High-throughput optical metrology

Abstract

A method for optical metrology of a sample, the method comprising: illuminating an area of ​​the sample with a set of pulses of different wavelengths while the sample is moving at variable speeds; collecting light reflected from the sample as a result of said illumination to provide a set of such frames, each set of frames comprising partially overlapping frames associated with said different wavelengths; and processing said frames to provide optical measurement results indicative of one or more evaluation parameters of an element of the area of ​​the sample, wherein said processing is based on a mapping between said set of frames and reference measurements obtained by another optical metrology process exhibiting a higher spectral resolution than the spectral resolution obtained by said illumination and said collection.

Description

[Background technology]

[0001] Optical metrology for semiconductor devices is the standard method for measuring critical dimensions on semiconductor wafers to promote high yields in the semiconductor manufacturing process. Many optical metrology-based techniques, such as spectral reflectometry, scatterometry, ellipsometry, and spectral ellipsometry, are commonly used to detect critical dimensions, film thickness, composition, and other parameters of semiconductor wafers during manufacturing.

[0002] White light reflectance measurements, scatterometry, and ellipsometry are relatively time-consuming techniques because they require acquiring and processing spectral information at hundreds or even thousands of wavelengths, which is why the measurements are performed on selected sites, which may form a negligible part of the wafer.

[0003] The measurement spot size for these techniques is typically less than 100 μm, and the measurements are performed sequentially, allowing measurements to be performed on only a very limited number of points without significantly impacting the throughput of the metrology system (the number of inspections that can be processed in a given time).

[0004] Hyperspectral imaging is a known optical measurement technique. A test wafer is illuminated with a broad light spectrum, and image data is generated that represents the intensity of light reflected or scattered from the test wafer. Each pixel in the detected image is analyzed individually for different spectral ranges. Hyperspectral imaging is expensive and complex. Analyzing hyperspectral data requires fast computers, sensitive detectors, and large data storage resources.

[0005] Scatterometry tools are widely used in semiconductor manufacturing process control. Scatterometry tools typically measure the reflectance spectrum at specific test points, and / or memory arrays, and / or other predefined in-die locations. The size of the area on the sample being measured (the area that returns (reflects) the light to be detected), or spot size, is typically small, ranging from about 10 to 50 microns in diameter.

[0006] Scatterometry tools are fast, with MAM (Move Acquire Measure) times well under a second, and TPT times typically exceed 100 watts per hour for standard sampling plans. Increasing sampling plans require measuring more locations, but in any case, small spot size measurements of large areas can be problematic for WID and / or WIW uniformity and / or when spatial wafer maps or extreme wafer edge performance are of interest.

[0007] There are several known methods for providing a complete wafer image / map, for example, by "one-shot" imaging of the entire wafer (e.g., Spark Nanda technology, Lars Markwort et al.'s "Full wafer macro-CD imaging for excursion control of fast patterning processes" (Proc. SPIE Vol. 7638, 2010)), or by scanning (US Patent Application US2019 / 0244374 A1). These tools are specifically prepared in advance to analyze the wafer image based on a DOE wafer using spectral filters or RGB cameras and define correlations between the image and parameters of interest. Scanning tools are proposed to be used as part of the polishing equipment, allowing wafer images to be obtained in close proximity to the process.

[0008] The concept of Integrated Metrology (IM) is described in detail in Nova's U.S. Patents 6,752,689 and 9,184,102, among others. The measurement units (MUs) of such IMs are typically mounted in the Equipment Front End Modules (EFEMs) of the process tool, and wafers are delivered to the IM system for measurement by the EFEM robots. A typical measurement sequence may include global and fine alignment of the wafer using an imaging system, as described in Nova's U.S. Patents 5,682,242 and 6,752,689, among others.

[0009] An optical metrology system would be highly desirable to provide a complete wafer metrology solution without detrimental throughput impact. Summary of the Invention

[0010] The subject matter of the present invention is particularly pointed out and distinctly claimed in the concluding portion of this specification, although the invention, both as to organization and method of operation, together with its objects, features, and advantages, may best be understood by reference to the following detailed description when read in conjunction with the accompanying drawings. [Brief explanation of the drawings]

[0011] [Figure 1] An example of a frame is shown below. [Figure 2] 1 shows an example of a lighting element. [Figure 3] 1 illustrates an example of an optical measurement system and its environment. [Figure 4] 1 illustrates an example of an optical measurement system and its environment. [Figure 5] 1 shows several examples of optical metrology systems and their environments. [Figure 6] 1 illustrates an example of components of an optical metrology system. [Figure 7] 1 shows an example of a wafer, a camera, and the effective area of ​​view of the camera. [Figure 8] 1 shows an example of a wafer, a camera, and the effective area of ​​view of the camera. [Figure 9] 1 illustrates an example of components of an optical metrology system and frame. [Figure 10] 1 illustrates an example of components of an optical metrology system. [Figure 11] An example of the method is shown below. [Figure 12] 1 illustrates at least a portion of a system. [Figure 13] 1 illustrates at least a portion of a system. [Figure 14] Show image. [Figure 15] 1 illustrates at least a portion of a system. [Figure 16] The optical measurement method is shown. [Figure 17] The optical measurement method is shown. [Example]

[0012] In the following detailed description, numerous specific details are set forth in order to provide a thorough understanding of the present invention. However, it will be understood by those skilled in the art that the present invention may be practiced without these specific details. Alternative embodiments, well-known methods, procedures, and components have not been described in detail so as not to obscure the present invention.

