High-contrast imaging in bonded sample metrology using oblique illumination
The dark-field imaging system with controlled angular illumination and homogenized light beams addresses the low contrast issue in bonded semiconductor samples by preventing specular reflections and enhancing image uniformity for accurate imaging of buried features.
Patent Information
- Application Number
- KR1020267035335
- Authority / Receiving Office
- KR · KR
- Patent Type
- Applications
- Current Assignee / Owner
- Priority Date
- 2024-03-26
- Filing Date
- 2025-03-25
- Publication Date
- 2026-11-09
AI Technical Summary
Current imaging techniques for bonded semiconductor samples suffer from low contrast when imaging buried features due to significant reflections from the top surface and absorption within the samples, limiting the accuracy of imaging systems.
A dark-field imaging system using multidirectional oblique illumination with controlled angular distribution of light beams to prevent specular reflections and enhance image contrast, employing lens assemblies with homogenizers to ensure uniformity and overlap of illumination beams at the buried target.
The system achieves high-contrast imaging of buried features by minimizing specular reflections and ensuring consistent illumination density across the field, improving signal-to-noise ratio and robustness against process variations.
Smart Images

Figure P1020267035335_ABST
Abstract
Description
Technology Field
[0001] The present disclosure generally relates to the measurement of bonded samples, and more specifically, to high-contrast imaging of features within bonded samples. Background Technology
[0002] Many modern semiconductor devices are formed from stacked samples (e.g., wafers, dies, or others)—which may also be referred to as bonded samples. During the fabrication of these devices, it is important to provide accurate imaging of buried features. However, current techniques, which are similar to but not limited to standard bright-field imaging, suffer from low contrast when imaging buried features due to significant reflections from the top surface of the bonded samples and / or absorption within the samples. Therefore, there is a need to develop systems and methods to address the above defects.
[0003] In the embodiments, the techniques described herein comprise an imaging subsystem comprising an objective lens and a detector configured to image a buried target within the volume of a sample; an illumination subsystem—the illumination subsystem comprises one or more illumination sources providing light having a first spatial profile; and two or more lens assemblies configured to direct light from one or more illumination sources as two or more illumination beams toward a buried target along optical paths outside the objective lens, wherein the two or more illumination beams completely overlap at the buried target to provide a common illumination field, and the two or more lens assemblies limit the numerical apertures of the two or more illumination beams to prevent specular reflection of the two or more illumination beams from entering the objective lens, and each of the two or more lens assemblies comprises a homogenizer to provide that each of the two or more illumination beams has a second spatial profile having higher spatial uniformity than the first spatial profile according to a selected metric—; The present invention relates to a darkfield imaging system comprising a controller communically coupled to a detector, wherein the controller comprises one or more processors, and the one or more processors are configured to execute program instructions that cause the one or more processors to receive one or more images of a buried target from the detector and to generate one or more measurements of a sample based on the one or more images.
[0004] In the embodiments, the techniques described herein relate to a dark-field imaging system in which each of one or more light sources comprises bare light-emitting diodes (LEDs), and each of two or more lens assemblies comprises: a field stop—a homogenizer within each lens assembly provides a second spatial profile at the field stop—; and a projection lens for imaging the field stop onto a buried target to provide each of two or more light beams.
[0005] In the embodiments, the techniques described herein relate to a dark-field imaging system in which a homogenizer within each lens assembly includes a waveguide.
[0006] In the embodiments, the techniques described herein relate to a dark-field imaging system in which the waveguide comprises a sealing waveguide.
[0007] In the embodiments, the techniques described herein relate to a dark-field imaging system in which the waveguide comprises a multimode waveguide.
[0008] In the embodiments, the techniques described herein relate to a dark-field imaging system in which the field aperture includes an aperture.
[0009] In the embodiments, the techniques described herein relate to a dark-field imaging system in which a field aperture is located on the output surface of a homogenizer.
[0010] In the embodiments, the techniques described herein relate to a dark-field imaging system in which the output plane of the field aperture and homogenizer is tilted to satisfy the Scheimpflug condition, and the projection lens images the field aperture onto the buried target using an image plane that is coplanar with the buried target.
[0011] In the embodiments, the techniques described herein relate to a dark-field imaging system in which each lens assembly further comprises an aperture stop located in the rear focal plane of the projection lens, and each illumination beam is telecentric to provide uniform angles of incidence across a common illumination field.
[0012] In the embodiments, the techniques described herein relate to a dark-field imaging system in which one or more light sources provide spatially coherent light, and the light subsystem comprises two or more optical fibers for coupling the spatially coherent light to two or more lens assemblies.
[0013] In the embodiments, the techniques described herein relate to a dark-field imaging system in which at least one of one or more light sources comprises a superluminescent diode.
[0014] In the embodiments, the techniques described herein relate to a dark-field imaging system in which at least one of two or more optical fibers is a multimode optical fiber.
[0015] In the embodiments, the techniques described herein relate to a dark-field imaging system in which a multimode optical fiber has at least one of a rectangular, square, and hexagonal core.
[0016] In the embodiments, the techniques described herein relate to a dark-field imaging system comprising: a collimator lens configured such that each lens assembly of two or more lens assemblies collects light from an associated optical fiber among two or more optical fibers; and two Powell lenses having vertices rotated 90 degrees relative to each other.
[0017] In the embodiments, the techniques described herein relate to a dark-field imaging system in which each lens assembly further comprises an additional collimator lens after two Powell lenses.
[0018] In the embodiments, the techniques described herein relate to a dark-field imaging system in which each lens assembly further comprises a field aperture after two Powell lenses.
[0019] In the embodiments, the techniques described herein relate to a dark-field imaging system configured to execute program instructions that cause one or more processors to receive a dark reference image without a sample; to receive a bright reference image of the plane of the sample at the depth of a buried target—wherein the buried target is outside the frame of the bright reference image—and to generate a natural uniformity calibration (NUC) reference image based on the bright reference image and the dark reference image.
[0020] In the embodiments, the techniques described herein relate to a dark-field imaging system comprising generating one or more measurements of a sample based on one or more images, subtracting a NUC reference image from one or more images to form one or more corrected images; and generating one or more measurements based on one or more corrected images.
[0021] In the embodiments, the techniques described herein relate to a dark-field imaging system in which a selected metric comprises at least one of the standard deviation and variance of intensity across a common illumination field.
[0022] In the embodiments, the techniques described herein relate to a dark-field imaging system in which one or more measurements include at least one of an instrumentation measurement, an inspection measurement, and an alignment measurement.
[0023] In the embodiments, the techniques described herein relate to an illumination system comprising: one or more illumination sources providing temporally incoherent light having a first spatial profile; and two or more lens assemblies configured to direct light from one or more illumination sources as two or more illumination beams along optical paths outside the objective lens of an imaging subsystem to a buried target— wherein the two or more illumination beams completely overlap at the buried target to provide a common illumination field, and the two or more lens assemblies limit the numerical apertures of the two or more illumination beams to prevent specular reflection of the two or more illumination beams from entering the objective lens, and each of the two or more lens assemblies includes a homogenizer to provide that each of the two or more illumination beams has a second spatial profile having higher spatial uniformity than the first spatial profile according to a selected metric.
