Diversification of SEM measurement schemes to improve accuracy

Abstract

A charged particle beam inspection method includes measuring a sample in a plurality of different signal acquisition modalities. Each modality may include a different set of inspection parameters that may be optimized for different purposes such as high resolution or high acquisition speed. Measurements from different signal acquisition modalities may be combined using deconvolution or other optimization tasks to form a composite image. The composite image may achieve a higher resolution at a higher yield than a single high resolution or high speed scan achievable.

Description

[0001] CROSS-REFERENCE TO RELATED APPLICATIONS

[0002] This application claims priority to U.S. application 63 / 443,832, filed February 7, 2023, and U.S. application 63 / 452,342, filed March 15, 2023, the entire contents of which are incorporated herein by reference. Technical Field

[0003] The description herein relates to measurement schemes that may be useful in the field of charged particle beam systems, and more particularly, to systems and methods that may be applicable to charged particle inspection systems such as scanning electron microscope (SEM) tools. Background Art

[0004] Inspection and measurement systems can be used to sense physically observable phenomena. For example, a charged particle beam tool (such as an electron microscope) can include a detector for receiving charged particles projected from a sample and outputting a detection signal. The detection signal can be used to reconstruct an image of the sample structure being inspected, and can be used to, for example, reveal defects in the sample. In the manufacture of semiconductor devices, it is becoming increasingly important to accurately image and detect defects in the sample, which may include a large number of densely packed, miniaturized integrated circuit (IC) components. For this reason, an inspection system can be provided.

[0005] As semiconductor devices continue to miniaturize, inspection systems continue to balance competing parameters such as speed and accuracy. For example, some inspections may use low beam currents to achieve high resolution at the expense of low throughput and a high signal-to-noise ratio (SNR). Other inspections may use higher beam currents, resulting in higher throughput and better SNR at the expense of lower resolution. Summary of the Invention

[0006] Some embodiments of the present disclosure provide a charged particle beam inspection method. The charged particle beam inspection method may include: measuring a first area of ​​a sample using a charged particle beam inspection device in a first signal acquisition modality to obtain a first signal profile; measuring a second area of ​​the sample using the charged particle beam inspection device in a second signal acquisition modality to obtain a second signal profile, wherein the second signal acquisition modality differs from the first signal acquisition modality in terms of inspection parameters of the charged particle beam device; and generating an inspection image using an optimization task based on a synthesis of the first signal profile and the second signal profile.

[0007] Some embodiments may include a non-transitory computer-readable medium having a set of instructions stored thereon. The set of instructions may be executed by at least one processor of a device to cause the device to perform the above method.

[0008] Some embodiments of the present invention provide a charged particle beam device. The charged particle beam device may include: a charged particle beam source configured to generate a primary charged particle beam; a charged particle optical system configured to guide the primary charged particle beam to a sample surface to inspect the sample surface; a charged particle detector configured to detect charged particles returning from the sample surface; and a controller including one or more processors and configured to cause the charged particle beam device to execute. BRIEF DESCRIPTION OF THE DRAWINGS

[0009] The above and other aspects of the present disclosure will become more apparent from the following description of exemplary embodiments with reference to the attached drawings.

[0010] Figure 1 is a schematic diagram of an exemplary electron beam inspection (EBI) system consistent with embodiments of the present disclosure.

[0011] Figure 2A-2B is a diagram illustrating a charged particle beam apparatus consistent with embodiments of the present disclosure, which may be an example of an electron beam tool.

[0012] Figure 3 is a diagram of an exemplary signal acquisition modality consistent with embodiments of the present disclosure.

[0013] Figure 4 is an illustration of an example measurement acquisition scheme consistent with embodiments of the present disclosure.

[0014] Figures 5A-5C is an illustration of an example measurement acquisition scheme consistent with embodiments of the present disclosure.

[0015] Figure 6 is an illustration of an example measurement acquisition scheme consistent with embodiments of the present disclosure.

[0016] Figure 7 is a flow chart illustrating an example method that may be used to generate composite images from multiple signal acquisition modalities using an optimization task, consistent with embodiments of the present disclosure.

[0017] Figure 8 is an illustration of an exemplary application of method 700 that may be used to generate composite images from multiple signal acquisition modalities using an optimization task, consistent with embodiments of the present disclosure.

[0018] Figure 9 is a flow chart illustrating an example method that may be used to generate a composite image from multiple signal acquisition modalities consistent with embodiments of the present disclosure. DETAILED DESCRIPTION

[0019] Reference will now be made in detail to exemplary embodiments, examples of which are illustrated in the accompanying drawings. The following description relates to the accompanying drawings, in which the same numerals in different figures represent the same or similar elements, unless otherwise indicated. The implementations set forth in the following description of exemplary embodiments do not represent all implementations consistent with the present invention. On the contrary, they are merely examples of devices, systems and methods consistent with aspects related to the subject matter that can be cited in the appended claims. For example, although some embodiments are described in the context of utilizing charged particle beams (e.g., electron beams), the present disclosure is not limited thereto. Other types of charged particle beams (e.g., photon beams) can be applied similarly. In addition, other imaging systems, such as optical imaging, photoelectric detection, x-ray detection, etc. can be used.

[0020] Electronic devices consist of circuits formed on a silicon wafer called a substrate. Many circuits can be built together on the same silicon wafer, and this is called an integrated circuit, or IC. As technology advances, the size of these circuits has drastically decreased, allowing more circuits to fit onto a substrate. For example, the IC chip in a smartphone can be as small as a fingernail, yet contain over 2 billion transistors, each less than 1 / 1000 the width of a human hair.

[0021] Manufacturing these ICs from extremely small structures or components is a complex, time-consuming, and expensive process, typically involving hundreds of individual steps. An error in even one step can cause a defect in the finished IC, rendering it unusable. Therefore, one goal of the manufacturing process is to avoid such defects in order to maximize the number of functional ICs produced in the process—in other words, to improve the overall yield of the process.

[0022] An integral part of improving yield is monitoring the chip manufacturing process to ensure that it is producing sufficient numbers of functional integrated circuits. One way to monitor the process is to inspect the chip's circuit structures at various stages of their formation. The inspection can be performed using a scanning charged-particle microscope (SCPM). For example, the SCPM can be a scanning electron microscope (SEM). The SEM can be used to image these extremely small structures, in effect taking a "picture" of the structure. The image can be used to determine whether the structure was formed correctly and in the correct location. If the structure is defective, the process can be adjusted to make the defect less likely to occur again. To increase throughput (for example, the number of samples processed per hour), it is best to perform inspections as quickly as possible.

[0023] The working principle of the SEM is similar to that of a camera. A camera takes pictures by receiving and recording the intensity of light reflected or emitted from a person or object. The SEM takes "pictures" by receiving and recording the energy or number of electrons reflected or emitted from a wafer structure. Before taking such a "picture", an electron beam can be projected onto the structure, and when electrons are reflected or emitted ("emitted") from the structure (e.g., from the wafer surface, from a structure below the wafer surface, or both), the detector of the SEM can receive and record the energy or number of these electrons to generate an inspection image. In order to take such a "picture", the electron beam can scan the wafer (e.g., in a line-by-line or zigzag manner), and the detector can receive the emitted electrons from the electron beam projected area (called a "beam spot"). The detector can receive and record the emitted electrons from each beam spot at a time, and merge the information recorded for all beam spots to generate an inspection image. Some SEMs use a single electron beam (referred to as "single-beam SEMs") to capture a single "picture" to generate an inspection image, while some SEMs use multiple electron beams (also referred to as "multi-beam SEMs") to capture multiple "pictures" of the wafer in parallel. These pictures can be used individually or stitched together to generate an inspection image. By using multiple electron beams, the SEM can deliver more electron beams to the structure to acquire these multiple "pictures," resulting in more electrons being emitted from the structure. As a result, the detector can receive more emitted electrons simultaneously, generating inspection images of the wafer structure with higher efficiency and speed.

[0024] Typically, the detection process involves measuring the magnitude of the electrical signal generated when electrons land on a detector. Alternatively, electron counting can be used, where the detector counts individual electron arrival events as they occur. In either approach, the intensity of the secondary beam can be determined based on the electrical signals generated in the detector, which are proportional to changes in the intensity of the secondary beam.

[0025] Various inspection parameters can affect competing interests in the inspection process, such as speed and resolution. For example, the landing energy and angle of incidence of the electron beam can have a significant impact on the sample's interaction volume (the area in which the incident electrons interact with the sample material to generate, for example, secondary and backscattered electrons). For example, the beam current can affect the detection spot size and surface charging effects. These characteristics can be important factors in the scanning speed and effective resolution of the inspection tool. For example, the size of the interaction volume can be related to the minimum pixel size of the image generated during the inspection process, and thus the finest level of detail that can be resolved.

[0026] For example, high-energy beams can generate larger interaction volumes. This can result in a large number of emitted electrons from the sample surface over a large area. The large number of electrons may be sufficient to achieve a high SNR and faster acquisition of larger sample pixels, resulting in faster scanning speeds. However, the large size of the interaction volume also limits the minimum achievable resolution of the resulting image. On the other hand, a smaller interaction volume may reverse these costs and benefits. A small interaction volume can be achieved, for example, using a low-energy or normal-incidence beam to generate images with finer resolution. However, the lower emitted electron yield may be more difficult to distinguish from noise, resulting in smaller pixels and slower scanning times, which in turn hurts throughput. Therefore, conventional inspection systems inevitably face a trade-off between speed and resolution.

[0027] Another trade-off that impacts the speed of the inspection process is the risk of damage to the features being inspected. High beam current inspection may not be suitable for all areas of the sample, especially those containing sensitive components. A low beam current setting can be selected to avoid damage to such sensitive areas, but this may come at the expense of slower scanning speed across the sample.

[0028] Embodiments of the present disclosure may provide an inspection device and an inspection method for producing high-resolution inspection images with a high throughput rate. The inspection device may include, for example, a charged particle beam device, such as a SEM tool or other electron beam tool. The device may be configured to scan an area of ​​a sample surface under multiple signal acquisition modalities. Each signal acquisition modality may include a different set of inspection settings or other inspection parameters to produce a different signal profile from the sample surface. Different signal acquisition modalities may be optimized for different purposes, such as to achieve high acquisition speed or high resolution. Optimization may include using different inspection settings to achieve different interaction volumes or other inspection parameters.