[0013] It will be appreciated that for simplicity and clarity of illustration, elements shown in the figures have not necessarily been drawn to scale. For example, the dimensions of some elements may be exaggerated relative to other elements for clarity. Further, where considered appropriate, reference numerals may be repeated among the figures to indicate corresponding or analogous elements.

[0014] Because the illustrated embodiments of the present invention can be implemented, for the most part, using electronic components and circuits known to those skilled in the art, no more detail will be described than is deemed necessary as illustrated above for the understanding of the underlying concepts of the present invention and to avoid obscuring or distracting from the teachings of the present invention.

[0015] References in the specification to a method should apply mutatis mutandis to a system capable of performing that method, and should apply mutatis mutandis to a non-transitory computer-readable medium storing instructions that are once executed by a computational result in performing that method.

[0016] References in the specification to a system should apply mutatis mutandis to a method executable by that system, and should apply mutatis mutandis to a non-transitory computer-readable medium storing instructions that once executed result from computation in performing that method.

[0017] In the following description, reference is made to wafers. Wafers, particularly semiconductor wafers, are just one example of a sample.

[0018] In the following description, we refer to a full wafer scan. Note that references to a full scan can, mutatis mutandis, apply to scanning only one or more portions of a wafer.

[0019] In the following description, we refer to light emitting diodes (LEDs), which are just one example of an illumination (illumination) source.

[0020] In the following description, wavelengths are mentioned. References to wavelengths should apply mutatis mutandis to ranges of wavelengths. Additionally or alternatively, references to wavelengths can apply mutatis mutandis to illumination and / or collection (light collection), e.g. polarization, angular content of the illumination and / or collection beams, etc.

[0021] In the following description, reference is made to an effective field of view (FOV). The effective FOV is the FOV that is considered during the metrology. The effective FOV can be the entire FOV of an area-scan camera or a portion of the FOV. For example, if only a portion of the camera pixels form a region of interest that is processed for metrology, the effective FOV is limited to that portion of the pixels.

[0022] The provided ultra-fast (eg, within 0.5, 1, or 2 seconds) whole wafer scanning metrology system can include: a. An illumination module capable of illuminating a region of the wafer, called a region of interest, using pulses of illumination elements, such as LEDs, that are of different wavelengths (one after the other), each illuminating a corresponding ROI during a set and measurement session. b. One or more area scan cameras with narrow and elongated fields of view (FOVs), which provide low-resolution spectral information about the sample because the spectral information is limited to the illumination wavelength. c. One or more additional optical elements, such as lenses, objectives, light guides, beam splitters, etc.

[0023] Relevant wafer areas are illuminated with different wavelengths (at different times) and frames are acquired.

[0024] The illumination can be performed outside the chamber (MU) and / or as the wafer moves from one chamber and / or tool to another. Examples of such moves include moving the wafer from a processing tool, such as a CMP polisher, to a metrology tool or from a metrology tool to a polisher, moving the wafer to or from a cassette, etc.

[0025] The wafer can be moved by the robot of the EFEM or by other means.

[0026] The robot movement can exhibit velocity fluctuations, as opposed to constant velocity stage movement. The robot movement does not have to be controlled by an optical metrology system. The maximum wafer movement speed must be known or estimated.

[0027] The effective FOV is narrow and elongated. The number of pixel rows (width, the narrow dimension of the effective FOV) represents a trade-off between the number of wavelengths that can be used in a metrology session (e.g., at least 5, 8, 10, 15, 20, 25, 30, 35) and the intensity of the reflected light. Increasing the effective FOV requires illuminating a larger area of ​​the wafer, which can reduce the illumination energy density.

[0028] Consecutive frames at different wavelengths are overlapped, thereby covering the complete wafer image at any of the different scanning wavelengths, allowing the method to obtain visual information of any point of interest on the wafer.

[0029] Any optical parameter can be selected, for example wavelength, polarization, etc.

[0030] The optical parameters can be selected based on metrology parameters (of the wafer or any portion thereof) to be evaluated, which can refer to one or more structural elements (e.g., one or more sub-micron structural elements, one or more nanometer-range structural elements, one or more sub-micron regions of a bare wafer) and one or more properties (e.g., critical dimension, film thickness, composition, etc.) of one or more structural elements.

[0031] The one or more optical parameters are selected based on a model-based simulation of the optical processes of illuminating (irradiating) a sample, collecting radiation from the sample, and generating a detected signal when applied to one or more structural elements, the model-based simulation revealing that the one or more optical parameters, once applied, provide results (e.g., detected signals) that are sensitive (e.g., most sensitive) to one or more tested metrology parameters.

[0032] The optical parameters (eg, wavelength) can be selected from a large set of parameters that can be provided by the system.

[0033] It should be noted that in the case of LED illumination, the optical metrology methods described above have coarser spectral resolution than spectral reflectometry, since they measure signals resulting from illumination with a limited number of wavelengths, each wavelength representing a narrow spectral range.

[0034] The optical parameters include six different wavelengths that can be emitted from one or more sets of six LEDs.

[0035] Referring to Figure 1, Figure 1 shows seven frames 91-97. The first six frames 91-96 form a set of frames that includes six frames 91-96 resulting from illuminating a wafer with six different illumination wavelengths. The seventh frame 97 represents the start of the next set of six frames, resulting from illuminating a wafer with the first wavelength of the set of six different illumination wavelengths.