[0024] In the embodiments, the techniques described herein relate to a lighting system in which a selected metric comprises at least one of the standard deviation and variance of intensity across a common lighting field.
[0025] In the embodiments, the techniques described herein include the step of illuminating a buried target within the volume of a sample using two or more illumination beams using an illumination subsystem, wherein the illumination subsystem comprises: one or more illumination sources providing temporally incoherent light having a first spatial profile; and two or more lens assemblies configured to direct light from one or more illumination sources as two or more illumination beams toward a buried target along optical paths outside the objective lens of an imaging subsystem comprising an objective lens and a detector, wherein the two or more illumination beams completely overlap at the buried target to provide a common illumination field, and the two or more lens assemblies limit the numerical apertures of the two or more illumination beams to prevent specular reflection of the two or more illumination beams from entering the objective lens, and each of the two or more lens assemblies includes a homogenizer to provide that each of the two or more illumination beams has a second spatial profile having higher spatial uniformity than the first spatial profile according to a selected metric; and the step of generating one or more images of the buried target from the detector. The invention relates to a dark-field imaging method comprising the step of generating one or more measurements of a sample based on one or more images.
[0026] In the embodiments, the techniques described herein relate to a dark-field imaging method further comprising the steps of: generating a dark reference image without a sample; generating a light reference image of the plane of the sample at the depth of a buried target, wherein the buried target is outside the frame of the light reference image; and generating a natural uniformity calibration (NUC) reference image based on the light reference image and the dark reference image.
[0027] In the embodiments, the techniques described herein relate to a dark-field imaging method comprising the steps of: generating one or more measurements of a sample based on one or more images; subtracting a NUC reference image from one or more images to form one or more corrected images; and generating one or more measurements based on one or more corrected images.
[0028] In the embodiments, the techniques described herein relate to a dark-field imaging method in which one or more measurements include at least one of an instrumentation measurement, an inspection measurement, and an alignment measurement.
[0029] It should be understood that both the general description above and the detailed description below are merely illustrative and descriptive and are not necessarily limiting to the invention as claimed. The accompanying drawings, incorporated into and constituting part of this specification, serve to illustrate embodiments of the invention and, together with the general description, explain the principles of the invention. Brief explanation of the drawing
[0030] Many of the advantages of the present disclosure can be better understood by those skilled in the art by referring to the accompanying drawings. FIG. 1a is a conceptual diagram of a measurement system according to one or more embodiments of the present disclosure. FIG. 1b is a simplified schematic diagram of a measurement system according to one or more embodiments of the present disclosure. FIG. 2a is a side view of a bonded sample according to one or more embodiments of the present disclosure. FIG. 2b is a top view of a first design of a landfill target according to one or more embodiments of the present disclosure. FIG. 2c is a top view of a second design of a landfill target according to one or more embodiments of the present disclosure. FIG. 3 is a simplified schematic diagram of a traditional ring light according to one or more embodiments of the present disclosure. FIG. 4a is a conceptual diagram of a first design of a lighting subsystem according to one or more embodiments of the present disclosure. FIG. 4b is a conceptual diagram of a second design of a lighting subsystem according to one or more embodiments of the present disclosure. FIG. 4c is a conceptual diagram of a third design of a lighting subsystem according to one or more embodiments of the present disclosure. FIG. 5 is a conceptual diagram of a fourth design of a lighting subsystem according to one or more embodiments of the present disclosure. FIG. 6 is a flowchart illustrating steps performed in a method (600) for imaging a buried target (106) according to one or more embodiments of the present disclosure. FIG. 7 is a flowchart illustrating steps performed in a method for correcting an image of a buried target according to one or more embodiments of the present disclosure. FIG. 8 is a simplified schematic diagram of a sample illustrating buried targets to be used as high-resolution reference points in a high-resolution alignment process according to one or more embodiments of the present disclosure. Specific details for implementing the invention
[0031] Now, we will refer in detail to the subject matter of the present disclosure as illustrated in the accompanying drawings. The present disclosure has been presented and described particularly in relation to specific embodiments and specific features thereof. The embodiments presented herein are taken as illustrative rather than restrictive. It will be readily apparent to those skilled in the art that various changes and modifications in form and detail may be made without departing from the spirit and scope of the present disclosure.
[0032] The embodiments of the present disclosure relate to systems and methods for providing high-contrast imaging of buried features within a sample. It is taken into consideration that high-contrast imaging of buried features may be difficult due to specular reflection from the top surface of the sample and / or absorption of light by a material on the features of interest (e.g., between the features of interest and the imaging device).
[0033] In the embodiments, the measurement tool includes an imaging device configured to image buried features within a sample, and a lighting device providing multidirectional oblique illumination (e.g., annular illumination, orthogonal illumination, dipole illumination, or others) having a controlled angle profile customized to prevent the collection of specular reflections by the imaging device. In this way, high-contrast imaging of the buried features can be achieved.
[0034] Under conditions where the top surface of the sample is polished and / or where features of interest are embedded beneath layers of absorbent material, it is becoming increasingly desirable to image features within the volume of the sample. As a non-limiting example, many modern semiconductor fabrication systems utilize bonded sample technology in which multiple components (e.g., wafers and / or dies) are bonded together. For example, a sample may be formed from two bonded wafers (wafer-to-wafer, W2W), two bonded dies (die-to-die, D2D), or a die bonded to a wafer (D2W).
[0035] As an example, it is often desirable to image targets located at or near the interface between these bonded components. For example, bonded samples may include metrology targets, such as but not limited to overlay targets, to facilitate registration measurements between the bonded components. As another example, bonded samples may include alignment targets to facilitate spatial navigation across the samples (e.g., to facilitate precise navigation to metrology targets for measurements). In either case, these targets are typically embedded beneath a substantial amount of semiconductor material (e.g., up to several hundred micrometers thick) that is typically doped and polished on one side. For example, the top surface of the bonded sample may correspond to the back surface of a die or wafer. Additionally, doped semiconductors typically absorb light used in imaging systems.
[0036] As a result, using typical imaging systems can severely limit image contrast. For example, typical through-the-lens (TTL) brightfield imaging, where a common objective lens is used for both illumination and acquisition, can suffer from low contrast due to significant specular reflections from the top surface of the sample that can overwhelm light from buried features of interest. As another example, darkfield TTL imaging can avoid specular reflections, but it may suffer from a limited acquisition NA because a large portion of the numerical aperture (NA) is used by illumination beams and is consequently unavailable for acquisition. As yet another example, outside-the-lens (OTL) illumination (e.g., where different optical systems are used for illumination and acquisition) using typical LED illumination (e.g., a typical LED ring illuminator) can suffer from poor angular uniformity and / or some specular reflections, particularly when used with non-telecentric imaging systems.