[0029] Embodiments of the present disclosure can combine information acquired under different signal acquisition modalities to generate a high-resolution inspection image. For example, using known information about the parameters of each respective signal acquisition modality, the acquired images can be combined by performing image synthesis or deconvolution. This can be achieved, for example, by solving an optimization task. Image synthesis / deconvolution can be used to combine images or features from different signal acquisition modalities, or to identify and remove systematic noise from an image. Embodiments of the present disclosure can allow charged particle inspection images to be inferred with greater accuracy than can be achieved by other systems at the same measurement speed, or at a higher measurement speed than can be achieved by other systems at the same resolution or accuracy.

[0030] In some embodiments, the inspection device can scan an entire area of ​​the sample in a first signal acquisition modality. The inspection device can then scan an entire area of ​​the sample in a second signal acquisition modality that is different from the first. This area can include, for example, a single scan line or the full field of view of the inspection device. By measuring the same location using multiple signal acquisition modalities, more information about the area can be obtained, resulting in a higher resolution image than could be obtained using a single signal acquisition modality.

[0031] In some embodiments, the inspection device can switch between signal acquisition modalities during an area scan. For example, the inspection device can change inspection settings on a pixel-by-pixel basis. In some embodiments, the inspection device can alternate between two or more signal acquisition modalities in a repeating sequence. In some embodiments, the inspection device can alternate between two or more signal acquisition modalities in an irregular or non-repeating sequence. In some embodiments, the inspection device can alternate between two or more signal acquisition modalities that are determined or updated in a feedforward or feedback manner based on, for example, real-time measurements or predetermined information (such as pattern data or a previous scan of a reference sample).

[0032] In some embodiments, pattern-aware sampling can be used to determine appropriate regions for switching between a first signal acquisition modality and a second signal acquisition modality. For example, an initial rough scan or pattern design file can be used to identify transition regions in a circuit pattern or other inspection sample. The transition region can be, for example, the edge of a circuit pattern feature, where there is a sharp change in the wafer topography. Such edge features may require higher resolution imaging than, for example, relatively flat regions on either side of the edge feature. Thus, pattern-aware sampling can be used to switch between a first signal acquisition modality suitable for flat features and a second signal acquisition modality optimized for edge features.

[0033] The objects and advantages of the present disclosure can be achieved by the elements and combinations set forth in the embodiments discussed herein. However, the embodiments of the present disclosure are not necessarily required to achieve such exemplary objects or advantages, and some embodiments may not achieve any of the objects or advantages.

[0034] Without limiting the scope of the present disclosure, some embodiments may be described in the context of providing detection systems and detection methods in systems that utilize electron beams (e-beams). However, the present disclosure is not limited thereto. Other types of charged particle beams (such as proton beams) may also be similarly applied. In addition, the systems and methods for detection may be used in other imaging systems, such as optical imaging, photon detection, proton detection, x-ray detection, ion detection, etc. Photon detection may include infrared, visible light, ultraviolet, DUV, EUV, x-rays, or light in any other wavelength range. Therefore, although the detectors in the present disclosure may be disclosed for electron detection, some embodiments of the present disclosure may be directed to detecting other charged particles or photons.

[0035] As used herein, unless otherwise specifically stated, the term "or" includes all possible combinations unless not feasible. For example, if a component is specified to include A or B, then unless otherwise specified or not feasible, the component may include A, or B, or A and B. As a second example, if a component is specified to include A, B, or C, then unless otherwise specified or not feasible, the component may include A, or B, or C, or A and B, or A and C, or B and C, or A, B, and C.

[0036] Now refer to Figure 1 , which illustrates an exemplary electron beam inspection (EBI) system 10 that may be used for wafer inspection consistent with embodiments of the present disclosure. Figure 1 As shown, the EBI system 10 includes a main chamber 11, a load / lock chamber 20, an electron beam tool 100 (e.g., a scanning electron microscope (SEM)), and an equipment front end module (EFEM) 30. The electron beam tool 100 is located within the main chamber 11 and can be used for imaging. The EFEM 30 includes a first load port 30a and a second load port 30b. The EFEM 30 may include additional load ports. The first load port 30a and the second load port 30b receive wafer front opening pods (FOUPs), which hold wafers (e.g., semiconductor wafers or wafers made of other materials) or samples to be inspected (wafers and samples may be collectively referred to as "wafers" herein).

[0037] One or more robotic arms (not shown) in the EFEM 30 can transport the wafer to the load / lock chamber 20. The load / lock chamber 20 is connected to a load / lock vacuum pump system (not shown) that removes gas molecules in the load / lock chamber to reach a first pressure below atmospheric pressure. After reaching the first pressure, one or more robotic arms (not shown) can transport the wafer from the load / lock chamber 20 to the main chamber 11. The main chamber 11 is connected to a main chamber vacuum pump system (not shown) that removes gas molecules in the main chamber 11 to reach a second pressure below the first pressure. After reaching the second pressure, the wafer is inspected by the electron beam tool 100. The electron beam tool 100 can be a single beam system or a multi-beam system. The controller 109 is electrically connected to the electron beam tool 100 and can also be electrically connected to other components. The controller 109 can be a computer configured to perform various controls of the EBI system 10. Although the controller 109 is Figure 1 1. The controller 109 is shown as being located outside of the structure including the main chamber 11, the load / lock chamber 20, and the EFEM 30, but it will be understood that the controller 109 may be part of the structure.

[0038] In some embodiments, the controller 109 may include one or more processors (not shown). A processor may be a general or specialized electronic device capable of manipulating or processing information. For example, a processor may include any number of any combination of the following: a central processing unit (or "CPU"), a graphics processing unit (or "GPU"), an optical processor, a programmable logic controller, a microcontroller, a microprocessor, a digital signal processor, an intellectual property (IP) core, a programmable controller array (PLA), a programmable array logic (PAL), a general array logic (GAL), a complex programmable logic device (CPLD), a field programmable gate array (FPGA), a system on a chip (SoC), an application specific integrated circuit (ASIC), and any other type of circuit capable of performing data processing. A processor may also be a virtual processor, which includes one or more processors distributed across multiple machines or devices coupled via a network.

[0039] In some embodiments, the controller 109 may also include one or more memories (not shown). The memory may be a general or specific electronic device capable of storing code and data accessible by the processor (e.g., via a bus). For example, the memory may include any combination of any number of the following: random access memory (RAM), read-only memory (ROM), optical disk, magnetic disk, hard disk, solid-state drive, flash drive, secure digital (SD) card, memory stick, compact flash (CF) card, or any type of storage device. The code and data may include an operating system (OS) and one or more application programs (or "apps") for specific tasks. The memory may also be virtual memory, which includes one or more memories distributed across multiple machines or devices coupled via a network.

[0040] Charged particle beam microscope (such as formed by EBI system 10 or can be included in the microscope therein) can be able to reach the resolution of for example nanometer level, and can be used as the practical tool for checking IC components on wafer.Using electron beam system, the electron of primary electron beam can be focused on the detection point on the wafer being inspected.The interaction of primary electron and wafer may cause to form secondary particle beam.Secondary particle beam may comprise backscattered electrons, secondary electrons or Auger electrons etc. produced by the interaction of primary electron and wafer.The characteristic (for example, intensity) of secondary particle beam can change based on the property of the internal or external structure or material of wafer, and therefore can indicate whether wafer has defect.

[0041] The intensity of secondary particle beam can be determined using detector. Secondary particle beam can form beam spot on the surface of detector. Detector can generate the electric signal (for example, electric current, electric charge, voltage etc.) representing the intensity of detected secondary particle beam. Electric signal can be measured with measuring circuit device, and this measuring circuit device can comprise other components (for example, analog-to-digital converter), to obtain the distribution of detected electron. The electron distribution data collected during the detection time window are combined with the corresponding scan path data of the primary electron beam incident on the wafer surface and can be used to reconstruct the image of the wafer structure or material being inspected. Reconstructed image can be used to disclose the various features of the inside or outer structure of wafer or material, and can be used to disclose the defect that may be present in wafer.

[0042] Figure 2A An example of a charged particle beam apparatus, which may be an electron beam tool 100 , is illustrated consistent with embodiments of the present disclosure. Figure 2A An apparatus is shown that uses multiple beams formed from a primary electron beam to simultaneously scan multiple locations on a wafer.

[0043] like Figure 2AAs shown, the electron beam tool 100A may include an electron source 202, a gun aperture 204, a condenser lens 206, a primary electron beam 210 emitted from the electron source 202, a source conversion unit 212, a plurality of beamlets 214, 216 and 218 of the primary electron beam 210, a primary projection optical system 220, a wafer stage ( Figure 2A ), a plurality of secondary electron beams 236, 238, and 240, a secondary optical system 242, and an electron detection device 244. The electron source 202 can generate primary particles, such as electrons of the primary electron beam 210. A controller, an image processing system, etc. can be coupled to the electron detection device 244. The main projection optical system 220 can include a beam splitter 222, a deflection scanning unit 226, and an objective lens 228. The electron detection device 244 can include detection sub-areas 246, 248, and 250.

[0044] The electron source 202, gun aperture 204, condenser lens 206, source conversion unit 212, beam splitter 222, deflection scanning unit 226, and objective lens 228 can be aligned with the primary optical axis 260 of the apparatus 100A. The secondary optical system 242 and electron detection device 244 can be aligned with the secondary optical axis 215 of the apparatus 100A.

[0045] The electron source 202 may include a cathode, an extractor, or an anode, where primary electrons may be emitted from the cathode and extracted or accelerated to form a primary electron beam 210 having a cross (virtual or real) 208. The primary electron beam 210 may be visualized as being emitted from the cross 208. The gun aperture 204 may block peripheral electrons of the primary electron beam 210 to reduce the size of the detection spots 270, 272, and 274.

[0046] The source conversion unit 212 may include an imaging element array ( Figure 2A not shown) and the beam limiting aperture array ( Figure 2A 218 ). Examples of the source conversion unit 212 can be found in U.S. Patent No. 9,691,586; U.S. Publication No. 2017 / 0021543; and International Application No. PCT / EP2017 / 084429, all of which are incorporated herein by reference in their entireties. The imaging element array may include a microdeflector array or a microlens array. The imaging element array may form a plurality of parallel images (virtual or real) of the intersection 208 with the plurality of beams 214, 216, and 218 of the primary electron beam 210. The beam limiting aperture array may limit the plurality of beams 214, 216, and 218.