[0036] The LED pulses are sequentially timed to provide overlapping frames. Note that Figure 1 shows the overlap obtained for a robot at a constant, maximum speed (scanning speed); in reality, the speed may change over time during the measurement session. The slower the speed, the greater the overlap between frames.

[0037] In FIG. 1, the distance between pulses can be predetermined, for example, 1 / 8 of the FOV (field of view in scan direction) divided by the maximum scan speed.

[0038] Figure 2 shows a linear array of lighting elements 112 that may use LEDs. Figure 2 also shows four sets of six LEDs each (emitting six different wavelengths). Figure 2 also shows three sets of eight LEDs each (emitting eight different wavelengths). Note that any arrangement of LEDs can be provided.

[0039] After six flashes (strobes), the LED is turned on again as the first wavelength illuminates the wafer, resulting in an overlap of 2 / 8 FOV between frames 91 and 97 (assuming a constant and maximum scan speed).

[0040] Variations in the velocity of the robot cause changes in the overlap between related frames, and such changes in the overlap between related frames need to be compensated for. Frames are related if they are acquired using illumination of the same wavelength.

[0041] A map of the entire wafer (for each group of related frames) needs to be generated, and combining (stitching) related frames is required to provide a map for each illumination wavelength.

[0042] The combination requires determining the location of the area of ​​the wafer imaged by each frame.

[0043] Stitching frames on bare wafers is more challenging than stitching frames on patterned wafers. The wafer pattern can include anchors, which can be used to determine the position of each frame. For bare wafers without anchors, the edge of the wafer can be used to detect the position of each frame's region. The edge position can be sensed by one or more cameras in an optical metrology system. Additionally or alternatively, other sensors can be used for edge detection and / or tracking of the robot position, for example, using visual or non-visual sensors to follow the wafer movement.

[0044] One example provides an effective FOV that is smaller than the camera's full FOV, which can include using a CMOS area sensor with ROI selection. Using such an area sensor allows a limited number of rows to be selected for image capture, resulting in a much higher frame rate than the full-frame equivalent. For example, the Basler a2A1920-160umBAS camera (1920 x 1200 pixels, 160 frames per second (fps)) operates at 2717 fps when used in ROI mode with 1920 x 40 pixels (20 μs exposure).

[0045] The wafer is assumed to move at a constant speed of 0.5 m / s. The pixel size on the wafer is 50 micrometers. The illumination system with multiple LEDs can provide uniform illumination over an area 300 mm long and 2 mm wide.

[0046] Four cameras are used to overlap the entire 300mm length, each imaging a 96mm long area (for a 1920 pixel Basler a2A1920-160umBAS camera), with 75mm of overlap per camera with no overlap. 40 pixels gives a width of 2mm. At a speed of 0.5m / s, scanning 300mm takes 0.6 seconds.

[0047] The maximum frame rate for the Basler a2A1920-160umBAS camera is 1630 frames in 0.6 seconds.

[0048] Another way to look at it is that about 200 frames are needed for each color of LED to get 0.5mm overlap (300mm / 1.5mm=200 frames, 2mm-1.5mm=0.5mm).

[0049] This means that eight different wavelengths (1600 frames / 200 frames per color = 8) can be used with a multi-area, multi-strobe LED imaging system without reducing scan speed for full wafer mapping.

[0050] 3 and 4 show an example of a metrology system 40 and its environment, such as an IM tool 31 integrated with an EFEM 30 of a processing device (CMP polisher), shown at 70 in FIG. 4. Other environments may also be provided, such as another environment that does not include an IM tool 31.

[0051] The environment also includes an IM tool 31 (or any other high-resolution optical metrology process, where high means higher than the process performed by optical metrology system 40) having a robot 60 (shown holding wafer 99) and a chamber 35 configured to receive the wafer from robot 60, perform high-resolution spectral reflectance processing, and then return the wafer to the robot.

[0052] The robot can place wafers into one or more cassettes (FOUP's) 62, 63 and 64 of the EFEM 30 and / or provide the wafers to another tool such as a polisher 74.

[0053] The robot 60 may be part of the EFEM and may move inside it. The cassette and IM tool can be connected to the EFEM through ports / openings. The IM tool is typically connected via a six-bolt attachment, and the cassette rests on a so-called load port that supports it. A metrology system 40 is assembled between the IM tool 31 and the EFEM. The metrology system 40 can be aligned by the Z axis according to the Z position of the robot arm (e.g., it can be configured to be partially or fully adjustable during installation according to the Z position of the wafer on the robot arm).

[0054] The optical metrology system 40 is positioned to perform measurements when a wafer is loaded into the IM tool 31 by the robot 60 and / or when a wafer is unloaded from the IM tool 31 .

[0055] The optical metrology system 40 may be an add-on system, in which case its shape and size must be determined based on the size constraints of other structural elements in the environment (e.g., configured to be insertable between the EFEM and IM tools 31).

[0056] The metrology system 40 can be configured "frame-like" around an opening / port with optics / lighting on top, at least partially sealing the connection of the IM tool to the EFEM. The metrology system 40 can be connected to the control unit / computer of the IM tool 31, or an additional separate control unit / computing for the metrology system 40 can be housed within or outside the IM tool.