[0037] In this invention, it is considered that buried features can be imaged with high contrast using multidirectional outside the lens (OTL) illumination with an angular distribution selected to prevent the collection of specular reflections from the top surface of the sample. Additionally, high image uniformity can be achieved when the illumination has a consistent angular distribution across the illumination field so that the relative contributions of illumination from different directions are consistent across the illumination field. In this way, the illumination density across the illumination field can be increased to improve contrast (e.g., by improving the signal-to-noise ratio associated with the features of interest) while avoiding the negative effects of specular reflections. A uniform angular distribution of light across the illumination field can also beneficially facilitate robust high-contrast imaging of buried features without artifacts, even when slight process variations are present (e.g., variations in sample height during imaging and / or variations in sample properties due to variations in the fabrication process).
[0038] Now, with reference to FIGS. 1a through 8, systems and methods for providing high-contrast imaging of buried features within a sample based on multidirectional illumination having a coherent angular distribution of light across an illumination field are described in more detail according to one or more embodiments of the present disclosure.
[0039] FIG. 1a is a conceptual diagram of a measurement system (100) according to one or more embodiments of the present disclosure.
[0040] In embodiments, the measurement system (100) comprises an imaging subsystem (102) having a detector (104) configured to image at least one buried target (106) within the volume of a sample (108), and further comprises an illumination subsystem (110) configured to illuminate the buried target (106) using a plurality of illumination beams (112) having a controlled angle distribution customized to prevent specular reflection from entering the imaging subsystem (102). In this way, the measurement system (100) can provide dark-field images of the buried target (106) based on diffracted or scattered light from the buried target (106)—referred herein as collected light (114).
[0041] In the present invention, it is considered that image contrast of a buried target (106) can be achieved through strict control of properties of the light beams (112), including but not limited to the size of the light field, the position of the light field, the light numerical aperture (NA), the spatial uniformity of all light beams (112) across the light field, and the angular uniformity of all light beams (112) across the light field.
[0042] In particular, strict control over the illumination NA enables illumination at angles that exclude the capture of specular reflections by the imaging subsystem (102). Strict control over the size and position of the illumination field also provides efficient utilization of light provided by the illumination subsystem (110) and high illumination density. Accordingly, in the embodiments, the illumination subsystem (110) can direct the illumination beams (112) into a common illumination field within the volume of the sample (108) to provide high illumination density.
[0043] Additionally, it is considered herein that uniform image contrast across the image can be achieved by providing multidirectional illumination having a consistent angular distribution along the illumination field such that the ratio of intensity from various illumination beams (112) is constant across the illumination field. In particular, this configuration provides a consistent response from the buried target (106) regardless of the orientation or design of the buried target (106). In the embodiments, a consistent angular distribution across the illumination field is achieved by providing a uniform distribution of illumination from each illumination beam (112) across the illumination field.
[0044] In another embodiment, the measurement system (100) includes a controller (116) that is communicably coupled to the measurement system (100). The controller (116) may be configured to instruct the imaging subsystem (102) to generate dark-field images based on one or more selected recipes. The controller (116) may also be configured to receive data, including but not limited to dark-field images from the imaging subsystem (102). Additionally, the controller (116) may be configured to determine an overlay associated with an overlay target based on the acquired dark-field images.
[0045] In another embodiment, the controller (116) includes one or more processors (118). For example, one or more processors (118) may be configured to execute a set of program instructions maintained in memory (120) or a memory device. One or more processors (118) of the controller (116) may include any processing element known in the art. In this sense, one or more processors (118) may include any microprocessor-type device configured to execute algorithms and / or instructions. Additionally, memory (120) may include any storage medium known in the art suitable for storing program instructions executable by the associated one or more processors (118). For example, memory (120) may include a non-transient memory medium. As additional examples, memory (120) may include, but is not limited to, read-only memory, random access memory, magnetic or optical memory devices (e.g., disks), magnetic tape, solid-state drives, and others. Additionally, note that the memory (120) may be housed within a common controller housing with one or more processors (118).
[0046] FIGS. 2a through 2c illustrate various non-limiting examples of buried targets (106) suitable for imaging by a measurement system (100).
[0047] The sample (108) may include any type of object or material comprising a buried target (106). In some embodiments, the sample (108) comprises one or more substrates (e.g., wafers, or others) formed of semiconductor or non-semiconductor materials. For example, semiconductor or non-semiconductor materials may include, but are not limited to, single-crystal silicon, gallium arsenide, and indium phosphide. The sample (108) may include one or more layers. For example, these layers may include, but are not limited to, resists, dielectric materials, conductive materials, and semiconductor materials. Many different types of these layers are known in the art, and as used herein, the term sample (108) is intended to encompass an object on which all types of these layers may be formed. One or more layers formed on the sample (108) may or may not be patterned. For example, the sample (108) may include a plurality of dies, each having repeatable patterned features. As a result of the formation and processing of these material layers, finished devices can finally be obtained. Many different types of devices may be formed on the sample (108), and as used herein, the term sample (108) is intended to encompass any object on which any type of device known in the art is being manufactured.
[0048] The buried target (106) may be located at any depth within the sample (108). In some embodiments, the buried target (106) is located below one or more sample layers disposed on the substrate. In some embodiments, the sample (108) is a bonded sample formed from two or more bonded substrates. In this configuration, the buried target (106) may include features on layers associated with any one of the substrates.
[0049] FIG. 2a is a side view of a bonded sample (108) according to one or more embodiments of the present disclosure. In particular, FIG. 2a illustrates a first substrate (202) having one or more first substrate layers (204) and a second substrate (206) having one or more second substrate layers (208), together with an interface (210) between the bonded portions. FIG. 2a also illustrates patterned first substrate features (212) within the first substrate layers (204) and second substrate features (214) within the second substrate layers (208). The bonded sample (108) in FIG. 2a may generally correspond to a W2W sample (108), a D2D sample (108), or a D2W sample (108). However, FIG. 2a is provided only for exemplary purposes and should not be interpreted as limiting. Rather, the buried target (106) may include any number of features at any location on any layer of the sample (108) or, more generally, within the volume of the sample (108). Additionally, the sample (108) may include any number of intermediate layers between any of the first substrate features (212) and any of the second substrate features (214) and / or between these features and their respective substrates.
[0050] In some embodiments, the embedded target (106) includes device features (e.g., patterned features that may be part of the device when fully formed). In some embodiments, the embedded target (106) includes dedicated features designed to be imaged by the measurement system (100) for a specific purpose. For example, the embedded target (106) may include features forming an alignment target, which may be suitable for correcting sample coordinates to chuck coordinates after bonding, for aligning the first substrate (202) and the second substrate (206) before bonding, or for any other suitable purpose. As another example, the embedded target (106) may include features forming an overlay target, which may be suitable for registration measurements between the first substrate features (212) and the second substrate features (214). For example, the overlay target may be used to determine registration errors between bonded substrates. Using FIG. 2a as an example, registration measurements (e.g., overlay measurements between first substrate features (212) on the first substrate layers (204) and second substrate features (214) on the second substrate layers (208)) may be used to evaluate the bonding process. In another example, an overlay target may be used to determine registration errors between features on different first substrate layers (204) and / or features on different second substrate layers (208).