[0047] The condenser lens 206 can focus the primary electron beam 210. The current of the beams 214, 216, and 218 downstream of the source conversion unit 212 can be changed by adjusting the focusing power of the condenser lens 206 or by changing the radial size of the corresponding beam limiting apertures within the beam limiting aperture array. The condenser lens 206 can be an adjustable condenser lens, which can be configured so that the position of its first principal plane is movable. The adjustable condenser lens can be configured to be magnetic, which can cause the off-axis beams 216 and 218 to land on the beam limiting aperture at a rotation angle. The rotation angle varies with the focusing power of the adjustable condenser lens and the position of the first principal plane. In some embodiments, the adjustable condenser lens can be an adjustable anti-rotation condenser lens, which involves an anti-rotation lens with a movable first principal plane. Examples of adjustable condenser lenses are further described in U.S. Publication No. 2017 / 0021541, the entire contents of which are incorporated herein by reference.

[0048] Objective lens 228 can focus beams 214, 216, and 218 onto wafer 230 for inspection and can form a plurality of detection points 270, 272, and 274 on the surface of wafer 230. Secondary electron beams 236, 238, and 240 can be formed that are emitted from wafer 230 and return toward beam splitter 222.

[0049] The beam splitter 222 can be a Wien filter type beam splitter that generates an electrostatic dipole field and a magnetic dipole field. In some embodiments, if the electrostatic dipole field is applied, the force exerted by the electrostatic dipole field on the electrons of the beams 214, 216, and 218 can be equal in magnitude and opposite in direction to the force exerted by the magnetic dipole field on the electrons. Therefore, the beams 214, 216, and 218 can pass directly through the beam splitter 222 with a zero deflection angle. However, the total dispersion of the beams 214, 216, and 218 generated by the beam splitter 222 can also be non-zero. The beam splitter 222 can separate the secondary electron beams 236, 238, and 240 from the beams 214, 216, and 218 and direct the secondary electron streams 236, 238, and 240 to the secondary optical system 242.

[0050] The deflection scanning unit 226 can deflect the beams 214, 216, and 218 to scan detection points 270, 272, and 274 over an area on the surface of the wafer 230. In response to the beams 214, 216, and 218 being incident on the detection points 270, 272, and 274, secondary electron beams 236, 238, and 240 can be emitted from the wafer 230. The secondary electron beams 236, 238, and 240 can include electrons having an energy distribution that includes secondary electrons and backscattered electrons. The secondary optical system 242 can focus the secondary electron beams 236, 238, and 240 onto detection sub-regions 246, 248, and 250 of the electron detection device 244. The detection sub-regions 246, 248, and 250 can be configured to detect the corresponding secondary electron beams 236, 238, and 240 and generate corresponding signals for reconstructing an image of the surface of the wafer 230.

[0051] The generated signals may represent the intensities of the secondary electron beams 236, 238, and 240 and may be provided to an image processing system (e.g., FIG. 1 ) in communication with the detection device 244, the primary projection optical system 220, and the actuated wafer stage. Figure 2B The image processing system 199 provided in FIG. 1 can synchronize and coordinate the movement speed of the actuated wafer stage with the beam deflection controlled by the deflection scanning unit 226 so that the movement of the scanning probe points (e.g., scanning probe points 270, 272, and 274) can orderly cover the region of interest on the wafer 230. The parameters of this synchronization and coordination can be adjusted to accommodate different materials of the wafer 230. For example, different materials of the wafer 230 can have different resistance-capacitance characteristics, which may result in different signal sensitivities to the movement of the scanning probe points.

[0052] The intensities of the secondary electron beams 236, 238, and 240 can vary depending on the external or internal structure of the wafer 230 and, therefore, can indicate whether the wafer 230 includes defects. Furthermore, as described above, the beams 214, 216, and 218 can be projected onto different locations on the top surface of the wafer 230, or onto different sides of a local structure of the wafer 230, to generate secondary electron beams 236, 238, and 240 that can have different intensities. Therefore, by mapping the intensities of the secondary electron beams 236, 238, and 240 to regions of the wafer 230, the image processing system can reconstruct an image that reflects the characteristics of the internal or external structure of the wafer 230.

[0053] Detection sub-regions 246, 248, and 250 may comprise separate detector packages, separate sensing elements, or separate regions of an array detector. In some embodiments, each detection sub-region may comprise a single sensing element.

[0054] Now refer to Figure 2BAnother example of a charged particle beam apparatus is discussed. An electron beam tool 100B (also referred to herein as apparatus 100B) may be an example of an electron beam tool 100 and may be similar to Figure 2A The electron beam tool 100A is shown. However, unlike the tool 100A, the tool 100B may be a single beam tool that uses only one primary electron beam to scan one location on the wafer at a time.

[0055] like Figure 2B As shown, the apparatus 100B includes a wafer holder 136 supported by an actuated stage 134 for holding a wafer 150 to be inspected. The electron beam tool 100B includes an electron emitter, which may include a cathode 103, an anode 121, and a gun aperture 122. The electron beam tool 100B also includes a beam limiting aperture 125, a focusing lens 126, a columnar aperture 135, an objective lens assembly 132, and a detector 144. In some embodiments, the objective lens assembly 132 may be a modified SORIL lens, which includes a pole piece 132a, a control electrode 132b, a deflector 132c, and an excitation coil 132d. During inspection or imaging, an electron beam 161 emitted from the tip of the cathode 103 may be accelerated by the anode 121 voltage, pass through the gun aperture 122, the beam limiting aperture 125, the focusing lens 126, and be focused by the modified SORIL lens into a detection point 170, and impinge on the surface of the wafer 150. The probe point 170 can be scanned across the surface of the wafer 150 by a deflector, such as deflector 132c or other deflectors in a SORIL lens. Secondary or scattered particles emitted from the wafer surface, such as secondary electrons or scattered primary electrons, can be collected by the detector 144 to determine the intensity of the beam, and thereby, an image of the region of interest on the wafer 150 can be reconstructed.

[0056] An image processing system 199 may also be provided, comprising an image collector 120, a storage device 130, and a controller 109. The image collector 120 may include one or more processors. For example, the image collector 120 may include a computer, a server, a mainframe, a terminal, a personal computer, any type of mobile computing device, or a combination thereof. The image collector 120 may be communicatively coupled to the detector 144 of the electron beam tool 100B via a medium such as an electrical conductor, a fiber optic cable, a portable storage medium, IR, Bluetooth, the Internet, a wireless network, wireless radio, or a combination thereof. The image collector 120 may receive signals from the detector 144 and construct an image. Thus, the image collector 120 may capture an image of the wafer 150. The image collector 120 may also perform various post-processing functions, such as image averaging, generating contours, overlaying indicators on the captured image, and the like. The image collector 120 may be configured to adjust the brightness and contrast of the captured image. The storage device 130 may be a storage medium such as a hard disk, random access memory (RAM), cloud storage, or other types of computer-readable memory. The storage device 130 can be coupled to the image collector 120 and can be used to save the scanned raw image data as a raw image and save the post-processed image. The image collector 120 and the storage device 130 can be connected to the controller 109. In some embodiments, the image collector 120, the storage device 130, and the controller 109 can be integrated together as an electronic control unit.

[0057] In some embodiments, image collector 120 may capture one or more images of the sample based on the imaging signal received from detector 144. The imaging signal may correspond to a scanning operation for charged particle imaging. The captured image may be a single image including multiple imaging regions, each of which may encompass various features of wafer 150. The single image may be stored in storage device 130. Imaging may be performed based on imaging frames.

[0058] The light collector and illumination optics of an electron beam tool may include or be supplemented by an electromagnetic quadrupole electron lens. Figure 2B As shown, the electron beam tool 100B may include a first quadrupole lens 148 and a second quadrupole lens 149. In some embodiments, the quadrupole lenses may be used to control the electron beam. For example, the first quadrupole lens 148 may be controlled to adjust the beam current, and the second quadrupole lens 149 may be controlled to adjust the beam spot size and beam shape.

[0059] Figure 2BThe diagram illustrates a charged particle beam apparatus that may use a single primary beam configured to generate secondary electrons by interacting with a wafer 150. The detector 144 may be positioned along the optical axis 105, such as Figure 2B In the embodiment shown, the primary electron beam can be configured to travel along the optical axis 105. Therefore, the detector 144 can include a hole at its center that allows the primary electron beam to pass through to reach the wafer 150. Figure 2B An example of a detector 144 is shown having an opening at its center. However, some embodiments may use a detector that is positioned off-axis relative to the optical axis along which the primary electron beam travels. For example, as described above Figure 2A In the embodiment shown, a beam splitter 222 may be provided to direct the secondary electron beam toward an off-axis detector. Figure 2A As shown, the beam splitter 222 may be configured to deflect the secondary electron beam at an angle α toward the electron detection device 244 .

[0060] The detector in the charged particle beam system may include one or more sensing elements. The detector may include a single element detector or an array having multiple sensing elements. The sensing element may be configured for charged particle counting. Sensing elements of a detector that can be used for charged particle counting are discussed in U.S. Publication No. 2019 / 0379682, the entire contents of which are incorporated herein by reference.

[0061] The sensing element may include a diode or a diode-like element that can convert incident energy into a measurable signal. For example, the sensing element in the detector may include a PIN diode. In this disclosure, for example, in the figures, the sensing element may be represented as a diode, although the sensing element or other components may deviate from the ideal circuit behavior of electrical components such as diodes, resistors, capacitors, etc.

[0062] Figure 3 A first example signal acquisition modality 353 and a second example signal acquisition modality 354 consistent with embodiments of the present disclosure are illustrated. The first signal acquisition modality 353 and the second signal acquisition modality 354 may be employed in an inspection device, such as, for example, Figure 1 Electron beam tool 100, Figure 2A Electron beam tool 100A or Figure 2B The electron beam tool 100B. Figure 3 A top view and a cross-sectional view of an area of ​​a sample 350 being inspected are shown. For example, the sample 350 may be a semiconductor wafer, and the area may be the field of view of the inspection equipment. Figure 3 In, such as Figure 4-Figure 6In other example embodiments, for illustrative purposes, sample 350 (or sample 450 / 550 / 650) may include a vertical line / space pattern. The line / space pattern may include a plurality of substantially flat areas 351 separated by edge features 352. In practice, the signal acquisition modality may be applied to the inspection of any sample, including integrated circuits, other semiconductor devices, photomasks, or other samples.