[0057] FIG. 5 shows front and side views of a measurement system 40, a spectral reflectance measurement tool 31, and its chamber 35. The measurement system 40 is compact, with dimensions (width and / or height and / or depth, and / or dimensions extending outside the spectral reflectance measurement tool) that are much less than one meter. For example, the measurement system 40 can extend 5-15 centimeters (or more) outside the spectral reflectance measurement tool, have a width and height of about 25-45 centimeters, etc. In some examples illustrated in FIG. 4, at least a portion of the measurement system 40 is disposed within the spectral reflectance measurement tool 31.

[0058] The spectral reflectance measurement tool 31 can further process the results of the measurement system 40 for various purposes, such as calibration, validation, selection of locations to be evaluated (where the results of the measurement system 40 may indicate the presence of a problem, a deviation from specification, etc.).

[0059] FIG. 6 shows an example of some components of an optical metrology system 40 .

[0060] The components include an LED-based illumination system 44, a beam splitter 43, a telecentric objective lens 42, and one or more cameras 41, such as line scan or area scan cameras.

[0061] A telecentric objective lens has a horizontal portion and a vertical portion.

[0062] The LEDs and beam splitter form a coaxial illumination system. The beam splitter can have a width and length of approximately 4 cm.

[0063] Light from LED 41 is directed by beam splitter 42 (e.g., at a normal incidence angle) onto wafer 99, where it is reflected onto the vertical portion of telecentric objective 43 and exits the horizontal portion of telecentric objective 43 towards camera 44.

[0064] Note that while the example in Figure 6 refers to a telecentric objective, a non-telecentric objective (e.g., a non-telecentric macro lens) can be used, which may require compensating for non-uniformities in the illumination angle caused by using a non-telecentric objective.

[0065] Figures 7 and 8 show a sequence of five cameras 41(1)-41(5), each with a narrow, elongated effective FOV 42(1)-42(5) that "covers" the entire length of the wafer 99. Figure 8 also shows a dedicated edge sensor 68 that detects the edge of the wafer. The edge sensor 68 is linear, with a field of view that exceeds the radius of the wafer and is oriented relative to (e.g., perpendicular to) the axis of wafer movement by the robot. The edge sensor can be positioned below the wafer and used for main system illumination.

[0066] 9 shows an example of some components of optical metrology system 40. These components include LED 55, beam splitter 53, and objective lens 52, which has an input FOV larger than its output FOV. Objective lens 52 collects light from multiple illumination segments 54(1), 54(2), and 54(3) that extend along the entire width of wafer 99 and directs the collected light to camera 51, which is much narrower than the width of the wafer.

[0067] FIG. 9 also shows a set of four frames 81(1)-81(4) and the first frame 81(5) in the next set of frames.

[0068] It should be noted that the metrology system 40 may use a contact image sensor (CIS) that is placed very close to the wafer and nearly touches it. Examples of CIS sensors that may be used include, for example, the VTCIS or VDCIS sensors from Tichawa Vision GmbH.

[0069] The wafer itself may not be perfectly flat, and the wafer may be moved by the robot in a path that deviates from a purely horizontal path. To prevent physical contact between the wafer and the CIS (or other optical elements of the metrology system 40), the metrology system 40 may be positioned at a safe distance from the wafer and / or may be movable to follow the wafer metrology movements and wafer flatness.

[0070] 10 shows an example of some components of the optical metrology system 40. These components include an illumination element 105 (located on a first board 104), a lens 106, a beam splitter 107, a light guide 103, and a camera 102 located on a second board 101.

[0071] The metrology system 40 can have multiple cameras with multiple pixels that "cover" the area of ​​the wafer with widths / lengths in the micron range (e.g., 20, 40, 60 microns) and can contain many pixels per row (e.g., for a full scan of a 300 mm wafer with a pixel width of 20 microns, there are 15,000 pixels per row).

[0072] It should be noted that metrology parameters may vary during a single scan of a wafer, for example, different metrology parameters may be applied to different portions of the wafer, e.g., memory areas may be scanned differently than logic areas.

[0073] Additionally, the metrology parameters may be evaluated and varied multiple times, for example, to increase the sensitivity of the metrology process to changes in the value of the evaluated parameter.

[0074] FIG. 11 illustrates an example method 200 for optical metrology of a sample.

[0075] The method 200 includes steps 210 , 220 and 230 .

[0076] Step 210 involves illuminating an area of ​​the sample with sets of pulses of different wavelengths while the sample is moving at variable speeds.

[0077] Step 220 includes collecting light reflected from the sample as a result of the illumination, and provides sets of such frames, each set including frames associated with different overlapping wavelengths.

[0078] Step 230 includes processing the frames to provide optical metrology results indicative of one or more evaluation parameters of elements of the areas of the sample, where the processing is based on a mapping between the set of frames and reference measurements obtained by another optical (or other reference) metrology process that exhibits a higher spectral resolution than the spectral resolution obtained by the illumination and collection.

[0079] The above processing may be performed by a computer system located at the metrology system 40, located at the IM tool 31, in communication with the metrology system 40, or in communication with the IM tool.

[0080] Two systems (also called units, modules, devices or tools) are provided. The first system has a first spectral resolution (e.g., processes or detects radiation at a first number of wavelengths) and a first throughput, and is capable of providing macroscale information. b. The second system has a second spectral resolution (e.g., processes or detects radiation at a second number of wavelengths) and a second throughput, and is capable of providing macroscale information.

[0081] The first spectral resolution is lower than the second spectral resolution.

[0082] For example, the first system described above may be capable of acquiring and processing optical information associated with 8, 10, 15, 20, and even tens (e.g., up to 30 or 40) wavelengths, whereas the second system may be capable of acquiring and processing optical information associated with hundreds or even hundreds of wavelengths.