[0051] FIGS. 2b and FIGS. 2c provide non-limiting examples of designs of a landfill target (106) according to one or more embodiments of the present disclosure.
[0052] FIG. 2b is a top view of a first design of a buried target (106) according to one or more embodiments of the present disclosure. In FIG. 2b, the first substrate features (212) include cross-shaped features and the second substrate features (214) include box features. This design may be particularly well suited for use as an alignment target, but is not limited thereto.
[0053] FIG. 2c is a top view of a second design of a buried target (106) according to one or more embodiments of the present disclosure. In FIG. 2b, the first substrate features (212) and the second substrate features (214) each include periodic features. Note that the first substrate features (212) and the second substrate features (214) are illustrated in FIG. 2c as having different periods, but this is merely exemplary and not limiting. Additionally, the number and orientation of any of the features are also merely exemplary and not limiting. This design may be particularly well suited for use as an overlay target, but is not limited thereto. Furthermore, the design illustrated in FIG. 2c may be characterized as an advanced imaging metrology (AIM) target.
[0054] Now, referring to FIG. 1b and FIG. 3 through 5, imaging using highly controlled illumination beams (112) according to one or more embodiments of the present disclosure is described in more detail.
[0055] FIG. 1b is a simplified schematic diagram of a measurement system (100) according to one or more embodiments of the present disclosure.
[0056] In embodiments, the imaging subsystem (102) includes various components configured to generate an image (e.g., a dark-field image) of a buried target (106) on a detector (104). For example, the imaging subsystem (102) may include an objective lens (122) for collecting collected light (114). The imaging subsystem (102) may further include one or more imaging lenses (124) for manipulating the collected light (114) and providing an image of the buried target (106) on the detector (104). As an example, although not explicitly illustrated, the imaging lenses (124) may relay one or more field planes and / or pupil planes. The imaging subsystem (102) may further include one or more imaging optical components (126) for manipulating the collected light (114), such as but not limited to one or more filters, one or more polarizers, one or more beam blocks, or one or more beam splitters. The imaging optical components (126) may be located at any suitable locations, including but not limited to field and / or pupil planes. The imaging lenses (124) and the imaging optical components (126) together can manipulate the collected light (114) to provide a desired dark-field image of the buried target (106).
[0057] The lighting subsystem (110) may include any number or type of lighting sources (128) for generating light associated with one or more lighting beams (112). In embodiments, the lighting source (128) includes one or more LEDs, which may provide light that is temporally and spatially incoherent. In embodiments, the lighting source (128) includes a superluminescent diode (SLD), which may provide light that is temporally incoherent but spatially coherent. In embodiments, the lighting source (128) includes a laser-sustained plasma (LSP) source. For example, the lighting source (128) may include, but is not limited to, an LSP chamber, an LSP bulb, or an LSP lamp suitable for containing one or more elements capable of emitting broadband light when excited to a plasma state by a laser source. In embodiments, the lighting source (128) includes a lamp source. For example, the light source (128) may include, but is not limited to, an arc lamp, a discharge lamp, an electrodeless lamp, or others.
[0058] The illumination beams (112) may include any desired spectral component. For example, the illumination beams (112) may include wavelengths within any spectral range, such as but not limited to the ultraviolet spectral range, the visible spectral range, the infrared (IR) spectral range, the short-wave infrared (SWIR) spectral range, the near IR spectral range, the mid IR spectral range, or the far IR spectral range. Additionally, the illumination beams (112) may have any bandwidth. In this way, the illumination beams (112) may be characterized as white light, broadband light, supercontinuous light, or narrowband light.
[0059] In some embodiments, the illumination beams (112) provided by the illumination subsystem (110) are temporally and / or spatially incoherent, which can avoid undesirable speckle patterns in the generated images. It is recognized herein that the coherence of light may be characterized by a characteristic temporal or spatial scale, beyond which interference is typically considered negligible. In this way, the term incoherent light may include having a characteristic temporal and / or spatial scale selected to prevent interference (at least within selected tolerances) on the sample (108) and / or detector (104).
[0060] In embodiments, the measurement system (100) includes a translation stage (130) for fixing and positioning a buried target (106) with respect to a lighting subsystem (110) and / or an imaging subsystem (102). The translation stage (130) may include any number or type of actuators, including but not limited to linear actuators (e.g., 3-axis linear actuators), rotary actuators, or tip / tilt actuators.
[0061] In the embodiments, the illumination subsystem (110) includes two or more lens assemblies (132) for forming two or more illumination beams (112) and directing the two or more illumination beams (112) into a common illumination field. In other words, the lens assemblies (132) can provide that the illumination beams (112) completely overlap at the location of the buried target (106) within the volume of the sample (108).
[0062] In the embodiments, each of the lens assemblies (132) includes a homogenizer (134) for homogenizing the spatial distribution of the corresponding illumination beam (112). For example, the homogenizer (134) may receive light from an illumination source (128) having a first spatial distribution and manipulate the light to provide an illumination beam (112) having a second spatial distribution different from the first spatial distribution, wherein the second spatial distribution has a higher spatial uniformity than the first spatial distribution according to a selected metric. Any suitable metric may be used, including but not limited to standard deviation, variation, or others.
[0063] In the embodiments, the imaging subsystem (102) provides two or more illumination beams (112) having different azimuthal angles of incidence, such as those measured from the surface normal of the sample (108) at common pole angles of incidence. For example, the illumination subsystem (110) may provide illumination beams (112) having a dipole distribution, a quadrupole distribution, or a ring distribution having any number of illumination beams (112). As an example, FIG. 1b illustrates two illumination sources (128) providing symmetrical illumination beams (112) at opposing azimuthal angles of incidence and common pole angles of incidence, but this does not limit the number or orientation of the illumination beams (112).
[0064] Additionally, each of the lens assemblies (132) may limit the numerical aperture of the corresponding illumination beam (112). For example, each of the lens assemblies (132) may limit the numerical aperture of the corresponding illumination beam (112) so that specular reflection of the corresponding illumination beam (112) does not enter the objective lens (122) of the imaging subsystem (102). In this way, specular reflection from the top surface of the sample (108) and / or from the buried target (106) can be prevented. In some embodiments, the lens assemblies (132) are configured together so that the angular illumination profile is constant across the illumination field within a selected tolerance (e.g., standard deviation, variance, or others). For example, the lens assemblies (132) may provide that the relative contribution of light as a function of angle is uniform across the illumination field. In other words, for each location within the illumination field, the ratio of the intensities of two or more illumination beams (112) is constant. In the present invention, it is considered that such a uniform angular distribution of light across the illumination field can mitigate hazing or angle-induced fluctuations in the image of the buried target (106) generated by the imaging subsystem (102).
[0065] FIGS. 3 to 5 illustrate various dark-field imaging configurations according to one or more embodiments of the present disclosure.