[0063] In the first signal acquisition modality 353, the sample 350 may be irradiated with the charged particle beam by continuously scanning the large pixel area row by row in a fast scan direction FS. The charged particle beam and the sample may be relatively displaced in a slow scan direction SS by electrical or mechanical means to irradiate subsequent rows of the large pixel area until substantially the entire area has been examined in the first signal acquisition modality 353. The second signal acquisition modality 354 may be used in a similar manner to the first signal acquisition modality 353, but may correspond to, for example, Figure 3 The smaller pixel areas shown. The two signal acquisition modalities can be further distinguished as described below. It should be noted that for clarity, the adjacent large pixel areas of the first signal acquisition modality 353 are depicted as being in a spaced relationship. In some embodiments, the large pixel areas can be adjacent to or overlap each other in the fast scan direction FS or the slow scan direction SS. Similarly, although for clarity, the adjacent small pixel areas of the second signal acquisition modality 354 are depicted as being in a spaced relationship, in some embodiments, the small pixel areas can be adjacent to or overlap each other in the fast scan direction FS or the slow scan direction SS. It should be further understood that the pixel areas shown and their corresponding interaction volumes 355 / 356 and signal profiles 357 / 358 are highly schematic in nature and are provided for illustrative purposes.

[0064] The size of the pixel area in the first signal acquisition modality 353 or the second signal acquisition modality 354 can depend on the size and other properties of the corresponding interaction volumes 355 and 356 generated by the selected inspection parameters of the signal acquisition modality. The interaction volume can be considered the volume of material at and below the surface of the sample 350 within which incident charged particles interact with the material of the sample 350 to generate secondary charged particles. For electron beam tools, secondary charged particles can include, for example, secondary electrons, backscattered electrons, Auger electrons, etc. A larger interaction volume may produce more secondary electrons at the detector surface, originating from a relatively larger sample area. Therefore, a larger interaction volume can correspond to higher signal intensity / higher SNR and faster signal acquisition time. However, a larger interaction volume can also result in poorer imaging resolution. Therefore, an inspection scan in the first signal acquisition modality 353 may result in a lower-noise, lower-resolution signal profile 357. On the other hand, a smaller interaction volume may produce fewer secondary electrons at the detector surface, originating from a relatively smaller sample area. Therefore, a smaller interaction volume can correspond to higher imaging resolution. However, the smaller number of secondary electrons may be difficult to distinguish from system noise, resulting in a poorer SNR and longer signal acquisition time.Thus, the inspection scan in the second signal acquisition modality 354 may result in a higher noise, higher resolution signal profile 358.

[0065] Inspection tool parameters that can significantly affect the interaction volume include beam current, accelerating voltage, landing energy, and beam incidence angle. For example, a first signal acquisition modality 353 can use a relatively high beam current, a high accelerating voltage or landing energy, or a low or normal incidence angle to produce a larger interaction volume 355. A second signal acquisition modality 354 can use a relatively low beam current, a low accelerating voltage or landing energy, or a high incidence angle to produce a smaller interaction volume 356.

[0066] In some embodiments, the first signal acquisition modality 353 and the second signal acquisition modality 354 can differ in other aspects besides interaction volume, resolution, SNR, pixel size, or acquisition speed. In general, scanning the sample under any number of different inspection tool settings can produce additional valuable information about the sample given the different signal profiles produced by each modality. When subjected to an image synthesis process such as discussed below, many signal acquisition modalities can be combined to produce an enhanced, higher quality inspection image. For example, the inspection tool settings can include beam current, landing energy, accelerating voltage, beam incidence angle, detection spot size, wafer orientation / beam scan angle, field of view size and shape, beam aperture settings, lens aberration values, focus, lens / deflector or other charged particle optical settings, or other charged particle inspection tool parameters.

[0067] In addition, some embodiments of the present disclosure are described with respect to only two signal acquisition modalities, such as the first signal acquisition modality 353 and the second signal acquisition modality 354 described above. However, the embodiments of the present disclosure are not limited thereto. For example, although some measurement schemes according to embodiments of the present disclosure are discussed with respect to only the first signal acquisition modality and the second signal acquisition modality, more than two signal acquisition modalities may be used. For example, a measurement scheme may employ 2, 3, 4, ... up to any number (N) of unique signal acquisition modalities.

[0068] Figure 4 An example measurement scheme 400 consistent with embodiments of the present disclosure is illustrated. The measurement scheme 400 may be performed using an inspection device, such as, for example, Figure 1 Electron beam tool 100, Figure 2A Electron beam tool 100A or Figure 2B The electron beam tool 100B.

[0069] The measurement scheme 400 can include a first measurement of an area of ​​a sample 450 in a first signal acquisition modality 453 and a second measurement of the area in a second signal acquisition modality 454. In other words, the same area of ​​the sample 450 can be scanned in multiple signal acquisition modalities. In some embodiments, the area can include the field of view of the inspection tool. In some embodiments, the area can include a portion of the field of view, such as a single scan line, multiple scan lines, or a portion of a scan line.

[0070] For example, the measurement scheme 400 may include: performing a scanning measurement of a first scan line under a first signal acquisition modality 453; and performing a scanning measurement of the first scan line under a second signal acquisition modality 454. Some embodiments may include further scanning of the first scan line up to an Nth scanning measurement under an Nth signal acquisition modality. The sample 450 may then be displaced relative to the charged particle beam spot in a slow scan direction SS, and the process may be repeated on the second scan line until, for example, the entire field of view has been scanned under all signal acquisition modalities. In some embodiments, the entire field of view may be scanned under one signal acquisition modality and then continued to be scanned under the next signal acquisition modality.

[0071] The first signal acquisition modality 453 may be configured for, for example, lower resolution, lower noise, faster measurements (similar to Figure 3 The first signal acquisition modality 353 of FIG. 45 is configured to generate a first signal profile 457. The second signal acquisition modality 454 can be configured for, for example, higher resolution, higher noise, slower measurements (similar to Figure 3 ) to generate a second signal profile 458.

[0072] Image synthesis can be performed using the first signal profile 457 and the second signal profile 458 to produce an enhanced composite image 459 of the surface of the sample 450. The composite image 459 can be used, for example, in an inspection process such as a metrology process, mask or wafer defect inspection, etc. Image synthesis can include solving convex or non-convex optimization tasks (such as those described below with respect to Figure 7-Figure 8 In some embodiments, the optimization task may include a deconvolution task configured to reduce or eliminate system noise, aberrations, or other undesirable imaging effects.

[0073] In the above Figure 4 In the discussion of , it is said that the "same" area can be scanned in a first image acquisition modality and a second image acquisition modality. However, as Figure 3 and Figure 4 As shown, if there are substantial differences in the interaction volume and the resulting pixel area, the areas irradiated in a single scan line under different signal acquisition modalities may not be the same. As described herein, in some embodiments, if, for example, the inspection beam spot or the resulting interaction volume is substantially centered on the same area in at least one planar direction, then two signal acquisition modalities can be considered to irradiate the same area. For example, Figure 4 Two linear scans are shown having significantly different widths in the slow scan direction SS, but their center positions in the slow scan direction SS are substantially the same.Thus, the first signal acquisition modality 453 and the second signal acquisition modality 454 can be said to irradiate the same area.

[0074] Although Figure 4 While the example of can achieve a high quality inspection image 459, in some embodiments it may be desirable to scan the sample with less redundancy. For example, while some embodiments of the measurement scheme 400 may include scanning the same portion of the sample 450 under multiple signal acquisition modalities, in some embodiments it may be desirable to scan different portions of the sample surface with different sets of signal acquisition modalities. The resulting partial signal profiles can then be synthesized to form a higher resolution image with a higher yield. Figures 5A-5C Some examples of these embodiments are described.

[0075] Figures 5A-5C Example measurement schemes 500A-500C consistent with embodiments of the present disclosure are illustrated. Except as described below, measurement schemes 500A-500C can be similar to measurement scheme 400. Specifically, some embodiments of measurement schemes 500A-500C can include multimodal measurements, in which different areas of a sample 550 surface can be measured under different signal acquisition modalities.

[0076] Figure 5AThe measurement scheme 500A may include performing a multimodal measurement sequence. For example, the multimodal measurement sequence may include a sequence similar to Figure 3 Mode 353 / 354 or Figure 4 The first signal acquisition modality 553 and the second signal acquisition modality 554 of the modality 453 / 454 are shown. However, instead of scanning the entire area under a single modality (such as a full scan line or full field of view), a multimodal measurement sequence can switch between signal acquisition modalities in real time during the scan. Figure 5A The resulting signal profile is shown on the right side of FIG, as a collection of partial signal profiles 557 and 558. Partial signal profile 557 can represent those sections of sample 550 scanned under a first (e.g., lower resolution, lower noise, faster) signal acquisition modality 553. Partial signal profile 558 can represent those sections of sample 550 scanned under a second (e.g., higher resolution, higher noise, slower) signal acquisition modality 554. Partial signal profiles 557 and 558 can be combined using, for example, an optimization task to form a composite image 559. Composite image 559 can have a higher resolution than could be achieved by, for example, scanning the entire area under the first signal acquisition modality 553. Furthermore, composite image 559 can have less noise and a faster acquisition time than could be achieved by, for example, scanning the entire area under the second signal acquisition modality 554. In some embodiments, the combination of two or more partial signal profiles can produce a composite image in which substantially all image characteristics are superior to any individual signal acquisition modality.

[0077] exist Figure 5B In the measurement scheme 500B of FIG5 , the multimodal measurement sequence can alternate in the slow scan direction SS rather than the fast scan direction FS. For example, the first row can be scanned in the first signal acquisition mode 553, and the second row can be scanned in the second signal acquisition mode 554. The signal profiles 557 and 558 can then be combined to form a higher resolution, lower noise composite image 559. Figure 4 Unlike the example of scanning the same area multiple times in FIG, measurement solution 500B can sequentially scan adjacent or partially overlapping areas under different signal acquisition modalities. This can achieve enhanced images 559 with a higher throughput.

