have other applications. Possible other applications include integrated optical systems for the manufacture of guidance and detection patterns for magnetic domain memories, flat panel displays, liquid crystal displays (LCDs), thin film magnetic heads, and the like.

[0408] Although detailed reference may be made herein to embodiments of the present invention in the context of inspection equipment or metrology equipment, embodiments of the present invention may be used in other equipment. Embodiments of the present invention may form part of a mask inspection equipment, a lithographic equipment, or any equipment that measures or processes an object such as a wafer (or other substrate) or a mask (or other patterning device).

[0409] Although specific reference may have been made above to the use of embodiments of the invention in the context of optical lithography, it will be appreciated that the invention is not limited to optical lithography and may be used in other applications, such as imprint lithography, where the context permits.

[0410] While the targets or target structures (more generally structures on a substrate) described above are metrology target structures that are specifically designed and formed for measurement purposes, in other embodiments, the property of interest may be measured on one or more structures that are functional parts of a device formed on the substrate. Many devices have regular grating-like structures. The terms structure, target grating, and target structure as used herein do not require that the structure has been specifically configured for the measurement being performed. Additionally, the pitch P of the metrology target is close to the resolution limit of the scatterometer's optical system, but can be much larger than the size of typical product features fabricated by the microlithographic process of the target portion C. In practice, the lines and / or spaces of the overlapping gratings within the target structure can be made to include smaller structures that are similar in size to the product features.

[0411] While specific embodiments of the present invention have been described above, it will be appreciated that the present invention may be practiced in other ways than those described above. The foregoing description is intended to be illustrative rather than restrictive. Therefore, those skilled in the art will appreciate that modifications may be made to the described invention without departing from the scope of the claims set forth below.

Claims

1. A metrology apparatus for determining a characteristic of interest associated with at least one structure on a substrate, the metrology apparatus comprising: an illumination branch configured to direct illumination radiation toward the substrate; a detection branch configured to detect scattered radiation from the at least one structure on the substrate; as well as A processor configured to: computationally determining phase and amplitude information of the electric field from said scattered radiation in a measurement acquisition, and computationally re-imaging the at least one structure after the measurement acquisition to obtain at least one computationally re-imaged image, and computationally re-imaging the measurement acquisition of the at least one structure after measurements for a plurality of different virtual settings of the parameters to obtain a plurality of computationally re-imaged images, The at least one computationally re-imaged image is obtained without using a structural model.

2. The measurement apparatus of claim 1 , wherein the processor is further configured to use a priori knowledge of the at least one structure to optimize the phase and amplitude of the electric field; and / or The processor is further configured to use at least a portion of the a priori knowledge of the at least one structure as a regularization and / or constraint on the optimization of the phase and amplitude of the electric field; and / or The processor is further configured to use the determined phase and amplitude to determine a characteristic of interest; and / or The processor is further configured to: The plurality of computationally re-imaged images are averaged to obtain an averaged computationally re-imaged image.

3. The metrology apparatus of claim 1 , wherein computationally re-imaging the measurement acquisition of the at least one structure comprises: digitally changing one or more illumination characteristics; and / or Said computationally re-imaging said measurement acquisition of said at least one structure comprises digitally modifying one or more characteristics of the detection optics of said detection branch.

4. The metrology apparatus of claim 3 , wherein the one or more illumination characteristics include: Illumination coherence, illumination coherence at a target level, illumination profile, illumination spectral shape, illumination polarization, and / or digitally applied polarization and / or illumination apodization; and / or The one or more characteristics of the detection optics include: a numerical aperture of the detection optics, optical components of the detection optics, aberration characteristics of optical components of the detection optics, filters of the detection optics and / or a focus setting of the detection optics.

5. The measurement device of claim 2 , wherein the processor is further configured to apply a positive weight, a zero weight, or a negative weight to each of the computationally reimaged images to obtain a weighted average computationally reimaged image, wherein the sum of the weights for all the computationally reimaged images is 1.

6. A method of determining a characteristic of interest associated with at least one structure on a substrate, the method comprising: computationally determining phase and amplitude information of the electric field from the scattered radiation in a measurement acquisition; computationally re-imaging the at least one structure after the measurement acquisition to obtain at least one computationally re-imaged image; as well as computationally re-imaging the measurement acquisition of the at least one structure after measurements for a plurality of different virtual settings of the parameters to obtain a plurality of computationally re-imaged images, The at least one computationally re-imaged image is obtained without using a structural model.

7. The method according to claim 6, further comprising: optimizing the phase and amplitude of the electric field using a priori knowledge of the at least one structure; and using at least a portion of the a priori knowledge of the at least one structure as a regularization or constraint on the optimization of the phase and amplitude of the electric field; and / or The method further includes using the determined phase and amplitude to determine a characteristic of interest; and / or The method further comprises: averaging the plurality of computationally re-imaged images to obtain an averaged computationally re-imaged image; and / or The method also includes applying a positive weight, a zero weight, or a negative weight to each of the computationally reimaged images to obtain a weighted average computationally reimaged image, wherein the sum of the weights for all of the computationally reimaged images is one.

8. The method of claim 6, wherein computationally re-imaging the measurement acquisition of the at least one structure comprises: digitally changing one or more illumination characteristics; and / or wherein said computationally re-imaging said at least one structure said measurement acquisition comprises digitally altering one or more characteristics of detection optics.

9. The method of claim 8, wherein the one or more illumination characteristics include: Illumination coherence, illumination coherence at a target level, illumination profile, illumination spectral shape, illumination polarization, and / or digitally applied polarization and / or illumination apodization; and / or The one or more characteristics of the detection optics include: a numerical aperture of the detection optics, optical components of the detection optics, aberration characteristics of optical components of the detection optics, filters of the detection optics and / or focus settings of the detection optics.

10. A non-transitory computer-readable medium comprising machine-readable instructions for causing a processor to execute the method according to any one of claims 6 to 9.

Images