[0066] Dark-field imaging relies on light deflected by features of interest through mechanisms such as, but not limited to, scattering or diffraction. In an ideal dark-field imaging system, contrast can generally be improved by increasing the illumination intensity or fluence across the illumination field. For example, the illumination intensity or fluence across the illumination field and the resulting image contrast can be increased by increasing the output power of the illumination beams (112), by decreasing the size of the illumination field for given power levels, and / or by ensuring high overlap between the illumination fields of the various illumination beams (112). In some applications, it may be more desirable to provide illumination with a consistent angular distribution across the illumination field, where the relative contributions of the various illumination beams (112) are consistent across the illumination field. This configuration can advantageously avoid image non-uniformity between features of the buried target (106) having different orientations.
[0067] However, actual systems may experience unwanted signals that can reduce image contrast and / or introduce image non-uniformity. As an example, FIG. 3 is a simplified schematic diagram of a traditional ring light according to one or more embodiments of the present disclosure. In FIG. 3, a ring of LEDs (302) is arranged around an objective lens (304) and projects light (306) onto a sample (308). As illustrated in FIG. 3, a typical LED illuminator can provide highly divergent light (306) with a relatively large angular distribution, which can result in a relatively large illumination field (310) on the sample (308). However, this configuration may result in the collection of some specular reflection from the top surface (312) of the sample (108). For example, FIG. 3 illustrates a ray (314) of light (306) associated with specular reflection from the top surface (312) of a sample (308) collected by an objective lens (304), which can have a negative effect on the image contrast of buried features, especially when using a non-telescentric imaging configuration.
[0068] Although not explicitly illustrated in FIG. 3, a typical LED (302) in a typical ring illuminator may also provide a highly non-uniform intensity of light (306) (e.g., a non-uniform illumination field) across a sample (308). For example, light from a typical LED (302) may have a Gaussian profile on the sample (308). In some cases, a typical ring illuminator may attempt to mitigate this field non-uniformity from the individual LEDs (302) by arranging the illumination fields of each LED (302) to overlap only partially in a pattern that provides a relatively uniform total intensity across at least a portion of the sample (308). However, this configuration may not be desirable for certain applications, such as, but not limited to, alignment or measurement. In particular, poor angular uniformity of illumination across the sample (308) (e.g., variation in relative intensity from individual LEDs (302) at different angles) can result in spatially varying image contrast variations for features having different orientations.
[0069] FIGS. 4a through 5 illustrate non-limiting examples of an illumination subsystem (110) suitable for high-contrast imaging of a buried target (106) according to one or more embodiments of the present disclosure. In particular, FIGS. 4a through 4c illustrate configurations incorporating bare LED illumination sources (128) together with lens assemblies (132) arranged to provide illumination beams (112) having high spatial and angular uniformity, whereas FIG. 5 illustrates a configuration incorporating SLDs and lens assemblies (132) arranged to provide collimated illumination beams (112) delivered by optical fibers.
[0070] FIG. 4a is a conceptual diagram of a first design of a lighting subsystem (110) according to one or more embodiments of the present disclosure.
[0071] In embodiments, the measurement system (100) comprises two or more light sources (128) composed of bare LEDs coupled to a field aperture (402) formed as an opening and a homogenizer (134). In this configuration, the homogenizer (134) within the lens assembly (132) provides a spatially uniform distribution of light at the opening of the field aperture (402) within the lens assembly (132). More specifically, the homogenizer (134) can receive a first spatial distribution of light and provide a second spatial distribution of light at the opening of the field aperture (402) that is more uniform than the first spatial distribution (e.g., based on a standard deviation metric, a dispersion metric, or any other suitable metric). Then, the lens assembly (132) may include a projection lens (404) configured to project (e.g., image) light from the field aperture (402) onto the plane of the buried target (106) within the volume of the sample (108). It is considered herein that this configuration may limit the angular distribution of light within each illumination beam (112) (e.g., to create a controlled angular profile), which may be customized to prevent specular reflection from entering the imaging subsystem (102) and, more generally, to facilitate high-contrast imaging.
[0072] Each homogenizer (134) may include any component or combination of components suitable for homogenizing the spatial distribution of light from a corresponding light source (128). In some embodiments, as illustrated in FIG. 4a, the homogenizer (134) includes a sealing waveguide. For example, the homogenizer (134) may include a glass structure (e.g., a glass cap) configured to contain light from the light source (128) and direct this light toward the top surface (312). Additionally, this sealing waveguide may include one or more coatings on the input and / or output surfaces to provide a desired level of power efficiency. In some embodiments, the homogenizer (134) includes a multimode optical fiber. This multimode optical fiber may have any core shape. For example, the multimode optical fiber may have a square shape, a rectangular shape, a hexagonal shape, or others configured to facilitate the spatial homogenization of the guided light. In some embodiments, the homogenizer (134) is provided by positioning the field aperture (402) at a selected distance from the light source (128). It is considered herein that this configuration may provide some spatial homogenization of light from the light source (128) at the cost of some light loss. However, in some applications, such loss may be acceptable.
[0073] The field aperture (402) may have any size or configuration. In a general sense, the size of the field aperture (402) (e.g., the size of the opening forming the field aperture (402)) can control the size of the illumination field. Thus, the field apertures (402) associated with different lens assemblies (132) may all have a common size to provide that the corresponding illumination beams (112) all have a common illumination field.
[0074] In the embodiments, the field aperture (402) is located on the output end of the homogenizer (134). For example, FIG. 4a illustrates a configuration in which the field aperture (402) is located on the output end of the sealing waveguide homogenizer (134). Similarly, the field aperture (402) may be located on the output end of the homogenizer (134) formed as a multimode optical fiber.
[0075] The projection lens (404) may include one or more lenses (e.g., a single lens or a composite lens assembly) or other focusing elements suitable for projecting (e.g., imaging) the field aperture (402) as an illumination beam (112) onto the plane of the buried target (106) within the volume of the sample (108). In this way, the distribution of light associated with the illumination beam (112) in the illumination field (e.g., on the plane of the buried target (106)) may correspond to the distribution of light in the field aperture (402).
[0076] In embodiments, the lens assembly (132) includes an aperture diaphragm (406) for controlling the numerical value (e.g., a range of angles of incidence) of the associated illumination beam (112). For example, the aperture diaphragm (406) may be formed as an aperture, wherein the size of the aperture controls the numerical value of the illumination beam (112). In some applications, the opening size of the aperture diaphragm (406) may be optionally controlled to minimize specular reflection and / or undesirable scattering entering the objective lens (122) of the imaging subsystem (102) in order to promote high-contrast imaging.
[0077] In some embodiments, the angle of incidence of the light beam (112) is adjustable. For example, the light subsystem (110) may include an angle control stage (136) for positioning a lens assembly (132) or a part thereof. Such an angle control stage (136) may include a rotary actuator, a tip / tilt actuator, a linear actuator, or any combination thereof. As an example, FIG. 4a illustrates an angle control stage (136) associated with each light source (128) and the associated lens assembly (132).