[0078] In addition, if Figure 5CAs shown, measurement scheme 500C can include multiple multimodal measurement sequences that alternate in both the fast scan direction FS and the slow scan direction SS. For example, first signal acquisition modality 553 and second signal acquisition modality 554 can alternate in multiple complementary sequences. The inspection device can, for example, irradiate a first scan line of sample 550 under a first multimodal measurement sequence and scan a second scan line of sample 550 under a second multimodal measurement sequence. In some embodiments, the first and second multimodal measurement sequences can be complementary to ensure that adjacent areas in the fast scan direction FS and the slow scan direction SS are measured under different signal acquisition modalities. For example, these sequences can be arranged to create a checkerboard or other 2D pattern.

[0079] The first multimodal measurement sequence may generate a plurality of first partial signal profiles 557a / 558a and the second multimodal measurement sequence may generate a plurality of second partial signal profiles 577b / 558b . Image synthesis may be used to merge all partial signal profiles to create a composite image 559 .

[0080] While embodiments of the present disclosure schematically depict instantaneous transitions between a first signal acquisition modality and a second signal acquisition modality, this may not always be the case in practice. In some embodiments, the change may be more gradual due to, for example, a mismatch between the scan speed in the fast scan direction FS and the time required to transition between the inspection tool settings of the first signal acquisition modality and the second signal acquisition modality. In some embodiments, such a gradual transition may include or be represented by one or more discrete signal acquisition modalities whose parameters take values ​​between the signal acquisition modalities on either side of them. For example, the act of switching between a first (lower resolution, lower noise, faster) signal acquisition modality and a second (higher resolution, higher noise, slower) signal acquisition modality may include a period that may be represented by a third (medium resolution, medium noise, medium speed) signal acquisition modality. Alternatively, the transition may include intentionally setting the inspection device to the third signal acquisition modality. Such a transition modality may allow the measurement scheme to be performed with greater knowledge of the tool settings at each exposure pixel, thereby enabling improved modeling and synthesis of the acquired signal profiles.

[0081] Figures 5A-5CThe multimodal sequences are depicted as regularly repeating and binary, but this is not necessarily the case. Some embodiments may include sequences of 2, 3, 4, ... to any number (N) of signal acquisition modalities. These sequences may be simple and monotonic (such as 1-2-3-1-2-3, 1-1-2-2-2, etc.), or more complex and oscillating (such as 1-2-1-3-1-4-1-3-1-2, etc.). In some embodiments, the multimodal measurements may have no repeating or discernible sequence. For example, the inspection device may be configured to alternate between different signal acquisition modalities in a random or pseudo-randomized manner.

[0082] Furthermore, in some embodiments, Figure 5A As shown, the multimodal measurement sequence may not correspond to the sequence of pattern features on the sample. For example, the repetition period of the multimodal measurement sequence may not coincide with the repetition period of the pattern features being inspected. This can be advantageous because it allows different sections of the repeating pattern to be scanned under different signal acquisition modalities. However, in other embodiments, the multimodal measurement sequence can be designed to conform to the pattern being inspected. For example, the multimodal measurement sequence can be configured to irradiate a flat area 551 with a first signal acquisition modality 553 and an edge feature 552 with a second signal acquisition modality 554.

[0083] In some embodiments, a multimodal measurement sequence can be designed based on known information (such as a GDS file or other pattern design data). In some embodiments, the pixel brightness measured at a first point can be used to infer information about the characteristics of the pattern being inspected at subsequent points (e.g., flat versus edge areas, material properties, surface height, or other topography). A dynamic multimodal measurement sequence can be determined or adjusted in real time or on a per-sample or per-batch basis. Dynamic determination can be based on, for example, feedforward or feedback information, a machine learning training set, or a deep learning system. In some embodiments, a reference area or reference sample can be scanned in a coarse signal acquisition mode (such as a lower resolution, faster acquisition mode 553) to identify key feature areas for scanning in a fine signal acquisition mode (such as a higher resolution, slower acquisition mode 554).

[0084] For example, Figure 6Another example measurement scheme 600 consistent with an embodiment of the present disclosure is illustrated. The measurement scheme 600 can include feature-aware sampling performed in a second signal acquisition modality 654 based on information acquired in a first signal acquisition modality 653. For example, a first scan can be performed on an area of ​​a sample 650 in the first signal acquisition modality 653 to roughly distinguish between key edge features 652 and flat areas 651. The first scan can include scanning the entire field of view or only a selected portion thereof. In some embodiments, the first scan can be performed on a different field of view or on a different sample than the second scan in the second signal acquisition modality 654. In some embodiments, a GDS file or other pattern design data can be used as an alternative to or in addition to the first scan.

[0085] Then, a second scan under the second signal acquisition modality 654 can be applied only to those areas where it is expected that key features 652 will be found. In this way, throughput can be improved by reserving the lower speed, higher resolution modality only for areas deemed necessary. The noise component in the high-resolution signal profile 658 can be mitigated by image compositing with the lower noise signal profile 657 to produce an enhanced composite image 659.

[0086] In some embodiments, pattern-aware sampling or other multimodal measurement sequences can be designed to accommodate sensitive structures on the sample surface. For example, certain components of the sample may be susceptible to damage if irradiated at higher beam currents. Therefore, any of the measurement schemes described above can be used to switch from, for example, a high-current signal acquisition modality to a low-current signal acquisition modality at sensitive structures. By reserving the slower signal acquisition modality only for areas requiring lower beam currents, throughput can be improved.

[0087] In some embodiments of the present disclosure, knowledge of the flat region 651 can be used to identify noise components in the acquired signal profile. For example, by scanning the same or similar flat region 651 of the sample surface under multiple signal acquisition modalities or at multiple locations within the field of view, aberrations and systematic noise components can be better decoupled from the measurement.

[0088] Figure 7 is a flow chart illustrating a method 700 that may be used to generate a composite image from multiple signal acquisition modalities using an optimization task, consistent with embodiments of the present disclosure. The method 700 may use, for example, Figure 1 Electron beam tool 100, Figure 2A Electron beam tool 100A or Figure 2B For example, some method steps may be performed using a controller, such as Figure 1 Controller 109 or Figure 2B In some embodiments, the method 700 may be combined with, for example, Figure 4-Figure 6 Any one of the measurement schemes 400-600 is used to perform the measurement.

[0089] In step 701, N locations on the sample can be irradiated, with each location being irradiated under m signal acquisition modalities to acquire m*N signal profiles as images of an area on the sample surface. As described above, each signal acquisition modality can correspond to a unique set of inspection tool settings. In some embodiments, different signal acquisition modalities can correspond to different areas within each of the N locations. For example, in some embodiments, the locations can correspond to a field of view of the sample, and each area can correspond to a portion of the field of view that is irradiated under a particular signal acquisition modality. Note that in some embodiments, not every location can be irradiated under the same number or type of signal acquisition modalities.

[0090] In steps 702 and 703, the m*N signal profiles acquired in step 701 can be combined with the signal acquisition modality in which the m*N signal profiles were captured to determine the numerical value of the composite image. Specifically, in step 702, a loss function can be formed based on the numerical parameters of the composite image, the acquired signal profiles, and the signal acquisition modality. In step 703, the numerical parameters of the composite image can be determined by minimizing the loss function with respect to the numerical parameters of the composite image (and optionally with respect to other parameters, as described below).

[0091] In step 704, using the composite image, an inspection process such as metrology or defect detection can be performed. For example, a determination can be made as to whether the product model meets anomaly criteria indicating the presence of a defect. Optionally, the location of the defect on the sample area can also be estimated. If a defect is detected, further metrology and / or defect detection can be performed. Alternatively or additionally, step 704 can include metrology (measurement) of the product model.

[0092] Figure 8 Further explained Figure 7 In step 701, a plurality of signal profiles may be acquired under a plurality of signal acquisition modalities. Each signal profile may constitute an acquired image of an area at a sample location. Figure 8 Three acquired signal profile images 857 are shown. For example, these images may be based on, for example Figure 4The image 857 may be acquired using the measurement scheme 400 and may correspond to an area on the sample field of view. The signal acquisition modality may include a first signal acquisition modality configured to produce a larger interaction volume and a second signal acquisition modality configured to produce a smaller interaction volume. The first signal acquisition modality may be configured for a higher acquisition speed, a higher SNR, and a lower resolution. The second signal acquisition modality may be configured for a lower acquisition speed, a lower SNR, and a higher resolution. Thus, the image 857 may be, for example, blurred and have a poor spatial resolution, or may have a higher resolution but a poor SNR. The acquired image is represented by {P k}, where k is an integer index, k=1, ...(m*N), N is the number of imaged locations, and m is the number of different signal acquisition modalities within each of the N images.

[0093] Each acquired image 857 may comprise an nxn array of pixels, where n is an integer (for simplicity, a square array of pixels is considered, but in some embodiments, the array need not be square), and each pixel may be associated with a corresponding brightness value.

[0094] The composite image in this example of method 700 may include a set of m*N images, such as Figure 8 859, which has a one-to-one correspondence with the acquired images 857. Each image 859 may include an array of pxp pixels, where p is an integer, and each pixel is associated with a brightness value. The brightness values ​​of all images 859 may be collectively formed into the numerical value of the composite image. Image 859 may be referred to as a "reconstructed" image. The reconstructed image is represented by {X k The reconstructed image 859 can have a higher resolution or SNR than the acquired image 857. Furthermore, p can be greater than n. The reconstructed image 859 can have a higher pixel resolution than the acquired image 857 because a given distance over the sampling area can span a higher number of pixels in the reconstructed image 859 than in the acquired image 857.

[0095] In this case, the imaging model can include a set 864 of convolutions (or any physics-based model capable of representing a model of the imaging system used to acquire the image 857) that are assumed to be known because each convolution 864 can correspond to the settings of a particular signal acquisition modality. Alternatively, the set 864 of convolutions can be viewed as unknown values ​​that can be learned from the available data. In this case, the set of convolutions or other models 864 can be part of the optimization problem. Each of the convolutions 864 can apply a function defined by the point spread function B. iThe point spread function can be a two-dimensional array of values ​​(kernel) followed by a pixel resolution reduction process 866 for reducing the pixel dimensions to nxn. In other words, the m*N reconstructed images 859 are images that would be identical if they were aligned with the appropriate kernel B. i Convolution, the resulting m*N arrays of corresponding convolution values ​​865 (which may also be of size pxp) are then generated, and if the pixel size of each array in the array 865 is reduced to nxn, to generate m*N images 867 that are similar to the corresponding image in the acquired image (e.g., by being blurred, having a lower spatial resolution or SNR, etc.). Here, these m*N images 867 can be referred to as "corrupted images". Note that each corrupted image 867 can correspond to one of the acquired images 857 and can be an image of the same area on the inspected sample.