[0083] The first throughput exceeds the second throughput, for example, by at least 10, 20, 30, 40, 50, 60, 70, 80, 90, or 100 (and even more) times.

[0084] The IM tool is an example of the second system, and the measurement system 40 is an example of the first system.

[0085] The first system can be used to provide a map of the entire wafer. The second system can be used to provide measurements of selected sites on the wafer. Figure 12 shows the first system (scan module, left side of the drawing) and the second system (IM OCD measurement unit, right side of the drawing).

[0086] The first system can be a scanning module and the second system can be a single shot optical module.

[0087] The first and second systems can complement each other, and information from the first and second systems can be processed together. Information from the first system can help with site selection. Information from one system can influence how the second system operates. Information from one system can be used to verify information in the other system.

[0088] The first system allows for rapid scanning of a large portion of a wafer, including the entire wafer, during transfer of the wafer from the EFEM to the IM tool or from the IM tool to the EFEM, and the first spectral resolution can be sufficient to find one or more metrology parameters of interest, for example, to capture WIW variations.

[0089] The second system can capture normal incidence reflected light and / or diffracted light, enabling flexibility and applicability for multiple metrology and inspection applications for various areas of semiconductor manufacturing, including CMP, deposition, and patterning.

[0090] Both systems can be based on illumination with a combination of LEDs, which can use any combination of LEDs of one or more wavelengths, with the second system providing coverage over a wide spectral range using LEDs that provide aggregated coverage, e.g., between UV and IR, between 265 and 960 nm, between 190 and 1000 nm, etc. The first system uses a limited number of wavelengths or a narrow wavelength range (the narrow wavelength range being the range emitted by an LED, such as a monochromatic LED, that is not completely monochromatic).

[0091] Both the first and second systems allow for polarization control (full or partial) of both the incident and collected light for operation in different regimes, including bright field BF and dark field DF.

[0092] The first system can be positioned above the wafer path from the processing equipment's EFEM to the IM tool, capturing "images" of the wafer as it moves toward the IM tool or returns to the processing equipment. In one possible implementation, the scanning module is used in a normal incidence (NI) configuration, where both the illumination (light from the light source toward the sample) and the collection (reflected light) are normal incidence (NI). The reflected light is collected by a lens and measured by a high-speed line camera (with a sampling rate of approximately 50 to 150 kHz, necessary for high-speed measurements). The line camera can have 10-20k pixels with a pixel size of 5-10 pm, allowing full-wafer measurements during loading and unloading. The optical scheme of the NI macro-optics module is shown in Figure 15.

[0093] Because the available volume in the first system is limited (due to the IM footprint), the optimal system for a complete wafer image is the normal-incidence line-scan concept, based on pulsed LED illumination, a line or area detector (scan camera), and an optical imaging system optimized for a line FOV. As shown in Figure 16, optical information can be acquired for the entire 300 mm wafer during a single scan, and half of the wafer, 150 mm, can be double-scanned (in and out). In this case, multi-wavelength sensing can be implemented using an illumination system with multiple LEDs and a camera with N lines. The time diagram for such sensing can be based on the exposure time of a single LED type, corresponding to the wafer movement over a length equivalent to the pixel size in the wafer plane. By synchronizing the wafer speed, LED switches, and the acquisition rate of the multi-line linear detector, multiple frames of the entire wafer can be acquired in multiple colors with minimal loss of resolution (Figure 5).

[0094] In another implementation, the first system is designed to measure diffracted or scattered light, where the sample illumination and collection angles are different, with the collection angle being normal incidence and the illumination being at an oblique angle.

[0095] The size of the optical module of the first system can be adapted to fit the load port, with minimal thickness (in the direction of wafer movement) and height and width defined by existing dimensions, so that both the footprint and integration (e.g., 6 volts) of the IM tool are not affected.

[0096] IM tools

[0097] The optical part of the IM tool can include movable optics (navigation over the wafer can also be implemented by a movable wafer stage (X, Y, R-theta, or a combination thereof)), allowing for a small footprint and confinement to the size of FOUPs (standard integration into polishers is via FE bolt connections, where the IM tool is placed as one of the standard ports). The movable optics enable at least one of two functions: spectral: enabling spectral measurement of targets of interest; and vision: enabling the capture of images used for multiple purposes, including pattern recognition, best focus, local and global alignment, etc. A vision system with a camera (monochrome, RGB, or hyperspectral) and a dedicated light source (capable of covering a wide wavelength range using filters, sets of LEDs, etc.) can be used as a micro-optical module to capture high-resolution images of predetermined locations on the wafer (see Figure 14).

[0098] The second system may include: a. Using the Vision Channel "micro" images collected in a standard measurement sequence spectrum, predict the variation of the profile geometric parameters in the area around the spectral measurement spot (scatterometry spot size less than 40 microns) (see Figure 14). b. Modify and optimize image acquisition conditions and sequences to collect images for prediction of the die maps to inspect problem areas on the wafer as defined by the full wafer map images and / or scatterometry data from the Scan module.

[0099] All proposed imaging solutions can also be applied to SA scatterometry tools.

[0100] Wafer Transfer

[0101] To obtain high quality images with the smallest possible pixel size, the movement of the wafer in the scanning system must be controlled.

[0102] Possible wafer motion uncertainties can be divided into three categories: uncertainties that the optical system is capable to account for without any HW additions (1), uncertainties that the optical system has to be flexible enough to accommodate for (2), and uncertainties that are easy to account for the wafer movement itself (3).