[0078] FIG. 4b is a conceptual diagram of a second design of a lighting subsystem (110) according to one or more embodiments of the present disclosure. FIG. 4b shows that the aperture diaphragm (406) is at the rear focal plane of the projection lens (404) (e.g., from the projection lens (404) The distance of - here It is substantially similar to FIG. 4a, except that it is positioned at the focal length. This configuration provides telecentric lighting, wherein the associated lighting beam (112) illuminates all points in the lighting field at the same angle of incidence (e.g., the same angle of inclination).
[0079] FIG. 4c is a conceptual diagram of a third design of a lighting subsystem (110) according to one or more embodiments of the present disclosure. FIG. 4b is substantially similar to FIG. 4b except that a field aperture (402) is tilted according to a Scheimpflug condition to provide a correspondingly tilted lighting field.
[0080] For example, the configurations of the lighting subsystem (110) illustrated in FIGS. 4a and 4b provide that both the field aperture (402) and the projection lens (404) are oriented perpendicular to the optical axis (138). In these configurations, the distribution of light in the field aperture (402) is projected by the projection lens (404) onto an image plane within the sample (108) that is also perpendicular to the optical axis (138). However, since the optical axis (138) is tilted relative to the buried target (106), the image of the field aperture (402) may be out of focus across the entire lighting field (e.g., across the buried target (106)). As a result, the distribution of light across the buried target (106) may exhibit non-uniformity, where the degree of non-uniformity may depend on factors such as the depth of field provided by the projection lens (404) or the tilt angle of the optical axis (138) relative to the buried target (106) (e.g., relative to the surface normal of the sample (108)), but is not limited thereto.
[0081] In contrast, FIG. 4c illustrates a configuration in which the image plane provided by the projection lens (404) is tilted so as to be coplanar with the buried target (106). As a result, the image of the field of view aperture (402) can be focused across the entire illumination field (e.g., across the buried target (106)).
[0082] In particular, the field of view aperture (402) is inclined on the buried target (106). and magnification When imaged, the Scheimpflug condition is the tilt angle of the field aperture (402). is satisfied when the following is satisfied:
[0083] . (1)
[0084] Now, referring to FIG. 5, in some embodiments, the lighting subsystem (110) provides collimated lighting beams (112). For example, a lighting source (128) providing spatially coherent (but perhaps temporally incoherent) light may be collimated to provide lighting beams (112) of very high spatial and angular uniformity. Any suitable spatially coherent lighting source (128), including but not limited to an SLD, may be used.
[0085] FIG. 5 is a conceptual diagram of a fourth design of a lighting subsystem (110) according to one or more embodiments of the present disclosure.
[0086] In FIG. 5, an SLD high-power light source (128) is used with a driver and a thermal controller to provide a stable short-wave infrared (SWIR) light beam (112). For example, the light beam (112) may have a center wavelength of approximately 1310 nm and a bandwidth of approximately 100 nm full-width half-maximum (FWHM). Additionally, light from this light source (128) may be transmitted using optical fibers or free-space optical systems. As an example, FIG. 5 illustrates a configuration in which a single SLD light source (128) is coupled to an optical fiber (502), wherein one or more splitters (504) split the light into multiple paths associated with a desired number of light beams (112). Any type of optical fiber (502), including but not limited to multimode or single-mode optical fibers (502), may be used. As another example, although not illustrated, the lighting subsystem (110) may include a separate lighting source (128) for each lighting beam (112).
[0087] In embodiments, the lens assembly (132) operates as both a homogenizer (134) and a collimator to provide an illumination beam (112) having high spatial and angular uniformity. For example, FIG. 5 illustrates a configuration in which each lens assembly (132) comprises a homogenizer (134) formed by two Powell lenses (508) having vertices rotated 90 degrees relative to each other, following a collimator lens (506) (e.g., a fiber collimator) for collimating light from an optical fiber (502). It is taken into account herein that the spatial profile of the light exiting the optical fiber (502) may depend on the mode profile of the guided light. In some optical fiber designs, the spatial profile may be Gaussian. Thus, the homogenizer (134) (here, illustrated as Powell lenses) can increase the spatial uniformity of the collimated light. FIG. 5 also illustrates an additional lens (510) to ensure collimation after the homogenizer (134) and a field aperture (512) to control the diameter of the illumination beam (112) and the consequent spot size on the buried target (106). It is considered herein that such a configuration may provide exceptionally high spatial and angular non-uniformity to facilitate high-contrast imaging of the buried target (106) as disclosed herein.
[0088] However, it should be understood that FIGS. 4a through 5 are provided merely for exemplary purposes and should not be interpreted as limiting the scope of the disclosure. For example, the lighting subsystem (110) may generate any number of lighting beams (112) in any arrangement of elevation or azimuth angles of incidence. In this way, the illustration of two lighting beams (112) in FIGS. 4a through 5 is merely exemplary. As another example, components of lens assemblies (132) that are similar to, but not limited to, the homogenizer (134) illustrated in FIGS. 4a through 5 are also merely exemplary. Any number or type of components may be used to provide lighting beams (112) having spatial and angular uniformity that satisfies or exceeds selected metrics (e.g., standard deviation metrics, variance metrics, or others).
[0089] Now, referring to FIG. 6, FIG. 6 is a flowchart illustrating steps performed in a method (600) for imaging a buried target (106) according to one or more embodiments of the present disclosure. The embodiments and enabling techniques previously described herein in the context of the measurement system (100) should be interpreted as extending to the method (600). For example, the processors (118) of the controller (116) may execute program instructions that cause the processors (118) to implement any or all of the steps of the method (600) directly or indirectly (e.g., through control signals to additional components of the measurement system (100). However, the method (600) is not limited to the architecture of the measurement system (100).
[0090] In embodiments, the method (600) includes the step (602) of illuminating a buried target (106) using an illumination subsystem (110) that provides two or more spatially uniform illumination beams (112) along optical paths outside the objective lens (122) of the imaging subsystem (102), wherein the illumination beams (112) completely overlap in the buried target (106) to provide a common illumination field, and wherein the illumination subsystem (110) includes lens assemblies (132) for limiting the numerical apertures of the illumination beams (112) to prevent specular reflection from entering the objective lens. In embodiments, the method (600) includes the step (604) of generating one or more images of the buried target (106) using the imaging subsystem (102). In embodiments, the method (600) includes the step (606) of generating one or more measurements of a sample (108) based on one or more images.
[0091] For example, the steps (602 and 604) of the method (600) may be implemented as the illumination subsystem (110) and imaging subsystem (102) of the measurement system (100) as described in relation to FIGS. 1a through 5, but are not required to be so. In this way, the descriptions associated with FIGS. 1a through 5 may be extended to apply to the method (600).
[0092] The measurements generated in step (606) may be of any type of measurement. In some embodiments, the measurements include overlay measurements or critical dimension (CD) measurements, but are not limited thereto. In some embodiments, the measurements include inspection measurements, such as the identification and / or characterization of defects, but are not limited thereto. In some embodiments, the measurements include alignment measurements. For example, in the case of a bonded sample (108), the embedded target (106) may correspond to an alignment target at the interface of the bonded components. In this configuration, the measurements of the embedded target (106) may indicate the accuracy of alignment between the bonded components before or after the bonding process.