[0096] In a simple form of the pixel resolution reduction process 866, p can be a multiple of n (e.g., p=an, where a is an integer), such that each pixel of the image 867 corresponds to a corresponding axa pixel patch of the convolution array 865. Thus, to perform the pixel resolution reduction process 866, the brightness value of each pixel of each corrupted image 867 can be obtained as the average of the brightness of the corresponding patch of the convolution array 865.

[0097] The reconstructed image 859 can be considered as a brightness image obtained if the sample area were imaged by an imaging process with a higher resolution and lower SNR than the imaging process that produced the corresponding image 857.

[0098] In principle, a large set of possible reconstructed images 859 has Figure 8 The properties shown (e.g., the problem of forming a reconstructed image 859 using the acquired image 857 can be ill-posed). However, this application of method 700 assumes that there is some prior knowledge about the reconstructed image 859. First, it is known that the complexity of the reconstructed image 859 is limited, with a relatively small number of top-down features visible on the patterned stack. In addition, the intended device structure can have smooth local variations (e.g., patterned lines are formed by spots of material rather than sharp features). In addition, there can be similarities between the reconstructed images 859 in that their corresponding imaging regions can contain the same or similar structures. This prior knowledge can be used to define the terms of the loss function as functions of the reconstructed image 859 (as well as the acquired image 857 and the signal acquisition modality), so that minimizing the loss function for the reconstructed image 859 can ensure that the reconstructed image 859 conforms to this prior knowledge, and also by Figure 8 The process illustrated in is matched with the measured data.

[0099] Specifically, the loss function can be in the form of:

[0100]

[0101] Here, square brackets [……] indicate the concatenation of the elements within the brackets. i represents the i-th process 866, such as applying a mask and, if necessary, reducing the pixel size. F represents the Fourier transform, F -1 P stands for inverse Fourier transform. k,i represents the kth acquired image 857 under the i-th signal acquisition mode. k represents the k-th reconstructed image 859. Therefore, F -1 (FX k ·FB i ) represents the reconstructed image X by applying the appropriate convolution 864 k The convolution value 865 array is obtained, M i F -1 (FX k ·FB i ) represents the damaged image 867. In some embodiments as described above, the set 864 may include other models based on physical properties. Thus, in general, the value array 865 may be represented by, for example, a function G(X k ,B i ) to achieve this. Note that the Fourier transform (i.e., the conversion from the spatial domain to the spatial frequency domain) is used because it is a computationally efficient method to perform the i In principle, the convolution operation can be implemented directly in the spatial domain instead of through the Fourier transform. Represents the Frobenius norm of the matrix, which is the squared l2 norm of the matrix in vector format. This quantifies the goodness of fit between the measured and reconstructed data.

[0102] is a regularization term, which includes the structural term D TV X k W represents wavelet transformation (there are several known wavelet transformations; the one used in this experiment is based on Haar wavelet). TV Indicates that X k A well-known operator that converts to the image gradient domain. It encodes this information using the differences between nearby pixels (in horizontal, vertical and / or diagonal directions). is the l1 norm, i.e. the sum of absolute values. Here, it is applied to [WX k ;D TV Xk ;X k ]This means WX k 、D TV X k and X k cascade.

[0103] D represents the matrix X k Operations that convert to vectors. ||·|| * Represents the nuclear norm operation. The norm is calculated as the sum of the absolute singular values ​​of the concatenation of all vectorized images, such as [DX1, DX2, ..., DX N ].

[0104] α and β are hyperparameters that determine the relative importance of each term in the loss function.

[0105] Then, the minimization algorithm can be expressed as:

[0106]

[0107] For all k and X k All points on the subject 0≤X k ≤M. (2) If deconvolution is performed on a single image X, the sum over k vanishes. Therefore, X k Each point in can be constrained to be in the range of 0 to an upper pixel intensity limit M. The image restoration task (e.g., obtaining a reconstructed image) can be described as a deconvolution task with smoothness and low-rank constraints.

[0108] Specifically, the item The corrupted image 867 is forced to be similar to the acquired image 857 to ensure data consistency. Note that equation (1) expresses this property in the Fourier domain because it is easier to i The convolution task is represented as a multiplication in the spatial frequency domain.

[0109] Regularization term Encourage X k Includes expected features of the reconstructed image, such as regions of uniform intensity and well-defined lines. The wavelet component WX k and based only on X k The weight of X k Has a low fill rate. Structural item D TV X k Optionally, to facilitate the recovery of horizontal and vertical lines (assuming that the x and y axes in image 857 are strongly related to the elongation directions of elongated elements in the product), D TV It can be defined as having higher weights in the vertical and horizontal directions than in the diagonal direction. A step d (e.g., D TVIt can be defined as computing the difference between horizontal / vertical pixels separated by a distance of d pixels, where d is greater than 1, rather than the nearest neighbor).

[0110] Item||[DX1, DX2, ..., DX N ]|| * Prompt request for image X k are similar because they are images of corresponding regions containing similar structures (e.g., based on the same design data, or design data that meets a similarity criterion). This requirement is encoded in Equation (1) by ensuring that the concatenated reconstructed images produce a low-rank matrix. This property is encoded via a convex relaxation of the low-rank property, namely the nuclear norm || [DX1, DX2, ..., DX N ]|| * .

[0111] The hyperparameters α and β can be selected by trail-and-error to produce a reconstructed image 859 with the desired properties. For example, if a higher resolution image is available for some product, the hyperparameters α and β can be selected to ensure the best match with that image. Note that the setting of the hyperparameters α and β may only need to be done once during the "setup" phase, so the cost of obtaining the higher resolution image may be worthwhile, and the hyperparameter values ​​thus obtained may be used in the Figure 8 The process is used to inspect more other products.

[0112] While the above-described optimization problem can produce high-quality images, in some embodiments, it may be desirable to reduce the time and computational burden required to achieve such images. For example, in some embodiments, alternative modeling using, for example, a neural network can be employed to improve the efficiency of the modeling process. A set of selected image pairs 857 / 867 can be fed to a faster neural network as training data generated from the relatively slow optimization process described above. The training set can, for example, be sampled at predetermined values ​​across the parameter space of interest. Using the initial training set, the network can construct a proxy model 864. The proxy model can then be analyzed to identify optimal conditions for generating further training image pairs 857 / 867 through a slow optimization process based on expected improvements on the training set. Additional image pairs 857 / 867 can then be used to enrich the training set, and the process can be repeated.

[0113] The proxy modeling process can be used to build a set of models 864 with much greater speed and less computational burden. After being trained, the proxy model can be applied to Figure 8 The optimization process is performed to produce enhanced synthetic images more quickly.

[0114] Further details of the optimization task are discussed in European Patent Application No. EP22185297, the entire contents of which are incorporated herein by reference.

[0115] Figure 9 is a flow chart illustrating a method 900 that may be used to generate a composite image from multiple signal acquisition modalities consistent with embodiments of the present disclosure. The method 900 may use, for example, Figure 1 Electron beam tool 100, Figure 2A Electron beam tool 100A or Figure 2B For example, some method steps may be performed using a controller, such as Figure 1 Controller 109 or Figure 2B In some embodiments, the method 900 may be combined with, for example, Figure 4-Figure 6 Any one of the measurement schemes 400-600 is used to perform the measurement.

[0116] In step 901, an inspection tool may measure a first area of ​​a sample in a first signal acquisition modality. The first area of ​​the sample may be, for example, the entire sample, a die area of ​​the sample, a field of view of the inspection tool on the sample, a scan line on the sample, or a portion of a scan line. The first signal acquisition modality may be a collection of inspection tool settings. The inspection tool settings may include, for example, beam current, landing energy, acceleration voltage, beam incidence angle, detection spot size, wafer orientation / beam scan angle, field of view size and shape, beam aperture setting, lens aberration values, focus, lens / deflector or other charged particle optical settings, or other charged particle inspection tool parameters. The first signal acquisition modality may be configured to produce a desired set of imaging parameters. For example, the first signal acquisition modality may be configured to produce a relatively large interaction volume, or may be configured to produce, for example, a signal profile with a higher acquisition speed, a higher SNR, or a lower resolution.

[0117] At step 902, the inspection tool may measure a second region of the sample in at least one inspection tool setting under a second signal acquisition modality different from the first signal acquisition modality. For example, in some embodiments, the second signal acquisition modality may be configured to produce a relatively smaller interaction volume, or may be configured to produce, for example, a lower acquisition speed, a lower SNR, or a higher resolution signal profile.

[0118] In some embodiments, the first region and the second region may be the same. For example, the inspection tool may scan an entire line in each of the first and second signal acquisition modalities, then proceed to the next line, and may continue until the entire field of view is exposed in both signal acquisition modalities. In this case, individual lines, the entire field of view, etc. may be considered the first and second regions. Figure 4An example of the above embodiment can be seen in the measurement acquisition scheme 400 of FIG.

[0119] In some embodiments, the first region and the second region can be different. For example, the first region and the second region can be adjacent, can be spaced apart from each other, or can slightly overlap. For example, the first region and the second region can correspond to different portions of one or more scan lines. The inspection tool can be configured to scan in a scanning direction (e.g., Figure 4-Figure 6 A single scan line may be irradiated in a multimodal measurement sequence in a fast scan direction (FS) as shown in FIG. 1 while alternating between a first signal acquisition modality and a second signal acquisition modality. In this case, the portion of the line irradiated in the first signal acquisition modality may correspond to a first region, and the portion of the line irradiated in the second signal acquisition modality may correspond to a second region. Alternatively, the inspection tool may scan multiple lines while simultaneously scanning in a non-scanning direction (such as FIG. Figure 4-Figure 6 In some embodiments, the inspection tool may employ multiple multimodal sequences on different scan lines in the field of view. Examples of the above embodiments may be employed in Figures 5A-5C The measurement acquisition schemes 500A-500C are seen.