[0103] The optical system must be able to account for X,Y position uncertainties without correction, and for very little wafer motion uncertainty. In addition to this, focus can be accounted for by a simple autofocus system or can be kept fixed by setting a specification for the wafer's Z position.

[0104] Even if the wafer's Z-average position can be fixed at the desired level, wafer warpage can affect imaging and must be taken into account. (Most modern wafers typically have warpage less than + / - 250 microns, but it can be up to 500-800 microns, and in extreme 3D NAND applications, even + / - 1 mm.) To accommodate the potentially different warpage from wafer to wafer, optical systems have a wide range of NAs, from large (0.1-0.2 or more for flat wafers) to very small (0.02 or less for measuring extreme warpage of + / - 1 mm). The trade-off in this case is a slight decrease in resolution and an increase in measurement time. In any case, imaging recipes can be optimized for warped wafers, including the NA and optimal wafer speed.

[0105] The translation system (scheme shown in Figure 16) must be capable of controllable and constant wafer speeds, as defined by the optimal wafer imaging recipe, without dynamic tilts and dynamic focus changes (both less than 50–100 microns). All of these requirements can be met by a robot, or, if the specifications of existing atmospheric point-to-point robots do not match the requirements, by designing a special retractable rail drawer that retrieves the wafer from the robot in front of the imaging system, controllably scans the wafer under the imaging system at the desired speed, and finally transports the wafer to the IM module for OCD measurement. In fact, this rail drawer or retractable semi-buffer should allow wafer exchange outside the MU. The drawer is always inside the IM and only exits to pick up and return wafers, so it does not interfere with the robot's movements within the EFEM.

[0106] Imaging Measurement

[0107] The goal of imaging metrology is to convert an image into a map of parameters of interest on the entire image, preferably a multi-spectral image. A sequence of operations that can be used to demonstrate the proposed approach is presented in the example of measuring the thickness of residues in a memory array. This approach can be used with both static and movable optical modules.

[0108] Imaging measurement recipe settings

[0109] Step 1. Imaging Conditions

[0110] A standard OCD recipe is created based on the spectral information collected on the feature of interest (memory array). Interpretation results include profile information containing all parameters of interest and can be used to set up image collection. The parameters of interest measured by the OCD are used together with the spectral information to define the optimal configuration for image collection (number of wavelength combinations for image collection) to obtain the best performance for the target parameters of interest. The image collection configuration can be used with the OCD recipe for all feature measurements.

[0111] Step 2. Image processing recipe

[0112] After preprocessing to improve image quality, an image processing recipe (which can also be defined as pattern recognition) is created to automatically identify the region of interest where measurement of the parameter of interest is desired and select the required pixels (for all captured images with all wavelength combinations). As a result, an image of the region of interest is created, for example, an image of a memory array is acquired. Additional processing can be performed on the pixels in the region of interest, including averaging, noise reduction, and / or other computational processes for more stable measurement performance. Averaging can be performed on a single array, a portion of a die, or the entire die, all using various averaging schemes to focus on different scales of variation based on requirements.

[0113] Step 3. Machine Learning (ML) Recipe for Image Parameters

[0114] The X,Y coordinates of such an array are then matched to the coordinates of the OCD measurements, allowing for a direct correlation between the image parameters and the OCD reference measurements, creating an ML recipe that can convert the image parameters to the metrology parameters of interest. The ML recipe undergoes standard training, testing, and validation.

[0115] Step 4. Optional fine-tuning for extreme edges

[0116] Special processing may be required for extreme edges: additional OCD measurements and / or additional OCD recipes (fine-tuning) may be required to obtain the best extreme edge description.

[0117] Imaging metrology measurements

[0118] For every wafer, a standard OCD recipe measures the parameters of interest at predefined locations (12-100 points per wafer). Images are collected and processed under the required conditions. The OCD results (all parameters of interest) and image parameters at the same locations are used together to perform accurate image metrology on each wafer.

[0119] The interpretation of the image into parameter maps can be done in several ways, including an instant ML approach, which builds on profile parameters measured at one of the image locations (standard scatterometry) on the current wafer, and / or a specially prepared DOE wafer, and / or existing knowledge about the wafer and its processing.

[0120] An imaging device is provided, the imaging device including an imaging device configured to image a wafer during movement of the wafer between a metrology device and another device at a first spectral resolution and a first throughput, where the first spectral resolution is coarser than a second spectral resolution of the metrology device and the first throughput exceeds the second throughput of the metrology device, and the imaging device includes a mechanical interface for mechanically coupling the imaging device to the metrology device.

[0121] The imaging device may be configured to measure features on the wafer.

[0122] The imaging device may include an illumination module configurable to scan the wafer with at least one illumination line while the wafer is moving.

[0123] The imaging device may include an illumination module that can be configured to scan the wafer while the wafer is moving with different illumination lines having different illumination frequencies, the different illumination lines being formed on the wafer at different and non-overlapping illumination periods.

[0124] Each of the illumination lines may be perpendicular to the direction of movement of the wafer.

[0125] The illumination can be normal to the wafer.

[0126] The imaging device may include a collection module that has an optical axis that may be normal to the wafer.

[0127] The illumination module may include an optical unit configured to convert an incident radiation beam of circular cross section into line radiation.

[0128] FIG. 16 illustrates the method 300.