[0093] The AMS of traditional measurement tools utilizes a low NA bright-field epi-illumination microscope configuration. However, this configuration suffers from low contrast due to strong specular reflection as previously described herein, as well as absorption by dopants. It is considered herein that a measurement system (100) can be used as an AMS to provide high-contrast imaging as described throughout this disclosure. Similarly, the method (600) can be used for the purpose of AMS calibration of the sample (108) and the chuck coordinate systems (e.g., the coordinate system associated with the translation stage (130)). In this way, high-contrast images associated with the systems and methods disclosed herein can advantageously enable more accurate calibration than traditional systems.
[0094] In some embodiments, the image of the buried target (106) is corrected to also improve the contrast and / or visibility of the buried target (106).
[0095] FIG. 7 is a flowchart illustrating steps performed in a method (700) for correcting an image of a buried target (106) according to one or more embodiments of the present disclosure. For example, FIG. 7 may illustrate a natural uniformity calibration (NUC) process. The embodiments and enabling techniques previously described herein in the context of the measurement system (100) should be interpreted as extending to the method (700). For example, the processors (118) of the controller (116) may execute program instructions that cause the processors (118) to implement any or all of the steps of the method (700) directly or indirectly (e.g., through control signals to additional components of the measurement system (100). However, the method (700) is not limited to the architecture of the measurement system (100).
[0096] In the embodiments, the method (700) includes the step (702) of generating an arm reference image without a sample (108). For example, the sample (108) may be moved away from the measurement field of the imaging subsystem (102) so that photons from the sample (108) are not captured by the objective lens (122).
[0097] In the embodiments, the method (700) includes the step (704) of generating a light reference image of a portion of the sample (108) at the depth of the buried target (106), wherein the buried target (106) is not within the frame of the light reference image. In this way, the light reference image may correspond to a fixed parasitic pattern that may include scattering or other parasitic signals from the sample (108).
[0098] In the embodiments, the method (700) includes the step (706) of generating a NUC reference image based on light and dark reference images. For example, the NUC reference image can be generated by subtracting a dark reference image from a light reference image.
[0099] In embodiments, the method (700) includes the step (708) of correcting an image of a buried target (106) using a NUC reference image to form a corrected image. For example, the NUC reference image may be subtracted from the image of the buried target (106) (e.g., the image generated in step (604) of the method (600). In this way, the contrast of the buried target (106) may also be improved. Additionally, this corrected image may be used to generate a measurement (e.g., alignment measurement, measurement, inspection measurement, or others). As an example, step (606) may include generating one or more measurements using the corrected image of the buried target (106).
[0100] As an example, marks forming the embedded targets (106) can be printed between the stacked wafers / dies at different locations on the bonded sample (108) for both overlay metrology and spatial navigation across the wafers. For site-to-site navigation at runtime, an initial alignment process of the axes of the bonded sample (108) and the axes of the tool chuck can be performed. Then, an alignment measuring system (AMS) can be used to image the embedded targets (106) between the bonded components and also to calibrate the sample and chuck coordinate systems. After calibration is complete, a user-provided map of the sample (108) can be used to navigate to desired locations on the sample (108).
[0101] For example, the method (600) may further include a step (not shown) of correcting sample coordinates associated with one or more images of one or more buried targets (106) to chuck coordinates associated with a translation stage that fixes the sample (108). In some cases, multiple buried targets (106) may be visible in the image. In some cases, multiple buried targets (106) across the sample (108) are imaged separately.
[0102] In some embodiments, the buried target (106) is used as a high-resolution reference point (HRRP) in a high-resolution alignment process, which can be performed after AMS calibration.
[0103] For example, buried targets (106) having common designs may be distributed across a sample (108) (e.g., distributed within different fields on the sample (108)). Then, the measurement system (100) may translate the sample (108) to the expected locations of these buried targets (106) (potentially after the first AMS calibration step as described above) and capture images of the buried targets (106) at these expected locations. Finally, high-resolution calibration of the sample coordinates with respect to the chuck coordinates may be generated based on deviations of the imaged locations of the buried targets (106) from the expected locations (e.g., deviations from the centers of the corresponding images).
[0104] FIG. 8 is a simplified schematic diagram of a sample (108) illustrating buried targets (106) to be used as high-resolution reference points in a high-resolution alignment process according to one or more embodiments of the present disclosure. In particular, FIG. 8 illustrates a sample (108) having various fields (802) associated with the exposure fields of a lithography system. Additionally, the buried targets (106) are distributed within selected fields (802), wherein the buried targets (106) are labeled as Site 0 through Site 5. In embodiments, the high-resolution alignment process may include generating images at expected locations of the buried targets (106) and recording deviations of the imaged locations of the buried targets (106) relative to the expected locations (e.g., image centers). In this way, these deviations may be used to generate a high-resolution correction for aligning the sample coordinates to the chuck coordinates. Additionally, as illustrated in FIG. 8, various fields (802) containing buried targets (106) may be intentionally distributed in a pattern suitable for capturing positioning errors associated with both large and small movements. For example, Step 1 may correspond to a relatively short step along the vertical direction, Step 2 may correspond to an intermediate step along the vertical direction, Step 3 may correspond to a long step along the vertical direction, Step 4 may correspond to a combination of horizontal and vertical motion, Step 5 may correspond to a long step along the horizontal direction, and Step 6 may correspond to an intermediate step along the horizontal direction. It should be understood that the pattern of buried targets (106) is provided only for exemplary purposes and should not be interpreted as being limited. Rather, the buried targets (106) may be distributed in any suitable pattern.
[0105] In some cases, the measurement system (100) includes two detectors (104) for performing both AMS calibration and HRRP calibration. For example, a first detector (104) having a first resolution may be used for AMS calibration, while a second detector (104) having a second resolution higher than the first resolution may be used for HRRP calibration. Additionally, the buried targets (106) used for AMS calibration and HRRP calibration may be the same or different. In some cases, the first measurement system (100) is configured to perform AMS calibration, where the second measurement system (100) is configured to perform HRRP calibration.
[0106] The subject matter described herein exemplifies different components that are sometimes contained within or connected to other components. It should be understood that the architectures described here are merely examples, and that many other architectures may be implemented to achieve the same function. In a conceptual sense, any arrangement of components to achieve the same function is effectively “associated” so that the desired function is achieved. Therefore, regardless of the architectures or intermediate components, any two components in the present invention combined to achieve a specific function may be seen as “associated” with one another. Likewise, any two components so associated may also be seen as “connected” or “combined” with one another to achieve the desired function, and any two components so associated may be seen as “combinable” with one another to achieve the desired function. Specific examples of combinable include, but are not limited to, components that may and / or are physically interactable and / or are wirelessly interactable and / or are logically interactable and / or are logically interactable.