[0120] In some embodiments, the second region can be included within the first region. For example, the inspection tool can irradiate substantially the entire line or the entire field of view in the first signal acquisition modality to obtain a rough measurement of the first region (such as a higher speed, lower resolution measurement). Using information about the key features (such as information obtained from the first scan, or known information such as previous scans, GDS files, or other pattern design data), the inspection tool can irradiate the area where the key features are expected to be found.

[0121] The measurements performed in steps 901 and 902 may produce a signal profile corresponding to an image of the measurement area.

[0122] In step 903, the signal profiles acquired in steps 901 and 902 may be synthesized to form an enhanced, higher quality image of a portion of the sample. For example, synthesis may include performing an optimization task to find a solution for the combination of signal profiles. In some embodiments, synthesis may be described as a deconvolution task. Examples of optimization tasks may include the above combined Figure 7 and Figure 8 Tasks discussed. In some embodiments, deep learning or machine learning techniques can be used to synthesize images.

[0123] The composite image may be used to perform an inspection process at step 904. For example, the composite image may be analyzed to perform defect detection, metrology operations, or other sample inspection processes.

[0124] A non-transitory computer-readable medium consistent with embodiments of the present disclosure may be provided that stores instructions for a controller (e.g., Figure 1 Controller 109 or Figure 2B The processor of the image acquisition unit 199) in the embodiment of the present invention is configured according to the embodiment of the present invention. Figure 4-Figure 6 Measurement acquisition solution 400-600, Figure 7 Example flowchart 700 or Figure 9 900 to detect charged particles. For example, instructions stored in a non-transitory computer-readable medium can be executed by circuitry of a controller to partially or fully perform the measurement acquisition schemes 400-600 or the methods 700 or 900. For example, common forms of non-transitory media include floppy disks, flexible disks, hard disks, solid-state drives, magnetic tapes or any other magnetic data storage media, compact disk read-only memories (CD-ROMs), any other optical data storage media, any physical media with a pattern of holes, random access memories (RAMs), programmable read-only memories (PROMs) and erasable programmable read-only memories (EPROMs), FLASH-EPROMs or any other flash memory, non-volatile random access memories (NVRAMs), caches, registers, any other memory chips or cartridges, and networked versions thereof.

[0125] The embodiments of the present disclosure may also be described in the following sections:

[0126] 1. A non-transitory computer-readable medium storing a set of instructions, the set of instructions being executable by at least one processor of a device to cause the device to perform a method, the method comprising:

[0127] measuring a first region of the sample with a charged particle beam inspection device in a first signal acquisition modality to obtain a first signal profile;

[0128] measuring a second region of the sample with the charged particle beam inspection apparatus in a second signal acquisition modality to obtain a second signal profile, the second signal acquisition modality differing from the first signal acquisition modality in terms of inspection parameters of the charged particle beam apparatus; and

[0129] An inspection image is generated using an optimization task based on a synthesis of the first signal profile and the second signal profile.

[0130] 2. The non-transitory computer-readable medium of Section 1, wherein the first region and the second region are the same region.

[0131] 3. The non-transitory computer-readable medium of Section 1, wherein the first region is different from the second region.

[0132] 4. The non-transitory computer-readable medium of Section 3, wherein the first region and the second region do not overlap.

[0133] 5. The non-transitory computer-readable medium of Section 1, wherein the first region comprises a field of view of the charged particle beam inspection device.

[0134] 6. The non-transitory computer-readable medium of Section 1, wherein:

[0135] The first region includes a first scan line in the field of view of the charged particle beam inspection apparatus, and

[0136] The second region includes a second scan line in the field of view of the charged particle beam inspection apparatus, the second scan line being different from the first scan line.

[0137] 7. The non-transitory computer-readable medium of Section 1, wherein:

[0138] The first region includes a first portion of a first scan line in the field of view of the charged particle beam inspection apparatus, and

[0139] The second region includes a second portion of the first scan line in the field of view of the charged particle beam inspection apparatus, the second portion being different from the first portion.

[0140] 8. The non-transitory computer-readable medium of Section 7, wherein the first region and the second region of the first scan line correspond to a first multimodal measurement sequence.

[0141] 9. The non-transitory computer-readable medium of Section 8, wherein the set of instructions executable by the at least one processor causes the device to further perform:

[0142] measuring a third region of the sample using the charged particle beam inspection device in the first signal acquisition modality to obtain a third signal profile; and

[0143] measuring a fourth region of the sample using the charged particle beam inspection device in the second signal acquisition mode to obtain a fourth signal profile,

[0144] The generation of the inspection image is further based on a synthesis of the third signal profile and the fourth signal profile.

[0145] 10. The non-transitory computer-readable medium of Section 9, wherein:

[0146] The third region includes a third portion of a second scan line in the field of view of the charged particle beam inspection apparatus, the second scan line being different from the first scan line, and

[0147] The fourth region includes a fourth portion of the second scan line in the field of view of the charged particle beam inspection apparatus, the fourth portion being different from the third portion.

[0148] 11. The non-transitory computer-readable medium of Section 10, wherein the third region and the fourth region of the second scan line correspond to a second multimodal measurement sequence that is different from the first multimodal measurement sequence.

[0149] 12. The non-transitory computer-readable medium of Section 8, wherein the multimodal measurement sequence does not correspond to a sequence of pattern features on the first region or the second region.

[0150] 13. The non-transitory computer-readable medium of Section 8, wherein the multimodal measurement sequence corresponds to a sequence of pattern features on the first region or the second region.

[0151] 14. The non-transitory computer-readable medium of Section 13, wherein the multimodal measurement sequence is based on a priori information of the pattern features on the first area or the second area.

[0152] 15. The non-transitory computer-readable medium of Section 13, wherein the at least one processor is configured to cause the device to further perform:

[0153] The multimodal measurement sequence is updated during the measurement of the first area or the second area based on information acquired from the first area or the second area.

[0154] 16. The non-transitory computer-readable medium of Chapter 1, wherein the second region corresponds to sample features within the first region.

[0155] 17. The non-transitory computer-readable medium of Chapter 16, wherein the sample features include expected locations of pattern edge features.

[0156] 18. The non-transitory computer-readable medium of Section 1, wherein the first signal acquisition modality is configured to generate a larger interaction volume in the sample than the second signal acquisition modality.

[0157] 19. The non-transitory computer-readable medium of Section 1, wherein the first signal acquisition modality is configured to achieve one of a higher signal acquisition speed, a higher signal-to-noise ratio, or a lower resolution than the second signal acquisition modality.

[0158] 20. A non-transitory computer-readable medium according to Chapter 1, wherein the inspection parameters of the charged particle beam device include one of beam current, landing energy, acceleration voltage, beam incidence angle, detection point size, chip orientation, beam scanning angle, field of view size, field of view shape, beam hole setting, lens aberration value, focus value and charged particle optical setting.

[0159] 21. The non-transitory computer-readable medium of Chapter 1, wherein the optimization task comprises a loss function.

[0160] 22. The non-transitory computer-readable medium of Chapter 1, wherein the optimization task comprises an inverse problem.

[0161] 23. The non-transitory computer-readable medium of Chapter 22, wherein the optimization task comprises a deconvolution task.

[0162] 24. A charged particle beam inspection method, comprising:

[0163] measuring a first region of the sample with a charged particle beam inspection device in a first signal acquisition modality to obtain a first signal profile;

[0164] measuring a second region of the sample with the charged particle beam inspection apparatus in a second signal acquisition modality to obtain a second signal profile, the second signal acquisition modality differing from the first signal acquisition modality in terms of inspection parameters of the charged particle beam apparatus; and

[0165] An inspection image is generated using an optimization task based on a synthesis of the first signal profile and the second signal profile.

[0166] 25. The method according to Chapter 24, wherein the first region and the second region are the same region.

[0167] 26. The method of Chapter 24, wherein the first region is different from the second region.

[0168] 27. The method of Chapter 26, wherein the first region and the second region do not overlap.

[0169] 28. The method of Chapter 24, wherein the first region comprises a field of view of the charged particle beam inspection device.

[0170] 29. The method of Section 24, wherein:

[0171] The first region includes a first scan line in the field of view of the charged particle beam inspection apparatus, and

[0172] The second region includes a second scan line in the field of view of the charged particle beam inspection apparatus, the second scan line being different from the first scan line.

[0173] 30. The method of Section 24, wherein:

[0174] The first region includes a first portion of a first scan line in the field of view of the charged particle beam inspection apparatus, and

[0175] The second region includes a second portion of the first scan line in the field of view of the charged particle beam inspection apparatus, the second portion being different from the first portion.

[0176] 31. The method according to Section 30, wherein the first region and the second region of the first scan line correspond to a first multimodal measurement sequence.

[0177] 32. The method according to Section 31, further comprising:

[0178] measuring a third region of the sample using the charged particle beam inspection device in the first signal acquisition modality to obtain a third signal profile; and

[0179] measuring a fourth region of the sample using the charged particle beam inspection device in the second signal acquisition mode to obtain a fourth signal profile,

[0180] The generation of the inspection image is further based on a synthesis of the third signal profile and the fourth signal profile.

[0181] 33. The method of clause 32, wherein:

[0182] The third region includes a third portion of a second scan line in the field of view of the charged particle beam inspection apparatus, the second scan line being different from the first scan line, and

[0183] The fourth region includes a fourth portion of the second scan line in the field of view of the charged particle beam inspection apparatus, the fourth portion being different from the third portion.

[0184] 34. The method according to Section 33, wherein the third region and the fourth region of the second scan line correspond to a second multimodal measurement sequence, the second multimodal measurement sequence being different from the first multimodal measurement sequence.

[0185] 35. The method of Section 31, wherein the multimodal measurement sequence does not correspond to a sequence of pattern features on the first region or the second region.

[0186] 36. The method of Chapter 31, wherein the multimodal measurement sequence corresponds to a sequence of pattern features on the first area or the second area.

[0187] 37. The method according to Chapter 36, wherein the multimodal measurement sequence is based on prior information of the pattern features on the first area or the second area.

[0188] 38. The method according to Section 36, further comprising:

[0189] The multimodal measurement sequence is updated during the measurement of the first area or the second area based on information acquired from the first area or the second area.

[0190] 39. A method according to Chapter 24, wherein the second region corresponds to sample features within the first region.

[0191] 40. The method of Chapter 39, wherein the sample features include expected locations of pattern edge features.