[0129] The method 300 includes acquiring 310 optical information about a wafer at a first spectral resolution and a first throughput by a first system during transfer of the wafer between a metrology system and another device, where the first spectral resolution (resolution) is coarser than a second spectral resolution of a second system, the first throughput exceeds a second throughput of the second system, and the first system is mechanically coupled to the second system by a mechanical interface. The first system may be an imaging system or a non-imaging system. The second system may be a metrology tool, an IM tool, or the like.

[0130] Step 310 is followed by step 320, in which the first system measures features of the wafer, which may be metrology parameters.

[0131] Step 310 can be performed using at least one line of illumination during the wafer movement while the illumination module of the first system scans the wafer in step 305. There can be multiple lines, but the number of lines is less than the number of pixels per line, such as 5-10 lines, 10-20 lines, 20-40 lines, 15-50 lines, etc.

[0132] Step 305 may include scanning the wafer with an illumination module of the first system while the wafer is moving, using different illumination lines with different illumination frequencies, the different illumination lines being formed on the wafer during different, non-overlapping illumination periods.

[0133] Each illumination line may be perpendicular to the direction of movement of the wafer.

[0134] The illumination may be normal to the wafer.

[0135] The first system may include a collection module having an optical axis perpendicular to the wafer.

[0136] Step 310 may include converting an incident circular cross-section radiation beam into line radiation by an optical unit of the first system.

[0137] FIG. 17 illustrates the method 400.

[0138] The method 400 may include a step 410 of acquiring optical information related to a wafer by a processing circuit, wherein an image is generated by a first system, the image being at a first spectral resolution, and the acquisition of the optical information is performed at a first throughput during movement of the wafer between a second system and another device.

[0139] The method 400 may include a step 420 of obtaining, by processing circuitry, second system results relating to one or more regions of the wafer, the metrology results being generated by a second system configured to perform measurements of features within the regions of the wafer at a second spectral resolution that is finer (finer) than the first spectral resolution and at a second throughput that is lower than the first throughput.

[0140] Steps 410 and 420 are followed by step 430, in which metrology results associated with one or more additional regions of the wafer are estimated based on (a) the mapping between the first system results and the second system results, (b) the results of the first system, and (c) the results of the second system, where the one or more additional regions are not included in the one or more regions of step 420.

[0141] This application provides significant technical improvements over the prior art, particularly in computer science.

[0142] References to the terms "comprising" or "having" should also be construed to refer to "consisting" of "essentially consisting of." For example, a method comprising certain steps can include additional steps, can be limited to certain steps, or can each include additional steps that do not materially affect the basic and novel characteristics of the method.

[0143] The present invention can also be embodied in a computer program executed on a computer system, the computer program including at least code portions that, when executed on a programmable device such as a computer system, perform the method of the present invention when the program is enabled to perform the functions of the device or system of the present invention. The computer program enables the storage system to assign disk drives to disk drive groups.

[0144] A computer program is a list of instructions, such as for a particular application program and / or operating system. A computer program may include, for example, one or more of the following: subroutines, functions, procedures, object methods, object implementations, executable applications, applets, servlets, source code, object code, shared libraries / dynamic load libraries, and / or any other sequence of instructions designed for execution on a computer system.

[0145] The computer program may be stored internally in a computer program product, such as a non-transitory computer-readable medium. All or part of the computer program may be provided on a permanent, removable, computer-readable medium, or remotely coupled to an information processing system. The computer-readable medium may be, for example, but not limited to, magnetic storage media, including disk and tape storage media; optical storage media, such as compact disc media (e.g., CD-ROM, CD-R) and digital video disc storage media; non-volatile memory storage media, including semiconductor-based memory units such as flash memory, EEPROM, EPROM, and ROM; ferromagnetic digital memory, MRAM; registers, buffers, or cache; main memory; RAM; and other volatile storage media. A computer process typically includes an executing (running) program, or portions of a program, current program values, state information, and resources used by an operating system that manages the execution of the process. An operating system (OS) is software that manages the sharing of computer resources and provides programmers with an interface used to access these resources. An operating system processes system data and user input and responds by allocating and managing tasks and internal system resources as a service to the system's users and programs. A computer system may include, for example, at least one processing unit, associated memory, and a number of input / output (I / O) devices. When executing computer programs, the computer system processes information according to the computer programs and generates resulting output information via the I / O devices.

[0146] In the foregoing specification, the invention has been described with reference to specific examples of embodiments thereof. It will, however, be evident that various modifications and changes can be made thereto without departing from the broader spirit and scope of the invention as set forth in the appended claims.

[0147] Furthermore, terms such as "front," "back," "top," "bottom," "over," "under," and the like, used in the specification and claims are used for descriptive purposes and do not necessarily describe permanent relative positions. Terms so used are interchangeable under appropriate circumstances, and it is understood that embodiments of the invention described herein can operate, for example, in orientations other than those illustrated or described herein.

[0148] Those skilled in the art will recognize that the boundaries between logical blocks are merely exemplary, and that alternative embodiments may merge logical blocks or circuit elements, or impose alternative decompositions of functionality on various logical blocks or circuit elements. That is, it should be understood that the architectures illustrated herein are merely exemplary, and that in fact many other architectures may be implemented which achieve the same functionality.

[0149] Any arrangement of components to achieve the same functionality is effectively "associated" such that the desired functionality is achieved. Thus, any two components combined in this specification to achieve a particular functionality can be considered to be "associated" with each other such that the desired functionality is achieved, regardless of architecture or intermediary components. Likewise, any two components so associated can also be considered to be "operably connected" or "operably coupled" with each other to achieve the desired functionality.