[0107] It is believed that the present disclosure and its numerous incidental advantages will be understood from the foregoing description, and it will be apparent that various modifications may be made in the form, configuration, and arrangement of components without departing from the disclosed invention or sacrificing all substantial advantages of the invention. The described forms are merely illustrative, and the following claims are intended to encompass and include such modifications. Furthermore, it should be understood that the present invention is defined by the appended claims.
Claims
Claim 1 As a darkfield imaging system, the imaging subsystem comprises an objective lens and a detector configured to image a buried target within the volume of a sample; an illumination subsystem - said illumination subsystem is: One or more light sources providing light having a first spatial profile; and The apparatus comprises two or more lens assemblies configured to direct light from one or more light sources as two or more light beams along optical paths outside the objective lens toward the buried target, wherein the two or more light beams completely overlap at the buried target to provide a common light field, and the two or more lens assemblies limit the numerical apertures of the two or more light beams to prevent specular reflection of the two or more light beams from entering the objective lens, and each of the two or more lens assemblies includes a homogenizer to provide that each of the two or more light beams has a second spatial profile having higher spatial uniformity than the first spatial profile according to a selected metric -; and a controller communically coupled to the detector - the controller comprises one or more processors, and the one or more processors enable the one or more processors: Receiving one or more images of the buried target from the detector; A dark-field imaging system comprising: a program command configured to generate one or more measurements of a sample based on one or more of the above images. Claim 2 A dark-field imaging system according to claim 1, wherein each of the one or more light sources comprises bare light-emitting diodes (LEDs), and each of the two or more lens assemblies comprises: a field stop—the homogenizer within each of the lens assemblies provides the second spatial profile at the field stop—; and a projection lens for imaging the field stop onto the buried target to provide each of the two or more light beams. Claim 3 In paragraph 2, the homogenizer within each lens assembly comprises: a waveguide, in a dark-field imaging system. Claim 4 In paragraph 3, the above waveguide is: a dark-field imaging system comprising a sealing waveguide. Claim 5 In paragraph 3, the above waveguide is: a dark-field imaging system comprising a multimode waveguide. Claim 6 In paragraph 2, the field aperture is: a dark-field imaging system comprising an aperture. Claim 7 A dark-field imaging system according to paragraph 2, wherein the field aperture is located on the output surface of the homogenizer. Claim 8 A dark-field imaging system according to claim 7, wherein the output surface of the field aperture and the homogenizer is tilted to satisfy the Scheimpflug condition, and the projection lens images the field aperture onto the buried target using an image plane coplanar with the buried target. Claim 9 A dark-field imaging system according to paragraph 2, wherein each lens assembly further comprises: an aperture stop located in the rear focal plane of the projection lens, and each illumination beam is telecentric to provide uniform angles of incidence across the common illumination field. Claim 10 A dark-field imaging system according to claim 1, wherein one or more lighting sources provide spatially coherent light, and the lighting subsystem comprises two or more optical fibers for coupling the spatially coherent light to two or more lens assemblies. Claim 11 A dark-field imaging system according to claim 10, wherein at least one of the above one or more light sources comprises: a superluminescent diode. Claim 12 A dark-field imaging system according to claim 10, wherein at least one of the two or more optical fibers is a multimode optical fiber. Claim 13 A dark-field imaging system according to claim 12, wherein the multimode optical fiber has at least one of a rectangular, square, and hexagonal core. Claim 14 A dark-field imaging system according to claim 10, wherein each lens assembly of the two or more lens assemblies comprises: a collimator lens configured to collect light from an associated optical fiber among the two or more optical fibers; and two Powell lenses having vertices rotated 90 degrees relative to each other. Claim 15 A dark-field imaging system according to claim 14, wherein each lens assembly further comprises an additional collimator lens after the two Powell lenses. Claim 16 A dark-field imaging system according to claim 14, wherein each lens assembly further comprises a field aperture after the two Powell lenses. Claim 17 A dark-field imaging system according to claim 1, wherein the one or more processors are also configured to execute program instructions that cause the one or more processors to receive a dark reference image without the sample; to receive a bright reference image of the plane of the sample at the depth of the buried target, wherein the buried target is outside the frame of the bright reference image; and to generate a natural uniformity calibration (NUC) reference image based on the bright reference image and the dark reference image. Claim 18 A dark-field imaging system according to claim 17, wherein generating the one or more measurements of the sample based on the one or more images comprises: subtracting the NUC reference image from the one or more images to form one or more corrected images; and generating the one or more measurements based on the one or more corrected images. Claim 19 A dark-field imaging system according to claim 1, wherein the selected metric comprises: at least one of the standard deviation and variance of the intensity across the common illumination field. Claim 20 A dark-field imaging system according to claim 1, wherein the one or more measurements include: at least one of a measurement, an inspection measurement, and an alignment measurement. Claim 21 An illumination system comprising: one or more illumination sources providing temporally incoherent light having a first spatial profile; and two or more lens assemblies configured to direct the light from the one or more illumination sources as two or more illumination beams along optical paths outside the objective lens of an imaging subsystem to a buried target—the two or more illumination beams completely overlap at the buried target to provide a common illumination field, the two or more lens assemblies limit the numerical apertures of the two or more illumination beams to prevent specular reflection of the two or more illumination beams from entering the objective lens, and each of the two or more lens assemblies includes a homogenizer to provide that each of the two or more illumination beams has a second spatial profile having higher spatial uniformity than the first spatial profile according to a selected metric. Claim 22 In paragraph 21, the lighting system wherein the selected metric comprises: at least one of the standard deviation and variance of the intensity across the common lighting field. Claim 23 As a dark-field imaging method, a step of illuminating a buried target within the volume of a sample using two or more illumination beams using an illumination subsystem - said illumination subsystem is: One or more light sources providing temporally incoherent light having a first spatial profile; and A dark-field imaging method comprising: two or more lens assemblies configured to direct light from one or more light sources as two or more light beams toward a buried target along optical paths outside the objective lens of an imaging subsystem including an objective lens and a detector, wherein the two or more light beams completely overlap at the buried target to provide a common light field, wherein the two or more lens assemblies limit the numerical apertures of the two or more light beams to prevent specular reflection of the two or more light beams from entering the objective lens, and each of the two or more lens assemblies includes a homogenizer to provide that each of the two or more light beams has a second spatial profile having higher spatial uniformity than the first spatial profile according to a selected metric; generating one or more images of the buried target from the detector; and generating one or more measurements of the sample based on the one or more images. Claim 24 A dark-field imaging method according to claim 23, further comprising: generating a dark reference image without the sample; generating a light reference image of the plane of the sample at the depth of the buried target, wherein the buried target is outside the frame of the light reference image; and generating a natural uniformity calibration (NUC) reference image based on the light reference image and the dark reference image. Claim 25 A dark-field imaging method according to claim 24, wherein the step of generating the one or more measurements of the sample based on the one or more images comprises: the step of subtracting the NUC reference image from the one or more images to form one or more corrected images; and the step of generating the one or more measurements based on the one or more corrected images. Claim 26 A dark-field imaging method according to claim 23, wherein one or more of the above measurements include: at least one of a measurement, an inspection measurement, and an alignment measurement.