[0192] 41. The method of Chapter 24, wherein the first signal acquisition modality is configured to generate a larger interaction volume in the sample than the second signal acquisition modality.

[0193] 42. The method of Chapter 24, wherein the first signal acquisition modality is configured to achieve one of a higher signal acquisition speed, a higher signal-to-noise ratio, or a lower resolution than the second signal acquisition modality.

[0194] 43. A method according to Chapter 24, wherein the inspection parameters of the charged particle beam device include beam current, landing energy, acceleration voltage, beam incidence angle, detection point size, chip orientation, beam scanning angle, field of view size, field of view shape, beam hole setting, lens aberration value, focus value and one of the charged particle optical settings.

[0195] 44. A method according to Chapter 24, wherein the optimization task includes a loss function.

[0196] 45. A method according to Chapter 24, wherein the optimization task comprises an inverse problem.

[0197] 46. ​​A method according to Chapter 45, wherein the optimization task includes a deconvolution task.

[0198] 47. A charged particle beam apparatus comprising:

[0199] a charged particle beam source configured to generate a primary charged particle beam;

[0200] a charged particle optical system configured to direct the primary charged particle beam to a sample surface to inspect the sample surface;

[0201] a charged particle detector configured to detect charged particles returning from the sample surface; and

[0202] A controller comprising one or more processors and configured to cause the charged particle beam device to perform:

[0203] measuring a first region of the sample with a charged particle beam inspection device in a first signal acquisition modality to obtain a first signal profile;

[0204] measuring a second region of the sample with the charged particle beam inspection apparatus in a second signal acquisition modality to obtain a second signal profile, the second signal acquisition modality differing from the first signal acquisition modality in terms of inspection parameters of the charged particle beam apparatus; and

[0205] An inspection image is generated using an optimization task based on a synthesis of the first signal profile and the second signal profile.

[0206] 48. A charged particle beam device according to Chapter 47, wherein the first region and the second region are the same region.

[0207] 49. A charged particle beam device according to Chapter 47, wherein the first region is different from the second region.

[0208] 50. A charged particle beam device according to Chapter 49, wherein the first region and the second region do not overlap.

[0209] 51. A charged particle beam device according to Chapter 47, wherein the first region includes a field of view of the charged particle beam inspection device.

[0210] 52. The charged particle beam apparatus of Section 47, wherein:

[0211] The first region includes a first scan line in the field of view of the charged particle beam inspection apparatus, and

[0212] The second region includes a second scan line in the field of view of the charged particle beam inspection apparatus, the second scan line being different from the first scan line.

[0213] 53. The charged particle beam apparatus of Section 47, wherein:

[0214] The first region includes a first portion of a first scan line in the field of view of the charged particle beam inspection apparatus, and

[0215] The second region includes a second portion of the first scan line in the field of view of the charged particle beam inspection apparatus, the second portion being different from the first portion.

[0216] 54. A charged particle beam device according to Chapter 53, wherein the first region and the second region of the first scan line correspond to a first multimodal measurement sequence.

[0217] 55. The charged particle beam device according to Section 54, wherein the controller is configured to cause the charged particle beam device to further perform:

[0218] measuring a third region of the sample using the charged particle beam inspection device in the first signal acquisition modality to obtain a third signal profile; and

[0219] measuring a fourth region of the sample using the charged particle beam inspection device in the second signal acquisition mode to obtain a fourth signal profile,

[0220] The generation of the inspection image is further based on a synthesis of the third signal profile and the fourth signal profile.

[0221] 56. The charged particle beam apparatus of Section 55, wherein:

[0222] The third region includes a third portion of a second scan line in the field of view of the charged particle beam inspection apparatus, the second scan line being different from the first scan line, and

[0223] The fourth region includes a fourth portion of the second scan line in the field of view of the charged particle beam inspection apparatus, the fourth portion being different from the third portion.

[0224] 57. The charged particle beam device according to Chapter 56, wherein the third region and the fourth region of the second scan line correspond to a second multimodal measurement sequence, which is different from the first multimodal measurement sequence.

[0225] 58. A charged particle beam device according to Chapter 54, wherein the multimodal measurement sequence does not correspond to a pattern feature sequence on the first area or the second area.

[0226] 59. A charged particle beam device according to Chapter 54, wherein the multimodal measurement sequence corresponds to a pattern feature sequence on the first area or the second area.

[0227] 60. The charged particle beam apparatus according to Chapter 59, wherein the multimodal measurement sequence is based on a priori information of the pattern features on the first region or the second region.

[0228] 61. The charged particle beam apparatus according to Section 59, wherein the controller is configured to cause the charged particle beam apparatus to further perform:

[0229] The multimodal measurement sequence is updated during the measurement of the first area or the second area based on information acquired from the first area or the second area.

[0230] 62. A charged particle beam apparatus according to Chapter 47, wherein the second region corresponds to sample features within the first region.

[0231] 63. A charged particle beam apparatus according to Chapter 62, wherein the sample features include expected positions of pattern edge features.

[0232] 64. A charged particle beam device according to Chapter 47, wherein the first signal acquisition modality is configured to generate a larger interaction volume in the sample than the second signal acquisition modality.

[0233] 65. A charged particle beam device according to Chapter 47, wherein the first signal acquisition modality is configured to achieve one of a higher signal acquisition speed, a higher signal-to-noise ratio, or a lower resolution than the second signal acquisition modality.

[0234] 66. A charged particle beam device according to Chapter 47, wherein the inspection parameters of the charged particle beam device include one of beam current, landing energy, acceleration voltage, beam incidence angle, detection point size, chip orientation, beam scanning angle, field of view size, field of view shape, beam hole setting, lens aberration value, focus value and charged particle optical setting.

[0235] 67. A charged particle beam apparatus according to Chapter 47, wherein the optimization task includes a loss function.

[0236] 68. A charged particle beam apparatus according to Chapter 47, wherein the optimization task comprises an inverse problem.

[0237] 69. A charged particle beam device according to Chapter 68, wherein the optimization task includes a deconvolution task.

[0238] The block diagrams in the figures may illustrate the architecture, functions and operations that may be implemented according to the systems, methods and computer hardware or software products of various exemplary embodiments of the present disclosure. In this regard, each box in the schematic diagram may represent certain arithmetic or logical operations that can be implemented using hardware such as electronic circuits. The box may also represent a code module, segment or portion that includes one or more executable instructions for implementing a specified logical function. It should be understood that in some alternative implementations, the functions indicated in the block may not appear in the order shown in the figure. For example, depending on the functions involved, two boxes shown in succession may be executed or implemented substantially simultaneously, or the two boxes may sometimes be executed in the opposite order. Some boxes may also be omitted. It should also be understood that each box in the block diagram and the combination of boxes may be implemented by a dedicated hardware-based system that performs the specified function or action, or by a combination of dedicated hardware and computer instructions.

[0239] It should be understood that the embodiments of the present disclosure are not limited to the exact configurations described above and shown in the accompanying drawings, but may be modified and altered without departing from the scope thereof. For example, the charged particle inspection system may be only one example of a charged particle beam system consistent with embodiments of the present disclosure.

Claims

1. A non-transitory computer-readable medium storing a set of instructions, the set of instructions being executable by at least one processor of a device to cause the device to perform a method, the method comprising: measuring a first region of the sample using a charged particle beam inspection device in a first signal acquisition modality to obtain a first signal profile; measuring a second region of the sample with the charged particle beam inspection device in a second signal acquisition modality to obtain a second signal profile, the second signal acquisition modality differing from the first signal acquisition modality in terms of inspection parameters of the charged particle beam device; as well as An inspection image is generated using an optimization task based on a synthesis of the first signal profile and the second signal profile. 2 . The non-transitory computer-readable medium of claim 1 , wherein the first region and the second region are the same region. The non-transitory computer-readable medium of claim 1 , wherein the first region is different from the second region. The non-transitory computer-readable medium of claim 3 , wherein the first region and the second region do not overlap. 5 . The non-transitory computer readable medium of claim 1 , wherein the first region includes a field of view of the charged particle beam inspection device.

6. The non-transitory computer-readable medium of claim 1 , wherein: The first region includes a first scan line in the field of view of the charged particle beam inspection apparatus, and The second region includes a second scan line in the field of view of the charged particle beam inspection apparatus, the second scan line being different from the first scan line.

7. The non-transitory computer-readable medium of claim 1 , wherein: The first region includes a first portion of a first scan line in the field of view of the charged particle beam inspection apparatus, and The second region includes a second portion of the first scan line in the field of view of the charged particle beam inspection apparatus, the second portion being different from the first portion. 8 . The non-transitory computer readable medium of claim 7 , wherein the first region and the second region of the first scan line correspond to a first multimodal measurement sequence.

9. The non-transitory computer-readable medium of claim 8, wherein the set of instructions executable by the at least one processor causes the device to further perform: measuring a third region of the sample using the charged particle beam inspection device in the first signal acquisition modality to obtain a third signal profile; and measuring a fourth region of the sample using the charged particle beam inspection device in the second signal acquisition mode to obtain a fourth signal profile, The generation of the inspection image is further based on a synthesis of the third signal profile and the fourth signal profile.

10. The non-transitory computer-readable medium of claim 9, wherein: The third region includes a third portion of a second scan line in the field of view of the charged particle beam inspection apparatus, the second scan line being different from the first scan line, and The fourth region includes a fourth portion of the second scan line in the field of view of the charged particle beam inspection apparatus, the fourth portion being different from the third portion. 11 . The non-transitory computer readable medium of claim 10 , wherein the third region and the fourth region of the second scan line correspond to a second multimodal measurement sequence that is different from the first multimodal measurement sequence. 12 . The non-transitory computer-readable medium of claim 8 , wherein the multimodal measurement sequence does not correspond to a sequence of pattern features on the first region or the second region. 13 . The non-transitory computer-readable medium of claim 8 , wherein the multimodal measurement sequence corresponds to a sequence of pattern features on the first area or the second area. 14 . The non-transitory computer-readable medium of claim 13 , wherein the multimodal measurement sequence is based on a priori information of the pattern features on the first area or the second area.

15. The non-transitory computer-readable medium of claim 13, wherein the at least one processor is configured to cause the device to further perform: The multimodal measurement sequence is updated during the measurement of the first area or the second area based on information acquired from the first area or the second area.

Images