[0150] Moreover, those skilled in the art will recognize that the boundaries between operations described above are merely exemplary. Operations may be combined into a single operation, a single operation may be distributed among additional operations, and operations may perform with at least partial overlap. Furthermore, alternative embodiments may include multiple instances of a particular operation, and the order of operations may be changed in various other embodiments.

[0151] Also, for example, in one embodiment, the illustrated examples may be implemented as circuits located on a single integrated circuit or within the same device. Alternatively, the examples may be implemented as any number of separate integrated circuits or separate devices interconnected with each other in any suitable manner.

[0152] Also, for example, embodiments or portions thereof may be implemented as a soft or coded representation of a physical circuit or a logical representation that can be translated into a physical circuit, such as in a hardware description language of any suitable type.

[0153] Furthermore, the present invention is not limited to physical devices or units implemented in non-programmable hardware, but may be applied to mainframes, minicomputers, servers, workstations, personal computers, notepads, personal digital assistants, electronic games, automobiles and other embedded systems, mobile phones, and various other wireless devices, which are generally referred to in this application as "computer systems," and which are programmable devices or units capable of performing desired device functions by operating in accordance with appropriate program code.

[0154] However, other modifications, variations, and alternatives are possible, and the specification and drawings are accordingly to be regarded in an illustrative rather than a restrictive sense.

[0155] In the claims, reference signs placed between parentheses shall not be construed as limiting the scope of the claim. The word "comprises" does not exclude the presence of elements or steps other than those recited in the claim. Furthermore, the terms "a" and "an" are defined herein as one or more. Furthermore, the use of introductory phrases such as "at least one" or "one or more" in a claim extends the introductory phrases in the claims, such as "a" or "an," to an invention containing only one of such elements, even if the same claim contains the introductory phrases "one or more" or "at least one" and an indefinite article such as "a" or "an." The same applies to the use of definite articles. Unless otherwise specified, terms such as "first" and "second" are used to arbitrarily distinguish between the elements they describe. Therefore, these terms are not necessarily intended to indicate a chronological or other priority of such elements. The presence of measures in mutually different claims does not indicate that a combination of these measures cannot be advantageously used.

[0156] While certain features of the invention have been illustrated and described, many modifications, substitutions, changes, and equivalents will occur to those skilled in the art. It is, therefore, to be understood that the appended claims are intended to cover all such modifications and changes as fall within the true spirit of the invention.

Claims

1. illuminating regions of the sample with a set of pulses of different wavelengths while the sample is moving at variable speeds; collecting light reflected from said sample as a result of said illumination to provide sets of such frames, each set of frames comprising partially overlapping frames associated with said different wavelengths; processing the frames to provide optical measurements indicative of one or more evaluation parameters of an element in the area of ​​the sample, the processing being based on a mapping between the set of frames and reference measurements obtained by another optical metrology process exhibiting a higher spectral resolution than that obtained by the illumination and the acquisition; the illumination and collection is performed while the sample is being moved by a robot towards an integrated metrology tool, or while the sample is being moved by a robot between an integrated metrology tool and an equipment front end module (EFEM) of a processing tool; Optical measurement method for samples.

2. The method of claim 1 , wherein the other optical metrology process is an integrated metrology process performed by an integrated metrology tool.

3. The method of claim 1 , wherein the mapping is provided by a machine learning process.

4. 4. The method of claim 3, wherein the machine learning process is trained by a training process that includes providing the machine learning process with (a) test measurements of one or more test samples obtained by other optical metrology processes and information, and (b) additional test measurements of one or more test samples obtained by the optical metrology method.

5. The method of claim 1 , wherein the number of different wavelengths is between 5 and 35.

6. The method of claim 1 , wherein the different wavelengths are selected based on a model-based analysis of the sensitivity of the optical metrology process to changes in the value of an evaluation parameter.

7. 2. The method of claim 1, wherein the different wavelengths are selected from among a number of wavelengths capable of being emitted by an illumination system that illuminates the sample.

8. means for illuminating regions of the sample with a set of pulses of different wavelengths while the sample is moving at variable speeds; means for collecting light reflected from said sample as a result of said illumination to provide sets of such frames, each set of frames comprising partially overlapping frames associated with said different wavelengths; means for processing the frames to provide optical measurement results indicative of one or more evaluation parameters of an element in the area of ​​the sample, the processing being based on a mapping between the set of frames and reference measurements obtained by another optical metrology process exhibiting a higher spectral resolution than that obtained by the illumination and the collection; the illumination and collection is performed while the sample is being moved by a robot towards an integrated metrology tool, or while the sample is being moved by a robot between an integrated metrology tool and an equipment front end module (EFEM) of a processing tool; Optical measurement system for the sample.

9. The optical metrology system of claim 8 , wherein the other optical metrology process is an integrated metrology process performed by an integrated metrology tool.

10. The optical metrology system of claim 8 , wherein the mapping is provided by a machine learning process.

11. The optical metrology system of claim 10, wherein the machine learning process is trained by a training process that includes providing the machine learning process with (a) test measurements of one or more test samples obtained by other optical metrology processes and information, and (b) additional test measurements of one or more test samples obtained by the optical metrology system.

12. The optical metrology system of claim 8 , wherein the different wavelengths are selected based on a model-based analysis of the sensitivity of the optical metrology process to changes in the value of an evaluation parameter.

Images